Creatinine Clearance vs eGFR: Which One Should You Use?
The same patient can have an eGFR of 51 and a creatinine clearance of 35. Both are correct. Here is why the two numbers diverge, which one drug labels actually ask for, and when the difference changes the dose you give.
Creatinine Clearance vs eGFR: Two Different Measurements
Creatinine clearance vs eGFR is not a question of which lab is right, because the two values answer different questions. Creatinine clearance estimates how much plasma the kidneys strip of creatinine each minute, reported in mL/min for that specific body. eGFR estimates glomerular filtration and is reported per 1.73 m² of body surface area, which is the surface area of a notional average adult from a 1927 paper. One is a measurement about a person. The other is a measurement about a person scaled to a standard body.
That scaling is deliberate and useful. It lets a nephrologist compare a 48 kg woman and a 110 kg man on the same chronic kidney disease staging chart. It is also the reason a pharmacist checking a dose threshold written in mL/min cannot use the lab value straight off the chart.
Why creatinine clearance vs eGFR diverge on the same blood draw
There is a second, quieter difference. The renal tubules secrete creatinine on top of filtering it, so creatinine clearance runs roughly 10 to 20 percent above true GFR, and the gap widens as function falls. Neither equation corrects for this. They were simply fitted to different reference standards, which is why they disagree even after you account for body size.
Cockcroft-Gault vs CKD-EPI: The Equations Side by Side
Cockcroft-Gault was published in 1976 and fitted to 249 men using 24-hour urine collections. CKD-EPI was developed in 2009 and refit in 2021 to remove its race coefficient. They take different inputs, and that alone explains part of the divergence.
| Feature | Cockcroft-Gault (CrCl) | CKD-EPI 2021 (eGFR) |
|---|---|---|
| Published | 1976 | 2009, refit 2021 |
| Units | mL/min | mL/min/1.73 m² |
| Uses weight | Yes — directly proportional | No |
| Uses race | No | Not since 2021 |
| Estimates | Creatinine clearance | Glomerular filtration rate |
| Primary use | Renal drug dosing | CKD staging and referral |
The weight row is the one that bites. Cockcroft-Gault is linear in weight, so it treats a 120 kg patient as clearing roughly twice as much as a 60 kg patient at the same creatinine. CKD-EPI ignores weight entirely. Run both through our Cockcroft-Gault calculator and our eGFR calculator on the same patient and the spread is usually obvious immediately.
One Patient, Three Different Numbers
Take a 78-year-old woman, 52 kg, 160 cm, serum creatinine 1.1 mg/dL. Nothing about her is unusual. Here is what each method reports.
Cockcroft-Gault: CrCl = [(140 − 78) × 52] / (72 × 1.1) × 0.85 = (62 × 52) / 79.2 × 0.85 = 3,224 / 79.2 = 40.7, then × 0.85 = 34.6 mL/min.
CKD-EPI 2021: the same creatinine returns roughly 51 mL/min/1.73 m².
De-indexed CKD-EPI: her body surface area is about 1.53 m², so 51 × 1.53 / 1.73 = 45 mL/min in absolute terms.
What the spread means at a dosing threshold
Three numbers — 35, 45, and 51 — from one blood draw, and all three are calculated correctly. Now suppose she needs a drug that requires dose reduction below 30 mL/min and is contraindicated below 15. At 51 she looks comfortably safe. At 34.6 she is four points from a dose change and one dehydrating illness away from a contraindication. That spread is the entire argument for knowing which equation the label meant.
The 1.73 m² Trap That Sends Large Patients the Wrong Dose
The de-indexing error runs in the opposite direction for big patients, and it is the more dangerous one because it causes underdosing of drugs that need to work.
A 110 kg, 185 cm patient has a body surface area near 2.33 m². An eGFR reported as 55 mL/min/1.73 m² is really 55 × 2.33 / 1.73 = 74 mL/min of actual filtration. If a clinician reads 55 and applies a moderate-impairment dose reduction, that patient receives less drug than a normally filtering body is clearing. For a renally cleared antibiotic, that is a subtherapeutic trough. Our creatinine clearance calculator reports in mL/min directly, which sidesteps the conversion entirely.
The conversion is simple enough to do at the bedside: eGFR × BSA / 1.73. The problem is not difficulty, it is that the units on a lab report are easy to skim past.
Which Number Does the Drug Label Actually Want?
Here is the rule that resolves most arguments: use the equation the label was validated against, because the threshold and the equation were approved as a pair. A cutoff of “CrCl below 30” means 30 by Cockcroft-Gault, not 30 by whatever your lab happens to print.
| Drug or setting | Use | Why |
|---|---|---|
| Dabigatran, rivaroxaban, edoxaban | CrCl (Cockcroft-Gault) | Trial dose bands were defined in CrCl |
| Vancomycin, aminoglycosides | CrCl (Cockcroft-Gault) | Pharmacokinetic models are built on CrCl |
| Metformin | eGFR | FDA relabeled from creatinine to eGFR in 2016 |
| CKD staging, nephrology referral | eGFR | KDIGO stages are defined per 1.73 m² |
| Carboplatin (Calvert formula) | Absolute GFR in mL/min | Dose is AUC × (GFR + 25) |
When the label is silent or you are working from an older reference, Cockcroft-Gault is the conservative default for dosing because it usually returns the lower number in exactly the patients who are most vulnerable: elderly, low weight, low muscle mass. For age-specific adjustments, our geriatric dose calculator layers renal changes on top of the age-related pharmacokinetic shifts.
When Both Equations Break Down
Every creatinine-based estimate assumes steady state and average muscle mass. Break either assumption and both equations fail together.
- Acute kidney injury. Creatinine lags the real drop in filtration by 24 to 72 hours. A patient whose GFR collapsed this morning still has this morning's creatinine.
- Very low muscle mass. Amputation, spinal cord injury, cirrhosis, or advanced frailty produce little creatinine. A creatinine of 0.5 mg/dL in a cachectic patient does not mean excellent kidneys.
- Obesity. Cockcroft-Gault with actual body weight above roughly BMI 30 inflates clearance, because fat mass adds weight but almost no creatinine production. Adjusted body weight of IBW + 0.4 × (actual − IBW) is the usual correction.
- Pregnancy. GFR rises 40 to 50 percent by the second trimester and neither equation was validated in pregnant patients.
- Extremes of age. Cockcroft-Gault returns nonsense below age 18, and its (140 − age) term becomes crude past 90.
When to reach for cystatin C instead
In these cases cystatin C is the usual escape hatch, since it is produced by all nucleated cells and is largely independent of muscle mass. A combined creatinine-cystatin C equation outperforms either marker alone. Our kidney function calculator puts the standard markers in one place for a broader picture.
What to Do With Your Number
If you are a patient reading a lab report, the practical takeaway is narrow: an eGFR between 60 and 89 with no protein in the urine and no other abnormality is not chronic kidney disease, and a single value never establishes a diagnosis, since CKD staging requires the abnormality to persist for at least three months.
If you are dosing a renally cleared drug
If you are dosing a drug, do three things. Check which unit the threshold uses. If it is mL/min and your lab reported mL/min/1.73 m², convert. And when the two equations straddle a threshold, as they did for the 78-year-old above, dose to the more conservative number and recheck the creatinine sooner than you otherwise would. Once you have the clearance figure, our mg to mL dose calculator converts the adjusted dose into an actual volume to draw up.
The equations disagree because they were built for different jobs. Knowing which job you are doing is most of the answer.
References
- National Kidney Foundation. GFR calculators and the 2021 CKD-EPI creatinine equation.
- U.S. Food and Drug Administration. Pharmacokinetics in Patients with Impaired Renal Function: Study Design, Data Analysis, and Impact on Dosing.
Frequently Asked Questions
Why is my eGFR higher than my creatinine clearance?
Two effects usually stack. eGFR is reported per 1.73 m² of body surface area, so a small person gets a number that is higher than their actual filtration in mL/min. On top of that, CKD-EPI and Cockcroft-Gault were fitted to different reference measurements. In a 52 kg woman with a body surface area of 1.53 m², an eGFR of 51 mL/min/1.73 m² is only about 45 mL/min in absolute terms, and Cockcroft-Gault may report 35 mL/min for the same blood draw.
Should I use Cockcroft-Gault or CKD-EPI for drug dosing?
Use whichever equation the drug label specifies, because the dose thresholds were validated against that equation. Most renally dosed drugs approved before roughly 2010 (dabigatran, rivaroxaban, gabapentin, most aminoglycosides) cite creatinine clearance by Cockcroft-Gault. Newer labels increasingly cite eGFR. Metformin is the clearest example of a label that switched: the FDA moved it from serum creatinine to eGFR in 2016.
How do I convert eGFR to mL/min?
Multiply the reported eGFR by the patient body surface area and divide by 1.73. For a patient with a BSA of 2.2 m², an eGFR of 60 mL/min/1.73 m² becomes 60 x 2.2 / 1.73, or roughly 76 mL/min. That de-indexed value is what you compare against a label threshold written in mL/min. Skipping this step is the single most common renal dosing error in large patients.
Is creatinine clearance the same as GFR?
No. The renal tubules secrete creatinine in addition to filtering it, so measured creatinine clearance typically overestimates true GFR by about 10 to 20 percent. The gap widens as kidney function falls, because secretion makes up a larger share of total creatinine elimination. This is why an iohexol or iothalamate clearance study, not a creatinine-based estimate, is used when an exact GFR matters.
What weight should I use in the Cockcroft-Gault equation?
Actual body weight works for patients near their ideal weight. Once BMI passes roughly 30, actual weight inflates the estimate because adipose tissue produces little creatinine, so most clinicians switch to ideal body weight or an adjusted body weight of IBW + 0.4 x (actual - IBW). For underweight patients, using actual weight is generally safer than ideal weight because it produces the lower, more conservative clearance.
At what eGFR should metformin be stopped?
The FDA label contraindicates metformin below an eGFR of 30 mL/min/1.73 m² and advises against starting it between 30 and 45. Patients already stable on metformin who drift into the 30 to 45 band are usually continued at a reduced dose with more frequent monitoring rather than stopped outright. Metformin is deliberately labeled in eGFR, not creatinine clearance.
Why did CKD-EPI remove the race coefficient in 2021?
The original 2009 CKD-EPI equation multiplied results by about 1.16 for Black patients, which systematically reported higher kidney function and could delay transplant referral and specialist care. A joint National Kidney Foundation and American Society of Nephrology task force recommended a race-free equation, and the 2021 refit removed the coefficient. Labs that switched will report slightly different values for the same creatinine than they did before.
Can I use Cockcroft-Gault during acute kidney injury?
No. Every creatinine-based equation assumes steady state, meaning creatinine production and elimination are balanced. In acute kidney injury creatinine lags the true drop in filtration by 24 to 72 hours, so an equation fed a still-normal creatinine will report a reassuring clearance for a patient who is barely filtering. Dose on the clinical trajectory and rising creatinine trend instead of a single calculated number.
Written by Jurica Šinko
Founder & CEO
Entrepreneur and health information advocate, passionate about making health calculations accessible to everyone through intuitive digital tools.
View full profile