乌鸦传媒

MUNICH -- The diet in the healthiest people around the globe is a balanced one, with lots of fruit, vegetables, and dairy but also including red meat and a modest amount of fat and carbohydrates, the PURE study showed.

The observational study of more than 138,000 people from 21 countries of wide-varying wealth, presented here at the European Society of Cardiology meeting by Andrew Mente, PhD, of McMaster University in Hamilton, Ontario, generated a diet score based on mortality associations.

The lowest quintile for mortality versus the highest quintile averaged:

• Carbohydrates: 54.0% versus 69.1% of daily caloric intake
• Fats: 28.3% versus 18.5% of daily calories
• Protein: 17.9% versus 11.9% daily calories
• Fruit and vegetables: 8.4 versus 1.8 servings per day
• Dairy: 3.0 versus 0.6 servings daily
• Unprocessed red meat: 1.4 versus 0.3 servings per day
• Nuts and legumes: 2.5 versus 0.7 daily servings

When applied to several large observational cohorts, the diet score showed graded associations with overall mortality and major cardiovascular disease, including MI and stroke individually, such that those with diets closer to those suggested as healthy in PURE had lower risk than the other cohorts.

For example, after adjustment for age, sex, education, waist-to-hip ratio, physical activity, total intake, diabetes, lipid and blood pressure lowering medication use, and clustering by center, those in the 31,546-patient ONTARGET study cohort with the best diet score (18 or more on the scale from 5 to 35 points) had a total mortality hazard ratio of 0.76 (95% CI 0.68-0.84) compared with those in the bottom scoring group of 11 points or less. For major cardiovascular disease, the hazard ratio was more modest (0.86) and significant for trend but not individual groups.

Implications

Despite the controversy around the prior report of the PURE data, "when looking at nutrients, looking at carbohydrates, protein, and fats, this is not conflicting with the guidelines," said Mente. "What is somewhat different from the guidelines is the two foods: dairy and unprocessed meats. But even those, our results don't show you consume outrageous amounts."

The analysis was less informative about the individual components than the overall pattern, he told 乌鸦传媒 in explaining the high fruit and vegetable intake level of the healthiest diet group compared with last year's findings from PURE that more than 3 to 4 servings didn't appear to be additionally beneficial.

But Mente noted, "if you eat more fruits and vegetables, you're eating less of something else. So it can't be a bad thing."

"This idea of moderation is a reasonable one," commented Fred Masoudi, MD, of the University of Colorado in Aurora. "The diet they're talking about is relatively balanced and moderate."

"It's certainly consistent with what the American Heart Association has been saying," commented Mariell Jessup, MD, incoming chief science and medical officer of the American Heart Association.

"When there are some slight inconsistencies, I think their answer is because low income countries may have different protein needs or even different sodium needs than high income countries. That's what they said in one of their slides," she added referencing a separate presentation by the PURE group here.

"The diets, they swing to extremes.... Exclusion diets are very popular," commented session co-moderator Barbara Casadei, MD, DPhil, of the University of Oxford, England. "It is easier to exclude a component of the diet entirely than it is to have a little bit of everything -- one ends up have too much of everything, right? In reality, there is no replacement for a balanced diet."

Limitations

The cohort studies included in the analysis -- PURE, ONTARGET/TRANSCEND, INTERHEART, and INTERSTROKE cohorts -- all had complimentary designs and looked at international populations ranging from 21 countries in PURE to 52 in INTERHEART.

While the researchers emphasized the benefits of a global look at diet compared with most prior studies looking only at Western nations, others saw it as impacting generalizability within those Western nations.

"It is very difficult to completely adjust for social factors, poverty," noted Casadei. "In an observational study, you're always hoping to adjust. To me the quintile 1 looks [like] very poor people who didn't have enough money for protein or enough money for buying high-quality food ... and they were doing badly because of that and probably because of a lot of things."

The borderline significance of the cardiovascular disease findings in the PURE analysis, may suggest that "competing morbidity, possibly inadequate caloric intake in developing countries, could be contributing to the results," agreed C. Noel Bairey Merz, MD, of the Preventive Cardiac Center at Cedars-Sinai Medical Center in Los Angeles.

"Specifically, if the population suffers from the malnutrition from inadequate food supply for some or all, then full fat dairy and unprocessed meat would counteract that as high calorie and protein dense food," she added. "It would be useful to see results by categorization of country or region with regard to developed versus emerging economies or other metric for adequacy of food for the population."

Mente argued that the findings are "widely applicable" and relevant to the U.S. population, for example, as it did include about 20,000 people in Canada, Sweden, and Poland.

He acknowledged, though, the limitations of observational data despite the extensive adjustment for potentially confounding factors. "That is about the best you can do with observational studies," he told 乌鸦传媒. "To definitively answer the question, you do need randomized trials," despite the difficulty.

Another constraint was the low statistical power for cancer during PURE's 8 years of follow-up so far, which is probably half what would be needed, Mente said.

Study discussant Eva Prescott, MD, of the University of Copenhagen, pointed out that the difference between even the healthiest and unhealthiest diet groups was only 7 points on the 35-point scale and both groups fell toward the bottom end of the scale, suggesting it "may have limited discriminatory power."

Mente countered that the differences between the top and bottom score quintiles was "pretty big" and the macronutrient difference was "not small."

He called the 28% fat intake level on par with most guidelines but not the American Heart Association's DASH diet, which is a low-fat diet.

"I think you can't argue against the DASH diet, which has been very, very successful in lifestyle intervention in helping to lower blood pressure mostly through weight reduction," Jessup argued, saying it has been "rigorously tested" in high-income countries.

"I think like everything else in medicine, you have to be specific, tailor all recommendations to individual patients," she concluded.

Comment