Trimethylamine N-oxide is associated with coronary atherosclerotic burden in non-ST-segment myocardial infarction patients: SZ-NSTEMI prospective cohort study

Advanced Technology, Chinese Academy of Sciences, 518055 Shenzhen, P. R. China Department of Cardiology, The First AffiliatedHospital of DalianMedical University, 116011Dalian, P. R. China Guangdong Cardiovascular Institute, GuangdongGeneralHospital, GuangdongAcademy ofMedical Sciences, 510080Guangzhou, P. R. China Tianjin Key Laboratory of Ionic-Molecular Function of CardiovascularDisease, Department of Cardiology, Tianjin Institute of Cardiology, SecondHospital


Introduction
Cardiovascular diseases (CVDs) remain a leading cause of death worldwide and their prevalence is increasing in the general population [1,2]. The non-ST-segment elevation myocardial infarction (NSTEMI) is a frequent type of CVDs undergoing coronary intervention and reported with twofold higher long-term adverse outcomes than patients with ST-segment elevation myocardial infarction (STEMI) [3,4]. Nevertheless, sufficient consideration attributed to the reduction of the traditional risk factors including hyperlipidemia, smoking, hypertension, diabetes mellitus, and treatment with novel pharmacotherapies only decreased 30% of adverse outcomes related with CVDs [5][6][7]. Hence, identification of novel pathogenic risk factor related to CVDs has important public health significance for disease prevention and early stratification [8].
Recently, trimethylamine N-oxide (TMAO) metabolite gained widespread attention and purposed to play a potential role in adverse outcomes and pathogenesis for CVDs [9][10][11][12][13]. A significant association of the TMAO with the atherosclerotic burden in coronary vessels has been reported in STEMI and stable coronary artery disease (CAD) [14,15]. However, the relationship between high coronary atherosclerotic bur-den and TMAO is yet to be explored in NSTEMI patients. In present study, we examined the relationship of the plasma TMAO levels with metrics that reflect coronary atherosclerotic burden, including the number of diseased coronary vessels and clinical risk scores including the SYNTAX and GENSINI scores in newly diagnosed NSTEMI patients.

Study design
This prospective cohort is a registered clinical study (ChiCTR1900022366) and received Ethics approval (YN201901) at the Fuwai Hospital Chinese Academy of Medical Sciences Shenzhen. All procedures were in accordance with the Declaration of Helsinki and each participant provided informed consent.

Study population
The main study protocol has been published in detail previously [16]. In brief, two groups of individuals were prospectively enrolled including NSTEMI group (n = 73) and age-sex matched Healthy group (n = 35) between 2019 and 2020. The NSTEMI group included patients with newly diagnosed NSTEMI, between 18 to 75 years old, underwent coronary angiography (CAG) within 24 hours, with significant lesions (≥ 70% stenosis) in a single or multiple coronary lesions vessels. Patients were excluded those having previous history of CAD including NSTEMI, STEMI, stable or unstable CAD or percutaneous coronary intervention or heart surgery, with < 70% stenosis lesions in a single coronary or any extent of a lesion in the left main coronary artery, current presentation with STEMI, stable or unstable CAD, having a history severe renal and hepatic diseases, atrial fibrillation, valvular heart disease, peripheral vascular disease, severe heart failure, chronic pulmonary disease, acute or chronic inflammatory disease, active cancer, past three months history with cerebrovascular event, pregnancy, or refused participation. The age-sex matched Healthy group individuals were those who visited our hospital for routine annual health examination and selected after completed screening with no CVD or/and non-CVD. The Healthy group main purpose was to compare their TMAO reference interval concentrations with NSTEMI group.
NSTEMI diagnosis was confirmed accordingly to European Society of Cardiology guidelines including the presence of specific symptoms, changes in the electrocardiogram (ECG) and myocardial biomarkers [1,3,4]. The symptoms were specific angina related symptoms (chest discomfort or dyspnea) lasting more than 30 minutes without accompanying persistent ST-segment elevation; changes in ECG presented with ST-segment depression ≥ 0.05 mV, T-wave inversion ≥ 0.3 mV or flat T wave, or transient ST-segment elevation ≤ 0.05 mV or normal ECG; and increased levels of high sensitive troponin T and I as defined by values exceeding the upper limits of normal [1,3,4].

Study procedures
The following baseline information including demographic, 12-lead ECG, physical examination, detailed present and past medical history, and blood samples were obtained in all individuals.
In the NSTEMI group, standard Fuwai hospital protocol was used to perform the diagnostic CAG within 24 hours following admission via the redial approach. A series of diagnostic angiogram projections were obtained including 4 views for the left anterior descending (LAD) and left circumflex coronary (Cx), and 2 views for the right coronary artery. All patients received dual antiplatelet therapy [300 mg of aspirin (follow by 75 to 100 mg daily) and 600 mg of clopidogrel (follow by 75 mg daily for ≥ 12 months)], and additionally unfractionated heparin infusion (70-100 IU/kg) during the intervention.
The coronary atherosclerotic burden was calculated using SYNTAX and GENSINI scores by two expert cardiologists separately who were independent from this study. The SYN-TAX score was calculated by (www.syntaxscore.com). The GENSINI score was obtained accordingly 1 point (1% to 25% stenosis), 2 points (26% to 50% stenosis), 4 points (51% to 75% stenosis), 8 points (76% to 90% stenosis), 16 points (91% to 99% stenosis), and 32 points for a complete occlusion [17]. These points were further multiplied according to the importance of the coronary artery as 2.5 for proximal LAD artery and proximal Cx, 1.5 for mid-LAD stenosis, and 1 for distal LAD, mid/distal Cx, and right coronary artery stenosis [17]. Additionally, SYNTAX and GENSINI scores were further categorized into intermediate-high risk (score ≥ 23) and low risk (score < 23). If any difference were present in calculated scores, then a third interventional cardiologist opinion was obtained. Moreover, the number of diseased coronary vessels were categorized into single vessel disease (SVD) or multiple vessel disease (MVD) by one or more major coronary arteries having ≥ 70% lesion following CAG respectively.

Blood sampling and biomarkers measurement
The blood samples were obtained from the peripheral veins with the subjects in the fasting state on the morning after the day of admission, or prior any procedure. Additionally, an extra EDTA sample was also collected at the same time and immediately centrifuged at 2500 g at room temperature for 10 minutes and obtained plasma sample was incubated at -80 • C for TMAO measurement. The isotope dilution high-performance liquid chromatography with online tandem mass spectrometry using a QTRAP 4500 triple quadrupole mass spectrometer (AB SCIEX, Framingham, MA) with a d9-(trimethyl)-labeled internal standard was used for the TMAO measurement as described previously [14,15,18]. An estimated glomerular filtration rate (eGFR) assessment was obtained from modified Diet in Renal Disease study. Laboratory personnel were blinded to individual characteristics who analyzed plasma TMAO and other routine laboratory tests.

Statistical analysis
The continuous and categorical data were displayed as mean ± standard deviation or median with interquartile range or percentages. Kolmogorov-Smirnov was performed for normality. Comparing two groups were analyzed by Mann-Whitney U test or t-tests or Fisher exact tests or Pearson χ 2 tests as appropriate. Spearmen's correlation was applied to test the correlation between the number of diseased coronary vessels, SYNTAX, and GENSINI scores. Logistic regression was applied to analyse the association of TMAO with the presence of MVD, and intermediate-high risk SYN-TAX and GENISI scores following adjustment of traditional risk factors including [age, sex, hypertension, diabetes mellitus, smoking, low-density lipoprotein cholesterol (LDL), triglyceride (TG), eGFR, high sensitivity C reactive protein (hs-CRP), and body mass index (BMI)]. Moreover, an area under receiver-operating curves (AUC) was examined for TMAO predictability for high coronary atherosclerotic burden as defined above. SPSS 25 (IBM) was used for all statistical analyses with P-value ≤ 0.05 as a significant.

Discussion
The main finding of this study was a significant association between plasma TMAO levels and high coronary atherosclerotic burden as stratified by number of disease vessels and clinical risk scores including SYNTAX and GENSINI scores in NSTEMI patients. Moreover, TMAO independently predicted the MVD, and intermediate-high risk (score ≥ 23) SYNTAX and GENSINI scores, even following traditional risk factors adjustment. In all, TMAO predicts high coronary atherosclerotic burden as quantified above.
Atherosclerosis is a significant clinical problem leading to ischemia in different parts of the vasculature [19][20][21]. In this occurs in the coronary vessels, then CAD with calcification [22], partial or total vessel occlusion [23][24][25], leading to gradual reduction or absent coronary blood flow [21,26]. This can manifest as silent asymptomatic disease, angina pectoris [27,28], NSTEMI or STEMI [29][30][31], with adverse consequences such as arrhythmias, heart failure, and death [32]. Therefore, there is an increasing interest to develop predictive risk models for accurate risk stratification, which may include clinical signs or symptoms, imaging results and biomarkers [33]. The TMAO is a gut metabolite, generated from dietary nutrients such as choline, L-carnitine, betaine,  and phosphatidylcholine [9,34,35]. There is growing appreciation that TMAO has mechanistic links to proatherogenic effect via increased macrophage foam cell formation, activation of the inflammatory and platelet hyperactivity pathways, impaired cholesterol and bile acid transport [9][10][11][12][34][35][36][37]. These all impaired factors are associated with high-risk CVDs including stroke, atherosclerosis, atrial fibrillation, heart failure, and chronic kidney disease [9-12, 20, 34-38]. Based on experimental animal studies, TMAO accelerates atherosclerotic progression by inhibiting reverse cholesterol transport, enhancing platelet activity and thrombosis, and activating macrophages, while targeting production of TMAO inhibit pathogenesis of atherosclerotic [9,10,[34][35][36][37]. Recent clinical data demonstrated a close relationship of TMAO with markedly high short and long-term adverse outcomes related to CVDs even following adjustment for confounding risk variables [11][12][13]38]. More recently, Yoriko et al. further appraised that long-term increases in TMAO were linked to a higher risk for CVDs and repeated evaluation of TMAO over 10 years increased the early identification of individuals with a greater CVD-related risk [39].   Continues data are expressed as means ± SD or medians (interquartile ranges), and categorical variables are expressed as counts (%). Hyperlipidemia was defined as a fasting total cholesterol level ≥ 5.2 (mmol/L), or/and triglyceride level ≥ 1.7 (mmol/L), or/and low density lipoprotein > 2.6 (mmol/L). BMI, body mass index; DBP, diastolic blood pressure; eGFR, estimated glomerular filtration rate; HDL, high density lipoprotein; hs-CRP, high sensitivity C Reactive protein; hs-TnI, high sensitivity Troponin I; hs-TnT, high sensitivity Troponin T; IS/TIA, ischemic stroke/ transient ischemic attack; LDL, low density lipoprotein; LVEF, left ventricular ejection fraction; MVD, multivessel disease; NT-proBNP, N-terminal pro-B-type natriuretic peptide; SBP, systolic blood pressure; TC, total cholesterol; TG, triglyceride; TMAO, trimethylamine N-oxide.  This study is the first, to the best of the authors' knowledge, to evaluate the relationship between TMAO and coronary atherosclerotic burden in NSTEMI patients. Our study further contributes to earlier findings those suggested a significant link between plasma TMAO and atherosclerotic burden in STEMI and stable CAD patients [14,15]. Sheng et al. showed a significant association between plasma TMAO levels and high atherosclerotic burden as ascertained by the SYNTAX score and MVD in STEMI [14]. Senthong et al. noted that strong predictability of TMAO for high coronary atherosclerotic burden as evaluated via SYNTAX score stable CAD [15]. Lately, Guo et al. also observed a positive linked of TMAO level with of a high atherosclerotic burden particularly severe coronary artery lesion (≥ 90% stenosis) compared to mild coronary artery lesion in CAD patients [40].
The strength of our study that TMAO was evaluated in multiple parameters for quantification and stratification of coronary atherosclerotic burden including clinical risk scores such as SYNTAX and GENSINI, and MVD as compared to prior studies evaluated TMAO in only SYNTAX score and/or MVD [14,15]. A high SYNTAX and GENSINI scores, and MVD are associated with poor prognosis and considered as clinical markers of coronary atherosclerotic burden [3,17,41]. Notably, the median SYNTAX score 23 (IQR: 11.5-29) in NSTEMI patients recruited in the present cohort were seemingly higher than STEMI patients described by ) and stable CAD patients by , which might be partially attributed to higher proportion of patients (56.1%) in intermediate-high risk SYNTAX score category in the present cohort compared to Sheng et al. (27.8%) and Senthong et al. (18.1%) [14,15]. Moreover, our study included NSTEMI patients with significant coronary lesion (≥ 70% stenosis) as compared to prior studies ≥ 50% [14,15] and that might further strengthened our observation for association of TMAO with significant coronary lesion (≥ 70% stenosis). In fact, Guo et al. reported strong correlation of TMAO for server coronary artery stenosis compared to mild coronary artery stenosis [40]. Furthermore, our study observed a lower concentration of TMAO in patients with NSTEMI than previously reported studies in other types of CAD [14,15,40]. Indeed, Wang et al. reported a wide range of TMAO concentrations (lower limit 0.05 µM to upper limit 200 µM) with ≥ 95% accuracy determination by isotope dilution high-performance liquid chromatography among 4000 subjects including both healthy individuals and cardiovascu-lar patients [18]. These differences may be caused by many factors associated with different populations particularly demographic characteristics including age, sex, and ethnicity, different types of CAD and criteria for severity, Eastern and Western dietary habits, and composition of gut microbiota [42][43][44][45][46]. Furthermore, Liu et al. reported significant alteration in gut microbiota profiles in different CAD subtypes of patients and atherosclerotic severity, however they did not observe these changes in TMAO but purposed accordingly to their findings [42]. Therefore, more investigations that cover all these points are required for further explanation.
Recently, there is accumulating evidence to suggest TMAO is a clinically relevant target or marker for risk stratification and short or long-term prognostic outcome related in CVDs [11,12,14,15]. Moreover, inhibition of TMAO has been identified as a novel therapeutic approach for the prevention of atherosclerosis progression compared with existing traditional treatment [10-12, 37, 39, 43-46]. In light of these promising findings, implication of TMAO may establish early diagnosis, risk stratification, optimal treatment, and preventive strategies for CVDs [10-12, 37, 39, 43-46].

Limitations
Firstly, we were unable to assess the nutritional and diet status before enrollment for each individual that may bias results as TMAO may be influenced by nutritional status as previously reported [43][44][45][46]. Secondly, we could not exclude selection bias as the present cohort is a single-center study with a comparatively small number of NSTEMI patients, particularly in all sub-categorized groups. Thirdly, the COVID-19 pandemic unable us to continue and complete our original SZ-NSTEMI trial including enrollment of NSTEMI patients about two hundred participants for the evaluation of TMAO in regards to short and long-term prognostic outcomes after primary percutaneous coronary intervention as previously described [16]. Therefore, we recommended a large prospective randomized control trial to evaluate the association between TMAO and coronary atherosclerotic burden as well as a prognostic biomarker for adverse cardiac outcomes in NSTEMI patients undergoing coronary intervention.

Conclusions
In conclusion, our study demonstrated that plasma TMAO levels were significantly associated with a high coronary atherosclerotic burden as quantified by the number of diseased vessels, and clinical risk scores including SYNTAX and GENSINI scores in patients with NSTEMI. We encour-aged and warranted future large, randomized trial to evaluate our observed results in NSTEMI patients for the clinical endpoints.

Author contributions
LlW, JLC, and KBW were involved to research design, protocol, submission to ethical committee, and trial registration. KBW, CNP, HT, LGD, YKL, HQZ, QYC, and AMS contributed to literature review, patients selection accordingly to inclusion and exclusion criteria, informed consent, and data collection and interpretation of data. YLX, SLW, GT, XTL, and AA underwent statistical analysis and manuscript writing. All authors are involved in final draft of the manuscript, read and approved the final manuscript.

Ethics approval and consent to participate
The study was conducted in accordance with the Declaration of Helsinki and the Ethics Committee of Fuwai Hospital Chinese Academy of Medical Sciences Shenzhen (Ethics approval number: YN201901) and registered clinical study (ChiCTR1900022366) as SZ-NSTEMI. The study can be identified http://www.chictr.org.cn/showprojen.aspx?pr oj=37821. Written informed consent was obtained from all participants.