Impact of Episiotomy on Postpartum Pain, Pelvic Floor Function, and Sexual Health: A Propensity Score-Matched Cohort Study and Prediction Model Development
摘要
To evaluate the association between episiotomy and postpartum pain, perineal healing, pelvic floor function, and sexual function, and to develop a prediction model for identifying women at high risk of moderate-to-severe perineal pain following vaginal delivery.
MethodsThis retrospective cohort study included 638 women who underwent vaginal delivery between January 2023 and December 2024. Propensity score matching (PSM; 1:1, caliper 0.02) balanced baseline characteristics between episiotomy (n = 268) and non-episiotomy groups (n = 268). Visual Analog Scale (VAS) pain scores were assessed at six time points up to 3 months postpartum. Pelvic floor dysfunction, perineal healing, and Female Sexual Function Index (FSFI) were evaluated at 3 months. A logistic regression prediction model was developed and internally validated.
ResultsAfter PSM, 516 patients completed 3-month follow-up (96.3% retention). The episiotomy group demonstrated significantly higher VAS pain scores at all time points (P < 0.001), with effect sizes decreasing from 1.2 at 24 h to 0.2 at 6 weeks. At 3 months, the episiotomy group showed poorer perineal healing (P < 0.001), higher stress urinary incontinence (32.8% vs 22.2%, P = 0.028, OR 1.58), and lower FSFI scores (24.2 vs 27.6, P < 0.001), with sexual dysfunction prevalence of 75.7% vs 55.0%. A prediction model for moderate-to-severe perineal pain (VAS ≥ 4) demonstrated good discrimination (training AUC 0.819, validation AUC 0.804) and calibration (Hosmer–Lemeshow P = 0.445). Key predictors included primiparity (OR 2.32), instrumental delivery (OR 2.1), and prolonged second stage (OR 1.9). Risk stratification identified three groups with observed pain incidence of 28.9%, 55.8%, and 72.0%, respectively.
ConclusionsEpisiotomy is associated with increased postpartum pain, impaired perineal healing, higher stress urinary incontinence risk, and reduced sexual function at 3 months. The prediction model shows promise for early postpartum risk stratification. External validation is needed before clinical implementation.