Should We Reconsider Antipyretics For Fever?

What is Fever?

Although often used interchangeably, the terms fever and hyperthermia refer to different processes, and the distinction is key. In fever the thermoregulatory set-point is elevated, and the body actively raises its temperature with chills and rigors to reach the new set-point. In hyperthermia the body’s temperature exceeds the set-point, due to increased heat production (eg hypermetabolic state) or decreased dissipation (eg high humidity or ambient temperature).1

Fever is generally defined as temperature ≥38°C (100.4°F) and results from a complex mechanism. The body produces pyrogens (specific cytokines) that act on the thermoregulatory center in the hypothalamus to increase the set-point. This is thought to occur by increased prostaglandin synthesis, and antipyretic drugs lower the set-point likely by inhibiting prostaglandin synthesis.2 There are also numerous endogenous antipyretics (cryogens).

Increased temperatures enhance immune function in many ways, including improved neutrophil migration and secretion of antibacterial substances, increased interferon, and increased T cell proliferation.1


Fever Anxiety

A 1980 study titled “Fever phobia: misconceptions of parents about fever” surveyed parents, and found 94% thought fever may have harmful effects, 18% thought brain damage or serious harm could result from fever <38.9°C (102°F), and 16% thought fever could rise up to 48.9°C (120°F) if untreated.3 A 2001 study re-examined similar questions, and found 76% believed serious harm could occur at ≤40°C (104°F).3

This phobia also exists among healthcare workers. A 1992 survey by the American Academy of Pediatrics in Massachusetts showed 65% of pediatricians thought fever alone is potentially dangerous, 72% “always or often” prescribed antipyretics for fever, and 89% recommended antipyretics for fever of 101-102°F.4 A 2000 study of pediatric emergency department nurses, with a median experience of 8 years, found 11% were unsure what temperature constituted fever, 29% thought permanent brain injury or death could occur from high fever, and 18% believed it is dangerous for children to be discharged from the emergency department if still febrile.5

Is Fever Harmful?

Some providers have concerns that the increased temperature or metabolic demand from fever will harm patients. Humans generally tolerate temperatures below 41°C (105.8F) without harm. In contrast to hyperthermia, it is extremely rare for fever as a host defense against infection to reach dangerous temperatures in neurologically normal patients, since the body is actively adjusting both the set-point and actual temperature.3 A 2011 American Academy of Pediatrics policy paper states “There is no evidence that fever itself worsens the course of an illness or that it causes long-term neurologic complications.” 6 Continue reading

CITW 13: Itch, itch, itch

Welcome back to another Clinical Image of the Week from the case files of the Brown EM Residency!

HPI/ROS: 37 year old male with no significant past medical history presents to the ED with a rash. He states that it began one month ago and has been getting worse. Associated symptom is intense pruritus. It is not painful and nothing of note has made it better or worse. He’s never had a rash like this before. He denies any fevers, chills, shortness of breath, chest pain, myalgia/arthralgias, abdominal pain, nausea, vomiting, diarrhea, or urinary symptoms. He denies any recent exposures (environmental or chemical), medication changes, recent infections, or sick contacts.

Vital Signs: T 98.6, HR 88, RR 14, BP 156/72, SpO2 99% on RA

Pertinent physical exam: Diffuse, papular rash along upper and lower extremities including trunk and back. The neck and face are spared. It is non-blanching, non-weeping, and there are no open sores. It spares the face, lower back, and calves. Patient appears well otherwise. No other pertinent exam findings.


TwoWhat’s the diagnosis?

Continue reading

You make me dizzy, Miss Lizzy: HINTS for assessing acute dizziness

The evaluation of the dizzy patient in the Emergency Department can be very complicated. The differential diagnosis is broad and misdiagnosis carries with it the potential for a high degree of morbidity and mortality. This summary will focus on patients with acute vestibular syndrome—patients with acute onset, continuous dizziness—and differentiating peripheral and central causes. We are not going to discuss vertigo or other causes of dizziness (see Table 1).

Screen Shot 2016-01-29 at 11.22.23 PM

Table 1: Timing and Trigger Approach to dizziness (from TiTrATE Approach to Dizziness, David E. Newman-Toker, MD, PhD)

One of the simplest ways to approach the dizzy patient is to categorize them based on the timing (episodic vs acute/continuous) and trigger (positional vs spontaneous; post-exposure vs spontaneous). Dizziness is difficult for patients to describe, so thinking of it in this way will minimize problems posed by vague and alternating histories to help narrow your differential and inform a targeted exam (Table 1).

A minority but significant percent of patients with acute vestibular syndrome have a stroke (population estimates range from 10-40%).1 Peripheral and benign causes includes cerumen impaction and benign positional vertigo. Symptoms are localized to the inner ear or CN8, are acute, benign, self-limited and often viral or post-viral. Vertigo with stroke is often abrupt, with maximum intensity at time of onset. Severe ataxia is common, but not always seen. Larger cerebellar strokes will produce symptoms localizing to the brainstem, like diplopia, dysarthria, limb ataxia, dysphagia, weakness or numbness, but 10% will have isolated vertigo without other symptoms.2

So, how can you tell the difference between central and peripheral causes?

Great reviews of vertigo can be found here, here, and here. While there are a number of articles in the Emergency Medicine literature, the most notable exam tool that is currently most broadly discussed is the HINTS exam.

The HINTS exam is an acronym for the three exam tools that help differentiate a non-hemorrhagic posterior cerebrovascular accident from other peripheral causes of acute vestibular syndrome. It stands for Head Impulse, Nystagmus, and Test of Skew. A number of derivation and validation studies, many of which were led by Dr. David Newman-Toker, have been published in recent years.

Kattah, et al6 found the HINTS. exam to be 100% sensitive and 96% specific for stroke in their 2009 publication in Stroke. In 2013, Newman-Toker, et al4 compared HINTS to the ABCD2 score in predicting ischemia in a prospective study among ED patients with one hour of dizziness at least one risk factor for stroke. They found HINTS to be superior to ABCD2 (ABCD2, SE61%, SP 62% vs HINTS SE 96.5% SP 84.4%), but both of these studies are limited by the fact that exams involved evaluations by specialists, primarily neuroophthalmologists, however Newman-Toker’s study included a vascular stroke trained emergency physician.

Finally, fresh off the presses is a new study just published in Neurology5 which used a combination of the ABCD2 score and HINTS exam to risk stratify patients into low, moderate, and high risk groups. The study was limited in that they used the HINTS exam in all patients, some who did not have nystagmus.   In the end, few patients with nystagmus who had a negative HINTS exam had a stroke.

Can EM physicians do it?

Obviously more studies remain to be done, but initial research is promising. Studies in EDs in Italy and Australia, for example, have shown high inter-evaluator reliability between emergency physicians and specialists7, 8 when assessing for stroke using exam findings of nystagmus, head impulse testing, and ability to ambulate.

What is the HINTS exam?

HINTS is a three part exam:

  1. Head Impulse
  2. Nystagmus
  3. Test of Skew
  • Head Impulse. This test relies on the vestibular-ocular reflex, which is a peripheral reflex that allows your eyes to focus while moving your head. When this is interrupted, your eyes lag. A POSITIVE test is consistent with peripheral vertigo. To

    Vestibulo-ocular Reflex

    say it another way, a positive head impulse test is reassuring.

    1. How to do it:
      1. Have patient fix eyes on your nose. And rapidly turn their head 20 degrees to one direction and back to center, then to other side.
        1. Eyes will stay fixed on your nose (if central)
        2. Eyes will first move to direction you moved their head, and then re-fixate on your nose. It is called “catch up” saccades. Eyes have to re-orient to your nose. = peripheral lesion


  • Nystagmus
    1. Peripheral vertigo should have fast beating nystagmus in only one direction.
    2. If it is direction changing = central
    3. Vertical nystagmus = central

  • Test of Skew
    1. You are looking for vertically disconjungate gaze (aka one eye is higher than other).
    2. This can be overcome by fixation, but can be ‘uncovered’ by the alternating cover test. Have them look at nose, cover one eye, will prevent covered eye from fixating. Then rapidly remove hand and see if eye misaligned will come back into alignment. Seeing realignment as eye is fixating on your nose. You are looking for that re-alignment.

To help guide your evaluation of the patient with acute vestibular syndrome, check out this new, handy guideline put together by Dr. Napoli and Dr. Siket:

Posterior stroke visio. AN 12#2

Guideline for the Acute Dizzy Patient in the Emergency Department, 2015

Try out the exam on your next dizzy patient and see how it goes. The best way to get comfortable with this exam is to practice!


Looking for more info?

Another EM overview of acute vestibular syndrome can be found on EMCrit.

Thank you to my Faculty Editors: Anthony Napoli, MD and Matt Siket, MD, MS

References/Further Reading:

  1. Tarnutzer AA, Berkowitz AL, Robinson KA, Hsieh Y, Newman-Toker DE. Does my dizzy patient have a stroke? A systematic review of bedside diagnosis in acute vestibular syndrome. CMAJ, June 14, 2011, 183(9).
  2. Nelson JA, Viirre K. Review: The Clinical Differentiation of Cerebellar Infarction from Common Vertigo Syndromes. West J Emerg Med. 2009;10(4):273-277.
  3. Newman-Toker DE, Kerber KA, Hsieh YH, et al. HINTS outperforms ABCD2 to screen for stroke in acute continuous vertigo and dizziness. Acad Emerg Med 2013;20:986–996.
  4. Seemungal BM, Bronstein AM. Review: A practical approach to acute vertigo. Pract Neurol 2008; 8: 211–221.
  5. Kerber KA, Meurer WJ, Brown DL, Burke JF, Hofer TP, Tsodikov A, Hoeffner EG, Fendrick AM, Adelman EE, Morgenstern LB. Stroke risk stratification in acute dizziness presentations: A prospective imaging-based study. Neurology 2015;85:1869–1878.
  6. Kattah, JC, Talkad AV, Wang DZ, Hsieh Y, Newman-Toker DE. HINTS to Diagnose Stroke in the Acute Vestibular Syndrome: Three-Step Bedside Oculomotor Examination More Sensitive Than Early MRI Diffusion-Weighted Imaging. Stroke 2009;40;3504-3510.
  7. Vanni S, Pecci R, Casati C, Moroni F, Risso M, Ottaviani M, Nazerian P, Grifoni S, Vannucchi P. STANDING, a four-step bedside algorithm for differential diagnosis of acute vertigo in the Emergency Department. Acta Otorhinolaryngol Ital 2014;34:419-426.
  8. Vanni S, Nazerian P, Casati C, Moroni F, Risso M, Ottaviani M, Pecci R, Pepe G, Vannucchi P, Grifoni S. Can emergency physicians accurately and reliably assess acute vertigo in the emergency department? Emerg Med Australas. 2015 Apr;27(2):126-31.

POCUS FOCUS: Lung Ultrasound for Pneumonia


The Case: A 5 year old previously healthy boy presents to the ED with three days of fever and progressive shortness of breath. His exam is notable for course rhonchi and decreased aeration in the right anterior lung. There was no personal or family history of asthma, eczema, or food allergies. He received an albuterol nebulizer with no effect.

The Diagnostic Intervention: Lung Ultrasound

Point of Care Lung Ultrasound:

Point of care lung ultrasound is highly useful in detecting pulmonary pathology commonly encountered in the emergency department, most notably pneumothorax, pulmonary edema, and pneumonia.  The basic technique of lung ultrasound in children is the same no matter what underlying process is suspected. The linear probe is used and oriented perpendicular to the rib in the longitudinal axis. This allows detailed visualization of the pleural line. This is perhaps the most important aspect of lung ultrasonography as 90% of respiratory dysfunction attributable to the lungs affects the pleura.1 Differences in the appearance of the pleural line help the clinician distinguish between causes of respiratory distress. There are several techniques described to adequately assess the lungs by ultrasound, the simplest of which is to image at three interspaces in the midclavicular, mid-axillary, and posterior lung fields.1,2

Figure 1.  Normal Lung Anatomy on Ultrasound

Screen Shot 2016-01-17 at 5.42.35 PM

Pneumonia on Ultrasound:

Pneumonia on lung ultrasound has a characteristic appearance. One can see “hepatization” of the lung as this normally echogenic and artifact filled tissue (see Figure 1) becomes echo-poor and increasingly tissue-like, resembling the liver on ultrasound (see Figure 2).  This is accompanied by “dynamic air bronchograms,” a branching lesion that courses through affected lung and moves with breathing. 3 This mobility helps distinguish pneumonia from atelectasis. A recent study found that dynamic air bronchograms have a 97% positive predictive value for pneumonia.4

How good is lung ultrasound for detecting pneumonia?

  • Sensitivity & Specificity in children > 90%5
  • When performed by emergency medicine physicians who have received a 1 hour training session, sensitivity drops to 86% but specificity is 97%.6
  • Ultrasound may also be better at detecting small pneumonias than standard chest x-ray.6 However, the clinical significance of this finding has yet to be determined.

Continue reading

Slow it or Fix it? Long-term Management Strategy for Atrial Fibrillation

This is part of a recurring series examining landmark articles in Emergency Medicine, in the style of ALiEM’s 52 Articles.

Main Points:

  1. This 2002 randomized multicenter trial found that the use of rate control drugs demonstrated similar mortality at five years (primary outcome) to antiarrhythmic strategy.
  2. Rate control strategy also demonstrated decreased hospitalization and decreased adverse drug effects (secondary outcome) compared to antiarrhythmic drugs.


At the time of the publication of this article, the initial management of atrial fibrillation remained unclear. The risk of stroke and increased mortality of atrial fibrillation were well recognized, but the relative risks and benefits of the different management approaches was widely contested. The AFFIRM (Atrial Fibrillation Follow-up Investigation of Rhythm Management) trial aimed to determine whether rate control or rhythm control was superior in long-term outcomes in patients with atrial fibrillation. The trial was a randomized multicenter trial enrolling a total of 4060 patients with a fib who were randomized to either rhythm control or rate control strategies. Both strategies had equal mortality, while the rate control strategy showed decreased hospitalization and adverse drug effects. The strengths of this study were its robust size, the generalizability and the study design. However there was significant cross over and cross back in this study as it was an intention to treat study.


This was a multicenter RCT that enrolled 4060 patients over the age of 65 with atrial fibrillation. The patients were randomized at each of the sties to either rhythm control or rate control strategy. Both groups had similar demographic and baseline characteristics including rates of CHF, duration of a fib, and ejection fraction. The rhythm control strategy allowed treating physicians to use cardioversion as necessary and treat with any of 9 options of antiarrhythmics. The rate control strategy allowed physicians to decide between beta blockers, calcium channel blockers or digoxin. The rate control group was required to be on Coumadin, while the rhythm group was encouraged to be on anticoagulation especially during the initial month to 12 weeks after rhythm control.

This was an intention-to-treat study. There was significant cross over and cross back observed during the study in both arms, though significantly greater cross over in the antiarrhythmic group. These patients were followed for an average of 3.5 years.

The primary outcome of the study was mortality. There was no significant difference in mortality between the rate control and rhythm control groups (p=.08). The authors state there is a trend towards lower mortality in rate controlled group at 3.5 years. Graphically they depict a divergence of mortality as time passes with rate control group trending towards lower mortality. They did find that it was significantly more likely for patients in rhythm control group to be hospitalized and experience adverse drug effect. p (p<.001).

Limitations to the study include the amount of cross over present, the variety in medications used in the rhythm treatment strategy. In addition this study is only applicable to those aged 65 or greater, and cannot be generalized to all patients with atrial fibrillation.

Level of Evidence:

Level II based on ACEP grading scheme for therapeutic questions.

Source Articles:

Wyse, D G et al. “A Comparison of Rate Control and Rhythm Control in Patients With Atrial Fibrillation.” The New England Journal of Medicine 2002; 347.23: 1825-1833.

Resident Reviewer: Dr. Kazakin
Faculty Reviewer: Dr. Siket

A True Orthopedic and Vascular Emergency

Case: A 76-year-old male presents after falling down a long flight of stairs. On exam the patient has multiple obvious external injuries, including a grossly deformed right shoulder with a large overlying hematoma. His chest x-ray and shoulder x-ray demonstrate a superiorly and laterally displaced right scapula, as well as a comminuted right scapular fracture and clavicle fracture. On further CT imaging, the patient has subtle widening of the scapulothoracic articulation.

OneQuestion: What potentially devastating injury should be considered in this patient?

Continue reading

CITW 12: The Extra Weight

Welcome back to another Clinical Image of the Week from the case files of the Brown EM Residency!

HPI: 27 year old male presents to the ED with left chest wall pain. He states he was bench pressing at the gym when he felt a “sudden twinge of pain and popping sensation” in his left upper arm and shoulder. He was unable to lift weights following this injury without experiencing excruciating pain. He subsequently noticed swelling and bruising on his left upper arm. He denies any other injuries, numbness, or weakness.

VS: HR 54, BP 112/64, RR 11, SpO2 100%, T 98.7

PE: Patient is noted to have weakness of the left arm with internal rotation and shoulder adduction. No other deficits appreciated. Neurovascularly intact.


What’s the diagnosis?

Continue reading

Coming Down the Pike: Zika Virus

From the Pan American Health Organization Zika virus website:

What is Zika Virus?
A single-stranded RNA virus of the Flavivirdae family, genus Flavivirus. The virus was first identified in 1947 in a rhesus monkey in the Zika Forest, Uganda.

What are the signs and symptoms?
Only about 1 in 4 people infected with Zika develop signs or symptoms, which include fever, maculopapular rash, arthralgias and conjunctivitis. Additionally, Zika causes headaches, myalgias, retro-orbital pain and vomiting.

How is it transmitted?
Zika Virus is primarily transmitted through the Aedes mosquito, which also transmits Dengue and Chikungunya. Transmission is also believed to take place vertically between mother and child, and through sexual contact.

Where has it been found?
As of the January 9, 2016, the following Pan American countries have seen confirmed cases of Zika virus: Brazil, Chile, Colombia, El Salvador, French Guiana, Guatemala, Honduras, Martinique, Mexico, Panama, Puerto Rico, Paraguay, Suriname, and Venezuela. 

Countries with confirmed Zika virus outside of the Americas include: Central African Republic, Egypt, French Polynesia, Gabon, India, Indonesia, Malaysia, Nigeria, The Philippines, Sierra Leone, Tanzania, Thailand, Uganda, and Vietnam

Why is it in the news?
Zika virus made national headlines in the United States in late December 2015 when Brazillian health officials advised would-be parents to delay pregnancy over concerns that Zika virus is contributing to a spike in microcephaly. The Brazil Ministry of Health reports a twenty-fold increase in the incidence of microcephaly over the past year in areas that have had confirmed Zika virus transmission (2,782 cases in 2015 versus 147 cases in 2014). The connection was made in November 2015 when Brazilian health officials found traces of Zika virus in a deceased newborn born with microcephaly.
From the Pan American Health Organization Zika virus website, Epidemiological Alert, December 1, 2015

Additionally, Brazil has reported an increase in neurological syndromes in patients infected with Zika virus, most notably Guillain-Barré syndrome.

What is the treatment?
Supportive care: rest, fluids, antipyretics, and analgesics. Hold aspirin or NSAIDs until Dengue has been ruled out to reduce the risk of hemorrhage.

Pan American Health Organization Zika virus website:

Brazil warns against pregnancy due to spreading virus, CNN, December 23, 2015

Foy BD, Kobylinski KC, Foy JLC, Blitvich BJ, Travassos da Rosa A, Haddow AD, et al. Probable non–vector-borne transmission of Zika virus, Colorado, USA. Emerg Infect Dis. 2011 May;

Pan American Health Organization Epidemiological Alert: Neurological syndrome, congenital malformations, and Zika virus infection. Implications for public health in the Americas; December 1, 2015. PDF Direct Link

A new mosquito-borne threat to pregnancy women in Brazil, The Lancet, published online December 23, 2015.

Hit Them Hard: The Use of H1 and H2 Antagonists for Acute Allergic Symptoms

This is part of a recurring series examining landmark articles in Emergency Medicine, in the style of ALiEM’s 52 Articles.

This blog post reviews the article by Lin, R. Curry, A. Pesola, G. et al. “Improved Outcomes in Patients With Acute Allergic Syndromes Who Are Treated With Combined H1 and H2 Antagonists.” Annals of Emergency Medicine, November 2000; 36(5):462-8.

Main Points:

  1. In this randomized, double blinded, placebo controlled trial of 91 patients presenting with acute allergic symptoms fewer patients in the active arm (ranitidine + diphenhydramine) had signs of cutaneous involvement such as urticaria at 2 hours compared to the placebo group (placebo + diphenhydramine).
  1. There was no significant difference, however, between the placebo group and active group with regards to the absence of erythema or angioedema at two hours.


Many patients present to the emergency department with acute allergic syndromes. Anti-histamines, primarily diphenhydramine, are the mainstay therapy in most mild cases and are both safe and cost effective. The addition of H2 antagonists such as ranitidine to diphenhydramine may help improve clinical outcomes and expedite management in the emergency department. The primary goals of this study were to look for resolution of urticaria and angioedema at two hours from presentation. This study was well balanced in its patient recruitment and overall provides insight into real-world application of a second agent for management of allergic symptoms.


The methodology within this study was rigorous, though the sample size was small. This trial was a randomized, double blinded, placebo controlled trial that enrolled 91 patients at an academic emergency department in New York, NY. Enrollment was based on a convenience sampling associated with the study physicians’ scheduling. Patients were enrolled in this study if they were adults who presented with acute urticaria, acute angioedema, acute unexplained stridor or acute pruritic rash following an exposure to a food, drug or contact with latex. Patients underwent vital sign monitoring, examination for physical findings such as: presence and extent of urticaria and erythema, presence of angioedema, wheezing, stridor, abdominal distention or tenderness, as well as symptom scoring. This data collection occurred at presentation, at 1 hour and 2 hours.

This study demonstrated a statistically significant difference, p=0.03, in the resolution of urticaria in the active group compared to the placebo group at two hours. One significant limitation in this study is that the treating physician was able to administer supplemental medications such as epinephrine, corticosteroids, bronchodilators and additional doses of antihistamine at their discretion with significantly more participants in the ranitidine arm receiving epinephrine, 17, compared to 9 in the placebo arm. The placebo arm had more use of additional antihistamine, 10, compared to 2 in the ranitidine arm. These additional therapies are documented in table 3; however, it is unclear if the severity of illness was equal between the two groups. The authors do note in their discussion that there was no observed covariate effect for epinephrine administration with respect to urticaria resolution.

Level of Evidence:

This study was graded a level I based on the ACEP Clinical Policy Grading Scheme


Within the sample groups there was significant history of asthma as well as other nonasthmatic atopic conditions which supports the theory that certain individuals are genetically predisposed to allergic syndromes.

Source Articles:

Lin, R. Curry, A. Pesola, G. et al. “Improved Outcomes in Patients With Acute Allergic Syndromes Who Are Treated With Combined H1 and H2 Antagonists.” Annals of Emergency Medicine, November 2000; 36(5):462-8.

Faculty Reviewer: Dr. Siket

Perusing the Literature (PTL): Steroids for Anaphylaxis?

This month we continue our Perusing of the Literature. Once again, this section consists of recent articles that residents and attendings have stumbled across that have raised an eyebrow. These posts are meant to spark a discussion and do not represent a change in the standard of care (unless otherwise noted).

January 2016:

The Article: Grunau BE, et al. Emergency Department Corticosteroid Use for Allergy or Anaphylaxis Is Not Associated With Decreased Relapses. Ann Emerg Med. 2015;66:381-389.

The One-Liner: Corticosteroid use does not appear to be associated with a decrease in allergy-related ED revisits in patients presenting with allergic reactions or anaphylaxis.

Background: Allergy-related ED visits account for approximately 1.0% of all visits. Aside from antihistamines, corticosteroids use in allergic reactions has increased from 22% to 50% from 1993-2004. The anti-inflammatory effect of steroid therapy intends to reduce risk of biphasic reactions, decrease the severity of reactions, and decrease ED return visits.

Methods: This was a retrospective cohort study in 2 urban EDs during a 5-year period, assessing patient encounters classified as ‘anaphylaxis’ or ‘allergic reaction’. The primary aim of this study was to determine if steroid administration in ED allergy patients decreased relapses to additional care within 7-days. Authors also aimed to identify potential benefits of steroids in decreasing death, clinically important biphasic reactions, or all-cause repeated ED visits. There were no defined protocols for allergic reactions and physicians managed the patients in an individualized manner. The patient’s health number was used to identify all patients who returned to any regional ED or died within the province during a 7-day follow-up period.
Anaphylaxis was defined as meeting 1 of the following 3 criteria:
• Skin or mucosal involvement AND respiratory compromise or SBP • Two of the following after exposure to ‘likely allergen’: 1) Skin or mucosal involvement, 2) respiratory compromise, 3) SBP • SBP An allergic reaction was defined as a clinical presentation in which criteria for anaphylaxis were not met, but provider deemed the cause of signs/symptoms to be result of allergic process.

• Total of 2701 ED patient encounters
• Corticosteroids administered to 1181 (44%) patients; 469 (17%) received parenteral, 786 (29%) received oral formulation; prescription for oral steroid at ED discharge in 819 (30%) encounters
• During the 7-day follow-up period, there were 170 (6.3%) allergy-related revisits – 75 (5.8%) patients in the steroid group and 95 (6.7%) in the nonsteroid group (95% CI 0.63 to 1.17).
• Anaphylaxis patients (n=473); there were 15/348 (4.3%) allergy-related re-visit in the steroid group, and 7/125 (5.6%) in the non-steroid group (95% CI 0.41 to 3.27)
• Allergic reaction patients (n=2228); there were 60/940 (6.4%) allergy-related revisits in the steroid group, and 88/1288 (6.8%) in the non-steroid group (95% CI 0.63 to 1.31)
• NNT with steroids to prevent 1 additional ED revisit was 176
• No deaths during any of index visits or within follow-up period for any patient (2698/2715; 99.4% patients able to be have provincial data linkage established)
• 5 clinically important biphasic reactions. 4 in steroid group, 1 in non-steroid group.

• Clinical impression was the basis for the diagnosis of allergic reaction
• Cannot confirm filling of prescriptions or tolerance of steroid course
• Physicians on index visit may have instructed patients to return for reassessment
• Patients with biphasic reactions may have represented out of the region

Author: Cameron Gettel, MD PGY1
Resident Section Editor: Adam Janicki, MD PGY4
Reviewed by Gita Pensa, MD, Clinical Assistant Professor, Department of Emergency Medicine