Cystatin C and frailty in older men

Allyson Hart, Misti L. Paudel, Brent C. Taylor, Areef Ishani, Eric S. Orwoll, Peggy M. Cawthon, Kristine E. Ensrud

Research output: Contribution to journalArticlepeer-review

26 Scopus citations


Objectives To determine whether higher cystatin C would be associated with greater frailty in men aged 65 and older. Design Cross-sectional cohort study. Setting Six U.S. sites. Participants A random sample of community-dwelling men aged 65 and older enrolled in the Osteoporotic Fractures in Men (MrOS) Study (mean age 73.8; 9.8% frail and 47.2% intermediate frailty; N = 1,602). Measurements Serum cystatin C, creatinine, and frailty were measured. Frailty was analyzed as an ordinal outcome of robust, intermediate frailty, and frail using a multinomial logistic regression model, and the base model was adjusted for age, race, and clinical site. Results Higher cystatin C was associated with seven times greater odds of being frail than being robust (odds ratio (OR) quartile 4 vs 1 = 7.12, 95% confidence interval (CI) = 3.76-13.46) and 2.4 times greater odds of intermediate frailty than robust (OR quartile 4 vs 1 = 2.38, 95% CI = 1.70-3.32). The association was attenuated but persisted after adjusting for multiple possible confounders. In contrast, neither higher serum creatinine (OR quartile 4 vs 1 = 1.36, 95% CI = 0.78-2.40) nor lower creatinine-based estimated glomerular filtration rate (OR quartile 4 vs 1 = 1.01, 95% CI = 0.54-1.87) was associated in a graded manner with greater odds of frailty. Conclusion Higher cystatin C, but not creatinine-based measures, was associated with greater odds of frailty in this cohort of older men.

Original languageEnglish (US)
Pages (from-to)1530-1536
Number of pages7
JournalJournal of the American Geriatrics Society
Issue number9
StatePublished - Sep 2013


  • creatinine
  • cystatin C
  • elderly
  • frailty
  • kidney function


Dive into the research topics of 'Cystatin C and frailty in older men'. Together they form a unique fingerprint.

Cite this