It’s that time again. Sure, there may be a lot to do during the month of December, but what better way to procrastinate than to grab a mug of hot chocolate, sit down in front of the fire, and read about some emergency medicine evidence based medicine….
(If that doesn’t sound appealing, you could toss in some earphones while you do your holiday shopping and listen to me and Casey ramble about these papers in the audio version on the BroomeDocs podcast.)
Can Dalton’s laws help dying patients?
Sakles JC, Mosier JM, Patanwala AE, Arcaris B, Dicken JM. First Pass Success Without Hypoxemia Is Increased With the Use of Apneic Oxygenation During Rapid Sequence Intubation in the Emergency Department. Academic emergency medicine. 23(6):703-10. 2016. PMID: 26836712
This is a single center observational study that looked at 635 adult intubations (done by resident physicians) in patients starting with an oxygen saturation ≥90%. They used a classic RSI technique, which means no positive pressure breaths were given during RSI unless necessitated by hypoxia. At the same time that this study was started, the teaching at this site was changed to suggest using apneic oxygenation routinely. Data was by self report post intubation. Over the study period, 380 patients were intubated with apneic oxygenation in place and 255 without. Apneic oxygenation was not standardized, with almost 30% of patients using less than 15 L/min (my suggestion). The primary outcome, first pass success without the oxygen saturation falling below 90%, occurred in 82% of the apneic oxygenation group and 69% of the no apneic oxygenation group (absolute difference 13%, 95%CI 6-20%). Ignoring first pass, the overall hypoxia rate was 13% with apneic oxygenation and 30 % without (difference 7% 95%CI 1-13%). Of course, these were non randomized groups, and there were differences between them in terms of the indication for intubation, the incidence of blood in the airway, the level of the learner, and the device used for intubation. Some of this could be personal preference, but a systematic choice to avoid an unfamiliar technique in sicker patients would also explain these results.
Bottom line: Apneic oxygenation makes sense to me, and these results can be used to support it’s use, but they are far from definitive.
Ask the makers of snake oil to prove its value and the result is modern evidence based medicine
Ioannidis JP. Evidence-based medicine has been hijacked: a report to David Sackett. Journal of clinical epidemiology. 73:82-6. 2016. PMID: 26934549
Evidence based medicine is confusing. Every month people send me papers that seem to contradict others I have included in the articles of the month. In large part, this is because the answer in life, science, and evidence based medicine is almost always “it’s complicated”. This is a beautifully written essay that also acted as a eulogy for the amazing Dr David Sackett that highlights some of the many difficulties we face when trying to practice efficiency based medicine in the modern era – primarily the result of industry money ‘hijacking’ the system of science creation. Some of his words on industry involvement in EBM:
“…the EBM movement has been hijacked. Even its proponents suspect that something is wrong. The industry runs a large share of the most influential randomized trials. They do them very well, they score better on “quality” checklists, and they are more prompt than nonindustry trials to post or publish results. It is just that they often ask the wrong questions with the wrong short term surrogate outcomes, the wrong analyses, the wrong criteria for success (e.g., large margins for noninferiority), and the wrong inferences, but who cares about these minor glitches? … corporations should not be asked to practically perform the assessments of their own products  . If they are forced to do this, I cannot blame them, if they buy the best advertisement (i.e., “evidence”) for whatever they sell.”
Bottom line: All of us need to be vocal about changing the system that relies on pharmaceutical companies to test their own products and then accepts the results of that ‘science’
Now if only we could actually get an urgent MRI
Ray JG, Vermeulen MJ, Bharatha A, Montanera WJ, Park AL. Association Between MRI Exposure During Pregnancy and Fetal and Childhood Outcomes. JAMA. 316(9):952-61. 2016. PMID: 27599330
This is a database study from Ontario looking at almost 1.5 million babies born over a decade. The rate of MRI during pregnancy was 4 in 1000. MRI during the first trimester was not associated with stillbirth or death (RR 1.68 95%CI 0.97 – 2.9). It was also not associated with congenital abnormalities, neoplasm, vision, or hearing loss. Gadolinium use was associated with an increase in stillbirths and death (adjusted RR, 3.70; 95% CI, 1.55 to 8.85) and rheumatological, inflammatory, or infiltrative skin condition. Of course, this is association not causation, and women getting MRIs are probably sicker than those who aren’t. I think that makes the non-association between MRI and outcomes stronger than the association between gadolinium and bad outcomes. However, there were 130 bad outcomes in 397 exposures to gadolinium (33%), which is a high enough rate to take this association seriously.
Bottom line: OK, I have never ordered a MRI in a pregnant woman, so this might be a rare situation (but maybe it shouldn’t be). For now it seems like MRI is safe, but that we might want to avoid gadolinium in the first trimester
No surprise: AEDs save lives. But we should probably discuss the costs
Kitamura T, Kiyohara K, Sakai T. Public-Access Defibrillation and Out-of-Hospital Cardiac Arrest in Japan. The New England journal of medicine. 375(17):1649-1659. 2016. PMID: 27783922 [free full text]
Cardiac arrest is relatively easy: push hard and shock early. And it works. This is a large prospective registry from Japan that records all out of hospital cardiac arrest. From 2005 to 2013, the number of patients who received defibrillation using a public access AED rose from 1% to 16.5%. The survival was much higher in patients who had bystander AED use. They estimate that the availability of public AEDs directly contributed to an extra 200 patients surviving neurologically intact over the course of 2013. That is fantastic news. Of course, nothing is free. In order to achieve that, they had to place 428,821 AEDs in the public. They don’t specifically mention a cost, but my guess is this represents well over 2 million dollars per life saved. I am all for AED availability, and I understand that cardiac arrest is an exciting area of medicine, but we should probably be having a discussion about the best ways to invest our limited health care dollars.
Bottom line: AEDs work, but you need to have one close at hand, and that costs money. (Maybe drone based AEDs can help?)
Ultrasound can’t be used for everything?
Rowlands R, Rippey J, Tie S, Flynn J. Bedside Ultrasound vs X-ray for the Diagnosis of Forearm Fractures in Children. The Journal of emergency medicine. 2016. PMID: 27814988
Does ultrasound have a role in diagnosing extremity fractures? This is a prospective study of 469 patients with suspected forearm fractures. Ultrasounds were performed by a group of doctors who had received a couple hours of training in identifying fractures on ultrasound. Ultrasounds had to be performed before x-ray. The treating doctor did not perform the ultrasound and was blinded to the result of the ultrasound. There were a total of 234 fractures, and ultrasound identified 214 of them (sensitivity 91.5%). The misses seem to be minor (buckle fractures) or the result of not looking in the right place (mid shaft fractures). They report of specificity of 88%, but that is using the x-ray as the gold standard. I imagine some of the fractures seen on ultrasound are real, even if they don’t show up on x-ray. All ultrasound studies have one key problem: the skill level of the sonographers in the study might not match the skill level of users in the real world. Although I use ultrasound a lot in the department, I don’t use it much for forearm fractures. My sense is that the pain of have the probe over the fracture isn’t much different than the pain of manipulation for x-ray and that the amount of time it takes for me to boot up the ultrasound isn’t much different from the time it takes to get an x-ray. On top of that, I am not too worried about the radiation dose associated with extremity xrays. There are a few scenarios I might look for fractures with an ultrasound. Currently I use it when x-ray isn’t great – like in rib fractures (and I might have to ultrasound out anyway to look for a pneumothorax). There may be a role for ultrasound if you don’t have easy access to x-ray. I think Casey Parker thinks that ultrasound is better than this study indicates, and is planning to tell me why on the podcast, but you’ll have to listen to hear his thoughts.
Bottom line: Based on this study, ultrasound doesn’t seem good enough for diagnosing extremity fractures.
Long term outcomes after PE
Sista AK et al. Persistent right ventricular dysfunction, functional capacity limitation, exercise intolerance, and quality of life impairment following pulmonary embolism: Systematic review with meta-analysis. Vasc Med 2016. PMID: 27707980
This is a systematic review looking at the long term outcomes after PE and the possible effects of thrombolytics. They identified 26 studies that included 3671 adult patients. The mean age was 59, which seems a little young for PE studies, but they don’t list comorbidities, which will have a big impact on functional outcomes. About a third of patients had some degree of functional impairment (NYHA II-IV) and 11% had moderate to severe impairment (NYHA III-IV) measured 9 months to 3 years after the PE. The median walk test was only at the 5th percentile for age matched norms. The point estimates were all improved in patients receiving thrombolytics, but not statistically so, so this data shouldn’t change practice. Usually, you will notice that I suggest that we should be doing less in medicine, especially when it comes to PE. However, my guess is this might turn out to be an area we need to do more. So far the thrombolytic trials have all focused on short term survival, but Jeff Kline has been arguing for a while that the real outcome we need to be focusing on in submassive PE is long term functional outcomes. This article gives us a better sense of that morbidity, but doesn’t answer the important question of what treatment is best.
Bottom line: PEs don’t just kill patients. They also result in significant long term disability, which we should probably be discussing with our patients.
Intra-articular lidocaine for shoulders
Wakai A, O’Sullivan R, McCabe A. Intra-articular lignocaine versus intravenous analgesia with or without sedation for manual reduction of acute anterior shoulder dislocation in adults. The Cochrane database of systematic reviews. 2011. PMID: 21491392
I have covered some individual articles on this topic before, but someone out there suggested I include the Cochrane review. (I think Patrick Bafuma?) Although it is a Cochrane review, they only found 5 quality studies that cover a total of 211 patients, so that is the best evidence that we get. When comparing intra-articular lidocaine to intravenous sedation for shoulder dislocations, there was no difference in the rate of success, but the intra-articular lidocaine resulted in fewer adverse events and patients spent less time in the ED. No trials reports any major harms – however that will be the biggest weakness of this data. There are more adverse events with the sedation group, but we know sedation is incredibly safe, so those are mostly trivial events. However, the adverse event from an intra-articular injection is going to be delayed; if you cause septic arthritis it will take a few days to develop. Those cases are less likely to show up in the data, especially because those patients might decide to go to a different ED if you were the one who screwed up their shoulder.
Bottom line: I love intra-articular injections for shoulders. I find they work well and I think there is probably a role for them. However, if you can get a patient sedated quickly, we know that is a safe and effective option (and the one I would probably prefer as a patient).
Croup gone wild
Chen IC et al. Croup-induced postobstructive pulmonary edema. The Kaohsiung journal of medical sciences. 26(10):567-70. 2010. PMID: 20604850 [free full text]
This is an interesting case report that describes a very rare condition, but one that is probably worth knowing about: croup induced pulmonary edema. Negative pressure pulmonary edema can occur after any severe, sudden upper airway obstruction. Other than croup, it might occur in aspirated foreign bodies, laryngospasm, epiglottitis, and angioedema. As the patient tries to breath against the obstructed airway, they create large negative intrathoracic pressures, resulting in pulmonary edema. Treatment includes oxygen, management of the underlying cause, and positive pressure ventilation as needed. These authors say diuretics are controversial. I would probably say they just shouldn’t be used. This pulmonary edema is not the result of fluid overload.
Bottom line: Negative pressure pulmonary edema is an interesting disease to keep in the back of your mind
The Beef Jerky Blues
Theobald JL, Spoelhof R, Pallasch EM, Mycyk MB. The Beef Jerky Blues: Methemoglobinemia From Home Cured Meat. Pediatric emergency care. 2016. PMID: 27749634
This is another interesting case report, this time brought to my attention by Ryan Radecki on EM Literature of Note. The authors report a case of a father and daughter who both presented to the emergency department with palpitations, dyspnea, hypoxia, and cyanosis. What happened? The family was making their own beef jerky, and made a calculation error when salting the meat. Instead of using 0.05 ounces of sodium nitrate, they used a full ounce. And sodium nitrate causes… methemoglobinemia. Both had levels over 35% and both improved with methylene blue.
Bottom line: Beef jerky is delicious; toxicology is all about the dose
No antibiotics for diverticulitis
Daniels L, Ünlü Ç, de Korte N. Randomized clinical trial of observational versus antibiotic treatment for a first episode of CT-proven uncomplicated acute diverticulitis. The British journal of surgery. 2016. PMID: 27686365
I’ve been following the topic of antibiotics for diverticulitis for a while now. In May 2015 I discussed the Cochrane review that first piqued my interest. In August 2015, I discussed an RCT showing no benefit with antibiotics. In December 2015 I included a prospective observational study indicating that is probably safe to forego antibiotics. In February 2016 I included the guideline from the American Gastroenterological Association Institute that indicated that antibiotics should be used selectively rather than routinely. However, the major shortfall of all that data is that it essentially all came out of a single centre in Finland. This is a new RCT from a different group in the Netherlands. They randomized 528 adult patients with CT proven uncomplicated acute diverticulitis to antibiotics (amoxicillin-clavulanate, or a combination of ciprofloxacin and metronidazole in case of allergy) or no antibiotics. These patients were almost all treated as inpatients. There were no statistical differences in time to recovery, complicated diverticulitis, recurrence, surgery, readmission, adverse events, or mortality. However, the study was not powered to detect these more rare events. Hospital stay was shorter in the no-antibiotics group (2 vs 3 days) but I am not sure any of these patients need to stay in hospital.
Bottom line: One more piece of evidence that antibiotics aren’t required in uncomplicated diverticulitis. I am relatively convinced. When I self-diagnosed myself with diverticulitis a few months back, I didn’t do anything except take ibuprofen and wait.
Here is the authors’ title: “Codeine: Time to say no”
Tobias JD et al. Codeine: Time To Say “No”. Pediatrics. 2016. PMID: 27647717
The FDA, Europeans Medicines Agency, American Academy of Pediatrics, and Health Canada have all issued warnings about the potential dangers of codeine. This paper covers the history of these warnings and the reported deaths. Even without the concern about super-metabolizers, we know codeine has no analgesic effect for some patients, and has never been shown to have any benefits over other formulations. In fact, in all the studies I am aware of, codeine has higher rates of side effects if you are trying to get to equianalgesic doses.
Bottom line: Although this is specifically about children, I don’t think we should be using codeine for any patients. There are much better options, like plain morphine.
Sorry – chocolate does cause zits
Delost GR, Delost ME, Lloyd J. The impact of chocolate consumption on acne vulgaris in college students: A randomized crossover study. Journal of the American Academy of Dermatology. 75(1):220-2. 2016. PMID: 27317522
OK, not much of an emergency medicine topic, but our friends and relatives expect us to be experts on all medical topics, right? This is a single blind, randomized cross-over study in which 54 college students were randomly assigned to receive either a 43 gram Hershey’s chocolate bar or 15 Jelly Belly jellybeans (to ensure the same glycemic load) and then had their skin re-examined 48 hours later. The number of acne lesions did not increase with jellybeans, but doubled from 4 to 9 lesions in the chocolate group. Now, these patients not not be representative of usual humans, because to get into the study they had to agree to abstain from chocolate for the entire duration of the study. I wasn’t aware robots got acne. Whether these results are extrapolatable to all chocolate, or if there is something specific about Hershey’s that results in acne is unclear.
Bottom line: Some old wives tales turn out to be true
Cheesy Joke of the Month
Too early for a Christmas joke? Too bad:
Why is it getting harder to buy Advent calendars?
Because their days are numbered!
Morgenstern, J. Articles of the month (November 2016), First10EM, November 30, 2016. Available at:
One thought on “Articles of the month (November 2016)”