Nonobstetric Risk Factors for Symptomatic Pelvic Organ Prolapse

OBJECTIVE: To identify possible nonobstetric risk factors for symptomatic pelvic organ prolapse in the general female population. METHODS: This was a population-based, cross-sectional study derived from a sample of 5,489 Stockholm women, 30 to 79 years old, who answered a validated questionnaire for the identification of symptomatic prolapse. The 454 women whose answers indicated the presence of such prolapse and the 405 randomly selected control participants with answers that gave no indication of prolapse received a 72-item questionnaire, which probed into a priori suspected risk factors. Only those women with intact uteri and no prior surgery for incontinence or prolapse were included. Multivariable logistic regression models estimated prevalence odds ratios (ORs) with 95% confidence intervals (CIs). RESULTS: In addition to age and parity, overweight (prevalence OR for body mass index [kg/m(2)] 26-30 compared with 19-25 was 1.9, 95% CI 1.2-3.1), history of conditions suggestive of deficient connective tissue (varicose veins/hernia/hemorrhoids, prevalence OR for positive history compared with no history 1.8, 95% Cl 1.2-2.8), family history of prolapse (prevalence OR for positive history compared with no history 3.3, 95% Cl 1.7-6.4), heavy lifting at work (prevalence OR for 10 kg or more compared with no heavy lifting 2.0,95% Cl 1.1-3.6), and presence of constipation, hard stools, or difficult evacuation (prevalence OR relative to normal bowel habits 2.1, 95% Cl 1.4-3.3) all were linked independently, significantly, and positively to the presence of symptomatic prolapse. CONCLUSION: In this nonconsulting population, age and parity were the dominating risk factors, but significant independent associations with markers suggestive of congenital susceptibility (family history and conditions signaling weak connective tissue) and nonobstetric strain on the pelvic floor (overweight/obesity, heavy lifting, and constipation) imply that individual predisposition and lifestyle/environment also may play an important role. The causal direction of the association with bowel habits remains uncertain, and the link to family history could be partly because of information bias.