REMSTARBC.ca    Rural Emergency Medicine Simulation Training and Resource - British Columbia


The following summaries are taken from various sources in the
Emergency Literature and for interest purposes.
For further reading the exact journal citation is indicated.

HOWEVER, any new update should be viewed through the perspective of the first journal cited below which is from 2014.
It was a review of how long it takes from a 'seminal' cardiac paperto be published to be adopted by the majority of clinicians. 
This can likely be translated to any area of medicine.
The depressing fact was the average time from a 'defining or pivotal' study being published to where is was adopted by the majority
of clinicians, took an average of 16 years!
So if you read what appears to be an 'amazing' and 'pivotal' studyin the ER literature, don't be depressed if it takes half your
medical career to become widely adopted.

2015 Feb;169(2):266-73. 2014.09.015. Epub 2014 Nov 12.

Translation of acute coronary syndrome therapies: from evidence to routine clinical practice.

(This article puts all the other articles here and others you read elsewhere, in a new perspective - SB)

Putera M1, Roark R2, Lopes RD3, Udayakumar K4, Peterson ED5, Califf RM6, Shah BR. Abstract BACKGROUND: The use of evidence-based therapies has improved the outcome of patients with acute coronary syndrome (ACS), but there is a time lag between the generation of clinical evidence and its application in routine clinical practice. We sought to quantify temporal lags in the lifecycle of American College of Cardiology (ACC)/American Heart Association (AHA) class IA ACS therapies. METHODS:

Using current and historical ACC/AHA guideline publications, we retrieved publication dates of pivotal clinical trials (PCTs) and class IA guideline-recommended therapies for patients with ST-elevation myocardial infarction (STEMI) and unstable angina (UA)/non-STEMI (NSTEMI). Clinical practice uptake data for each therapy were retrieved from the National Registry for Myocardial Infarction, Can Rapid Risk Stratification of Unstable Angina Patients Suppress Adverse Outcomes with Early Implementation of the ACC/AHA Guidelines, and Acute Coronary Treatment and Intervention Outcomes Network Registry-Get with the Guidelines, which are registries containing publicly available peer-reviewed data. Descriptive data were calculated and compared for each phase of the evidence life cycle for both STEMI and UA/NSTEMI drug classifications.

RESULTS:

We identified 11 class IA- and 4 class IB/IC-recommended therapies for acute, inhospital, and discharge use for patients with STEMI or UA/NSTEMI. The median time lags were 2 years (interquartile range [IQR], 1-4 years) from PCT to practice guideline recommendation, 14 years (IQR, 11-15 years) from guideline recommendation to 90% practice uptake, and overall, a 16-year median (IQR, 13-19 years) from PCT to 90% practice uptake.

CONCLUSIONS:

The time of PCT publication to meaningful uptake of class IA ACS therapies into clinical practice took a median of 16 years. This significant time lag indicates systemic barriers to the translation of therapeutics into routine clinical practice.

Electrical or Chemical Cardioversion for Atrial Fibrillation in the Emergency Department?

Daniel M. Lindberg, MD reviewing

Either approach prevents admission for most patients with acute AF.

Emergency department (ED) management of patients with atrial fibrillation (AF) is highly variable. As part of a sustained effort to determine best practices, this research group compared pharmacologic cardioversion followed by electrical cardioversion, if needed, versus electrical cardioversion alone.

Stable adult ED patients with acute AF (clear symptom onset 3–48 hours prior to arrival or 3–168 hours prior to arrival if anticoagulated for at least 4 weeks) were randomized to infusions of procainamide (15 mg/kg) or a placebo given over 30 minutes. Patients with persistent AF 30 minutes after the infusion ended received up to three biphasic shocks of at least 200 J each. The main outcome was achievement of sinus rhythm for at least 30 minutes.

The authors enrolled 396 patients presenting to 11 Canadian EDs from 2013 to 2018. Sinus rhythm was achieved in 96% of patients in the procainamide-shock group and 92% in the placebo-shock group (difference not significant). Among patients who received procainamide, 52% converted without need for shock, compared with 9% who received the placebo. More than 95% of patients in each group were discharged home, and there were no serious adverse events during 14-day follow-up. Patients receiving electrical cardioversion were further randomized to anteroposterior or anterolateral pad positioning; conversion rates were similar (92% and 94%)

The Single Syringe Technique for Adenosine in Supraventricular Tachycardia

Lauren M. Westafer, DO, MPH, MS reviewing

In this small study, patients who received adenosine and saline in a single syringe had a higher rate of conversion after the first dose than patients who received them in separate syringes.

Adenosine is frequently used to convert supraventricular tachycardia (SVT) to sinus rhythm. Because it has a half-life of <10 seconds, administration is typically followed by a saline flush to deliver the medication bolus to the heart quickly. Historically, clinicians have used a cumbersome technique involving two syringes, and in some cases a stopcock, to administer adenosine and saline quickly. Some clinicians now combine adenosine and saline in a single syringe, but there are few studies of this technique.

Investigators performed a prospective study of 53 hemodynamically stable adults who presented to a single emergency department with SVT. At the treating clinician's discretion, patients were given 6 mg of adenosine one of two ways: in a single syringe combined with 18 mL of saline (26 patients) or in two separate syringes, one containing adenosine and the other 20 mL of saline (27 patients). More patients in the single-syringe group than the two-syringe group converted to sinus rhythm after the first dose (73.1% vs. 40.7%)

August 19, 2019

Unraveling the Safety Profile of Proton-Pump Inhibitors

John R. Saltzman, MD and David J. Bjorkman, MD, MSPH (HSA), SM (Epid.)

A large randomized trial has shown no association between PPI use and most of the adverse outcomes noted in observational studies.

Proton-pump inhibitors (PPIs) provide important clinical benefits for many patients. However, observational studies have suggested an association between PPI use and a variety of adverse events, including dementia, osteoporosis, bone fractures, micronutrient deficiencies, pneumonia, spontaneous bacterial peritonitis, kidney disease, and enteric infections. We have observed that some patients with documented indications for PPIs have stopped the drugs on their own after hearing media reports about frightening potential side effects such as dementia; other patients have discontinued them on the advice of their physicians. As an illustration, we provide the following example — a composite of several cases seen in our gastroenterology practices:

The postulated adverse effects of PPIs generally have been noted in retrospective analyses of databases created for other purposes. But now, a rigorously conducted, prospective, randomized trial has cast doubt on the credibility of most of those claims (NEJM JW Gastroenterol Aug 9 2019; [e-pub] and Gastroenterology 2019; 157:403). Nearly 18,000 patients with stable cardiovascular disease were randomly assigned to receive either a PPI (pantoprazole, 40 mg once daily) or placebo. During average follow-up of 3 years, there were no associations between PPI therapy and nearly all safety outcomes (dementia, bone fracture, new diagnoses of diabetes mellitus, chronic kidney disease, pneumonia, chronic obstructive pulmonary disease, gastric atrophy, cancer overall, and cancers of specific primary sites). The one exception was a small, barely significant association with enteric infections (1.4% with pantoprazole vs. 1.0% with placebo). For Clostridium difficile specifically, there were only nine cases in the pantoprazole group and four cases in the placebo group, a nonsignificant difference.

Methodologic limitations of observational studies are often ignored or poorly explained by the lay media, and many physicians who prescribe PPIs may not be familiar with the strengths and weaknesses of the relevant literature on PPI side effects. However, even before the aforementioned randomized trial was published, there was reason to doubt whether the reported associations between PPIs and the publicized adverse effects were causal. In a 2017 review of observational studies that showed these associations, researchers applied standard epidemiologic criteria for causation (Hill criteria). They concluded that these studies were prone to residual confounding (i.e., a high likelihood that links between PPIs and adverse events were mediated by other factors) and that the evidence for causality was weak.

Additionally, observational studies can vary in their rigor and conclusions; the alleged association between PPI use and dementia provides an instructive example. The report that initially attracted media attention in 2016 was a retrospective study drawn from an insurance claims database, and the researchers were unable to control for key confounding variables (NEJM JW Gen Med Jun 15 2016 and JAMA Neurol 2016; 73:410). Subsequently in 2017 and 2018, methodologically stronger observational studies — which included prospective assessments of cognitive impairment and dementia and more-rigorous controlling for potentially confounding factors — failed to corroborate the PPI–dementia link (NEJM JW Gen Med Sep 15 2017 and Am J Gastroenterol 2017; 112:1802 and Gastroenterology 2017; 153:971; NEJM JW Gen Med Mar 15 2018 and J Am Geriatr Soc 2018; 66:247; JW Gen Med Nov 15 2017 and J Am Geriatr Soc 2017; 65:1969).

For Rapid Sedation of the Agitated Patient, Consider Midazolam

Adequate sedation at 15 minutes was better with intramuscular midazolam than ziprasidone, olanzapine, or haloperidol.

Agitated patients often require sedation, but what is the best medication? Investigators conducted a prospective observational study in which all patients in one emergency department who required intramuscular sedation for agitation were treated with a particular agent during a 3-week period; after 15 weeks, five agents (haloperidol 5 mg, haloperidol 10 mg, ziprasidone 20 mg, olanzapine 10 mg, and midazolam 5 mg) had been used and average patient responses were compared.

Results from 737 encounters revealed midazolam to be the best agent, with 71% of patients adequately sedated at 15 minutes, versus 40% for haloperidol 5 mg, 42% for haloperidol 10 mg, 52% for ziprasidone, and 61% for olanzapine. However, 40% of the midazolam patients required an additional medication to achieve or maintain sedation, versus 33%, 20%, 24%, and 21%, respectively, for the other agents.

Comment by NEJM editors

In my experience, midazolam is woefully underused in multiple contexts, because it is viewed as a procedural sedation agent in the category of propofol. This study suggests that it is a great agent for sedation of the agitated patient. However, its short half-life means that it should probably be coadministered with another agent, such as haloperidol 5 mg

A Sacred Cow Slaughtered: Eliminating the Glasgow Coma Scale from Trauma Field Triage

A large two-center validation study provides good evidence that “patient does not follow commands” is just as predictive as the more-complex Glasgow Coma Scale.

A signature contribution of modern prehospital care has been to regionalize trauma care. Under regionalization, severely injured patients are not transported to the nearest hospital — instead, they go to the region's designated trauma center (they are “up-triaged”). But how do prehospital personnel decide? They use the CDC's Guidelines for Field Triage of Injured Patients. These guidelines direct prehospital personnel to up-triage when a patient has a Glasgow Coma Scale (GCS) score of <14 or meets one of the other numerous criteria for up-triage.

Because the GCS is a hassle for paramedics to learn and calculate, investigators have proposed using instead a simpler triage tool: if an injured person is not following commands, up-triage. Now investigators report on a two-center validation study comparing the predictive value of this dichotomous triage tool with that of the full GCS. This analysis of nearly 48,000 trauma patients found that “patient does not follow commands” had the same predictive value as GCS <14 for all studied trauma outcomes, with slightly superior positive likelihood ratios (e.g., those for mortality, 2.37 vs. 2.13) offsetting slightly inferior negative ones (e.g., those for mortality, 0.25 vs. 0.24).

Comment by nejm editors

Take a look at the Field Triage Guidelines linked above. They could stump a rocket scientist (or a brain surgeon, for that matter). To minimize time in the field, the expert panel that owns these guidelines should simplify them. The first step? Eliminate the Glasgow Coma Scale, and instead just up-triage if the patient is not following commands

Ketamine for Delirium Tremens

Dan

This study suggests that ketamine can safely be used to avoid intubation and may decrease length of intensive care unit stay.

Severe alcohol withdrawal, or delirium tremens (DT), is a life-threatening condition that can require massive doses of benzodiazepines or barbiturates (GABA agonists), which can require intubation and prolonged intensive care unit (ICU) care. These authors studied a retrospective sample of adult patients admitted to a single ICU with DT to determine whether adjunctive therapy with ketamine improved outcomes.

They compared outcomes in 29 patients who received symptom-triggered therapy with GABA agonists with outcomes in 34 patients who were treated after initiation of a guideline that added an intravenous ketamine infusion (0.15–0.3 mg/kg/hour) to GABA agonist therapy. Using multivariable modeling that accounted for initial ethanol level and the total amount of GABA agonist required for treatment, patients who received ketamine had significantly lower rates of intubation (29% vs. 76% for patients who did not receive ketamine) and shorter ICU stay (5.7 days vs. 11.2 for patients who did not receive ketamine). There were no reported adverse events.

Comment by nejm editors

Is ketamine the “new” old wonder drug? Pain, asthma, nausea, severe depression, suicidality, cannabinoid hyperemesis, post-traumatic stress disorder, and now DTs? The findings of this small series support the need for a randomized trial. Nonetheless, I plan to use ketamine for my patients when ethanol withdrawal requires ICU admission.

It's Better with a Bougie

A large randomized trial finds higher first-attempt success rates when a bougie is used for endotracheal intubation.

The bougie is traditionally reserved for anticipated difficult airways or for rescue after a failed attempt. Now investigators report on a large trial in which 757 adult patients were randomized to intubation with a bougie or with an endotracheal tube plus stylet. The main outcome was first-attempt success among patients with at least one difficult airway characteristic, and the secondary outcome was first-attempt success for all comers.

All comers to a single emergency department who were undergoing planned intubation with a Mac style blade were eligible. (Because the hyperangulated GlideScope makes bougie passage difficult, patients intubated with this device were excluded.) Ninety-six percent of patients were intubated with the CMAC video laryngoscope, although the screen was not utilized in about half the cases.

Among the 380 patients with at least one difficult airway characteristic, first-attempt success was significantly more common in patients intubated with the bougie (96%) than in those intubated without the bougie (82%). Among all comers, the first-attempt success rates were also significantly higher in the bougie group (98% vs. 87%).

Children with Isolated Vomiting After Head Injury Are Unlikely to Have Intracranial Injuries

Only 0.2% of such children had intracranial injuries.

In a planned secondary analysis of data from the Australasian Pediatric Head Injury Rule Study, investigators determined the prevalence of clinically important traumatic brain injury (ciTBI) and traumatic brain injury identified on computed tomography (TBI-CT) among children presenting after a head injury with isolated vomiting. The definition of ciTBI was head injury requiring any of the following: neurosurgical procedure, intubation, or admission ≥2 nights because of persistent signs or symptoms from head injury. Isolated vomiting was defined as vomiting without other predictors of head injury from the Pediatric Emergency Care Applied Research Network (PECARN) and the Children's Head Injury Algorithm for the Prediction of Important Clinical Events (CHALICE) rules.

Of 19,920 children <18 years of age, 3389 (17%) had vomiting. Of the 1006 children (30%) with isolated vomiting, only 1 (0.1%) had ciTBI and 2 (0.2%) had TBI-CT. No child <2 years of age with isolated vomiting had ciTBI or TBI-CT. In multivariate regression analysis, significant predictors of ciTBI in patients with isolated vomiting were: signs of skull fracture (odds ratio, 80), altered mental status (OR, 2.4), headache (OR, 2.3), and acting abnormally (OR, 1.9); predictors of TBI-CT in children with isolated vomiting were: signs of skull fracture (OR, 113), concern for nonaccidental injury (OR, 6.8), headache (OR, 2.6), and acting abnormally (OR, 1.8).

Comment by NEJM editors

Children with head injury and isolated vomiting are very unlikely to have intracranial injuries. Unless non-accidental trauma is suspected, vomiting alone should not instigate head imaging. These children should be observed until they are able to tolerate oral intake and the treating clinician feels comfortable that the patient is stable without additional symptoms of head injury.

Does Fentanyl Cause Hypotension in Patients Undergoing Rapid Sequence Intubation in the ED?

A secondary database analysis finds an association between fentanyl use and postintubation hypotension.

Opioids are sometimes given as pretreatment agents to mitigate the catecholamine release associated with laryngoscopy, although benefit from this practice has not been proven. Fentanyl is an opioid often chosen for its lack of histamine-releasing properties and its short duration. Nevertheless, its lessening of adrenergic tone may predispose recipients to hypotension. Investigators conducted a secondary analysis using a database of intubations in 14 Japanese emergency departments to determine whether pretreatment with fentanyl was associated with postintubation hypotension, defined as a systolic pressure ≤90 mm Hg, during rapid sequence intubation.

Among 1263 patients without baseline hypotension, 37% had pretreatment with fentanyl, and the rest comprised the nonfentanyl group. Postintubation hypotension was more common in the fentanyl than the non-fentanyl group (17% vs. 6%) although there was no distinction as to the clinical significance.

Comment by NEJM editors

Like other opioids and sedatives that can reduce adrenergic tone, fentanyl may cause hypotension. Blunting of adrenergic tone during laryngoscopy has not been proven beneficial — nor has any “pretreatment” regimen, for that matter. If you choose to pretreat with fentanyl, do so only in patients not at risk for hypotension.

Fewer Adverse Kidney Events in Patients Who Are Given Balanced Crystalloids (Ringe's Lactacte)

Acutely hospitalized patients had better outcomes with lactated Ringer's solution than with normal saline.

What is the right crystalloid for fluid resuscitation? Because limited data guide this decision, choices traditionally have been driven by culture: saline in medical units, and lactated Ringer's solution for surgical patients. With emerging evidence showing that the hyperchloremia associated with administration of normal saline might cause harm, interest in the effects of various crystalloids has increased.

During 2 years in two parallel studies, investigators at Vanderbilt University assigned patients to receive either normal saline or a more-balanced crystalloid solution (i.e., lactated Ringer's or Plasma-Lyte). One cohort comprised 15,802 patients who were admitted to the intensive care unit (ICU), and the other cohort comprised 13,347 patients admitted from the emergency department (ED) to non-ICU acute care. The study design was unique: For each month of the trials, the ED and ICU were assigned to use either saline or a balanced crystalloid, and computer order entry systems directed providers to use the assigned fluid. The unassigned fluid also was available, but few patients received it. Most patients who received balanced crystalloid solutions were given lactated Ringer's.

In the ICU trial, significantly more patients in the saline group (15% vs. 14%) suffered major adverse kidney events (a composite of death, renal replacement therapy, or a doubling of creatinine at discharge). This effect was most pronounced in patients with sepsis and in those who received larger volumes of fluid. The primary outcome (hospital-free days) in the ED trial was similar in the two groups; however, major adverse kidney events were also significantly higher for the ED saline group (6% vs. 5%).

Comment by nejm editors

Trials from a single center should almost never be the definitive word on practice. However, these two trials push us toward defaulting to lactated Ringer's rather than saline in most patients. These data suggest the benefit of balanced crystalloid solution is greatest in septic patients and in those who require large volume resuscitation; these might be the populations where grabbing lactated Ringer's makes the most sense.

Inhaled Isopropyl Alcohol Superior to Oral Ondansetron as an Antiemetic

Astonishing results from a small, well-designed study may have far-reaching implications.

Though ondansetron is viewed by many as the first-line agent for nausea in the emergency department (ED), there is evidence it doesn't work in noncancer patients (NEJM JW Emerg Med Aug 2014 and Ann Emerg Med 2014; 64:526). An alternative agent, inhaled isopropyl alcohol, has shown promise (NEJM JW Emerg Med Feb 2016 and Ann Emerg Med 2016; 68:1).

In the current trial, 120 adult ED patients with nausea or vomiting who did not require intravenous access were randomized to inhaled isopropyl alcohol plus 4 mg oral ondansetron; inhaled isopropyl alcohol plus oral placebo; or inhaled saline plus 4 mg oral ondansetron. Isopropyl alcohol was provided in the form of a standard alcohol swab. Patients received a single dose of the oral intervention but could sniff alcohol or saline swabs repeatedly. Nausea was measured on a 100-mm visual analog scale at baseline and 30 minutes.

Mean nausea scores decreased by 30 mm in the alcohol/ondansetron group, 32 mm in the alcohol/placebo group, and 9 mm in the saline/ondansetron group. Rescue antiemetic therapy was given to 28%, 25%, and 45% of each group, respectively. Differences between alcohol and saline groups were statistically significant. Patients in the inhaled alcohol groups also had better nausea control at the time of discharge and reported higher satisfaction with nausea treatment. No adverse events occurred. The mechanism of action is currently unknown.

Comment by nejm editors

It is uncommon for us to assign a rating of “Practice Changing” to a small, single-center study, but these results are truly remarkable and are consistent with prior research. For patients not obviously requiring IV therapy, we should treat nausea with repeated inhalations from an isopropyl alcohol swab instead of administering any other drug. And, although this study provides no direct evidence of benefit to patients who do require IV therapy, there would seem to be little downside to trying this simple and safe intervention in that group, too.

Cervical Spine MRI in Trauma Patients with Normal CT Is Not Cost Effective

An analysis suggests that outcomes are worse and costs higher when MRI is performed.

A prior meta-analysis found that no important injuries were missed by cervical spine magnetic resonance imaging (MRI) in blunt trauma patients with altered mental status (NEJM JW Emerg Med Apr 2015 and Ann Intern Med 2015; 162:429). The same investigators now report the results of a cost-effectiveness analysis on this topic.

The analysis considered a hypothetical 40-year-old trauma patient with a normal cervical spine computed tomography (CT) scan and a Glasgow Coma Scale (GCS) score <13. Estimates of the likelihood and consequences of foreseeable outcomes were derived from the literature, and decision analytic techniques were used to calculate the costs and benefits of two follow-up strategies after normal CT findings: cervical spine MRI or no MRI. The cost of the MRI approach averaged US$14,185 with a health benefit of 24.02 quality-adjusted life-years (QALY); the cost of the no-MRI approach averaged $1059 with a health benefit of 24.11 QALY. This means MRI was both more costly and less beneficial.

Comment by nejm editors

Editorialists aptly note that “the devil is in the details.” This cost-effectiveness analysis is filled with assumptions that make its generalizability and validity subject to debate. As one example, they defined patients with GCS score <13 as “obtunded,” while I would reserve that term for patients with GCS scores lower than this. Nevertheless, considering the totality of evidence published to date, my opinion is that it is not necessary to obtain cervical spine MRI for trauma patients who have normal cervical spine CT scans, unless there are documented focal deficits. Depressed mental status and midline cervical spine tenderness should not be indications for cervical spine MRI

Does Bolus-Dose Phenylephrine Cause Hypertension or Bradycardia in Emergency Department Patients?

A chart review study finds no evidence of harm from this intervention.

Bolus-dose phenylephrine at a dose of roughly ½ mg (500 μg) is a convenient way to support an emergency department (ED) patient's blood pressure while more-definitive measures are being prepared. But is it safe? Investigators reviewed charts at a single academic ED to determine whether patients had hypertension or reflex bradycardia after receiving bolus-dose phenylephrine. The ED had systematically facilitated the practice by providing pre-filled syringes with 1 mg phenylephrine in 10 mL saline.

Among 181charts reviewed, there were no serious adverse events. There were five mild adverse events: three cases of hypertension (systolic blood pressure >180 mm Hg) and two of bradycardia (heart rate <50 beats per minute) within 30 minutes of phenylephrine administration.

Comment by nejm editors

This case series suggests that bolus-dose phenylephrine may be safe in the ED setting but does not provide evidence of benefit and therefore is not practice changing. As the authors note, prospective research would be required to provide evidence of efficacy but given the commonness of this intervention in the operating room setting and the strong rationale for its use in the ED as a bridging measure, there seems little reason to avoid it. The fact that the study ED provided pre-filled syringes is notable because ad hoc bedside preparation of bolus-dose pressors is subject to measurement error.

Symptoms of Acute Myocardial Infarction in Young Women and Men

Beyond identifying differences between the sexes, this study highlights the need for clinicians to include AMI in the differential diagnosis of young adults.

It has been reported that women and men suffering acute myocardial infarction (AMI) present with different symptomatology. To ascertain this in adults aged ≤55, researchers analyzed data from standardized patient interviews in the multisite VIRGO study of 2985 patients hospitalized with AMI; patients were enrolled in a 2:1 female-to-male ratio (mean age, 47; 76% white).

Chest pain, pressure, tightness, or discomfort was a presenting symptom in 87% of women and 90% of men. Overall, women presented with a greater number of additional non–chest-pain symptoms than men; 62% of women and 55% of men presented with ≥3 such symptoms. In adjusted analyses, women aged >45 had a 40% greater likelihood of presenting without chest pain than men. Women with ST-segment elevation MIs were 50% more likely than men to present without chest pain. More than half of patients initially thought that their symptoms were due to a non–heart-related condition, with indigestion and acid reflux the most common self-diagnoses. Women were more likely than men to attribute their symptoms to stress and anxiety (21% vs. 12%); men were more likely to attribute them to muscle pain (21% vs. 15%). Women waited longer to seek medical attention (3.2 hours vs. 2.4 hours). Prehospitalization, more women than men had sought medical attention for similar symptoms (30% vs. 22%); over half of the women reported that the clinicians did not deem their symptoms to be heart-related (53% vs. 37% of men)

Use of Opioids for Migraine Treatment Is Common (but Inappropriate!)

In a study of three different emergency department settings, opioids were prescribed for 36% of migraine visits.
    Opioids are not indicated for treatment of migraines and should be avoided assiduously. Investigators assessed opioid use for migraine at three Connecticut emergency departments (EDs), one each in an academic medical center, an urban nonacademic hospital, and a community hospital.
    Review of a medical records database identified 1222 ED visits for migraine from 2014 to early 2015. Opioids were prescribed at 36% of these visits (12% of academic medical center visits, 41% of urban nonacademic ED visits, and 69% of community ED visits). Patients who received opioids had more repeat visits and longer hospital stays and required more rescue therapy than those who did not receive opioids.

Comment by nejm editors

Opioids should not be used for migraine headaches, or any other primary headache syndrome for that matter (e.g., tension, cluster). If you suspect a secondary headache syndrome (subarachnoid hemorrhage, glaucoma, trauma) or any other headache with a specific reversible anatomic etiology, opioids may be appropriate. Otherwise, rely on the extensive evidence-based armamentarium available for treatment of primary headache disorders. For migraine, use a triptan if the onset is within 1–2 hours. For migraines that have been present longer or do not respond to a triptan, an ideal cocktail is parenteral prochlorperazine (perhaps with diphenhydramine), oral nonsteroidal anti-inflammatory drugs, and oral caffeine. For non-responders, use dexamethasone (single dose), and for those patients with severe unremitting symptoms, consider subdissociative-dose ketamine. Do not use opioids for primary headache disorders, period.

Ottawa Subarachnoid Hemorrhage Rule Is 100% Sensitive and Can Rule Out the Diagnosis

However, with a specificity of only 13.6%, its impact on imaging and lumbar puncture rates may be low.
    Given the potential severity of missed subarachnoid hemorrhage (SAH), the diagnosis is an ideal one for a highly sensitive clinical decision rule. We previously summarized the Ottawa SAH Rule in 2013 (NEJM JW Emerg Med Nov 2013 and JAMA 2013; 310:1248) but noted its lack of validation and low specificity (15%) as barriers to use. Briefly, the rule is meant for neurologically intact patients 16 years or older with an atraumatic headache that peaked within 1 hour of onset. No further evaluation is needed if none of these six variables are present: age >40, neck pain or stiffness, witnessed loss of consciousness, exertional onset, thunderclap onset (immediate peaked pain), and limited neck flexion on exam.
    The researchers validated their rule in 1153 new patients at the same six Canadian emergency departments previously used. Of these patients, 67 (5.8%) were found to have SAH. Those discharged without evaluation were called, and no missed SAHs were found. Overall, the sensitivity of the rule was 100%, but the specificity was only 13.6%

Use a Non-Rebreather Mask for Preoxygenation for Rapid Sequence Intubation

With flush-rate oxygen, a non-rebreather mask was at least as good as a bag-valve-mask, and notably better when the BVM had a leak.
    Many intubators using the rapid sequence intubation protocol use a bag-valve-mask (BVM) for pre-oxygenation. But in the absence of a good mask seal, a BVM may not be superior to a non-rebreather mask (NRBM). In two crossover studies in healthy volunteers, investigators compared these devices using flush-rate oxygen (supply valve wide open) and measuring expired oxygen (FeO2) after pre-oxygenation as the primary outcome. Each study enrolled about 30 volunteers and each participant served as his or her own control.
    The first study assessed pre-oxygenation using the two devices (modified with a 1-way exhalation port) with and without a simulated mask leak. Mean FeO2 values were 81% for the NRBM, 76% for the BVM, and 30% for the BVM with a mask leak. The difference in FeO2 values between the NRBM and BVM was of borderline statistical significance (95% confidence interval, −1% to 10%), whereas the difference in values between the NRBM and the BVM with leak was highly significant. The second study compared use of the two devices with inspiratory assistance. In this study, mean FeO2 values were 83% for the NRBM with assistance versus 77% for the BVM with assistance, and this difference was statistically significant.

Comment by nejm editors

This investigation suggests that when flush-rate oxygen is used, better preoxygenation is achieved with a non-rebreather mask than with a bag-valve-mask, and markedly so when there is a BVM leak. I think the best approach to preoxygenation is a high-flow nasal cannula (to permit apneic oxygenation; NEJM JW Emerg Med Sep 2017 and Am J Emerg Med 2017; 35:1184) plus a NRBM with flush-rate oxygen. Not only will this achieve better oxygenation, but it's also more convenient and avoids the temptation to bag the patient, which is not indicated in uncomplicated rapid sequence intubation and risks filling the stomach with air.

Cricoid Pressure During Laryngoscopy — What Is It Good For?

Absolutely nothing. A small study finds that it is biomechanically impossible to maintain adequate cricoid pressure, even with real-time monitoring using a scale.
    A number of studies have called into question the practice of cricoid pressure (CP) — meant to occlude the esophagus and prevent aspiration during intubation but often resulting in worsened glottis views and more difficult tube passage — during intubation (NEJM JW Emerg Med Mar 2016 and Br J Anaesth 2016; 116:413; NEJM JW Emerg Med Sep 2014 and Anaesthesia 2014; 69:878; NEJM JW Emerg Med Aug 2010 and Resuscitation 2010; 81:810). Australian researchers used weighing scales to measure CP in real time during direct laryngoscopy, randomizing CP operators to be able or unable to see the scale in order to achieve the target pressure of 30 to 40 newtons. Their primary outcome was the difference in pepsin detection rates as a marker of esophageal occlusion.
    The study targeted inclusion of 212 patients but was terminated early after enrollment of only 70, once it was determined that neither group of CP operators could maintain the desired minimum 30 N of CP during direct laryngoscopy — despite both groups maintaining pressure in the target range during the induction phase (prior to insertion and elevation of the laryngoscope)

Comment by nejm editors

This should be the final nail in the coffin of cricoid pressure as it goes to dogma heaven (or in the other direction). The counterforce of direct laryngoscopy clearly makes application of adequate CP impossible, even with real-time monitoring. While video laryngoscopy might decrease the counterforce and allow for adequate CP to prevent aspiration, prior studies have already demonstrated that CP worsens the glottis view during intubation. Regardless of the laryngoscopic technique used, CP should not be.

Cuffed vs. Uncuffed Tracheal Tubes for Children: Time to Let Go of Old Dogma

In a randomized trial, cuffed tubes had better tidal volumes and less leakage.
Dogma suggests using uncuffed endotracheal tubes for children <8 years old. As the teaching goes, because the cricoid is the narrowest part of the airway, cuffs are unnecessary and may lead to tracheal stenosis. However, modern high-volume/low-pressure cuffs have become widely accepted for children.
    These authors compared leakage, ventilation, and complications with cuffed versus uncuffed endotracheal tubes in a randomized trial of 104 children aged ≤16 years undergoing elective surgery at a single pediatric center in Australia. Initial tube size was 0.5-mm smaller for cuffed than uncuffed tubes. Cuff pressure was set to ≤20 cm H2O and monitored continuously. Participants were followed for 3 to 5 years for long-term outcomes.
    Of children randomized to uncuffed tubes, 34% had limited data because airway leak was too large, or they required switching to a cuffed tube. During the first intubation attempt, a correctly sized tube was successfully placed in 80% of patients randomized to cuffed tubes, compared with only 30% of patients randomized to uncuffed tubes. Airway leakage (the primary endpoint) was significantly higher with uncuffed tubes (median, 0.92 vs. 0.20 mL/kg) and expiratory tidal volume was lower (median, 5.48 vs. 6.10 mL/kg). Findings were similar in the subgroup of patients <6 years old. Serious complications were rare, and minor complications (e.g., sore throat, hoarse voice) were more common with uncuffed tubes

Ketamine Is Effective Adjunct to Morphine for Controlling Acute Pain

Morphine plus ketamine performed better than morphine alone in this randomized, double-blind study.

Researchers performed a double-blind, randomized, placebo-controlled trial to evaluate the safety and effectiveness of subdissociative dose ketamine as an adjunct to morphine for acute pain. Sixty adult patients presenting to a level II community teaching hospital with acute pain (defined as pain with a duration of 15 days or less) were randomized to receive 0.1 mg/kg of morphine plus either 0.3 mg/kg of ketamine or placebo infused over 15 minutes. Pain was assessed on a traditional 10-point scale before administration and every 15 minutes after administration for up to 2 hours.

Pain scores were significantly lower in the morphine/ketamine group than in the morphine/placebo group at both 15 minutes (median scores, 3.6 vs. 6.0) and 30 minutes (3.0 vs. 5.0) after administration. There was no statistically significant difference in adverse events between groups and no occurrence of emergence phenomena in the ketamine group. Patients in the ketamine group also reported a higher rate of satisfaction with pain control (mean satisfaction score, 8.57 vs. 6.05 on a 10-point Likert scale)

Low-Dose Ketamine for Pain: IV Push or Slow Infusion?

Neuropsychiatric side effects were more common with intravenous push administration, but does it matter?

At intravenous (IV) doses of 0.1–0.3 mg/kg, ketamine is a safe and effective analgesic. Adverse effects are minimal and include sedation and a feeling of disconnection from reality. Investigators compared the effectiveness and adverse effect burden of ketamine administration by IV push versus slow infusion in a randomized, placebo-controlled, double-dummy trial.

Forty-eight emergency department patients with acute pain received 0.3 mg/kg of IV ketamine given either over 5 minutes by push or over 15 minutes by infusion. Feelings of unreality (quantified using the Side Effects Rating Scale for Dissociative Anesthetics) were more common in the IV push group overall (92% vs. 54%). Sedation was more pronounced in the IV push group, with median Richmond Agitation-Sedation Scale values at 5 minutes of −2 versus 0. The groups were comparable in terms of analgesic efficacy, but feelings of unreality were more pronounced in the IV push route at all time points

No Threshold for Hypotension in Patients with Traumatic Brain Injury

Mortality decreased with increasing prehospital blood pressures, even for patients who would not be considered hypotensive.

Prior studies have shown that even a single episode of prehospital hypotension or hypoxia increases mortality in patients with traumatic brain injury (TBI). Based on studies that dichotomized patients using various thresholds of systolic blood pressure (SBP), some guidelines define hypotension as SBP <90 mm Hg and provide corresponding treatment recommendations.

These authors questioned whether that threshold of 90 mm Hg represents a true inflection point for mortality or whether mortality continues to decline with even higher SBPs. In an observational study, they examined the relationship between in-hospital mortality and prehospital systolic blood pressures ranging from 40 to 119 mm Hg in a cohort of 3844 patients aged 10 years and older with TBI who were transported by Arizona emergency medical services.

Using robust statistical methods, the investigators demonstrated no threshold or inflection point for the relationship between mortality and SBP. At every 10-mm Hg interval of SBP from 40 to 119 mm Hg, mortality decreased when pressure was higher and increased when pressure was lower (adjusted odds ratio, 0.81 for each 10-mm Hg pressure increase)

Crank up the Oxygen and Skip the Bag-Valve-Mask

 Driver BE et al. Ann Emerg Med 2017 Jan.
High-flow oxygen through a non-rebreather mask is non-inferior to bag-valve-mask for pre-oxygenation.

Pre-oxygenation is a key step in rapid sequence intubation and involves allowing the patient to breath spontaneously with supplemental oxygen. The classic method, thought to achieve the highest alveolar oxygen concentration, is to use a bag-valve-mask (BVM) with high-flow oxygen and a good mask seal but no positive pressure (i.e., no squeezing of the bag). These investigators assessed whether pre-oxygenation with a nonrebreather mask delivering high-flow oxygen is non-inferior to pre-oxygenation with a BVM delivering oxygen at 15 L/minute.

Twenty-six healthy volunteers were pre-oxygenated with each of four methods in random order: nonrebreather mask with 15 L/minute oxygen flow; nonrebreather mask with the wall-mounted oxygen valve wide open (>40 L/minute flow); BVM with 15 L/minute oxygen flow, and standard mask with the wall-mounted oxygen valve wide open. The corresponding mean fractions of expired O2 in a single breath were 54%, 86%, 77%, and 72%

Comment by  Daniel J. Pallin, MD, MPH

The traditional BVM method is uncomfortable for patients because someone has to hold the mask tightly over the patient's face. Also, in my experience, this method is rarely used correctly, in that everyone seems to feel a compulsion to squeeze the bag, which risks filling the stomach with air and increasing the incidence of vomiting and aspiration.

This study shows that high-flow oxygen through a nonrebreather mask is not inferior to the BVM approach. However, it is unknown if optimizing preoxygenation influences patient outcomes, so it's hard to know what to recommend — a nonrebreather mask at 15 L/minute might be adequate for most intubations. Having read this study, I will no longer preoxygenate with a BVM. Instead, I will use a nonrebreather mask at 15 L/minute and then will open the oxygen valve to maximize flow during the last 3 minutes (or 8 tidal volume breaths for cooperative patients)

Should We Treat Patients with Isolated Calf Deep Venous Thrombosis?

In a randomized trial, treatment did not improve outcomes.

For isolated deep venous thrombosis (DVT) of the calf, American College of Chest Physicians guidelines suggest (1) anticoagulation in patients with severe symptoms or risk factors for proximal extension, and (2) in lower-risk patients, repeat ultrasound imaging during 2 weeks, plus anticoagulation for those whose DVTs extend proximally (Chest 2016; 149:315). However, data to inform these decisions are limited. In this double-blind trial, researchers in Europe randomized 259 adults with first acute isolated symptomatic calf DVTs, diagnosed by ultrasound, to receive either the low-molecular-weight heparin nadroparin (not available in the U.S.) or placebo for 6 weeks; follow-up visits with repeat ultrasound were scheduled at 3 to 7 days and at 6 weeks. Patients with active cancer or previous venous thromboembolic disease were excluded. About half of the patients had reversible risk factors (i.e., estrogen therapy or recent surgery, immobilization, or prolonged travel).

The primary composite outcome (proximal extension of calf DVT, contralateral proximal DVT, or pulmonary embolism at 6 weeks) occurred in 3% of nadroparin patients and 5% of placebo patients — a non-significant difference. Outcomes were similar at 3 months. Major bleeding occurred in 4% of nadroparin patients and no placebo patients (P=0.03). No participant died.

Comment by NEJM Editors

According to the authors, this study is the first placebo-controlled, randomized treatment trial for isolated calf DVT. The results suggest that, for patients like those enrolled in this trial, initial anti-coagulation is unnecessary if a second ultrasound examination is done within a week to screen for proximal extension. With its sample size, the study was underpowered to detect a several percentage point benefit for anti-coagulation, but it did remind us of bleeding risks during an even relatively brief course of anti-coagulation

Does Nasal cannula oxygen improve preoxygenation for intubation?

Hayes-Bradley C et al. Ann Emerg Med 2015 Dec 31.
Adding nasal cannula to nonrebreather mask or bag-valve-mask preoxygenation improved end-tidal oxygen levels.

Preoxygenation is used to fill the lungs with oxygen and wash out nitrogen before emergency endotracheal intubation. Investigators assessed the contribution of supplemental nasal cannula–delivered oxygen to end-tidal oxygen levels in a randomized crossover trial of 60 healthy volunteers who received either bag-valve-mask or nonrebreather mask ventilation in each of four scenarios: with and without nasal cannula oxygen and with and without mask leak. End-tidal oxygen levels were measured after 3 minutes of preoxygenation.

Oxygenation was not improved by the addition of nasal cannula oxygen to bag-valve-mask ventilation in the absence of mask leak. However, the addition of nasal cannula oxygen did improve oxygenation in the other scenarios: bag-valve-mask with leak (mean end-tidal oxygen, 66% with nasal cannula oxygen vs. 41% without) nonrebreather mask with leak (65% vs. 48%), and nonrebreather mask without leak (67% vs. 52%).

- See more at: http://www.jwatch.org/na40214/2016/01/22/does-nasal-cannula-oxygen-improve-preoxygenation#sthash.lkxDFsqh.dpu

This article came out in August 2015, basically showing that ANY interruption in cardiac compressions were bad.

Brouwer TF et al. Circulation 2015 Aug 7.

Longer pauses in chest compressions were associated with lower survival in patients with ventricular fibrillation cardiac arrest. Current cardiopulmonary resuscitation (CPR) guidelines emphasize continuous compressions with minimal interruptions. The focus has been on minimizing perishock pauses, as these were the types of pauses most easily measured during ongoing resuscitations. However, impedance recording capabilities of newer defibrillators allow measurement of all interruptions. To determine whether pauses for any reason are associated with reduced survival, researchers reviewed all cases of ventricular fibrillation out-of-hospital cardiac arrest in Amsterdam in 2009 for which all continuous impedance recordings were available (including from public access defibrillators). A total of 319 patients were included, with an overall survival to hospital discharge of 38%. A multivariate regression model demonstrated decreased odds of survival for each 5-second increase in the longest chest compression pause (odds ratio, 0.89). The longest pause was for reasons other than shock delivery in 36% of patients; survival was lower in these patients than in those whose longest pause was associated with defibrillation (27% vs. 44%). Comment – NEJM reviewer Ali Raja CPR shouldn't be paused for anything except defibrillation, and even those pauses should be very brief. Non-shock pauses (e.g., for intubation) are clearly associated with worse patient outcomes, and we should aim to minimize or eliminate them from our patients' resuscitations

Then this article, published in November 2015, showed a completely different result? Similar Survival Rates with Continuous or Interrupted CPR for Out-of-Hospital Cardiac Arrest

Nichol G et al. N andEngl J Med 2015 Nov 9. Koster RW. N Engl J Med 2015 Nov 9.

Both groups in this randomized trial received very high-quality compressions. Recent American Heart Association guidelines recommend changes to cardiopulmonary resuscitation (CPR) including that compressions be fast (100–120/minute), deep (5–6 cm), and continuous (NEJM JW Emerg Med Dec 2015 and Circulation 2015; 132[Suppl 2]:S315). However, the recommendation about compressions was based on retrospective cohort and animal studies and prospective studies involving multi-intervention bundles. To tease out whether continuous compressions independently improve survival in out-of- hospital cardiac arrest, Resuscitation Outcomes Consortium investigators performed a cluster randomized trial involving 114 emergency medical services agencies in North America. They randomized adults with nontraumatic cardiac arrest to receive continuous compressions (100 compressions/minute with asynchronous positive-pressure ventilation at 10 ventilations/minute) or interrupted compressions (30 compressions followed by 2 positive-pressure ventilations). Other basic and advanced life support interventions were not standardized. Of 23,711 patients analyzed, 53% received continuous compressions. Rates of survival to hospital discharge (the primary outcome) were similar in the continuous- and interrupted-compression groups (9.0% and 9.7%, P=0.07), as were rates of survival with favorable neurologic function (7.0% and 7.7%). Compressions were delivered in the two groups at similar depth (mean, 4.8 and 4.9 cm, respectively) and speed (mean, 110 and 109/minute, respectively). Mean pre- and post-shock pauses were the same in the two groups (pre-shock, 12 seconds; post-shock, 6 seconds). Comment: NEJM reviewer Ali Raji This well-done study demonstrates that continuous and interrupted high-quality compressions are equally effective. The take-home message is that high-quality CPR saves lives, regardless of compression strategy.