摘要
目的 回顾性分析胸部CT值结合血清肿瘤标志物对胸腔积液良恶性诊断价值.方法 选取2022 年 2 月至 2023 年 2 月本院收治的 96 例胸腔积液患者作为研究对象,经病理学检查确诊胸腔积液的良恶性,其中良性 62 例,恶性 34 例;分为良性组与恶性组,所有患者均进行多层螺旋CT扫描检查,获取平均CT值,采取电化学发光技术检测癌胚抗原(CEA)、神经元特异烯醇化酶(NSE)和鳞状细胞癌抗原(SCC)水平.采用t检验比较 2 组胸部CT值和肿瘤标志物水平;运用连续性变量相关性pearson分析胸部CT值、CEA、NSE和SCC水平与恶性胸腔积液的相关性;受试者工作特性(ROC)曲线分析胸部CT值结合肿瘤标志物对胸腔积液良恶性的诊断价值.结果 恶性组胸部CT值(16±7)、CEA[(240.4±34.7)μg/L]、NSE[(50.2±6.3)ng/L]和SCC[(7.02±1.21)ng/L]水平分别为均高于良性组胸部CT值(12±8)、CEA[(20.6±2.8)μg/L]、NSE[(6.0±3.1)ng/L]和SCC[(1.7±0.3)ng/L],差异有统计学意义(P<0.05);使用连续性变量相关性pearson分析得出,胸部CT值与CEA(r=0.38,P<0.001)、NSE(r=0.34,P<0.001)、SCC(r=0.38,P<0.001)呈正相关;ROC曲线分析显示,当胸部CT值、CEA、NSE和SCC的AUC分别为 0.75、0.84、0.81 和 0.78 时,约登指数最大,其截断值分别为 17.71 Hu、23.26 μg/L、26.63 ng/L、3.93 ng/L时,联合诊断AUC(95%CI)为 0.94(0.90,0.99)、灵敏度为 82.4%、特异度为 96.8%、阳性预测值 96.3%、阴性预测值 84.6%均高于个参数单项检测(P<0.05),综合诊断效能最佳.结论 胸部CT值结合CEA、NSE和SCC血清肿瘤标志物对鉴别胸腔积液良恶性具有良好的诊断效能,可在临床推广.
Abstract
Objective To retrospectively evaluate the diagnostic value of chest CT number combined with serum tumor markers in differentiation of malignant from benign pleural effusions.Methods Ninety-six patients with pleural effusion admitted to our hospital between February 2022 and February 2023 were selected,and classified into benign group(n=62)and malignant group(n=34)according to pathological examinations.All patients were examined by multi-slice spiral CT scanning to obtain mean CT numbers.Serum levels of carcinoembryonic antigen(CEA),neuron-specific enolase(NSE),and squamous cell carcinoma antigen(SCC)were measured using electrochemilumine-scence.An independent sample t-test was performed to compare the differences in CT numbers and tumor markers between groups.Then the correlation of CT number,CEA,NSE,and SCC with malignant pleural effusion was discussed using the Pearson correlation coefficient,and the receiver operator characteristic(ROC)curve was plotted to evaluate the diagnostic value of chest CT value combined with tumor markers in differentiation of malignant from benign pleural effusions.Results The CT numbers and levels of CEA,NSE,and SCC were(16±7),(240.4±34.7)μg/L,(50.2±6.3)ng/L,and(7.02±1.21)ng/L in the malignant group,which were all larger than that(12±8),(20.6±2.8)μg/L,(6.0±3.1)ng/L and(1.73±0.31)ng/L in the benign group,with statistical difference(all P<0.05).Pearson correlation analysis denoted that chest CT number was positively correlated with CEA(r=0.38,P<0.001),NSE(r=0.34,P<0.001),and SCC(r=0.38,P<0.001),with statistical difference.ROC curve analysis revealed that the area under the curve(AUC)of 0.751,0.839,0.809,and 0.778 for chest CT number,CEA,NSE,and SCC obtained the largest Youden index.A cut-off value of 17.71 Hu,23.26 μg/L,26.63 ng/L,and 3.93 ng/L for chest CT number,CEA,NSE,and SCC resulted in an AUC(95%CI)of 0.945(0.900,0.990),the sensitivity of 82.4%,specificity of 96.8%,positive predictive value of 96.3%,and negative predictive value of 84.6%in the differentiation of malignant from benign pleural effusions through combined test,which were all larger than those of separate test(P<0.05).Conclusion Chest CT numbers combined with serum tumor markers CEA,NSE,and SCC have good diagnostic efficacy in the differentiation of malignant from benign pleural effusions,which is worthy of application.
基金项目
河南省医学科技攻关计划(LHGJ20210568)
NETL
NSTL
维普
万方数据