Abstract MP54: Sex-Related Differences in the Association of Social Isolation With Plasma Leptin Concentration

Abstract

Introduction: Studies have reported a positive associations of depression and psychosocial stress and plasma leptin concentration. Social isolation has been linked to increased leptin signaling in male rats, but this association has not been investigated in humans. Methods: We investigated the cross-sectional association of social isolation and plasma leptin concentration among 2,729 Framingham Offspring Cohort participants at the 7 th examination (1999-2001, mean age = 60.1 years, 42% men). Social isolation was measured using the Berkman-Syme Social Network Index: we assigned 1 point for being married, 1 point each for >2 close friends and >2 close relatives, 1 point for participation is social groups, and 1 point for participation in religious meetings ≥1-2 times per month. Social isolation was defined as a total of 0 points. Non-fasting plasma leptin levels were measured using a commercial immunoassay. The primary outcome was plasma leptin concentration. Secondary outcomes included caloric intake, measured by food frequency questionnaire, and volumes of abdominal subcutaneous adipose tissue (SAT), and visceral adipose tissue (VAT), measured by CT among 1,210 of the subjects between 2002 and 2005. Leptin concentration and adipose tissue volume were log-transformed due to skewed distributions. We estimated the associations of social isolation with all outcomes using linear regression. Models included age, body mass index, years of education, smoking status, physical activity index, hypertension, high blood glucose, and CES-D score as covariates. Because leptin levels and responses to social stress may vary among men and women, analyses were stratified by sex. Results: Socially isolated participants were younger and had higher CES-D scores than those who were not. Other characteristics were similar between the two groups. Plasma leptin concentration was 20% lower among socially isolated men (95% CI: -31, -6%) compared with men who were not isolated. Social isolation was also associated with lower caloric intake as well (β = -144.5 kcal, 95% CI: -287.1, -1.87 kcal). Among the subset of men for whom we had adipose tissue depot volume measurements (n=552), VAT volume appeared to be 10% lower among socially isolated men compared with non-isolated men (95% CI: -19, 1%), but this was not true of SAT (β = 1%, 95% C: -9, 10%). We did not observe differences in leptin concentration (β = -1%, 95% CI: -10, 15%), caloric intake (β = -39.1 kcal, 95% CI: -136.0, 57.7 kcal) or SAT β=4%, 95% CI: -3, 14%) according to social isolation among women. Socially isolated women appeared to have higher VAT (β = 10, 95% CI: -1, 10%) Conclusions: In our population of middle-aged and older adults, social isolation was associated with lower leptin levels among men, but not women. The lower leptin concentration observed among isolated men may reflect decreased caloric intake in response to social stress.