Feasibility and reproducibility of new technique for measurement of transverse diameter of levator ani muscle hiatus using two‐dimensional transperineal ultrasound in nulliparous women at term

ABSTRACT Objectives The aim of the present study was twofold; firstly to develop and describe the technique for measurement of the transverse diameter of the levator ani muscle hiatus (TD) in the coronal view using 2D transperineal ultrasound (TPUS) in nulliparous women at term of pregnancy. Secondly, we aimed to assess the feasibility and reproducibility of 2D TPUS assessment of TD, in addition to the inter‐method agreement between 2D TPUS and the 3D measurement of TD on the axial plane, the latter considered as the gold standard in nulliparous women at term of pregnancy. Methods For the aim of the study, we recruited a group of nulliparous women at term of pregnancy before the onset of labor. The study was conducted in two phases: phase 1 involved developing and describing the 2D TPUS technique for measuring TD, while phase 2 focused on assessing the technique's feasibility, reproducibility, and inter‐method agreement. In phase 1, we enrolled 30 women. Each woman underwent the acquisition of a 3D TPUS volume, which was analyzed using the multiplanar mode to identify the appearance of the lateral borders of the levator ani muscle at the level of the plane of minimal hiatal dimensions in the coronal plane. These borders were used as landmarks for TD measurement. Additionally, we measured the distance between the line indicating TD and the center of the urethra in the axial view. Phase 2 involved recruiting 100 women. Each woman underwent the acquisition of three 2D TPUS clips in the coronal plane, each encompassing a sweep of the entire levator hiatus, and a 3D volume, all obtained during rest. On the 2D clips, TD was measured twice by one operator and once by another operator. TD was measured once in the 3D volume in the axial plane, considered the gold standard. Each operator was blinded to all other measurements during their assessments. We analyzed intraobserver, interobserver, and intermethod (2D vs. 3D) reproducibility. Bland‐Altman analysis was conducted, and Levene's W0 and Student t‐tests were performed to explore clinical factors that might contribute to systematic differences. Results In phase 1, we successfully identified the landmarks denoting the lateral borders of TD in the coronal view. These appeared as two symmetrical hypoechoic indentations located at the inner border of the hyperechoic structure of the levator ani muscle, at the point of maximum distance between the two sides of the levator ani muscle. Additionally, the distance between the urethra and the plane where TD was measured using 3D TPUS in the axial plane had a median of 4 mm and varied from 0 to 9 mm. This enabled us to describe the method for assessing TD in the coronal plane through the use of 2D TPUS. In phase 2, TD was successfully measured in all 2D and 3D acquisitions from the entire group of 100 women. The analyses for intraobserver, interobserver reproducibility, and intermethod comparison (2D vs. 3D) revealed almost perfect agreement in TD measurements using 2D TPUS, with Intraclass Correlation Coefficients (ICCs) of 0.95 (95% CI, 0.92 to 0.96), 0.87 (95% CI, 0.78 to 0.92), and 0.85 (95% CI, 0.78 to 0.90), respectively. The average differences between measurements were 0.1 mm for intraobserver, 1.0 mm for interobserver, and 0.2 mm for intermethod repeatability. No systematic differences were observed in any of the measurement sets, except in the inter‐operator analysis, although the difference was clinically insignificant (38.2 vs. 37.1 mm, P= 0.01). None of the examined clinical factors (maternal body mass index and maternal age) exhibited a statistically significant impact on intraobserver, interobserver, or intermethod reliability. Conclusions Utilizing our described technique to measure the transverse diameter of the levator hiatus in the coronal view using 2D TPUS is not only feasible but also highly reproducible and accurate in nulliparous women at term of pregnancy. Moreover, it yields measurements that are comparable to those obtained in the reconstructed axial plane generated by 3D TPUS. This article is protected by copyright. All rights reserved.