Predictive value of endocervical curettage at cervical conization and provider practices in management of high-grade cervical dysplasia

OBJECTIVE: To determine the utility of endocervical curettage (ECC) at cervical conization for predicting residual or recurrent dysplasia and to evaluate differences in management between general gynecologists and gynecologic oncologists. STUDY DESIGN: From February 1999 to November 2007, 192 patients with high-grade dysplasia on conization were retrospectively identified. Data were analyzed for 54 patients who underwent repeat conization or hysterectomy to evaluate predictors of disease. Data for all patients were analyzed based on provider. RESULTS: Among patients who underwent secondary procedures, 68.5% (37/54) had residual or recurrent disease. Eighty-six percent of patients with a positive ECC had residual or recurrent disease compared to 48% of patients with a negative ECC (OR 6.91, CI 1.595-30.00, p = 0.01). Among all patients, 77% (148/192) were managed by a generalist, and 23% (44/192) by a gynecologic oncologist. Oncologists were significantly more likely to perform a hysterectomy (45.5% vs. 14.2%, OR 5.04, CI 2.38-10.69, p<0.0001). CONCLUSION: Endocervical curettage at the time of conization with high-grade dysplasia is a simple and reliable predictor of residual or recurrent disease and should be performed routinely. Gynecologic oncologists are more likely than general gynecologists to perform a hysterectomy in the management of high-grade dysplasia on conization.