CrCl vs eGFR: What's the Difference?

A laboratory report says eGFR 58. A drug label says "reduce the dose if creatinine clearance is below 50 mL/min." Are you below the threshold or not?

This question comes up constantly, and the honest answer is that you cannot tell from the eGFR alone. The two numbers measure related things in different units for different purposes.

The short version

| | Creatinine clearance (CrCl) | Estimated GFR (eGFR) | | --- | --- | --- | | Units | mL/min | mL/min/1.73 m² | | Indexed to body size? | No | Yes | | Typical equation | Cockcroft-Gault | CKD-EPI 2021 | | Primary use | Renal drug dosing | CKD staging and diagnosis | | Relationship to true GFR | Overestimates (tubular secretion) | Estimates directly | | Needs body weight? | Yes | No |

Why eGFR is indexed and CrCl is not

Glomerular filtration rate scales with body size. A 110 kg adult filters more plasma per minute than a 45 kg adult with identical kidney health, simply because they have more kidney.

For diagnosis and staging, that is noise. You want to know whether this person's kidneys are working well for a person, so you normalise to a standard body surface area of 1.73 m². That is what the "/1.73 m²" means, and it is what makes an eGFR of 58 mean the same thing in both patients.

For drug dosing, that normalisation is exactly wrong. A drug is cleared by the actual kidneys of the actual patient, at their actual size. What you want is absolute clearance in mL/min — not a size-adjusted comparison figure.

This is the crux. eGFR is indexed because comparison is the point. CrCl is not indexed because dosing is the point.

Converting between them

The conversion is straightforward arithmetic:

eGFR → absolute (mL/min) :  eGFR × BSA / 1.73
CrCl → indexed (/1.73 m²):  CrCl × 1.73 / BSA

BSA (Du Bois) = 0.007184 × height(cm)^0.725 × weight(kg)^0.425

Worked through for a 68-year-old man, 75 kg, 178 cm, creatinine 1.1 mg/dL:

BSA                      = 1.93 m²
eGFR (CKD-EPI 2021)      = 73.1 mL/min/1.73 m²
De-indexed               = 73.1 × 1.93 / 1.73 = 81.4 mL/min
CrCl (Cockcroft-Gault)   = 66.5 mL/min

De-indexing moved the eGFR figure by 8.3 mL/min. It still does not equal the Cockcroft-Gault result — a 15 mL/min gap remains.

The eGFR and CrCl conversion calculator does all of this at once, showing both figures and the conversion.

Why they still disagree after conversion

De-indexing aligns the units. It does not make the numbers equivalent, for two reasons.

Different equations, different populations. Cockcroft-Gault was fitted to 249 patients in 1976. CKD-EPI 2021 was fitted to a far larger, more recent and more diverse dataset. They were built to predict slightly different things and will not agree.

Creatinine clearance is not GFR. Creatinine is filtered at the glomerulus and actively secreted by the proximal tubule. Clearance therefore counts both, and exceeds true GFR — typically by 10–20%, and by more as kidney function declines. eGFR equations are calibrated to estimate GFR itself.

So creatinine clearance is systematically the larger number, by design. That is not an error in either equation.

Which should you use?

Use eGFR when you are asking about kidney disease. Staging CKD, tracking progression, deciding on nephrology referral, applying KDIGO categories. The 2021 CKD-EPI equations are current standard, and they are race-free — the race coefficient in older equations was removed on NKF-ASN Task Force recommendation.

Use CrCl when you are asking about a dose. Particularly when the product label specifies creatinine clearance, which most do. The pharmacokinetic studies behind those thresholds used Cockcroft-Gault, so matching the measure matters.

Use cystatin C when creatinine is unreliable. Amputees, cachexia, paraplegia, very high muscle mass, creatine supplementation. Cystatin C is produced by all nucleated cells rather than muscle, so it sidesteps the muscle-mass problem entirely. The CKD-EPI 2021 combined creatinine–cystatin C equation is the most accurate of the three.

The mistake to avoid

The most common error is applying a threshold from one measure to a number from the other without conversion.

Reading eGFR 58 and concluding "below 60, reduce the dose" is wrong twice over: the label threshold was CrCl, and 58 mL/min/1.73 m² is not 58 mL/min unless the patient happens to have a BSA of exactly 1.73 m².

The error runs in a predictable direction. In a large patient, eGFR understates absolute clearance — you risk underdosing. In a small, elderly patient, eGFR overstates absolute clearance — you risk overdosing, in exactly the population least able to tolerate it.

Practical summary

1. Check which measure the drug label specifies. Usually creatinine clearance.
2. If it specifies CrCl, calculate CrCl. Do not substitute the laboratory eGFR.
3. If you must convert, de-index by BSA — and know the two figures will still differ.
4. Use eGFR for staging, CrCl for dosing.
5. At the extremes of body size, the gap between the two is largest and matters most.

Both figures are available side by side on the eGFR and CrCl calculator, along with the BSA conversion and the KDIGO category.

Where the two diverge most — a worked comparison

The size of the disagreement depends almost entirely on how far the patient is from an average body.

| Patient | BSA | eGFR (mL/min/1.73 m²) | De-indexed (mL/min) | Difference | | --- | --- | --- | --- | --- | | Large adult, 110 kg | 2.30 m² | 55 | 73 | +18 | | Average adult, 75 kg | 1.93 m² | 55 | 61 | +6 | | Standard reference | 1.73 m² | 55 | 55 | 0 | | Small elderly woman, 45 kg | 1.40 m² | 55 | 45 | −10 |

Three patients with an identical laboratory eGFR of 55, and a 28 mL/min spread in absolute clearance. Two of them sit either side of the 50 mL/min threshold used for direct oral anticoagulant dose reduction — from the same number on the same report.

Note the direction of the error. In a large patient, eGFR understates absolute clearance, so dosing off the eGFR risks underdosing. In a small, elderly patient, eGFR overstates it, so the same shortcut risks overdosing — in exactly the population least able to tolerate it.

Adjacent question: which one do you monitor over time?

For tracking a patient's kidney function across months or years, use eGFR — the indexing is what makes two readings comparable as weight fluctuates. A patient who loses 15 kg has a genuinely different creatinine clearance, but their kidney function may be unchanged; only the indexed figure tells you that.

For the dose you are writing today, recalculate CrCl each time. Weight changes, and so does the answer.

When neither is the right tool

Both equations assume a steady state. In evolving acute kidney injury, serum creatinine lags the true filtration rate by a day or more, and both figures will read reassuringly high while the kidneys are failing. There is no version of this calculation that is valid mid-AKI — use the trend and the urine output instead.

Where creatinine itself is unreliable — amputation, cachexia, very high muscle mass, a creatine supplement — cystatin C sidesteps the problem, because it is produced by all nucleated cells rather than by muscle. Where the dosing decision is high-consequence, a timed urine collection measures the patient rather than estimating from a population.