Value of using ultrasound features to improve the Ovarian-Adnexal Image Reporting and Data System Category 4 in the benign-malignant differential diagnosis of ovarian-adnexal masses
10.3760/cma.j.cn131148-20241014-00526
- VernacularTitle:基于超声特征改良2022版O-RADS 4类鉴别诊断卵巢-附件肿物良恶性的价值
- Author:
Lei WU
1
;
Yingnan WU
;
Jing ZHAO
;
Liping GONG
;
Shuang ZHANG
;
Jiawei TIAN
;
Zhirong HE
;
Litao SUN
Author Information
1. 锦州医科大学研究生培养基地（浙江省人民医院），杭州 310014
- Publication Type:Journal Article
- Keywords:
Ultrasonography;
Ovarian Adnexal Reporting and Data System;
Ovarian cancer;
Risk of malignancy
- From:
Chinese Journal of Ultrasonography
2025;34(3):232-238
- CountryChina
- Language:Chinese
-
Abstract:
Objective:To explore the value of ultrasound features modified version 2022 of the Ovarian-Adnexal Imaging Reporting and Data System（O-RADS）Category 4 in the differential diagnosis of benign and malignant ovarian-adnexal tumors.Methods:Retrospective analysis was conducted in 501 cases with ovarian masses classified into 4 categories according to the 2022 version of O-RADS who were collected from 4 clinical centers［the Second Afliated Hospital of Harbin Medical University（188 cases），Zhejiang Provincial People's Hospital（146 cases），Sichuan Provincial Maternity and Child Health Care Hospital（90 cases），and Fuling Hospital of Chongqing University（77 cases）］from January 2018 to July 2024 with concomitant surgical resection.The 424 cases from 3 of the clinical centers（the Second Hospital of Harbin Medical University，Zhejiang Provincial People's Hospital，and Sichuan Maternal and Child Health Hospital）were randomly divided into a training group（339 cases）and an internal validation group（85 cases）according to an 8∶2 randomization，while the cases from the other clinical center（Fuling Hospital of Chongqing University）were selected as the external validation group（77 cases），and the pathological diagnosis was used as the “gold standard”.Univariate and multifactorial logistic regression analyses were performed on the ultrasound characteristics of the training group to screen the independent predictors associated with ovarian carcinogenesis，and to formulate the stratification rules for the 4 types of masses in O-RADS. The ROC curve of this stratification method was plotted and the area under the curve（AUC）was calculated，and it was validated in the internal validation group and the external validation group；and the diagnostic accuracy was compared with that of the 2022 version of O-RADS.Results:Univariate logistic analysis showed that cysts with solid components，≥ 4 papillary projections，smooth inner wall of the cyst，color flow score ≥ 3 points，and acoustic shadowing were independent predictors of ovarian cancer（all P < 0.05）；while multifactorial logistic analysis showed that cysts with a solid component and a color flow score ≥3 points were independent risk factors of ovarian cancer（all P < 0.05），and smooth cyst walls and acoustic shadows were independent protective factors（all P < 0.05）.The diagnostic accuracies of the modified training group，internal validation group，and external validation group were 73.7%，68.2%，70.1%，respectively，which were significantly higher than the diagnostic accuracies of the 2022 version of the O-RADS（38.9%，37.6%，33.8%）（all P < 0.05）.The diagnostic sensitivity，specificity and AUC of the training group were 0.871，0.652，0.762，respectively，while the internal validation group were 0.844，0.585，0.714，and 0.846，0.627，0.737 in the external validation group. Conclusions:Improvement of the 2022 version of O-RADS category 4 using ultrasound features may improve the identification of benign and malignant ovarian-adnexal tumors.
