[Clinical outcomes of people with abnormal cervical cancer screening results but normal colposcopy findings].

Objective: To analyze the clinical outcomes of people with abnormal cervical cancer screening results but normal colposcopic findings, and explore the reasons for missed detection high-grade cervical intraepithelial neoplasia Ⅱ and worse (CINⅡ+). Methods: People who underwent colposcopy at the Department of Obstetrics and Gynecology, Peking University First Hospital from January 1st, 2019 to December 31st, 2019 were selected. Those with normal colposcopic findings were included, some of them underwent biopsy, others were followed up for one year. Data including thinprep cytologic test (TCT) and high-risk human papillomavirus (HR-HPV)screening results, colposcopic image characteristics, and histopathological outcomes were collected for statistical analysis. Based on cytology results, people with normal colposcopic findings were divided into two groups: the low-risk group [negative for intraepithelial lesion and malignancy (NILM), atypical squamous cell of undetermined signification (ASCUS), and low grade squamous intraepithelial lesion (LSIL)] and the high-risk group [high-grade squamous intraepithelial lesion (HSIL), atypical squamous cells cannot exclude high-grade lesion (ASC-H), and atypical glandular cells (AGC)]. The detection performance of normal colposcopic findings was evaluated by calculating sensitivity, specificity, positive predictive value, negative predictive value and overall accuracy. Furthermore, based on follow-up results, the missed detection rate of CINⅡ+in these people and the reasons for missed detection were analyzed. Results: A total of 1 584 women with normal colposcopic findings were included in this study; the median age was 39.0 years. Among them, 1 419 cases were in low-risk group, and 165 cases were in high-risk group. (1) All of the women in high-risk group underwent cervical biopsy, the CINⅡ+ missed detection rate was 37.58% (62/165). In contrast, 133 cases in low-risk group underwent cervical biopsy, of which 14 cases were immediate missed detection, while 1 068 cases who did not undergo cervical biopsy completed one year of follow-up, 16 and 6 new cases of CINⅡ+ were detected at 6 months and 1 year, respectively. Among the 16 new detected cases at 6 months, 12 of them were CINⅡ (1 was focal CINⅡ and 11 were multifocal CINⅡ) and 4 of them were CINⅢ. Among the 6 new detected cases at 1 year, 4 of them were CINⅡ (2 were focal CINⅡ and 2 were multifocal CINⅡ) and 2 of them were CINⅢ. (2) The immediate missed detection rate of CINⅡ+ in low-risk group was 1.17% (14/1 201), and the cumulative missed detection rate at one year was 3.00% (36/1 021). Normal colposcopic findings demonstrated a high negative predictive value (92.83%) for identifying CINⅡ+, particularly in low-risk group (97.00%). However, the specificity was only 74.97%. (3) The TCT and HR-HPV results of women with missed detection of colposcopy were analyzed. All of the 36 cases in low-risk group had HR-HPV infection, 28 cases had HPV 16/18 infection, and 8 people had non-16/18 HR-HPV infection. In the low-risk group with missed detection, 20 cases had the TCT result of NILM, 9 had ASCUS and 7 had LSIL. Among the 62 cases in the high-risk group with missed detection of colposcopy, 56 had HR-HPV infection and 6 did not have HR-HPV infection. Among 56 cases who had HR-HPV infection, 26 cases had HPV 16/18 infection, 30 cases had non-16/18 HR-HPV infection. (4) Univariate and multivariate analyses of factors associated with missed detection of CINⅡ+ lesions identified HR-HPV infection, transformation zone type, and the location of endocervical lesion as independent influencing factors (all P<0.05). Conclusions: The risk of missed detection of cervical CINⅡ+ lesions in in low-risk group people is extremely low, particularly in HPV negative people. Follow-up observation is recommended for these people, which could reduce unnecessary cervical biopsies.