Accuracy of endometrial sampling compared to conventional dilatation and curettage in women with abnormal uterine bleeding

Autor: Amr H. Yehia, Mohamed Al-Kadi, Khaled M. Abdelrazak, Ibrahim A. Abdelazim, Assem A. M. Elbiaa
Rok vydání: 2014
Předmět:
Zdroj: Archives of Gynecology and Obstetrics. 291:1121-1126
ISSN: 1432-0711
0932-0067
DOI: 10.1007/s00404-014-3523-y
Popis: To compare the diagnostic accuracy of brush endometrial sampling with conventional dilatation and curettage in women with abnormal uterine bleeding.Two hundred and twenty (220) women with abnormal uterine bleeding were included in this comparative study; endometrial sampling was done before cervical dilatation using Tao Bruch followed by conventional dilatation and curettage (DC). The histopathology report of the Tao Bruch samples was compared with that of the DC samples and the DC results were considered as the gold standard.100% of samples obtained by conventional DC, while 98.2% of the samples obtained by Tao Brush were adequate for histopathology examination. In this study; Tao Brush had 100% sensitivity, 100% specificity, 100% predictive values and accuracy for diagnosing endometrial hyperplasia, endometrial carcinoma, proliferative and secretory endometrium, also, it had 86.7% sensitivity, 100% specificity, 100% positive predictive value (PPV) and 99% negative predictive value (NPV) and accuracy for diagnosing endometritis (no significant difference compared to conventional DC), while, it had 77.8% sensitivity, 100% specificity, 100% PPV and 99% NPV and accuracy for diagnosing endometrial polyps (no significant difference compared to conventional DC) CONCLUSION: Endometrial sampling using endometrial brush cytology (EBC) is safe, accurate, cost-effective outpatient procedure, avoids general anesthesia with high sensitivity and specificity for detection of endometrial hyperplasia and endometrial carcinoma. EBC could be used as complementary diagnostic tool when hysteroscopic biopsies or other blinded procedures for endometrial sampling are unwanted or not available.
Databáze: OpenAIRE