M. Quinn*¹, A. Rainey², K. J. Cairns², A. H. Marshall², F. Kee³, G. Savage³, D. G. Fogarty^1
1Department of Nephrology, ²Centre for Statistical Science and Operational Research, ³Department of Epidemiology, Queen's University, Belfast, United Kingdom

Introduction: Chronic kidney disease (CKD) guidelines have focused on the utility of the modified 4-variable MDRD equation (traceable by isotope dilution mass spectrometry IDMS). This formula accounts for variance in creatinine measured by an analyser different to that used when the original MDRD equation was devised.
Aims: To assess theoretically and in practice the effect sizes of IDMS correction over the 4-variable MDRD equation in eGFR calculation with a range of UK laboratories and the subsequent impact of this on CKD prevalence.
Methods: MATLAB generated a range of creatinine data (30-300umol/l) for male and female patients aged 20-100 years. The maximum differences between the IDMS and MDRD equations for all 14 UK laboratory techniques were explored with an averaged (IDMS + MDRD)-eGFR less than 60mls/min and 30mls/min. Similar procedures were applied to 712,540 samples (reflecting 5 methods in Northern Ireland), belonging to patients 18 years+, to explore graphically maximum differences in techniques. CKD prevalence using both estimation equations was compared.
Results: Simulated data indicates that the majority of laboratories in the UK demonstrate small differences between the IDMS and MDRD methods in stages 4 and 5 CKD (where the averaged maximum difference for all laboratory methods was 1.27mls/min for females and 1.59mls/min for males).MDRD deviated furthest from the IDMS results for the Endpoint Jaffe method: the maximum difference of 9.93mls/min for females and 5.42mls/min for males occurred at extreme ages and in those with eGFR reflecting stage 3 or higher disease. The real data graphically agreed with these theoretical results.
Existing data for 93,870 patients yielded a first MDRD eGFR<60mls/min in 2001. 66,429 (71%) had a second test >3months later of which 47,093 (71%) continued to have an eGFR<60mls/min. Estimated crude prevalence was 3.97% for laboratory detected CKD in adults using the MDRD equation which reduced to 3.69% when applying the IDMS equation. Over 95% of this difference in prevalence was explained by older females with stage 3 CKD; yet close to the stage 2 interface, reemphasising the need for further research into the subcategorisation of stage 3 CKD.

Conclusions: Improved accuracy of eGFR is obtainable by using IDMS corrected eGFR especially in early stage CKD; however our data suggests this will have little practical impact on stages 4-5 considering the current referral guidelines.

1st Image/Graph:
2nd Image/Graph: