Recommendations for Implementing the CKD-EPI 2021 Race-Free eGFR Calculation: Guidelines for Clinical Laboratories
Table of Contents
- When introducing the new eGFR equations for eGFRcr and eGFRcr-cys and the testing is not reported by the laboratory
- When introducing the new eGFR equations for eGFRcr and for eGFRcr-cys
- Suggested laboratory report comment messages
- Table S1: Values to verify appropriate calculation of eGFRcr
- Table S2: Values to verify appropriate calculation of eGFRcr-cys
- Table S3. Equations to calculate eGFRcr with creatinine in SI units
- Table S4. Equations to calculate eGFRcr-cys with creatinine in SI units
- Programming the eGFR cystatin C equation
- Resources available from the National Kidney Foundation
When introducing the new eGFR equations for eGFR_{cr} and eGFR_{cr-cys} and the testing is not reported by the laboratory
This text can be used or modified to meet the needs of the laboratory, health care professionals and patients.
Effective <DATE>, our laboratory is changing the calculation of estimated glomerular filtration rate (eGFR) from creatinine to the new CKD-EPI 2021 equation that does not include a race coefficient. The new equation is recommended by the National Kidney Foundation and the American Society of Nephrology’s Task Force on Reassessing the Inclusion of Race in Diagnosing Kidney Disease. The new eGFR_{cr} equation has similar overall performance characteristics to the older equations and has been assessed to not have potential consequences that disproportionately affect any one group of individuals. For most patients the eGFR_{cr} result will be similar, however, for some, the values may differ by more than 10% particularly at higher values of eGFR_{cr} and for younger adult ages. Please go to eGFR Calculator | National Kidney Foundation to compare the new CKD-EPI 2021 eGFR_{cr} result with a previously calculated eGFR based on older equations.
The eGFR_{cr} values using the new equation will only report one value and will use nomenclature to distinguish results from the older equations. Results will not trend with those using the older equations. The new report name is eGFR_{cr} (calculated using only creatinine)
<INCLUDE THE NEXT SENTENCE IF APPLICABLE FOR A LABORATORY.> We are also introducing the Kidney Profile order that includes both creatinine with calculated eGFR_{cr} and urine albumin-creatinine ratio (uACR). Clinical practice recommendations suggest ordering uACR with serum creatinine to facilitate appropriate classification of patients with chronic kidney disease, to assess risk for progression and to monitor patients at risk to develop CKD.
When evaluating a patient’s GFR_{,} it is important to remember that eGFR is an estimate of the patient’s GFR. For eGFR_{cr} and eGFR_{cys} (calculated from cystatin C), 80-90% of values are within 30% of measured GFR. eGFR_{cr-cys} (calculated using both creatinine and cystatin C) is more accurate than either eGFR_{cr} or eGFR_{cys} alone.
eGFR_{cr} values need to be interpreted based on clinical context. Clinical practice recommendations suggest ordering cystatin C as a confirmatory test for patients with eGFR_{cr} of 45-59 mL/min/1.73m^{2} with uACR <30 mg/g, and in patients for whom the creatinine may be a less reliable indicator of GFR near decision points. eGFR_{cr-cys} provides a more accurate estimate of GFR. In addition, eGFR_{cys} may be more accurate in conditions when creatinine is a less reliable test for estimating the GFR. Situations in which non-GFR factors may have a large affect on serum creatinine include alterations in creatinine generation (muscle wasting diseases, amputees, body builders, vegan diet), drugs that affect tubular secretion of creatinine (cimetidine, cobicistat, dolutegravir, fenofibrate, ritonovir, trimethoprim and others), and conditions with extra-renal elimination of creatinine (gastrointestinal and “third-space” losses). Clinicians must also consider non-GFR factors affecting cystatin C, including smoking, obesity, inflammation and disorders of thyroid or adrenal hormones.
References
Delgado C, Baweja M, Crews DC, et al. A Unifying Approach for GFR Estimation: Recommendations of the NKF-ASN Task Force on Reassessing the Inclusion of Race in Diagnosing Kidney Disease. Am J Kidney Dis. 2021 DOI: 10.1053/j.ajkd.2021.08.003
Inker LA, Eneanya ND, MCorsh J, et al. New Creatinine- and Cystatin C–Based Equations to Estimate GFR without Race. New England J Med. 2021: DOI: 10.1056/NEJMoa2102953
When introducing the new eGFR equations for eGFR_{cr} and for eGFR_{cr-cys}
This text can be used or modified to meet the needs of the laboratory, health care professionals and patient.
Effective <DATE>, our laboratory is changing the calculation of estimated glomerular filtration rate (eGFR) from creatinine and from combined creatinine-cystatin C to the new CKD-EPI 2021 equations that do not include a race coefficient. The new equations are recommended by the National Kidney Foundation and the American Society of Nephrology’s Task Force on Reassessing the Inclusion of Race in Diagnosing Kidney Disease. The new eGFR_{cr} equation has similar overall performance characteristics to the older equations and has been assessed to not have potential consequences that disproportionately affect any one group of individuals. For most patients the eGFR_{cr} result will be similar, however, for some, the values may differ by more than 10% particularly at higher values of eGFR_{cr} and for younger adult ages. Please go to eGFR Calculator | National Kidney Foundation to compare the new CKD-EPI 2021 eGFR_{cr} result with a previously calculated eGFR based on older equations.
The eGFR values using the new equations will only report one value and will use nomenclature to distinguish results from the older equations. Results will not trend with those using the older equations. The new report names are:
- eGFR_{cr} (calculated using only creatinine)
- eGFR_{cr-cys} (calculated using both creatinine and cystatin C)
- eGFR_{cys} (calculated using only cystatin C)
<INCLUDE THE NEXT SENTENCE IF APPLICABLE FOR A LABORATORY.> We are also introducing the Kidney Profile order that includes both creatinine with calculated eGFR_{cr} and urine albumin-creatinine ratio (uACR). Clinical practice recommendations suggest ordering uACR with serum creatinine to facilitate appropriate classification of patients with chronic kidney disease, to assess risk for progression and to monitor patients at risk to develop CKD.
When evaluating a patient’s GFR_{,} it is important to remember that eGFR is an estimate of the patient’s GFR. For eGFR_{cr} and eGFR_{cys}, 80-90% of values are within 30% of measured GFR. eGFR_{cr-cys} is more accurate than either eGFR_{cr} or eGFR_{cys} alone.
eGFR_{cr} values need to be interpreted based on clinical context. Clinical practice recommendations suggest ordering cystatin C as a confirmatory test for patients with eGFR_{cr} of 45-59 mL/min/1.73m^{2} with uACR <30 mg/g, and in patients for whom the creatinine may be a less reliable indicator of GFR near decision points. eGFR_{cr-cys} provides a more accurate estimate of GFR. In addition, eGFR_{cys} may be more accurate in conditions when creatinine is a less reliable test for estimating the GFR. Situations in which non-GFR factors may have a large affect on serum creatinine include alterations in creatinine generation (muscle wasting diseases, amputees, body builders, vegan diet), drugs that affect tubular secretion of creatinine (cimetidine, cobicistat, dolutegravir, fenofibrate, ritonovir, trimethoprim and others), and conditions with extra-renal elimination of creatinine (gastrointestinal and “third-space” losses). Clinicians must also consider non-GFR factors affecting cystatin C, including smoking, obesity, inflammation and disorders of thyroid or adrenal hormones.
References
Delgado C, Baweja M, Crews DC, et al. A Unifying Approach for GFR Estimation: Recommendations of the NKF-ASN Task Force on Reassessing the Inclusion of Race in Diagnosing Kidney Disease. Am J Kidney Dis. 2021 DOI: 10.1053/j.ajkd.2021.08.003
Inker LA, Eneanya ND, MCorsh J, et al. New Creatinine- and Cystatin C–Based Equations to Estimate GFR without Race. New England J Med. 2021: DOI: 10.1056/NEJMoa2102953
Suggested laboratory report comment messages
Circumstance | Message |
Laboratory report, prior to change (with implementation date not determined) | Reported eGFR is based the [ADD CURRENT] equation. The laboratory is working to implement the new CKD-EPI 2021 equation that does not use a race coefficient. In the interim, to calculate eGFR based on the CKD-EPI 2021 equation, go to CKD-EPI Creatinine Equation (2021) | National Kidney Foundation. |
Laboratory report, prior to change (perhaps one month prior to change) | Reported eGFR is based the [ADD CURRENT] equation. Effective DATE, the laboratory will implement the new CKD-EPI 2021 equation. In the interim, to calculate eGFR based on the CKD-EPI:2021 equation, go to CKD-EPI Creatinine Equation (2021) | National Kidney Foundation. |
Laboratory report, after implementing the change | Reported eGFR is based the CKD-EPI 2021 equation that does not use a race coefficient. |
Laboratory report reported eGFR of 45-59 mL/min/1.73m^{2}. Laboratories need to identify a referral laboratory or offer eGFR_{cr-cys }prior to listing this comment | NKF KDOQI and KDIGO guidelines recommend confirming eGFR of 45-59 mL/min/1.73m^{2} with uACR <30 mg/g based on eGFR calculated using both creatinine and cystatin C. |
Table S1: Values to verify appropriate calculation of eGFR_{cr}
AGE (years) | Sex |
Serum Creatinine mg/dL |
eGFR_{cr} (CKD-EPI 2021) mL/min/1.73m^{2} |
<18 | Male | do not calculate | |
<18 | Female | do not calculate | |
18 | Male | 0.90 | 127 |
18 | Male | 0.91 | 125 |
18 | Female | 0.70 | 128 |
18 | Female | 0.71 | 126 |
90 | Male | 0.50 | 97 |
90 | Male | 1.50 | 44 |
90 | Female | 0.50 | 89 |
90 | Female | 1.50 | 33 |
not available | do not calculate | ||
not available | do not calculate | ||
Creatinine below or above the measuring interval (analytical measurement range) |
do not calculate (see notes) |
NOTES
The creatinine values for age 18 years will confirm the correct equation is used at the splice or knot value in the equations. The creatinine values at other ages recheck the calculation both below and above the splice or knot value in the equations.
Table S1 is based on non-SI units (mg/dL). To convert serum creatinine values to SI units (μmol/L) multiply by 88.4.
When creatinine is below or above the measuring interval (analytical measurement range), a laboratory may choose to report eGFR_{cr} as less than or greater than the eGFR_{cr} value that is calculated for concentrations at the respective limits of the measuring interval.
Table S2: Values to verify appropriate calculation of eGFR_{cr-cys}
AGE (years) | Sex |
Serum Creatinine mg/dL |
Serum Cystatin C mg/L |
eGFRcr-cys (CKD-EPI:2021) mL/min/1.73m^{2} |
<18 | Male | do not calculate | ||
<18 | Female | do not calculate | ||
18 | Male | 0.90 | 0.80 | 126 |
18 | Male | 0.90 | 0.81 | 125 |
18 | Female | 0.70 | 0.80 | 121 |
18 | Female | 0.70 | 0.81 | 120 |
18 | Male | 0.91 | 0.80 | 125 |
18 | Male | 0.91 | 0.81 | 124 |
18 | Female | 0.71 | 0.80 | 120 |
18 | Female | 0.71 | 0.81 | 119 |
30 | Male | 1.50 | 0.80 | 91 |
30 | Male | 1.50 | 0.81 | 90 |
30 | Female | 1.50 | 0.80 | 76 |
30 | Female | 1.60 | 0.81 | 73 |
not available | do not calculate | |||
not available | do not calculate | |||
Creatinine below or above the measuring interval (analytical measurement range) |
do not calculate (see notes) |
NOTES
The creatinine values for age 18 years will confirm the correct equation is used at the splice or knot value in the equations. The creatinine values at other ages recheck the calculation both below and above the splice or knot value in the equations.
Table S2 is based on non-SI units (mg/dL). To convert serum creatinine values to SI units (μmol/L) multiply by 88.4.
When creatinine and/or cystatin C is below or above the measuring interval (analytical measurement range), a laboratory may choose to report eGFR_{cr-cys} as less than or greater than the eGFR_{cr-cys} value that is calculated for concentrations at the respective limits of the measuring intervals.
Table S3. Equations to calculate eGFR_{cr} with creatinine in SI units
Logic^{a} | |||
Age | Sex | S_{Cr}^{b}_{ (µmol/L)} | eGFR Equation |
>=18 | Female | <=61.9 | = 142 x (S_{cr} /61.88)^{-0.241} x 0.9938^{Age} x 1.012 |
>61.9 | = 142 x (S_{cr} /61.88)^{-1.200} x 0.9938^{Age} x 1.012 | ||
Male | <=79.6 | = 142 x (S_{cr} /79.56)^{-0.302} x 0.9938^{Age} | |
>79.6 | = 142 x (S_{cr} /79.56)^{-1.200} x 0.9938^{Age} |
^{a} Programming logic for “If” statements to select the correct equation for each set of parameters.
^{a} Scr = serum creatinine in µmol/L
Table S4. Equations to calculate eGFR_{cr-cys} with creatinine in SI units
Logic^{a} | ||||
Age | Sex | S_{Cr}^{b}_{ (}_{µmol/L)} | S_{Cys (mg/L)} | eGFR_{Cr-Cys} |
>=18 | Female | ≤61.9 | ≤0.80 | 135 x (S_{cr}/61.88)^{-0.219 }x (S_{cys}/0.8)^{-0.323} x 0.9961^{Age} x 0.963 |
>0.80 | 135 x (S_{cr}/61.88)^{-0.219} x (S_{cys}/0.8)^{-0.778} x 0.9961^{Age} x 0.963 | |||
>61.9 | ≤0.80 | 135 x (S_{cr}/61.88)^{-0.544 }x (S_{cys}/0.8)^{-0.323} x 0.9961^{Age} x 0.963 | ||
>0.80 | 135 x (S_{cr}/61.88)^{-0.544 }x (S_{cys}/0.8)^{-0.778} x 0.9961^{Age} x 0.963 | |||
Male | ≤79.6 | ≤0.80 | 135 x (S_{cr}/79.56)^{-0.144 }x (S_{cys}/0.8)^{-0.323} x 0.9961^{Age} | |
>0.80 | 135 x (S_{cr}/79.56)^{-0.144} x (S_{cys}/0.8)^{-0.778} x 0.9961^{Age} | |||
>79.6 | ≤0.80 | 135 x (S_{cr}/79.56)^{-0.544 }x (S_{cys}/0.8)^{-0.323} x 0.9961^{Age} | ||
>0.80 | 135 x (S_{cr}/79.56)^{-0.544 }x (S_{cys}/0.8)^{-0.778} x 0.9961^{Age} |
^{a} Programming logic for “If” statements to select the correct equation for each set of parameters.
^{b} Scr = serum creatinine in µmol/L
Programming the eGFR cystatin C equation
From reference 14, the equation for ages <=18 years in a single expression for eGFR_{cys} is:
eGFR_{cys} = 133 x min(S_{cys}/0.8, 1)^{-0.499 }x max(S_{cys}/0.8, 1)^{-1.328} x 0.996^{Age} x 0.932 [if female]
where S_{cys} = serum cystatin C in mg/L
Age (years)
The coding for implementing the single equation is nuanced. The “min(S_{cys}/0.8, 1)” factor indicates the minimum of S_{cys}/0.8 or 1.0 and “max(S_{cys}/0.8, 1)” indicates the maximum of S_{cys}/0.8 or 1.0.
Coding may be simpler if the single equation is expressed as four equations applicable to four logical conditions as in Table S5.
Table S5. Equations to calculate eGFR_{cys}
Logic^{a} | |||
Age | Sex | S_{Cys}^{b}_{ (mg/L)} | eGFR Equation |
>=18 | Female | <=0.8 | = 133 x (S_{cys}/0.8)^{-0.499} x 0.996^{Age} x 0.932 |
>0.8 | = 133 x (S_{cys}/0.8)^{-1.328} x 0.996^{Age} x 0.932 | ||
Male | <=0.8 | = 133 x (S_{cys}/0.8)^{-0.499} x 0.996^{Age} | |
>0.8 | = 133 x (S_{cys}/0.8)^{-1.328} x 0.996^{Age} |
^{a} Programming logic for “If” statements to select the correct equation for each set of parameters.
^{b} S_{cys} = serum cystatin C
Resources available from the National Kidney Foundation
eGFR calculator website version
eGFR app available free from Apple and Google stores
Laboratory Engagement Initiative
Patient educational resources
Changes to eGFR Calculation and What that Means for People Living with Kidney Disease
Social Determinants of Kidney Disease
Family History and Kidney Diseases
Free professional CME
Not by Muscle, Race or Ethnicity: Practical use of Cystatin C to estimate GFR