Cross-national disparities in eGFR and CKD classification in HF trials: a comparative analysis of five equations in QUEST, PARADIGM-HF, and DAPA-HF
L Shen, I Cheang, X Zhu, M Petrie, C S P Lam, P Tolomeo, P S Jhund, S Solomon, M Packer, J McMurray, X L LiAbstract
Background
Estimated glomerular filtration rate (eGFR) is central to trial eligibility and therapeutic decision-making in heart failure (HF). However, landmark HF trials have used different equations to estimate eGFR. The impact of these varying equations on eGFR distribution and chronic kidney disease (CKD) classification across diverse global populations remains uncertain.
Purpose
To compare the eGFR distribution and CKD staging between patients enrolled in China and those from other countries in the QUEST, PARADIGM-HF, and DAPA-HF trials using five validated creatinine-based equations.
Methods
This analysis included patients with symptomatic HFrEF from QUEST (N=3110; China only), PARADIGM-HF (N=8399; China n=353), and DAPA-HF (N=4744, China n=237). Renal exclusion criteria were serum creatinine >2 mg/dL in QUEST, and eGFR <30 mL/min/1.73m² in PARADIGM-HF and DAPA-HF. In the original trials, eGFR was calculated using CKD-EPI 2021 in QUEST, modified MDRD in PARADIGM-HF, and CKD-EPI 2009 in DAPA-HF. For this analysis, eGFR was recalculated using five equations: CKD-EPI 2009, CKD-EPI 2021, modified MDRD, European Kidney Function Consortium (EKFC), and Xiangya (Li et al Kidney Int. 2019;95:636-646). eGFR distributions and KDIGO CKD stages were compared across equations and regions, stratified by age (<60, 60-69, and ≥70 years) and sex.
Results
Across all trials, mean eGFR was highest with CKD-EPI 2021 and lowest with the Xiangya equation in Chinese cohorts or EKFC in non-Chinese cohorts, with a maximum between-equation difference of 7-10 mL/min/1.73m² (Figure 1). The Xiangya equation yielded the narrowest distribution (smallest standard deviation), and classified most patients as CKD stage 2 (66-86%). Conversely, CKD-EPI 2021 categorized more patients as stage 1, while EKFC consistently identified a greater proportion of stage ≥3 CKD. Mean eGFR was 6-11 mL/min/1.73m² higher in Chinese patients than those from other countries across all trials and equations (Figure 1). Accordingly, the prevalence of CKD stage ≥3 was significantly lower in Chinese cohorts. Although patients enrolled in China were generally younger and more frequently male, the higher eGFR in Chinese participants persisted across all age strata and in both sexes, with regional differences attenuating at older ages.
Conclusions
eGFR values and CKD classifications in HFrEF trials vary meaningfully depending on the equation used. Patients enrolled in China consistently exhibit higher eGFR and lower CKD stages than their global counterparts, differences only partially explained by age and sex. These findings highlight the need for standardized renal assessment in global HF trials and have important implications for trial eligibility criteria and external validity.Mean eGFR values and CKD classifications