Diagnostic Accuracy of Sonography for Detection of a Fistula on the Birth Day in Neonates With an Imperforate Anus: Comparison of Diagnostic Performance Between Suprapubic and Perineal Approaches.

OBJECTIVES

To evaluate the diagnostic accuracy of sonography for detection of an internal fistula on the birth day in neonates with an imperforate anus and to compare the diagnostic performance between the suprapubic and perineal approaches.

METHODS

We included 46 neonates with an imperforate anus (29 low type and 17 intermediate/high type) who underwent sonography by both the suprapubic and perineal approaches on the birth day. Thirty-nine neonates had internal fistulas, and 12 did not, as surgically proven. Two blinded radiologists evaluated the suprapubic and perineal sonograms for the presence of the internal fistula in consensus. A final diagnosis of the internal fistula was determined on the basis of the findings of both approaches. A receiver operating characteristic analysis was used to compare the diagnostic performance for detection of an internal fistula between the suprapubic and perineal approaches.

RESULTS

The sensitivity, specificity, and accuracy of the final diagnosis based on the findings of suprapubic, perineal, and both approaches were 52.9%, 79.4%, and 79.4%; 75.5%, 75.5%, and 75.5%; and 58.7%, 78.3%, and 78.3%, respectively. The diagnostic performance of the perineal approach was significantly better than that of the suprapubic approach (P < .0001).

CONCLUSIONS

The diagnostic accuracy of sonography for detection of an internal fistula on the birth day exceeded 75% in neonates with an imperforate anus, and sonography on the birth day is feasible. The perineal approach had superior diagnostic performance over the suprapubic approach. Thus, when evaluating an internal fistula by sonography, we recommend using the perineal approach in addition to the suprapubic approach.