Hysteroscopy and cytology in endometrial cancer

To estimate the effect of preoperative diagnostic hysteroscopy on peritoneal cytology in patients with endometrial cancer. A total of 256 charts were reviewed. Two cohorts were established based on diagnosis by hysteroscopy or blind endometrial sampling via either endometrial biopsy or dilatation and curettage (D&C). Malignant or suspicious peritoneal cytology was the primary outcome. Cohorts were compared using logistic regression to correct for potential confounders of stage and grade. A total of 204 cases were diagnosed by endometrial biopsy or D&C, whereas 52 were identified by hysteroscopy. In the endometrial biopsy or D&C arm, 14 of 204 (6.9%) patients had malignant or suspicious cytology compared with 7 of 52 (13.5%) patients in the hysteroscopy arm (P = .15). After logistic regression controlling for stage and grade, the odds ratio for positive cytology after hysteroscopy was 3.88 (95% confidence interval 1.11,13.6; P = .03). Four of the 52 (7.7%) cases diagnosed by hysteroscopy were stage IIIA due to cytology alone compared with 3 of the 204 (1.4%) cases diagnosed by endometrial biopsy or D&C (P = .03). Hysteroscopy appears to be associated with an increased rate of malignant cytology after controlling for confounders of stage and grade. Further, there appears to be an association between hysteroscopy and upstaging patients due to cytology alone. II-2.