Artificial Sweeteners Linked to CVD, Stroke in French Adults: Observational Study

Artificial sweeteners used as sugar substitutes in food and drinks may be associated with increased risk of CVD events and strokes, suggest data from the large NutriNet-Santé study of French adults.

The analysis, led by PhD candidate Charlotte Debras (Sorbonne Paris Nord University, France) and published in the BMJ, assessed dietary intake of artificial sweeteners (from beverages to a variety of food products and tabletop sweeteners) and measures of individual sweetener consumption (aspartame, acesulfame potassium, and sucralose).

“In the NutriNet-Sante cohort, total artificial sweetener intake was associated with increased risk of overall CVD and cerebrovascular disease,” the researchers write. “Aspartame intake was associated with increased risk of cerebrovascular events, and acesulfame potassium and sucralose were associated with increased coronary heart disease risk.”

As for how the artificial sweeteners might lead to adverse CVD and cerebrovascular events, Debras and colleagues say part of the association might be related to body weight, noting that some prospective cohort studies have reported links between artificially sweetened beverage consumption and increased risk of obesity or weight gain.

However, John L. Sievenpiper, MD, PhD (St. Michael's Hospital, Toronto, Canada), who was not involved in the study but has conducted research on artificial sweeteners and health, said one of the issues with interpreting epidemiological observations is that the data are at high risk of reverse causality.

“Basically, what that means is people may be taking artificial sweeteners as a means to mitigate risk: to lose weight, for example, or to maintain a healthier body weight, and they're doing that because they're at risk. That background risk typically is what you're seeing with the higher event rates in people consuming sweeteners,” he said.

People may be taking artificial sweeteners as a means to mitigate risk: to lose weight, for example, or to maintain a healthier body weight. John L. Sievenpiper

The researchers acknowledge that while they cannot exclude reverse causality, they don’t think it can entirely explain the observed associations. Their analysis excluded CVD events occurring during the first 2 years of follow-up and adjusted for baseline body mass index (BMI), weight-loss diet, and weight change during follow-up, “which did not substantially change the results,” they write.

In a recent systematic review and meta-analysis, Sievenpiper and colleagues addressed sources of bias in artificial sweetener studies and reached a different conclusion than the NutriNet-Santé investigators. There, after adjusting for confounders, people drinking diet, low-, or no-calorie beverages had lower risk of cardiometabolic outcomes including lower weight, body fat, CVD events, and total mortality when they consumed them as substitutes for sugar-sweetened beverages. Additionally, no impact on CVD events was seen when the drinks were substituted for plain water.

Aspartame Most Common

For the new study, Debras and colleagues analyzed data from 103,388 participants (mean age 42 years; 80% female) recruited from the general French population and enrolled in NutriNet-Santé, which looks more broadly at nutrition and health. Upon enrollment, participants complete detailed online questionnaires about their diet (including 24-hour dietary records), personal and family health history, height and weight, lifestyle and sociodemographic data, and physical activity levels (including minutes of activity per week). All of the information from the questionnaires were used to assess dietary intakes and consumption of artificial sweeteners.

Consumption of sugar substitutes was reported by 37% of all participants, with an average daily intake of 42.46 mg (range, 7.46 to 77.62 mg/ day), which the authors say is the equivalent of one individual packet of sweetener or 100 mL of a carbonated diet beverage. The most commonly consumed sweetener was aspartame (58%), followed by acesulfame potassium (29%), and sucralose (10%), with other artificial sweeteners such as saccharin and steviol glycoside, extracted from the stevia plant, accounting for the remaining 3%.

Daily intake ranged from a low of 7.46 mg/day to a high of 77.62 mg/day. Compared with participants reporting no use of sugar substitutes, those in the highest range of consumption were younger, had a higher BMI, were more likely to smoke, were less physically active, and were more likely to be following a diet for weight loss. High artificial sweetener users also had lower total calorie intake, drank less alcohol, and ate less saturated and polyunsaturated fats, fiber, carbohydrates, fruits, and vegetables. They also had higher intakes of sodium, red and processed meat, dairy products, and beverages with no added sugar.

Over 9 years of follow-up, there were 730 CV events and 777 TIAs and strokes among the participants. CV events occurred at a rate of 314 per 100,000 person-years among nonconsumers of artificial sweeteners and in 346 per 100,000 in those with the highest daily consumption levels.

Compared with nonconsumers, those with high intakes of aspartame intake had increased risk of cerebrovascular events (P = 0.02). Similarly, those with high intakes of acesulfame potassium and sucralose had increased risk of CVD compared with nonusers (P = 0.02 and P = 0.05, respectively).

Getting at Causation

Debras and colleagues say other underlying mechanisms could explain their results, including metabolic syndrome, which has been linked to artificially sweetened beverage consumption in several studies. There also is a possibility, they add, that glucose and energy homoeostasis could be altered by artificial sweeteners and that gut microbiota also could be affected, leading to increased glucose intolerance.

“Vascular dysfunction, which contributes to CVD onset and development, after the ingestion of artificial sweeteners, has been observed in experimental studies (rodent models) and in vitro (human cellular model), and could also play a part in the risk of CVD,” the researchers write. They also note that one study has suggested that artificial sweetener consumption might be associated with increased inflammation.

Sievenpiper said while the researchers adjusted for many important known confounders, there likely are many more unknown confounders at play.

“This study does have an advantage [over some others] because they have people report what they're consuming. They recorded the actual food and the amount based on a 24-hour recall to get at individual sweetener use, which I think was useful,” he said. “But you still have the issue that most sweeteners are not used in isolation. They’re in foods, they’re in drinks, . . . so looking at the sweeteners individually may not allow us to get at important interactions needed for causation.”

Finally, he cautioned that communicating to the general public about epidemiological studies can be tricky.

“If the message that people get is to go back to sugar-sweetened beverages, the evidence is pretty clear that that would not be a good thing,” Sievenpiper said. “Ideally, you would want people to drink neither sugar-sweetened nor artificially sweetened beverages. We prefer them to drink water and nonsweetened beverages like tea or coffee. But a lot of people are going to want a sweet substitute, and it’s probably better [based on the totality of evidence] that they drink the low- and no-calorie artificially sweetened beverage versus the sugar-sweetened beverage.”

In response to the new paper, the International Sweeteners Association (ISA) released a statement on its website saying that numerous other studies, including RCTs, have found “no evidence of a plausible mechanism to support potential effects of low/no-calorie sweeteners on cardiometabolic health.” Additionally, the ISA noted, the amounts of sweetener consumed in NutriNet-Santé are “extremely low, even for higher consumers,” increasing the likelihood that confounders, rather than a small amount of sugar substitute use each day, play a greater role in health outcomes. The group added that the safety of all approved low- and no-calorie sweeteners available to consumers has been confirmed by numerous organizations, including the European Food Safety Authority and the US Food and Drug Administration.