Coronary Vasomotor Function Is Abnormal in First-Degree Relatives of Patients With Type 2 Diabetes

The study subjects were healthy volunteers recruited by advertisements for the University of Malaya. All gave written informed consent to participate in the study, which was approved by the local ethics committee. To be eligible as a first-degree relative, subjects had to have one or more parent with type 2 diabetes. To avoid any confounding influence on endothelial function, subjects were excluded if they had a history of smoking, manifest type 2 diabetes, hypertension, hypercholesterolemia (serum cholesterol ≥6.5 mmol/l), and concurrent CVD, including current or previous history of myocardial infarction and angina. None were on regular medication. At the screening visit, subjects underwent a standard 75-g oral glucose tolerance test (OGTT), and those with impaired glucose tolerance (IGT) according to American Diabetes Association criteria (7) were identified.

On the first visit, subjects were studied following an overnight fast and had the following procedures: a blood draw, an OGTT, and measurement of FMD. Blood was drawn for measurement of serum glucose, lipids, serum inflammatory markers (interleukin-6 [IL-6] and high-sensitivity C-reactive protein), and soluble intercellular adhesion molecule-1 (sICAM-1), a measure of endothelial cell activation (8). Following the blood draw, a standard OGTT was performed with venous blood sampling at 0, 15, 30, 60, 90, and 120 min to measure plasma glucose and serum insulin and to allow determination of insulin resistance using the homeostatic model of insulin resistance (9). Following the OGTT, peripheral endothelial function was determined using the FMD method as described by Celermajer et al. (10). Subjects returned for a second study visit to assess the coronary flow reserve. Coronary flow reserve was assessed by transthoracic Doppler echocardiography as we previously described (11,12). In women, investigations were made during the menstrual phase because the menstrual cycle may affect vascular reactivity (12). In brief, echocardiography was performed with the Acuson Sequoia 512 (Siemens Medical Solutions USA, Mountain View, CA) initially using a 3.5-MHz probe to assess left ventricular structure and function. None had left ventricular hypertrophy, and all had normal left ventricular function. Thereafter, a 7.0-MHz transducer was used to visualize coronary blood flow in the left anterior descending coronary artery by color Doppler echocardiography. Baseline spectral Doppler signals were recorded in the distal portion of the left anterior descending coronary artery over five cardiac cycles at end expiration by transthoracic Doppler echocardiography. Intravenous adenosine was then administered (140 μg · kg⁻¹ · min⁻¹ i.v.) for 2 min to record spectral Doppler signals during hyperemic conditions. Mean diastolic velocities were measured at baseline and peak hyperemic conditions measured from the Doppler signal recordings. Measurements were averaged over three cardiac cycles. Coronary flow reserve velocity (CFVR) was defined as the ratio of hyperemic to basal mean diastolic velocity. All echocardiographic assessments were performed and analyzed by an experienced investigator (K.H.) who was blinded to the clinical information of the subjects.

Of the 32 first-degree relatives who were screened, 8 were excluded because of IGT. Altogether, 24 eligible first-degree relatives and 26 healthy control subjects with normal OGTTs participated in the study. When compared with control subjects, CFVR and FMD were significantly lower in first-degree relatives (Table 1). sICAM-1, together with serum markers of inflammation, were elevated in first-degree relatives (Table 1). There was significant inverse correlation between CFVR and FMD with serum sICAM-1 and IL-6 (Table 1). In a multivariate analysis including age, sex, BMI, and homeostatic model assessment of insulin resistance, first-degree relative remained a significant determinant of CFVR and FMD (P = 0.028 and P = 0.007, respectively). In the entire study population, CFVR and FMD were significantly correlated (r = 0.34, P = 0.01) (Fig. 1).

Prior et al. (13) recently demonstrated that position emission tomography–derived endothelium-dependent coronary vasoreactivity was significantly diminished in insulin-resistant individuals, as well as in patients with IGT and normotensive hypertensive patients with type 2 diabetes. They also showed that total vasodilator capacity, a measure of coronary flow reserve, was decreased in type 2 diabetes. An impaired coronary flow reserve has been reported in patients with type 2 diabetes (14). We now extend this observation to apparently healthy nonobese first-degree relatives. In this study, coronary flow reserve was assessed by adenosine-induced increases in coronary flow, which is mediated largely through vascular smooth muscle relaxation, although endothelium-dependent mechanisms may contribute to the hyperemic response because inhibition of nitric oxide synthase decreases hyperemic flow in both peripheral and coronary circulation (15,16). Therefore, a restricted CFVR in first-degree relatives may reflect functional alterations of the endothelium, vascular smooth muscle cell, or both. The significant correlation between FMD and CFVR suggests that there is endothelial dysfunction of both peripheral and coronary circulation in first-degree relatives. Indeed, we found that the decreased CFVR was related to increases in serum sICAM-1 and IL-6, markers of generalized endothelial dysfunction (7,17). It should be emphasized that the presence of correlations between variables neither proves causality nor defines the manner in which they are related. Nevertheless, they can provide the basis to offer some hypothesis. It may suggest a role for inflammation in the altered CFVR. In this respect, a link between inflammation and endothelial dysfunction has been suggested in type 2 diabetes (17). Oxidative stress may also play a pathophysiological role in the endothelial dysfunction in first-degree relatives (18), although we did not measure parameters of it. On the other hand, it could be argued that the increase in sICAM-1, which is a marker of endothelial cell activation (7), could be secondary to subclinical CVD in first-degree relatives, a possibility consistent with evidence that concentrations of sICAM-1 are correlated with the degree of intima-media thickness (19), which is increased in first-degree relatives (20). Our study was not designed to compare the value of FMD and CFVR as early markers of cardiovascular risk in first-degree relatives. It should be noted that the noninvasive determination of CFVR is a relatively new technique with emerging evidence of its prognostic value in clinical settings (21). In conclusion, our study shows that the coronary vasomotor function is abnormal in first-degree relatives. These findings, when taken together with previous reports of subclinical CVD in first-degree relatives, strongly support the existence of a link between genetic predisposition to endothelial dysfunction and insulin resistance and/or type 2 diabetes and increased risk of CVD.