Proteinuria or Albuminuria as Markers of Kidney and Cardiovascular Disease Risk: An Individual Patient-Level Meta-analysis

Hiddo J L Heerspink; Morgan E Grams; Yingying Sang; Shoshana H Ballew; Josef Coresh; Aditya Surapaneni; Natalia Alencar de Pinho; Nigel J Brunskill; Alexander R Chang; Elizabeth Ciemins; Laura M Dember; Keiko Kabasawa; Lindsey Kornowske; Adeera Levin; Rupert Major; Patrick B Mark; Eric McArthur; James Medcalf; Marie Metzger; Girish N Nadkarni; David M J Naimark; Cassianne Robinson-Cohen; Keiichi Sumida; Robin W M Vernooij; Ron T Gansevoort; Bengt Fellström; Steven Chadban

doi:10.7326/ANNALS-25-02117

Proteinuria or Albuminuria as Markers of Kidney and Cardiovascular Disease Risk: An Individual Patient-Level Meta-analysis

Hiddo J L Heerspink
, Morgan E Grams^*
, Yingying Sang
, Shoshana H Ballew
, Josef Coresh^*
, Aditya Surapaneni
, Natalia Alencar de Pinho
, Nigel J Brunskill
, Alexander R Chang
, Elizabeth Ciemins
, Laura M Dember
, Keiko Kabasawa
, Lindsey Kornowske
, Adeera Levin
, Rupert Major
, Patrick B Mark
, Eric McArthur
, James Medcalf
, Marie Metzger
, Girish N Nadkarni

David M J Naimark, Cassianne Robinson-Cohen, Keiichi Sumida, Robin W M Vernooij, Ron T Gansevoort, Bengt Fellström, Steven Chadban,

^*Corresponding author for this work

Central Diagnostic Laboratory

Research output: Contribution to journal › Article › Academic › peer-review

Abstract

BACKGROUND: Urinary albumin-creatinine ratio (UACR) and urinary protein-creatinine ratio (UPCR) are both used in clinical practice to diagnose and monitor chronic kidney disease (CKD). Which measure exhibits stronger associations with clinical outcomes and whether this varies by patient characteristics are unknown.

OBJECTIVE: To assess and compare the performance of UACR and UPCR across CKD-related clinical outcomes.

DESIGN: Individual patient-level meta-analysis.

SETTING: 38 research and clinical cohorts.

PARTICIPANTS: 148 994 participants with same-day measurements of UACR and UPCR.

MEASUREMENTS: We quantified the associations of UACR and UPCR with subsequent clinical outcomes, including kidney failure and cardiovascular events, using Cox proportional hazards regression. Analyses were done in each cohort, followed by random-effects meta-analysis. Subgroups included those based on severity of proteinuria, type 2 diabetes, estimated glomerular filtration rate (eGFR) less than 60 mL/min/1.73 m2, and glomerular disease.

RESULTS: There were 148 994 participants and 9773 kidney failure events during a median of 3.8 years of follow-up. Higher UACR and UPCR both had a log-linear association with increased risk for kidney failure. The association with kidney failure was somewhat stronger for UACR (adjusted hazard ratio [HR] per SD increment, 2.55 [95% CI, 2.36 to 2.74]) than UPCR (HR, 2.40 [CI, 2.28 to 2.53]; P for comparison<0.001). Results were consistent to slightly stronger in subgroups with UACR greater than 30 mg/g, UPCR greater than 500 mg/g, eGFR less than 60 mL/min/1.73 m2, diabetes, and glomerular disease. Associations between UACR and UPCR were generally similar for cardiovascular outcomes but favored UACR in subgroups with moderately to severely elevated UACR.

LIMITATION: Assessment of UACR and UPCR in spot urine samples.

CONCLUSION: Overall, UACR was more strongly associated with kidney failure than UPCR (particularly in subgroups with higher UACR), supporting the use of UACR rather than UPCR to diagnose and risk-stratify patients.

PRIMARY FUNDING SOURCE: National Kidney Foundation and National Institute of Diabetes and Digestive and Kidney Diseases.

Original language	English
Pages (from-to)	32-41
Journal	Annals of Internal Medicine
Volume	179
Early online date	4 Nov 2025
DOIs	https://doi.org/10.7326/ANNALS-25-02117
Publication status	Published - Jan 2026

Access to Document

10.7326/ANNALS-25-02117

Embargoed Document

proteinuria-or-albuminuria-as-markers-of-kidney-and-cardiovascular-disease-risk
Final published version, 295 KB
Embargo ends: 4/05/26

Cite this

Heerspink, H. J. L., Grams, M. E., Sang, Y., Ballew, S. H., Coresh, J., Surapaneni, A., Alencar de Pinho, N., Brunskill, N. J., Chang, A. R., Ciemins, E., Dember, L. M., Kabasawa, K., Kornowske, L., Levin, A., Major, R., Mark, P. B., McArthur, E., Medcalf, J., Metzger, M. (2026). Proteinuria or Albuminuria as Markers of Kidney and Cardiovascular Disease Risk: An Individual Patient-Level Meta-analysis. Annals of Internal Medicine, 179, 32-41. https://doi.org/10.7326/ANNALS-25-02117

@article{19fd45cd2df648d4b5ff43a1215764af,

title = "Proteinuria or Albuminuria as Markers of Kidney and Cardiovascular Disease Risk: An Individual Patient-Level Meta-analysis",

abstract = "BACKGROUND: Urinary albumin-creatinine ratio (UACR) and urinary protein-creatinine ratio (UPCR) are both used in clinical practice to diagnose and monitor chronic kidney disease (CKD). Which measure exhibits stronger associations with clinical outcomes and whether this varies by patient characteristics are unknown.OBJECTIVE: To assess and compare the performance of UACR and UPCR across CKD-related clinical outcomes.DESIGN: Individual patient-level meta-analysis.SETTING: 38 research and clinical cohorts.PARTICIPANTS: 148 994 participants with same-day measurements of UACR and UPCR.MEASUREMENTS: We quantified the associations of UACR and UPCR with subsequent clinical outcomes, including kidney failure and cardiovascular events, using Cox proportional hazards regression. Analyses were done in each cohort, followed by random-effects meta-analysis. Subgroups included those based on severity of proteinuria, type 2 diabetes, estimated glomerular filtration rate (eGFR) less than 60 mL/min/1.73 m2, and glomerular disease.RESULTS: There were 148 994 participants and 9773 kidney failure events during a median of 3.8 years of follow-up. Higher UACR and UPCR both had a log-linear association with increased risk for kidney failure. The association with kidney failure was somewhat stronger for UACR (adjusted hazard ratio [HR] per SD increment, 2.55 [95\% CI, 2.36 to 2.74]) than UPCR (HR, 2.40 [CI, 2.28 to 2.53]; P for comparison<0.001). Results were consistent to slightly stronger in subgroups with UACR greater than 30 mg/g, UPCR greater than 500 mg/g, eGFR less than 60 mL/min/1.73 m2, diabetes, and glomerular disease. Associations between UACR and UPCR were generally similar for cardiovascular outcomes but favored UACR in subgroups with moderately to severely elevated UACR.LIMITATION: Assessment of UACR and UPCR in spot urine samples.CONCLUSION: Overall, UACR was more strongly associated with kidney failure than UPCR (particularly in subgroups with higher UACR), supporting the use of UACR rather than UPCR to diagnose and risk-stratify patients.PRIMARY FUNDING SOURCE: National Kidney Foundation and National Institute of Diabetes and Digestive and Kidney Diseases.",

author = "Heerspink, \{Hiddo J L\} and Grams, \{Morgan E\} and Yingying Sang and Ballew, \{Shoshana H\} and Josef Coresh and Aditya Surapaneni and \{Alencar de Pinho\}, Natalia and Brunskill, \{Nigel J\} and Chang, \{Alexander R\} and Elizabeth Ciemins and Dember, \{Laura M\} and Keiko Kabasawa and Lindsey Kornowske and Adeera Levin and Rupert Major and Mark, \{Patrick B\} and Eric McArthur and James Medcalf and Marie Metzger and Nadkarni, \{Girish N\} and Naimark, \{David M J\} and Cassianne Robinson-Cohen and Keiichi Sumida and Vernooij, \{Robin W M\} and Gansevoort, \{Ron T\} and Bengt Fellstr{\"o}m and Steven Chadban",

year = "2026",

month = jan,

doi = "10.7326/ANNALS-25-02117",

language = "English",

volume = "179",

pages = "32--41",

journal = "Annals of Internal Medicine",

issn = "0003-4819",

publisher = "American College of Physicians",

}

Heerspink, HJL, Grams, ME, Sang, Y, Ballew, SH, Coresh, J, Surapaneni, A, Alencar de Pinho, N, Brunskill, NJ, Chang, AR, Ciemins, E, Dember, LM, Kabasawa, K, Kornowske, L, Levin, A, Major, R, Mark, PB, McArthur, E, Medcalf, J, Metzger, M, Nadkarni, GN, Naimark, DMJ, Robinson-Cohen, C, Sumida, K, Vernooij, RWM, Gansevoort, RT, Fellström, B, Chadban, S 2026, 'Proteinuria or Albuminuria as Markers of Kidney and Cardiovascular Disease Risk: An Individual Patient-Level Meta-analysis', Annals of Internal Medicine, vol. 179, pp. 32-41. https://doi.org/10.7326/ANNALS-25-02117

TY - JOUR

T1 - Proteinuria or Albuminuria as Markers of Kidney and Cardiovascular Disease Risk

T2 - An Individual Patient-Level Meta-analysis

AU - Heerspink, Hiddo J L

AU - Grams, Morgan E

AU - Sang, Yingying

AU - Ballew, Shoshana H

AU - Coresh, Josef

AU - Surapaneni, Aditya

AU - Alencar de Pinho, Natalia

AU - Brunskill, Nigel J

AU - Chang, Alexander R

AU - Ciemins, Elizabeth

AU - Dember, Laura M

AU - Kabasawa, Keiko

AU - Kornowske, Lindsey

AU - Levin, Adeera

AU - Major, Rupert

AU - Mark, Patrick B

AU - McArthur, Eric

AU - Medcalf, James

AU - Metzger, Marie

AU - Nadkarni, Girish N

AU - Naimark, David M J

AU - Robinson-Cohen, Cassianne

AU - Sumida, Keiichi

AU - Vernooij, Robin W M

AU - Gansevoort, Ron T

AU - Fellström, Bengt

AU - Chadban, Steven

PY - 2026/1

Y1 - 2026/1

N2 - BACKGROUND: Urinary albumin-creatinine ratio (UACR) and urinary protein-creatinine ratio (UPCR) are both used in clinical practice to diagnose and monitor chronic kidney disease (CKD). Which measure exhibits stronger associations with clinical outcomes and whether this varies by patient characteristics are unknown.OBJECTIVE: To assess and compare the performance of UACR and UPCR across CKD-related clinical outcomes.DESIGN: Individual patient-level meta-analysis.SETTING: 38 research and clinical cohorts.PARTICIPANTS: 148 994 participants with same-day measurements of UACR and UPCR.MEASUREMENTS: We quantified the associations of UACR and UPCR with subsequent clinical outcomes, including kidney failure and cardiovascular events, using Cox proportional hazards regression. Analyses were done in each cohort, followed by random-effects meta-analysis. Subgroups included those based on severity of proteinuria, type 2 diabetes, estimated glomerular filtration rate (eGFR) less than 60 mL/min/1.73 m2, and glomerular disease.RESULTS: There were 148 994 participants and 9773 kidney failure events during a median of 3.8 years of follow-up. Higher UACR and UPCR both had a log-linear association with increased risk for kidney failure. The association with kidney failure was somewhat stronger for UACR (adjusted hazard ratio [HR] per SD increment, 2.55 [95% CI, 2.36 to 2.74]) than UPCR (HR, 2.40 [CI, 2.28 to 2.53]; P for comparison<0.001). Results were consistent to slightly stronger in subgroups with UACR greater than 30 mg/g, UPCR greater than 500 mg/g, eGFR less than 60 mL/min/1.73 m2, diabetes, and glomerular disease. Associations between UACR and UPCR were generally similar for cardiovascular outcomes but favored UACR in subgroups with moderately to severely elevated UACR.LIMITATION: Assessment of UACR and UPCR in spot urine samples.CONCLUSION: Overall, UACR was more strongly associated with kidney failure than UPCR (particularly in subgroups with higher UACR), supporting the use of UACR rather than UPCR to diagnose and risk-stratify patients.PRIMARY FUNDING SOURCE: National Kidney Foundation and National Institute of Diabetes and Digestive and Kidney Diseases.

AB - BACKGROUND: Urinary albumin-creatinine ratio (UACR) and urinary protein-creatinine ratio (UPCR) are both used in clinical practice to diagnose and monitor chronic kidney disease (CKD). Which measure exhibits stronger associations with clinical outcomes and whether this varies by patient characteristics are unknown.OBJECTIVE: To assess and compare the performance of UACR and UPCR across CKD-related clinical outcomes.DESIGN: Individual patient-level meta-analysis.SETTING: 38 research and clinical cohorts.PARTICIPANTS: 148 994 participants with same-day measurements of UACR and UPCR.MEASUREMENTS: We quantified the associations of UACR and UPCR with subsequent clinical outcomes, including kidney failure and cardiovascular events, using Cox proportional hazards regression. Analyses were done in each cohort, followed by random-effects meta-analysis. Subgroups included those based on severity of proteinuria, type 2 diabetes, estimated glomerular filtration rate (eGFR) less than 60 mL/min/1.73 m2, and glomerular disease.RESULTS: There were 148 994 participants and 9773 kidney failure events during a median of 3.8 years of follow-up. Higher UACR and UPCR both had a log-linear association with increased risk for kidney failure. The association with kidney failure was somewhat stronger for UACR (adjusted hazard ratio [HR] per SD increment, 2.55 [95% CI, 2.36 to 2.74]) than UPCR (HR, 2.40 [CI, 2.28 to 2.53]; P for comparison<0.001). Results were consistent to slightly stronger in subgroups with UACR greater than 30 mg/g, UPCR greater than 500 mg/g, eGFR less than 60 mL/min/1.73 m2, diabetes, and glomerular disease. Associations between UACR and UPCR were generally similar for cardiovascular outcomes but favored UACR in subgroups with moderately to severely elevated UACR.LIMITATION: Assessment of UACR and UPCR in spot urine samples.CONCLUSION: Overall, UACR was more strongly associated with kidney failure than UPCR (particularly in subgroups with higher UACR), supporting the use of UACR rather than UPCR to diagnose and risk-stratify patients.PRIMARY FUNDING SOURCE: National Kidney Foundation and National Institute of Diabetes and Digestive and Kidney Diseases.

U2 - 10.7326/ANNALS-25-02117

DO - 10.7326/ANNALS-25-02117

M3 - Article

C2 - 41183334

SN - 0003-4819

VL - 179

SP - 32

EP - 41

JO - Annals of Internal Medicine

JF - Annals of Internal Medicine

ER -

Proteinuria or Albuminuria as Markers of Kidney and Cardiovascular Disease Risk: An Individual Patient-Level Meta-analysis

Abstract

Access to Document

Embargoed Document

Fingerprint

Cite this