詹建福 邱磷安 艾斯① 杨景辉 林涛
【摘要】 目的:探讨血清C-反应蛋白(CRP)、乳酸脱氢酶(LDH)、血清铁蛋白(SF)及血沉(ESR)的线性判别分析(LDA)模型对三类发热青少年的诊断价值。 方法:回顾性分析249例首诊为发热原因待查青少年患者的临床资料,根据出院确诊结果分为急性感染组(78例)、风湿组(84例)及肿瘤组(87例),并进行统计分析。结果:三组患者血清CRP、ESR、LDH和血浆SF水平比较差异均有统计学意义(P<0.05)。以CRP、ESR、LDH及和SF建立LDA模型,该模型对感染组、风湿组和肿瘤组的诊断正确率分别为100%、78.6%和74.7%,预测正确率分别为100%、80.0%和79.3%。结论:血清CRP、LDH、SF及血浆ESR的LDA模型有助于感染组、风湿组和肿瘤组发热青少年的鉴别诊断。
【关键词】 C-反应蛋白; 血沉; 乳酸脱氢酶; 血清铁蛋白; 线性判别分析
doi:10.14033/j.cnki.cfmr.2017.22.028 文章编号 B 文章编号 1674-6805(2017)22-0056-02
The Diagnostic Value of CRP,LDH,SF and ESR for Adolescent Patients with the Symptom of Fever Based on the Linear Discriminant Analysis Model/ZHAN Jian-fu ,QIU Lin-an,AI Si,et al.//Chinese and Foreign Medical Research,2017,15(22):56-57
【Abstract】 Objective:To evaluate the diagnostic value of serum C-reactive protein(CRP),lactate dehydrogenase(LDH),serum ferritin(SF) and erythrocyte sedimentation rate(ESR) for adolescent patients with the symptom of fever based on linear discriminant analysis(LDA) model.Method:Clinical information of 249 adolescent patients with the initial symptom of fever were analyzed retrospectively,and data of serum levels of CRP,LHD,SF and ESR in the infection group(n=78),rheumatism group(n=84) and the tumor group(n=87) were analyzed based the diagnosis at discharge.Result:Significantly elevated serum level of CRP,ESR,SF,and LDH in plasma were found among the infection group,rheumatism group and the tumor group,the differences were statistically significant(P<0.05).The diagnostic accuracy of combined serum CRP,ESR,SF and plasma LDH in the LDA model for the diagnosis of adolescent patients either with infection,rheumatism and rheumatism was 100%,78.6% and 74.7%,and the prediction accuracy was 100%, 80.0% and 79.3%, respectively.Conclusion:The LDA model based on serum CRP, LDH, SF and ESR in plasma was helpful for the diagnosis of patients with the symptom of fever among the infection group,rheumatism group and the tumor group.
【Key words】 C-reactive protein; Erythrocyte sedimentation rate; Serum ferritin; Lactate dehydrogenase; Linear discriminant analysis
First-authors address:Peoples Hospital of Fujian University of Traditional Chinese Medicine,Fuzhou 350004,China
感染性疾病,風湿性疾病和肿瘤是导致青少年出现发热症状最主要的原因[1]。早期、及时诊断发热患者,对患者及时采取最佳治疗方案和减少其经济负担有重要意义[2-3]。血清C-反应蛋白(C-reactive protein,CRP)、乳酸脱氢酶(lactate dehydrogenase,LDH)、血清铁蛋白(serum ferritin,SF)和血沉(erythrocyte sedimentation rate,ESR)是较常用于青少年发热相关疾病的筛查指标,但单项指标较难实现多种疾病的鉴别诊断[4]。鉴于此,本研究以CRP、ESR、LDH和SF建立线性判别分析(linear discriminant analysis,LDA)模型,探讨其对感染,风湿病和肿瘤引起青少年发热的鉴别诊断价值。endprint
1 资料与方法
1.1 一般资料
回顾性分析福建省人民医院、福建省福州儿童医院和福建医大附属协和医院2013年1月-2017年2月出院的青少年患者(联合国规定为满13周岁至24周岁)的临床信息,其中首诊为发热待查患者共249例,根据患者出院诊断结果分为感染组(78例)、风湿组(84例)和肿瘤组(87例),三组患者性别构成比及年龄分布比较差异有统计学意义(P<0.05),见表1。
1.2 方法
患者血清CRP、LDH、SF和血浆ESR均为入院首次检测结果。血清CRP采用日立HITACHI7060,试剂为京九强公司试剂盒,采用速率散射比浊法,正常参考值为0~9.0 mg/L。……