Disease Persistence in Patients with Cervical Intraepithelial Neoplasia Undergoing Electrosurgical Conization
Abstract & Commentary
Synopsis: Following loop electrical excision of CIN when the transformation zone is visible at follow-up, the chance of persistent CIN is low.
Source: Costa S, et al. Gynecol Oncol. 2002;85: 119-124.
Costa and colleagues reviewed their 3-year follow-up data on 699 women who underwent loop electroexcision for treatment of CIN. Nineteen percent of the treated women had an initial diagnosis of CIN 1, 17% CIN 2, and 64% CIN 3. The median width of the excised specimen was 19 mm and the median depth was 15 mm. Twenty seven percent of the specimens had positive margins. Over the study period, a total of 38 women (5.4%) had persistent CIN detected. Most of this was found at the first follow-up visit. Indeed, by the time of the third visit, the prevalence of CIN in the treated group was lower than the prevalence of CIN in the general population in the region of the study.
The presence of a positive margin was not associated with a higher rate of persistence in this study. The only 2 factors that predicted persistence were the pretreatment Pap smear (a high-grade Pap was more highly associated with persistent disease than a low-grade Pap), and a visible squamocolumnar junction at the time of the follow-up visit.
Costa et al noted that their results—the presence of cone margin involvement did not predict persistent disease—are different from most published studies. They strongly suspect that the reason for this difference is that they conservatively estimated cone margin involvement. In addition, the size of their specimens was considerably larger than that reported by many other authors.
Comment by Kenneth L. Noller, MD
This paper is interesting for only one reason: Costa et al found that if the squamocolumnar junction (really, the entire transformation zone) is visible at the time of follow-up after loop electroexcision for CIN, the chance of persistent disease is small. This finding is somewhat unique but does have great appeal. It will be interesting to see if it is confirmed by other studies, or by reanalysis by previously published work.
I am not terribly surprised that there was no association with disease persistence and specimen margin status since Costa et al removed such large pieces of tissue. The fact that their median depth of excision was 1.5 cm means that they excised far more tissue than is reported in most other papers. Thus, if Costa et al had a positive margin it would be more likely very nearly to include the full extent of the lesion than previous reports that might have removed only the central portion of the lesion and left far more disease behind.
The observation that, after 2 negative postprocedure visits, the prevalence of CIN was actually lower than the general population is interesting. We have probably been following our post-loop excision patients far too closely (and far too expensively).
Dr. Noller is Professor and Chairman, Department of OB/GYN, Tufts University School of Medicine, Boston, Massachusetts.