eGFR vs creatinine clearance: what each one tells you
eGFR and creatinine clearance both start from serum creatinine, but they are not interchangeable. eGFR is scaled to 1.73 m² and stages kidney disease; Cockcroft-Gault CrCl is an absolute figure in mL/min that many drug labels still dose on. Two worked cases show how far apart they can land.
At a glance
| eGFR (CKD-EPI 2021) | CrCl (Cockcroft-Gault) | |
|---|---|---|
| What it estimates | Glomerular filtration rate | Creatinine clearance (filtration plus some tubular secretion) |
| Unit | mL/min/1.73 m² | mL/min |
| Inputs | Creatinine, age, sex | Creatinine, age, sex, weight |
| Scaled to body size? | Yes, to a standard 1.73 m² | No, absolute value for this person |
| Main use | Staging and following chronic kidney disease | Renal dose bands in many product labels |
| Usually reported by | The laboratory, automatically | Calculated at the bedside or by pharmacy |
| Weak spots | Extremes of body size when used for dosing | Obesity, choice of weight, older adults |
Both are estimates built on serum creatinine, so both inherit its limits: creatinine depends on muscle mass and diet, and it only reflects kidney function once it has settled.
eGFR with CKD-EPI 2021
The CKD-EPI 2021 equation was refitted without a race coefficient. In Europe the EFLM advises laboratories to keep CKD-EPI 2009 for now, so it is worth checking which one your lab reports. It uses creatinine, age and sex only; there is no weight term. You can try it with the eGFR calculator.
The result is indexed: it is expressed per 1.73 m² of body surface area, roughly the surface of an average adult in the 1920s. Indexing makes results comparable between people of different sizes, which is exactly what you want for the G1–G5 categories of chronic kidney disease. It is less helpful when a drug is cleared by the kidney, because a larger body really does filter more millilitres per minute in absolute terms.
De-indexing eGFR
To turn eGFR into an absolute rate, multiply by the person’s body surface area and divide by 1.73:
eGFR (mL/min) = eGFR (mL/min/1.73 m²) × BSA ÷ 1.73
BSA can be estimated with the body surface area calculator (Mosteller by default). For someone close to 1.73 m² the change is small; for a very small or very large person it can be 20–45% or more.
Creatinine clearance with Cockcroft-Gault
Cockcroft and Gault published their equation in 1976, from a group of hospitalised men:
CrCl (mL/min) = (140 − age) × weight (kg) ÷ (72 × creatinine mg/dL), × 0.85 if female
Creatinine in µmol/L is divided by 88.4 first. The creatinine clearance calculator shows each step.
Because weight sits in the numerator, the result changes in direct proportion to the weight you enter. That is the equation’s main strength (it gives an absolute value) and its main problem (fat mass raises the result without producing more creatinine). Creatinine clearance also slightly overestimates true GFR, because a fraction of creatinine is secreted by the tubules.
Which weight?
- Actual (total) body weight was used in the original study and suits people near their ideal weight or below it.
- Ideal body weight (IBW calculator) tends to underestimate clearance in people with obesity.
- Adjusted body weight with a 0.4 factor (adjusted weight calculator) sits between the two; in one large comparison with measured 24-hour clearance, it was the least biased option in overweight and obese patients.
The guide on actual, ideal and adjusted body weight explains the three weights in more detail.
Same patient, different answers
Patient A is an 84-year-old woman, 50 kg, 158 cm, creatinine 1.0 mg/dL (88.4 µmol/L). Her ideal body weight is 50.6 kg, so the weight choice hardly matters.
| Estimate | Result |
|---|---|
| eGFR, CKD-EPI 2021 | 56 mL/min/1.73 m² (G3a) |
| BSA (Mosteller) | 1.48 m² |
| eGFR de-indexed (56 × 1.48 ÷ 1.73) | 48 mL/min |
| CrCl, Cockcroft-Gault, 50 kg | 33 mL/min |
The same creatinine produces 56, 48 and 33. A product label with a band at 50 mL/min or at 30–50 mL/min could put her in different categories depending on which figure is read. The gap comes from the (140 − age) term, which falls steeply in old age, and from her low weight.
Patient B is a 45-year-old man, 130 kg, 175 cm (BMI 42.4), creatinine 1.0 mg/dL.
| Estimate | Result |
|---|---|
| eGFR, CKD-EPI 2021 | 95 mL/min/1.73 m² (G1) |
| eGFR de-indexed (BSA 2.51 m²) | 137 mL/min |
| CrCl with actual weight, 130 kg | 172 mL/min |
| CrCl with adjusted weight, 94.3 kg | 124 mL/min |
| CrCl with ideal weight, 70.5 kg | 93 mL/min |
Here the weight entered into Cockcroft-Gault moves the answer from 93 to 172 mL/min. These are the numbers the site’s calculators return for the same inputs.
eGFR or creatinine clearance: which should you use?
- Classifying and following chronic kidney disease: eGFR. KDIGO 2024 staging is built on it, and the laboratory reports it automatically.
- Renal dose adjustment: it is advisable to use the measure the product information is written in. Many labels, including those of several direct oral anticoagulants, define their bands with Cockcroft-Gault CrCl.
- Regulators: the European Medicines Agency asks that renal dosing recommendations be based on an absolute measure in mL/min, not on GFR normalised to 1.73 m², and that labels state when creatinine clearance was the measure used.
- KDIGO 2024 accepts validated eGFR equations for drug dosing in most settings, suggests de-indexing when body size is far from average, and points to cystatin C or measured GFR when precision really matters, for example with narrow-therapeutic-index drugs.
The equations are aids to judgement; we recommend following local protocols and the product information for each drug.
Common pitfalls
- Treating mL/min/1.73 m² as mL/min. In a small person this overstates absolute clearance; in a large person it understates it.
- Plugging total weight into Cockcroft-Gault in obesity without checking what the protocol specifies.
- Rounding a low creatinine up (for example to 1.0 mg/dL in a frail older adult). It is a common habit, but the study by Winter and colleagues found it did not improve accuracy.
- Using either estimate during acute kidney injury. A rising creatinine has not caught up with the fall in filtration, so the estimate looks better than reality; on recovery, the opposite happens.
- Ignoring muscle mass. Amputation, paralysis, cachexia or very high muscle mass shift creatinine production and both estimates with it.
- Mixing creatinine units. 1.0 mg/dL is 88.4 µmol/L; entering µmol/L where mg/dL is expected makes Cockcroft-Gault about 88 times too small and CKD-EPI more than 200 times too small.
- Applying adult equations to children, where paediatric formulas are needed.
Frequently asked questions
Are eGFR and creatinine clearance the same thing?
No. eGFR estimates glomerular filtration rate scaled to a standard body surface area of 1.73 m², while Cockcroft-Gault estimates creatinine clearance in mL/min for the actual person, using their weight. They often give similar numbers in average-sized middle-aged adults and drift apart at the extremes.
Why do many drug labels use creatinine clearance?
Because most older drugs were studied when Cockcroft-Gault was the standard bedside estimate. Their renal dose bands in the product information are therefore written in CrCl, in mL/min. A label threshold only makes sense with the measure it was derived from, so it is advisable to check which one the label names before comparing numbers.
How do I convert eGFR from mL/min/1.73 m² to mL/min?
Multiply the eGFR by the person’s body surface area and divide by 1.73. For someone with a BSA of 2.0 m², an eGFR of 60 becomes about 69 mL/min; for a BSA of 1.5 m² it becomes about 52 mL/min. Close to 1.73 m², the change is small.
Which weight goes into the Cockcroft-Gault equation?
The original study used actual body weight, which still suits people near or below their ideal weight. In obesity, actual weight overestimates clearance, so many protocols specify ideal or adjusted body weight instead. In one large comparison, adjusted weight was the least biased. The right choice depends on the drug and its dosing guidance.
Can either equation be used when creatinine is changing quickly?
No, neither is reliable then. Both assume creatinine is in a steady state. During acute kidney injury or recovery, the serum value lags behind the true filtration rate. While creatinine is rising, the estimate looks better than reality; during recovery, it looks worse. Either way, the number can be far off.
Can I use these equations in children?
No. CKD-EPI 2021 and Cockcroft-Gault were developed in adults, and the site’s calculators show a warning for any age under 18. Children’s kidney function is estimated with paediatric equations such as the bedside Schwartz formula, which uses height instead of age and weight.
Sources
- Inker LA, et al. New creatinine- and cystatin C-based equations to estimate GFR without race. N Engl J Med 2021
- Cockcroft DW, Gault MH. Prediction of creatinine clearance from serum creatinine. Nephron 1976
- KDIGO 2024 Clinical Practice Guideline for the Evaluation and Management of Chronic Kidney Disease. Kidney Int 2024
- EMA. Guideline on the evaluation of the pharmacokinetics of medicinal products in patients with decreased renal function
- Winter MA, et al. Impact of various body weights and serum creatinine concentrations on the bias and accuracy of the Cockcroft-Gault equation. Pharmacotherapy 2012
- Delanaye P et al. The new, race-free, CKD-EPI equation to estimate glomerular filtration rate: is it applicable in Europe? A position statement by the EFLM. Clin Chem Lab Med 2023;61(1):44–47