Elevated plasma dimethylglycine predicts cardiovascular risk only when plasma methylmalonic acid is also high

Abstract

Background:
We have previously identified elevated plasma dimethylglycine (DMG) as an individual risk predictor for acute myocardial infarction (AMI) and mortality [1], and activation of peroxisome proliferator-activated receptor (PPAR) α was suggested as an underlying mechanism. Notably, PPARα agonism also increases systemic concentrations of methylmalonic acid (MMA) in rodents, making both DMG and MMA potential endogenous biomarkers of PPARα activation.
Objective: To investigate whether the risk associations between DMG and risk of incidence AMI and cardiovascular and total mortality were modified by plasma MMA.

Design:
In this prospective cohort study, 4154 patients with suspected stable angina pectoris (72% men, median age 62) were followed for a median of 4.7 y. The patients were divided into four groups according to baseline median concentrations of plasma DMG and MMA. Risk associations were evaluated by Cox regression, usingthe low DMG/low MMA group was as reference. The model was adjusted for age, gender, diabetes, hypertension and smoking.

Results:
During follow-up 343 (8.2%) patients experienced an AMI, and 306 (7.4%) patients died, whereof 166 (4.0%) from CVD. The hazard ratios (HRs) (95% confidence intervals (CI)) were 1.90 (1.42-2.53; p<0.001) for AMI, 2.85 (1.78-4.54; p<0.001) for CVD death and 2.00 (1.45-2.77; p<0.001) for total death when comparing the high DMG/high MMA group against the low DMG/high MMA group. Elevated plasma DMG was not associated with increased risk of the three end points in patients with low plasma MMA, whereas patients with elevated MMA had increased risk of total death also when DMG was low (HR (95% CI) 1.49 (1.02-2.18; p=0.038).

Conclusion:
In this prospective cohort study elevated plasma DMG was associated with increased risk of AMI and cardiovascular and total mortality only among those who also had high plasma MMA. Elevated plasma MMA was associated with increased total mortality also when plasma DMG was low. This suggests that risk previously demonstrated with plasma DMG among cardiovascular patients is modified,and perhaps better explained by plasma MMA.
1. Svingen GF, Schartum-Hansen H, Ueland PM et al: Elevated plasma dimethylglycine is a risk marker of mortality in patients with coronary heart disease. European journal of preventive cardiology 2014.

Full text

Elevated plasma dimethylglycine
predicts cardiovascular risk only
when plasma methylmalonic acid is
also high
Vegard Lysne
Gard Frodahl Tveitevåg Svingen
Elin Strand
Per Magne Ueland
Eva Ringdal Pedersen
Ottar Nygård

Elevated plasma dimethylglycine (DMG)
and increased cardiovascular risk
Hazard ratio
Cardiovascular mortality
SAP AMI
Total mortality
Plasma dimethylglycine, µmol/L
Svingen, GFT. et al. Arterioscler Thromb Vasc Biol. 2013. Svingen, GFT. et al. Eur j prev cardiol. 2014

PPARα activation and elevated plasma
DMG
Sheikh, K. et al. Am J Physiol Endocrinol Metab(2007). Chu, R. et al. Mol Cell Biol(2004). Wrzesinski, K. et al. J Proteomics.(2013)
Choline
Hcy
Met Glycine Serine DMG
Betaine PPARα ↑
Sarcosine
÷ ÷

PPARα activation increases both DMG
and methylmalonic acid (MMA)
GASTRIC
DMG MMA
Standardized
mean
difference
Control group
Control group
-1
0
1
2
3
4
PPAR 2014 TTA
PPARγ (Roziglitazone)
PPARα (WY 14.643)
DMG MMA

MMA, DMG and risk
PPARα ↑
MMA ↑
DMG ↑ Risk ↑
?

MMA, DMG and risk
Population
4154 patients with
suspected stable
angina pectoris,
median follow-up
4.7y
End points
•Acute
myocardial
infarction
•Cardiovascular
mortality
•Total mortality
Groups
Plasma MMA
and DMG
above/below
median

Incidence & mortality rates
Total
population
Low MMA High MMA
Low DMG
High DMG
Low DMG
High
DMG
n
4154 1360 960 720 1114
MMA (
pmol/L) 160 130 140 210 210
DMG (µmol/L)
4.1 3.3 5.0 3.5 5.0
Acute
myocardial
infarction
n = 343
(8.2 %) (5.6 %) (7.2 %) (6.1 %) (13.8 %)
Cardiovascular
mortality
(1.8 %) (2.4 %) (3.5 %) (8.4 %)
Total
mortality n = 306
(7.4 %) (4.1 %) (5.3 %) (7.5 %) (13 %)

Low MMA / Low DMG
Low MMA / High DMG
High MMA / Low DMG
High MMA / High DMG
The risk is confined to those with
high MMA / high DMG
Total
mortality *
***
***
***
Cardiovascular
mortality
Acute
myocardial
infarction
*** *
Adjusted for age, gender, diabetes,
hypertension and smoking.
p < 0.05, p < 0.001
1 3 4 2 5

So, what is different about this group?
Total
population
Low MMA High MMA
Low DMG
High DMG
Low DMG
High
DMG
p*
Male
72 % 73.5 % 74.5% 69.2 % 69.7 % 0.018
Age
61.7
(10.4)
58.9 (9.1) 61.0 (9.4) 64.1 (9.2) 66.3 (9.7)
<0.001
BMI
26.8 (4.0)
27.2 (3.6)
26.9 (3.8)
26.7 (3.7)
26.7 (3.7)
0.066
Smoking
31.7 % 32.6 % 36.8 % 23.6 % 31.3 %
<0.001
Diabetes
11.8 % 12.8 % 9.5 % 10.6 % 13.6 % 0.014
Hypertension
46.8 % 41.9 % 43.0 % 45.7 % 56.6 %
<0.001
1
-vessel
disease
23.2 % 24.9 % 26.5 % 21.8 % 19.3 %
<0.007
2
-vessel
disease
22.3 % 22.4 % 21.4 % 23.5 % 22.3 % 0.784
3
-vessel
disease
29.3 % 26.2 % 27.6 % 26.8 % 36.4 %
<0.001
Estimated
GFR
87.8 (17.3)
95.8 (12.3)
90.7 (14.1)
85.7 (14.5)
77.0 (20.4)
<0.001
Total
cholesterol
5.1 (1.2) 5.2 (1.2) 4.9 (1.1) 5.1 (1.1) 5.1 (1.3) 0.004
*Between-group comparisons. One-Way ANOVA for means, Chi-Square for counts

Total
population
Low MMA High MMA
Low DMG
High DMG
Low DMG
High
DMG
p*
Age
61.7 (10.4)
58.9 (9.1)
61.0 (9.4)
64.1 (9.2)
66.3 (9.7)
<0.001
Hypertension
46.8 % 41.9 % 43.0 % 45.7 % 56.6 %
<0.001
3
-vessel
disease
29.3 % 26.2 % 27.6 % 26.8 % 36.4 %
<0.001
Estimated
GFR
87.8 (17.3)
95.8 (12.3)
90.7 (14.1)
85.7
(14.5)
77.0 (20.4)
<0.001
They are older and sicker
*Between-group comparisons. One-Way ANOVA for means, Chi-Square for counts

Further adjustment
No attenuation of
the risk estimates
Fasting status
BMI
B-vitamin treatment
Blood lipids
Statin treatment
Plasma tHcy
Extent of coronary heart disease

Further adjustment
Fasting status
BMI
B-vitamin treatment
Blood lipids
Statin treatment
Plasma tHcy
Extent of coronary heart disease
Estimated GFR
Only slight
attenuation of risk
estimates

To summarize
vegard.lysne@helse-bergen.no
The cardiovascular risk
associations reported with
elevated plasma dimethylglycine
is confined to those who also
have elevated plasma
methylmalonic acid.

To summarize
The cardiovascular risk
associations reported with
elevated plasma dimethylglycine
is confined to those who also
have elevated plasma
methylmalonic acid.
Based on animal data, there is
reason to suspect PPARα
activation as a possible
underlying mechanism.
vegard.lysne@helse-bergen.no