Evaluation of Perineal Descent Measurements on Pelvic Floor Imaging

Objectives: The aim of this study is to validate a uniform method for measuring perineal descent which can be used for different imaging methods, to establish cut-off values for this measurement, and to assess diagnostic test accuracy (DTA) of imaging techniques using these cut-off values. Secondly, the study aims to correlate perineal descent to symptoms, signs and imaging findings in women with obstructed defaecation syndrome (ODS) to assess its clinical relevance. Methods: Cross-sectional study of 131 women with symptoms of ODS. Symptoms and signs were assessed using validated methods. These women underwent evacuation proctography (EP), magnetic resonance imaging (MRI), transperineal ultrasound (TPUS) and endovaginal ultrasound (EVUS). Perineal descent was measured on EP and MRI as the position of anorectal junction (ARJ) with respect to the pubococcygeal line (PCL) at rest (i.e., static descent) and during evacuation (i.e., descent at Valsalva). Dynamic perineal descent was measured on all four imaging techniques as the difference between the position of the ARJ at rest and Valsalva. DTA of dynamic perineal descent was estimated using Latent Class Analysis in the absence of a reference standard. Results: Interobserver agreement of dynamic perineal descent measurements was good for MRI and EVUS (ICC 0.86 and 0.85) and moderate for EP and TPUS (ICC 0.61 and 0.59). The systematic differences in measurements between imaging techniques show the need for individual cut-off values. New established cut-off values for dynamic descent are for EP 20 mm, MRI 35 mm, TPUS 15 mm and EVUS 15 mm. Sensitivity was moderate for EP (0.78) and MRI (0.74), fair for TPUS (0.65) and poor for EVUS (0.58). Specificity was similar for all imaging techniques (0.73-0.77). Static perineal descent correlated with symptoms of pelvic organ prolapse (POP) (r = 0.19), prolapse of all three compartments (r = 0.19-0.36), presence of levator ani muscle avulsion (p = 0.01) and increased hiatal area (r = 0.51). Dynamic perineal descent correlated with excessive straining (r = 0.24) and use of laxatives (r = 0.24). Classic symptoms of ODS (incomplete evacuation and digitation) did not correlate with perineal descent measurements. Static and dynamic perineal descent were associated with presence of rectocele, enterocele, intussusception, and absence of anismus. Conclusions: Dynamic perineal descent is a reliable measurement that can be applied to different imaging techniques to allow standardisation. Static descent is more often present in women with POP and dynamic descent is more often present in women with constipation. Perineal descent does not correlate with typical symptoms of ODS. Specificity of TPUS and EVUS is comparable to EP and MRI, hence ultrasound could be used for the initial assessment of pelvic floor dysfunction.