Author + information
- Torbjørn Holm, MD, PhD⁎ (, )
- Thor Ueland, PhD,
- Bente Halvorsen, PhD,
- Aina Hognestad, MD,
- Anne G.V. Holm, MD,
- Ragnhild Wergeland, MD,
- Lars Gullestad, MD, PhD,
- Pål Aukrust, MD, PhD and
- John Kjekshus, MD, PhD
- ↵⁎Department of Medicine, Diakonhjemmet Hospital ASA and Department of Cardiology Rikshospitalet HF, Vindern, 0032 Oslo, Norway
To the Editor:Patients with hypercholesterolemia, even with additional risk factors, may not during a lifetime develop coronary artery disease (CAD) (1). These observations suggest that protective mechanisms may balance those directly related to hypercholesterolemia and delay, or even abolish, the atherosclerotic process in otherwise high-risk individuals. To further elucidate these issues, we examined endothelial function as well as soluble markers of endothelial cell activation and inflammation in selected patients referred to coronary angiography with low or high levels of plasma low-density lipoprotein (LDL) cholesterol.
We consecutively screened all individuals attending to our coronary care unit for clinically indicated diagnostic coronary angiography in order to identify high-risk patients (LDL cholesterol >4.5 mmol/l; high-cholesterol group) and low-risk patients (LDL cholesterol <2.7 mmol/l; low-cholesterol group) with and without CAD, resulting in four different study groups: 1) Low LDL without CAD (n = 26), 2) low LDL with CAD (n = 21), 3) high LDL without CAD (n = 22), and 4) high LDL with CAD (n = 24). Significant CAD was predefined as at least two diseased vessels (>70% narrowing of luminal diameter). The exclusion criteria included previous or current statin treatment, previous episodes of acute coronary syndrome (myocardial infarction or unstable angina), signs of left ventricular dysfunction (left ventricular ejection fraction <50%), plasma homocysteine >14 mmol/l, smoking or other known risk factor for development of CAD, extensive use of alcohol, and signs of any acute inflammatory condition. All individuals were receiving aspirin (160 mg/day) and 72% were receiving beta-blocker therapy. The study was approved by the regional ethical committee. Signed informed consent was obtained from each individual.
The determination of endothelium-dependent and -independent vasodilatation in the forearm skin microcirculation was performed by iontophoresis of acetylcholine (Ach) and sodium nitroprusside (SNP) with laser-Doppler perfusion measurements (2). Plasma levels of soluble markers of endothelial cell activation and inflammation were measured by enzyme immunoassays (2–4).
For comparisons of two groups of individuals, the Mann-Whitney rank-sum test was used. When comparing four groups of individuals, the Kruskal-Wallis test was used. Probability values of p < 0.05 (two-sided) were considered statistically significant.
With the exception of plasma lipid levels, there were no significant differences in clinical variables between the four groups of individuals. Within the low- and the high-cholesterol groups, the lipid levels were comparable between those with and without CAD.
The vasodilatory response to Ach, but not the endothelial-independent response to SNP, was significantly lower in both groups of CAD patients compared to those with similar LDL levels and normal coronary angiograms (Fig. 1).Coronary artery disease patients in both the high- and the low-LDL groups had markedly elevated plasma levels of von Willebrand factor when compared to those with normal angiograms, but with similar LDL levels, showing a similar pattern as the endothelial-dependent Ach response (Fig. 1).
The highest plasma levels of high-sensitivity C-reactive protein (hs-CRP) and tumor necrosis factor (TNF)-alpha were found in those with CAD, regardless of cholesterol levels (Fig. 2).In contrast, the prototypical anti-inflammatory cytokine interleukin (IL)-10 showed a different pattern with raised levels in those with hypercholesterolmia and normal coronary angiogram comparing to the other groups (Fig. 2). This “high-risk” cholesterol group without CAD was also characterized by a low TNF-alpha/IL-10 ratio, suggesting anti-inflammatory net effects in these individuals compared with the other groups (Fig. 2).
Similarly to markers of endothelial cell activation and systemic inflammation, those with CAD had enhanced plasma levels of soluble CD40 ligand (sCD40L) compared with those with normal angiograms, regardless of cholesterol levels, suggesting enhanced platelet activation in these individuals (3) (Fig. 1). A similar pattern was also seen for soluble P-selectin, reflecting both endothelial cell and platelet activation (2,3) (Fig. 1).
High levels of hs-CRP (odds ratio [OR] for CAD: 1.89 [95% confidence interval (CI) 1.06 to 3.33], p = 0.031), high levels of P-selectin (OR: 1.11 [95% CI 1.03 to 1.20], p = 0.009), and low vasodillatory responses to Ach (OR: 0.96 [95% CI 0.91 to 0.99], p = 0.035) were all independently associated with CAD in the low-risk group. In the high-risk group, high levels of IL-10 were independently associated with a normal coronary angiography (OR: 0.035 [95% CI 0.002 to 0.50], p = 0.01), and high levels of sCD40L were associated with CAD (OR: 6.57 [95% CI 2.00 to 313.1], p = 0.02).
In contrast to the increased levels of inflammatory markers in those with CAD, we found that the only group with raised levels of the anti-inflammatory cytokine IL-10 was that with hypercholesterolemia and normal coronary angiograms. This group also had particularly low ratios of TNF-alpha/IL-10, suggesting anti-inflammatory net effect in this “high-risk” group without CAD. Interleukin-10 has been found to inhibit the release of inflammatory cytokines, tissue factor, and matrix metalloproteinases from leukocyte subsets in CAD patients, possibly promoting plaque stabilization in these patients (4–5). These findings suggest an attenuating role of IL-10 in atherogenesis, possibly involving protective effect on the endothelium. Indeed, we found that the raised IL-10 levels in patients with hypercholesterolemia and normal coronary angiograms were accompanied by preserved endothelial function. Such a role has also been shown in various animal models, and IL-10 has recently been proposed as an “immunological scalpel” in the atherosclerotic process (5). Our findings herein suggest that IL-10 also could represent a protective phenotype in those individuals with high LDL levels who do not develop CAD.
- American College of Cardiology Foundation
- Ridker P.M.,
- JUPITER Study Group
- Aukrust P.,
- Muller F.,
- Ueland T.,
- et al.
- Terkeltaub R.A.