Background: This study aimed to assess the value of endocervical curettage (ECC) in detecting high-grade squamous intraepithelial lesion or worse (HSIL+) in women with type 3 transformation zone (TZ3) lesions, and to identify the clinical characteristics of patients with TZ3 lesions who benefit most from ECC. Methods: This retrospective, multicenter study included 1,905 women with TZ3 lesions who attended cervical screening in one of seven tertiary hospitals in China between January 2020 and November 2021. All participants had received abnormal results and had been referred to colposcopy. Risk factors were identified through univariate and multifactorial logistic analyses. Results: In total, 20.5% (n = 391) of HSIL+ cases with TZ3 lesions had been diagnosed with biopsy and ECC. ECC detected 0.8% (n = 15) HSIL+ cases otherwise missed by biopsy alone. Multivariate analysis identified four factors which influenced detection performance. The probability of detecting HSIL+ with ECC is 2.653 (95% confidence interval [CI] 1.009–6.977) times greater in women aged 40–49 years and 2.545 (95% CI 0.965–6.716) times greater for those aged 50 years and older compared to those younger than 30 years. The probability of ASC-H (atypical squamous cells, cannot exclude high-grade squamous intraepithelial lesion) and HSIL cytologies were respectively 2.415 (95% CI 1.213–4.808) and 2.933 (95% CI 1.648–5.220) times higher than for NILM (negative for intraepithelial lesion or malignancy). Women with human papillomavirus 16/18 infections were 2.299 (95% CI 0.942–5.613) times more likely to be HSIL+. Precancerous lesions were 35.884 (95% CI 12.214–105.426) times more likely in women who had high-grade colposcopic impressions compared to those with normal impressions. Conclusions: ECC should be performed for patients with ASC-H or HSIL cytologies, human papillomavirus 16/18 infections, and for those with high-grade colposcopic impressions. This will increase the number of HSIL+ cases identified using biopsy by reducing the number of false negatives.
