Positive margins after cervical conization as an indicator of residual dysplasia

Abstract

Objective: To determine the significance of positive margins of resection after cervical conization as an indicator of residual dysplasia. Materials and Methods: A retrospective analysis for patients who underwent cervical conization either by loop electrosurgical excision procedure or cold knife conization between 1986 and 1997 at Montefiore Medical Center and North Central Bronx Hospital. The factors evaluated included grade of dysplasia with respect to positive or negative margins and post-procedure follow-up. Differences among groups were evaluated using the χ 2 and Fisher’s Exact test. Results: Of a total of 179 patients, 99 had positive margins of resection and 80 had negative margins of resection. Twenty-six patients had CIN I, of which 10 had positive margins of resection and 16 had negative margins of resection. There were 30 patients with CIN II, of which 13 had positive margins and 17 had negative margins. One hundred fourteen patients had CIN III/carcinoma in situ (CIS), of which 68 had positive margins of resection while 46 had negative margins of resection. Nine patients had microinvasive disease (MIC), of which 8 had positive margins of resection and 1 had negative margin of resection. The correlation between higher grades of dysplasia and the likelihood of having positive margins was noted to be statistically significant ( P = .02). Patients were followed up from a period of 6 weeks to 5 years. Of the patients with CIN I and positive margins, 5 had a normal post-cone Papanicolaou smear while 2 had an abnormal post-cone Papanicolaou smear. Seven of 9 patients with CIN II and positive margins had normal initial post-procedure Papanicolaou smear while only 2 had abnormal initial post follow-up Papanicolaou smear. Twenty-five patients with CIN III/CIS and positive margins had normal Papanicolaou smears at their initial post-procedure follow-up while 36 patients had an abnormal initial follow-up. Six of 8 patients with MIC and positive margins had documented follow-up. Of these 6, 2 had normal post-procedure Papanicolaou smears while 4 had abnormal post-procedure Papanicolaou smears. This approaches statistical significance. Additionally, the incidence of residual disease was analyzed in hysterectomy specimens with respect to grade of dysplasia. No patients with CIN I and positive margins were treated with hysterectomy. Of the 3 patients with CIN II treated with hysterectomy, 2 had residual dysplasia. Of the 21 patients with CIN II/CIS who underwent hysterectomy, 10 had residual disease. Of the 5 patients with MIC who underwent hysterectomy, 3 had residual disease. Conclusion: The likelihood of positive margins of resection increases with higher dysplasia. The incidence of abnormal initial post-procedure Papanicolaou smear appears to be increased with increasing grade of dysplasia. Interestingly, there does not appear to be an increase in the incidence of residual dysplasia when hysterectomy is performed for positive margin of resection after conization for high grade dysplasia.