Key takeaways:

• Environmental exposures such as air quality, greenspace and climate impact cardiometabolic risk.
• Predictive tools combining satellite imaging and AI could guide localized cardiometabolic disease intervention.

PHILADELPHIA — Environmental exposures increase risk for cardiometabolic disease, but improved predictive tools combining satellite imaging, AI and time-variable considerations for risk could guide future interventions, a speaker reported.

Sanjay Rajagopalan, MD, MBA, FACC, FAHA, chief of cardiovascular medicine and chief academic and scientific officer at University Hospitals, Harrington Heart & Vascular Institute, discussed environmental exposures that impact cardiometabolic health and how future predictive models could account for factors such as built environment and changes in risk factors and exposures over time.

Environmental exposures such as air quality, greenspace and climate impact cardiometabolic risk. Image: Adobe Stock

“The importance of the environment can’t be understated. The Global Burden of Disease estimates that roughly 12 million deaths of the overall 60 million deaths or so globally occur because of environmental reasons, and this is likely an underestimate, given the fact that a number of environmental exposures are currently not measured or assessed but have credible evidence that they contribute significantly to the pathogenesis of risk factors and eventually to CVD,” Rajagopalan told Healio.

Rajagopalan said social determinants of health and environmental risk factors both impact the American Heart Association’s Life’s Essential 8, which includes diet, physical activity, nicotine exposure, sleep, body weight, lipids, blood glucose and BP, but social determinants are conditions while the environment represents a meta-exposure. Meta-exposures impacting Life’s Essential 8 include air quality, greenspace, walkability, transportation access, food environment, chemical exposures and climate, he said.

Sanjay Rajagopalan, MD, MBA, FACC, FAHA, discussed environmental exposures that impact cardiometabolic health and how future predictive models could account for factors such as built environment and changes in risk factors and exposures over time.

“An exposure like air pollution, for instance, is estimated to contribute to roughly 9 million deaths a year, and we now know that air pollution not only contributes to CV mortality and all-cause mortality, but also is responsible for the genesis of the very risk factors that give rise to CVD, such as increases in BP, poor glycemic control, etc,” Rajagopalan said. “Now, if you add some of the other risk factors, including lead poisoning, for instance, occupational exposures and heat, which is becoming increasingly important as an exposure related to climate change, the Global Burden of Disease numbers of 12 million deaths a year to environmental reasons are likely an underestimate.”

As Healio previously reported, the burden of heat-attributable CVD in the U.S. is projected to increase through 2050, primarily in years of life lost, and will disproportionately impact older adults and those in lower-income counties.

In addition, data published in The Lancet Planetary Health and Journal of the American Heart Association both showed a significant global burden of type 2 diabetes and BP elevation, respectively, associated with exposure to high levels of fine particulate matter (PM_2.5) air pollution.

Areas with higher levels of vegetation and environmental greenness also had a significant impact on lowering all-cause and CV mortality vs. areas with lower levels of greenness, according to studies published in Environmental Pollution and The Lancet Planetary Health.

“Although it is mentioned in the guidelines in terms of context for a number of risk factors, there is not much emphasis on defining measurable exposures in any of the guideline documents, and I believe that’s a big shortcoming,” Rajagopalan told Healio. “They could be part of the risk equation, because we now know ... that a number of risk exposures primordially might arise because of unhealthy environments.”

Rajagopalan said phenotyping external environmental exposures using AI analysis of satellite and street view imagery could be used to guide interventions to reduce the impact of those exposures on cardiometabolic health.

As an example, Rajagopalan cited prior data published in the European Heart Journal in which researchers conducted an AI-based assessment of built environment on Google Street View to estimate the localized prevalence of coronary artery disease.

The researchers reported that built environment feature-AI analysis predicted 63% of the census tract variation in CAD prevalence and improved accuracy of a model that previously only included age, sex, race, income and education or a composite of indices of social determinants of health.

A “time” element could also be incorporated into future predictive models as well, as current models only provide a single snapshot of an individual’s CVD risk and environmental exposures are not static and do not exist in isolation, according to the presentation.

“Currently, most of the risk estimates that we have, including the PREVENT equation, are static estimates at one time point, and that’s simply not going to perform very well, because you can’t expect measurements of a limited number of risk factors to predict your health very well down the road,” Rajagopalan said. “What we need are much more informed metrics that are incorporated into predictive risk models that also give you a time element, and hopefully this should really improve risk estimation in the future.”

For more information:

Sanjay Rajagopalan, MD, MBA, FACC, FAHA, is chief of cardiovascular medicine and chief academic and scientific officer at University Hospitals, Harrington Heart & Vascular Institute and director of the Case Cardiovascular Research Institute, Herman K. Hellerstein MD, Professor of Cardiovascular Medicine and professor in the department of internal medicine and biomedical engineering at Case Western Reserve University. Rajagopalan can be reached at sxr647@case.edu.