Aims, Objectives and Background Anaphylaxis is seen relatively frequently in the emergency department, and if not managed promptly and appropriately, is associated with significant morbidity and mortality. Often patients are admitted for extended periods of observation due to fears that they may develop a biphasic reaction. NICE has recently updated its guidelines on this but acknowledges the sparsity of research investigating the appropriate length of the observation period. The purpose of this study was to identify the frequency of biphasic reactions and what period of observation would be sufficient for patients following attendance with anaphylaxis.

Method and Design This was a retrospective observational study which looked at the presentation, treatment and outcomes of patients presenting to Royal Derby Hospital emergency department with anaphylaxis over the last 10 years. The patients were identified from clinical coding completed at the time of discharge and the information was attained from a combination of electronic and paper notes. We defined a biphasic reaction as a recurrence of anaphylaxis without re-exposure to an allergen following a complete resolution of symptoms and normalisation of observations.

Results and Conclusion A total of 377 patients were identified as presenting with anaphylaxis from the discharge coding. After case note review, 346 of these were concluded to have been true anaphylaxis (examples of those excluded were hereditary angioedema and ACE-inhibitor related angioedema). 321 (93%) were given intra-muscular adrenaline at some point and 244 (70%) were admitted to hospital for further treatment or observation. Of the 346 cases included in the study only 5 (1.4%) met the criteria for a biphasic reaction. Interestingly, 1 patient was responsible for 3 of those 5 attendances. The results of our study strongly suggest that the risk of a biphasic reaction is significantly less than previously thought and that admission for prolonged observation may be unnecessary in the majority of cases.