Background:
While trans-fatty acids (TFA) influence CHD, their effects on type 2 diabetes mellitus (DM) are not established, with mixed findings of experimental, short-term intervention, and observational studies. Effects may vary depending on specific TFA subtype or method of assessment (circulating biomarkers vs. diet).
Objectives:
To examine prospective associations of circulating and estimated dietary TFA with risk of incident DM in older adults.
Methods:
Plasma phospholipid trans-(t-)16:1n9, total t-18:1, and cis/trans-(c/t-), t/c- and t/t-18:2(n6,9) were measured in blood stored among 3,076 adults in the Cardiovascular Health Study (CHS), aged 74±5y and free of prevalent DM in 1992. Dietary TFA was estimated among 4,246 adults free of prevalent DM when dietary questionnaires were initially administered in 1989 (n=3,917) or in 1996 (n=329). Incident DM up to 2009 was defined as new use of insulin or hypoglycemic drugs, fasting glucose≥126 mg/dL, nonfasting glucose≥200 mg/dL, or 2-hour post-challenge glucose≥200 mg/dL. The relative risk of incident DM associated with each TFA subtype was assessed using multivariate Cox proportional hazards regression.
Results:
Levels of each circulating TFA subtype varied from 2.00±0.73 (% of fatty acids) for t-18:1 to 0.05±0.02 for t/t-18:2. TFA subtypes were moderately to highly intercorrelated (r=0.4 to 0.8), except for t/t-18:2 which weakly correlated with all other TFAs (r<0.1). During 30,927 person-years, 364 DM cases occurred among participants with plasma phospholipid TFA measures. Adjusting for demographics, lifestyle factors, and medical history, lower DM risk was associated with higher levels of t-16:1n9 (Quartile 4 vs. Quartile 1 HR=0.76, p trend=0.03), total t-18:1 (HR=0.71, p trend=0.02) and t/t-18:2 (HR=0.73, p trend=0.04). However, further mutual adjustment for the different TFA subtypes attenuated these inverse associations, and none of the 5 circulating TFA biomarkers were independently related to incident DM (p trend≥0.14 for all). During 50,508 person-years in the dietary analyses, 453 DM cases occurred. Adjusting for demographics, lifestyle, medical history, and other dietary habits, increased DM risk was observed among participants with higher consumption of total TFA (Quartile 4 vs. Quartile 1 HR=1.40, p trend=0.04) and t-18:2 (HR=1.49, p trend=0.006), and t-18:1 consumption (HR=1.32, p trend=0.08), although the latter was not statistically significant.
Conclusions:
Plasma phospholipid TFA subtypes were not associated, whereas dietary total TFA and t-18:2 were positively associated, with incident DM among older adults. These findings highlight the need to understand how dietary TFA may influence DM and why associations may differ for circulating versus dietary TFAs.
Alcohol Consumption and Type 2 Diabetes Among Older Adults: The Cardiovascular Health Study*
