Using Z-scores to evaluate levator hiatal dimensions with four-dimensional translabial ultrasound.

AIM: This study represents the first attempt to use 'Z-scores' to assess levator hiatal dimensions and quantify hiatal expansion.

METHODS: We undertook a retrospective study on the data of 110 nulliparae and 73 pelvic organ prolapse (POP) patients who had a clinical POP exam and translabial 4-D ultrasound. We used Z-scores to normalize the dimensions of nulliparae levator hiatus and to build a Z-score model to evaluate the 'hiatal ballooning' against the POP staging.

RESULTS: Valid data were gathered from 102 nulliparae and 64 POP patients. We built the Z-score models as follows: Z-Av = (measured value - 17.19) / 2.98, Z-Lv = (measured value - 55.65) / 5.48, where 'Av' and 'Lv' represented the 'hiatal area' and the 'anterior-posterior diameter' on Valsalva, respectively. The '90% reference range' of Z-Av was (-1.8-+1.8) and that of Z-Lv was (-2.0-+2.0). On receiver-operator curve analysis, the cut-off against POP stage 2 for Av was 20 cm2 (sensitivity of 79%, specificity of 66%), and that for Lv was 6.0 cm (sensitivity of 68%, specificity of 65%), which were approximately equal to a Z-Av of 1.0 and a Z-Lv of 1.0, correspondingly. For POP stage 3, the cut-off was an Av of 24 cm2 (sensitivity of 85%, specificity of 83%) or an anterior-posterior diameter of 6.3 cm (sensitivity of 77%, specificity of 80%), which was approximately equal to a Z-Av of 2.0 or a Z-Lv of 1.5, respectively.

CONCLUSION: 'Normal hiatal dimension' was defined as Z-Av ≤ 1.0 or Z-Lv ≤ 1.0, and 'hiatal ballooning' was defined as Z-Av ≥ 2.0 or Z-Lv ≥ 1.5.