Resensitization of β-Lactams After Negative Initial Standard Evaluation: A Systematic Review and Meta-Analysis

Abstract

Background: β-Lactam (BL) allergy workup varies across studies because of methodological heterogeneity, which affects the estimated risk of BL resensitization after a negative allergy test. Consequently, controversy remains regarding recommendations for retesting. Objective: This systematic review and meta-analysis aimed to quantify the prevalence, severity, and determinants of BL resensitization to support safe and individualized retesting strategies. Methods: PubMed, Embase, Scopus, and CINAHL were searched from inception to August 4, 2024, in accordance with Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA). Eligible studies enrolled patients with documented BL allergy who achieved a negative initial standard evaluation confirming tolerance and subsequently underwent retesting. Random-effects models generated pooled prevalence with 95% confidence intervals (CIs); subgroup analyses examined retest modality, reaction chronology, geography, and age. The strength of evidence was graded with Grading of Recommended Assessment, Development, and Evaluation (GRADE). Results: Thirty-two studies comprising 5766 retests met eligibility criteria. The overall pooled resensitization rate was 3.80% (95% CI, 2.35-5.50; I² = 82.96%). Limiting to studies using the sequential or direct drug provocation test (DPT) across 3414 retesting evaluations, the resensitization rate was 2.44% (95% CI, 0.99-4.43; I² = 86.08%), equivalent to 1 case detected per 41 retests. Severe reactions during retesting with these methods occurred at a rate of 0.32% (95% CI, 0.18-0.58; I² = 0.0%). The overall strength of evidence for resensitization prevalence was graded as low. Conclusions: In DPT-based studies, the pooled resensitization risk was low (approximately 1%-4%) with substantial heterogeneity. Serious reactions during retesting were very rare. These findings do not support routine retesting after a negative evaluation, as the observed risk is in the range of de novo BL reactions in the general population.