Development of estimating equations for glomerular filtration rate among Southeast Chinese population

Abstract Purpose: The accuracy of newly reported estimated glomerular filtration rate (eGFR) equations has not been evaluated among Southeast Chinese, and whether there is a better eGFR equation that could improve GFR assessment was unknown. Thus, we aimed to explore an eGFR equation with high accuracy among Southeast Chinese population. Patients and methods: A total of 2449 Southeast Chinese participants with gold standard of measured GFR (mGFR) were enrolled. All participants were randomly divided into development and validation datasets to develop new eGFR equations by linear regression analysis. The performance of the new and reported equations was evaluated in both datasets, and subgroups with different age, sex, diabetes, and CKD status. Results: A total of three new eGFR equations were developed, and named as SC-GFRcr, SC-CKD-EPIcr, SC-GFRcr-cys equations. The lowest bias (-0.32, 0.26 mL/min/1.73 m2) was observed for SC-GFRcr equation. The highest P30 level (79.44%, 78.45%) and lower root mean squared error (RMSE) (17.18, 17.79) were observed for SC-CKD-EPIcr equation in the development and validation datasets. The correlation coefficients between the new equations and mGFR were all >0.7, the sensitivity of SC-CKD-EPIcr equation was as high as 76.33%. Meanwhile, four previously reported equations were selected and compared with the 3 new equations. In the development and validation datasets, and subgroup analyses, newly developed SC-CKD-EPIcr equation performed superior than others, and eGFRcr was best among all previously reported equations. Conclusions: The new-developed eGFR equation, SC-CKD-EPIcr, could improve GFR assessment, especially in Southeast China. Still, external validation of the equation is needed.