RESUMO
AIMS: Postoperative urinary retention (POUR) is a common complication of urogynecological surgery. Our study aimed to identify demographic and perioperative risk factors to construct a prediction model for POUR in urogynecology. METHODS: Our retrospective cohort study reviewed all patients undergoing pelvic reconstructive surgeries at our tertiary care center (Jan 1, 2013-May 1, 2019). Demographic, pre-, intra- and postoperative variables were collected from medical records. The primary outcome, POUR, was defined as (1) early POUR (E-POUR), failing initial trial of void or; (2) late POUR (L-POUR), requiring an indwelling catheter or intermittent catheterization on discharge. Risk factors were identified through univariate and multivariate logistic regression analyses. A clinical prediction model was constructed with the most significant and clinically relevant risk factors. RESULTS: In 501 women, 182 (36.3%) had E-POUR and 61 of these women (12.2% of the entire cohort) had L-POUR. Multivariate logistic regression revealed preoperative postvoid residual (PVR) over 200 ml (odds ratio [OR]: 3.17; p = 0.026), voiding dysfunction symptoms extracted from validated questionnaires (OR: 3.00; p = 0.030), and number of concomitant procedures (OR: 1.30 per procedure; p = 0.021) as significant predictors of E-POUR; preoperative PVR more than 200 ml (OR: 4.07; p = 0.011) and antiincontinence procedure with (OR: 3.34; p = 0.023) and without (OR: 2.64; p = 0.019) concomitant prolapse repair as significant predictors of L-POUR. A prediction model (area under the curve: 0.70) was developed for E-POUR. CONCLUSIONS: Elevated preoperative PVR is the most significant risk factor for POUR. Alongside other risk factors, our prediction model for POUR can be used for patient counseling and surgical planning in urogynecologic surgery.
