Glomerular filtration rate

Definition and the trajectory with age. GFR is the volume of plasma filtered by the glomeruli per unit time, expressed as mL/min/1.73 m². It peaks at approximately 140 mL/min/1.73 m² in the third decade and then declines at about 8 mL/min/1.73 m² per decade. Effective renal plasma flow falls proportionally more — about 10% per decade, from 600 mL/min in youth to 300 mL/min by age 80 — so the filtration fraction (GFR/ERPF) actually increases in the elderly. Importantly, in approximately one-third of older adults GFR does not decrease with age at all.

Why serum creatinine misleads in old age. The fall in creatinine clearance is accompanied by a concomitant decrease in creatinine production owing to senile sarcopenia, so serum creatinine does not increase despite the progressive decrease in GFR. Healthy older people additionally show net creatinine tubular reabsorption, so measured creatinine clearance underestimates true glomerular filtration. The consequence to quote: a serum creatinine of 1 mg/dL reflects a GFR of 120 mL/min in a 20-year-old but only 60 mL/min in an 80-year-old.

The estimating equations. Cockcroft–Gault: creatinine clearance = (140 − age) × body weight ÷ (72 × serum creatinine), and 15% lower in women. MDRD and CKD-EPI are the two other standard formulae. Regarding cystatin C as a GFR marker, it has been demonstrated that in elderly people with glomerular filtration below 60 mL/min it is not superior to that calculated by the Cockcroft–Gault and MDRD formulae. Keller's simple approximation is GFR = 130 − age (mL/min) for people between 25 and 100 years, though this has not been validated against cimetidine creatinine clearance.

Choosing the right equation — a favourite examination point. Cockcroft–Gault underestimates GFR in people over 80, while MDRD has the advantage of not requiring the patient's weight. CKD-EPI tends to be the most accurate choice at higher levels of GFR and is recommended for estimating creatinine clearance in adults without known chronic kidney disease, and for risk stratification before percutaneous coronary intervention. However, when dosing medications based on a patient's GFR, the Cockcroft–Gault equation remains the standard of care and should be used.

Consequences of senile hypofiltration. (a) Senile hypofiltration and diastolic cardiac failure predispose healthy old people to cardiac failure and to lung congestion after a saline load. (b) The dose of every prescribed drug must be adjusted to the estimated GFR in the elderly. (c) Senile hypofiltration differs from chronic kidney disease because the glomerular filtration value is stable over 6–12 months and there is no haematuria and no significant proteinuria (>0.30 g/day) — a distinction that prevents the over-labelling of healthy older people as having CKD.

KDIGO 2024 CKD guideline — combined creatinine–cystatin C estimation becomes the preferred standard. The guideline recommends using a validated GFR estimating equation to derive GFR from serum filtration markers rather than relying on the markers alone; that in adults at risk of CKD, creatinine-based eGFR should be used, and if cystatin C is available the GFR category should be estimated from the combination of creatinine and cystatin C (eGFRcr-cys); and specifically that eGFRcr-cys should be used in clinical situations where creatinine-based eGFR is less accurate and GFR affects clinical decision-making — which describes the sarcopenic, malnourished, amputee, cirrhotic or very old patient exactly. The equations are now race-free. Two further points of practical relevance: point-of-care testing for creatinine and urine albumin may be used where laboratory access is limited — directly applicable to rural India — and measured GFR using an exogenous filtration marker remains the reference standard when a more accurate assessment is needed for clinical decision-making.