Gut microbial metabolite trimethylamine N-oxide as a novel predictor for adverse cardiovascular events after PCI: a systematic review and dose-response meta-analysis

Abstract

Background

Cardiovascular diseases are the leading cause of mortality worldwide, with acute coronary syndrome (ACS) being particularly fatal. Percutaneous coronary intervention (PCI) is a key treatment for ACS; however, major adverse cardiovascular events (MACE) frequently occur postoperatively. Trimethylamine N-oxide (TMAO), a gut microbiota-derived metabolite, has been proposed as an emerging risk factor for cardiovascular disease. This study aims to systematically evaluate TMAO’s predictive value for MACE post-PCI and explore its dose-response relationship.

Methods

A comprehensive literature search was conducted in four databases (PubMed, Web of Science, Embase, and the Cochrane Library), including retrospective or prospective cohort studies involving patients undergoing PCI. The primary outcome was MACE, and the secondary outcome was all-cause mortality. A dose-response analysis was conducted using a restricted cubic spline model to explore potential nonlinear associations between TMAO levels and outcomes. Heterogeneity was assessed using the Cochrane Q test and the I² statistic. Subgroup analysis and meta-regression were performed to identify sources of heterogeneity.

Results

Eleven studies (comprising 13 independent cohorts) with 11,279 participants were included. Pooled analysis showed a significant association between elevated plasma TMAO levels and an increased risk of MACE after PCI (HR: 1.99, 95%CI: 1.68–2.35, 95%PI: 1.64–2.40, I² = 0%, p < 0.00001). Similarly, elevated plasma TMAO levels were significantly associated with an increased risk of all-cause mortality after PCI (HR: 1.76, 95%CI: 1.32–2.35, 95%PI: 0.79–3.90, I² = 65.1%, p < 0.00001). The dose-response analysis did not reveal a nonlinear relationship between TMAO and MACE or all-cause mortality. The linear model showed that each 1 µmol/L increase in plasma TMAO was associated with an 8.95% increased hazard of MACE (HR = 1.0895, 95%CI: 1.03–1.15), while all-cause mortality increased by 4% (HR = 1.04, 95%CI: 0.99–1.09).

Conclusions

This study demonstrates that elevated plasma TMAO levels are significantly associated with an increased risk of MACE and all-cause mortality after PCI, with a dose-dependent effect on MACE risk. As a potential biomarker, TMAO may be used to predict the risk of adverse cardiovascular events after PCI, and future studies should further validate its clinical utility.

Registration

PROSPERO CRD42024557486.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12937-025-01159-9.

Introduction

Cardiovascular disease (CVD) ranks as the leading cause of mortality worldwide, far exceeding cancer and other diseases, accounting for over 40% of disease-related deaths [1], with acute coronary syndrome (ACS) being the most severe type [2]. With the increasing burden of ischemic heart disease, percutaneous coronary intervention (PCI) has become the primary strategy for treating ACS [3]. However, due to specific pathophysiological conditions such as endothelial injury, PCI-treated patients remain at risk of developing in-stent restenosis (ISR) [4], no-reflow phenomenon (NR) [5], myocardial ischemia-reperfusion injury (MIRI) [6], and malignant ventricular arrhythmias [7]. These complications can further contribute to adverse cardiovascular events [8] and pose a significant threat to survival and quality of life. Therefore, identifying patients who undergo PCI but remain at high risk for adverse cardiovascular events is crucial.

Trimethylamine N-oxide (TMAO) arises from the metabolism of dietary choline, betaine, and L-carnitine, predominantly synthesized by gut bacteria and further metabolized into TMAO in the liver by flavin-containing monooxygenase 3 (FMO3) [9]. Recently, TMAO has garnered significant attention as an emerging cardiovascular risk factor. Studies indicate that elevated plasma TMAO levels are closely linked to cardiovascular events, including atherosclerosis, thrombosis, and heart failure [10]. Potential mechanisms include TMAO’s effects on cholesterol metabolism, bile acid homeostasis, platelet activation, and inflammatory pathways [11]. Notably, recent studies have highlighted the emerging role of TMAO in predicting outcomes after PCI. Tan et al. [12] demonstrated a significant association between plasma TMAO and new atherosclerosis and plaque rupture in patients experiencing very late stent thrombosis (VLST). Zhao et al. [13] reported a close relationship between pre-PCI plasma TMAO levels and atherosclerotic burden in patients. However, Yasushi et al. [14]found no significant correlation between acute-phase TMAO levels and the incidence of MACE in STEMI patients undergoing PCI. Previous studies focused on TMAO in general or cardiovascular populations, but systematic analyses targeting PCI patients remain limited. Moreover, the predictive value of TMAO for MACE following PCI remains inconsistent. Hence, we conducted a meta-analysis to investigate the association between TMAO and MACE post-PCI through the synthesis of current research, offering insights for future studies and clinical applications.

Methods

Search strategy

This review adheres to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines, as well as the most up-to-date guidelines for systematic reviews and meta-analyses of prognostic factor research [15, 16]. The PRISMA checklist is provided in Supplementary Table (1) This systematic review protocol has been registered in the International Prospective Register of Systematic Reviews (PROSPERO), with the registration number CRD42024557486. The meta-analysis encompassed cohort studies examining TMAO’s predictive utility for MACE following PCI. Articles were retrieved from four English databases (PubMed, Web of Science, Embase, and Cochrane Library) up to December 25, 2024, using keywords including “trimethylamine oxide,” “trimethylammonium oxide,” “trimethylamine N-oxide,” “TMAO,” “coronary artery disease,” “myocardial infarction,” “acute coronary syndrome,” “percutaneous coronary intervention,” “STEMI,” “NSTEMI,” “PCI,” “AMI,” “ACS,” and “major adverse cardiovascular events”. No language restrictions were applied, and we also conducted a manual search, including screening the reference lists of previous systematic reviews and meta-analyses to retrieve additional relevant articles for further analysis. Detailed search strategies are shown in Supplementary Table (2) MACE is defined as a composite of clinical events, encompassing endpoints related to “safety” (death, myocardial infarction, stroke) and “efficacy” (target vessel revascularization, any repeat revascularization, or re-hospitalization). Based on the above information and referencing recommendations from multiple authoritative sources [17], we defined MACE as all-cause death, non-fatal myocardial infarction, non-fatal stroke, revascularization, and cardiac rehospitalization. Our study design and analytical methods were structured according to this definition, and findings were interpreted and analyzed correspondingly.

Study selection

Following our meta-analysis objectives, we established a “Population, Index prognostic factors, Control prognostic factors, Outcome, Time, Environment” (PICOTS) framework, adapted from the guidelines proposed by Riley et al. [16].

The inclusion criteria for this study were as follows: (1) Study type: retrospective or prospective cohort studies; (2) Prognostic factor: Fasting plasma TMAO concentration was measured before treatment. (3) Study subjects: patients undergoing PCI; (4) The primary outcome measure was MACE, defined as a composite of all-cause death, non-fatal myocardial infarction, non-fatal stroke, revascularization, and cardiac rehospitalization; (5) Secondary outcome measure: all-cause death; (6) Dose-response analysis required three or more TMAO categories (with 95% confidence intervals [CIs] or information to calculate these intervals), and quantitative TMAO measurements and the number of subjects in each group must be provided. (6)

Only literature reporting adjusted effect estimates was included.

The exclusion criteria for this study were as follows: (1) Cross-sectional studies, reviews, preclinical studies, and studies irrelevant to the purpose of the Meta-analysis; (2) Animal experiments, conference abstracts, case reports, and duplicate publications; (3) Studies lacking data on plasma TMAO levels and their association with outcomes; (4) Studies where the outcomes of interest were not reported or could not be extracted or calculated from the published results.

Data extraction

Initially, duplicate articles were excluded, and the remaining retrieved articles were independently screened by two authors. Articles meeting the inclusion and exclusion criteria were comprehensively screened based on reviewing their titles and abstracts.

The extracted data include the surname of the first author, year of publication, country or region where the study was conducted, number of participants and cases, follow-up time, mean or median TMAO concentration among total study participants, or the highest category, range, or interquartile range (IQRs) of TMAO risk assessment, plasma TMAO measurement methods, adjustment for variables and other study characteristics. The original article used the tertile or quartile as a classification of TMAO levels. We extracted the number of cases/N, exposure level, HR, and 95% CI for TMAO categories, as well as confounding factors in the adjusted model. We only included publications that could calculate HR and 95% confidence intervals using published methods [18].

Quality assessment

Two authors independently evaluated the methodological quality of the included studies using the Quality in Prognosis Factor Studies (QUIPS) tool [19]. The QUIPS tool evaluates six domains: study participation, attrition, prognostic factor measurement, outcome measurement, confounding adjustment, and statistical analysis and reporting. The risk of bias for each domain was categorized as low, moderate, or high [20]. Any discrepancies between authors were addressed in a consensus meeting, and unresolved disagreements were settled through discussion with a third author.

Statistical analysis

Statistical analysis was performed using Stata/SE 15.1 software and R 4.4.1. The high TMAO concentration group referred to the highest quartile, triquartile, or higher, while the low concentration group referred to the lowest or lower. The random-effects model was employed to compute HR and 95% CI for MACE and all-cause mortality risk associated with the highest TMAO concentration category versus the lowest. This model addresses potential heterogeneity across included studies [21]. Heterogeneity among the included cohort studies was assessed using Cochrane’s Q test, and the I² statistic was calculated [22]. Heterogeneity was deemed significant if I² exceeded 50%. Subgroup analyses were conducted based on disease status, nationality, age, prevalence of hypertension, diabetes, hyperlipidemia, smoking rate, and follow-up duration. Meta-regression was performed to further explore potential sources of heterogeneity. Sensitivity analysis was conducted using a leave-one-out approach to test the robustness of the results [23]. A p-value < 0.05 was considered statistically significant. Potential publication bias was evaluated through visual inspection of funnel plot symmetry and Egger’s test [24].

Dose-response analysis requires extracting the median or mean TMAO concentration for each category. If the average TMAO number was not reported for each group Or median, the median of each Interzone group was taken as the mean dose of the group. For open intervals of TMAO concentrations, we assume the interval is similar to the adjacent one and calculate its midpoint [25]. The “mvmeta” command in Stata was used, employing restricted cubic splines with three knots, to explore the potential nonlinear association between TMAO levels in PCI patients and the risk of MACE and all-cause mortality [26]. A nonlinear random-effects model was constructed using a combination of generalized least squares and multivariate maximum likelihood approaches [27]. The model included all categories of TMAO, TMAO HR levels, and the number of patients in each study and group. If the non-linear model tests p < 0.05, a dose-response relationship with curve nonlinearity is considered. If the non-linear model test p > 0.05, a linear model is used for curve fitting, displaying the dose-response results for every 1 µmol/L increase in plasma TMAO.

Results

Study selection and study characteristics

Using a predefined search strategy, we retrieved 1,124 records from four databases. Four additional studies were identified via manual searching. The study selection process flowchart and detailed exclusion reasons are depicted in Fig. 1. Initially, 657 duplicate publications were removed using EndNote X7. Subsequently, 368 articles were excluded due to being animal experiments, case reports, reviews, or meta-analyses. Finally, 103 publications underwent full-text review. Following further screening, 11 studies [14, 28–37] (comprising 13 independent cohorts) were included in the subsequent meta-analysis. The studies encompassed a total of 11,279 participants, with average/median ages ranging from 56.4 to 74 years. The studies were conducted across 8 countries: Poland, China, Japan, the USA, Spain, Norway, Switzerland, and New Zealand. All included studies used plasma samples, of which 7 studies used liquid chromatograph-tandem mass spectrometry (LC-MS/MS), 2 studies used liquid chromatograph-mass spectrometry (LC-MS), and 1 study used rapid separation liquid chromatograph-Quadrupole time-of-flight mass spectrometry (RRLC-QTOF/MS) for plasma TMAO analysis [38]. The median plasma TMAO levels in the included studies ranged from 2.4 to 9.98 µmol/L. The follow-up duration for calculating HRs of MACE and all-cause mortality ranged from 6 months to 9.8 years. The follow-up time for MACE and all-cause mortality ranged from 6 months to 9.8 years. The characteristics of the included studies are presented in Table 1, while the population characteristics are provided in Supplementary Table 3. Based on the bias risk assessment using the QUIPS tool, two studies [28, 30] were classified as “high risk of bias,” six studies [29, 32–35, 37]) as “moderate risk of bias”, and three studies [14, 31, 36] as “low risk of bias”. The detailed assessment results are presented in Table 2.

Fig. 1.

Flow diagram of the study selection process

Table 1.

Characteristics of included studies

Study	Country	Year	involved centers	Disease status	TMAO (umol/L)	TMAO measure method	Sample Source	Sample size	Follow-up (years)	Outcome
Ceren Eyileten [28]	Poland	2021	Single	ACS	4.7(3.1–8.3)	LC-MS/MS	Serum	292	7	MACE
Tan [29]	China	2021	Single	ACS	2.40 (1.40–4.05)	LC-MS/MS	Serum	444	0.5	MACE all-cause mortality
Yasushi Matsuzawa [14]	Japan	2019	Single	ACS	6.76 (3.82–12.53)	LC-MS	Serum	112	5.4	MACE
Tang [30]	USA	2017	Single	CAD	4.4 (2.8–7.7)	LC-MS/MS	Serum	1216	5	MACE all-cause mortality
Raul Sanchez-Gimenez [31]	Spain	2022	Single	ACS	9.98 (7.42–19.19)	LC-MS/MS	Serum	309	6.7	MACE
Zhaoshuhong [32]	China	2023	Single	ACS	High TMAO (> 2.86)	RRLC-QTOF/MS	Serum	1004	1	MACE all-cause mortality
Nan Li [33]	China	2022	Single	ACS	6.7 (4.0-11.7)	LC-MS/MS	Serum	985	1.96	MACE all-cause mortality
Espen Ø. Bjørnestad [34]	Norway	2022	Single	CAD	5.7(3.6–9.7)	LC-MS/MS	Serum	4132	9.8	all-cause mortality
Li [35]	USA	2017	Single	ACS	4.28 (2.55–7.91)	LC-MS/MS	Serum	530	7	MACE all-cause mortality
Li [35]	Swiss	2017	multi- centre	ACS	2.87 (1.94–4.85)	LC-MS/MS	Serum	1683	1	MACE all-cause mortality
Michael Lever (A) [36]	New Zealand	2014	Single	ACS	4.80 (3.00-9.10)	LC-MS	Serum	396	4.96	MACE all-cause mortality
Michael Lever (B) [36]	New Zealand	2014	Single	ACS	7.50 (4.40–12.10)	LC-MS	Serum	79	4.82	MACE all-cause mortality
Zhaoxiaoxiao [37]	China	2024	Single	ACS	3.82 ± 3.36	LC-MS/MS	Serum	97	2.03	MACE

Abbreviations: TMAO, Trimethylamine-N-Oxide; ACS, acute coronary syndrome; CAD, coronary artery disease; LC-MS/MS, stable isotope dilution liquid chromatography with online tandem mass spectrometry; RRLC-QTOF/MS, rapid-resolution liquid chromatography quadrupole time-of-flight mass spectrometry; LC-MS, Liquid Chromatography-Mass Spectrometry

Table 2.

Risk of Bias using the QUIPS tool for assessing bias in prognostic studies

Study ID	Study Participation	Study attrition	Prognostic factor measurement	Outcome measurement	Adjustment for other prognostic factors	Statistical analysis and reporting	Overall rating
Ceren Eyileten 2021	low	high/moderate	low	high/moderate	high	high/moderate	HIGH
Tan 2021	moderate	low	moderate	moderate	low/moderate	moderate	MODERATE
Yasushi Matsuzawa 2019	low	low	low	low	low/moderate	low	LOW
Tang 2017	low	high/moderate	low	high/moderate	high	high/moderate	HIGH
Raul Sanchez-Gimenez 2022	low	low	low	low	low/moderate	low/moderate	LOW
Zhao 2023	low/moderate	low/moderate	low	low	moderate	moderate	MODERATE
Nan Li 2022	low	low/moderate	low	low	high/moderate	low	MODERATE
Espen Ø. Bjørnestad 2022	low	low/moderate	low	low	moderate	moderate	MODERATE
Li 2017	low	low	low/moderate	low	moderate	high/moderate	MODERATE
Michael Lever 2014	low	low	low	low	low/moderate	low	LOW
Zhaoxiaoxiao 2024	low	low	low	low	moderate	moderate	MODERATE

Major adverse cardiovascular events

The association between TMAO and postoperative MACE in PCI patients was systematically analyzed across ten studies (including 11 independent cohorts) comprising a total of 6,617 participants. The aggregated results show a significant positive correlation between elevated plasma TMAO concentration and increased risk of postoperative MACEs in PCI patients (HR: 1.99, 95%CI: 1.68–2.35, 95%PI: 1.64–2.40, I²=0%, p < 0.00001) (Fig. 2). Subgroup analyses were conducted based on disease status, nationality, age, prevalence of hypertension, diabetes, hyperlipidemia, smoking rate, and follow-up duration. The results demonstrated no significant impact on the overall stability of the findings (Supplementary Table 4).

Fig. 2.

Forest plots for the meta-analysis of the association between TMAO and risk of subsequent MACE in patients with PCI

All-cause mortality

The relationship between TMAO and all-cause mortality in PCI patients was systematically analyzed across seven studies (including 9 independent cohorts) comprising a total of 10,469 participants. The aggregated results from the random effects model show a significant positive correlation between elevated plasma TMAO levels and increased risk of all-cause mortality in PCI patients (HR: 1.76, 95%CI: 1.32–2.35, 95%PI: 0.79–3.90, I²=65.1%, p < 0.00001) (Fig. 3). Subgroup and meta-regression analyses suggested that disease status, nationality, and hypertension prevalence might be potential sources of heterogeneity (Supplementary Table 5).

Fig. 3.

Forest plots for the meta-analysis of the association between TMAO and risk of all-cause mortality in patients with PCI

Publication bias

Publication bias in studies on MACE outcome measures was assessed. A visual inspection of the funnel plot indicated symmetry, and both Egger’s test (p = 0.087) and Begg’s test (p = 0.276) showed no significant publication bias (Fig. 4). Less than 10 studies were available for assessing publication bias about TMAO and all-cause mortality risk, hence no publication bias assessment was conducted.

Fig. 4.

Funnel plots for the meta-analysis of TMAO and MACE risk

Sensitivity analysis

Sensitivity analysis for MACE and all-cause mortality showed that removing one study at a time still resulted in reliable overall risk estimates (Fig. 5).

Fig. 5.

A: Sensitivity analysis for TMAO and risk of subsequent MACE in patients with PCI; B: Sensitivity analysis for TMAO and risk of subsequent all-cause mortality in patients with PCI

Dose-response meta-analysis of associations between circulating TMAO concentrations and MACE and all-cause mortality

Dose-response analysis of TMAO and MACE used 16 h estimates from 5 studies, while analysis of TMAO and all-cause mortality used 10 h estimates from 3 studies. Each study’s TMAO concentration included at least 3 categories. The dose-response analysis revealed no nonlinear associations between TMAO and MACE or all-cause mortality (p = 0.07 and p = 0.749, respectively)(Fig. 6). The linear model showed that each 1 µmol/L increase in plasma TMAO was associated with an 8.95% increased hazard of MACE (HR = 1.0895, 95%CI: 1.03–1.15), while all-cause mortality increased by 4% (HR = 1.04, 95%CI: 0.99–1.09).

Fig. 6.

(a) Dose-response plot of TMAO and MACE in patients with PCI. (b) Dose-response plot of TMAO and all-cause mortality in patients with PCI. (Used a restricted cubic splines in a non-linear random-effects model, The solid line and the long dashed line represents the estimated HR and its 95% CI of the nonlinear.relationship, while the short dashed line represents the linear relationship)

Discussion

This meta-analysis examined the relationship between TMAO levels and the risk of postoperative MACE and all-cause mortality in PCI patients. Results indicated that patients with high plasma TMAO levels had a 1.99-fold increased risk of MACE and a 1.76-fold increased risk of all-cause mortality compared to those with low TMAO levels. Dose-response analysis showed that each 1 µmol/L increase in plasma TMAO was associated with an 8.95% increased hazard of MACE, while all-cause mortality increased by 4%. These findings suggest that TMAO can serve as a marker for identifying high-risk populations after PCI.

The results of this study indicate that high TMAO levels are significantly associated with the occurrence of MACE after PCI, with a pooled HR of 1.99 (95%CI: 1.68–2.35, 95%PI: 1.64–2.40, I² = 0%). Furthermore, no significant heterogeneity was observed in the pooled analysis of MACE. Sensitivity analysis indicated that the overall risk estimate remained robust after removing any single study, suggesting a consistent association among studies, which may be attributed to standardized PCI protocols and outcome definitions. For all-cause mortality, the pooled HR was 1.76 (95%CI: 1.32–2.35, 95%PI: 0.79–3.90, I² = 65.1%), suggesting a significant association between high TMAO levels and increased all-cause mortality. However, the 95% prediction interval was relatively wide, with moderate heterogeneity (I² = 65.1%), suggesting potential uncertainty in this association in future studies. To explore potential sources of heterogeneity, we conducted subgroup and meta-regression analyses. The results indicated that study location, hypertension, and disease type might contribute to heterogeneity. Previous studies have demonstrated that hypertension [39] can lead to alterations in gut microbiota. Similarly, dietary differences between Eastern and Western populations may contribute to variability in TMAO distribution across different study locations [40]. Therefore, although the meta-analysis supports the association between elevated TMAO levels and all-cause mortality, the high heterogeneity suggests that further high-quality prospective studies are needed to validate the stability of this relationship.

PCI is widely used in clinical practice due to its advantages of minimal trauma and rapid recovery [41]. However, various pathogenic risk factors contribute to the occurrence of adverse cardiovascular events such as acute heart failure [42], malignant arrhythmias [43–45], cardiogenic shock [46], and all-cause mortality [47] after PCI. Therefore, identifying prognostic predictors in PCI-treated patients is crucial for achieving personalized treatment and improving risk stratification [48]. Currently, clinical risk prediction for MACE in PCI patients commonly relies on scoring systems such as the GRACE score, SYNTAX score, and Gensini score [49–51]. However, these scoring systems do not consider metabolic factors and inflammatory markers, often resulting in inadequate identification of high-risk patients. Numerous studies have shown that gut microbiota and its metabolites, such as TMAO, bile acids (BAs), and short-chain fatty acids (SCFAs), are involved in the occurrence and development of cardiovascular adverse events through various molecular pathways [52]. TMAO, a hallmark product of gut microbiota-related metabolites, is closely associated with the occurrence of MACEs in patients [53]. Animal experiments demonstrate that elevated plasma TMAO levels increase the risk of atherosclerotic cardiovascular diseases [54] and thrombosis [55], thereby amplifying the likelihood of cardiovascular adverse events. A study by Zhu et al. further suggested that high TMAO levels increase the incidence of MACEs in patients with myocardial infarction [56]. A meta-analysis conducted in 2017 found that participants with higher TMAO levels had a 62% increased risk of MACEs and a 63% higher risk of all-cause mortality [57]. Compared with other traditional predictive factors, TMAO also has significant value in predicting adverse cardiovascular events after PCI. The study by Tang et al. showed that TMAO exhibits superior predictive capability for adverse cardiovascular events compared to LDL, cholesterol, and C-reactive protein [53]. Additionally, studies have indicated that TMAO is closely associated with traditional risk-scoring systems. A prospective study by Senthong et al. found that TMAO levels are not only an independent predictor of adverse cardiovascular events in patients with coronary artery disease but also positively correlate with the SYNTAX score [58]. Tan et al. showed that incorporating plasma TMAO levels into the Global Registry of Acute Coronary Events (GRACE) risk score could more accurately predict the occurrence of cardiovascular events in PCI patients [29]. In conclusion, the results of this study indicate that plasma TMAO levels can predict the occurrence of MACE and all-cause mortality after PCI. In the future, existing prognostic scoring systems can be further compared with TMAO levels to comprehensively evaluate the potential impact of TMAO on the incidence of adverse cardiovascular events in patients undergoing PCI treatment.

Further dose-response analysis indicates a linear correlation between TMAO levels and both MACEs and all-cause mortality. Previous studies have demonstrated a risk relationship between TMAO and cardiovascular adverse events. In 2017, Schiattarella et al. [59]found a positive dose-dependent relationship between high TMAO levels and increased cardiovascular risk and mortality; for every 10 µmol/L increase in plasma TMAO, the relative risk (RR) of all-cause mortality increased by 7.6%. Li et al. demonstrated a dose-dependent relationship between TMAO levels and the occurrence of MACEs in ACS patients undergoing PCI [35]. However, in 2020, a meta-analysis by Yao et al. showed a “J-shaped” association between TMAO concentration and MACE incidence (P = 0.033) [60], When the TMAO concentration increased to 3.9 µmol/L, its adverse effects on the cardiovascular system began to appear. The results of this study are consistent with most other studies [57, 59, 61], showing a linear correlation between plasma TMAO concentration and the risk of MACE and all-cause mortality in PCI patients. For every 1µmol/L increase in TMAO concentration, the relative risk of MACE increases by 8.95%, and the relative risk of all-cause mortality increases by 4%. The main reason is that this study adopted stricter inclusion criteria and primarily focused on a specific population undergoing PCI treatment. In contrast, Yao et al.‘s study focused on adverse cardiovascular event outcomes, including various ACS patients such as STEMI, NSTEMI, and unclassified ACS. Additionally, there is no unified standard reference value for plasma TMAO levels, and different studies may use varying cut-off points to define elevated TMAO levels, often leading to inconsistencies in observed results. When aggregating data from multiple studies using different methods, this inconsistency may lead to a “J-shaped” association.

The potential mechanism through which TMAO contributes to the occurrence of cardiovascular adverse events after PCI is likely related to its pro-thrombotic effects. Although PCI can effectively restore blood flow by opening the infarct-related artery, the procedure induces endothelial injury and stent implantation, which can create a prothrombotic microenvironment [62]. Animal studies have shown that TMAO enhances platelet hyperreactivity by activating Ca²⁺ -dependent signalling pathways, inducing intracellular calcium overload, thereby promoting abnormal platelet aggregation in response to adenosine diphosphate (ADP), thrombin, and collagen [55, 63]. Witkowski et al. also demonstrated that TMAO directly promotes arterial thrombosis by upregulating tissue factor and thrombin generation [64]. This mechanism may increase the risk of in-stent thrombosis, a key trigger of MACE after PCI [65]. These findings align with the results of the present study’s meta-analysis: for every 1 µmol/L increase in plasma TMAO, the risk of MACE increases by 8.95% demonstrating a linear dose-response relationship.

Currently, the methods to control TMAO levels have become a significant focus in ongoing research. Wang et al. [66] found that long-term consumption of choline-rich foods, such as red meat and fish, can significantly increase TMAO levels in the body, while dietary fibre consumption can reduce plasma TMAO levels by lowering total choline intake [67]. Tenore et al. [68] effectively controlled plasma TMAO levels by reshaping the gut microbiota with probiotics. In terms of drug therapy, Brunt et al. [69] inhibited the activity of gut microbiota in elderly mice using broad-spectrum antibiotics, which reduced TMAO levels and reversed age-related endothelial dysfunction and aortic sclerosis. However, Tang et al. [53] found that while antibiotic use reduced plasma TMAO levels in healthy humans, TMAO levels increased again one month after discontinuation. Additionally, the use of broad-spectrum antibiotics can disrupt the normal function of gut microbiota, and long-term use may lead to the emergence of resistant bacteria, so the use of antibiotics still needs further verification. Furthermore, aspirin [63], TMA inhibitors [70], FMO3 inhibitors [71]intestinal immune modulators [72], etc., have evidence showing they can effectively reduce plasma TMAO levels. However, current research is mostly limited to animal experiments and studies on cardiovascular and cerebrovascular disease populations. Future research needs to expand the scope of study subjects and clarify the underlying mechanisms.

This study also has several limitations. First, the standard reference value for plasma TMAO has not been established, and the criteria for elevated TMAO in the included studies are inconsistent. Larger, multicenter, prospective studies are needed to further determine the standard reference value. Second, blood samples were collected at a single time point before emergency interventional surgery, and we lack information on the dietary history and previous antibiotic use of the enrolled patients, which may affect plasma TMAO levels. Finally, due to the small number of included studies, the interpretation of the results should be approached with caution. We anticipate more high-quality studies in the future to draw more precise conclusions.

Conclusions

This study found that elevated TMAO levels were significantly associated with a higher risk of MACE in PCI-treated patients. Although a positive association with all-cause mortality was observed, the presence of moderate heterogeneity warrants a cautious interpretation of these findings. While many interventions targeting TMAO are currently available, their exact efficacy requires further clinical research.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Abbreviations

TMAO

Trimethylamine N-Oxide

ACS

Acute coronary syndrome

PCI

Percutaneous coronary intervention

MACE

Major adverse cardiovascular events

CVD

Cardiovascular disease

NR

No-reflow phenomenon

ISR

In-stent restenosis

MIRI

Myocardial ischemia-reperfusion injury

FMO3

Flavin-containing monooxygenase 3

VLST

Very late stent thrombosis

NOS

Newcastle-Ottawa Scale

LC-MS

Liquid chromatograph-mass spectrometry

LC/MS/MS

Liquid chromatography with on-line tandem mass spectrometry

RRLC-QTOF/MS

Rapid separation liquid chromatograph-Quadrupole time-of-flight mass spectrometry

BAs

Bile acids

SCFAs

Short-chain fatty acids

ADP

Adenosine diphosphate

PC

Prospective cohort

STEMI

ST­segment elevation myocardial infarction

NSTEMI

Non-ST-segment elevation myocardial infarction

Author contributions

Chunyu Zhang writing original draft preparation; Jinyu He, Yujia Huo, Lin Liu and Yufei Meng conducted initial literature search; Yong Xie and Gang Wei ploted the figures. Li Deng edited the manuscript, Yang Jiang and Jian Feng revised the manuscript. All authors have read and agreed to the published version of the manuscript.

Funding

This research was funded by grants from Sichuan Science and Technology Program (2022YFS0610), Luzhou Municipal People’s Government - Southwest Medical University Science and Technology Strategic Cooperation (2021LZXNYD-J33), Hejiang County People’s Hospital - Southwest Medical University Science and Technology Strategic Cooperation Project (2021HJXNYD13), Gulin County People’s Hospital - Affiliated Hospital of Southwest Medical University Science and Technology strategic Cooperation (2022GLXNYDFY13), Xuyong County People’s Hospital - Affiliated Hospital of Southwest Medical University Science and Technology strategic Cooperation (2024XYXNYD18) and 2022-N-01-33 project of China International Medical Foundation.

Data availability

No datasets were generated or analysed during the current study.

Declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.

Clinical trial number

Not applicable.