Dtsch Arztebl Int. 2009 Dec;106(51-52):849-54.

BACKGROUND: Chronic renal disease is common, and its prevalence is rising. Its main causes are hypertension and diabetes mellitus. An abnormally low glomerular filtration rate (GFR) often escapes medical notice in the earliest, most treatable stage, so that an increasing number of patients progress to end-stage renal failure. Early recognition of low GFR would thus be an important clinical advance. METHODS: The authors selectively review the literature retrieved by a PubMed search on the topic and also present their own clinical and laboratory data. RESULTS: Chronic renal failure can be detected early by direct measurement of the GFR with the aid of an exogenous filtration marker. Such techniques are costly and time-consuming and are therefore indicated only for patients at special risk. Chronic renal disease can also be diagnosed early with the aid of the endogenous filtration markers creatinine and cystatin C, which serve as indicators of a low GFR. The serum levels of these two substances are not taken as measures of GFR in themselves, but are rather entered into predictive equations for the estimation of GFR. Cystatin C-based equations seem to be more sensitive indicators of low GFR than creatinine-based equations. CONCLUSIONS: Creatinine- and cystatin C-based equations for the estimation of GFR are valuable tools for the early diagnosis of chronic renal disease and for disease staging according to the US National Kidney Foundation criteria.

Key Messages
•The diagnostic sensitivity of serum creatinine determination is too low for early detection of CRD.
•In addition to measurement of serum creatinine the MDRD equation should be used to calculate eGFR, allowing early diagnosis of CRD.
•In individuals without CRD the MDRD equation underestimates GFR, but in CRD the agreement is acceptable.
•Reductions in GFR are detected earlier by means of cystatin C and cystatin C-based eGFR than by serum creatinine. Because of the higher costs, however, cystatin C determination should be requested only in particular indications.
•When a reduction in eGFR is found, direct measurement of GFR (mGFR) should be used to establish the exact base value of GFR and assess the progression of CRD.

J Nephrol. 2010 Mar-Apr;23(2):125-32.

Kidney function should be evaluated by procedures including the calculation of glomerular filtration rate (GFR) estimates and the assessment of albuminuria or proteinuria as creatinine-normalized urinary ratios for albumin or total protein. GFR estimates are an approximation of true GFR, which circumvent the limitations of serum creatinine and creatinine clearance without increasing costs and time of diagnostic work-up. Estimates by Cockcroft-Gault equation tend to be higher than true GFR and estimates by other equations, because this equation predicts creatinine clearance, hence true GFR plus creatinine excretion via tubular secretion. The inclusion of a weight coefficient in the equation causes a GFR overestimation in the presence of large adiposity or edema. Estimates by equations of the Modification of Diet in Renal Disease (MDRD) study can be unreliable for high-normal GFR because that study did not enroll individuals without kidney disease. The Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) group has reported a new equation to overcome this limitation. GFR estimates can be biased by interassay creatinine differences or unusual levels of creatinine generation (muscle mass) or of renal tubular creatinine secretion. The urinary ratio of albumin (or total protein) to creatinine is measurable in untimed spot urine and reflects the urinary excretion rate of albumin (or total protein). Low muscle mass could imply borderline elevation in the ratio merely because of low urinary creatinine. Vice versa, high muscle mass could imply normal ratios even in the presence of high urinary albumin, because of high urinary creatinine due to high creatinine generation.