Abstract
OBJECTIVE: To identify the risk factors for persistent HR-HPV infection in patients with cervical intraepithelial neoplasia (CIN). METHODS: A total of 312 patients with cervical intraepithelial neoplasia were followed up for six months. Among them, 164 patients with persistent HPV infection during re-examination were categorized into the persistent infection group, while 148 patients with negative HPV results were classified into the negative conversion group. RESULTS: Multivariate logistic regression analyses identified the following independent risk factors for persistent HR-HPV infection: age ≥50 years (95% CI: 3.037-11.447; P<0.001), multiple HPV infections (95% CI: 4.250-18.417; P<0.001), HPV viral load ≥100 (95% CI: 1.529-5.673; P=0.001), reproductive tract inflammation (95% CI: 1.186-4.696; P=0.014), and thyroid dysfunction (95% CI: 8.346-17.207; P<0.001). A prediction model was developed based on the logistic regression analysis: Logit(P) = -102.56 + (age × 1.774) + (HPV multiple infections × 2.180) + (HPV viral load ≥100 × 1.080) + (reproductive tract inflammation × 0.859) + (thyroid dysfunction × 3.650). Receiver operating characteristic (ROC) curve analysis showed an area under the curve (AUC) of 0.800 for the model in predicting persistent high-risk HPV infection, with sensitivity of 81.00% and specificity of 79.46%. CONCLUSION: Age ≥50 years, multiple HPV infections, HPV viral load ≥100, reproductive tract inflammation, and thyroid dysfunction are independent risk factors for persistent high-risk HPV infection in patients with CIN.