Author + information
- Received May 10, 2006
- Revision received June 19, 2006
- Accepted June 26, 2006
- Published online October 17, 2006.
- Giuseppe Patti, MD⁎,
- Massimo Chello, MD⁎,
- Vincenzo Pasceri, MD, PhD, FACC†,
- Diego Colonna, MD‡,
- Annunziata Nusca, MD⁎,
- Marco Miglionico, MD⁎,
- Andrea D’Ambrosio, MD⁎,
- Elvio Covino, MD⁎ and
- Germano Di Sciascio, MD, FACC⁎,⁎ ()
- ↵⁎Reprint requests and correspondence:
Dr. Germano Di Sciascio, Department of Cardiovascular Sciences, Campus Bio-Medico University, Via E. Longoni, 83, 00155 Rome, Italy
Objectives The goal of this work was to investigate whether protection from myocardial injury during percutaneous coronary intervention (PCI) by atorvastatin is related to reduction of endothelial inflammatory response.
Background In the randomized ARMYDA (Atorvastatin for Reduction of MYocardial Damage during Angioplasty) trial, 7-day pre-treatment with atorvastatin before PCI significantly reduced procedural myocardial injury; mechanisms underlying this effect are not characterized.
Methods In a planned subanalysis of the ARMYDA trial, a subgroup of 76 patients was blind-tested for measurement of plasma levels of vascular cell adhesion molecule-1 (VCAM-1), intercellular cell adhesion molecule-1 (ICAM-1), and E-selectin: 38 patients belonged to atorvastatin (40 mg/day) and 38 to the placebo arm. Adhesion molecules were evaluated 7 days before intervention, immediately before PCI, and after 8 and 24 h.
Results Reduction of procedural myocardial injury after statin pre-treatment was also confirmed in this subgroup. Intercellular cell adhesion molecule-1, E-selectin, and VCAM-1 levels were not different at randomization and before intervention in either arm. At 8 h, increase of ICAM-1 levels was similar in the 2 arms, whereas 24-h levels were significantly lower in the atorvastatin versus placebo group (282 ± 56 vs. 325 ± 70 ng/ml; p = 0.007). Attenuation of E-selectin elevation occurred at 8 h in the atorvastatin group (50 ± 8 vs. 59 ± 13 ng/ml; p = 0.002) and became even more significant at 24 h (57 ± 9 vs. 73 ± 18 ng/ml; p = 0.0008). Vascular cell adhesion molecule-1 levels were not different at any time point in the 2 arms.
Conclusions In patients undergoing PCI, reduction of procedural myocardial injury after 7-day pre-treatment with atorvastatin is paralleled by concomitant attenuation of post-procedural increase of ICAM-1 and E-selectin levels; thus, reduction of endothelial inflammatory response may explain this protective effect of statins.
Optimization of periprocedural pharmacologic therapy has a crucial role in patients undergoing percutaneous coronary intervention, influencing occurrence of early cardiac events (1,2). In the randomized ARMYDA (Atorvastatin for Reduction of MYocardial Damage during Angioplasty) trial (1), pre-treatment with atorvastatin 40 mg/day initiated 7 days before percutaneous intervention in patients with stable angina was associated with 80% risk reduction of periprocedural myocardial infarction. The mechanisms underlying this protective action are unknown, but are possibly due to the pleiotropic effects of statins; this class of drugs may, in fact, have anti-inflammatory effects by influencing cytokines and growth-factors release (3), may decrease apoptosis (4), improve endothelial function (5), reduce adhesion molecules expression and adhesiveness of leucocytes to vascular endothelium (6,7), and have direct protective effects on myocardial cells (8); these actions may be independent of low-density lipoprotein reductions.
A planned subgroup analysis was prospectively performed among patients enrolled in the ARMYDA trial to investigate whether pre-treatment with atorvastatin reduces procedural myocardial damage by mechanisms linked to reduction of inflammatory endothelial activation (ARMYDA-CAMs [Atorvastatin for Reduction of MYocardial Damage during Angioplasty-Cell Adhesion Molecules] substudy); in particular, the study was designed to evaluate whether variations in the post-procedural levels of soluble adhesion molecules (vascular cell adhesion molecule-1 [VCAM-1], intercellular cell adhesion molecule-1 [ICAM-1], and E-selectin) are influenced by pre-treatment with atorvastatin in patients undergoing elective percutaneous coronary intervention.
Patient population and study design
The main ARMYDA study was a randomized, multicenter, prospective, double-blind, placebo-controlled trial (1). Patients with stable angina, significant coronary artery disease, and an indication to elective coronary angioplasty were enrolled. The study included 153 patients from 2 institutions randomized to atorvastatin (40 mg/day, n = 76) or placebo (n = 77) starting 7 days before planned percutaneous revascularization. The ARMYDA-CAMs was designed as a prospective substudy of the main trial to be performed in all patients enrolled at a single institution (Campus Bio-Medico University of Rome). Inclusion and exclusion criteria were the same as the original ARMYDA trial (1), although patients were required to give an additional specific informed consent for participation to the ARMYDA-CAMs. Thus, the study included all 76 consecutive patients enrolled at Campus Bio-Medico University: 38 were randomized to atorvastatin and 38 to placebo. Vascular cell adhesion molecule-1, ICAM-1, and E-selectin plasma levels were blindly measured at randomization (i.e., 7 days before intervention), immediately before the procedure, and after 8 and 24 h (Fig. 1).According to the design of the ARMYDA trial, blood samples were also drawn before intervention and at 8 and 24 h after the procedure to assay creatine kinase-MB (mass), troponin-I (mass), myoglobin, and C-reactive protein (CRP) levels.
All interventions were performed via the femoral approach with standard technique, as previously described (1). A weight-adjusted dose of heparin (100 IU/kg) was given in the catheterization laboratory before the procedure. Blood samples for adhesion molecule determinations were collected in EDTA-K3 syringes and immediately centrifuged at 10°C (5 min, 3,000 × g), and the plasma was stored at −70°C until analysis. Soluble adhesion molecules VCAM-1, ICAM-1, and E-selectin were assessed by blind investigators using a double primary antibody sandwich enzyme-linked immunoabsorbant assay (R&D Systems, Abingdon, United Kingdom). The test is based on simultaneous reaction of the adhesion molecule to 2 monoclonal antibodies directed against different epitopes on the adhesion molecule. All assays are standardized against purified forms of recombinant VCAM, ICAM, or E-selectin. In our laboratory, sensitivity (minimal detectable dose) for VCAM-1 is lower than 2.0 ng/ml, for ICAM-1 it is lower than 0.35 ng/ml, and for E-selectin it is lower than 1.0 ng/ml. There was no cross-reactivity with other adhesion molecules. Dilution curves of the serum samples were parallel to standard dilution curves. The inter-assay and intra-assay coefficients of variation were <8%, as determined in human serum. Measurements of creatine kinase-MB, troponin-I, and myoglobin levels were performed using the Access 2 Immunochemiluminometric assay (Beckman Coulter, Fullerton, California) (9). Upper normal limits were defined as the 99th percentile of normal population with a total imprecision of <10%, according to Joint European Society of Cardiology/American College of Cardiology guidelines (10). Normal limits were ≤4 ng/ml for creatine-kinase MB, ≤0.08 ng/ml for troponin-I, and ≤80 ng/ml for myoglobin. High-sensitivity CRP determination was obtained by the KRIPTOR-ultrasensitive immunofluorescent assay (BRAHMS, Hennigsdorf/Berlin, Germany), with a detection limit of 0.06 mg/l.
The primary end point of the ARMYDA-CAMs substudy was to compare changes of adhesion molecule levels after coronary interventions according to pre-treatment with atorvastatin or placebo. The secondary end point was the correlation of such levels to the occurrence of procedural myocardial damage.
The study was approved by the institutional review board of our institution and was not supported by any external source of funding.
Values are expressed as mean ± SD, unless otherwise specified. Non-parametric Friedman analysis of variance for repeated measures followed by pairwise comparisons (Wilcoxon signed rank test with Bonferroni’s adjustment for the 3 comparisons of adhesion molecule levels at each time point) was applied to detect changes in VCAM-1, ICAM-1, and E-selectin levels over time within the same group (atorvastatin or placebo). The Mann-Whitney Utest was used to compare continuous data between the 2 arms; proportions were compared by Fisher exact test when the expected frequency was <5; otherwise, the chi-square test (Yates’ corrected) was applied. Correlations between variables were assessed by Spearman’s rank test. All calculations were performed by SPSS 12.0 (SPSS Inc., Chicago, Illinois), and p values <0.05 (2-tailed) were considered significant.
Clinical and procedural data of the ARMYDA-CAMs population are indicated in Tables 1 and 2,⇓⇓respectively. The 2 randomization arms were similar for age, gender, cardiovascular risk factors, clinical presentation (all patients had stable angina by protocol design), left ventricular function, serum creatinine levels, and medical therapy at the time of intervention. Coronary anatomy, procedural characteristics, use of drug-eluting stents, diameter and length of implanted stents, and use of glycoprotein IIb/IIIa inhibitors were also similar. Procedural success was obtained in all 76 patients; no patient had no-reflow phenomenon or significant (≥2 mm) side branch closure during the intervention. There were no in-hospital major complications (death or need for urgent revascularization).
Primary end point
Intercellular cell adhesion molecule-1, E-selectin, and VCAM-1 levels drawn at the time of randomization were similar in both arms (Fig. 2).
Intercellular cell adhesion molecule-1 levels did not change during the 7 days of pre-treatment in either arm (atorvastatin 216 ± 40 ng/ml, placebo 204 ± 38 ng/ml, p = 0.07). At 8 h post-procedure, the levels rose significantly from baseline, in a similar fashion in both groups (atorvastatin 243 ± 46 ng/ml, placebo 239 ± 43 ng/ml; +12 ± 9% vs. +17 ± 14%; p ≥ 0.1 for groups comparison) (Figs. 2 and 3);⇓at 24 h this increase was significantly lower in the atorvastatin versus placebo arm (282 ± 56 vs. 325 ± 70 ng/ml, p = 0.007; +30 ± 6% vs. +59 ± 9%, p = 0.0001).
As illustrated in Figure 2, E-selectin levels before intervention were also similar in the 2 arms (atorvastatin 40 ± 9 ng/ml, placebo 41 ± 10 ng/ml, p = 0.72) and increased significantly after the procedure. Unlike ICAM-1 levels, a significant attenuation of E-selectin rise was already observed at 8 h in the atorvastatin group versus placebo (50 ± 8 vs. 59 ± 13 ng/ml, p = 0.002; +25 ± 5% vs. +44 ± 8% from baseline, p = 0.0001); this became even more significant at 24 h (57 ± 9 vs. 73 ± 18 ng/ml, p = 0.0008; +42 ± 9% vs. +78 ± 18%, p = 0.0001) (Figs. 2 and 3).
Vascular cell adhesion molecule-1 levels increased significantly from baseline at 8 and 24 h, but no differences at any time point were observed between the 2 arms (Figs. 2 and 3).
Low-density lipoprotein levels after the 1-week pre-treatment were not different in the 2 groups (atorvastatin 120 ± 28 mg/dl; placebo 125 ± 30 mg/dl, p = 0.39); no correlation was found between lipid levels and circulating levels of adhesion molecules (p > 0.40).
Procedural myocardial injury
In agreement with the results obtained in the global ARMYDA population, in this subset of patients myocardial infarction (defined as post-procedural creatine kinase-MB increase >2× above the upper normal limit) occurred in 5% of patients in the atorvastatin versus 21% of those in the placebo group (p = 0.08); any elevation of creatine kinase-MB above normal limits was detected in 11% of patients in the atorvastatin arm (vs. 34% in the placebo arm; p = 0.028); similarly, the proportions of patients with troponin-I (18% vs. 45%; p = 0.026) and myoglobin increases (21% vs. 47%; p = 0.030) were significantly lower.
According to the secondary end point, post-procedural adhesion molecule peak levels were also correlated to occurrence of procedural myocardial damage, defined as troponin-I elevation above normal limits after intervention. As illustrated in Figure 4,patients with E-selectin levels below the median (67 ng/ml) had a significantly lower incidence of myocardial damage in the atorvastatin versus the placebo arm (3% vs. 21%, p = 0.03). No significant increase in E-selectin levels was observed in patients of either arm with procedural myocardial damage (p ≥ 0.33). Similar findings were obtained with ICAM-1 determinations (data not shown).
C-reactive protein levels at the time of the procedure were not different (2.4 ± 2.4 mg/l in the treatment vs. 6.3 ± 13.8 mg/l in the placebo group, p = 0.54). There was a non-significant trend toward attenuation of 8-h post-procedural increase of CRP values in patients randomized to atorvastatin (3.1 ± 2.5 vs. 7.4 ± 13.9 mg/l, p = 0.10), whereas 24-h CRP levels were not significantly different between the 2 groups (3.7 ± 2.7 vs. 7.5 ± 14.1 mg/l, p = 0.78).
The ARMYDA-CAMs substudy shows that 7 days of pre-treatment with atorvastatin of patients with stable angina significantly attenuates the increase in levels of adhesion molecules after coronary intervention; these findings suggest that this effect may contribute to the procedural myocardial protection provided by atorvastatin, first described in the ARMYDA trial (1).
Adhesion molecules are expressed on the membrane of activated cells, representing specific surface receptors for interaction between leukocytes and endothelial cells (11–13). Soluble circulating forms of adhesion molecules can be measured from peripheral blood samples, and their levels reflect the amount of membrane-bound adhesion molecules and the degree of local endothelial activation (14). Previous studies on small numbers of patients have shown that coronary angioplasty is followed by a transient increase in adhesion molecule levels (15) due to local endothelial activation/damage (16). Our results confirm and expand these findings by showing, in a larger number of patients, that percutaneous intervention is invariably followed by an increase in adhesion molecule levels, in agreement with the activation kinetics of these molecules (13) (although the absence of measurements beyond the 24 h does not allow evaluation of their decrease pattern).
Anti-inflammatory actions of statins are recognized, but studies on the effects of these drugs on adhesion molecule concentrations in subjects with hypercholesterolemia yielded controversial results (17–19); to date, no previous study has assessed the effect of short-term treatment with statins on adhesion molecule levels in patients undergoing percutaneous coronary intervention. In keeping with previous experimental data showing that statins do not influence baseline expression of adhesion molecules in vitro (20), in our study atorvastatin 40 mg/day for 1 week did not modify baseline levels of soluble adhesion molecules. However, statins were demonstrated to decrease CAM expression after different inflammatory stimuli (including tumor necrosis factor-alpha, lipopolysaccharide, interferon-gamma or CRP) (6,21–23). Recent data from our institution (24) showed that pre-treatment with simvastatin significantly decreases post-operative ICAM-1 and E-selectin elevations after coronary bypass surgery; likewise, the ARMYDA-CAMs study now suggests a drug-mediated attenuation of the adhesion molecules increase after the inflammatory stimulus represented by percutaneous coronary intervention.
The protective effects of atorvastatin on myocardial damage during coronary intervention observed in the main randomized ARMDYDA trial (1) has been confirmed in the subpopulation of the present study. The mechanism(s) underlying this clinical benefit are not clear; the ARMYDA-CAMs substudy shows a significant decrease of post-intervention levels especially of E-selectin, expressed by endothelial cells; thus, attenuation of endothelial activation may, at least in part, explain the protective role of atorvastatin. In fact, patients of the atorvastatin arm with lower post-procedural CAM levels had a significantly less incidence of myocardial damage, suggesting that reduction of adhesion molecule release may help prevent myocardial damage by limiting local recruitment of inflammatory cells, by improving small vessel function (25), and, possibly, by stabilizing coronary plaque and reducing microembolization. Another explanation of the parallelism between endothelial activation and myocardial injury is that both these findings could be an expression of a common mechanism of protection induced (or potentiated) by statin therapy (i.e., increased nitric oxide bioavailability). It may be also argued that the attenuation in the increase in CAMs may be due to the lower micro-infarct rate (i.e., inflammation due to micro-infarcts causes an elevation in CAMs, and a lower infarct rate may lead to lower CAMs). However, no significant increase of adhesion molecule levels was observed in patients of either arm with procedural myocardial damage, and atorvastatin significantly attenuated CAM levels irrespective of the occurrence of myocardial damage. These results would support the hypothesis that the beneficial effect of atorvastatin is indeed due to attenuation in the increase of adhesion molecules, and that procedural myocardial necrosis was not a significant effect modifier that drives the results.
In our study, atorvastatin did not affect post-procedural VCAM-1 levels; this was also observed in a previous study of different design, not involving coronary intervention, after prolonged therapy with statins (19). Indeed, although common regulatory mechanisms have been recognized, different factors may be involved in the regulation of adhesion molecules on different cell types (26). Moreover, pre-treatment with atorvastatin was not associated with significant attenuation of CRP levels before and after intervention; in fact, effects of statins on adhesion molecules may be independent from CRP (27) or may precede systemic cytokine cascade activation (5).
In conclusion, the ARMYDA-CAMs substudy shows that pre-treatment with atorvastatin reduces myocardial damage and adhesion molecule levels after coronary angioplasty. These findings may help clarify the mechanisms underlying myocardial protection provided by statin therapy before percutaneous coronary interventions, as well as the biological basis of the pleiotropic effects of statins.
- Abbreviations and Acronyms
- Atorvastatin for Reduction of MYocardial Damage during Angioplasty
- Atorvastatin for Reduction of MYocardial Damage during Angioplasty-Cell Adhesion Molecules substudy
- C-reactive protein
- intercellular cell adhesion molecule-1
- vascular cell adhesion molecule-1
- Received May 10, 2006.
- Revision received June 19, 2006.
- Accepted June 26, 2006.
- American College of Cardiology Foundation
- Pasceri V.,
- Patti G.,
- Nusca A.,
- Pristipino C.,
- Richichi G.,
- Di Sciascio G.,
- ARMYDA Investigators
- Patti G.,
- Colonna G.,
- Pasceri V.,
- Lassandro Pepe L.,
- Montinaro A.,
- Di Sciascio G.
- Paajarvi G.,
- Roudier E.,
- Crisby M.,
- Hogberg J.,
- Stenius U.
- Weber C.,
- Erl W.,
- Weber K.S.,
- Weber P.C.
- Di Napoli P.,
- Taccardi A.,
- Grilli A.,
- et al.
- Uettwiller-Geiger D.,
- Wu A.H.,
- Apple F.S.,
- et al.
- Alpert J.S.,
- Thygesen K.,
- Antman E.,
- Bassand J.P.
- Jang Y.,
- Lincoff A.M.,
- Plow E.F.,
- Topol E.J.
- Ferns G.A.,
- Forster L.A.,
- Williams J.C.,
- et al.
- Wagner A.H.,
- Gebauer M.,
- Guldenzoph B.,
- Hecker M.
- Pasceri V.,
- Cheng J.S.,
- Willerson J.T.,
- Yeh E.T.