Abstract:
With the increasing recognition of the several pitfalls of serum creatinine (SCr) and creatinine-based glomerular filtration rate (GFR) estimates as measures of kidney function, a novel biomarker, cystatin C (Cys C) which is claimed to be superior to the creatinine-based estimates has been proposed. However, this claim is yet to be validated in an indigenous black setting. This study was therefore, to verify this claim and to assess the best method to estimate GFR in a typical African black population. Three hundred (150 males and 150 females) apparently healthy black subjects, aged 20 to 82 years, in Enugu metropolis, Enugu state were used for the study, which was conducted between April and August 2010. Demographic and anthropometric measurements were taken, fasting blood samples were collected and analyzed for urea, albumin, creatinine and cystatin C using standard procedures. GFR was estimated using six-published GFR equations —the Filler’s Cys C, standard Cockcroft and Gault (S-CG), body surface area (BSA)-modified CG (B-CG) , original Modification of Diet in Renal Disease (MDRD1) study, abbreviated MDRD (MDRD2) study, and the Mayo Clinic Quadratic (MCQ) formulae. The performance of the SCr-based GFR equations relative to the Filler’s equation was evaluated using four modalities — bias, precision, accuracy and correlation. The non-parametric reference intervals (RI) for Cys C were 0.65 – 1.12 mg/L (median 0.86 mg/L) in males, 0.62 – 1.12 mg/L in females (median 0.85 mg/L) and 0.64 – 1.12 mg/L (median 0.86 mg/L) in all the subjects. The non-parametric RI for SCr were 73 – 110 µmol/L (median 89 µmol/L) in males, 65 -102 µmol/L (median 82 µmol/L) in females and 66 – 106 µmol/L (median 86 µmol/L) in all the subjects. There was no significant difference between males and females in the Cys C levels and reference intervals (p > 0.05). Overall, body indices showed an independent association with Cys C levels, except for a modest association with body mass index (r = 0.15; p < 0.05), whereas SCr levels were strongly associated with gender and body indices. Cys C- based GFR equation appeared to estimate GFR best with a median GFR of 110 ml/min/1.73m2 (p < 0.0001), whereas the CG formula appeared to perform best amongst the SCr-based GFR equations, and the MDRD2 study equation was the worst performer. Our results suggest but cannot prove that Cys C may be a better marker of kidney function than creatinine, and that the CG equation remains the best SCr-based estimate of GFR in our setting. More work however, is recommended to address other reported limitations of Cys C, and a GFR equation specially tailored to the needs of the indigenous black population is urgently needed.
