Thursday, August 6, 2020

COVID-19 Shift Decontamination Routine

While the primary form of transmission of the novel coronavirus, SARS-CoV-2, is through respiratory droplets, there remains the possibility of transmission via fomites, particularly the clothing and gear we wear in the emergency department.
Whether a stringent decontamination routine adds much on top of diligent handwashing, mask-wearing, and personal protective equipment (PPE) is unknown. Below is my personal routine
The 3 basic principles I abide by are:
  • Keep clean things clean 
    • Have a 'hot' station at desk for ECGs that need to be signed, pen, phone
  • Have a system that's sustainable (mine is above)
    • I use a bag and bin system - have a clean bag and dirty bag (bought cheap washable bags and use shopping bags and clean and dirty bins in my car.

  • Don't cross-contaminate
    • Ways I prevent cross-contamination - the phone is a key example. I place it in a plastic bag on arrival (pictured below), I sanitize hands after touching, place it on the 'hot' part of my work station, wipe down after removing from bag and place in clean bag to go home. 

Sunday, February 16, 2020

Fluid Resuscitation in Patients with Sepsis and Heart Failure, ESRD, or Cirrhosis

The Gist: While an article in Chest by Khan et al reports no association between 30 cc/kg of fluids in patients with congestive heart failure (CHF), end-stage renal disease (ESRD), and cirrhosis and intubation, this study should not be used to justify administration of 30 cc/kg of fluids in these patients. Rather, the decision to administer volume should be made according to the individual patient scenario, particularly given the lack of evidentiary basis for this specific volume of fluid.

In the United States (US), a standard part of resuscitating patients with severe sepsis and septic shock includes a crystalloid fluid bolus of 30 cc/kg within the first 3 hours. The Surviving Sepsis Guidelines give this a strong recommendation based on low-quality evidence. In fact, the guideline states that this recommendation is based on essentially no evidence other than this was the average volume of fluids given in trials (PROCESS, ARISE) [2]. In the US, clinicians are incentivized to administer this volume in the first 3 hours when treating patients with septic shock as it is part of the SEP-1 core measure. This quality measure is an "all or none" measure, in which many components of treating patients with severe sepsis/septic shock must be met to satisfy the measure. It may seem that 30 cc/kg of fluids is not that much, only ~2 L for a 70 kg patient. However, fluids are not benign and may pose a risk in patients with a tenuous fluid balance at baseline, particularly heart failure (HF), end-stage renal disease (ESRD), and cirrhosis. 

In a study published in Chest (February 2020), Khan et al boldly conclude "We detected no difference in the incidence of intubation in the high-risk cohort of cirrhotic, heart failure, and ESRD
patients with sepsis who received SSC guideline-concordant (30 mL/kg) vs restricted fluid resuscitation. Based on our results, we suggest that guideline-based early fluid resuscitative efforts should not be omitted in patients with cirrhosis, CHF, and ESRD solely for concern of acute respiratory failure" [1]. The basics of the paper can be seen below:

Unfortunately, this study provides limited insight into whether it is actually safe to administer 30 cc/kg of fluids to patients with HF, ESRD, and cirrhosis. 

This study is a retrospective cohort study, with limited propensity matching. This study design may provide some tentative insight into associations but can certainly not provide a definitive conclusion. Additionally, for this type of study design, there are some limitations with this particular study. 
  • Few methods regarding data extraction. It's clear exactly which parts of the study were performed using chart review methods and was administrative database extraction. Study methods should be transparent and replicable.
  • While patients were matched according to age, sex, weight, mean arterial pressure, lactate level at the time of first fluid bolus administration, APACHE III score, and presence of septic shock, cirrhosis, ESRD, and heart failure there are other variables that could be important. Heart failure is variable and it's probable that those with reduced ejection fractions may be most susceptible to fluid overload. In the restricted fluids cohort, 20 patients (28%) had HFrEF vs 12 (16%) in the liberal fluids cohort. Additionally, seemingly important confounders, such as diagnosis of pneumonia, acute lung injury, or ARDS were not accounted for in the propensity matching or in the multivariable model.
The binary cut-point of 30 cc/kg is an understandable cut-off based on guidelines, but also somewhat arbitrary. In this study patients were grouped according to fluid administered as boluses (excluding those administered in the prehospital setting) in the first 6 hours as either "restrictive" (<30 cc/kg) or standard (≥30 cc/kg). Thus, a patient receiving 29 cc/kg would be classified as restrictive even though they received 70 cc (2.4 oz) less than a patient in the standard group. This is a trivial volume of fluid. However, the standard group did receive, on average, about 2L more crystalloid in the first 2 hours which does suggest an actual difference between groups. Yet, it might make more sense to analyze ≥30 cc/kg and <15 cc/kg. 

The data actually suggests the groups are fundamentally different. The Kaplan-Meier curve in the article, replicated below, actually shows a trend towards increased probability in intubation after ~18 hours. While propensity matching attempts to mitigate selection bias, it cannot fully do so. The restricted fluid group had a higher probability of intubation throughout much of the study period, either due to their baseline disease or uncaptured clinical factors. Patients who were given ≥30 cc/kg were probably deemed by clinicians to be able to "tolerate" the fluids. This inflated probability may make a difference between the two groups disappear. 

  1. Khan RA, Khan NA, Bauer SR, et al. Association Between Volume of Fluid Resuscitation and Intubation in High-Risk Patients With Sepsis, Heart Failure, End-Stage Renal Disease, and Cirrhosis. Chest. 2020;157(2):286-292.
  2. Rhodes A, Evans LE, Alhazzani W, et al. Surviving Sepsis Campaign: International Guidelines for Management of Sepsis and Septic Shock: 2016. Crit Care Med. 2017;45(3):486-552.

Friday, January 31, 2020

D-Dimer in Patients at "Intermediate Risk" of Pulmonary Embolism

The Gist: A D-dimer in the evaluation of suspected pulmonary embolism (PE) in patients with an intermediate probability of PE is recommended in many professional society guidelines [1-4]. The intermediate probability group is the most high-value group for application of a D-dimer, particularly in places such as the United States where the prevalence of PE among those tested is typically <10% [5].

Risk stratification of pulmonary embolism is complex, partially due to the presence of several cut-offs in Wells, one of the most popular risk stratification scores in the United States [6].

The origin: The original Wells score used a trichotomized risk scoring system: Low (<2), Intermediate (2-6), and High (>6), however, this was prior to the introduction of computed tomographic pulmonary angiogram (CTPA) and used compression ultrasound and ventilation-perfusion (V/Q) scans [7]. The diagnostic alogrithm proposed by Wells in 2001 was complex yet many patients in the intermediate-risk group who had a negative d-dimer were considered to have PE excluded (all of these patients got a V/Q and then high probability V/Q scans were treated as positive for PE regardless of dimer result) [78]. Further, this iteration employed a D-dimer assay not widely used currently (SimpliRED), a qualitative assay rather than the high sensitivity quantitative assays [8,9].

The simplification: The initial Wells algorithm was cumbersome and quickly became outdated with CTPA and new d-dimer assays. In 2006, the Christopher Study, evaluated a dichotomized Wells Score that stratified patients into "PE likely" (Wells >4) or "PE Unlikely" (Wells ≤ 4). This study incorporated more relevant diagnostic tests, the CTPA and the VIDAS or Tinaquant quantitative D-dimer assays and found that "PE unlikely" patients with a negative D-dimer had very low risk of PE at 90 days (0.5%; 95%CI 0.2%-1.1%). Approximately 37.1% had PE in the "PE likely" group vs 12.1% in the "PE unlikely" group [10].

The "intermediate" risk group incorporates patients for whom the D-dimer is the most helpful. Evaluation patterns for suspected PE vary across the world; however, CTPA yield (# positive/#ordered) is particularly low in the United States, generally <10% but often in the 3-5% range [5,11]. In a recent US study the prevalence of PE was 4% and, contrasted with the aforementioned Christopher study, in which the overall prevalence of PE was 20%, demonstrates that in the US the patients we evaluate are at even lower risk of PE. Thus, in the US, the dichotomized Wells score likely moves patients with a probability of PE <15-20% to the "PE likely" group, inflating their perceived risk of PE.

In the US, we have a problem with overtesting for PE, and the use of the D-dimer in the intermediate-risk group, in addition to clinically adjusted D-dimer thresholds, may help improve the quality of care we deliver to patients [12].

1. Wolf SJ, Hahn SA, Nentwich LM, Raja AS, Silvers SM, Brown MD. Clinical Policy: Critical Issues in the Evaluation and Management of Adult Patients Presenting to the Emergency Department With Suspected Acute Venous Thromboembolic Disease. Ann Emerg Med 2018;71(5):e59–109. 
2. Raja AS, Greenberg JO, Qaseem A, Denberg TD, Fitterman N, Schuur JD. Evaluation of patients with suspected acute pulmonary embolism: Best practice advice from the Clinical Guidelines Committee of the American College of Physicians. Ann Intern Med 2015;163(9):701–11.
3. Lim W, Le Gal G, Bates SM, et al. American Society of Hematology 2018 guidelines for management of venous thromboembolism: diagnosis of venous thromboembolism. Blood Adv 2018;2(22):3226–56. 
4. Konstantinides S V, Meyer G, Becattini C, et al. 2019 ESC Guidelines for the diagnosis and management of acute pulmonary embolism developed in collaboration with the European Respiratory Society (ERS): The Task Force for the diagnosis and management of acute pulmonary embolism of the European Society of. Eur Respir J. 2019;1–61. 
5. Venkatesh AK, Agha L, Abaluck J, Rothenberg C, Kabrhel C, Raja AS. Trends and Variation in the Utilization and Diagnostic Yield of Chest Imaging for Medicare Patients With Suspected Pulmonary Embolism in the Emergency Department. Am J Roentgenol  2018;210(3):572–7. 
6. Westafer LM, Kunz A, Mazor KM, Schoenfeld EM, Stefan MS, Lindenauer PK. Provider Perspectives on the Use of Evidence-based Risk Stratification Tools in the Evaluation of Pulmonary Embolism : A Qualitative Study. Acad Emerg Med 2020; In Press.
7. Wells PS, Ginsberg JS, Anderson DR, et al. Use of a clinical model for safe management of patients with suspected pulmonary embolism. Ann Intern Med. 1998;129(12):997–1005. 
8. Wells PS, Anderson DR, Rodger M, et al. Excluding Pulmonary Embolism at the Bedside without Diagnostic Imaging : Management of Patients with Suspected Pulmonary Embolism Presenting to the Emergency Department by Using a. 2001;5(3):98–107. 
9. Riley RS, Gilbert AR, Dalton JB, Pai S, McPherson RA. Widely used types and clinical applications of D-dimer assay. Lab Med 2016;47(2):90–102. 
10. Van belle A, Büller HR, Huisman MV, et al. Effectiveness of managing suspected pulmonary embolism using an algorithm combining clinical probability, D-dimer testing, and computed tomography. JAMA. 2006;295(2):172-9.
11. Kline JA, Garrett JS, Sarmiento EJ, Strachan CC, Courtney DM. Over-Testing for Suspected Pulmonary Embolism in American Emergency Departments. Circ Cardiovasc Qual Outcomes. 2020;13(1):1–10. 
12. Kearon C, De Wit K, Parpia S, et al. Diagnosis of Pulmonary Embolism with D-Dimer Adjusted to Clinical Probability. N Engl J Med 2019;381(22):2125–34. 

Thursday, November 28, 2019

Probability Adjusted D-Dimer for Pulmonary Embolism

The Gist: Kearon and colleagues found that the PEGeD strategy, which employs an elevated D-dimer threshold of 1000 ng/mL in patients with a Wells score of ≤4, resulted in less imaging than the standard approaching using Wells, age-adjusted D-Dimer with Wells, or the YEARS algorithm while  “missing” ≤2% of venous thromboembolic events (VTE) at 90 days [1]. This multi-center study indicates that this approach is likely safe in low-prevalence settings such as the US and Canada and could dramatically reduce the burden of imaging.

The background: Risk stratification is the first step in the evaluation of pulmonary embolism (PE) and this can be assessed using gestalt or a risk stratification tool such as the Pulmonary Embolism Rule Out Criteria (PERC), Wells Score, Revised Geneva Score, or the YEARS algorithm. In conjunction with the age-adjusted d-dimer, these are recommended in guidelines to reduce imaging in patients with suspected PE [2-4]. While provider gestalt yields similar estimates of the probability of PE, risk stratification tools have demonstrated benefit in improving imaging yield (and reducing unnecessary imaging) [5].

The problem: Unfortunately, the lack of specificity of the D-Dimer renders many patients who are low or intermediate risk requiring imaging. Literature has increasingly supported elevated D-Dimer thresholds in certain patients (such as those > 50 years old) and in pregnant patients [2-4, 6, 7]. However, studies have recently examined, with success, risk adjusting the D-dimer to use higher thresholds in low probability patients [8-10]. 

This study is an evaluation of a modified Wells + D-Dimer pathway, allowing for patients with a Wells score of ≤4 to have a D-dimer threshold of 1000 ng/mL. 

The study basics are depicted below.

This study found that VTE at 90 days, a surrogate for "missed PE" was very uncommon. This occurred in 0 patients with a D-Dimer <1000 ng/mL (double the typical threshold!). Two patients, both with low pretest probability and d-dimer >1000 ng/m, had VTE at 90 days. One patient, initially diagnosed with PE on imaging, was also diagnosed with a PE on day 77. This is obviously not a "miss" of the algorithm. Another patient, with active malignancy, was diagnosed with a DVT on day 4. 

Assuming the worst-case scenario, VTE at 90 days still occurred in <2% of patients. Unfortunately, 9 of the 1285 patients with a Wells ≤4 and D-Dimer <1000 ng/mL were lost to follow up. Assuming all of these patients had VTE (which is quite unlikely), 0.7% (95% CI 0.32,1.33) would have had VTE at 90 days with an exact binomial confidence interval below the test threshold of 2%. Treating ALL patients, including those who got imaging for PE who were lost to follow up, as having 90-day VTE, would result in 0.7% (95% CI 0.4,1.2) with VTE, again well within the safe testing margins.

MANY (253) patients were excluded because they received imaging despite a Wells  ≤ 4  and D-Dimer <1000 ng/mL.  It is interesting that these patients weren't handled as "protocol violation" and were simply excluded. There are several possible reasons patients received imaging despite their Wells score and low D-Dimer, including something about the patient's clinical course changed or, perhaps, providers just couldn't' help themselves.  Data were available for 127 of these patients (from 2 of the centers) who would otherwise have been eligible for the study, and only 2 had PE on imaging (D-Dimer 720 ng/mL and 650ng/mL), resulting in an overall event rate of 0.19% (95%CI 0.05,0.48%) not accounting for those lost to follow up. Assuming all patients lost to follow up had VTE at 90 days, 0.8% (95% CI 0.47,1.25) of patients would have VTE events, still below the 2% threshold. 

The outcome is quite conservative but is a proxy for "missed PE," which is likely lower. Throughout the PE literature, 90-day deep vein thrombosis (DVT) or PE has been used as a surrogate for missed PE. This is an overly conservative measure that attempts to account for not all patients receiving confirmatory testing (which would be unethical in low-risk patients). This is interesting, though, because this treats risk stratification, D-Dimer, and imaging as a fortune-telling test rather than markers of a disease process at point of time it is tested. Patients with risk factors necessitating testing, will continue to have these risk factors.

This study is also interesting because the "low risk" group in the Wells score is ≤ 4 and the intermediate group is 4.5-6 The Wells scoring systems are complex, with both dichotomous (≤ 4 as "Unlikely, > 4 as "Likely) and trichotomous (<2 as "Low", 2-6 as "Intermediate", or >6 as "High") scoring systems. The major appeal to this cut point is that a patient can still have "Alternative diagnosis less likely than PE," (3 points) and still be classified as "low risk" and have the <1000 ng/mL D-Dimer threshold. While overall 21% of patients in this study met this criterion, most of these patients were in the Wells >4 group and only 11% of patients in the "low risk group" met this criterion. The PEGed algorithm may perform less optimally in a cohort where more low risk patients satisfy this criterion; however, I suspect that this is indicative of the less conservative evaluation in Canada compared with the US (but this is strictly conjecture).

While this study has limitations, it is a multicentered trial in many centers in Canada, the findings are consistent with other findings suggesting the acceptability and safety of higher D-Dimer thresholds. These findings may not be generalizable to settings in Europe and Asia where PE prevalence is higher than the 7.4% in this study. However, in the United States where the prevalence of PE in those evaluated is less than in Canada, these findings likely apply. Additionally, the PEGeD approach is similar to the YEARS algorithm which has been studied in Europe and the United States with many D-Dimer assays.

No algorithm or test will result in 0% miss rate. Further, the harms caused by excess testing and treatment if the likelihood is <2% would outweigh the benefit (5-25% false positive rate of CTPA for PE, anticoagulation, harms from contrast, etc). Given the frequency of PE evaluation and the incremental benefit in reducing imaging, the potential to reduce imaging could be profound.

1. Kearon C, De Wit K, Parpia S, et al. Diagnosis of Pulmonary Embolism with D-Dimer Adjusted to Clinical Probability. N Engl J Med 2019;381(22):2125–34. 
2. Raja AS, Greenberg JO, Qaseem A, Denberg TD, Fitterman N, Schuur JD. Evaluation of patients with suspected acute pulmonary embolism: Best practice advice from the Clinical Guidelines Committee of the American College of Physicians. Ann Intern Med 2015;163(9):701–11.
3.Wolf SJ, Hahn SA, Nentwich LM, Raja AS, Silvers SM, Brown MD. Clinical Policy: Critical Issues in the Evaluation and Management of Adult Patients Presenting to the Emergency Department With Suspected Acute Venous Thromboembolic Disease. Ann Emerg Med. 2018;71(5):e59–109. 
4. Fesmire FM, Brown MD, Espinosa JA, et al. Critical issues in the evaluation and management of adult patients presenting to the emergency department with suspected pulmonary embolism. Ann Emerg Med.  2011;57(6):628-652.e75. 
5. Wang RC, Bent S, Weber E, Neilson J, Smith-Bindman R, Fahimi J. The Impact of Clinical Decision Rules on Computed Tomography Use and Yield for Pulmonary Embolism: A Systematic Review and Meta-analysis. Ann Emerg Med. 2016;67(6):693–701. 
6. van der Pol LM, Tromeur C, Bistervels IM, et al. Pregnancy-Adapted YEARS Algorithm for Diagnosis of Suspected Pulmonary Embolism. N Engl J Med [Internet] 2019;380(12):1139–49. 
7. Kline JA, Williams GW, Hernandez-Nino J. D-Dimer concentrations in normal pregnancy: New diagnostic threshold are need. Clin Chem 2005;51(5):825–9. 
8. Kabrhel C, Van Hylckama Vlieg A, Muzikanski A, et al. Multicenter Evaluation of the YEARS Criteria in Emergency Department Patients Evaluated for Pulmonary Embolism. Acad Emerg Med [Internet] 2018;25(9):987–94. 
9. van der Hulle T, Cheung WY, Kooij S, et al. Simplified diagnostic management of suspected pulmonary embolism (the YEARS study): a prospective, multicentre, cohort study. Lancet 2017;390(10091):289–97. 
10. Kline JA, Hogg MM, Courtney DM, Miller CD, Jones AE, Smithline HA. D-dimer threshold increase with pretest probability unlikely for pulmonary embolism to decrease unnecessary computerized tomographic pulmonary angiography. J Thromb Haemost. 2012;10(4):572-81.

Wednesday, September 18, 2019

Tips for Pumping While Working in the Emergency Department

This post is a bit different than the typical evidence-based post. Rather than a post on how to take care of patients, it's a compilation of tips for expressing breastmilk while working in the Emergency Department

Why is this important?

This is an important issue for everyone, not just women and not only those who are breastfeeding. 
There are tangible ways everyone can contribute to a supportive atmosphere. This includes being supportive of colleagues and reaching out to those who are returning from leave. If your colleagues don't have a place to pump, join them in advocating for a space (by the way, a bathroom is NOT an appropriate place to pump). Know that residents may be especially reticent to ask for time to pump due to being perceived as "weak" or "lazy." 
Pumping on shift can be stressful and come with guilt for stepping away from immediate patient care. There are things that you can do to minimize the time away. 

 Pumping in the car (or on a commute) can be a game changer to ensure you start the shift at time 0, effectively resetting the clock until your next pump.

Pumping on a single coverage shift may seem impossible but it is do-able. One of the most important things is discussing this with your team. Some techs and RNs may feel comfortable coming to you with ECGs or patient updates while pumping, others may prefer to call. Establish a game plan ahead of time.

Know that being able to pump at work is a legal right in the United States and that the American College of Emergency Physicians (ACEP) released a statement in 2013 supporting workplace support for lactating clinicians. 

Saturday, March 23, 2019

Risk Adjusted Dimer in the Workup of Pulmonary Embolism in Pregnancy

The Gist: Use of the YEARS algorithm in pregnant patients with possible pulmonary embolism (PE) appears safe and results in fewer computed tomographic pulmonary angiograms (CTPAs), particularly in low-prevalence testing environments [1]. Given the natural course of d-dimer throughout pregnancy, gestation adjusted versions of this algorithm may better curb overtesting.

The workup of pregnant patients for suspected pulmonary embolism is challenging for some of the following reasons:

  1. Risk of ionizing radiation to the fetus.
  2. Risk of ionizing radiation to maternal breast tissue.
  3. Hormones during pregnancy thought to place pregnant patients at higher risk for venous thromboembolism.
  4. The d-dimer is known to increase throughout pregnancy making this test difficult to interpret or useless based on one's approach. A study by Murphy et al found a steady rise in median d-dimer values in healthy pregnant patients throughout pregnancy. This group found that the median d-dimer in pregnant patients crossed the threshold for non-pregnant patients prior to 20 weeks of gestation [2].
Researchers are tackling this issue with various algorithms. In October 2018, Righini and colleagues published an algorithm rooted in the Geneva Score which would have resulted in fewer CTs. Now, van der Pol and colleagues have published the results of using the YEARS algorithm in pregnant patients. The article is summarized below but briefly, this protocol was implemented in ED and OB units at several centers in Europe and patients were followed up. The results demonstrated safety in this population with overall low prevalance of PE (4% overall).

YEARS in Pregnancy Study Protocol

Notably, a standard YEARS approach to testing for PE in the pregnant patient results in imaging many patients for PE who do not have the disease in question due to the standard trajectory of d-dimer in a healthy pregnant patient. Additionally, the YEARS criteria represent a modified Wells criteria with the most commonly positive criteria being "PE is the most likely diagnosis." Many providers find this criterion frustrating. Future approaches may examine the effects of a gestation adjusted version of the YEARS algorithm. Yet, in the era of overtesting for PE in pregnant patients, use of this algorithm may reduce imaging in these patients.

1. van der Pol LM, Tromeur C, Bistervels IM, et al. Pregnancy-Adapted YEARS Algorithm for Diagnosis of Suspected Pulmonary Embolism. N Engl J Med. 019;380(12):1139–49. 1. 
2. Murphy N, Broadhurst D, Khashan A, Gilligan O, Kenny L, O ’donoghue K. Gestation-specific D-dimer reference ranges: a cross-sectional study. BJOG. 2015;122:395–400. 
3. Righini M, Robert-ebadi H, Elias A, Sanchez O, Moigne E Le. Diagnosis of Pulmonary Embolism During Pregnancy. Annals of Internal Medicine. 2018;
4. Kline J a, Richardson DM, Than MP, Penaloza A, Roy P-M. Systematic Review and Meta-analysis of Pregnant Patients Investigated for Suspected Pulmonary Embolism in the Emergency Department. Acad Emerg Med. 2014;21(9):949–59. 

Thursday, March 7, 2019

Slaying Pre-historic Practice: Epinephrine in OHCA

The Gist: For nearly 15 years, the American Heart Association (AHA) Guidelines have stated epinephrine in out of hospital cardiac arrest (OHCA) is reasonable, not that it is standard of care to administer epinephrine routinely [1-3]. A large randomized controlled trial recently confirmed that while epinephrine increases the return of spontaneous circulation (ROSC), it may do so at the expense of a good neurologic outcome [4]. We likely believe we should administer epinephrine routinely because of difficulties unlearning low-value practices.

Administration of epinephrine 1mg every 3-5 minutes is ubiquitous in OHCA. In fact, the pocket algorithm card included in the AHA guidelines merely lists "Epinephrine q 3-5 minutes." The AHA guidelines, however, give epinephrine a relatively low-level recommendation (Class IIb) and merely state it is "reasonable to consider" using epinephrine in OHCA.

A recent randomized controlled trial, the PARAMEDIC-2 trial, is the largest trial comparing epinephrine with placebo in OHCA. This trial brought renewed attention to the debate as to whether to administer epinephrine routinely in OHCA as patients survived more frequently with epinephrine but did not achieve better neurologically favorable survival [4]. The study was not powered to detect differences in neurologic survival, which may be difficult to do given the low numbers with a favorable outcome; however, does demonstrate this continued trend toward reviving the heart at the expense of the brain.

Why do we have a difficult time unlearning that epinephrine is "standard of care" in OHCA?
We rarely read the guidelines in depth.
The pocket algorithm card lists "Epinephrine q 3-5 minutes" and does not go into details about the risks and benefits of epinephrine administration. Guidelines are often long, cumbersome and difficult to read so we rely upon the synopses provided in simplified algorithms or on the simplified distillation by others.
It's harder to unlearn NOT to do something than it is to add something to our repertoire.
Many of us have a bias towards action, particularly in resuscitation. We want to save patients, we want to do whatever we can.  In OHCA, survival is poor with a mere 2-10% survival and often depends on shockable rhythm, etiology, and defibrillation [4,5]. Epinephrine may give us the satisfaction of feeling as though we are doing something in a typically futile situation.
We may be less likely to fear the consequences when they aren't proximate to our intervention.
We often don't see the downstream consequences that happen after achievement of ROSC in the emergency department - including poor neurologic function.

1. ILCOR. Part 4: Adult Cardiac Life Support. Circulation. 2005 Nov 28;112(24_suppl). Available from:
2. Neumar RW, Callaway CW, Sinz E, Davis D, Otto CW, McNally B, et al. Part 8: Adult Advanced Cardiovascular Life Support: 2010 American Heart Association Guidelines for Cardiopulmonary Resuscitation and Emergency Cardiovascular Care. Circulation. 2010;122(18_suppl_3):S729–67. 
3. Link MS, Berkow LC, Kudenchuk PJ, Halperin HR, Hess EP, Moitra VK, et al. Part 7: Adult advanced cardiovascular life support: 2015 American Heart Association guidelines update for cardiopulmonary resuscitation and emergency cardiovascular care. Circulation. 2015;132(18):S444–64. 
4.  Perkins G, Ji C, Deakin C, Quinn T, et al. A Randomized Trial of Epinephrine in Out-of-Hospital Cardiac Arrest. N Engl J Med. 2018; 
5. Benjamin EJ, Virani SS, Callaway CW, et al. Heart Disease and Stroke Statistics-2018 Update: A Report From the American Heart Association. Circulation. 2018; 137(12)

Friday, November 2, 2018

Droperidol vs ondansetron vs placebo for nausea

The Gist: In the United States, droperidol has been hard to come by in many locations for years. Many lament the loss for a myriad of indications - agitated delirium, nausea, vomiting, and migraines.  Meek et al found that in nauseated emergency department (ED) patients, droperidol 1.25 mg IV did not outperform ondansetron 8 mg IV or placebo IV in a randomized, blinded trial in 3 Australian EDs [1]. Common anti-emetics or placebo, reduce nausea scores by similar amounts; however, more patients receiving ondansetron or droperidol reached the desired effect and did not receive rescue medications.

Nausea is a common condition in the emergency department; yet ED-based studies of anti-emetics have not found consistent benefit [2-5]. Droperidol received a controversial black box warning in the US in 2001 after 71 adverse events were reported on a single day. Since that time, many providers have lamented to lack of availability of droperidol in the US and thought...this would be the clear winner for our nauseated patients. Hence Meek and colleagues aimed to settle the debate as to whether any of these drugs were superior to placebo in nauseated ED patients. Details pictured below.

Unfortunately, this trial leaves several questions unanswered. It's not all bad news in anti-emetic therapy. Recent trials of inhaled isopropyl alcohol have been promising and this intervention appears to be more effective than oral ondansetron alone [6,7]. It's uncertain how long these effects last and how this therapy would perform compared with intravenous therapies.

  1. Meek R, Mee MJ, Egerton-Warburton D, et al. Randomized placebo-controlled trial of droperidol and ondansetron for adult emergency department patients with nausea. Acad Emerg Med. 2018;In Press. doi: 10.1111/acem.13650 
  2. Braude D, Soliz T, Crandall C, Hendey G, Andrews J, Weichenthal L. Antiemetics in the ED: a randomized controlled trial comparing 3 common agents. Am J Emerg Med 2006; 24: 177–82. 
  3. Barrett TW, DiPersio DM, Jenkins CA et al. A randomized, placebo-controlled trial of ondansetron, metoclopramide, and promethazine in adults. Am J Emerg Med 2011; 29: 247– 55.
  4. Egerton-Warburton D, Meek R, Mee MJ, Braitberg G. Antiemetic use for nausea and vomiting in adult emergency department patients: randomized controlled trial comparing ondansetron, metoclopramide, and placebo. Ann Emerg Med 2014; 64: 526–32.
  5. Furyk JS, Meek R, Egerton-Warburton D. Drugs for the treatment of nausea and vomiting in adults in the emergency department setting (Review). Cochrane Database Syst Rev 2015: CD010106. 
  6. Beadle KL, Helbling AR, Love SL, April MD, Hunter CJ. Isopropyl Alcohol Nasal Inhalation for Nausea in the Emergency Department: A Randomized Controlled Trial. Ann Emerg Med. 2016;68(1):1-9.e1.
  7. April MD, Oliver JJ, Davis WT, et al. Aromatherapy Versus Oral Ondansetron for Antiemetic Therapy Among Adult Emergency Department Patients: A Randomized Controlled Trial. Ann Emerg Med. 2018;

Wednesday, October 24, 2018

Pulmonary Embolism Workup in Pregnancy: Is D-dimer a Thing?

The Gist: D-dimer rises during pregnancy and few patients will have a normal d-dimer by the third trimester. However, integration of d-dimer into a diagnostic algorithm for pulmonary embolism (PE) could reduce the need for CT scan by ~11% [1]. The next step - gestational age adjusted d-dimers.

The workup of pregnant patients for PE is complicated. Pregnancy is considered a risk factor for PE. However, the algorithms and risk stratification tools widely used excluded pregnant patients in the development phase. Additionally, the use of the d-dimer in pregnancy is controversial in pregnancy. It's known that the d-dimer rises through pregnancy, making the utility questionable. While some have attempted to determine gestational age adjusted d-dimer cut offs and one expert recommends a first trimester cutoff of 750 ng/mL, a second trimester, cutoff of 1000 ng/mL, and a third trimester cut off of 1250 ng/mL, this has not been formally studied in PE[2,3]. Further, the risks of the objective tests in the patients may be greater due to radiation to the fetus and maternal breast tissue.

A group in Switzerland (also the creators of the Geneva Score) tested an algorithm on nearly 400 patients which appears promising. See algorithm and study summary below. Further research will be needed to ascertain if an elevated dimer cut-off based on gestational week can be used and if another risk stratification tool performs better in this population. For example, the Revised Geneva has age >65, which is essentially incompatible with pregnancy as well as heart rate cut offs of >75 or >95 beats per minute, when the average heart rate in pregnancy rises to >90 beats per minute.

1. Righini M et al. Diagnosis of Pulmonary Embolism During Pregnancy. Ann Internal Med. 2018; In press. doi: 10.7326/M18-1670
2. Murphy N, Broadhurst D, Khashan A, Gilligan O, Kenny L, O ’donoghue K. Gestation-specific D-dimer reference ranges: a cross-sectional study. BJOG . 2015;122:395–400. 
3. Kline JA, Kabrhel C. Emergency Evaluation for Pulmonary Embolism, Part 2: Diagnostic Approach. J Emerg Med. 2015;49(1):104-17.

Wednesday, August 1, 2018

Breastfeeding and Alcohol: A sensational but flawed study in Pediatrics

The Gist: A recent study by Gibson et al in Pediatrics claims to show a link between cognitive delays and alcohol intake while breastfeeding [1]. Despite wild uptake by lay and medical news outlets, this study is seriously flawed by confounding, multiple comparisons, and inappropriate extrapolation of survey data. As a result, the study findings do not support this conclusion and do not support a causal link between drinking alcohol while breastfeeding and cognitive delays. Breastfeeding while inebriated may pose risks due to safely handling the baby and safe sleeping, but unlikely from the alcohol contained within the breastmilk in otherwise healthy term babies. Further, high-risk alcohol use (binge drinking, dependence) may be associated with other factors that could potentially influence the cognitive and emotional development of children. However, this study only shows our obsession with p values and statistical significance.

Gibson et al hypothesized that alcohol use during lactation would be associated with lower cognitive scores in children [1]. The lead author has said, "it suggests that there is no safe level of alcohol for a breastfeeding mother to drink" [2]. However, the study does not actually suggest this. An overview of the study can be found in the graphic above.

1. Misunderstanding the term breastfeeding
The authors inappropriately describe "drinking while breastfeeding." Consuming alcohol while lactating is not the same as drinking while breastfeeding, although it is unlikely that either have clinical significance in most scenarios. The questionnaire administered at Wave 1 asked about general current alcohol intake and whether or not the infant was breastfed (in general). They did not ask about the specific timing of drinking, not that this necessarily matters. Further, the authors did not measure alcohol concentration in the milk they were feeding their infants, which would really be the only way to demonstrate causation (which would be an impossible study to conduct, yet the authors used the word "cause" in the manuscript).

Alcohol intake during breastfeeding has been controversial as it is known that alcohol does transfer into breast milk [2]. Official recommendations are to refrain from drinking while breastfeeding or limit to 1 drink per day based on this notion [3]. However, the amount of alcohol an infant would ingest is minimal as the percentage of alcohol in the breastmilk approximates that of the mother's blood alcohol level (or approximately 5% of the intake) [2]. Further, milk production and composition is a dynamic process such that when an individual is no longer inebriated, the breastmilk does not contain alcohol [2]. As such consuming alcohol and directly feeding the baby at the time of peak ethanol effects is different than consuming alcohol at a time period when the mother is not feeding the baby (ex: immediately after nursing or at night).

2. Multiple comparisons
Only one transient, statistically significant association was found, as seen below. Children who were ever breastfed, the bulk of whom were not being breastfed at the time the alcohol questionnaire was administered (~age 9.7 months), with mothers who had scores associated with high or risky alcohol consumption performed less well on the Matrix Reasoning subtest of the Wechsler Intelligence Scale at Wave 4 (~age 7) but not Wave 5 (~age 10). The more comparisons, the greater likelihood of finding a positive association due to chance. While these authors did use the Benjamini and Hochberg procedure to control the false discovery rate (proportion of false positives), the bulk of the data including the vocabulary test and the early literacy/numeracy tests do not support this association.

3. Data dredging: p-value without clinical significance
Researchers have an obsession with p values. This study had greater than 99% power to detect a small effect size. Even with this power, the study did not detect an association in 6 of the 7 time-outcome combinations they evaluated. Yet,the authors focused on the single statistically significant association that was not found on follow up tests (i.e. a child whose mom had ever breastfed them and engaged in high/risky alcohol use at Wave 1 performed less well on the matrix reasoning test at Wave 4 but performed on par with peers at Wave 5). Additionally, if it were the alcohol that resulted in the lower scores, we would expect to see the same (or greater) results in infants who were currently being breastfed.

Even if this association was real rather than a statistical anomaly, the kids who were being breastfed at the time of the alcohol questionnaire had no measurable deleterious cognitive outcome at follow up.

4. Lack of an adequate control group and missing data. 
Just over 8% of infants were classified as "never breastfed." This is a tiny group with few expected outcomes. Additionally, this cohort had significant missing data, notably 17.2% missing data for alcohol consumption.

Most physicians lack significant training regarding the care of lactating patients and advice regarding "pumping and dumping" (pumping and discarding milk instead of feeding it to the baby) is rampant, even when medications are safe in lactating patients [3]. For example, as recently as 2017 major anesthesia texts recommended empiric "pumping and dumping" for 24 hours following anesthesia [4,5]. Studies such as the one by Gibson et al, which has been picked up by news outlets, only serve to increase fears without actually contributing meaningfully to the knowledge base [1]. It is time to move from the overly cautious approach to the lactating woman to a reasonable, evidence-based approach in order to support breastfeeding without creating hysteria amongst parents.


  1. Gibson L and Porter M. Drinking or Smoking While Breastfeeding and Later Cognition in Children. Pediatrics; In Press.
  2. Rapaport L. Drinking while breastfeeding tied to cognitive problems in young kids. July 31, 2018.
  3. Haastrup MB, Pottegård A, Damkier P. Alcohol and breastfeeding. Basic Clin Pharmacol Toxicol. 2014;114(2):168-73
  4. CDC. "Alcohol."
  5. Cobb B, Liu R, Valentine E, Onuoha O. Breastfeeding after Anesthesia: A Review for Anesthesia Providers Regarding the Transfer of Medications into Breast Milk. Transl Perioper Pain Med. 2015;1(2):1-7
  6. Dodd SE, Sharpe EE. Pump and Dump; Anesthesiologists Lead the Feed. Anesthesiology. 2018;128(5):1046-1047.
  7. Wijeysundera D, Sweitzer BJMiller RDPreoperative Evaluation, Miller’s Anesthesia. 2015, pp 8th Edition. Edited by Elsevier, 1085–1155

Wednesday, September 13, 2017

Quarterly Literature Update - September 2017

Do abscesses need antibiotics after incision & drainage (I&D)
Two recent studies have rendered many with the conclusion that antibiotics improve cure and recurrence rates when added to I&D.  Talan et al randomized people with abscesses after I&D to TMP-SMX or placebo and found improved clinical cure rates in the TMP-SMX group (80.5% vs 73.6%). This year, Daum et al randomized patients to TMP-SMX, clindamycin, or placebo after I&D and specifically looked at small abscesses (<5 cm) and again found higher rates of 10-day clinical cure in patients receiving antibiotics (TMP-SMX 83.1%, Clindamycin 81.7%, Placebo 68.9%). Many people have pushed the idea that all abscesses need antibiotics after drainage based on these studies. However, these studies included patients with significant surrounding cellulitis (on average, >25 cm2) and most recommendations have historically recommended antibiotics to those with more than a 5 cm diameter of surrounding cellulitis
  • Bottom Line: After I&D of an abscess, give antibiotics if there is >5 cm diameter of surrounding cellulitis (~30 cm2 total).  We probably do not need to do this if less cellulitis.
  • Stats/Study Design point: With any study check out the baseline/demographics table and ask - are these patients like my patients.
FAST is fast, but is it good?
The FAST exam, initially developed to assess for free fluid in unstable trauma patients, is widely used and may be an area where indication creep (i.e. we start expanding the indication for it) has really taken hold. Holmes et al performed an RCT on 945 stable pediatric blunt trauma patients who were randomized to standard trauma evaluation + FAST or just a standard trauma evaluation. There was no difference in rates of CT scans (~50% in each arm), missed intra-abdominal injuries (n=1 in FAST arm), or hospital charges. Even things like length of stay and hospitalization rates were about the same between groups.
  • Bottom Line: Keep FASTing trauma patients who are unstable or borderline unstable and management may change based on the FAST results (i.e. thoracotomy or OR). Find a negative FAST on a trauma patient? That’s called a false sense of security.
  • Stats/Study Design point: Beware of indication creep (using a test in patients it was not originally designed or validated in - this can change parameters such as sensitivity and specificity)

A magic sepsis cocktail?
Marik et al have made a wave in public media for a magic sepsis cocktail of vitamin C, hydrocortisone, and thiamine. In this paper, they published their findings of a paltry 47 patients who underwent treatment and report an incredible mortality reduction in the cocktail arm (8.5 % 4/47 vs 40.4% 19/47). The primary author, very well respected in the sepsis/critical care community, has argued that this treatment should be adopted immediately because prior studies show biological plausibility. However, this is a supremely tiny hypothesis generating study subject to incredible bias (see this podcast) due to no blinding and other factors. Further, prior studies of individual components (ex: steroids) have been negative.
  • Bottom Line: Vitamin C, hydrocortisone, thiamine is an inexpensive and relatively benign therapy; however, a single retrospective study of 47 patients cannot inform us of efficacy or harm.  Medicine is rife with examples of adopting therapies too early because they look promising based on terrible studies only to later find out that they don’t work. Further studies will be needed to determine whether or not this practice should be adopted.
  • Stats/Study Design pearl: Before/after studies are subject to significant bias and confounding. In this study we worry about the lack of blinding as all providers and patients knew what they were getting and that there was this new "life saving" cocktail - perhaps this could have influenced other care.

Haloperidol HUGS and Love
Gastroparesis and cyclic vomiting patients are frustrating and often their medication allergy lists are extensive. Enter haloperidol, a butyrophenone antipsychotic (dopamine antagonist), a sister medication of droperidol. Roldan et al randomized 15 patients with gastroparesis to 5 mg IV haloperidol + conventional treatment and 18 patients to conventional treatment alone. Pain and nausea dropped from 8.5/10 to 1.83 (pain) and 4.53/10 to 1.83 (nausea) at 1 hour in the haloperidol group compared with a drop from 8.28 to 7.17 (pain) and 4.11 to 3.39 (nausea) in the placebo group. Another small, not especially well done observational study, termed HUGS, by Ramirez et al echo these findings.
  • Bottom Line:  These studies aren't necessarily practice changing but correlate with significant prior literature for pain and nausea treatment in migraines and in the anesthesia literature.It’s worth a shot to try haloperidol 2.5-5 mg IV in patients with gastroparesis (or even cyclic vomiting/cannabinoid hyperemesis although this is based on case series). It's nice to have literature out there saying "hey, this is a thing we are doing."
  • Stats/Study Design Pearl: Size does matter when it comes to the methods section. In a retrospective study you should be able to replicate the study by reading the methods. The Ramirez study had a paltry 11 sentences.

PERC - Love it but ignore it?
It’s well accepted that in a patient with low pretest probability (15% or less, i.e. Low Risk by Wells Criteria), we can rule out pulmonary embolism (PE) using the PERC criteria. The miss rate using this approach is <2%, which is the cut point where further workup and treatment would cause more harm than good (test-treatment threshold). Apparently, we suck at actually following this evidence based approach according to a prospective observational study by Buchanan et al of patients with chest pain and/or shortness of breath looked at PE testing rates in PERC negative patients. Over 25% of patients who were PERC negative underwent further PE testing and 7.2% went straight to CTPA or V/Q, skipping the d-dimer.
  • Bottom Line: If you think of working up PE, consider the patient’s risk of PE, then think of PERC, and then d-dimer before you jump to CTPA. In low-risk patients (ex: meets Wells Low Criteria) with possible PE that are PERC negative, do not do further testing to assess for PE.
  • Stats/Study Design Pearl: The test-treatment threshold is important. This is the idea that there is a point at which the harms of further workup and treatment of a disease process equal or become greater than harm of not pursuing the test. For PE this has been established at about 1.8%.

The End of Apneic Oxygenation?
Placing a nasal cannula at 15 LPM (or more) during RSI has become pretty standard practice in many ED rapid sequence intubations (RSI), often deemed apneic oxygenation (ApOx). The ENDAO trial randomized 206 ED patients undergoing RSI who had at least 3 minutes of preoxygenation to ApOx or no ApOx. Nobody was blinded. The study was looking for a difference in lowest mean oxygen saturation before groups, which they did not find (92% vs 93%).
  • Bottom Line: This study, and others like it, are negative, probably because the desaturations we are looking to prevent with ApOx are rare. ApOx may be helpful in difficult intubations or those with poor physiologic parameters who are likely to desaturate, unfortunately, we sometimes don’t know who these people are.
  • Stats/Study Design Pearl: Choice of primary outcome is important. It may seem like a difference in oxygen saturation is clinically important but when oxygen saturations are globally high, this is not going to make a difference to the patient.

Steroids for all that wheezes?
Many patients with lower respiratory tract infections (LRTI), particularly with wheezing, receive empiric steroids, inhaled beta-agonists, and possibly antibiotics (such as azithromycin), yet the evidence behind this is scare in patients without underlying lung disease (asthma/COPD). Hay et al randomized 401 patients presenting to outpatient clinics with cough and signs of LRTI who had no history of asthma or COPD to either 40 mg prednisolone daily for 5 days or placebo. They found no difference in duration of moderately bad or worse cough (~5 days in each group) or severity of symptoms on days 2-4. Of note, about 40% of the patients in this study were wheezy.
  • Bottom Line: Just because a patient with a cough wheezes, it doesn’t mean they will benefit from steroids. Keep giving steroids to patients with flares of asthma and COPD but we can be much more selective in the those without these lung diseases.

Bonus Throwback(Cutting Edge from 2010) - Stop the pulse checks in codes
We are terrible at pulse checks - they are inaccurate and insensitive. The pulse check during a code (i.e. after the initial <10 sec pulse check by a healthcare provider to establish arrest) is nowhere in ACLS. Nada. The 2010 AHA Guidelines state: "Because of difficulties with pulse assessments, interruptions in chest compressions for a pulse check should be minimized during the resuscitation, even to determine if ROSC has occurred." The ACLS guidelines recommend pulse checks during CPR only if an organized rhythm is seen
  • Bottom Line: Continue to assess for rhythm changes during CPR - this can take place in 2-3 seconds. There’s no need to stop for a pulse check but you may if you see ETCO2 rise significantly (just be aware it’s not very accurate). You can take a few second ultrasound clip during a rhythm check to assess for cardiac contractility.

Bonus Non-Clinical Paper - Boys > Girls?

There’s been a lot of talk about the gender pay gap, but surely it doesn’t apply to academic emergency medicine, right? Wrong. Masden et al used 2015 survey data from full-time faculty academic emergency physicians and found that women made > $19,000 less than men. You may say that doesn’t account for rank, fellowship status, administrative roles, location, clinical hours, etc. But again, that would be incorrect. That gap was after adjusting for all of these potential confounders.
  • Bottom Line: Women make less than men in academic emergency medicine. Transparency and awareness may help so share the paper with your colleagues and chairs and don’t be afraid to negotiate.