Puswhisperer 5

by Mark Crislip

Contents

Dedication

To my brother David, as I have run out of people to dedicate a book to.

Foreword

Yet another volume of interesting and unusual infectious disease cases.

Pus never ends.

Chapter One

Feel Bad Yet?

The patient is a middle-aged male with lung cancer, getting his first round of chemotherapy. He comes in with severe nausea, vomiting, sore mouth, headache, and fevers. He is admitted to the hospital and started on the usual therapy and mostly gets better.

The call me when the fevers persist, barely, when the white cell count returns after falling during the chemo. 38 ºC is a fever. A review of systems is negative except for an improving headache with photophobia. But he doesn't appear ill and there is no stiff neck.

The white cell count is now 3000 cells per microliter, edging towards the lower limit of normal which is 4500. The maximum temperature is 38 degrees, and the swab of the mouth shows herpes simplex virus (HSV-1). A CT scan of the head is negative. He is reasonably upset that he has a herpes infection in the mouth and wonder where it came from.

"So," I explained, "many Americans have herpes but most never have symptoms. It was your immune suppression from the chemo that led to it reactivating." Unless, I suppose, he had been kissing his oncologist---and while I quite like said physician, he is not particularly smoochable from my perspective.

The more difficult issue is the headache. It is probably due to HSV meningitis, and so he has been on intravenous acyclovir, an antiviral, for 2 days by the time I see him and is much improved. The headache could, of course, have yet another etiology, but it is hard to suggest a spinal tap for someone who is improving.

Where to go from here? There are zero standards for the treatment of HSV meningitis, a common manifestation of herpes viri.

"HSV-2 meningitis presents most often without a history of genital herpes, recurrent meningitis, or genital symptoms. Current management practices are highly variable and may lead to unnecessary hospitalization and prolonged intravenous therapy."

And suppressive therapy not only doesn't work but makes things worse:

"Suppressive treatment with 0.5 g of valacyclovir twice daily was not shown to prohibit recurrent meningitis and cannot be recommended for this purpose after HSV meningitis in general. Protection against mucocutaneous lesions was observed, but the dosage was probably inappropriate for the prevention of HSV activation in the central nervous system. The higher frequency of meningitis, after cessation of active drug, could be interpreted as a rebound phenomenon."

Although both studies were in normal hosts and with HSV-2. There is a lot less information with HSV- 1. Probably the same, but who knows. How long to treat intravenously? I'll stop it after an arbitrary duration when he feels better and has good immune reconstitution.

Should we give him chronic suppressive therapy? I don't know. To judge from the HSV-2 study, I am going to vote no, but I am uncertain. It would help the oral disease but perhaps worsen the central nervous system infection.

My partner suggested that when you make an arbitrary decision based on inadequate data and extrapolation, you know you have done the correct thing if you feel bad about it. There is some truth to that. I feel less bad about no long-term prophylaxis than I would if I sent him out on long-term acyclovir. Maybe I will split the difference and suggest acyclovir around chemo times. Then I can fell doubly bad.

Rationalization

Am J Med. 2009 Jul;122(7):688-91. Herpes simplex type-2 meningitis: presentation and lack of standardized therapy.

http://www.ncbi.nlm.nih.gov/pubmed/19559173

Clin Infect Dis. 2012 May;54(9):1304-13. Epub 2012 Mar 28. Long-term valacyclovir suppressive treatment after herpes simplex virus type 2 meningitis: a double-blind, randomized controlled trial.

http://www.ncbi.nlm.nih.gov/pubmed/22460966

POLL RESULTS

Making arbitrary decisions

• is done arbitrarily. 26%

• makes me uneasy. 29%

• is why I get the heavy discounted Medicare bucks. 6%

• MY decisions are not arbitrary, I ALWAYS know what do to. 9%

• Harvey Dent is my role model, so no problemo. 24%

• Other Answers 6%

  • makes the day go by quickly.

  • is what I get paid to do and my consulters apparently get paid to ignore. It's been a hard week.

Chapter Two

Things Fall Apart

Things fall apart; the centre cannot hold;

Mere anarchy is loosed upon the world,

The blood-dimmed tide is loosed, and everywhere

Are emboli.

William Butler Yeats. Sort of.

Some diseases are simply disasters, and there is little that can be done to mitigate the inevitable complications. You do your best and fail.

The patient is an elderly female who is admitted with syncopal episodes secondary to what looks to be a supraventricular tachycardia (SVT), or an abnormally high heart rate. This is worrisome since the patient has both a pacemaker and a prosthetic mitral valve. As best as can be determined, she has had no constitutional symptoms such as fevers, chills, or sweats. She lives an uneventful life in a care facility. The valve and pacemaker are about three years old, and as part of the evaluation, the patient gets an echocardiogram.

There is a vegetation on the valve and the wire. Not good. And it gets worse when the blood cultures are negative at outside hospital after three days. Crap. Culture-negative endocarditis is not good. Prosthetic valve culture-negative endocarditis is even worse. Bad bugs in a bad place:

"Among 759 patients with BCNE, a causative microorganism was identified in 62.7%, and a noninfective etiology in 2.5%. Blood was the most useful specimen, providing a diagnosis for 47.7% of patients by serological analysis (mainly Q fever and Bartonella infections). Broad-range polymerase chain reaction (PCR) of blood and Bartonella-specific Western blot methods diagnosed 7 additional cases. PCR of valvular biopsies identified 109 more etiologies, mostly streptococci, Tropheryma whipplei, Bartonella species, and fungi."

There is nothing on the physical exam to suggest endocarditis, so perhaps this is acute. On the other hand, she developed this vegetation while on the anti-coagulant warfarin, so it may be of longer duration since it takes time to make a 1.5 cm clot.

With the lack of risks, I am expecting the cultures to grow Abiotrophia or what's called HACEK: Haemophilus, Aggregatibacter (previously Actinobacillus), Cardiobacterium, Eikenella, Kingella, all fastidious Gram-negative organisms. But to be complete, I send off all the culture-negative endocarditis serologies. And the next day the blood cultures grow yeast. Double crap. Time for amphotericin and 5-fluorouracil, and the next day, while we are rounding up the surgeons, it is identified as C. albicans. Odd organism; there are fewer than two dozen cases of prosthetic valve infections reported in the literature:

"The overall mortality rate was 46.6%. Mortality was higher in patients with PVE (5 of 8 cases, 62.5%) than in patients with NVE (2 of 5 patients, 40%). A better outcome was observed in patients treated with a combined medical and surgical therapy."

And no good reason according to the history for a Candida valve infection.

Then triple crap: she becomes hemiparetic, and a CT scan shows too-numerous-to-count emboli with early herniation. I am sure the anticoagulation didn't help, and the vegetation broke apart and showered the brain. Emboli have usually been the reason for death in my endocarditis patients. One was on the last day of therapy for streptococcal endocarditis with a negative echocardiogram, when I told the patient everything looked good for a cure. She went home, had a massive embolic event, and died.

The reported mortality of endocarditis is high. I should not be surprised when it happens. Yet every time I am. And every time I say crap. Or worse.

Rationalization

Clinical Infectious Diseases Volume 51, Issue 2Pp. 131-140. Comprehensive Diagnostic Strategy for Blood Culture-Negative Endocarditis: A Prospective Study of 819 New Cases

http://cid.oxfordjournals.org/content/51/2/131.full

Candida prosthetic valve endocarditis: prospective study of six cases and review of the literature. Nguyen MH, Nguyen ML, Yu VL, McMahon D, Keys TF, Amidi M.Clin Infect Dis. 1996 Feb;22(2):262-7. Review

https://www.ncbi.nlm.nih.gov/pubmed/8838182

Medicine (Baltimore). 2009 May;88(3):160-8. Candida infective endocarditis: report of 15 cases from a prospective multicenter study.

http://www.ncbi.nlm.nih.gov/pubmed/19440119

Chapter Three

I'm back but do not want to be

Four days in Vegas with my wife, our first vacation in 19 years without a kid. So much fun and, as a tree-hugging Oregonian, I am always surprised how much I like Vegas. I just want to be back poolside in the heat.

I started the day with a lecture to the second-year medical students on endocarditis, and there was one student who glowered at me the whole time. Boredom I am used to. Laughter often. But glowering? A little unsettling.

But back to work, looking at the electronic medical records (EMR) and trying vaguely to remember how to order a CAT scan and spell iliopsoas so I can link a diagnosis to a test and blah blah blah.

It is always a challenge to pick up a service after time away and, if I may continue to whinge, the EMR has completely lost the ability to document the patient narrative. It is much more difficult to discover from the electronic chart what a patient really does or does not have, as opposed to what we are treating.

One of the patients has bilateral nodular infiltrates. He is on biologics for rheumatoid arthritis and gets the full workup for infection. The only thing positive is the galactomannan, a test for Aspergillus, on the bronchoscopy---and the test is not much of a positive at that. No cultures, not a classic CT scan, nothing else to suggest the bronchoscopy result is either a false positive or a true positive. A lifelong two-pack-a-day smoker, he might have enough airway colonization by Aspergillus to have a true positive test that is unimportant clinically---or at least that's what I surmise.

I have never had a galactomannan-positive bronchoscopy specimen before, and I wonder about the validity of the test outside of serum. A PubMed later finds a meta-analysis that suggests I should take the test seriously.

"In all, we came to the conclusion that BAL-GM test was an appropriate technique for diagnosing IA, using the cutoff value of 1.0. Compared with GM detection in serum, BAL-GM test has a higher SEN but a lower SPE, and with PCR assay, BAL-GM test has a higher SEN and a similar SPE."

My patient's value was 1.6. I hate treating tests when there are a lot of clinical indicators that make invasive Aspergillus unlikely. And I hate meta-analysis, where it is so often the case that the underlying theory is if you collect all the cow pies into one big pile you get gold instead of one big cow pie. But this appears to be a good meta-analysis.

And safe is better than sorry, so treat it is.

Postscript

It is odd to re-read this in 2019, as the galactomannan assay is now a standard test on bronchoscopy; but in 2012 it was the first time I had used the test. What was once new is now mundane.

Rationalization

PLoS One. 2012;7(8):e43347. Epub 2012 Aug 14. Systematic review and meta-analysis of detecting galactomannan in bronchoalveolar lavage fluid for diagnosing invasive aspergillosis.

http://www.plosone.org/article/info%3Adoi%2F10.1371%2Fjournal.pone.0043347

POLL RESULTS

I

• am uncomfortable treating a number. 8%

• think numbers are valid, people can't be trusted. 22%

• believe that the synthesis of all the clinical information is the goal. Single tests be damned. 16%

• think to err is human, to really screw up takes an EMR. 14%

• find vacation fades fast. 41%

Chapter Four

Other Explanations Are Needed

I am asked to see a patient for a positive syphilis serology obtained for peripheral neuropathy: the fluorescent treponemal antibody (FTA) is 2+, rapid plasma regain (RPR) is 1:2.

But it is never simple.

First, it is a classic diabetic neuropathy by exam and history, and she has no findings of syphilis. She is 70, from rural Southern Mexico, having been in the US for a decade. She has only been sexually active with her husband of 50 years, and his serology is negative. Hm. As best I can tell through an interpreter, she has never had symptoms of lues, never had unusual illnesses as a child, and has no other medical problems except related to diabetes, having been on dialysis for three years.

So is this the Great Pox?

It could be she has not been monogamous, but I suspect not. She had been inadvertently treated with a course of penicillin for another reason, but she can't recall why.

The RPR is false positive in about 2% of dialysis patients, but not the FTA. Old studies suggest an FTA false positive rate at 3% and another where 3 of 250 nuns had a false positive FTA. Yeah. Nuns are not sexually active. Right. Antiphospholipid antibodies and lupus can also cause a false positive FTA.

It may be the diabetes. In one series, 23 of 476 patients had a positive FTA for no other reason than diabetes. Color me skeptical; be it diabetics or nuns, people may have reasons for denying risks for syphilis.

She also grew up in the land of pinta, a non-venereal treponemal disease (Treponema pallidum carateum). Another one in this category is yaws (Treponema pallidum pertenue), but not where she is from. There was a great hamburger joint in NE Portland when I was young called Yaw's. I wonder now what was in the meat. Much to my delight, Yaw's is coming back to Portland. The burger, not the spirochete. It is possible that the serology is from pinta as a child, although history neither confirms nor denies that hypothesis.

And it could be a cross reaction from one or another Borrelia species, again not confirmed by history.

If the history is true, and there is no reason to suspect it isn't, then the serology is anything but syphilis. After a normal spinal tap, she was treated for late latent syphilis anyway.

Rationalization

The significance of syphilis serology tests on long-term hemodialysis patients. Lee CT, Lam KK, Liao SC, Chen JB, Hsu KT. Changgeng Yi Xue Za Zhi. 1998 Dec;21(4):447-52. J Clin Microbiol. 1979 March; 9(3): 369 372. PMCID: PMC273032

https://www.ncbi.nlm.nih.gov/pubmed/379033

False-positive reactions in the rapid plasma reagin-card, fluorescent treponemal antibody-absorbed, and hemagglutination treponemal syphilis serology tests. Calif Med. 1973 September; 119(3): 47. PMCID: PMC1455247

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC1455247/

False positive reactions in confirmatory tests for syphilis in presence of antiphospholipid antibodies: misdiagnosis with prognostic and social consequences. Gabriela de Larranaga, Luis Trombetta, Silvia Peres Wingeyer, Graciela Remondino. Dermatology Online Journal 12 (4): 22.

https://www.ncbi.nlm.nih.gov/pubmed/17083877

J Diabetes Complications. 1994 Jan-Mar;8(1):57-62. False-positive treponemal serology in patients with diabetes mellitus. Brauner A, Carlsson B, Sundkvist G, Ostenson CG.

https://www.ncbi.nlm.nih.gov/pubmed/8167389

Acta Derm Venereol. 1991;71(4):306-11. Evaluation of the fluorescent treponemal antibody-absorption (FTA-Abs) test specificity. Carlsson B, Hanson HS, Wasserman J, Brauner A.

[https://www.ncbi.nlm.nih.gov/pubmed/1681646]{.underline}

POLL RESULTS

I suspect the patient may not be entirely forthcoming about

• sex 41%

• drugs 8%

• rock and roll 23%

• diet 18%

• literacy 10%

Chapter Five

Dodging Bullets

I do not like repeat business. Most businesses thrive when the customer comes back, but after my patients finish a course of antibiotics, I never want to see them again---at least not in the hospital. Maybe running into them at Starbucks would be OK.

As an aside, it still rankles when a nurse refers to one of my patients as a "client." Lawyers and accountants have clients. I have patients.

This week I get two re-consults.

The first had enterococcal bacteremia from a diabetic foot infection, as best I could determine. I treated him with a course of ampicillin and, despite having a prosthetic valve, he showed zero evidence of endocarditis. He was readmitted with fevers and chills a month later with Streptococcus in the blood.

The second patient had Streptococcus salivarius endocarditis as a complication of a bicuspid valve, and three weeks after finishing a course of therapy is back with fevers and streptococcal bacteremia.

Both elicit that sinking feeling. Did I give the wrong therapy in the first case? Should I have treated for endocarditis? And did the treatment fail in the second case, as sometimes happens, and the now the patient needs a valve replacement?

But it is better to be lucky than to be good.

The first patient had Group G Streptococcus in his blood---S. equisimilis, from a thigh cellulitis, a common cause of bacteremic cellulitis and a very rare cause of endocarditis.

The second had a S. mutans. I questioned the lab as to whether it was really a different Streptococcus, since identification of the viridans streptococci always seems problematic. As one review understated

"...the often confusing changes that have taken place in the classification, identification, and nomenclature of the member species of the oral streptococci."

They assured me it was sufficiently different, including the resistance patterns and penicillin minimum inhibitory concentration (MIC), that they thought it was a new bug, not a relapsed bug. A new record for me, about two weeks to get a reinfection of a heart valve.

All I can say is whew. The patient, and I, dodged that bullet.

And who knew? As I was searching PubMed to write this up, I discovered that endocarditis is common in slaughtered pigs, 0.025%, most commonly from Streptococcus suis, Erysipelothrix rhusiopathiae, and a hodgepodge of other organisms. Why pigs are prone to endocarditis, I could not find, there is no information on swine underlying valve pathology, living conditions or illicit intravenous drug use.

Rationalization

Current classification of the oral streptococci. Whiley RA, Beighton D. Oral Microbiol Immunol. 1998 Aug;13(4):195-216

https://www.ncbi.nlm.nih.gov/pubmed/10093535

Aortic annular abscess complicating prosthetic valve endocarditis with group G streptococcus: detection during life with transesophageal echocardiography. Pollack BD, Belkin RN, Rubin DA, Moggio RA. J Heart Valve Dis. 1993 Sep;2(5):558-60.

https://www.ncbi.nlm.nih.gov/pubmed/8269167

J Comp Pathol. 2011 May;144(4):289-95. Epub 2010 Dec 17. Endocarditis-associated brain lesions in slaughter pigs. Karstrup CC, Jensen HE, Aalbæk B, Leifsson PS, Boye M, Agerholm JS.

[https://www.ncbi.nlm.nih.gov/pubmed/21168147]{.underline}

Acta Pathol Microbiol Immunol Scand B. 1984 Aug;92(4):237-8. The bacteriology of endocarditis in slaughter pigs. Pedersen KB, Henrichsen J, Perch B.

[https://www.ncbi.nlm.nih.gov/pubmed/6516848]{.underline}

Bacterial Isolation from Slaughtered Pigs Associated with Endocarditis, Especially the

Isolation of Streptococcus suis

https://www.jstage.jst.go.jp/article/jvms/59/1/59_1_75/_pdf

POLL RESULTS

I have

• patients. 67%

• clients. 11%

• customers. 2%

• patrons. 2%

• users. 10%

• Other Answers 8%

  • peeps

  • Issues and concerns and trouble and strife.

  • friends - I'm not a Dr., but like your blog.

Chapter Six

Looking for patterns in all the wrong places.

I tend to operate as if the world, and infectious diseases, has patterns and is predictable. That, of course, is a delusion. Any pattern I put on the world is a temporary pareidolia. The bugs certainly ignore patterns, why don't I?

Endocarditis in non-users of IV drugs is usually left-sided. Right-sided disease is reserved for shooters and people with central lines. If it is right-sided disease, it is going to be on the tricuspid valve, not the pulmonic.

Left-sided disease in the elderly is usually a viridans streptococcus; right-sided disease is almost always a Staphylococcus of some sort.

So I see an elderly male with a history of both a ventricular septal defect repair and an aortic aneurysm repair 30 plus years ago, but who still has native valves. He has had seven days of fever and during that time has had two sets of blood cultures on separate days that are positive for gram-positive cocci in chains.

I expect, from the pattern that I hallucinate as existing, that it will be a viridans streptococcus on his aortic valve. The graft in his heart is a worry, but 30 years out I would hope it has endothelialized and is not involved. I hope. The PubMeds would suggest that infections of aortic graft with oral Streptococci are rare; references with "unique" in the title are oddly reassuring.

It is a Streptococcus mutans. OK. So far so good. The transesophageal echocardiogram shows a pulmonic vegetation; the graft looks fine. Right-sided viridans streptococcal endocarditis. Wrong bug on the wrong valve. So much for patterns.

A review from 2000 suggests

"the literature from 1960 to 1999 identified only 36 reported cases of pulmonic valve endocarditis in structurally normal hearts."

The patient is not structurally normal, being old and with previous cardiovascular repairs, but all the work was on the left side of his heart.

A less than exhaustive review by me of the 99 PubMed reports suggests that of reported cases, less than a half-dozen are due to viridans streptococci, most being staphylococci of one sort or another.

Why is tricuspid endocarditis rare?

"It is assumed that its rarity is due to the low-pressure gradients within the right heart, the low prevalence of congenital malformations, the lower oxygen content of venous blood and the differences in the covering and vascularization of the right heart endothelium."

Sounds good to me. I can kill Streptococcus, so I am cautiously optimistic I will cure him.

I did.

Rationalization

Rev Esp Cardiol. 2008 Dec;61(12):1253-9. Isolated right-sided valvular endocarditis in non-intravenous drug users.

https://www.ncbi.nlm.nih.gov/pubmed/19080963

Clin Ter. 2012;163(1):27-9. An unique case of thoracic endovascular aortic repair (TEVAR) graft infection with Streptococcus viridans.

https://www.ncbi.nlm.nih.gov/pubmed/22362230

POLL RESULTS

Patterns

• are an illusion. 5%

• exist in medicine, but just at the threshold of detection. 18%

• are real, but we do not recognize that, as an example, clouds really are shaped like bunnies. 21%

• are helpful, but only for making clothes. 21%

• are quantum fractals. Of course, that means nothing. Like patterns. 29%

• Other answers 5%

• were Generals during WWII

  • are useful but massively simplified and ultimately illusory neurological mechanism that has evolved (to some degree) in most animals.

Chapter Seven

Protect Yourself

Flu season is upon us. If there is such a thing as flu season. H1N1 started at the furthest point in time you could get from the traditional start of the flu season. It is an interesting question as to whether global warming will alter the flu season, as it has the season for respiratory syncytial virus (RSV). Classically influenza is a fall/winter disease, and fall started today. It is perhaps worthwhile to review what is known about influenza.

1) The disease influenza, the gasping oppression, is a disease that has been plaguing humans for about 500 years. It is a predominantly respiratory infection, causing a protracted cough with fevers and myalgias. The term "flu" is used loosely to cover almost any febrile illness, and I remain uncertain as to what a "stomach flu" might be. I tend towards the picky side; flu is due to the influenza virus. There are many viri that can cause a flu-like illness, but only the influenza virus causes the flu.

2) The flu is due to a virus, the influenza virus. There are influenza A and B and C. There are those who deny germs as a cause of disease, and I will admit that as an infectious disease doctor I have a vested interest in germs causing disease. After all, my job is "me find bug, me kill bug, me go home."

3) Not every flu-like illness is due to the influenza virus. In a given season, the influenza virus may account for around 10-15% of flu-like illness in the community, although during epidemics influenza can account for 60% of flu-like illnesses.

4) Influenza kills. It kills directly; it kills by acute secondary infections, it kills by worsening underlying diseases like heart failure, it kills long-term by increasing vascular events such as heart attacks and stroke. Influenza kills the young, the old, the obese and the pregnant. Deaths by influenza also depend on the virulence of the circulating strains. Some strains, such as H5N1 (the bird flu) or H3N2 are more likely to kill (60% for H5N1) than strains like H1N1, which was of moderate virulence. Influenza deaths, both direct and indirect, are difficult to measure and are at best an estimate. As an example, somewhere between 151,700 to 575,400 people died worldwide from H1N1 the first year, with a mean of about 250,000.

5) When exposed to influenza, either as an infection or as a vaccine, the body responds in part by making antibody. Roughly, the better the antibody response to the influenza antigens, the better the protection to that strain of influenza. I say "roughly" as it depends on what part of the antigenic structure the body responds to: antibody to some antigenic sites on the virus offers better protection to influenza than others. If you are lucky enough to respond to some highly conserved sites on the virus you could conceivably be immune to all strains of influenza.

6) The influenza vaccine offers moderate protection against influenza. Unfortunately, the antigens of the virus change year to year (antigenic drift) and decade to decade (antigenic shift), while the vaccine strains are fixed prior to a given season and may not optimally match the circulating strains. If there is a good match between vaccine and circulating strains, in a healthy population the protection from the vaccine can be high: for H1N1 it was 87%.

7) Unfortunately, those who need to be protected from the influenza virus are those most likely not to respond to the vaccine: the obese, the pregnant, the elderly and those with chronic medical diseases.

8) The benefit of the vaccine extends beyond the prevention of flu. Of course, if you don't get influenza, you cannot spread it to others. If you do not get influenza, then, of course, you can't die of its complications, like pneumonia and heart attack. Influenza vaccination could potentially reduce the occurrence of sudden death, acute myocardial infarction, and stroke by 50%. Babies born to vaccinated mothers have fewer cases of influenza. Vaccinated mothers are less likely to have stillborn or small babies. Not getting influenza or being vaccinated against flu has many short-term and long-term benefits.

9) Influenza spread can be decreased by mechanical/environmental interventions: handwashing and masks, although these interventions have variable reported efficacy depending on the population studied. They are not as effective as not getting the virus.

10) Health care workers are particularly problematic. For a variety of reasons they are likely to come to work ill, influenza can be potentially infectious before people are symptomatic, and spread in the hospital occurs, leading to true stories such as this:

"Patient Story: Spreading the Flu 4-2012 As a reminder, we have added patient stories to meetings as a way of bringing the patient into the room, clarifying the context for our quality plan, and emphasizing the complexities and the importance of the work we are undertaking.

Today's story is about a group of patients, a nurse, and influenza. It starts with Patient 1, a 57-year-old woman admitted through the emergency department (ED) to one of our hospitals in mid-March with fever and shortness of breath. She was transferred to an inpatient unit with a mask on, which triggered the staff on the receiving unit to implement droplet precautions. Initially thought to have pneumonia, testing confirmed her symptoms were the result of influenza type A, H1N1. After four nights in the hospital, she was discharged home after an uneventful hospital stay and a flu shot.

Patient 2, next door to Patient 1, is a 58-year-old man who was admitted in early March for a GI bleed with multiple comorbidities. His progress was steady until nine days after admission, when he developed a new fever and respiratory symptoms. These symptoms developed on the same day of Patient 1's admission. Influenza was suspected two days following the development of his fever, and staff implemented droplet precautions. Lab testing confirmed influenza type A. He remained hospitalized for two more days and received a flu shot before being transferred to a skilled nursing facility.

Down the hall, Patient 3, a 77-year-old man, was admitted two days after Patient 1 for acute stroke and urinary tract infection. On day 3 of his hospitalization, he developed a fever and cough. Lab testing confirmed influenza type A. Droplet precautions were ordered with the lab test for influenza. He remained hospitalized an additional four nights and received a flu shot before being discharged. Patient 4, a 76-year-old man, down the hall from the first two patients and around the corner from Patient 3, was admitted on the same day as Patient 1 following a fainting event at home. Due to his long-standing heart issues, he was kept overnight for observation and discharged the following morning. However, he returned to the ED three days later with continued symptoms. He was discharged from the ED only to return the next day with shortness of breath. Six hours after being readmitted, staff suspected influenza and ordered droplet precautions. His lab tests returned positive for influenza type A. After spending three nights in the hospital, he was discharged home after receiving a flu shot. The following day, he was admitted to the intensive care unit and continued receiving treatment as an inpatient for secondary pneumonia, a complication of his influenza type A infection.

The fifth person in our story is a nurse on the unit where these four patients were admitted. She works on a nursing unit whose hand hygiene performance is currently 67%, and where 85% of the unit staff were vaccinated for this year's seasonal flu. The particular nurse in this case, however, was 1 of only 9 on the unit who chose not to be vaccinated. Her manager stated that the reason the nurse gave for not receiving the vaccine was that she was not convinced of the evidence that the vaccine protects patients from transmission & she said she would get the vaccine if she truly believed it protected her patients, but that she didn't. This nurse cared for Patient 1 on her first day of admission. She cared for Patient 2 on the eighth and ninth day (when he developed flu symptoms) of his stay. She also cared for Patient 3 on the first two days of his inpatient stay.

There does not appear to be any direct contact with this nurse and Patient 4. The nurse in our story developed symptoms consistent with influenza three days after working with Patient 1 and Patient 2 (which is the usual 1- to 4-day incubation period for influenza). Due to symptoms, she only worked a partial shift that day. Suspecting her symptoms may be influenza, she used a mask until relief staff was available. She returned home and was able to care for herself without medical intervention. She was not tested for influenza and remained off work for one week. She is still undecided about receiving the flu vaccine."

11) Because those in the hospital are particularly vulnerable to the ravages of infections, mortality from hospital-acquired influenza is remarkably high: 25%.

12) Influenza vaccine is safe as well as effective. There have been some rare complications to the vaccine: Guillain-Barré in the 1970s, and the recent reports that prior vaccination may have made disease from H1N1 worse. Biological systems are complex, but I look at vaccines as much safer than seat belts and air bags, both of which can cause injury and death. I would still prefer to be in a car accident with seat belts and airbags and work during the flu season with vaccination.

13) Influenza vaccination for health care workers runs at best 70% in the US, which if it were my kid's math grade would be cause for some 'splanin. Doctors and nurses run a bit higher, with an 80% influenza vaccination rate in some institutions.

14) It would be nice to prevent the spread of influenza from health care workers to patients. There is no data for hospitals, although there are nursing home studies to suggest that when staff is vaccinated against flu there are a decrease in flu in the inmates, er, nursing home patients. One retrospective study suggested that units that had higher vaccination rates had fewer cases of flu, but the definitive study has yet to be done. There are buckets of biologic plausibility to suggest that vaccinating health care workers would be of benefit to patients under their care.

15) I have little (actually no) respect for health care workers who do not get vaccinated. We have a professional and moral obligation to place our patients first. I have a career in infection control and quality and have seen too much morbidity and mortality from health care workers not putting their patients first---from poor hand hygiene, to not following isolation policy, to not "scrubbing the hub" to prevent central line infections, to not giving pre-op antibiotics, to... I could go on and on about the failings of some in the medical field.

I think those who do not get vaccinated, except for a minority with a valid allergy, are putting their patients at risk and are dumbasses.

This part of the essay is, I would like to clarify, directed at healthcare providers, not patients. Healthcare providers have no excuse to avoid the flu vaccine: they have access to the world's medical knowledge and should be able to rise above superstition and ignorance. Yes, I too am a dumbass, but for different reasons.

16) And this leads to my final thought. There is a tremendous amount of medical literature pointing to the safety and wide-ranging benefits of the influenza vaccine as well as the morbidity and mortality that influenza inflicts on humans every year. Despite that information, when you are admitted to the hospital you have a greater than one in three chance that the health care worker taking care of you is ignoring that information and going unvaccinated.

I have long been of the opinion that you judge a person by the company they keep. If your health care worker is a big enough dumb ass to avoid the flu vaccine, what other areas of medicine are they equally incompetent in? Do you want to drive in a car with no seat belts or airbag, whose brakes are of uncertain maintenance? Do you want you or your loved ones to be cared for by someone who is dumbass enough to not get the flu vaccine, putting you and yours at risk when most vulnerable? Can you expect that person to do the rest of their job correctly when they cannot understand and implement a core competency of medicine? Not me.

Here is my suggestion. When you and yours are in the clinic or the hospital, request care only from practitioners that have had the flu vaccine. Put a sign on your hospital room door: No entry unless you are influenza vaccinated or put a copy of this essay on the door. You do have the right to refuse care, especially from a dumbass.

I know this idea is a non-starter. Sick, vulnerable people are in no condition to potentially antagonize their providers. No one is likely to want to piss off their health care worker, especially if that person holds the key to the morphine. There is a bankrupt idea that it is OK for patients to ask their provider if they washed their hands. It never worked. I took an informal poll of patients on one of my medical floors and asked if they would ever tell their doctor or nurse to wash their hands, and not a one said they would. It would be like asking your pilot if he put the wheels down as they stated a landing. Patients need to trust that we are doing what is in their best interests. When it comes to influenza vaccination, you can't.

I do not really expect anyone will actually ask to be cared for only by influenza-vaccinated providers. But I can dream.

Rationalization

Of sorts.

I had the opportunity to play golf with my brother and son and the time I would normally spend linking to references was spent on the links. Golfing not internet. I hope to update it, but really, you can't search the PubMeds?

Dolan GP, Harris RC, Clarkson M, Sokal R, Morgan G, Mukaigawara M, et al. Vaccination of health care workers to protect patients at risk for acute respiratory disease. Emerg Infect Dis [serial on the Internet]. 2012 Aug [date cited].

http://dx.doi.org/10.3201/eid1808.111355 J

Hosp Infect. 2012 Jul;81(3):202-5. Epub 2012 Jun 1. Nosocomial H1N1 infection during 2010-2011 pandemic: a retrospective cohort study from a tertiary referral hospital.

http://www.ncbi.nlm.nih.gov/pubmed/22658238

Poll Results

When it comes to flu vaccination of health care workers:

• It should be a condition of employment. 81%

• can be skipped despite the data. I am more important than my patients. 1%

• until the data is incontestable and the vaccine is 100 safe and efficacious, it can be skipped. 8%

• I prefer my practitioners to go unvaccinated and catch flu from them. Real infection leads to better immunity than a vaccine. 2%

• I take vitamin D, C and oscillococcinum. The reality of flu is not a concern of mine. 3%

• Other Answers 5%

  • It is the duty of each health care worker to get vaccinated. making vaccination conditional on employment should not be necessary (although, unfortunately it is)

  • It should be left up to the individual. Flu vaccine is not 100% effective

  • How does a vaccinated person get flu from an unvaccinated person if the vaccine works?

  • These are very biased responses on an article that pretends "science."

  • If I exercise, eat right, and wash hands I will be okay.

  • If you see patients you should get it, but should not be mandatory.

  • Effective prevention: by strengthening the immune system, healthy lifestyle choices and diet, some supplementation maybe. Plus specific essential oils

  • Is this poll supposed to be humor? I sure hope it's not supposed to be taken seriously.

  • hopelessly biased poll

  • Should be vaccinated unless strong medical reasons not to

  • If I get sick, I'll stay home.

  • Healthcare workers should be educated or shamed into getting vaccinated, but not be forced to do so.

Chapter Eight

Trying to tie one disease to another

The patient returns after a trip to Bangladesh and presents with fevers, rigors, myalgias, headache behind both eyes. It hurt to move both his eyes. He is from Bangladesh and spent a month with his family in a small city, eating the food and drinking the water. Like most expatriates, he took no precautions for his trip nor any while he was there.

As an outpatient, he had a mild transaminitis, but a complete blood count, urinalysis, hepatitis serologies, chest X-ray, and malaria smears were all negative.

Because of the ongoing fevers and rigors, he was admitted, and the initial fever workup was negative.

As his fevers resolved he developed motor weakness--legs worse than arms and face, with a bilateral 7^th^ nerve palsy. An MRI was negative, and the lumbar puncture was negative except for a protein level of 550 milligrams per deciliter in the cerebrospinal fluid, when normal is less than 60.

The clinical diagnosis was Guillain-Barré syndrome (GBS), and he improved with steroids and plasmapheresis.

He has GBS, but from what? He had no diarrheal illnesses while in Bangladesh. Then tests trickled in.

Dengue: way positive for IgM and IgG.

West Nile: way positive for IgM and but only slightly positive for IgG.

Polio type 3: IgG way positive, type one and two slightly positive.

Blood and stool cultures are all negative.

So how to put it together? To the Googles and the PubMeds.

GBS is common in Bangladesh; it is hypothesized to be due to the high prevalence of Campylobacter. I can neither confirm nor deny a recent Campylobacter exposure in this patient.

It is unusual to have a bulbar form of polio, and I interpret the labs as he has been immunized (he has) and then exposed and boosted.

As best I can tell there is no West Nile in Bangladesh and Japanese encephalitis is rare.

West Nile and Dengue are both flaviviruses and the serologies can cross-react, as they can with other flaviviruses. BTW, just look at all the flaviviruses (cut and pasted from Wikipedia):

Tick-borne viruses

Mammalian tick-borne virus group

Alkhurma virus (ALKV)

Deer tick virus (DT)

Gadgets Gully virus (GGYV)

Kadam virus (KADV)

Karshi virus

Kyasanur Forest disease virus (KFDV)

Langat virus (LGTV)

Louping ill virus (LIV)

Omsk hemorrhagic fever virus (OHFV)

Powassan virus (POWV)

Royal Farm virus (RFV)

Tick-borne encephalitis virus (TBEV)

Turkish sheep encephalitis virus (TSE)

Seabird tick-borne virus group

Meaban virus (MEAV)

Saumarez Reef virus (SREV)

Tyuleniy virus (TYUV)

Mosquito-borne viruses

Avian tembusu-related virus

Calbertado virus

Duck tembusu virus

Without known vertebrate host

Aedes flavivirus

Calbertado virus

Cell fusing agent virus

Culex flavivirus

Culex theileri flavivirus

Kamiti River virus

Nakiwogo virus

Quang Binh virus

Aroa virus group

Aroa virus (AROAV)

Dengue virus group

Dengue virus (DENV)

Kedougou virus (KEDV)

Japanese encephalitis virus group

Bussuquara virus

Cacipacore virus (CPCV)

Koutango virus (KOUV)

Ilheus virus (ILHV)

Japanese encephalitis virus (JEV)

Murray Valley encephalitis virus (MVEV)

Rocio virus (ROCV)

St. Louis encephalitis virus (SLEV)

Usutu virus (USUV)

West Nile virus (WNV)

Yaounde virus (YAOV)

Kokobera virus group

Kokobera virus (KOKV)

Ntaya virus group

Bagaza virus (BAGV)

Ilheus virus (ILHV)

Israel turkey meningoencephalomyelitis virus (ITV)

Ntaya virus (NTAV)

Tembusu virus (TMUV)

Spondweni virus group

Zika virus (ZIKV)

Yellow fever virus group

Banzi virus (BANV)

Bouboui virus (BOUV)

Edge Hill virus (EHV)

Jugra virus (JUGV)

Saboya virus (SABV)

Sepik virus (SEPV)

Uganda S virus (UGSV)

Wesselsbron virus (WESSV)

Yellow fever virus (YFV)

Viruses with no known arthropod vector

Entebbe virus group

Entebbe bat virus (ENTV)

Yokose virus (YOKV)

Modoc virus group

Apoi virus (APOIV)

Cowbone Ridge virus (CRV)

Jutiapa virus (JUTV)

Modoc virus (MODV)

Sal Vieja virus (SVV)

San Perlita virus (SPV)

Rio Bravo virus group

Bukalasa bat virus (BBV)

Carey Island virus (CIV)

Dakar bat virus (DBV)

Montana myotis leukoencephalitis virus (MMLV)

Phnom Penh bat virus (PPBV)

Rio Bravo virus (RBV)

Whoa. That's a lot of viri. Cowbone ridge is in Florida as best I can tell. I didn't think there was enough elevation in Florida to make a ridge.

There are multiple reports of GBS following dengue, as well as West Nile virus, the latter which is more of a polio-like syndrome.

So at this point, given the clinical presentation and labs, I am calling it post-Dengue Guillain-Barré and the West Nile serology represents a cross-reaction. Anyone want to argue?

Rationalization

Guillain-Barré syndrome in Bangladesh

http://www.icddrb.org/what-we-do/publications/doc_download/3583-guillain-barre-syndrome-in-bangladesh

Rev Inst Med Trop Sao Paulo. 2011 Jul-Aug;53(4):223-5.Acute inflammatory demyelinating polyradiculoneuropathy (Guillain-Barré syndrome) following dengue fever.

http://www.ncbi.nlm.nih.gov/pubmed/21915467

Front Neurol. 2012;3:37. Epub 2012 Mar 21. Neuromuscular manifestations of west nile virus infection.

http://www.ncbi.nlm.nih.gov/pubmed/22461779

J Med Virol. 2011 Oct;83(10):1861-5. doi: 10.1002/jmv.22180.Acute West Nile virus neuroinvasive infections: cross-reactivity with dengue virus and tick-borne encephalitis virus.

http://www.ncbi.nlm.nih.gov/pubmed/21837806

J Gen Virol. 2011 Dec;92(Pt 12):2821-9. Epub 2011 Sep 7. Flavivirus-induced antibody cross-reactivity.

http://www.ncbi.nlm.nih.gov/pubmed/21900425

Chapter Nine

Maybe My Second Case

The patient had a knee replacement a year ago, and shortly thereafter it started to trouble her. Pain, stiffness, moderately red and warm. Not an "I can't bear weight" knee, but an "It hurts a lot after I walk" knee. No significant constitutional signs. More time passes, and the knee is scoped: not much in the way of infection, minimal white blood cells, but the cultures grow P. acnes.

Hmm. It was, given the lack of other objective findings, considered a contaminant, or at least a non-pathogen, and ignored. Several months pass and the knee is no better. The patient has another washout, and this time there is dead bone/osteomyelitis, and again there is not much on the gram stain: a few white blood cells and no organisms.

This time they call me, and I suspect it has been the P. acnes all along. I have written about his beast, which is most excellent at not causing much of an inflammatory response when it is not in a zit. I was going to bet was all due to P. acnes, but all the current cultures are growing Peptostreptococcus.

Hmm again.

Peptostreptococcus is not a common cause of prosthetic joint infection--you can cut off a hand and still count the number of reports in the literature. Unless, of course, you start with one hand.

On the other hand (if that hand is still present) if you culture clean ortho cases, you can grow bugs that do not go on to cause infection:

"Of the forty patients, twenty-three (58 per cent) had a positive culture on at least one of the media that were used and seventeen (43 per cent) had negative cultures. Of the forty specimens that were obtained from swabbing of the wound, eight (20 per cent) were positive on culture, compared with twenty (50 per cent) that were obtained from biopsy of tissue. Of these twenty-eight positive cultures, thirteen (46 per cent) were on routine blood-agar plates and fifteen (54 per cent), in broth only. Of the thirty-three bacterial organisms that were identified in the twenty-eight positive cultures of the wound, nineteen (58 per cent) were coagulase-negative Staphylococcus; eight (24 per cent), Propionibacterium acnes; two (6 per cent), Peptostreptococcus; and four (12 per cent), miscellaneous organisms."

Of course, knowing what I know now about how surgeries were done then, would I consider an ortho case from 1991 to be clean? Nope.

One of the PubMed cases was also infected with P. acnes, so can I suppose the inflammation of the P. acnes set up the joint for a secondary infection with Peptostreptococcus? Seems a bit of a stretch, but why didn't the Peptostreptococcus grow the first time? (Besides the fact that Peptostreptococcus, like all anaerobes, is difficult to grow?) She has no gastrointestinal/odontogenic source over and above the daily bacteremia we all encounter.

So the knee is out (too much bone involvement for salvage) and she is on a course of penicillin G; so old school.

I had a case of Peptostreptococcus in a normal knee many years ago, so this is the second case, at least that I can remember. But it is not satisfying for the root cause.

Rationalization

Int J Med Microbiol. 2005 Jan;294(7):465-70. Micromonas (Peptostreptococcus) micros: unusual case of prosthetic joint infection associated with dental procedures. http://www.ncbi.nlm.nih.gov/pubmed/15715175

Clin Rheumatol. 1996 Jul;15(4):399-402. Infection of a total knee joint prosthesis by peptostreptococcus micros and propionibacterium acnes in an elderly RA patient: implant salvage with longterm antibiotics and needle aspiration/irrigation. Stoll T, Stucki G, Brühlmann P, Vogt M, Gschwend N, Michel BA.

http://www.ncbi.nlm.nih.gov/pubmed/8853177

J Bone Joint Surg Am. 1991 Sep;73(8):1200-7. The importance of positive bacterial cultures of specimens obtained during clean orthopaedic operations. Dietz FR, Koontz FP, Found EM, Marsh JL.

http://www.ncbi.nlm.nih.gov/pubmed/1890121

Chapter Ten

Poly Want a Morphism?

The last week has been slow in regards to cases with an answer. Most of the patients have ended up better or continue to be ill (the only two options when you think about it) without a reason. I assume no one wants to read "All work and no answers makes Mark a dull boy" over and over.

But the week has not been entirely bereft of cases of interest. In the scheme of toxicities, antibiotics are mostly benign; it is rare to see a severe reaction to antimicrobials. I have to wonder about the ones I do not see. The recent New England Journal of Medicine report of an increase in deaths of patients prescribed azithromycin and maybe some quinolones does give one pause. Use of azithromycin results in an

"estimated 47 additional cardiovascular deaths per 1 million courses; patients in the highest decile of risk for cardiovascular disease had an estimated 245 additional cardiovascular deaths per 1 million courses."

Although a rare side effect, azithromycin does mess with the conduction system. And that is a lot of azithromycin--a drug I rarely suggest.

Usually, drug reactions are a bit more blatant. A young female has a urinary tract infection and is treated with trimethoprim-sulfamethoxazole (TMP-SMX). She is admitted with pancytopenia, a rash that looks like a boiled lobster, hyperemic eyes, and lymphadenopathy. Quite the reaction, not quite a Stevens-Johnson. But why? Why does she have such an excessive reaction to SMX, more akin to what you see in HIV patients?

Maybe it is in her genes:

"The molecular genetics results reported here led to the prediction that NAT1 10 and 11 may be protective against SMX-induced hypersensitivity. In contrast to loss of function polymorphisms in NAT2, the effect of which appears to be confounded by reduced liver enzyme activity caused by HIV infection [40], gain of function polymorphisms could have a persistent effect even in HIV/AIDS subjects. Our result indicate for the first time that NAT1 fast acetylators (10/10 and 11 carriers) are protected against SMX-induced hypersensitivity in HIV/AIDS, but only in slow NAT2 metabolizer subjects."*

Ouch. My brain hurts.

So maybe she is unable to metabolize the SMX efficiently and thence the allergic reaction. It would be so cool to be able to read everybody's genome inexpensively and know for sure. So many polymorphisms, so many variable risks.

Rationalization

N Engl J Med. 2012 May 17;366(20):1881-90. Azithromycin and the risk of cardiovascular death. Ray WA, Murray KT, Hall K, Arbogast PG, Stein CM.

http://www.ncbi.nlm.nih.gov/pubmed/22591294

Pharmacogenet Genomics. 2011 Oct;21(10):652-64 Human N-acetyltransferase 1 10 and 11 alleles increase protein expression through distinct mechanisms and associate with sulfamethoxazole-induced hypersensitivity. Wang D, Para MF, Koletar SL, Sadee W.

http://www.ncbi.nlm.nih.gov/pubmed/21878835

Biochem Pharmacol. 1993 Mar 24;45(6):1277-82. Role of polymorphic and monomorphic human arylamine N-acetyltransferases in determining sulfamethoxazole metabolism. Cribb AE, Nakamura H, Grant DM, Miller MA, Spielberg SP.

http://www.ncbi.nlm.nih.gov/pubmed/8466547

Chapter Eleven

Ghost in the Machine

Those whom the gods would destroy, they first make senior physicians. ~ Euripides

For the last 4 or 5 years, I have been at my intellectual peak. For here there is nothing but decline. Yippee. But for the time being my brain, my ID brain, has never been better. I know diagnoses like that (imagine the snap of a finger).

Most diagnoses come in a flash, although from where I do not know. At citywide conference today there were presenting an unknown and as the case progressed (a young female with peritonitis) I ideas burbled up from somewhere and were rapidly discarded. At the end, the presenter said, "Anyone want to guess the organism?"

Guess, my ass. I KNEW.

Group A Streptococcus.

Right for a dollar. How did I know it was Group A Streptococcus? Got me. I did see a case a year ago and I can give a post hoc, ergo propter hoc explanation of how I got there, but all the processing of the information was well below consciousness. This happens more and more as my career progresses, and over the years has time-shifted. I used to get these flashes hours after I saw the patient, but now the latest they pop into my head is while I do the dictation.

Just where these ideas come from is unknown, and a little creepy since it is not the conscious me who is doing the work. It is part of the reason why, along with the placebo effect and free will, I suspect consciousness is a minor and unimportant aspect of the human condition. So much goes on over which I have no conscious control.

"Most of the brain's work is done while the brain's owner is ostensibly thinking about something else, so sometimes you have to deliberately find something else to think and talk about." ~ Neal Stephenson, Cryptonomicon

Of course, those flashes are not infallible, not even close, and I make sure on each case that I go through the routine and consider the differential diagnosis so I don't miss anything. And sometimes the machine breaks down completely, or maybe I needed something else to think about.

Years ago I saw a lady in clinic with fevers to 40 ºC, severe arthralgias and myalgias, and who looked really sick. I admitted her, suspecting bacteremia, and left for the weekend. When I returned on Monday, I was surprised that the patient was discharged.

Oh, said the hospitalist, it was an obvious case of argle-bargle. (I'll reveal argle-bargle at the end, some of you will undoubtedly recognize argle-bargle long before that). It wasn't obvious to me. Well, having seen my first case of argle-bargle I figured I would not miss the next case. A mere 6 months later I saw another case and missed it again. 0 for 2.

But now evaluation for argle-bargle is part of my work-up for fever of unknown origin (FUO).

Fast forward to this month. The patient is a young male with ulcerative colitis (UC) who is on Remicade, an immunosuppressant. The last dose was over a month ago and the UC is quiescent. Now he has fevers to 40 ºC, a severe sore throat, a painful red bump on the leg that was thought to be a boil and leukocytosis; he feels terrible. At first I thought it was going to be a common illness, but the fevers persisted. He developed three skin lesions: multiple bumps that were erythema nodosum, a lacy rash on the chest that looked like a drug reaction to the cefazolin, and a bright red bilateral rash in the inguinal area.

He got the gold star FUO workup and nothing. I was worried it was some opportunistic infection from the Remicade suppressing his immune system, but nothing panned out. Finally, someone (not me) thought of argle-bargle, and he received a gram of solumedrol, an anti-inflammatory glucocorticoid. Everything got better like that (another snap of the finger).

Yeah, you know it. Argle-bargle is Still's disease: an idiopathic (no known cause) inflammatory condition characterized by fevers, joint pain (arthralgia), and a bumpy rash.

Surprised the heck out of me. Missed it again. 0 for 3. I need the machine to recognize Still's because I sure can't.

There were some unusual features: no arthralgias and the rash was not typical. The ferritin was only 500 micrograms per liter, elevated but not extremely so; in Still's, it is often greater than 3000. Erythema nodosum and Still's? One case in the PubMeds. And there is no relationship between Adult Still's disease and ulcerative colitis. So I will cut the machine a wee bit o' slack for not recognizing it.

Rationalization

J Dermatol. 1996 Mar;23(3):216-7. Adult Still's disease manifesting as erythema nodosum. Torinuki W, Funyu T.

http://www.ncbi.nlm.nih.gov/pubmed?term=8935635

Ann Rheum Dis 1992;51:683-685 doi:10.1136/ard.51.5.683 Evaluation of serum ferritin as a marker for adult Still's disease activity. M Schwarz-Eywill, B Heilig, H Bauer, A Breitbart, A Pezzutto

[http://ard.bmj.com/content/51/5/683.short]{.underline}

POLL RESULTS

I strongly suspect there is no

• free will. 7%

• consciousness 2%

• placebo effect 2%

• free lunch 35%

• unbiased poll 47%

• Other Answers 6%

  • Tooth Fairy

  • all of the above

Chapter Twelve

Ghost in the Machine

There are lots of awful diseases in medicine. Currently, we often have the ability to ameliorate or cure a remarkable number of diseases, but that was not always the case. There was a time before antibiotics when people routinely died of what are now curable infections (assuming access to care---a whole different issue in large parts of the world like Africa, Asia, and the US).

One of the classic ID cases of horribleness was Alfred S. Reinhart, a medical student who developed subacute endocarditis in 1931 that, over the course of 7 months, slowly killed him from a combination of infection, embolic events, and heart failure. He kept a careful diary of his disease that was published along with notes from his treating physician.

It is a depressing read, but at least we rarely get to watch the natural history of bacterial infections, although with increasing resistance of many bugs that may change in the future. I am putting my bet on tuberculosis as the first totally resistant organism to make a comeback, although gonorrhea is giving it a run for the money.

These days you see the natural history of disease in people who do not have insurance, or who lack the wherewithal to interact with the system to acquire insurance. I see a case or two of advanced HIV a year in people who have let their treatment slide for reasons of no insurance. And there are the occasional unnecessary progressive hypertensive or diabetic complications that maim or kill. You can't get your hypertension or HIV treated in an emergency room.

It has only been the last month that I have had the spirit of Alfred S Reinhart return, a progressive endocarditis because the patient preferred the siren call of heroin over that of nafcillin.

A young, homeless, uninsured heroin user was my whole service this week. Saying he was young is almost redundant---there are no old IV drug users, for good reason. They die. Although I once saw a Candida endocarditis in a 69-year-old whose son turned him on to heroin at age 68. (Father-son bonding is all well and good, but that might not be an optimal example. But I play golf with my kids. Is that really any better?)

My young patient was diagnosed with left-sided endocarditis about eight weeks ago. He left the hospital and disappeared. Came in a couple of weeks later with lots of emboli to the leg and positive blood cultures, then disappeared again. Every once in a while over the last month I would get calls from different emergency rooms where he would show up with a fever, get blood cultures and a dose of an antibiotic, and disappear again.

For the last two weeks, he has had progressive back pain that eventually became refractory to heroin, and showed up in one of my hospitals with this CT.

{width="3.5256944444444445in" height="4.309722222222222in"}

Ouch

He has been debrided and still has positive blood cultures for methicillin-sensitive Staph aureus (MSSA). We are two months into a barely-treated MSSA endocarditis and he is walking down the path of Reinhart. Now that his back is opened up, he will not be able to skip out against medical advice and maybe I will be able to get a course of antibiotics into him so he does not recapitulate the final days of Reinhart:

"Beginning with October 18 the patient developed transient attacks of aphasia and on October 21 complete right hemiplegia occurred. On October 26 he developed pulmonary edema and died."

I still think death is too high a price to pay for drug addiction and stupidity, but further morbidity with mortality seems the likely end. I find these lives wasted to be so sad.

Nope. He left against medical advice after a week, never to be seen again.

Rationalization

CMAJ. 2002 December 10; 167(12): 13791383. Subacute bacterial endocarditis observed: the illness of Alfred S. Reinhart Kenneth M. Flegel

http://www.ncbi.nlm.nih.gov/pmc/articles/PMC137358/

Clin Infect Dis. 2012 Feb 15;54(4):579-81. Epub 2011 Dec 21. Totally drug-resistant tuberculosis in India.Udwadia ZF, Amale RA, Ajbani KK, Rodrigues C.

http://www.ncbi.nlm.nih.gov/pubmed/22190562

Alarms Sounded over Totally Drug-Resistant Tuberculosis, Gonorrhea

http://www.microbemagazine.org/index.php?option=com_content&view=article&id=4684:alarms-sounded-over-totally-drug-resistant-tuberculosis-gonorrhea&catid=948:04-2012-current-topics&Itemid=1222

POLL RESULTS

The most likely pan-resistant infectious resurgence will be due to

• TB 43%

• gonorrhea 7%

• E. coli 7%

• S. aureus 14%

• What does it matter, we can't afford to treat susceptible infections 21%

• Other Answers 7%

  • Bloviation-induced stupidity.

  • Enterobacteriaceae in general due to beta-lactam resistance

Chapter Thirteen

Common plus Common equals Uncommon

There are lots of common diseases. They occur commonly, hence the designation. They are the common infections. At some level, ID is easy in that most infections are caused by a handful of the organisms that have evolved to pester humans---only around 200 when I once counted. Not bad and certainly manageable. I also have to know many of the uncommon organisms as well. That number is around 1800. But combine a common syndrome with a common bug and you can get a very unusual result.

The patient is an elderly male, in remission from a T-cell lymphoma, but overall not doing well. He has acute pharyngitis: Fevers, painful swallowing, leukocytosis. Exam shows the right side of this throat is red with some exudates and the CT scan shows pharyngeal inflammation but no abscess or clot. There are certainly a wide variety of organisms that can cause pharyngitis, and he is an immunocompetent host with lots of recent interactions with the medical-industrial complex, so he is put on the pipercillin-tazobactam and vancomycin. Who isn't? He is around grandchildren, so I would expect a Group A Streptococcus.

The next day he has gram-positive cocci in chains in the blood and they call me and I ramble on about Group A Streptococcus and the odd Streptococci that can cause a soft tissue infection: the Streptococcus milleri group and Group B Streptococcus and blah blah blah.

The blood cultures grew S. pneumonaie. Huh. Didn't expect that. Quick reassessment: none of the standard infections from S. pneumoniae, such as in the lung. Unfortunately, only a rapid strep screen was sent, not a throat culture. His throat pain got better with the antibiotic. So pharyngitis and bacteremia from S. pneumoniae it is.

How many cases in the literature? One. There are more cases of pneumococcal cellulitis/soft tissue infections reported; I have seen maybe a half-dozen in my career. This was more of a soft tissue infection of the throat than a typical pharyngitis, so a little bit less odd.

Two common problems adding up to a real oddity.

Rationalization

J Med Microbiol. 2008 May;57(Pt 5):674-5. A case of pharyngitis caused by Streptococcus pneumoniae

http://www.ncbi.nlm.nih.gov/pubmed/18436606

Eur J Clin Microbiol Infect Dis. 2007 Apr;26(4):247-53. Epub 2007 Mar

20. Streptococcus pneumoniae skin and soft tissue infections: characterization of causative strains and clinical illness.

http://www.ncbi.nlm.nih.gov/pubmed/17372776

Chapter Fourteen

An Onion on my Belt

It is remarkable how rapidly bacteria can lead to morbidity. My old partner, since retired, used to observe that the patients that did the worst with staph infections were those who were the least likely to get a staph infection. Normal people would get S. aureus and go to stink, while the IV drug user, the diabetic, the dialysis patient--with their constant exposure to S. aureus--tended to tolerate the infection a bit better.

It may have been confirmation bias (the tendency to favor information that confirms their beliefs and ignore information that does not), but it seemed to be true.

In the last couple of months, I have had a pair of normal middle-class males with spontaneous MSSA infections that were a disaster. Multiple abscesses, vegetations, prolonged fevers, septic joints, and osteomyelitis. What they had in common was no medical or social reasons for a S. aureus infection, and when the beast gained access to their blood, it ran wild.

It is not the first time I have witnessed this phenomenon, and I have presumed that needle users are chronically being exposed to S. aureus with their injections and have a primed immune system. It is almost like a classic vaccination for smallpox by inoculating cowpox: the etymology of the word vaccination being "from vaccine (adj.) 'pertaining to cows, from cows' (1798), from Latin vaccinus from cows,' from vaccacow'".

There may be something to this idea. For example

"the risk of sepsis was significantly lower in those patients with higher levels of IgG against α-hemolysin (Hla), δ-hemolysin (Hld), Panton-Valentine leukocidin (PVL), staphylococcal enterotoxin C-1 (SEC-1), and phenol-soluble modulin α3 (PSM α3).

Conclusions. Our results suggest that higher antibody levels against Hla, Hld, PVL, SEC-1, and PSM- α 3 may protect against sepsis in patients with invasive S. aureus infections."

It would be nice to have a vaccine to S. aureus, and a perusal of the PubMeds suggests they have a long way to go before they have a useful product. Too many antigens to choose from.

In the meantime, I do not advise autoinoculation with heroin or meth. It is not an optimal lifestyle.

My partner also used to say the summer was S. aureus season. He thought it was due to people stewing in their own sweat and as a result, there was increased carriage and more infections. He may have been right there as well:

"Although most studies published to date are methodologically weak, some seasonal variation in the occurrence of S. aureus infection appears to exist, particularly an association of warm-weather months with S. aureus skin and soft-tissue infections."

Now that I am the old geezer, pay attention to my observations. They may be wisdom there. Or I may just be babbling.

Rationalization

J Infect Dis. 2012 Sep;206(6):915-23. Epub 2012 Jul 17. Lower Antibody Levels to Staphylococcus aureus Exotoxins Are Associated With Sepsis in Hospitalized Adults With Invasive S. aureus Infections.

http://www.ncbi.nlm.nih.gov/pubmed?term=22807524

Clin Microbiol Infect. 2012 Oct;18(10):927-33. doi: 10.1111/j.1469-0691.2012.03955.x. Seasonality of staphylococcal infections.

http://www.ncbi.nlm.nih.gov/pubmed/22958212

Chapter Fifteen

Aesthetic

Everyone has an aesthetic. I like simple, uncluttered, clean lines, black and white. Like my website. Not that you would know it to look at any other space for which I am responsible. Or my brain. That's a mess.

One of the hospitals I go to has to have one of the ugliest interior designs ever for a hospital. Of any building for that matter. And remember, I live in the city that has the Portland Building. We know how to do ugly in the Northwest.

This week also marked the roll out of a new version of Power Chart at one of my hospitals. The Power Chart upgrade has a color scheme that looks like one of those websites run by an apparently crazy person---all distracting and obfuscating colors that, if not seizure inducing, certainly result in nausea.

But there are people who feel that way about my prose. One person's art is another's garbage.

Some consults are as aesthetically appealing as Power Chart. Today's patient has pleural metastasis (of what cancer?) with recurrent effusions, and to manage the symptoms he has a pleural catheter. Every now and then he has to drain off fluid when he gets short of breath.

Now he has fevers, rigors, and pleuritic chest pain, and the pleural fluid turns cloudy. A CT scan shows fluid, air, masses and a rind in the pleural space. A Gram stain of the pleural fluid shows white blood cells and gram positive cocci, and grows a methicillin-resistant Staphylococcus lugdunensis. Infected tumor and pleura.

I have written about S. lugdunensis in the past: a coagulase-negative Staphylococcus that is more towards the S. aureus end of the virulence-o-meter. There are no reports of this organism in an empyema; not much of a surprise as it lives mostly below the waist.

What to do?

Thoracic surgery, for good reason, does not want to do a VAT (video-assisted thoracoscopy).

I doubt antibiotics will work without a debridement, and it would be optimal for the pleural catheter to be removed, but he needs it.

Try lytics in the space? Thrombolytics are used for malignant effusions and infected effusions, but no one has tried that intervention for infected malignant pleural effusions.

It is not aesthetically pleasing; damned if you do and damned if you do. For now, they/we have elected to see what happens with a course of antibiotics. No matter the choice, it is ugly.

Postscript

He did OK. We elected, for quality-of-life reasons, for suppressive antibiotics, and at least he was not readmitted for infection.

Rationalization

Intrapleural use of tissue plasminogen activator and DNase in pleural infection. Engl J Med. 2011 Aug 11;365(6):518-26.

[https://www.nejm.org/doi/full/10.1056/NEJMoa1012740]{.underline}

Portland Building

"...something designed by a Third World dictator's mistress's art-student brother."

https://en.wikipedia.org/wiki/Portland_Building

POLL RESULTS

My aesthetic is summed up best as

• Zen. 27%

• Northwest: jeans, boots and flannel. 21%

• Goth. 6%

• Trump. 6%

• Goodwill 27%

• Other Answers 12%

  • plaid top, striped bottom

  • organic

  • scrubs

Chapter Sixteen

Dirty Laundry

I have been involved with infection control for 6 hospitals for my entire career. Because of infection control and its links to quality, I know things. I aware of people's dirty medical laundry. Part of infection control/quality is identifying issues and correcting them, and over the years we have washed a lot of laundry, metaphorically speaking, with excellent results. Every year has seen a decline of not only infections but all manner of complications. Hospitals, at least my hospitals (I am very possessive), are markedly safer today than two decades ago.

It has not been without pain. One of the aspects of my job that I do not understand is the pushback from some physicians. There was an article that clearly demonstrated that full barriers at the time of central venous catheter insertion led to fewer infections. This was 1994 and for years a few docs, mostly anesthesiologists, refused to apply the information to their practice. They didn't need to wear a gown because they "never had a catheter infection," as if they had any data to confirm or deny the assertion. The issue was eventually resolved, but life would be so much easier if people would apply the evidence to their practice.

What then really burns my cookies is they would then ask me for the data. What, you don't have PubMed? I am supposed to read YOUR literature for YOU and let YOU know how best to practice? It is in the practice guidelines of your own society. What is wrong with this picture? Shouldn't you be the one who is up to snuff and bring these practices to me? Never happens. Ever.

Walking through operating rooms can be an interesting experience if you have an eye for infection control. Masks worn incorrectly, backpacks in the room, food on the anesthesiologist cart. Almost all of these issues were back in the day, and all hospitals have tightened up practice so that these kinds of infractions are a thing of the past. Well. Almost.

At one of my hospitals, an anesthesiologist was observed to access a catheter four times in a row and never once scrubbed the hub. ARRGH. I still do not understand why people do things like this. They just forgot in the heat of the moment, which I can somewhat understand. Somewhat. At least this time there is no pushback; the person involved is more than a little chagrined when chastised.

It is not as if the potential danger of unscrubbed hubs are a mystery. From the anesthesia literature:

"One hundred sixty-four cases (82 case pairs) were studied. We identified intraoperative bacterial transmission to the IV stopcock set in 11.5 % (19/164) of cases, 47% (9/19) of which were of provider origin. We identified intraoperative bacterial transmission to the anesthesia environment in 89% (146/164) of cases, 12% (17/146) of which were of provider origin. The number of rooms that an attending anesthesiologist supervised simultaneously, the age of the patient, and patient discharge from the operating room to an intensive care unit were independent predictors of bacterial transmission events not directly linked to providers.

CONCLUSION:

The contaminated hands of anesthesia providers serve as a significant source of patient environmental and stopcock set contamination in the operating room. Additional sources of intraoperative bacterial transmission, including postoperative environmental cleaning practices, should be further studied."

and

"Bacterial contamination of the anesthesia work area increased significantly at the case conclusion, with a mean difference of 115 colonies per surface area sampled (95% confidence interval [CI], 62-169; P < 0.001). Transmission of bacterial organisms, including vancomycin-resistant enterococcus, to intravenous stopcock sets occurred in 32% (95% CI, 20.6-44.9%) of cases. Highly contaminated work areas increased the odds of stopcock contamination by 4.7 (95% CI, 1.42-15.42; P = 0.011). Contaminated intravenous tubing was associated with a trend toward increased nosocomial infection rates (odds ratio, 3.08; 95% CI, 0.56-17.5; P = 0.11) and with an increase in mortality (95% CI odds ratio, 1.11-infinity; P = 0.0395).

CONCLUSION:

Potentially pathogenic, multidrug-resistant bacterial organisms are transmitted during the practice of general anesthesia to both the anesthesia work area and intravenous stopcock sets. Implementation of infection control measures in this area may help to reduce both the evolving problem of increasing bacterial resistance and the development of life-threatening infectious complications."

I have long supposed the route of wound infection is by way of the open wound. Now I wonder. It is hard to infect people hematogenously. IV drug abusers shoot up bacteria every day, we are bacteremic every day, but infections are rare---due to a combination of a good immune system and no place for the bacteria to go. Bacteria love clot and damaged tissue. I wonder how many of the infections I see, especially the coagulase-negative staphylococcus prosthetic infections, get there by way of the unscrubbed hub. If such an epidemiologic study has been done, I can't find it.

I am of the opinion that accessing a line 4 times in a row without cleaning off the hub is probably the same infection risk as an unrecognized failed autoclave load where the clean, but not sterile, instruments were used. I have seen that maybe 2 or 3 times over the years. Each time there was no infection in the patients, but each time we told the patient of the problem. Perhaps we should have told the patient here as well.

Next time I have surgery I am going to pay someone to ensure that my hub is scrubbed each and every time.

Addendum

All our IN hubs now soak in an alcohol cap.

Rationalization

Raad II, Hohn DC, Gilbreath BJ, et al. Prevention of central venous catheter-related infections by using maximal sterile barrier precautions during insertion. Infect Control Hosp Epidemiol 1994; 15:231--8.

http://www.ncbi.nlm.nih.gov/pubmed/8207189

Anesth Analg. 2011 Jan;112(1):98-105. Epub 2010 Aug 4. Hand contamination of anesthesia providers is an important risk factor for intraoperative bacterial transmission. Loftus RW, Muffly MK, Brown JR, Beach ML, Koff MD, Corwin HL, Surgenor SD, Kirkland KB, Yeager MP.

http://www.ncbi.nlm.nih.gov/pubmed/20686007

Anesthesiology. 2008 Sep;109(3):399-407. Transmission of pathogenic bacterial organisms in the anesthesia work area. Loftus RW, Koff MD, Burchman CC, Schwartzman JD, Thorum V, Read ME, Wood TA, Beach ML.

[http://www.ncbi.nlm.nih.gov/pubmed/18719437]{.underline}

POLL RESULTS

In my experience, the group most likely to think infection control does not apply to them are

• surgeons 32%

• hospitalists 12%

• anesthesiologists 33%

• Ob 9%

• residents 9%

• Other Answers 6%

  • All of them

  • nurse

  • Hospital Administrators

  • dentists

  • House staff

Chapter Seventeen

I have seen the only two cases. Ever.

The patient is obese, diabetic and admitted to the hospital with acute rubor, dolor, calor, and tumor of the leg. He is a touch septic and requires a bit of levophed to maintain his blood pressure.

His leg is red and swollen, but areas are more cool than hot, always a concern for vascular insufficiency instead of infection. Because he had been in the hospital a month ago and met the criteria for the systemic inflammatory response syndrome (SIRS), he was placed on antibiotics---

vancomycin and cefepime.

Next day both sets of blood cultures are growing P. aeruginosa and they call me.

I don't find much else. There is, by exam and panscan, no other reason for the bacteremia. No urinary tract infection or bowel source or pneumonia. The arteries to his leg are patent. There had been some minor trauma to the leg, but he has not been in any odd water sources or used unusual ointments that would have inoculated his leg with Pseudomonas. I find nothing by labs or exam to suggest a necrotizing soft tissue infection.

There are soft tissue infections from Pseudomonas: hot tub folliculitis, abscesses after a nail through the sneaker (no one calls them sneakers anymore), necrotizing soft tissue infections in the profoundly immunocompromised.

As best I can tell, there are no cases of cellulitis due to P. aeruginosa that do not have one of the aforementioned clinical syndromes. It looked like a group A Streptococcus clinically, and I will confess I would not have given the cefepime on admit if I had been the one looking at the leg. Pseudomonas doesn't do this. Except it does.

Two years ago I wrote about a morbidly obese (750 lbs.) diabetic who had multiple admissions for bacteremic Pseudomonas cellulitis of the leg for which I could find no nidus of infection, exposure or immunologic defect to account for the problem. The current case is the same, except his obesity is of a more conventional size, 300 lbs.

A search of the Pubmeds again fails to find any cases of cellulitis from this bug. So I have the only two cases ever. I think I have a resident who is going to write it up. Even though no one will ever see a case, I think I will include one of my pet peeves from the literature and mention that Pseudomonas needs to be considered as a cause of cellulitis in the obese diabetic. Probably at the same level you have to worry about being hit by lightning on a sunny day.

POLL RESULTS

Archaic terms I still use:

• sneaker. 27%

• ice box. 12%

• davenport. 8%

• patient (instead of client). 44%

• admits (instead of endorses, as in patient admits to fever) 6%

• Other Answers 4%

  • Shrubbery

  • johnny pump

Chapter Eighteen

100%

We have been at 100% for hand hygiene compliance for the last couple of months. Really. 100%. You do not believe me? Allow me tell you about our hand hygiene program.

A while back I decided that our hand hygiene compliance program was flawed. We had observers watch for compliance with hand hygiene and chart whether or not they met criteria, and then report the results a month or two later at various meetings. No effort was made in the moment to make sure people washed/foamed their hands if they failed. Unlike every other quality measure, we spent our time measuring failure. If someone forgets to give prophylactic preoperative antibiotics, we do not chart the failure, we make sure the antibiotic is given. Depending on the population measured, we were anywhere from 50 to 100%, with an average around 80% for hand hygiene. Barely a B. Pathetic.

I think the reason most people do not wash/foam is they are so preoccupied with the too-numerous-to-count actions at work that they forget. A couple of years ago I was taking the kids to school. It was 7 am, mostly dark. At the bottom of a hill I had to make a quick right then an immediate left across five lanes of traffic. The corner had a Crusty Cream, a bus stop, and a 7-11, so you had to watch for people. I was talking to the kids, trying to get the damn hip-hop off the radio, look left, look right---everything good, double check, no cars, no people, make the right then the left turn and piece of cake. Done.

Police lights behind me. What the? It turned out the I ran a stoplight, and on my left as I made the illegal turn was a police car. Did not see the stop light or the police car. That, combined with three speeding tickets, led to my losing my driver's license for a month. No fun, let me tell you. I tried to blame my toxoplasmosis seropositivity, but no luck. I was too busy with too many tasks to note the red light and police car at a turn I have made hundreds of times.

It's the same issue with hand hygiene. We are just too busy to remember everything, and asking patients to remind us is pathetic. If we have to ask our patients to tell us to clean our hands, we are bankrupt as a profession and might as well pack it in.

So I proposed that we do the following: hand hygiene is simple. We want to stop the spread of contagion in our hospital. Consider the doorway like the portal at the City on the Edge of Forever. Walking through can change the future, at least of infection control. Foam/wash when you enter the room and when you leave. While there are ample other opportunities to foam, that is going to be the basic core competency everyone will meet.

If you forget to foam/wash, the hand hygiene observer, who will be visible (no hidden shopper) will ask you to "foam please."

And the expectation is that everyone is going to smile, say "Thank you," and foam.

Would you believe it? Some docs can be real jerks when asked to foam. Not allowed. No bitching, no whining, no kvetching, no excuses, no abuse, no nothing. Smile. Say "Thank you." Foam. Go on about your day. Because seriously, you have time to argue about foaming? Quit being a baby. Just do it.

In preparation, I talked to almost every MD group and meeting in the hospital and made a short film to show all the staff.

And it has worked. So far over 300 people have been observed, about 1 in 6 need to be reminded and everyone was nice and foamed. 100% compliance. No jerks.

And no contagion entering or leaving the rooms on someone's hands.

Rationalization

BMC Infect Dis. 2009 May 26;9:72. Increased incidence of traffic accidents in Toxoplasma-infected military drivers and protective effect RhD molecule revealed by a large-scale prospective cohort study. Flegr J, Klose J, Novotná M, Berenreitterová M, Havlícek J.

http://www.ncbi.nlm.nih.gov/pubmed/19470165

http://en.wikipedia.org/wiki/The_City_on_the_Edge_of_Forever

POLL RESULTS

When in a car accident I blame

• Toxoplasmosis. 26%

• a bee in the car 14%

• Obama. 12%

• texting. 8%

• I don't. I am no weasel. 33%

• Other Answers 8%

  • space aliens

  • lack of foam

  • my wife

  • The other driver who is at fault!!

Chapter Nineteen

Playing the Odds

I get a call about a patient with urosepsis. Blood and urine cultures with Escherichia fergusonii, a close relative of E. coli. The patient has a prosthetic valve, and they assume the patient needs to be treated for endocarditis, right?

Got me. So I went and saw the patient. Classic urinary tract infection symptoms with rapid progression to sepsis and equally rapid resolution of the symptoms. No stigmata whatsoever of endocarditis.

There are what, like 250,000 cases of E. coli bacteremia a year? That is the number in my head. I can't find the exact number, but with 750,000 cases of sepsis a year and 54,000 cases of urosepsis in the elderly a year, it seems like a reasonable ballpark.

Contrast that with the rarity of E. coli endocarditis in general (about 31 cases in the literature between 1909 and 2002) and prosthetic valve involvement (16 cases reported before 1960, 5 between 1960 and 1980, and 11 after 1980) in particular.

Most patients with the E. coli endocarditis have sustained bacteremia (repeat cultures negative), abnormal ECHO (ECHO negative) and do poorly clinically (this patient looked like a rose).

So the odds of E. coli seeding the valve are pretty remote, and if the valve is infected the odds of resolving the infection without surgery are also remote.

So do you give a worthless therapy for an extremely unlikely disease that is not without complications, unlike a line infection with S. aureus that really would seed the valve? One of my superstitions is that the less the indication for treatment, the greater the chance of a complication. I suppose there will be one person out there who will say do it for legal reasons, but that is one sorry excuse to practice bad medicine.

And what is E. fergusonii? There are a hodgepodge of Escherichia: E. albertii, E. blattae, E. coli, E. fergusonii, E. hermannii, E. vulneris to cut and paste a few.

There are three clinical cases of E. fergusonii in the PubMeds: one urosepsis and two cases of biliary sepsis, so it appears to have minimal pathogenicity in humans. Apparently, it is more of an issue in farm animals, of which she has no contact.

So I treated the patient for urosepsis, and she did just fine.

Rationalization

J Infect Dis. 2005 May 1;191(9):1523-9. Epub 2005 Mar 22. Burden of community-onset Escherichia coli bacteremia in seniors. Jackson LA, Benson P, Neuzil KM, Grandjean M, Marino JL.

http://www.ncbi.nlm.nih.gov/pubmed/15809912

Scand J Infect Dis. 2011 Jul;43(6-7):545-6. Epub 2011 Feb 10. Infectious endocarditis caused by Escherichia coli. Lauridsen TK, Arpi M, Fritz-Hansen T, Frimodt-Møller N, Bruun NE.

http://www.ncbi.nlm.nih.gov/pubmed/21309637

Clin Microbiol Infect. 2006 May;12(5):401-3. Escherichia coli native valve endocarditis. Micol R, Lortholary O, Jaureguy F, Bonacorsi S, Bingen E, Lefort A, Mémain N, Bouchaud O, Larroche C.

http://www.ncbi.nlm.nih.gov/pubmed/16643514

Eur J Clin Microbiol Infect Dis. 2005 Aug;24(8):537-41. Escherichia coli endocarditis: seven new cases in adults and review of the literature. Branger S, Casalta JP, Habib G, Collard F, Raoult D.

http://www.ncbi.nlm.nih.gov/pubmed/16133408

POLL RESULTS

At what point do the odds warrant a course of therapy

• 1 in 100 17%

• One in a million. 6%

• Always. Treatment is always the safer course. 14%

• I don't play odds, I wait to confirm a diagnosis. 37%

• I let a coin flip decide, like Dr. Harvey Dent. 20%

• Other Answers 6%

  • I consult with "Magic 8 Ball"

  • When the patient's niece is an attorney

Chapter Twenty

Confirmation Bias

Almost every patient I see has an infection, and if not an infection, then some sort of inflammatory state. So there is ample opportunity for me to confirm the bee lodged firmly in my bonnet (there's a mental picture for you) that infections of all kinds lead to vascular events.

The sequence is simple. Infections are inflammatory. Inflammation is prothrombotic. Thrombosis leads to clot going where it doesn't belong.

The patient is a middle-aged male who is admitted with a near stroke---a transient ischemic attack, or TIA. As an outpatient, he was being treated for mastoid and sphenoid sinusitis along with otitis media that was not responding to oral antibiotics. He developed an expressive aphasia that lasted several hours. He had the big evaluation: MRI, MRA (magnetic resonance angiogram), TEE (transesophageal echocardiogram), ultrasound of carotids. Nothing but the sinusitis.

Of note in the past medical history was head and neck radiation for oral cancer years ago.

Maybe it was just the prior radiation that caused the stroke:

"The relative risk of transient ischemic attack or ischemic stroke is at least doubled by head and neck radiotherapy."

But the literature is filled with vascular events after infections (stroke, heart attack, pulmonary embolism) after a wide variety of infections: influenza, urinary tract infection, zoster, HIV, and pneumonia, to name a few. Generally speaking, if you get an infection, your risk for a vascular event of pops up and stays up for up to a year. Including sinusitis:

"We found that patients with rhinosinusitis were more likely to suffer strokes than the control population, after adjusting for potential confounders (adjusted HR, 1.39; 95% confidence interval [CI], 1.28~1.50). The HR of stroke was 1.39 (95% CI, 1.28~1.51) for acute sinusitis patients, and 1(95% CI, 1.04~1.74) for chronic sinusitis patients."

Even dental work is dangerous:

"The rate of vascular events significantly increased in the first 4 weeks after invasive dental treatment (incidence ratio, 1.50 [95% CI, 1.09 to 2.06]) and gradually returned to the baseline rate within 6 months."

It is why I quit going to the dentist. I look better with green teeth anyway. It is also why if the immune system really could be boosted (it can't) you would probably get more vascular events.

And there is this ongoing hint that being admitted with infection or getting an infection in the hospital increases mortality after discharge. A representative example:

"For Medicare beneficiaries undergoing elective open abdominal vascular procedures, the development of any IC significantly increased not only the in-hospital mortality rate but also the mortality rates 30 and 90 days after discharge from the hospital."

Although how much post-discharge mortality is due to vascular events has not been worked out. Infections are bad, both acutely and long term for vascular events and mortality. The TIA was probably a combination of badness: the sinusitis plus the prior radiation.

A diverse literature points to the long-term thrombotic dangers (perhaps leading to death) from infections, although there are no studies to indicate benefits from interfering with the thrombosis. My prediction is that someday everyone with an acute infection will be on some sort of anti-thrombotic agent. Me? If I get an infection, I am taking aspirin for six months.

Rationalization

Stroke. 2011 Sep;42(9):2410-8. Epub 2011 Aug 4. Ischemic stroke and transient ischemic attack after head and neck radiotherapy: a review. Plummer C, Henderson RD, O'Sullivan JD, Read SJ.

https://www.ncbi.nlm.nih.gov/pubmed/21817150

Invasive dental treatment and risk for vascular events: a self-controlled case series. Minassian C, D'Aiuto F, Hingorani AD, Smeeth L. Ann Intern Med. 2010 Oct 19;153(8):499-506.

https://www.ncbi.nlm.nih.gov/pubmed/20956706

Am J Rhinol Allergy. 2012 Jul-Aug;26(4):278-82. Risk of stroke among patients with rhinosinusitis: a population-based study in Taiwan. Wu CW, Chao PZ, Hao WR, Liou TH, Lin HW.

https://www.ncbi.nlm.nih.gov/pubmed/22801014

Surg Infect (Larchmt). 2012 Oct 19. Impact of Infectious Complications after Elective Surgery on Hospital Re-Admission and Late Deaths in the U.S. Medicare Population. Vogel TR, Dombrovskiy VY, Lowry SF.

https://www.ncbi.nlm.nih.gov/pubmed/23082877

POLL RESULTS

The fallacy I most like to participate in is

• confirmation bias 20%

• strawman 3%

• cherry picking 14%

• denial 43%

• ad hominem 17%

Chapter Twenty-One

It's been a long time

Thank goodness for prosthetic joints. With large parts of my practice slowly evaporating thanks to practice bundles, SKIP, and aggressive infection control, I can always count on the occasional prosthetic joint to get infected.

The patient had a prosthetic joint placed seven months ago, and for the first six months had no issues. Then he had increasing pain for a week or three before admission, culminating in the inability to walk. A tap of the hip demonstrated white blood cells and gram-positive rods, and he was admitted for an incision and drainage. Sometimes I am asked if I am the incision and drainage doctor, and I have to reply, "No." And if you see me with a scalpel in my hands, run---I am probably about to go all Phillip Reynolds.

He had no other medical problems of note and was very nontoxic on review of systems. Just a painful hip.

Well, you know, and I know, that an indolent infection with a gram-positive rod is likely to be Propionibacterium acnes, and indeed it was. The nice thing about P. acnes, at least from the perspective of an ID doc with one son in college, is that it is probably not preventable. It lives in oil of hair follicles, protected from both the skin prep and the pre-op antibiotics and then is dragged into the wound at surgery.

Not without surprise, it is more often seen in men: we are the greasy, hairy gender, although less so as I age.

The hallmark of this infection is the often long periods of time between the surgery and the infection. My record is seven years.

"Fifty patients with prosthetic hip (34), knee (10) or shoulder (6) infections were included and analyzed according to their symptom-free interval: < or = 2 years for 35 and > 2 years for 15 (mean interval: 11+/-6 years). The numbers of previous prostheses (p=0.04) were higher for the shorter-interval group, which had more frequent signs of infection (p=0.004). These findings suggest infection in most of the patients whose PJI symptoms appeared: < or = 2 years after the index operation, and colonization in the majority of those whose symptoms appeared > 2 years after index surgery. Treatment combining exchange arthroplasty with prolonged intravenous antibiotics was successful for 92% of the patients.

*CONCLUSION:*

P. acnes can cause different types of PJI: late chronic infections, colonization of loosened prostheses and, exceptionally, acute postoperative infections."

I find this a weird report: "colonization" with joint loosening? Sounds like infection à moi.

P. acnes is often there if you look for it, although you may increase the yield of cultures if you sonicate the prosthesis (I have never done this) and hold the culture for 14 days (I usually do).

The incision and drainage doc did his thing then I am going to do mine: six weeks of penicillin.

Rationalization

J Infect. 2007 Aug;55(2):119-24. Epub 2007 Apr 5. Propionibacterium acnes: an agent of prosthetic joint infection and colonization.

http://www.ncbi.nlm.nih.gov/pubmed/17418419

J Clin Microbiol. 2009 Jun;47(6):1878-84. Epub 2009 Mar 4. Microbiologic diagnosis of prosthetic shoulder infection by use of implant sonication.

http://www.ncbi.nlm.nih.gov/pubmed/19261785

J Clin Microbiol. 2011 Jul;49(7):2490-5. Epub 2011 May 4. Optimization of periprosthetic culture for diagnosis of Propionibacterium acnes prosthetic joint infection.

http://www.ncbi.nlm.nih.gov/pubmed/21543562

Chapter Twenty-Two

What infects humans, crocodiles and dolphins?

I get called for a necrotizing fasciitis of the perineum, aka Fournier's gangrene. Icky disease, since the treatment of necrotizing fasciitis is mostly surgical, and no one wants to have their perineum debrided.

Fournier's is usually seen in diabetics (as here, poorly controlled and a smoker as well) and is usually a mixed infection (Streptococci and anaerobes and the odd Enterobacteriaceae). I see a case or two a decade.

There were a couple of odd features:

First, he came in not because of symptoms related to his infection but due to supra-ventricular tachycardia/palpitations. The infection was found as a part of the ER exam.

Cardiac events, including arrhythmias, are common with community-acquired pneumonia. I know I keep harping on the vascular complications of infections, and perhaps you can add arrhythmia to the list, although the etiology is probably multifactorial. Still, my confirmation bias is set to 11.

The other odd thing was the microbiology. Group B streptococcus, S. agalactiae, was the only organism seen and grown from the extensive necrosis. And to complicate matters, the blood and urine grew MRSA, but nowhere else. So perhaps the MRSA was the cause of the arrhythmia, since he may have endocarditis, although the ECHO is negative. But bacteremia is bad for the heart.

"The frequency of new-onset AF in bacteremia is substantial. Initial CRP levels or white blood cell count do not seem to predict new-onset AF, as opposed to systemic inflammatory response syndrome. On the other hand, in patients with bacteremia, new-onset AF should be viewed as an indicator of increased mortality and morbidity."

Why not MRSA bacteremia and necrotizing fasciitis annoying the myocardium as well?

I had not thought of Group B streptococcus as a cause of necrotizing fasciitis, but there are multiple cases (about 40 reports in the PubMeds) in humans, saltwater crocodiles, and dolphins, occasionally with toxic shock syndrome as well.

Incision and drainage and antibiotics and he is getting better. The crocs and Flipper did not fare as well.

Rationalization

J Infect. 2012 Sep 11. pii: S0163-4453(12)00254-X. doi: 10.1016/j.jinf.2012.09.003. Risk stratification and prognosis of acute cardiac events in hospitalized adults with community-acquired pneumonia.

http://www.ncbi.nlm.nih.gov/pubmed/22981899

PLoS Med. 2011 Jun;8(6):e1001048. Epub 2011 Jun 28. Cardiac complications in patients with community-acquired pneumonia: a systematic review and meta-analysis of observational studies.

http://www.ncbi.nlm.nih.gov/pubmed/21738449

Cardiology. 2008;111(3):171-80. Epub 2008 Apr 25. New-onset atrial fibrillation in bacteremia is not associated with C-reactive protein, but is an indicator of increased mortality during hospitalization.

http://www.ncbi.nlm.nih.gov/pubmed/18434721

Epidemiol Infect. 2007 Nov;135(8):1248-55. Epub 2007 Apr 20. Necrotizing fasciitis in captive juvenile Crocodylus porosus caused by Streptococcus agalactiae: an outbreak and review of the animal and human literature.

http://www.ncbi.nlm.nih.gov/pubmed/17445318

J Vet Diagn Invest. 2005 Nov;17(6):617-22. Fatal necrotizing fasciitis and myositis in a captive common bottlenose dolphin (Tursiops truncatus) associated with Streptococcus agalactiae.

http://www.ncbi.nlm.nih.gov/pubmed/16475528

Chapter Twenty-Three

Can is not Ought

My goal of late would appear to be to try and alienate as many of my readers as I possibly can. It isn't. Really. But one of the principles by which I try and live is to be straightforward, within the boundaries of propriety. I have gathered a bunch of teaching awards from the residents over the years, but the one for which I am most proud was being voted "Attending most likely to tell it like it is."

So with that as background, I will confess to a serious lack of fondness for dog owners. Partly it's my neighbors, who think I want to listen to their dogs bark and howl despite have been given information to the contrary. And I am never happy that dogs use my yard as a toilet. I don't crap on your yard; why does your dog crap on mine? And I hate it when some dog comes running up barking at me or jumps on me. Sorry, I do not want to be in contact with your shaggy vermin or its excrement. If I behaved as most dog owners let their not-so-adorable pests act, I would be in jail.

But what really gripes my cookies are people bringing their dogs into the hospital. I am not talking Seeing Eye dogs or a family member bringing Fifi in so Granny can see her beloved poodle before dying. I am not even talking Pet Therapy animals, which I am always voting against at the infection control meetings. It is the other dogs---the pets and companions that appear to be showing up with increasing frequency in the hospital. Animals do not belong in hospitals.

Dogs are a source for infection, including MDRO:

"This study shows the emergence of conjugative plasmid-borne ESBLs among E. coli strains from companion animals in the United States, which may compromise the effective therapeutic use of ESCs in veterinary medicine."

and pets have been implicated in E. coli urinary tract infections in households.

"Within-household sharing of E. coli, including in households in which a member has a UTI, is common and can involve any combination of humans and pets. Identification of the underlying mechanism(s) could lead to novel preventive measures against UTI."

Unlike humans, dogs lick their, um, er, butts, and then lick their fur and you---not a potential vector I want in the hospital. At least I can get humans to wash their hands, but I can't alcohol foam a dog, although I keep trying to autoclave them. Without success.

Besides E. coli, dogs can carry Pasteurella, MRSA, Salmonella, Capnocytophaga and C. difficile, to name a few. If you ever watch an animal in the hospital, everyone wants to pet it and rarely do people think to wash their hands afterward.

In the case of companion animals, their handlers always insist that their vermin are covered by the Americans with Disabilities Act, and that if we try to remove the dog from the hospital we will be reported. That always stops us in our tracks.

It's a hospital--filled with the ill, the immunoincompetent and perhaps even the allergic who have no choice but to be in the hospital. You, with your yippy little beast, are visiting. Just because you CAN doesn't mean you OUGHT. What kind of self-centered human can't leave their contaminated and colonized dog for an hour while they go to the hospital? Or the grocery store? Or the restaurant?

But if I have learned one thing in 55 years, it is that "dog owner" is often a synonym for "inconsiderate."

Rationalization

Antimicrob Agents Chemother. 2011 Dec;55(12):5666-75. Epub 2011 Sep 26. Molecular characterization of resistance to extended-spectrum cephalosporins in clinical Escherichia coli isolates from companion animals in the United States.

http://www.ncbi.nlm.nih.gov/pubmed/21947397

Vet Rec. 2011 Apr 2;168(13):354. Epub 2011 Mar 23. Prevalence of antimicrobial-resistant Escherichia coli in dogs in a cross-sectional, community-based study.

http://www.ncbi.nlm.nih.gov/pubmed/21498238

J Infect Dis. 2008 Jan 15;197(2):218-24. Escherichia coli colonization patterns among human household members and pets, with attention to acute urinary tract infection. http://www.ncbi.nlm.nih.gov/pubmed/18179385

J Hosp Infect. 2009 Jul;72(3):268-9. Epub 2009 Mar 28. Contamination of pet therapy dogs with MRSA and Clostridium difficile.

http://www.ncbi.nlm.nih.gov/pubmed/19329224

ADA

http://www.nh.gov/disability/information/community/serviceanimals.htm

POLL RESULTS

The animal I would let in the hospital is

• dog. 37%

• cat. 3%

• children. 7%

• none. I would ban human visitors if I could. 32%

• only as meals. 17%

• Other Answers 5%

  • raccoon

  • leeches. Sterile ones are available and they have some therapeutic utility.

  • miniature goats

  • if the animal/children do not affect the other patients

  • Mark Crislip

  • Only those that have paid admission to enter.

Chapter Twenty-Four

Sleepy Head

The phone awakens me at 1 am. It's the ER calling. They evidently have a higher opinion of my neurological function upon awakening from deep sleep than I do. My dad was a cardiologist, and I remember the phone ringing all night growing up. About six months after retiring he became this whole new sharp, energetic person, probably finally getting caught up in his sleep. Fortunately, ID is nowhere as bad as cardiology for night time phone calls. It is existential angst and guilt that keeps me from sleeping.

They have a patient with an infection of the soft tissues around his eye, a periorbital cellulitis with an abscess. The best I can tell from the phone he has no risks outside of HIV and a lowish CD4 T-cell count, but no trauma or exposures to make me worry about an unusual organism, at least as the neurons are currently firing. He had a CT scan that was negative for anything evil.

Cellulitis with an abscess? Probably MRSA. Streptococci usually do not cause abscess except for the rare (at least in adults) Pneumococcal cellulitis, and the Streptococcus milleri group tends to avoid the face. He is allergic to penicillin, so I say start him on vancomycin. My son, who overhears many phone conversations, says ID is easy. All you ever say, he notes, is "Get cultures and start vancomycin."

I did not remember the patient's name the next morning, but how many MRSA facial cellulitis cases are in the hospital? Turns out none, as his blood cultures grow Acinetobacter baumannii.

That bug would not have been on my list even if I were fully awake. There is one case like it in the literature and three total of three with this beast causing cellulitis.

Why? My history yielded nothing, and he had no sinus infection to be a source. Oddly, I wanted to credit his dog as the source, but no luck. In dogs, Acinetobacter baumannii is a nosocomial infection. I suppose I will have to blame the confluence of HIV and bad luck since even in the HIV population it is a rare pathogen: Ten patients out of 1,923 (0.52%).

It was pan-susceptible, and he responded rapidly once we gave antibiotics that killed the bug.

Rationalization

Optometry. 2005 Mar;76(3):176-80. Acinetobacter as a causative agent in preseptal cellulitis.

http://www.ncbi.nlm.nih.gov/pubmed?term=15786636

J Formos Med Assoc. 2003 Sep;102(9):650-2. Community-acquired bacteremic cellulitis caused by Acinetobacter baumannii.

http://www.ncbi.nlm.nih.gov/pubmed/14625612

J Clin Microbiol. 2002 Feb;40(2):685-6. Community-acquired bacteremic Acinetobacter pneumonia in tropical Australia is caused by diverse strains of Acinetobacter baumannii, with carriage in the throat in at-risk groups.

http://www.ncbi.nlm.nih.gov/pubmed/11825997

Chemotherapy. 2001 Jan-Feb;47(1):19-28. Acinetobacter infections in patients with human immunodeficiency virus infection: microbiological and clinical epidemiology. http://www.ncbi.nlm.nih.gov/pubmed/11125229

POLL RESULTS

I am kept awake at night by

• phone calls. 9%

• guilt. 18%

• coffee. 18%

• the dread of certain death and an IRS audit. 18%

• I sleep like a baby and just as content. 24%

• Other Answers 13%

  • Hydraulic pressure

  • the boogeyman.

  • just as awaken as if it were 11 am...

  • Teenagers, regardless of whether in or out of the house.

  • existential angst and post-menopausal insomnia

Chapter Twenty-Five

Baader-Meinhof

I go through a ton of medical articles. For each Puscast I scan several hundred articles, whittle them down to about 50 that I read, and end up discussing maybe 25. I have collected over 10,000 references representing 3.6 gigabytes of data. How much of that can I actually recall? Uncertain, but I bet only a fraction. It is why I am glad to have Google as my alternative memory.

One weird thing about reading the medical literature is how occasionally I read an article about a rare infection, and shortly after that I see a case. It happened after I read an article on S. lugdunensis bacteremia likely being a manifestation of endocarditis and that Osler-Weber-Rendu patients get S. aureus septic joints.

I know this is simply a manifestation of the Law of Large Numbers:

"That a particular specified event or coincidence will occur is very unlikely. That some astonishing unspecified events will occur is certain. That is why remarkable coincidences are noted in hindsight, not predicted with foresight."--David G. Myers

The law of truly large numbers says that with a large enough sample many odd coincidences are likely to happen.

I read a lot. I am bound to occasionally see a disease after I read a paper on the topic.

The patient was admitted with S. pneumoniae bacteremia three weeks ago. No focal infection was found, but he had had a splenectomy from trauma years ago. He responded immediately to antibiotics, and his exam was negative as was an ECHO, so he was treated with IV antibiotics while in the hospital and discharged on oral amoxicillin. As he approached the end of this antibiotic therapy, he had two transient neurologic events: one where his vision whited out and another where his arm became painful. Both lasted a few minutes.

Nothing was changed on exam but his ECHO. Now, where it had been pristine, it was thickened but no vegetation. A couple of weeks ago I came across the following:

Streptococcus pneumoniae endocarditis is uncommon. It has a predilection for the aortic valve and is associated with high mortality. We present 3 consecutive cases of pneumococcal endocarditis, each preceded by a different extracardiac infection but all causing destructive aortic valve lesions associated with severe regurgitation, in the absence of vegetations on transesophageal echocardiography. This case series illustrates the aggressive nature of pneumococcal endocarditis and the need for early diagnosis. Echocardiography should be considered in all individuals with persistent extracardiac pneumococcal infections.

So my reading the article somehow caused the disease in this patient to come into existence. Magic. Learning about a topic and then seeing it is called the Baader-Meinhof Phenomenon, named for some reason after a German terrorist group. Also known as the "Frequency illusion -- the illusion in which a word, a name, or other thing that has recently come to one's attention suddenly appears everywhere with improbable frequency."

So is it Pneumococcal endocarditis? I can't disprove it short of looking at the valve in the operating room. The Pneumococcal urine antigen is negative, which kind of mediates against active infection. It could be that he had endocarditis and is having late, post-infectious embolic events. I had one patient, clinically cured, who had a fatal cerebral embolism on the last day of treatment for S. viridans endocarditis.

Safe is better than sorry, so I suggested a course of therapy for endocarditis.

Rationalization

Can J Cardiol. 2012 Aug 24. Pneumococcal Endocarditis Causing Valve Destruction in the Absence of Vegetations on Transesophageal Echocardiography: A Series of Three Consecutive Cases.

http://www.ncbi.nlm.nih.gov/pubmed/22926036

Stroke in infective endocarditis. doi: 10.1161/01.STR.21.5.695 Stroke. 1990;21:695-700

http://stroke.ahajournals.org/content/21/5/695.full.pdf

Chapter Twenty-Six

First One Thing...

Thanksgiving is over with minimal weight gain, although I did make my best turkey ever and my best pumpkin pie ever. As much as I enjoy my work, it is not better than four days on the Oregon Coast. As I have mentioned before, I have yet to have a patient tell me at the end of their days, "Doc, I wish I had spent more time at work." But if you have to work, there is no more interesting job than ID.

The patient has myelodysplastic syndrome and runs a WBC of 0.9 thousand cells per microliter on a good day, or about five times the lower limit of normal. He comes in with a progressive cough, mostly nonproductive, and fevers. A chest X-ray and CT scan of the chest demonstrate extensive pneumonia. He is placed on hospital-acquired pneumonia therapy and does not get better.

All the routine studies are negative. Gram stain is negative, and cultures grow nothing, so off to bronchoscopy for a bronchoalveolar lavage (BAL). His low platelet count precludes a safe biopsy.

All the stains and cultures are negative on the BAL, but the galactomannan is quite positive, indicating fungi. So it is presumptive Aspergillus.

This is the first time I have made the diagnosis of pulmonary aspergillosis (actually it was the pulmonologist who sent off the test) on a BAL. He had negative everything else for Aspergillus: negative cultures, no crescent sign on the CT scan and a negative blood galactomannan.

I was trained in the old school, where you liked to have a positive biopsy to prove invasive molds, but the old school had to close due to an inadequate tax base.

The BAL is a better than serum for galactomannan and there are multiple studies to demonstrate its operational characteristics for diagnosing Aspergillus pneumonia.

He was started on voriconazole, an anti-fungal, and he was slowly improving both clinically and radiographically, much to my delight. I remember the old days when all we had was amphotericin B (any idea why the "B" is capitalized?), and I cannot remember ever curing an Aspergillus without immune reconstitution or cutting it out. Now? I can't remember a failure. Of course, my memory is not what it used to be.

And then...

Next entry: Then Another.

Rationalization

BMC Infect Dis. 2010 Mar 3;10:44. Galactomannan testing of bronchoalveolar lavage fluid is useful for diagnosis of invasive pulmonary aspergillosis in hematology patients.

http://www.ncbi.nlm.nih.gov/pubmed/20199673

The turkey:

http://www.finecooking.com/recipes/bacon-wrapped-smoked-turkey.aspx

The pie:

http://www.myrecipes.com/recipe/pumpkin-caramel-ice-cream-50400000123748/

POLL RESULTS

For Thanksgiving I excel at

• the turkey. 19%

• the pie. 22%

• the bread. 7%

• getting drunk while others cook. 22%

• depression, eating alone in my apartment. 14%

• Other Answers 16%

  • washing the dishes after the feast

  • Being Australian and not knowing what thanksgiving is.

  • shopping, what else?

  • just eating.

  • eating out and tipping big

  • eating

Chapter Twenty-Seven

Then Another

When last we left our patient, he was slowly improving on voriconazole for his Aspergillus pneumonia and was discharged home.

Now he returns a week later septic-ish: still neutropenic, febrile, tachycardiac, hypotensive and some mild abdominal pain.

He stabilizes on antibiotics and the usual ICU care, and the work-up shows what looks to be a right-sided colitis with a very thick bowel wall.

Typhlitis, aka neutropenic enterocolitis, is the presumptive diagnosis. Due to a low platelet count and general debility, the patient is not a good candidate for surgery, and he does not want it anyway. Since he is clinically improved, we continue the antibiotics and give granulocyte colony stimulating factor (G-CSF) to boost his white blood cell count. There are the odd case reports of patients surviving typhlitis without resection, so maybe.

Then the blood cultures turn positive...For Clostridium septicum, an anaerobic bacterium. Didn't see that coming. Usually, C. septicum is associated with bowel cancer, where necrosis of the tumor gives a toehold for the anaerobe to take off. I have also seen it post-chemo for lymphoma, where I suppose that necrosis of lymphoma in the Peyer's patches results in the same fertile soil for Clostridia growth.

C. septicum is a cause of typhlitis, and it also has the odd habit of causing infections in arteries-- mycotic aneurysms--neither of which are good.

On antibiotics he continues to improve, and to see what is going on in this bowel a CT is repeated about ten days later. Now what was thought to be cecal colitis looks classic for intussusception. The patient is eating and having no bowel symptoms, still pancytopenic, still not a surgical candidate.

So he probably never had typhlitis, which is why he did well. Where did the Clostridia come from? The intussusception? The only similar case was in a dog, and if intussusception were the source, that would imply bowel necrosis that would not have improved. Occasionally Clostridia show up in a sterile site and, while real, are not clinically significant. I don't think so in this case, though, given the clinical presentation. An endoscopic laparotomy sure would have been nice to clarify.

So we are going to treat him for a while and see what happens. He has had enough bad luck with weird infections. You do not want to be an interesting case for me.

Rationalization

BMJ Case Rep. 2012 Sep 7;2012. pii: bcr2012006167. doi: 10.1136/bcr-2012-006167. Clostridium septicum sepsis and its implications. http://www.ncbi.nlm.nih.gov/pubmed/22962388

Dig Dis Sci. 2005 Feb;50(2):215-20. Neutropenic enterocolitis in adults: case series and review of the literature.

http://www.ncbi.nlm.nih.gov/pubmed/15745075

J Clin Pathol 1984;37:335-343 Neutropenic enterocolitis due to Clostridium septicum infection

http://www.ncbi.nlm.nih.gov/pmc/articles/PMC498711/pdf/jclinpath00174-0103.pdf

Br Vet J. 1989 May-Jun;145(3):291-2. Double intussusception fatally complicated by clostridial infection in a dog (a case report).

http://www.ncbi.nlm.nih.gov/pubmed/2736386

Chapter Twenty-Eight

Half of My Favorite Quartet

The patient has HIV, controlled for years with a low viral load and a CD4 count that hovers around 260 cells per cubic millimeter---low, but not AIDS-defining low. What medicine gives, it takes away, and a renal transplant has resulted in medications that make his immune function suspect. He has been fine for years with both, however, until this admission--where he has several days of weakness and the sudden onset of a painful rapidly growing mass on his chest. No rubor, no calor, just dolor and tumor.

He also has nothing else of note: no fevers, chills, trauma, seat belt injury, or any other reason for an infection of his chest wall.

On exam, it is a well-circumscribed, golf-ball-sized tender mass about three inches lateral to the nipple. It is neither red nor hot.

What could it be?

Cold Staphylococcal abscess is most likely, I suppose, but only because common things are common and tautologies are tautological. But there is no reason for a Staph abscess, and the chest/breast is an odd place for a spontaneous abscess.

Malignancy? The ultrasound suggests it is vascular, but that is an odd place. Same problem for fungus and acid-fast bacilli, the other possibilities: odd place and no reason for the infection.

So they tap it and yep, it is S. aureus. It is a bug that likes to occasionally show up as a mass with no classic signs of infection. The usual is a Brodie's abscess, a S. aureus of long bone that presents more like malignancy, but I have seen lung and liver abscess due to S. aureus that has none of the typical inflammatory reactions that you expect with S. aureus. Other bugs can cause a cold abscess as well, such as M. tuberculosis.

Of course, it is better to have an infection than a tumor, and he had a definitive incision and drainage the next day. Never use only aspiration drainage on an MRSA abscess (although his was MSSA) as it is inferior:

" Patients with CA-MRSA (n=33) were less likely to receive successful drainage with needle aspiration (8% versus 55%) or incision and drainage (61% versus 89%). The difference for needle aspiration and incision and drainage was 47% (95% CI 15% to 57%) and 28% (95% CI 4% to 45%), respectively.

CONCLUSION:

Ultrasonographically guided needle aspiration is insufficient therapy for skin abscesses."

On the other hand, since I never do a procedure, if you see me coming at you with a scalpel, run --I have likely snapped. One hypothesis for his infection was that he recently started spironolactone, which induced unilateral gynecomastia that became infected. Nothing on the PubMeds or the Googles to support or deny that hypothesis, but it's like an iPad mini: I don't buy it, but I sure like it.

Rationalization

Ann Emerg Med. 2011 May;57(5):483-91.e1. Epub 2011 Jan 15. A randomized controlled trial of incision and drainage versus ultrasonographically guided needle aspiration for skin abscesses and the effect of methicillin-resistant Staphylococcus aureus.

http://www.ncbi.nlm.nih.gov/pubmed/21239082

POLL RESULTS

I like but will not buy

• iPad mini. 35%

• Nexus 7. 9%

• Kindle Fire. 15%

• Microsoft Surface. 9%

• iPhone 5. 24%

• Other Answers 9%

  • Vote

  • all the above

  • Mail order brides.

  • Lamb chops.

  • Diamonds

Chapter Twenty-Nine

I See All The Weirdness

I have been the sole, and the soul (no, not really), doctor at 3 or 4 hospitals for the last twenty-three-plus years. So I see all the weird cases. I think the cases were odder back in the day. For example, in my first decade of practice I saw one each of the HACEK group of bacteria causing endocarditis, but not a one this century. It was always better in the golden past.

I have patients who will say something to the effect that "If something weird is going to happen, it is going to happen to me." Nothing weird ever happens to me, but I certainly see enough weirdness infecting others.

It started commonly enough. MSSA bacteremia without a source or a reason, not that S. aureus needs either. No murmur, no emboli, but it should be endocarditis. Usually is. But the transthoracic echocardiogram?

There was a 1x1 cm mass coming out of the right atrial wall, consistent with a myxoma---a non-cancerous primary tumor of the heart. Nothing on the valves. Wrong place for a vegetation or clot.

Is it infected? Could be. About a decade ago, I had a patient who had a left atrial myxoma infected with Streptococcus viridans. Kind of rare, but there are a couple dozen cases on the PubMeds. It is the kind of thing people evidently like to write up. I (or preferably a cardiac surgeon) would have to take it removed, as we had to do on the first case I saw, to prove it. It hardly seems worth it at the moment.

The PubMeds have a hodgepodge of bacteria infecting myxomas, more commonly left-sided since 75% of myxomas are left-sided, as was the first case I saw.

For now, he is hemodynamically stable, so there will be a course of antibiotics before the myxoma is resected. Sometimes I wish unusual things would happen to me. Powerball comes to mind.

Rationalization

Medicine (Baltimore). 1998 Sep;77(5):337-44. Infected cardiac myxoma. Case report and literature review.

http://www.ncbi.nlm.nih.gov/pubmed/9772922

HACEK = *Haemophilus parainfluenzae, Haemophilus aphrophilus, Haemophilus paraphrophilus, Aggregatibacter actinomycetemcomitans (which no one can say correctly), Cardiobacterium hominis, Eikenella corrodens, and **Ki*ngella.

I pronounce HACEK with a soft "c" while my colleague from the East Coast uses a hard "c." I am the correct one, of course.

POLL RESULTS

Weirdness

• happens to others. 11%

• happens to me. 18%

• weirdly, never happens. 5%

• is found in Portland. Keep it there. 46%

• is only noticed by those who can't have it or be it. 19%

Chapter Thirty

The Joint Ain't Jumping Any Time Soon

The joints of the pelvis are a pain. Infection is rare, and as a result, the diagnosis of pelvic septic arthritis can be missed or delayed.

The patient is an IV drug user coming in with hip pain and a fever. First they MRI the hip: Negative. Then MRI the spine. Negative. The fields of the MRI are too narrow, and for both studies the sacroiliac joint can be just outside the imaging. The same can be true of other areas in the pelvis that are prone to infection: paraspinous muscles, iliopsoas, femoral head, joint, and bursa.

A thorough exam and history can usually pinpoint which anatomical space is involved, but a withdrawing heroin addict demanding narcotics often is not the most cooperative patient, and an MRI is easier.

A pelvic CT scan is a blunter implement---imaging all the pertinent structures, but not sensitive as in this case, where it showed no pathology. Then the patient had a bone scan that glowed like Chernobyl in the right sacroiliac joint. That, combined with the methicillin-sensitive S. aureus in the blood, made the diagnosis. Of interest, the bone scan glowed in the femoral neck on the other side, and upon review of the CT it was decided that there was a Brodie's abscess on the right as well.

It always pays to go over the films with the radiologist.

Usually, you like to incise and drain infected joints---but the sacroiliac joint, the surgeons assure me, is difficult to get to and you have to chop through a bunch of muscles to get to it. Sounds unpleasant.

There was a recent review of infected sacroiliac joints. More Baader-Meinhof phenomena, where I read about a topic and then shortly thereafter see a case. Or perhaps a plate of shrimp:

"A lot o' people don't realize what's really going on. They view life as a bunch o' unconnected incidents 'n things. They don't realize that there's this, like, lattice o' coincidence that lays on top o' everything. Give you an example; show you what I mean: suppose you're thinkin' about a plate o' shrimp. Suddenly someone'll say, like, plate, or shrimp, or plate o' shrimp out of the blue, no explanation. No point in lookin' for one, either. It's all part of a cosmic unconciousness."

The article really did not answer the question as to the necessity of incision and draining of the joint. The best they say is

"While exceptional, surgical treatment may be proposed in the case of abscesses, proven osteomyelitis, bone involvement or necrosis, and failed antibiotic therapy ... Drug addicts, however, have a higher risk of abscesses, osteomyelitis, and relapse after stopping antibiotic therapy."

I tend not to want to skimp on therapy just because the patient is an IV drug abuser. Death or worse seems too high a price to pay for bad decisions and addiction, but for the moment he is somewhat more interested in heroin than antibiotics.

I always fret about using ceftriaxone for methicillin-sensitive staph, although there is data to suggest it is fine for osteoarticular infections.

"In this comparison of ceftriaxone versus oxacillin for MSSA osteoarticular infections, there was no difference in treatment success at 3-6 and > 6 months following the completion of intravenous antibiotics."

I will give the ceftriaxone a try, and we will do what we can to maximize the compliance with the adherence.

He left against medical advice, never to be seen again. I assume he died.

Rationalization

BMC Infect Dis. 2012 Nov 15;12(1):305. [Epub ahead of print] Infectious sacroiliitis: a retrospective, multicentre study of 39 adults.

http://www.ncbi.nlm.nih.gov/pubmed/23153120

Clin Infect Dis. 2000 Jan;30(1):205-7. Ceftriaxone therapy for staphylococcal osteomyelitis: a review. Guglielmo BJ, Luber AD, Paletta D Jr, Jacobs RA.

http://www.ncbi.nlm.nih.gov/pubmed/10619757

Clin Infect Dis. 2012 Mar 1;54(5):585-90. Epub 2011 Dec 5. A retrospective comparison of ceftriaxone versus oxacillin for osteoarticular infections due to methicillin-susceptible Staphylococcus aureus.

[http://www.ncbi.nlm.nih.gov/pubmed/22144536]{.underline}

The Baader-Meinhof Phenomenon

[https://www.damninteresting.com/the-baader-meinhof-phenomenon/]{.underline}

Repo Man

https://en.wikiquote.org/wiki/Repo_Man_(film)

POLL RESULTS

After reading these blog entries, I:

• See a similar case. 11%

• Eat a plate of shrimp. 17%

• Doze fitfully, with bad dreams. 17%

• Complain they just don't teach proper English any more. 19%

• Wish I had done ID instead of my current speciality. 25%

• Other Answers 11%

  • Read Pharyngula

  • have mixed feelings about being retired

  • Always learn something

Chapter Thirty-One

Phone Call

I'm on call this weekend, and I cover seven hospitals, although fortunately not all hospitals have active patients.

I finish rounds, come home, go out for a burger with the family, and get home to watch the end of the Blazers game. Not looking good. The phone rings. I answer the page, and it is a doc wanting to discuss one of the cases I am covering.

The case is a transfer in from another hospital, very ill, and at first I think it is one of the referring docs wanting to discuss the case. I am a bit put off as the area code is in Washington but the transfer is from down the Valley, and the person on the phone is badly mispronouncing the name of the city the patient is from.

He suggests a disease the patient did not have, and what I discover after my initial confusion is that this is some doctor (I think) who has read about the case on the internet and decided to call me with advice on the diagnosis.

What the hell?

Now I don't mind help on a case; I ask for it all the time. Medicine is a team effort, and many a time a resident or attending will suggest a diagnosis that I did not consider, and I will run with it. A great idea should always be considered, no matter the source.

But who makes a diagnosis from an internet/newspaper report? I can't. I do not always have confidence that I know what is going on in a patient after a history and physical, much less after reading about it on the web. And then who decides to pester someone with their idea at 9 o'clock at night?

Needless to say, once I figured out it wasn't a treating physician calling me, I only said I could not discuss the case with a random phone caller. I tried to be polite, but I do not think I succeeded. I was more than a little pissed and irritated, and it is hard to get me really pissed. I try hard to always exhibit the 5 As of a consultant: Ability, Availability, Affability, Accountability, and Appearance. Well, 4 out of 5 ain't bad.

And it is not that we do not know what is going on; we do. We have a diagnosis, and it is a classic case. The caller? Let's say he did now wow me going on about a case about which he is Jon Snow. At least he didn't suggest iron overload. I would have really lost it.

Seriously. I am as arrogant as the next doctor, but I would never think I could diagnose a case accurately based on a newspaper report. The media always get it wrong. If you are reading stories on the web and you think you know what is going, you don't. You want to help? Come in and do history and physical. Then we will talk. Otherwise, let me watch the game. Jeeze.

Chapter Thirty-Two

A Great Idea

The patient is an elderly-ish male with a prosthetic hip that almost never worked right. It was slow to recover function, and over the next six months he had increasing pain and decreasing function and a little loosening.

A tap of the hip showed white blood cells, possibly indicating infection, but cultures were negative. So it was off to the operating room where more pus was found, it was washed out, and the plastic on the prosthesis exchanged.

Gram stain and cultures were negative initially, but after three days the plates started to grow an anaerobic gram-positive rod. OK. Propionibacterium acnes. Been there, killed that. I put him on penicillin G. But it wasn't. It was Clostridium bifermentans.

What? Never heard of that Clostridia before. No surprise. The PubMeds suggest a smattering of cases in humans with mostly infections after trauma, a few cases of endocarditis, and one case of osteomyelitis. No infected joints, prosthetic or otherwise. Evidently, Clostridium bifermentans is not so good to grass carp either.

He has done well on IV penicillin G, and I plan for a three-month course of oral metronidazole to play it safe and kill off any residual bacteria.

I have an amazing number of what I think are great ideas for projects that other people should do. Here is one.

I have been writing these essays since September 2008, two or three entries a week. That is well over 500 cases, mostly different, some unique, all interesting. It is ID. By definition it is interesting. I will never have the time or inclination to do a formal write up of these cases, and if my practice is representative, there are thousands of curious ID cases each year that are unreported.

What is needed is a simple searchable internet database where clinicians can report their curious cases. Just the basics: the bug, the organ involved, the susceptibilities, the risks, the treatments, and outcome. Like my C. bifermentans case.

Of course, that requires money, an institution, and support: servers, programmers, and above all someone to keep the database in people's minds so that they use it. It would be a great resource.

I have suggested this to the Infectious Diseases Society of America and others several times in the past; since I was no one from nowhere, they understandably ignored me.

This database is not something I want to do but instead I want to see done by someone else. If anyone runs with the idea, let me know. I will quit my occasional harping on the topic. But it is a great idea.

Rationalization

J Clin Microbiol. 1994 Nov;32(11):2867-8. Clostridium bifermentans bacteremia with metastatic osteomyelitis.

http://www.ncbi.nlm.nih.gov/pubmed/7852591

[Clostridium bifermentans infection in grass carp (Ctenopharyngodon idella)]. Hoffmann RW, Stolle A, Eisgruber H, Kölle P. Berl Munch Tierarztl Wochenschr. 1995 Feb;108(2):55-7. German.

http://www.ncbi.nlm.nih.gov/pubmed/7786278

Chapter Thirty-Three

A Ticket to the Fungal Ball

The patient has the sudden onset of unilateral painful red eye and a decrease in visual acuity.

A dialysis patient due to diabetes, the patient has no bad habits or risks and is sent off to the ophthalmologist. He is found to have endophthalmitis with what looks to be a small fungus ball in the back of the eye, but no involvement of the optic nerve or the vitreous.

The ophthalmologist (if he worked for Apple would he called an iDoc?) is pretty adamant that it looks like Candida, although it sure would be nice if there was a more typical risk for Candida endophthalmitis---like heroin use or a recent stay in the ICU.

Otherwise, the patient is stable, so I suggest blood cultures (negative) and a beta-D-glucan test for invasive fungal infections. It is modestly, but really, positive.

On a course of the anti-fungal fluconazole, his vision improves and the fungus ball goes away.

I spent part of my fellowship, way back in the mists of time, working on the pathogenesis of disseminated Candida. It is safe to say that there is no worser bench researcher than me, ever. So I am a clinician. But it did result in my one publication, which is not worth the electrons to read.

I spent a fair amount of my waking hours thinking about Candida and its propensity for infecting the eye. Now I just have PTSD-like flashbacks to my fellowship years.

Candida involves the eye in about 20% of patients with positive blood cultures, if followed prospectively, and is often asymptomatic. It is not supposed to be helpful as a diagnostic tool without a prior diagnosis, although every couple of years I have a high-risk patient in whom I suspect disseminated Candida and I hire an ophthalmologist and they find a fungus ball for me.

The "right" approach to candidemia and whether they should all have eye exams is open to discussion. I don't unless they have symptoms or suspicion of complicated disease (prolonged fevers/bad host), but I tend towards the "give lots of antifungals for a minimum of two weeks" school so I am certainly inadvertently treating a lot of mild eye disease. I try not to order tests if they are not going to change treatment.

Since we could not find any source for the Candida, I assume the patient had a transient fungemia as a result of a dialysis needle stick and was very, very unlucky.

Rationalization

Infection. 2012 Dec 2. [Epub ahead of print] Duration of treatment for candidemia and risk for late-onset ocular candidiasis http://www.ncbi.nlm.nih.gov/pubmed/23212461

Clin Infect Dis. 2011 Aug 1;53(3):262-8. doi: 10.1093/cid/cir355. Ocular manifestations of candidemia. http://www.ncbi.nlm.nih.gov/pubmed/21765074

POLL RESULTS

I still have nightmares about

• high school. 30%

• college. 6%

• medical school. 16%

• residency. 16%

• today. 20%

• Other Answers 12%

  • IRON GUY

  • my teen's prom night

  • Falling and/or flying, without outside assistance.

  • work

Chapter Thirty-Four

Close Enough

Endocarditis can be a difficult disease to diagnose definitely. OOhhhhh. Dr. Alliteration is in the house. Often the data are suggestive, but short of taking the valve out and looking at it, there is a pall of uncertainty over some cases.

Take today's case. In favor of the diagnosis: a month of decline, several neurologic events (can't talk for an hour, vertigo with falling), splinter hemorrhages on the right nail beds and Enterococcus in the blood 24 hours apart with no other source evident. There is nothing like a sustained bacteremia with no source to make an ID doc fret about an endovascular infection.

Against the diagnosis: female (enterococcal endocarditis more often male), the documented fever began 48 hours after admission, the urine culture was negative at admit, but grew Enterococcus 48 hours later (no Foley), the splinters looked more traumatic than embolic and the ECHO has no vegetations, only the standard valve abnormalities of aging. Cannot rule out endocarditis mwa mwa mwa mwa. I read these canned reports filled with disclaimers and all I hear are the adults in Charlie Brown.

And there are no cardiac abnormalities or murmurs that would predispose for endocarditis. I know endocarditis can involve normal valves for no reason. Years ago I took care of a thirtysomething female who destroyed her valve with enterococcal endocarditis and she too had no reason for enterococcal endocarditis.

So how much is due to endocarditis and how much is a red herring?

What tipped me to treating for endocarditis was that the patient may indeed have some underlying cardiac abnormalities. The patient has carcinoid, the source of which is as of yet undiscovered, but all the symptoms are present: the flushing, diarrhea and high serotonin levels. Carcinoid can lead to endocardial lesions:

"Carcinoid heart disease is characterized by endocardial plaque-like deposits found predominantly on right-sided heart valves, leading to the combination of valvular stenosis and regurgitation. Left-sided cardiac involvement can also occur in <10% data-preserve-html-node="true" of patients."

These plaques are not just on the valve but can occur on the endocardium, although the smattering of cases in the literature (smattering is less than 6) are all infections of the tricuspid valve or Eustachian valve.

So is this carcinoid-related endocarditis? Close enough that I think I should treat. But I am not happy with the decision. Safe is better than sorry, but the patient has hives to ampicillin and ceftriaxone and gentamicin resistance so it will have to be vancomycin and streptomycin with a CrCl of 30. Argh. Not so safe, but somehow I am thinking I will avoid linezolid.

Rationalization

Cardiol Rev. 2012 Jul-Aug;20(4):167-76. doi: 10.1097/CRD.0b013e31824c866e. Carcinoid heart disease.

http://www.ncbi.nlm.nih.gov/pubmed/22314145

Br Med J (Clin Res Ed). 1988 Mar 5;296(6623):682. Infective endocarditis complicating tricuspid valve disease in the carcinoid syndrome.

http://www.ncbi.nlm.nih.gov/pubmed/3128367

Ann Pharmacother. 2005 May;39(5):956-61. Epub 2005 Apr 12. Serotonin toxicity associated with concomitant use of linezolid.

http://www.ncbi.nlm.nih.gov/pubmed/15827071

mwa mwa mwa mwa

http://www.youtube.com/watch?v=ss2hULhXf04

POLL RESULTS

My motto for the practice of medicine is

• First, do no harm. So as a rule I do nothing. 6%

• Safe is better than sorry. 22%

• It is always 5 p.m. somewhere. 12%

• It is easier to get forgiveness than permission. 20%

• Once you've got a task to do, it's better to do it than live with the fear of it. - Logen Ninefingers 34%

• Other Answers 6%

  • If you don't have the time to do it right when will you have time to do it over, and how are you going to explain yourself to the next-of-kin?

Chapter Thirty-Five

Avoid the 3&3s

We have been having a ton of gastrointestinal symptoms around my hospitals. Given the volume of vomiting and diarrhea, "a ton" may be an underestimate.

There have been outbreaks in a few long-term care facilities that appear to be norovirus (some confirmed), and we just got over an outbreak at one my hospitals of norovirus where several dozen staff and patients were affected and we had to shut down a unit for ten days.

I am paranoid about norovirus, since I have now lived through three hospital outbreaks in my 23 years and it is not fun. I hate the smell of bleach.

I remain convinced that one outbreak was associated with a large bowel W&Ws that was shared on the unit.

Norovirus is the most infectious disease we know. Infectious. It is in my title, and I take it very seriously:

"We estimate the average probability of infection for a single Norwalk virus particle to be close to 0.5, exceeding that reported for any other virus studied to date."

And the virus can spread in the vapors of vomit to infect others:

"The connection (for the norovirus) turned out to be a reusable grocery tote bag filled with the cookies and other food items like chips and grapes that had been sitting on the floor of the bathroom where the first girl had repeatedly gotten sick."

And can hang around for a long time:

"This study demonstrates that Norwalk virus in groundwater can remain detectable for over three years and can remain infectious for at least 61 days. "

Part of the problem is nausea and vomiting and diarrhea are common manifestations of many illnesses, so people may not be as compulsive with infection control for symptoms as they would be if we had a name of the beast. Diagnosis is currently delayed, as the Norovirus test is a send- out with a three-day turnaround. So I fret when I am told that staff are calling in with nausea and vomiting. So far no one has been at work when the symptoms hit, but 'tis the season for sharing food.

If the food is not wrapped in cellophane or deep-fried, I am not going to eat it. Especially the E&E's.

Rationalization

J Med Virol. 2008 Aug;80(8):1468-76. doi: 10.1002/jmv.21237. Norwalk virus: how infectious is it?

http://www.ncbi.nlm.nih.gov/pubmed/18551613

Appl Environ Microbiol. 2011 Oct;77(19):6884-8. doi: 10.1128/AEM.05806-11. Epub 2011 Aug 19. Norovirus infectivity in humans and persistence in water.

http://www.ncbi.nlm.nih.gov/pubmed/21856841

Epidemiol Infect. 2002 Oct;129(2):355-60. An outbreak of viral gastroenteritis following environmental contamination at a concert hall.

http://www.ncbi.nlm.nih.gov/pubmed/12403111

J Infect Dis. 2012 Jun;205(11):1639-41. doi: 10.1093/infdis/jis250. Epub 2012 May 8. A point-source norovirus outbreak caused by exposure to fomites.

http://www.ncbi.nlm.nih.gov/pubmed?term=22573873

Chapter Thirty-Six

Gas

Everything makes gas. The end result of metabolism is to make gas, as my kids have proven on innumerable occasions.

The patient is an intramuscular heroin user who came in with an abscess. It was incised and drained in the ER and he was sent out on trimethoprim-sulfamethoxazole for presumed methicillin-resistant Staphylococcus aureus.

He is back in less than twenty-four hours with a marked increase in pain, and a CT shows this on the scout:

{width="5.375in" height="6.041666666666667in"}

and this

{width="6.916666666666667in" height="5.666666666666667in"}

and this

{width="6.125in" height="4.875in"}

and this.

{width="6.125in" height="4.875in"}

Lots of gas. Not good. Off to the OR, they find all the muscles that are surrounded by gas to be dead, and they are removed. No particularly repulsive odor to suggest anaerobes.

Necrotizing myositis. Group A Streptococcus? A Clostridium? MRS.? Perhaps a mixed synergistic necrotizing infection. All are certainly possible, especially the Clostridium, which is found in heroin.

But no. Klebsiella pneumoniae with a whiff of Enterococcus is all that grows. It is not, the lab tells me, the hyper-viscous strain, but the run of the mill Klebsiella.

Supposedly rare as of 2007:

"Necrotizing fasciitis caused by Klebsiella pneumoniae is rare, with 11 documented cases in the literature, all occurring in Asia and the Middle East."

Klebsiella pneumoniae has been increasingly common in some parts of the world:

"Of 134 NF cases, 88 were monomicrobial, of which the most common pathogens were GAS (n = 16) and K. pneumoniae (n = 15). Monomicrobial KP-NF entailed a moderate risk of limb loss (20% vs 25%; P = 1.000) and high mortality (47% vs 19%; P = .135), and it was more likely to involve bacteremia (80% vs 31%; P = .011), concomitant distant abscesses (27% vs 0%; P = .043), and underlying immunocompromising conditions (100% vs 63%; P = .018), compared with GAS-NF. The isolated K. pneumoniae strains (n = 10) were of capsular polysaccharides genotype K1 (n = 4), K54/K20/K5 (n = 4), K2 (n = 1), and K16 (n = 1). All strains carried rmpA, iucABCDiutA, and iroA. Genotype K1 strains had a significantly higher risk of concomitant distant abscesses, compared with non-K1 strains (75% vs 0%; P = .033)."

He has not been out of Oregon, but his heroin has. It is my understanding that we get much of our heroin from Asia, so perhaps that is the source, although I can find nothing on the interwebs to suggest Klebsiella is found in black tar heroin, just Clostridium.

It was resistant only to ampicillin and from an ID perspective he is doing well. He has sustained extensive damage to his ability to use his leg and will be a long time recovering.

Rationalization

Clin Infect Dis. 2012 Oct;55(7):930-9. doi: 10.1093/cid/cis565. Epub 2012 Jun 19. Recent trend of necrotizing fasciitis in Taiwan: focus on monomicrobial Klebsiella pneumoniae necrotizing fasciitis.

http://www.ncbi.nlm.nih.gov/pubmed/22715175

Surg Infect (Larchmt). 2007 Apr;8(2):227-32. Klebsiella pneumoniae necrotizing fasciitis and septic arthritis: an appearance in the Western hemisphere.

http://www.ncbi.nlm.nih.gov/pubmed/17437368

J Forensic Sci. 2007 Jul;52(4):920-3. Epub 2007 May 25. Necrotizing fasciitis: manifestations, microbiology and connection with black tar heroin.

http://www.ncbi.nlm.nih.gov/pubmed/17524065

Chapter Thirty-Seven

Two in a Row

Do events come in threes? I doubt it, but events do tend to occur one right after the other, and that includes infections. I note that when I am consulted for an infection, it is always the case that I get another consult for an infection.

Last week I mentioned a probable enterococcal endocarditis. This week I saw a male who a month ago had a lithotripsy for kidney stones. At the time he had enterococcus in the urine, for which he had a short course of ciprofloxacin.

A month later he comes in with progressive decline and 48 hours where he just couldn't get out of bed. Now the blood cultures have Enterococcus, but the urine is clean. He has a nice heart murmur but no other stigmata of endocarditis.

It is a variation of a case that frequently occurs in the Medical Knowledge Self-Assessment Program materials: an elderly male who gets enterococcal endocarditis after a urinary procedure. Bacteremia is common with lithotripsy:

"In a clinical study performed on 49 consecutive patients undergoing extracorporeal shock wave lithotripsy a 14.3% rate of bacteremia was detected during treatment. "

We get an echocardiogram, which shows vegetations on the aortic and pulmonic valves. That is odd. Pulmonic vegetations are quite rare, although there are a smattering of enterococcal infections involving the pulmonic valve.

He also has lung cancer, and given the multi-valvular involvement with a relatively odd organism, I wonder---mostly for hoots and giggles---if this is marantic endocarditis that is secondarily infected. Now usually called "nonbacterial thrombotic endocarditis," marantic endocarditis comes from the Greek "wasting away" and is a growth of sterile vegetations on the valve leaflets associated with wasting diseases such as cancer, systemic lupus erythematosus, and previous rheumatic fever. Marantic vegetations would be fertile soil in which to grow bacteria, but there would be no way to prove it. So probably not, but fun to think about, although the pulmonic valve is the least likely to be involved with marantic endocarditis.

Bacteremia is common, endocarditis not so much---so while it is nice to have reasons for the valves getting infected, no good prior valve pathology is evident in this case except for the changes of age.

He also "dies" from amoxicillin, so it is a course of vancomycin and gentamicin for now. Another unhappy choice of antibiotics in a patient already hard of hearing.

My second case of enterococcal endocarditis this month and the second case of pulmonic endocarditis this year.

Rationalization

Infection. 2004 Jun;32(3):170-5. Isolated pulmonic valve infective endocarditis: a persistent challenge.

http://www.ncbi.nlm.nih.gov/pubmed/15188078

An Med Interna. 2006 Oct;23(10):487-9. [Non-bacterial thrombotic endocarditis and lung neoplasm].

http://www.ncbi.nlm.nih.gov/pubmed/17134312

Postgrad Med J. 1996 Jan;72(843):51-2. Enterococcal endocarditis after extracorporeal shock wave lithotripsy for nephrolithiasis.

http://www.ncbi.nlm.nih.gov/pubmed/8746286

J Urol. 1991 Sep;146(3):733-6. Bacteremia during extracorporeal shock wave lithotripsy of renal calculi.

http://www.ncbi.nlm.nih.gov/pubmed/1875482

POLL RESULTS

Events occur

• in threes. 10%

• randomly and order is an illusion. 54%

• in order and randomness is an illusion. 15%

• in a Fibonacci Series. 15%

• in a preordained manner. 3%

• Other Answers 3%

  • Period. Get used to it.

Chapter Thirty-Eight

Probably not, but fun to think about

Four days off in a row. Wow. And another four days off over New Year's. So much free time, I do not know what to do with myself. But they must have saved up the infections for me over the holidays as it was quite busy today.

"What' is always good, "why" is even better. The fun part of the job is trying to discover why something occurs, even if it is a bit of a stretch. If nothing else, the search for why leads to a little learnin'.

Three weeks ago, the patient had nausea, vomiting, rigors, and abdominal pain. His father had the same symptoms, so they passed it off as food poisoning, and while the father improved, the patient only worsened.

He eventually came to the ER (after three weeks, no insurance) when pus started flowing out of his belly button. Weird, he thought, and it smelled most foul. No constipation or diarrhea for the whole time.

A CT that showed an enormous abscess, one of the biggest I have seen, that was called diverticular--although at 51 diverticulitis with perforation is not that common, and he did have a smattering of tics.

It was drained for 1500 cc pf pus, with a marked improvement of the pain, and, to my disappointment, not sent for culture. Sigh. But the odor from the navel is indeed most foul.

Why did it drain out the belly button? Path of least resistance, I suppose. I have wondered if it is an infected urachal cyst or the pus drained out a urachal sinus. An infected urachal cyst can be large. The only abscesses that drain through the belly button I can find are due to urachal cysts and can find no case reports of other abscesses finding egress by that route.

He has not yet gone to the OR, which may confirm the diagnosis, and he may not be operated on if he responds to drainage and antibiotics. Although it is my understanding if it is an infected cyst, it needs resecting.

So maybe a why, but probably not.

Note the year of the first rationalization. Worth a read. Makes me glad I live in modern times.

Postscript.

He did fine with conservative therapy.

Rationalization

Lond J Med. 1850 May; 2(17): 473--478. PMCID: PMC2544018 Case of Muco-Enteritis, with Abscess at Umbilicus. Relieved by Profuse Serous Discharge from a Spontaneous Opening of the Umbilicus by Ulceration, Followed by Prolonged Suppuration, Repeated Hæmorrhage, and Stercoraceous Vomiting

http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2544018/?page=1

Infected urachal cyst in an adult: a case report and review of the literature Kingsley C Ekwueme and Nigel J Parr Cases Journal 2009, 2:6422

http://www.casesjournal.com/content/2/1/6422

J Korean Surg Soc. 2012 Oct;83(4):254-7. doi: 10.4174/jkss.2012.83.4.254. Epub 2012 Sep 25. Urachal cyst presenting with huge abscess formation in adults.

http://www.ncbi.nlm.nih.gov/pubmed/23091800

Abscess of urachal remnants presenting with acute abdomen: a case series Journal of Medical Case Reports 2012, 6:226

http://www.jmedicalcasereports.com/content/6/1/226

Chapter Thirty-Nine

Leukocytosis

I occasionally get called to see patients with a leukocytosis (high white blood cell count). Why, they ask, is there a leukocytosis?

That's easy, I respond. It's because the white cells are up. For some reason, they don't find that answer satisfying, so I do the full consult.

The patient was admitted with an exacerbation of chronic obstructive pulmonary disease (COPD) and has improved, but his white cell count is chronically high--in the upper teens, lower twenties of thousands of cells per microliter, where the upper limit of normal is about 11,000. Nothing abnormal appears on the differential count.

He is not particularly forthcoming with symptoms, being primarily interested in being discharged from "this nut house." On the face of it that doesn't appear to be an unreasonable request or conclusion.

The review of systems is negative, the labs are negative except the WBC count, and he had the pan-scan during this admission which was unimpressive for pathology.

So no infection. What else causes leukocytosis? Tumor on occasion, especially bladder cancer, which can make granulocyte colony stimulating factor (G-CSF). No tumors are apparent in this case, though. Drugs? When in doubt, blame the medications. The prednisone would be nice to blame, but his WBC count is elevated when he is not on prednisone. The only other drug that I know causes elevated WBCs is lithium, and to make that diagnosis would result in nirvana. No lithium, so Nevermind.

So on to the Googles which leads to Medscape which leads to

Heparin: Heparin induces leukocytosis, mainly lymphocytosis, but in some cases, neutrophilia as well. One in every 230 patients treated with heparin had leukocytosis.

Didn't know that about heparin.

Medications that are known to cause leukocytosis and leukemoid reaction along with eosinophilia are antiepileptic drugs, including carbamazepine,[11] phenobarbitol, and phenytoin. [12] Minocycline, which is commonly used for the treatment of acne.[13] have been reported to cause a severe hypersensitivity reaction called drug rash with eosinophilia and systemic symptoms (DRESS) syndrome. The neutrophilia and leukocytosis are secondary to hypersensitivity reaction to the drug. Some patients developed lymphocytosis or leukopenia instead of neutrophilia and leukocytosis with antiepileptic drugs. The antipsychotic drug clozapine has been known to cause agranulocytosis, but it also causes dose-related elevation in leukocytes and neutrophil counts.[14]"

And wouldn't you know it, he is on chronic clozapine.

So that led to PubMed and

"Interestingly, patients who receive clozapine treatment may occasionally experience elevations in their total white blood cell count (WBC). In some of these patients, the leukocytosis may be persistent. We report the case of a patient with refractory schizophrenia who is treated with clozapine and who experienced chronic leukocytosis."

And about a half-dozen similar cases, much less common than agranulocytosis from the drug. Maybe. There was a series out of China that reported.

"The main side effects of 7921 hospitalized patients taken clozapine from July in 1980 to October in 1988 were investigated. In these cases, there were 600 patients with the main side effects caused by clozapine. They included 312 patients with leukocytosis (52.0%), 114 patients with leukopenia (19.0%), (included 16 patients with agranulocytosis)"

Much higher rates of leukocytosis than the official publications, but association is not causation. The increase in the WBC is benign, and with no other reason, I am blaming the drug.

Rationalization

http://emedicine.medscape.com/article/956278-clinical

Clin Schizophr Relat Psychoses. 2010 Jul;4(2):141-4. doi: 10.3371/CSRP.4.2.6. Chronic leukocytosis associated with clozapine treatment.

http://www.ncbi.nlm.nih.gov/pubmed/20643637

Zhonghua Shen Jing Jing Shen Ke Za Zhi. 1991 Apr;24(2):71-4, 123. [Clinical analysis in the main side effects of clozapine: enclosed 600 cases report].

[http://www.ncbi.nlm.nih.gov/pubmed/1860384]{.underline}

POLL RESULTS

At the end of the year, I wish everyone a

• Merry Christmas and Happy New Year 47%

• Happy Holiday 10%

• Happy Newtonmas 3%

• Happy Festivus 18%

• Generic non-denominational seasonal greeting 16%

• Other Answers 6%

  • Happy Trails to you, till we meet again

  • Felix sit annus novus

  • Bah-Humbug.

Chapter Forty

Won't get fooled again. I hope.

I remember my first consult in practice, twenty-three-plus years ago. The patient had Bactrim- induced neutropenia, developed fevers, progressive multiorgan system failure, and died. At post mortem the patient had disseminated miliary tuberculosis (TB).

There were features that made the diagnosis opaque at the time for a newbie: no risks for TB except a distant history of a positive PPD skin test, and no miliary pattern on the chest X-ray. While I am not pleased with missing the diagnosis, it was an unusual presentation of an unusual disease. Of course, years later, the case was presented as an unknown to a visiting professor, and he made the diagnosis. Me bitter much?

My second consult was not much better. An end-stage alcoholic with multiorgan system failure who at autopsy had cardiac aspergillosis. Even more odd.

My practice has become a wee bit easier over the years, but my failures stay with me far more than my successes. Being right is easy; being wrong is hard.

Today's patient is middle-aged, homeless, on Remicade (infliximab, an immunosuppressant) for the last five years for Crohn's Disease. He had been in the hospital for fever a few weeks back, and nothing was found. No focal infection after an extensive workup. We fretted about TB, but he was PPD skin test negative five years ago. We also worried about reactivation of endemic fungi since he had lived in both the Central Valley of California and the upper Midwest. After testing, we found that he had a normal white blood cell count and a normal procalcitonin (PCT) level. PCT is used as a tool to characterize systemic inflammation, infection, and sepsis, and should be elevated in both bacterial and fungal infections, so we then thought maybe there was a viral process.

Ten days later he is back in the ICU with fevers, multiorgan system failure, and new lung infiltrates--the right middle lobe more than the upper lobes. The PCT level is off the wall. The only finding on all his labs is 3 of 3 pulmonary specimens, including a bronchoscopy, are positive for acid-fast bacilli.

Crap. I think this is disseminated TB, and the patient is being treated accordingly. I wonder if this is actually progressive primary infection (rather than reactivation) since the patient was PPD negative, homeless, and it occurred after years on immunosuppressive therapy. Plus, a right middle lobe infiltrate would be the most likely area for primary TB to infect.

"Increased susceptibility to TB, often with extrapulmonary or disseminated disease, occurs following treatment with all anti-TNF-alpha biological agents and amounts to a four- to 20-fold increased risk with infliximab. TB usually occurs shortly after anti-TNF-alpha initiation suggesting reactivation of latent infection."

Besides the non-focality, we put a lot of weight on the low procalcitonin for lack of a serious infection, but

"...serum PCT in HIV-negative PTB [pulmonary tuberculosis] patients is basically low and is a useful biomarker for discriminating PTB and CAP [community-acquired pneumonia]; however, when serum PCT is above the normal cut-off point (0.5 ng·mL1), it is a poor prognostic marker in PTB patients."

And the suggestion is as the PCT goes up, so does mortality. There is only one reference in the literature to procalcitonin and disseminated TB that I cannot access in our library. Anyone get bored, send me the summary.

Rationalization

J Crohns Colitis. 2012 Oct;6(9):946-9. doi: 10.1016/j.crohns.2012.02.018. Epub 2012 Jun 28. Life-threatening disseminated tuberculosis as a complication of treatment by infliximab for Crohn's disease: report of two cases, including cerebral tuberculomas and miliary tuberculosis.

http://www.ncbi.nlm.nih.gov/pubmed/22749231

Aliment Pharmacol Ther. 2008 Jan 1;27(1):19-30. Epub 2007 Oct 16. Review article: minimizing tuberculosis during anti-tumour necrosis factor-alpha treatment of inflammatory bowel disease.

http://www.ncbi.nlm.nih.gov/pubmed/17944997

Int J Tuberc Lung Dis. 2011 Feb;15(2):251-6, i. Serum procalcitonin in pulmonary tuberculosis.

http://www.ncbi.nlm.nih.gov/pubmed/21219690

Eur Respir J. 2011 Feb;37(2):371-5. doi: 10.1183/09031936.00011910. Epub 2010 Jun 7. Usefulness of serum procalcitonin levels in pulmonary tuberculosis.

http://www.ncbi.nlm.nih.gov/pubmed/20530033

Disseminated tuberculosis following unrelated cord blood transplantation for refractory peripheral T-cell lymphoma: Clinical role of serum procalcitonin levels. Koya J, Nannya Y, Yoshizato T, Ono K, Seo S, Nakagawa M, Ichikawa M, Kurokawa M. J Infect. 2011 Mar;62(3):237-40. doi: 10.1016/j.jinf.2011.01.010. Epub 2011 Jan 31. No abstract available.

http://www.ncbi.nlm.nih.gov/pubmed/21281675

POLL RESULTS

The best Who album is

• Tommy 32%

• Quadrophenia. 15%

• Live at Leeds. 15%

• Who's Next 26%

• American Woman 9%

Chapter Forty-One

The wonderful cases for me are often the worst cases for the patient.

O wonderful, wonderful, and most wonderful wonderful! and yet again wonderful, and after that, out of all whooping!

As You Like It

Kind of sums up Infectious Diseases, doesn't it? At least I think so. Even when work is slow, and it has been slower and slower every year, I still get to see one curiosity after another.

The patient is an 86-year-old with a slow decline for five months. For the last week, he has been bedridden and is brought in by family.

A review of systems reveals little: slow, progressive malaise. No fevers, no chills, no sweats, no nothing.

Exam: a nice aortic stenosis murmur, long standing, and poor dentition, but little else of note.

Screening labs show little beyond a slightly elevated white blood cell count of 13,000 cells/microliter (upper limit is 11,000) and an erythrocyte sedimentation rate.

(ESR) > 100 mm/hr, highly elevated and indicative of infection. So blood cultures are done, and they tell all: all the sets grow Abiotrophia defectiva. It is almost certainly endocarditis.

Abiotrophia is an odd cause of endocarditis. Abiotrophia used to be called "nutritionally deficient Streptococcus," but it took affront at being called deficient and the microbiologists that be gave it a new name. I am not so certain that doing from deficient to defective was a step up.

Abiotrophia is a mouth bug, so that the poor dentition was the probable source.

Abiotrophia is also a hard bug to kill, mostly due to a funky cell wall, and carries a high risk of complications and mortality relative to other streptococci. I am going to try and kill it with penicillin and gentamicin, although it would not grow briskly enough to check susceptibilities. I wonder if it is the delay in diagnosis that allows it to cause so much damage.

In my fellowship, I had a patient with Abiotrophia on a bicuspid aortic valve who was doing clinically well and the valve blew out on the last day of therapy, requiring emergent valve repair. She was in the hospital at the time, as it predated home antibiotics. If the valve had blown at home, she would have died.

Years ago I had an IV drug abuser with aortic valve endocarditis due to Abiotrophia. Evidently, her mother had died of endocarditis, and as a consequence, she was exceedingly anxious about her infection. She completed her penicillin and gentamicin, with apparent success, and then was lost to follow-up, returning to the world of homelessness and heroin.

A year---yes, a year---later she is found down with a massive stroke, wide open aortic insufficiency, dilated congestive heart failure, and she codes and dies. All her blood cultures grow Abiotrophia again. Given the rarity of the infection, I have to suppose it was a failure of therapy and not a new infection, but an endocarditis festering for a year boggles the mind.

Awful results, maybe not so wonderful after all.

So I am a wee bit apprehensive with this one. So far, everything seems to be going fine, but Abiotrophia makes me nervous.

Rationalization

Infez Med. 2012 Jun;20(2):67-74.

Endocarditis caused by nutritionally variant streptococci: a case report and literature review.

http://www.ncbi.nlm.nih.gov/pubmed/22767303

Chapter Forty-Two

The World is a Litter Box

The patient is back from a trip to Asia. While there he developed some "bites" on his foot. He does not remember anything biting him, but it looked like a bite. Then over the next week, he developed two red lines on his foot, both of which were intensely pruritic and did not respond to the cephalexin he had in the medicine cabinet.

While in Asia he spent at most 15 minutes in freshwater, but lots on time on the Thailand beaches.

One exam there are two red, non-fluctuant areas on his foot that look like they had an aggressive scratching.

I figure it is one of two things. First is swimmer itch, Schistosomiasis. His time in fresh water was minimal, but it doesn't take long. Swimmers itch is often due to avian schistosomes and is in the waters throughout the world. However, Schistosomiasis tends to cause a more nodular reaction.

More likely it is cutaneous larva migrans, the creeping eruption, due to one hookworm or another. Most likely it is Ancylostoma braziliense, the cat tapeworm. The dog tapeworm, Ancylostoma caninum, can cause cutaneous larva migrans as well, including one outbreak in a temple where devotees rolled on the ground acquiring the parasite. And avoid eating live pond loaches, they will cause cutaneous larva migrans from Gnathostoma.

"Freshwater fish, specifically loaches, are potential hosts of Gnathostoma species. Consumption of uncooked freshwater loaches may result in cutaneous larva migrans. We report the case of a 70-year-old Taiwanese man who presented with a serpiginous skin rash on his abdomen. He had eaten live loaches intermittently (sometimes daily) for 5 months before his presentation."

The beaches of Asia are one big litter box, and the patient spent a lot of time barefoot in the sand. Since humans are not the normal host, the parasite wanders around in the skin, unable to burrow deeper, causing pain and itching until it finally dies. You can hasten its demise with a little albendazole or ivermectin.

Gives me the willies, the thought of a parasite wandering about, waiting to die. At least it is not going to emerge like the parasite in Alien.

Rationalization

Folia Parasitol (Praha). 2007 Jun;54(2):81-7. Schistosomes causing cercarial dermatitis: a mini-review of current trends in systematics and of host specificity and pathogenicity.

http://www.ncbi.nlm.nih.gov/pubmed/17894034

Lancet Infect Dis. 2008 May;8(5):302-9. doi: 10.1016/S1473-3099(08)70098-7. Epidemiological and clinical characteristics of hookworm-related cutaneous larva migrans.

http://www.ncbi.nlm.nih.gov/pubmed/18471775

Clin Infect Dis. 2000 May;30(5):811-4. Epub 2000 May 18. Treatment of cutaneous larva migrans.

http://www.ncbi.nlm.nih.gov/pubmed/10816151

PLoS One. 2012;7(1):e30516. doi: 10.1371/journal.pone.0030516. Epub 2012 Jan 25. Cutaneous larva migrans among devotees of the Nallur temple in Jaffna, Sri Lanka. http://www.plosone.org/article/info%3Adoi%2F10.1371%2Fjournal.pone.0030516

Clin Exp Dermatol. 2011 Dec;36(8):878-81. doi: 10.1111/j.1365-2230.2011.04116.x. Epub 2011 Jun 21. Cutaneous larva migrans induced by swallowing live pond loaches.

http://www.ncbi.nlm.nih.gov/pubmed/21689146

POLL RESULTS

The most horrifying scene in the movies was

• the alien emerging from the chest in Alien. 42%

• the amputation in Gone with the Wind. 10%

• the end of Buried where he is, well, buried. Alive. Arrrgghh. 4%

• the pecked out eyes in The Birds 21%

• the organ playing in the Ghost and Mr. Chicken. 13%

• Other Answers 10%

  • the beaches of D-Day in Saving Private Ryan

  • Something in Play Misty for Me that about gave me a stroke.

  • The kid gunsmith getting his arm shot off in The Jackal.

Chapter Forty-Three

Swing and a Miss

The patient has a whopping Candida infection of the mouth and esophagus, so extensive on endoscopy that the gastroenterologist likened it to the severe disease seen in HIV. The problem is, there is no reason that had been found for such severe Candida in this patient. And it hasn't been touched by the anti-fungal fluconazole.

No diabetes, no HIV, the CD4 cell count is normal, no steroids, no findings of lymphoma or leukemia.

When I see her, there is indeed very bad thrush. I hate saying "oral thrush" since thrush is by definition oral. I sent off cultures, wondering if it were either an atypical Candida or a resistant C. albicans. Neither was the case.

The heck. Why was she having such severe Candida? I was on the verge of a big shrug and giving up when I thought, What the hell.

I spent my fellowship years at Harbor-UCLA under the direction of John E. Edwards, one of the Candida mavins in the world, so in the distant past I had spent an excessive amount of my mental energy thinking about all things Candida.

Thymomas are one of the weird predisposing conditions that result in severe mucocutaneous Candida. I had never seen a case, mind you, but maybe this was a first. So one CT scan later and there is a thymic mass. My first thymoma.

So it was removed, and it wasn't a thymoma---just a really big thymus. But when I see her in follow up, the severe Candida is mostly gone. And her mother volunteers that while growing up, the patient also had issues with mucocutaneous Candida. So maybe not acquired, but genetic. There are several genetic mutations that lead to mucocutaneous Candida, but there is no history to suggest a genetic disease.

"Mutations in the CC domain of STAT1 underlie autosomal dominant CMC and lead to defective Th1 and Th17 responses, which may explain the increased susceptibility to fungal infection."

Many papers point to an issue in the IL-17 pathway with resultant cutaneous Candida, but the pathophysiology is beyond my pay grade. And there are 17 Interleukins?!? Since when? No, wait. I count 36 !?!?! Man, I am behind in my interleukins.

Acquired or genetic, I will never know. I do not have access to the testing. I look forward to the day when I can schmear my patient's genome on a slide and find all their mutations. Or a tricorder. I'd take a tricorder.

Rationalization

J Exp Med. 2010 Feb 15;207(2):299-308. doi: 10.1084/jem.20091669. Epub 2010 Feb 1. Chronic mucocutaneous candidiasis in APECED or thymoma patients correlates with autoimmunity to Th17-associated cytokines.

http://www.ncbi.nlm.nih.gov/pubmed/20123959

J Am Acad Dermatol. 1994 Sep;31(3 Pt 2):S14-7. Chronic mucocutaneous candidiasis.

http://www.ncbi.nlm.nih.gov/pubmed/8077500

N Engl J Med. 2011 Jul 7;365(1):54-61. doi: 10.1056/NEJMoa1100102. Epub 2011 Jun 29. STAT1 mutations in autosomal dominant chronic mucocutaneous candidiasis

http://www.ncbi.nlm.nih.gov/pubmed/21714643

Curr Opin Allergy Clin Immunol. 2012 Dec;12(6):616-22. doi: 10.1097/ACI.0b013e328358cc0b. Inborn errors of human IL-17 immunity underlie chronic mucocutaneous candidiasis.

http://www.ncbi.nlm.nih.gov/pubmed/23026768

Arthritis Res Ther. 2012 Jul 23;14(4):217. [Epub ahead of print]Mucocutaneous candidiasis: the IL-17 pathway and implications for targeted immunotherapy.

http://www.ncbi.nlm.nih.gov/pubmed/22838497

Chapter Forty-Four

Eagle Effect

The patient comes in with sepsis and necrotizing fasciitis/myositis. This is the third case in the last month, and the second to grow Streptococcus pyogenes. Initially, he is on "too numerous to count" antibiotics (cefepime, linezolid, and clindamycin) and intravenous immunoglobulin (IVIG), but when the cultures come back, he is changed to penicillin alone. The bug is resistant to clindamycin, and on penicillin the infection clinically advances, so they call me.

Perhaps you have heard of the Eagle effect. It is where one of the blandest bands of all time has managed to sell 42 million albums.

Wait. Wrong Eagle effect.

"In 1952, Harry Eagle demonstrated the failure of penicillin treatment to cure S. pyogenes(STREPA) myositis in a mouse model. Subsequent animal model studies have confirmed this Eagle effect and have shown that clindamycin is a more effective treatment...The Eagle effect is relevant in children with invasive STREPA infections and clindamycin in combination with a beta-lactam (with surgery if indicated) appears to be the most effective treatment."

So what to do when the bug is resistant to clindamycin? Besides surgery?

I went searching and found no solution. Tetracyclines? No data. But it does have the advantage of interfering with protein synthesis. Susceptibilities are pending and tetracycline resistance can follow macrolide resistance:

"A high frequency of macrolide-tetracycline coresistance in S. pyogenes is found in many countries including Denmark; hence tetracycline use must be considered as a co-factor in selection of MRSP."

Vancomycin? Always the wrong answer. Linezolid? Tempting, as it also messes with protein synthesis, but I have even less experience with it.

I opted for adding doxycycline. I am never a fan of making decisions when there is no good (or any) data, but that is part of my job.

And as my grandmother loved to point out, a job is what a dog does on the carpet.

Rationalization

1998 Abstracts The American Pediatric Society and The Society for Pediatric Research Pediatric Research (1998) 43, 161161; doi:10.1203/00006450-199804001-00957

http://www.nature.com/pr/journal/v43/n4s/full/pr19981084a.html

Published July 1, 1948 // JEM vol. 88 no. 1 99-131 THE RATE OF BACTERICIDAL ACTION OF PENICILLIN IN VITRO AS A FUNCTION OF ITS CONCENTRATION, AND ITS PARADOXICALLY REDUCED ACTIVITY AT HIGH CONCENTRATIONS AGAINST CERTAIN ORGANISMS.

http://jem.rupress.org/content/88/1/99

Microb Drug Resist. 2004 Fall;10(3):231-8.Tetracycline and macrolide co-resistance in Streptococcus pyogenes: co-selection as a reason for increase in macrolide-resistant S. pyogenes?

http://www.ncbi.nlm.nih.gov/pubmed/15383167

Pharmacotherapy. 2003 May;23(5):638-42. Targeting bacterial virulence: the role of protein synthesis inhibitors in severe infections. Insights from the Society of Infectious Diseases Pharmacists

http://www.ncbi.nlm.nih.gov/pubmed/12741438

POLL RESULTS

The best-selling album that most amazes me is ([http://en.wikipedia.org/wiki/List_of_best-selling_albums]{.underline})

• Eagles. Their Greatest Hits 9%

• Shania Twain. Come on Over 16%

• Bruce Springsteen. Born in the U.S.A. Not Born to Run. Born in the USA. What is the matter with you people? 19%

• James Horner. Titanic: Music from the Motion Picture 34%

• Linkin Park. Hybrid Theory 22%

Chapter Forty-Five

Unlike NW Forests, ID is Never Clear Cut

Sometimes I feel like I am are forcing a round peg into a square hole. All the information suggests a diagnosis and treatment plan, but there are enough outliers and uncertainty that I am not completely satisfied with the plan, but safe is often better than sorry.

Four weeks ago, the patient was admitted with fevers, rigors and severe weakness. He had diabetes and was elderly and was given the diagnosis of community-acquired pneumonia mostly on the basis of the chest X-ray, not signs or symptoms. I looked at said X-ray. Meh. Yeah, it was abnormal, some slight infiltrate/atelectasis, but not the nice consolidation I like to call a pneumonia. Still, chest X-ray often underrepresents the extent of pulmonary disease when compared to a CT scan, but with the retrospectoscope I did not think the X-ray accounted for the blood cultures: 1/2 for Group G streptococcus. Not a common cause of community-acquired pneumonia.

"Group G streptococci have been recognized as a cause of endocarditis, septic arthritis, puerperal sepsis, and cellulitis. The organism has not previously been implicated as a pneumonic pathogen in adults."

He was sent home on pneumonia therapy.

Fast forward 4 weeks. His symptoms recur, he calls 911, and he is back in the hospital. This time both sets of blood cultures are positive for Group G streptococcus and his leg is red.

Cellulitis with the same bug a month later? Seems hinky, and the patient swears the front of his leg did not turn red until after he had an ultrasound for deep vein thrombosis. I'm skeptical, as Group G streptococcus is a common cause of bacteremic cellulitis.

" Group G Streptococcus (Streptococcus dysgalactiae subsp. equisimilis) was found most often and was isolated from 22% of patient samples of either skin lesions or blood, followed by group A Streptococcus, which was found in 7% of patients... The recurrent nature of cellulitis became evident during this study."

Allergic reactions to ultrasound gel are uncommon but reported.

But the leg does have that contact dermatitis look, as if I can tell. An echocardiogram shows old person's valves but no vegetations.

I just cannot ignore a sustained bacteremia, and I fret given the recent Medicine article on gram positive bacteremia, although he doesn't fit the demographic:

"We performed a prospective study in 115 patients with Staphylococcus aureus or Streptococcus species bacteremia with at least 1 risk factor for the development of metastatic foci, such as community acquisition, treatment delay, persistently positive blood cultures for > 48 hours, and persistent fever > 72 hours after initiation of treatment. An intensive search for metastatic infectious foci was performed including ¹⁸F-fluorodeoxyglucose-positron emission tomography in combination with low-dose computed tomography scanning for optimizing anatomical correlation (FDG-PET/CT) and echocardiography in the first 2 weeks of admission. Metastatic infectious foci were detected in 84 of 115 (73%) patients. Endocarditis (22 cases), endovascular infections (19 cases), pulmonary abscesses (16 cases), and spondylodiscitis (11 cases) were diagnosed most frequently. The incidence of metastatic infection was similar in patients with Streptococcus species and patients with S. aureus bacteremia. Signs and symptoms guiding the attending physician in the diagnostic workup were present in only a minority of cases (41%)."

It is that last sentence that gives me pause, although I am not inclined to get a PET scan, I am going to treat him for endocarditis. Group G is equally rare as a cause of endocarditis (I remember maybe 2 in 25 years) but the most potentially catastrophic if untreated. It is not clear-cut. As always. I hate it when Occam fails me. That, I suppose, is why I get the big denial of payment.

Rationalization

Clin Infect Dis. 1998 Aug;27(2):358-63. High-resolution computed tomography for the diagnosis of community-acquired pneumonia. <Http: data-preserve-html-node="true"//www.ncbi.nlm.nih.gov/pubmed/9709887>

South Med J. 1982 Nov;75(11):1427.bacteremic group G streptococcal pneumonia. <http: data-preserve-html-node="true"//www.ncbi.nlm.nih.gov/pubmed/7146979>

Clin Infect Dis. 2008 Mar 15;46(6):855-61. doi: 10.1086/527388. Acute bacterial, nonnecrotizing cellulitis in Finland: microbiological findings. <http: data-preserve-html-node="true"//www.ncbi.nlm.nih.gov/pubmed/18260753>

Reg Anesth Pain Med. 2006 Sep-Oct;31(5):480-1. Allergic contact dermatitis caused by ultrasonic gel. <http: data-preserve-html-node="true"//www.ncbi.nlm.nih.gov/pubmed/16952823>

Medicine (Baltimore). 2012 Mar;91(2):86-94. doi: 10.1097/MD.0b013e31824d7ed2. Metastatic infectious disease and clinical outcome in Staphylococcus aureus and Streptococcus species bacteremia. <http: data-preserve-html-node="true"//www.ncbi.nlm.nih.gov/pubmed?term=22391470>

J Infect. 1988 Mar;16(2):169-76. Group G streptococcal endocarditis: two case reports, a review of the literature and recommendations for treatment. <http: data-preserve-html-node="true"//www.ncbi.nlm.nih.gov/pubmed/3280691>

Chapter Forty-Six

Epiphenomena

All diseases are not necessarily infectious in etiology. Which is a shame. Years ago, Dr. David Gilbert suggested at a conference that all diseases were infectious, genetic, or due to wear and tear, and for the most part that is right. Non-infectious diseases are usually not as interesting as those caused by bugs, but are sometimes a curiosity.

I am called to see a patient, past medical history with diet-controlled diabetes and quiescent ulcerative colitis, with a hand/arm cellulitis. He has the sudden onset of fevers, spreading erythema of the arm and severe pain when he makes a fist, suggesting a tenosynovitis. On exam, his hand and forearm have rubor, dolor, calor, tumor. There is mild leukocytosis. Cellulitis, right?

But.

There is no good reason for the infection. Usually, there is at least some trauma or exposure to set off an infection this severe. Only trauma is finger sticks to check his sugars.

And.

He has a patch of rubor, dolor, calor, and tumor over his heel. Bacteremic spread? Blood cultures are negative at 24 hours.

And.

Six months ago he had a similar episode at another hospital involving his foot that was debrided but never grew a pathogen, and the pathology was not diagnostic.

We start antibiotics and prednisone, and he is off the OR where inflammation is found in the soft tissue, tendons, and wrist joints, but gram stain and cultures remain negative.

So is this infectious? Maybe, but I bet this is an odd epiphenomenon of his ulcerative colitis, mimicking infections. I think it is a variant of this case:

"This patient demonstrates many of the manifestations of Sweet's syndrome with a recurrent, self‐limiting, painful, blistering rash occurring in a middle‐aged female in association with a fever, myalgia and a migratory, asymmetrical, polyarticular, large joint arthritis. The histology of affected skin, a peripheral neutrophilic leucocytosis and an elevated ESR are also typical. However, a significant clinical feature of her episodes has been tenosynovitis at multiple sites. This has not previously been included in descriptions of classical Sweet's syndrome, although tenosynovitis is recognized in the bowel bypass syndrome. Histological findings in this condition are the same as in Sweet's syndrome. Bowel‐associated dermatosis arthritis syndrome has been used to describe the Sweet's syndrome which is found in bowel bypass syndrome and other bowel conditions, including ulcerative colitis, Crohn's disease and following partial gastrectomy, including Bilroth II gastrectomy for peptic ulceration."

Patients with UC can get a wide variety of odd skin manifestations, and a subsequent endoscopy revealed more disease than was suspected.

As is so often the case, the patient had been on antibiotics more than 24 hours before OR specimens, so excluding infection is impossible. The inflammation is involving the dominant hand so I am going to treat for infection, although I think the clinical improvement is due more to the steroids than the antibiotics.

Rationalization

Rheumatology (Oxford). 2002 Sep;41(9):1067-9. Recurrent tenosynovitis in Sweet's syndrome.

http://www.ncbi.nlm.nih.gov/pubmed/12209043

Ann Intern Med. 2005 Jan 4;142(1):47-55. Narrative review: diseases that masquerade as infectious cellulitis.

http://www.ncbi.nlm.nih.gov/pubmed/15630108

Aliment Pharmacol Ther. 2004 Oct;20 Suppl 4:50-3. Review article: skin complications associated with inflammatory bowel disease.

http://www.ncbi.nlm.nih.gov/pubmed/15352894

POLL RESULTS

All disease

• is due to the wrong diet. We need to photosynthesize. 21%

• is due to infection. 13%

• is due to genetics. Blame my parents. 23%

• is due to unballanced energy. 19%

• is an illusion. 9%

• Other Answers 15%

  • is infectious, genetic or due to wear and tear

  • is evolutionary pressure to improve the species by eliminating you, or that other guy in the office that needs to shape up or ship out.

Chapter Forty-Seven

Never Dull

My main, and rather selfish, goal in life is to avoid being bored. I get bored very easily and like to keep my brain entertained with novelty pretty much 24-7. Infectious diseases is a good job for the ADHD prone like me as it is different every day. Even diseases that do not appear to be that different at first glance have variations that at least keep me entertained.

Take flu. Please. Flu is going gangbusters at the moment. Going to Google "Flu trends," it looks like flu activity is much greater than the height of the H1N1 epidemic of 2009. We have lots of flu-like illness, and per the CDC there are buckets of confirmed flu---but so far the season, while probably having more cases than 2009, is clinically not the same.

The H1N1 epidemic was due to, well, H1N1. Only about 1% of the isolates are currently H1N1, so presumptively most everyone is immune from infection or vaccination. It is H3N2 and B that are causing all the disease.

Despite the increase in cases, it is remarkable for what we are not seeing. While people are being admitted to the hospital with flu, unlike during the H1N1 epidemic, the ICU is not filled with influenza patients on ventilators.

That was the most remarkable aspect of the 2009 outbreak: we were maxed out in every ICU in my system. All beds and ventilators were in use. If a patient came through the door needing a ventilator, we did not have one. And right as we reached our surge capacity, the epidemic peaked. Dodged that bullet.

So far we have no patients on ventilators, no young people dying (meaning 20-30-year-olds; there have been 20 pediatric deaths), no flu encephalitis (we had two deaths from central nervous system involvement in young people with H1N1), no one on extracorporeal membrane oxygenation (ECMO), no pregnant females with advanced influenza.

It is curious how the strains differ in their effects on populations. Maybe the H1N1 has modified the disease for H3N2, since infection and vaccination to H1N1 in some people leads to a more "universal" antibody against flu. Or maybe there are other factors to the H3N2 virus that result in different clinical manifestations.

One would expect increased mortality this year:

"The magnitude of the seasonal component was highly correlated with traditional measures of excess mortality and was significantly larger in seasons dominated by influenza A(H2N2) and A(H3N2) viruses than in seasons dominated by A(H1N1) or B viruses.

But not from infection but from an increase in cardiac events:

"These data suggest that influenza infections, particularly by A/H3N2 viruses are directly associated with acute IHD[ischemic heart disease]-related events in older individuals."

And vaccination can prevent some of those events.

Of course, I speak from the perspective of a hospital-based doctor in Portland, Oregon, and as such, have a narrow and not representative experience. It will be interesting for someone with access to all the data to compare and contrast the two flu seasons in the years to come. If your experience is different, comment away.

Rationalization

Am J Epidemiol. 2004 Sep 1;160(5):492-502. Influenza and the winter increase in mortality in the United States, 1959-1999

http://www.ncbi.nlm.nih.gov/pubmed?term=15321847

J Infect Dis. 2012 Sep 15;206(6):821-7. doi: 10.1093/infdis/jis435. Epub 2012 Jul 12. The relationship between influenza outbreaks and acute ischemic heart disease in Maryland residents over a 7-year period.

http://www.ncbi.nlm.nih.gov/pubmed?term=22798684

Future Cardiol. 2012 May;8(3):345-8. doi: 10.2217/fca.12.1. Examining the potential of the influenza vaccine for secondary prevention: a myocardial infarction vaccine? . http://www.futuremedicine.com/doi/full/10.2217/fca.12.1

POLL RESULTS

My primary purpose in life is

• to avoid boredom. 28%

• to lose that extra weight. 13%

• to do unto others before they do unto me. 13%

• avoid taxes 3%

• avoid lawsuits. 23%

• Other Answers 21%

  • Have a helluva lot of fun while I can.

  • Embarrass my grandchildren

  • to avoid being hit by crazy drivers in south Florida.

  • to keep my spouse happy.

  • to exit on my own terms, or at least terms that are favorable, mostly.

  • avoid the void

Chapter Forty-Eight

Causing Trouble

My turn to take weekend call. The one drag in an otherwise great job. But there are upsides. When you work in hospitals, by and large, you work with great people. So it is usually a pleasure to go to work. There are a few flies in the ointment, some unpleasant people, but they can generally be avoided. And the pathophysiology is wonderful, the always interesting variety of ID. The subjects discussed in this book are cases from three hospitals. Weekends I cover seven hospitals, so there is sure to be some great cross-coverage. I need to be a little careful as I am just the cover guy--so while I get to go in and kibitz about the case, on Monday the responsibility will revert to the primary ID doc, and I don't want to make their day too difficult.

Two cases of cryptococcal meningitis, both with possible variations on a theme. The first is an AIDS patient. No surprise there: Cryptococcus is a known manifestation of AIDS. Ah, but which one? C. neoformans? Could be. C. gattii? Maybe. Vancouver Island was colonized with C. gattii just this century, and in a spectacular failure of homeland security, the yeast has been moving south into the Great Pacific Northwest. But the patient has been in the US for six years, having been born and raised in Thailand.

There is a third (actually there are over 37, more than half a Heinz) variety of Cryptococcus in SE Asia (and other parts of the world): Cryptococcus neoformans variety grubii, which is more virulent than Cryptococcus neoformans var neoformans. I half wonder if the patient brought the fungus with her to the US where it reactivated. There is some data to suggest that it can cause disease years after being acquired:

"A total of 24 (60%) isolates could be linked to C. gattii--endemic regions in Brazil, the United States, Africa, and the Vancouver Island outbreak region. These observations demonstrate that C. gattii infections can be imported subclinically and can cause infections after being dormant for many years."

I have asked the State to see if they can tell if the beast is local or imported, not that it will make any clinical difference, but there is nothing quite as satisfying as the joy of finding things out.

The other case is a non-HIV case of cryptococcal meningitis. What is curious is the cerebrospinal fluid (CSF) is not terribly abnormal: 300 white blood cells per cubic millimeter, when normal has <5, data-preserve-html-node="true" a glucose of 20 mg/dL when normal should be somewhat higher, protein of 210 mg/dL when normal is <45. data-preserve-html-node="true" The gram stain for Cryptococcus is negative, and a CSF cryptococcal antigen is only 1:1. Not much meningeal involvement, yet she has a 6th nerve paralysis and papilledema on exam. There is more neurologic involvement than one would expect for the CSF parameters. But the MRI suggests cyptococcomas, which suggests it will be C. gattii.

"The slow response to therapy is largely because C. gattii infection causes a disproportionate number of cryptococcomas in the brain (up to 30% of cases) and/or lung, compared with C. neoformans"

But what to do about the physical findings and increased intracranial pressure? Gamma interferon is suggested by the Guidelines in refractory cases, and I wonder if it would be of benefit in those with auto-antibodies against gamma interferon. Will have to suggest that diagnosis to the usual ID doc.

Steroids? Tempting but little data.

"Clinical outcomes were reviewed for 4 patients with Cryptococcus gattii central nervous system infection who received dexamethasone for the treatment of persisting mental status abnormalities and focal lesions on brain scan despite culture-negative cerebrospinal fluid and the management of intracranial pressure. Favorable clinical responses were observed in 3 patients. Although corticosteroids are not recommended for the treatment of cryptococcal meningitis, these observations suggest that dexamethasone should be further evaluated in this subset of patients."

For good or ill, I am the weekend guy so that I can stir the pot, but it will be up to my poor colleagues to taste the stew. I just have to remember to get the follow-up.

Rationalization

PLoS Pathog. 2011 Apr;7(4):e1001343. doi: 10.1371/journal.ppat.1001343. Epub 2011 Apr 28. Low diversity Cryptococcus neoformans variety grubii multilocus sequence types from Thailand are consistent with an ancestral African origin.

http://www.ncbi.nlm.nih.gov/pubmed/21573144

J Clin Microbiol. 2011 Feb;49(2):658-64. doi: 10.1128/JCM.01985-10. Epub 2010 Dec 15. Most cases of cryptococcal meningitis in HIV-uninfected patients in Vietnam are due to a distinct amplified fragment length polymorphism-defined cluster of Cryptococcus neoformans var. grubii VN1.

http://www.ncbi.nlm.nih.gov/pubmed/21159929

Autochthonous and dormant Cryptococcus gattii infections in Europe Emerg Infect Dis [Internet]. 2012 Oct [date cited].

http://dx.doi.org/10.3201/eid1810.120068

Clin Infect Dis. 2010 Feb 1;50(3):291-322. doi: 10.1086/649858. Clinical practice guidelines for the management of cryptococcal disease: 2010 update by the infectious diseases society of america.

http://www.ncbi.nlm.nih.gov/pubmed/20047480

N Engl J Med. 2012 Aug 23;367(8):725-34. doi: 10.1056/NEJMoa1111160. Adult-onset immunodeficiency in Thailand and Taiwan.

http://www.ncbi.nlm.nih.gov/pubmed/22913682

Trans R Soc Trop Med Hyg. 1997 Jan-Feb;91(1):50-2. The effect of corticosteroids on visual loss in Cryptococcus neoformans var. gattii meningitis.

http://www.ncbi.nlm.nih.gov/pubmed/9093628

AIDS. 2012 Jun 1;26(9):1105-13. doi: 10.1097/QAD.0b013e3283536a93. Adjunctive interferon-γ immunotherapy for the treatment of HIV-associated cryptococcal meningitis: a randomized controlled trial.

http://www.ncbi.nlm.nih.gov/pubmed/22421244

Clin Infect Dis. 2009 Aug 15;49(4):591-5. doi: 10.1086/603554. Dexamethasone in Cryptococcus gattii central nervous system infection.

http://www.ncbi.nlm.nih.gov/pubmed/19589083

Chapter Forty-Nine

The Problem with Being a Klutz

It is with good reason I am in a field that has no procedures. If you look up the word "klutz" in the Oxford English Dictionary, you will see my picture. An example is the time I spent all morning putting together a bookshelf, hammering and gluing, and when I put it upright, all the shelves were in upside down, the unfinished sides up. Sigh. Thank goodness for contact paper. I have no hand-eye coordination at all and always seem to misread the directions. Just imagine what would happen if I were a surgeon.

To my mind, no good could ever come out of my doing a procedure. Well, with one exception. As a second-year resident, I was post-call, unsuccessfully trying to put a line in a demented old lady. I called for a nurse to come and give me a hand. Just as she walked in the door, I hit the vein, threaded the CVP, and was done. I turned to the nurse, who I had never seen before, and said, "Thanks. You brought me good luck. Would you marry me?" and went home to sleep. We will have been together 30 years this summer. While I define klutz, she defines patience.

But I do not do even minor procedures, putting me at the mercy of others if I want a biopsy-proven diagnosis.

One patient I saw this weekend has fevers, multiple organ-system failure, and culture-negative sepsis after an ascending aortic aneurysm (AAA) repair. The systemic inflammatory response syndrome (SIRS) is not uncommon after AAA repair of all kinds:

"The majority of patients developed SIRS postoperatively. Elective patients with a cumulative SIRS score of > or =10 during postoperative days 1-4 were more likely to die, compared to patients with a SIRS score of <10 data-preserve-html-node="true" (p=0.02). The development of SIRS late in the postoperative period (day 5-10) was associated with adverse outcome (death) in elective patients (p=0.01). The actual number of SIRS criteria present did not significantly correlate with either outcome or the incidence of organ failure."

But on the thigh is a patch of petechiae, and I wonder if these are cholesterol emboli, which would explain the clinical syndrome nicely.

Today I saw a patient with culture-negative sepsis who has multiple petechiae in his feet. He has a slowly growing abdominal aneurysm that had a vascular stent repair two months ago. I suspect embolism, cholesterol or otherwise, as the cause.

"Atheroembolization occurred in eight patients (12 iliac artery stents and 1 aortic stent) at intervals ranging from 9 to 43 months (mean 22 months) following aortoiliac stent placement."

In both cases, I am trying to browbeat someone into doing a biopsy for me. Yeah, I know, what a wimp. Won't even do a skin biopsy. But as Mr. Callahan said, "A man's got to know his limitations."

Never got the biopsies. Life is filled with disappointments.

Rationalization

Eur J Vasc Endovasc Surg. 2004 Mar;27(3):292-8. The clinical value of the systemic inflammatory response syndrome (SIRS) in abdominal aortic aneurysm repair.

http://www.ncbi.nlm.nih.gov/pubmed/12618699

J Vasc Surg. 2003 Mar;37(3):600-6. The systemic inflammatory response syndrome, organ failure, and mortality after abdominal aortic aneurysm repair.

http://www.ncbi.nlm.nih.gov/pubmed/12618699

Cutaneous Manifestations of Cholesterol Embolism

http://emedicine.medscape.com/article/1096593-overview

J Surg Res. 2002 Apr;103(2):153-9. Late complication of aortoiliac stent placement- atheroembolization of the lower extremities.

http://www.ncbi.nlm.nih.gov/pubmed?term=11922729

POLL RESULTS

The first words to my eventual spouse were

• what's your sign? 3%

• come here often? 11%

• you must be joking. 23%

• No. 11%

• Yes. 18%

• Other Answers 34%

  • thanks

  • Will you marry me

  • hi

  • Do you like karaoke?

  • Cheers!

  • Thanks. You brought me good luck. Would you marry me?

  • Hi

  • I'm inviting you to church

  • leave me alone

Chapter Fifty

Asymptomatic

I am called to see a recurrent urinary tract infection (UTI). But there is more to it than that. The patient has diabetes, had a kidney transplant three months ago, and has had three episodes of the same Pseudomonas in her urine.

But no symptoms. No fevers, no chills, no increase in frequency or urination. For two of the episodes, she received first a 10-day then a 14-day course of IV antibiotics (cipro resistant) and now she is admitted with flu-like symptoms: fevers, cough, and myalgias, although not severe. If we had influenza PCR assay kits, I might be able to say whether she had flu or not, but we have done run out. She had the vaccine in October, and there was a recent upswing in coughing at her facility, so it may be an attenuated influenza or one the many illnesses that cause flu-like symptoms. I sure wish we had cheap, easy diagnostics for respiratory infections. But no urinary symptoms--just bacteria, and 20-50 white blood cells per high-power field. No signs of any mechanical causes of the UTI on an unenhanced CT scan.

What to do?

It is increasingly clear that in some populations treating symptomatic bacteriuria is counterproductive: it increases the risk for subsequent symptomatic UTIs:

"At baseline, the 2 most commonly isolated pathogens were Escherichia coli (group A, 38.4%; group B, 39.3%) and Enterococcus faecalis (group A, 32.7%; group B, 33.2%). At the first follow-up visit, there was no difference between the 2 groups (relative risk [RR], 1.05; 95% confidence interval [CI], 1.01-1.10), whereas after 6 months, 23 (7.6%) in group A and 98 (29.7%) in group B showed recurrence with a statistically significant difference (RR, 1.31; 95% CI, 1.21-1.42; P < .0001). At the last follow-up, 41 (13.1%) in group A and 169 (46.8%) in group B showed recurrence (RR, 3.17; 95% CI, 2.55-3.90; P < .0001). One patient in group A and 2 patients in group B were found to have pyelonephritis.

CONCLUSIONS:

This study shows that AB should not be treated in young women affected by UTI, suggesting it may play a protective role in preventing symptomatic recurrence."

Why is interesting. Are the bacteria protective? Filling adherence sites or modulating local immunity? Or is it more akin to C. difficile, where the antibiotics are messing with endogenous bladder flora allowing pathogen overgrowth? Because yes, Virginia, the bladder has bacteria, we just can't grow them:

"Clinical urine specimens are usually considered to be sterile when they do not yield uropathogens using standard clinical cultivation procedures. Our aim was to test if the adult female bladder might contain bacteria that are not identified by these routine procedures. An additional aim was to identify and recommend the appropriate urine collection method for the study of bacterial communities in the female bladder. Consenting participants who were free of known urinary tract infection provided urine samples by voided, transurethral, and/or suprapubic collection methods. The presence of bacteria in these samples was assessed by bacterial culture, light microscopy, and 16S rRNA gene sequencing. Bacteria that are not or cannot be routinely cultivated (hereinafter called uncultivated bacteria) were common in voided urine, urine collected by transurethral catheter (TUC), and urine collected by suprapubic aspirate (SPA), regardless of whether the subjects had urinary symptoms. Voided urine samples contained mixtures of urinary and genital tract bacteria. Communities identified in parallel urine samples collected by TUC and SPA were similar. Uncultivated bacteria are clearly present in the bladders of some women. It remains unclear if these bacteria are viable and/or if their presence is relevant to idiopathic urinary tract conditions."

It will be interesting to see what role, if any, this bladder microbiome plays in a variety of bladder diseases.

Of course, bacteriuria in a diabetic renal transplant is a whole different ball of wax. Odd phrase.

The whole ball of wax means the whole thing, everything. The whole ball of wax is an American idiom of uncertain origin, so far it has been traced back to at least the 1880s. Many apocryphal stories have sprung up to explain the origin of the phrase the whole ball of wax, but it is most likely a mondegreen of the idiom the whole bailiwick, meaning the whole territory. According to Google's Ngram, the popularity of the term the whole ball of wax has risen quickly since the 1960s, which coincides with a science-fiction book published at that time, The Big Ball of Wax, written by Shepherd Mead.

Perhaps.

"We retrospectively analysed the outcome of 334 asymptomatic Escherichia coli (E. coli) and Enterococcus faecalis (E. faecalis) bacteriuria that occurred in 77 RTR later than 1 month post-transplantation. We distinguished: Type I, high-grade bacteriuria with pyuria; Type II, high-grade bacteriuria without pyuria; Type III, low-grade bacteriuria with pyuria and Type IV, low-grade bacteriuria without pyuria.

RESULTS:

None of the 334 episodes was followed by acute rejection or chronic pyelonephritis. One hundred and one (30%) episodes were treated [32 (62%) Type I, 38 (45%) Type II, 13 (36%) Type III and 18 (11%) Type IV]. Evolution to symptomatic urinary tract infection (UTI) was similar between treated and untreated episodes (0/101 versus 4/233, P = 0.32). The four UTI resolved favourably without further complication upon treatment. Persistent asymptomatic bacteriuria occurred in 45 (46%) treated episodes (2 Type I, 27 Type II, 8 Type III and 9 Type IV), with selection of resistant pathogen in 35 cases (78%). Spontaneous bacterial clearance occurred in 138 (59%) untreated episodes (15 Type I, 23 Type II, 9 Type III and 91 Type IV). Negative control cultures tended to be more frequent in treated Type I (P = 0.09) and in untreated Type II episodes (P = 0.08).

CONCLUSION:

Restricting antibiotic treatments for asymptomatic low-grade bacteriuria and high-grade bacteriuria in the absence of pyuria, occurring later than 1 month posttransplantation, might be safe in RTR."

Not quite the same as my patient, and other studies have suggested asymptomatic bacteriuria increases the risk of pyelonephritis in transplant patients.

"However, pyelonephritis incidence was more than seven times higher in the group of patients who presented asymptomatic bacteriuria."

Insanity: doing the same thing over and over again and expecting different results. With no structural issues and no response to two courses of IV antibiotics, I am inclined to watch and wait but it makes me nervous. Very nervous.

We did nothing. The patient did fine.

Rationalization

Clin Infect Dis. 2012 Sep;55(6):771-7. doi: 10.1093/cid/cis534. Epub 2012 Jun 7. The role of asymptomatic bacteriuria in young women with recurrent urinary tract infections: to treat or not to treat?

http://www.ncbi.nlm.nih.gov/pubmed/22677710

J Clin Microbiol. 2012 Apr;50(4):1376-83. doi: 10.1128/JCM.05852-11. Epub 2012 Jan 25. Evidence of uncultivated bacteria in the adult female bladder.

http://www.ncbi.nlm.nih.gov/pubmed/22278835

Kidney Int. 2010 Oct;78(8):774-81. doi: 10.1038/ki.2010.286. Epub 2010 Aug 18. Systematic screening and treatment of asymptomatic bacteriuria in renal transplant recipients.

http://www.ncbi.nlm.nih.gov/pubmed/20720526

Nephrol Dial Transplant. 2011 Dec;26(12):4109-14. doi: 10.1093/ndt/gfr198. Epub 2011 May 17. Outcome of treated and untreated asymptomatic bacteriuria in renal transplant recipients.

http://www.ncbi.nlm.nih.gov/pubmed/21592976

Chapter Fifty-One

Primum non nocere. Yeah, right.

Often in medicine, there is not a good option but the least bad option---the lesser of two evils.

The patient is in his mid-eighties and has some hematuria. A CT scan shows nothing to cause the hematuria but does show an almost perfectly round mass in the right lower lobe. He has had some mild right flank pain for several months, but no other signs or symptoms. Looks like cancer. So it is biopsied, and no cancer is found. Otherwise, the patient is healthy except for "mild" chronic lymphocytic leukemia (CLL).

When the pulmonologist called me about the case for some advice, before he got to the denouement I said, "They found Cryptococcus on the biopsy."

It flabbered his gaster, but been there, done that. The only infection in the great Pacific Northwest that will be mistaken for a cancer and be otherwise asymptomatic is Cryptococcus, and it is fun play Sherlock Holmes getting the answer by putting together all the clues.

I said check a serum cryptococcal antigen, start high dose fluconazole and send him my way.

When I see the patient all the chest pain is gone, he has no other symptoms, the CLL is indeed mild to judge from the blood count, and he is just getting over a whopping rash from the fluconazole, and the Cryptpococcal antigen in the blood is negative.

Now what? As the guidelines say

"Immunocompetent patients who are asymptomatic and who have a culture of the lung that is positive for C. neoformans may be observed carefully or treated with fluconazole, 200--400 mg/d for 3--6 months... Immunocompetent patients who present with mild-to-moderate symptoms should be treated with fluconazole, 200--400 mg/d for 6--12 months. The toxicity of amphotericin B limits its utility as a desired agent in the treatment of mild-to-moderate pulmonary disease among immunocompetent hosts. However, if oral azole therapy cannot be given, or the pulmonary disease is severe or progressive, amphotericin B is recommended, 0.4--0.7 mg/kg/d for a total dose of 1000--2000 mg."

So he is immunoincompetent: elderly and CLL. But a course of amphotericin B means toxicity and expense, neither of which the patient can afford for what was a serendipitous finding. The literature really doesn't address well what to do for an incidentaloma, an asymptomatic pulmonary cryptococcoma, since most patients in series had symptoms.

"Our findings suggest that an initial period of observation without the administration of antifungal therapy is a reasonable option for non immunocompromised subjects with pulmonary cryptococcosis in the absence of systemic symptoms or evidence of dissemination, as well as after surgical resection for focal cryptococcal pneumonia."

I do not like treating and I do not like not treating, so for now, I can temporize and do a further evaluation for evidence of dissemination. It will be a much easier decision if I can find a cryptococcal antigen positive in one fluid or another. I don't mind bankrupting people and destroying their kidneys if it is for a good cause. Well, I do, but I want a good reason to do potential and real harm.

Primum non nocere. It always seems to suggest you know in advance down which path no harm lies. The real phase, at least by Google translate from English to Latin, should be quod commodum non nocere praeponderat.

Rationalization

Practice Guidelines for the Management of Cryptococcal Disease

http://cid.oxfordjournals.org/content/30/4/710.full

Chest. 2003 Dec;124(6):2143-7. Pulmonary cryptococcosis in nonimmunocompromised patients.

http://www.ncbi.nlm.nih.gov/pubmed/14665493

POLL RESULTS

First

• do no harm. 10%

• do no harm that outweighs the benefit. 25%

• check the insurance. 7%

• check the legal precedent. 0%

• realized you are damned if you do, damned if you don't. 56%

• Other Answers 2%

  • Don't get hurt. Make good parts. Be as productive as is sustainable without getting hurt or making bad parts.

Chapter Fifty-Two

Less than Perfect

It will come as no surprise to anyone that I am less than perfect. My kids will confirm that observation. I try to always do the right thing, but try is not succeed. Fortunately, I am quick to change direction when I discover I am on the wrong path and learned a long time ago that I am far from infallible.

However, I cringe when I make an error.

I cover three hospitals (seven on the weekend) that are twelve or more miles apart and can spend an inordinate amount of time in my car. That goodness for Audible. And hospitalists. As a result, I am often in motion from one institution to another with meetings and other commitments.

It was 2:30 and I was heading out the door for a three o'clock meeting at another hospital when a resident says, Oh by the way, we have a discitis slash osteomyelitis.

Me: Any epidural abscess on MRI?

Resident: No.

Me: Any cultures?

Resident: No, but neurosurgery is on the case, and we are holding antibiotics until we get a biopsy. The patient is clinically stable and no neurologic problems.

Me: Sounds good. Why the epidural? Shooter or diabetic?

Resident: No good reason.

Me: Once you get the biopsy, start the patient on vancomycin and ceftriaxone. I'm running late for a meeting, I'll see them tomorrow if they are stable.

Resident: No problem.

What is wrong with this conversation? I didn't ask the level. Somehow I assumed it was a lumbar infection; most of them are. I didn't ask, they didn't tell.

So that night I get a call from Neurosurgery with a polite question, and here I paraphrase, "Are you a moron for not suggesting antibiotics right away for a cervical discitis?"

I hate that feeling of the blood draining out of my face when you realize, expletive deleted, I screwed up. For a cervical spine infection, I would be late for my meeting. I have seen a few spine infections go from zero to quadriparesis in less than a day, even on antibiotics.

Although there was no epidural abscess, patients can develop rapidly obstructing masses on the cord, and occasionally infections can lead to vertebral artery thrombosis and spinal cord infarction, mimicking cord compression:

"We report 2 patients with SEA [spinal epidural abscess], in whom, on neuropathological examination, the neurological signs were found to be caused by spinal cord ischemia due to thrombosis of leptomeningeal vessels and compression of spinal arteries, respectively, while evidence of spinal cord compression was absent."

In the lumbar spine that isn't a worry---there is no spinal cord down there. But if there is cord involvement and you cannot get cultures ASAP, then you start antibiotics and worry about the cultures later. There were 36 hours between admission and antibiotics. Way too long.

Fortunately for the patient, there was no harm. Nothing evil occurred as a result of the delay in antibiotics. We got a biopsy, and antibiotics were begun. The bone showed osteomyelitis, but unfortunately, the cultures and gram stains were negative and I was stuck with best guess therapy despite the best attempts to get the microbiologic diagnosis. And next time I will ask the level.

Rationalization

Clin Neuropathol. 2004 May-Jun;23(3):102-6. Myelopathy due to spinal epidural abscess without cord compression: a diagnostic pitfall.

http://www.ncbi.nlm.nih.gov/pubmed/15200287

Chapter Fifty-Three

Horse, Zebra or Unicorn?

I have an affinity for the weird. I have to be careful as I will always look for the zebra and could ignore the horse. I try to only go oddity hunting only if really indicated.

The patient has pleural effusions. Not too impressive: 500 cells per cubic millimeter when 10,000 or more would indicate infection, lactate dehydrogenase (LDH) 90 IU per liter when only above 1000 is very worrisome, pH 7.9. But they are recurrent, bilateral, and symptomatic.

He had a thoracentesis which led to a pneumothorax, a video-assisted thoracoscopy, and chest tube all on the left, all of which finally mostly resolved. But now he is re-admitted a week later with fevers, leukocytosis, new left infiltrate and a new right-sided effusion. It has a rind, and there are other areas where the fluid is loculated.

Because of prosthetic valves, I am asked to weigh in as to whether or not he may have endocarditis and I think the answer, short of a diagnostic valve replacement, is no. I think it is all pulmonary.

However, the pleural effusions are a puzzle. A tap of the new right effusion has similar parameters to the left. No heart failure, no infection, no hypothyroidism, none of the other reasons for fluid in the pleural space.

I do not know how I did medicine in the era before PubMed. Nothing like a computer search to find a zebra, or in this case, a unicorn.

I go on a search and find the answer, going on the hypothesis that the top cause of any opaque diagnostic problem is an unusual medication reaction:

"This case report emphasizes that amiodarone should be considered in the differential diagnosis of patients with exudative effusions after a thorough workup has excluded other causes."

Amiodarone is an anti-arrythmic drug that he has been on for [amount of time? Reason?]

Is his amiodarone the reason for the effusions? Maybe. Maybe not. His chest on the left has been surgurized, so it could be a new infection in the pleural space on top of what looks like an aspiration pneumonitis. Being an Occam's kind of guy, I hate coming up with two processes to account for a patient's symptoms. So I will punt to cardiology.

In the end it was the amiodarone.

Rationalization

Pharmacotherapy. 2010 Feb;30(2):218. doi: 10.1592/phco.30.2.218. Amiodarone-induced loculated pleural effusion: case report and review of the literature.

http://www.ncbi.nlm.nih.gov/pubmed/20099996

POLL RESULTS

I look for

• horses. 14%

• zebras. 8%

• unicorns. 10%

• hic sunt dracones. 8%

• evil lurking in the hearts of men. 55%

• Other Answers 4%

  • Puns

Chapter Fifty-Four

Wild Denouement

I hope that everyone in medicine is doing what they like. I, of course, like ID. I like figuring things out. Why is there a particular infection in a particular patient? That is intellectually fun, but not always possible. And sometimes, after being flummoxed, I get an answer. A wild answer.

It started nine months ago with a coronary artery bypass graft (CABG) that appeared to be a success. Then six months after the CABG, the patient developed a methicillin-sensitive Staph. aureus (MSSA) bacteremia that was credited to his dialysis catheter. The catheter was removed and he was discharged on a two-week course of nafcillin.

One week later he showed up in my hospital with a tender, red mass on the inferior aspect of his sternum. A CT scan didn't show much; it was debrided and gram stain and culture were negative. I saw the patient at that time of the sternal mass and I had a lot of uncertainty about what was going on. His transesophageal echocardiogram (TEE) was negative; the CT scan showed no pathology outside the sternum, and repeat blood cultures were negative. Was it more S. aureus? Probably. But I couldn't find where it was, and I was uncertain I really knew what was going on. So I stretched out his therapy to four weeks.

He did well for the next two months, and then got admitted to another hospital and was found to have a ruptured pseudoaneurysm/mycotic aneurysm of the ascending aorta. Now there was all new pus tracking in the mediastinum, sternum, and chest wall.

He was taken to the OR and the infected aorta removed and replaced.

Years ago I had a patient I diagnosed with endocarditis, placed on therapy, and transferred to surgery to get his valve replaced as he went into heart failure. Every year the surgeon gets bottles of wine to commemorate his survival, 7 bottles for seven years. I tell the cardiovascular surgeon he gets all the glory and the internist gets squat. But really, sometimes they really do earn their pay. No way could I do an emergent aorta replacement. I don't have the temperament, much less the skill set.

Evidently, when they cannulate the aorta as part of a CABG procedure, they use a felt pledget to help repair the rent in the aorta. The surgeon thinks it was the pledget that was infected, probably from the bacteremia two months ago. He had a similar case back in his training and this would explain why the infection looked like it went from the outside-in, not inside-out, especially as the patient never had any aortic pathology to seed.

There are not many cases like it on the PubMeds:

"Surgical findings showed vegetation of the aortic valve and a subannular type mycotic aneurysm. In the aneurysm, the infected pledget used in a previous surgery was found."

Wild case. That was a why I would have never, ever, suspected.

Rationalization

Nihon Kyobu Geka Gakkai Zasshi. 1993 Aug;41(8):1373-7. [A case report of aortic valve and VSD Dacron patch infective endocarditis after VSD patch closure 15 years ago].

http://www.ncbi.nlm.nih.gov/pubmed/8360540

Chapter Fifty-Five

Gee

The patient has cellulitis of the leg. A common enough reason for admission. The rest? Not so common.

First, the blood cultures. Group G Streptococcus. S. equisimilis. While this is a more common cause of bacteremic cellulitis than S. pyogenes, I can't say I see all that many bacteremic soft tissue infections. A least I know what to kill.

Second is the erythroderma: it is on the leg. And the right groin. And the opposite arm. And the back. But not the abdomen or chest, and the eyes and tongue are normal. So it is not the typical rash of toxic shock syndrome (TSS) or other streptococcal erythroderma. She does not have TSS clinically, and Group G makes TSS toxins only very rarely. The picture doesn't quite fit and there is nothing to suggest a necrotizing fasciitis.

Is the erythroderma from metastatic infection? Maybe, although my Google-fu fails me as I can find only references to Helicobacter causing multifocal/metastatic cellulitis. I know such a thing can happen, but I can't find a reference, so maybe it is a false memory. All the skin lesions have improved at the same rate on a beta-lactam, suggesting a multifocal infection.

Lastly, because the patient had some hemodynamic instability, she got an echocardiogram, and there is a large vegetation on the mitral valve---although nothing on exam to suggest endocarditis. She has only been ill for a few days, so this vegetation grew very rapidly.

Group G streptococci are only rarely reported as a cause of endocarditis, but I will confess to some microbiologic ignorance. There are a number of Group C streptococcal endocarditis series with Streptococcus dysgalactiae subsp. equisimilis, and Streptococcus dysgalactiae subsp. equisimilis can act more like a S. pyogenes. Streptococci keep being randomly assigned new names and classifications. As a result, while the Lancefield typing has remained the same, I am less certain how much the prior literature on Streptococcus dysgalactiae subsp. equisimilis, usually Group C, applies to my Group G Streptococcus equisimilis.

And how often should you get an ECHO for bacteremic cellulitis? Like almost never? About a decade ago, I saw a routine Group A Streptococcus bacteremic cellulitis in an otherwise normal young woman. The hospitalist ordered an ECHO, and I pooh-poohed the need, but there was a big vegetation on the mitral valve--the only Group A Streptococcus endocarditis of my career. Sometimes you hit pay dirt with kinda-sorta-not-really indicated tests, and it colors your future approach.

She is making slow improvement with her infections: fevers and blood cultures down, erythroderma evolving away. Nothing in the case is typical except I can still kill a Streptococcus.

Update:10 days later

She was probably heading for valve replacement but didn't really have the classic indications: refractory congestive heart failure, recurrent emboli, abscess, nothing. The surgeon was not thrilled about taking the patient to surgery due to comorbidities, and I finally got around to looking at the ECHO. I look at most ECHOs out of curiosity, but when the cardiologist calls a vegetation, who am I to argue?

This time I wonder. It is really odd, and definitely not there on an ECHO from 8 months ago. It looks like a 5 cm piece of spaghetti attached to the underside of the mitral valve, and it snaps around like a whip into the atrium and thuds against the aortic valve. Doesn't look like any infectious vegetation I have seen, and I have seen a fair number.

"I worked with vegetations; I knew vegetations. Senator, you're no vegetation."

So we have a very unusual cause of endocarditis: a vegetation that grew way too fast, attached to the wrong side of the valve, which whips around in a matter that should cause a platelet-thrombin collection to shatter, but no emboli on exam and it has not changed a whit in 10 days.

So if not a vegetation, then what? Not chordae tendinae rupture; cardiology is emphatic on that point.

So I PubMed "mitral valve tumor" and what do you know, there is a cardiac tumor called a papillary fibroelastoma, which can present like this. They can be

"perhaps as large as 3 to 4 cm; and a highly mobile mass, with a pedicle attached to the valve or endocardium."

There are too many atypical features for this to be acute endocarditis, so I have to wonder if we have a zebra. Papillary fibroelastoma can cause emboli, especially if large, but she has multiple comorbidities that make open heart surgery very high risk. I may never know for sure, except if it someday melts away. Then it was a vegetation.

If I get the answer, there will be a third thought.

I never did get follow-up.

Rationalization

J Cardiovasc Ultrasound. 2012 Dec;20(4):213-5. doi: 10.4250/jcu.2012.20.4.213. Epub 2012 Dec 31. Papillary fibroelastoma mimicking vegetation of the mitral valve.

http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3542518/

Tex Heart Inst J. 2012;39(5):731-5. Incidental papillary fibroelastoma multimodal: imaging and surgical decisions in 2 patients.

http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3461690/

J Infect. 1988 Mar;16(2):169-76. Group G streptococcal endocarditis: two case reports, a review of the literature and recommendations for treatment.

http://www.ncbi.nlm.nih.gov/pubmed/3280691

J Clin Microbiol. 2010 Apr;48(4):1484-7. doi: 10.1128/JCM.01737-09. Epub 2010 Feb 10. Two unusual cases of severe soft tissue infection caused by Streptococcus dysgalactiae subsp. equisimilis.

http://www.ncbi.nlm.nih.gov/pubmed/20147644

Chapter Fifty-Six

Really Short Course

My (medical) partner is gone for two weeks and I somehow managed to put my call weekend in the middle, so it is going to be twelve days where I get to cover everything. Five hospitals. Lucky me. That goodness for housestaff and hospitalists.

Twice the work, but that means twice the fun. I am a little like Sherlock Holmes. No, not an IV cocaine addict, thank you very much. When work was slow Holmes always fell into a depressed torpor, and when he had work he became energized. I am the same way. Twice the work means four times the energy. I am an ID addict.

Weird story. The patient has known cardiopulmonary disease: chronic obstructive pulmonary disease (COPD), coronary artery disease (CAD), pulmonary hypertension, and diastolic dysfunction. He has had three days of shortness of breath, which is what gets him to the hospital. No fevers, no chills, no signs or symptoms of infection, and to check out his heart he gets an ECHO.

There is a vegetation on the aortic valve that was not there on admission exactly a year ago to the day. I am asked if he has endocarditis, and besides the vegetations, he has no other markers of the disease. Blood cultures are negative, but he was put on azithromycin for exacerbation of COPD before they were drawn. Curses.

Here is where it gets odd.

For his admission a year ago, he had cellulitis with a beta-hemolytic Streptococcus. He had two days of IV ceftriaxone and was discharged on a 10-day course of amoxicillin. That was the admission with the vegetation-free ECHO. After discharge, his bug was finally identified as S. bovis. Thanks to the vagaries of the electronic medical record, which has the "feature" of removing patients from your list upon discharge whether you want that or not, this went unnoticed.

That was probably his real endocarditis, and it was cured with the short course of IV and po beta-lactams. Not unheard of. In the old days, sensitive streptococcal endocarditis would be cured with oral penicillins. Wild. In this day and age, I would never even think of trying oral antibiotics for endocarditis.

Six months later he had a screening colonoscopy, and multiple polyps were, of course, found and removed. S. bovis in the blood means cancer in the colon.

And now I see him and I do not think he has endocarditis, but short of a valve replacement, I can't say for sure. And since he does not have the physiologic reserve for a relapse, I am going to give him a more typical definitive course of antibiotics.

Years ago, I had a patient with a bicuspid aortic valve with a vegetation and a sensitive Streptococcus who, after 2 doses of ceftriaxone, remembered he was a Christian Scientist and refused further therapy. He came back to the ICU several months later to demonstrate he was well and cured by an alternative means, although I credit the antibiotics. What the record for shortest successful course of IV antibiotics for endocarditis is I cannot find on either the Googles or the PubMeds.

For a sensitive Streptococcus, I wonder what the least amount of antibiotics can be curative. Two doses of ceftriaxone are the record. Has anyone succeeded with one?

Rationalization

Can Med Assoc J. 1961 March 11; 84(10): 535--539.

PMCID: PMC1939332 Treatment of Bacterial Endocarditis by Oral Phenethicillin Potassium (Syncillin)

http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1939332/

Union Med Can. 1963 Aug;92:873-80. [Oral treatment of malignant streptococcal endocarditis by potassium phenethicillin: 10 cases].

http://www.ncbi.nlm.nih.gov/pubmed/14018160

Rocky Mt Med J. 1968 Apr;65(4):33-5. Oral therapy in bacterial endocarditis.

whttp://www.ncbi.nlm.nih.gov/pubmed/4967211

Chapter Fifty-Seven

Evolution

It has been a while. I have suffered from the gruesome twosome. First, my partner was off trying to get typhoid fever and multi-drug-resistant tuberculosis (India). Second, I was ill with a moderate upper respiratory infection. Twice the work and half the health, and even someone with my work ethic has to slow down. But I am healed, my partner is back, and I have a wealth of cases that have accumulated over the last couple of weeks.

The patient is a middle-aged male with a month of progressive fevers, night sweats, and fatigue. He has a partial work up as an outpatient and has a CT scan that shows multiple solid nodules throughout the lung with hilar adenopathy. Because of the fevers, they call me.

He is a smoker, so maybe it is cancer. But he has spent a month in the rural Yucatan, and the symptoms started shortly after returning.

Exposures? He ate the food, drank the water. He spent the time clearing brush and was around goats. He spent one morning on his hands and knees, exploring a cave in the area.

His exam is normal and so are most of his labs, except for a moderately elevated white blood cell count of 15,000 cells per microliter and an elevated erythrocyte sedimentation rate (> 100 mm/hr).

The CT? Looked like cancer or fungus to me. So I sent off all the studies for Mexico, and it was all negative. Histoplasma urine antigen, Coccidiomycosis, Cryptococcus all negative. TB? QuantiFERON negative. No sputum. Darn. I was betting on histoplasmosis since it is found in the caves of Mexico.

"Three Austrian travelers (a 37-year-old woman, a 47-year-old woman and a 47-year-old man) presented with fever, dyspnea, thoracodynia, cephalea, arthralgia and fatigue 4 weeks after visiting a bat cave in Mexico. Computed tomography of the lungs showed bilateral nodular infiltrates in all three patients and enlarged mediastinal lymph nodes in two patients."

So he went for lymph node biopsy, and it showed budding yeast and this week it was confirmed as growing Histoplasma.

Why was the antigen negative?

"Disease in patients in Mexico and Latin America has been described as being caused by isolates of Histoplasma capsulatum var. capsulatum that exhibit distinct genetic profiles in comparison to those from the United States"

Which is what you would expect. Basic evolution. ID is nothing if not applied evolution. African Histoplasma is very different, haven taken a different path when the Atlantic opened up. New World Histoplasma have not been separate as long.

"Until recently, Histoplasma capsulatum was believed to harbour three varieties, var. capsulatum (chiefly a New World human pathogen), var. duboisii (an African human pathogen) and var. farciminosum (an Old World horse pathogen), which varied in clinical manifestations and geographical distribution. We analysed the phylogenetic relationships of 137 individuals representing the three varieties from six continents using DNA sequence variation in four independent protein-coding genes. At least eight clades were identified: (i) North American class 1 clade; (ii) North American class 2 clade; (iii) Latin American group A clade; (iv) Latin American group B clade; (v) Australian clade; (vi) Netherlands (Indonesian?) clade; (vii) Eurasian clade and (viii) African clade. Seven of eight clades represented genetically isolated groups that may be recognized as phylogenetic species. The sole exception was the Eurasian clade which originated from within the Latin American group A clade. The phylogenetic relationships among the clades made a star phylogeny. Histoplasma capsulatum var. capsulatum individuals were found in all eight clades. The African clade included all of the H. capsulatum var. duboisii individuals as well as individuals of the other two varieties. The 13 individuals of var. farciminosum were distributed among three phylogenetic species. These findings suggest that the three varieties of Histoplasma are phylogenetically meaningless. Instead we have to recognize the existence of genetically distinct geographical populations or phylogenetic species. Combining DNA substitution rates of protein-coding genes with the phylogeny suggests that the radiation of Histoplasma started between 3 and 13 million years ago in Latin America."

As a result, Panamanian histoplasmosis does not always react with the US antigen test; perhaps the same with Mexican histoplasmosis:

"Isolates from patients in the United States are predominantly found to be class 2 isolates when typed using the nuclear gene YPS3, while isolates from Latin America are predominantly typed as class 5 or class 6... In this study, we have compared the sensitivity of antigen detection for AIDS patients from Panama who had progressive disseminated histoplasmosis to that for those in the United States. Antigenuria was detected in the MVista Histoplasma antigen enzyme immunoassay (EIA) in 95.2% of Panamanian cases versus 100% of U.S. cases."

So it is now a course of itraconazole (taken with Coke of course) and I expect he will get better.

Rationalization

Nihon Kokyuki Gakkai Zasshi. 2001 Apr;39(4):293-7. [Cave-associated acute pulmonary histoplasmosis in two Japanese returning from Mexico].

http://www.ncbi.nlm.nih.gov/pubmed/11481831

Infection. 2008 Jun;36(3):282-4. Epub 2007 Sep 28. Pulmonary histoplasmosis in three Austrian travelers after a journey to Mexico.

http://www.ncbi.nlm.nih.gov/pubmed/17906839

Clin Vaccine Immunol. 2008 April; 15(4): 681--683. Detection of Histoplasma capsulatum Antigen in Panamanian Patients with Disseminated Histoplasmosis and AIDS.

http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2292648/

Mol Ecol. 2003 Dec;12(12):3383-401. Phylogeography of the fungal pathogen Histoplasma capsulatum. Link

Eur J Clin Pharmacol. 1997;52(3):235-7. Influence of an acidic beverage (Coca-Cola) on the absorption of itraconazole.

http://www.ncbi.nlm.nih.gov/pubmed/9218932

Chapter Fifty-Eight

Bad Bug

Bad bug bad bug/Watcha gonna do, watcha gonna do when I come for you/Watcha gonna do, whaddya gonna do when I overthrow you.

Zorak

Gram-positive cocci in the blood is a common cause of consultation --- bread and butter ID. But I always get that "oh crap" feeling when I look at the chest X-ray, and there is a pacer/arrythmia control device present.

How do you give a rabbit endocarditis? Besides making it a junkie? Put a wire across the valve, wait a few days, and then give it a big bolus of bacteria and the valve will nearly always get infected.

The valve, whacking against the wire 40,50,60 times a minute, gets slightly traumatized and is fertile soil for infection. Unlike rabbits in an endocarditis study, humans rarely (or so I hope) get infected with millions of colony-forming units of bacteria. We may get the damaged valve, but we do not get the high-grade bloodstream infection that seeds the valve. Usually. But everyone once in a while some bacteria manage to evade the immune system and settle on a tricuspid valve and the adjacent wire.

The patient presents with fevers and chills during dialysis and has gram-positive cocci in the blood.

The initial assumption is that it is from the dialysis catheter, but the call me, I see the chest X-ray and, "oh crap" there's the pacemaker.

And the next day it is identified as Staphylococcus lugdunensis, and I know there is going to be endocarditis. And the transesophageal echocardiogram shows a vegetation on the tricuspid valve, which means the pacer wire needs to go. Double "oh crap," as taking out a pacer, especially when the patients is dependent on it, is no simple task. In almost thirty years, I have yet to see a patient die from pacer extraction, but one of these days it is going to happen.

I have written about S. lugdunensis before. It is a bad bug; a coagulase-negative Staphylococcus that is closer to S. aureus in virulence. When in the blood it often means endocarditis, it is usually beta-lactam susceptible, it smells like bleach (but not teen spirit), and its preferential habitat on the human body is the great toe.

And wouldn't you know it, there is a chronic ulcer under the first metatarsal that is the likely source.

There is a smattering of cases in the literature. Complications are the norm, and pacer removal is a good thing:

"10 cases of IE[infectious endocarditis] caused by S lugdunensis were identified at our institutions, representing 0.8% (four of 467), 1.5% (two of 135), and 7.8% (four of 51) of cases of native valve, prosthetic valve, and pacemaker lead endocarditis in the non-drug misusers. Native valve IE was present in four patients (two aortic, one mitral, and one pulmonary), prosthetic valve aortic IE in two patients, and pacemaker lead IE in the other four patients. All patients with left sided IE had serious complications (heart failure, periannular abscess formation, or shock) requiring surgery in 60% (three of five patients) of cases with an overall mortality rate of 80% (four of five patients). All patients with pacemaker IE underwent combined medical treatment and surgery, and mortality was 25% (one patient). In total 59 cases of IE caused by S lugdunensis were identified in a review of the literature. The combined analysis of these 69 cases showed that native valve IE (53 patients, 77%) is characterised by mitral valve involvement and frequent complications such as heart failure, abscess formation, and embolism. Surgery was needed in 51% of cases and mortality was 42%. Prosthetic valve endocarditis (nine of 60, 13%) predominated in the aortic position and was associated with abscess formation, required surgery, and high mortality (78%). Pacemaker lead IE (seven of 69, 10%) is associated with a better prognosis when antibiotic treatment is combined with surgery."

The pacer is out, the patient is on nafcillin and doing well. So far.

Rationalization

Heart. 2005 Feb;91(2):e10. Staphylococcus lugdunensis infective endocarditis: description of 10 cases and analysis of native valve, prosthetic valve, and pacemaker lead endocarditis clinical profiles.

http://www.ncbi.nlm.nih.gov/pubmed/15657200

N Z Med J. 2012 May 11;125(1354):51-9. Unusually virulent coagulase-negative Staphylococcus lugdunensis is frequently associated with infective endocarditis: a Waikato series of patients.

[http://www.ncbi.nlm.nih.gov/pubmed/22595924]{.underline}

Zorak. Bad Bug

https://www.youtube.com/watch?v=1AUuZ021ppQ

Chapter Fifty-Nine

Now What?

The patient has leukemia as a consequence of prior chemotherapy for a solid tumor years ago. So sad. Cure one cancer, get another. Oncology tells me this form of leukemia is more refractory to remission than the run-of-the-mill, as if leukemia can be run-of-the-mill, acute myelogenous leukemia.

So he has been neutropenic for months but has not had chemotherapy for the leukemia yet, and is on prophylactic quinolone to help prevent infection. For the week prior to admission he has had fevers and cough. The chest X-ray shows some patchy infiltrates, but the CT scan is more impressive, showing nodules with adjacent interstitial infiltrates.

The CT looks more fungal to me, but he isn't all that ill, complaining mostly of cough. Usually, patients with fungal pneumonia are ill.

We send off cultures and fungal blood tests and call pulmonary for bronchoscopy. They also suggest that given the lack of toxicity, what we are seeing are leukemic infiltrates. I didn't consider that option as I did not know that was a thing, even in neutropenics.

All the cultures are negative and neutropenic fever trends down on antibiotics.

The bronchoscopy? Fungal elements. Good thing we started voriconazole. And there are other studies that come back. As long as they are down there, the pulmonologist sends off everything, and some of those everythings come back positive.

A PCR test for HSV (Herpes Simplex Virus) is positive. Ignore that. To have HSV is to excrete HSV and if you test for it, you will find it. It is not the cause of this disease.

A PCR for respiratory syncytial virus (RSV) is positive. Now what? I have seen just one RSV in an adult, so what are the CT findings?

"RSV presented with an airway-centric pattern of disease (13/19 cases [68%]) characterized by varying mixtures of tree-in-bud opacities and bronchial wall thickening, with or without peribronchiolar consolidation."

That tells me nothing, but it does not look like the photos in the reference.

RSV is not a common cause of respiratory disease in the elderly, although he is wheezy and coughing and not that ill, so maybe it is the real deal, and perhaps he has more than one process. Occam's razor is not that reliable in the profoundly immunocompetent.

And RSV is a bad disease in patients with leukemia:

"RSV infection is associated with significant morbidity and mortality in leukemia patients; treatment with aerosolized ribavirin at the stage of URI may prevent pneumonia in some subsets of patients."

And if it is real, now what? Treat? Ribavirin is used in those with severe illness, which he doesn't have, and you can say ribavirin is safe. But efficacious? Oral therapy is unlikely to help:

"Thirty-day mortality and underlying respiratory death rates were 31% (35/114) and 12% (14/114), respectively, for the oral ribavirin group versus 19% (6/31) and 16% (5/31), respectively, for the nonribavirin group (P = 0.21 and P = 0.56)...Our data suggest that oral ribavirin therapy may not improve clinical outcomes in hematologic disease patients infected with paramyxovirus."

And since the patient is neutropenic without having had chemotherapy yet (most of the literature is patients with leukemia on chemotherapy), there will be no immune reconstitution to save the day. Inhaled ribavirin is $14,000 for a 10-day course and in adults there is little information to suggest benefit.

Clinically he is doing fine with symptom treatment, and no progression on the X-ray findings or hypoxia, so we nervously watch and wait.

Rationalization

AJR Am J Roentgenol. 2011 Nov;197(5):1088-95. doi: 10.2214/AJR.11.6501. CT of viral lower respiratory tract infections in adults: comparison among viral organisms and between viral and bacterial infections.

http://www.ncbi.nlm.nih.gov/pubmed/22021500

Haematologica. 2007 Sep;92(9):1216-23. Epub 2007 Aug 1. Characteristics and outcome of respiratory syncytial virus infection in patients with leukemia.

http://www.ncbi.nlm.nih.gov/pubmed/17666367

Antimicrob Agents Chemother. 2013 Feb;57(2):983-9. doi: 10.1128/AAC.01961-12. Epub 2012 Dec 10. Efficacy of oral ribavirin in hematologic disease patients with paramyxovirus infection: analytic strategy using propensity scores.

http://www.ncbi.nlm.nih.gov/pubmed/23229488

Chapter Sixty

Bloody Nose

Noses are dangerous. Keep your fingers away from them as it can lead to trouble.

Years ago, I had a patient with a brain abscess from excessive nose picking. Completely removed the nasal septum and picked away the base of the skull to the brain. Remarkable.

Not every case is that odd.

This patient has fevers, as almost all my patients do, and a bloody nose and came to the hospital. Blood cultures grew S. aureus, and they called me.

It was quite a bloody nose. She was on warfarin for atrial fibrillation, and home oxygen for chronic obstructive pulmonary disease. The theory was the O~2~ dried out her nose so much it caused bleeding, and because of the warfarin the bleeding didn't stop.

When I saw her the exam was negative except for a particularly unpleasant appearing large inflated balloon in the nose.

The exam was negative, as was an ECHO and repeat blood cultures. So why the S. aureus bacteremia? The bloody nose.

Patients with Osler-Weber-Rendu have an increase in S. aureus infections because of recurrent nosebleeds in the presence of S. aureus colonization, which occurs in 25% of people at any given time.

"Extracerebral infections accounted for 67% of all infections, and most involved Staphylococcus aureus and were associated with prolonged epistaxis."

At least in children, a bloody nose leads to an increase in S. aureus colonization.

"Children with epistaxis are more likely to have nasal colonization with S aureus than controls. Our data would support the hypothesis that S aureus replaces existing nasal flora and causes inflammation and new vessel formation."

And there is one case on a bloody nose leading to S. aureus bacteremia.

So I blame the epistaxis. And in the future, I am going to keep my fingers as far from my nose, and anyone else's, as is possible.

Rationalization

An unusual case of epistaxis and Staphylococcus aureus bacteremia in an older Chinese woman. Ruan W, Yap P, Foo SS, Lee KK, Low JA, Yap KB. J Am Geriatr Soc. 2010 Sep;58(9):1815-6

http://www.ncbi.nlm.nih.gov/pubmed/20863354

Clin Infect Dis. 2007 Mar 15;44(6):841-5. Epub 2007 Feb 1. Hemorrhagic hereditary telangiectasia (Rendu-Osler disease) and infectious diseases: an underestimated association.

http://www.ncbi.nlm.nih.gov/pubmed/17304458

Otolaryngol Head Neck Surg. 2008 Mar;138(3):307-10. doi: 10.1016/j.otohns.2007.10.029. Childhood epistaxis and nasal colonization with Staphylococcus aureus.

http://www.ncbi.nlm.nih.gov/pubmed/18312876

Chapter Sixty-One

Big Picture: Why?

A 68-year-old woman presents with two months of fevers, chills, and night sweats. The physical exam shows an easily heard (grade 3/6) aortic murmur, emboli in the conjunctivae, and very poor dentition. The labs are remarkable for an elevated white blood cell count (16,000 cells per cubic millimeter) and hematuria (blood in the urine). Blood cultures yield Streptococcus mutans with a penicillin minimum inhibitory concentration (MIC) 0.02 micrograms/milliliter, meaning it was penicillin-sensitive. The echocardiogram shows a vegetation on the aortic valve. So it's endocarditis. But why?

Not little picture "why." Big picture "why," like initial conditions at the Big Bang why.

It gets down to why we have cavities. Cavities are not part of the human condition. Hunter-gatherers did not need to floss:

A comparative study of the frequency of caries in certain periods indicates the following: in the hunting-gathering period it was 1%-2%, in the Early Neolithic it was 3%-5% (Catal Hoyuk), in the Neolithic (beginning of agriculture) it was 5.6% (Cayonu) in the Chalcolithic it was 11.7% (Norsuntepe), in the Roman period it was 11.1% (Panaztepe), and 16% (Datca), in the Late Byzantine it was 10.9% (Iznik) and in the Medieval it is 14.2% (Arslantepe). These findings contribute to understanding how dietary change and life conditions are interrelated with the changing patterns of dental diseases in Anatolian populations.

Of course, they didn't live very long either, so it perhaps the lack of cavities was due to an early death --- not a good trade-off.

There was an interesting article in Science that reviewed two papers suggesting that a change in diet led to dental caries.

Streptococcus mutans started to be found when agriculture started:

Researchers looked at DNA in tarter of teeth from 400 to 7500 years ago. Ick. They found

"Hunter-gatherers had fewer species that cause cavities and periodontal disease, and different percentages of all 15 phyla of bacteria found in modern teeth... Early farmers showed a sharp increase in bacteria that cause tooth decay, such as Veillonellaceae strain and a dramatic surge in Porphyromonas which causes periodontal disease.

And S. mutans did not show up until 4300 years ago, and then 400 years ago diversity dropped again, with S. mutans predominating.

Another study looked at S. mutans across the world, and by using genetic techniques (i.e., magic), it looks like S. mutans started living in human mouths about 10,000 years ago.

So a shift to an agricultural society led to a selection of S. mutans in human mouths, cavities, and eventually endocarditis in my patient. Blame it on wheat.

One of the many reasons ID is just so much cooler than the rest of medicine: it can provide insights into the beginning of human culture. No other part of medicine does that so often or so well.

Rationalization

Implications of Dental Caries in Anatolia: From HuntingGathering to the Present Human Evolution (August 2006), 21 (3-4), pg. 215-222

http://journals2.scholarsportal.info/details.xqy?uri=/03939375/v21i3-4/215_iodciafhttp.xml

Science 22 February 2013: Vol. 339 no. 6122 pp. 896-897 DOI: 10.1126/science.339.6122.896 How Sweet It Is: Genes Show How Bacteria Colonized Human Teeth

https://science.sciencemag.org/content/339/6122/896

POLL RESULTS

ID. Bah. I am so glad I am

• not a doctor. 25%

• a surgeon. 4%

• anything but an ID doc. 11%

• medicated. 11%

• drunk. 21%

• Other Answers 29%

  • a retired microbiologist

  • reading, not doing.

  • RETIRING SOON

  • I am ID

Chapter Sixty-Two

A Big Scare

Some diseases worry me more than others. When my kids were young and had a fever, I would check their skin about every five minutes for purpura. Meningococcus is/was my biggest ID fear.

Then there are infection control fears. I have never worried too much about catching diseases at work because I am punctilious about infection control. Go figure. Knowing how a disease is spread is most of the battle of not acquiring it.

If you have an unusual presentation of an unusual (for the US) disease, then there is the potential for the inadvertent spread of contagion. In the old days, ID was called "communicable diseases"; I like the implied warning in the term communicable.

Today's patient is admitted with what looks to be a progressive multi-focal cavitary pneumonia of maybe three weeks' durations after a sore throat. The history is not great as he is from Moldova, and there is a language barrier. There are cavities in the left lower lobe of his lung, additional lesions in the left middle lobe, left upper lobe, and right upper lobe. The thought was septic emboli, perhaps endocarditis or Lemmier's syndrome, which is thrombophlebitis of the internal jugular vein usually caused by anaerobic organisms from an oropharyngeal infection.

.

He has a bronchoscopy and, goodness gracious great balls of fire, it is 3+ acid-fast bacilli---active tuberculosis. And he has spent a day or two without isolation in the hospital. It happens. I hate it when it happens. But it happens.

But put TB and Moldova in the search box, and the return gives you a sense of dread.

The first hit from the WHO is not My Generation, but

"The Republic of Moldova is among the 27 high multidrug-resistant tuberculosis (MDR-TB) burden countries in the world"

And the PubMeds reveal that in Moldova

"Of 3463 cases, 57.1% were recorded as 'new' and 24.6% as 'retreatment' cases; previous treatment status was not recorded for 18.3%. Of the 'new' cases, 1655 were correctly classified according to international recommendations and 322 were misclassified. The number of cases increased from 443 in 1995 to 939 in 1999; the proportion of 'retreatment' increased from 17.4% to 35.5%, 'any drug resistance' from 20.3% to 41.6%, and 'multidrug resistance' from 2.7% to 11.2%. In 1998-1999, 'any drug resistance' and 'multidrug resistance' in 800 previously untreated cases were respectively 29.1% and 5.3%, and respectively 61.0% and 21.9% in 521 'retreatment' cases. Of a total of 216 'multidrug-resistant' cases in 1998-1999, 21.8% were reported resistant to ethambutol and 81.5% to streptomycin."

Not good. Fortunately, public health is quite helpful, and I have access to some modern testing. With their blessing we had the acid-fast bacilli tested by genetic methods (BEACON), and the test came back sensitive to all the known TB drugs.

Dodged that bullet.

There are a variety of molecular tests available for TB, although due to the expense they only use them when there is a worry about multidrug-resistant TB. Like Moldovans.

The tests are not perfect, but then no test is. But it does give a jump on the ever so slow growth and testing that we usually rely on. And I do not have to fret about what to do if there was an exposure/conversion from a multidrug-resistant TB.

But it is only a matter of time before such a beast makes it to Portland. Resistance is not futile; it is inevitable.

Rationalization

Int J Tuberc Lung Dis. 2003 Apr;7(4):336-42. Resistance to anti-tuberculosis drugs and practices in drug susceptibility testing in Moldova, 1995-1999.

http://www.ncbi.nlm.nih.gov/pubmed/12729338

Report of Expert Consultations on Rapid Molecular Testing to Detect Drug-Resistant Tuberculosis in the United States

http://www.cdc.gov/tb/topic/laboratory/rapidmoleculartesting/background.htm

POLL RESULTS

My biggest fear is

Multidrug-resistant TB 12%

Multidrug-resistant GC 6%

Meningococcus 17%

People will see me for the fraud I truly am. 31%

fear itself 27%

Chapter Sixty-Three

More Acid-Fast Bacilli

The balance is not quite there. We have steadily eradicated nosocomial infections at my hospitals with all the practice bundles and other interventions.

We have had over 800 fewer infections in the last five years compared to historical controls. That is a real decline, as those would have been consults, and my work has steadily declined this century. Highly active antiretroviral therapy (HAART), granulocyte colony-stimulating factor (G-CSF), hand hygiene, and other medical advances have slowly eroded my practice, but I am old enough, with just a handful of full-time practice years ahead of me, that it suits me just fine.

You would think that biologics would help make up the difference, with all of the new immunosuppressant drugs--but Remicade and its numerous cousins are not really causing the odd infections to fill the void.

Today's patient had a knee replacement 14 years ago for rheumatoid arthritis (RA). His RA has been well controlled on low-dose prednisone and methotrexate. Five years ago his artificial knee began to bother him: pain and swelling that slowly progressed. Several taps yielded nothing, and he lived with it.

As his RA became more symptomatic, he was started on Remicade, and his knee worsened while the rest of his joints improved.

Osteomyelitis was noted on plain films, and so he came in for a two-stage exchange. Bone infection and synovitis were noted when the knee was removed.

All the specimens were positive for acid-fast bacilli.

I had one case of Mycobacterium tuberculosis in a prosthetic knee years ago, but it was in a PPD skin test-positive Filipino. This patient is PPD negative and he had no risks for TB.

And it wasn't TB.

It is growing Mycobacterium kansasii.

Odd.

An M. kanasii-like organism may have been the precursor to Mycobacterium tuberculosis:

Genomic studies have provided a refined understanding of the genetic diversity within the Mycobacterium genus, and more specifically within Mycobacterium tuberculosis. These results have informed a new perspective on the macro- and micro-evolution of the tubercle bacillus. In the first step, a M. kansasii-like opportunistic pathogen acquired new genes, through horizontal gene transfer, that enabled it to better exploit an intracellular niche and ultimately evolve into a professional pathogen. In the second step, different subspecies and strains of the M. tuberculosis complex emerged through mutation and deletion of unnecessary DNA.

There is one other case in the PubMeds of a prosthetic joint infection due to this beast and a smattering of normal joint infections, mostly in the immunoincompetent.

Susceptibilities are pending, and I send all my infections with atypical acid-fast bacilli to Dr. Kevin Winthrop at OHSU--a former resident who is now one of, if not the expert on atypical mycobacterial infections associated with biologics. He told me we don't have this bug in Oregon, so it will be interesting to see what he discovers that I did not.

Rationalization

Adv Exp Med Biol. 2013;783:81-91. doi: 10.1007/978-1-4614-6111-1_4. Evolution of Mycobacterium tuberculosis.

http://www.ncbi.nlm.nih.gov/pubmed/23468104

J Clin Microbiol. 2006 Jul;44(7):2648-9. Septic arthritis caused by Mycobacterium kansasii in a prosthetic knee joint.

http://www.ncbi.nlm.nih.gov/pubmed/16825405

Clin Infect Dis. 1999 Dec;29(6):1455-60. Mycobacterium kansasii septic arthritis: French retrospective study of 5 years and review.

http://www.ncbi.nlm.nih.gov/pubmed/10585795

Chapter Sixty-Four

Man. I Hate Failure.

One of the many nice things about ID is I get to cure people. Really cure. When I am done with you, your infection is gone for good. Most of the time. It is why ID is a simple job as well: me find bug, me kill bug, me go home.

I tend to be conservative in treating patients, since failure has the potential to be catastrophic.

The patient was admitted three weeks ago with the abrupt onset of fevers and weakness. She was brought into the ER where a blood sugar of 400 was found and the blood grew Group B Streptococcus--aka S. agalactiae.

She had a skinned knee, but no other source for the bacteremia: the physical exam was negative, the review of systems negative, a transthoracic echocardiogram was normal. We even MRIed her back as she had marked leg weakness, but nothing.

Spontaneous bacteremia, often in diabetics, alcoholics and cancer patients, is a common manifestation of Group B streptococcal bacteremia in adults.

Bacteraemia without focus (43.4%) and bone and joint infections (18.7%) were the main clinical manifestations.

and

We found that 39% of isolates from adult blood cultures were group B, a frequency nearly identical to that of group A streptococcal bacteremia. Most (66%) adult patients were more than 50 years old. Primary bacteremia was the most frequent clinical diagnosis, occurring in 7 (22%) of 32 patients. Nonhematologic cancer was the most frequently associated condition (25%). Nineteen percent of the patients had diabetes mellitus. The overall mortality rate was 31% and was significantly associated with increasing age.

Been there, done that many times. No suggestion of a drainable focus, and the knee skin trauma seemed a reasonable source for the bacteremia. So we, and by we I mean I, gave her IV antibiotics while in the hospital and then home on a course of oral antibiotics. There is surprisingly little data for the duration of treatment for bacteremic infections, but if there is no focus, I can't see the reason for longer courses. The endpoint is cure, which is usually clinical.

Two days after stopping the antibiotics, she is back with a fever and Group B streptococci in the blood and, like last time, responds quickly to antibiotics. Unlike last time we get a transesophageal echocardiogram, and there is a vegetation on the aortic valve. Expletive deleted.

I can't remember a prior case of Group B streptococcal endocarditis, but then I can't remember yesterday all that well. Group B Streptococcus is an unusual cause of endocarditis nonetheless, with maybe 150 in the literature. Years ago, the green edition of Mandell, Douglas, and Bennett's Principles and Practice of Infectious Diseases had a table showing the odds that a given streptococcal bacteremia might be due to endocarditis. The table was removed from subsequent editions, and I have not seen the table reproduced elsewhere, but it was a handy reference.

Now the patient is on a long course of beta-lactam for endocarditis.

Man. I hate failure.

Rationalization

Clin Microbiol Infect. 2011 Oct;17(10):1587-9. doi: 10.1111/j.1469-0691.2011.03628.x. Epub 2011 Aug 29. Invasive group B streptococcal infections in adults, France (2007-2010).

http://www.ncbi.nlm.nih.gov/pubmed/21883671

Medicine (Baltimore). 1995 Jul;74(4):176-90. Group B streptococcal bacteremia in adults. Five years' experience and a review of the literature.

http://www.ncbi.nlm.nih.gov/pubmed/7623653

Crit Care. 2011;15(6):R267. doi: 10.1186/cc10545. Epub 2011 Nov 15. Duration of antibiotic therapy for bacteremia: a systematic review and meta-analysis.

http://www.ncbi.nlm.nih.gov/pubmed/22085732

Clin Infect Dis. 2002 Jun 15;34(12):1576-84. Epub 2002 May 24. Streptococcus agalactiae infective endocarditis: analysis of 30 cases and review of the literature, 1962-1998.

http://www.ncbi.nlm.nih.gov/pubmed/12032892

Chapter Sixty-Five

I Don't Believe the Diagnosis

The patient has Parkinson's and has an aspiration event at a restaurant. A Heimlich fixes it, but over the next five days, she does not do so well. Progressive cough and low-grade fevers, she is admitted, and a CT scan shows a fuzzy, roundish, multifocal pneumonia, almost embolic looking, but not quite.

Her only other problem is diabetes. Blood cultures are drawn and the patient is started on pneumonia therapy.

Next day the blood cultures are positive for budding yeast. Huh. I figure it must be cryptococcal pneumonia with fungemia. No other yeast is going to do this. So I start her on amphotericin. She has no reason for a Candida in the blood: No indwelling lines, no IV drug abuse, no neutropenia, no nothing. It has been six months since she has been in the clutches of the medical-industrial complex.

She does better, and the next day they tell me it is Candida tropicalis. No way. I look the patient over from head to toe. Normal ECHO, no abdominal pathology, again no nothing.

I think this is aspiration Candida pneumonia with fungemia and I do not believe it.

Candida pneumonia almost never exists; Candida in the sputum is almost always contamination from the gastrointestinal tract. The rule is you have to have pathology to confirm Candida pneumonia.

I have seen two cases of real deal Candida pneumonia. One was a massive aspiration in a patient with thrush who, on autopsy, had alveoli full of Candida. The other was a fungus ball that was resected due to recurrent hemoptysis, and rather than the usual Aspergillus, it was a C. albicans ball.

There is only one similar case in the PubMeds and that is with a C. glabrata. The only C. tropicalis pneumonias are from references that date from 50 years ago, so color me skeptical.

This is too weird even for me. I like to come up with the odd diagnosis, the unusual bug causing a unique infection, but I have trouble buying this one. But I canna come up with another explanation.

And whether it was due to treatment or true-true and unrelated, she improved on antifungals.

Rationalization

Intern Med. 2005 Nov;44(11):1191-4. A probable case of aspiration pneumonia caused by Candida glabrata in a non-neutropenic patient with candidemia.

http://www.ncbi.nlm.nih.gov/pubmed/?term=16357460

Medicine (Baltimore). 1993 May;72(3):137-42. Primary Candida pneumonia. Experience at a large cancer center and review of the literature.

http://www.ncbi.nlm.nih.gov/pubmed/8502166

Mycoses. 2002;45 Suppl 3:22-6. [Candida pneumonia in patients without definitive immunodeficiency].

http://www.ncbi.nlm.nih.gov/pubmed/12690966

POLL RESULTS

When confronted with a unique diagnosis I

believe it. 2%

want it named after me. 30%

caution that it should be considered in similar presentations in the future. 2%

keep looking for a more realistic explanation. 54%

know that it is the great doctors that make the great diagnoses. 13%

Chapter Sixty-Six

Rearrange Your Liver

"And rearrange your liver to the solid mental grace."

Yes. Close to the Edge

Not all the internal organs have been well served by rock lyrics.

The patient comes in with two weeks of fevers, sweats, weight loss and moderate abdominal pain. The white blood cell count is 11,000 cells per cubic millimeter, at the upper limit of normal, and the alkaline phosphatase (ALP, a measure of liver function) is barely elevated. There is no past medical of note and so off to the CT scan, with this result:

{width="4.097222222222222in" height="3.2777777777777777in"}

{width="4.180555555555555in" height="3.0972222222222223in"}

I get a call from the radiologist asking if I had ever seen an infection that looked like the above. No. If these are liver abscesses, it will be the most impressive set I have seen. Usually they are one or two lesions. No biliary problems, no reason for a hematogenous infection. Although the CT does show lots of diverticulosis, there has never been any -itis. I would really worry about tumor, but this is odd for that as well, especially with no risk factors or lab abnormalities of note.

So they tap the lesions, and the pus is returned but the gram stain is negative. No tumor on pathology. Blood cultures remain negative. Ceftriaxone, metronidazole and patience are recommended and after three days we finally grow Fusobacterium nucleatum from the pus.

Odd. But then that describes most of my practice.

Most Fusobacterium cases are associated with dental disease, of which there is none in this case, or the occasional septic thrombophlebitis--of which is there none on close re-review of the CT. So I am stuck with the diverticulosis as a reason, and not a very good one.

Of the 24 or so case reports on the PubMeds, multiple abscesses are common with this organism, which is not the case with other causes of liver abscesses.

Drainage and antibiotics and the patient is slowly getting better.

Rationalization

Indian J Gastroenterol. 2012 Jul;31(4):198-200. Epub 2012 Aug 9. Fusobacterium nucleatum infection mimicking metastatic cancer.

[http://www.ncbi.nlm.nih.gov/pubmed/22875742]{.underline}

POLL RESULTS

The organ best represented in Rock & Roll is

• heart 35%

• hand 0%

• liver 16%

• skin 4%

• libido. You know what I mean :) 39%

• Other Answers 6%

  • duh. the electric organ.

  • tongue

  • spleen. Vented all over the place

Chapter Sixty-Seven

A Brilliant Idea

Back from vacation. Two unexpected days in Denver thanks to a late March snowstorm and the rest in Santa Fe, some of the most blood-drenched land per capita in the US, at least historically. A wonderful time, the first vacation without kids in 19 years. But coming back I am now a week behind in everything. Everything. The world moves on without me, and it shouldn't. Getting back, I wish I was paid by the signature.

I was sitting in the computer room when I was curbsided about a patient I have seen many times in the past. He is a quadriplegic who has been getting an increasing number of urinary tract infections with, no surprise, increasingly resistant organisms. Was there anything that could be done to prevent these infections? Nope.

There is little to be offered to prevent UTIs in neurogenic bladders requiring chronic catheters. Good bladder emptying and careful, clean care are about all we have to offer. Maybe probiotics. And quit treating the bacteria in the bladder if you can. Antibiotics generally only increase the risk for symptomatic UTIs.

At least in young women, treating asymptomatic bacteriuria leads to increased rather than decreased infections. Non-pathogenic bacteria are protective against pathogenic bacteria.

Proof of this concept is that non-pathogenic Escherichia coli 83972 prevents infection when instilled into neurogenic bladders. Too bad I don't have access to non-pathogenic E. coli.

One of the themes in ID the last several years has been the importance of the normal microbiota in preventing infections. If the normal bacteria are gone, it gives pathogens a chance to move in.

I am intrigued by the study that demonstrated that at least in females there is a normal, if unculturable, bacterial flora of the bladder. Perhaps giving lots of antibiotics also alters this flora, increasing the risk for pathogenic bacteria to gain a toe, or perhaps bladder, hold.

And that is when inspiration hit.

Perhaps the problem is a lack of normal flora wiped out by too many antibiotics. The patient needs a urine transplant. If it can work for C. difficile, why not for UTIs?

Brilliant. To judge from Google and PubMed, no one has ever used the words "urine" and "transplant" together, at least in the context of repleting a bladder biome. Someone did write a song with that title; I did not listen to it. Not that I am going to try a urine transplant. Wouldn't be ethical. But it is a great idea. Like all my too-numerous-to-count great ideas, I will leave it to others to do the actual work.

Oh, and although this is April Fool's day, I am as serious as sepsis.

Rationalization

Clin Infect Dis. 2012 Sep;55(6):771-7. doi: 10.1093/cid/cis534. Epub 2012 Jun 7. The role of asymptomatic bacteriuria in young women with recurrent urinary tract infections: to treat or not to treat?

http://www.ncbi.nlm.nih.gov/pubmed/22677710

Urology. 2001 Sep;58(3):339-44. Pilot trial of bacterial interference for preventing urinary tract infection.

http://www.ncbi.nlm.nih.gov/pubmed/11549475

J Clin Microbiol. 2012 Apr;50(4):1376-83. doi: 10.1128/JCM.05852-11. Epub 2012 Jan 25. Evidence of uncultivated bacteria in the adult female bladder.

http://www.ncbi.nlm.nih.gov/pubmed?term=22278835

Chapter Sixty-Eight

Leaky

The patient is out walking the dog and, SQUIRREL!!!, the animal makes a lunge, giving the owner an unexpected yank and a groin pull.

Over the next week, the patient has increasing right groin pain but no fevers, chills or rigors, and he seeks care when the pain becomes unbearable.

Past medical history is important for diabetes cured with a pancreas/kidney transplant a decade ago; he has been doing well since.

The CT scan shows a large, complex fluid collection in the iliopsoas, and I expect it to be S. aureus. I see a minor groin pull that transitions into an iliopsoas abscess once or twice a year. The presumption is that the injury leads to a small bleed in the muscle that is then seeded with the bacteremia that is part of life.

The lack of constitutional signs are a puzzle, but sometimes S. aureus causes a cold abscess, although cold abscess is more often associated with TB. Maybe it is the immunosuppression caused by his anti-rejection medications that is causing a lack of fever.

So we tap it --- lots of white blood cells and no organisms on gram stain.

It grows pure Candida parapsilosis. Huh? That makes no sense from the history. The surgeon, being more clever than I, sends off a request for an amylase level (a mark of pancreatic function) and it is sky high.

So the infection is probably coming from a leaking 10-year-old pancreas transplant. Not unheard of:

"The mean time from transplant to the late leak was 20.5 months (range = 3.5-74 months). A direct predisposing event or risk factor occurring in the 6 wk preceding leak diagnosis was identified in 10 (40%) of the recipients. Such events or risk factors included a biopsy-proven episode of acute rejection (n = 4, 16%), a history of blunt abdominal trauma (n = 3, 12%), a recent episode of cytomegalovirus infection (n = 2, 8%), and obstructive uropathy from acute prostatitis (n = 1, 4%). "

I guess a jerk from the dog is the reason; it fits temporally. And Candida is a common pathogen, at least acutely after a pancreas transplant.

"The overall incidence of wound complications was similar (34% vs 20%, P = NS) in the two groups. Deep abscess formation occurred more frequently in the midline group (27% vs 0%, P = 0.02). Staphylococcus epidermidis and Candida albicans were the most common microbial isolates from deep abscesses."

The infection is drained and getting better on micafungcin. The surgeon is understandably hesitant to go back in to try and fix the leak. It would be a mess under the best of circumstances, but I am not optimistic that conservative therapy will work since acute leaks usually require reoperation. Insert cliche re: time here.

Time actually healed the wound.

Rationalization

Clin Transplant. 2005 Apr;19(2):220-4. Late anastomotic leaks in pancreas transplant recipients - clinical characteristics and predisposing factors.

http://www.ncbi.nlm.nih.gov/pubmed/15740558

Transpl Int. 1996;9(1):62-7. The impact of midline versus transverse incisions on wound complications and outcome in simultaneous pancreas-kidney transplants: a retrospective analysis. https://www.ncbi.nlm.nih.gov/pubmed/8748413

SQUIRREL!!!

https://www.youtube.com/watch?v=SSUXXzN26zg

Chapter Sixty-Nine

Not Enough Time

I am the only ID doctor at three hospitals that are 10 to 30 miles apart, so depending on when I get a consult it may be a while before I can get in to see a patient. Hospitalists make this less of an issue as I can often get the ball rolling over the phone.

I was as far as I could be from one of my hospitals when I got the call:

The patient is an elderly woman with chronic lymphocytic leukemia (CLL), who just had her first dose of Rituxan (rituximab) a week ago. Rituximab is a monoclonal antibody that attacks both normal and malignant B cells that bear the CD20 marker, and can have a wide range of side effects. She is admitted early in the morning afebrile, short of breath, confused, and showing signs of a hemolytic anemia: her hemoglobin level is 4 mg/dL.

Yes, four. Normal is 12 mg/dL or higher. The only time I have seen hemolysis result in a hemoglobin of 4 from infections was a case years ago of Mycoplasma in a 19-year-old girl. She was pale but otherwise remarkably asymptomatic at rest.

But this patient now has all the blood cultures growing gram-positive rods in about 4 hours. Not good.

It will be Clostridium perfringens, I say. Start penicillin and clindamycin and I will be there as soon as I can, but I expect she will die before I get there.

It was and she did. Crap.

C. perfringens with massive hemolysis

"...usually kills patients within hours of presentation."

It is sobering how fast some infections can kill.

Hemolytic anemia caused by Clostridium perfringens is a rare complication in patients with neoplastic diseases. According to the literature, in one case

our patient seems to be the first patient with underlying malignancy (multiple myeloma) who has survived C. perfringens septicaemia complicated with acute haemolysis and acute anuria.

The presumed pathogenesis, at least in the patients I have seen with Clostridial sepsis and malignancy, is that chemotherapy leads either to bowel wall breakdown or death of tumor in or near the bowel. With lymphoma it is perhaps the malignancy in Peyer's patches that serves as a portal for entry. The Clostridia gain a toehold in the dead tissue and then cut loose like almost no other organism.

However, CLL does not involve bowel, right? Wrong. CLL can involve the bowel and

"Gastrointestinal CLL manifestations can also form a route for infectious complications. Sadullah et al. have reported a case, where life-threatening gram-negative sepsis developed in a patient with CLL in association with postchemotherapy neutropenia."

Neither CLL nor Rituxan has, as best I can find after a Google and PubMed search, ever been reported associated with C. perfringens massive hemolysis.

Good. It is the first case of C. perfringens with massive hemolysis of my career and I hope never to be called about this again.

Rationalization

Ugeskr Laeger. 1999 May 3;161(18):2678-80. [Acute hemolytic anemia in Clostridium perfringens infection].

http://www.ncbi.nlm.nih.gov/pubmed/10434790

Gastroenterol Res Pract. 2008; 2008: 742146. Published online 2008 June

19. doi: 10.1155/2008/742146 PMCID: PMC2533105 Colonic Involvement in a Patient with Chronic Lymphocytic Leukaemia

http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2533105/

Chapter Seventy

Names

What's in a name? that which we call an anaerobe By any other name would smell as foul;

~ Edward de Vere, 17th Earl of Oxford

After a decade of mostly science and history, I have returned to my literary roots, reading a lot of fantasy of late. Right now, I am going through the Dresden Files on Audible as I drive from hospital to hospital. A recurring theme in all fantasy is knowing the true name of an object brings with it power over that object.

It's true in ID. I need to know the name of the bug before I can optimally kill it and know the significance of the culture.

Today's patient is elderly, moderately demented, and admitted with fevers, rigors, and altered mental status.

The chest X-ray shows a right lower middle lobe infiltrate in the lung.

Pneumonia. Seems simple enough. And the blood cultures grow...

An anaerobic gram-negative rod. Its name? Not known for three days, and today it is called Bacillus fragilis.

Now I am not so certain. The patient has neither teeth nor abdominal pain. So anaerobic pneumonia is unlikely given his lack of teeth. B. fragilis is relatively unusual in respiratory samples, depending on who is sampled:

A prospective study of 54 cases of pulmonary infection following aspiration was performed. Specimens utilized for bacteriologic study were either transtracheal aspirates, empyema fluid or blood. Appropriate anaerobic bacteriologie [sic] methods were employed. Anaerobic bacteria were recovered in 50 patients (93 per cent) and were the only pathogens in 25 (46 per cent). The predominant species were Bacteroides melanino-genicus, Fusobacterium nucleatum and anaerobic or microaerophilic gram-positive cocci. Bacteroides fragilis, which is resistant to many commonly used antibiotics, was recovered in nine patients (17 per cent)

vrs

"A total of 39 clinical specimens, which consisted of 19 bronchial aspirates, 10 percutaneous lung aspirates, 2 transtracheal aspirates, 6 pleural fluid specimens, and 2 pus specimens, were tested. All four culture-positive samples, of which two were bronchial aspirates, one was pleural fluid, and one was pus, were positive by PCR. Among 35 culture-negative samples, 2 bronchial aspirates were positive by PCR."

vrs

The materials were obtained by percutaneous aspiration of lung and heart immediately after death or directly from lung and heart after opening the thorax at autopsy. Anaerobic bacteriologic methods were employed for isolation. Twenty-one strains of anaerobic bacteria were recovered (3.4% of the 603 total isolates) from the lungs for 368 cases, and 16 strains (6.5%) of anaerobic bacteria were isolated among 247 strains from heart blood for 314 cases. Bacteroides fragilis was the most frequently isolated anaerobe from both the lungs and heart blood.

Diverticular or other gastrointestinal disease is unlikely on clinical grounds in this patient.

And he got all better on a beta-lactam and metronidazole. For a variety of reasons, we will give him a course of antibiotics and see how he does.

As to names, he introduced a female in the room as "my" friend. Not "a" friend---"my" friend, with an odd emphasis on my. That is how my Dad would introduce his girlfriend. At what point do adults stop referring to others as boyfriend/girlfriend? "My," friend doesn't really tell the nature of the relationship---how much financial, emotional, and physical support will be available. With husband and wife, at least there is a general idea.

The permutations of human relationships and commitment are far greater than our language and legal system can describe. There needs to be a box on the Physician Orders for Life-Sustaining Treatment (POLST) form for relationships so I can quickly ascertain what "my" friend is. "She/he is a type 1a companion."

I learned long ago, after presuming that the female in the room was a daughter when she was a wife, not to assume anything about the relationships of people in a room with the patient. I only wish we had better names for them.

Rationalization

J Clin Microbiol. 1994 Mar;32(3):679-83. Detection of Bacteroides fragilis in clinical specimens by PCR.

http://www.ncbi.nlm.nih.gov/pubmed/8195378

Rev Infect Dis. 1984 Mar-Apr;6 Suppl 1:S128-31. Pulmonary infection due to anaerobes in a hospital autopsy survey.

http://www.ncbi.nlm.nih.gov/pubmed/6718932

The bacteriology of aspiration pneumonia John G. Bartlett, M.D., Sherwood L. Gorbach, M.D. Sydney M. Finegold, M.D.

http://www.sciencedirect.com/science/article/pii/0002934374905981

POLL RESULTS

My preferred term is

• "my" friend. 4%

• spouse. 16%

• partner. 14%

• what's his/her name. 27%

• my reason for living. 14%

• Other Answers 25%

  • Beloved

  • the person with whom i have shared 12 years of wedded bliss in 44 years of marriage. (10 years of bliss for him, 2 for me)

  • My sun and stars

  • husband

  • the Grand Poo-Bah

  • She who must be obeyed.

  • That guy

  • sperm donor

  • $HERSELF

  • my better half

  • sweetie pie

Chapter Seventy-One

In Case After Case

"IN MY EXPERIENCE"... Once

"IN CASE AFTER CASE"... Twice

"IN A SERIES OF CASES"... Thrice

Once upon a time, I was taught that gram-negative rods do not, as a rule, cause cellulitis in ostensibly healthy people without necrosis.

Now? I am not so sure.

I saw a case of Serratia cellulitis years ago in a severe vasculopathy. It happens.

In 2010 I wrote about a case of recurrent Pseudomonas bacteremic cellulitis in a patient with morbid obesity. Best as I could tell at the time, there was no such thing as Pseudomonas cellulitis.

Last year another case was admitted to the hospital: bacteremic Pseudomonas cellulitis. No reason found; no necrosis.

And now a third. Like the other two, the risk was severe chronic lymphedema and vasculopathy. No abscess, no necrosis, just rubor, dolor, calor, tumor of a leg and bacteremia and they all got better very quickly on an anti-pseudomonal antibiotic. The other striking thing was how relatively non-septic they were.

It is generally believed that in my experience that it can clearly be shown in case after case a definite trend is evident: Pseudomonas can cause a bacteremic non-necrotic cellulitis.

And there is still nothing to support this on PubMed.

Proof omitted.

Rationalization

Infectious Diseases in Clinical Practice: December 2002 - Volume 11 - Issue 9 - pp 550-554 Case Report Serratia marcescens Cellulitis: A Case Report and Review of the Literature

[http://journals.lww.com/infectdis/Fulltext/2002/12000/Serratia_marcescens_Cellulitis__A_Case_Report_and.5.aspx]{.underline}

• Milliways

[http://boards.medscape.com/forums?128@@.29fa5116!comment=1]{.underline}

Scientific Jargon --- Humor

https://www.smart-jokes.org/research-phrases-meaning.html

Chapter Seventy-Two

More is not always better

There are diseases of which I am seeing more in patients in whom I do not expect it. Ick. What a sentence. Normally I would delete it and start again, but I thought you might like to see the first thought put to pixel.

I am increasingly seeing some diseases in patient populations that classically are not considered at risk. More testing? Better testing? A real increase in disease? I am not certain.

Today's patient is an elderly-ish female receiving TAC for breast cancer. TAC, from the inconsistent acronyming of oncologists (mixing generic and brand names so they can come up with a catchy title. They need to either pick generic or brand, but stop the mix and match.) is Taxotere (docetaxel), adriamycin (doxorubicin), and cyclophosphamide.

She is on the third cycle and not tolerating it very well. This time she finishes the chemo and, while not neutropenic, has progressive shortness of breath, a non-productive cough, and fevers. Labs are unimpressive except for a lactate dehydrogenase level of 365 U/L, somewhat elevated over the upper limit of normal at 280 U/L, which indicates tissue damage of some kind. A chest X-ray shows bilateral patchy infiltrates.

I am betting drug lung. Only two pulmonary processes give you this clinical pattern: pulmonary fibrosis and Pneumocystis jirovecii pneumonia (PJP), and she does not have any risk for PJP. No HIV, no steroids, no hematologic malignancy, no protein malnutrition, no lymphocytopenia.

Wrong.

Bronchoscopy shows PJP.

There are a smattering of cases of PJP associated with breast cancer chemotherapy, but the patients have usually been on steroids, a classic risk factor.

As best as I can determine, there are two prior cases of PJP with only TAC as a risk. So it is rare indeed.

I am not the only one who is seeing, or thinks I am seeing, more PJP. So are the English:

In this study, we found an increasing trend in rates for clinical cases recorded in HES and microbiologically confirmed and reported cases in England during 2000--2010. This finding suggests a real increase in the numbers of cases of P. jirovecii pneumonia diagnosed. We also found an association between P. jirovecii infection and a variety of chronic lung diseases not described in the literature as being associated with P. jirovecii infection.

I am no longer surprised when we isolate PJP in patients with atypical risks. It seems to be lurking around the edges of late.

Rationalization

J Oncol Pharm Pract. 2012 Jun;18(2):311-5. doi: 10.1177/1078155211429384. Epub 2012 Jan 4. Chemotherapy-induced infiltrative pneumonitis cases in breast cancer patients.

http://www.ncbi.nlm.nih.gov/pubmed/22217649

Pneumocystis Carinii Pneumonia During Dose-Dense Chemotherapy for Breast Cancer

http://jco.ascopubs.org/content/24/33/5330.full.pdf

Increasing Pneumocystis Pneumonia, England, UK, 2000--2010. Volume 19, Number 3---March 2013

http://wwwnc.cdc.gov/eid/article/19/3/12-1151_intro.htm

Rev Pneumol Clin. 2011 Sep;67(4):191-8. doi: 10.1016/j.pneumo.2011.06.001. Epub 2011 Aug 5. [Pneumocystis jirovecii pneumonia in non-HIV infected patients: a study of 41 cases].

http://www.ncbi.nlm.nih.gov/pubmed/21920277

Clin Microbiol Rev. 2012 April; 25(2): 297--317. doi: 10.1128/CMR.00013-12 PMCID: PMC3346301 Colonization by Pneumocystis jirovecii and Its Role in Disease

http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3346301/

Chapter Seventy-Three

No Problem

I was seeing a patient today, did the usual aich and pea, talked with the patient about the plan, and left the room.

As I left the room, I foamed. Or tried to. I am a compulsive foamer at work. I foam upon entering the room, before I examine the patient, after I examine the patient (and I add the stethoscope), and as I leave the room.

Four times with each patient. Fortunately, I do not get dermatitis from the product we have. This time, as I accessed the foam, the canister was empty.

As I have written about before, we have been rolling out a different way of doing hand hygiene. The minimum requirement is to foam as you walk in and foam as you walk out of the patient room. Period. Every time. If you are witnessed not foaming you are reminded in real time by the hand hygiene monitor. I think it is unethical to watch people fail in a key infection control intervention without intervening. Not doing hand hygiene could spread infection and potentially harm and/or kill someone.

When you think about it, that is just what our regulatory bodies do: The Joint, the State, OHSA. Over the years, I have been a party to multiple reviews by these organizations, and evidently, they are specifically told not to intervene if they witness a failure of proper care. Interesting. I really wonder about the ethics of not intervening. As an example, at one of my many hospitals, the observer watched the anesthesiologist not "scrub the hub" three times before injecting medications and did nothing. That is the kind of culture we are striving to avoid, and our regulatory bodies need to do the same and stop being complicit in patient harm.

As part of the rollout of hand hygiene, I have gone from physician group to physician group telling them of the new hand hygiene program. I warn them that they are going to be asked to foam and it is the expectation that when asked to foam they will smile, say thank you and foam. No whinging allowed.

It is working quite well. In the first six months at one hospital, we have asked about 1 in 6 to foam, everyone has been nice and infection rates--which had been flat--fell by another 2.1 a month with an estimated cost savings of $425,000.

There are two consistent comments at the physician groups. The first is that foam is not always convenient. I do not think that is true, but apparently for some docs convenient means always having foam at arm's length. Can't walk a few feet, no sir. I remember in the days before foam a doc saying the reasons she could not wash her hands every time is that the sink was on the other side of the patient room. Really. It turns out that the sinks were a mile away, uphill, through snow and then she had to walk barefoot over broken glass to reach it.

The other comment is that sometimes they go to use the foam and, like today for me, the canister is empty. I imagine them sitting on the toilet and realizing there is no toilet paper. They probably sit there, baffled by what to do next.

Here is my recommendation. Show a wee bit of initiative. Take the empty canister to the nursing station. Ask for a new one. Put it in the wall mount. Then foam. Took me less than a minute.

How hard is it?

POLL RESULTS

When I note there is no alcohol or TP in the dispenser, I

• ignore it and more on 0%

• yell for someone to take care of it 8%

• take care of the problem myself 28%

• I'm a doctor not a janitor, damn it. ~ Bones McCoy 19%

• use soap and water 42%

• Other Answers 3%

  • Use the 100-proof whiskey in my hip flask.

  • tell housekeeping

Chapter Seventy-Four

Rendering

I often get called to render an opinion on what to do about a culture. Sometimes this is not unlike rendering animals. Is the organism real, i.e., a cause of disease, or is it a contaminant?

Sometimes the answer is easy: diptheroids growing on day 5 after a lumbar puncture. Ignore it. 1 of 6 cultures show coagulase-negative Staphylococcus? Probably not the real deal. How about one of three blood cultures with alpha Streptococcus? Probably a real bacteremia, and probably not clinically important. To have streptococcal bacteremia is likely part of the human condition, at least for those who have teeth and brush them, and occasionally we catch it on blood cultures.

Some cases are not so straightforward.

Two patients in the last week presented from the community with fevers, chills, and altered mental status. One had 1/4 blood cultures with E. coli. The other had 1/4 with Proteus. Source? Completely negative. No urinary tract infection, no intra-abdominal pathology on CT, no line, no nothing.

I get one or two of these kinds of cases a year: a patient with a gram negative bacteremia and no source. They respond quickly to antibiotics, and the best I have to offer is a Gallic shrug. Can't tell you why, just pick an antibiotic, count the number of fingers on both hands for a duration and see how they do. To date they have all done well, but I would really like to have a reason or a source for what is clinically a real bacteremia. I can't ignore it. I presume they have all had a reversible/transient translocation from the bowel. Perhaps a self-sealing puncture, who knows.

Sometimes I get asked to render an opinion on bugs that no one but me knows about, as evidently, I am the only one with access to the Googles, my secret weapon.

Today's patient is on Imuran (azathioprine, an immunosuppressant) and prednisone for a collagen vascular disease, and is admitted with pneumonia and maybe/maybe not aspiration. A bronchoscopy shows normal flora and greater than 100,000 Corynebacterium per milliliter. What, I am asked, do you make of the latter?

Depends. I ask the lab to identify it further and it is Corynebacterium pseudodiphtheriticum. Now that's a lung pathogen:

The conclusions are that non-diphtheriae Corynebacterium species are an emerging cause of respiratory infection among patients with chronic respiratory disease and/or immunosuppression, and cannot always be considered as mere colonisers. The microorganism's predominance in Gram-stained purulent respiratory samples together with abundant growth in the culture is the key for the microbiological diagnosis.

We (meaning me) treated with Vancomycin until it came back pan-sensitive, and I changed him to levofloxacin (he had multiple allergies).

Real pathogen, colonizer, both? I can render. I can convert cultures into stable, value-added material.

Rationalization

Clin Microbiol Infect. 2003 Aug;9(8):793-802. Source of infection and other factors associated with case fatality in community-acquired bacteremia--a Danish population-based cohort study from 1992 to 1997.

http://www.ncbi.nlm.nih.gov/pubmed/14616699

Crit Care. 2011;15(6):R267. doi: 10.1186/cc10545. Epub 2011 Nov 15. Duration of antibiotic therapy for bacteremia: a systematic review and meta-analysis.

http://www.ncbi.nlm.nih.gov/pubmed/22085732

Eur J Clin Microbiol Infect Dis. 2012 Dec 28. [Epub ahead of print] Non-diphtheriae Corynebacterium species: an emerging respiratory pathogen.

http://www.ncbi.nlm.nih.gov/pubmed/23271676

Chapter Seventy-Five

Cheeky

We don't get many ventilator infections. Or surgical wound infections. Or central line infections. With the application of the various bundles, we have driven hospital-acquired infections into the dirt--and with them my practice. Work volumes are down about 60% for me over the last decade, a consistent decline year to year. In case anyone wonders if the decline in hospital-acquired infections is real, I always say I have the W-2s to prove it.

So when we get a line infection, I am surprised. Doubly so when it is on the oncology floor where we have the most meticulous nurses I have ever seen. And triply surprised when it is a yeast, C. albicans.

The patient is a homeless heroin user, and she has been in the hospital for two weeks being treated for an extensive soft tissue infection that was not a necrotizing infection, although we did not know it at the time. It had progressed fairly rapidly, and given the patient's toxicity, her doctors were worried about Group A Streptococcus. It was a severe S. equisimilis infection, but she received a few fasciotomy incisions when she was septic.

She was receiving a course of antibiotics when she spiked a temperature and the blood cultures were obtained. It was a one-time fever; repeat cultures were negative but we pulled the midline, and it grew Candida as well.

Not quite typical for a Candida line infection, but close enough; she had the risks of antibiotics, central line, and recent surgery.

Then the residents told me she was diverting her pain meds and collecting them in a rubber glove she kept under the covers, between her legs. There was a little altercation when it was discovered as she did not want to either admit to the stash or give it up.

Aha.

Years ago I had a patient who was cheeking his pain meds, saving them up and injecting them, and he developed Candida discitis. I bet the Candida came from his own mouth oral flora.

So, I asked, is she doing the same?

We asked the patient and the response was an artful reply that neither confirmed nor denied the practice. She really should run for president.

There are no infections reported on either the PubMeds or the Googles from cheeking medications. And maybe the Candida was picked up from the environment; she was keeping the glove near the groin. It probably happens more often than realized--you just need to remember the extent to which people will go to get their narcotics IV.

I suppose there are worse cheeks in which you could hide your meds.

Rationalization

An Unusual Case of Cheeking

http://www.jailmedicine.com/an-unusual-case-of-cheeking/

Chapter Seventy-Six

That's Just Swell

The patient is admitted with 25 hours of fevers and chills.

He has insulin-dependent diabetes mellitus and a foot ulcer that is a likely source for the soon-to- be-diagnosed Group B streptococcal bacteremia.

And he is on immunosuppressive drugs: methotrexate and Remicade (infliximab) for psoriatic arthritis.

And he had Hodgkin's lymphoma cured in his youth.

And he has known severe disc disease, with a big lumbar extrusion and bad back pain that is now much worse.

After three days the fever does not resolve. A workup shows no discitis, and there are no vegetations on the transesophageal echocardiogram. After 4 days the fevers resolve, but the back pain does not.

A week into the course of IV antibiotics, I repeat the MRI and now it shows discitis and a bit of new white in the old disc bulge. There is edema in the bulge that was not there before. The radiologists yammer on about T1- and T2-weighted images, but I look for "white where it should be black" and "black where it should be white."

Radiology lets me know it is not an abscess, but does not know what to make of the edema in the bulge.

I assume that the initial MRI was a false negative due to his immunomodulatory medications and extend his course of IV antibiotics. I have seen delays in the MRI in patients who are slow to make an inflammatory response. But besides the typical pain of discitis, he is without change.

And then five days later he is admitted for cauda equina syndrome. Cauda equina syndrome is when the bundle of nerves below the end of the spinal cord, called the cauda equina (horse's tail), is injured. Patients have low back pain, pain that radiates down the leg, numbness around the anus, and loss of bowel or bladder function. He has urgent surgery where all this find is the big, bulging disc. No epidural abscess, just discitis. Whew.

I think what happened is that the discitis led to swelling of the disc and increased its bulge into the spinal canal. There was no wiggle room in the canal, and so he became symptomatic.

There are no cases on the PubMeds that I can find of discitis in a preexisting herniation that progressed with signs of obstruction not due to abscess, but from the tumor from infection. Tumor as in rubor, dolor, calor, tumor.

Post op he is doing quite well. Whew again.

Rationalization

Enferm Infecc Microbiol Clin. 2005 Feb;23(2):71-5. [Spondylodiscitis and sacroiliitis due to Streptococcus agalactiae in adults: clinical case and literature review].

http://www.ncbi.nlm.nih.gov/pubmed/15743577

Chapter Seventy-Seven

Scary Bug

Gram-positive rods in the blood are usually a contaminant, a Bacillus or Diphtheroids that grow out on day 5.

4 out of 4 sets of anaerobic blood cultures all with gram-positive rods that grow in 12 hours does tend to get one's attention.

The patient is an elderly male who had his prostate microwaved three weeks ago. Besides some urinary issues, he has otherwise been stable when he had the abrupt onset of fevers, rigors, altered mental status, and mid-thoracic back pain.

He came to the hospital and was admitted with SIRS (systemic inflammatory response syndrome). He had a little leukocytosis but little else on the labs, and by next morning his blood cultures were all growing Clostridium perfringens.

Unlike the patient who had massive hemolysis and died several weeks back, he looked ill but not trying-to-die ill. Besides the severe back pain, he had no other focal complaints.

The C. perfringens had to be coming from somewhere, so we went looking.

The spine was clean on MRI.

The CT scan showed a biliary system filled with air (a longstanding problem since his cholecystectomy years ago). His prostate was boggy, but the radiologists tell me the prostate is a hard organ to assess on CT. And there was hepatic vein thrombosis.

The patient, who was penicillin allergic, did well on clindamycin and metronidazole. I always have a touch of magical thinking with C. perfringens. I cannot do susceptibility testing, and there are penicillin and clindamycin resistance, so I always add oral metronidazole. Belt and suspenders.

Was the Clostridia from the prostate? The biliary tract? And what about the clot? No answer.

We, meaning I, elected to treat him for 4 weeks with antibiotics and anticoagulation, and I remain very unsatisfied.

My partner has had a pair of Clostridial gas gangrenes this week in black tar heroin users, one of whom died with typical Clostridial rapidity and one of whom has Clostridium sordellii. Scary bugs.

Rationalization

J Infect. 2003 May;46(4):253-5. Clostridium perfringens septicemia with thrombophlebitis of the portal vein.

http://www.ncbi.nlm.nih.gov/pubmed/12799155

Anaerobe. 2012 Apr;18(2):254-9. doi: 10.1016/j.anaerobe.2011.11.001. Epub 2011 Nov 20. Life-threatening clostridial infections.

http://www.ncbi.nlm.nih.gov/pubmed/22120198

Int J Infect Dis. 2013 Apr 8. pii: S1201-9712(13)00113-6. doi: 10.1016/j.ijid.2013.03.001. [Epub ahead of print] Clinical significance and outcomes of Clostridium perfringens bacteremia-a 10-year experience at a tertiary care hospital.

http://www.ncbi.nlm.nih.gov/pubmed/23578849

POLL RESULTS

The bug I most fear is

• Clostridium perfringens 21%

• XDRTB 27%

• GC. Er, honey, I need to tell you something.... 10%

• Zorak 6%

• A fly with a human head. Help meeeeeeeeeee. 22%

• Other Answers 13%

  • The mosquito

  • Spiders

  • all MDRO

Chapter Seventy-Eight

Clot Happens

It still amazes me how you, yes you, can be living the vida loca, no problems, no risks, when, wham, out of nowhere you can get a severe infection.

I have this ongoing sense that infections are a sword of Damocles hanging over everyone's head by a very frayed rope. And you do not want to be a frayed.

The patient is a healthy male, middle-aged (by my standards) who has fevers and teeth-chattering rigors, but is otherwise non-focal.

Most infections are infections of something--some organ or other is involved--and as a result, there are symptoms in that organ. He has none. Symptoms, not organs.

He goes to the ER, where the white blood cell count is found to be normal but contains 51% bands---an immature form of neutrophils seen in infection and some other conditions. The comprehensive metabolic panel shows a minimally elevated alkaline phosphatase, a marker of liver function. A large bandemia--a term I hate, I would say bandosity--should always give one pause:

"Even with normal total white blood cells, patients with moderate and high bandemia (moderate (11%-19%), or high (≥20%)) on admission had significantly increased odds of having positive cultures, including blood cultures, and of in-hospital mortality."

After the basics are ruled out, he has a CT scan looking for pus and they find what looks to be infected clot in the portal vein, pylephlebitis, and they call me.

I bet it will grow Fusobacterium. It doesn't. At day three the blood grows Streptococcus salivarius. Is this the etiology of the presumptively infected clot? Could be.

Pylephlebitis

"occurs most commonly in association with diverticulitis and appendicitis, and in most cases more than 1 type of bacterial organism is found. The most common isolated microbes include Bacteroides, Enterococcus, Escherichia coli, Klebsiella, Staphylococcus, and Streptococcus. Treatment includes empiric therapy with broad-spectrum antibiotics, until cultures are available, with total treatment duration of 4 to 6 weeks.

An associated hypercoagulable condition, malignancy, or acquired immune deficiency syndrome is not an uncommon finding. Anticoagulation therapy may reduce the incidence of septic embolization to the liver from infected portal thrombi and may prevent liver abscesses. The optimum duration of anticoagulation therapy is unclear."

He has none of the above risks and is slowing improving on antibiotics and anticoagulation. I have seen two infected abdominal clots and two C. perfringens the last two weeks. Do diseases come in twos? Or threes? If the latter you should be very afraid.

Rationalization

West J Emerg Med. 2011 Nov;12(4):575-6. doi: 10.5811/westjem.2011.1.2197. Pylephlebitis: an uncommon complication of intra-abdominal infection. Hagopian T, Zuniga F, Surani SR.

http://www.ncbi.nlm.nih.gov/pubmed/22224165

BMJ Case Rep. 2011 Nov 15;2011. pii: bcr0720114527. doi: 10.1136/bcr.07.2011.4527. Fusobacterium necrophorum--beyond Lemierres syndrome. Shahani L, Khardori N.

http://www.ncbi.nlm.nih.gov/pubmed/22674593

Am J Med. 2012 Nov;125(11):1124.e9-1124.e15. doi: 10.1016/j.amjmed.2012.04.039. Epub 2012 Aug 28. Bandemia with normal white blood cell counts associated with infection.

http://www.ncbi.nlm.nih.gov/pubmed/22939096

POLL RESULTS

My least favorite medical word(s):

• bandemia 5%

• dehydration 2%

• SIRS 7%

• strong, big gun and powerful 56%

• denied 24%

• Other Answers 5%

  • metastases

  • Compromise.

Chapter Seventy-Nine

Killing Emotions?

Different diseases lead to the same result.

The first patient has Streptococcus viridans in a prosthetic valve and is getting better on therapy when, 10 days in, she has the abrupt onset of left upper quadrant pain.

I have seen this many times over the years. One day you have a vegetation, the next day you do not and your spleen is infarcted. Emboli, especially late emboli, are what makes endocarditis a nerve-wracking disease. I had one patient who had a fatal cerebral embolus on the last day of therapy. Very rare, but it is what keeps me up at night.

The CT confirmed the clinical diagnosis with a large wedge-shaped defect in the spleen. Hopefully, it will neither rupture nor abscess.

The other patient has myelofibrosis and a positively enormous spleen. Bigger than the liver. He has the new onset of increased splenic pain and fevers. Evaluation for an infectious source of fevers was negative but for the first time ever he had Howell-Jolly bodies on the complete blood count. Howell-Jolly bodies are leftover nuclear remnants in red blood cells that are usually removed in the spleen. They occur where there is a non-functioning spleen. I called splenic infarctions as a cause of fevers.

"We identified 26 cases with a mean age of 52 years. Common causes were hematologic malignancy (six cases) and intracardiac thrombus (five cases). Only three cases were associated with bacterial endocarditis. In 21 cases the splenic infarction brought a previously undiagnosed underlying disease to attention. Only half the subjects complained of localized left-sided abdominal pain, 36% had left-sided abdominal tenderness; 31% had no signs or symptoms localized to the splenic area, 36% had fever, 56% had leukocytosis and 71% had elevated lactate dehydrogenase levels."

We did not get a CT to confirm the diagnosis, although heme-onc agreed with me. Good enough, especially as myelofibrosis is a risk for spontaneous splenic infarction.

In the old days, the spleen was thought to be the seat of emotions and to be "splenic" was to be melancholy. I wonder if the infarct would have been thought a treatment for depression? I'll stick with SSRIs.

Rationalization

Isr Med Assoc J. 2010 Jun;12(6):362-5. Splenic infarction: an update on William Osler's observations.

http://www.ncbi.nlm.nih.gov/pubmed/20928991

Clin Nucl Med. 1983 Aug;8(8):335-6. Radiocolloid redistribution and multiple splenic infarcts in myelofibrosis.

http://www.ncbi.nlm.nih.gov/pubmed/6627802

Chapter Eighty

I Hate Having to be Clever

I hate health care in this country. I have no idea what to do to make the health care system better. Personally, I doubt there is a good solution, just the least bad solution. To my mind, the least bad solution would be a single-payer system, so that instead of being screwed up in uncountable ways, health care would just be screwed in one way. One consistent FUBAR is better than multiple inconsistent FUBARs.

All I want to do is make a diagnosis, get the patient on therapy, and get them better. Sometimes I have to try and be clever in the process, and I am uncomfortable being clever, at least when it comes to therapeutic interventions.

The patient is on Medicare, diabetic with triopathy---the triad of neuropathy, retinopathy, and nephropathy associated with diabetes. She is a mother of 2 young children. She needs to be home to take care of her kids.

She gets a complicated foot infection that is debrided, but the podiatrist tells me he is worried they may be some residual bone infection. Further debridement will lead to loss of function. From a structural point of view it is between an attempt at salvage with antibiotics and an amputation.

The problem is which antibiotics to give. Bone cultures show methicillin-resistant S. aureus (MRSA) and a gram-negative bacterium that is only susceptible to carbapenems. She needs either a once-a-day antibiotic, so she can get it paid for by coming into the hospital, or she has to go to a nursing home--the latter not a good option with the aforementioned children. Home antibiotics are out of the question as Medicare does not pay for them.

The MRSA I can deal with: daptomycin. The gram negative? That is where I get clever, and as I told the patient, clever may not cut it for curing her infection.

With her slightly decreased glomerular filtration rate, maybe we can turn meropenem into a once-a-day drug with the addition of probenecid. Maybe. I told her it may not work, and the best bet was going to a nursing home for standard therapy, but the nursing home could not take her kids.

Most of my use of probenecid has been through the ER, where I suggest it combined with cefazolin as an inexpensive way to treat cellulitis.

Three weeks in it seems to be working, but I am not satisfied with the solution.

Postscript

It worked.

Rationalization

J Antimicrob Chemother. 1989 Sep;24 Suppl A:311-20. The pharmacokinetics of meropenem in volunteers.

http://www.ncbi.nlm.nih.gov/pubmed/2808215

Clin Infect Dis. 2002 Jun 1;34(11):1440-8. Epub 2002 May 7. Once-daily intravenous cefazolin plus oral probenecid is equivalent to once-daily intravenous ceftriaxone plus oral placebo for the treatment of moderate-to-severe cellulitis in adults.

http://www.ncbi.nlm.nih.gov/pubmed/12015689

POLL RESULTS

I want

• a good solution 27%

• a just solution 10%

• a cost-effective solution 8%

• the least bad solution 33%

• a solution that is best for me, everyone else be damned 6%

• Other Answers 16%

  • a good, just, cost-effective solution.

  • A solution that's not in the bottom quintile

  • A solution provided by one who is always right and never lies. Where is George Leroy Tirebiter when we need him?

  • Home IV therapy and a wound vac

  • A good precipitate.

  • a 7 percent solution.

Chapter Eighty-One

Ramble On

Weekend call. It is not so bad. The nice thing about ID is they rarely call me to see trivial cases, so I keep entertained. Every institution has an electronic medical record (EMR)--I have three I have to navigate, and being the covering doctor I have to rely on the medical record to ascertain what is going on with the patient. Good luck. What really struck me this weekend is the death of the clinical narrative.

The medical record is a hodgepodge of cut & paste and automatic importation, each note organized to suit the specific doctors' psychopathology and designed to maximize billing but not the dissemination of information. I miss the old simple subjective, objective, assessment, and plan (SOAP) note.

I also note that as the EMR has become ubiquitous, younger physicians do not have the ability to verbally communicate in a linear, comprehensive way. When referring docs call for a consult, they give the case in what appears to be a random recalling of the facts: history, labs, and physical mentioned in as disorganized a manner as possible. There is no ID consult that can't be summarized in a sentence limited to five words, and I make the interns tell me their question in five words. They usually start with the "the patient is a four-five year old..." and I cut them off. You reached your limit.

If you can't summarize your question in five words, you have no understanding of the case. Often two words will do: Why fever? Or Which antibiotic?

People have even mastered how to ramble in a text message, which is quite the accomplishment given the 150-character limit. I like large swaths of the EMR, just not the progress notes.

I'm fortunate in that my job is easy: me find bug, me kill bug, me go home. But I pity anyone who wants to get an easy narrative from the EMR.

One of my cases from several weeks ago got readmitted this weekend. An older male with urinary retention who self-catheterizes initially came in with a lumbar epidural abscess/ discitis/ osteomyelitis.

It was an odd bug: Enterococcus. Fewer than a dozen cases on the PubMeds. I presume it got to the spine by way of Batson's plexus, as I always do when there is a lumbosacral spine discitis/osteomyelitis/epidural abscess with a urinary pathogen. Especially since there was no endocarditis as best we can determine, given the issues of proving a negative.

I treated him with ampicillin and gentamicin and two weeks later he is no better. His pain is unchanged. Is he just slow to get better? Or is it a failure of antibiotics? Given the inability of gentamicin to penetrate into and work in bone, maybe he is failing what might be one drug therapy.

So I opted to change the gentamicin to ceftriaxone. Most of the favorable data is for endocarditis, but there are reasonable pharmacokinetics reasons to suspect it would be better for bone.

I hope I explained my reasoning in the EMR with a coherent narrative.

Postscript

And he got better.

Rationalization

Case Rep Med. 2013;2013:513920. doi: 10.1155/2013/513920. Epub 2013 Mar

20. Insidious Onset of Tetraparesis due to Cervical Epidural Abscess from Enterococcus faecalis.

http://www.ncbi.nlm.nih.gov/pubmed/23573096

Surg Neurol. 2008 Feb;69(2):121-5. Epub 2007 Aug 27. Infective endocarditis associated with spondylodiscitis and frequent secondary epidural abscess.

http://www.ncbi.nlm.nih.gov/pubmed/17720227

Antimicrob Agents Chemother. 2009 Oct;53(10):4305-10. doi: 10.1128/AAC.00444-09. Epub 2009 Aug 10. Pilot study of ampicillin-ceftriaxone combination for treatment of orthopedic infections due to Enterococcus faecalis.

http://www.ncbi.nlm.nih.gov/pubmed/19667290

POLL RESULTS

The best part of the EMR

• the progress notes 6%

• charting from anywhere 23%

• never needing to go to radiology to see films 31%

• improved billing 3%

• the opportunity to endlessly rewrite my notes 26%

• Other Answers 11%

  • it may be rambling and incoherent, but at least I can read what they wrote.

  • SAVE A TREE, NOT LIKELY

  • legibility!

  • I can read what others have written - no more calling consultants: "what the.....?"

Chapter Eighty-Two

Eventually Every Bug Show Up Everywhere

Common things are common. In my practice, uncommon things are common, and I often end up saying, "Huh. Never saw that bug there before." But if you are in practice long enough, by random chance, you will eventually see every organism infecting every organ. Hopefully not all at once.

Today's patient is admitted with fevers and altered mental status. In the last year, she had some major head trauma that resulted in meningitis. They worked her up for a cerebrospinal fluid (CSF) leak, found nothing, and the patient went back to life--including IV drug abuse.

Now she is in the ICU on a vent with an impressive spinal tap: 18,000 polymorphonuclear leukocytes (PMNs) per microliter when anything over a few is abnormal, glucose very low at < 5 mg/dL, protein highly elevated at 750 mg/dL. Gram stain negative, which is odd since all of these values scream bacterial meningitis. Meningococcus? Sometime gram negatives are missed against a sea of white blood cell pink, and it is an organism that is excellent at inciting an inflammatory response.

But it is her second case of meningitis, so I am betting on Streptococcus pneumoniae and her doctors just missed the CSF leak. Maybe endocarditis, although nothing on exam to suggest it. By 12 hours the blood and CSF are growing gram-positive cocci in chains. The lab tells me it doesn't look like S. pneumoniae. Not Lancet-shaped, looks more like the NEJM.

I'm not so certain now. Some kind of Streptococcus. Group A? She seems too ill for an alpha-hemolytic strep. 18 hours later it is identified as a Group B Streptococcus.

Huh. Never saw that bug there before.

Meningitis with Streptococcus agalactiae is distinctly odd in the normal, non-partum (is that a word?) adult, but there are cases, although none due to a CSF leak. I still wonder if there is an issue with the integrity of the skull, but that evaluation will have to wait.

Postscript

No leak was found; the patient eventually discharged never to be seen again.

Rationalization

J Med Assoc Thai. 2002 Mar;85(3):385-7. Streptococcus agalactiae meningitis in adults: report of two cases.

http://www.ncbi.nlm.nih.gov/pubmed/12117031

BMC Infect Dis. 2011 May 25;11:149. doi: 10.1186/1471-2334-11-149. Streptococcus agalactiae in adults at Chiang Mai University Hospital: a retrospective study.

http://www.ncbi.nlm.nih.gov/pubmed/21612629

Chapter Eighty-Three

The Great Race

Which bug will be the first to be both widespread and resistant to all antibiotics? I am betting Neisseria gonorrhoeae, although it is an organism that shows up in neither my personal or professional life. Acute STDs are not part of a hospital-based practice, and I have not seen a disseminated gonorrhea this century. I wonder where it went.

There have been four carbapenem-resistant Enterobacteriaceae is my various hospitals in the last week. Two E. coli, a Klebsiella and a Proteus. Hate to see a whole class of bacteria become pan-resistant, especially one that lives in the colon. But at the end of the day all we are is sentient transport for bacteria and their resistance genes.

And now I have a S. aureus that I should be able to kill but may have difficulty.

The patient has methicillin-sensitive S. aureus mitral valve endocarditis (not an IV drug abuser, thank you very much, just unlucky), started on nafcillin at another institution and admitted to one of my hospitals with acute interstitial nephritis.

Get this. While this S. aureus is methicillin-sensitive, its minimum inhibitory concentration (MIC) of vancomycin is 2.0 micrograms per milliliter---borderline high. That means that not only is the nafcillin toxic, it may be less effective. There is more to an increased vancomycin MIC than just vancomycin resistance:

"We reported an association between elevated vancomycin MIC and 30-day mortality in patients with Staphylococcus aureus bacteraemia (SAB), including patients with methicillin-susceptible S. aureus (MSSA) treated with flucloxacillin. A detailed analysis of comorbidities and disease severity scores in the same cohort of patients was performed to ascertain if unknown clinical parameters may have influenced these results. The association between elevated vancomycin MIC and 30-day mortality in SAB remained significant (p 0.001) on multivariable logistic regression analysis even when accounting for clinical factors. In addition, the association persisted when restricting analysis to patients with MSSA bacteraemia treated with flucloxacillin. This suggests that elevated vancomycin MIC is associated with but not causally linked to an organism factor that is responsible for increased mortality."

Now what? Daptomycin, 8 mg/kg a day, dosed from renal failure:

"The clinical success rate in patients treated with ≥ 8 mg/kg daptomycin was 90% [n=72 (RIE, 91%; LIE, 89%)]."

It is sensitive. Or is it? He has been infected for a while and that may not bode well:

"Among daptomycin-susceptible MRSA isolates from patients who had never received daptomycin, higher daptomycin MICs tracked with increased resistance to killing by platelet-derived but not neutrophil-derived HDPs. These findings support the notion that endogenous exposure of MRSA to specific HDPs may play a role in selecting strains with an intrinsically higher daptomycin MIC phenotype."

And what is endocarditis but a slug of infected clot on a valve?

Every time I treat a patient I am probably breeding resistance, contributing to the Great Race.

In the end the daptomycin cured the infection.

Rationalization

Clin Microbiol Infect. 2013 Feb 26. doi: 10.1111/1469-0691.12168. [Epub ahead of print] Vancomycin minimum inhibitory concentration, host comorbidities and mortality in Staphylococcus aureus bacteraemia.

http://www.ncbi.nlm.nih.gov/pubmed/23441652

J Infect Dis. 2011 Aug 1;204(3):340-7. doi: 10.1093/infdis/jir270. Antibiotic choice may not explain poorer outcomes in patients with Staphylococcus aureus bacteremia and high vancomycin minimum inhibitory concentrations.

http://www.ncbi.nlm.nih.gov/pubmed/21742831

J Infect Dis. 2012 Oct;206(8):1160-7. Epub 2012 Aug 16. Reduced susceptibility to host-defense cationic peptides and daptomycin coemerge in methicillin-resistant Staphylococcus aureus from daptomycin-naive bacteremic patients.

http://www.ncbi.nlm.nih.gov/pubmed?term=22904338

J Antimicrob Chemother. 2013 Apr;68(4):936-42. doi: 10.1093/jac/dks467. Epub 2012 Nov 28. Daptomycin for the treatment of infective endocarditis: results from a European registry.

http://www.ncbi.nlm.nih.gov/pubmed/23190763

POLL RESULTS

The first but that will be ubiquitous and totally resistant will be

• Gonococcus 25%

• S. aureus 11%

• E. coli 7%

• TB 25%

• Some unexpected bioterrorist bug stolen from a old decrepit Soviet lab 20%

• Other Answers 13%

  • Wasn't the cockroach the 1st bug resistant to everything? They're already ubiquitous.

  • the cockroachisteria

  • Paris Hilton

  • Acinetobacter

Chapter Eighty-Four

But they can obfuscate

Which two don't belong and why?

• Friends

• Family

• The ball

• Spouses

• Children

• Cultures

• Acquaintances

The ball and cultures. Cultures don't lie and ball don't lie. All the other others will lie, although the cultures can be obscure.

If you know what the typical habitat of a bacteria is and you find it outside of that habitat, like in the blood of the consult today, then you know where it should have originated. In theory at least.

A couple of times a year I see a patient who gets septic and has one or another Enterobacteriaceae in the blood: E. coli, Klebsiella or, in this case, Citrobacter. Totally non-focal on review of systems and exam. Nothing on labs to indicate a likely source.

If a Citrobacter is in the blood, there has some anatomical issue in the gastrointestinal tract. And there isn't. As is typical. Yet another elderly person with gram-negative sepsis for no good reason at all.

I presume, without data, that there was a small perforation of the colon that self-sealed. It has to happen now and then that a bone or other sharp object causes a small rent in the colon and it seals up on its own. Just look at sausage casings--they are not all that thick. These bowel tears do not always seal. I had a patient with a bread tag in the middle of his colonic abscess.

And now they want to know how long to treat, as if I know. Most antibiotic durations are multiples of 5 (five fingers on a hand) and 7 (7 days in a week) and based on very little data. Good thing we are not Babylonian; they had a base 60 counting system.

For many infections, the right answer is "Until they are all better," and that usually means the source is gone. He has no source to get better. Just because he is bacteremic (S. aureus and Candida not included) doesn't mean he needs a longer course of therapy:

The optimal duration of antibiotic therapy for bloodstream infections is unknown. Shorter durations of therapy have been demonstrated to be as effective as longer durations for many common infections; similar findings in bacteremia could enable hospitals to reduce antibiotic utilization, adverse events, resistance, and costs.

RESULTS: Twenty-four eligible trials were identified, including one trial focusing exclusively on bacteremia, zero in catheter-related bloodstream infection, three in intra-abdominal infection, six in pyelonephritis, thirteen in pneumonia and one in skin and soft tissue infection. Thirteen studies reported on 227 patients with bacteremia allocated to 'shorter' or 'longer' durations of treatment. Outcome data were available for 155 bacteremic patients: neonatal bacteremia (n = 66); intra-abdominal infection (40); pyelonephritis (9); and pneumonia (40). Among bacteremic patients receiving shorter (5-7 days) versus longer (7-21 days) antibiotic therapy, no significant difference was detected with respect to rates of clinical cure (45/52 versus 47/49, risk ratio 0.88, 95% confidence interval [CI] 0.77-1.01), microbiologic cure (28/28 versus 30/32, risk ratio 1.05, 95% CI 0.91-1.21), and survival (15/17 versus 26/29, risk ratio 0.97, 95% CI 0.76-1.23).

CONCLUSIONS: No significant differences in clinical cure, microbiologic cure and survival were detected among bacteremic patients receiving shorter versus longer duration antibiotic therapy. An adequately powered, randomized trial of bacteremic patients is needed to confirm these findings.

Just always give the most antibiotic for the shortest duration you can, and if you are uncertain, call your local ID doc to count fingers and days for you.

I will count the number of days in a week for this patient and recommend accordingly.

Rationalization

Crit Care. 2011;15(6):R267. doi: 10.1186/cc10545. Epub 2011 Nov 15. Duration of antibiotic therapy for bacteremia: a systematic review and meta-analysis.

[http://www.ncbi.nlm.nih.gov/pubmed/22085732]{.underline}

Ball Don't Lie

https://www.nytimes.com/2012/12/05/sports/basketball/rasheed-wallace-keeps-technicals-fouls-flowing-with-ball-dont-lie.html

Chapter Eighty-Five

One of the Bees in my Bonnet

Feel sorry for the housestaff that have to rotate on ID with me. I have the occasional idée fixe and they are a captive audience. So I rant.

One of the topics I like to bloviate about is the fact that infections are bad for people. There are two steams of evidence I like to talk about, the streams have yet to cross. That is a Ghostbusters allusion, in case you missed it.

In one stream, patients admitted with infections have higher mortality in the next year compared to patients with similar co-morbidities who do not have acute infections. Why they die at higher rates is not known; perhaps they have more fatal car accidents driving to the ID doctor.

In the other stream, we have vascular events. No matter what infection is evaluated (pneumonia, cystitis, shingles, influenza, even having your teeth cleaned) infections are associated with short- and long-term vascular events. Get infected and have a stroke, a heart attack, a pulmonary embolism or a transient ischemic event in the next month or more.

Infections are inflammatory, inflammation is prothrombotic, and thrombosis leads to clot in all the wrong places. It is quite an interesting literature, and if you are a fan of confirmation bias, you will soon recognize that many who are admitted with a vascular event have had a preceding infection.

I suspect that the reason people have increased death after infections is due to an increased number of vascular events, but that is an epidemiology that as yet to be completely elucidated.

As Mark Twain said, "Everybody complains about infection induced thrombosis, but nobody does anything about it." For the most part. The only intervention that we routinely do is influenza vaccination, which prevents many a heart attack and stroke.

Imagine my surprise when I came across what should be a key article. Why it was published in the Journal of No One Will Ever Find Me, rather than a more high-impact journal I do not know. Here is the whole abstract:

Coron Artery Dis. 2013 May;24(3):231-237. Does aspirin use prevent acute coronary syndrome in patients with pneumonia: multicenter prospective randomized trial.

*OBJECTIVES: The aim of this study was to test the hypothesis that aspirin would reduce the risk for acute coronary syndromes (ACSs) in patients with pneumonia.*

BACKGROUNDS: Pooled data suggest that pneumonia may trigger an ACS as a result of inflammatory reactions and the prothrombotic changes in patients with pneumonia. Hypothetically considering its antiaggregating and anti-inflammatory effects, aspirin might also be beneficial for the primary prevention of ACS in patients with pneumonia.

METHODS: One hundred and eighty-five patients with pneumonia who had more than one risk factor for cardiovascular disease were randomized to an aspirin group (n=91) or a control group (n=94). The patients in the aspirin group received 300 mg of aspirin daily for 1 month. ECGs were recorded on admission and 48 h and 30 days after admission to assess silent ischemia. The level of high-sensitivity cardiac troponin T was measured on admission and 48 h after admission. The primary endpoint was the development of ACS within 1 month. The secondary endpoints included cardiovascular death and death from any cause within 1 month.

RESULTS: The χ-test showed that the rates of ACS at 1 month were 1.1% (n=1) in the aspirin group and 10.6% (n=10) in the control group (relative risk, 0.103; 95% confidence interval 0.005-0.746; P=0.015). Aspirin therapy was associated with a 9% absolute reduction in the risk for ACS. There was no significant decrease in the risk of death from any cause (P=0.151), but the aspirin group had a decreased risk of cardiovascular death (risk reduction: 0.04, P=0.044).

CONCLUSION: This randomized open-label study shows that acetyl salicylic acid is beneficial in the reduction of ACS and cardiovascular mortality among patients with pneumonia."

Bad epidemiology math time. Let us say there are about 1.2 million cases of pneumonia a year in the US, which is a reasonable estimate. 120,000 may have had an acute coronary syndrome as a result. If they all took a month of aspirin, about 12,000 would have had an acute coronary syndrome.

This trial needs to be redone with larger numbers, to be double-blind and placebo controlled, and to look at other common infections as well.

Someday, when all the studies are done, I bet everyone with an acute infection will go home on a month of aspirin. Then I can retire that bee and move on to other occupants of the hive.

POLL RESULTS

My medical idée fixe is

• Infection-induced thrombosis 10%

• hand hygiene 50%

• better reimbursement 5%

• early retirement 17%

• finding an EMR that makes life easier 14%

• Other Answers 5%

  • single-payer health system

  • Better beer at lower prices

Chapter Eighty-Six

Schmear

I was asked to see an elderly male with a positive PPD skin test for TB and a positive QuantiFERON blood test. The nursing home sent him in for us to give him a clean mycobacterial bill of health.

He is in isolation, and I take a history: no signs of TB, no consumption. No risks for TB except for age. He is old enough to be from a time where TB was a common problem, so he could easily have been exposed in his youth.

His chest X-ray is unimpressive, and he has a CT scan. A small nodule in the right middle lobe and a small left pleural effusion. One sputum test for acid-fast bacilli is negative.

So what?

Sputum acid-fast-bacilli tests are useful when there is a cavitary pneumonia. There are an enormous number of acid-fast bacilli in cavitary TB, so a positive AFB test is the norm.

Non-cavitary TB is less likely to be AFB positive and will turn culture-negative faster, but they have infiltrates.

Can you have a positive AFB with an essentially normal CT and no symptoms of TB?

I don't think so. AFB in the sputum with a normal chest X-ray is an extremely rare situation. Keeping him in isolation and in the hospital while we try and get a sputum that doesn't exist looking for AFB that shouldn't be there doesn't seem like a rational approach to the patient.

But we have to say he doesn't have TB before the skilled nursing facility will take him back, so we get three sputum tests that are a waste of time and declare him TB-free.

What I really need is a Giant African Pouched Rat. Let them sniff the sputum, say it is negative, and call it good.

Rationalization

Am Rev Respir Dis. 1984 Feb;129(2):264-8. Acid-fast bacilli in sputum smears of patients with pulmonary tuberculosis. Prevalence and significance of negative smears pretreatment and positive smears post-treatment.

http://www.ncbi.nlm.nih.gov/pubmed/6421211

Am Rev Respir Dis. 1977 Jan;115(1):147-9. Persistence of Mycobacterium tuberculosis in sputum without chest roentgenographic evidence of active disease.

http://www.ncbi.nlm.nih.gov/pubmed/?term=402095

Chest. 1988 Aug;94(2):316-20. Chest roentgenogram in pulmonary tuberculosis. New data on an old test.

http://www.ncbi.nlm.nih.gov/pubmed/2456183

Saudi Med J. 2006 May;27(5):633-6. The pattern of active pulmonary tuberculosis in adults at King Hussein Medical Center, Jordan.

http://www.ncbi.nlm.nih.gov/pubmed/16680251

Pan Afr Med J. 2011;9:28. Epub 2011 Jul 18. Using giant African pouched rats to detect tuberculosis in human sputum samples: 2010 findings.

http://www.ncbi.nlm.nih.gov/pubmed/22145062

Chapter Eighty-Seven

Pesky Veins

The patient has a UTI with bacteremia from K. pneumoniae. Not all that uncommon as a cause of admission. The bug is the usual pan-sensitive, although I have heard tell of 5 carbapenem-resistant Enterobacteriaceae in my fair city. I pretty much expect any and all infection control attempts will fail to halt the spread of this organism. When the most common normal bacteria are both resistant and in the community, they will spread with amazing rapidity. Watching the New Delhi metalo beta-lactamase (NDM) spread had given me pause.

"Since the first report in 2009, cases involving NDM-producing Enterobacteriaceae have been reported in every continent except South America and Antarctica."

To date, most of the resistant bugs--MRSA, Pseudomonas, Acinetobacter--have not been part of the normal human microbiome, so we at least had a chance of stopping them. Stopping Carbapenem-resistant Enterobacteriaceae will likely be like trying to stop the wind. We will be able to stop spread in the hospital with good infection control, but unless these bacteria have a real fitness disadvantage, there is no reason for them not to wander from colon to colon in the community.

There are two features that are curious about the case. One is while the patient's blood cultures were negative on admission, she is still febrile on day three, and it was those blood cultures that are positive. She has a peripherally inserted central catheter (PICC) due to short gut, so perhaps bacteria seeded the line.

The other is her baseline back pain has become markedly worse, she has new urinary retention and bilateral thigh weakness, and although she has known severe spinal stenoses, it is much worse.

Cue the music from Jaws.

Good news is that the MRI shows no epidural abscess. The bad news is there is discitis. K. pneumoniae is not high on my list as etiologies of spondylodiscitis, with maybe 3 or 4 cases on the PubMeds. It is not the hyperpviscous strains associated with more severe and metastatic disease, and as usual with a urine bug in the lumbosacral spine, I blame Batson's plexus.

Now a third-generation cephalosporin and time.

And she got better.

Rationalization

Carbapenem-Resistant Enterobacteriaceae Containing New Delhi Metallo-Beta-Lactamase in Two Patients --- Rhode Island, March 2012 MMWR June 22, 2012 / 61(24);446-448

http://www.cdc.gov/mmwr/preview/mmwrhtml/mm6124a3.htm

Infection. 1999 Nov-Dec;27(6):368-9. Spontaneous spondylodiscitis caused by Klebsiella pneumoniae.

http://www.ncbi.nlm.nih.gov/pubmed/10624600

Neurosurgery. 2011 Nov;69(5):1007-14; discussion 1014. doi: 10.1227/NEU.0b013e3182274865. History of the vertebral venous plexus and the significant contributions of Breschet and Batson.

http://www.ncbi.nlm.nih.gov/pubmed/21654535

Chapter Eighty-Eight

More Questions Than Answers

There are diseases I have yet to see, and there are diseases I have seen but have yet to diagnose. The fun in medicine, at least for me, is figuring out what the patient has. Treating them? It is often the matter of looking it up. With today's case, that has led to more questions than answers.

The patient has a disease I have seen twice before, but have yet to make the diagnosis myself.

Three years ago, in her mid-fifties, the patient was diagnosed with Polymyalgia Rheumatica (PMR). Kind of young, but she had muscle pain/fatigue, and it melted away with 20 milligrams of prednisone. The clinical pattern, except for the age, was perfect. Over the next three years, they slowly weaned her off the prednisone, and then in January there was a clinical change: fevers, diarrhea, fatigue, forty-pound weight loss, malabsorption. She had the million-dollar workup as an outpatient, with no diagnosis, and was admitted after a stroke. The CT and MRI show multifocal involvement of the brain with infarctions.

On Chief of Service rounds, our head of medicine called it "Whipple's disease" and later that day the small bowel biopsy came back with Whipple's. Score one for Steve, and they then called me for treatment. Whipple's disease is a very rare infection by a bacterium called Tropheryma whipplei; only about 1000 total cases have been described. Although the infection was first described in 1907, the bacteria were not cultured until 2000, and many questions about the disease remain. Several different presentations are possible: classic Whipple's disease presents with lesions in the gastrointestinal tract, but the disease may also cause culture-negative endocarditis and neurologic infection.

There are some curious questions about the case.

Stroke is not a common manifestation of the disease:

"Cerebral Whipple's disease resembling a stroke syndrome has so far been reported in only two other patients and in both cases it represented the first presentation of the disease."

Was it PMR, then Whipple's starting in January? I think so. I can't imagine Whipple's brewing on prednisone with no symptoms for 2 or 3 years, but perhaps I lack imagination.

Is this Whipple's endocarditis? With the multifocal involvement of the brain it looks like an embolic shower, and while Whipple's is a common cause of culture-negative endocarditis, it can also present as stroke. So while probably the latter, since the transesophageal echocardiogram was pristine, I am going to treat for the former. Plus, given the degree of gastrointestinal malabsorption, I can't trust the absorption of antibiotics after 2 weeks of ceftriaxone.

The recommendations are 2 to 4 weeks of ceftriaxone, then a year of Bactrim (trimethoprim/ sulfamethoxazole) one tablet twice a day for a year. And the relapse rate is 20 to 40%.

Anyone surprised? Give a piddly dose of Bactrim, no matter how long, and you expect the disease to go away? I don't think so. Doxycycline plus hydroxychloroquine is also suggested for central nervous system disease. For antibiotics, ice cream and beer, more is usually better, and the way you breed resistance is to expose an organism to a small dose for a long time. Just to make sure I get drug into every tissue I would be inclined to use higher doses if she tolerates it, although there is nothing in the literature to support the use of higher dose antibiotics. One twice a day would be fine for a Whipple's urinary tract infection, I suppose, but not a systemic disease.

And why? Whipple's can be found in normal stool, but what little epidemiology there is suggests sewage is the source. And she has no swage exposure.

More questions than answers. That's what you get with a rare disease.

Rationalization

Whipple's Disease Florence Fenollar, M.D., Ph.D., Xavier Puéchal, M.D., Ph.D., and Didier Raoult, M.D., Ph.D. N Engl J Med 2007; 356:55-66January 4, 2007DOI: 10.1056/NEJMra062477

http://www.nejm.org/doi/full/10.1056/NEJMra062477

Scand J Gastroenterol. 2005 May;40(5):607-9. A patient with cerebral Whipple's disease and a stroke-like syndrome.

http://www.ncbi.nlm.nih.gov/pubmed/16036516

J Infect. 2013 Mar;66(3):263-70. doi: 10.1016/j.jinf.2012.12.004. Epub 2013 Jan 3. Intravenous ceftriaxone, followed by 12 or three months of oral treatment with trimethoprim-sulfamethoxazole in Whipple's disease.

http://www.ncbi.nlm.nih.gov/pubmed/23291038

Whipple's DiseaseFlorence Fenollar, M.D., Ph.D., Xavier Puéchal, M.D., Ph.D., and Didier Raoult, M.D., Ph.D

NEJM

Chapter Eighty-Nine

Expletive Deleted

"If I had all the answers, I'd run for God." Max Klinger

I wish I were half as smart as I like to think I am. Medicine is often about making decisions with the data at hand, and sometimes those decisions are wrong. Maybe if I were smarter, I would be closer to perfect.

The patient is a new hire as a nursing assistant, and part of the evaluation is a PPD skin test for tuberculosis. Since the patient is from Africa, it is not a surprise that the PPD is a whopping positive, so they send him to me.

He has no symptoms of TB--no consumption, no pulmonary complaints. He has no good history of TB exposure, but coming from sub-Saharan Africa, it is likely he had multiple unrecognized exposures in his life.

The chest X-ray is interesting. I would have called it normal. Radiology, and this is why we have radiologists, say there is a faint right upper lobe infiltrate. I see it only after reading the radiologist's report.

The patient does not have a cough, much less sputum production, and I have to know what is in that right upper lobe. With hesitation, I arrange a bronchoscopy.

The specimens are negative on smear, and the cultures are negative at three weeks, and the patient is without symptoms and is anxious to start earning an income. So I think it's not TB, probably not infectious TB given no cavity or cough, so it is OK to return to work. If this patient had been in the hospital to rule out TB, I would have called him good to go.

You know where this is going.

At the start of week five, acid-fast bacilli are growing from the cultures and it is Mycobacterium tuberculosis. Expletive deleted.

It is probably early reactivation TB and he is mostly controlling it. A repeat chest X-ray shows the infiltrate is a tich worse, a slightly nodular smudge, like a Messier object.

He is still entirely without symptoms and off work.

Active TB remains challenging to diagnose with alacrity, even after 400 years of infecting humans. Can't even trust a negative chest X-ray to rule out the disease:

"The rate of normal CXR among persons with culture-confirmed pulmonary TB was high. Respiratory specimen cultures should be obtained in TB suspects with a normal CXR, particularly HIV-infected persons."

Oh well. Live and learn. At least no harm done.

Rationalization

Int J Tuberc Lung Dis. 2008 Apr;12(4):397-403. Normal chest radiography in pulmonary tuberculosis: implications for obtaining respiratory specimen cultures.

http://www.ncbi.nlm.nih.gov/pubmed/18371265

Chapter Ninety

Fatherly Advice

My Dad was not one for giving advice. I assume he preferred to lead by example, so the few pieces of advice he did offer I usually, but not always, paid attention to.

One was to get a pre-nuptial agreement. I ignored that one: thirty years and no need for it to date.

Another was to go into Internal Medicine instead of Family Practice as I would be bored as a family practitioner. Good call on his part.

Another was that if the tests do not support the diagnosis, it is the tests that are wrong. That one is often true. As I have mentioned many times in the past, friends, family, and acquaintances will lie, but the cultures do not lie. Perhaps they mislead and confuse, but they do not lie.

Today's patient has extensive upper gastrointestinal tract disease and surgery this admission: erosive esophagitis, peptic ulcer disease, duodenal stricture, an antrectomy (removal of the distal half of the stomach), Roux-en-Y (although how they brown the flour in the abdomen, I have not discovered), fundoplication, and an esophagogastroduodenoscopy (EGD) during the procedure.

Post-op she has fevers and leukocytosis and new bilateral pleural effusions, which are tapped on consecutive days. Both grow Candida albicans.

Not much in the way of pneumonia, and besides, there is almost no such thing as Candida pneumonia much less with a subsequent empyema; Candida pneumonia is the Coelacanth of infections.

There has to be, or perhaps past tense, had to have been, a hole in the esophagus. There is no other way for the Candida to get to that space:

"Candida empyema thoracis can be a consequence of operation, gastropleural fistula, and esophageal perforation."

Surgeons, as is their nature, are hesitant to consider that the disease could be due to something they did. I totally understand that approach. The only way you could be a surgeon is, at some level, the ability to deny that you can hurt people.

So we get a contrast study and... ?

No leak. I, of course, think the leak self-sealed, as the effusions have not reaccumulated. Others believe it was never there. But I will stick with my dad's aphorism.

Time and antifungals and the patient got all better.

Rationalization

Mycopathologia. 2010 Jun;169(6):451-5. doi: 10.1007/s11046-010-9277-6. Epub 2010 Feb 9. On a fatal case of Candida krusei pleural empyema in a pregnant woman with spontaneous esophagus perforation.

[http://www.ncbi.nlm.nih.gov/pubmed/20143192]{.underline}

The pun referenced: How to Make a Roux: A Step-by-Step Guide

https://www.foodnetwork.com/how-to/articles/how-to-make-a-roux-a-step-by-step-guide

POLL RESULTS

The best advice I ever had from a parent was

• get a pre-nup. 2%

• all men are pigs. 10%

• I owe my success to having listened respectfully to the very best advice, and then going away and doing the exact opposite. 6%

• Stocks may rise and fall, utilities and transportation systems may collapse. People are no damn good, but they will always need land, and they will pay through the nose to get it 27%

• The only thing to do with good advice is to pass it on. It is never of any use to oneself. 35%

• Other Answers 20%

  • Don't spend your capital. And stop that sniffling!

  • Don't trouble trouble till trouble troubles you.

  • beauty is skin deep, but ugly is to the bone

  • Don't try to understand women. Women understand women, and they hate each-other.

  • Don't make me stop this car.....

  • NEVER UNDERESTIMATE THE POWER OF HUMAN STUPIDITY

  • Go to Med school. Didn't listen. My bad.

  • Don't sh*t where you eat.

Chapter Ninety-One

Glucan Free?

The patient is admitted to the ICU with multilobar consolidated pneumonia and is septic. She is on immunosuppression, and a test for 1-3 beta-D glucan is sent to look for invasive fungal infection. I am not quite certain why the test is done, but it comes back off the wall,

500 pg/mL when anything above 80 is a definite positive. So they call me.

Not much else in the history: renal transplant, pacemaker, just back from the central valley of California.

A CT scan of the chest shows consolidation but no cavitation.

What I have is a whole lot of confusion as to the etiology of the beta-glucan.

A bronchoscopy shows one colony of Aspergillus and scant Candida albicans. No Pneumocystis jirovecii.

Blood cultures at day three grow Propionibacterium acnes.

Serologies are negative for Coccidiomycosis, the galactomannan is negative on the bronchoscopy, and the Cryptococcal antigen is negative.

After a week on micafungin and community-acquired pneumonia therapy, the repeat beta-D glucan is negative. The chest X-ray is much improved, and she is off the ventilator and moved to the floor.

So what is the source of the beta-D glucan?

Too high for a false positive.

Could it be Candida pneumonia? Seems unlikely, given the rarity of this organism causing pulmonary disease. The Aspergillus is a likely contaminant/colonization, given the negative galactomannan and the CT scan.

I got all excited about the P. acnes. I had read a recent report of Listeria causing a false positive for the galactomannan assay. Could a P. acnes infection cause a similar reaction? So I went looking, and it turns out that P. acnes also makes beta-D glucan. I think. Biochem-fu is not my strong point.

Unfortunately, the therapy for community-acquired pneumonia also treated P. acnes, so repeat blood cultures were negative. The transesophageal echocardiogram showed no pacer wire involvement, and off antibiotics, she has not relapsed. Although if P. acnes makes a beta-glucan, could micafungin have had an effect? As a fellow, I inadvertently found out that Candida species inhibited by aminoglycosides. I'll leave this to the biochemists; it is beyond my pay grade. If micafungin does work against bacteria, you heard it here first and the company can send me a check. Ha. As I mentioned recently, I have lots of ideas. About 1 in 100 are good ones.

So I am left with an arbitrary course of therapy for a blood test whose provenance I remain uncertain about. I need a glucan-free practice.

Rationalization

Diagn Microbiol Infect Dis. 2013 Jun;76(2):250-1. doi: 10.1016/j.diagmicrobio.2013.02.005. Epub 2013 Mar 19. Cross-recognition of aspergillus galactomannan caused by Listeria monocytogenes infection.

http://www.ncbi.nlm.nih.gov/pubmed/23518185

Contribution of Surface β-Glucan Polysaccharide to Physicochemical and Immunomodulatory Properties of Propionibacterium freudenreichii Appl. Environ. Microbiol. March 2012 vol. 78 no. 6 1765-1775

http://aem.asm.org/content/78/6/1765.abstract

Appl Microbiol Biotechnol. 2005 Aug;68(2):163-73. Epub 2005 Apr 8. Curdlan and other bacterial (1--> 3)-beta-D-glucans.

http://www.ncbi.nlm.nih.gov/pubmed/15818477

J Agric Food Chem. 2010 May 26;58(10):6149-56. doi: 10.1021/jf904529q. Screening and selection of 2-branched (1,3)-beta-D-glucan producing lactic acid bacteria and exopolysaccharide characterization.

http://www.ncbi.nlm.nih.gov/pubmed/20423154

Chapter Ninety-Two

Wil do ID for Food

Yesterday a half-dozen people sent me links to the NEJM article Targeted versus Universal Decolonization to Prevent ICU Infection and wanted to know what I thought of it and whether we should do the same thing at my hospitals. I guess they think I don't read the NEJM :)

I read the article over. They had three groups:

"Group 1 implemented MRSA screening and isolation; group 2, targeted decolonization (i.e., screening, isolation, and decolonization of MRSA carriers); and group 3, universal decolonization (i.e., no screening, and decolonization of all patients)."

and they showed benefit:

"In the intervention versus baseline periods, hazard ratios for bloodstream infection with any pathogen in the three groups were 0.99 (crude rate, 4.1 vs. 4.2 infections per 1000 days), 0.78 (3.7 vs. 4.8 infections per 1000 days), and 0.56 (3.6 vs. 6.1 infections per 1000 days), respectively (P<0.001 data-preserve-html-node="true" for test of all groups being equal). Universal decolonization resulted in a significantly greater reduction in the rate of all bloodstream infections than either targeted decolonization or screening and isolation."

I don't know. My reaction is that the intervention, the decolonization, was that it is a surrogate intervention for sloppy baseline practice.

Let me digress.

I have been in charge of Infection Control at my hospitals for 23 years. In 2007-2008 it was decided that the acceptable infection rate in our hospitals was going to be zero. I had perhaps been a little bit jaded, thinking that infections were the occasional price we paid for the interventions we did in the ICU. That was wrong.

With a very supportive administration and a very dedicated staff at all levels, we were aggressive at implementing surgical care improvement projects, various bundles (collections of evidenced-based interventions to prevent infections) and maximizing hand hygiene adherence.

We have been extraordinarily successful:

{width="12.48611111111111in" height="8.333333333333334in"}

Year after year, infections have fallen. We estimate that we have prevented over 1800 infections across the system compared to if we had continued to practice according to the standards of 2007. That is a lot of prevented infections.

That is a real drop in infections, not playing with numbers; I have the W-2s to prove it. My practice, which used to be based in part on nosocomial infections, has slowly evaporated over the last decade. I just do not see nosocomial infections anymore. The drop in infections includes an inner-city hospital, a burn center, and a level one trauma center. One thing I have learned is that the plea, "but our patients are sicker" is the last refuge of the lazy. Zero is a realistic infection rate if you have a health care system with a minimal number of ABPs*.

To put those rates in perspective, that sort of parallel what is reported in the NEJM, bloodstream infections in my main ICU for 2012 were about 0.4 per 1000 patient days. At the end of the intervention, their best rate was 3.6 per 1000 attributable days. Their best rates would be deeply embarrassing if we had them in my hospital. They have rates we haven't seen in a decade, from back in the bad old days. In 2013 we started routine chlorhexidine baths in the ICU. I expect infections rates to fall further, and I will shortly be on a freeway off-ramp with a sign saying "Will do ID for Food."

So when I see ghastly infection rates that, after an intervention, is only slightly less ghastly, I really wonder what the compliance rate with core infection control competencies are. How compliant are they at SCIP, at implementing bundles, at hand hygiene? It is not reported, but given "One bloodstream infection was prevented per 54 patients who underwent decolonization" it can't be all that good.

I don't like to diss other institutions, which I know I am doing, but 3.6/1000? Ow.

The key to infection control and preventing the spread of infection is being punctilious with basics every time. Not antibiotics. At least, that is what has been effective in my hospitals, much to the distress of my bottom line. And a good thing too.

Rationalization

Targeted versus Universal Decolonization to Prevent ICU Infection May 29, 2013DOI: 10.1056/NEJMoa1207290

http://www.nejm.org/doi/full/10.1056/NEJMoa1207290

*ABD: arrogant butt-head practitioners. You know, those expletive deleteds who somehow think standards do not apply to them. They are few and far between in my hospitals.

Chapter Ninety-Three

Triad, Pentad, or Septad

The patient is admitted with fevers and altered mental status. At 24 hours the cultures are positive for a gram-negative rod in the blood and an E. coli in the urine, so they give me a holler.

Could the prosthetic aortic valve be involved?

Over the phone I said, "Unlikely." There is the odd case of E. coli going to heart valves, but it is rare. There are maybe 54,000 episodes of E. coli bacteremia a year in the US elderly population, even more if you include all ages, but seeding valves is seen once a career event. I have already seen my case. No need for a second, but one never knows.

When I see the patient, I am less sanguine. A good murmur, both conjunctiva have had a nice shower of emboli, and there is hematuria. So we have the classic triad of endocarditis.

We get a transesophageal echocardiogram to let us see which valve is involved, while meanwhile the organism is identified as H. parainfluenza. The E. coli was a red herring and, given the condition of the dentition, it is not a surprise she is infected with a mouth organism.

H. parainfluenza is part of the HACEK group of organisms: Haemophilus parainfluenzae, Haemophilus aphrophilus, Haemophilus paraphrophilus, Actinobacillus actinomycetemcomitans, now Aggregatibacter aphrophilus, Cardiobacterium hominis, Eikenella corrodens, and Kingella kingae.

Is that a pentad? 5 species in HACEK, but seven bugs. Septad? Anyway, last century I completed my septad and as best as my enfeebled memory suggests, I haven't seen endocarditis from these organisms this century except for an Eikenella. So two down and 3-5 to go.

Prosthetic valve endocarditis from H. parainfluenza is rare, maybe a septad's worth of reports on the PubMeds, so not a lot of guidance for treatment. The good news in this case was that the prosthetic valve was clean, the only anomaly being something flickering on the tricuspid valve of all things. There isn't even a triad of cases of the tricuspid valve being involved with H. parainfluenza.

So a course of ceftriaxone and a short course of gentamicin and hope we can give the patient a non-surgical cure.

We did.

Rationalization

J Infect Dis. 2005 May 1;191(9):1523-9. Epub 2005 Mar 22. Burden of community-onset Escherichia coli bacteremia in seniors.

http://www.ncbi.nlm.nih.gov/pubmed/15809912

Diagn Microbiol Infect Dis. 2009 Jun;64(2):216-9. doi: 10.1016/j.diagmicrobio.2009.02.015. Epub 2009 Apr 18. Tricuspid valve endocarditis caused by Haemophilus parainfluenzae: a case report and review of the literature.

http://www.ncbi.nlm.nih.gov/pubmed/19376668

Heart Lung. 2005 Mar-Apr;34(2):152-4. Haemophilus parainfluenzae mitral prosthetic valve endocarditis in an intravenous drug abuser.

http://www.ncbi.nlm.nih.gov/pubmed/15761462

Clin Infect Dis. 1997 Jun;24(6):1087-94. Haemophilus endocarditis: report of 42 cases in adults and review. Haemophilus Endocarditis Study Group.

http://www.ncbi.nlm.nih.gov/pubmed/9195062

Chapter Ninety-Four

Haemophali

The patient is 53, mostly healthy, and is admitted with fevers, slight disorientation, and the worst headache of her life.

A lumbar puncture is consistent with bacterial meningitis, the gram stain is negative and she is started on the usual empiric therapy.

The next day the blood and cerebrospinal fluid are growing a gram-negative rod, and so the call goes out to ID.

She has been ill for about 48 hours and looks impressively nontoxic for someone with a gram-negative meningitis. Ill, but not the usual "I'm going to die soon" look that most people have with these kinds of organisms in the spinal fluid.

Risk factors are minimal; she takes care of her grandchildren, and they are up to date on their vaccinations. I am puzzled. Gram-negative rods of all kinds do cause meningitis, but she has nary a risk nor a source.

So we wait.

Next day we grow Haemophilus influenza, beta-lactamase negative. Do not see that bug often in the adult cerebrospinal fluid. The last, and perhaps only other, I can remember had a fistula between his inner ear and brain of some sort.

Today I found out it was a nontypeable H. influenza. Maybe from the sinuses, as she had minimal ethmoid thickening showing on the CT scan, but hardly an impressive sinusitis. It does not appear to be a typical presentation even for this atypical organism:

"Clinical presentations of 396 cases included mainly invasive pneumonia (190 cases [48%]) and bacteremia without a clinical focus (75 cases [19%]). Comorbidities were present in 327 [83%] and immunodeficiency in 173 [44%] of all cases. The overall case fatality rate within the first month after diagnosis was 12% and the lowest (2%) was among patients aged 5-54 years. The highest extrapolated age-specific incidence rates occurred within the first 6 weeks of life (19.0 cases per 100,000 persons), concerning mostly prematurely born infants with bacteremia within 24 h after birth, and in the first year of life (5.6 cases per 100,000 persons). The highest rate in adults was among elderly patients aged > 65 years (2.2 cases per 100,000 persons). CONCLUSIONS: This study provides a detailed overview of invasive ntHi disease cases in the Netherlands. Risk groups are prematurely born infants, elderly patients aged > 65 years, and immunocompromised patients."

Meningitis is not a common manifestation either:

"Bacteremia accounted for 83% of all cases, meningitis for 12.4%, and other sterile sites for the remaining 4.6%. Nontypeable strains, Hib, and Hif accounted for 13 of the 15 cases of meningitis (Table). Infection with Hib was significantly more likely to be associated with meningitis than nontypeable strains (odds ratio [OR] 6.1, 95% CI 1.3--29.0)."

It was remarkable, given the reports, just how relatively well she was and how rapidly she improved. I am an adult ID doc (who I treat, not how I behave) and I have never seen a case of invasive H. influenza b thanks in large part to the vaccine, so in 30 years I never had an opportunity to evaluate why these patients are in droplet isolation.

Secondary cases are common in kids and the younger they are, the more likely they are to acquire the disease:

"The secondary attack rate in children under 5 years or less in the month after exposure to an index case was thus 2.3%, 800 times the endemic attack rate for H. influenzae meningitis. This is a conservative estimate..."

Impressive for kids, but my Google-fu is lacking. I can't find where adults or health care workers are at increased risk from the index case. It is a reasonable assumption, but you know the old saw about assume. But safe > > sorry.

That's two Haemophilium this week.

Rationalization

Clin Infect Dis. 2011 Jul 1;53(1):e1-7. doi: 10.1093/cid/cir268. Nontypeable Haemophilus influenzae invasive disease in The Netherlands: a retrospective surveillance study 2001-2008.

http://www.ncbi.nlm.nih.gov/pubmed/21653293

Emerg Infect Dis. 2011 September; 17(9): 1645--1650. doi: 10.3201/eid1709.101991 PMCID: PMC3322072 Increasing Incidence of Invasive Haemophilus influenzae Disease in Adults, Utah, USA

http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3322072/

Secondary Haemophilus influenzae Type b in Day-care Facilities Risk Factors and Prevention David W. Fleming, MD; Mark H. Leibenhaut, MD; Demetrius Albanes, MD; Stephen L. Cochi, MD; Allen W. Hightower, MS; Sue Makintubee, RN; Steven D. Helgerson, MD, MPH; Claire V. Broome, MD JAMA. 1985;254(4):509-514.

http://jama.jamanetwork.com/article.aspx?articleid=399670

Br Med J. 1980 March 29; 280(6218): 899--901. PMCID: PMC1601106 Haemophilus influenzae type B meningitis: a contagious disease of children.

http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1601106/

Chapter Ninety-Five

Toy Boat, Toy Boat, Toy Boat

Say "toy boat" three times as fast as you can. I can't. "Shoulder surgery" is another pair of words that thuds off my tongue with confused syllables.

The patient had shoulder surgery to repair a fracture, and over the last half a year had not recovered. Painful and stiff, the shoulder required manipulation under anesthesia and still did not improve.

Then something came to a head over the incision, which ruptured to drain pus, but he never had any constitutional symptoms and the shoulder lacked rubor, calor, and tumor.

The patient had an incision and drainage of the shoulder. There was not an impressive infection in the OR and the gram stain was negative, but a week in all the cultures are growing Propionibacterium acnes. Again.

I am starting to suspect this is now the most common post-op infection associated with hardware in my practice. I need to run the numbers someday, but the question is, what can we do to prevent it?

P. acnes sits in hair follicles, surrounded by grease, which is probably why the bug is seen more often in men--we are hairier and greasier than women as a general rule. Sebum protects the bacteria from topical scrubs and parenteral antibiotics.

P. acnes is not uncommon in shoulder surgery and is slow to manifest and slow to grow, so you have to look for it, asking the lab to hold the cultures for two weeks. I assume that it causes so little reaction as it is part of normal flora and we are used to living in quasi-harmony with it, so it goes unrecognized by the immune system.

One group that is probably protected is actors. They all seem to lack body hair; glabrousness is apparently a job requirement. And I am the only one who finds those commercials for an electric razor for chest hair grooming kind of creepy? Or is that yet another issue, like tattoos and extra-auricle piercing, that I am too old to understand? Probably.

A washout and long-term antibiotics, and I hope for a cure.

And a cure it was.

Rationalization

Clin Infect Dis. 2008 Jun 15;46(12):1884-6. doi: 10.1086/588477. Propionibacterium acnes postoperative shoulder arthritis: an emerging clinical entity. Levy PY, Fenollar F, Stein A, Borrione F, Cohen E, Lebail B, Raoult D.

http://www.ncbi.nlm.nih.gov/pubmed/18462110

J Infect. 2007 Aug;55(2):119-24. Epub 2007 Apr 5. Propionibacterium acnes: an agent of prosthetic joint infection and colonization. Zeller V, Ghorbani A, Strady C, Leonard P, Mamoudy P, Desplaces N.

http://www.ncbi.nlm.nih.gov/pubmed/17418419

POLL RESULTS

I most have trouble saying

• toy boat toy boat toy boat 11%

• shoulder surgery shoulder surgery shoulder surgery 11%

• rubber baby buggy bumpers 24%

• I don't know 7%

• I am wrong 33%

• Other Answers 13%

  • No, no! put away your money. Dinner is my treat.

  • unique new york

  • Memorial Union.

  • Unique New York

  • She sits and shines she shines and sits.

Chapter Ninety-Six

Another Learning Experience

The patient is an elderly male with type two diabetes who has severe rigors, high fevers, myalgias, and confusion. I get a call from his wife to discuss the symptoms.

The patient is totally non-focal and I suggest they get to the ER. Now.

So they did, and he was admitted. Completely non-focal on exam, review of systems, and labs.

Doesn't matter. Rigors mean bacteremia:

"In a multivariate analysis, several factors were independently associated with community-onset bacteremia, including an age of > 65 years (odds ratio [OR], 2.81; 95% confidence interval [CI], 1.25-6.33), the presence of rigor (OR, 13.7; 95% CI, 4.47-42.0) or chills (OR, 6.04; 95% CI, 1.10-32.9), a body temperature > 39.9 °C (OR, 2.68; 95% CI, 1.03-6.94),"

But after 24 hours, the blood cultures are negative, the illness has completely resolved either from the antibiotics or the natural history of the disease, and he is discharged.

36 hours after the onset of the fevers and shortly after discharge he has 8 very watery diarrheal stools, which are sent for cultures and grow Campylobacter.

I do not read Up to Date as a rule, since I am usually Way Behind, but it was pointed out to me that it says:

"In about one-third of cases, a prodromal period characterized by high fever accompanied by rigors, generalized aches, dizziness, and delirium is observed. It may last for one day (rarely two or three days) prior to onset of gastrointestinal symptoms. Patients presenting with prodromal symptoms tend to have more severe disease than those presenting with diarrhea."

Didn't know that.

But where did he get it from? The only recent diet exposures are that he makes his own boudin sausages. Pork can be a source of Campylobacter. And Oregon strawberries. There is so much in Oregon that is superior: the beer, the food, the hikes, and, mmmmm, the strawberries. Everything is better, in fact--everything but an elegant sense of style. In Portland, even the beautiful people dress dumpy. But the strawberries this time of year define luscious, and he has been eating them unrinsed. But there are no reports I can find of Campylobacter from strawberries, so the boudin it may be.

Or just bad luck. All the food we eat has a layer of animal or human excrement on it. It is a matter of how thick and how well it is washed off. It is why I only eat fried food.

Which reminds me. What is the difference between Portland and yogurt? Yogurt has culture. I'll stick with PDX.

Addendum

Subsequently there are been reports of Campylobacter from strawberries.

Rationalization

Diagn Microbiol Infect Dis. 2012 Jun;73(2):168-73. doi: 10.1016/j.diagmicrobio.2012.02.009. Epub 2012 Mar 29. Prediction of community-onset bacteremia among febrile adults visiting an emergency department: rigor matters.

http://www.ncbi.nlm.nih.gov/pubmed/22463870

Chapter Ninety-Seve

Yet Another Thing I Do Not Understand

The patient has end-stage liver disease (ESLD) and is at baseline when he has the abrupt onset of rigors and fevers. Completely non-focal, but it persists, and after 24 hours he is admitted, has blood cultures drawn, is started in ceftriaxone and gets all better.

Next day both sets of blood cultures, drawn 55 minutes apart, are growing. Streptococcus parasanguis, a viridans streptococcus, and they call me.

The physical exam shows nothing but a long-standing murmur, and the labs are those of ESLD. The echocardiogram is negative for a vegetation. S. parasanguis is not a terribly common cause of endocarditis in the viridans family, maybe 4% of cases. Of course, given the vagaries of streptococcal identification, who knows really knows. Often the lab calls it viridans Streptococcus, and that is that.

It is not a pleasant bug. The minimum inhibitory concentration (MIC) to penicillin is 3.0 micrograms per milliliter, and to ceftriaxone it is 1.5 micrograms per milliliter. Both of these indicate resistance. What are the options? Liver disease is a risk for renal and ototoxicity from aminoglycosides.

" By logistic regression analysis, the relative odds of renal dysfunction developing during tobramycin treatment in a patient were 6.0 (95 percent confidence interval: 3.8 to 9.5). The relative odds of renal dysfunction developing in a patient receiving tobramycin and having liver disease were 31.8 (95 percent confidence interval: 19.7 to 51.4). This analysis demonstrates an interaction between tobramycin use and liver disease for increasing the risk of renal dysfunction."

If I am going to commit this patient to a course of antibiotics with a real risk of harm, I want to know if there is endocarditis, and the TEE showed a one-centimeter-sized vegetation on the aortic valve. Given the culture's MICs, it is going to be 6 weeks of ceftriaxone and gentamicin.

And this is what I did not understand. The MIC to penicillin was 3.0, and the interpretation was "intermediate." So I called the lab. It is not 'I.' Anything greater than or equal to 1.0 is resistant and means 6 weeks of gentamicin for endocarditis.

It turns out that the new interpretation guidelines say viridans Streptococci are "resistant" if the MIC is > 4 micrograms per milliliter, 'intermediate' if 0.1 to 2.5 and 'sensitive' if < 0.1. They do not even have a classification for an MIC of 3.0, so the lab called it 'I.'

The only time you really care about the MIC of a viridans Streptococcus is endocarditis. If the MIC is between 0.1 and 0.5 (intermediate clinically), they get 2 weeks of gentamicin; if it is 1.0 or greater (resistant clinically), they get 6 weeks of gentamicin.

Why they came up with interpretations that are so removed from clinical practice I cannot discover. Hopefully, one my readers knows why the powers that be decided to confuse people and have the interpretation of virdans Streptococci susceptibility completely divorced from clinical practice.

Or am I missing something? There is probably a sophisticated and clever explanation. I just can't find it.

A reader responded:

"Hi, microbiologist here, I enjoy reading your blog a lot. Perhaps I can bring some light to MIC interpretation. Does your lab do E-tests? The scale on the strip represents a continuous gradient, but standard MIC interpretive criteria (S, I, R) are derived from broth dilution testing which uses two-fold dilution steps. Therefore, if the E-test MIC falls between two-fold dilutions, you are supposed to round up to the next highest two-fold dilution and then interpret it. So, the MIC of 3.0 should have been reported as 4 and interpreted as R (> =4, according to CLSI). EUCAST, which we use, considers any viridans streptococci with a penicillin MIC > 2 as R, by the way. "

Rationalization

J Med Microbiol. 1993 Sep;39(3):179-82. Identity of viridans streptococci isolated from cases of infective endocarditis.

http://www.ncbi.nlm.nih.gov/pubmed/8366515

Am J Med. 1986 Jun;80(6):1093-7. Increased risk of renal dysfunction due to interaction of liver disease and aminoglycosides.

http://www.ncbi.nlm.nih.gov/pubmed/3728506

Chapter Ninety-Eight

Friday Complaining

Acute medicine is messy. The fog of war and all that. The history and physical are not always suggestive of a diagnosis, and certain syndromes require a shotgun approach. The patient comes in septic, you culture everything and kill what might be trying to kill the patient and hope you get it right. Sometimes you need to shotgun. Sometimes all you need is a BB gun.

Medicine also has an aesthetic. You want to do the least amount to the patient to get the maximal diagnostic certainty. We are, at some level, resource stewards, and I hate to order unneeded tests. The longer I am in medicine, the fewer tests I need for certainty in both the diagnosis and to follow the patient. When you are young, it is "Don't just stand there, do something." And when you are old, it is "Don't just do something, stand there."

The patient is a middle-age male, with a 30-pound weight loss in a month, fevers, progressive shortness of breath, bilateral interstitial infiltrates, and lactate dehydrogenase of 750 units per liter, indicating tissue damage of some kind (normal values are < 280).

I try to consider three lists in every patient: the list of likely diseases, the list of diseases I do not want to miss as they could kill or hurt the patient, and the list of cool diseases--the oddity that would be fun to diagnose.

For this patient, all three have Pneumocystis jirovecii pneumonia at the top of the list. Number two on the list? Nothing really. There can be other processes that present like this, but the odds are about 99.99999999999999999% in favor of Pneumocystis jirovecii, PJP. Remember, do NOT say PJP pneumonia. What is the second 'P' in PJP? Pneumonia. Do not be a member of the Department of Redundancy Department.

The patient is started on Bactrim (trimethoprim/sulfamethoxazole). Fine. And azithromycin. Well, OK, it could be an atypical pneumonia, but if it is, it is an atypical atypical pneumonia. And ceftriaxone because that is part of the protocol for community-acquired pneumonia (CAP). Sigh. If there is a ceftriaxone-sensitive pneumonia that has this presentation, let me know, because I certainly am unaware of it. But that is the Epic order set for CAP and ceftriaxone is given. That is aesthetically challenged.

Initially, he is too unstable for bronchoscopy, so a beta-D-glucan test is sent out to assay for fungi. It comes back three days later off the wall, but respiratory failure occurs, and he is intubated and bronched. And every test imaginable was sent on the bronchoscopy specimen. Every respiratory virus that could be tested for was, evidently as the order set popped up on the Epic software--and a simple click, and viola, or perhaps violin, a dozen unneeded tests were ordered. The PJP was positive and, I am happy to report he did not have influenza, parainfluenza, metapneumovirus, rhinovirus, RSV, adenovirus, pertussis, coronavirus, Chlamydia or Mycoplasma. But I knew that without testing. Seeing all those unneeded tests annoys me, all the more so knowing they were generated by the click of a button.

And then we ran out of IV Bactrim for a day. There is still a shortage, going on for a couple of years now. Evidently, it is made by one company, and I cannot find out why they can't make it. What is this, Cuba? Three years and we still can't make a sufficiency of IV Bactrim? Someone needs to make me philosopher-king so I can fix the too-numerous-to-count stupidities that define US health care.

And now, after a week of Bactrim and steroids, he is not that much better. The lactate dehydrogenase is still 760 units per liter. And he is still febrile. And no other diagnosis.

Switch antibiotics (there is some rare Bactrim resistance), even though Bactrim is the best drug. Switching always ends badly. Add another agent like caspofungin? Or Clindamycin/primaquine? Be patient? I have opinions and cases but no right answers. And maybe it is the bug, not the antibiotic. Some strains of PJP are worse than others.

http://www.wavsource.com/snds_2013-06-09_7806069158793874/tv/mpfc/complaining.wav

I know the feeling.

In the end we scrounged up some Bactrim and he did fine.

Rationalization

Emerg Infect Dis. 2013 Jan;19(1):21-8; quiz 186. doi: 10.3201/eid1901.120140. Pneumocystis jirovecii genotype associated with increased death rate of HIV-infected patients with pneumonia.

http://www.ncbi.nlm.nih.gov/pubmed/23260763

J Antimicrob Chemother. 2009 Dec;64(6):1282-90. doi: 10.1093/jac/dkp372. Epub 2009 Oct 26. Clinical efficacy of first- and second-line treatments for HIV-associated Pneumocystis jirovecii pneumonia: a tri-centre cohort study.

[http://www.ncbi.nlm.nih.gov/pubmed/19858161]{.underline}

Poll Results

It is good to complain as

• it clears the soul. 29%

• the squeaky wheel gets the oil 29%

• it is the nail that sticks up that gets the hammer. 6%

• keep a stiff upper lip and quit your whinging (I love the word whinging). 17%

• it makes you attractive to the opposite sex. 6%

• Other Answers 14%

  • Misery loves company

  • It establishes your position relative the issue in question, so you are cleared of blame when those who screwed up go up against the wall.

  • It discourages others from using more unnecessary redundancy than they need to.

Chapter Ninety-Nine

Is the Bark Worse?

The patient is a young female, no medical problems, who had a baby 6 months ago. It was an uneventful pregnancy and delivery. Actually, they are all eventful, just not necessarily in a bad way. I certainly remember my wife's pregnancies as eventful.

For a month she has had a cough, fevers, malaise, and shortness of breath. Several rounds of antibiotics in the outpatient setting do nothing, and a chest X-ray is finally done that shows a thick-walled cavitary pneumonia.

She gets a bronchoscopy that has no organisms and cultures nothing even at two weeks. Gram stain did show a yeast. She gets the cavitary pneumonia workup; all studies are negative except the serum cryptococcal antigen at 1:32.

So is this cryptococcal pneumonia? The cerebrospinal fluid is negative, and there is no reason for a titer. She is improving on fluconazole, so I suppose so.

Pregnancy is a weird immunologic state; I do not even pretend to understand it. Infections are not the same during pregnancy and can get worse after delivery. I wonder if this is an odd form of postpartum immune reconstitution (inflammatory) syndrome (called IRS or IRIS) :

"In patients with cryptococcosis, a Th2 response facilitates the establishment of infection [21, 53--55]. On the other hand, a Th1 profile confers protection and is characterized by inflammatory lesions, which are a hallmark of cryptococcal-related IRS observed in diverse immunocompromised hosts, such as HIV-infected individuals and transplant recipients [22, 56]. Case reports document a similar phenomenon in pregnant women, in whom quiescent cryptococcal infection acutely worsened during the postpartum period;"

Her case is probably a little late in the postpartum period to be IRIS, but it could have been festering for 6 months, having worsened with the return of immunity. This is described for HIV IRIS. Or not. Cryptococcus gattii is endemic in the Great Pacific Northwest, which is more likely to cause severe disease, and here is the curious part. She works in the office of a nursery --- plants, not kids. There is lots of bark dust at work, filling the air.

I had a case of C. gattii pneumonia and meningitis in a person who is an animal trainer and spends a lot of time in dusty barns filled with bark dust. The bark dusk comes from Canada.

Pacific Northwest fir and pine bark are one of the ecologic niches of C. gattii. There are no reported cases of C. gattii acquired from bark dust but

"Two of 64 parks had positive samples for C. gattii. One park had a positive tree and the other park, 60 miles away, had positive bark mulch samples from a walkway.

Ha!

As I have mentioned before, I am waiting for a post-Christmas outbreak, with the trees as a source.

So postpartum IRIS? Occupational bark dust exposure? Both? Neither? I suppose I should go culture the bark dust, but there are just so many hours in the day, and I am more of a Mycroft than a Sherlock. That is my story, and I am sticking to it.

Postscript

It took 6 months of fluconazole for the chest X-ray to clear.

Rationalization

Clin Infect Dis. 2007 Nov 1;45(9):1192-9. Epub 2007 Sep 21. Immune reconstitution syndrome and exacerbation of infections after pregnancy.

http://www.ncbi.nlm.nih.gov/pubmed/17918082

Clin Infect Dis. 2002 Dec 15;35(12):e128-33. Epub 2002 Nov 20. Immune reconstitution cryptococcosis after initiation of successful highly active antiretroviral therapy.

http://www.ncbi.nlm.nih.gov/pubmed/12471589

Mycopathologia. 2013 Apr;175(3-4):351-5. doi: 10.1007/s11046-013-9614-7. Epub 2013 Jan 26. Detection of Cryptococcus gattii in selected urban parks of the Willamette Valley, Oregon.

http://www.ncbi.nlm.nih.gov/pubmed/23354596

Chapter One Hundred

You Probably Know This

And if you don't, you really should pay attention here.

First. Oral vancomycin has zero absorption. Zero. You can't use it to treat methicillin-resistant Staph aureus (MRSA). Like the patient I saw in the clinic who had a MRSA urinary tract infection and was given a 10-day course of oral vancomycin.

Bad idea. And the patient was not thrilled to pay 1200 dollars out of pocket for a 10-day course of placebo.

As to the cost of vancomycin, here is another pearl: Roll your own. I know compounding pharmacies have a bad rep at the moment, but IV vancomycin is relatively inexpensive. If you mix the IV vancomycin with a flavored base (I like watermelon) you can get a course of vancomycin at a fraction of the cost of the pills, all the more important if you are giving it as a placebo.

There are two issues to consider when dosing an antibiotic. Well, more than two. But I have two in mind today.

One is weight. Imagine a person as a sack of water that you have to fill with antibiotic to get a therapeutic concentration. If the patient weighs 500 pounds, do you really think 250 of cephalexin four times a day is going to do anything? We dose antibiotics for the mythical 70 kg human, and most US citizens haven't been that light since 8th grade. The larger the patient, the more the antibiotic you should give, unless you subscribe to the magic of homeopathy where increasing dilution leads to increasing potency.

The other is glomerular filtration rate, related to kidney function. If the patient is old and female and has an elevated creatinine level and the glomerular filtration rate is 40 mL/min/17.3m^2^, nitrofurantoin has zero efficacy. Or so I have always been told. Turns out there is not a lot of data:

"Concerns of increased risks of serious adverse reactions in patients with reduced renal function have further limited the use of nitrofurantoin. However, although not completely clear, these complications seem to be linked most often to prolonged treatment, genetic variability, and predisposition to hypersensitivity.

CONCLUSIONS: Data supporting the contraindication of nitrofurantoin for patients with a CrCl less than 60 mL/min are nonexistent."

Surprised me. It always pays to review topics you are sure you have mastered. But nitrofurantoin has ample opportunity for toxicity.

Today's patient was admitted to the ICU with new onset of fevers, rash, altered mental status and eosinophilia. The patient had been on 6 months of nitrofurantoin for urinary tract infection prophylaxis. Besides retroperitoneal and pulmonary fibrosis, nitrofurantoin can cause DRESS: drug rash, eosinophilia and systemic symptoms.

We stopped the drug and the patient got all better.

That's four pearls today; you are on your way to a very nice necklace.

Rationalization

Pharmacoepidemiol Drug Saf. 2013 Jun 4. doi: 10.1002/pds.3461. [Epub ahead of print] The effect of obesity on antibiotic treatment failure: a historical cohort study. [http://www.ncbi.nlm.nih.gov/pubmed/23733599]{.underline}

Ann Pharmacother. 2013 Jan;47(1):106-11. doi: 10.1345/aph.1R352. Epub 2013 Jan 22. Nitrofurantoin contraindication in patients with a creatinine clearance below 60 mL/min: looking for the evidence. [http://www.ncbi.nlm.nih.gov/pubmed/23341159]{.underline}

Neth J Med. 2009 Apr;67(4):147-9. DRESS syndrome caused by nitrofurantoin. [http://www.ncbi.nlm.nih.gov/pubmed/19581659]{.underline}

Chapter One Hundred One

I Hate Chest X-rays

I really do. They so often miss disease that I wonder why we rely on them so much. Probably because they are relatively inexpensive and lower radiation exposure than a CT scan, which gives so much more information.

Spend a lot of time in the ICU, and you rapidly realize the superiority of a CT scan. Pneumonias and fluids and emboli, oh my. Whether that extra information is actually helpful is beside the point.

Last week two patients presented more or less the same ways: fevers, rigors, myalgias, slight shortness of breath, and severe watery diarrhea for about 24 hours.

I thought for sure they were going to have a bacillary diarrhea, probably Camylobacter or Shigella, but stool studies were negative, and fevers persisted. It was odd that in both cases, a test for Clostridium difficile was sent but not standard bacterial cultures.

After 48 hours, the patients were no better, the stool cultures were negative, and the chest X-ray was repeated. Both had a lobar pneumonia. Another negative X-ray, and not from dehydration-- from diarrhea.

To quote me from a prior entry:

"I suspect dehydration delaying pneumonia findings on CXR is a myth, like we use 10% of our brain (the presidential debates not withstanding) and Eskimos having 300 words for snow, although in Oregon we have 300 words for rain, and they all start with damn."

Diarrhea does occur with many other infectious diseases, including pneumonia:

"In a study of 1,812 patients with community-acquired pneumonia (CAP), diarrhea was noted at presentation in 24%. Diarrhea occurred in 29% of younger patients, between 18 and 44 years of age, compared with 18-22% of patients older than 44 years of age. Cough was the first symptom of disease, occurring 7 days before presentation; fever started 3 days before presentation, and diarrhea, vomiting and abdominal pain occurred 2 days before presentation."

Confuses the HE-double-toothpicks out of the presentation, that's for sure.

But why? My just so story as to why we have diarrhea? I figure that early in evolution, many common infections came from consuming contaminated food and water, and the key to all infections is source control. Cytokines cause diarrhea. Diarrhea is a good source control for gastrointestinal pathogens as well as a good way to get reading done. So the evolved response to GI pathogens has bled (pooped?) over into other infections that were less common on the plains of Ethiopia, the cytokines not knowing any better, and we get diarrhea with non-GI infections. And it was the cause of human literacy.

Seems reasonable to me, but then I think that of all my ideas.

Rationalization

Nat Clin Pract Gastroenterol Hepatol. 2005 May;2(5):216-22. Diarrhea caused by primarily non-gastrointestinal infections.

http://www.ncbi.nlm.nih.gov/pubmed/16265204

POLL RESULTS

My least favorite test is the

• CXR 8%

• ESR 23%

• MMPI. They use it to spy on my brain. 20%

• troponin. Always elevated in every disease. 28%

• ABIM recertification. Good money spent to no practical end. 8%

• Other Answers 15%

  • ddimmer

  • CSU

  • colonoscopy

  • sed rate

  • 24 hour fecal fat.

Chapter One Hundred Two

At the End of the Rumsfeld Scale

ID is an odd subspecialty. No one would deliver a baby if obstetrics was available, remove an appendix if a surgeon was around, or treat a myocardial infarction if there were an available cardiologist. But people treat infections all the time without the help of ID.

Which is as it should be. Most common infections are relatively straightforward. If the average hospitalist can't treat a urinary tract infection, a community-acquired pneumonia, or a cellulitis without ID help they need to pack it in and go into retail.

The problem occurs when a disease occurs at the end of the Rumsfeld scale:

"There are known knowns. These are things we know that we know. There are known unknowns. That is to say, there are things that we know we don't know. But there are also unknown unknowns. There are things we don't know we don't know."

~Donald Rumsfeld

It is those tricky unknown unknowns that can throw you for a loop.

Combine that with a touch of Dunning--Kruger:

"The Dunning--Kruger effect is a cognitive bias in which unskilled individuals suffer from illusory superiority, mistakenly rating their ability much higher than average. This bias is attributed to a metacognitive inability of the unskilled to recognize their mistakes."

and you have people treating infections, or mistreating infections, without help and not realizing they need it.

Like Staphylococcus aureus, the bacterium voted most likely to be inadequately treated. If I had a nickel for every time S. aureus bacteremia was mistreated, I could probably purchase a new MacBook Air. Like what happened this week.

The patient is a diabetic, admitted two weeks ago for heart failure. One of two blood cultures at the time were positive for S. aureus. The blood cultures were done as part of a pneumonia evaluation, and the patient was better and discharged less than 48 hours later on no antibiotics, the culture was thought to be a contaminant.

Now she is admitted after 2 weeks of this lack of treatment with fevers and 3/3 positive blood cultures with the same S. aureus. S. aureus is never a contaminant, especially in the blood of a diabetic and when the cultures are positive in less than 24 hours.

S. aureus in the blood is always real and (almost) always treated with IV antibiotics.

Ignored, treated with oral antibiotics or IV clindamycin, S. aureus causes people to flail around under the mistaken assumption that all they need to know in treating an infection is the susceptibility on the lab report.

There are now several studies to show that calling ID for S. aureus bacteremia leads to better outcomes. Calling someone who knows what they are doing is of benefit in patient care? Huh. Who would have figured?

Rationalization

J Infect. 2010 Sep;61(3):197-204. doi: 10.1016/j.jinf.2010.06.001. Epub 2010 Jun 12. Time to blood culture positivity in Staphylococcus aureus bacteremia: association with 30-day mortality.

http://www.ncbi.nlm.nih.gov/pubmed/20547181

Clin Infect Dis. 2013 Feb;56(4):527-35. doi: 10.1093/cid/cis889. Epub 2012 Oct 19. Telephone consultation cannot replace bedside infectious disease consultation in the management of Staphylococcus aureus Bacteremia.

http://www.ncbi.nlm.nih.gov/pubmed/23087397

Chapter One Hundred Three

At the end of the day, safe > > > sorry.

A problem in medicine is trying to prove a negative. Get an echocardiogram to "rule out endocarditis" and you don't rule it out. You exclude a vegetation, but given the false negative rates of both transesophageal and transthoracic ECHO, given the right clinical scenario the safe thing, or perhaps safer thing to do, is treat for endocarditis.

Today's patient had an uneventful laminectomy and then returned 48 hours later with a fever, disorientation, and a peripheral leukocytosis.

And a lumbar puncture, like a guitar solo, is performed. What's in the cerebrospinal fluid?

1970 white blood cells per microliter (<5 data-preserve-html-node="true" is normal)

128,000 red blood cells per microliter (none is normal, but there is bleeding from the surgery)

The differential on the white blood cells shows 74% segmented forms, 7% lymphocytes and 19% monocytes

Glucose 28 mg/dL (normal is > 60% of serum glucose; low levels indicate bacterial or viral meningitis)

protein 498 mg/dL (<45 data-preserve-html-node="true" is normal)

The gram stain shows no organisms, and the cultures are negative. No abscess on imaging. And she got completely better in 24 hours on antibiotics and steroids.

The neurosurgeon says it's probably chemical meningitis, from the blood in the CSF. Really? How can one tell? One being me, and the loneliest number if memory serves.

"We reviewed the records of 70 consecutive adult patients with meningitis after a neurosurgical procedure, to determine the characteristics that might help to distinguish a sterile postoperative chemical meningitis from bacterial infection. The spinal fluid profiles in bacterial and chemical meningitis are similar. The exceptions are that a spinal fluid white blood cell count > 7500/microL (7500 x 10(6)/L) and a glucose level of < 10 mg/dL were not found in any case of chemical meningitis. The clinical setting and clinical manifestations were distinct enough that no antibiotic was administered after lumbar puncture to 30 (43%) of the 70 patients with postoperative meningitis. Chemical meningitis was infrequent after surgery involving the spine and sinuses. Patients with chemical meningitis did not have purulent wound drainage or significant wound erythema or tenderness, coma, new focal neurological findings, or onset of a new seizure disorder. They rarely had temperatures > 39.4 degrees C or cerebrospinal fluid rhinorrhea or otorrhea."

So by the parameters it is closer to chemical than bacterial, and the authors note that response to antibiotics is unreliable. They did not give patients steroids part of the confounding presentation. And what lead to the chemical reaction? No one knows:

"It has been proposed that, after transection of the dura and the arachnoid, substances derived from the breakdown of RBCs or from surgical materials lead to an inflammatory reaction (chemical meningitis). Experimental injection of various substances into the subarachnoid space (including saline solution, air, casein, and blood) has produced a neutrophilic pleocytosis and clinical illness."

But but but. Not treating a bacterial meningitis could be catastrophic for the patient and is a negative culture enough to prove a negative? It makes me leery.

"Kill thy physician, and the fee bestow

Upon the foul disease."

Not the Leary. So I suggested a course of antibiotics lest my patient starts agreeing with Kent.

Rationalization

Clin Infect Dis. 2001 Jan 15;32(2):179-85. Epub 2001 Jan 15. Characterization of chemical meningitis after neurological surgery.

http://www.ncbi.nlm.nih.gov/pubmed/11170905

Chapter One Hundred Four

Perfect World?

No way. I practice in the real world, and there is often a disconnect between what I would do, what needs to be done, and what we can actually do, limited by physiology and patients' wishes. Which is as it should be. As much as I would like the universe to revolve around me and what I want, it is not the case.

Today's patient has multiple medical problems: diabetes, dialysis, mitral, and aortic artificial valves and is admitted with a Non-ST-elevation myocardial infarction (NSTEMI). Some sort of heart thing--I do not really keep up on the cardiology nomenclature.

As an outpatient, she had received moxifloxacin for sinusitis just prior to admit. An ECHO was done as part of the cardiac evaluation, and serendipitously a vegetation was found on both valves along with an aortic ring abscess.

The blood cultures are negative, probably from the prior antibiotics, and there were no good symptoms that suggested endocarditis before admission.

So now what?

The patient does not want surgery.

The surgeons think she is very high risk and are loath to operate. In a perfect world, she would be off to the OR, but, as mentioned above, physiology and preference preclude that intervention.

I have cured two prosthetic valve abscesses in my career, both in patients who had streptococcal ring abscesses and were deemed inoperable. I embarked on an attempt of medical cure with no optimism that I would succeed, yet the antibiotics actually worked.

The literature does have cases of myocardial ring abscesses cured without surgery in some patients:

"Outcome was favorable (mean follow-up 33 +/- 10 months) in all patients with an abscess surface area < 1.5 cm2, no signs of heart failure, no grade III or IV valvular regurgitation, apyrexia after less than 8 days on antibiotics and no staphylococcus positive blood culture."

This mostly describes the current patient. Just because you can get away with an intervention does not mean it is the best intervention, but the real world supersedes the perfect world. So medical therapy it is. I expect failure, but nothing ventured, nothing gained. Or lost.

Postscript

And fail we did; she died of her infection.

Rationalization

Presse Med. 1996 Sep 28;25(28):1276-80. [Outcome of cardiac valve ring abscesses after medical treatment: attempt to identify criteria of favorable prognosis].

http://www.ncbi.nlm.nih.gov/pubmed/8949787

Chapter One Hundred Five

I Love Me a Good Gram Stain

The patient is an older male diabetic admitted with a week of fevers, chills, and malaise. He has had new shortness of breath, and a chest X-ray revealed no pneumonia, just bilateral pleural effusions, so the diagnosis of pulmonary embolism was entertained (dinner and a movie) and a chest CT scan was done.

There was something in the right lobe of the liver. Further investigation demonstrated a probable abscess.

Over a year ago, he had a cholecystectomy and treatment for a peri-hepatic abscess, in the gallbladder fossa. It was treated at Outside Hospital, and it appears he had appropriate drainage and antibiotics for mixed gastrointestinal flora.

He had been completely asymptomatic until a week ago.

Recurrence? A year later seems a bit much for a relapsed hepatic infection, but not outside the limit of my experience. His pathology last time was on the right side, not the left, so it's wrong anatomically for a recurrence, too. New infection? More likely, and having gallbladder issues is a good reason to get a liver abscess.

So IR (baboon? I often add that noun when I hear I.R., and not because of anything the radiologists do; it is one of the many minor forms of PTSD left over from raising my kids). Anyway, Interventional Radiology drains it, and after three days, the cultures are negative.

That would be odd if not for the gram stain. I love the gram stain. Medically ancient, hard to read unless you do it routinely, it is a comforting diagnostic test that let me know there are bacteria there, even if they cannot be grown, and a hint of what may be there. And here the gram stain shows branching gram-positive rods.

It has to be Actinomyces. Just has to be. Not that there are a lot of cases of actinomycosis causing liver abscesses:

"We report a case of isolated hepatic actinomycosis and review 35 previously reported cases. Three-fourths of the reported patients were male, and more than one-half were between 30 and 50 years of age. Although some patients had oral disease or intraabdominal infections, the majority of cases were cryptogenic. Common presenting symptoms included fever, abdominal pain, and anorexia with weight loss. Findings on physical examination included pyrexia, abdominal tenderness, and hepatomegaly. Leukocytosis with a left shift, anemia, an elevated serum erythrocyte sedimentation rate, and an elevated level of alkaline phosphatase were almost universally present. Diagnosis was frequently made at the time of exploratory laparotomy, but percutaneous diagnostic procedures obviated the need for surgery in many recent cases. Microbiological diagnosis involved visualization of branching gram-positive Actinomyces organisms or recovery of organisms in anaerobic culture. Treatment most commonly consisted of prolonged administration of penicillin or tetracycline and was associated with an excellent outcome in the majority of cases."

It could be Nocardia, but as branching gram-positive rods go Nocardia is even rarer, less than a smattering of cases reported.

I usually treat liver abscess with ceftriaxone and metronidazole and Actinomyces can be susceptible to ceftriaxone, but penicillin is the standard and I am a standards kind of guy.

Eventually it went away.

Rationalization

Clin Infect Dis. 1993 Feb;16(2):303-9. Pyogenic liver abscess involving Actinomyces: case report and review.

http://www.ncbi.nlm.nih.gov/pubmed/8443315

J Antimicrob Chemother. 2005 Aug;56(2):407-9. Epub 2005 Jun 21. Antimicrobial susceptibility testing of Actinomyces species with 12 antimicrobial agents.

http://www.ncbi.nlm.nih.gov/pubmed/15972310

Chapter One Hundred Six

I leave for a few days and look what happens

One day of vacation time equals three days of work. Or so it seems. I was gone for 4 days to talk at The Amazing Meeting in Las Vegas, and I get back, and the work that piled up in such a short absence seems so much more than what would have occurred had I been at work. Vacation is like Fairie, it moves at a different rate than the mortal world. I am spending all my reading time in the Dresden Files, so some of you may understand.

Two consults today were cases I have railed about in the past.

The first was a chronic cellulitis. Remember, cellulitis--diffuse erythroderma from bacteria--is never, ever, never, ever, never, ever chronic. Never. Ever.

I cured the alleged chronic infection the same way I always do: I held the leg higher than the heart, and within seconds, the redness is gone. Post-infectious swelling and redness.

Listen up: you cannot tell anything about a cellulitis if the leg is not higher than the heart. The rubor, dolor, calor, tumor of cellulitis does not recede with elevation. The patient was happy to stop the antibiotics and go on his way.

The other was less benign. The patient has methicillin-resistant Staph. aureus bacteremia without an obvious source. Maybe skin, maybe dialysis site. A transthoracic echocardiogram is negative, and she is sent out on two weeks of vancomycin, despite the fact the MIC is 1.0 micrograms per milliliter (the upper end of being considered sensitive) and the patient has both an aortic valve replacement and a mitral valve repair.

Needless to say, the patient relapsed with MRSA bacteremia, and the echocardiogram shows a mitral valve vegetation. The aortic valve replacement looks clean. It isn't. If the patient has endovascular hardware, a prosthetic valve, or a pacer, you have to evaluate and treat as if the endovascular hardware is infected.

"In this investigation, approximately half of all patients with prosthetic valves who developed S. aureus bacteremia had definite endocarditis. The risk of endocarditis was independent of the type, location, or age of the prosthetic valve. The mortality of prosthetic valve endocarditis is high. All patients with a prosthetic valve who develop S. aureus bacteremia should be aggressively screened and followed for endocarditis."

Two weeks of IV therapy is limited to the totally normal people with MSSA.

And call an ID doc for S. aureus bacteremia. An ID consult is beneficial. It improves morbidity and mortality.

"Only one third of patients with S. aureus bacteremia in this cohort had an infectious diseases specialist consultation. Infectious diseases consultation was independently associated with a reduction in 28-day mortality. Routine infectious diseases consultation should be considered for patients with S. aureus bacteremia, especially those with greater severity of illness or multiple comorbidities."

Don't wing it unless your name is Dunning-Kruger.

Look what happens when I go away for even a short time:

"Things fall apart; the centre cannot hold;

Mere bacteremia is loosed upon the world,

The blood-dimmed tide is loosed, and everywhere

The correct diagnosis and treatment is drowned;

The best lack all knowledge, while the worst

Have prescription privileges."

William Butler Yeats. Kind of.

Rationalization

Am J Med. 2010 Jul;123(7):631-7. doi: 10.1016/j.amjmed.2010.01.015. Epub 2010 May 20. The value of infectious diseases consultation in Staphylococcus aureus bacteremia.

http://www.ncbi.nlm.nih.gov/pubmed/20493464.

One of many.

Am J Med.]2005 Mar;118(3):225-9. Risk of endocarditis among patients with prosthetic valves and Staphylococcus aureus bacteremia.

http://www.ncbi.nlm.nih.gov/pubmed/15745719

Chapter One Hundred Seven

Flummoxed

Usually, I/we get to a diagnosis. Usually. Not so far this time. The patient is a middle-aged, well, old, 58 is old, male.

I was in line at Southwest Airlines coming home from Vegas on Sunday, and the young man directing the line turned to me when I was next in line and said, "You are up, young fella." Young fella = old as dirt. And I am a mere fifty-six.

The patient presents with a stroke rendering him hemiparetic and aphasic. Poor guy. About 8 days into his hospitalization he starts to have daily fevers to 104, and his white blood cell count climbed to 50,000 per microliter, almost five times the upper limit of normal.

He received the five-star work up and nothing. No infection, no tumor, no collagen vascular disease, no nothing. He is from south of the border, and so I worried about miliary tuberculosis. It is on the differential for a white blood cell count of 50,000.

My very first consult was a patient who had miliary TB that was diagnosed at, well, autopsy. My second consult had cardiac Aspergillus. Diagnosed at autopsy. Not an auspicious start to an infectious disease career, but let us say I am sensitive to the atypical presentations of miliary TB.

So we biopsied the bone marrow and liver and started four-drug TB therapy, and over the next week the fevers mostly went away, and the white blood cell count normalized.

Response to antibiotics is an untrustworthy way to make a diagnosis. If you have never read Observations on spiraling empiricism: its causes, allure, and perils, with particular reference to antibiotic therapy, get it. It is one of the classics of medicine and should be required reading by all health care providers. Number 6 on the list Fallacies in Antibiotic Therapy is"'Response implies diagnosis."

Because the liver and bone marrow show no granulomas and the cultures are negative at two weeks, I stop the antibiotics. And then after 10 days his fever went up to 104 ºF, in one day his white blood cell count jumped to from 14,000 to 50,000 and the workup was, again, negative. And a day later the fever is gone, and the white count is declining.

I am totally flummoxed as to what is going on in this patient. Which means it is probably a rheumatology process.

Rationalization

Am J Med. 1989 Aug;87(2):201-6. Observations on spiraling empiricism: its causes, allure, and perils, with particular reference to antibiotic therapy

http://www.ncbi.nlm.nih.gov/pubmed/2667357

POLL RESULTS

When I can't figure out a case

• it is probably rheumatologic 22%

• it can't be figured out 6%

• I follow the saying: muddy water, let stand, will clear. 52%

• discharge the patient to a skilled nursing facility near a competitor hospital 9%

• give up. There are some things man is meant to know 5%

• Other Answers 6%

  • consult my peers

  • I ask a colleague for help.

  • Refer to psych

  • I run it by a colleague.

Chapter One Hundred Eight

Sour Rant After Vacation

I just spent nine days in Eastern Oregon for summer vacation. There is no better part of the world for hiking, biking, golf, rafting, or eating.

So you would think that after such an excellent vacation in one of the most beautiful spots in the world, I would come back rested and ready for endless pus. And there are some great cases to write about, the usual unusual collection of infectious diseases that make my workday so much fun.

But no. I'm going to bitch, moan and complain. I sometimes think that no one pays any attention to their ID education, or medical schools and residency programs are just doing a lousy job.

The last follow-up of the day was a coronary artery bypass graft patient, slow to get off the vent due to lung disease, who had a witnessed aspiration about day 7 of the hospitalization and over the next day has lung infiltrates, fevers, and leukocytosis.

He gets put on piperacillin-tazobactam, vancomycin and because it is an aspiration, metronidazole for anaerobic organisms.

There is a simple evaluation to know if you really need to worry about anaerobes: on the basis of dentition, and only on the basis of dentition, would you feel comfortable kissing the patient? There are a lot of other factors that go into the decision to kiss another human on the mouth, and in the hospital, with patients, they all lead to "No." But forget about those other issues. Just the condition of the dentition. If you see meth teeth or their moral equivalent and want to gag, then anaerobes are a consideration. Otherwise, forget about it.

My patient has no teeth. No teeth means no anaerobes. Repeat again. No teeth means no anaerobes. Say it a third time; it is the charm.

Does the patient lose consciousness from heroin or alcohol, and do they have a lung abscess and empyema with breath that reeks of putrefying meat, which you can smell down the hall? Then consider anaerobes.

But in a hospitalized patient, aspiration does NOT NOT NOT NOT NOT NOT NOT NOT NOT NOT NOT NOT NOT NOT NOT NOT NOT NOT NOT NOT NOT NOT NOT NOT equal an anaerobic condition. That is NOT equals.

Read the literature:

Despite painstaking effort, we were able to isolate only one anaerobic organism (nonpathogenic) from this group of patients. The spectrum of aerobes in patients with VAP was similar to that reported in the literature.

and

HAP, VAP, and HCAP are commonly caused by aerobic gram-negative bacilli, such as P. aeruginosa, K. pneumoniae, and Acinetobacter species, or by gram-positive cocci, such as S. aureus, much of which is MRSA; anaerobes are an uncommon cause of VAP.

and

The need for specific anaerobic coverage for CAP is generally overestimated. Anaerobic bacteria cannot be detected by diagnostic techniques in current use. Anaerobic coverage is clearly indicated only in the classic aspiration pleuropulmonary syndrome in patients with a history of loss of consciousness as a result of alcohol/drug overdose or after seizures in patients with concomitant gingival disease or esophogeal motility disorders. Antibiotic trials have not demonstrated a need to specifically treat these organisms in the majority of CAP cases. Small volume aspiration at the time of intubation should be adequately handled by standard empirical severe CAP treatment and by the high oxygen tension provided by mechanical ventilation.

So stop with the anaerobic coverage for hospitalized aspirations. I would like to say it is stupid, but that would be an insult to the truly stupid the world over. But it just flabbers my gaster I still have to whinge about a topic that to my mind is such basic medicine.

And you should see me when I am grumpy before vacation. Maybe I should go back to Sunriver.

Rationalization

Chest. 1999 Jan;115(1):178-83. The role of anaerobes in patients with ventilator-associated pneumonia and aspiration pneumonia: a prospective study.

http://www.ncbi.nlm.nih.gov/pubmed/9925081

Infectious Diseases Society of America/AmericanThoracic Society Consensus Guidelines on theManagement of Community-Acquired Pneumoniain Adults

https://www.ncbi.nlm.nih.gov/pubmed/17278083

Guidelines for the Management of Adults with Hospital-acquired, Ventilator-associated, and Healthcare-associated Pneumonia.

[https://www.ncbi.nlm.nih.gov/pubmed/15699079]{.underline}

Med Oral Patol Oral Cir Bucal. 2015 Mar 1;20(2):e205-10.

Methamphetamine abuse and "meth mouth" in Europe.

https://www.ncbi.nlm.nih.gov/pubmed/25662544

Chapter One Hundred Nine

Premature Closure

The patient is a young female who had a complicated cholecystectomy and a resultant Pseudomonas aeruginosa perihepatic abscess. It was drained percutaneously, and she was sent home on a long course of ciprofloxacin.

After a month she was no better and was sent to me. She was ill-appearing and complained of fevers, sweats, weight loss, and severe fatigue, in bed 16 hours a day. Examination had some mild right upper quadrant pain, and she had a mild leukocytosis.

Must be a residual abscess, and I bet it had become resistant to the cipro. It is an easy mutation for the organism, especially if there is an ongoing collection of infection.

"Ps aeruginosa developed resistance to ciprofloxacin in 25 of 96 infections."

As the Borg said, resistance is inevitable. Sorry, I said, we need to repeat the CT and look for the abscess to drain and culture. I apologized, as the patient has no insurance. And the CT revealed? Nothing. The infection was all gone.

And the day after the CT I got a call from her GI doc who examined her neck--which he does routinely, looking for "Virchow's node," the left supraclavicular lymph node. This node is near the junction of the thoracic duct and the left subclavian vein, and inflammation is one of the first signs of gastric cancer. Her Virchow's node was fine, but he noted a large thyroid and found her to be very hyperthyroid from thyroiditis. That was the cause of her symptoms, not her right upper quadrant.

Totally missed it, in part due to premature closure. I did my usual evaluation but remained focused from the beginning on the right upper quadrant. I really did not broaden my differential beyond the known perihepatic abscess. I usually try to make a conscious effort to consider a differential beyond infection, although this is easier with the more leisurely pace of inpatient medicine. I did not think beyond the problem at hand.

"Premature closure may be the consequence of several biases and plays a central role by preventing physicians to consider other possibilities and remain open-minded. It is also the target of several strategies aiming at reducing error rates, as developed below."

She still needed the CT, but in retrospect, when I see her in the mind's eye, she looked ill, but not infected ill. I know that sounds kind of goofy, but she had a nontoxic worn-out look, like a mother of newborn triplets.

The hardest part of medicine is not the massive amounts of information we have to learn or the vagaries of disease presentation. The hardest part is thinking clearly and recognizing the innumerable ways in which humans do not think clearly. Rational thinking is an unnatural act, and like all unnatural acts, requires conscious effort to do well or even at all.

Although I suspect Rita Mae Brown had it right, "Nothing is unnatural - just untried."

Read the last reference, a sobering reminder of all the ways we go wrong.

Rationalization

Lancet. 1986 Apr 12;1(8485):819-22. Oral ciprofloxacin therapy of infections due to Pseudomonas aeruginosa.

http://www.ncbi.nlm.nih.gov/pubmed/2870313

Arch Pathol Lab Med. 1995 Aug;119(8):727-30. Virchow's node revisited. Analysis with clinicopathologic correlation of 152 fine-needle aspiration biopsies of supraclavicular lymph nodes.

http://www.ncbi.nlm.nih.gov/pubmed/

Swiss Med Wkly. 2012 Oct 23;142:w13706. doi: 10.4414/smw.2012.13706. Diagnostic errors and flaws in clinical reasoning: mechanisms and prevention in practice.

http://www.ncbi.nlm.nih.gov/pubmed/23135902

Chapter One Hundred Ten

My Powers in Action

Or perhaps inaction. As I have noted before, I have a creepy superpower. Every two weeks, I scan hundreds of titles in the ID literature looking for interesting papers to both broaden my education and as grist for the podcast mill.

What is weird is I will learn something new, some medical pearl of which I was not aware, and shortly thereafter, I will see a case. My acquisition of new knowledge changes reality. I hope to see a paper on a Portland ID doc winning Powerball.

You probably recognize the logical fallacies involved in the above paragraphs. It is just the Baader-Meinhof Phenomenon in action---the frequency illusion where after seeing something for the first time, you start seeing it everywhere and thinking it's more common than it is.

Still, my twice-monthly review of ID literature always reminds me that despite 30 years in ID, I still feel like Jon Snow from Game of Thrones.

The patient has a leak after a partial colectomy and develops an abscess. Cultures are obtained and grow the usual hodgepodge of organisms. Methicillin-resistant Staph. aureus (well, maybe not so usual), gram-negative aerobes, and Candida. She is put on vancomycin, ciprofloxacin, metronidazole, and fluconazole, and I am consulted for optimal antibiotics.

Six of one, half-dozen of another, what is the best treatment for an abscess? Drainage and source control. Do that and antibiotics probably have little to do with the final cure, although they certainly prevent the patient from dying of acute sepsis. There is more than one way to skin a cat, or so they say. I have never had a cat stay still long enough for me to try. I did not quibble over the choice of antibiotics, although I tend to be a cephalosporin rather than quinolone kind of guy.

Two days later, the creatinine has gone from .66 to 2.42 mg/dL, indicating kidney damage. Vancomycin, right? The decline in kidney function seemed a bit fast to me to be due to vancomycin and the PubMeds agree:

"Onset of nephrotoxicity ranged from 4 days to 8 days from the start of therapy. The degree of renal dysfunction was modest, with a reported decrease of 35-45% in creatinine clearance from baseline."

Maybe this is too much too soon.

The nice thing about the PubMed era is I do not actually have to remember information. I just have to remember that I should remember something and I had a tickle in the brain. Didn't I read something recently about ciprofloxacin increasing the risk for acute kidney injury?

Yep. A two-fold increased risk in acute kidney injury. In men. Especially if on renin angiotensin-system blockers, which she is not. And a review of the literature suggests, it takes weeks to develop ciprofloxacin renal failure. Not even close; my powers are not what they once were. But it helps to imprint the information into my brain, which at the end of the day, is why I blog and podcast. It is all about me, as it should be.

So in this case, it is probably the usual cause of renal failure in the hospital, namely multifactorial: drugs plus infection plus vascular comorbidities. But I changed the vancomycin/ciprofloxacin to ceftriaxone and daptomycin, although as I write this, I should have gone with ceftaroline. Two stones with one bird and all that. I am getting old.

Pardon me while I log on to Epic and make the change. The pharmacist is going to think I am an intern the way I keep changing antibiotics.

Rationalization

Int J Antimicrob Agents. 2011 Feb;37(2):95-101. doi: 10.1016/j.ijantimicag.2010.10.013. Epub 2010 Dec 3. Vancomycin-associated nephrotoxicity: a critical appraisal of risk with high-dose therapy.

https://www.ncbi.nlm.nih.gov/pubmed/21130609

Risk of acute kidney injury associated with the use of fluoroquinolones. CMAJ 2013. DOI:10.1503

[http://www.cmaj.ca/content/185/10/E475]{.underline}

POLL RESULTS

When I learn new information

• it alters reality 14%

• it is rapidly forgotten 14%

• it cancels out old information leaving no net increase in information 27%

• it is usually something I 'learned' last year 32%

• there is nothing new to be learned, only variations on the past 7%

• Other Answers 7%

  • It increases knowledge that can be useful

  • It sinks into the swamp I call a "mind" only to bubble up years later with a vile-smelling "burp!"

Chapter One Hundred Eleven

Festering Zits

It may be confirmation bias, but at my institutions, the microbiology of surgical infections has changed. Thanks to the Surgical Care Improvement Project (SCIP) and just about 100% hand hygiene, I see fewer and fear infections due to S. aureus. I still see a few cases, but not like the old days, and most are occurring late--week three or four after surgery--suggesting that they are acquired as an outpatient rather than a direct result of surgical inoculation. Anastomotic leaks remain an issue that I think is due to more primary anastomosis now than was done back in the day, but I have no data, only suspicion, that this is the case.

So I suspect while I am seeing fewer infections overall, I am seeing more infections due to the one organism that we can do little to if anything to prevent.

Three months prior to admission, the patient had an emergent repair of an ascending aortic dissection and had done well in the interim. Because there was some descending aorta dissection he had a follow-up CT scan and there was a pseudoaneurysm at the suture line, found serendipitously.

He was taken to the operating room, and there was infection/pus/adhesions around the graft, which was removed and, obviously, replaced. Reading the operation note, it is safe to say I am glad I am not a surgeon. The gram stain was negative for organisms, and ID was called. Hey. You. Yeah You. ID. Get over here and give an antibiotic.

After 5 days several of the op cultures grew Propionibacterium granulosum. No surprise since Propionibacterium, living in the grease of hair follicles, is not killed by either the pre-op scrub or the peri-op antibiotics, and it settles into artificial material to fester silently for weeks and months and years before it becomes symptomatic.

There are no reported graft infections with this organism, in part I am sure due to the difficulty in identifying bacteria. Bacteria are like quantum wave functions that occasionally collapse to give a probability of species, in this case 85% Propionibacterium granulosum. There are a few cases of P. acnes graft infection in the literature.

I doubt the patient would be up for a third repair, so after a course of IV penicillin, I will keep him on lifetime oral Penicillin V Potassium. But I swear I have seen more Propionibacterium deep infections in the last five years or so than the prior twenty-five years. Increased surveillance? Increased awareness? Better cultures? Confirmation bias? Probably a bit of each.

The question to which I have no answer is prevention. Wait, that's not a question. How can P. acnes be prevented?

Rationalization

Vasc Endovascular Surg. 2013 Jul;47(5):394-6. doi: 10.1177/1538574413487262. Epub 2013 Apr 29. Infected Aortic Stent Graft With Propionibacterium acnes.

[https://www.ncbi.nlm.nih.gov/pubmed/23632775]{.underline}

Chapter One Hundred Twelve

Going Up

The patient is a 90-year-old male admitted with altered mental status and hypotension. He had been in the hospital for pneumonia three weeks prior, where he developed C. difficile, and has just finished a course of vancomycin for the diarrhea.

He had a few loose stools at admission, but no abdominal pain, minimal fevers, and the abdominal exam was normal. Over the next 48 hours, the C. difficile was positive, oral vancomycin started, the diarrhea resolved, and the abdominal exam remained normal.

And the white cells started to climb: 10, 20, 40, 55 thousand cells per microliter, where 11,000 is the upper limit of normal. All with no abdominal issues or diarrhea. I have seen a lot of bad C. difficile in my day with marked leukemoid reaction. And to a person, they had tender bellies, thick walls of the colon on CT, and/or toxic megacolon. I had trouble believing there could be such a totally negative abdominal exam and CT with C. difficile driving such a white cell count.

The differential for a leukemoid reaction is narrow: dead bowel, miliary TB, and an infection the size of my ego. Well, a few more things; but that is the big three. He apparently had none of the above. But a quick endoscopy showed lots of pseudomembranes in the colon.

C. difficile can cause leukocytosis without diarrhea or other abdominal symptoms:

"The majority of patients in our hospital who had unexplained leukocytosis had C. difficile infection. Unexplained leukocytosis in hospitalized patients should prompt a search for symptoms and signs consistent with C. difficile infection and a study to detect C. difficile."

And can be a cause of leukemoid reactions:

"The mean peak leukocyte count was 52 +/- 18.2 x 10(9)/L (+/- SD) in the case group and 14.9 +/- 6.5 x 10(9)/L in the control group. Patients with a leukemoid reaction had a lower temperature, a lower serum albumin level, and a higher hematocrit value. Multivariable logistic regression showed respiratory tract infection and lower temperature to be independent predictors of a leukemoid reaction. There were 10 deaths (50%) in the leukemoid reaction group and 5 deaths (7.7%) in the control group. All seven patients with a peak leukocyte count greater than 50 x 10(9)/L died, compared with eight deaths (10.3%) among the remaining 78 patients whose peak leukocyte count was less than 50 X 10(9)/L."

What gets me is the combination of the leukemoid reaction and minimal GI complaints by exam or CT. Maybe it is advanced age blunting the response.

Rationalization

Am J Med. 2003 Nov;115(7):543-6. Clostridium difficile infection in patients with unexplained leukocytosis.

http://www.ncbi.nlm.nih.gov/pubmed/14599633

South Med J. 2004 Oct;97(10):959-63. Leukemoid reactions complicating colitis due to Clostridium

http://www.ncbi.nlm.nih.gov/pubmed/15558922difficile.

Chapter One Hundred Thirteen

Glee

I have to admit I get a little pissy when I get a consult at 2 in the afternoon on a Friday. I have 5 (yes 5) hospitals to cover and hate to have to double back to see consults, but I can't dump them off the weekend coverage. And the consult is often not someone who is acutely and unexpectedly ill, but someone who has been in for days and the attending doesn't get around to telling house staff to call a consult until late in the afternoon. Seriously. It is all about me.

At 2:30, long after I had left the hospital in question, I get called back to see a consult that had been in all week. They wanted to know if I had anything to add to the case. Besides miles on my car and the subsequent increase in my carbon footprint.

The patient was a young female who had been admitted to trauma with a blunt abdominal injury with bowel perforation, which they fixed. At two in the am, the patient had a dose of Dilaudid, got itchy all over, and then her uvula grew. The patient had no fevers or other signs of infection, and in talking with her there was nothing to suggest an infection by history. They treated her with steroids and the growing uvula stabilized.

And then I looked in the throat.

I have never seen anything quite like it. The uvula, and just the uvula, was enormous, the size of my pinky (I wear size 6.5 surgical gloves) that hung down the back of the throat, almost vanishing down the retropharynx like a large worm or a meat stalagmite. It was so amazing I burst out in laughter out of sheer delight at seeing such a uvula.

I apologized to the patient for laughing, but I told her it was just so cool and unusual. She did not take it amiss. And then she showed me a photo on her phone where the uvula curled forward and rested on the posterior third of her tongue, like some deep-sea fish lure. It was delightful. Really.

Maybe I am the only one who delights in curious pathology. I have two large masses of scar tissue on my flanks from where my wife elbows me. I do have a bad habit of staring at pathology in public, and my wife has a pointy reminder that it is not polite to stare, especially at oddities.

It is always the late Friday consult that initially irritates me that ultimately leads to glee at some unique finding or diagnosis. Medicine is always fun.

Rationalization

J Investig Allergol Clin Immunol. 2010;20(1):80-3.

Edema of the uvula: etiology, risk factors, diagnosis, and treatment.

http://www.ncbi.nlm.nih.gov/pubmed/20232778

Chapter One Hundred Fourteen

Euripides is Setting Me Up

The patient is admitted with fevers and mild abdominal pain. After a day, one of the blood cultures is growing Streptococcus milleri and, looking for a source, a CT is done.

Five years ago she had an abdominal aneurysm repair and now, right where it crosses the duodenum, there is a good five centimeters of air in and around the graft, but minimal inflammation.

Not good. But.

The patient has no teeth. No teeth usually means no endovascular infection from alpha-hemolytic streptococci. It is at the interface of teeth and gums that is the presumed access point by which various oral alpha streptococci get into the bloodstream.

It is an old graft, and it would be odd for a graft to be seeded late, especially in the middle of the graft. Usually, the infections occur at the anastomotic site, although I speak from experience, not from data.

Streptococcal graft infections are rare, fewer than two handfuls in the literature.

And the gas. Where did the gas come from? Anyone who has teenage boys at home knows the vast amounts of gas that some living organisms can make. But not S. milleri. Clostridia? S. aureus? E. coli? Candida? All can make gas faster than tissues can absorb it, but not Streptococcus. The gas had to be coming from the bowel due to an incipient aorto-duodenal fistula.

And in the OR that is what they found.

And then I found this today:

"We described seven patients with Streptococcus milleri group aortic (six patients) or vena cava (one patient) graft infection secondary to a vasculo-digestive fistula. Time between vascular graft setting and first clinical signs varied from eight months to more than thirteen years. Six patients had fever. Three patients presented with recurrent fever for more than nine months and in two of these cases, delay before diagnosis was long because repeated blood cultures were sterile. Three patients had abdominal pain and/or digestive haemorrhage. Abdominal CT-scan S. milleri was not contributive for the diagnosis in four patients. Streptococcus anginosus was isolated in four patients, Streptococcus constellatus in three patients. One patient died before surgical management. The other six patients were cured by a surgical management associated with a prolonged antibiotic (lactams) treatment. S. milleri group graft infections are rare (or misdiagnosed) while we found only 4 similar cases in the English medical literature. We conclude that a peri-prosthetic infection secondary to a digestive fistula must be insistently searched (and blood cultures must be repeated many times) in any patient with an aortic (or any other vascular) graft presenting prolonged or recurrent fever or acute digestive symptoms."

There is a smug satisfaction from getting to the correct answer by reasoning. Of course, that just means I am going to get slapped down big time in the near future. Those whom the gods would destroy, they first make smug diagnosticians.

Rationalization

Infection. 2007 Jun;35(3):182-5. Streptococcus milleri group infection associated with digestive fistula in patients with vascular graft: report of seven cases and review.

[https://www.ncbi.nlm.nih.gov/pubmed/17565461]{.underline}

Euripides

http://en.wikiquote.org/wiki/Euripides

Chapter One Hundred Fifteen

A belt and suspenders

Diffuse erythroderma with fevers and leukocytosis is usually Group A Streptococcus. All the medical literature suggests if no abscess, no MRSA, and you do not need to cover MRSA. They need either cefaZOlin (my pronunciation) or ceFAZolin (how the pharmacists mistakenly pronounce it); cephalexin as an outpatient.

"...guidelines do not recommend CA-MRSA coverage for cellulitis, except purulent cellulitis, which is uncommon. Despite this, antibiotics targeting CA-MRSA are prescribed commonly and increasingly for skin infections, perhaps due, in part, to lack of experimental evidence among cellulitis patients (I have a more cynical explanation)...Among patients diagnosed with cellulitis without abscess, the addition of trimethoprim-sulfamethoxazole to cephalexin did not improve outcomes overall or by subgroup."

No one does that, of course. I got a call from the ER about a patient with chronic lymphedema who gets recurrent cellulitis. Cultures are negative. Should be group A Streptococcus, and the patient needs a long-acting beta-lactam and perhaps suppressive penicillin. He was getting vancomycin and was taking dicloxicillin when he had the outbreak. I suggested a more literature-based approach.

The 250 mg twice a day of Penicillin V Potassium suggested in the NEJM article for prophylaxis is a wee bit on the homeopathic side, given the patient populations' BMI. Although we often dose based upon creatinine clearance, many docs to not take into consideration the volume of distribution of some patients, explaining in part why the obese have a higher failure rate for infections. There is an understandable, but irrational, hesitancy to give large amounts of oral antibiotics. Most oral antibiotics are dosed for the 70 kg human, a weight I have not seen since high school.

I recently ran across a diffuse erythroderma on both vancomycin and cefazolin. It is good that rolling eyes is not fatal, as I would have coded and died.

There are times I use vancomycin and nafcillin empirically: epidural abscess and meningitis pending susceptibility, mostly on the basis of studies that demonstrate that patients with MSSA bacteremia who receive vancomycin up-front have poorer outcomes than those who receive a beta-lactam.

"The cases for the case-control study were defined as patients who received vancomycin treatment for MSSA-B; the controls, who were patients that received beta-lactam treatment for MSSA-B, were selected at a 1:2 (case:control) ratio according to the objective matching scoring system and the propensity score system. In the cohort study, SAB-related mortality in patients with vancomycin treatment (37%, 10/27) was significantly higher than that in those with beta-lactam treatment (18%, 47/267) (P = 0.02). In addition, multivariate analysis revealed that vancomycin treatment was associated with SAB-related mortality when independent predictors for SAB-related mortality and propensity score were considered (adjusted odds ratio of 3.3, 95% confidence interval of 1.2 to 9.5). In the case-control study using the objective matching scoring system and the propensity score system, SAB-related mortality in case patients was 37% (10/27) and in control patients 11% (6/54) (P < 0.01). Our data suggest that vancomycin is inferior to beta-lactam in the treatment of MSSA-B."

Given how lousy vancomycin is and how poorly it crosses the blood-brain barrier, I prefer to have both agents onboard for central nervous system infections. The brain is not a forgiving organ when damaged physically. Or emotionally and intellectually for that matter. Most sepsis/bacteremia patients are on a beta-lactam as part of protocols, although not an anti-staphylococcal antibiotic. But for uncomplicated cellulitis? No way.

It is one of the many signs of my growing senescence: I get irritated when antibiotics are not used optimally.

Rationalization

Clin Infect Dis. 2013 Jun;56(12):1754-62. doi: 10.1093/cid/cit122. Epub 2013 Mar 1. Clinical trial: comparative effectiveness of cephalexin plus trimethoprim-sulfamethoxazole versus cephalexin alone for treatment of uncomplicated cellulitis: a randomized controlled trial.

https://doi.org/10.1093/cid/cit122

N Engl J Med. 2013 May 2;368(18):1695-703. doi: 10.1056/NEJMoa1206300. Penicillin to prevent recurrent leg cellulitis.

https://www.nejm.org/doi/10.1056/NEJMoa1206300

Clin Exp Dermatol. 2011 Jun;36(4):351-4. doi: 10.1111/j.1365-2230.2010.03978.x. Epub 2010 Dec 24. Local complications of erysipelas: a study of associated risk factors.

https://doi.org/10.1111/j.1365-2230.2010.03978.x

Antimicrob Agents Chemother. 2008 Jan;52(1):192-7. Epub 2007 Nov 5. Outcome of vancomycin treatment in patients with methicillin-susceptible Staphylococcus aureus bacteremia.

https://www.ncbi.nlm.nih.gov/pubmed/17984229

Chapter One Hundred Sixteen

The world would be a better place if, upon awakening every morning, people would ask themselves, how can I make Mark Crislip's life easier today? Seems a simple thing to me, but evidently it is too much of a challenge for most people most days.

There are numerous ways for that simple wish to be fulfilled, but on a day-to-day basis, it is probably blood cultures that give me the most headaches. One out of two bottles are growing coagulase-negative Staphylococcus, and the patient was placed on vancomycin. Now what? they ask.

So some simple considerations to make my life easier, and by extension, the lives of all ID docs.

If you get called for positive blood cultures, especially those that are growing something round and purple (you know, gram-positive cocci)--before you knee-jerk the vancomycin, repeat the rassen-frassen-frick-frackin blood cultures.

It all gets down to the fact that people get blood cultures without considering the question they are trying to answer. I sometimes think that if you order the damn lab, you interpret it. Especially when I get called after someone sends off Epstein-Barr virus serology for symptoms that have nothing to do with EBV, and I have to tell the ordering doc what the serologies mean.

When you are ordering blood cultures, think about what you are trying to prove. Blood cultures are more than just part of the pan-culture for fever.

There are, to my mind, three questions you might be trying to answer with blood cultures.

Is the patient bacteremic? Then all you probably want is a large volume of blood incubating to catch what is probably a transient and low inoculum condition. 6 liters of blood would be optimal, but the patient would tolerate it poorly. Since it takes around 90 minutes to go from bacteremia to fever, by the time you get the blood cultures you have missed your window of opportunity to catch the organism in the blood and it partially explains why apparently bacteremic patients have negative blood cultures. It is probably best to time the blood cultures draw at an hour and a half before the fever*.

Does the patient have a central line infection? Then draw simultaneous cultures from the line (probably one from each port) and one peripherally. If the line is infected, then that draw has a higher inoculum and the blood culture will turn positive 120 minutes before the peripheral draw.

Does the patient have an endovascular infection, usually endocarditis? Then get blood cultures at least 4 hours apart if clinically safe. The sine qua non of endocarditis is a sustained bacteremia.

So make my life easier. Think about why you are ordering vancomycin or blood cultures, then act in a manner that will make me happy.

*Yes. I know. At least you thought about it.

Rationalization

Clin Infect Dis. 2010 Jun 15;50(12):1575-9. doi: 10.1086/652766. How many lumens should be cultured in the conservative diagnosis of catheter-related bloodstream infections?

https://doi.org/10.1086/652766

J Clin Microbiol. 2008 April; 46(4): 1381--1385. Timing of Specimen Collection for Blood Cultures from Febrile Patients with Bacteremia

http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2292961/

Chapter One Hundred Seventeen

BMD with Collateral Damage

The amount of damage bacteria can do to the host can be quite remarkable, especially given their size relative to human cells. Of course, much of the damage, depending on the organism, may be due more to the host response than the bacteria. Polymorphonuclear leukocytes have their own destructive ability, and the tissues next to infection are not always treated kindly by white cells.

The patient is a young IV drug abuser who has progressive shortness of breath, to the point where she cannot even stand without dyspnea, and thence to the ER where wide open aortic insufficiency was found with vegetations destroying the valve.

She had an emergent valve repair, and the findings in the OR were remarkable for the destruction. Not only was the valve destroyed, but there was a large bifurcated abscess in the annulus that had to be patched with Dacron before the prosthesis could be placed.

Not the best of clinical scenarios. The gram stain showed gram-positive cocci and I expected S. aureus, perhaps a S. milleri group given the abscess and destruction, although I was surprised at the complete lack of emboli on exam and the relative lack of constitutional symptoms.

It was finally identified as a pan-sensitive S. epidermidis. Did not see that coming. Coagulase-negative staphylococci can be a cause of native valve endocarditis, and the IV drug abuse would be a good way for the infection to gain access to the blood and then to the bicuspid aortic valve.

I do not see a lot of native valve endocarditis due to coagulase-negative staphylococci. I think this is my first this century that was not a S. lugdenesis. About half the time coagulase-negative endocarditis is due to nosocomial infection, and most of the time it is S. epidermidis.

"Of the CoNS NVE cases with an etiologic agent specified, 55 (85%) of 65 were attributed to Staphylococcus epidermidis; while the remainder of cases were due to Staphylococcus hominis (4), Staphylococcus lugdenesis (3), Staphylococcus capitis (1),Staphylococcus capris (1), and Staphylococcus simulans (1)."

And it is a bad disease given the relative lack of virulence.

"Heart failure was reported in 49 (49%) of 99 patients, and intracardiac abscess was reported in 15 (15%) of 99 patients. More than half of the patients (53 of 99 54%) underwent surgery during the index hospitalization. The in-hospital mortality rate was 19% (19 of 99 patients)."

I assume the infection had been festering a long time to result in such a BMD (bacteria of mass destruction) and it caused few constitutional symptoms being part of the normal flora.

Usually, valve replacement is curative for endocarditis but the heart is not an organ that tolerates a wide debridement, and the worry is that the repair and valve may have been placed into active infection. So I am going to treat as if prosthetic valve endocarditis.

She, like so many, was lost to follow-up and heroin.

Rationalization

Clin Infect Dis. 2008 Jan 15;46(2):232-42. doi: 10.1086/524666. Emergence of coagulase-negative staphylococci as a cause of native valve endocarditis.

http://www.ncbi.nlm.nih.gov/pubmed/18171255

Clin Infect Dis. 2004 Nov 15;39(10):1527-30. Epub 2004 Oct 27. Native valve endocarditis due to coagulase-negative staphylococci: report of 99 episodes from the International Collaboration on Endocarditis Merged Database.

http://www.ncbi.nlm.nih.gov/pubmed/15546091

Chapter One Hundred Eighteen

A Quadruple Whinge

I like the term whinging. It is English, and I first heard it listening to Ricky Gervais. This year marks 30 years out of medical school, and I am starting my 24th year in practice. The old get the right to whinge at length on any topic they want. Now that summer is over, and the clouds and rain have moved in to match my mood, I am going to whinge.

Whinge 1

I want a curious doctor. One who asks why to everything, who has to see and poke all the pathology that comes their way. Curiosity does not seem to be an important characteristic in many docs, sometimes to the point where it flabbers my gaster.

It was a weekend evening, and I was out and about when I get a call from one of the seven hospitals I cover. A (non-internal medicine) hospitalist called me. According to the radiology report, one of her patients had pneumonia, and she wondered what to do about it.

Well, what does the film look like? I asked. What kind of pneumonia is it?

She replied she was not a radiologist and does not read X-rays.

Besides the fact that a career of looking at abnormal and normal X-rays is an excellent way to develop an understanding of radiographic variations in human anatomy and pathology, wouldn't you at least be a wee bit curious? I look at every X-ray I order and most of the ones that I do not. Always have and always well, if for no other reason that I am curious what the X-ray looks like. Kids today. They just do not have what it takes.

Whinge 2

There needs to be a handout to patients on how to give a history. I am interested in your symptoms: how they started, how they progressed, their current status. I do not care about the route you took to the hospital, what your labs and X-rays showed (it is in the chart) and above all I do not care about the details of your diet. It is amazing how people will spend scant seconds on the history of present illness when given an open question, but will go on for what seems like forever about the minutia of what they have eaten in the last few days.

Whinge 3

Yesterday the NEJM, as a gift for resubscribing, sent me a circular mouse pad with their logo on it. Who in hell needs a mouse pad in the era of optical mice and trackpads?

And what about the carbon footprint to produce this worthless and unneeded piece of plastic and send it to me? In a plastic bubble wrap envelope, as if a rubber trackpad can break. And for $2.24 in postage. Next time, why not spend the money on something of social and medical value? Donate to African vaccines. Give money to the Red Cross. Hire some editors that understand supplements and complementary and alternative medicine (which I call SCAM). Do something, anything, useful and responsible with that money.

A couple of years ago, when they sent the last mouse pad, I sent them a letter with the same suggestions, more respectfully phrased. I am not holding my breath this time.

Whinge 4

I used to think ID docs were the smartest docs. Now, I am not so sure. Survey says...

http://montypython.joolsc.net/downloads/sounds/COMPLAIN.WAV

Chapter One Hundred Nineteen

This Little Piggie

The patient is a middle-aged male who had a week of fevers, night sweats, malaise and went to the ER. Work up showed a normal complete blood count and composition, but the CT showed some vague nodules in the lung and liver, maybe a malignancy.

No risk factors for much of anything, infectious or otherwise. He was sent home, and 48 hours later, his blood was growing a gram-negative coccobacillus. HACEK of some sort? Maybe.

Next day the lab says it is a probably Brucella. Didn't see that coming. In 23 years, I have seen two other cases of Brucella. Oregon is not, as a rule, a place where I think of Brucella.

Or is it?

Because it was identified as B. suis and evidently the powers that be can get a little hinky about Brucella:

"Classification of the United States as free of swine brucellosis is based on surveillance of domestic swine populations. However, many states classify swine into 3 categories: domestic swine that have no contact with feral swine, transitional swine that might have contact with feral swine, and feral swine. If a domestic swine herd is infected with B. suis, many states then reclassify that herd as transitional. Therefore, classification of a state as free of swine brucellosis does not mean that transitional herds or infected feral swine do not exist. Our results indicate possible underestimation of the role of feral swine in the sylvatic transmission of B. suis."

We have feral swine in Oregon, although not many, and there are some in the Willamette Valley. It is a growing problem across the US and, like many issues in Oregon, imported from California. As best I can tell it has been forever since there has been an endogenous case of B. suis in Oregon. It may be a potential problem in your state even if you are not in the South.

Where did the patient acquire the Brucella? He occasionally purchases a local pig that he slaughters for a pig roast. The investigation is starting to investigate the provenance of the pigs.

And therapy? As this is an intracellular organism with a high relapse rate, I tend towards "more is better" and "longer is better." Six weeks of doxycycline-rifampin and a week of gentamicin looks best to me.

Postscript

Eventually they sequenced the Brucella and found it to be closely related to Brucella from his home island in Polynesia. So it was likely reactivation, years after leaving an endemic area. See the third reference.

Rationalization

Volume 17, Number 12---December 2011 Brucella suis Infection in Dogs, Georgia, USA

http://wwwnc.cdc.gov/eid/article/17/12/11-1127article.htm

Trop Doct. 2012 Jan;42(1):13-7. doi: 10.1258/td.2011.110284. Doxycycline-rifampin versus doxycycline-rifampin-gentamicin in treatment of human brucellosis.

[http://www.ncbi.nlm.nih.gov/pubmed/22290107]{.underline}

Identification of Source of Brucella suis Infection in Human by Whole-Genome Sequencing, United States and Tonga

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4696693/

Chapter One Hundred Twenty

Pain

I saw two patients in the last week with pain. Well, more than two. Pain is a common presentation for many processes, and recognizing pain patterns is often key to knowing the likely diagnosis. Pain and wives/girlfriends are the two issues that drive men to the hospital.

The first was groin pain. I have mentioned this before, and I mention it again, as I want someone who is not me to study this comprehensively. As best I can tell with my Google-fu, the following has yet to be reported in the real medical literature.

The patient has recurrent lower extremity cellulitis as a consequence of lymphedema. And before his leg gets red, hot, swollen, and tender, he has 12 hours or so of severe inguinal pain.

I think it is lymphadenitis, although I have yet to find a swollen node on exam. Inguinal pain is a sentinel complaint for streptococcal cellulitis in some patients. I just need someone to publish it. And name it after me. Crislip's sign.

The other patient presents with severe back pain, fevers, hypotension, and rigors for half a day. A dialysis patient he has the usual systemic inflammatory response syndrome (SIRS) workup that is initially negative, including an MRI of his spine and the blood cultures grow...

Staphylococcus aureus.

Of course. The echocardiogram is negative, there is no discitis, osteomyelitis, epidural abscess or pyomyositis, and the back pain resolves rapidly with antibiotics.

Severe back pain without demonstrable pathology is not an uncommon presentation of S. aureus bacteremia.

"During the 12-year period, there were 106 cases, for a prevalence of 0.34 per 1000 admissions. Ninety-three (87.7%) of these patients were seen by me. The patients' ages ranged from 12 to 83 years (median, 61 years). Eighteen cases were nosocomial (15 were associated with intravascular catheters). Twenty-one patients were injecting drug users. Severe back pain was the chief complaint in nine patients. "

I am uncertain why. A vasculitis, I suppose, or a false negative MRI, although the latter is extremely rare. It seems a common enough presentation of S. aureus bacteremia that I can predict the pathogen in the right clinical situation. Impresses the hell out of people, let me tell you.

For those of you new in practice, a word of advice. Do not blather on with an extended differential without committing to a diagnosis and/or organism. No one will remember if you predict with authority a weird infection or organism and are wrong. People are wrong all the time in medicine. But if you are right, they will remember you forever as a genius.

Rationalization

Arch Intern Med. 1994 Oct 24;154(20):2330-5. Staphylococcus aureus endocarditis at a community teaching hospital, 1980 to 1991. An analysis of 106 cases.

http://www.ncbi.nlm.nih.gov/pubmed/7944855

Chapter One Hundred Twenty-One

A Surfeit of Diseases, a Paucity of Letters

The patient is an IV drug abuser with Staphylococcus aureus in the blood, MRSA, MIC to vancomycin of 1.0 microgram per milliliter. Common enough problem for my practice. The examination yields nothing to suggest endocarditis, but he has pain, swelling, and tenderness in the left upper chest.

I go to the hospitalist in charge and tell him that it looks like the SC is infected.

Subclavian? He asks with a raised eyebrow. How can you tell?

I laughed. No, the sternoclavicular joint. I think it is infected. The pain and swelling are right over the joint, and IV drug abusers are one of the groups that can get infections in that joint. If so it needs an incision and drainage.

That is a problem in medicine. Letters and diseases mean different things in different contexts. Sometimes I am asked if I am the I and D doctor. Nope. That would be the surgeon.

So we get a CT of the area and call the surgeons to have the joint cleaned out, and what to you know? The joint was not infected, but there was inflamed clot in the subclavian. An ultrasound confirms clot and air in the vein extending into the upper arm.

So it is septic thrombophlebitis. While not uncommon, it usually occurs near where the veins are being accessed, and he has not been using the veins in the upper arm, not yet having to go above the antecubital fossa for heroin injection.

But it is not called coagulase-positive Staphylococcus for nothing. S. aureus loves to clot blood, which is part of why it is virulent. I love the language in older journals. From 1947:

"The pathogenicity of staphylococci exhibits certain features peculiar to this organism which have never been adequately explained, such as the tendency to the production of circumscribed local lesions, the inadequacy of the immune response to infection, the difficulty of artificial immunization, etc. Out interest in staphylocoagulase was aroused in the first instance because circumstantial evidence suggested it might be a major factor in the initiation, and possibly also in the later development of staphylococcal infections... The results do show that the initial infectivity of staphylococci is largely dependent upon their power of coagulase production, so that without this power the other virulence factors are probably of little avail."

The whole article is online and well worth your time. Unfortunately, we do not write like that in journals anymore. It is why it should no longer be referred to as the medical literature. No literature in what I read.

Once we make sure his valve has no vegetations, it will be a course of antibiotics and anticoagulation.

Rationalization

Swiss Med Wkly. 2002 Jul 13;132(27-28):386-92. Septic deep venous thrombosis in intravenous drug users.

http://www.ncbi.nlm.nih.gov/pubmed/12428193

PLoS Pathog. 2010 Aug 5;6(8):e1001036. doi: 10.1371/journal.ppat.1001036. Contribution of coagulases towards Staphylococcus aureus disease and protective immunity.

http://www.ncbi.nlm.nih.gov/pubmed/20700445

Br J Exp Pathol. 1947 Feb;28(1):57-67. The role of coagulase in staphylococcal infections.

http://www.ncbi.nlm.nih.gov/pubmed/20290308