EvidenceHealthVitaminsSupplementsCardiometabolic
Vitamin E & C and Cardiovascular Disease in Men
- Study type:
- Large randomised, double‑blind, placebo‑controlled 2×2 factorial trial in male physicians
- Topic:
- High‑dose vitamin E and vitamin C supplementation and cardiovascular disease in men
- Added:
- 2026-01-10
- Status:
- published
Strength of evidence
Very low – robust randomisation and blinding with clinically relevant endpoints, but no true Null‑control arm and diagnostically defined cardiovascular events; under WorthDay's strict evidence criteria this yields only very low‑quality evidence for causality, while providing a strong negative signal against cardiovascular benefit and a potential harm signal for vitamin E.
Source
Sesso HD, et al. Vitamins E and C in the prevention of cardiovascular disease in men: the Physicians' Health Study II randomized controlled trial. JAMA. 2008;300(18):2123–2133.
Open original studyIn this 8‑year trial, over 14,000 male physicians were randomised to high‑dose vitamin E (400 IU every other day), vitamin C (500 mg daily), both, or placebos. Neither vitamin reduced heart attacks, strokes or cardiovascular deaths. Vitamin E was associated with a higher risk of haemorrhagic stroke (brain bleeding). Under WorthDay's criteria, the study does not establish causality, but it clearly fails to support antioxidant megadoses as heart‑protective and raises concerns about possible harm from vitamin E.
Key findings
- 14,641 male physicians (≥50 years) were randomised in a 2×2 factorial design to vitamin E, vitamin C, both, or matching placebos.
- High‑dose vitamin E (400 IU every other day) did not reduce the composite of non‑fatal myocardial infarction, non‑fatal stroke, and cardiovascular death (HR ≈ 1.01; 95 % CI around 0.90–1.13).
- High‑dose vitamin C (500 mg/day) likewise showed no meaningful effect on major cardiovascular events (HR ≈ 0.99; 95 % CI around 0.89–1.11).
- Vitamin E was associated with a significantly increased risk of haemorrhagic stroke (HR ~1.7; confidence interval excluding 1.0), suggesting potential harm.
- Under WorthDay's criteria, the absence of a true Null‑control arm and the use of diagnostically defined endpoints limit causal interpretation, but the data strongly argue against meaningful cardiovascular benefit and highlight a possible bleeding risk with vitamin E.
Background
- Antioxidant vitamins such as vitamin E and vitamin C have long been promoted as potential protectors against atherosclerosis and oxidative vascular damage. Observational studies and mechanistic arguments suggested these vitamins might reduce the risk of heart attacks and strokes. The Physicians' Health Study II (PHS II) antioxidant component was designed to test, in a rigorous randomised setting, whether long‑term supplementation with high‑dose vitamin E and vitamin C lowers cardiovascular risk in middle‑aged and older men.
Methods
- Species / population: 14,641 male U.S. physicians aged 50 years or older, including a subgroup with pre‑existing cardiovascular disease at baseline. Design: Randomised, double‑blind, placebo‑controlled 2×2 factorial trial. Interventions: Vitamin E: 400 IU synthetic α‑tocopherol every other day vs matching placebo. Vitamin C: 500 mg ascorbic acid daily vs matching placebo. Participants were assigned to one of four arms: vitamin E only, vitamin C only, both vitamins, or double placebo. Follow‑up: Mean follow‑up was about 8 years. Endpoints (as defined by the investigators): Primary endpoint: first major cardiovascular event, defined as non‑fatal myocardial infarction, non‑fatal stroke, or cardiovascular death. Secondary endpoints: total myocardial infarction, total stroke, cardiovascular mortality, and all‑cause mortality, plus other vascular outcomes. All events were identified and adjudicated through medical records, diagnostic procedures and classification rules, rather than purely direct mechanical observation.
Results
- No cardiovascular protection from vitamin E: The rate of major cardiovascular events was essentially identical in the vitamin‑E and placebo groups. The hazard ratio for vitamin E vs placebo was about 1.01, with a confidence interval centred around 1.0, excluding any substantial benefit. No cardiovascular protection from vitamin C: Vitamin C supplementation likewise did not reduce the incidence of major cardiovascular events. The hazard ratio for vitamin C vs placebo was approximately 0.99, with confidence intervals crossing 1.0. Harm signal for haemorrhagic stroke: Men assigned to vitamin E had a significantly higher risk of haemorrhagic stroke (intracranial bleeding) compared with those not receiving vitamin E. This suggests that high‑dose vitamin E may adversely affect bleeding risk in the brain. No improvement in overall survival: All‑cause mortality was not significantly reduced by vitamin E or vitamin C. Any absolute differences in death rates between groups were numerically small and not clinically meaningful. Overall, the trial provided no evidence that high‑dose vitamin E or C supplementation lowers cardiovascular risk in this population of male physicians, and it raised specific safety concerns for vitamin E.
Interpretation
- For WorthDay, S0032 delivers a clear message: High‑dose vitamin E and vitamin C, taken for many years by well‑nourished middle‑aged and older men, did not prevent heart attacks, strokes, or cardiovascular deaths. Vitamin E, in particular, may carry a non‑trivial risk of haemorrhagic stroke, undermining the notion that it is a harmless "heart‑protective" antioxidant. Under WorthDay's strict criteria, important limitations are emphasised: No true Null‑control arm: Every participant ingested at least one capsule (active or placebo). This makes it impossible, under WorthDay's logic, to cleanly separate capsule‑ or expectation‑related effects from the specific biochemical actions of the vitamins. Diagnostic endpoints: Myocardial infarction, stroke and cardiovascular death are defined via complex diagnostic protocols and classifications, not through simple, direct mechanical observation. WorthDay regards these as soft endpoints when used for strong causal inferences. Focus on relative effects: Results are expressed mainly as hazard ratios and relative risks. While they show a lack of benefit, absolute risk differences and their precision are crucial for judging real‑world relevance; here, they are small and clinically unconvincing. From a practical WorthDay perspective, however, the pattern is robust: In a large, well‑conducted RCT, antioxidant vitamin megadoses did not deliver the promised cardiovascular benefits and instead produced a plausible harm signal (increased brain bleeding) for vitamin E.
Limitations
- Selective population: Male physicians are not representative of the general population; they likely have good baseline nutrition, healthcare access and health literacy.
- No focus on deficiency: The trial did not specifically recruit individuals with documented vitamin E or C deficiency; effects in truly deficient or malnourished populations were not tested.
- Endpoints based on medical diagnosis: Event ascertainment depends on clinical diagnostics, coding and adjudication, which WorthDay regards as indirect and potentially assumption‑driven.
- Lack of Null‑control: There is no group without any capsule intake, so generic capsule and expectation effects cannot be fully disentangled.
WorthDay perspective
- S0032 supports a sceptical stance toward routine high‑dose antioxidant supplementation (vitamin E and C) for heart protection in already well‑nourished adults.
- For WorthDay users, the key takeaway is: more isolated antioxidants are not automatically better – in this case, they did not reduce cardiovascular events and may increase the risk of serious bleeding.
- WorthDay can use this study to highlight the importance of whole‑food patterns, lifestyle and recovery instead of relying on high‑dose antioxidant pills for cardiovascular prevention.
Permalink: https://www.worth-day.com/science-compass/s0032-vitamin-e-c-and-cardiovascular-disease-in-men
Related studies
Science meets practice