Atherosclerosis is the primary cause of coronary artery disease (CAD), one of the most common causes of illness and death worldwide. There is increasing recognition that lesion composition rather than size determines the acute complications of atherosclerotic disease in humans. Several studies suggested that thin-cap fibroatheroma (non-obstructive plaques) are prone to rupture and result in acute coronary artery occlusions [1]–[3], whereas obstructive, calcified plaques result in clinically stable angina pectoris.

Initiation and progression of the atherosclerotic lesion are highly complex processes, and many aspects of atherogenesis remain incompletely understood. Ectopic visceral adipose tissue was linked to the pathogenesis of atherosclerosis due to secretion of a multitude of pro- and anti-atherogenic cytokines and adipokines [4]. More recently, pericardial adipose tissue (PAT) was also reported to play an important role in coronary atherosclerosis [5], presumably through paracrine and vasocrine signaling of adipokines [6].

Adiponectin is the most abundant adipokine produced by adipose tissue. Serum levels of adiponectin are markedly decreased in patients with visceral obesity and states of insulin resistance such as non-alcoholic fatty liver disease and type 2 diabetes [7], [8]. There is an ongoing debate regarding adiponectin's significance for CAD. Although experimental data do suggest an atheroprotective effect [9], existing epidemiological data connecting adiponectin and cardiovascular disease are controversial. Low adiponectin levels have been linked to the presence of CAD [10] and were shown to be a risk factor for CAD [11] and cardiovascular events [12]. Low adiponectin levels were reported to be associated with a higher risk of acute coronary syndrome, independent of other traditional metabolic and cardiovascular risk factors [13]. Low adiponectin levels were also reported to be associated with the progression of coronary artery calcification as determined by electron-beam CT [14]. Furthermore, low serum adiponectin levels were shown to be an independent predictor of the extent of CAD and coronary lesion complexity as determined by coronary angiography [15]–[17]. In contrast, other studies including a recent meta-analysis of 7 prospective reports on adiponectin and coronary heart disease in Western populations failed to show an association between adiponectin and incident coronary heart disease as well as secondary cardiovascular events in patients with known CAD [18], [19].

The angiographic assessment of coronary luminal stenosis has been considered a surrogate marker of the severity of atherosclerosis. However, coronary angiography has low predictive value to assess atherosclerotic plaque burden or to predict acute coronary syndrome events [20]–[22]. It does not allow to identify non-obstructive coronary plaques or to determine the composition of atherosclerotic plaques. Therefore, in the present study, we used dual-source multi-slice CT (DSCT)-angiography to quantitatively and qualitatively assess coronary artery plaques to test whether 1) adiponectin serum levels are associated with coronary atherosclerotic plaque burden and 2) adiponectin levels are associated with coronary atherosclerotic plaque morphology.

Baseline characteristics of the study population are shown in Table 1. 303 consecutive patients with stable typical or atypical chest pain underwent DSCT-coronary angiography to exclude coronary artery stenosis.

In 60 patients no coronary plaques could be detected. The median number of coronary plaques was 3 (IQR: 1–6; range: 0–26). The median numbers of calcified, mixed and non-calcified plaques were 1 (IQR: 0–3; range: 0–22, number of patients with calcified plaques: 154), 0 (IQR: 0–1; range: 0–10, number of patients with mixed plaques: 102) and 0 (IQR: 0–2; range: 0–9, number of patients with non-calcified plaques: 126), respectively.

Our study was designed to examine the relationship of adiponectin with total coronary atherosclerotic plaque burden and atherosclerotic plaque morphology in humans. We demonstrate that 1) adiponectin levels are predictive of total coronary plaque burden, and 2) adiponectin levels are inversely correlated with the number of mixed and non-calcified plaques. No significant correlation was seen with calcified plaques. Our data suggest that the number of mixed and non-calcified plaques decrease by 1 for each 11.5 µg/ml and 13.2 µg/ml increase in serum adiponectin levels, respectively.

Pundziute et al. [23] reported that the number of coronary segments with mixed plaques as determined by MSCT was an independent predictor of acute cardiac events including cardiac death, nonfatal myocardial infarction and unstable angina requiring hospitalization. Hoffmann et al. [24] demonstrated that in contrast to calcified plaques, non-calcified plaques as determined by MSCT were consistently present in culprit lesions of patients with acute coronary syndrome. We have recently shown that rapid progression of angiographic stenosis severity occurred most frequently in coronary segments with non-calcified or predominantly non-calcified plaques as determined by MSCT [25]. Wolk et al. [13] reported that higher plasma adiponectin levels were associated with a lower risk of acute coronary syndrome, suggesting that the pathophysiological role of adiponectin may be related to the stability of atherosclerotic plaque rather than atherosclerotic burden. Our data indicate that adiponectin may reduce the number of non-calcified and mixed plaques without affecting stable calcified plaques. Consistent with these observations, Marso et al. [26] showed in a very recent study that low adiponectin levels were associated with lipid-rich yet not calcified coronary plaques in non-diabetic patients in bivariate analysis as determined by IVUS. Furthermore, Otake et al. [27] demonstrated that the necrotic core component ratio of culprit lesions as determined by IVUS in patients with acute coronary syndrome was negatively correlated with adiponectin levels in bivariate analysis. Therefore, the finding of an independent association between adiponectin and non-calcified and mixed coronary plaques may have important therapeutic and preventive implications for decreasing the risk of acute coronary events.

Although the exact mechanisms of the negative association between adiponectin and mixed and non-calcified plaques remain to be determined, several possible explanations should be considered. In addition to its beneficial effects on insulin sensitivity and lipid metabolism [28], adiponectin exerts its vasculoprotective effects through its direct actions on endothelial cells, monocytes, macrophages and other inflammatory cells, platelets and smooth muscle cells, thus modulating initiation and progression of atherosclerosis. Adiponectin augments endothelial NO production [29], [30], inhibits ox-LDL–induced endothelial ROS generation [31], suppresses the expression of endothelial adhesion molecules [32]–[34], attenuates leucocyte-endothelium interactions [35] and protects endothelial cells from apoptosis [36]. Furthermore, adiponectin inhibits macrophage activation [37], [38] and foam cell formation [39], promotes the clearance of early apoptotic cells by macrophages [40], inhibits smooth muscle cell proliferation [32], [41] and antagonizes the stimulatory effect of TNF-α on vascular smooth muscle cell calcification [42].

The present study has several strengths and limitations. Our data support the concept that adiponectin is an important marker in the pathogenesis of atherosclerosis since adiponectin remains significantly associated with plaque morphology in a fully adjusted multivariate model containing age, sex, BMI, hypertension, diabetes mellitus, smoking, family history of CAD, LDL-C, HDL-C, triglycerides, hsCRP, medication and PAT volume. However, since this was an association study, our study does not establish a causal relationship between adiponectin and coronary plaque morphology. Furthermore, the clinical implications of our results still need to be determined.

There are conflicting epidemiological data regarding the role of adiponectin in atherosclerosis with some studies showing strong inverse associations between adiponectin levels and CAD [10]–[17] and others failing to detect any association [18], [19]. Differences in study design, population characteristics and statistical adjustments may in part explain the apparent discrepancies. In addition, different methods of CAD assessment (CT-angiography or IVUS versus conventional angiography or CT-based assessment of coronary artery calcification) may yield distinct results. Adiponectin may modulate early stages of atherogenesis thus mainly affecting the development of non-calcified or mixed plaques.

In the present study adiponectin is not correlated with calcified plaques, the majority of coronary plaques in our study cohort. Therefore, although statistically highly significant, adiponectin levels are associated with a very modest decrease in total coronary plaque burden accounting for only 3% of the variability in total number of coronary plaques. In contrast, adiponectin accounts for approximately 20% of the variability in mixed and non-calcified plaque burden. Determination of the bioactive high molecular weight form of adiponectin may have possibly resulted in an even stronger association of adiponectin with plaque burden and morphology. Overall, our data emphasize the importance of additional risk factors in the pathogenesis of atherosclerosis.

Storage conditions of blood samples (e.g. storage time and temperature) were suggested to alter adiponectin levels [43], [44]. However, comparing adiponectin concentrations in serum aliquots stored for 21 and 33 months at −70°C, adiponectin levels were not significantly different (p = 0.45) suggesting that storage conditions did not impact on our results due to adequate specimen stability.

Finally, our study results may only be applicable to non-diabetic patients. Diabetics were clearly underrepresented in the present study (7% of patients). The small number of diabetic patients may also explain the lack of association between adiponectin levels and diabetes.

In summary, we demonstrate that adiponectin is inversely associated with coronary plaque burden and mixed and non-calcified plaques, suggesting an important role of adiponectin in the pathogenesis of atherosclerosis and possibly in the pathophysiology of acute coronary syndrome.