A re-evaluation of the PURE study shows that consumption of foods with a high glycemic index or high glycemic load increases the risk of cardiovascular events and cardiovascular death by up to 50%.
Our Western dietary style is dominated by soft drinks, snacks and convenience foods. Such foods cause blood sugar to rise rapidly because they contain many simple carbohydrates, and are therefore considered unhealthy. Some studies have already shown a link between a diet rich in "bad" carbohydrates and an increased risk of cardiovascular disease. Now a new, large observational study comes to the same conclusion. It is based on data from the Prospective Urban Rural Epidemiology (PURE) study, one of the world's largest nutrition studies. The PURE study aims to identify diets associated with increased cardiovascular risk. Unlike previous studies, it has a key advantage: more than 137,000 subjects from 20 countries worldwide (from all five continents) were recruited. This makes it far larger and more diverse than previous studies.
A few facts about the study design: the average age of the study participants was 51 years, 43% of the subjects were male, 53% lived in an urban area, 21% were smokers, and 8% suffered from a previous cardiovascular disease. All participants completed 28 questionnaires about their dietary behavior at the start of the PURE study, from which the foods that subjects consumed regularly could be identified.
To determine the quality of the carbohydrates in a food, the researchers used the glycemic index (GI). It indicates how much a food causes blood glucose to rise compared to glucose. For example, cornflakes have a GI of 86%, while an apple has a GI of only 36%. The average GI of all participants' foods was 83%. The researchers then divided the subjects into 5 groups based on their individual GI: from a low GI (76%) to a high GI (91%). However, since GI does not take into account the carbohydrate density of a food (for example, watermelon has a fairly high GI, but is considered a healthy food), the measure of glycemic load (GL) was also used. In addition to the quality of the carbohydrates, GL also takes into account the quantity supplied. The average GL was 281 g/day. Also based on GL, subjects were divided into 5 groups.
After the baseline visit, participants were followed up for 9.5 years. The primary end point was a combination of cardiovascular event (cardiovascular death, nonfatal myocardial infarction, stroke, heart failure) or death from any cause. The researchers then calculated whether the primary endpoint occurred more frequently in subjects with high GI or high GL. In their analysis, they accounted for the usual demographic and anthropometric confounders.
After about 10 years, 11.7% of the subjects reached the primary endpoint. Of these, 6.8% suffered at least one cardiovascular event and 7.3% died from a cardiovascular event. The highest GI group had a 25% increased risk of cardiovascular event or death compared with the lowest GI group (hazard ratio [HR]: 1.25; 95% confidence interval [95% CI]: 1.15-1.37). When combined with two other studies, the authors arrive at a mean risk increase of 26% for cardiovascular death from a high-GI diet (HR: 1.26; 95% CI: 1.12 - 1.41). The risk was also increased in those without prior cardiac disease (HR: 1.21; 95% CI: 1.11 - 1.34). However, in patients with known cardiovascular disease, it was more than twice as high at 51% (HR: 1.51; 95% CI: 1.25 - 1.82).
Subgroup analyses showed that the relationship between high GI and primary endpoint was most pronounced in overweight subjects (BMI > 25 kg/m2) (HR: 1.38; 95% CI: 1.22 - 1.55), whereas normal-weight subjects had only a borderline increased risk (HR: 1.14; 95% CI: 1.00 - 1.30). Physical activity, nicotine use, or use of antihypertensives and statins had no effect on the association between GI and primary endpoint.
With regard to GL, only subjects with known cardiovascular disease showed a positive effect on the primary endpoint (HR: 1.34; 95% CI: 1.08 - 1.67), whereas study participants without prior cardiovascular disease and high GL did not show an increased risk.
"Despite the compelling data, the study also has some weaknesses. Surveying participants via questionnaires at a single time point is imprecise, and a shortcoming of many observational studies. As a result, we also do not know whether subjects changed their dietary patterns during the course of the study. Further, the data for GL were not as clear as those for GI. However, GL in particular can identify unhealthy, high-sugar foods better than GI. Thus, further studies are warranted to add strength to existing dietary recommendations that emphasize consumption of complex carbohydrates and to enforce them worldwide.
Source:
Jenkins et al. Glycemic Index, Glycemic Load, and Cardiovascular Disease and Mortality. NEJM. February 24, 2021.