Abstract
OBJECTIVE: The role of serum fatty acids in Type 2 diabetes mellitus (T2DM) remains unclear. We aimed to assess and compare the associations of multiple serum unsaturated fatty acids with the prevalence of T2DM to elucidate their heterogeneous relationship profiles. METHODS: This study used the data from the National Health and Nutrition Examination Surveys (2003-2004 and 2011-2012). Weighted proportional and multivariate logistic regression analyses were conducted to evaluate the associations of serum polyunsaturated fatty acids (PUFAs) and monounsaturated fatty acids (MUFAs) with T2DM status and to adjust for potential confounders. RESULTS: The study included 3760 individuals. Results of the multivariate logistic regression analyses showed that among the serum PUFAs, docosapentaenoic acid (DPAn3) (22:5 n3) was associated with lower odds of T2DM (odds ratio [OR]: 0.595, 95% CI: 0.375-0.942, P (trend) = 0.028). Arachidonic acid (AA) (20:4(20: 4 n6) and linoleic acid (LA) (18:2 n6) were negatively associated with T2DM prevalence (OR in quintile 5: AA [20:4 n6] = 0.396, 95% CI: 0.232-0.674, P (trend) = 0.002; LA (18:2 n6) = 0.277, 95% CI: 0.163-0.472, P (trend) < 0.001). Moreover, among the MUFAs, oleic acid (OA) (18:1 n9) and eicosenoic acid (EA) (20:1 n9) were associated with increased odds of T2DM, whereas nervonic acid (NRA) (24:1 n9) was associated with decreased odds of T2DM. Furthermore, nonlinear analyses showed that the n-6 PUFAs dihomo-γ-linolenic acid (DGLA) (20:3 n6) and AA (20:4 n6), as well as the n-9 MUFA EA (20:1 n9), had nonlinear associations with T2DM. CONCLUSIONS: Our findings suggest that several serum n-3 and n-6 PUFAs are inversely associated with T2DM prevalence, whereas some n-9 MUFAs are positively associated with T2DM prevalence. These findings provide comprehensive descriptive data on the serum fatty acid profile in relation to T2DM and underscore the heterogeneity of associations across individual fatty acid species.