Association of NEFA composition with insulin sensitivity and beta cell function in the Prospective Metabolism and Islet Cell Evaluation (PROMISE) cohort
Our aim was to determine the longitudinal associations of individual NEFA with the pathogenesis of diabetes, specifically with differences in insulin sensitivity and beta cell function over 6 years in a cohort of individuals who are at risk for diabetes.
In the Prospective Metabolism and Islet Cell Evaluation (PROMISE) longitudinal cohort, 477 participants had serum NEFA measured at the baseline visit and completed an OGTT at three time points over 6 years. Outcome variables were calculated using the OGTT values. At each visit, insulin sensitivity was assessed using the HOMA2 of insulin sensitivity (HOMA2-%S) and the Matsuda index, while beta cell function was assessed using the insulinogenic index over HOMA-IR (IGI/IR) and the insulin secretion-sensitivity index-2 (ISSI-2). Generalised estimating equations were used, adjusting for time, waist, sex, ethnicity, baseline age, alanine aminotransferase (ALT) and physical activity. NEFA were analysed as both concentrations (nmol/ml) and proportions (mol%) of the total fraction.
Participants’ (73% female, 70% with European ancestry) insulin sensitivity and beta cell function declined by 14–21% over 6 years of follow-up. In unadjusted models, several NEFA (e.g. 18:1 n-7, 22:4 n-6) were associated with lower insulin sensitivity, however, nearly all of these associations were attenuated in fully adjusted models. In adjusted models, total NEFA, 16:0, 18:1 n-9 and 18:2 n-6 (as concentrations) were associated with 3.7–8.0% lower IGI/IR and ISSI-2, while only 20:5 n-3 (as mol%) was associated with 7.7% higher HOMA2-%S.
Total NEFA concentration was a strong predictor of lower beta cell function over 6 years. Our results suggest that the association with beta cell function is due to the absolute size of the serum NEFA fraction, rather than the specific fatty acid composition.
KeywordsBeta cell function Diabetes pathogenesis Fatty acid composition Insulin sensitivity Longitudinal cohort Non-esterified fatty acids
False discovery rate
Generalised estimating equations
HOMA2 of insulin sensitivity
Impaired fasting glucose
Insulinogenic index over HOMA-IR
Impaired glucose tolerance
Insulin sensitivity index (Matsuda index)
Insulin secretion-sensitivity index-2
Modified activity questionnaire
Metabolic equivalent of task
Partial least squares regression
Prospective Metabolism and Islet Cell Evaluation Cohort
Polyunsaturated fatty acids
Quasi-likelihood information criterion
The authors thank J Neuman, PV Nostrand, S Kink and A Barnie (all of the Leadership Sinai Centre for Diabetes, Mount Sinai Hospital, Canada) and S Porter and M Marin (both of the Centre for Studies in Family Medicine, University of Western Ontario, Canada) for their expert technical assistance and dedication in their work for PROMISE.
The authors had the following responsibility: LWJ analysed and interpreted the data and drafted the article; RR and SBH contributed to the conception and design and revised the article intellectual content; ZL acquired the data and revised the article intellectual content; AG assisted with interpretation of the data and revised the article intellectual content; AJH and RPB substantially contributed to the study conception and design, assisted with interpretation of the data and revised the article intellectual content. All authors read and approved the final version. LWJ and AJH have primary responsibility for final content and are the guarantors of this work.
Duality of interest
The authors declare that there is no duality of interest associated with this manuscript.
- 12.Miller MR, Pereira RI, Langefeld CD et al (2012) Levels of free fatty acids (FFA) are associated with insulin resistance but do not explain the relationship between adiposity and insulin resistance in Hispanic Americans: The IRAS Family Study. J Clin Endocrinol Metab 97:3285–3291CrossRefPubMedPubMedCentralGoogle Scholar
- 15.WHO, IDF (2006) Definition and diagnosis of diabetes mellitus and intermediate hyperglycaemia: Report of a WHO/IDF consultation. Available from: http://whqlibdoc.who.int/publications/2006/9241594934_eng.pdf
- 33.R Core Team (2015) R: A Language and Environment for Statistical Computing. R Foundation for Statistical Computing, ViennaGoogle Scholar
- 34.Højsgaard S, Halekoh U, Yan J (2006) The R package geepack for generalized estimating equations. J Stat Softw 15(2):1–11Google Scholar
- 35.Benjamini Y, Hochberg Y (1995) Controlling the false discovery rate: A practical and powerful approach to multiple testing. J R Stat Soc Ser B Stat Methodol 57:289–300Google Scholar