Reproducibility of a method to evaluate the pelvic floor in normal pregnancy and puerperium using 3D ultrasound and 3D-Power Doppler angiography to assess endopelvic vascularization

Introduction and hypothesis: We tested inter/intra-observer reproducibility of ten different three-D ultrasound (3D-US) pelvic floor variables and endopelvic vascularization using 3D-power Doppler angiography (3D-PDA) as a reliable method to assess pelvic floor status during pregnancy and puerperium. We hypothesized that the method would show good reproducibility. Materials and methods: A prospective study was conducted in two university hospitals, with 162 nulliparous pregnant women evaluated at weeks 12, 28 and 36, and at 48 hours postpartum and three months post-delivery. At each visit, we performed transperineal 3D-US and 3D-PDA of the pelvic floor. Images corresponding to 30 cases were electronically sent between the 2 observers to test the reproducibility of the results. Results: Correlation coefficients > 0.70 for all 10 variables were obtained. Intra-observer reproducibility for each observer was very good, with intra-class correlation coefficients > 0.86. Inter-observer reproducibility of urethral sphincter volume and vascularization, 3D-PDA, VI, FI and VFI measured by VOCAL technique with automatic threshold mode showed good correlation (ICC > 0.80), considered sufficiently high to be clinically applicable. Conclusions: The present study showed good reproducibility and high inter- and intra-observer correlation coefficients for all the variables used to assess the pelvic floor during normal pregnancy and puerperium, including vaginal anterior wall anchors (AWA) as new biometric measures. This makes them reliable parameters for diagnosis and monitoring of pelvic floor disorders in normal pregnancies. We found that use of the VOCAL technique with automatic threshold mode for measurements was superior to the manual mode due to its accuracy and shorter time requirement.