Accuracy of preoperative histology and macroscopic assessment of cervical involvement in endometrial carcinoma

Objective: To compare the diagnostic accuracy between the preoperatively and macroscopically established determination of cervical involvement in endometrial carcinoma. Study design: During the period 1987 to 1991, 154 patients with endometrial cancer were evaluated in a retrospective blind manner with the objective of assessing the diagnostic accuracy of the preoperative (prehysterectomy curettage) and the macroscopic (sectioned surgical specimen) determination of the involvement of the cervix. The results were compared with the histological findings of the hysterectomy specimen (gold standard). Results: Preoperative pre-hysterectomy curettage established a sensitivity of 38% and a specificity of 91% whereas the macroscopic findings (gross appearance) confirmed a sensitivity of 50% and a specificity of 95%. A prevalence of 17% for cervical involvement was found. In patients with cervical involvement diagnosed on pre-hysterectomy curettage, a positive predictive value of 45% and negative predictive value of 88% were established. A positive predictive value of 68% and a negative predictive value of 90% resulted from the judgement of the cervical gross appearance. Conclusions: We conclude that the cervical involvement of endometrial carcinoma diagnosed on pre-hysterectomy tissue is less predictive than the judgement of the intraoperative gross appearance of the cervix. (C) 1997 Elsevier Science Ireland Ltd.