The levator ani muscle forms a sling that borders the levator hiatus, the largest potential hernial opening in the female body. A larger levator hiatus area measured with three- and four-dimensional transperineal ultrasound in women with symptomatic pelvic organ prolapse (POP) has been found. An increase in the levator hiatus area following the first vaginal delivery has been described by several research groups, but less is known about how subsequent vaginal deliveries impact its size. The few existing ultrasound studies assessing the levator anatomy after subsequent vaginal delivery claim an insignificant role of subsequent vaginal delivery on the levator hiatus area (1). On the other hand, large epidemiological studies found a strong association between subsequent vaginal deliveries and POP(2). Given the link between increased levator hiatus area following childbirth and symptomatic POP later in life, longitudinal data on the levator hiatus in normal childbearing populations may give insights into the mechanism of prolapse development.

The study aims to assess the levator hiatus area from mid-pregnancy to 8 years after first delivery in women with and without subsequent vaginal deliveries.

In this single center longitudinal cohort study, 300 nulliparous women were recruited in gestational weeks 17-19 of their first pregnancy and followed up to 8 years after their first delivery. For the present analysis only data from women with vaginal delivery at first delivery were used. Women who had undergone pelvic floor surgery during the follow-up and women with missing data on subsequent delivery were excluded. A three- and four-dimensional transperineal ultrasound examination using the GE Kretz Voluson E8 (GE Medical Systems) with 4-8MHz curved array 3D/4D ultrasound transducer (RAB4-8l/obstetric) was performed at 21 weeks of gestation during first pregnancy and 1 and 8 years after first delivery. Ultrasound volumes were acquired at rest, maximum pelvic floor muscle (PFM) contraction and maximum Valsalva maneuver. The levator hiatus area was measured as the area bordered by the levator ani muscle, the symphysis pubis and the inferior ramus pubis in render mode (3). Two trained investigators performed the examination and analysed ultrasound volumes. The inter-investigator reliability for levator hiatus area measures was excellent (Intra-class correlation >0.96). The investigators were blinded to previous ultrasound findings and women’s demographic and obstetric data.
Exposure measurement for the study was subsequent vaginal delivery. Women with at least one subsequent vaginal delivery after vaginal delivery at first delivery were allocated to the “subsequent vaginal group”. Women with no further deliveries or with subsequent cesarean section after first vaginal delivery were allocated to the “primiparous vaginal group”. 
Outcome measure was the change of the levator hiatus area over time. Linear mixed models were estimated. Post hoc analyses were performed to derive the mean between-group differences and between-group differences in within-group changes. The mean levator hiatus area (with 95% confidence intervals (CIs)) and mean differences (with 95% CIs) from the post hoc analysis are presented.

In this study, 166 participants were included. One hundred and thirty-nine women were allocated to the subsequent vaginal group, 113 (81.3%) had one subsequent vaginal delivery, and 26 (18.7%) had two subsequent vaginal deliveries. Twenty-seven women were allocated to the primiparous vaginal group, out of them 22 (81.5%)women had no subsequent delivery and five (18.5%) of them had one subsequent cesarean section. There was no difference in the following background data between study groups: pre-pregnancy BMI, age, and follow-up time points. The mean levator hiatus area and 95% CI for the two study groups at the respective time points are illustrated in figures (fig.1-3). In the subsequent vaginal group, the levator hiatus area at rest did not change from 1 to 8-year follow-up. In the primiparous vaginal group, the levator hiatus area at rest decreased from 1- to 8-year follow-up (fig. 1). The change in levator hiatus area at rest from 1 to 8 years was statistically significant between study groups (mean difference -1.27 cm2 (CI -2.18; -0.35)). No statistically significant differences between study groups in the levator hiatus area at rest at 1- and 8-year follow-up were found. Levator hiatus area at PFM contraction increased from 1- to 8-year follow-up in women with subsequent vaginal delivery but not in women in the primiparous vaginal group (fig. 2). The difference in change in levator hiatus area at PFM contraction between time points was not statistically significant when comparing study groups. At 8-year follow-up, women with subsequent vaginal delivery had statistically significantly larger levator hiatus area at PFM contraction compared to women without subsequent vaginal delivery (mean difference 0.83 cm2 (CI 0.003; 1.66)). In women with subsequent vaginal delivery, the levator hiatus area at Valsalva decreased slightly from 1- to 8-year follow-up, whereas it increased slightly in the primiparous vaginal group (fig. 3). The difference in the change of levator hiatus area at Valsalva from 1 to 8 years was not statistically significant between the study groups. No statistically significant differences between study groups in the levator hiatus area at Valsalva at 1- and 8-year follow-up were found.

Our longitudinal study reveals new insights in the development of the levator hiatus area when comparing women with and without subsequent vaginal delivery. Findings from previous ultrasound studies on the effect of subsequent vaginal deliveries on levator anatomy stand in contrast to epidemiological studies, where vaginal parity plays an important role in prolapse development. The reasons for those contrasting results might be the cross-sectional design comparing primipara and multipara and the focus on the levator hiatus at the Valsalva maneuver. In our longitudinal study, associations between subsequent vaginal delivery and levator hiatus area at rest and PFM contraction but not at Valsalva were found 8 years after first vaginal delivery.

Subsequent vaginal delivery was associated with levator hiatus area at rest and contraction, but not at Valsalva maneuver 8 years after first vaginal delivery.

Horak TA, Guzman-Rojas RA, Shek KL, Dietz HP. Pelvic floor trauma: does the second baby matter? Ultrasound Obstet Gynecol 2014; 44: 90-94.
Patel DA, Xu X, Thomason AD, Ransom SB, Ivy JS, DeLancey JO. Childbirth and pelvic floor dysfunction: an epidemiologic approach to the assessment of prevention opportunities at delivery. Am J Obstet Gynecol 2006; 195: 23-28
Dietz HP, Wong V, Shek KL. A simplified method for determining hiatal biometry. Aust N Z J Obstet Gynaecol 2011; 51: 540-543