Risk stratification of invasive cervical cancer diagnosed after cervical conization

Background Cervical intraepithelial neoplasia (CIN) diagnosis is based on colposcopy-aided histological examination. However, its accuracy in CIN diagnosis is limited. Some invasive cervical cancers (ICCs) are diagnosed after cervical conization. Therefore, risk stratification of undetected ICC is particularly important for the management of patients with CIN. This study aimed to identify the risk factors for undetected ICC. We especially focused on the association of human papillomavirus (HPV) genotypes. Methods We retrospectively reviewed the clinicopathological characteristics (including age, parity, and preoperative diagnosis) and HPV genotypes of 348 patients diagnosed with CIN or adenocarcinoma in situ (AIS) who underwent cervical conization at our hospital between 2008 and 2016. The relationship between preoperative factors, including HPV genotypes and post-conization ICC, was evaluated. Results Among the 348 patients, 322, 7, and 19 had preoperative CIN3, CIN2, and AIS, respectively; 181 were nulliparous. The median patient age was 41 (23–83) years. HPV genotyping was performed on 237 patients. Overall, post-conization ICC was detected in 16 patients (4.6%). Multivariate analysis showed that nulliparity and HPV16 positivity were the independent risk factors for post-conization ICC (OR: 6.01, P = 0.0302; OR: 5.26, P = 0.0347, respectively). The combination of HPV16 status and parity improved diagnostic accuracy. Seven of 53 HPV16-positive cases (13%) without childbirth history were diagnosed with post-conization ICC. In contrast, none of the HPV16-negative cases with childbirth history was diagnosed with post-conization ICC. Conclusion HPV16 positivity and nulliparity were identified as risk factors for undetected ICC. Careful treatment selection and preoperative scrupulous examination are especially important in these cases.