An Tang MD, Presenter: Nothing to Disclose

Claude Kauffmann PhD, Abstract Co-Author: Nothing to Disclose

Stephanie Madill PhD, Abstract Co-Author: Nothing to Disclose

Stephanie Pontbriand-Drolet, Abstract Co-Author: Nothing to Disclose

Chantale Dumoulin PhD, Abstract Co-Author: Nothing to Disclose

1. To evaluate the morphology and movement of the levator ani (LA) muscle with a shape-deformable high-resolution model during dynamic MRI. 2. To compare 2D pelvimetry measurements with 3D measurements of the LA.

IRB approval was obtained for this prospective study in 31 women: 12 continent asymptomatic volunteers and 19 with mixed urinary incontinence (mean age, 68.0 ± 4.1 years; range 60-76). Pelvic floor MRI were performed at 3.0 T. Imaging sequences included thin-slice T2w FSE axial, coronal and sagittal sequences at rest, followed by T2w HASTE axial and sagittal sequences during pelvic floor muscle maximal voluntary contraction and Valsalva maneuver. Pelvimetry measurement of the M-line was performed in the mid-sagittal plane. LA muscle segmentation included: 1) manual segmentation of muscle at rest, contraction and Valsalva, 2) deformation of a high-resolution 3D LA model previously created from an asymptomatic woman to fit each segmentation result, 3) 3D reconstruction of the LA geometry and 4) calculation of LA surface area and encircled volume. Segmentation time was recorded. Statistical analysis consisted of independent sample t-tests to compare measurements between continent and incontinent subjects.

The M-line was significantly longer in the incontinent group (4.0±7.11mm) than in the continent group (-0.2±3.7mm) during contraction (p<0.05), but not Valsalva. The LA muscle surface and volume differences, as calculated by segmentation, between incontinent and continent women were non significant during maximal contraction and Valsalva. Average segmentation time per case was 5 min 55 s ± 68 s.

A longer M-line suggests greater pelvic floor laxity in incontinent women. From our preliminary experience, pelvic floor segmentation using a shape-deformable model is feasible, but not predictive of the incontinence status.

Despite advances in 3D levator ani segmentation technique, measurement of the M-Line remains a convenient predictor of pelvic floor laxity.

Tang, A, Kauffmann, C, Madill, S, Pontbriand-Drolet, S, Dumoulin, C, Levator Ani Muscle Movement in Older Continent and Urinary Incontinent Women: Comparison of 2D and 3D Measurements.  Radiological Society of North America 2010 Scientific Assembly and Annual Meeting, November 28 - December 3, 2010 ,Chicago IL. http://archive.rsna.org/2010/9013470.html