|Home | About | Journals | Submit | Contact Us | Français|
Anaphylaxis guidelines recommend that patients with a history of anaphylactic reaction should carry self-injectable epinephrine and should be referred to an allergist.
To evaluate how frequently patients dismissed from the emergency department after treatment for anaphylaxis received a prescription for self-injectable epinephrine or allergist referral.
A retrospective medical record review identified patients with anaphylaxis in a community-based study from 1990 through 2000. Records of patients with Hospital Adaptation of the International Classification of Diseases, Second Edition or International Classification of Diseases, Ninth Revision codes representing anaphylaxis were reviewed, and a random sample of patients with associated diagnoses was also reviewed. Patients who met the criteria for diagnosis of anaphylaxis were included in the study.
Among 208 patients identified with anaphylaxis, 134 (64.4%) were seen in the emergency department and discharged home. On dismissal, 49 patients (36.6%; 95% confidence interval [CI], 28.4%–44.7%) were prescribed self-injectable epinephrine, and 42 patients (31.3%; 95% CI, 23.5%–39.2%) were referred to an allergist. Treatment with epinephrine in the emergency department (odds ratio, 3.6; 95% CI, 1.6 –7.9; P = .001) and insect sting as the inciting allergen (odds ratio, 4.0; 95% CI, 1.6 –10.5; P = .004) were significantly associated with receiving a prescription for self-injectable epinephrine. Patient age younger than 18 years was the only factor associated with referral to an allergist (P = .007).
Most patients dismissed after treatment for anaphylaxis did not receive a self-injectable epinephrine prescription or allergist referral. Emergency physicians may be missing an important opportunity to ensure prompt treatment of future anaphylactic reactions and specialized follow-up care.
Anaphylaxis is a systemic allergic reaction. Symptoms of anaphylaxis can range from mild and self-limited to rapidly fatal.1,2 Studies in children have shown that anaphylaxis is most commonly treated in emergency departments (EDs).3 In this study of children and adults, we also found that most of the patients were seen in the ED, making emergency physicians the most likely physicians to diagnose, treat, and facilitate follow-up. Guidelines recommend that all patients with a history of anaphylaxis to an allergen that they may encounter in a nonmedical setting should be given a prescription for self-injectable epinephrine. They should also be referred to an allergist for follow-up.2,4 However, studies evaluating patients presenting to the ED specifically with food allergies and insect sting allergies both found that less than 25% of patients with anaphylaxis were prescribed self-injectable epinephrine when discharged from the ED and even fewer were referred to an allergist.5,6
We investigated the clinical presentation, ED management, home-going treatment, and follow-up of a community-based cohort of patients presenting to the ED with anaphylaxis. We determined how frequently patients were dismissed with a prescription for self-injectable epinephrine and referred to an allergist and which factors were associated with receiving a prescription for self-injectable epinephrine or allergist referral.
This retrospective study was performed by identifying cases of anaphylaxis occurring in patients in Olmsted County, Minnesota, from 1990 through 2000. Patients were seen in the ED at Saint Mary’s Hospital, a tertiary care academic medical center affiliated with Mayo Clinic, with approximately 70,000 visits per year, and at Olmsted Medical Center, a community-based ED, with approximately 19,000 visits per year. The study was approved by the institutional review boards at both centers. Informed consent was obtained from all research participants. The study linked and indexed records of virtually all medical care professionals in this defined geographic region. 7–9 All diagnoses were coded using either Hospital Adaptation of the International Classification of Diseases, Second Edition (HICDA) codes or International Classification of Diseases, Ninth Revision (ICD-9) codes. We searched for all patients who had a new diagnostic code related to anaphylaxis from 1990 through 2000, who were residents of Olmsted County, and who had not refused permission for their medical records to be used for research. Only the initial episode of anaphylaxis during the study period was included for each patient. All 248 patients with ICD-9 codes (or corresponding HICDA codes) representing the following diagnoses were reviewed: anaphylactic shock; anaphylactic shock due to food; anaphylactic shock not elsewhere classified; and shock, anaphylactic, following sting. In addition, a random sample of 600 patients from 2,442 patients with associated diagnoses was also reviewed: 300 patients diagnosed as having either venom from bee sting or toxic effect of venom and 300 patients diagnosed as having allergy, foodstuff; adverse effect, food; dermatitis due to food taken internally; or toxic effect of specific food. Of the 248 patients, 157 met criteria for anaphylaxis, and of these, 92 were treated and dismissed from the ED. Of the randomly chosen 600 cases, 54 met the initial criteria for anaphylaxis, and of these, 43 were treated and dismissed from the ED. Therefore, of the total of 211 cases of anaphylaxis initially reviewed, 135 were treated and dismissed from the ED.
This study was started before the second symposium on the definition and management of anaphylaxis. Initially, the criteria for anaphylaxis used in the study by Yocum et al were used to identify patients. These criteria are similar to those developed by the second symposium on the definition and management of anaphylaxis.2,7,10 To ensure that the cases included in this study met the most recent and accepted criteria for anaphylaxis, all 211 cases that had been identified using the Yocum criteria were reanalyzed using the criteria proposed by the second symposium. There were 3 cases that did not meet the criteria proposed by the second symposium and these were removed. This left a total of 208 remaining cases of anaphylaxis. One hundred thirty-four of these cases were treated and dismissed from the ED. The criteria for diagnosis of anaphylaxis proposed by the second symposium are given in Table 1.
Records were screened to determine if the date and assigned code met inclusion criteria. Nurse abstractors reviewed medical records and abstracted demographic data, presenting symptoms, ED treatment, home-going treatments, and follow-up referrals.
There were 2 abstractors during the study. Abstractor 1 conducted a pilot study of 22 medical records to test the criteria for defining a case of anaphylaxis, to determine the amount of time needed to collect each data point, and to finalize the data collection forms. Abstractor 2 joined the study 2 months after abstractor 1 and reviewed 10 medical records that had been reviewed previously by abstractor 1. Time needed to review medical records was approximately the same for both abstractors, and data comparability was excellent (>95%). Abstractors and investigators met monthly to discuss progress and to resolve abstracting or coding questions. Investigators reviewed medical records for which collection of data points was unclear and were available on an ad hoc basis to resolve questions.
The primary outcomes of interest were prescriptions for self-injectable epinephrine and allergist referrals for patients presenting to the ED with anaphylaxis who were dismissed from the ED.
Statistical analysis was based on data obtained from reviewed cases only. Univariate and multivariate logistic regression models were fit to evaluate factors for an association with outcomes of interest on ED dismissal: (1) prescription of self-injectable epinephrine and (2) referral to an allergist. Odds ratios (ORs) and corresponding 95% confidence intervals (CIs) were calculated for each association. P values were 2-sided, and P < .05 was considered statistically significant. Statistical analyses were performed using the SAS software package (version 9.0) (SAS Institute Inc, Cary, North Carolina).
Among 208 reviewed patients with anaphylaxis, 134 (64.4%) were seen in the ED and discharged home. The demographics of this cohort are given in Table 2. The mean age of the patients was 29.3 years, 55.2% were female, and 84.3% were white. The inciting allergens are also summarized in Table 2. The inciting agent was food in 38.1%, insect stings in 19.4%, medication in 8.2%, unspecified in 24.6%, and other agents in 9.7%.
Table 3 summarizes the symptoms of the 134 reviewed cases. Twenty-six patients (19.4%; 95% CI, 12.7%–26.1%) presented with symptoms involving 3 organ systems (oral and gastrointestinal tract, cardiovascular system, and respiratory tract) and symptoms of mucocutaneous mediator release.
Forty-nine patients (36.6%; 95% CI, 28.4%–44.7%) were prescribed self-injectable epinephrine before ED discharge, and 42 patients (31.3%; 95% CI, 23.5%–39.2%) were referred to an allergist. Those prescribed epinephrine on dismissal were more likely to be referred to an allergist compared with those who were not prescribed epinephrine (46.9% vs 22.4%; P = .003).
Table 4 summarizes the factors evaluated for an association with prescription of self-injectable epinephrine on ED discharge. On the basis of multivariate logistic regression analysis, treatment with epinephrine in the ED (OR, 3.6; 95% CI, 1.6 –7.9; P = .001) and insect sting as the inciting allergen (OR, 4.0; 95% CI, 1.6 –10.5; P = .004) were identified as being associated with receiving self-injectable epinephrine at ED dismissal. Patient age, history of asthma, other inciting allergens, other types of ED treatment, and presenting symptoms were not associated with receiving a self-injectable epinephrine. Patient age was the only factor associated with referral to an allergist; among patients younger than 18 years, 18 to 50 years, and older than 50 years, 49%, 27%, and 11%, respectively, were referred (P = .007).
We also evaluated how frequently the 74 patients who were diagnosed as having anaphylaxis but not dismissed from the ED received a prescription for self-injectable epinephrine. Forty-two of these patients were admitted to the hospital, and 26 (62%) received a prescription for self-injectable epinephrine. The remaining 32 patients were evaluated in other outpatient settings (urgent care or outpatient clinics) and 16 (50%) received a prescription for self-injectable epinephrine.
We found that 36.6% of patients who experienced anaphylaxis were prescribed self-injectable epinephrine and 31.3% were referred to an allergist before dismissal. These results are slightly higher than, although generally consistent with, those of prior studies evaluating patients specifically with food allergies or with insect sting allergies.5,6 We also found that patients treated in other outpatient settings and those who had been admitted to the hospital for anaphylaxis were more likely to receive prescriptions for self-injectable epinephrine than those who were dismissed from the emergency department.
Epinephrine has been shown to be an effective treatment for anaphylaxis,11 and guidelines recommend that all patients who have anaphylaxis to an allergen that may be encountered in nonmedical settings should be prescribed self-injectable epinephrine.2,4 Studies have shown that patients who do not have epinephrine available at the time of the anaphylactic reaction are at increased risk of a fatal reaction.12–14 In an analysis of 32 patients with fatal anaphylaxis reported to a national registry established by the American Academy of Allergy, Asthma, and Immunology, 90% did not have epinephrine available at the time of their reaction.12 Consistent with studies in children, we found that most patients with anaphylaxis were treated in an ED.3 Hence, proper diagnosis, treatment, and referral in the ED are important.
The low rate of prescriptions for self-injectable epinephrine and allergist referrals among patients dismissed from the ED may have several explanations. Lack of documentation may be partially responsible. Failure to diagnose anaphylaxis may be another reason. A study by Klein and Yocum10 found that many patients who are diagnosed as having an allergic reaction in the ED were subsequently, on allergist review, considered to have experienced an anaphylactic episode.
The most important factor contributing to failure to diagnose an anaphylactic reaction may be the lack of a universally accepted definition. This was addressed during the second symposium on the definition and management of anaphylaxis, and criteria for the diagnosis of anaphylaxis were proposed.2 There is growing consensus that anaphylaxis is a systemic allergic reaction that typically involves more than 1 organ system but can present with cardiovascular compromise as the only symptom. The criteria for diagnosis of anaphylaxis proposed by the second symposium are given in Table 1.
On the basis of these criteria, anaphylaxis need not be life-threatening. Respiratory tract and cardiovascular symptoms are not mandatory for diagnosis of anaphylaxis. This is an important distinction because historically many practitioners have believed that cardiovascular or respiratory tract symptoms were invariably present in anaphylaxis.
It is imperative that ED physicians diagnose even mild anaphylaxis (for example, rash and vomiting after an allergen exposure). These patients should be provided with self-injectable epinephrine and follow-up referrals because subsequent reactions may be life-threatening. In a study of 139 fatalities, 78% of deaths attributed to food allergy and 82% of deaths attributed to venom allergy occurred in patients with no previous history of severe allergic reactions.13
Recent studies have shown that biphasic reactions occur in approximately 5% to 20% of anaphylactic reactions. The second phase can occur 4 to 72 hours after the initial reaction. 14–20 The second reaction can be more severe than the initial reaction and can involve organ systems not involved in the initial reaction. Biphasic reactions can be fatal.14 Thus, patients need to have self-injectable epinephrine available to them during this early period. Usually, the patient is able to see a primary physician or allergist only at a later date.
We found that patients who received treatment with epinephrine in the ED were more likely to be prescribed self-injectable epinephrine. These findings are consistent with prior studies that have evaluated predictors for self-injectable epinephrine prescriptions in patients with food allergies5 or insect stings.6 However, we did not find that patients with a history of asthma were more likely to be prescribed self-injectable epinephrine. This is in contrast to a study of patients presenting with food allergies (including anaphylaxis) by Clark et al,5 who found that patients with history of asthma were more likely to be prescribed self-injectable epinephrine on dismissal from the ED.
Asthma and history of allergy to peanuts or tree nuts are known risk factors for fatal anaphylaxis.12,14 Bock et al12 analyzed a national registry of fatal anaphylactic reactions and found that 93% of patients who died had a history of asthma and 94% were allergic to peanuts or tree nuts. It is important that emergency physicians recognize these risk factors.
Although guidelines recommend referral to an allergist, this may not be feasible in many health care settings. It may be more realistic to refer patients to their primary care physician, who can then make appropriate referrals. Patients must understand that follow-up care is needed. One study showed that many patients treated in the ED for anaphylaxis did not realize that follow-up care was needed.10
The primary limitation of this study is its retrospective nature. Another limitation imposed by our study setting is that the characteristics of the study population are similar to the US white population and may not be generalizable to minority or ethnic populations.7 Finally, the clinical criteria used to diagnose anaphylaxis will affect the results. All patients included in this study met the criteria proposed by the second symposium on the definition and management of anaphylaxis. However, to date, no universally accepted set of criteria for diagnosis of anaphylaxis is available.
Most patients treated for anaphylaxis are not prescribed self-injectable epinephrine or referred to an allergist on ED discharge. Emergency physicians may be missing an important opportunity to ensure prompt treatment of future anaphylactic reactions and specialized follow-up care. Collaboration between emergency physicians and allergists with a goal toward agreement on criteria for diagnosis of anaphylaxis and indications for prescription of self-injectable epinephrine will ensure better management of anaphylaxis. These efforts are already ongoing as evidenced by the Be SAFE guide, a tool developed by a panel of allergists and emergency physicians to improve ED management of anaphylaxis. 21
Editing, proofreading, and reference verification were provided by the Section of Scientific Publications, Mayo Clinic.
Funding Sources: This study was supported in part by a research grant provided by the Food Allergy and Anaphylaxis Network and Mayo Foundation for Medical Education and Research.
Disclosures: Authors have nothing to disclose.