Author + information
- Received September 19, 1996
- Revision received January 31, 1997
- Accepted February 13, 1997
- Published online May 1, 1997.
- Carlos V. Serrano Jr., MDA,* (, )
- José Antônio F. Ramires, MD, FACCPA,
- Margareth Venturinelli, BScB,
- Siguemituso Arie, MDA,
- Elbio D’Amico, MDB,
- Jay L. Zweier, MD, FACCC,
- Fulvio Pileggi, MD, FACCA and
- Protasio L. da Luz, MD, FACCA
- ↵*Dr. Carlos V. Serrano, Jr., Heart Institute, School of Medicine, University of São Paulo, Clinical Division, Avenida Enéas de C. Aguiar, 44, São Paulo, SP 05403-000, Brazil.
Objectives. This study sought to characterize leukocyte and platelet activation and adhesion molecule expression after coronary angioplasty.
Background. Coronary angioplasty can be regarded as a clinical model of postischemic inflammation because this intervention leads to the release of inflammatory mediators as a result of plaque rupture and endothelial injury.
Methods. In 13 patients with stable angina (mean [±SEM] age 56.0 ± 2.4 years, range 44 to 79), blood samples were drawn from the aorta and coronary sinus immediately before and immediately and 15 min after coronary angioplasty. Subsequently, leukocyte and platelet functions were determined. Eleven control patients (57.5 ± 2.3 years, range 52 to 78) underwent coronary arteriography.
Results. Coronary arteriography and angioplasty showed no difference in number of leukocytes between the coronary sinus and the aorta. However, 15 min after coronary angioplasty, there was an increase in neutrophil CD18 and CD11b, monocyte CD14 and platelet glycoprotein IIb/IIIa expression and a decrease in neutrophil L-selectin expression (189 ± 25%, 163 ± 27%, 158 ± 35%, 141 ± 22% and 31 ± 10%, respectively, p < 0.01). In the control subjects, no change in adhesion molecule expression occurred. Superoxide production and aggregation in ex vivo-stimulated neutrophils collected from the coronary sinus 15 min after coronary angioplasty was significantly decreased compared with that after coronary arteriography (54 ± 12% vs. 106 ± 30% and 58 ± 11% vs. 102 ± 29%, respectively, p < 0.01). The reduced responses to phorbol ester stimulation may be explained by previous in vivo activation of neutrophils during coronary angioplasty.
Conclusions. Coronary angioplasty increases neutrophil, monocyte and platelet adhesion molecule expression and induces a significant decrease in ex vivo-stimulated neutrophil superoxide generation and aggregation. These findings suggest that coronary angioplasty triggers cellular activation with an inflammatory response that could contribute to restenosis.
(J Am Coll Cardiol 1997;29:1276–83)
Although percutaneous transluminal coronary angioplasty (PTCA) is widely accepted for the treatment of patients with critical obstructive coronary artery disease, its pathophysiologic implications are not fully understood. For instance, coronary angioplasty can be regarded as a clinical model of myocardial ischemia/reperfusion because it is characterized by successive short periods of myocardial ischemia ([1–3]). Therefore, postischemic inflammation may be induced. Another peculiar aspect of this intervention is that it leads to the release of inflammatory mediators and chemoattractant factors because of plaque rupture, arterial wall damage and endothelial injury ([4, 5]). The recruitment of activated leukocytes, such as neutrophils and monocytes, and activated platelets to the inflammatory site is related to the regulation of adhesion molecules expressed on the cellular surface. Optimal neutrophil–endothelial cell interactions involve at least two classes of adhesion molecules that are expressed on the surface of neutrophils. These include L-selectin, which is constitutively expressed on nonactivated leukocytes, and the beta2-integrins, which are upregulated when neutrophils are stimulated. L-selectin appears to be responsible for the initial attachment and rolling of stimulated neutrophils on the endothelium. Firm adhesion, neutrophil aggregation and transendothelial migration are mediated by beta2-integrins. These consist of a family of heterodimers, each of which has a common beta-subunit, designated CD18, and one of three known alpha-subunits, designated CD11a (LFA-1), CD11b (Mac-1) and CD11c (p150,95) ([6–8]). In monocytes, besides the CD11/CD18 complex, the specific adhesion molecule CD14 is also expressed (). In contrast, adherent platelets contain the integrin glycoprotein (GP) IIb/IIIa receptor on the surface, the final common pathway of platelet aggregation ().
In addition, activated leukocytes release a variety of mediators, for example superoxide radicals, that are potentially deleterious to the heart and are capable of inducing platelet activation and aggregation, vasoconstriction and direct inflammatory tissue damage. Neutrophils may thus exert direct cytotoxic effects on coronary vascular and myocardial cells ([11, 12]).
However, it is still not known whether coronary angioplasty results in leukocyte and platelet activation and adhesion molecule expression, which could contribute to restenosis. Therefore, this study sought to characterize cellular activation and the surface expression of adhesion molecules of coronary circulating leukocytes and platelets after coronary angioplasty.
1.1 Study patients.
Thirteen patients with stable coronary artery disease underwent elective coronary angioplasty; their clinical profile is shown in Table 1. All patients had a history of effort angina pectoris, with no evidence of a previous myocardial infarction, and all had viable myocardium at risk as defined by the development of typical chest pain with ST segment depression ≥1 mm during an exercise stress test or reversible perfusion defect on exercise thallium-201 or MIBI scintigraphy. Approval was obtained from our institutional review board, and written consent from each patient.
To ensure that neither the technical maneuvers involved in coronary angioplasty (e.g., vessel puncture and catheter manipulation) nor the contrast medium used resulted in changes in cell function, 11 control patients underwent routine coronary arteriography, without coronary angioplasty. For this purpose, seven men and four women years were studied (mean [±SEM] age 57.5 ± 2.3 years); their clinical profile is shown in Table 2.
1.2 Study protocol.
All medications, including aspirin, heparin, nitrates, calcium channel blocking agents and beta-adrenergic blocking drugs, were not discontinued for the study; however, fasting of at least 12 h was required before the procedure. After premedication with prophylactic antibiotics and local anesthesia, a 7F catheter was advanced either from the brachial or the femoral vein to the coronary sinus. A sheath was placed into either the brachial or the femoral artery for insertion of coronary angiographic and angioplastic catheters. Before intracoronary manipulations, an intravenous bolus of 15,000 IU of heparin was administered. Blood samples were obtained at the beginning of the procedure from the coronary sinus and from the guiding catheter in the aortic root. Coronary angioplasty was performed with balloon dilation catheters of 2.0- to 3.5-mm diameter; the balloon size chosen for each patient was based on the diameter of normal segments adjacent to the stenosis. The balloon was inflated for 45 to 90 s at a pressure of 4 to 10 atm. Coronary angioplasty was regarded as successfulwhen the stenotic segment was dilated so that lumen narrowing was <50%. Coronary artery diameter was measured with calipers at the site of maximal obstruction and at the proximal and distal portions of the obstruction, and percent lumen narrowing was calculated. Blood samples were drawn from the aortic root and from the coronary sinus immediately before as well as immediately and 15 min after the coronary angioplasty procedure. Subsequently, the blood samples were processed immediately for cellular adhesion molecule expression and ex vivo-stimulated neutrophil superoxide generation and aggregation, as indicated later. In addition, total leukocyte count was determined in each blood sample.
The control patients also received an intravenous injection of 5,000 IU of heparin and isosorbide dinitrate during the procedure. Coronary arteriography was performed by the Sones technique. A 7F catheter was advanced either from the brachial or the femoral vein to the coronary sinus. Blood samples were obtained from the aortic root and from the coronary sinus immediately before as well as immediately and 15 min after the coronary arteriography procedure. All cardiovascular medications were discontinued on the day of the procedure.
1.3 Immunofluorescence and flow cytometric assays.
Cellular surface expression of the adhesion molecules CD11b, CD18, L-selectin (neutrophil), CD14 (monocyte) and GP IIb/IIIa (platelet) were measured by direct immunofluorescence techniques associated with flow cytometry. According to a previously described technique (), adhesion molecule labeling was performed for 30 min at 4°C, in the dark, using aliquots of whole (unseparated) blood and saturating concentrations of respective fluorescent, isotype monoclonal antibodies.
This technique avoided in vitro manipulation capable of activating cells. The samples were washed with phosphate-buffered saline (PBS), and the erythrocytes were hemolyzed and leukocytes and platelets fixed with commercially available solutions (). The samples, once centrifuged and dissolved in PBS, were used for flow cytometric analysis. Samples were run through a FACStarPLUSflow cytometer (Becton-Dickinson Immunocytometry Systems) to measure the fluorescent signal. Gating of the forward and side scatter diagrams was used to circumscribe the analysis to a homogeneous cell population, excluding clusters and debris. The laser excitation wavelength was 488 nm, and the emission light was 578 nm. A frequency distribution histogram of the fluorescent signal was obtained from at least 5,000 cells for each sample with a Hewlett-Packard computer programmed with LYSYS II Profile Analyzer software.
Negative control for nonspecific flow cytometric background was performed with carefully isotype-matched, fluorescent, nonbinding monoclonal antibodies.
Neutrophils could be distinguished from lymphocytes and monocytes by the combination of low angle forward and right-angle scattered laser light, and a cytogram of each cell population could be generated. The region of interest in the neutrophil, monocyte or platelet population was identified by an observer using a computer-aided outlining technique. A histogram of fluorescence distribution was constructed with the fluorescence intensity on the abscissa and the number of cells on the ordinate. The relative mean intensity of fluorescence was obtained from the histogram and expressed as an index of membrane surface expression. Results are expressed as either arbitrary units or as percentage of the label-matched coronary sinus/aortic ratio, to account for variability of basal values among different periods.
1.4 Neutrophil preparation and purification.
Isolation of neutrophils was executed by Ficoll-Hypaque centrifugation, dextran sedimentation of red cells and removal of remaining red cells by hypotonic lysis (). This procedure routinely produced a neutrophil preparation >98% purity and viability. Determination of neutrophil superoxide production and aggregation was carried out after ex vivo 30 min stimulation with phorbol myristate acetate (PMA) (final concentration 100 ng/ml).
1.5 Determination of neutrophil superoxide production.
Rates of superoxide production were determined by measuring the reduction of cytochrome c(reduced–oxidized 21.100 mol/liter per cm) in the presence of 30 μg/ml of superoxide dismutase. Rates were measured by continuously monitoring A550with 50 μmol/liter cytochrome cand 106cells/ml, or the total amount of superoxide produced during the course of the reaction was determined from the overall ΔA550measured with 150 μmol/liter cytochrome cand 0.25 × 106or 0.5 × 106cells/ml (Beckman DU-70 Spectrophotometer) ().
1.6 Measurement of neutrophil aggregation.
Neutrophil suspensions (20,000 cells/mm3) were suspended in PBS containing 0.1% bovine serum albumin (BSA), 1 mmol/liter calcium chloride and 0.7 mmol/liter magnesium chloride. Once stimulation was performed with phorbol ester, aggregation was initiated in a Chrono-Log 440 Dual Channel aggregometer (Chrono-Log Corp.) with continuous stirring at 1,000 rpm. The aggregation measures were done by the turbidometric method described by Born and Cross (). Changes in optical density were recorded on a Chrono-Log recorder model 707. Results are expressed in amplitude (cm/min).
1.7 Chemicals and reagents.
The chemicals and reagents used in this study were purchased from the sources indicated: whole-blood lysing reagent kit (Coulter Corp.); monoclonal antibodies anti-CD11b, anti-CD18, anti-CD14, anti-GP IIb/IIIa, immunoglobulin G R-phycoerythrin-conjugated secondary antibody and negative control nonbinding monoclonal antibodies (Dako Corp.); monoclonal antibody anti–L-selectin (Becton-Dickinson Immunocytometry Systems); and, PMA, Ficoll, Histopaque 1083, cytochrome c, dextran, BSA, calcium chloride and magnesium chloride (Sigma Chemical Co.).
1.8 Statistical analysis.
Results are presented as mean value ± SEM. The variables from each group (control and PTCA groups) were analyzed separately by the Student ttest for independent samples. Statistical analysis of the time-course data was performed by two-factor analysis of variance for repeated measures. When analysis of variance revealed a statistical significance, the Newman-Keuls test for matched samples was used. A p value <0.05 was conventionally assumed to indicate statistical significance.
2.1 Coronary angioplasty and control groups.
The procedure was successful in all patients, and no significant complications occurred. The degree of coronary stenosis was significantly reduced, from 82.4 ± 1.5% to 35.8 ± 1.7% (p < 0.001). Balloon inflation time was 284.6 ± 11.1 s (range 240 to 360). During balloon inflation, 10 patients had ST segment transitory changes, 5 of whom complained of chest pain. However, no serious complications leading to progression or recurrence of myocardial infarction occurred. The average amount of contrast medium used during coronary angioplasty procedure was 320 ml.
There were no substantial differences between the patients in the coronary angioplasty group and the control subjects in their baseline clinical characteristics (Table 1and Table 2). The control group included 11 patients with coronary artery disease: 2 had two-vessel disease; 6 had one-vessel disease; 2 had angiographically normal coronary arteries; and 1 had mild aortic stenosis associated with mild coronary artery stenosis <50%. The average amount of contrast medium used during coronary arteriography was 180 ml, which was not significantly different from that used during the coronary angioplasty.
2.2 Leukocyte count.
No differences were observed in total leukocyte count between the aortic root and the coronary sinus in patients undergoing either coronary arteriography or angioplasty. Control patients had a leukocyte count of 5.4 × 106cells/ml before, 6.3 × 106cells/ml immediately after and 5.7 × 106cells/ml 15 min after coronary arteriography. Patients undergoing coronary angioplasty had leukocyte counts of 5.3 × 106, 5.9 × 106and 5.9 × 106cells/ml, respectively.
2.3 Surface expression of leukocyte adhesion molecules.
Fig. 1and Fig. 2show percent changes in mean fluorescence intensity of CD18, CD11b, L-selectin, CD14 and GP IIb/IIIa in patients who underwent coronary arteriography or coronary angioplasty. In the 11 control subjects who underwent routine coronary arteriography, no change in mean fluorescence intensity was observed among cells collected from the coronary sinus compared with those from the aortic root. Before coronary arteriography, control patients had mean fluorescence intensities of aortic and coronary sinus neutrophils for CD18 of 147.9 ± 8.7 and 149.2 ± 8.2; for CD11b of 70.0 ± 5.3 and 70.8 ± 5.4; and for L-selectin of 56.2 ± 2.6 and 55.2 ± 3.0, respectively. Aortic and coronary sinus monocytes and platelets demonstrated mean fluorescence intensities before coronary arteriography for CD14 of 73.2 ± 6.3 and 72.4 ± 6.4 and for GP IIb/IIIa of 94.2 ± 5.3 and 93.8 ± 6.4, respectively.
In contrast, in the 13 patients undergoing routine coronary angioplasty, surface expression of the adhesion molecules of leukocytes and platelets from the coronary sinus was markedly changed 15 min after coronary angioplasty compared with that from the aorta. Patients with coronary angioplasty had adhesion molecule surface expression in aortic and coronary sinus cells preceding the procedure similar to that observed in control patients. Aortic and coronary sinus mean fluorescence intensities were 90.5 ± 12.8 and 92.3 ± 10.9 for neutrophil CD18, 110.7 ± 16.8 and 116.2 ± 21.5 for CD11b, and 76.8 ± 27.7 and 72.2 ± 15.5, for L-selectin, respectively. Aortic and coronary sinus monocytes and platelets demonstrated mean fluorescence intensities before coronary angioplasty of 70.8 ± 12.5 and 72.2 ± 13.5 for CD14 and 100.9 ± 15.6 and 95.9 ± 20.0 for GP IIb/IIIa. After coronary angioplasty, adhesion molecule expression of cells from the coronary sinus was significantly higher than that after coronary arteriography. No correlation with the amount of contrast medium or balloon catheter size used during coronary angioplasty was detected with the adhesion molecule expression of leukocytes and platelets.
Fig. 3shows typical flow cytometric histograms observed from those experiments measuring CD18, CD11b and L-selectin expression of neutrophils collected from the coronary sinus before and 15 min after coronary angioplasty of a 90% stenosis of the left anterior descending coronary artery. Fig. 4shows representative histograms of CD14 and GP IIb/IIIa expression from the coronary sinus of the same patient.
2.4 Measurement of neutrophil superoxide production.
Fig. 5(left panel) shows changes in ex vivo phorbol ester–stimulated neutrophils of control subjects and patients who underwent coronary angioplasty. Superoxide production from neutrophils collected from the coronary sinus did not change compared with cells collected from the aortic root immediately and 15 min after coronary arteriography (from 4.7 ± 0.2 to 4.7 ± 0.3 and from 4.6 ± 0.3 to 4.6 ± 0.3 nmol·O2−/106cells, respectively, p = NS), whereas superoxide production after coronary angioplasty significantly decreased (from 4.2 ± 2.2 nmol·O2−/106cells to 3.1 ± 1.9 and 2.3 ± 1.9 nmol·O2−/106cells immediately and 15 min after angioplasty, respectively, p < 0.01). Coronary sinus neutrophil superoxide production after coronary angioplasty was significantly lower than that after coronary arteriography (coronary angioplasty, p < 0.01).
2.5 Neutrophil aggregation.
Fig. 5(right panel) shows changes in aggregation curves obtained from neutrophils collected from the aortic root and the coronary sinus in the control subjects and in patients who underwent coronary angioplasty. Neutrophil aggregation in the coronary sinus did not change compared with that in the aorta immediately and 15 min after coronary arteriography (from 4.7 ± 0.3 to 4.8 ± 0.3 and from 4.8 ± 0.2 to 4.9 ± 0.2 cm/min, respectively, p = NS), whereas neutrophil aggregation in the coronary sinus decreased significantly after coronary angioplasty (from 4.9 ± 2.9 to 3.2 ± 1.8 and 2.8 ± 1.9 cm/min immediately and 15 min after angioplasty, respectively, p < 0.01).
There was no correlation between the changes in neutrophil superoxide production and aggregation and inflation time, amount of contrast medium or balloon catheter size used during coronary angioplasty.
Although coronary angioplasty is a commonly used treatment for obstructive coronary artery disease, little is known regarding leukocyte and platelet function during this intervention. The present study demonstrates that an acute inflammatory response is provoked by coronary angioplasty. We observed that this intervention induced a significant alteration in membrane surface expression of the adhesion molecules CD18, CD11b, L-selectin (neutrophil), CD14 (monocyte) and GP IIb/IIIa (platelet) and induced a significant decrease in the generation of superoxide in ex vivo phorbol ester–stimulated neutrophils as well as neutrophil aggregation. No such changes in leukocyte and platelet functions were found in the control subjects, who underwent routine coronary arteriography.
3.1 Leukocyte adhesion molecules.
Our study describes sequential changes in the fluorescence intensity of L-selectin, CD18 and CD11b induced by coronary angioplasty. L-selectin adhesion molecule expression of neutrophils collected from the coronary sinus decreased immediately after coronary angioplasty, preceding the marked increase of CD18 and CD11b expression. During the neutrophil adhesion process, a transition from L-selectin–mediated adhesion to CD18-mediated adhesion must take place—L-selectin rapidly sheds from the neutrophil surface before the upregulation of Mac-1 ([18–21]). In agreement with our study, Haught et al. () recently identified abnormalities in circulating intercellular adhesion molecule-1 (ICAM-1) and soluble L-selectin (shed form of L-selectin) levels in patients with coronary artery disease regardless of white blood cell counts, clinical status or severity of disease ().
From our study, it is uncertain whether the changes in leukocyte and platelet adhesion molecules could persist for >15 min in patients undergoing coronary angioplasty. However, other studies have provided evidence for the presence of chronic inflammation in coronary artery disease, suggesting that inflammation is not merely a manifestation of acute myocardial ischemia (). In addition, in an experimental study, local inflammatory activation of endothelial cells and leukocytes lasts up to 30 days after balloon injury in the rabbit aorta ().
3.2 Leukocyte and platelet activation after coronary angioplasty.
Two explanations can be proposed as to why coronary angioplasty activated coronary circulating leukocytes and platelets in our patients:
1. Coronary angioplasty can be viewed as a clinical model of myocardial ischemia/reperfusion because its procedures are characterized by successive short periods of myocardial ischemia ([1–3]). Reperfusion of ischemic myocardium initiates a deleterious cascade of events leading to further myocardial injury ([23, 24]). Leukocytes, mainly neutrophils, are important mediators of this phenomenon. An essential initiating step in the process of neutrophil-mediated injury within the heart involves the activation and adhesion of circulating neutrophils to the coronary endothelium ().
2. Coronary angioplasty causes plaque rupture, arterial wall damage and endothelial injury, which in turn provokes the release of inflammatory and chemoattractant factors ([4, 5]). These factors provide the signal for leukocytes to enter the inflamed tissue ([6, 19]). In addition, platelet adhesion and aggregation may also take place as a result of arterial wall damage as part of the inflammatory response (). This potential interaction between platelets, leukocytes and endothelial cells is a fertile ground for the upregulation of CD11b/CD18. Perhaps most directly, thrombin stimulates the enzymatic pathway leading to the formation of platelet-activating factor (PAF), which is a potent chemotactic agent that in addition, to platelet activation, increases surface expression and activation of CD11b/CD18 and promotes neutrophil adhesion to endothelial cells as well as homotypic aggregation of neutrophils ([10, 27–29]).
3.3 Leukocyte adhesion.
Even though changes in leukocyte adhesion molecule expression occurred after coronary angioplasty, no differences were observed in the leukocyte count between the aortic root and the coronary sinus. Although some infiltration of leukocytes may occur, one presupposes that most of the activated leukocytes are released to move downstream. It is likely that leukocytes escape from adhesion complexes induced in this turbulent circulation because of hemodynamic variability, shifting shear stress conditions, alterations in levels of generated thrombin and the intrinsic instability of the selectin adhesion molecules ([30, 31]). Thus, in a phasic condition such as after coronary angioplasty, it is easy to imagine that the high pressure flow (shear stress) system could ultimately project the leukocytes downstream out of the postobstructive chamber.
3.4 Neutrophil superoxide release and hyperaggregability after coronary angioplasty.
The reduced responses of neutrophil superoxide generation and aggregation to phorbol ester stimulation observed after coronary angioplasty may be explained by previous in vivo activation (priming) of neutrophils. Phorbol ester is considered to induce maximal neutrophil activation (). Accordingly, Dinerman et al. () have demonstrated in patients with acute myocardial ischemia that ongoing in vivo neutrophil activation results in diminished in vitro activity. They suggest that neutrophil stimulation, specifically enhanced release of elastase, after interventions that attempt to restore blood flow to jeopardized myocardium may contribute to the progression and severity of ischemic heart disease.
Our study suggests a pathogenic link between the surface expression of leukocyte adhesion molecules and the generation of superoxide in patients who underwent coronary angioplasty. Superoxide radicals are important mediators of leukocyte–endothelial cell interactions observed after reperfusion of the ischemic heart (). In addition to triggering the upregulation of CD18, there are several other mechanisms by which superoxide radicals might trigger reperfusion-induced leukocyte adherence. Superoxide radicals react with the endothelial cell to produce and release humoral mediators, including leukotriene B4and PAF, which in turn activate and promote the adherence of leukocytes (). Superoxide may also induce the expression of endothelial adhesion molecules, such as P-selectin and ICAM-1 ([35, 36]). Another possible mechanism by which superoxide might further promote neutrophil adhesion is by inactivation of nitric oxide, an endothelial-derived vasodilator that interferes with neutrophil adherence to microvascular endothelium ([37, 38]). Furthermore, reactive oxygen metabolites secreted by activated neutrophils, such as those after coronary angioplasty, can amplify the response to injury and may potentiate platelet activation and vasoconstriction ().
The increased aggregation present in the neutrophils of our patients who underwent coronary angioplasty is consistent with previous reports (). Homotypic aggregation may occur through a CD11/CD18-dependent mechanism and result in leucostatic plugs, leading to microvascular occlusion ([30, 34, 41]). However, because the coronary circulation has an unusual microcirculation, activated leukocytes as well as activated platelets may be trapped not in the coronary circulation but in either the pulmonary circulation or the spleen and thus disappear from the systemic circulation ().
3.5 Clinical implications.
Regardless of the mechanism triggering neutrophil and monocyte activation after coronary angioplasty, the generation of oxygen free radicals along with the release of other inflammatory mediators may aggravate the endothelial damage induced by coronary angioplasty and further stimulate platelets.
Thus, leukocyte and platelet activation after coronary angioplasty in humans may trigger the pathophysiologic chain reaction that eventually results in coronary restenosis. However, further studies are required to evaluate the implications of these findings and to define fully the mechanisms by which the complex interaction of leukocyte–endothelial cell–platelet contributes to restenosis.
We gratefully thank Izabel K. Tutiya, BSc for technical assistance in flow cytometry.
☆ This work was supported in part by a research grant from the Fundação E. J. Zerbini and by Grant SF 0752/94 from the Financiadora de Estudos e Projetos (FINEP).
- bovine serum albumin
- intercellular adhesion molecule-1
- platelet-activating factor
- phosphate-buffered saline
- phorbol myristate acetate
- percutaneous transluminal coronary angioplasty
- Received September 19, 1996.
- Revision received January 31, 1997.
- Accepted February 13, 1997.
- The American College of Cardiology
- Kloner RA,
- Giacomelli F,
- Alker KJ,
- Hale SL,
- Matthews R,
- Bellows S
- De Servi S,
- Mazzone A,
- Ricevuti G,
- et al.
- Sluiter W,
- Pietersma A,
- Lamers JMJ,
- Koster JF
- Berliner JAB,
- Navab M,
- Fogelman AM,
- et al.
- Jang Y,
- Lincoff M,
- Plow EF,
- Topol EJ
- Haziot A,
- Chen S,
- Ferrero E,
- Low MG,
- Silber R,
- Goyert SM
- Dinerman JL,
- Mehta JL
- Harlan JM
- Kuijpers TW,
- Tool ATJ,
- van der Schoot,
- et al.
- Kishimoto TK,
- Jutila MA,
- Berg EL,
- Butcher EC
- Tanaka H,
- Sukhova GK,
- Swanson SJ,
- et al.
- Hansen PR
- Forman MB,
- Virmani R,
- Puett DW
- Lefer DJ,
- Shandelya SML,
- Serrano CV Jr.,
- Becker LC,
- Kuppusamy P,
- Zweier JL
- Merhi Y,
- L-Lacoste L,
- Lam JYT
- Kaikita K,
- Ogawa H,
- Yasue H,
- et al.
- Kereiakes DJ,
- Kleiman NS,
- Ambrose J,
- et al.
- Entman ML,
- Ballantyne CM
- Ikeda H,
- Koga Y,
- Kuwano K,
- et al.
- Dinerman JL,
- Mehta JL,
- Saldeen TGP,
- et al.
- Suzuki M,
- Inauen W,
- Kvietys PR,
- et al.
- Kubes P,
- Suzuki M,
- Granger DN
- Laurindo FRM,
- Pedro MA,
- Barbeiro HV,
- et al.
- Mehta J,
- Dinnerman J,
- Mehta P,
- et al.
- Ricevutti G,
- De Servi S,
- Mazzone A,
- Angoli L,
- Ghio S,
- Specchia G
- Anderson DC,
- Miller LJ,
- Schmalstieg FC,
- Rothlein R,
- Springer TA