Observer variation in the radiological measurement of the anorectal angle

Abstract

Determination of the anorectal angle (ARA) and the position of the pelvic floor is, theoretically, very important in understanding the mechanisms of anorectal continence and defaecation. The variability in the measurement of the ARA was analyzed. Nine experts drew the rectal axis either as a line along the posterior wall of the distal rectum or as the central axis of the rectal lumen on the outlines of 18 representative proctographic images. The standard deviations and ranges of the mean values of each ARA were comparable but large in both methods. On average, the S.D. was 8° and the range value about 23°. Inter-observer variation was not related to the magnitude of the ARA, but rather to the anorectal configuration. Drawing a line along the posterior distal rectal wall is difficult when it is irregular or when the puborectalis impression is indistinct. The central rectal axis is difficult to draw when the junction between the upper and lower rectum is ill defined or when the outlines of the distal rectum are asymmetric e.g by the presence of a rectocele. Thus, the variability of both methods was not strongly interrelated (r=0.68 for the median values). It is concluded that, in general, radiologic assessment of the ARA is not reliable enough for comparative investigation of the dynamics of the anorectum.