目的 探讨卵巢-附件超声报告和数据风险分层系统(O-RADS US)结合卵巢恶性肿瘤风险算法(ROMA)鉴别卵巢-附件肿瘤良恶性的价值。方法 回顾性分析2021年6月至2023年5月于青岛市市立医院妇产科住院治疗的卵巢-附件肿瘤患者89例的临床资料。采用经阴道超声观察病灶的大小、回声、形态、内部分隔及血流分布等超声特征，按照O-RADS US对病灶进行分类，并通过糖类抗原125(CA125)和人附睾蛋白4(HE4)表达水平计算ROMA值。采用受试者操作特征(ROC)曲线分析O-RADS US、ROMA及其联合诊断卵巢癌的效能。结果 本研究共收集到93个病灶，卵巢癌病灶的最大径及ROMA值均高于良性病灶，差异有统计学意义(P＜0。05)。O-RADS US分类为2、3、4、5类的病灶分别占21。5%(20/93)、26。9%(25/93)、33。3%(31/93)和18。3%(17/93);联合ROMA对O-RADS US分类进行校正，则O-RADS US分类为2、3、4、5类的病灶分别占34。4%(32/93)、29。0%(27/93)、14。0%(13/93)、22。6%(21/93)。联合ROMA后，O-RADS US 2、3、4类卵巢-附件病灶分别有4、6、9个升级为O-RADS US 3、4、5类;同样，O-RADS US 5、4、3类卵巢-附件病灶分别有5、20、16个降级为O-RADS US 4、3、2类。O-RADS US诊断卵巢癌的敏感度、特异度、准确度、曲线下面积分别为80。0%、75。3%、76。3%、0。861。ROMA诊断绝经前和绝经后患者卵巢癌的敏感度、特异度、准确度及曲线下面积分别为85。0%、82。2%、82。8%、0。876及90。0%、89。0%、89。2%、0。904。O-RADS US联合ROMA诊断卵巢癌的敏感度、特异度、准确度及曲线下面积分别为95。0%、91。8%、92。5%、0。926。以O-RADS US联合ROMA诊断卵巢癌的曲线下面积最大，其次是ROMA，差异有统计学意义(P＜0。05)。结论 O-RADS US分类系统可以有效识别卵巢癌，与ROMA联合应用，能够克服单独应用的不足，提高卵巢癌诊断的性能，减少不必要的穿刺活检。
Value of O-RADS US combined with ROMA in differentiating benign and malignant ovarian-adnexal tumors
Objective To ovarian-adnexal reporting and data system lexicon for ultrasound(O-RADS US)combined with the risk of ovarian malignancy algorithm(ROMA)in differentiating benign and malignant ovarian-adnexal tumors.Method Retrospective analysis of clinical data of 89 patients with ovarian adnexal tumors admitted to the Obstetrics and Gynecology Department of Qingdao Municipal Hospital for treatment from June 2021 to May 2023.Transvaginal ultrasound was used to observe the size,echo,morphology,internal separation,and blood flow distribution of lesions.The lesions were classified according to the O-RADS US,and the ROMA value was calculated based on the expression levels of carbohydrate antigen 125(CA125)and human epididymal protein 4(HE4).Using receiver operating characteristic(ROC)curves to analyze the efficacy of O-RADS US,ROMA,and their combination in diagnosing ovarian cancer.Result This study collected a total of 93 lesions,and the maximum diameter and ROMA value of ovarian cancer lesions were higher than those of benign lesions,with statistical significance(P＜0.05).The lesions classified as Class 2,3,4,and 5 by O-RADS US accounted for 21.5%(20/93),26.9%(25/93),33.3%(31/93),and 18.3%(17/93),respectively;Combined with ROMA for progressive correction of O-RADS US classification,lesions classified as class 2,3,4,and 5 accounted for 34.4%(32/93),29.0%(27/93),14.0%(13/93),and 22.6%(21/93),respectively.After combining with ROMA,4,6,and 9 ovarian adnexal lesions of O-RADS US 2,3,and 4 were upgraded to O-RADS US 3,4,and 5,respectively;Similarly,there were 5,20,and 16 ovarian adnexal lesions downgraded to O-RADS US 4,3,and 2,respectively,in O-RADS US 5,4,and 3 categories.The sensitivity,specificity,accuracy,and area under the curve of O-RADS US in diagnosing ovarian cancer are 80.0%,75.3%,76.3%,and 0.861,respectively.The sensitivity,specificity,accuracy,and area under the curve of ROMA in diagnosing ovarian cancer in premenopausal and postmenopausal patients were 85.0%,82.2%,82.8%,0.876 and 90.0%,89.0%,89.2%,0.904,respectively.The sensitivity,specificity,accuracy,and area under the curve of O-RADS US combined with ROMA in diagnosing ovarian cancer were 95.0%,91.8%,92.5%,and 0.926,respectively.The area under the curve for the diagnosis of ovarian cancer using O-RADS US combined with ROMA is the largest,followed by ROMA,and the difference is statistically significant(P＜0.05).Conclusion O-RADS US classification system can effectively identify ovarian cancer when combined with ROMA,which can overcome the shortcomings of using O-RADS US alone to diagnose the ovarian cancer.The adjusted O-RADS US can effectively reduce unnecessary needle biopsy of the ovarian-adnexal mass.
UltrasonographyOvarian-adnexal reporting and data system lexicon for ultrasoundOvarian CancerRisk of ovarian malignancy algorithm
NETL
NSTL
万方数据
