To estimate the accuracy of vaginal cytology in postoperative surveillance for detecting recurrent endometrial cancer and to estimate the optimal management of squamous abnormalities detected in this setting.
This review included women who underwent hysterectomy for endometrial cancer between January 1, 2006, and December 31, 2010, and had at least one postoperative Pap test. Clinical and demographic data were collected and outcomes including abnormal vaginal cytology, results of colposcopic examination, and endometrial cancer recurrence were assessed. A Cox regression model to estimate the risk of abnormal cytology was created. Sensitivity, specificity, and negative and positive predictive values of detecting vaginal recurrences were calculated.
Four hundred thirty-three women contributed 2,378 Pap tests. At least one abnormal cytology result was found during follow-up of 55 (13%) women, representing 3% of all Pap tests. No recurrent endometrial cancers were diagnosed on the basis of isolated abnormal cytology. No cases of recurrent cancer were diagnosed in women with atypical squamous cells of undetermined significance or low-grade squamous intraepithelial lesion (LSIL) Pap test results. In multivariable analysis, abnormal cytology was highly associated with prior postoperative radiation therapy (P<.001). The sensitivity, specificity, and positive and negative predictive values of an abnormal Pap test result in detecting a local recurrence are 40%, 87.9%, 7.3%, and 98.4%, respectively.
Colposcopy is not needed after a Pap test result read as atypical squamous cells of undetermined significance or LSIL.
Vaginal cytology is poorly predictive of recurrent endometrial cancer; colposcopic examination of low-grade squamous lesions is unnecessary and of low yield.
Division of Gynecologic Oncology, Department of Obstetrics and Gynecology, Washington University School of Medicine and Siteman Cancer Center, St. Louis, Missouri.
Corresponding author: Akiva P. Novetsky, MD, MS, Division of Gynecologic Oncology, Department of Obstetrics and Gynecology, Washington University School of Medicine, 4566 Scott Avenue, Campus Box 8064, St Louis, MO 63110; e-mail: firstname.lastname@example.org.
The Siteman Cancer Center is supported by NCI Cancer Center Support Grant P30 CA91842.
The authors thank Gongfu Zhou for his statistical assistance.
Financial Disclosure The authors did not report any potential conflicts of interest.