Estimating equations

Creatinine based equations

Creatinine measured in the blood is the most commonly measured endogenous filtration marker. It is a 113 Dalton amino acid derivative that is generated from the breakdown of creatinine in muscle, distributed throughout total body water, and excreted by the kidneys by glomerular filtration and tubular secretion. Although the blood level is affected primarily by the level of GFR, it is also affected by three other physiological processes (non-GFR determinants): generation by muscle mass and diet, tubular secretion by active transport, and extra-renal elimination by gastrointestinal or third space losses. Due to variation in these unmeasured processes, particularly creatinine generation, amongst individuals and within individuals over time, the normal values for creatinine differs among groups. The estimated GFR based on creatinine (eGFRcr) adjusted for demographic factors (age, sex and race) or clinical factors (weight and others) that are related to these processes, provides more information that the creatinine alone. All creatinine based estimating equations are limited by non-GFR determinants of creatinine that cannot be accounted for by the demographic and clinical factors. This is especially important in conditions associated with muscle wasting, as can be seen with severe acute or chronic illnesses.

The 2012 KDIGO CKD guidelines recommended the 2009 CKD-EPI creatinine equation as the first step in GFR evaluation. As of 2020, the three most commonly used equations are the 2009 CKD-EPI creatinine equation and the Modification of Diet in Renal Disease (MDRD) Study equation, which include age, sex and race, and Cockcroft and Gault equation, which includes age, sex and weight. In 2021, the NKF-ASN Task Force on Reassessing the Inclusion of Race in Diagnosing Kidney Diseases concluded that race should not be included in GFR estimating equations, and recommended use of the 2021 CKD-EPI creatinine equation that includes age and sex but not race.
CKD-EPI creatinine equation (2021)

In 2021, to address the concerns about use of a race variable in medical algorithms, CKD-EPI published a new equation for estimating GFR from serum creatinine (2021 CKD-EPI Creatinine equation ) that was developed without a term for Black race. The equation was developed in the same dataset used for development of the 2009 CKD-EPI creatinine equation and validated in a new dataset comprised of the 2012 CKD-EPI creatinine and cystatin C validation dataset plus additional studies (a total of 4050 participants in 12 studies). Compared to the 2009 CKD-EPI Creatinine equation, the 2021 equation is less accurate but acceptable for clinical use in many circumstances. It gives lower eGFR results for Black individuals, and higher eGFR results for non-Black individuals, compared to the 2009 creatinine equation, leading to higher estimated prevalence of CKD among Black individuals, and lower among non-Black individuals.

CKD-EPI Equation GFR
CKD-EPI equation (2009)

The CKD-EPI equation was developed in 2009 using a diverse population of 8254 people with and without chronic kidney disease from Europe and North America. It estimates GFR from serum creatinine, age, sex and race (self-reported Black vs. other peoples). The CKD-EPI equation is as accurate as the MDRD Study equation in the subgroup with estimated GFR less than 60 mL/min/1.73 m2 and substantially more accurate in the subgroup with estimated GFR greater than 60 ml/min/1.73 m2. It is also more accurate than the Cockcroft-Gault equation. Other more recently developed equations derived from European populations, the Lund-Malmo Revised (LMR) and Full-Age Spectrum equations, estimate GFR from creatinine, age and sex without race, and have similar but not better accuracy than the CKD-EPI in white individuals.

The table shows the CKD-EPI equation expressed as separate equations by level of serum creatinine, sex and race. The footnote at the bottom of the tables shows the equation expressed as a single equation.

CKD-EPI Equation GFR
MDRD study equation (2005)

The 4-variable MDRD Study equation was developed in 1999 using data from 1,628 patients with CKD with GFR from approximately 5 to 90 milliliters per minute per 1.73 m2. It estimates GFR adjusted for body surface area and is more accurate than measured creatinine clearance from 24-hour urine collections or estimated by the Cockcroft and Gault formula. The equation was re-expressed in 2005 for use with a standardized serum creatinine assay, which yields 5 percent lower values for serum creatinine concentration. The equation is:

GFR = 175 x Standardized Scr-1.154 x Age-0.203 x (0.742 if female) x (1.210 if African American)

GFR is expressed in mL/min/1.73 m2, SCr is serum creatinine expressed in mg/dL, and age is expressed in years.
Cockcroft and gault formula (1976)

The Cockcroft and Gault formula was developed using data from 249 men with creatinine clearance (CCr) from approximately 30 to 130 mL/m2. It is not adjusted for body surface area. CCr is expressed in milliliters per minute, age in years, weight in kilograms:

CCr= {((140–age) x weight)/(72xSCr)}x 0.85 (if female)

This formula is no longer recommended for use because it has not been expressed using standardized creatinine values and has been shown to be highly inaccurate, especially in the elderly and people who are obese, but is still frequently used for drug dosing.
 

Additional resources

National Kidney Foundation GFR Calculator 
Cockcroft, D.W. and M.H. Gault. Prediction of creatinine clearance from serum creatinine. Nephron. 1976. 16(1):31-41.

Cystatin C based equations

Cystatin C is less affected by muscle mass and diet than is creatinine, and therefore provides an alternative to creatinine based estimation. Its level is also affected by other processes, in particular inflammation, thyroid, obesity and smoking, possibly related to alterations in generation. eGFRcys is not more accurate compared to eGFRcr. The combination of creatinine and cystatin C (eGFRcr-cys) provide a more accurate estimate than either eGFRcr or eGFRcys.

The 2012 KDIGO CKD guidelines recommended the 2012 CKD-EPI cystatin C, which includes age and sex, and the 2012 CKD-EPI creatinine-cystatin C equation, which includes age, sex and race, as confirmatory tests in GFR evaluation. In 2021, the NKF-ASN Task Force on Reassessing the Inclusion of Race in Diagnosing Kidney Diseases concluded that race should not be included in GFR estimating equations, and recommended use of the 2021 CKD-EPI creatinine-cystatin C equation that includes age and sex but not race.

Cystatin C based equations
CKD-EPI Creatinine-Cystatin C equation (2021)

In 2021, to address the concerns about use of a race variable in medical algorithms CKD-EPI published a new equation for creatinine and cystatin C (2021 CKD-EPI Creatinine-Cystatin C equation) that was developed without a term for Black race. The equation was developed in the same dataset used for development of the 2012 CKD-EPI creatinine-cystatin C equation and validated in a new dataset comprised of the 2012 CKD-EPI creatinine and cystatin C validation dataset plus additional studies (a total of 4050 participants in 12 studies). The 2021 equation has similar accuracy as the 2012 CKD-EPI creatinine-cystatin C equation, and greater accuracy than the 2021 creatinine equation. It gives different eGFR results for both Black and non-Black individuals compared to the 2012 CKD-EPI creatinine-cystatin C equations.

ckd epi equation gfr 2021 c
CKD-EPI Creatinine-Cystatin equation (2012)

CKD-EPI 2012 Creatinine-Cystatin C equations were developed in parallel with the 2012 cystatin C equations. They estimate GFR from serum creatinine, cystatin C, age, sex, and race. The equation had similar bias to the creatinine or cystatin C equations but was more precise and therefore had greater overall accuracy. 

The table shows the equation expressed as a separate equation by level of serum creatinine, sex and race. The footnote at the bottom of the tables shows the equation expressed as a single equation.

eGFR-equations-4
CKD-EPI Cystatin C equation (2012)

CKD-EPI Cystatin C equation was developed in 2012 using a diverse population of 5352 people with and without chronic kidney disease from Europe and North America. The table shows the equation expressed as a separate equation by level of serum cystatin C, age and sex. The footnote at the bottom of the tables shows the equation expressed as a single equation.

eGFR equations 5

Panel eGFR equations
Contribution of B2M and BTP to Performance of Race-Free CKD-EPI Equations 2023

CKD-EPI Technical Report, 2023 - PDF
CKD-EPI B2M and BTP equations (2020)

Β-2 microglobulin (B2M) and β-Trace protein (BTP) and are low molecular weight protein filtration markers. Like cystatin C, they are less affected than creatinine by muscle, and are less strongly associated than creatinine with age, sex and race than creatinine. 

In 2020 CKD-EPI published a three-marker panel (cys, BTP and B2M) and a four marker panel (cr, cys, BTP and B2M). Both panels included age and sex, but did not require a coefficient for Black race for maximal accuracy compared to the reference. The three marker panel was more accurate than eGFRcys and the four-marker panel without race was as accurate as eGFRcr-cys. BTP is not commercially available, thus these results demonstrate proof of concept that a multiple marker panel can result in improved accuracy even with less demographic factors.

eGFR equations
CKD-EPI Metabolite Equations (2019)

A GFR-estimating equation based on a panel of metabolites - N-acetylthreonine, pseudouridine, phenylacetylglutamine, and tryptophan – has been developed. It was shown to be more accurate than eGFRcr and eGFRcys and is independent of creatinine, cystatin C, or demographic characteristics. It is currently not available for use outside of research settings but provides proof of concept. CKD-EPI investigators are working on more refined versions of a metabolite based approach.

ckd-epi metabolite equations

Approach to GFR evaluation

Click on any of these to learn more about our approach to GFR evaluation.