Predictive factors for residual disease in subsequent hysterectomy following conization for CINIII

Objective. The aim of this study was to determine predictive factors for po st-cone residual disease in subsequent hysterectomy for CIN III. Methods. From June 1994 to June 1999, 120 patients with CIN III who receive d hysterectomy within 6 months of conization regardless of marginal status were identified from 1450 conization cases. The demographic features and pa thologic parameters were analyzed for the predictive rate of post-cone resi dual disease. Results. Age greater than or equal to 50 years and parity greater than or e qual to5 were significant factors associated with residual disease. The inc idence of residual disease was 56.5 and 29.3% in patients greater than or e qual to 50 and <50 years, respectively, and 61.8 and 36.0% in patients with parity <greater than or equal to>5 and <5. Post-cone endocervical curettag e (ECC) and multiple-quadrant disease were the only pathologic predictive f actors identified. The incidence of residual disease was 64.6 and 29.2% in patients with positive ECC and negative ECC, respectively, and 48.4 and 25. 9% in patient with multiple-quadrant disease and one- or two-quadrant disea se. Other pathologic parameters, including endocervical margins, ectocervic al margins, endocervical gland involvement, and depth of conization, were n ot predictive of residual disease. When ECC was combined individually with age, endocervical margins, or multiple-quadrant disease, there was no incre ase of positive predictive rate; Conclusions. (1) Age 50 years or more and parity <greater than or equal to> 5 were two demographic features that predicted post-cone residual disease. (2) ECC and multiple-quadrant disease were the only pathologic parameters t hat predicted post-cone residual disease. (3) With the appropriate applicat ion of the predictive factors, post-cone hysterectomy may be further decrea sed. (C) 2000 Academic Press.