Creatinine and cystatin C-based estimated glomerular filtration rate estimates of kidney function in Black people with HIV on antiretroviral therapy

Abstract

Background: To reduce health inequalities, the creatinine-based chronic kidney disease epidemiology collaboration 2021 formula for estimated glomerular filtration rate (eGFR) is replacing the 2009 formula, which required adjustment specifically for Black individuals. We compared the 2021 and 2009 creatinine-based formulae with cystatin C-based eGFR in Black people on antiretroviral therapy (ART) with HIV RNA <200 c/ml. Methods: Cross-sectional analysis of paired serum creatinine and cystatin C measurements. Bias, imprecision, accuracy, and performance for identifying individuals with eGFR cystatin C <60 (units: ml/min per 1.73 m2) were determined. The effects of ART with no, mild-moderate, or marked effect on tubular creatinine secretion on the performance of the 2021 formula was assessed. Results: We included 362 individuals (mean age 51 years, 56% female, mean eGFR-cystatin C 88.3). Overall, the 2021 (vs. the 2009 race-adjusted) formula was less biased and had improved imprecision and accuracy compared with eGFR-cystatin C but underestimated eGFR-cystatin C in those with eGFR ≥90 and overestimated eGFR-cystatin C in those with eGFR <60. The 2021 (vs. the 2009) formula had high specificity (95% vs. 97%) and negative predictive value (97% vs. 96%), but low sensitivity (56% vs. 52%) and positive predictive value (44% vs. 54%) for identifying individuals with eGFR-cystatin C <60 (P > 0.25). Performance at the eGFR <60 cut-off was minimally affected by ART exposure group. Conclusion: The CKD-EPI 2021 creatinine-based formula was better aligned with eGFR-cystatin C than the 2009 formula. eGFR-cystatin C may provide clinically useful information in Black people with eGFR <60 irrespective of ART regimen.