Assessing and improving creatinine-based glomerular filtration rate equations to evaluate kidney disease in the United Kingdom (AIM CKD UK)

Research summary

Primary Aim: 1. To assess the bias,precision,limits of agreement and 30% accuracy of eGFR equations compared to measured GFR for people of different ethnicities (e.g. South Asian,Chinese,South-East Asian and Black) in the United Kingdom,compared to people of reported White ethnicity. Design: Retrospective observational data only multi-centre cohort study. Study Population: People with both normal kidney function and chronic kidney disease,who have had a measured GFR test and a paired eGFR test (within 30 days),between 2009 and 2022 inclusively. The study population will include people from diverse ethnic backgrounds in the United Kingdom. Sample Size: 1200 participants,with a minimum of 240 in each of the following ethnicity groups (White,Black,South Asian,Other,Not specified) Analysis: Estimated GFR (eGFR) will be calculated using multiple equations including: the CKD-EPI equation with serum creatinine (2009 and 2021),MDRD with serum creatinine,the FAS (full age spectrum) equation,Lund Malmo revised equation and the European Kidney Function Consortium equation. eGFR for each equation,will be compared to mGFR by calculating bias (median and percentage),precision,30% accuracy (P30; as this is the most commonly used percentage for assessment of eGFR compared to mGFR) and Bland Altman plots with limits of agreement. Misclassification of CKD and CKD staging will be reported as counts and percentages for each eGFR equation. Bias between eGFR equations will be compared using ANOVA / Kruskal Wallis. Subgroup analyses will be performed stratified by ethnicity group (White,Black and South Asian),CKD / non-CKD cohorts,normal / low albumin concentrations and time between eGFR and mGFR testing (3 groups: same day,1 - 7 days,8-30 days).

Principal Investigator

Dr Edward John Sharples

Contact us

Email: renalandtransplanttrials@ouh.nhs.uk

IRAS number

320215