Transvaginal ultrasonography and endometrial cytology as a diagnostic schema for endometrial cancer

To determine whether the combined use of transvaginal ultrasonography and endometrial cytology is an effective diagnostic schema for endometrial cancer and hyperplasia of the uterus.

Five hundred fifty-two women were enrolled in this study. For all subjects, endometrial thickness was evaluated by transvaginal ultrasonography. Endometrial cytology was carried out according to the following criteria: all women with atypical uterine bleeding, for asymptomatic postmenopausal women with endometrial thickness >or=5 mm, and for asymptomatic premenopausal women with endometrial thickness >or=20 mm. When the cytological findings were abnormal, we performed a hysteroscopy-guided biopsy within 2 weeks. Women who received transvaginal ultrasonography alone, or those who showed negative cytology, underwent repeated gynecological examination and transvaginal ultrasonography 3, 6, and 12 months after the first examination.

Endometrial cytology was done on 129 women (23.4% of all subjects), of whom 14 were diagnosed as 'positive' cytology, 20 as 'suspicious positive', and 95 as 'negative'. A total of 34 women with 'positive' or 'suspicious positive' cytological result underwent hysteroscopy-guided endometrial biopsy. The histological diagnosis of the endometrium included 13 endometrial cancers, 9 endometrial hyperplasias (one atypical hyperplasia and 3 hyperplastic polyps), and 10 normal endometria. As a diagnostic schema for endometrial cancer, this combined method resulted in 100% sensitivity, 99.1% specificity, 92.9% positive predictive value, and 100% negative predictive value. For endometrial hyperplasia, the method resulted in 100% sensitivity, 89.6% specificity, 40.0% positive predictive value, and 100% negative predictive value.

A combination of transvaginal ultrasonography and endometrial cytology may be an effective diagnostic schema for endometrial cancer and hyperplasia.