Positive penicillin allergy testing results: a systematic review and meta-analysis of papers published from 2010 through 2015

Abstract

ABSTRACTβ-lactam antibiotics are the most widely used group of antibiotics, given their effectiveness for the most common bacterial pathogens and their relatively low price. Adverse reactions, mainly cutaneous, are often reported to be associated with their use and hence, less effective and usually more costly alternative antibiotics are prescribed. However, it is not clear what is the risk of immediate immune-mediated (i.e. developing within one hour of administration) and potentially life-threatening reactions among those using β-lactam antibiotic. We conducted a systematic review to assess the prevalence of immediate adverse reactions to β-lactam antibiotics, specifically penicillin derivatives, in patients with a reported adverse reaction to β-lactam antibiotics. In addition, we determined the effect of age on the prevalence of immediate reactions. Assessing the true risk of using β-lactam antibiotics in patients with a reported allergy could prevent physicians from unnecessarily discouraging the use of β-lactam antibiotics. We conducted a systematic review and a meta-analysis using the PubMed, OVID, and Embase databases of work published in English and in French in the last 5 years. Studies were only eligible if they established the prevalence of immediate penicillin reactions with skin testing or challenges in case of negative skin tests. The meta-analysis was conducted using Stata version 12.0. The prevalence of immediate reactions to penicillin derivatives in patients reporting a β-lactam hypersensitivity is 1.98% (95%CI; 1.35%, 2.60%) in the pediatric (under 18 years old) group, 7.78% (95%CI; 6.53%, 9.04%) in the adult group, and 2.84% (95%CI; 1.77%, 3.91%) in the combined group, as tested in various studies, using skin tests and oral challenges. The I2 value ranged between 87.2% and 97.0%. Our results indicate that the prevalence of immediate reactions is higher in adults than in children. However, wide confidence intervals and a large study heterogeneity preclude conclusive estimates.