Evaluation of the CKD-EPI creatinine based glomerular filtration rate estimating equation in Black African and Indian adults in KwaZulu-Natal, South Africa.

Background: The estimation of glomerular filtration rate (GFR) plays a vital role in assessment of the renal function. This study evaluated the performance of the CKD-EPI and MDRD equations in the South African Kwa-Zulu Natal population.

Objectives: The objectives if the study were to compare the of CKD-EPI and MDRD equations in the selected population to the measured GFR using Sodium Technetium-99 m-diethylene-triamine-pentaacetate clearance derived GFR.

Method: Records of adult patients with measured GFR performed at the Nuclear Medicine Department at Inkosi Albert Luthuli Central Hospital, Durban, South Africa from 1 April 2014 to 31 March 2016 were reviewed. eGFR for all included patients was calculated using the MDRD equation without African American correction factor and the CKD-EPI equation with and without the African American correction factor for the Black African patients. Statistical comparison of the eGFR with measured GFR was performed with Bland Altman bias plots, Wilcoxon match pairs signed ranks test and accuracy within 10% and 30%.ROC curve analysis assessed the sensitivity and specificity at eGFR <90 and < 60 ml/min/1.73m.

Results: After exclusion, 287 patients were included for analysis with sufficient numbers for only the Black African and Indian patients. None of the equations showed accuracy of eGFR within 30% of measured GFR for 90% of patients. In the Black African population, the CKD-EPI equation without the correction factor performed best. 17% and 14.4% of the Black African participants would be reclassified with the CKD-EPI equation without and with the African American correction factor respectively compared to mGFR at a cut-off of 60 mls/min/1.73m.

Conclusion: None of the evaluated equations attained the 2002 KDOQI benchmark of P >90%. 11.1-17% of individuals would have been incorrectly classified using the CKD-EPI equation.