Prevalence and Mortality Implications of Differences in Creatinine- and Cystatin C- Based Estimated Glomerular Filtration Rate in U.S. Adults, 1999-2004

Abstract

Background: Discrepancies between creatinine-based and cystatin C-based estimated glomerular filtration rate (eGFR) are common in clinical practice. New eGFR equations without race have been developed. Little is known of the discrepancies between creatinine-based and cystatin C-based eGFR without race in the general population. Design, Setting, and Participants: We conducted both cross-sectional and prospective analyses of adults with serum creatinine and serum cystatin C available in the United States participating in the National Health and Nutrition Examination Survey (NHANES) from 1999 to 2004. Analyses were stratified by age categories. Exposures: eGFRdiff (eGFRcys – eGFRcr), was calculated based on the 2021 CKD-EPI cystatin C and creatinine equations both as a continuous variable and categorized into 3 groups: negative, <-15 mL/min/1.73m2 (eGFRcys lower than eGFRcr); reference, -15 to <15 mL/min/1.73m2 (eGFRcys similar to eGFRcr); and positive, ≥15 mL/min/1.73m2 (eGFRcys higher than eGFRcr). Outcomes and Measures: All-cause and cardiovascular deaths were linked to National Death Index through 31 December 2015. Results: Among 10,457 participants, mean eGFRcys was 106.8 (SE 0.5) mL/min/1.73m2, and mean eGFRcr was 95.8 (SE 0.4) mL/min/1.73m2, corresponding to an eGFRdiff of 11.0 (SE 0.4) mL/min/1.73m2. In adjusted linear regression models, older age (≥ 60 yrs), female, smoking, obesity, elevated C-reactive protein, elevated ACR, diabetes, hypertension, history of cardiovascular disease, and history of liver disease were associated with a lower eGFRcys compared to eGFRcr (eGFRdiff <0); non-Hispanic Blacks, drinking, and higher socioeconomic status were associated with a higher eGFRcys compared to eGFRcr (eGFRdiff >0). During a median follow-up of 13 years, 2135 participants died including 375 of cardiovascular causes. In adjusted Cox proportional hazards models, the negative eGFRdiff group was significantly associated with higher all-cause mortality (HR: 1.50; 95% CI: 1.23-1.84); positive eGFRdiff group was significantly associated with lower all-cause mortality (HR: 0.69; 95% CI: 0.60-0.80) and cardiovascular mortality (HR: 0.58; 95% CI: 0.41-0.81) compared to the reference group. Conclusion: Discrepancies between eGFR estimated from serum cystatin C and creatinine are common, with differences in magnitude across population subgroups. The discrepancies are associated with mortality, suggesting prognostic significance of these differences.