Hydrogenated vegetable oils and heart attack risk
What it found
In a study of 521 people in the Boston area, those who ate the most trans fats from partially hydrogenated vegetable oils were more likely to have had a first heart attack.
The link was strongest for trans fats from vegetable sources, not animal sources.
What they found
Studies in people
Heart attack risk
People in the highest fifth of trans-fat intake had a higher risk of heart attack than those in the lowest fifth. The link remained after accounting for age, sex, smoking, high blood pressure, family history, alcohol, exercise, body mass index, and other fats in the diet.
Source of trans fats
Hydrogenated vegetable fats supplied 74% of the trans fats in this group. The higher heart attack risk was almost entirely explained by trans fats from hydrogenated vegetable fats. No significant link was seen for trans fats from animal fats.
Margarine intake
Margarine was the main source of trans fats. People who ate more than 2.5 pats of margarine a day had a higher risk than those who ate less than 1 pat a day. The link did not change much after accounting for butter intake.
Blood cholesterol
The link between trans-fat intake and heart attack stayed strong even after accounting for blood levels of HDL and LDL cholesterol. The authors note this result should be interpreted with caution because cholesterol levels in patients may have changed after the heart attack.
Men and women
The link between trans-fat intake and heart attack was seen in both men and women. After accounting for standard risk factors, the risk comparing highest with lowest intake was higher in women than in men.
Studies in people and animals
Possible mechanisms
In lab tests, trans fats have been reported to affect platelet clumping, cell membrane enzymes, and receptors in rat heart membranes. Diets high in trans fats from partially hydrogenated vegetable fat have raised LDL cholesterol and lowered HDL cholesterol in randomized trials.
What the authors conclude
“These data support the hypothesis that intake of partially hydrogenated vegetable oils may contribute to the risk of myocardial infarction.”
Also in their conclusions
- They say their data support the hypothesis that intake of partially hydrogenated vegetable oils may contribute to the risk of myocardial infarction.
- They suggest that if the link were causal, people in the top fifth of trans-fat intake could halve their risk of heart attack by reducing their intake of trans-isomers.
- They note that this benefit would be comparable to what is expected from a 20% reduction in plasma cholesterol.
How it was done
Concern that trans-fatty acids formed when vegetable oils are partially hydrogenated might raise the risk of coronary disease had existed for decades, but direct evidence in people was limited.
They compared 239 patients admitted to six Boston-area hospitals with a first heart attack and 282 control subjects of similar age and sex from the same towns. Trans-fat intake was estimated from a food frequency questionnaire covering the previous year.
What it can’t tell you
- This was a case-control study, so it can show a link between trans-fat intake and heart attack but cannot prove that trans fats caused the heart attacks.
- Some misreporting of margarine and trans-fat intake is likely, which could only have weakened the link seen.
- The authors cannot rule out that patients overreported or controls underreported margarine use, though they found indirect evidence against this.
Who paid
- Authors work at
- Harvard School of Public Health, MA
The paper
- Title
- Trans-fatty acids intake and risk of myocardial infarction
- Type
- Study
- Evidence
- Studies in people · animals · lab
- Summarised from
- Full text
- Cite
- Ascherio A, Hennekens CH, Buring JE, Master C, Stampfer MJ, Willett WC (1994). Trans-fatty acids intake and risk of myocardial infarction. Circulation. doi:10.1161/01.cir.89.1.94Free full textPubMed 8281700DOI
Summary written 26 Sep 2026. Check it against the paper before it changes what you eat. How we summarise papers · Report an error