Login to view PhD Thesis

Enter your username and password here in order to log in on the website:

Forgot your password?

Cardiovascular Disease in Type 2 Diabetes Mellitus: Prevalence of and Relationship Between Myocardial Ischemia, Carotid Arterial Disease, Peripheral Arterial Disease, and Left Ventricular Dysfunction

Mikael Kjaer Poulsen


IntroductionThis Ph.D. thesis comprises three original articles based on a single observational clinical trial conducted at the Departments of Endocrinology, Nuclear Medicine, and Cardiology, Odense University Hospital, in collaboration with the Department of Statistics, University of Southern Denmark during the 1st of January, 2006, to the 1st of June 2009. 

The main purpose was to obtain the prevalence of myocardial ischemia, carotid, and peripheral arterial disease, and left ventricular (LV) diastolic dysfunction in type 2 diabetes mellitus (T2DM) patients without known or suspected cardiovascular disease (CVD). Secondly, to assess the interrelationship between each of these parameters. Thirdly, to establish an algorithm identifying T2DM patients with a low, intermediate, or high risk of myocardial ischemia on myocardial perfusion scintigraphy (MPS). Finally, to assess the relationship between abnormal LV diastolic function, myocardial perfusion, and vascular function to further investigate the causes of LV diastolic dysfunction in T2DM patients. 

MethodsT2DM patients without known or suspected CVD referred consecutively to a diabetes clinic for the first time (n=305, aged: 58.6 ± 11.3 years, diabetes duration: 4.5 ± 5.3 years) and age-matched non-diabetic reference subjects (n=40, aged: 55.7 ± 5.5 years) were screened for myocardial ischemia, carotid, and peripheral arterial disease, by means of MPS, carotid artery B-mode ultrasonography, peripheral ankle and toe systolic blood pressure measurements. Only the T2DM patients were further screened for left ventricular systolic and diastolic function by echocardiography. 

ResultsIn the T2DM patients, the prevalence of myocardial ischemia, carotid, and peripheral arterial disease was 30% (95% CI: 25-36), 42% (95% CI: 37-48), and 15% (95% CI: 11-19), respectively, almost three times higher than in the age-matched reference subjects (p=0.007, p=0.001, and p=0.09, respectively). The T2DM patients with CVD in one vascular territory had a significantly increased risk of CVD in other territories as well (OR: 1.99, 2.09, and 3.09 for patients with myocardial ischemia, carotid, and peripheral arterial disease, respectively). The interrelations between CVD manifestations were, however, not strong enough to allow us to examine only one vascular territory and then identify all cases of CVD, since 40%, 52%, and 22% of the T2DM patients demonstrated exclusively myocardial ischemia, carotid, or peripheral arterial disease. 

An algorithm was established in the attempt to identify which T2DM patients had a low (n=96), intermediate (n=65), or high risk (n=115) of myocardial ischemia on MPS. For each of these three groups the prevalence of myocardial ischemia was 15%, 23%, or 43%, respectively. Overall the algorithm reduced the number of T2DM patients referred to MPS from 305 to 144. The sensitivity and specificity of the algorithm to detect myocardial ischemia on MPS was 68% and 62%, respectively. 

In the T2DM patients the prevalence of left ventricular (LV) diastolic dysfunction and left atrial (LA) volume index >32 ml/m2 was 40% (95% CI: 34-46) and 32% (95% CI: 27-38), respectively. Predictors of moderate or severe LV diastolic dysfunction were total arterial compliance, valvulo-arterial impedance, Nterminal pro-type B natriuretic peptide, LV mass index, and summed stress score on MPS, whereas pulse pressure and carotid arterial compliance were not, after adjusting for age, male gender, diabetes duration, and hypertension. On multivariable modeling, summed stress score on MPS (p=0.001) and valvulo-arterial impedance (p=0.03) remained predictors of grade 2, LV diastolic dysfunction, and only summed stress score (p=0.001) was a predictor of moderate to severe LA dilatation. 

ConclusionsMyocardial ischemia and carotid arterial disease were common in the T2DM patients and almost three times more common than in the age-matched non-diabetic reference subjects. In the T2DM patients, signs of CVD in one vascular territory carried a significantly increased risk of CVD in other territories as well, however, the interrelationship between CVD manifestations were not strong enough to allow us to examine one single vascular territory and then identify all cases of CVD. Therefore, if screening for CVD in T2DM patients without known or suspected CVD, one should examine all three vascular territories. The constructed algorithm was able to stratify, which T2DM patients had a low, intermediate, or high risk of myocardial ischemia on MPS using both clinical and “sophisticated” predictors. However, the algorithm had a rather low sensitivity and specificity to detect myocardial ischemia in combination with a high cost and time consumption. 

The prevalence of LV diastolic dysfunction was high in the T2DM patients. The study showed a close association between the presence of moderate or severe abnormal LV filling and abnormal myocardial perfusion on MPS. In contrast, the association with abnormal vascular function was considerably less prominent. Thus, supporting that moderate or severe LV diastolic dysfunction and LA dilatation in the early phase of T2DM is related to intrinsic LV dysfunction in a greater extent than arterial stiffening with abnormal ventriculo–arterial coupling.