Objectives: The primary objective was to evaluate the ability of different anatomic cut-off points, as established in specialist urogynecology populations, to identify clinically relevant prolapse in a population of postmenopausal women with pelvic floor symptoms recruited from primary care.
Study design: Cross-sectional study among 890 women (>= 55 years) screened for pelvic floor symptoms.
Main outcome measures: The Pelvic Floor Distress Inventory 20 was used to measure symptoms, and the Pelvic Organ Prolapse Quantification (POP-Q) system was used to assess prolapse. Areas under the curves, sensitivity, and specificity were calculated for the hymen as a cut-off point for symptomatic prolapse of the anterior and posterior vaginal wall. For the apical compartment, a cut-off point of -5 cm relative to the hymen was used.
Results: Vaginal bulging was the only symptom reported more often with increasing POP-Q stages. Areas under the curves (95% confidence intervals) to discriminate between women with and without vaginal bulging symptoms were 0.66 (0.61-0.72), 0.56 (0.50-0.63), and 0.61 (0.55-0.66) for the anterior (Ba), posterior (Bp) and apical (C) compartment, respectively. When the hymen was used as the cut-off point, Ba had a sensitivity of 38.1% and a specificity of 82.4%, and Bp had a sensitivity of 13.3% and a specificity of 96.5%. For C, the cut-off point of -5 cm relative to the hymen had a sensitivity of 37.9% and a specificity of 73.1%.
Conclusions: The anatomic cut-off points for clinically relevant prolapse established in the specialist urogynecology population cannot adequately identify symptomatic prolapse in a population of postmenopausal women with pelvic floor symptoms recruited from primary care. (C) 2017 Elsevier B.V. All rights reserved.
- Pelvic floor symptoms
- Pelvic organ prolapse
- Pelvic Organ Prolapse Quantification
- ROC curve
- Sensitivity and specificity
- Spectrum bias
- SUPPORT DEFECTS
- FLOOR DISORDERS
- SPECTRUM BIAS