Author + information
- Received February 9, 2004
- Revision received April 22, 2004
- Accepted May 2, 2004
- Published online August 18, 2004.
- Alexander Hansen, MD†,* (, )
- Anjali Kumar, PhD*,
- David Wolf, BS†,
- Kamilla Frankenbergerova, BS†,
- Arthur Filusch, MD†,
- Marie-Luise Gross, MD‡,
- Sebastian Mueller, MD†,
- Hugo Katus, MD† and
- Helmut Kuecherer, MD†
- ↵*Reprint requests and correspondence:
Dr. Alexander Hansen, Department of Cardiology, University of Heidelberg, Bergheimerstr. 58, 69115 Heidelberg, Germany
Objectives The purpose was to examine the cardioprotective effects of recombinant P-selectin glycoprotein ligand-immunoglobulin (rPSGL-Ig) in ischemia-reperfusion injury by real-time myocardial contrast echocardiography (MCE).
Background P-selectin mediates leukocyte recruitment into areas of inflammation.
Methods Sixteen pigs underwent 45 min of left anterior descending coronary artery occlusion followed by reperfusion and received rPSGL-Ig or vehicle. To assess changes in myocardial perfusion (A × β), MCE was performed.
Results After 120 min of reperfusion, A × β in the risk area was higher (0.84 ± 0.15 dB/s vs. 0.28 ± 0.1 dB/s, p < 0.0001), and the infarct size was lower (30.3 ± 12% vs. 57 ± 14%, p = 0.002) in the rPSGL-Ig group compared with the vehicle group.
Conclusions Recombinant PSGL-Ig improved postischemic reflow accurately detected by real-time MCE.
Microcirculatory reflow after myocardial infarction is intimately related to functional recovery (1). Leukocytes can induce microvascular damage in the reperfused myocardium, known as “ischemia-reperfusion injury” (2,3). P-selectin glycoprotein ligand-1 (PSGL-1) is the physiologic ligand for endothelial P-selectin expressed on leukocytes, as well as a key molecule responsible for recruitment of neutrophils into areas of inflammation (4,5). Recombinant PSGL-Ig, a recombinant immunoglobulin form of PSGL-1, acts as a competitive inhibitor for P-selectin (6,7). Therefore, immunoneutralization of P-selectin should reduce the extent of leukocyte-mediated tissue damage. The purpose of this study was to investigate whether real-time myocardial contrast echocardiography (MCE) can be used to evaluate microvascular reflow and the cardioprotective effects of rPSGL-Ig.
Sixteen farm pigs (20 to 25 kg) received 2,500 IU heparin and 500 mg acetylsalicylic acid and underwent left anterior descending coronary artery (LAD) occlusion below the first branch of the LAD for 45 min, followed by 120-min reperfusion. A dose of 1 mg/kg rPSGL-Ig (6) or vehicle was given during ischemia.
Myocardial contrast echocardiography
Real-time MCE was performed serially with an HDI5000 (ATL, Bothell, Washington) during intravenous infusion of SonoVue (60 ml/h; Bracco, Milan, Italy) at a low mechanical index of 0.09 and analyzed using HDI software (ATL). The risk area was identified as the area of absent myocardial opacification during LAD occlusion and determined by planimetry in end-diastolic frames 10 cardiac cycles after flash. Perfusion defects during reperfusion were expressed as the percentage of risk area. The increase of signal intensities in the risk area and the posterolateral wall (control area) in end-systolic frames was fitted to an exponential function: y = Ax (1 − e−βt), as described (8). Short-axis fractional area shortening (FAS) was measured by planimetry (EchoCom, Fulda, Germany).
Histologic examinations and myocardial blood flow (MBF)
The LAD was reoccluded at the end of the experiment, and 1 mg/kg monastral blue dye (Sigma, St. Louis, Missouri) was injected into the left atrium to outline the area at risk. The heart was cut into 1-cm-thick slices. From the myocardial slice corresponding to the MCE imaging plane transmural myocardial tissue, samples were excised from the central portion of the risk area, the ischemic border zone (unstained by blue dye), and the control area, divided, and processed either for analysis of capillary density, as described (9), or measurements of MBF. The MBF was measured with 15-μm fluorescent microspheres (Molecular Probes, Eugene, Oregon) with five different wavelengths injected into the left atrium at the same time points as MCE imaging (except 30 min reperfusion), as described previously (10).
Myocardial slices were immersed in 1.3% triphenyltetrazolium chloride (Sigma), which stains viable myocardium red. The infarct and risk areas in each slice were determined by planimetry. The risk area by histology was expressed as the percentage of the total left ventricle (LV) (equal to sum of all slices), and the infarct size was expressed as the percentage of risk area.
Data are expressed as the mean value ± SD, and statistical significance was set at p < 0.05. Correlations between continuous variables were performed by linear regression analysis. The Student ttest, paired or unpaired, as appropriate, was used for group-to-group comparisons.
Hemodynamics and histologic analysis
Mean arterial pressure and FAS decreased during reperfusion but were higher in the rPSGL-Ig group after 120-min reperfusion (Table 1).Left ventricular weight was similar between the rPSGL-Ig and vehicle groups (66 ± 16 g vs. 61 ± 13 g, p = NS), as well as the risk area (% total LV) by monastral blue staining (19.4 ± 4% vs. 22.3 ± 6%, p = NS). The infarct size to risk area by TTC staining was 57 ± 14% and decreased to 30.3 ± 12% (p = 0.002) in the rPSGL-Ig group. Microscopic analysis demonstrated extensive myocardial injury in the risk area, but the numerical capillary density was higher in the rPSGL-Ig group in the central portion of the risk area (1,395 ± 150 vessels/mm2vs. 1,170 ± 155 vessels/mm2, p = 0.03) and the ischemic border zone (2,190 ± 2,650 vessels/mm2vs. 1,435 ± 10 vessels/mm2, p = 0.01), but similar in the nonischemic area (3,270 ± 6,100 vessels/mm2vs. 3,590 ± 830 vessels/mm2, p = NS).
Determination of reflow by real-time MCE
Real-time MCE allowed direct visualization of the risk area during LAD occlusion and postischemic reflow (Fig. 1).The risk area by MCE was similar in both groups (42 ± 6% vs. 38 ± 4%, p = NS). Similar to histologic findings, rPSGL-Ig reduced the perfusion defect size to risk area by 49%, from 68 ± 10% to 35 ± 9% (p = 0.001) (Fig. 2).Postischemic “now-reflow” zones correlated well with the infarct size by TTC staining of the corresponding myocardial slice (Fig. 2).
After initial hyperemia, a progressive reduction in myocardial perfusion was observed during reperfusion, which could be correctly quantified by MCE. A, β, and A × β gradually decreased during reperfusion but were higher in the rPSGL-Ig group, indicating preserved MBF in the risk area (Fig. 3).Recombinant PSGL-Ig reduced the decrease of A by 61% (59 ± 15% vs. 23 ± 14%, p = 0.003) and A × β by 49% (67 ± 13% vs. 34 ± 15%, p = 0.008), but not β (relative reduction of 12%, from 47 ± 8% to 40 ± 9%, p = NS). Myocardial blood flow by microspheres confirmed the cardioprotective effects of rPSGL-Ig on microvascular flow (Table 1). A × β correlated with MBF (r = 0.66, p < 0.001), and A correlated with capillary density (r = 0.8, p = 0.007).
Repeated measurements of A × β showed a good reproducibility (r = 0.93, p < 0.0001), with a mean ± 2 SD of 0.013 ± 0.38 dB/s. The risk area was not significantly different when assessed 3 or 10 heart cycles after flash, which demonstrates low collateral flow in pigs.
The present study outlines the pathophysiologic implication of P-selectin in the development of ischemia-reperfusion injury. Immunoneutralization of P-selectin by rPSGL-Ig reduced the infarct size and preserved microvascular integrity accurately detected by real-time MCE.
Assessment of ischemia-reperfusion injury by MCE.
Triggered MCE techniques have been used extensively in acute coronary syndromes and shifted the interest from infarct-related artery patency to microvascular integrity (1–3,8). However, practical considerations have limited its broader clinical use. Real-time MCE allows simultaneous evaluation of myocardial perfusion and contractility and obviates the need for background subtraction, but little information is available on the detection of postischemic microvascular dysfunction. In this study, myocardial perfusion could be easily visualized and quantified by real-time MCE. Furthermore, MCE confirmed a fully reperfused risk area characterized by absent perfusion defects and hyperemic reflow. During follow-up, the MBF index A × β decreased with reappearance of a marked perfusion defect. The infarct size and MBF by MCE corresponded well with standard techniques. These findings are in agreement with previous reports based on triggered and real-time techniques (11,12).
Kunichika et al. (12) recently demonstrated the usefulness of real-time MCE to assess the cardioprotective effects of glycoprotein IIb/IIIa inhibition in a canine model of 3-h ischemia. In contrast to our model, no significant changes in perfusion during reperfusion were reported, most likely reflecting extensive irreversible injured myocardium occurring during ischemia, as well as the inability to reperfuse the myocardium. The duration of ischemia was shorter in our study (45 min), resulting initially in hyperemic reflow with a gradual decrease in myocardial perfusion patterns A and β. This supports the existence of ischemia-reperfusion injury characterized by serial changes in the postischemic microcirculation.
Cardioprotective effects of rPSGL-Ig
In the present study, we used rPSGL-Ig, a recombinant immunoglobulin form of PSGL-1, which has been mutated to reduce both complement activation and Fc receptor binding with a half-life of ∼11 days in pigs (4–6). As a result, P-selectin blockade by rPSGL-Ig administrated at the time of reperfusion, on top of standard therapy with heparin and acetylsalicylic acid, reduced myocardial ischemia-reperfusion injury. This was manifested by a marked reduction in infarct size of ∼50%, improved myocardial capillary density, and blood flow, in agreement with previous studies (7).
PSGL-1 binds to P-selectin and is a key molecule responsible for recruitment of neutrophils into areas of inflammation and adhesion of platelets to neutrophils. Therefore, a proposed mechanism of rPSGL-Ig is the preservation of endothelial function through inhibition of leukocyte adhesion, thus potentially maintaining nitric oxide release and vasodilation (2,6,13). Although we did not directly investigate the pathophysiologic implications of leukocytes, we could observe that plateau A, an index of capillary volume, was elevated for a prolonged time after reperfusion in the rPSGL-Ig group. Hence, failure to respond to hyperemia after ischemia may be an ominous sign of a dysfunctional and damaged microvasculature, triggering a lethal reperfusion injury. Besides, P-selectin mediates interactions between platelets and neutrophils, which has potential implications, as leukocyte-platelet aggregates contribute to microvascular obstruction. Indeed, P-selectin blockade reduced platelet deposition on injured arterial surfaces (13) and enhanced thrombolysis (6) without an increase in bleeding rate.
Our study provides strong evidence of the detrimental effects of postischemic reperfusion and suggests a potential role of P-selectin in the pathogenesis of ischemia-reperfusion injury. However, it is not easy to speculate whether rPSGL-Ig may provide a beneficial adjunct to standard reperfusions strategies. The selectin antagonist CY-1503 reduced the extent of reperfusion lung injury (14), but rPSGL-Ig, currently studied in phase I/II clinical trials, failed to improve thrombolysis in patients with myocardial infarction (15). However, soluble P-selectin and intercellular adhesion molecule-1 concentrations were consistently elevated after percutaneous transluminal coronary angioplasty but not after thrombolysis with tissue-type plasminogen activator (16). Additional studies are needed to further elucidate this clearly exciting approach to preserve microvascular integrity. Because real-time MCE has been improved both qualitatively and quantitatively, it may become invaluable in the diagnosis and management of acute ischemic syndromes.
- Abbreviations and acronyms
- A × β
- myocardial perfusion
- fractional area shortening
- left anterior descending coronary artery
- left ventricle
- myocardial blood flow
- myocardial contrast echocardiography
- P-selectin glycoprotein ligand-1
- recombinant P-selectin glycoprotein ligand-immunoglobulin
- Received February 9, 2004.
- Revision received April 22, 2004.
- Accepted May 2, 2004.
- American College of Cardiology Foundation
- Ito H.,
- Maruyama A.,
- Iwakura K,
- et al.
- Ambrosio G.,
- Weisman H.F.,
- Mannisi J.A.,
- Becker L.C.
- Canfield S.M.,
- Morrison S.L.
- Kumar A.,
- Villani M.P.,
- Patel U.K,
- et al.
- Wei K.,
- Jayaweera A.R.,
- Firoozan S,
- et al.
- Weibel E.R.
- Van Osterhout A.F.M.,
- Willigers H.M.M.,
- Reneman R.S,
- et al.
- da Galiuto L.,
- DeMaria A.N.,
- Balzo U,
- et al.
- Kunichika H.,
- Yehuda O.B.,
- Lafitte S,
- et al.
- Merhi Y.,
- Provost P.,
- Chauvet P,
- et al.
- Tanguay J.F.,
- Krucoff M.W.,
- Gibbons R.J,
- et al.