Do ultrasound findings of levator ani “avulsion” correlate with anatomical findings: A multicenter cadaveric study

Abstract

Aims: This study aimed to validate the levator “avulsion” injury as seen on ultrasound against anatomical dissection in the same cadaver. Methods: Puboviseral muscle (PVM) anatomy of female cadavers was studied using 3D-translabial ultrasonography and an “avulsion” confirmed per standard recommendations [Dietz HP. Aust N Z J Obstet Gynaecol 53:220–230, 2013]. Cadavers were then dissected to determine the macroscopic attachment or detachment of the PVM and the dimensions including the PVM symphysis gap and PVM attachment depth. Intra and inter-observer reliability of USS findings and anatomical measurements were assessed using the Cohen's κ and Bland & Altman plots respectively. McNemar's and Mann–Whitney U tests were used to compare imaging and cadaveric dissection findings. Results: “Avulsions” were seen on imaging in 11/30 (36.7%) cadavers; the defect was bilateral in 1/30 (3.3%) and unilateral in 10/30 (33.3%). No “avulsion” was found at dissection (McNemar's χ2 = 60.0, P < 0.001). An additional thirty-nine cadavers were dissected with no “avulsion” identified. A narrower PVM insertion depth was strongly associated with “avulsion” on ultrasound (mean: 4.79 mm vs. 6.32 mm, Z = −3.191, P = 0.001). Intra- and inter-observer agreement was perfect (K = 1.0 ± 0.0) and good (K = 0.85 ± 0.142) for anatomical “avulsions” and USS, respectively. Conclusions: There is a clear difference between anatomical and USS findings. The imaged appearance of an “avulsion” does not represent a true anatomical “avulsion” as confirmed on dissection. The term “avulsion” is misrepresentative and should not be used to describe this imaging finding. Moreover, further attempts at surgically repairing this defect should be avoided, at least until there is a better understanding of its pathophysiology. Neurourol. Urodynam 35:683–688, 2016. © 2015 Wiley Periodicals, Inc.