Labeling male anorectal malformations: objective evaluation of radiologic imaging before surgery.

Abstract

Prone cross-table lateral x-ray (CTLxR) and colostogram aid surgical planning for anorectal malformations (ARMs) without perineal fistulas. We suggest objective imaging tools to classify ARMs. Three observers prospectively evaluated CTLxR and colostograms of male ARM patients (2012-2022) without perineal fistulas. The level of the rectal pouch was estimated with pubococcygeal (PC) and ischiatic (I) lines. On CTLxR, we described the "pigeon sign", defined as the rectal pouch ending with a beak-like image, suspicious for a rectourinary fistula. ARM was defined as rectobulbar when the rectal pouch was below the I line, rectoprostatic when between PC and I lines, and rectovesical when above the PC line. Concordance was assessed with Fleiss' kappa. Sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) of the "pigeon sign" were calculated. Thirteen patients were included in this study. The interobserver agreement on CTLxR was 69.2% (k = 0.54) on pouch ending, 84.6% (k = 0.69) on the "pigeon sign", and 76.9% (k = 0.69) on diagnosis; concordance between observers and intraoperative diagnosis was 66.6% (k = 0.56). The "pigeon sign" had 75% sensitivity, 100% specificity, 100% PPV, and 50% NPV. The interobserver agreement on colostograms was 84.6% (k = 0.77) on pouch ending and 89.7% (k = 0.86) on diagnosis; concordance between observers and intraoperative diagnosis was 92.3% (k = 0.90). PC and I lines and the "pigeon sign" are useful tools in examining CTLxR and colostograms. Adequate CTLxR interpretation may modify surgical strategy.