标题：超声GI-RADS分类对卵巢良恶性肿瘤的鉴别诊断价值

      作者：张姣 1，鲁琳 1，李超 2    陕西省核工业二一五医院超声科 1、科教科 2，陕西 咸阳 712000

      卷次： 2022年33卷5期

      【摘要】 目的 探讨超声妇科影像报告与数据系统 (GI-RADS)分类对卵巢良恶性肿瘤的鉴别诊断价值。方法 回顾性分析 2018年 12月至 2020年 12月陕西省核工业二一五医院收治的140例卵巢肿瘤患者的临床资料，所有患者入院后均行超声检查。根据病理检查结果将卵巢肿瘤患者分为恶性肿瘤(恶性组) 43例和良性肿瘤(良性组) 97例，比较两组患者的超声特征、彩色多普勒血流信号分级及血流参数[收缩期峰值速度(PSV)、舒张末期流速(EDV)、阻力指数(RI)、博动指数(PI)]，以病理检查结果作为金标准，采用四格表卡方检验分析GI-RADS对卵巢肿瘤良恶性定性诊断的临床价值。结果 良性组患者的超声检查影像学特征以肿块多发(87.63%)、直径<3 cm(79.38%)并伴随高回声(76.29%)为主，而恶性组患者超声检查影像学特征以肿块单发(81.40%)、直径≥3 cm(79.07%)并伴随低回声(72.09%)为主，两组在数量、直径和回声方面比较差异均有统计学意义(P均<0.05)；良性组患者的血流信号分级以Ⅰ级(42.27%)~Ⅱ级(31.96%)为主，恶性组患者的血流信号分级以Ⅲ级(25.58%)~Ⅳ级(41.86%)为主，良性组血流信号分级明显低于恶性组，差异均有统计学意义(P<0.05)；良性组患者的 PSV、EDV分别为(15.14±5.58) cm/s、(6.24±2.10) cm/s，明显低于恶性组的(24.24±6.47) cm/s、(11.24±4.15) cm/s，差异均有统计学意义(P<0.05)；良性组患者的RI、PI分别为(0.62±0.17)、(0.92±0.24)，明显高于恶性组的(0.35±0.16)、(0.66±0.23)，差异均有统计学意义(P<0.05)；超声GI-RADS分类对卵巢肿瘤良恶性定性诊断的敏感度、特异度、准确率和阳性预测值、阴性预测值分别为 86.67% (39/45)、95.79% (91/95)、92.85% (130/140)、90.70% (39/43)、93.81% (91/97)。结论 超声GI-RADS分类对卵巢良恶性肿瘤的定性诊断具有重要指导意义。
      【关键词】 卵巢肿瘤；超声；超声妇科影像报告与数据系统；肿瘤性质；血流参数；鉴别价值
      【中图分类号】 R737.31 【文献标识码】 A 【文章编号】 1003—6350（2022）05—0606—04

Value of ultrasonic gynecology imaging reporting and data system classification in differential diagnosis ofovarian benign and malignant tumors.
ZHANG Jiao 1, LU Lin 1, LI Chao 2. Department of Ultrasonography 1, Departmentof Science and Education 2, the 215th Hospital of Shaanxi Nuclear Industry, Xianyang 712000, Shaanxi, CHINA
【Abstract】 Objective To explore the value of ultrasound gynecology imaging reporting and data system(GI-RADS) classification in the differential diagnosis of ovarian benign and malignant tumors. Methods The clinicaldata of 140 patients with ovarian tumor admitted to the 215th Hospital of Shaanxi Nuclear Industry from December 2018to December 2020 were retrospectively analyzed. All patients underwent ultrasound examination after admission. Ac-cording to the result of pathologic examination, patients with ovarian tumors were divided into malignant tumors (malig-nant group, n=43) and benign tumors (benign group, n=97). The ultrasonic characteristics, color Doppler blood flow sig-nal classification and blood flow parameters [peak systolic velocity (PSV), end diastolic velocity (EDV), resistance in-dex (RI) and pulsatility index (PI)] were compared between the two groups. Using pathological examination results asthe gold standard, the clinical value of GI-RADS in the qualitative diagnosis of benign and malignant ovarian tumorswas analyzed by chi-square test of four grids. Results The ultrasonic imaging features of patients in the benign groupwere mainly multiple masses (87.63%), diameter<3 cm (79.38%) with hyperechogenicity (76.29%), while those in themalignant group were mainly single mass (81.40%), diameter≥3 cm (79.07%) with hypoechogenicity (72.09%); the twogroups were significant in number, diameter and echo (P<0.05). The blood flow signal grading of benign group wasmainly grade Ⅰ (42.27% ) to Ⅱ (31.96% ), and that in the malignant group was mainly grade Ⅲ (25.58% ) to Ⅳ(41.86%); the blood flow signal grading of benign group was significantly lower than that in malignant group (P<0.05).The PSV and EDV of benign group were (15.14±5.58) cm/s and (6.24±2.10) cm/s, respectively, which were significantlylower than (24.24±6.47) cm/s and (11.24±4.15) cm/s of malignant group (P<0.05). The RI and PI in benign group were0.62±0.17 and 0.92±0.24, respectively, which were significantly higher than 0.35±0.16 and 0.66±0.23 in malignantgroup (P<0.05). The sensitivity, specificity, accuracy, positive predictive value, and negative predictive value of ULTRA-SONIC GI-RADS classification for benign and malignant qualitative diagnosis of ovarian tumor were 86.67% (39/45),95.79% (91/95), 92.85% (130/140), 90.70% (39/43), 93.81% (91/97), respectively. Conclusion Ultrasound GI-RADSclassification has important guiding significance in qualitative diagnosis of ovarian benign and malignant tumors.
      【Key words】 Ovarian tumor; Ultrasound; Ultrasonic gynecology imaging reporting and data system;