Author + information
- Received June 6, 2003
- Revision received August 18, 2003
- Accepted August 27, 2003
- Published online January 21, 2004.
- Hideki Kunichika, MD, PhD*,
- Ori Ben-Yehuda, MD, FACC*,
- Stephane Lafitte, MD*,
- Naomi Kunichika, MD, PhD*,
- Barry Peters, MD* and
- Anthony N. DeMaria, MD, MACC*,* ()
- ↵*Reprint requests and correspondence:
Dr. Anthony N. DeMaria, Division of Cardiology, UCSD Medical Center, 200 West Arbor Drive, San Diego, California 92103-8411, USA.
Objectives We assessed the effect of glycoprotein IIb/IIIa inhibition (GPI) on microvascular flow after coronary occlusion/reperfusion using quantitative myocardial contrast echocardiography (QMCE).
Background Platelets may play a major role in the dissociation of epicardial artery recanalization and tissue-level reperfusion, referred to as the “no-reflow phenomenon.” Therefore, GPI might improve myocardial reperfusion, distinct from its effects on epicardial patency.
Methods Three-hour occlusion of the left anterior descending coronary artery (LAD) was followed by 3-h reperfusion in 16 open-chest dogs: 8 controls and 8 given a continuous infusion of the GPI tirofiban, starting 45 min before LAD reopening. Perfusion of the LAD bed was quantified by the rate of intensity rise (b)by QMCE; myocardial blood flow (MBF) was assessed by fluorescent microspheres.
Results No differences in bor MBF were observed within the risk area between the control and GPI groups at baseline or occlusion. However, band MBF were higher in GPI dogs than in controls during reperfusion, despite similar epicardial flow (p < 0.05 at 30, 60, and 90 min; p = NS at 180 min). Infarct area size was significantly reduced in GPI dogs compared with non-treated dogs (26.9 ± 10.5% vs. 49.0 ± 11.1% of at-risk area, respectively).
Conclusions As demonstrated by QMCE, GPI improves microvascular flow and reduces the infarct area after coronary occlusion/reperfusion, independent of epicardial flow. These data demonstrate the usefulness of QMCE in assessing microvascular flow, provide novel evidence for the role of platelets in the early phase of reperfusion injury, and show that GPI is of value in preserving microvascular perfusion after coronary reperfusion.
Reperfusion strategies have led to a substantial improvement in the prognosis of patients with acute myocardial infarction (MI). However, the beneficial effect of successful restoration of epicardial coronary flow on myocardial salvage may be offset by inadequate tissue perfusion, a condition referred to as the “no-reflow phenomenon” (1–3). Myocar- dial contrast echocardiography (MCE), which can provide information on tissue-level perfusion, has demonstrated that even in the presence of Thrombolysis In Myocardial Infarction (TIMI) flow grade 3, tissue-level perfusion may still be severely reduced. Moreover, significant improvement in myocardial function has been demonstrated to occur only in the presence of MCE reflow.
The mechanism responsible for the no-reflow phenomenon is uncertain and is likely multifactorial. Due to the potential for thrombus formation and release of vasoactive substances, it has been suggested that platelets may play a major role in the dissociation of epicardial artery recanalization and tissue-level reperfusion (4). However, the role of platelets in reperfusion injury, independent of epicardial thrombosis and embolization, remains to be determined. Similarly, although data exist that inhibition of the platelet glycoprotein IIb/IIIa receptor prevents abrupt re-occlusion after percutaneous coronary revascularization procedures and improves TIMI flow rates in acute coronary syndrome patients, the ability of glycoprotein IIb/IIIa inhibition (GPI) to enhance flow at the microvascular level after coronary reperfusion has not been determined (5,6).
We have recently shown that quantitative parameters derived from refilling curves generated from quantitative myocardial contrast echocardiography (QMCE) are closely correlated to myocardial blood flow (MBF). Accordingly, QMCE enables evaluation of flow in the coronary microvasculature (7,8). In the present study, we employed QMCE to examine whether glycoprotein IIb/IIIa inhibition (GPI) improves microvascular flow and thereby reduces infarct size in an animal model of coronary occlusion/reperfusion.
The present study was approved by the University of California-San Diego, Animal Research Committee and conformed to the “Position of the American Heart Association on Research Animal Use,” adopted by the Association in November 1984. Sixteen mongrel dogs (26.1 ± 1.9 kg) were anesthetized and ventilated to keep arterial blood gases and pH within normal limits. The right femoral artery and vein were cannulated for arterial pressure monitoring and contrast agent injection, respectively. The heart was exposed through a left lateral thoracotomy and suspended in a pericardial cradle. The proximal portion of the left anterior descending coronary artery (LAD) was dissected free from the surrounding tissue. An atraumatic vascular clamp produced coronary occlusion of the proximal LAD.
After a period of stabilization after instrumentation, 3-h proximal LAD occlusion was followed by 3-h reperfusion in 16 open-chest dogs: 8 controls and 8 given a continuous infusion of tirofiban (Aggrastat, Merck & Co., Whitehouse Station, New Jersey) at 3 μg/kg/min, starting 45 min before LAD reopening (9–11). Hemodynamic measurements, MBF, coronary flowmeter, and real-time MCE data were acquired at baseline, immediately before release of occlusion (180 min), and at 30, 60, 90, and 180 min of reperfusion.
Real-time imaging with MCE
Echocardiography was performed with a commercial instrument (HDI 5000, Philips Ultrasound, Andover, Massachusetts) using a broadband 4- to 2-MHz transducer. Color-coded harmonic power-pulse inversion images were obtained with ultrasound transmitted at 2 MHz and received at 4 MHz in the short-axis papillary muscle view, using low-energy (mechanical index = 0.1) real-time imaging at 15 frames/s (12,13). The pulse repetition frequency was fixed at 2,500 Hz. Instrument settings were held constant for each experiment. A latex bag filled with degassed saline functioned as an acoustic interface between the heart and transducer, which was positioned to image the LAD perfusion territory. SonoVue (Bracco Inc., Geneva, Switzerland) was continuously infused at a rate of 30 ml/h by a gently agitated infusion pump. Recordings were obtained 2 min after initiating infusion to ensure that plateau intensity had been reached. The MCE refilling sequences were recorded; they consisted of high-energy (mechanical index = 0.8) fast low-angle shot (FLASH) frames to destroy microbubbles, followed by 15 cardiac cycles of refilling.
The raw image data were digitally captured and analyzed off-line using HDI laboratory software (Philips Ultrasound). Perfusion of the LAD bed was quantified by fitting intensity data of end-systolic images to an exponential function: y = A (1 − e−bt), where y is the signal intensity at any given time; A is the plateau signal intensity that reflects the microvascular cross-sectional area or myocardial blood volume; bis the rate of signal intensity rise (slope of curve) that reflects myocardial microbubble velocity; and t is the time after FLASH. The bparameter was used to estimate perfusion. Transmural regions of interest that encompassed the LAD perfusion territory were selected, excluding high-intensity signals from the epicardium and endocardium.
The myocardial risk area was identified as the region of unopacified myocardium by MCE during coronary occlusion. To measure the risk area size from MCE images, the area of the largest clearly demarcated myocardial opacification defect at 180 min of LAD occlusion was manually traced for the final three end-diastolic images of the 15-cycle FLASH refilling sequence (14). Also, to measure the infarct area size from MCE, the myocardial opacification defect at 180 min of reperfusion was traced for the final three end-diastolic images. The infarct area size from MCE was expressed as the percent of the area at risk.
Measurements of LAD flow and MBF
Epicardial LAD flow was measured by a Doppler flowmeter. A transit-time flow probe connected to a digital flowmeter (series 2RB and model T201, Transonics System, Ithaca, New York) was placed snugly around the proximal LAD. Epicardial LAD flow and MBF were measured at baseline, immediately before release of occlusion (180 min), and at 30, 60, 90, and 180 min after release of occlusion by injection of fluorescent microspheres (Molecular Probes, Eugene, Oregon) into the left atrium, while reference blood samples were withdrawn from the femoral artery. For every injection, a different fluorescent wavelength was used to allow for independent assessment of flow at the different time points. After the animal was euthanized, the heart was sliced, and the cross-sectional segment corresponding to the short-axis image was cut into 12 wedge-shaped transmural tissue pieces, each of which was divided into endocardial and epicardial segments. Transmural MBF to 12 wedge-shaped pieces was calculated as the quotient of the summed flows to the individual segments within that piece and their combined weight. The MBF to the LAD beds, defined by monastral blue dye injection, was then calculated by averaging the transmural MBF in the pieces from LAD bed.
Infarct and risk area size
At the end of the experiment, the LAD was re-occluded, and blue dye was injected into the left atrium. With the guidance of long needles placed as markers, a left ventricular short-axis slice (6 to 8 mm thick) was cut out at the same level at which the echocardiogram was recorded. The risk area was delineated as that without blue stain. To determine the infarct area, each slice was incubated in 1.5% triphenyltetrazolium chloride (TTC) at 37°C for 15 min, and the unstained region with TTC was planed (15). Blinded quantitative histologic analysis of the risk segment was also performed by light microscopy to look for platelets.
Data are expressed as the mean ± SD. Analyses of the data were performed in a blinded fashion. Comparisons of data among all stages were performed using repeated-measures analysis of variance. Comparisons of multiple linear regression data between control and GPI groups were performed using analysis of co-variance. Differences were considered significant at p < 0.05.
All experiments were completed successfully without significant hemodynamic changes or adverse events. Recordings were technically adequate to allow full measurement and analysis.
Signal intensity data provided by MCE
Measurements of the bparameter by MCE, representing the rate of intensity rise within the risk area, are summarized in Table 1for each group at baseline, occlusion, and after coronary reperfusion. No significant differences in bwere observed between the control and GPI groups either at baseline or with occlusion. However, bwas significantly reduced during occlusion, compared with baseline, in both groups. During reperfusion, bimproved in both groups but was greater in GPI dogs than in controls (Fig. 1). This difference reached statistical significance at 30, 60, and 90 min but not at 180 min of reperfusion.
Assessment of size of infarct area
The percent infarct area of the total risk area derived by MCE was significantly reduced in GPI dogs compared with controls (26.9 ± 10.5% vs. 49.0 ± 11.1% of area at risk, p < 0.01). Figure 2shows representative changes in end-systolic images after FLASH obtained after 180 min of reperfusion in GPI and control dogs. To validate MCE measurements of infarct size, the percent myocardium of the total left ventricle manifesting an opacification defect was compared with that measured by TTC at 180-min reperfusion. Measurements of the size of the MCE opacification defect closely correlated with those of the infarct area by TTC (y = 1.0x − 1.3, r = 0.94) (Fig. 3). Quantitative histologic analysis of the risk segment, including the no-reflow zones, failed to reveal platelet plugging of the microvasculature in either control or GPI dogs.
LAD flow and MBF
Data on epicardial LAD flow by Doppler and on MBF by microspheres at baseline, 180-min LAD occlusion, and after coronary reperfusion are summarized in Table 1for each group. The MBF data during occlusion and reperfusion are expressed as a percentage of the baseline value. During LAD occlusion, Doppler flow and MBF within the risk area were severely reduced from baseline in both groups. During reperfusion, epicardial LAD flow returned to baseline in both groups. Although MBF within the risk area continued to be reduced in the control group after reperfusion, it returned to baseline levels in the GPI dogs. Thus, no differences in risk-area MBF were observed between control and GPI groups at baseline or occlusion. However, at 30, 60, and 90 min of reperfusion, MBF in GPI dogs was significantly higher than in controls (Fig. 4). At 180 min, the difference in MBF between the GPI and control dogs was no longer statistically significant.
Infarct size and MBF during occlusion
The canine model has variable coronary collateral flow. Therefore, infarct size, normalized as a percentage of the area at risk, was plotted against the mean MBF of the risk area during coronary occlusion for the control and GPI groups (Fig. 5). A significant downward shift in the line of the inverse relationship was observed in the GPI-treated group (p < 0.05). This result clearly shows that GPI has a true effect to reduce infarct size.
In this study, we used QMCE, which reflects blood velocity and volume in the microcirculation, to assess the effect of tirofiban, a specific GPI (16), on tissue-level perfusion in a dog model of occlusion/reperfusion. We demonstrated enhanced microvascular flow that was independent of epicardial flow with GPI, particularly in the initial 90 min of reperfusion. Furthermore, we observed that this increased flow was associated with a reduction in infarct size. These data are the first to directly establish that GPI exerts a favorable effect on capillary perfusion after restoration of flow in epicardial vessels.
Our study also provides important evidence for the role of platelets in the early phase of reperfusion. Assessing the role of platelets in reperfusion injury is confounded clinically and, in some experimental models of MI, by the presence and possible microembolization of epicardial thrombus (17). Indeed, using labeled platelets, microembolization has been documented in a model of coronary thrombosis. Our data implicate platelets in the mechanism of reperfusion injury, even in the absence of coronary thrombosis.
Infarct vessel patency has traditionally has been viewed as the hallmark of successful reperfusion. However, increasing evidence indicates that, even in the presence of vessel patency and seemingly normal flow, up to one-third of patients do not have adequate reperfusion at the tissue level. Inadequate tissue-level perfusion has been shown to correlate with a lack of myocardial salvage. Therefore, optimal reperfusion can be redefined as restoration of both epicardial and tissue-level perfusion (18–20). This study demonstrates a role for GPI in enhancing the latter.
Previous studies have provided evidence of the beneficial effect of GPI in acute coronary syndromes. It has been shown that GPI improves the outcomes of patients with acute coronary syndromes undergoing a primary percutaneous coronary intervention (21,22). In the Platelet Receptor inhibition for Ischemic Syndrome Management in Patients Limited by Unstable Signs and symptoms (PRISM-PLUS) trial, the use of tirofiban was shown to reduce intracoronary thrombus and improve TIMI flow (23). In the setting of ST-segment elevation MI, GPI was shown to improve the speed and efficacy of epicardial reperfusion in combination with lytic therapy. In addition, GPI reduced the 30-day incidence of death, re-infarction, and urgent target vessel revascularization in the setting of primary stenting, albeit with an increased risk of bleeding (24,25). Flow velocity in the infarct-related artery has also been shown to be improved with GPI with stenting (5).
Although the beneficial effects of GPI on the microvasculature have also been suggested, current studies have used only indirect assessment, such as improvement in coronary flow, resolution of ST-segments, and improvement in myocardial salvage, to gain insight into the effects on the microcirculation (5). The most direct evidence, to date, has been in the setting of elective percutaneous coronary intervention. In a substudy of the Enhanced Suppression of the Platelet GP IIb/IIIa Receptor with Integrilin Therapy (ESPRIT) trial, Gibson et al. (26)demonstrated that coronary flow reserve was improved in the eptifibatide-treated group, compared with placebo. Myocardial blush was also assessed using a visual scale and also demonstrated improvement. The QMCE data in our study directly demonstrate that GPI enhances microcirculatory flow during reperfusion.
Mechanism of the no-reflow phenomenon
The proposed mechanism of the no-reflow phenomenon is multifactorial. Animal and postmortem histologic studies have demonstrated varying degrees of small-vessel vasospasm, endothelial gap and bleb formation, neutrophil plugging of capillaries as well as microvascular compression from myocytes, interstitial edema, and hemorrhage after recanalization (1,27,28). Platelets have also been implicated as major contributors to the no-reflow phenomenon by inducing vascular injury and the loss of capillary autoregulation.
Platelet effects may be particularly important in the first stage of reperfusion, referred to as the “microvascular obstruction stage” (29). Additional reperfusion injury, mediated by neutrophils, free radical release, and edema formation, is thought to occur in the second stage. Platelet activation can lead to the formation of thrombi or distal microemboli, and obstructive platelet aggregates within myocardial capillaries have been observed both experimentally and clinically in reperfused ischemic tissue (30–32). Platelet degranulation has also been observed during reperfusion. Released platelet granules contain multiple vasoactive and chemotactic mediators that can compound tissue ischemia and increase neutrophil infiltration. Distal vasoconstriction due to platelet activation may mediate the deleterious influence of nonocclusive coronary thrombosis on distal tissue perfusion (33,34). After experimental ischemia/reperfusion, coronary microvessels have been demonstrated to exhibit increased vascular resistance, a finding that led to the so-called “injury-spasm hypothesis,” perhaps mediated by platelets (35). Thus, several mechanisms exist by which GPI may be beneficial in the prevention of the no-reflow phenomenon.
Quantifying tissue perfusion by QMCE in response to GPI therapy
Quantitative MCE is a novel method to assess myocardial perfusion; it can be performed repeatedly throughout the course of acute MI. Contrast microbubbles are excellent tracers of red blood cell kinetics. The method to quantify MBF velocity is based on rapid destruction of microbubbles by ultrasound and subsequent assessment of the rate at which they are replenished into the myocardial microcirculation contained in the imaging field.
Our results using QMCE show that GPI improves microvascular flow and reduces infarct size after coronary occlusion/reperfusion. Importantly, the improvement in tissue-level perfusion in the GPI group compared with the control group was demonstrated, despite equivalent epicardial flow, as measured by Doppler. Presumably, the epicardial flow not transiting the microcirculation in the controls was accommodated by the extensive collateral channels known to be present in dogs. The QMCE measurements were further validated by microsphere and TTC data, with a good correlation between MBF derived by microspheres and bby QMCE.
Although the bparameter at 180 min of reperfusion was greater in GPI dogs than in controls, this difference did not reach statistical significance. These data are consistent with the concept that platelets play a greater role in the no-reflow phenomenon early after reperfusion rather than later. The late loss of a GPI effect on tissue perfusion seen in our study may have been caused by reperfusion injury, which is thought to occur after the microvascular obstruction stage (18). The use of other concomitant therapies, such as free radical scavengers and anti-adhesion molecules, may be necessary to protect the microcirculation from this late phenomenon. Although of lesser significance than the difference at earlier time points, the effect at 180 min may have been affected by our small sample size and may still be of clinical importance, as evidenced by the reduction in infarct size.
Infarct size and GPI
The data from this study also demonstrated that GPI reduced infarct size. The favorable effects on infarct size observed with GPI might have been related to a more rapid restoration of MBF. The enhancement of early reflow by GPI may promote the delivery of blood to the risk area, thus reducing MI size, speeding healing, and decreasing infarct expansion. In the TIMI-14 trial, the improvement in perfusion of the infarct-related artery seen with additional administration of GPI was greater at 60 min than at 90 min of reperfusion, suggesting that GPI may accelerate the speed as well as enhance the extent of epicardial reperfusion (24). Our results suggest that the beneficial effects of GPI may be mediated not only by a faster recanalization of the culprit epicardial vessel but also by enhanced microvascular flow, resulting in a reduced size of MI.
There are several limitations of this study. These results were obtained using a canine model, which has abundant natural collateral channels, and so may not apply to patients with more severely impaired multivessel coronary disease. Heparin and/or lytic therapy was not administered in this model, and hence the contribution of GPI in combination with these agents was not assessed. We did not measure platelet aggregation in our study animals; therefore, the exact degree of platelet inhibition that corresponds with the improvements in MBF and infarct size cannot be determined. However, the dose of tirofiban chosen has been shown to inhibit platelet aggregation in the dog by >80%, a level documented to be necessary for clinical benefit in both animal and human studies (9,36,37). Because we assessed infarct size at 3-h reperfusion, the ultimate infarct size may be larger, especially given the possibility of further non–platelet-mediated reperfusion injury. In addition, the infarct size by QMCE may change with variation of the duration of ischemia. Finally, these results were observed in a acute experimental setting, and no long-term data are available.
Our study supports a role for platelets in early reperfusion injury and demonstrates a benefit of GPI on microvascular flow, even in the absence of overt coronary thrombosis. Furthermore, coronary recanalization strategies are focusing on new end points, with microvascular perfusion being particularly important. This study also demonstrates that QMCE can provide a direct approach to assessing changes in tissue perfusion in response to GPI therapy and, moreover, that it has the unique ability to delineate the myocardial and microvascular level at which these structural and physiologic changes occur after coronary reperfusion.
We thank Peter DiBattiste, MD, and Jacquelynn Cook, PhD, of Merck Research Laboratories for their assistance and advice.
☆ Dr. DeMaria has received grants and been a sponsored speaker or occasional ad hoc consultant for Bracco Pharmaceuticals as well as virtually all other ultrasound contrast manufacturers. Both Drs. DeMaria and Ben-Yehuda have been sponsored speakers for Merck Inc., which provided a research grant for this study. Jonathan R. Lindner, MD, FACC, acted as the Guest Editor of this paper.
- rate of signal intensity rise
- fast low-angle shot
- glycoprotein IIb/IIIa inhibition/inhibitor
- left anterior descending coronary artery
- myocardial blood flow
- myocardial contrast echocardiography
- myocardial infarction
- quantitative myocardial contrast echocardiography
- Thrombolysis In Myocardial Infarction
- triphenyltetrazolium chloride
- Received June 6, 2003.
- Revision received August 18, 2003.
- Accepted August 27, 2003.
- American College of Cardiology Foundation
- Ito H.,
- Okamura A.,
- Iwakura K.,
- et al.
- Anderson K.M.,
- Califf R.M.,
- Stone G.W.,
- et al.
- Neumann F.J.,
- Blasini R.,
- Schmitt C.,
- et al.
- Masugata H.,
- Peters B.,
- Lafitte S.,
- et al.
- Galiuto L.,
- DeMaria A.N.,
- del Balzo U.,
- et al.
- Lynch J.J.,
- Cook J.J.,
- Sitko G.R.,
- et al.
- Hammerman H.,
- Kloner R.A.,
- Briggs L.L.,
- Braunwald E.
- Lafitte S.,
- Higashiyama A.,
- Masugata H.,
- et al.
- Lele M.,
- Sajid M.,
- Wajih N.,
- et al.
- Roe M.T.,
- Ohman E.M.,
- Maas A.C.P.,
- et al.
- Ito H.,
- Maruyama A.,
- Iwakura K.,
- et al.
- Montalescot G.,
- Barragan P.,
- Wittenberg O.,
- et al.
- Zhao X.-Q.,
- Theroux P.,
- Snapinn S.M.,
- et al.
- de Lemos J.A.,
- Antman E.M.,
- Gibson C.M.,
- et al.
- Brener S.J.,
- Barr L.A.,
- Burchenal J.E.,
- et al.
- Manciet L.H.,
- Poole D.C.,
- McDonagh P.F.,
- et al.
- Topol E.J.,
- Yadav J.S.
- Saitoh S.,
- Onogi F.,
- Aikawa K.,
- et al.
- Shen Y.-T.,
- Wiedmann R.T.,
- Lynch J.J.,
- et al.
- Jordan R.E.,
- Wagner C.L.,
- Mascelli M.,
- et al.
- Steinhubl S.R.,
- Talley J.D.,
- Braden G.A.,
- et al.