Cohort studycited by the Tayebat archivefull text checked★ key study A-EGG-13 in-depth review
Gut bacteria, TMAO and heart risk
Intestinal microbial metabolism of phosphatidylcholine and cardiovascular risk.
Overview
Tang WH et al.. N Engl J Med 2013 · PMID 23614584 · DOI 10.1056/NEJMoa1109400
Gut bacteria make TMAO from egg-yolk components; heart patients with the highest TMAO had more events (about 1.4-fold after adjustment, 2.5-fold unadjusted); egg eating itself was not studied against outcomes.
Does the study support the specific claim? Partly · Methodological quality: moderate
Plain-language summary
Study explained simply
- What did the researchers want to find out?
Is gut-microbial metabolism of dietary phosphatidylcholine to TMAO linked to adverse cardiovascular events?
- How was the study done?
Human feeding challenge with antibiotic suppression plus prospective cohort of cardiology patients
- How many people, animals or samples were studied?
40 healthy adults (6 given antibiotics) for the egg challenge; 4,007 heart-catheterisation patients for outcomes
- What was actually found?
TMAO rose after eating eggs and was blocked by antibiotics. Highest vs lowest TMAO quartile: HR 2.54 unadjusted, 1.43 (1.05-1.94) after full adjustment, for death, heart attack or stroke within 3 years.
- How large was the observed effect?
Cox regression; HR
- What do the results mean in everyday life?
Shows a mechanism, not the effect of eating eggs.
- What are the study's limitations?
Egg intake not measured in the outcome cohort; effect shrinks markedly after adjustment (incl. kidney function); high-risk population; antibiotic test in only 6 people.
- What does it say about Elawady's specific claim?
Partly supports; does not show eggs cause events.
- Do the results apply to Germany and Europe?
Not region-specific.
- What can explicitly not be concluded from this study?
Cannot show that eating eggs increases cardiovascular events, nor a 2.5-fold risk from eggs.
Scientific in-depth review
1. Full reference
- Authors
- Tang WH, Wang Z, Levison BS, Koeth RA, Britt EB, Fu X, Wu Y, Hazen SL
- Journal
- N Engl J Med
- Year
- 2013
- Identifier
- PMID 23614584 DOI 10.1056/NEJMoa1109400 · PMC3701945
- Verification
- verified – PubMed E-utilities efetch, 2026-10-25
- Retracted
- no
- Corrections
- none
- Extent checked
- full text
- Conflicts of interest and funding
NIH grants (incl. Office of Dietary Supplements), Cleveland Clinic/CWRU CTSA, Leonard Krieger Fund gift; mass spectrometry facility partly supported by AB SCIEX. Individual author disclosures are in NEJM disclosure forms not included in the retrieved text.
2. Research question
Is gut-microbial metabolism of dietary phosphatidylcholine to TMAO linked to adverse cardiovascular events?
3. Study design and population
- Study design
- Cohort study – Two studies: (1) uncontrolled human feeding challenge (2 hard-boiled eggs + 250 mg d9-phosphatidylcholine tracer) with antibiotic suppression in a subset; (2) prospective cohort (Cleveland Clinic GeneBank) relating a single fasting plasma TMAO measurement to 3-year MACE
- Population
(1) 40 healthy adults without chronic illness or recent antibiotics/probiotics (6 received antibiotics); (2) 4,007 stable patients undergoing elective diagnostic cardiac catheterisation without acute coronary syndrome (mean age 63, 64% men, 32% diabetes, 72% hypertension, 42% prior MI)
- Country
USA (Cleveland Clinic)
- Period
3-year follow-up for MACE; enrolment dates not given in the retrieved text
- Sample size
Challenge: 40 (antibiotic sub-study n=6); outcomes cohort: 4,007 (513 MACE)
- Inclusion and exclusion criteria
Challenge: healthy adults, no heart, renal, pulmonary or haematological disease, no active infection, no antibiotics/probiotics in the past month. Cohort: elective coronary angiography, no acute coronary syndrome (troponin T <0.1 ug/l if available).
4. Methods and endpoints
- Exposure / intervention
(1) Dietary phosphatidylcholine challenge (two eggs, ~250 mg total choline each, plus d9-PC) before/after 1 week of metronidazole + ciprofloxacin; (2) fasting plasma TMAO quartiles (no dietary or egg intake data collected)
- Comparator
(1) Same participants before vs after antibiotics and after recolonisation; (2) lowest TMAO quartile (<2.43 uM)
- Endpoints
(1) Plasma/urine TMAO, d9-TMAO, choline, betaine kinetics; (2) adjudicated MACE (death from any cause, nonfatal MI, nonfatal stroke) at 3 years
- Statistical methods
Cox regression unadjusted and in three adjusted models (Model 1: traditional risk factors + hsCRP; Model 2: + myeloperoxidase, eGFR, white-cell count, BMI, medications; Model 3: + angiographic extent of CAD); per-SD analysis; NRI, IDI, C statistic; subgroup analyses.
5. Actual results
- Results
Challenge: TMAO and d9-TMAO rose after eggs + d9-PC; nearly completely suppressed after 1 week of antibiotics (n=6) and reappeared >=1 month after withdrawal (variable recovery). Cohort: median TMAO 5.0 uM in those with events vs 3.5 uM without. Highest vs lowest quartile (>6.18 vs <2.43 uM): unadjusted HR 2.54 (1.96-3.28); Model 1 HR 1.88 (1.44-2.44); Model 2 (incl. eGFR) HR 1.49 (1.10-2.03); Model 3 HR 1.43 (1.05-1.94). Per SD: unadjusted HR 1.40 (1.29-1.51), adjusted 1.30 (1.20-1.41). Death HR 3.37 (2.39-4.75); nonfatal MI/stroke HR 2.13 (1.48-3.05) (unadjusted). NRI 8.6%; C statistic 68.3% vs 66.4%.
- Absolute risks
513 of 4,007 (12.8%) had MACE within 3 years; quartile-specific absolute event rates only shown graphically (Kaplan-Meier), not tabulated in retrieved text.
- Authors' conclusion
TMAO production from dietary phosphatidylcholine depends on gut microbiota; higher TMAO associated with MACE.
6. Strengths
- Stable-isotope tracer demonstrates conversion of dietary phosphatidylcholine to TMAO in humans
- Antibiotic suppression and recolonisation show gut-microbial dependence
- Large cohort with adjudicated hard outcomes and stepwise adjustment, including kidney function
7. Weaknesses
- Cohort measured plasma TMAO only, not egg or phosphatidylcholine intake
- Effect size shrinks from HR 2.54 (unadjusted) to 1.43 (fully adjusted), indicating substantial confounding (age, kidney function, disease severity)
- Single TMAO measurement; high-risk cardiology population
- Antibiotic sub-study only 6 participants; challenge uncontrolled
- Reverse causation (sicker patients with lower eGFR have higher TMAO) cannot be excluded
- TMAO is also raised by fish intake, as authors note
8. Risk of bias
- Orienting tool
- ROBINS-E (cohort component); challenge study has no standard tool
- Type of assessment
- structured assessment based on the full text by a single reviewer (no formal double review)
- Full formal assessment
- no
- Overall risk of bias
- some concerns
- Rationale
Unchanged (some_concerns) for the TMAO-MACE association; for any inference about eggs the evidence is very indirect. Single-reviewer structured judgement; a formal ROBINS-E needs two independent reviewers and the protocol.
| Domain | Judgement | Rationale |
|---|---|---|
| Confounding | some concerns | Stepwise adjustment incl. eGFR and CAD extent reduces HR from 2.54 to 1.43; residual confounding by renal function and comorbidity likely. |
| Exposure measurement | some concerns | Validated LC-MS/MS but single fasting TMAO; no dietary exposure measured. |
| Selection of participants | some concerns | Elective angiography patients at one centre (high-risk referral population). |
| Post-exposure interventions | some concerns | Revascularisation after enrolment; sensitivity analysis excluding 30-day revascularisation consistent (HR 2.47 unadjusted). |
| Missing data | low | Outcomes ascertained for all participants per authors. |
| Outcome measurement | low | Adjudicated death, MI, stroke. |
| Selection of reported result | some concerns | Abstract headlines the unadjusted HR 2.54; adjusted HRs reported only in table. |
Full-text review (25 October 2026): Added challenge n=40 (antibiotics n=6), cohort characteristics, adjusted HRs (Model 1 1.88; Model 2 1.49; Model 3 1.43 (1.05-1.94)) vs unadjusted 2.54, per-SD HRs, 513 events, funding incl. AB SCIEX facility support; directness set very_indirect; archive derivation strengthened to overinterpreted. Bias unchanged.
9. Assessment of the statistical interpretation
The often-quoted '2.5-fold' risk is the unadjusted quartile HR; after adjustment for risk factors, eGFR and angiographic disease the HR is 1.43 (1.05-1.94). The association is between a blood marker and events in a high-risk population, not between egg consumption and events.
10. Relevance to the specific claim
11. Independent contrary findings
12. Transferability to Germany and Europe
Not region-specific.
13. Overall methodological judgement
Important mechanistic and prognostic-marker study: dietary phosphatidylcholine (from eggs plus tracer) is converted to TMAO by gut microbes, and high TMAO predicts MACE in cardiology patients. It does not measure egg intake versus outcomes, and the adjusted association (HR 1.43) is much smaller than the 2.5-fold figure.
| Multidimensional assessment | |
|---|---|
| Authenticity of the source | verified |
| Methodological quality | moderate |
| Risk of bias | some concerns |
| Precision of results | moderate |
| Directness | very indirect |
| Reproducibility | partly replicated |
| Consistency with the overall evidence | mixed |
| Regional transferability | moderate |
| Adequacy of the interpretation (by the archive) | partly correct but over-interpreted |
14. Open questions
Whether TMAO is causal; whether habitual egg intake raises fasting TMAO meaningfully (later trials suggest not); kidney function as confounder.
15. Linked fact checks
Use by the Tayebat archive
Note: according to the archive, the studies were compiled by the archive operators. Whether Dr. Elawady himself cited this study is not established.
- What the archive says about the study
Egg-yolk phosphatidylcholine is converted by gut bacteria into TMAO, and high TMAO levels were tied to a 2.5-fold higher risk of major cardiovascular events.
- A. What was actually studied?
TMAO formation after eggs+PC; TMAO levels vs events.
- B. What did the authors find?
Eggs + d9-PC raised TMAO, which depended on gut microbes (n=40; antibiotics n=6). In 4,007 cardiology patients, highest vs lowest TMAO quartile HR 2.54 unadjusted, 1.43 (1.05-1.94) fully adjusted; egg intake not measured.
- C. What conclusions do the authors themselves draw?
TMAO depends on microbiota; associated with events.
- D. What does the archive derive from it?
Links eggs to 2.5-fold cardiovascular risk.
- E. Is this derivation scientifically justified?
Overinterpreted: the 2.5-fold figure is the unadjusted HR for blood TMAO quartiles (fully adjusted 1.43) in high-risk patients; the study did not measure egg intake against outcomes. Later trials (I-EGG-10) found eggs did not raise fasting TMAO.
- Adequacy of the interpretation (by the archive)
- partly correct but over-interpreted