Prospective human studies of anaphylaxis and its mechanisms have been limited, with few severe cases or examining only 1 or 2 mediators.

We wanted to define the clinical patterns of anaphylaxis and relationships between mediators and severity.

Data were collected during treatment and before discharge. Serial blood samples were taken for assays of mast cell tryptase, histamine, anaphylatoxins (C3a, C4a, C5a), cytokines (IL-2, IL-6, IL-10), soluble tumor necrosis factor receptor I, and platelet activating factor acetyl hydrolase. Principal component analysis defined mediator patterns, and logistic regression identified risk factors and mediator patterns associated with reaction severity and delayed reactions.

Of 412 reactions in 402 people, 315 met the definition for anaphylaxis by the National Institute of Allergy and Infectious Diseases/Food Allergy and Anaphylaxis Network. Of 97 severe reactions 45 (46%) were hypotensive, 23 (24%) were hypoxemic, and 29 (30%) were mixed. One patient died. Severe reactions were associated with older age, pre-existing lung disease, and drug causation. Delayed deteriorations treated with epinephrine occurred in 29 of 315 anaphylaxis cases (9.2%) and were more common after hypotensive reactions and with pre-existing lung disease. Twenty-two of the 29 delayed deteriorations (76%) occurred within 4 hours of initial epinephrine treatment. Of the remaining 7 cases, 2 were severe and occurred after initially severe reactions, within 10 hours. All mediators were associated with severity, and 1 group (mast cell tryptase, histamine, IL-6, IL-10, and tumor necrosis factor receptor I) was also associated with delayed deteriorations. Low platelet activating factor acetyl hydrolase activity was associated with severe reactions.

The results suggest that multiple inflammatory pathways drive reaction severity and support recommendations for safe observation periods after initial treatment.