Prediction model for residual high-grade cervical intraepithelial lesions

Area of Science:

• Gynecology
• Oncology
• Pathology

Background:

• High-grade cervical intraepithelial lesions (HSIL) require precise management to prevent progression.
• Positive surgical margins after loop electrosurgical excision procedures (LEEP) necessitate further evaluation for residual disease.

Purpose of the Study:

• To identify risk factors for residual HSIL in patients undergoing a second cervical excision.
• To develop a predictive model for residual HSIL after initial LEEP with positive margins.

Main Methods:

• Retrospective analysis of 290 patients with HSIL positive margins post-LEEP (March 2015-August 2019).
• Multivariate logistic regression and stepwise analysis of 14 variables, including demographics, clinical data, pathology, and HPV genotypes.
• Residual disease defined as HSIL in the second cervical excision specimen.

Main Results:

• Of 290 patients, 124 (42.8%) had residual HSIL (≥CIN2) in the second excision.
• A prediction model incorporating gravida, endocervical canal involvement (CIN2-3) in the first LEEP, and high-risk HPV (excluding 16/18) showed an AUC of 0.605.
• The model's predictive capability for residual disease was limited.

Conclusions:

• A statistically significant, yet poorly discriminative, prediction model for residual HSIL was developed.
• Identified risk factors include gravida, endocervical involvement, and specific HR-HPV types.
• Further studies are needed to refine prediction models for guiding clinical decisions in managing cervical intraepithelial neoplasia.