Transvaginal Sonographic Criteria for the Diagnosis of Adenomyosis Based on Histopathologic Correlation

Abstract

To evaluate the accuracy of different transvaginal sonographic criteria in the diagnosis of adenomyosis, and to determine the most useful sonographic feature by comparison with histopathologic results. A total of 213 consecutive patients scheduled for hysterectomy underwent preoperative transvaginal sonography in this retrospective study. The diagnosis of adenomyosis was made if one or more of the following sonographic findings were present: (1) a globular uterine configuration; (2) poor definition of the endometrial-myometrial interface; (3) sub-endometrial echogenic linear striations; (4) myometrial anterior-posterior asymmetry; (5) myometrial cysts; and (6) a heterogeneous myometrial echotexture. These sonographic findings were then compared with the histopathologic findings. The prevalence of adenomyosis was 39.9%. The sensitivity, specificity, positive and negative predictive values, and accuracy of transvaginal ultrasound for the diagnosis of adenomyosis were 87.1, 60.1, 59.2, 87.5 and 70.9%, respectively. We found that subendometrial echogenic linear striations, a heterogeneous myometrial echotexture, and myometrial anterior-posterior asymmetry showed greater accuracy for the diagnosis of adenomyosis. Further evaluation of these findings showed that subendometrial echogenic linear striations had the best sensitivity, and positive and negative predictive values for the diagnosis of uterine adenomyosis (91.8, 67.8 and 92.9%, respectively). The presence of a globular uterine configuration was the most specific sonographic feature (78.1%), but showed poor specificity (50.6%). The presence of subendometrial echogenic linear striations, a heterogeneous myometrial echotexture, and myometrial anterior-posterior asymmetry on transvaginal ultrasonography supports the diagnosis of adenomyosis. Among the transvaginal ultrasonographic findings consistent with the diagnosis of adenomyosis, subendometrial linear striations had the highest diagnostic accuracy.