Anaphylaxis in an emergency department: a retrospective 10-year study in a tertiary hospital.

Background. Anaphylaxis is a potentially fatal medical emergency. The frequency of hospital admissions for anaphylaxis seems to be increasing in the recent decades. Objective. Characterize the patients admitted for anaphylaxis to the adult emergency department (ED) of a tertiary care hospital over a 10-year period, discriminating aetiologies, clinical features and therapy administered. Methods. Retrospective, descriptive and inferential study, evaluating age, sex, Manchester triage system, suspected allergen, site of allergen exposure, comorbidities, cofactors, clinical findings and symptoms, treatment and management. Patients admitted between January 2007 and December 2016 were included. Results. Forty-three patients were enrolled: 23 males, mean age 54.3 ± 16.2 years, n = 22 had history of allergic disease. Two patients were triaged as non-urgent. The most frequently suspected causes of anaphylaxis were: drugs (33%, n = 14), Hymenoptera venoms (23%, n = 10), foods (21%, n = 9) and iodinated contrast products (12%, n = 5). Adrenaline was used in 88% of the episodes (n = 38), 55% of which (n = 21) intramuscularly. Mortality was registered in one case. At discharge, adrenaline auto-injector was prescribed in 7% (n = 3) of the patients, and Allergy and Clinical Immunology consultation (ACIC) was requested in 65% of the episodes (n = 28). Statistically significant associations (p minor 0.05) were established: a, anaphylaxis to drugs associated with a low intramuscular adrenaline use and with frequent oxygen therapy; b, anaphylaxis to food associated with intramuscular adrenaline administration; c, anaphylaxis to Hymenoptera venom associated with male sex; and d, anaphylaxis to iodinated contrasts associated with referral to ACIC and with shock. All obese patients developed shock. Conclusions. Anaphylaxis is a life-threatening condition that requires early recognition. Although most patients received adrenaline, administration was not always performed by the recommended route and only a few patients were prescribed adrenaline auto-injector.