Background. The purpose of this study was to evaluate the diagnostic value of transvaginal ultrasound measurement of endometrial thickness, cytology obtained by Gynoscannr`, and histology of the endometrium sampled by Uterine Explora Curetter` compared with histology of the uterine specimen as the gold standard.
Methods. Consecutive patients admitted for hysterectomy had transvaginal ultrasound, sampling by Gynoscannr`, and Uterine Explora Curetter` done just before surgery, after informed consent.
Results. A total of 181 women entered the study. Sixteen had endometrial cancer, seven had atypical hyperplasia and nine had complex hyperplasia. A total of 168 patients had a transvaginal ultrasound done. At a cutoff limit of 4mm (endometrial thickness of 4mm or less indicating normal endometrium), the sensitivity was 90.3%, the specificity 24.8%, the positive predictive value 21.4% and the negative predictive value 91.9%. One endometrial cancer, one atypical and one complex hyperplasia were missed. The Gynoscannr` method showed a sensitivity of 62.5%, a specificity of 94.0%, a positive predictive value of 69.0% and a negative predictive value of 92.1%. Two cancers, three atypical and six complex hyperplasia were missed. The Uterine Explora Curetter` showed a sensitivity of 90.6%, a specificity of 100.0%, a positive predictive value of 100.0% and a negative predictive value of 98.0%. One endometrial cancer and two complex hyperplasia were missed.
Conclusion. Transvaginal ultrasound is a reliable method in excluding endometrial pathology. The Uterine Explora Curetter` was superior to Gynoscannr` in diagnosing neoplasia of the endometrium. It was found to have the same diagnostic accuracy as conventional dilatation and curettage.