High Betaine, a Trimethylamine N-Oxide Related Metabolite, Is Prospectively Associated with Low Future Risk of Type 2 Diabetes Mellitus in the PREVEND Study

Erwin Garcia*, Maryse C. J. Oste, Dennis W. Bennett, Elias J. Jeyarajah, Irina Shalaurova, Eke G. Gruppen, Stanley L. Hazen, James D. Otvos, Stephan J. L. Bakker, Robin P. F. Dullaart, Margery A. Connelly

*Corresponding author for this work

Research output: Contribution to journalArticleAcademicpeer-review

26 Citations (Scopus)
86 Downloads (Pure)


Background: Gut microbiota-related metabolites, trimethylamine-N-oxide (TMAO), choline, and betaine, have been shown to be associated with cardiovascular disease (CVD) risk. Moreover, lower plasma betaine concentrations have been reported in subjects with type 2 diabetes mellitus (T2DM). However, few studies have explored the association of betaine with incident T2DM, especially in the general population. The goals of this study were to evaluate the performance of a newly developed betaine assay and to prospectively explore the potential clinical associations of betaine and future risk of T2DM in a large population-based cohort. Methods: We developed a high-throughput, nuclear magnetic resonance (NMR) spectroscopy procedure for acquiring spectra that allow for the accurate quantification of plasma/serum betaine and TMAO. Assay performance for betaine quantification was assessed and Cox proportional hazards regression was employed to evaluate the association of betaine with incident T2DM in 4336 participants in the Prevention of Renal and Vascular End-Stage Disease (PREVEND) study. Results: Betaine assay results were linear (y = 1.02X - 3.75) over a wide range of concentrations (26.0-1135 mu M). The limit of blank (LOB), limit of detection (LOD) and limit of quantitation (LOQ) were 6.4, 8.9, and 13.2 mu M, respectively. Coefficients of variation for intra- and inter-assay precision ranged from 1.5-4.3% and 2.5-5.5%, respectively. Deming regression analysis of results produced by NMR and liquid chromatography coupled to tandem mass spectrometry(LC-MS/MS) revealed an R-2 value of 0.94 (Y = 1.08x - 1.89) and a small bias for higher values by NMR. The reference interval, in a cohort of apparently healthy adult participants (n = 501), was determined to be 23.8 to 74.7 mu M (mean of 42.9 +/- 12.6 mu M). In the PREVEND study (n = 4336, excluding subjects with T2DM at baseline), higher betaine was associated with older age and lower body mass index, total cholesterol, triglycerides, and hsCRP. During a median follow-up of 7.3 (interquartile range (IQR), 5.9-7.7) years, 224 new T2DM cases were ascertained. Cox proportional hazards regression models revealed that the highest tertile of betaine was associated with a lower incidence of T2DM. Hazard ratio (HR) for the crude model was 0.61 (95% CI: 0.44-0.85, p = 0.004). The association remained significant even after adjusting for multiple clinical covariates and T2DM risk factors, including fasting glucose. HR for the fully-adjusted model was 0.50 (95% CI: 0.32-0.80, p = 0.003). Conclusions: The newly developed NMR-based betaine assay exhibits performance characteristics that are consistent with usage in the clinical laboratory. Betaine levels may be useful for assessing the risk of future T2DM.

Original languageEnglish
Article number1813
Number of pages14
JournalJournal of Clinical Medicine
Issue number11
Publication statusPublished - 1-Nov-2019


  • betaine
  • trimethylamine N-oxide related metabolites
  • nuclear magnetic resonance spectroscopy
  • type 2 diabetes mellitus

Cite this