The present cross-sectional study demonstrated increased serum RBP4 levels in patients with CAD (either AMI or SA) compared to age- and sex-matched individuals without CAD, but with at least 2 classical cardiovascular risk factors. In our study cohort, the presence of significant CAD was independently related to RBP4 and hsCRP circulating levels. This is the first study demonstrating the independent association of RBP4 with CAD severity indexes (e.g. Gensini score and number of narrowed vessels).
In spite of the rather established relation to insulin resistance and cardiovascular risk factors, the actual association of RBP4 levels with atherosclerotic-related cardiovascular disease is still controversial. Previous studies have shown the positive association of RBP4 with subclinical atherosclerosis -. In elderly men, RBP4 levels correlated with hypertriglyceridemia and prior cerebrovascular disease . Furthermore, plasma RBP4 concentration proved to be related to the presence of cardiovascular disease in non-obese, non-diabetic subjects . A recent prospective trial suggested its value in predicting CAD in a large women cohort during a follow-up period of 9 to 16 years . In contrast, another large prospective trial doubted the additive prognostic value of RBP4 among CAD-free men and women . Further recent studies in experimental models of high cardiovascular risk in humans have actually exhibited a protective role of RBP4 in vasculature ,, or even reduced levels in men with AMI , which add to the controversy concerning the actual role of RBP4 in atherosclerosis. However, in these studies the samples are smaller, they mainly include men, and do not distinguish between subjects according to the actual vessel pathology but rather according to cardiovascular risk. In our study cohort, consisting of patients with SA, AMI and non-CAD subjects, RBP4 and hsCRP levels were independently associated with the presence of significant CAD. It is worth mentioning that the non-CAD patients had at least two cardiovascular risk factors, presented with high suspicion of CAD. Thus, this is the first study implicating the potential of serum RBP4 levels in discriminating patients with established CAD from high risk patients, using coronary angiographic criteria. Notably, the presence of AMI did not affect RBP4 serum levels among CAD patients, implicating the dissociation of RBP4 from acute coronary event. Perhaps, circulating RBP4 is predominantly influenced by the presence of coronary atherosclerotic lesions rather than the atherosclerotic plaque destabilization. However, this postulation needs further investigation.
Accumulating data support the relationship of novel adipokines with CAD severity . To our knowledge this is the first study indicating the association of CAD severity, expressed by Gensini score, with RBP4 levels in addition to hsCRP and HDL. Interestingly, RBP4 levels significantly increased across the number of diseased vessels. Those striking findings suggest an interaction between RBP4 and the pathophysiological process of coronary atherosclerosis. The latter notion has been recently supported by the higher RBP4 expression in epicardial fat derived from CAD rather than non-CAD patients . Those authors hypothesized higher protein released in close proximity to coronary arteries, implying a causative role in the pathogenesis of coronary atherosclerosis. Perhaps, measuring serum RBP4 could contribute to patients’ risk stratification in order to avoid diagnostic procedure that bares risks itself, such as coronary angiography. Unambiguously, future trials will clarify the emerging role of serum RBP4 as a valid biomarker of CAD extent.
Regarding the underlying mechanisms, we observed the independent correlation of RBP4 with insulin resistance indices and established markers of inflammation, like hsCRP. More recently, RBP4 levels independently predicted early endothelial dysfunction, linking adipose tissue inflammation and subclinical atherosclerosis in non-diabetic individuals . The association of RBP4 with markers of inflammation is supported by several studies. RBP4 was found to induce in vitro inflammation in endothelial cells, by stimulating expression of proinflammatory molecules, such as vascular cell adhesion molecule 1 (VCAM-1), E-selectin, intercellular adhesion molecule 1 (ICAM-1), monocyte chemoattractant protein 1 (MCP-1), and interleukin-6 (IL-6) . Those effects may be mediated via the activation of NADPH oxidase and NF-κB leading to endothelial inflammation. In another study involving patients with diabetes and CAD, RBP4 levels rose in subjects with both conditions, and were rather correlated with TNFa than with markers of insulin resistance .
The role of retinoids in lipid metabolism is well known and is mediated through the regulation of ApoC-III and VLDL production and fatty acid oxidation . A relatively large study of patients with type 2 diabetes or CAD previously reported the relation of RBP4 levels to an unfavorable lipid profile . In the diabetic state, a positive association of RBP4 with plasma triglycerides levels and VLDL-apoB100 total fractional catabolic rate has also been found . The latter evidence suggests a potential interaction between RBP4 and CAD through pro-atherogenic lipoproteins and their enzymes. Moreover, RBP4 has been recently identified as an HDL-associated protein; it is demonstrated that in patients with acute coronary syndrome, HDL shifts to an inflammatory profile, which can in turn, alter the protective effects of HDL on the atherosclerotic plaque. Thus, in this inflammatory milieu, RBP4 could also share such properties . In parallel, RBP4 has exhibited modest heritability and sexual dimorphism (higher levels in men) , while it is considered to represent a link between visceral adiposity and cardiovascular disease . Taken together, our study failed to reveal any association of RBP4 with lipids, gender or BMI. Perhaps, the lipid-lowering medications, the low percentage of women and the vast majority of overweight, but non-obese participants, might have confounded the relationship of the above parameters, respectively, with RBP4 levels.
The major limitation of the present investigation was the cross-sectional design, which prevented us from inferring cause-effect relationship of RBP4 with CAD. Although we did not recognise differences between acute and stable condition of CAD, the cross-sectional design of our study did not allow us to evaluate the association of RBP4 with either AMI occurrence or long-term clinical outcomes. Since the majority of patients with classical cardiovascular risk factors (e.g. diabetes, dyslipidemia, hypertension etc.) were already treated, we cannot rule out the plausible effects of pharmaceutical agents (e.g. statins) on RBP4, leading to underestimation of its predictive power. Another important limitation was the considerable differences in some biochemical parameters between CAD and non-CAD groups, which might have affected RBP4 fluctuations. Despite the independent association between RBP4CAD and CAD diagnosis, the absence of matching for baseline characteristics may weaken our conclusions. Finally, as our control group sample comprised of patients with cardiovascular risk factors, we couldn’t extrapolate our conclusions to healthy subjects. Another study limitation is the potential influence of the transthyretin TTR-RBP4 complex in the affinity towards RBP4 that could be interfering with the ELISA measurement of RBP4. However, a previous study has shown that circulating RBP4 and TTR were not affected by human obesity or T2DM, compared to lean controls . The same could apply to our population as well, considering that a low grade inflammation often coexists with T2DM and obesity and that in our study we did not recognize differences between acute and stable condition of CAD.