Risk factors associated with anaphylaxis and other allergic-like events following receipt of 2009 monovalent AS03-adjuvanted pandemic influenza vaccine in Quebec, Canada

IntroductionIn Quebec, Canada, receipt of the 2009 AS03-adjuvanted pandemic H1N1 vaccine was associated with increased risk of anaphylaxis and other allergic-like events (ALE), especially among women of childbearing age. In response to this safety signal, a case–control study was conducted to identify potential risk factors.MethodsA total of 435 ALE (50 anaphylaxis) occurring <24 h following pandemic vaccination were compared to 849 age-gender matched controls randomly selected from the provincial Pandemic Influenza Vaccination Registry. More than 60 potential risk factors were evaluated through phone interviews and included demographic information, medical history, medication use or acute respiratory illnesses (ARI) concurrent with vaccination and other risk factors associated with general allergy. Odds ratios (ORs) with 95% confidence intervals were estimated with unconditional logistic regression.ResultsFactors associated with increased risk of anaphylaxis included concurrent ARI (18% cases vs. 4% controls, ORadj 7.67, 95%CI: 3.04–13.37), food allergy (26% cases vs. 4% controls, ORadj 3.84, 95%CI: 1.51–9.74) and vaccination during the first four weeks of the campaign (66% cases vs. 50% controls, ORadj 2.16, 95%CI: 1.10–4.25) whereas alcohol exposure (≥1 drink/week) was associated with reduced risk (29% cases vs. 42% controls, ORadj 0.26, 95%CI: 0.13–0.57). These factors were also significantly associated with any ALE but the strength of association was weaker. Allergy to components found in the vaccine (e.g., egg, thimerosal) was infrequent and did not significantly differ between cases and controls.ConclusionIncreased anaphylaxis and other allergic-like events observed in association with AS03-adjuvanted pandemic H1N1 vaccine remain mostly unexplained despite extensive risk factor review. However, prior to mass vaccination with similar formulations this safety signal warrants further consideration and better understanding. In particular, the predominance among women of childbearing age may be a clue to underlying biological or hormonal influences on adverse immunological responses to vaccine.