Risk factors for postoperative recurrence of abdominal wall endometriosis: a meta-analysis and a 24-year clinical cohort study

Abdominal wall endometriosis (AWE) is a common iatrogenic condition following cesarean delivery, yet robust evidence on factors associated with its recurrence remains limited. This study utilizes an integrated evidence synthesis design, combining a primary single-center institutional cohort with a systematic review and meta-analysis. The study employed a dual-axis complementary design. A retrospective cohort of 259 consecutive AWE patients from Peking University People's Hospital (2001–2025, with complete follow-up in 202 patients) was analyzed. A systematic meta-analysis of global literature (January 2020 to March 2026, 7 external studies plus the institutional dataset) was performed. Pooled odds ratios (OR) or mean differences (MD) with 95% confidence intervals (CI) were calculated using fixed- or random-effects models as appropriate. Sensitivity analyses (leave-one-out and model comparison) assessed robustness. The meta-analysis identified three factors associated with recurrence: short latent period (MD = –12.98 months, 95% CI: –20.71 to –5.24, P < 0.01), subcutaneous (Type I) localization (OR = 1.87, 95% CI: 1.16–3.01, P = 0.01), and multifocal lesions (OR = 2.33, 95% CI: 1.45–3.77, P < 0.01). Age, Body mass index (BMI), lesion size, number of previous cesarean deliveries, incision type, and postoperative hormonal medication showed no significant association (all P > 0.05). The institutional recurrence rate was 3.47% (7/202). Short latent period was the most consistently associated factor with recurrence across sensitivity analyses. Type I localization and multifocality also showed associations in the pooled analysis, but these findings were sensitive to exclusion of individual studies (significance lost after omitting Zang et al. and Song et al., respectively). Short latent period was the most consistently associated factor with recurrence. Type I localization and multifocality showed associations in the pooled analysis, but these associations were sensitive to individual study exclusion. Our findings suggest that complete excision with an adequate surgical margin appears favorable, whereas limited excision may be associated with higher recurrence risk.