Establishment Of New Formulae For Evaluating GFR And Analysis Of Risk Factors For GFR Decline With Age In Healthy Individuals

Backgrounds and Objective:Kidneys are fragile organs subject to aging. Glomerular filtration rate(GFR)is a good indicator for evaluating renal function which declines with age.Cross-sectional studies and longitudinal study show that, for the persons over 40 years old,the creatinine clearance rate declines with age at a rate of 7.5 to 8 ml per 10 years.The decline of GFR with age has been recognized from studies which had excluded only overt kidney diseases,but not the potential undetected diseases that might still exert influence on GFR, Therefore,the variation of GFR with age of healthy individuals is not really clear yet.Currently,there are several methods for the evaluation of renal functions,but none has been proved to be perfect in healthy individuals.Creatinine clearance is the most frequently employed method,but urine collecting is cumbersome and susceptible to error.24-hour urine collection is no longer recommended for routine estimatation of kidney functions as the measurement of creatinine clearance is strongly influenced by urine output.As a simple and fast method, renal dynamic visualization with 99myc DTPA does not require blood and urine samples,and simultaneously provides information on renal functions.However, the clinical application of renal dynamic visualization is limited by its uncertainties due to various influencing factors.In 2003,K/DOQI guideline suggested that MDRD and CG formulae be adopted for the assessment of GFR. Nevertheless,the two formulae were obtained based on studies on CKD,and hardly applicable to healthy individuals.The modified MDRD formula in China has been found to be more accurate and precise than the MDRD one,but whether it is suitable for healthy individuals is unclear.In this study,the GFR assessment of healthy individuals was performed with the double plasma method as a reference.The investigations were carried out on a group of northern Chinese.The report covers the following aspects:(1)assessing the performance of five available methods for GFR assessment in healthy individuals;(2)establishment of new formulae for estimating GFR based on the double plasma method;(3)risk analysis of the GFR decrease with age based on the longitudinal study.Methods:1.The GFR of 144 community-dwelling healthy adults were measured with dual plasma method and renal dynamic visualization of 99m Tc DTPA.Creatinine clearance was calculated with blood and urinary creatinine.GFR was estimated with Cockcroft-Gault equation,abbreviated MDRD equation and the modified MDRD equation as well.The performance of the five methods for GFR estimate was assessed with dual plasma method.2.Establishment of new equations for the GFR determination of healthy individuals:the GFR measured with dual plasma method was used as the dependent variable,while the independent variables included sex,age,serum creatinine,urea,cystatin C and the body surface area.Multivariable linear regression model were used to establish new equations for renal assessment.At the same time,the equation with the dual plasma method and the renal dynamic visualization was established3.Risk analysis of the decrease of GFR with age:the baseline and follow-up data of the study population were collected.GFR in 2004 and 2009 were estimated with Cockcroft-Gauit equation,abbreviated MDRD equation,the modified MDRD equation,and the newly established equation of GFR estimate.△GFR was calculated with 2009 GFR minus the 2004 GFR,and△GFR less than zero indictes GFR decline.Multiple linear regression model was used to analyze the risk factors of GFR decline with age,and△GFR being the dependent variable,and the baseline index the independent variable.Results: 1.For Ccr,CG GFR,Tc GFR,aGFR,and c-a GFR,the bias were 0.11,13.57, -6.11,-8.89 and -3.64 ml/min.l.73m~2,respectively.All the GFR estimating methods underestimated GFR except Ccr.The precision for the five GFR estimating methods were 24.5,26.2,17.5,12.9 and 11.8 ml/min.1.73m~2, respectively.The accuracy and the 30%relative accuracy for Ccr,CG GFR,Tc GFR,a GFR and c-a GFR were 24.7%,19.3%,16.7%,16.1%and 14.9%,70.1%, 85.4%,81.9%,90.2%and 90.2%,respectively.The areas under the Receiver Operating Characteristic Curve(ROC)for CG GFR were the biggest as 0.897 and 0.856,respectively.2.The new equation were established with renal dynamic visualization as follows:GFR=88.287+0.43×Tc GFR-0.634×age(r=0.727,r~2=0.529 )for female and GFR=103.789+0.275×Tc GFR-0.645×age(r=0.601,r~2=0.362)for male. The equations for creatinine,cystatin C,and the combination of both were GFR=170.729-0.733×age-0.705×Scr+14.016(male)(r=0.714, r~2=0.509);GFR=146.64-0.665×age-28.225×cystatin C(r=0.642, r~2=0.410)and GFR=172.538-0.595×age-0.631×Scr-18.521×cystatin C +12.942(male)(r=0.728,r~2=0.530).The performance of the newly established equations was better in the healthy individuals.3.The associated factors for GFR decline with age:Although the associated factors were not the same for Cockcroft-Gault equation,abbreviated MDRD equation,the modified MDRD equation,and the newly established equation of GFR estimate,smoke and systolic pressure were associated with GFR decline with age in male,but cystatin C for female.Conclusions:1.The five GFR-estimating methods being used at present can not evaluate the GFR in healthy individuals perfectly.2.The four newly established equations for GFR assessment were better than the GFR-estimating methods used at present,although further verification is still needed before application to clinical use. 3.The associated factors of GFR decline with age were different between the genders,which should be taken into consideration separately.