Association of labor epidural analgesia with early postpartum pelvic floor function in primiparous women undergoing vaginal delivery: a retrospective cohort study.

BACKGROUND: The association between labor epidural analgesia (LEA) and postpartum pelvic floor function remains uncertain. We evaluated whether LEA was associated with pelvic floor outcomes in primiparous women undergoing vaginal delivery. METHODS: This single-center retrospective cohort study included 200 primiparous women with singleton, cephalic vaginal deliveries. Participants were classified according to LEA exposure. Baseline and peripartum characteristics were compared in the overall cohort. Early postpartum pelvic floor outcomes at 6-8 weeks were analyzed in 191 women, and multivariable regression was performed in 175 complete cases. Outcomes included type I and type II pelvic floor surface electromyography, bladder neck mobility, posterior bladder angle during Valsalva, levator hiatus area during Valsalva, urinary incontinence, and pelvic organ prolapse. The multivariable models additionally adjusted for second-stage labor duration and oxytocin use, which were considered potential labor-process variables in the interpretation of the adjusted estimates. RESULTS: No statistically significant differences were identified in the measured baseline characteristics between the LEA and non-LEA groups. Compared with non-LEA, LEA was associated with longer first-, second-, and third-stage labor, longer total labor duration, and more frequent oxytocin use. No significant detectable between-group differences were observed in any pelvic floor parameter, urinary incontinence, pelvic organ prolapse, prolapse stage, or levator avulsion. After adjustment for maternal age, pre-pregnancy body mass index, gestational age, neonatal birth weight, second-stage labor duration, and oxytocin use, LEA was not significantly associated with any continuous pelvic floor outcome, urinary incontinence (adjusted odds ratio, 1.305; 95% confidence interval, 0.629-2.710), or pelvic organ prolapse (1.771; 0.717-4.373). CONCLUSIONS: LEA was associated with altered labor characteristics, whereas no statistically detectable association with worse pelvic floor outcomes at 6-8 weeks postpartum was observed after adjustment for the measured covariates. Given the observational design, potential residual confounding, and limited precision for infrequent outcomes, modest associations cannot be excluded.