From the Gage Occupational and Environmental Health Unit, and the Health Sciences Research Centre, St. Michael's Hospital, Toronto; University of Toronto, Toronto, Ontario.
Reprinted from the Canadian Journal of Allergy & Clinical Immunology. Zimmerman B, Urch B, Mercado B. Food Allergy: Frequency of Adrenaline Administration. Can J Allergy Clin Immunol 6: 159-61;2001.
Objectives:To determine whether the teaching and prescription of an epipen to children who are sensitized to peanut leads to appropriate use of the epipen with subsequent reactions.
Design: Analysis of data from consecutive peanut skin-test positive patients returning for re-evaluation in one doctor's office practice from January 1, 1999 to December 31, 1999.
Setting: Non-university office practice
Subjects: 96 children previously found to be skin test positive to peanut who returned for re-testing with peanut at least 1 year later.
Main outcome measures: Type and severity of allergic reactions to peanut and other foods, results of skin prick tests, and incidence of other allergic diseases and associated allergies.
Results: Ninety-six patients who tested positive to peanut returned for re-assessment and 10 of them had lost the positive skin tests and successfully underwent oral challenge without reacting. Of the 86 patients who retained the positive skin test to peanut, 26 had a total of 41 further reactions to food. Twenty-one of the reactions were associated with some form of respiratory symptoms but in only 6 instances (29%) was the epipen used.
Conclusion: The data suggest that teaching and prescription of epipens to peanut allergic patients does not guarantee the appropriate administration of adrenaline.
Peanut sensitivity is a common clinical problem in children (1,2). We wished to determine in a group of food-allergic patients who had been prescribed and taught the use of adrenaline auto-injection whether the epipen was actually administered when they developed further reactions to foods. We chose to study patients who had been taught the use of epipen autoinjector because of sensitivity to peanut. We reasoned that peanut sensitive patients would be a good group of patients to determine usage of the epipen with reactions since they are often a highly atopic group who are allergic to other foods and they have been taught an aggressive approach to epipen usage (2, 3).
All children previously seen in our non-university clinical allergy practice who tested skin test positive to peanut, whether they had had a clinical reaction or not and who returned over one year later for retesting in 1999 were characterized prospectively. The passage of time meant that most parents would need to decide about renewal of the epipen prescription and would have thought about their response to reactions either directly by the child or parent or by a caregiver.
Skin Tests: Skin testing to peanut was always performed in duplicate with two different lots of peanut allergen (4,5). The skin test with the largest diameter was used in further calculations. Skin tests were performed by the prick method and were measured in two directions at the longest diameter and then at a right angle to that. A skin test was said to be positive if it measured at least 3 mm in one direction. The size of the peanut positive presented in the tables was an average of the two diagonal measurements. Patients were routinely tested to milk, egg, wheat, soy, peanut, cat, dog, horse, tree, grass, ragweed, alternaria, cladosporium, aspergillus, D. farinae and D. pteronyssinus. At visit 2, they were tested to a series of nuts including almond , Brazil nut, cashew, coconut, hazelnut, pecan, pistachio and walnut ( Bencard Ltd, Toronto, Canada) . They were also tested to any other food identified by history as a problem. At visit 2 if the peanut test solutions were negative, patients were tested to peanut butter itself and then a RAST test was done, (CAP-RAST, Pharmacia Upjohn Ltd., Montreal Canada). If the RAST was also negative, an open oral challenge was undertaken in the office.
Patient Instructions: All patients with a positive skin test to peanut were instructed to strictly avoid peanut (unless they routinely ate peanut without problem) and carry an epipen. Epipen teaching was provided with a trainer. All parents and patients were instructed to go to an emergency department at the first sign of a reaction and the epipen should be administered immediately if: 1) the patient had asthma; 2) the reaction occurred away from home; 3) if the reaction progressed beyond hives; 4) if there were any cough, wheeze or choking.
In 1999, 96 patients who tested positive to peanut returned for re-assessment. Of these, 66 patients had a history of clinical reaction to peanut while the rest were found to be skin test positive without ever knowingly having had peanut (Table 1)(4). They were usually found during assessment for reaction to another food or eczema. Of the 66 "clinical reactors", 13 or almost 20% reported respiratory symptoms (Table 2).
At visit 2 it was found that 10 patients had lost the positive skin tests (8 of whom had had a previous clinical reaction) were negative on RAST testing and successfully underwent oral challenge without reacting. These patients have been described elsewhere (6).
Of the 86 patients who retained the positive skin test to peanut, 26 went on to have a total of 41 reactions to various foods (Table 3). Of the 41 reactions 7 were to peanut and 6 involved respiratory symptoms of cough or wheeze or gasping. All patients were taken to an emergency setting but only one received adrenaline prior to be taken to the emergency department. It is noteworthy that of the 6 patients with respiratory symptoms during the reaction, 3 were patients who had not previously had a reaction to peanut but were from the 30 patients detected prior to being given peanut. Of the 34 non-peanut reactions (which occurred with diverse foods including nuts, fish, shellfish, milk, egg, kiwi, spice, rice, mushroom, cantaloupe, turkey, sesame, celery, candy), 15 reactions were associated with respiratory symptoms but the epipen was administered on only 5 occasions overall. One mother, in the panic surrounding the reaction, held the auto-injector up side down and injected her own thumb.
In this study of 96 children sensitized to peanut, 26 had 41 reactions to different foods after having been identified as peanut sensitive and taught the use of an epipen by demonstration, given information on food sensitivity and the use of an epipen along with a prescription for epipen or epipen junior. Of the further reactions, 21 were associated with respiratory symptoms (51.2%). In only 6 instances (28.6%) was the epipen administered. All children were transported to an emergency department and all survived.
Our practice has been to train all families with children who are skin test positive to peanut in the use of adrenaline injection with an auto-injector (epipen or epipen junior, Allerex Laboratory Ltd; Kanata Ontario) using a blank training device. Since the publication of a series of cases of death and near death with peanut sensitivity (2) and subsequent guideline recommendations (7-9), we have taught that patients with asthma reacting to peanut should be given adrenaline even in the absence of respiratory symptoms associated with the reaction. All reacting children should be taken to the nearest emergency department. These recommendations are perhaps slightly more aggressive in the use of the epipen than recommended elsewhere (10, 11) but the guidelines are still in flux. We have recommended that even children who have a positive skin test to peanut detected before they were knowingly given peanut should be regarded as peanut sensitive. Sampson in an editorial (7) has suggested that such children should strictly avoid peanut and carry an epipen although it is not clear that they would necessarily react clinically. That has been the standard of practice in our office and the present data indicated that of 7 children having a second reaction to peanut after visit1, 6 had respiratory symptoms and 3 were from the group identified as skin-test positive without ever having knowingly ingested peanut.
At the initial visit only 13 of 60 patients with clinical symptoms on ingestion of peanut had respiratory symptoms (20%) while at follow-up 6 of 7 patients reporting a reaction to peanut had respiratory symptoms (85.7%). The data collected in the present study indicated that training and provision of epipens do not guarantee that the adrenaline will be administered in children where it would be appropriate i.e. those children reacting with respiratory symptoms. These results are similar to a number of published studies demonstrating that only a small fraction of patients reacting to a food who have been trained in epipen usage actually use the device when the time comes (12 ). In the study by Gold and Sainesbury (12) only 29% of patients who were taught the use of an epipen after they had suffered a previous anaphylactic episode actually used the device. When it was used, it seemed to reduce morbidity and hospital admissions.
Evidence in the literature suggests that a poor outcome from anaphylaxis is associated with late administration of adrenaline (2, 13 ). As a result current guidelines suggest that adrenaline is the first-line treatment of choice in the management of anaphylaxis and it should be administered as soon as possible (8,9,14,15).
In the present study we did not formally ascertain why patients failed to use the epipen but other studies have demonstrated that children often fail to have the device available or it is outdated or they do not have the skills to administer the epipen properly. While the epipen is the simplest auto-injector device for patients to use, studies have shown that at the time of prescription of the epipen, the majority of patients were not taught how to administer the device and its use was not demonstrated (16, 17). This is not surprising since it has been found that health-care professionals often do not know how to use the epipen appropriately (18) and are therefore not in a position to teach its use. Moreover even an outdated epipen can be life-saving (19). Since many of the threatening reactions occur when the child is away from home, it is necessary to educate the school system on the management of anaphylaxis and the use of the epipen (9,20,21).
Huang (16 ) listed the following recommendations for epipen instruction: (1) give proper instructions to the patients or their parents (this should include the use of a demonstrator of Epipen); (2) give parents or patients the opportunity to practice with an Epipen trainer on each follow-up visit to their physician's office; (3) tell parents and patients that even the application of Epipen may not always rescue the patient from a severe anaphylaxis and that they should go to an emerhency department. He also suggested that the Epipen distributor should provide better teaching and demonstration materials, such as videos and illustrated pamphlets for physicians and pharmacists.
Clearly at every visit to the doctor's office for whatever reason it is appropriate to ask about epipen use and provide refresher teaching. After having done that, if there is still failure to use the epipen when appropriate, some other strategy must be devised to improve compliance. None of the patients in the present study failed to survive. That might be attributable to the fact that they went to emergency departments but at the same time that this study was underway, we saw two patients who were not part of the study and who had asthma that was under poor control at the time of their reactions. They also failed to use an epipen, which was not prescribed in one case and was used very late in the second case. They survived but barely and both had to be intubated and admitted to an intensive care unit.
| Number 96 | Age at first visit (months) 28.5 | Number reacting to peanut (%) 66(68.75) | Size of skin test to peanut (mm) 9.4 | Total number of Positive Skin tests excluding peanut 2.2 |
| Number reacting 66 | erythema (%) 19 (28.8) | hives (%) 37 (56.1) | swelling (%) 24 36.4 | vomiting (%) 11 (16.7) | respiratory (%) 13 (19.7) |
| Time elapsed to second visit (months) 42.1 | Total number of Positive Skin tests 6.1 |
Number positive to nuts (%) 56(58.3) | Number with Asthma (%) 65(67.7) | Number with Eczema (%) 47 (49.0) | Number having a second reaction to any food 26 |