不明原因胸腔积液性质预测模型的构建及其诊断价值  

Construction of a model for predicting the nature of unexplained pleural effusion and its diagnostic value

作　　者：赵志刚 李向南 刘海洋 刘姿 况红艳 ZHAO Zhigang;LI Xiangnan;LIU Haiyang;LIU Zi;KUANG Hongyan(Department of Respiratory and Critical Care Medicine,Henan Provincial People's Hospital,Zhengzhou University People's Hospital,Zhengzhou,Henan 450003,China)

机构地区：[1]河南省人民医院呼吸与危重症医学科郑州大学人民医院,河南郑州450003

出　　处：《中华实用诊断与治疗杂志》2025年第3期271-276,共6页Journal of Chinese Practical Diagnosis and Therapy

摘　　要：目的构建预测不明原因胸腔积液良恶性质的模型,探讨其诊断价值。方法2021年1月—2023年12月河南省人民医院收治不明病因胸腔积液患者234例,均行内科胸腔镜胸膜活检组织病理检查,未明确诊断者进一步行支气管镜、浅表淋巴结活检或经验性治疗明确诊断。依据临床最终诊断结果,将234例患者分为恶性组和良性组。比较2组基线资料(年龄、吸烟情况、肿瘤病史等)、血液指标[红细胞沉降率(ESR)、纤维蛋白原降解产物(FDP)、细胞角蛋白片段21-1(Cyfra21-1)、C反应蛋白(CRP)、D-二聚体(D-D)、结核分枝杆菌T淋巴细胞(T-SPOT.TB)等]、胸腔积液相关指标[颜色、积液量、癌胚抗原(CEA)、腺苷脱氨酶(ADA)等];采用单因素及多因素logistic回归分析不明原因胸腔积液为恶性的影响因素;依据多因素logistic回归分析得到的影响因素和回归系数构建预测模型,绘制ROC曲线,评估预测模型诊断不明原因胸腔积液为恶性的价值。结果(1)恶性组年龄[66(56,72)岁]大于良性组[58(42.5,69)岁](U=4455.50,P<0.001),男性(53.09%)、发热(9.88%)比例及血ESR(55.56%)、血CRP(49.38%)、血D-D(85.19%)、血FDP(51.85%)、血T-SPOT.TB(29.63%)、胸腔积液ADA(2.47%)阳性率均低于良性组(71.90%、42.48%、73.86%、71.90%、97.39%、81.05%、56.86%、31.37%)(χ^(2)=8.088~26.333,P均<0.05),恶性肿瘤(20.99%)、血性胸腔积液(41.98%)比例及血CEA(30.86%)、血Cyfra21-1(56.79%)、胸腔积液CEA(48.15%)阳性率均高于良性组(5.23%、12.42%、4.58%、15.03%、1.31%)(χ^(2)=13.783~80.409,P均<0.05)。(2)胸腔积液颜色(OR=5.044,95%CI:1.615~15.752,P=0.005)、年龄(OR=1.050,95%CI:1.011~1.090,P=0.010)、恶性肿瘤病史(OR=7.511,95%CI:1.903~29.648,P=0.004)、ESR(OR=0.236,95%CI:0.087~0.641,P=0.005)、血Cyfra21-1(OR=6.336,95%CI:2.335~17.196,P<0.001)、血FDP(OR=0.283,95%CI:0.095~0.842,P=0.023)、胸腔积液CEA(OR=89.593,95%CI:10.233~784.430,P<0.001)、胸腔积液ADA(OR=0.154,95%CI:0.025~0.941,P=0.043)Objective To construct a model for predicting benign and malignant unexplained pleural effusion,and to explore its diagnostic value.Methods Totally 234 patients with unexplained pleural effusion were treated in Henan Provincial People's Hospital from January 2021 to December 2023.All patients underwent thoracoscopy and pleural biopsy for pathological examination.For those without a clear diagnosis,further procedures such as bronchoscopy,superficial lymph node biopsy,or empirical treatment were performed for a definitive diagnosis.Based on the final clinical diagnosis results,234 patients were divided into the malignant and benign groups.The baseline data(age,smoking status,tumor history),blood indicators[erythrocyte sedimentation rate(ESR),fibrinogen degradation product(FDP),cytokeratin fragment 21-1(Cyfra21-1),C-reactive protein(CRP),D-dimer(D-D),T-cell spot of tuberculosis(T-SPOT.TB)],and pleural effusion related indicators[color,amount of effusion,carcinoembryonic antigen(CEA),adenosine deaminase(ADA)]were compared between two groups.Univariate and multivariate logistic regression analyses were used to investigate the influencing factors of malignant unexplained pleural effusion.The prediction model was constructed according to the influencing factors and regression coefficients obtained from multivariate logistic regression analysis.ROC curve was plotted to evaluate the value of the prediction model to the diagnosis of malignant unexplained pleural effusion.Results(1)The age of the malignant group[66(56,72)years]was greater than that of the benign group[58(42.5,69)years](U=4455.50,P<0.001).The proportions of males(53.09%)and those with fever(9.88%),as well as the positive rates of blood ESR(55.56%),blood CRP(49.38%),blood D-D(85.19%),blood FDP(51.85%),blood T-SPOT.TB(29.63%),and pleural effusion ADA(2.47%)in the malignant group were all lower than those in the benign group(71.90%,42.48%,73.86%,71.90%,97.39%,81.05%,56.86%,31.37%)(χ^(2)=8.088-26.333,all P values<0.05).The proportion of malignant tumors(20.99%)

关 键 词：不明原因胸腔积液 预测模型 诊断价值 

分 类 号：R561[医药卫生—呼吸系统]