Current issues of ACP Journal Club are published in Annals of Internal Medicine


Vitamin E and β-carotene were not effective in men at high risk for coronary events

ACP J Club. 1998 Jan-Feb;128:7. doi:10.7326/ACPJC-1998-128-1-007

Source Citation

Rapola JM, Virtamo J, Ripatti S, et al. Randomised trial of α-tocopherol and β-carotene supplements on incidence of major coronary events in men with previous myocardial infarction. Lancet. 1997 Jun 14;349:1715-20.



To determine the effectiveness of vitamin E, β-carotene, or both in men at high risk for major coronary events.


Randomized, double-blind, placebo-controlled trial with median follow-up of 5.3 years.


Southwestern Finland.


1862 men who smoked, were 50 to 69 years of age (median age 59 y), and had had a myocardial infarction (MI). Exclusion criteria were malignant conditions, severe angina pectoris, chronic renal insufficiency, liver cirrhosis, alcoholism, anticoagulant therapy, or use of vitamins A or E or β-carotene.


Patients were allocated to vitamin E, 50 mg/d (n = 466); β-carotene, 20 mg/d (n = 461); both (n = 497); or placebo (n = 438).

Main outcome measure

First major coronary event.

Main results

The groups did not differ for major coronary events. The rate of nonfatal MI was lower in the vitamin E group than in the placebo group after multivariate adjustment (relative risk [RR] 0.62, 95% CI 0.41 to 0.96). Fatal coronary heart disease (CHD) was increased in the β-carotene group (P = 0.007) and the combined group (P = 0.03) but not in the vitamin E group compared with the placebo group (Table). Analysis that used MI as the only end point showed no differences between study groups, although an increase in fatal MI was seen in the β-carotene (RR 3.44, CI 1.70 to 3.94, P < 0.001) and combined groups (RR 2.67, CI 1.30 to 5.48, P = 0.007).


Vitamin E and β-carotene did not reduce total coronary events in men who smoked and had had a previous MI. Fatal CHD was increased in the β-carotene and combined groups.

Sources of funding: Finnish Foundation for Cardiovascular Research; Ida Montin Foundation; Academy of Finland; United States National Cancer Institute.

For article reprint: Dr. J.M. Rapola, National Public Health Institute, Mannerheimintie 166, FIN-00300 Helsinki, Finland. FAX 358-9-4744591.

Table. Fatal coronary heart disease with β-carotene, vitamin E, or both supplements vs placebo*

Treatment EER CER RRI (95% CI) ARI |EER-CER| NNH (CI)
β-Carotene 16.1% 8.9% 80% (26 to 160) 7.2% 14 (9 to 35)
Vitamin E 11.6% 8.9% 30% (-12 to 92) 2.7% (CI-1.3 to 6.7) Not significant
Combined 13.5% 8.9% 51% (5 to 120) 4.6% 22 (12 to 194)

*Abbreviations defined in Glossary; RRI, ARI, NNH, and CI calculated from data in article.


Extensive laboratory data show the importance of oxidative modification of cholesterol in the pathogenesis of atherosclerosis. Epidemiologic studies report strong inverse associations between intake of antioxidant vitamins and CHD (1).

A previous trial with 6-year follow-up of > 29 000 middle-aged men who were targeted for primary prevention reported no benefits from vitamin E and potential for harm with β-carotene (2). The current substudy of patients with a history of MI extends these findings to secondary prevention and reports no benefits from either vitamin E or β-carotene. A small reduction in nonfatal MI but not in fatal CHD events was seen with vitamin E. β-Carotene had no beneficial effects on any end point evaluated. A major study limitation is that the vitamin E dose (50 mg/d) is lower than that suggested by epidemiologic investigations to be cardioprotective.

2 other recent clinical trials of vitamin E in cardiovascular disease are the Cambridge Heart Antioxidant Study (CHAOS) (3), which reported reductions in nonfatal MI with 400 to 800 IU/d of vitamin E but no reduction in cardiovascular mortality, and a Chinese study of low-dose vitamin E combined with other vitamins and minerals (4), which found a modest reduction in total mortality, although deaths from CHD were not reported. The current study adds to the extensive and disappointing data on β-carotene's lack of efficacy in prevention of cardiovascular events.

Use of antioxidant vitamins to prevent cardiovascular events deserves further evaluation, and large randomized trials are ongoing. Currently, vitamin E has shown the most promise, although the data remain inconclusive. β-Carotene, on the other hand, clearly failed to show benefit. Until further evidence becomes available, the widespread use of vitamin E or any other antioxidant vitamin should not be encouraged and the aggressive and consistent use of proven preventive strategies should be pursued, especially in high-risk persons targeted for secondary prevention.

Eva M. Lonn, MD
Hamilton General Hospital McMaster ClinicHamilton, Ontario, Canada


1. Jha P, Flather M, Lonn E, et al. The antioxidant vitamins and cardiovascular disease. A critical review of epidemiologic and clinical trial data. Ann Intern Med. 1995;123:860-72.

2. The Alpha-Tocopherol Beta Carotene Cancer Prevention Study Group. N Engl J Med. 1994;330:1029-35.

3. Stephens NG, Parsons A, Schofield PM, et al. Randomised controlled trial of vitamin E in patients with coronary disease: Cambridge Heart Antioxidant Study. Lancet. 1996;347:781-6.

4. Ross RK, Yuan JM, Henderson BE, et al. Prospective evaluation of dietary an other predictors of fatal stroke in Shanghai, China. Circulation. 1997;96:50-5.