Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2024 May 9.
Published in final edited form as: Am J Kidney Dis. 2023 Sep 17;83(2):272–276. doi: 10.1053/j.ajkd.2023.06.008

Performance of GFR estimating equations in young adults

Lesley A Inker 1, Hocine Tighiouart 2, Ogechi M Adingwupu 1, Derek K Ng 3, Michelle M Estrella 4, David Maahs 5, Wei Yang 6, Marc Froissart 7, Michael Mauer 8, Roberto Kalil 9, Vicente Torres 10, Martin de Borst 11, Goran Klintmalm 12, Emilio D Poggio 13, Jesse C Seegmiller 14, Peter Rossing 15, Susan Furth 16, Bradley Warady 17, George J Schwartz 18, Ruben Velez 19, Josef Coresh 3, Andrew Levey 1
PMCID: PMC11080956  NIHMSID: NIHMS1989962  PMID: 37717845

Introduction

In the United States, GFR is commonly estimated using serum creatinine and the 2021 Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) equation for individuals older than 18 years or the 2021 Chronic Kidney Disease in Children Study under 25 (CKiD-U25) equation for those between 1 and 25 years of age with CKD (supplemental background). 1,2 These equations may result in different eGFR values at 18 years and older, leading to uncertainty in assessment of severity of disease, progression rate, and clinical decisions based on level of GFR. The CKiD-U25 has not been externally validated in a diverse population of young adults.

We compared the CKD-EPI and CKiD-U25 equations in young adults prior to the generally accepted age related GFR decline (aged 18 to 40 years) in the 2023 CKD-EPI creatinine external validation dataset (1491 participants from 21 studies) with measured GFR (mGFR) using urinary or plasma clearance of exogenous filtration markers(Table S1–2, Figure S1).1,2 We hypothesized that the CKiD-U25 equation would perform better in young adults with lower GFR, similar to the population in whom the CKiD-U25 equation was developed (mean GFR of 49 (SD 23.0) ml/min/1.73m2), compared to those of older age and higher GFR, similar to population in whom the CKD-EPI equation was developed [mean GFR of 67.6 (SD 39.6 ml/min/1.73 m2)]. We evaluated bias and precision (median and interquartile range of the difference between mGFR and eGFR, respectively), and accuracy (percentage of eGFR within 15% or 30% of mGFR, agreement of eGFR to mGFR categories).1,3,4In sensitivity analyses, we calibrated mGFR to account for potential differences between measurement methods in validation versus the development datasets (Table S3)5–21. We also evaluated performance of the European Kidney Function Consortium (EKFC) equation, which can estimate GFR across the full age spectrum, but was developed in a predominantly white population (Table S2).22

Mean (SD) age was 31.7 (6.0) years and mean (SD) mGFR was 92.7 (32.7) ml/min/1.73m2(Table S4). Younger age was associated with higher mGFR(Figure S2). The equations provided similar estimates for participants with eGFR less than 60 ml/min/1.73m2. At higher values, CKD-EPI yielded generally higher GFR estimates(Figure 1 top panel). Magnitude of the difference in eGFR between equations was larger at younger age and shorter height (Figure S3).

Figure. Difference between eGFR computed using the CKD-EPI and CKiD U25 equations.

Figure

Open in a new tab

Left panel: Agreement between the CKD-EPI and CKiD U25 equations in the study population. Each grey dot represents a participant.

Middle and bottom panels: Comparison of the difference between measured GFR and eGFR creatinine and the average of the two for the CKD-EPI (middle) and CKiD U25 (bottom) equations in the study population. Each grey dot represents a participant. Solid black line is a loess curve. +/−30% and +/−15% lines represent P30 and P15, respectively

For the CKD-EPI equation, there was minimal bias between mGFR and eGFR overall ([−0.5 (95%CI −1.5 to 0.7) mL/min/1.73m2], with small variation by GFR(Figure 1 middle panel, Figure S4, Table S5). In contrast, the CKiD-U25 equation moderately underestimated mGFR overall [7.2 (6.1, 8.3) ml/min/1.73m2], with large underestimation at higher levels of eGFR (Figure 1 bottom panel, Figure S4, Table S5). There was greater variation by age groups with CKiD-U25 than CKD-EPI, with greater underestimation at younger adult ages (Table 1). The CKiD-U25 equation also had greater underestimation, compared to CKD-EPI, across sex and race groups as well BMI >20 kg/m2, but smaller bias for the BMI <20 kg/m2 group (Table S6). P30 was similar for both equations in all subgroups, except for BMI <20 kg/m2 in which P30 was higher for the CKiD-U25 equation. Adjustment for possible differences in measurement methods for GFR attenuated the bias in CKiD-U25 (Table S7). The EKFC equation underestimated mGFR compared to the CKD-EPI equation (Table S6–S8 and Figure S5) and was similar to CKiD-U25.

Table 1.

Performance of CKD-EPI 2021 and U25 equations by age groups compared to measured GFR

Equation Metric Overall Age
18-25 >25-30 >30-35 >35-40
N 1491 276 294 421 500
mGFR 95.2 (72.6, 114.0) 110 (93.0, 126.9) 100 (84.0, 117.0) 94 (71.0, 112.0) 86.5 (62.7, 106.0)
CKD-EPI (Reference equation) Bias −0.5 (−1.5, 0.7) −3.3 (−5.0, 0.0) −3.5  (−5.5, −2.6) 1.1 (−0.5, 2.5) 1 (−0.3, 2.2)
IQR 22.5 (21.0, 23.6) 25.9 (23.2, 29.2) 22 (19.0, 25.4) 22.8  (19.8, 25.3) 19.4 (17.0, 21.4)
P15 57.7  (55.2,60.2) 56.2 (50.0, 62.3) 61.2 (55.4, 66.7) 57.2 (52.5, 62.0) 57 (52.5, 61.2)
P30 88.9 (87.3, 90.5) 90.2 (86.6, 93.5) 90.5 (87.1, 93.5) 89.5  (86.5, 92.4) 86.6  (83.6, 89.4)
Concordance 55.9 (53.3, 58.5) 55.4 (49.6, 61.6) 55.1 (49.3, 60.5) 56.1 (51.1, 60.8) 56.4 (52.0, 60.8)
CKiD-U25 Bias 7.2 (6.1, 8.3) 12.0 (7.7, 15.5) 8.3 (6.6, 10.2) 6.7 (4.3, 10.7) 4.8 (2.8, 6.7)
IQR 23.9 (22.6, 24.9) 29.4 (24.6, 33.1) 22.7 (19.6, 26.1) 24.4 (21.6, 26.8) 20.6 (18.1, 23.5)
P15 52.3 (49.8, 54.9) 48.6 (42.8, 54.7) 53.7 (48.0, 59.5) 51.1 (46.3, 55.8) 54.6 (50.2, 58.8)
P30 87.8 (86.1, 89.4) 87 (83.0, 90.6) 87.4 (83.3, 91.2) 87.9  (84.8, 91.2) 88.4  (85.6, 91.2)
Concordance 50.9 (48.6, 53.5) 46.7 (40.9, 52.5) 51 (45.2, 56.8) 49.4 (44.4, 54.4) 54.4 (50.0, 58.6)
Open in a new tab

mGFR, is reported as median (IQR). Bias (median difference, 95% CI) was expressed as the median difference between mGFR and eGFR. A negative bias indicates overestimation of the measured GFR, and a positive bias indicates underestimation of the measured GFR. IQR (95% CI) is the distance between the 25th and 75th percentile of differences between measured GFR and estimated GFR. P15 (95% CI) is the percentage of individuals with estimated GFRs within 15% of measured GFR. P30 (95% CI) is the percentage of individuals with estimated GFRs within 30% of measured GFR. P30 from 75-80 to 90% has been considered to be adequate for decision making in many clinical circumstances; P30 >90% is considered optimal.24 Concordance (95% CI) was defined as the agreement between measured and estimated GFR categories (<30, 30 – 59, 60 – 89 and ≥90 mL/min/1.73m2). Units for bias is mL/min/1.73m2and for concordance, P15 and P30 are percent.

Colored font indicates non-overlapping confidence intervals (from the use of absolute values for bias) compared with the CKD-EPI equation (reference equation). Red font indicates worse performance compared with the CKD-EPI equation.

Abbreviations: mGFR, measured glomerular filtration rate; CKD-EPI 2021, chronic kidney disease epidemiology creatinine equation published in 2021; CKiD-25, chronic kidney disease in children under 25 serum creatinine equation; mGFR, measured GFR, P30, percentage of estimates within 30% of measured GFR

For young adults with CKD, the transition from pediatric to adult care can occur over a wide age range. In addition, young adults without previously diagnosed CKD may have need for evaluation of GFR. Providers have choices for GFR estimation in these settings. In this study, we found that the CKiD-U25 equation, developed in children and young adults with CKD, had minimal bias in young adults with lower GFR, similar to the CKD-EPI equation, but underestimated mGFR at higher values. The CKD-EPI equation had consistent performance across GFR and age subgroups. In contrast, the EKFC equation performed similarly to the CKiD-U25 equation, as was noted in a European cohort of young adults with higher GFR.23 Differences between study populations in which the equations were developed, especially level of GFR, should be considered when using these equations in clinical practice.2

Strengths of this study are the diverse population across range of GFR, disease and race group, separate from the population in which the equations were developed. A limitation are that the healthy individuals in CKD-EPI development and validation populations included people with type 1 diabetes or kidney donor candidates, who may differ from young adults in the general population.

The results support use 2021 CKD-EPI equation for reporting of eGFR by clinical laboratories in individuals older than 18 years of age. For young adults with childhood CKD, our results support continuing use of the CKiD-U25 equation to maintain consistency of eGFR. This study reinforces the need for additional research in young US adults to resolve differences observed at high levels of GFR and refine recommendations for use of eGFR equations.

Supplementary Material

Supplementary Materials

Acknowledgements

We acknowledge data contribution from Anders Grubb, MD (study investigator, the University of Lund Study), as well as collaborators from other studies included in our analysis. A full list of study collaborators is outlined in the supplementary material. We also acknowledge Shiyuan Miao, MS for his assistance with figures.

Support

Research reported in this manuscript was primarily supported by Grant 1R01DK116790 to Tufts Medical Center from the National Institute of Diabetes and Digestive and Kidney Diseases. Support for studies included in analyses are listed in the supplementary material. The CKiD Study is funded by the National Institute of Diabetes and Digestive and Kidney Diseases, with additional funding from the National Institute of Child Health and Human Development, and the National Heart, Lung, and Blood Institute (U01-DK-66143, U01-DK-66174, U24-DK-082194, U24-DK-66116).

Role of the funder/sponsor:

The funding sources had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.

Financial Disclosures

Dr. Inker report receiving grants to Tufts Medical Center from the National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK). Dr. Levey reports being on the advisory board for AstraZeneca clinical trials for Dapagliflozin and receiving NIH and NKF grants to his institution. Dr. Wei Yang reports grant U24-DK060990 from NIDDK and consulting fees as statistical editor for AJKD. Dr. Martin de Borst reports consulting fees and honoraria to institution from Astra Zeneca, Bayer, Pharmacosmos, and Sanofi Genzyme Amgen and Kyowa Kirin Pharma, and Vifor Pharma, respectively. Dr. Maahs reports grants and contracts from NIH, Helmsley Charitable Trust, and NSF and consulting fees from Medtronic, Provention, Lifescan, and Eli Lilly. Dr. Rossing reports receiving honoraria to his institution from Astra Zeneca, Bayer, Boehringer Ingelheim, Novo Nordisk, Gilead, Sanofi, Abbott and Eli Lilly. Dr. Schwartz reports grants and honoraria from NIDDK and Children’s Mercy Hospital. Dr. Velez reports support from Dallas Nephrology Associates. Dr. Klintmalam reports consulting fees from Immucor and Honoria from UCLA. Dr. Kalil reports research grant from Eurofins. Dr. Torres reports grants from Palladio Biosciences, Mironid, Blueprint Medicines, Tribune, Sanofi, Palladio, Reata and Regulus and honoraria to institution from Otsuka Pharmaceuticals and Vertex Pharmaceuticals. Dr. Seegmiller reports NIDDK to institution. Dr. Coresh reports grants from NKF and consulting fees from Healthy.io.

Data Sharing

Data of studies used in the paper were shared with the CKD-EPI GFR group under strict data use agreements which prohibit the group from sharing data with parties external to the agreement. However, analysis of bio-specimens are shared back to respective study groups and could be accessed via their repositories.

References

  • 1.Inker LA, Eneanya ND, Coresh J, et al. New Creatinine- and Cystatin C-Based Equations to Estimate GFR without Race. The New England journal of medicine. 2021;385(19):1737–1749. doi: 10.1056/NEJMoa2102953. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2.Pierce CB, Muñoz A, Ng DK, Warady BA, Furth SL, Schwartz GJ. Age- and sex-dependent clinical equations to estimate glomerular filtration rates in children and young adults with chronic kidney disease. Kidney international. 2021;99(4):948–956. doi: 10.1016/j.kint.2020.10.047. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3.Inker LA, Schmid CH, Tighiouart H, et al. Estimating glomerular filtration rate from serum creatinine and cystatin C. The New England journal of medicine. 2012;367(1):20–29. doi: 10.1056/NEJMoa1114248. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Inker LA, Couture SJ, Tighiouart H, et al. A New Panel Estimated GFR, Including beta2-Microglobulin and beta-Trace Protein and Not Including Race, Developed in a Diverse Population. Am J Kidney Dis. 2021;77(5):673–683. doi: 10.1053/j.ajkd.2020.11.005. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Kaslow RA, Ostrow DG, Detels R, Phair JP, Polk BF, Rinaldo CR Jr. The Multicenter AIDS Cohort Study: rationale, organization, and selected characteristics of the participants. American journal of epidemiology. 1987;126(2):310–318. doi: 10.1093/aje/126.2.310. [DOI] [PubMed] [Google Scholar]
  • 6.Mathiesen ER, Hommel E, Giese J, Parving HH. Efficacy of captopril in postponing nephropathy in normotensive insulin dependent diabetic patients with microalbuminuria. Brit Med J 1991;303(6794):81–87. doi: 10.1136/bmj.303.6794.81. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Tarnow L, Rossing P, Jensen C, Hansen BV, Parving HH. Long-term renoprotective effect of nisoldipine and lisinopril in type 1 diabetic patients with diabetic nephropathy. Diabetes care. 2000;23(12):1725–1730. doi: 10.2337/diacare.23.12.1725. [DOI] [PubMed] [Google Scholar]
  • 8.Hansen HP, Tauber-Lassen E, Jensen BR, Parving HH. Effect of dietary protein restriction on prognosis in patients with diabetic nephropathy. Kidney international. 2002;62(1):220–228. doi: 10.1046/j.1523-1755.2002.00421.x. [DOI] [PubMed] [Google Scholar]
  • 9.Mauer M, Drummond K. The early natural history of nephropathy in type 1 diabetes: I. Study design and baseline characteristics of the study participants. Diabetes. 2002;51(5):1572–1579. doi: 10.2337/diabetes.51.5.1572. [DOI] [PubMed] [Google Scholar]
  • 10.Mauer M, Zinman B, Gardiner R, et al. ACE-I and ARBs in early diabetic nephropathy. Journal of the renin-angiotensin-aldosterone system : JRAAS. 2002;3(4):262–269. doi: 10.3317/jraas.2002.048. [DOI] [PubMed] [Google Scholar]
  • 11.Chapman AB, Guay-Woodford LM, Grantham JJ, et al. Renal structure in early autosomal-dominant polycystic kidney disease (ADPKD): The Consortium for Radiologic Imaging Studies of Polycystic Kidney Disease (CRISP) cohort. Kidney international. 2003;64(3):1035–1045. doi: 10.1046/j.1523-1755.2003.00185.x. [DOI] [PubMed] [Google Scholar]
  • 12.Feldman HI, Appel LJ, Chertow GM, et al. The Chronic Renal Insufficiency Cohort (CRIC) Study: Design and methods. J Am Soc Nephrol. 2003;14(7):S148–153. doi: 10.1097/01.asn.0000070149.78399.ce. [DOI] [PubMed] [Google Scholar]
  • 13.Jacobsen P, Andersen S, Rossing K, Jensen BR, Parving HH. Dual blockade of the renin-angiotensin system versus maximal recommended dose of ACE inhibition in diabetic nephropathy. Kidney international. 2003;63(5):1874–1880. doi: 10.1046/j.1523-1755.2003.00940.x. [DOI] [PubMed] [Google Scholar]
  • 14.Gonwa TA, Jennings L, Mai ML, Stark PC, Levey AS, Klintmalm GB. Estimation of glomerular filtration rates before and after orthotopic liver transplantation: evaluation of current equations. Liver Transpl. 2004;10(2):301–309. doi: 10.1002/lt.20017. [DOI] [PubMed] [Google Scholar]
  • 15.Bosma RJ, Doorenbos CR, Stegeman CA, van der Heide JJ, Navis G. Predictive performance of renal function equations in renal transplant recipients: an analysis of patient factors in bias. American journal of transplantation : official journal of the American Society of Transplantation and the American Society of Transplant Surgeons. 2005;5(9):2193–2203. doi: 10.1111/j.1600-6143.2005.00982.x. [DOI] [PubMed] [Google Scholar]
  • 16.Froissart M, Rossert J, Jacquot C, Paillard M, Houillier P. Predictive performance of the modification of diet in renal disease and Cockcroft-Gault equations for estimating renal function. J Am Soc Nephrol. 2005;16(3):763–773. doi: 10.1681/asn.2004070549. [DOI] [PubMed] [Google Scholar]
  • 17.Grubb A, Nyman U, Bjork J, et al. Simple cystatin C-based prediction equations for glomerular filtration rate compared with the modification of diet in renal disease prediction equation for adults and the Schwartz and the Counahan-Barratt prediction equations for children. Clinical chemistry. 2005;51(8):1420–1431. doi: 10.1373/clinchem.2005.051557. [DOI] [PubMed] [Google Scholar]
  • 18.Rook M, Hofker HS, van Son WJ, van der Heide J, Ploeg R, Navis G. Predictive capacity of pre-donation GFR and renal reserve capacity for donor renal function after living kidney donation. Am J Trans. 2006;6:1653–1659. doi: 10.1111/j.1600-6143.2006.01359.x. [DOI] [PubMed] [Google Scholar]
  • 19.Inker LA, Wyatt C, Creamer R, et al. Performance of creatinine and cystatin C GFR estimating equations in an HIV-positive population on antiretrovirals. Journal of acquired immune deficiency syndromes (1999). 2012;61(3):302–309. doi: 10.1097/QAI.0b013e31826a6c4f. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Kasiske BL, Anderson-Haag T, Israni AK, et al. A prospective controlled study of living kidney donors: three-year follow-up. Am J Kidney Dis. 2015;66(1):114–124. doi: 10.1053/j.ajkd.2015.01.019. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.Afkarian M, Polsky S, Parsa A, et al. Preventing Early Renal Loss in Diabetes (PERL) Study: A Randomized Double-Blinded Trial of Allopurinol-Rationale, Design, and Baseline Data. Diabetes care. 2019;42(8):1454–1463. doi: 10.2337/dc19-0342. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.Pottel H, Bjork J, Courbebaisse M, et al. Development and Validation of a Modified Full Age Spectrum Creatinine-Based Equation to Estimate Glomerular Filtration Rate : A Cross-sectional Analysis of Pooled Data. Annals of internal medicine. 2021;174(2):183–191. doi: 10.7326/M20-4366. [DOI] [PubMed] [Google Scholar]
  • 23.Nyman U, Björk J, Berg U, et al. The Modified CKiD Study Estimated GFR Equations for Children and Young Adults Under 25 Years of Age: Performance in a European Multicenter Cohort. Am J Kidney Dis. 2022. doi: 10.1053/j.ajkd.2022.02.018. [DOI] [PubMed] [Google Scholar]
  • 24.Kidney Disease: Improving Global Outcomes (KDIGO). KDIGO 2012 Clinical Practice Guideline for the Evaluation and Management of Chronic Kidney Disease. Kidney Int Suppl. 2013;3(1):1–150. doi: 10.1038/ki.2013.243. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplementary Materials
NIHMS1989962-supplement-Supplementary_Materials.docx (840.3KB, docx)

Data Availability Statement

Data of studies used in the paper were shared with the CKD-EPI GFR group under strict data use agreements which prohibit the group from sharing data with parties external to the agreement. However, analysis of bio-specimens are shared back to respective study groups and could be accessed via their repositories.

RESOURCES