|
|
|
| Comparison of efficacy of transvaginal ultrasound and hysteroscopy for diagnosing perimenopausal abnormal uterine bleeding |
| Wenling Dongfang Hospital, Zhejiang Province, 317525 |
|
|
|
|
Abstract To evaluate the values of transvaginal ultrasound and hysteroscopy for diagnosing perimenopausal abnormal uterine bleeding. Methods: A retrospective survey was conducted in 175 patients with perimenopausal abnormal uterine bleeding from January 2018 to January 2020. These patients were all given examinations by transvaginal ultrasound and hysteroscopy, and all these patients had the pathological diagnosis results. Paired design and McNemar test were used to analyze the difference between the examinations by transvaginal ultrasound and hysteroscopy and the pathological diagnosis results. Kappa test was performed to analyze the detection consistency of pathogenesis. The diagnostic efficacies of transvaginal ultrasound and hysteroscopy were calculated. Results: There was significant difference in diagnosing the abnormal perimenopausal uterine bleeding between vaginal ultrasonography and pathological diagnosis (P＜0.05), which Kappa value was 0.409, with poor diagnostic consistency. There was no significant difference in diagnosing the abnormal perimenopausal uterine bleeding between hysteroscopy diagnosis and pathological diagnosis (P＜0.05), which Kappa value was 0.844, with certain diagnostic poor consistency. The Kappa value of vaginal ultrasound combined with hysteroscopy for diagnosing the abnormal perimenopausal uterine bleeding was 0.931, and which efficiency had increased significantly. The sensitivity, the positive predictive value, the specificity, and the negative predictive value of transvaginal ultrasound for detecting abnormal perimenopausal uterine bleeding were 84.6%, 93.3%, 65.4%, and 42.5%, respectively. The sensitivity and the positive predictive value of hysteroscopic for detecting abnormal perimenopausal uterine bleeding were over 98%, and the specificity and the negative predictive value of which were over 85%. The sensitivity, the specificity, the positive predictive value, and the negative predictive value of transvaginal ultrasound combined with hysteroscopic for detecting abnormal perimenopausal uterine bleeding were 98.7%, 92.3%, 99.3%, and 96.0%, respectively. In the coincidence rate of etiology detection, the efficacy in diagnosing endometrial malignancy and dysplasia, anovulatory uterine blood, local abnormalities of endometrium, and other factors of hysteroscopy was significantly higher than those of transvaginal ultrasound (P＜0.05). When the endometrial thickness <4mm, there was no significant difference in the detection rate of etiology of the endometrial lesions of the patients between hysteroscopy and transvaginal ultrasound. When the endometrial thickness >4mm, the advantages of hysteroscopy in detecting local endometrial abnormalities and physiological changes of endometrium were significantly better than those of transvaginal ultrasound. Conclusion: Transvaginal ultrasound for diagnosing perimenopausal abnormal uterine bleeding has some value, especially for detecting the uterine shape and size. However, compared with those of the gold standard, the examination by transvaginal ultrasound still has differences in the poor consistency, the low specificity, and the negative predictive value. The value of hysteroscopy for diagnosing perimenopausal abnormal uterine bleeding is relatively high, and is consistent with the pathological diagnosis, especially for endometrial malignancy and atypical hyperplasia, the diagnosis coincidence rate of which is higher than that of transvaginal ultrasound, and the detection rates of local abnormalities and physiological changes of which are higher than those of transvaginal ultrasound when endometrial thickness >4mm. The accuracy of the combined hysteroscopy and transvaginal ultrasound is higher, which suggests that the examination of transvaginal ultrasound or hysteroscopy can be added according to the condition of perimenopausal abnormal uterine bleeding and other factors of the patient.
|
|
|
|
|
|
|
|
|
