Moving to a Population-Based Approach to Find Links Between Diabetes, Heart Disease

Published on: 

Identifying which patients with diabetes will develop heart disease is not as straightforward as it might seem. This session explored the use of biomarkers, imaging, and how a population health-based model will do a better job of identifying women at risk.

Connections between diabetes and heart disease are well-known, but figuring out exactly which patients with diabetes will develop conditions such as congestive heart failure or cardiovascular disease is less straightforward. Can tools such as biomarkers or imaging provide help? And how can moving away from “disease-based” to a “population-based” approach to identifying at-risk diabetics, especially women, play a role?

Speakers at Saturday's symposium, "Imaging the High Risk Diabetic Patient," was a collaboration between several professional associations and the American College of Cardiology, which is holding its 64th Annual Scientific Sessions March 14-16, 2015, in San Diego, California.

Matthew J. Budoff, MD, of the University of California at Los Angeles, started the discussion by looking at what genetics is starting to tell medicine about the connections between diabetes and heart disease, with the caveat that diabetes is affected by a mix of genetics, environmental factors, nutrition, and whether a person is obese. The vast majority—92% of patients—have insulin resistance, and many develop hypertension. Inflammation appears to be key, with damage at both the microvascular and macrovascular level, and Budoff discussed markers related to inflammation.

It appears inflammation sets in motion the process that leads to cardiovascular disease. Budoff also discussed C-reactive protein, which is found in blood plasma and rises in response to inflammation; while a weaker marker it is still associated with diabetes, he said. Meanwhile, plasma levels tend to drop in patients with diabetes, also in response to inflammation.

Finding good markers is important, Budoff said, because “There is a rising tide of cardiovascular disease in diabetes,” that is hard to explain even with the rise in diabetes itself.


Robert S. Rosenson, MD, of Mount Sinai Hospital in New York City, noted the weaknesses of current risk assessment tools: most are based on information reported by patients themselves, and have been developed across a diverse set of populations worldwide. It’s already known that assessment tools that work with one population are not transferable to another population with a different ethnic background, diet, or culture.

Examinations of the components of metabolic syndrome are leading to explorations on how to connect risk scores between different conditions. “There’s been debate about A1C and cardiovascular disease (CVD), and those relationships are quite striking he said. He went through several CVD risk prediction models, but it’s not as simple as it appears to apply them to a population with diabetes.

A genetic-based tool for coronary artery disease (CAD) did not prove to be a good predictor for events when applied to a diabetes population. New tools are needed that can take existing models and apply them across populations using known biomarkers, while taking ethnic differences into account, he said.

W. Guy Weigold III, MD, of MedStar Washington Hospital Center, said that while there is clearly a relationship between diabetes and heart disease, “a one size fits all paradigm is not accurate.” And this is very true when using imaging to test for calcium in persons with diabetes.

Getting a calcium score would seem a straightforward biomarker. While imaging can help, it’s not a simple process, Weigold said. He reviewed several studies of this phenomenon, including a 2003 study by Wong, et al.1

“Patients who had diabetes with metabolic syndrome have some degree of coronary calcium, but some of these diabetics don’t have any calcium,” he said. “Right off the bat, it’s not necessarily the case that diabetes always equals coronary artery disease.”

Diabetics also accumulate calcium differently; their maximum scores tend to reach a peak, while nondiabetics who have heart disease tend to keep growing. Calcium scores can be used to stratify diabetics for treatment, but some still have higher mortality even with calcium scores of zero.

Prem Soman, MD, of the University of Pittsburgh, found that routine nuclear imaging of diabetics for silent ischemia yielded very little benefit, while Vera Rigolin, MD, of Northwestern University found that stress echocardiography provided excellent prognostic information; especially given how rapidly conditions can change for persons with diabetes.

Throughout the session, presenters took note of the different indicators and outcomes for men and women. Regina Druz, MD, of iVisitMD, of Long Island, New York, said that ensuring better care for women will require a different way of identifying patients who are at risk of heart disease. Until now, the process has been screening, taking history, taking a patients’ baseline, and hoping for early detection and treatment.

Today’s model of population health, Druz said, calls for physicians to deploy a risk stratification model that identifies groups of patients who need care before cardiac events occur. Among persons with diabetes, that may involve biomarkers or calcium tests but also an analysis of environmental and nutrition factors that may affect certain groups of patients. “This is risk-guided treatment,” she said. “We are moving from the disease-centric model to the patient-centric model.”

Brand-new indicators will emerge for identifying patients, because even the 2013 ACC/AHA guidelines don’t offer cardiologists as much as they could on risk assessment, Druz said. “Everything boils down to identification of global CAD risk,” she said. As physicians learn more about the effects of genetics and other factors that propel diabetes to trigger heart disease, “this is a framework that will grow with us into the future.”


1. Wong ND, Sciammarella MG, Polk D, et al. The metabolic syndrome, diabetes, and subclinical atherosclerosis assessed by coronary calcium. JACC. 2003;41:1547.