Cardiac phenotyping with contrast cardiac magnetic resonance imaging in a cohort of patients with type 2 diabetes: Non-invasive assessment of fibrosis, left ventricular filling and the effect of GLP-1 treatment

Patients with type 2 diabetes (T2DM) have an increased risk of developing heart disease. Even though decades of improved multifactorial treatment have reduced this risk, a considerable
difference between T2DM patients and controls of the same age and sex remains. The first step toward improved care for these patients is to gain further knowledge of the underlying mechanisms leading to cardiac disease and identify risk markers. Previous studies have found that patients with T2DM show signs of impaired filling as well as signs of cardiac remodeling.
Animal studies, human autopsy studies, and smaller human imaging studies suggest that increased myocardial fibrosis and impaired myocardial blood flow are parts of the pathogenesis. These important factors, however, remain understudied in larger human cohorts. A proposed treatment for impaired filling is with a glucagon-like peptide-1 (GLP-1) receptor agonist. Treatment with this drug class has been shown to reduce cardiovascular
death, but the underlying mechanisms are unclear. One mechanism could be improved cardiac function such as left ventricular filling.


In this thesis, we aimed at quantifying localized fibrosis in T2DM patients and assessing the relationship to the cardiac phenotype and other diabetes complications. Further we aimed at, studying the association of interstitial fibrosis and myocardial blood flow with left ventricular filling. In addition to studying the effect of liraglutide, a GLP-1 receptor agonist, on the cardiac diastolic function.


A large cross-sectional study of 296 patients was conducted. Between 2016-2019, patients with T2DM were recruited from the outpatient clinic at Slagelse/Næstved Hospital. We performed an extensive medical examination of the participants, including a cardiac magnetic resonance imaging scan with the examination of myocardial blood flow and assessment of myocardial fibrosis. Additionally, a double-blinded randomized placebocontrolled trial was conducted. 40 patients with signs of diastolic dysfunction were randomized to liraglutide or placebo for 18 weeks of treatment. At baseline and at follow-up, cardiac magnetic resonance imaging was performed to assess our primary outcome of improved left ventricular diastolic filling.


We found that about 10% of patients with type 2 diabetes had localized non-ischemic LGE lesions with a distinct pattern, which could be localized fibrosis. The affected patients had signs of impaired cardiac filling and a higher prevalence of diabetes complications; retinopathy and autonomic neuropathy. Diffuse fibrosis were associated with signs of decreased left ventricular compliance and impaired myocardial blood flow at stress and
myocardial perfusion reserve were associated with early relaxation of the cardiac myocyte and restoring forces of the left ventricle. Liraglutide improved blood glucose levels in the patients and induce a weight loss but did not significantly improve cardiac filling parameters.


In conclusion, we found a relatively high prevalence of what could be focal scar tissue that displayed a distinct pattern, that has not been systematically described previously. These lesions seemed to be existing in more severely ill patients. We demonstrated that the signs of diffuse fibrosis and myocardial blood flow are independently associated with different parts of cardiac filling in patients with type 2 diabetes. Further, we found that short-term
treatment with liraglutide, a GLP-1 receptor agonist, did not improve cardiac filling. Hence, the cardioprotective effect of liraglutide likely stems from other underlying mechanisms.