Abstract
OBJECTIVE: To identify risk factors for a final diagnosis of cervical cancer in patients initially diagnosed with high-grade cervical intraepithelial neoplasia and to evaluate a combined Cytology-Endocervical Curettage (ECC) index for predicting cancer risk. METHODS: This retrospective study included 11,651 patients at West China Second University Hospital (WCSUH) with biopsy-confirmed high-grade cervical intraepithelial neoplasia over a five-year period. Patients were divided into two groups based on whether they ultimately received a diagnosis of cervical cancer following surgery. Multivariate logistic regression was used to identify independent risk factors, and receiver operating characteristic (ROC) curves were constructed to assess the performance of the Cytology-ECC index. RESULTS: Of the 11,651 patients, 229 were subsequently diagnosed with cervical cancer. Multivariate analysis identified age (OR = 1.10, 95% CI: 1.06-1.14), abnormal vaginal bleeding (OR = 2.94, 95% CI: 1.75-4.92), human papillomavirus (HPV)16/18 infection (OR = 2.56, 95% CI: 1.62-4.03), single HPV type infection (OR = 1.68, 95% CI: 1.03-2.75), atypical squamous cells, cannot exclude HSIL (ASC-H) (OR = 3.77, 95% CI: 1.68-8.45), high-grade squamous intraepithelial lesion (HSIL) (OR = 4.65, 95% CI: 2.16-10.05), and endocervical glandular involvement (OR = 1.59, 95% CI: 1.01-2.49) as independent risk factors (all P < 0.05). The Cytology-ECC index demonstrated a predictive accuracy with an area under the curve (AUC) of 0.787, outperforming cytology or ECC alone. CONCLUSIONS: Age, abnormal vaginal bleeding, HPV16/18, single HPV type infection, ASC-H, HSIL, and endocervical glandular involvement were independent predictors of cervical cancer in patients with an initial biopsy of high-grade cervical intraepithelial neoplasia. The combined Cytology-ECC index showed enhanced predictive value in clinical decision-making and may aid in earlier identification of patients at elevated risk for undetected invasive disease.