Effects of GFR estimating equations derived from different reference methods on staging and long term mortality risks of chronic kidney disease in a Southeast Asian cohort.

Aim: There are limited data on the risks of chronic kidney disease (CKD) in Southeast Asian populations. Several GFR estimating equations have been developed in diverse Asian populations, but they produce markedly discrepant results. We investigated the impact of Asian equations on the mortality risk of CKD in a Thai cohort during long term follow-up, and explored the differences between equations grouped according to the reference GFR methods used to develop them. Methods: Employees of the Electricity Generating Authority of Thailand (n=3430) were enrolled in a health survey and followed up for 22 years. The risks for all-cause mortality for each GFR stage classified by CKD-EPI or different Asian equations were assessed by using Cox proportional hazard models. Results: Equations derived from DTPA clearance (Chinese MDRD, Thai GFR, Singapore CKD-EPI) produced higher GFR, whereas equations from inulin clearance (Japanese CKD-EPI, Taiwan MDRD or Taiwan CKD-EPI) produced lower GFR compared to CKD-EPI. (Average Delta GFR: inulin, -14.9 vs. DTPA +5.80 mL/min per 1.73 m(2), P<0.001). CKD prevalence varied widely (0.7 to 24 %) with inulin-based equations being higher than DTPA-based. GFR stage concordance was over 80% for equations using similar reference method compared to less than 40% between inulin and DTPA-based equations. Low GFR (<45) was an independent mortality risk factor when DTPA-based equations were used, but not when inulin-based equations were used. Conclusion: Chronic kidney disease prevalence and prognosis in Thais varied widely depending on the equation used. Differences in the reference GFR methods could be an important cause for the discrepancies between Asian equations.