Thermal artifacts and inter-gynecologist variation of laser cone biopsies of the cervix.

Abstract

To evaluate the extent of thermal artifacts and the inter-gynecologist variation in CO2 laser cone biopsies for cervical intraepithelial neoplasia (CIN) using high power density. All 1180 histologic sections from 93 CO2 laser cone biopsies performed by nine gynecologists were evaluated by morphometry. Power density of 19,700 W/cm2 and spot size of 0.43 mm were used. The lengths of different epithelia, the distance from CIN to the excision margins, the thermal artifacts, and the extension of denudation were analyzed. Mean values and standard deviations were analyzed for all indices. The inter-gynecologist variation was analyzed using one-way analysis of variance, Tukey-Kramer Honestly Significant Difference test, and chi 2 contingency table analysis regarding cones classified as "good" or "not good". The mean length of analyzed surface per cone was 134 mm. In 3% of the sections, the sum of artifacts at the external excision margin exceeded the average "minimum free distance" from CIN to the margin (2.88 mm). At the endocervical excision margin, the free distance was exceeded in 17% of the sections. Eighteen percent of the cones had CIN at the excision margins, with no significant differences between gynecologists (P = .83). Differences existed between gynecologists regarding free minimum distance and minor thermal artifacts at the external margin, but when clinically relevant groups were applied, no significant differences were found (P > .05). We found that a high power density at CO2 laser cone biopsy resulted in specimens with minimal thermal artifacts compared to other authors who used low power density. No clinically relevant differences were found between the gynecologists regarding free margins or artifacts.