摘要
目的应用Logistic回归分析和受试者工作特征曲线(ROC曲线)探讨血清癌胚抗原(CEA)、神经特异性烯醇化酶(NSE)、细胞角蛋白19片断(CYFRA21-1)以及性别在肺癌诊断中的临床应用价值。方法收集134例肺癌患者、107例肺良性病患者和105例正常人的静脉血,用全自动化学发光免疫分析仪测定血清CEA、NSE和CYFRA21-1含量,通过Logistic回归建立回归模型,绘制ROC曲线并计算曲线下面积(AUC)来评价各指标的诊断价值。结果肺癌组三种肿瘤标记物的水平与肺良性病组和正常对照组比较CEA和CYFRA21-1有统计学差异(P<0.01),而NSE无统计学差异(P>0.05)。在肺癌/正常对照组中,联合检测Y2的AUC大于Y1和各单项检测的AUC,且Y1、CYFRA21-1、CEA检测差异有统计学意义(P<0.01),而性别检测差异无统计学意义(P>0.05)。在肺癌/肺良性病组中,单项检测的CEA、CYFRA21-1的差异有统计学意义(P<0.01),联合检测Y4的AUC大于各单项检测的AUC,而NSE和性别检测差异无统计学意义(P>0.05)。结论单项检测CEA、CYFRA21-1、NSE对鉴别肺癌的意义不大,肿瘤标记物联合性别检测可为临床鉴别肺癌提供有效的参考。
Objective To explore the diagnostic value of serum CEA, NSE, CYFRA21-1 and SEX for lung cancer with Logistic regression and the receive operating characteristic curve. Methods Blood samples from 134 patients with lung cancer, 107 patients with benign lung diseases and 105 disease-free controls were measured by the automated chemiluminescence immunoassay analyzer. We established regression equations, drew ROC curve and compared the area under curve to evaluate the diagnostic value of each indicator of tumors. Results The levels of serum CEA and CYFRA21-1 in patients with lung cancer were significantly higher than other two groups(P<0.01), but NSE was no statistically significant(P>0.05). In the lung cancer/normal control group, the AUC of Y2 was larger than any AUC of Y1, CEA, NSE, CYFRA21-1 and SEX alone, but the AUC of Y1, CYFRA21-1 and CEA alone(P<0.01), SEX was no statistically significant(P>0.05). In the lung cancer/benign lung diseases group, the AUC of CEA and CYFRA21-1 alone(P<0.01), the AUC of Y4 was larger than AUC of CEA, NSE, CYFRA21-1 and SEX alone, but NSE and SEX were no statistically significant(P>0.05). Conclusion Detection of CEA, NSE, CYFRA21-1 alone has slight significance for diagnosis of lung cancer, combined detection of CEA, NSE, CYFRA21-1 and SEX can provide effective reference for clinical identification of lung diseases.
出处
《中华临床医师杂志（电子版）》
CAS
2017年第3期388-391,共4页
Chinese Journal of Clinicians(Electronic Edition)
关键词
肺肿瘤
肿瘤标记
生物学
诊断
Lung neoplasms
Tumor markers
biological
Diagnosis
