Office hysteroscopy, transvaginal ultrasound and endometrial histology: a comparison in infertile patients

Objective. To evaluate accuracy of transvaginal sonography (TVS)and hysteroscopy in detection of intrauterine pathology in infertilewomen. Subjects and methods. This retrospective study was conductedin 56 infertile women with abnormal transvaginal ultrasoundfindings of the uterine cavity which was performed during the midfollicularphase as a part of routine infertility workup. Hysteroscopy wasperformed between 6th and 10th day of cycle. Results. The mean age ofthe subjects was 31.9±4.0. The most frequent ultrasound finding wasendometrial polyp in 34 (60.7%) patients, septate uterus in 8 (14.3%)patients, submucosal myoma in 7 (12.5%) patients, endometrial hyperplasiain 5 (8.9%) patients and Syndroma Ascherman in 2 (3.6%)patients. Hysteroscopy confirmed 20 (35.7%) polyps, the same numberof myomas, septate uterus and Syndroma Ascherman as detectedby ultrasound, (7 (12.5%), 8 (14.3%) and 2 (3.6%), respectively) and19 (33.9%) endometrial hyperplasia. In 46 women with histologicalexcamination, the sensitivity of TVS and hysteroscopy in the diagnosisof endometrial polyps were identical - 100%, while the specificity washigher in hysteroscopy than in TVS (92.3% versus 56.4%, p<0.001).The sensitivity of TVS in the diagnosis of endometrial hyperplasia washigher than that of hysteroscopy (86.4% versus 22.7%, p<0.001), whilespecificity was identical, of 100%. Accordance between hysteroscopyand histology was good (k=0.79), between ultrasound and hystologywas moderete (k=0.59). Conclusion. Hysteroscopy appeared tobe more reliable in diagnosis than TVS. The use of a high frequencyultrasound probe leads us to a lack of diagnostic clarity between endometrialpolyps and hyperplasia.