HPV genotype distribution and the risk of high-grade cervical lesions in a hospital-based gynecology population in Henan, China: a retrospective cross-sectional study

Background China’s organised cervical screening programme covers women aged 35 to 64 years, yet much human papillomavirus (HPV) testing still takes place opportunistically when women attend hospital, including women outside that age range. The genotype- and age-specific epidemiology of this setting is poorly characterised. Methods We retrospectively analysed 9,941 women aged 14 years or older who underwent genotyping of 23 HPV types, with or without liquid-based cytology, at a tertiary centre in Henan during 2025, classified as opportunistic screening (n = 7,116) or abnormal-result re-examination (n = 2,825). Among 850 women with histology, we estimated odds ratios (OR) for cervical intraepithelial neoplasia grade 2 or worse (CIN2+), and estimated the whole-cohort absolute risk per genotype-positive woman as a range between a naive and a verification-bias-adjusted estimate. Results Overall HPV prevalence was 23.2% (95% confidence interval [CI] 22.3 to 24.0) and high-risk HPV prevalence 17.8% (17.1 to 18.6) using the 14-type assay panel, or 16.3% restricted to IARC Group 1 types; prevalence was three times higher on re-examination than in opportunistic screening (44.5% versus 14.7%). The commonest genotypes were HPV52, HPV16 and HPV58. Prevalence was bimodal by age (52.2% at 20 years or younger, 19.1% at 30 to 39 years, 40.2% at 60 years or older). Of the 117 women with CIN2+, 13.8% of 850 women with histology, 54.7% carried HPV16 or HPV18. HPV16/18 was independently associated with CIN2+ (adjusted OR 10.67, 95% CI 4.94 to 23.05, versus women negative for all 14 high-risk types), whereas no statistically significant association was detected for the commonest type, HPV52 (genotype-specific adjusted OR 1.23, 0.72 to 2.13). Absolute CIN2+ risk per genotype-positive woman was 18.9% to 32.9% for HPV16 and 9.1% to 18.7% for HPV58, against 4.3% to 10.2% for HPV52, and this ranking was unchanged across a sensitivity analysis of the verification-bias correction. Conclusion In this hospital-based population the commonest genotype was not the one most strongly associated with prevalent CIN2+, and HPV prevalence did not fall with age. Genotype-specific risk stratification and the upper age limit for screening warrant evaluation in prospective, population-based studies.