常规超声结合剪切波弹性成像对大涎腺肿瘤良恶性的诊断价值

目的 探讨常规超声声像图结合剪切波弹性成像（SWE）在诊断和鉴别大涎腺肿瘤良恶性中的价值。方法 回顾性研究。纳入2019年12月—2022年2月蚌埠医学院第一附属医院经手术治疗的69例大涎腺肿瘤患者的临床资料。其中男48例、女21例,年龄12~78（54.3±9.6）岁。患者肿瘤术前未做任何治疗,均行常规超声及SWE检查。以患者手术病理诊断结果为金标准进行分组：良性组41例,恶性组28例;多形性腺瘤组23例,沃辛瘤组11例。比较良性组与恶性组之间、沃辛瘤组与多形性腺瘤组之间的SWE各弹性参数,即平均弹性值（Emean）、最大弹性值（Emax）以及取样框内弹性数据离散度值（Esd）的差异。分析常规超声与常规超声结合SWE图像诊断的结果与病理诊断结果的符合率;采用受试者操作特征曲线（ROC曲线）下面积（AUC）比较这2种方法诊断大涎腺肿瘤良恶性的效能。绘制SWE各弹性参数的ROC曲线,计算AUC,比较Emean、Emax、Esd值的诊断效能。结果 良性组的Emean、Emax以及Esd值（89.45±49.59）、（120.58±57.59）、（15.91±11.57）kPa]均小于恶性组（216.33±53.28）、（252.68±56.42）、（30.49±13.31）kPa],差异均有统计学意义（t=10.13、9.43、4.83,P值均<0.001）,沃辛瘤组的Emean、Emax以及Esd值（37.46±21.57）、（62.35±24.64）、（8.55±5.46）kPa]也均小于多形性腺瘤组（108.51±28.46）、（156.44±37.51）、（18.26±11.27）kPa],差异均有统计学意义（t=7.31、7.55、3.38,P值均<0.01）。常规超声诊断结果与病理结果符合45例（65.2%,45/69）,其中恶性18例、良性27例;常规超声结合SWE诊断结果与病理结果符合58例（84.1%,58/69）,其中恶性23例、良性35例。常规超声与常规超声结合SWE图像诊断大涎腺肿瘤良恶性的特异度、灵敏度及准确度分别为65.9%、64.3%、65.2%和85.4%、82.1%、84.1%,AUC分别为0.651、0.838,差异有统计学意义（Z=2.67,P<0.05）。Emean、Emax、Esd值鉴别大涎腺肿瘤良恶性的AUC分别为0.941、0.929、0.762,Emean、Emax优于Esd。分别以Emean=169.5 kPa、Emax=196.4 kPa、Esd=22.4 kPa为最佳临界值时,其相应特异度分别为92.7%、90.2%、80.5%,灵敏度分别为92.9%、89.3%、64.3%,阴性预测值分别为95.0%、92.5%、76.8%,阳性预测值分别为89.7%、86.2%、69.2%。结论 SWE能在常规超声检查的基础上进一步定量测得大涎腺肿瘤硬度,两者相结合诊断大涎腺肿瘤良恶性准确率明显提高,其中Emean及Emax诊断效能均较好,且均优于Esd。

The value of conventional ultrasound combined with shear wave elastography in the diagnosis of benign and malignant major salivary gland tumors

Objective This study aims to investigate the value of conventional ultrasonography combined with shear wave elastography (SWE) in the diagnosis and differentiation of benign and malignant major salivary gland tumors. Methods This is a retrospective study. The clinical data of 69 (48 males and 21 females, 12-78 54.3±9.6] years old) patients with major salivary gland tumors who underwent surgery in the First Affiliated Hospital of Bengbu Medical College from December 2019 to February 2022 were included. All patients received conventional ultrasonography and SWE examinations without any treatment before tumor surgery. Patients were divided into groups based on the results of surgical and pathological diagnosis as the gold standard: 41 and 28 cases in the benign and malignant groups, respectively, and 23 and 11 cases in the pleomorphic adenoma and Worthing's tumor groups, respectively. The elastic parameters of SWE between the benign and the malignant groups, and between the Worthing's tumor and the pleomorphic adenoma groups, i.e., the mean elasticity value (Emean), the maximum elasticity value (Emax), and the elastic data dispersion value in the sampling frame were compared (Esd). To analyze the coincidence rate between the results of conventional ultrasound and conventional ultrasound combined with SWE image diagnosis and the results of pathological diagnosis. The area under the receiver operating characteristic (ROC) curve (AUC) was used to compare the efficacy of the two methods in diagnosing benign and malignant major salivary gland tumors. The ROC curve of each elastic SWE parameter was drawn, the AUC calculated, and its diagnostic performance was compared. Results The Emean, Emax, and Esd values of the benign group (89.45±49.59] kPa, 120.58±57.59] kPa, and 15.91±11.57] kPa, respectively) were lower than those of the malignant group (216.33±53.28] kPa, 252.68±56.42] kPa, and 30.49±13.31] kPa, respectively), and the differences were statistically significant (t=10.13, 9.43, and 4.83; all P values <0.001). The Emean, Emax, and Esd values of Worthing's tumor group (37.46±21.57] kPa, 62.35±24.64] kPa, and 8.55±5.46] kPa, respectively), were also lower than those of the pleomorphic adenoma group (108.51±28.46] kPa, 156.44±37.51] kPa, and 18.26±11.27] kPa, respectively), and the differences were statistically significant (t=7.31, 7.55, and 3.38; all P values <0.01). Forty-five (65.2%, 45/69) cases were consistent with the results of the conventional ultrasound diagnosis and pathology, including 18 and 27 malignant and benign cases, respectively. The diagnostic results of the conventional ultrasonography combined with SWE were consistent with the pathological results in 58 (84.1%, 58/69) cases, including 23 and 35 malignant and benign cases, respectively. The specificity, sensitivity, and accuracy of the two methods in diagnosing benign and malignant major salivary gland tumors were 65.9% and 64.3%, 65.2% and 85.4%, 82.1% and 84.1%, respectively, and the AUC was 0.651 and 0.838, respectively, and the differences were statistically significant (Z=2.67, P<0.05). The AUC of Emean, Emax, and Esd in distinguishing benign and malignant major salivary gland tumors were 0.941, 0.929, and 0.762, respectively. Emean and Emax were better than Esd. When Emean=169.5 kPa, Emax=196.4 kPa, and Esd=22.4 kPa were the optimal critical values, the corresponding specificities were 92.7%, 90.2%, and 80.5%, and the sensitivities were 92.9%, 89.3%, and 64.3%, respectively. The negative predictive values were 95.0%, 92.5%, and 76.8%, respectively, and the positive predictive values were 89.7%, 86.2%, and 69.2%, respectively. Conclusion SWE can further quantitatively measure major salivary gland tumor stiffness bases on the conventional ultrasonography, and the combination of both can significantly improve the accuracy of diagnosing benign and malignant major salivary gland tumors. Among the elastic parameters of SWE, Emean, and Emax have better diagnostic performance, and both are better than Esd.