CrCl in Amputees, Children & Race

Three groups where applying the standard equation unmodified gives a wrong answer — each for a different reason, and each requiring a different response.

Amputees

Cockcroft-Gault uses body weight as a proxy for muscle mass. After an amputation, measured weight understates the muscle mass the equation was calibrated against, so clearance is underestimated — and the patient may be underdosed.

The correction estimates pre-amputation weight:

Estimated pre-amputation weight = measured weight / (1 − missing fraction)

Divide rather than adding a percentage back. The measured weight already represents the remaining fraction of the original body, so adding a percentage of the reduced weight undershoots.

| Segment | % of total body weight | | --- | --- | | Hand | 0.7% | | Forearm with hand | 2.3% | | Entire arm | 5.0% | | Foot | 1.5% | | Lower leg with foot (below knee) | 5.9% | | Entire leg | 16.0% |

Source: Osterkamp LK. J Am Diet Assoc. 1995;95(2):215–18.

Worked through — a 62-year-old man, below-knee amputation, measured 68 kg, 175 cm, creatinine 1.0 mg/dL:

Missing fraction  = 0.059
Estimated weight  = 68 / (1 − 0.059) = 72.3 kg

Uncorrected CrCl  = [(140 − 62) × 68]   / (72 × 1.0) = 73.7 mL/min
Corrected CrCl    = [(140 − 62) × 72.3] / (72 × 1.0) = 78.3 mL/min

A 4.6 mL/min difference here — modest, because a below-knee amputation is a small fraction. A bilateral above-knee amputation is 32% of body mass, and the correction becomes large enough to cross dosing thresholds.

Two important caveats. First, these are population averages; individual variation is substantial. Second, muscle atrophy in the remaining limbs is common after amputation, so even a corrected weight can overstate current muscle mass.

Where the dosing decision matters, this is a strong indication for either a timed urine collection or a cystatin C-based eGFR — cystatin C is produced by all nucleated cells rather than muscle, so it sidesteps the problem entirely.

The special populations calculator applies these corrections and shows the corrected and uncorrected figures together.

Children

Do not use Cockcroft-Gault in anyone under 18.

It was derived entirely in adults, has never been validated in a paediatric population, and applying it is a recognised dosing error. The correct tool is the bedside Schwartz equation:

eGFR (mL/min/1.73 m²) = 0.413 × height(cm) / Scr(mg/dL)

Source: Schwartz GJ et al. J Am Soc Nephrol. 2009;20(3):629–37.

Three things distinguish it:

It uses height, not weight. In growing children, height correlates better with muscle mass than weight does.

It returns a BSA-indexed figure. Result is in mL/min/1.73 m², not mL/min. Paediatric dosing is usually per kg or per m² anyway, so this is generally what you want — but do not compare it directly against an adult mL/min threshold.

It assumes IDMS-traceable enzymatic creatinine. The 0.413 coefficient is specific to modern standardised assays. Older Jaffe-based values need a different coefficient — check what your laboratory reports.

Neonates and infants under one year need different equations again. Renal function is still maturing, and neither Schwartz nor any adult equation applies.

Because this is a genuine safety issue rather than a preference, the calculator on this site suppresses the adult equations entirely when age is under 18 rather than trusting the user to know.

Race-based adjustments

There is no race adjustment for creatinine clearance. Cockcroft-Gault has never contained one.

Some eGFR equations did. MDRD applies a 1.212 multiplier for Black patients; the 2009 CKD-EPI equation had a similar term. These were empirical: the derivation cohorts showed, on average, higher serum creatinine at a given measured GFR among Black participants, most plausibly reflecting differences in average muscle mass across the specific populations studied.

The 2021 CKD-EPI equations removed it. The NKF-ASN Task Force on Reassessing the Inclusion of Race in Diagnosing Kidney Disease recommended race-free equations, and the 2021 creatinine and creatinine–cystatin C equations were developed in response.

The reasoning:

• Race is a social category, not a biological variable. Using it as a proxy for muscle mass is imprecise at the individual level and unjustifiable at the conceptual level.
• The coefficient raised estimated GFR for Black patients, which systematically delayed CKD diagnosis, specialist referral and transplant waitlisting.
• Assignment was frequently made by clinician observation or an administrative field, and multiracial patients had no coherent category at all.

The effect was not trivial. For a patient with an eGFR of 55 mL/min/1.73 m² without the coefficient, applying it gives 67 — moving them from KDIGO G3a to G2, and potentially out of eligibility for referral.

In practice: use the 2021 race-free CKD-EPI equations. Where muscle mass is genuinely in question for an individual patient, measure it with cystatin C rather than inferring it from race. This site implements only the race-free 2021 equations for clinical use; MDRD with its coefficient is shown solely for comparison with historical results, and labelled as legacy.

Summary

| Population | Problem | Response | | --- | --- | --- | | Amputees | Weight understates muscle mass | Correct weight, or measure clearance / use cystatin C | | Children | Equation never validated | Bedside Schwartz; neonates need different equations again | | Race-based adjustment | Race is not a biological variable | Use race-free CKD-EPI 2021; use cystatin C if muscle mass is in question |

Adjacent question: pregnancy

Pregnancy is the fourth population where the standard approach fails, and it is often overlooked because the patients are young and otherwise well.

Glomerular filtration rate rises substantially in pregnancy — by roughly 40–50% by the second trimester — and serum creatinine falls accordingly. A creatinine that would be unremarkable in a non-pregnant adult can represent genuinely reduced function in pregnancy.

Two practical consequences:

• Estimating equations are not validated in pregnancy. Cockcroft-Gault, MDRD and CKD-EPI were all derived in non-pregnant populations, and weight gain during pregnancy breaks the weight-as-muscle-mass assumption directly.
• A timed urine collection is the more defensible measure where a real clearance figure is needed, and remains standard in obstetric nephrology.

Adjacent question: dialysis and residual function

In a patient on dialysis, neither an estimating equation nor a spot clearance means what it usually means. Dosing follows dialysis-specific references and depends on the modality, the schedule, and how much residual renal function remains.

Residual function matters more than it is usually credited: a patient with 5 mL/min of their own clearance handles renally cleared drugs differently from an anuric patient on an identical schedule. Where it is relevant, it is measured with a timed collection between sessions, not estimated.

Choosing an approach across the four groups

| Population | Standard equation | Better option where it matters | | --- | --- | --- | | Obese | Cockcroft-Gault with adjusted body weight | Salazar-Corcoran; timed collection | | Elderly | Cockcroft-Gault, smaller of actual/ideal weight | Cystatin C | | Amputee | Corrected weight | Cystatin C; timed collection | | Paediatric | Bedside Schwartz | Specialist paediatric nephrology input | | Pregnancy | None validated | Timed collection |

The pattern is consistent: correcting an equation buys you a usable estimate, but where the dosing decision is consequential, measuring beats correcting.