Limits...
Betaine and Trimethylamine-N-Oxide as Predictors of Cardiovascular Outcomes Show Different Patterns in Diabetes Mellitus: An Observational Study.

Lever M, George PM, Slow S, Bellamy D, Young JM, Ho M, McEntyre CJ, Elmslie JL, Atkinson W, Molyneux SL, Troughton RW, Frampton CM, Richards AM, Chambers ST - PLoS ONE (2014)

Bottom Line: In subjects with diabetes (n = 79), high plasma betaine was associated with increased frequencies of events; significantly for heart failure, hazard ratio 3.1 (1.2-8.2) and all cardiovascular events, HR 2.8 (1.4-5.5).In subjects without diabetes (n = 396), low plasma betaine was associated with events; significantly for secondary myocardial infarction, HR 2.1 (1.2-3.6), unstable angina, HR 2.3 (1.3-4.0), and all cardiovascular events, HR 1.4 (1.0-1.9).Adding the estimated glomerular filtration rate to Cox regression models tended to increase the apparent risks associated with low betaine.

View Article: PubMed Central - PubMed

Affiliation: Clinical Biochemistry Unit, Canterbury Health Laboratories, Christchurch, New Zealand; Department of Pathology, University of Otago Christchurch, Christchurch, New Zealand.

ABSTRACT

Background: Betaine is a major osmolyte, also important in methyl group metabolism. Concentrations of betaine, its metabolite dimethylglycine and analog trimethylamine-N-oxide (TMAO) in blood are cardiovascular risk markers. Diabetes disturbs betaine: does diabetes alter associations between betaine-related measures and cardiovascular risk?

Methods: Plasma samples were collected from 475 subjects four months after discharge following an acute coronary admission. Death (n = 81), secondary acute MI (n = 87), admission for heart failure (n = 85), unstable angina (n = 72) and all cardiovascular events (n = 283) were recorded (median follow-up: 1804 days).

Results: High and low metabolite concentrations were defined as top or bottom quintile of the total cohort. In subjects with diabetes (n = 79), high plasma betaine was associated with increased frequencies of events; significantly for heart failure, hazard ratio 3.1 (1.2-8.2) and all cardiovascular events, HR 2.8 (1.4-5.5). In subjects without diabetes (n = 396), low plasma betaine was associated with events; significantly for secondary myocardial infarction, HR 2.1 (1.2-3.6), unstable angina, HR 2.3 (1.3-4.0), and all cardiovascular events, HR 1.4 (1.0-1.9). In diabetes, high TMAO was a marker of all outcomes, HR 2.7 (1.1-7.1) for death, 4.0 (1.6-9.8) for myocardial infarction, 4.6 (2.0-10.7) for heart failure, 9.1 (2.8-29.7) for unstable angina and 2.0 (1.1-3.6) for all cardiovascular events. In subjects without diabetes TMAO was only significant for death, HR 2.7 (1.6-4.8) and heart failure, HR 1.9 (1.1-3.4). Adding the estimated glomerular filtration rate to Cox regression models tended to increase the apparent risks associated with low betaine.

Conclusions: Elevated plasma betaine concentration is a marker of cardiovascular risk in diabetes; conversely low plasma betaine concentrations indicate increased risk in the absence of diabetes. We speculate that the difference reflects control of osmolyte retention in tissues. Elevated plasma TMAO is a strong risk marker in diabetes.

No MeSH data available.


Related in: MedlinePlus

Hazard ratios.Hazard ratios for low (left) or high (middle) plasma betaine concentrations, and for high plasma trimethylamine N-oxide concentrations (right). Model 1: just low and high marker (2 variables). Model 2: baseline eGFR added to model. Model 3: Model 2 plus baseline LVEF and plasma NT-proBNP added to model. A: in subjects without diabetes; B: in subjects with diagnosed diabetes. Events: MI: secondary myocardial infarction; HF: hospitalization for heart failure; UA: unstable angina; All: all cardiovascular events. Left side: low plasma betaine; middle: high plasma betaine; right side: high plasma trimethylamine N-oxide. Vertical dotted red lines: hazard ratio of 1.0. Ratios significantly (p<0.05) different from 1 are shown in red. All ratios are shown with 95% confidence intervals.
© Copyright Policy
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4262445&req=5

pone-0114969-g003: Hazard ratios.Hazard ratios for low (left) or high (middle) plasma betaine concentrations, and for high plasma trimethylamine N-oxide concentrations (right). Model 1: just low and high marker (2 variables). Model 2: baseline eGFR added to model. Model 3: Model 2 plus baseline LVEF and plasma NT-proBNP added to model. A: in subjects without diabetes; B: in subjects with diagnosed diabetes. Events: MI: secondary myocardial infarction; HF: hospitalization for heart failure; UA: unstable angina; All: all cardiovascular events. Left side: low plasma betaine; middle: high plasma betaine; right side: high plasma trimethylamine N-oxide. Vertical dotted red lines: hazard ratio of 1.0. Ratios significantly (p<0.05) different from 1 are shown in red. All ratios are shown with 95% confidence intervals.

Mentions: Hazard ratios were estimated to give easy to interpret summaries of the risk associated with high or low (top or bottom quintile) plasma betaine and high (top quintile) plasma TMAO. When “high plasma betaine” (top quintile of gender-adjusted plasma betaine concentrations compared with all other results) and “low plasma betaine” (bottom quintile compared with all other results) were assessed as predictors of secondary outcomes in two-variable Cox proportional hazards regression models, the hazard ratios (Fig. 3) were concordant with the Kaplan-Meier analyses (Figs. 1 & 2). High plasma betaine appeared as a significant predictor in the subjects with diabetes, and despite the small numbers the hazard ratios were significant for hospital admission for heart failure (hazard ratio 3.1, CI 1.2–8.2; p = 0.019) and all events (hazard ratio 2.8, CI 1.4–5.5; p = 0.004). In contrast, in the subjects without diabetes, low plasma betaine was a risk marker, notably for secondary myocardial infarction (hazard ratio 2.1, CI 1.2–3.6; p = 0.009), unstable angina (hazard ratio 2.3, CI 1.3–4.0; p = 0.003) and all events (hazard ratio 1.4, CI 1.0–1.9; p = 0.031).


Betaine and Trimethylamine-N-Oxide as Predictors of Cardiovascular Outcomes Show Different Patterns in Diabetes Mellitus: An Observational Study.

Lever M, George PM, Slow S, Bellamy D, Young JM, Ho M, McEntyre CJ, Elmslie JL, Atkinson W, Molyneux SL, Troughton RW, Frampton CM, Richards AM, Chambers ST - PLoS ONE (2014)

Hazard ratios.Hazard ratios for low (left) or high (middle) plasma betaine concentrations, and for high plasma trimethylamine N-oxide concentrations (right). Model 1: just low and high marker (2 variables). Model 2: baseline eGFR added to model. Model 3: Model 2 plus baseline LVEF and plasma NT-proBNP added to model. A: in subjects without diabetes; B: in subjects with diagnosed diabetes. Events: MI: secondary myocardial infarction; HF: hospitalization for heart failure; UA: unstable angina; All: all cardiovascular events. Left side: low plasma betaine; middle: high plasma betaine; right side: high plasma trimethylamine N-oxide. Vertical dotted red lines: hazard ratio of 1.0. Ratios significantly (p<0.05) different from 1 are shown in red. All ratios are shown with 95% confidence intervals.
© Copyright Policy
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4262445&req=5

pone-0114969-g003: Hazard ratios.Hazard ratios for low (left) or high (middle) plasma betaine concentrations, and for high plasma trimethylamine N-oxide concentrations (right). Model 1: just low and high marker (2 variables). Model 2: baseline eGFR added to model. Model 3: Model 2 plus baseline LVEF and plasma NT-proBNP added to model. A: in subjects without diabetes; B: in subjects with diagnosed diabetes. Events: MI: secondary myocardial infarction; HF: hospitalization for heart failure; UA: unstable angina; All: all cardiovascular events. Left side: low plasma betaine; middle: high plasma betaine; right side: high plasma trimethylamine N-oxide. Vertical dotted red lines: hazard ratio of 1.0. Ratios significantly (p<0.05) different from 1 are shown in red. All ratios are shown with 95% confidence intervals.
Mentions: Hazard ratios were estimated to give easy to interpret summaries of the risk associated with high or low (top or bottom quintile) plasma betaine and high (top quintile) plasma TMAO. When “high plasma betaine” (top quintile of gender-adjusted plasma betaine concentrations compared with all other results) and “low plasma betaine” (bottom quintile compared with all other results) were assessed as predictors of secondary outcomes in two-variable Cox proportional hazards regression models, the hazard ratios (Fig. 3) were concordant with the Kaplan-Meier analyses (Figs. 1 & 2). High plasma betaine appeared as a significant predictor in the subjects with diabetes, and despite the small numbers the hazard ratios were significant for hospital admission for heart failure (hazard ratio 3.1, CI 1.2–8.2; p = 0.019) and all events (hazard ratio 2.8, CI 1.4–5.5; p = 0.004). In contrast, in the subjects without diabetes, low plasma betaine was a risk marker, notably for secondary myocardial infarction (hazard ratio 2.1, CI 1.2–3.6; p = 0.009), unstable angina (hazard ratio 2.3, CI 1.3–4.0; p = 0.003) and all events (hazard ratio 1.4, CI 1.0–1.9; p = 0.031).

Bottom Line: In subjects with diabetes (n = 79), high plasma betaine was associated with increased frequencies of events; significantly for heart failure, hazard ratio 3.1 (1.2-8.2) and all cardiovascular events, HR 2.8 (1.4-5.5).In subjects without diabetes (n = 396), low plasma betaine was associated with events; significantly for secondary myocardial infarction, HR 2.1 (1.2-3.6), unstable angina, HR 2.3 (1.3-4.0), and all cardiovascular events, HR 1.4 (1.0-1.9).Adding the estimated glomerular filtration rate to Cox regression models tended to increase the apparent risks associated with low betaine.

View Article: PubMed Central - PubMed

Affiliation: Clinical Biochemistry Unit, Canterbury Health Laboratories, Christchurch, New Zealand; Department of Pathology, University of Otago Christchurch, Christchurch, New Zealand.

ABSTRACT

Background: Betaine is a major osmolyte, also important in methyl group metabolism. Concentrations of betaine, its metabolite dimethylglycine and analog trimethylamine-N-oxide (TMAO) in blood are cardiovascular risk markers. Diabetes disturbs betaine: does diabetes alter associations between betaine-related measures and cardiovascular risk?

Methods: Plasma samples were collected from 475 subjects four months after discharge following an acute coronary admission. Death (n = 81), secondary acute MI (n = 87), admission for heart failure (n = 85), unstable angina (n = 72) and all cardiovascular events (n = 283) were recorded (median follow-up: 1804 days).

Results: High and low metabolite concentrations were defined as top or bottom quintile of the total cohort. In subjects with diabetes (n = 79), high plasma betaine was associated with increased frequencies of events; significantly for heart failure, hazard ratio 3.1 (1.2-8.2) and all cardiovascular events, HR 2.8 (1.4-5.5). In subjects without diabetes (n = 396), low plasma betaine was associated with events; significantly for secondary myocardial infarction, HR 2.1 (1.2-3.6), unstable angina, HR 2.3 (1.3-4.0), and all cardiovascular events, HR 1.4 (1.0-1.9). In diabetes, high TMAO was a marker of all outcomes, HR 2.7 (1.1-7.1) for death, 4.0 (1.6-9.8) for myocardial infarction, 4.6 (2.0-10.7) for heart failure, 9.1 (2.8-29.7) for unstable angina and 2.0 (1.1-3.6) for all cardiovascular events. In subjects without diabetes TMAO was only significant for death, HR 2.7 (1.6-4.8) and heart failure, HR 1.9 (1.1-3.4). Adding the estimated glomerular filtration rate to Cox regression models tended to increase the apparent risks associated with low betaine.

Conclusions: Elevated plasma betaine concentration is a marker of cardiovascular risk in diabetes; conversely low plasma betaine concentrations indicate increased risk in the absence of diabetes. We speculate that the difference reflects control of osmolyte retention in tissues. Elevated plasma TMAO is a strong risk marker in diabetes.

No MeSH data available.


Related in: MedlinePlus