The quality of healthier foods is a more important factor in preventing major cardiac events than whether a diet is low in fat or carbohydrates, according to a large study of three cohorts averaging 30 years of data.

The study, published on February 11 in the Journal of the American College of Cardiology, follows earlier studies of low-carb and low-fat diets that have had mixed results. This study included almost 200,000 participants.

“Low-carb and low-fat diets are widely practiced in the US,” said Zhiyuan Wu, PhD, a coauthor of the study carried out at the Harvard T.H. Chan School of Public Health in Boston. “However, the long-term cardiovascular impacts and the role of diet quality in these dietary patterns remained unclear.”

Study Details

This study looked at the risk for major cardiac outcomes — coronary revascularization, myocardial infarction, and death due to coronary heart disease — linked to diet, determined from diet questionnaire data.

Wu and colleagues found that people with the lowest-carbohydrate diet, compared with those eating the highest-carb diet, did not benefit (hazard ratio [HR] for cardiac outcomes, 1.05). The same comparison for low-fat diets fared only a little better (HR, 0.93).

But when the diets were broken down into those that included healthier foods vs less healthy foods, the healthy versions of low-carb diets had a protective effect (HR, 0.85), as did the healthy versions of low-fat diets (HR, 0.87). There were also somewhat better outcomes for plant-based versions of the diets vs animal-based versions.

“I think the most surprising result is the similarity of the health benefits between the healthy versions of low-carb and low-fat diets,” said Wu, who is a postdoctoral research fellow at the Harvard T.H. Chan School of Public Health. “The diets emphasizing plant-sourced fats, proteins and whole grains, regardless of whether the diet patterns are low in carbs or low in fats, were associated with a very comparable risk reduction.”

Quality Over Quantity

“These findings indicate that the quality of the foods making up the diet, for example, whole grains vs refined grains, rather than the macronutrient composition, confer the health benefits,” said Alice H. Lichtenstein, DSc, a professor of nutrition science and policy at Tufts University in Boston who was not involved in the study.

“The findings emphasize the importance of diet quality when formulating dietary guidance,” added Lichtenstein, who chaired the American Heart Association’s dietary guidance in 2021.

Advice to patients should emphasize not only healthy foods but also avoidance of sugars and refined grains, Wu said. “We should promote a healthy eating pattern rather than very strict macronutrient restriction. That should be the central strategy for the primary prevention of heart disease.”

A more flexible approach to diet can improve long-term adherence, he said. For example, people can continue to enjoy their traditional cultural foods by eating whole-grain versions.

“For Chinese people or Italian people, carbs give you happiness. It’s a boost to your mental health as well,” said Wu. “But we should choose the healthier source of carbs” such as whole-grain pasta and brown rice.

Strength of Evidence

“There is always concern about the strength of the evidence with data from observational studies,” Lichtenstein said. However, the three Harvard cohort studies “are studied as rigorously as possible,” she said.

To add to the evidence, the researchers analyzed metabolites from blood samples taken from a subset of participants. They found distinct metabolites associated with healthy and unhealthy dietary patterns.

They also used machine learning to develop scores that included several metabolites, based on the survey information, Wu said. They found that the scores were linked to biomarkers of heart health, such as blood lipid levels and C-reactive protein.

“Metabolic data continues to grow in importance. Having the metabolomic and biomarker data strengthens the observational conclusions,” Lichtenstein said. “However, it does not change that the observations are of an association; they can’t establish causation.”

No disclosures were reported.

Carolyn Brown is a freelance medical writer in Ottawa, Canada.