Abstract MP57: Associations of Pentraxin 3 with Frailty: The Cardiovascular Health Study (CHS)

Abstract

Background: Pentraxin 3 (PTX3) is an inflammatory biomarker primarily reflecting vascular damage. PTX3 is associated with cardiovascular disease (CVD); however, associations with frailty, another inflammation-related disease of aging, are unknown. Methods: At baseline, we examined these associations in 1,948 white and black men and women (mean age 73 y) in the Cardiovascular Health Study (CHS) who had measures of PTX3 and were free of dementia, Parkinson’s disease and depression. PTX3 was determined by immunoassay (coefficient of variation 10%). Robust regression models were minimally adjusted for age, sex, ethnicity and clinic site with additional adjustments for obesity, smoking, education, C-reactive protein (CRP) and prevalent subclinical or clinical CVD. Results: Levels of PTX3 ranged from 0.24 - 23.05 ng/ml, mean 2.36 ng/ml (standard deviation 2.00 ng/ml), in these older adults. PTX3 was significantly associated with unintentional weight loss (p<0.001), slow walk time (p=0.04) and low grip strength (p=0.03), but not the other domains of frailty, exhaustion and physical activity, both p>0.2. Increased PTX3 was also associated with frailty status comparing those who were not frail (0 domains of frailty, N=879) to those who were intermediate frail (presence of 1-2 domains of frailty, N=926) or frail (≥3 domains, N=143). In minimally adjusted models, the relative risk (RR) for intermediate frailty was 1.22 (95% confidence interval 1.04-1.44) and 1.70 (1.25-2.29) for prevalent frailty, comparing both to non-frail. Results were slightly attenuated but remained significant in fully adjusted models; RR 1.19 (1.01-1.41) for intermediate and 1.59 (1.16-2.18) for frail. As some biomarkers show sex-dependent associations, we stratified on sex. In fully adjusted models, increased PTX3 was associated with intermediate frailty (RR 1.3; 1.00-1.69) and frailty (RR 2.14; 1.24-3.67) in men, but did not appear to be associated in women (RR 1.13; 0.91-1.42 for intermediate and 1.28; 0.85-1.92 for frail); however, the interaction term was not significant (p=0.2). Conclusions: In these older adults, PTX3 was associated with some domains of frailty and frailty status. Associations were independent of the presence of clinical and/or subclinical CVD and CRP, a biomarker of general inflammation. PTX3 may reflect ongoing vascular damage in multiple aging-related diseases such as CVD and frailty. Further study is needed to evaluate the potential for PTX3 to act as a biomarker of healthy aging.