Monday, July 23, 2012

The Wrong Way To Quit Drinking

This week's NEJM Case Records is an Emergency Department & Toxicology patient – I won't ruin the final diagnosis for you – who uses a mysterious South American alcoholism cure to attempt to rehabilitate himself after a night of heavy drinking.

Obviously, it wouldn't be a case report if the patient in question didn't end up in the ICU!  There are nice, educational discussions of several toxodromes, antidotes, and various options for preventing end-organ damage in a subset of uncommonly seen poisonings.

"Case 22-2012: A 34-Year-Old Man with Intractable Vomiting after Ingestion of an Unknown Substance"

Friday, July 20, 2012

Empiric Measurement of Bias in Unblinded Trials

This lovely article was passed along to me by David Newman during a discussion of IST-3 – the recently infamous, massive randomized trial of thrombolysis for acute stroke.  There are two ways of thinking about IST-3, and how the results are viewed in the literature seems to depend how much funding you receive from Boehringer or Genentech.  The first way of thinking seems to be accept the results as published, pick apart the subgroups, do statistical contortions, and then either come out in the "pro" camp (Boehringer) or the "con" camp.

The second way of thinking, supported by this article, is "garbage-in, garbage-out".  The key issue for this approach is that IST-3 is an unblinded, open trial, which introduces bias - treating clinicians and patients who believe TPA is a "promising, yet unproven" treatment (from the uncertainty principle of the study) are perceived as more likely to contribute to favorable reported outcomes when receiving the experimental intervention.  This effect is probably even more pronounced given that much of the follow-up scoring for the Oxford Handicap Scale was performed by mail-in questionnaire, rather than standardized expert evaluation – which has rather poor kappa to begin with.

Page three of this article delves into the empiric analysis of the impact of blinding, and the relative likelihood of unblinded trials to report favorable outcomes.  Essentially, the relative chance of reporting both favorable and unfavorable outcomes are significantly affected.  In clinical terms, this leads to presentation of results in which the benefits are exaggerated and the harms are minimized.  In the context of IST-3, this essentially means the likelihood of any hidden positive effects vanishes, while the poor outcomes are underreported – and it's more "negative" than "neutral".

The authors also note they are preparing a systematic review of trials with blind and non-blind outcome assessors, which would be particularly apt to IST-3, as well.

"Blinding in Randomized Clinical Trials: Imposed Impartiality"

Wednesday, July 18, 2012

Keeping Children Happy

When I started in medicine – hardly long ago – Child Life, if it existed at all in the Emergency Department, might have consisted of a few plastic toys and perhaps a Nintendo Entertainment System.  Now, the staple of every department is an iPad, filled with apps and distractions for children.

This is a short article from the Pediatric literature reviewing a few cases in which tablet computers proved useful, along with a review of several apps worth loading on for distraction during potentially troubling procedures.  Most of the apps reviewed are for iPad, but equivalent exist for Android devices and iPhone.

I've definitely gotten mileage out of the movie "Toy Story 3" on my iPhone – perfect for the 3 AM laceration repair when Child Life has gone home for the night.

"Using a Tablet Computer During Pediatric Procedures - 
A Case Series and Review of the 'Apps'"

Monday, July 16, 2012

Massive Overtesting for Febrile Seizures

Frightening, yet benign, febrile seizures are seen frequently in the Emergency Department.  The American Academy of Pediatrics recommends minimal evaluation for uncomplicated febrile seizures, and invasive testing only in complex cases or those with other indications for testing.

Despite this, the real-world experience documented by these authors at a community hospital in New York is slightly different.  Rather than minimal testing, 100% of patients - mostly aged greater than 12 months - received a CBC and Chem7.  94% received a blood culture, 94% received a urine culture, and 85% had a chest x-ray.  24% had CSF cultures and 21% had CT scans - mostly the complex febrile seizures.

The yield of all this testing - they diagnosed a few UTIs, and one blood culture grew out salmonella.  The authors appropriately feel this testing strategy is excessively wasteful - and confirms the AAP recommendations.

"Current Role of the Laboratory Investigation and Source of the Fever in the Diagnostic Approach"

Friday, July 13, 2012

It's Nothing Like "ER"

This is a fun little article regarding the realism of the Emergency Medicine environment showcased on the popular U.S. television show "ER".  As the authors state in their introduction, the viewers of the show have been surveyed, and a significant portion of the viewers believe the content of the show to be valid clinical information.  However, the televised outcomes are frequently unrealistic (CPR success rates, patients emerging from comas, etc.), and lead to inaccurate public perceptions.

This team of authors watched all 22 episodes from a single season of "ER" to evaluate the types of patient encounters depicted, and then compared their findings with representative data from the NHAMCS dataset.  Overall, there were 192 patients during the 22 episodes, and they differed from the real-world by:
 - Weighted heavily towards 25-44 years of age, rather than infants and elderly.
 - More male and white, rather than black and female.
 - Depictions of lower pain levels.
 - Far more traumatic injuries.

And, this analysis only observed the patients - the responsibilities and skills of the treating medical students, residents, and attendings are also wildly dramatized, of course.

So, it's nothing like "ER".  It's really more like "Scrubs"....

"ER vs. ED: A comparison of televised and real-life emergency medicine."

Wednesday, July 11, 2012

Flumazenil - Seizures, But Not Frequently

It seems as though, when teaching trainees about benzodiazepine overdose, flumazenil is discussed - and in the same breath, the commandment to use it never or with extraordinary caution.  The fundamental issue is whether an underlying pro-convulsant state can be unmasked if the protective effect of benzodiazepines is removed.

The answer from this study, a retrospective review of a decade of flumazenil use in California, is clearly yes.  However, the "yes" is only 13 seizures out of 904 reviewed cases, most of whom had some sort of co-ingestant that contributed to the pro-convulsant state.  The authors also note, for the cases in which data was available, flumazenil was therapeutic (and potentially diagnostic as well) in 53% of administrations, with return of alertness from unresponsiveness or drowsiness.

So, the answer to the clinical question - whether flumazenil use should be as taboo as current dogma - is more complex, and, unfortunately, descends into that dark area where risks must be weighed against benefits.  Is the risk of poor clinical outcome secondary to resuscitative efforts in the field, delayed/missed intubations, etc. greater than the 1-2% risk of seizures?  Or can the patient be safely observed with minimal intervention in a monitored setting?  Or, if flumazenil is effective, how much money was saved by reducing the need for the expansive medical testing performed on unresponsive individuals?  I don't believe a single blanket answer suffices to cover each individual clinical situation.

"A poison center's ten-year experience with flumazenil administration to acutely poisoned adults."

Monday, July 9, 2012

NICE Agrees - No PPI in UGIB

It's hard to fight this battle in the United States.  It's like hyperkalemia - where you carefully talk down the rotating IM intern from giving albuterol, terbutaline, bicarbonate, insulin, Kayexalate, and calcium to the K+ of 5.7 in your dialysis patient - and then the nephrology fellow on-call tells 'em to give it anyway.  Sigh.

But, in any event, despite the lack of evidence for benefit in patient-oriented outcomes for intravenous proton-pump inhibitors in UGIB, invariably the GI fellow wants it.  There's even a suggestion of harms associated with IV PPIs in some of these studies - in addition to everything we're learning about how gastric acidity contributes to the total body immune defense.  For all its criticisms, I think NICE - the clinical effectiveness consensus group in the United Kingdom - has gotten it right for UGIB.  Terlipressin, which isn't available in the United States, appears to be beneficial in variceal bleeding.  Somatostatin analogues, not included in this guideline, may or may not be beneficial, and I agree that it was appropriate for them to be excluded.

In the meantime, I'll keep fighting the inanity, one patient and one resident at a time....

"NICE clinical guideline 141 - Acute upper gastrointestinal bleeding: management"

Friday, July 6, 2012

Put Hydroxyethyl Starch Away

The use of colloid solutions as volume expanders is tempting - massive crystalloid resuscitation suffers from third-spacing, limiting the practical intravascular volume provided.  Colloid resuscitation, in theory, uses the oncotic pressure of the solute to favor intravascular expansion.  One of the alternatives that I'd seen use, but was unaware it was widely used, are hydroxyethyl starches.  Earlier studies, at least, the ones I was familiar with, linked the high-molecular weight HES to renal failure.

This trial, from Denmark, evaluated a low-molecular weight HES (Tetraspan) with Ringer's acetate resuscitation in an intensive care setting, enrolling patients diagnosed with sepsis in need of fluid resuscitation.  The trial was randomized and blinded, with the resuscitation fluids being hung in identical black bags.  Each enrolled patient could receive up to 33 mL per kg ideal body weight of the trial fluid, and additional fluid was unmasked Ringer's acetate.

With 800 well-matched patients between groups, 51% of the HES group was dead at 90 days, compared with 43% in the Ringer's acetate group (RR 1.17, CI 1.01 to 1.36).  Renal replacement therapy was needed in 22% of the HES group, compared with 16% of Ringer's acetate group - and was a predictor of death.

Investigators did not see any particular fluid volume advantage to the HES solution, and the toxic effects of the hydroxyethyl starch molecules, unfortunately, were associated with greater morbidity and mortality.

Seems like another great-sounding idea that needs to be rapidly curtailed until a better safety profile and outcome benefit can be demonstrated.

"Hydroxyethyl Starch 130/0.4 versus Ringer’s Acetate in Severe Sepsis"

Wednesday, July 4, 2012

Could Ordering Reprints Help You Get Published?

Medical journals, to a certain extent, require independent sustainable business models.  The full-time editorial staff, the administrative personnel, and the printing costs must be defrayed by elements such as advertising, subscription fees, or other largess.  One of these sources of largess - particularly for journals with high impact factors - is the ordering of reprints.  After gifts, the major promotional material circulated by pharmaceutical companies among physicians is reprints of publications.

This recent study in the BMJ queried the most prominent medical journals regarding their reprints, hoping to gauge the scope of the reprint requests, as well as the financial windfall these might represent.  JAMA, NEJM, and Annals of Internal Medicine all declined to provide data, so these authors were left with the Lancet and the BMJ family of journals.  Of the most-frequently reprinted articles in these journals, they were far more likely to be industry-sponsored, and represented significant sources of income for the journals - up to a $2.4 million USD order from the Lancet.

There are significant limitations to this study, but, clearly, the revenue stream from reprints may be substantial enough that it may further influence and bias the publication of medical literature.

"High reprint orders in medical journals and pharmaceutical industry funding: case-control study"

Monday, July 2, 2012

Warfarin and tPA Mix - If They're Subtherapeutic

These authors almost have a conclusion I can't quibble with - but, rather than "Among patients with ischemic stroke, the use of intravenous tPA among warfarin-treated patients (INR ≥1.7) was not associated with increased sICH risk compared with non-warfarin-treated patients" I would add the caveat to say "after multiple adjustments".

This is a retrospective registry review published in JAMA, comparing the rate of sICH in warfarin-treated patients with non-warfarin-treated patients who received tPA for ischemic stroke.  And, 5.7% of warfarin patients developed sICH vs. 4.6% in the non-warfarin group.  However, after adjustments for multiple variables - the warfarin group tended to be older, had more previous strokes, and had higher NIHSS - the OR was 1.01.  Not terribly surprising there wasn't much difference, considering the mean INR in the warfarin cohort was only 1.2.  Their confidence intervals start getting very wide above 1.6, but there's suggestion of a clear association with increasing sICH as the INR increases.

There are plenty of reasons not to give tPA, but subtherapeutic warfarin use probably should not exclude patients from consideration.

"Risks of Intracranial Hemorrhage Among Patients With Acute Ischemic Stroke Receiving Warfarin and Treated With Intravenous Tissue Plasminogen Activator"

Friday, June 29, 2012

Xigris Isn't Dead - Just Hibernating

Activated Protein C, also known as Xigris, which has had an infamous and circuitous career of sorts, is back.

After a short life of use in severe sepsis, the continued investigations into its efficacy have finally been unable to establish its benefit.  Although many expensive therapies without conclusive benefit are still in use in medicine, we'll score this one (belatedly) for the good guys.

This early animal research, published as a letter in Nature Medicine, reports on interventions targeting the aPC pathway to prevent lethal radiation injury to hematopoietic cells.  They say that starting infusions of aPC within 24 hours of lethal radiation exposure mitigated radiation mortality in mice.  Probably quite a long way off for real-world usage, but any potential treatment is better than none.

"Pharmacological targeting of the thrombomodulin–activated protein C pathway mitigates radiation toxicity"

Wednesday, June 27, 2012

Failings of Modern Medicine

A brilliant piece that eloquently states many of the ideas espoused on this blog, focusing on pulmonary embolism as the poster child for over-testing, over-diagnosis, and lack of sound evidence underlying treatment.

These authors, in the Archives of Internal Medicine, accurately describe the chimeric nature of pulmonary embolism - historically described as a dreaded disease, diagnosed clinically from the manifestations of pulmonary infarction, to the modern manifestation of filling defects noted on CTA during an episode of pleuritic chest pain.  They discuss the handful of patients who benefited from the first heparinization for treatment, and argue the disease for which anticoagulation is the treatment is not the disease we are diagnosing today.

This article covers so many excellent points, and ties the clinical problems so tightly into the underlying principles, that it's almost the sort of must-read article to which medical students should be exposed - in order to bring about that frightening moment of maturity in medicine in which you realize the emperor is distinctly lacking in clothes.

Lovely work!

"The Diagnosis and Treatment - of Pulmonary Embolism: A Metaphor for Medicine in the Evidence-Based Medicine Era"

Monday, June 25, 2012

Impaled in a Rowing Accident

This article I dredged up from the archives is mostly of sentimental value - although, I could claim it's related to Olympic sport-related trauma with the upcoming Games.

This is from the series "Case records of the Massachusetts General Hospital", which run the gamut all the way out to some of the most esoteric diagnoses possible.  This particular article describes the management and outcomes of a man impaled by a rowing shell while on the Charles River.  Eight-person rowing shells are ~17 meters in length, have a crewed weight of nearly 1,000 kg, and travel fast enough that a water skier may be towed behind.  There is a small rubber bumper affixed to the, otherwise sharp, wooden or carbon-fiber bow that is meant to reduce the potential for injury in event of a collision.  In this incident, the momentum of a head-on impact dislodged the bow ball and resulted in the unfortunate impalement incident described.  A fascinating little read.

Rowing collisions are uncommon, injuries are rare, and this is probably nearly unique.

"Case records of the Massachusetts General Hospital. Case 10-2007. A 55-year-old manimpaled in a rowing accident."