MR imaging of the anal sphincter in multiparous women using an endoanal coil: correlation with in vitro anatomy and appearances in fecal incontinence.

Objective: The aim of this study was to correlate the components of the normal female anal sphincter seen on high-resolution MR images with the in vitro anatomy and to describe the change in appearances of these components in multiparous women with fecal incontinence.

Subjects And Methods: Ten asymptomatic female volunteers (32-72 years old; mean, 54 years old) and 22 women with fecal incontinence were studied. In six patients (26-68 years old; mean, 49 years old) fecal incontinence began immediately after childbirth; in the remaining 16 patients (45-77 years old; mean, 58 years old) fecal incontinence developed 15- 30 years after childbirth. In the latter group of patients, terminal motor latencies of the pudendal nerve were measured. Imaging was done on a 0.5-T Picker Asset unit and on a 1.0-T Picker HPQ unit. A saddle geometry endoanal receiver coil was used for all imaging. T1-weighted spin-echo (720-820/20 [range of TR/TE]), T2-weighted spin-echo (2500/80 [TR/TE]), fast spin-echo (4500/96 [TR/ effective TE]), and short inversion time inversion recovery (2500/80 [TR/TE]; inversion time, 107 msec) MR images were obtained in transverse, coronal oblique, and sagittal planes. Images were assessed for integrity of the sphincter components. A nonpaired separate-variance t test was used to compare thickness of individual muscle components between patients with delayed-onset fecal incontinence and asymptomatic age-matched volunteers. Degree of muscle atrophy was correlated with degree of delay in the terminal motor latency of the pudendal nerve.

Results: The high resolution obtained with an endoanal coil allowed differentiation of the various muscle components of the anal sphincter complex. The internal sphincter was seen as a ring of homogeneously high signal intensity with a low-signal-intensity rim that was rich in collagen and contained neurovascular bundles. The external anal sphincter, which had low signal intensity on T1- and T2-weighted images, was shown as three components: subcutaneous, superficial, and deep. In six patients who had fecal incontinence that began immediately after childbirth, endoanal MR imaging revealed the site and extent of a tear. All tears were confirmed at surgery. In the 16 patients who had fecal incontinence that began several years after childbirth, atrophy of the external sphincter was revealed in all cases in the superficial and deep components. The internal sphincter remained normal. However, we found that the degree of atrophy of individual components of the external sphincter did not correlate with the degree of delay in pudendal nerve conduction.

Conclusion: MR imaging with an endoanal coil reveals the integrity and bulk of individual muscle components of the anal sphincter in multiparous women with fecal incontinence.