Risk index for pelvic organ prolapse based on established individual risk factors

• Published in: Archives of gynecology and obstetrics Vol. 293; no. 3; pp. 617 - 624
• Main Authors: Mothes, A. R., Radosa, M. P., Altendorf-Hofmann, A., Runnebaum, I. B.
• Format: Journal Article
• Language: English
• Published: Berlin/Heidelberg Springer Berlin Heidelberg 01.03.2016; Springer Nature B.V
• Subjects: Adult; Aged; Body Mass Index; Case-Control Studies; Endocrinology; Female; General Gynecology; Gynecology; Health risk assessment; Health Status Indicators; Human Genetics; Humans; Logistic Models; Medicine; Medicine & Public Health; Menopause; Middle Aged; Obstetrics; Obstetrics/Perinatology/Midwifery; Pelvic organ prolapse; Pelvic Organ Prolapse - diagnosis; Pelvic Organ Prolapse - epidemiology; Pregnancy; Prevalence; Prognosis; Regression analysis; Retrospective Studies; Risk Factors; ROC Curve; Sensitivity and Specificity; Surveys and Questionnaires; Womens health; Prolapse risk index; Pelvic organ prolapse; Risk factors
• ISSN: 0932-0067; 1432-0711
• DOI: 10.1007/s00404-015-3863-2

Purpose To identify known risk factors for pelvic organ prolapse (POP) in a hospital cohort and to develop a prolapse risk index (PRI). Methods Risk factors for POP were recorded in women who underwent surgery with symptomatic POP ( n  = 500) or non-POP gynaecological conditions ( n  = 236). Descriptive statistics were determined by Chi-squared and Mann–Whitney U tests. Stepwise multivariate regression analysis was performed for all patients and subgroups by age (<60 and ≥60 years). Primary outcome measures were variables with the strongest impact on prolapse and PRI development. Secondary: specificity, sensitivity, positive and negative predictive values (PPV and NPV, respectively), and Cohen’s kappa statistic ( κ ). Results Stepwise multivariate regression analysis ( n  = 736) showed difficult obstetric history [odds ratio (OR) 10.04], family history of POP (OR 7.28), and ≥10 years since menopause (OR 4.53) were independent risk factors for prolapse ( P  < 0.001). When one of the three variables with the strongest influence on POP development was present, the PRI for all women showed a PPV of 82 %, NPV of 68 %, and κ of 0.47 for predicting symptomatic POP requiring treatment. In women under 60 years ( n  = 349), logistic regression revealed difficult obstetric history (OR 9.108), positive family history (OR 8.016), and body mass index (OR 2.274) as independent risk factors. Conclusions Eighty-seven percent of our patient cohort with symptomatic POP requiring therapy could be identified by the PRI, which may be useful for counselling and education.