Chronic kidney disease prevalence estimates among racial/ethnic groups: the Multi-Ethnic Study of Atherosclerosis

Abstract

Background and objectives: Muscle mass is not a major determinant of serum cystatin C levels, and its use to estimate GFR may lead to more congruent estimates of chronic kidney disease (CKD) across gender and racial/ethnic groups.

Design, setting, participants, & measurements: The Multi-Ethnic Study of Atherosclerosis is a population-based study of 6814 men and women who are aged 45 to 85 yr and do not have clinical cardiovascular disease. Estimated CKD prevalence, defined as an estimated GFR <60 ml/min per 1.73 m(2) body surface area, was compared using three different GFR prediction equations: The abbreviated Modification of Diet in Renal Disease (MDRD) equation and two equations based on serum cystatin C.

Results: Among women, CKD prevalence estimates across the four racial/ethnic groups using the MDRD- or the cystatin C-based GFR equations, which include gender and race coefficients, varied by approximately two-fold (P < 0.0001) but were more congruent with use of a serum cystatin C-based equation without the use of coefficients (P = 0.3). CKD prevalence estimates did not differ significantly across racial/ethnic groups among men with the MDRD (P = 0.07) or cystatin C formula without coefficients (P = 0.05) but did differ significantly with the cystatin C formula, which incorporates gender and race coefficients (P = 0.006).

Conclusions: CKD prevalence estimates vary across racial/ethnic groups, and the degree of variability depends on the method used to estimate GFR, especially among women. Further research is needed to determine the accuracy and precision of GFR prediction equations in racially diverse populations.

Figure 1.

Distribution of serum creatinine (mg/dl) and cystatin C (mg/L) concentrations among men and women and by racial/ethnic group.

Figure 2.

Unadjusted chronic kidney disease (CKD) prevalence estimates by GFR prediction equation and by racial/ethnic group among men (top) and women (bottom). *P < 0.008 versus white men (if male) or white women (if female).

Figure 3.

Unadjusted and adjusted odds ratios of CKD by gender, race/ethnicity, and GFR prediction equation compared with white individuals. *Adjusted for age, race/ethnicity, body mass index, presence of diabetes, hypertension, and current smoking.