Effect of TMAO on the incidence and prognosis of cerebral infarction: a systematic review and meta-analysis

Objective: This study aimed to evaluate the effect of trimethylamine oxide (TMAO) on the incidence and prognosis of cerebral infarction. Methods: We searched PubMed, Embase, and Cochrane databases for all clinical studies on the association of TMAO with cerebral infarction incidence and prognosis from inception to April 2023. A systematic review and meta-analysis were conducted using the meta-analysis of observational studies in epidemiology (MOOSE) declaration list. The Newcastle-Ottawa Scale (NOS) was used to assess the quality of the study. This study protocol was registered on the PROSPERO database with the ID: CRD42023459661. The extracted data included the OR value of the effect of TMAO on the incidence and prognosis of cerebral infarction, the HR value between TMAO and underlying diseases, the RR value, 95% confidence intervals, and the AUC value of TMAO in the prediction model of cerebral infarction. Results: Fifteen studies including 40,061 patients were included. All the patients were from China or Germany. The TMAO level was significantly correlated with the Modified Rankin Score (mRS) 3 months after the onset of cerebral infarction (OR, 1.581; 95% CI, 1.259-1.987; p < 0.01). The TMAO level was significantly correlated with the rate of first-time incidence and recurrence of cerebral infarction (OR, 1.208; 95% CI, 1.085-1.344; p < 0.01 and HR, 1.167; 95% CI, 1.076-1.265; p < 0.01, respectively). The TMAO level was also highly correlated with disease severity at onset (National Institutes of Health Stroke Scale, NIHSS >5) (OR, 5.194; 95% CI, 1.206-22.363; p < 0.05), but had no significant correlation with mortality after cerebral infarction (p > 0.05). Correlation analysis of TMAO with underlying diseases in the population indicated that TMAO had a significant correlation with histories of hypertension, diabetes mellitus, coronary artery disease, and cerebral infarction (p < 0.05), but not with hyperlipidemia (p > 0.05). Six risk prediction models of TMAO for cerebral infarction reported in four studies were systematically evaluated; five of them had good predictive value (AUC >= 0.7). Conclusion: TMAO is an independent risk factor affecting the onset, prognosis, and severity of cerebral infarction.