Vacuum-Assisted Breast Biopsy and Excision for Breast Lesions: A Systematic Review and Reproducible Meta-analysis of 26 Reports

Vacuum-assisted breast biopsy (VABB) and vacuum-assisted excision (VAE) serve different diagnostic and therapeutic purposes. We reconstructed a report-level and outcome-level dataset from 26 published reports (references 8-33), using 18 reports now reviewed from full-text PDF or full-text HTML and 8 reports that remain abstract-only or only partly accessible through indexed publisher text after targeted retrieval. Each value was classified as directly reported, uniquely reconstructed from published counts or rounded percentages, or unavailable; overlapping reports were mapped before pooling. Random-effects meta-analyses of proportions used a logit scale, Paule-Mandel between-study variance, modified Hartung-Knapp 95% confidence intervals (CIs), exact study CIs, and 95% prediction intervals (PIs). Six studies (137/1,105 lesions) yielded a pooled malignant-upgrade proportion after diagnostic VAB/VABB of 12.07% (95% CI, 8.55%-16.79%; PI, 5.83%-23.36%; I²=56.9%). Four studies (45/625 lesions) yielded a pooled malignant-upgrade proportion detected at second-line VAE of 9.03% (95% CI, 2.70%-26.22%; PI, 0.36%-73.28%; I²=85.5%). Across six heterogeneous follow-up cohorts, the exploratory lesion-free/complete-excision proportion was 96.33% (95% CI, 88.86%-98.86%; PI, 51.67%-99.85%; I²=94.5%). Three clinically heterogeneous diagnostic cohorts produced exploratory pooled sensitivity of 89.15% (95% CI, 67.49%-97.02%) and specificity of 99.02% (95% CI, 74.70%-99.97%). In one retrospective microcalcification cohort, the false-negative risk was lower with VAB than CNB (risk ratio, 0.19; 95% CI, 0.07-0.55), but outcome and population differences precluded a universal VAB-versus-CNB pooled effect. These results support lesion- and pathway-specific interpretation rather than claims of general superiority or equivalence.