Author + information
- Received May 17, 1996
- Revision received December 11, 1996
- Accepted December 20, 1996
- Published online April 1, 1997.
- Mario F Meza, MDA,
- Marc A Kates, DOA,
- R.Wayne Barbee, PhDA,
- Susan RevallA,
- Bret PerryA,
- Joseph P Murgo, MD, FACCA and
- Jorge Cheirif, MD, FACCA,*
- ↵*Dr. Jorge Cheirif, Ochsner Medical Institutions, 1514 Jefferson Highway, New Orleans, Louisiana 70121.
Objectives. This study tested whether the combination of dobutamine echocardiography (DE) and myocardial contrast echocardiography (MCE) was superior to either technique alone in identifying postischemic myocardium and in differentiating it from necrotic myocardium.
Background. Wall motion abnormalities at rest occur in postischemic myocardium in the presence of infarction, stunning or hibernation, alone or in combination. Various investigators have suggested that either DE or MCE can be used to identify the presence of myocardial viability.
Methods. We studied a total of 53 mongrel dogs in an open chest model of coronary occlusion of various durations followed by reperfusion and dobutamine administration (10 μg/kg body weight per min). MCE with aortic root injections of Albunex (area under the curve) and DE (percent thickening fraction) were performed at the different stages. Postmortem triphenyltetrazolium chloride (TTC) staining was used to identify myocardial necrosis.
Results. Thirteen dogs underwent brief (15 min) occlusions and developed no necrosis (Group I). Of 40 dogs that underwent prolonged (30 to 360 min) occlusions, 14 had no infarction (Group II), whereas 26 did (Group III: 12 papillary muscle, 7 subendocardial, 7 transmural). MCE (expressed as percent change from baseline) demonstrated changes that paralleled the blood flow changes observed by radiolabeled microspheres at all interventions (r = 0.67, p < 0.0001). Regional ventricular function improved with dobutamine administration in the ischemic region in all three groups. The sensitivity (88%) for detecting myocardial viability was superior when the two techniques were combined; however, a poor specificity (61%) was observed.
Conclusions. Contractile reserve and perfusion data are complementary when assessing regional wall motion abnormalities in postischemic myocardium. DE alone cannot differentiate postischemic from infarcted myocardium; simultaneous data on myocardial perfusion are required. The combination of DE and MCE is superior to either technique alone for identifying the absence of myocardial necrosis.
(J Am Coll Cardiol 1997;29:974–84)
© 1997 by the American College of Cardiology
Early reperfusion (by means of thrombolysis or primary angioplasty, or both) in the setting of acute myocardial infarction (MI) has resulted in the reduction of morbidity and mortality. However, the success of either intervention has been difficult to assess because wall motion abnormalities at rest in this setting could relate to infarcted as well as postischemic or hibernating myocardium, or both ([1–5]). To compound this problem, the duration of ischemia could affect the degree of postischemic recovery of function ([6, 7]). Thus, diagnostic techniques able to distinguish infarcted from noninfarcted, yet asynergic, myocardium would be clinically useful. Various investigators have previously suggested that myocardial contrast echocardiography (MCE) ([8–15]) and dobutamine echocardiography (DE) ([16–24]) are useful techniques for identifying the presence of myocardial viability in the setting of acute MI. Whether these two techniques can be used to distinguish necrotic from reperfused (postischemic) myocardium remains to be determined.
Because differentiation between stunned, ischemic and infarcted myocardium relies on the knowledge of myocardial perfusion and potentially the degree of response to inotropic stimulation, we hypothesized that the combination of MCE (with its ability to demonstrate preserved myocardial perfusion) and DE (with its ability to demonstrate contractile reserve) is superior to either one individually in distinguishing viable from nonviable myocardium and in distinguishing postischemic dysfunction (i.e., “stunning,” from infarction of various degrees of transmurality).
The study protocol conformed to the “Position of the American Heart Association on Research Animal Use” adopted by Association in November 1984 and was approved by the Animal Care Committee of the Ochsner Medical Institutions.
1.1 Surgical preparation.
Fifty-three mongrel dogs weighing 29.5 ± 4.7 kg (mean ± SD) were anesthetized with 30 mg/kg body weight of sodium pentobarbital intravenously (supplemented as necessary), intubated and artificially ventilated with a Harvard respirator. Dogs were instrumented as previously described by us ([6, 25–28]). The heart was exposed through a left thoracotomy and suspended in a pericardial cradle. A cannula was placed in the left atrium to inject microspheres. Catheters were placed in the carotid artery for continuous blood pressure monitoring, in the femoral artery for microsphere reference blood sampling, in the jugular vein for fluids and drug administration and in the femoral artery to place a modified pigtail catheter in the proximal aortic root just above the aortic valve for administration of sonicated albumin.
The proximal left anterior descending (LAD) and left circumflex (LCx) coronary arteries were dissected free, and silk ties were loosely placed around them. Distal to the ties, a Doppler flow velocity probe was placed snugly around each vessel. Ties were used to occlude the LAD in 44 dogs and the LCx in 9. Continuous recordings of arterial and pulmonary artery pressure, electrocardiogram and Doppler flow velocity signals were obtained using a five-channel strip chart recorder. The chest cavity was filled with warm saline, and a ziplock bag filled with saline was placed over the heart and used as an acoustic interface. The echocardiographic short-axis slice imaged was identified with the use of an epicardial suture as previously described ([6, 25–28]).
1.2 Assessment of regional blood flow.
Regional myocardial blood flow (RMBF) was assessed with radiolabeled microspheres (cerium-141, tin-113, scandium-46, ruthenium-103) as previously described ([6, 25–29]). One to two million spheres were injected into the left atrium. Thirty seconds before injection, reference blood was drawn at a constant rate of 4.5 ml/min. Withdrawal continued for 2.5 min after injection of spheres. After the protocol, the reference blood and tissue samples were counted in a gamma counter. RMBF was calculated for each nuclide using the standard formula ().
In this study, RMBF, at the different stages of the protocol, was expressed as the percent change between events in the ischemic region (i.e., baseline to coronary occlusion).
1.3 Myocardial contrast echocardiography.
MCE was performed as previously described ([26–28]). Briefly, 0.5 to 1.0 ml of sonicated 5% human serum albumin (Albunex, Molecular Biosystems, Inc.) was injected into the aortic root. The dose that produced optimal myocardial opacification at baseline in each dog was used and kept unchanged during the remainder of the protocol ([26–28]). Three to five end-diastolic frames were digitized before contrast injection, and enough frames were placed on cine-loop format to allow the myocardial contrast effect to pass and completely disappear from the myocardium.
1.4 Assessment of regional myocardial perfusion by MCE.
Regional myocardial perfusion was analyzed with previously described and validated on-line software (Acoustic Densitometry, Sonos 1500, Hewlett-Packard) ([28, 30]). This software allows the operator to capture sequential end-diastolic frames to analyze changes in intensity on-line with injection of echo contrast medium. The computer subtracts the intensity values (preinjection) from those obtained in the frames demonstrating the contrast effect, constructing a time–intensity curve. As in previous studies, ([25, 26, 28]) the area under the curve (AUC) was used as the MCE variable of myocardial perfusion. The investigator (M.F.M.) performing the MCE analysis had no knowledge of the results of RMBF and regional myocardial function.
1.5 Regional function assessment.
Quad screen cine loops of digitized cardiac cycles were created with Freeland-Tomtec software. Regional function (thickening fraction) was assessed as described by Porter et al. (). The advantage of thickening fraction is that it allows not only for the differences in wall thickness among dogs, but also for changes that occur with ischemia and reperfusion (). Regional function was assessed in the center of the ischemic zone by an investigator (J.C.) with no knowledge of the results of RMBF and MCE and to which group the dogs belonged.
1.6 Tissue processing.
After the experimental protocol, the heart was removed from the chest. To identify each individual coronary artery perfusion territory, the dual-perfusion technique was used (). The coronary arteries were perfused simultaneously at identical pressures with two dyes: macrodex (transparent dye) into the artery perfusing the ischemic territory; monastral blue (blue dye) into the artery perfusing the control territory, allowing macroscopic identification and measurement of the perfusion territories of each coronary artery.
The heart sections paralleled the coronary sinus at distances of 1 cm toward the apex. The slice corresponding to the echocardiographic region of interest was incubated for 20 to 30 min in triphenyltetrazolium chloride (TTC) (pH 7.4 at 38°C) () and then photographed. Results of staining were classified as positive(absence of necrosis) or negative(necrotic myocardium). Papillary muscle infarctswere those showing any necrotic pale areas within the papillary muscle. Subendocardial infarctswere those showing any necrosis by TTC in the subendocardium (inner half of the myocardium). Transmural infarctswere those showing any degree of subepicardial necrosis. To find the amount of tissue necessary to maintain contractile reserve and baseline myocardial perfusion, the relation between amount of necrosis (percent area at risk), perfusion and thickening was obtained.
The slice of tissue corresponding to the echocardiographic region of interest was sectioned into six equidistant segments to obtain a large number of radiolabeled microspheres per tissue sample, a factor known to increase its reliability (). The resulting samples were then placed in the gamma counter for quantitation of radioactivity and calculation of RMBF.
1.7 Assessment of myocardial necrosis.
Sensitivities and specificities for identification of absence of myocardial necrosis were calculated for MCE and DE (independently and in combination). Myocardial necrosiswas defined as the absence of red staining by TTC in the postmortem analysis.
1.7.1 Absence of necrosis by MCE.
The criterion for absence of necrosis by MCE was return of myocardial perfusion within 20% of the preocclusion value of the AUC 1) after release of the coronary occlusion or 2) during dobutamine administration. This criterion was selected on the basis of two findings: Injection to injection variability of this variable previously shown by us using the videointensity analysis system used in this study (); and a hypothesis that infarcted tissue might be accompanied by a lower blood flow than control tissue, as suggested by Gewirtz et al (). Thus, for MCE, true positiveregions were considered those that demonstrated a return of perfusion after release of the occlusion with or without dobutamine administration and had no myocardial necrosis by TTC staining. True negativeregions were considered those that demonstrated no return of perfusion to within 20% of baseline at release of the occlusion with or without dobutamine administration and had evidence of myocardial necrosis. False positiveregions were considered those with return of perfusion at release of the occlusion or during dobutamine administration, yet demonstrated myocardial necrosis by TTC staining. False negativeregions were considered those with no return of perfusion with or without dobutamine administration and no evidence of myocardial necrosis at postmortem analysis.
1.7.2 Absence of necrosis by DE.
Criteria for absence of necrosis on regional function consisted of improvement in thickening fraction (≥10% based on reproducibility in 18 separated measurements) after reperfusion in the ischemic region versus the values observed during coronary occlusion. This improvement was observed after reperfusion in 13 of 35 frankly dyskinetic regions (10 became hypokinetic, 3 became normal). For DE, true positiveregions were considered those that demonstrated improvement in thickening fraction after release of the coronary occlusion with or without dobutamine administration and no evidence of myocardial necrosis by TTC staining. True negativeregions were considered those with no improvement in thickening fraction, yet had evidence of necrosis. False positiveregions were considered those with improvement in thickening fraction and evidence of necrosis. False negativeregions were considered those without improvement in thickening fraction and no evidence of myocardial necrosis by TTC staining.
1.7.3 Absence of necrosis by MCE and DE combined.
When the two techniques were combined, true positiveregions consisted of those that demonstrated perfusion at levels within 10% of baseline values or an increase in wall thickening after release of the occlusion with or without dobutamine administration, with no evidence of myocardial necrosis by TTC staining. True negativeregions were those with no return of perfusion and no lessening of wall thickening at release of the occlusion with or without dobutamine administration, with evidence of necrosis by TTC staining. Finally, false positiveand false negativeregions were those that met either set of criteria for each individual technique, as previously described.
Hemodynamic variables were constantly monitored. After instrumentation, baseline injections of radiolabeled microspheres and Albunex were performed in the left atrium and aortic root, respectively. Intravenous lidocaine (1.5 to 2 ml/kg) was followed by occlusion of one coronary artery. Absence of anterograde flow was confirmed by Doppler. A second set of injections was performed 10 min after occlusion.
To create a model of short ischemia, occlusions were released after 15 min (Group I, 13 dogs). To create models of prolonged ischemia and infarction, occlusions were kept between 30 and 360 min (prolonged occlusion without [Group II, 14 dogs] and with infarction [Group III, 26 dogs]). Release of occlusion was verified by Doppler. A third set of injections was performed after 10 to 15 min of stable reperfusion. The last set of injections was performed 30 min after reperfusion during intravenous administration of 10 μg/kg per min of dobutamine. The animals were killed with an overdose of potassium chloride. Assessment of regional function, myocardial perfusion (by MCE) and RMBF was performed at baseline, after coronary occlusion, after reperfusion and during intravenous administration of dobutamine.
1.9 Statistical analysis.
Linear regression analysis was performed for RMBF by radiolabeled microspheres and AUC in the region of interest, both expressed as percent change between events and baseline. Analysis of variance for multiple observations was performed between the different interventions (baseline, occlusion, release, dobutamine) for each of the measured variables (microspheres, MCE, wall thickening fraction). Multivariate analysis of variance was used for comparison of variables at each different intervention for analysis of repeated measures. Between-group mean values were compared by post hoc tests. Sensitivities and specificities for DE and MCE alone and in combination and their 95% confidence intervals (using p-normal theory confidence interval estimations) were calculated. Super Anova Version 1.11 software (Abacus Concepts, Inc.) was used. Statistical significance was established at p < 0.05.
2.1 Hemodynamic variables (Table 1).
Heart rate and mean arterial pressure were stable during coronary occlusions. Compared with baseline measurements, a significant decrease in heart rate was observed during reperfusion, followed by a significant increase during dobutamine infusion. A significant decrease in pressure, compared with that at baseline, was observed during reperfusion, with a return to baseline values when dobutamine was added.
2.2 Correlation between regional myocardial perfusion and RMBF.
A total of 126 paired measurements in 53 dogs were available in the ischemic zone for comparison between RMBF by radiolabeled microspheres and AUC, both expressed as percent change (%Δ) between events in the ischemic region, as previously described. The linear regression plot is shown in Fig. 1(r = 0.67, p < 0.0001).
2.3 Regional perfusion and function.
This model mimics the classic model of stunning (i.e., 15-min occlusion followed by reperfusion) ([34, 35]). All 14 dogs demonstrated a normal contractile pattern and lack of perfusion defects at baseline. As expected, during occlusion, a significant decrease in RMBF was observed (%Δ −75 ± 17 [mean ± SD] from baseline). A similar decrease in AUC was observed (%Δ −97 ± 62), accompanied by the development of severe wall motion abnormalities (wall thickening fraction −11 ± 17%).
After release of the occlusion, an increase (i.e., reactive hyperemia) in RMBF (%Δ 27 ± 46, p < 0.0001 from baseline) and AUC (%Δ 62 ± 127, p < 0.0009 from baseline) was observed. Regional function showed a significant improvement compared with that during occlusion; however, significant dysfunction persisted relative to baseline measurements (8.7 ± 14% at release, p < 0.05 from baseline).
After dobutamine administration, further improvement in function was noted with return to preocclusion levels (thickening fraction 23 ± 14%, p = NS from baseline). In this group, 6 (54%) of 11 dogs showed a return of function to preocclusion levels during dobutamine administration. One dog developed ventricular fibrillation requiring multiple defibrillations before dobutamine administration. Thus, regional function was not analyzed in this animal. TTC staining revealed no evidence of myocardial necrosis in the perfusion territory of the occluded vessel.
Thirteen dogs underwent prolonged occlusions (30 to 360 min), yet developed no evidence of infarction on TTC staining. One dog had fibrillation and was not analyzed during dobutamine administration. Significant decreases (%Δ from baseline) in RMBF (%Δ −58 ± 22), AUC (%Δ = −69 ± 26) and thickening fraction (−14 ± 34%, p < 0.0001) were observed during occlusion in the ischemic region. After release of the occlusion, RMBF (%Δ −32 ± 44, p = NS from baseline) and AUC (%Δ −18 ± 94, p = NS) returned toward baseline values; however, despite this return, the severe abnormality in function observed during occlusion persisted (thickening fraction −11 ± 20, p < 0.0001 from baseline).
During dobutamine administration, an increase in RMBF (%Δ 42 ± 82, p < 0.0009) and AUC (%Δ 100 ± 18, p < 0.004) from postrelease levels was seen. Improvement in thickening fraction from postrelease levels was also observed (14 ± 17%, p < 0.0003 from release). However, a depressed function prevailed when compared with preocclusion levels (p = NS); only 7 (50%) of 14 dogs showed improvement in function to preocclusion levels. TTC staining revealed no evidence of myocardial necrosis in any dog in this group.
Twenty-six dogs undergoing prolonged coronary occlusions (180 to 360 min) demonstrated the presence of myocardial necrosis by TTC staining (12 papillary muscle infarcts, 7 subendocardial infarcts, 7 transmural infarcts). Similar to the short-occlusion and prolonged ischemia groups, RMBF and AUC demonstrated a significant decrease during occlusion (%Δ −67 ± 16 and −81 ± 22 from baseline, respectively), and thickening fraction became severely impaired (−11.8 ± 24%, p < 0.0001 from baseline). After release of the occlusion, RMBF and AUC returned to preocclusion levels. No significant relation was observed between the amount of necrosis (percent area at risk) and perfusion during dobutamine administration (expressed in absolute and relative terms compared with postreperfusion function). Thickening fraction was significantly impaired compared with that at baseline (−7.5 ± 22%, p < 0.0001).
During dobutamine administration, a marked hyperemic response was observed by microspheres (%Δ 97 ± 109) and MCE (%Δ 45 ± 15) compared with occlusion levels (p < 0.0001); however, thickening fraction during dobutamine administration remained severely impaired relative to that at baseline (14 ± 13%, p < 0.003). Two dogs with a papillary muscle infarct and one with a transmural infarct were not included in the analysis of function during dobutamine administration because they had fibrillation after release of occlusion; the defibrillations could have affected ventricular function. Only 3 (12%) of 26 dogs showing an infarct demonstrated recovery of function to baseline levels. No significant relation was observed between the amount of necrosis (percent area at risk) and the contractile reserve with dobutamine (expressed in absolute and relative terms vs. postreperfusion function) (Fig. 6).
2.4 Transmurality of infarcts.
2.4.1 Papillary muscle infarcts.
During coronary occlusion in this subgroup of dogs, RMBF and AUC decreased significantly in the ischemic region compared with that at baseline (%Δ −58 ± 15 and −77 ± 22, respectively). After release of occlusion, RMBF (%Δ 15 ± 52, p < 0.03) and AUC (%Δ 52 ± 92, p < 0.001) increased. During dobutamine administration, both showed an increase from predobutamine levels (%Δ 83 ± 107 and 114 ± 19, p < 0.0001, respectively).
In this subgroup, a decrease in percent thickening fraction was seen during occlusion (p < 0.0001) and persisted throughout the protocol. Although dobutamine administration improved wall thickening fraction, it did not reach baseline levels. Two dogs were not included in the analysis of function during dobutamine administration because of ventricular fibrillations requiring multiple cardioversions.
2.4.2 Subendocardial infarcts.
During occlusion, RMBF and AUC decreased from baseline levels (%Δ −70 ± 170 and −88 ± 15, respectively). After release, an insignificant increase in RMBF (%Δ −7.2 ± 44) and AUC (%Δ −54 ± 50) was observed. During dobutamine infusion, RMBF increased (%Δ 124 ± 107, p < 0.0001 from baseline); however, no increase in AUC was noted (%Δ −0.079 ± 0.844, p = 0.08). Wall thickening fraction remained depressed after release (p = 0.01) but returned to near preocclusion levels during dobutamine (p = 0.49 from baseline).
2.4.3 Transmural infarcts.
During occlusion, RMBF and AUC decreased compared with baseline levels (%Δ −79 ± 8 and −80 ± 27, respectively). RMBF increased after release (%Δ 0.07 ± 52, p = 0.02), further increasing after dobutamine (%Δ 98 ± 113, p = 0.0001). However, AUC did not show a change after release (%Δ −19 ± 87, p = NS) or during dobutamine (%Δ −19.8 ± 64.5, p = NS). A decrease in thickening fraction persisted after release (p = 0.001) and remained abnormal during dobutamine infusion. One dog was excluded from analysis of function during dobutamine because ventricular fibrillation after the occlusion was released.
2.4.4 Comparison between infarct subgroups.
RMBF alone did not allow differentiation among the three infarct subgroups. A comparable decrease in AUC was seen during occlusion in the subgroups (%Δ −58 ± 15 for papillary, −79 ± 8 for transmural and −72 ± 17 for subendocardial infarcts, p = NS), followed by insignificant hyperemia (%Δ 15 ± 52 for papillary, −7 ± 41 for subendocardial and 7 ± 52 for transmural infarcts, p = NS). Although hyperemia was accentuated by dobutamine, no significant differences among the subgroups were observed (%Δ 83 ± 107 for papillary, 124 ± 107 for subendocardial and 98 ± 132 for transmural infarcts).
AUC demonstrated changes similar to RMBF during all interventions only in the papillary muscle infarction group. In contrast, the MCE results observed in the subendocardial and transmural groups showed no noticeable hyperemia by MCE after release of the occlusion or during dobutamine administration, or both. Thus, MCE was able to distinguish between the papillary muscle group and the subendocardial and transmural groups after release of the coronary occlusion.
All subgroups demonstrated a significant decrease in wall thickening during occlusion that persisted after release of the occlusion. Despite this decrease, the papillary muscle infarction subgroup had slightly better wall thickening recovery after release than the other two subgroups (p = 0.04 vs. subendocardial infarcts, p = 0.05 vs. transmural infarcts). Compared with preocclusion wall thickening values, only the subendocardial infarct group demonstrated a near normal wall thickening recovery with dobutamine stimulation.
In 1935, Tennant and Wiggers () demonstrated a strong relation between ischemia and the development of regional wall motion abnormalities. More recent studies have shown ([34–38]) that even relatively short occlusions (<15 min) followed by reperfusion result in prolonged myocardial dysfunction (i.e., myocardial stunning). Characteristically, the stunned myocardium recovers in hours, days or even weeks after the initial ischemic insult (). It is this rather slow recovery of function that could potentially lead to an erroneous conclusion that irreversible myocardial damage has occurred despite adequate reperfusion. To circumvent this problem, one can use techniques that can simultaneously demonstrate the presence of normal or near normal perfusion () and recovery of function during inotropic stimulation ([34, 39–41]).
The present study was performed to critically examine under experimental conditions the ability of MCE and DE, independently and in combination, to determine the amount of myocardium with potential for functional recovery after reperfusion (i.e., absence of necrosis) in a canine model of coronary occlusions of varying duration.
3.1 MCE and DE in stunned myocardium.
MCE has been used to identify stunned myocardium in the post-MI period ([8, 9, 14, 15]). Using MCE in acute anterior MI, Ito et al. () observed the no-reflow phenomenon (i.e., no evidence of perfusion in the area at risk after reperfusion) in 9 patients and restoration of perfusion in the remaining 30. At follow up, the patients showing reflow after reperfusion showed a greater improvement in function than those showing no reflow. Agati et al. () observed similar results in a group of patients studied 1 week post-MI.
Sabia et al. () observed an improvement in function in patients with, but not in those without, collateral flow by MCE. Furthermore, despite prolonged coronary occlusions, when collateral flow is present, patients and animals exhibit marked improvement in function after percutaneous transluminal coronary angioplasty (PTCA) and thrombolysis, respectively ([6, 15]). In the current study, MCE in Group I documented a significant decrease in flow during coronary occlusion, hyperemia after the release of the occlusion and a further increase in flow in response to dobutamine administration, paralleling the flow changes observed with radiolabeled microspheres. Thus, our results confirm the value of MCE in documenting blood flow changes in the classic setting of myocardial stunning.
In Group I, we observed regional dysfunction after restoring adequate flow during reperfusion and improvement in regional function during dobutamine administration. These findings are characteristic of myocardial stunning and in agreement with previous studies in dogs ([24, 32, 34, 35, 38]).
Pierard et al. () studied 17 patients treated with thrombolysis within 3 h of an acute MI. During dobutamine administration, patients found to have normal perfusion and glucose uptake by positron emission tomography showed an improvement in function in the asynergic regions and an improvement in function at follow-up, suggesting that DE can identify viable stunned myocardium. Marzullo et al. () studied 14 patients with wall motion abnormalities at rest after MI. Correct preoperative identification of myocardial viability (i.e., postoperative recovery of function) was achieved with DE in 82% of viable segments and 92% of nonviable segments. Smart et al. () studied 63 patients 2 days after MI. PTCA or coronary artery bypass graft surgery was performed without knowledge of the results of DE. Low dose DE correctly predicted improvement in regional function in 19 of 22 patients. These studies support the use of low dose dobutamine in detecting stunned myocardium early after MI. As previously reported ([6, 7]), we observed a significant improvement in regional function after reflow in Group I only, suggesting a relation between duration of ischemia and degree of postischemic recovery of function.
During coronary occlusion, we observed similar changes in regional function in the short (Group I) and prolonged ischemia (Group II) groups. However, after reperfusion function remained more severely impaired in the prolonged ischemia group. During dobutamine administration, regional function improved in both groups after reperfusion.
Our data agree with previous data ([13, 16–18, 20–22, 42]) suggesting that myocardium subjected to short as well as prolonged ischemia can potentially demonstrate an improvement in function with dobutamine. However, to distinguish between the short and the prolonged ischemia groups (and from necrosis), simultaneous analysis of flow and perhaps metabolism is required. Along these lines, a recent study by Vanoverschelde et al. () suggests that wall motion abnormalities at rest in humans are more likely to be due to repetitive stunning. This phenomenon could potentially explain the positive responses to dobutamine observed in studies where “hibernation” was suspected, but no simultaneous flow data were obtained. Thus, DE appears to be a very useful test for identifying myocardial stunning.
3.2 MCE in prolonged ischemia.
In the prolonged ischemia group, a decrease from baseline in RMBF (58%) and perfusion (69%) was observed during coronary occlusion by radiolabeled microspheres and MCE, respectively. This decrease was of a lesser magnitude than that observed in the infarction and short-occlusion groups, suggesting that the prolonged ischemia group had better collateral vessels and was able to maintain myocardial viability despite prolonged occlusion times. In this respect, the dogs in this group are similar to those studied by Schultz et al. (). After release of the occlusion, flow and perfusion remained below baseline values as opposed to the results observed in Group I, which, in the absence of necrosis, suggests the “low reflow” phenomenon described by Pryzklenk et al. (). Thus, knowledge of flow and perfusion helped separate Groups I and II.
During dobutamine administration, both flow and perfusion increased in Groups I and II. The magnitude of this change was greater in Group I than in Group II and suggests better microvascular reserve with shorter, rather than longer, occlusions; however, this did not reach statistical significance. This finding has also been previously described by Pryzklenk et al. ().
3.3 Dobutamine and MCE in infarctions.
In the present study, a marked decrease in regional function, similar to the short and prolonged ischemia groups, occurred after coronary occlusions in the infarct group. A trend toward worsening regional function was observed as the extent of infarction increased. After release of the occlusion, regional function remained abnormal in all groups of infarcts, and no difference in function was observed relative to Group 2. In contrast, a significant difference between short occlusions and infarction was observed. Dobutamine administration, after release of the occlusion, resulted in a significant improvement in regional function in papillary muscle and subendocardial infarcts but did not allow separation between infarcts as a group from Groups I and II.
Our data suggest that the presence of contractile reserve in itself does not exclude the presence of regional necrosis. This finding has also been reported by Sklenar et al. () in the dog model. In their study, in some dogs with infarcts of considerable sizes (40% to 53%), dobutamine administration at the dose we used in our study (10 μg/kg per min) resulted in an improvement of percent wall thickening when compared with occlusion and reflow values. Although, as a group, dogs with smaller infarcts demonstrate a return of percent wall thickening to preocclusion levels during dobutamine infusion, a significant degree of variability can be observed when the data are analyzed in individual dogs.
The observation that dobutamine administration can elicit a contractile response in infarcted myocardium has been reproduced in the clinical arena. In a recent study, Barilla et al. () administered low dose dobutamine in 21 patients presenting with an acute MI. All but one of the patients enrolled demonstrated an improvement in wall motion score with dobutamine administration. Only two patients with a Q wave MI demonstrated functional improvement in the ischemic zone at 40 ± 15 days of follow-up, despite having demonstrated improvement during dobutamine administration. All but one of the patients who underwent revascularization had functional improvement at follow-up. No improvement in function was seen in medically treated patients. That study exemplifies not only the benefits of revascularization after MI, but it also illustrates the lack of specificity of the contractile response elicited by dobutamine administration in the setting of an MI.
One potential explanation for the improvement of regional function that we observed in some subendocardial and transmural infarcts may reside in our definition of these types of infarcts (see “Assessment of myocardial necrosis”). Islands of normal (or relatively normal) tissue coexisted with necrotic tissue as assessed by TTC staining (Fig. 5). These islands of viable tissue showed perfusion and thus were capable of showing contractile reserve as well. This probably explains the improved specificity observed when the use of perfusion data was added to the data on regional function (Table 4). In addition to this, a tethering effect from the subepicardium or from normal adjacent areas to the ischemic zone, or both, could have led to an improvement in regional function. In contrast, the differentiation of short and prolonged ischemia and infarction improved when a more strict definition of myocardial viability (i.e., return of thickening fraction during dobutamine administration to preocclusion values) was used.
Our data suggest that reactive hyperemia, immediately after reperfusion, limits the use of quantitativedata of myocardial perfusion for the identification of necrosis. In this regard, our data support the qualitativefindings of Ito et al. () in humans. These investigators found that despite the presence of an acute MI, no perfusion abnormalities were visually identified after successful reperfusion in 30 of 39 patients. Similarly, in a previous study of myocardial perfusion patterns from our laboratory () in the same group of animals as those used in the present study, we visually observed perfusion abnormalities early after reperfusion in only 13 of 26 dogs with infarction.
Our results suggest that hyperemia observed in response to dobutamine administration is reduced in asynergic regions subjected to prolonged ischemia and infarction compared with short ischemia. Furthermore, this response worsens as the transmurality of infarction progresses, a finding that could be explained by worsened microvascular reserve with more prolonged and severe ischemia. These results could potentially be used to determine not only the severity of the ischemia, but the amount of myocardium irreversibly injured.
3.4 Combined use of DE and MCE.
The present study suggests that the combination of DE and MCE can identify the causes of myocardial asynergy (i.e., stunning vs. infarction) better than either technique alone. However, the poor specificity of the combination limits their potential clinical applicability.
The effect of an inotropic agent on contractile reserve was found to decrease with longer ischemic insults independent of the severity of underperfusion. The large variability in regional function observed in response to dobutamine administration in Group I dogs has previously been observed by other investigators using other inotropes (). This variability created significant overlap with the response observed in the prolonged ischemia and infarction groups and was the main reason for the failure of dobutamine to effectively differentiate the three groups studied.
3.5 Limitations of the study.
We used a single dose of dobutamine (10 μg/kg per min) in the present study because this dose has been found to be effective in determining the presence of myocardial stunning in animals () and humans ([18, 22]). Afridi () and Smart et al. (), in patients, and Sklenar et al. (), in dogs, found that dobutamine (≤10 μg/kg per min) correctly identified the majority of segments that eventually showed an improvement in function. Although potentially higher doses of dobutamine could have further differentiated the three groups of animals studied, the safety of this approach in the acute MI period needs further study. Furthermore, loss in specificity could have resulted with higher doses of dobutamine.
In the present study, unlike the clinical situation, no coronary stenosis is present on relief of the coronary occlusion. After successful thrombolysis or PTCA, or both, during acute MI in humans, the severity of the stenoses can vary from mild to critical. A severe stenosis could limit the reactive hyperemia and thus potentially result in reduced regional flow in the infarct zone. Thus, our data might apply to those patients in whom the severity of the remaining stenoses does not limit hyperemia.
The definition of absence of necrosis by MCE used in the present study was based on the hypothesis that infarcted segments would exhibit lower flow than viable segments, as suggested by a recent study by Gewirtz et al. (). In that study, 96% of viable regions had regional blood flow ≥0.39 ml/g per min and >75% had flow >0.58 ml/g per min. However, significant overlap of viable and nonviable segments has recently been observed in a study of 14 patients with a previous MI (). Those investigators found that in 45% of segments with wall motion abnormalities at the site of a previous MI, baseline flow was not different from that in noninfarcted segments. A better separation of viable and nonviable segments was observed only after dipyridamole administration, when 84% of the viable segments (but only 20% of the infarcted ones) maintained a normal coronary reserve. Thus, our results are more in agreement with the latter study and suggest that the presence of adequate rest perfusion after MI does not rule out the presence of infarction.
3.6 Clinical implications.
When evaluating regional wall motion abnormalities after reperfusion, knowledge of contractile reserve, as well as perfusion data, is imperative to enhance prediction of absence of necrosis and the etiology of the underlying asynergy. The use of low dose dobutamine by itself can lead to overestimation or underestimation of the presence or absence of necrosis.
We conclude that although the combination of low dose DE and MCE is superior to either technique alone in differentiating necrotic from postischemic myocardium, significant limitations for these techniques remain in this setting. The ability of the combination of DE and MCE to offer more accurate information in predicting functional recovery than either technique alone after revascularization procedures in patients with coronary artery disease remains to be investigated.
☆ This study was presented in part at the 44th Annual Scientific Session of the American College of Cardiology, New Orleans, Louisiana and was performed during the tenure of Clinician-Scientist Award 92004390 from the American Heart Association, Dallas, Texas (Dr. Cheirif) and was supported by a generous educational grant from Molecular Biosystems, Inc., San Diego, California.
- area under the curve
- dobutamine echocardiography
- left anterior descending coronary artery
- left circumflex coronary artery
- myocardial contrast echocardiography (echocardiographic)
- myocardial infarction
- percutaneous transluminal coronary angioplasty
- regional myocardial blood flow
- triphenyltetrazolium chloride
- percent change
- Received May 17, 1996.
- Revision received December 11, 1996.
- Accepted December 20, 1996.
- The American College of Cardiology
- Brunken R,
- Tillisch J,
- Schwaiger M,
- et al.
- Kolibask AJ,
- Goodenow JS,
- Busk CA,
- Teltalman MR,
- Lewis RP
- Chatterjee K,
- Swan JHC,
- Parmley WW,
- Sustaita H,
- Marcus HS,
- Matloff J
- Cheirif J,
- Narkiewicz-Jodko JB,
- Hawkins HK,
- Bravenec JS,
- Quiñones MA,
- Mickelson JK
- Tennant RL,
- Wiggers CJ
- Ito H,
- Tomooka T,
- Sakai N,
- Yu H,
- Higashino Y,
- Fujii K
- Agati L,
- Voci P,
- Bilotta F,
- et al.
- Ito H,
- Iwakura K,
- Oh H,
- et al.
- Sklenar J,
- Ismail S,
- Villanueva FS,
- Goodman C,
- Glasheen WP,
- Kaul S
- Sabia PJ,
- Powers ER,
- Jayaweera AR,
- Ragosta M,
- Kaul S
- Pierard LA,
- De Landsheere CM,
- Berthe C,
- Rigo P,
- Kulbertus HE
- Smart SC,
- Sawada S,
- Ryan T,
- et al.
- Barilla F,
- Gheorghiade M,
- Alam M,
- Khaja F,
- Goldstein S
- Cigarroa CG,
- deFilippi CR,
- Brickner ME,
- Alvarez LG,
- Wait MA,
- Grayburn PA
- La Canna G,
- Alfieri O,
- Giubbini R,
- Gargano M,
- Ferrari R,
- Visioli O
- Afridi I,
- Kleiman NS,
- Raizner AE,
- Zoghbi WA
- Panza JA,
- Dilsizian V,
- Laurienzo JM,
- Curiel RV,
- Katsiyiannis PT
- Cheirif J,
- Zoghbi WA,
- Bolli R,
- O’Neil PG,
- Hoyt BD,
- Quiñones MA
- ↵Desir RM, Cheirif JB, Bolli R, Zoghbi WA, Hoyt BD, Quiñones MA. Assessment of regional myocardial perfusion with MCE in a model of varying degrees of coronary stenosis. Am Heart J 1994;127:56–3.
- Cheirif J,
- Desir RM,
- Bolli R,
- et al.
- Porter TR,
- D’Sa A,
- Turner C,
- et al.
- Porter TR,
- Xie F,
- Kricsfeld A,
- Kilzer K
- Bolli R,
- Zhu WX,
- Thornby JI,
- O’Neill PG,
- Roberts R
- Gewirtz H,
- Fischman AJ,
- Abraham S,
- Gilson M,
- Strauss HW,
- Alpert NM
- Bolli R,
- Myers M,
- Zhu W,
- Roberts R
- Bolli R
- Ross J Jr.
- Myers JH,
- Stirling MC,
- Choy M,
- Buda AJ,
- Gallagher KP
- Ito BR,
- Tate H,
- Kobayashi M,
- Schaper W
- Vanoverschelde JLJ,
- Wijns W,
- Depre C,
- et al.
- Schulz R,
- Guth BD,
- Pieper K,
- Martin C,
- Heusch G
- ↵Kates MA, Meza MF, Barbee RW, et al. Potential clinical implications of abnormal myocardial perfusion patterns immediately following reperfusion in a canine model: a MCE study. Am Heart J. In press.
- Marzullo P,
- Parodi O,
- Sambuceti G,
- et al.