Background: The prevalence and intensity of persistent post-surgical pain (PPSP) after breast cancer surgery are uncertain. We conducted a systematic review and meta-analysis to further elucidate this issue. Methods: We searched MEDLINE, Embase, CINAHL, and PsycINFO, from inception to November 2018, for observational studies reporting persistent pain (3 months) after breast cancer surgery. We used random-effects meta-analysis and the Grading of Recommendations, Assessment, Development and Evaluations approach to rate quality of evidence. Results: We included 187 observational studies with 297 612 breast cancer patients. The prevalence of PPSP ranged from 2% to 78%, median 37% (inter-quartile range: 22e48%); the pooled prevalence was 35% (95% confidence interval [CI]: 32e39%). The pooled pain intensity was 3.9 cm on a 10 cm visual analogue scale (95% CI: 3.6e4.2 cm). Moderate-quality evidence supported the subgroup effects of PPSP prevalence for localized pain vs any pain (29% vs 44%), moderate or greater vs any pain (26% vs 44%), clinician-assessed vs patient-reported pain (23% vs 36%), and whether patients underwent sentinel lymph node biopsy vs axillary lymph node dissection (26% vs 43%). The adjusted analysis found that the prevalence of patient-reported PPSP (any severity/location) was 46% (95% CI: 36e56%), and the prevalence of patientreported moderate-to-severe PPSP at any location was 27% (95% CI: 10e43%).