Comparison of cross-sectional renal function measurements in African Americans with hypertensive nephrosclerosis and of primary formulas to estimate glomerular filtration rate

Abstract

Renal function measurements were obtained in 1,703 African Americans with presumed hypertensive nephrosclerosis who were screened for entry into the African-American Study of Hypertension and Kidney Disease (AASK). We examined the effect of race on relationships involving renal variables by comparing African Americans enrolled into the AASK with non-African Americans enrolled into the Modification of Diet in Renal Disease (MDRD) study. We examined the effect of gender on renal variables by comparing African American men and women. We compared various methods for estimating glomerular filtration rate (GFR) with iodine 125-labeled ((125)I)-iothalamate GFR. AASK data were also used to derive a new formula for estimating GFR in African Americans. After adjusting for age, sex, and baseline GFR, African American patients on the AASK study were heavier and had larger body surface areas and body mass indices than either MDRD African Americans or non-African Americans. African Americans had greater serum creatinine levels and urinary creatinine excretions for any given level of GFR. Mean GFR was greater in African American men than African American women (59.7 versus 51.7 mL/min/1.73 m(2)), although serum creatinine levels were also greater in men (1.91 versus 1.73 mg/dL). Seventy-eight percent of women with serum creatinine levels between 1.2 and 1.5 mg/dL had GFRs less than 65 mL/min/1.73 m(2). For African Americans in the AASK, GFR was overestimated by the 24-hour creatinine clearance and underestimated by the Cockcroft-Gault formula. A prediction formula developed in the MDRD study more accurately predicted GFR in AASK patients than these measurements. AASK data were also used to derive a new five-term formula for estimating GFR that was slightly more accurate in the African Americans in the AASK than the MDRD formula (median percentage of error, 12.4% for the MDRD formula versus 12.1% for the AASK formula). Important differences exist in renal variables between African Americans and non-African Americans and between African American men and African American women. Formulas using demographic data and readily measured serum values estimate (125)I-iothalamate GFR.