Performance characteristics of basophil activations tests for diagnosing penicillin allergy: a meta-analysis

Approximately 10% of the global population identify themselves as penicillin allergic, yet 90% are not truly allergic and could safely tolerate penicillin. There is no simple way to identify these people. Current in vitro diagnostics include specific immunoglobulin E (with a sensitivity of 19% and specificity of 97%) and basophil activation testing (BAT) with undefined sensitivity and specificity. PubMed and EMBASE searched from inception to 03/05/2022 for original studies evaluating the performance characteristics of basophil activation test for penicillin allergy in adults. Study selection, data extraction, risk of bias (RoB) assessment with QUADAS-2 tool and certainty assessment with GRADE methodology were performed independently, in duplicate. Meta-analysis was performed using Reitsma methodology. Twenty-two studies fulfilled the inclusion criteria. Twelve used the same positive threshold giving a summary point sensitivity 51% (95% CI, 46%-56%) and specificity 89% (95% CI, 85%-93%). Significant RoB identified, due to patient selection. GRADE certainty of evidence rated sensitivity “very low” due to imprecision and specificity as “low”. Use of 1000 basophils per test did not improve performance above 500 basophils. BAT has improved sensitivity but lower specificity over other in vitro diagnostics in penicillin allergy. BAT methods need to be standardised to limit heterogeneity in results.