通过血清胱抑素C建立慢性肾脏病患者肾小球滤过率预测模型的临床研究

Clinical Study on Prediction Model of Glomerular Filtration Rate in Patients with Chronic Kidney Disease by Serum Cystatin C

目的:研究血清胱抑素C(serum cystatin C,s-cystatin C)与慢性肾脏病(chronic kidney disease,CKD)患者临床估算的肾小球滤过率(estimated glomerular filtration rate,eGFR)之间的关系,建立基于s-cystatin C的CKD患者eGFR预测模型。方法:随机选取住院的84例符合纳入标准的CKD患者,并随机分为试验组和验证组,每组各42例。采用乳胶免疫比浊法检测血清s-cystatin C浓度及酶法测定同步血肌酐浓度,再依据针对中国人简化的肾脏病饮食改良研究公式计算出eGFR(ml/min·1.73 m^2)=1.233×186×血肌酐-1.154×年龄-0.203×[女性×0.742],在试验组中分析s-cystatin C与eGFR之间的相关关系,然后在验证组中对其关系进行验证。结果:①在试验组中发现1/s-cystatin C与eGFR之间存在直线相关性,采用多重线性回归建立的回归方程为:eGFR(ml/min·1.73 m^2)=93.531/s-cystatin C-4.631,方程相关系数为0.86,决定系数为0.74、校正的决定系数为0.734,剩余标准差为11.188,说明该方程的拟合度较好。②在验证组中采用该方程进行验证,发现方程组计算的GFR值样本与MDRD方程计算的eGFR值样本分布无差异,且经ROC曲线分析,eGFR在90、60、30、15 ml/min·1.73 m^2对应的血清s-cystatin C浓度诊断点为:1.15、1.805、2.6、3.805 mg/L。结论:本研究建立的直线回归方程可通过血清胱抑素C来预测肾小球滤过率,具有相对简便、经济、准确的特点,可推荐用于帮助临床医师对CKD患者的临床分期进行初步预判。

Objective To study the relationship between serum cystatin C(s-cystatin C)and the estimated glomerular filtration rate(eGFR)in patients with chronic kidney disease(CKD)and establish an eGFR prediction model for CKD patients based on s-cystatin C.Method 84 CKD patients who met the inclusion criteria were randomly selected and randomly divided into the experimental group and the validation group,with 42 patients in each group.Latex immune turbidimetric method was used to detect serum s-cystatin C concentration and enzymatic determination of serum creatinine concentration,and then on the basis of study the Chinese simplified kidney diet improved formula to calculate the eGFR(ml/min 1.73 m 2)=1.233×186×serum creatinine-1.154×age-0.203[women×0.742],analysis in the experimental group s-relation between cystatin C and eGFR,and then in the validation set to validate the relationship.Results①linear correlation between 1/s-cystatin C and eGFR was found in the experimental group.The regression equation established by multiple linear regression was eGFR(ml/min·1.73 m 2)=93.531/s-cystatin C-4.631,the correlation coefficient was 0.86,the determination coefficient was 0.74,the corrected determination coefficient was 0.734,and the residual standard deviation was 11.188,indicating that the fit degree of the equation was good.②in the validation group,this equation was adopted for verification,and it was found that there was no difference between the GFR value samples calculated by the equations and the eGFR value samples calculated by the MDRD equation.Moreover,ROC curve analysis showed that the diagnosis points of serum s-cystatin C concentration corresponding to eGFR at 90,60,30,15ml/min·1.73 m 2 were:1.15,2.6,1.805,2.6、3.805 mg/L.Conclusion The linear regression equation established in this study can be used to predict the glomerular filtration rate through serum cystatin C,which is relatively simple,economical and accurate,and can be recommended to help clinicians to preliminarily predict the clinical stages of CKD patients.