Hysterosalpingographic accuracy of peritubal adhesion.

The purpose of this study was to determine diagnostic accuracy of hysterosalpingography by using different diagnostic criteria in peritubal adhesion diagnosis. The authors retrospectively reviewed cases in which both hysterosalpingography and laparoscopy were performed. Fifty-nine of 84 cases had laparoscopy proved peritubal adhesion. Five hysterosalpingographic signs (convoluted tube, vertical tube, ampullary dilatation, peritubal halo and loculation of the spillage of contrast material) defined by Karasick and Goldfarb were used to diagnose peritubal adhesion. All cases were analyzed by two different diagnostic criteria: first diagnostic criterion, presence of one or more signs means abnormal; second diagnostic criterion, presence of two or more signs means abnormal. Peritubal adhesion was diagnosed in 70 of 84 cases by using the first diagnostic criterion, 53 of 84 cases by using the second diagnostic criterion. The first diagnostic criterion displayed 94.9 per cent sensitivity, 44 per cent specificity, 80 per cent positive predictive value, 79.76 per cent accuracy and the likelihood ratio of 1.69. The second diagnostic criterion showed 74.6 per cent sensitivity, 64 per cent specificity, 83 per cent positive predictive value, 71.43 per cent accuracy and the likelihood ratio of 2.07. The authors conclude that using the 2nd diagnostic criterion is more appropriate than using the 1st diagnostic criterion in diagnosing peritubal adhesion.