Clinical application of ultrasound modified gynecology imaging reporting and data system(mGI-RADS) in the diagnosis of ovarian tumors / 中华超声影像学杂志

ABSTRACT

Objective:To classify the ultrasound features of ovarian tumors by modified gynecology imaging reporting and data system(mGI-RADS), and explore the clinical value of mGI-RADS in differentiating of benign and malignant tumors.Methods:The 242 ultrasound images of the adnexal mass from 221 patients with ovarian tumor who underwent ultrasound scan and proceeded surgeries in the Second Affiliated Hospital of Harbin Medical University from September 2017 and December 2019 were involved in the retrospective analysis and compared with the pathological results. According to the latest ultrasound terminology for adnexal masses proposed by the International Ovarian Tumor Analysis (IOTA), the adnexal masses were classified. The suspeted malignant sings were screened for mGI-RADS. Using GI-RADS and mGI-RADS classification systems, all the adnexal masses were double-blindly classified by two senior doctors and their diagnostic efficiencies were evaluated.Results:Among the suspected malignant signs proposed by IOTA, resistance index (RI)<0.5, central blood flow signal, ascites, irregular cystic wall with uneven thickness separation, mass composition (solid component ≥30% and silent shadow attenuation), and papillary structure≥7 mm, showed retatively higher OR values(14.282, 10.372, 9.653, 8.832, 5.851, 4.506, respectively. Using GI-RADS and mGI-RADS classification systems, the diagnostic consistency by the two senior doctors in differentiating benign and malignant ovarian neoplasms was good(Kappa=0.767, P<0.05). Grade Ⅳ was divided into three subtypes (Ⅳa, Ⅳb, Ⅳc) by the mGI-RADS, and their positive predictive values for malignant ovarian masses were 69.0%, 90.4% and 92.9% respectively. When a cut-off value >mGRⅣ was used to distinguish the benign and malignant adnexal tumors, the AUC was the largest, and its sensitivity, specificity, positive predictive value, negative predictive value and accuracy were 93.4%, 93.1%, 90.4%, 95.2% and 93.2%, respectively. Conclusions:The mGI-RADS has a high clinical value in the differential diagnosis of ovarian tumors.