Risk factors for postpartum urinary retention: a systematic review and meta‐analysis

Please cite this paper as: Mulder F, Schoffelmeer M, Hakvoort R, Limpens J, Mol B, van der Post J, Roovers J. Risk factors for postpartum urinary retention: a systematic review and meta‐analysis. BJOG 2012;119:1440–1446. Background Postpartum urinary retention (PUR) is a common condition with varying prevalence. Measurement of the post‐void residual volume (PVRV) is not regularly performed. Various studies have been published on overt (the inability to void after giving birth, requiring catheterisation) and covert (an increased PVRV after spontaneous micturition) PUR. To evaluate which clinical prognostic factors are related to PUR, the identification of independent risk factors for covert and overt PUR is needed. Objectives We performed a systematic review and meta‐analysis of observational studies reporting on risk factors for PUR. Search strategy Systematic search of MEDLINE and EMBASE to September 2011. Selection criteria Articles that reported on women diagnosed with PUR or with an abnormal PVRV. Data collection and analysis The included articles were selected by two authors. We constructed two‐by‐two tables for potential risk factors of overt and covert PUR and calculated pooled odds ratios (ORs) with 95% confidence intervals. Main results Twenty‐three observational studies with original data were eligible for data extraction, of which 13 could be used for meta‐analysis. Statistically significant risk factors for overt PUR were epidural analgesia (OR 7.7), instrumental delivery (OR 4.5), episiotomy (OR 4.8) and primiparity (OR 2.4). For covert PUR, variety in the definitions used resulted in heterogeneity; no significant prognostic factors were found. Conclusions Instrumental delivery, epidural analgesia, episiotomy and nulliparity are statistically significantly associated with a higher incidence of overt PUR. The same factors were identified for covert PUR, but without statistical significance. Uniformity in definitions in future research is essential to create a prognostic model.