High-grade CIN on cervical biopsy and predictors of the subsequent cone histology results in women undergoing immediate conization

2015 
Abstract Objective To identify the clinical/colposcopic variables that associate with low-grade/negative cone histology in screening-age women undergoing conization for high-grade cervical intraepithelial neoplasia (CIN). The follow-up outcomes of study participants were also compared. Study design In this retrospective cohort study, 585 consecutive screening-age women who underwent immediate conization for CIN2-3 were divided according to cone histology (CIN2+ versus ≤CIN1) and assessed in relation to clinical/colposcopic variables by univariate and multivariate analyses. Results Low-grade [adjusted odds ratio (AOR) = 52.67, 95% confidence interval (CI) 22.49–123.34] or normal (AOR = 9.81, 95% CI 2.38–40.44) colposcopic impression and CIN2 on cervical biopsy (AOR = 19.59, 95% CI 6.62–57.92) associated with CIN1/negative cone histology. Multivariate analysis also showed that Eastern European ethnicity (AOR = 0.13, 95% CI 0.03–0.52) and high-risk-Human Papillomavirus (hr-HPV)-positivity (AOR = 0.38, 95% CI 0.17–0.87), associated with CIN2+ cone histology. Overall, there were no significant differences between the two groups in terms of high-grade recurrence during the 2-year follow-up. Conversely, a higher rate of high-grade recurrence was present in CIN2-3 (positive cone margins) than in CIN1/negative cone histology (21.9% versus 7.4%, P  = 0.008, respectively). Conclusion The presence of CIN2 on cervical biopsy and a low-grade colposcopic impression were predictive of a minor cone histology, unless the subject was of East European ethnicity or was positive for hr-HPV test. Given the follow-up outcomes, the same women need to perform a close monitoring. However, positive cone margins in women with CIN2-3 cone histology seem to define a population at greater risk of high-grade recurrence.
    • Correction
    • Source
    • Cite
    • Save
    • Machine Reading By IdeaReader
    31
    References
    10
    Citations
    NaN
    KQI
    []