Cardiovascular Magnetic Resonance in Diabetic Patients

In addition, the workup of these patients is challenging because of the high prevalence of advanced disease, various overlapping expression of abnormalities, their rapid progression over time, their atypical presentation (eg, atypical symptoms caused by autonomous neuropathy), and the overweight frequently encountered in patients with type 2 diabetes mellitus, as well as limited exercise capacity because of peripheral vascular disease in ≈30% of patients requiring stress testing. Various techniques are used in clinical practice and research to assess the above named pathologies, leading to a continuous improvement in understanding the natural history and relative importance of different biomarkers for prognosis and risk stratification. Especially, the detection of epicardial coronary artery disease and its prognostic relevance has been an area of discussion given the difficulties to perform exercise testing in these patients, the overlap of significant epicardial coronary artery disease with diffuse coronary artery disease, microvascular disease, and endothelial dysfunction, and the specific difficulties the diabetic population poses on echocardiography and single photon emission computed tomography (SPECT). Echocardiography is frequently the first-line technique in assessing patients with cardiovascular presentations because of its immediate availability. However, in the diabetic population, image quality is frequently suboptimal because of obesity, and exercise echocardiography is often impossible. The need for dobutamine as a pharmacological stressor has been shown to be an independent predictor of adverse outcome. Although dobutamine stress echocardiography allows to riskstratify patients, it remains far from perfect, with a yearly mortality rate in patients with a negative dobutamine stress echocardiography of 4%. Echocardiography is unable to differentiate between epicardial and microvascular coronary disease and misses smaller MIs not causing significant wall motion abnormalities. SPECT imaging is the most frequently used method to assess myocardial ischemia, especially in the United States. In patients with diabetes mellitus, SPECT has several distinct disadvantages. Its spatial resolution of≈1 cm×1 cm does not allow detection of subendocardial ischemia. It relies on regional differences in myocardial blood flow and is thus less accurate in patients with triple vessel disease and balanced ischemia or diffuse microvascular dysfunction. The use of radioactive tracers is unfavourable in the relatively young diabetic cohort with rapid progress of disease and atypical presentations, frequently requiring repeat examinations. Obese patients tend to have lower signal in the inferolateral wall, causing false-positive findings. Finally, SPECT imaging has a low sensitivity for nontransmural MI. Despite these limitations, a recent meta-analysis of SPECT studies confirmed its ability for risk classification in diabetes mellitus. Averaged across patients with and without symptoms, a normal SPECT scan conferred a low annualized event rate for cardiac death and MI of 1.9%. Computed tomography plays an increasing role in assessing patients with diabetes mellitus, and there is an ongoing debate on its use in asymptomatic patients or calcium screening as a gatekeeper to ischemia testing. However, recent data from the Factor 64 study show that in diabetic patients with or without symptoms, computed tomographic angiography does not improve patient management or outcome. As a result of the difficulties to sufficiently work up the cardiovascular status of patients with diabetes mellitus, these patients often undergo invasive angiography frequently without therapeutic consequence. Cardiovascular magnetic resonance (CMR) imaging has made continuous progress in assessing patients with diabetes mellitus. There is convincing data on the ability of CMR in detecting MI beyond other tests. The ICELAND MI study revealed a prevalence of 27% MI in the general population aged 67 to 93 years and 32% in patients with diabetes mellitus. Importantly, the number of unrecognized MIs was high in both groups (17% and 21%), demonstrating significant (Circ Cardiovasc Imaging. 2016;9:e004699. DOI: 10.1161/CIRCIMAGING.116.004699.) © 2016 American Heart Association, Inc.