Postoperative MRI evaluation of anorectal malformations with clinical correlation.

Sixteen postoperative patients with anorectal malformation were evaluated by MRI, and the results compared with the clinical assessment. Patients were classified into three groups--good (group 1, n = 10), fair (group 2, n = 3) and poor (group 3, n = 3)--on the basis of Kelly's clinical score of incontinence. The degree of development of the puborectalis and external sphincter muscles and the levator hammock was evaluated on MRI in comparison with patients without anorectal disease. The proportions of fair or poor development of the muscles were 37% in group 1,22% in group 2 and 67% in group 3. Although fair or poor development of the muscles was seen more frequently in group 3, there was no statistically significant difference between groups. However, poorly developed muscles were seen only in patients with fair or poor clinical scores. The difference in the anorectal angle measured on sagittal MRI images between patients in group 1 and groups 2 or 3 was significant. Our study indicates that MRI evaluation based solely on muscle development can be misleading, and measurement of the anorectal angle should be included in the MRI evaluation.