MRI versus 3-dimensional ultrasound: a comparative study of levator hiatal dimensions in women with pelvic organ prolapse [corrected] [published erratum appears in AUST NZ CONTINENCE J 2006;12(4):85].

Magnetic resonance imaging (MRI) has been used to identify both normal and abnormal female pelvic anatomy, permitting resolution and tissue definition not previously possible and faster acquisition times adding to our understanding of dynamic pelvic anatomy. These systems are, however, expensive and limited to tertiary centres. The alternative, three-dimensional ultrasound (3D USS), has undergone major advances in development, so that, today, dynamic axial plane imaging has become feasible.This study aimed to compare the two imaging modalities in the assessment of the levator hiatus. Eleven women undergoing prolapse surgery were assessed using MRI and 3D USS. Data volumes were acquired at rest and on maximum Valsalva, and were stored and reviewed independently at a later date to measure the dimensions of the levator hiatus. Similar measurements were undertaken in the equivalent axial slice on MRI using fast acquisition T2 weighted scanning. Independent, clinical pelvic examination was performed using the ICS POP-Q system. The mean age of participants was 58 years, median parity was 3, with 46% (5) having had a hysterectomy. A total of 82% (9) had a sensation of a lump, 55% (6) had constipation. Clinical grading on POP-Q was 2 (median). All had rectoceles at surgery, with 46% (5) having coexistent cystoceles, 18% (2) level 1 prolapse, and 82% (9) having enteroceles. The intra-class correlations were relatively poor-0.22 for levator area at rest and 0.47 on Valsalva. Possible explanations for the poor correlations observed were the use of different points of reference and the difficulties of obtaining equivalent Valsalva manoeuvres. In particular, it was virtually impossible to consistently image the plane of minimal dimensions on Valsalva by single-slice MR imaging. Further larger studies comparing the two imaging modalities are required before final conclusions can be drawn. [ABSTRACT FROM AUTHOR]