摘要
目的:探讨血清肌酐(CR)联合胱抑素C(CYC)对慢性肾衰竭(CRF)患者的预测价值。方法:选取90例CRF患者为研究对象,依据慢性肾脏疾病(CKD)不同进展分为CKD3期(n=28)、CKD4期(n=33)及CKD5期(n=29);另选取同期健康体检者作为对照组(n=45)。采用受试者工作特征(ROC)曲线验证CR联合CYC的预测价值;Logistics回归分析独立影响因素;建立人工神经网络模型并验证。结果:CRF组高血压、糖尿病患病率、HBA1c、GLU、URCA、BUN、WBC、TG、hs-CRP、CR、CYC等指标均高于对照组(P<0.05)。CR联合CYC的AUC为0.895,CR、CYC截断值分别为51.32μmol/L及0.76 mg/L。HBA1c、GLU、BUN、WBC、hs-CRP、CR、CYC水平在不同CKD分期中,差异有统计学意义(P<0.05),HBA1c、BUN、WBC、hs-CRP、CR、CYC水平升高是CRF进展的独立危险因素(P<0.05)。人工神经网络模型显示,CR、CYC、BUN权重较高,模型AUC为0.869(95%CI:0.786~0.915,P<0.05),一致性指数为0.873;校准曲线Hosmer-Lemeshow检验结果差异无统计学意义(P>0.05)。结论:血清CR联合CYC对CRF有较高预测价值,HBA1c、BUN、WBC、hs-CRP、CR、CYC水平是CRF患者CKD进展的危险因素。
Objective:To explore the predictive value of serum creatinine(CR)combined with cystatin C(CYC)in patients with chronic renal failure(CRF).Methods:A total of 90 CRF patients were selected as the research subjects.According to the staging criteria of chronic kidney disease(CKD),the patients were divided into three stages:CKD 3(n=28),CKD 4(n=33)and CKD 5(n=29).Another 45 patients with sound renal function in the same period were included in the control group.The predictive value of CR combined with CYC was verified by receiver operating characteristic(ROC)curve.Independent influencing factors were analyzed by multivariate Logistics regression.The artificial neural network model was established and verified.Results:The proportion of hypertension and diabetes and the levels of HBA1c,GLU,URCA,BUN,WBC,TG,hs-CRP,CR and CYC in the CRF group were higher than those in the control group(P<0.05).CR in conjunction with CYC s AUC=0.895.The optimal critical values of CR and CYC were 51.32μmol/L and 0.76mg/L,respectively.There were significant differences in the levels of HBA1c,GLU,BUN,WBC,hs-CRP,CR and CYC in different CKD stages(P<0.05),and increased levels of HBA1c,BUN,WBC,hs-CRP,CR and CYC were independent risk factors for the progression of CRF(P<0.05).The artificial neural network model showed that CR,CYC,and BUN had higher weights,with a model AUC of 0.869(95%CI:0.786~0.915,P<0.05).The consistency index was 0.873.There was no significant difference in the results of Hosmer-Lemeshow test(P>0.05).Conclusion:The levels of CR combined with CYC has high predictive value for CRF.The levels of HBA1c,BUN,WBC,hs-CRP,CR and CYC are all risk factors for CKD progression in patients with CRF.
作者
罗小宝
黄丽
黄德喜
LUO Xiao-bao;HUANG Li;HUANG De-xi(Department of Laboratory,Tongling Municipal Hospital,Tongling 244099,Anhui,China)
出处
《川北医学院学报》
CAS
2023年第6期822-826,共5页
Journal of North Sichuan Medical College
关键词
血清肌酐
胱抑素C
慢性肾衰竭
人工神经网络模型
Serum creatinine
Cystatin C
Chronic renal failure
Artificial neural network model
