This article is more than 5 years old. Information may no longer be current.
Different approach to CVD risk assessment should be taken in patients with diabetes
Click Here to Manage Email Alerts
PHILADELPHIA — Models to assess CV risk in patients with diabetes often underestimate risk, but the addition of clinical tests including imaging tests and biomarkers may improve this, according to a presentation at the Heart in Diabetes Medical Conference.
Christie M. Ballantyne, MD, professor of medicine, chief of the section of cardiovascular research and director of the Center for Cardiovascular Disease Prevention at Baylor College of Medicine in Houston, and diplomate for the American Board of Clinical Lipidology, discussed the various results often seen in patients with diabetes with current models of risk assessment.
Ballantyne provided a case study of an older woman who is overweight with a history of hypertension and diabetes, and no CVD. Based on her BP, lipid profile and other pertinent information, the Adult Treatment Panel III guideline would not require a risk calculation, according to the presentation, but with the Framingham Risk Score, she had a 11% 10-year risk for MI. Reynolds Risk Score showed an 8% 10-year risk, and with the Pooled Cohort Equations, the patient had a 30.4% 10-year risk.
“The controversy is not which risk equation,” Ballantyne said. “The controversy, which everyone is ignoring, is what health conditions do we include for risk prediction? Framingham is just fatal and nonfatal MI. Reynolds Risk Score includes stroke and so does the Pooled Cohort Equation. That’s important for women because stroke is more common in this woman and having an MI, probably at least as common. When you only give a risk for MI, you’re underestimating her events. Neither one of them gives the risk for revascularization and angina.”
According to unpublished data from the ARIC study, the most common events in patients with a mean age of 76 years were stroke (4.5 per 1,000 person-years), CHD including revascularization (14.5 per 1,000 person-years) and HF hospitalization (19.6 per 1,000 person-years). Most of the of the risk assessment models don’t address the common cardiac problems, even though patients want to know their global risk score, Ballantyne said.
“This is the epidemic as we talk about diabetes,” Ballantyne said. “According to health care expenditures, the most expensive [CV] event is now [HF], and ... something that can be prevented. The other issue is this whole approach towards older patients. It is irrational. What is the single greatest risk factor? It’s age, and yet, as people get older, our guidelines take an interesting attitude with less intensive therapy.”
Less aggressive treatment guidelines suggest a BP treatment goal below 150 mm Hg systolic/90 mm Hg diastolic in patients 60 years and older. In patients 75 years and older, there are no firm recommendations for statin therapy.
“At the age of 76, you’re not worried about 20-year risk, but you are worried about a 5-year risk, so you need to have a different approach towards older individuals,” Ballantyne said.
In the SPRINT trial, patients aged 75 years and older who were at high risk had the greatest benefit with treatment. Although there were some adverse effects, most of them were manageable, Ballantyne said. Similar results were seen in the JUPITER trial, where troponin and N-terminal pro-B-type natriuretic peptide were measured.
In the ARIC study, troponin-T and NT-proBNP predicted HF hospitalization, CHD and all-cause mortality better than C-reactive protein, according to the presentation. About 75% of patients in this cohort were considered at-risk.
“We can predict more accurately who had [HF] hospitalizations than we can who’s going to have MI or stroke, yet we don’t do it in practice nor does anybody try to prevent [HF],” Ballantyne said.
In a study published in Circulation, patients with diabetes were more likely to have higher levels of high-sensitivity cardiac troponin T over a 6-year period, and the levels of HF were more dramatic, he said.
According to Ballantyne, NT-proBNP is a good biomarker, but its levels go down as a patient becomes obese, and troponin I and T increase with both diabetes and obesity; also, when a clinician adds troponin elevation or NT-proBNP to BP measurements, the ability to identify patients at high risk for CV or HF improves.
“In terms of atherosclerotic events, calcium scoring is better, but for [HF] hospitalization, these two are very useful markers,” Ballantyne said. “There’s also an association with atrial fibrillation.”
ECGs can also improve global CVD risk prediction in patients with diabetes and prediabetes, as it can show elevations of strain. Other imaging modalities that can increase risk prediction are eye exams and neuropathy assessments.
In the MESA study, adding biomarker tests to base models of risk assessment had a “robust change in risk assessment,” Ballantyne said.
The Dallas Heart Study showed that younger populations benefit more from traditional risk factors, but as patients get older, those become less effective, he said.
New data from the ARIC study showed that lipoprotein(a) was a strong predictor for incident CHD, ischemic stroke and CVD for individuals with diabetes or impaired fasting glucose. Similar results were found in the BiomarCaRE consortium, in which researchers measured lipoprotein(a) in patients with and without diabetes.
“[Lipoprotein(a)] might be something that is actually not just a risk marker, but a targeted therapy for individuals who have diabetes in the future,” Ballantyne said.
Ballantyne said accurate risk assessment and prevention is very important.
“[The studies] that we need to do [are] not taking people with advanced [CVD] risk, but taking diabetics at high risk and trying to prevent [HF] and other outcomes in those populations.” – by Darlene Dobkowski
Reference:
Ballantyne CM. Diabetes – Assessing CV Risk and Approach to Therapy. Presented at: Heart in Diabetes Medical Conference; July 14-16, 2017; Philadelphia.
de Lemos JA, et al. Circulation. 2017;doi:10.1161/CIRCULATIONAHA.117.027272.
Everette BM, et al. Circulation. 2015;doi:10.1161/CIRCULATIONAHA.114.014522.
Saunders JT, et al. Circulation. 2011;doi:10.1161/CIRCULATIONAHA.110.005264.
Selvin E, et al. Circulation. 2014;doi:10.1161/CIRCULATIONAHA.114.010815.
The SPRINT Research Group. N Engl J Med. 2017;doi:10.1056/NEJMoa1511939.
Waldeyer C, et al. Eur Heart J. 2017;doi:10.1093/eurheartj/ehx166.
Disclosure: Ballantyne reports receiving grant and research support from Abbott Diagnostic, ADA, Amarin, American Heart Association, Amgen, Esperion, Ionis, NIH, Novartis, Pfizer, Regeneron, Roche Diagnostic and Sanofi-Synthelabo; and consulting for Abbott Diagnostics, Amarin, Amgen, AstraZeneca, Boehringer Ingelheim, Eli Lilly, Esperion, Ionis, Matinas BioPharma Inc., Merck, Novartis, Pfizer, Regeneron, Roche Diagnostic and Sanofi-Synthelabo.