POS-230 COMPARISON OF MEASURED GLOMERULAR FILTRATION RATE BY PLASMA IOHEXOL CLEARANCE WITH EXISTING EQUATIONS OF ESTIMATING GLOMERULAR FILTRATION RATE IN INDIAN SUBJECTS

Abstract

Glomerular filtration rate (GFR) is a basis of diagnosis and severity of renal diseases. Currently used estimating GFR equations are ethnicity- specific and have not been validated in Indian population. We evaluated the performance of the existing eGFR equations for assessment of GFR in Indians using plasma clearance of Iohexol method. We measured GFR in 208 subjects who were being evaluated as healthy prospective kidney donors (HPKDs) or diagnosed as chronic kidney disease (CKD). The demographic details, anthropometric measurements and clinical data were recorded and baseline blood sample was collected. 3-5ml Iohexol was intravenously injected and blood sampling was done at 60, 120, 180 and 240 mins. The Iohexol were analysed using UHPLC. Serum creatinine was measured by Jaffe’s kinetic Assays and serum Cystatin C using ELISA kit. The GFR was measured by plasma disappearance rate of Iohexol. The dietary protein intake was measured using 24-hour diet recall method. The precision and accuracy of estimating GFR equations was calculated against measured GFR. The bias was expressed as mean difference between measured GFR and estimated GFR and precision was expressed as 95% confidence interval of bias. The accuracy (P30) was calculated as number of subjects with eGFR within ±30% of mGFR. Study participants with mean age 48.4±11.1 years included 130 CKD patients and 78 prospective renal donors. 52% were male and half of the total subjects were vegetarian. The average daily protein intake was 49.9±14.0 g/day. The mean mGFR was 49.0±25.0 ml/min/1.73m2. The mean estimated GFR was 64.6±36.1 ml/min/1.73m2, 46.9±22.2 ml/min/1.73m2and 64.1±40.7 ml/min/1.73m2by CKD-EPIcr, CKD-EPIcys and MDRD, respectively. Barring CKD-EPIcys equation, a significant difference was observed between mGFR and other two eGFR equations (CKD-EPIcr: p<0.001 and MDRD: p<0.001). The mean bias was -15.6±22.4 ml/min/1.73m2(95% CI: -18.6 to -12.5), 2.1±16.8 ml/min/1.73m2(95% CI: -0.1 to 4.4) and -15.1±26.4 ml/min/1.73m2 (95% CI: -18.7 to -11.8) for CKD-EPIcr, CKD-EPIcys and MDRD equations. The P30 was 46.2% for CKD-EPIcr, 57.7% for CKD-EPIcys and 45.2% for MDRD equation. When data were stratified by CKD and HPKD groups, the mean mGFR was 35.1±14.4 ml/min/1.73m2 in CKD and 72.3±21.4 in HPKD. 32% of HPKDs had measured GFR <60ml/min/1.73m2. The mean bias for CKD-EPICr, CKD-EPICys and MDRD were -5.6±14.4 ml/min/1.73m2 (95%CI: -8.1 to -3.1), 1.6±12.9 ml/min/1.73m2 (95%CI: -0.62to 3.9) and -3.2 ± 14.5 ml/min/1.73m2 (95%CI: -5.7 to -0.6) in CKD patients and it was -32.2±23.6 ml/min/1.73m2(95% CI: -37.4 to -26.8), 3.0±21.9 ml/min/1.73m2 (95% CI: -1.9 to 7.9) and -35.0±29.9 ml/min/1.73m2 (95% CI: -41.7 to -28.3 in HPKD. In CKD group, P30 for CKD-EPIcr, CKD-EPICys and MDRD was 53.8%, 51.5% and 55.3% whereas it was 33.3%, 67.9% and 28.2% in HPKDs. Creatinine-based eGFR equations overestimated GFR. However, eGFR by CKD-EPIcys was close to mGFR; exhibited lowest bias and highest accuracy. The accuracy of eGFR equations against mGFR in Indians will become clear in future study in healthy general population to assess the actual GFR.