How to Calculate eGFR With the CKD-EPI 2021 Equation
eGFR (estimated glomerular filtration rate) is a number that tells you how much blood your kidneys filter per minute, scaled to a standard body size. This tool uses the CKD-EPI 2021 creatinine equation, the current race-free standard, and turns three inputs (age, sex, serum creatinine) into a value in mL/min/1.73m² plus a KDIGO stage from G1 to G5.
What the equation needs, and why race is gone
CKD-EPI 2021 takes only three variables: age in years, biological sex, and serum creatinine in mg/dL. If your lab reports creatinine in µmol/L (common outside the US), divide by 88.4 to convert, or just pick that unit in the calculator below and let it do the conversion.
The full formula:
eGFR = 142 × min(Scr/κ, 1)^α × max(Scr/κ, 1)^-1.200 × 0.9938^Age × (1.012 if female)κ (kappa) is 0.7 for women and 0.9 for men. α (alpha) is -0.241 for women and -0.302 for men. Scr is serum creatinine in mg/dL. The min and max terms split the curve at the κ threshold, which is why the equation looks piecewise rather than a single clean exponent.
That split exists because the relationship between creatinine and filtration rate is not linear across the whole range. Below the κ threshold, a small rise in creatinine barely moves the estimate; above it, the same rise drags the result down faster. Two patients half a point apart in creatinine can land in different eGFR bands depending on which side of κ they sit on, which is one reason the raw creatinine number alone is a poor substitute for running it through the equation.
Earlier versions, the 2009 CKD-EPI equation and the older MDRD formula, multiplied the result by a fixed coefficient (1.159) for patients coded as Black, based on population-average muscle mass and creatinine generation assumptions. The 2021 refit, developed with the National Kidney Foundation and American Society of Nephrology, removed that coefficient. Race is not a reliable proxy for muscle mass or creatinine generation at the individual level, and the adjustment is no longer recommended in US clinical guidance. That is the whole reason this version needs only age, sex, and creatinine: sex stays in the formula because the underlying biology (muscle mass distribution, creatinine generation rate) differs by sex on average, but race was never a direct biological input in the first place.
Worked example
Take a 65-year-old man with a serum creatinine of 1.4 mg/dL. His ratio to κ (0.9 for males) is 1.556, above 1, so the equation uses the -1.200 exponent branch and the male α value of -0.302.
| Age | Creatinine | Sex | eGFR | Stage |
|---|---|---|---|---|
| 65 | 1.4 mg/dL | Male | 55.8 mL/min/1.73m² | G3a |
An eGFR of 55.8 falls in the 45 to 59 range, KDIGO stage G3a: mildly to moderately decreased kidney function. On its own this is a data point, not a diagnosis; more on that below.
Sex changes the result even at identical age and creatinine, because κ, α, and the 1.012 multiplier are all sex-specific. A 60-year-old with creatinine of 1.1 mg/dL scores an eGFR of 76.9 (stage G2) as a man, but 57.5 (stage G3a) as a woman, same age, same lab value. That gap is the reason sex stayed in the equation after race was dropped: it reflects an average physiological difference in creatinine generation, not a demographic label.
Calculate it with your own values
If your lab reports creatinine in µmol/L, the calculator handles the conversion for you. A 55-year-old woman with creatinine of 120 µmol/L (about 1.36 mg/dL after dividing by 88.4) comes out to an eGFR of 46.1, also stage G3a.
KDIGO stages
KDIGO (Kidney Disease: Improving Global Outcomes) splits eGFR into six bands. Stage 3 is further split into 3a and 3b because outcomes and monitoring differ meaningfully within that range.
| Stage | eGFR (mL/min/1.73m²) | Meaning |
|---|---|---|
| G1 | ≥ 90 | Normal or high |
| G2 | 60 to 89 | Mildly decreased |
| G3a | 45 to 59 | Mildly to moderately decreased |
| G3b | 30 to 44 | Moderately to severely decreased |
| G4 | 15 to 29 | Severely decreased |
| G5 | < 15 | Kidney failure |
For contrast with the G3a examples above: a 40-year-old man with creatinine of 1.0 mg/dL lands at eGFR 97.6, comfortably in G1. A 72-year-old man with creatinine of 3.5 mg/dL lands at 17.8, in G4, severely decreased.
The stage itself has practical weight beyond a label. It feeds decisions like how often to recheck kidney function, when to adjust the dose of a renally cleared drug, and when a referral to nephrology makes sense. G1 and G2 are usually monitored in primary care; G4 and G5 typically involve a specialist and planning around dialysis or transplant options if the trend keeps declining. None of that is decided from a single eGFR figure in isolation, which is the point of the next section.
Common mistakes and limitations
The equation assumes stable, steady-state creatinine. In acute kidney injury, creatinine is rising or falling and has not equilibrated with the current filtration rate, so a single eGFR calculated during that window is unreliable. The same goes for people at either extreme of muscle mass: creatinine is a byproduct of muscle metabolism, so very low muscle mass (frailty, amputation, long-term immobility) inflates the estimate, and very high muscle mass (bodybuilders) deflates it. The equation is not validated for pregnancy, where kidney physiology changes substantially, or for children, who need a pediatric formula such as the Schwartz equation instead.
The biggest misread is treating one low eGFR result as a CKD diagnosis. KDIGO defines chronic kidney disease as abnormalities in kidney structure or function present for more than three months, which means the eGFR reduction (or albuminuria, or both) has to persist, not just show up once. Full CKD staging also brings in the urine albumin-to-creatinine ratio (ACR) alongside eGFR: the two together, not eGFR alone, place a patient in the KDIGO risk grid. A single low number is a prompt to retest and investigate, not a verdict.
Frequently asked questions
What is a normal eGFR?
90 or above is KDIGO stage G1, normal or high. Values naturally decline somewhat with age even in healthy kidneys, which is part of why age is a direct input in the equation.
Why doesn’t the CKD-EPI 2021 equation use race anymore?
The earlier 2009 CKD-EPI and MDRD equations applied a fixed multiplier for patients coded as Black, based on population-level assumptions about muscle mass and creatinine generation. That adjustment is not a reliable individual-level marker, and the 2021 refit, backed by the NKF-ASN task force, removed it. The equation now runs on age, sex, and creatinine alone.
Does a low eGFR always mean chronic kidney disease?
No. CKD requires the abnormality to persist beyond three months, and staging also uses the urine albumin-to-creatinine ratio (ACR) alongside eGFR. A single low reading, especially during acute illness, can reflect a temporary drop in creatinine clearance rather than chronic disease. Retesting and clinical context decide the diagnosis.
Can I use this calculator for a child?
No. CKD-EPI 2021 is built and validated for adults. Pediatric kidney function uses different equations, such as the Schwartz equation, which account for a child’s height and growth rather than the adult muscle-mass assumptions baked into CKD-EPI.