Abstract
Introduction
Trimethylamine-N-oxide (TMAO) is a well characterized pro-atherogenic metabolite derived from the microbial processing of phosphatidylcholine and carnitine (usually present in red meat) and subsequent hepatic oxydation, which promotes endothelial dysfunction, platelet activation and thrombosis initiation. Its role concerning cerebral and cardiovascular adverse events has been assessed in various patient subpopulations but not for long term in patients with atrial fibrillation.
Methods
Baseline TMAO plasma levels were measured by high-performance liquid chromatography/mass spectrometry in plasmas of 2,379 subjects from our multicentric study. Among them, 1,722 participants at time of recruitment underwent brain MRI. Participants were stratified into TMAO tertiles and Cox PH models, linear mixed effect models or logistic mixed effect models were employed adjusting for several risk factors (age, sex, BMI, active and past smoke habit, cystatin c levels, heart failure, diabetes mellitus, hypertension, coronary artery disease and history of TIA/stroke). Subjects were prospectively followed with a median observation time of 4 years.
Results
Subjects in the highest tertile of TMAO were older (75.4 vs. 70.6 years in low tertile p<0.001) and had significantly more often comorbidities, (26.9% of subjects were diabetic vs. 9.1% in low tertile p<0.001), with higher BMI (28.1 vs 27.0, p<0.001) and worse renal function as assessed by serum cystatin C (1.46 vs 1.07, mg/dl; p<0.001). Heart failure was present in 37.9% participants in the upper compared to 15.8% in the lower tertile. (p<0.001). As shown in Figure 1, Kaplan Meier estimates showed increased cardiovascular mortality with increasing TMAO tertiles (p<0.0001). After adjustment for the abovementioned factors the upper tertile (T3) had an increased hazard ratio (HR) compared to the lowest one (HR 2.36 95% CI 1.56–3.58 p<0.01). Similar trends for global and ischemic stroke occurrences were not found although TMAO levels positively weakly correlated with NIHSS severity (Spearman's coefficient 0.31 p=0.02). Concerning brain MRI findings, TMAO tertiles identified individuals with different prevalence of small non-cortical infarcts (30.5%, 18.1% and 17.4% in high, middle and low tertiles respectively; p<0.001) and when present, larger white matter lesions volumes (5061 mm3, 4158 mm3 and 2970 mm3; p<0.001). After adjustment, the association with small non-cortical infarcts with TMAO levels remained significant in the highest tertile (T3) (OR 1.48 95% CI 1.07–2.05; p=0.02) and a trend towards larger white matter lesions volumes was observed (estimate 1307 95% CI −90–2705; p=0.07).
Conclusions
TMAO represents a robust prognostic independent biomarker identifying multimorbid, high risk patients for cardiovascular mortality and brain damage.
FUNDunding Acknowledgement
Type of funding sources: Public Institution(s). Main funding source(s): Swiss National Science Foundation (SNSF) and Theodor und Ida Herzog-Egli Foundation Figure 1. CV mortality according to TMAO Figure 2. Brain lesions assessment
Neonatal Brain MRI and Motor Outcome at School Age in Children with Neonatal Encephalopathy: A Review of Personal Experience
