Metabolic Clusters and Outcomes in Older Adults

The Cardiovascular Health Study

Kenneth J. Mukamal, David S. Siscovick, Ian H. de Boer, Joachim H. Ix, Jorge Kizer, Luc Djoussé, Annette L. Fitzpatrick, Russell P. Tracy, Edward J. Boyko, Steven E. Kahn, Alice M. Arnold

Research output: Contribution to journalArticle

3 Citations (Scopus)

Abstract

Background/Objectives: Few studies have the requisite phenotypic information to define metabolic patterns that may inform our understanding of the pathophysiology and consequences of diabetes in older adults. We sought to characterize clusters of older adults on the basis of shared metabolic features. Design: Population-based prospective cohort study. Setting: Four U.S. Cardiovascular Health Study field centers. Participants: Individuals aged 65 and older taking no glucose-lowering agents (N = 2,231). Measurements: K-means cluster analysis of 11 metabolic parameters (fasting and postload serum glucose and plasma insulin, fasting C-peptide, body mass index, C-reactive protein (CRP), estimated glomerular filtration rate (eGFR), albuminuria, carboxymethyl lysine (an advanced glycation end-product), procollagen III N-terminal propeptide (a fibrotic marker)) and their associations with incident cardiovascular disease, diabetes, disability, and mortality over 8 to 14.5 years of follow-up and with measures of subclinical cardiovascular disease. Results: A 6-cluster solution provided robust differentiation into distinct, identifiable clusters. Cluster A (n = 739) had the lowest glucose and insulin and highest eGFR and the lowest rates of all outcomes. Cluster B (n = 419) had high glucose and insulin and intermediate rates of most outcomes. Cluster C (n = 118) had the highest insulin. Cluster D (n = 129) had the highest glucose with much lower insulin. Cluster E (n = 314) had the lowest eGFR and highest albuminuria. Cluster F (n = 512) had the highest CRP. Rates of CVD, mortality, and subclinical atherosclerosis were highest in clusters C, D, and E and were similar to rates in participants with treated diabetes. Incidence of disability was highest in Cluster C. Conclusion: Clustering according to metabolic parameters identifies distinct phenotypes that are strongly associated with clinical and functional outcomes, even at advanced age.

Original languageEnglish (US)
Pages (from-to)289-296
Number of pages8
JournalJournal of the American Geriatrics Society
Volume66
Issue number2
DOIs
StatePublished - Feb 1 2018

Fingerprint

Insulin
Glucose
Glomerular Filtration Rate
Health
Albuminuria
C-Reactive Protein
Cluster Analysis
Fasting
Cardiovascular Diseases
Procollagen
Advanced Glycosylation End Products
Mortality
C-Peptide
Atherosclerosis
Body Mass Index
Cohort Studies
Prospective Studies
Phenotype
Incidence
Serum

Keywords

  • aging
  • cardiovascular disease
  • chronic kidney disease
  • diabetes
  • epidemiology
  • inflammation
  • insulin resistance

ASJC Scopus subject areas

  • Geriatrics and Gerontology

Cite this

Mukamal, K. J., Siscovick, D. S., de Boer, I. H., Ix, J. H., Kizer, J., Djoussé, L., ... Arnold, A. M. (2018). Metabolic Clusters and Outcomes in Older Adults: The Cardiovascular Health Study. Journal of the American Geriatrics Society, 66(2), 289-296. https://doi.org/10.1111/jgs.15205

Metabolic Clusters and Outcomes in Older Adults : The Cardiovascular Health Study. / Mukamal, Kenneth J.; Siscovick, David S.; de Boer, Ian H.; Ix, Joachim H.; Kizer, Jorge; Djoussé, Luc; Fitzpatrick, Annette L.; Tracy, Russell P.; Boyko, Edward J.; Kahn, Steven E.; Arnold, Alice M.

In: Journal of the American Geriatrics Society, Vol. 66, No. 2, 01.02.2018, p. 289-296.

Research output: Contribution to journalArticle

Mukamal, KJ, Siscovick, DS, de Boer, IH, Ix, JH, Kizer, J, Djoussé, L, Fitzpatrick, AL, Tracy, RP, Boyko, EJ, Kahn, SE & Arnold, AM 2018, 'Metabolic Clusters and Outcomes in Older Adults: The Cardiovascular Health Study', Journal of the American Geriatrics Society, vol. 66, no. 2, pp. 289-296. https://doi.org/10.1111/jgs.15205
Mukamal, Kenneth J. ; Siscovick, David S. ; de Boer, Ian H. ; Ix, Joachim H. ; Kizer, Jorge ; Djoussé, Luc ; Fitzpatrick, Annette L. ; Tracy, Russell P. ; Boyko, Edward J. ; Kahn, Steven E. ; Arnold, Alice M. / Metabolic Clusters and Outcomes in Older Adults : The Cardiovascular Health Study. In: Journal of the American Geriatrics Society. 2018 ; Vol. 66, No. 2. pp. 289-296.
@article{efc3687102b6471d96594e62e2ac91ee,
title = "Metabolic Clusters and Outcomes in Older Adults: The Cardiovascular Health Study",
abstract = "Background/Objectives: Few studies have the requisite phenotypic information to define metabolic patterns that may inform our understanding of the pathophysiology and consequences of diabetes in older adults. We sought to characterize clusters of older adults on the basis of shared metabolic features. Design: Population-based prospective cohort study. Setting: Four U.S. Cardiovascular Health Study field centers. Participants: Individuals aged 65 and older taking no glucose-lowering agents (N = 2,231). Measurements: K-means cluster analysis of 11 metabolic parameters (fasting and postload serum glucose and plasma insulin, fasting C-peptide, body mass index, C-reactive protein (CRP), estimated glomerular filtration rate (eGFR), albuminuria, carboxymethyl lysine (an advanced glycation end-product), procollagen III N-terminal propeptide (a fibrotic marker)) and their associations with incident cardiovascular disease, diabetes, disability, and mortality over 8 to 14.5 years of follow-up and with measures of subclinical cardiovascular disease. Results: A 6-cluster solution provided robust differentiation into distinct, identifiable clusters. Cluster A (n = 739) had the lowest glucose and insulin and highest eGFR and the lowest rates of all outcomes. Cluster B (n = 419) had high glucose and insulin and intermediate rates of most outcomes. Cluster C (n = 118) had the highest insulin. Cluster D (n = 129) had the highest glucose with much lower insulin. Cluster E (n = 314) had the lowest eGFR and highest albuminuria. Cluster F (n = 512) had the highest CRP. Rates of CVD, mortality, and subclinical atherosclerosis were highest in clusters C, D, and E and were similar to rates in participants with treated diabetes. Incidence of disability was highest in Cluster C. Conclusion: Clustering according to metabolic parameters identifies distinct phenotypes that are strongly associated with clinical and functional outcomes, even at advanced age.",
keywords = "aging, cardiovascular disease, chronic kidney disease, diabetes, epidemiology, inflammation, insulin resistance",
author = "Mukamal, {Kenneth J.} and Siscovick, {David S.} and {de Boer}, {Ian H.} and Ix, {Joachim H.} and Jorge Kizer and Luc Djouss{\'e} and Fitzpatrick, {Annette L.} and Tracy, {Russell P.} and Boyko, {Edward J.} and Kahn, {Steven E.} and Arnold, {Alice M.}",
year = "2018",
month = "2",
day = "1",
doi = "10.1111/jgs.15205",
language = "English (US)",
volume = "66",
pages = "289--296",
journal = "Journal of the American Geriatrics Society",
issn = "0002-8614",
publisher = "Wiley-Blackwell",
number = "2",

}

TY - JOUR

T1 - Metabolic Clusters and Outcomes in Older Adults

T2 - The Cardiovascular Health Study

AU - Mukamal, Kenneth J.

AU - Siscovick, David S.

AU - de Boer, Ian H.

AU - Ix, Joachim H.

AU - Kizer, Jorge

AU - Djoussé, Luc

AU - Fitzpatrick, Annette L.

AU - Tracy, Russell P.

AU - Boyko, Edward J.

AU - Kahn, Steven E.

AU - Arnold, Alice M.

PY - 2018/2/1

Y1 - 2018/2/1

N2 - Background/Objectives: Few studies have the requisite phenotypic information to define metabolic patterns that may inform our understanding of the pathophysiology and consequences of diabetes in older adults. We sought to characterize clusters of older adults on the basis of shared metabolic features. Design: Population-based prospective cohort study. Setting: Four U.S. Cardiovascular Health Study field centers. Participants: Individuals aged 65 and older taking no glucose-lowering agents (N = 2,231). Measurements: K-means cluster analysis of 11 metabolic parameters (fasting and postload serum glucose and plasma insulin, fasting C-peptide, body mass index, C-reactive protein (CRP), estimated glomerular filtration rate (eGFR), albuminuria, carboxymethyl lysine (an advanced glycation end-product), procollagen III N-terminal propeptide (a fibrotic marker)) and their associations with incident cardiovascular disease, diabetes, disability, and mortality over 8 to 14.5 years of follow-up and with measures of subclinical cardiovascular disease. Results: A 6-cluster solution provided robust differentiation into distinct, identifiable clusters. Cluster A (n = 739) had the lowest glucose and insulin and highest eGFR and the lowest rates of all outcomes. Cluster B (n = 419) had high glucose and insulin and intermediate rates of most outcomes. Cluster C (n = 118) had the highest insulin. Cluster D (n = 129) had the highest glucose with much lower insulin. Cluster E (n = 314) had the lowest eGFR and highest albuminuria. Cluster F (n = 512) had the highest CRP. Rates of CVD, mortality, and subclinical atherosclerosis were highest in clusters C, D, and E and were similar to rates in participants with treated diabetes. Incidence of disability was highest in Cluster C. Conclusion: Clustering according to metabolic parameters identifies distinct phenotypes that are strongly associated with clinical and functional outcomes, even at advanced age.

AB - Background/Objectives: Few studies have the requisite phenotypic information to define metabolic patterns that may inform our understanding of the pathophysiology and consequences of diabetes in older adults. We sought to characterize clusters of older adults on the basis of shared metabolic features. Design: Population-based prospective cohort study. Setting: Four U.S. Cardiovascular Health Study field centers. Participants: Individuals aged 65 and older taking no glucose-lowering agents (N = 2,231). Measurements: K-means cluster analysis of 11 metabolic parameters (fasting and postload serum glucose and plasma insulin, fasting C-peptide, body mass index, C-reactive protein (CRP), estimated glomerular filtration rate (eGFR), albuminuria, carboxymethyl lysine (an advanced glycation end-product), procollagen III N-terminal propeptide (a fibrotic marker)) and their associations with incident cardiovascular disease, diabetes, disability, and mortality over 8 to 14.5 years of follow-up and with measures of subclinical cardiovascular disease. Results: A 6-cluster solution provided robust differentiation into distinct, identifiable clusters. Cluster A (n = 739) had the lowest glucose and insulin and highest eGFR and the lowest rates of all outcomes. Cluster B (n = 419) had high glucose and insulin and intermediate rates of most outcomes. Cluster C (n = 118) had the highest insulin. Cluster D (n = 129) had the highest glucose with much lower insulin. Cluster E (n = 314) had the lowest eGFR and highest albuminuria. Cluster F (n = 512) had the highest CRP. Rates of CVD, mortality, and subclinical atherosclerosis were highest in clusters C, D, and E and were similar to rates in participants with treated diabetes. Incidence of disability was highest in Cluster C. Conclusion: Clustering according to metabolic parameters identifies distinct phenotypes that are strongly associated with clinical and functional outcomes, even at advanced age.

KW - aging

KW - cardiovascular disease

KW - chronic kidney disease

KW - diabetes

KW - epidemiology

KW - inflammation

KW - insulin resistance

UR - http://www.scopus.com/inward/record.url?scp=85041946831&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85041946831&partnerID=8YFLogxK

U2 - 10.1111/jgs.15205

DO - 10.1111/jgs.15205

M3 - Article

VL - 66

SP - 289

EP - 296

JO - Journal of the American Geriatrics Society

JF - Journal of the American Geriatrics Society

SN - 0002-8614

IS - 2

ER -