Novel MRI approach to quantify perineal membrane and perineal body impairments

Objectives: An enlarged genital hiatus after vaginal delivery confers a 3-fold increased odds of prolapse at 1 year and a persistently enlarged genital hiatus after prolapse surgery carries a 6-fold increase odds of surgical failure. Hiatal closure is maintained through complex interactions between the levator ani (LA) muscles and the connective tissues of the perineal membrane (PM) and perineal body (PB). LA injury is present in half of women with prolapse but explains 12% hiatus size variation. Little is known about how PM/PB impairment contribute to the prolapse. Objective: Test the feasibility of quantifying PM/PB structural parameters at maximum Valsalva on MRI in women with and without prolapse and assess their failure frequency in women with prolapse.

Methods: This was a secondary analysis of previously acquired 3D stress MRI images in 13 women with prolapse and five controls with normal support. MRIs from women with prolapse and without major LA defects were selected. The PM was identified on the coronal plane (Fig1) at the distal location of its bony attachment. Two PM/PB parameters were measured: 1) PM ‘swing angle’ was defined as the angle between the PM plane and a line through the left and right bony attachments. 2) PB separation was quantified as the transverse distance between distal PM/PB connection points. LA muscle length defined as the distance from muscle insertion in the pubic bone to its distal vaginal attachment to evaluate “LA extension”. Urogenital AP diameter and area were made on a tipped plane through the pubic symphysis and PB (Fig 1). The difference (Δ) between rest and maximal Valsalva was calculated. Control measurements were used to estimate normal distribution. Z-scores related to each measurement were calculated and the proportion of subjects in each group with parameters outside of the normal range was used to define “failure frequency.”

Results: : On average, cases had a 15° larger swing angle (45°±16° vs 30°±4°, p=.06, effect size=1.0) and 64% larger PB separation (46±12 mm vs 28±10 mm, p=.01, effect size=1.2) at Valsalva. The structural failure frequency in the prolapse group was 77% and 60% for PM swing angle and PB separation. Despite all subjects having intact LA, 30% of prolapse women showed over-extension of the LA muscle length. There was also a significant patient-specific variance in number and combination of perineal complex impairment sites (Fig 2). Cases all had abnormal UGH measures at Valsalva, yet only 69% had abnormal resting UGH.

Conclusions: This pilot study demonstrated the feasibility of identifying patient-specific perineal complex structural impairments that lead to hiatal closure failure. PM /PB impairment play an important role in enlarged GH and further mechanistic studies are warranted.