Loop electrosurgical excision procedures (LEEPs) are negative for high-grade cervical intraepithelial neoplasia (CIN 2+) after a hematoxylin and eosin–based CIN 2+ colposcopic biopsy diagnosis in 14% to 24% of cases. This may be due to diagnostic errors or biopsy-related regression of the dysplasia. Because p16 immunohistochemical staining of cervical biopsies improves diagnostic accuracy, we hypothesized that p16-based cervical biopsy diagnoses may reduce the frequency of negative LEEPs.Materials and Methods
We performed a retrospective cross-sectional study of all cervical LEEPs completed at our institution from 2002 to 2012. We recorded patient age, sexual history, smoking history, pathologic diagnoses (including whether the diagnosis was p16 based), the number of days from biopsy to follow-up LEEP, and clinical follow-up. This yielded 593 study subjects meeting inclusion criteria of CIN 2+ colposcopic diagnoses with follow-up LEEP and 2 years of clinical follow-up. Colposcopic biopsies and follow-up LEEPs were reviewed and p16 immunostaining was performed on all samples to provide criterion standard results. Data were analyzed by χ2 and regression modeling.Results
Our practice employed p16 to aid cervical biopsy diagnoses by 2006. The frequency of negative LEEPs before 2006 was 12 (10%) of 126. The frequency dropped during the p16 era (2006–2012) to 23 (5%) of 467. Overall, we observed an inverse relationship between the frequency of p16 employment and the frequency of negative LEEP outcomes (R2 = 0.71; p < .001), independent of potential covariates.Conclusions
Our data suggest that more accurate p16-based diagnoses may reduce the frequency of unnecessary LEEPs.