- Email: [email protected]
Myocardial viability in patients with chronic coronary artery disease and previous myocardial infarction: Comparison of myocardial contrast echocardiography and myocardial perfusion scintigraphy
Myocardial viability in patients with chronic coronary artery disease and previous myocardial infarction: Comparison of myocardial contrast echocardiography and myocardial perfusion scintigraphy Sarah Vernon, MD, Sanjiv Kaul, MD, *Eric R. Powers, MD, Gustavo Camarano, MD, Lawrence W. Glmple, MD, and Michael Ragosta, MD Charlottesville, Va.
The aim of this study was to compare perfusion patterns on myocardial contrast echocardiography with those on myocardial perfusion scintigraphy for the assessment of myocardial viability in patients with previous myocardial infarction. Accordingly, perfusion scores with the two techniques were compared in 91 ventricular regions in 21 patients with previous {>6 weeks old) myocardial infarction. Complete concordance between the two techniques was found in 63 {69%) regions; 25 (27%) reg!ons were discordant by only 1 grade, and complete discordance (2 grades) was found in only 3 (3%) regions. A kappa statistic of 0.65 indicated good concordance between the two techniques. Although the scores on both techniques demonstrated a relation with the wall motion score, the correlation between the myocardial contrast echocardiography and wall motion scores was closer {r = -0.63 vs r = -0.50, p = 0.05). It is concluded that myocardial contrast echocardiography provides similar information regarding myocardial viability as myocardial perfusion scintigraphy in patients with coronary artery disease and previous myocardial infarction. (Am Heart J 1997;134:835-40.)
The extent of myocardial viability in patients with chronic coronary artery disease, previous myocardial infarction (biD, and reduced left ventricular systolic function has both prognostic and therapeutic significance) Its assessment is therefore important in the clinical treatment of such patients, especially w h e n a revascularization procedure is being considered. Myocardial perfusion scintigraphy (MPS) is an established method for the assessment of myocardial viability in the clinical setting. 2,3 It is usually performed by measuring the uptake of thallium-201 (which requires functional myocardial cell membranes 4) or, less frequently, of 99nrrc-labeled agents such as sestamibi (which requires intact myocardial mitochondrial functionS). Recently, myocardial contrast echocardiography (MCE) with intracoronary injecFrom the Cardiovascular Division, University of Virginia. *Supported in part by a grant from the National Institutes of Health, Bethesda, Maryland IRO1-HL48890), and an Established Investigator Award from the American Heart Association, Dallas, Texas. Presented in part at the Seventh Annual Scientific Session of the American Society of Echocardlography, Chicago, Ill., June10-12, 1996. Submitted Feb. 16, 1997; accepted Aug. 15, 199Z Reprint requests: Michael Ragosta, MD, Cardiovascular Divisbn, Box 158, Medical Center, University of Virginia, Charlottesville, VA 22908. Copyright 9 by Mosby- Yeor Book, Inc. 0002-8703/97/$5.00 + 0 411186009
tion of microbubbles has b e e n demonstrated to be useful for the assessment of myocardial viability in patients with both recent MI6~ and those with p o o r left ventricular systolic function associated with chronic coronary artery disease. 9 The aim of this study was to compare perfusion patterns on MCE with those on MPS for the assessment of myocardial viability in patients with chronic coronary artery disease and previous MI.
Methods Patient population The study was approved by the Human Investigation Committee at the University of Virginia, and all patients gave written informed consent. Patients with previous MI (>6 weeks old) who had a corresponding wall motion abnormality on left ventriculography underwent MCE at the time of diagnostic catheterization. Twenty-one of these patients were also referred by their physicians for MPS within 4 weeks of catheterization. These 21 patients form the basis of this report, and their clinical characteristics are provided in Table I.
MCE We lmve previously described the method of performing MCE in the cardiac catheterization laboratory.1~ In brief, sonicated Renografin-76 (Squibb), which contains 500,000 +200,000 microbubbles of air with a mean diameter of 6 lam,
836
Vernon et al.
Table I. Clinical characteristics of the patient population No.
Age (yr) Male Time from MI (weeks) Indications for catheterization Unstable angina Exertional angina Congestive heart failure Extentof coronary artery disease Single-vessel disease Muhivessel disease
33-75
i Figure I .
.
.
.
.
.
.
.
.
.
21 (median = 63) 61% 41 (median)
kappa=0.65
43% 19% 38% 14% 86%
3O
0ocl
# of regions S'
was injected into the left main (1.5 ml) and right coronary (1.0 ml) arteries during simultaneously performed transthoracic two-dimensional echocardiography in multiple views (mid-papillary muscle short-axis and apical four- and twochamber views).6 In patients with a totally occluded artery, contrast patte~'ns within the occluded bed were noted during injection of We nonoccluded arteries, which result from collateral flow as described by us previously.7
MPS MPS was performed with planar 2~ imaging in 17 patients and '~mTc-sestamibi single photon emission computed tomography in the remaining 4. For 17 patients 5 mCi of z~ was injected during exercise 5 minutes before the initial images were obtained. The delayed images were obtained 2 hours later. Images were obtained in the anterior, 45-degree, and 70-degree left anterior oblique projections and were analyzed with a computer-assisted approach. 2 For four patients 8 mCi of 99mTc-sestamibiwas injected at rest, and images were acquired 1 hour later. Exercise images were acquired later in the day 1 hour after injection of 25 to 30 mCi of 99mTc-sestamibi during peak exercise. Data were processed with Ramp and low-pass filters and back-projection, after which tomographic images were created in the horizontal and vertical long-axis views. Quantitative measurements were then performed in different myocardial regions. Only the delayed planar 2~ and the rest 99mTc-sestamibiimages were analyzed for this study.
Image interpretation The left ventricle was divided into five regions for both forms of imaging: apex and interventricular septum and lateral, inferior, and anterior walls. Each region was evaluated by two observers on MCE and MPS. Similar scores were used for both methods of imaging: 0 = no opacification on MCE and severe defect (<50% of maximal counts) on MPS; 0.5 = partial or patchy opacification (including that seen only in the epicardial rim) on MCE and mild to moderate defect (25% to 50% of maximal counts) on MPS; and 1 = homogeneous opacification on MCE and normal uptake on MPS.6 The interobserver and intraobserver errors for MCE and b,IPS are small and have been previously reported. 7,n42 Each region was
0 vov
good
w
ID
poor
MCE score Concordance between MCE and MPS scores.
also graded for wall motion as follows: 1 = normal, 2 = mild hypokinesia, 3 = severe hypokinesia, 4 = akinesia, and 5 = dyskinesia.13 Our interobserver and intmobserver errors for wall motion score calculation are also small and have been previously reported. 13 The wall motion scores were assigned while the observers were blinded to the MCE and MPS data.
Statistical methods All data were analyzed with RS/1 (Bolt, Bemnek, and Newman). A weighted kappa statistic was used to assess concordance between MCE and MPS. The weight favored fewer and penalized greater differences in scores between the methods. Correlations between scores were performed with Spearman's rank statistic.
Results Ninety-one of the 105 ventricular regions were visualized by both techniques. O f the 14 regions not visualized, 10 were because of inadequate MCE images where the interventricular septum was not seen in 4, the lateral wall in 3, the anterior wall in 2, and the apex in 1. Four regions were not well seen o n MPS and included the interventricular septum in two and the lateral and anterior walls each in one. Fig. 1 illustrates the concordance between the two tests. In the 91 regions visualized by both techniques, complete concordance was noted in 63 (69%) between the two methods; 25 regions (27%) were discordant by only 1 grade, and complete discordance (2 grades) was found in only three (3%) regions. A kappa statistic of 0.65 indicated good concordance between the two techniques.
AmericanHearl JournaT November 1997
Vernon etal.
Figure 3
Figure 2
J
Ae
AO
Apical 2C View no
Apical 4C V i e w no
6
.! IIll
Anterior View
45 LAO View
Example of concordance between MCE (A) and MPS (B) in patient with previous inferior MI. Arrows in A indicate endocardium. Opacification is localized in epicardial rim on MCE, whereas mild perfusion defect involving entire inferior wall is seen on planar 2mTI imaging.
Example of discordance between MCE (A) and MPS {B) in patient with previous anteroapical MI. Clear apical defect (arrows) is seen on MCE, whereas MPS shows normal 2~ uptake in apex. Other views also showed normal 2~ uptake in apex.
Fig. 2 is an example of concordance b e t w e e n the two techniques a n d demonstrates an inferoapical defect in a patient with previous inferior MI. Although planar 2~ shows overall low counts in the inferior wall, the MCE image shows opacification limited only to the epicardial
rim, with n o n e seen in the rest of the myocardium. The inferior wall received a score of 0.5 by both techniques. In 16 regions the MCE score was higher than the score on hIPS, whereas in 12 regions the converse occurred. Most o f the discordance occurred by 1
American Heart Journal Volume 134, Number 5, Parl ]
837
838
Vernon etal.
Discussion
: Asur,4
5
3
r h o = 0.61 p<0.~l
Our study demonstrates that in patients with previous MI, blCE provides similar information reghrding viability as MPS, a more established method. These findings potentially broaden the role of this technique for the assessment of myocardial viability in the cardiac catheterization laboratory.
n=17
n=~
1
~ 1
03
good
0 ~ poor
M C E score B
rho = 0.S p < 0.001
t~
~ o
n=16 n=33
3: n=42 2
_| 1
i
O.5
good
0
_
b- poor
M C E score
Relations between wall motion score on echocardiography (yaxis) and perfusion scores {x-axis) on MCE (A) and MPS (B).
grade. The three regions with complete discordance (>1 grade) included a lateral wall defect seen on MPS but not on MCE and two apical defects noted on blCE but not on MPS. Thus discordance was noted in only 3 of the 21 patients. An example of complete discordance between the two techniques is depicted in Fig. 3 in a patient with a previous anteroapical MI. A large perfusion defect is seen in the apex on MCE, and no such defect is seen in the apex on planar 2~ imaging. Although the 2~ data are shown only in the 45-degree left anterior oblique projection, other projections also did not s h o w a perfusion defect in the apex. Although the scores on both MCE and MPS showed a correlation with wall motion score (Fig. 4), the correlation between the MCE and wall motion scores was closer (p = 0.63 vs p = 0.50, p = 0.05).
Reasons for concordance The delayed 201tl image reflects the amount of 201'I] sequestered in the myocardium. Because 2~ enters the myocytes mostly by w a y of the Na+/K+ pump, 4 its presence within the myocardium indicates that the cell membrane is intact. Similarly, at rest, 99mTc-sestamibi diffuses into the extravascular space and passively enters the myocyte before binding to the negatively charged mitochondrial membrane. 5 Consequently, its retention within the myocyte is dependent on intact mitochondrial function. The relative activity of both tracers therefore provides an approximate estimate of the number of myocytes that are viable within a region. In comparison, microbubbles used for MCE reside entirely within the vascular space. 14 They do not enter the extravascular space, nor are they extracted by myocytes. They do not enter regions where microvessels are absent such as in scar tissue. 6,8 Microbubbles are also not seen in regions with other forms of microvascular disruption, plugging, and obliteration, which is frequently noted in areas with infarction. The high degree of concordance between MCE and MPS should therefore not be surprising and has also been found in studies with venous injections of microbubbles. n blild differences in score (1 grade) noted in our study can be explained on the different methods of image representation (tomographic vs planar). Reasons for discordance Although complete discordfince between the two techniques was exceptional, it could be most dramatic as in the example shown in Fig. 3. There are several possible reasons for this discordance. Although no clinically documented events occurred between cardiac catheterization and hIPS, there is a small chance that a new blI could have occurred. The likelihood of an artifact on MCE is small, particularly if it encompasses the entire apex. It is possible that the microvasculature within the apex was so sparse as not to be detected on MCE. The low level of myocardial perfusion may, however, still have allowed accumulation of 2~ over several hours within viable myocardium. Thus unless very low levels of flow are detectable on MCE, an
American Heart Journal November 1997
Vernonelal. 839
extractable tracer may have an advantage in terms of assessing myocardial viability. It is likely, however, that more sensitive methods of detecting the presence of microbubbles in tissue such as intermittent harmonic imaging will make it possible to measure low levels of flow o n M C E . 11 Microbubbles are destroyed on ultrasound exposure. 15 During the process of destruction, they produce an "acoustic noise," which contains many frequencies including the frequency to which they were exposed (fundamental frequency). Selective acquisition of the nonfundamental frequencies (including harmonics of the fundamental frequency) results in increased signal-to-noise ratio and better imaging of the microbubbles. 15,16 The signal-to-noise ratio is further increased when microbubble destruction is minimized by obtaining images intermittently rather than continuously "i15,16 Experimental data from our laboratory indicate that this method can be used to measure very low levels of flow. 17
AdvantagesOf contrast echocardiography In the cardiac catheterization laboratory MCE offers a practical advantage over other techniques in that it can provide more immediate viability assessment, which can then be used to guide clinical decision making and obviate the need for a noninvasive viability test done on another day, which can increase hospital stay and cost. Because of its superior spatial resolution, blCE can define the transmural location of viable tissue, which can have prognostic implications. 1 For instance, in Fig. 2 both techniques demonstrated partial viability. However, on blCE it is clear that opacification is limited only in the epicardial rim. Because wall thickening is dependent on endocardial viability,1 this patient is unlikely to demonstrate improvement in regional function after revascularization. The 2~ image does not provide this potentially important information because it cannot distinguish between reduced perfusion caused by low flow from that caused by partial MI.
Limitations of the study Except for four patients, the MPS data were not tomographic. Data registration between the two techniques was therefore not optimal. It is for this reason that we reduced the data to regions of the left ventricular myocardium instead of comparing segments to each other. This approach, however, did not detract from the way we would have read each study clinically. Despite the use of two different isotopes, no differences were seen in the results, although the number of patients studied was small.
American Heart Journal Volume 134, Number 5, Part }
We did not perfom~ a revascularization procedure to determine which method predicts recovery in function. Revascularization was not clinically indicated in many of these patients, and follow-up studies were not performed in those who received such a procedure. We and others have demonstrated the predictive value of both techniques for improvement of regional function after revascularization in similar kinds of patients. 2,3,9 This is the first study, however, that compares the results of the two techniques in the same patients. This study was performed in the cardiac catheterization laboratory with selective coronary injections. Recent experimental data indicate that similar results can be obtained with aortic root injections when intermittent harmonic imaging is used, 18 which will make the technique easier and simpler to perform.
Conclusions MCE provides similar information as MPS for the putative assessment of myocardial viability in patients with coronary artery disease and old MI. \X~ and others have previously demonstrated similar results in patients with recent blI.68 These findings potentially broaden the role of this technique for the assessment of myocardial viability in the cardiac catheterization laboratory.
References 1. Kaul S. There may be more to myocardial viability than meets the eye! Circulation 1995;92:2790-3. 2. Ragosta M, Beller GA, Watson DD, Kaul S, Gimple I.W. Quantitative planar rest-redistribution2~ imaging in detection of myocardial viability and prediction of improvement in left ventricular function after coronary bypass surgery in patients with severely depressed left ventricular function. Circulation 1993;87:1630-41. 3. Udelson ]E, Coleman PS, Melherall ], Pandian NG, Gomez AR, Griffith JL, el al. Predicting recovery of severe regional ventricular dysfunction: comparison of resting scintigraphy with 2~ and 99mTc-sestamibi. Circulation 1994;89:2552-61. 4. Goldhaber SZ, Newell JB, Ingwall JS, Pohosl GM, AIpert NM, Fossel El'. Effectsof reduced coronary flow on thallium-201 accumulation and release in an in vitro rat heart preparation. Am J Cardiol 1983;51:891-6. 5. Piwnica-Worms D, Kronauge JF, Chiu ML Uptake and retention of hexakis (2 methoxyisobutyl isonitrile) technetium (I) in cultured chick myocardial cells. Mitochondrial and plasma membrane potential dependance. Circulation 1990;82:1826-38. 6. Ragosta M, Camarano GP, Kaul S, Powers E, Sarembock U, Gimple LW. Microvascular integrity indicates myocellular viability in patients with recent myocardial infarction: new insights using myocardial contrast echocardiography. Circulation 1994;14;89:2562-9. Z Sabia PJ, Powers ER, Ragosta M, Sarembock IJ, Burwell tR, Kaul S. An association belween collateral blood flow and myocardial viability in patients with recent myocardial infarction. N Engl J Med 1992;12;372:1825-31. 8. Ito H, Tomooka 1",Sakai N, Yu H, Higashino Y, Fujii K, el al. lack of
840
Vernon et al.
myocardial perfusion immediately after successful thromboIysis:a predictor of poor recovery of left ventricular function in anterior myocardial infarction. Circulation 1992;85:1699-705. 9. deFilippi CR, Willett DWL, Irani WN, Eichhorn EJ,Velasco CE, Grayburn PA. Comparison of myocardial contrast echocardiography and low
ment using two-dimensional echocordiography. J Am Coil Cordial 1989;13:1506-13. 14. Keller MW, Segal SS, Koul S, Duling B. The behavior of sonicated albumin mlcrobubbles within the microcirculation:a basis for their use during myocardial contrast echocardiography. Circ Res 1989;65:458-67 15. Wei K, Skyba DM, FirschkeC, LindnerJR,Jayaweera AR, Kaul S. Interaction between microbubbtes and ultrasound. In vitro and in viva observatiot!s. J Am Coil Cardio11997;29:1081-8. 16. FirschkeC, I.indnerJR,Wei K, Skyba D, Goodman NC, Kaul S. Myocardial perfusion imaging in the setting of coronary artery stenosis and acute myocardial infarction using venous iniection of FS-069, a second generation echocardiographic contrast agent. Circulation 1997;96:95967. 1Z Wei K, Firoozan S, Jayaweera AR, I.inka A, Skyba DM, Koul S. Ultrasound-induced microbubble destruction during venous infusion: a novel method for the quantification of myocardial blood flow with echocardiography. Circulation 1997. In press. 18. FirschkeC, I.indnerJR, Goodman NC, Skyba DM, Wei K, Kaul S. Myocardial contrast echocardiography in acute myocardial infarction using aortic root injections of microbubbles: potential application in the cardiac catheterization laboratory. J Am Call Cardio11997;29:207-16.
American Heart.loumal November 1997
The aim of this study was to compare perfusion patterns on myocardial contrast echocardiography with those on myocardial perfusion scintigraphy for the assessment of myocardial viability in patients with previous myocardial infarction. Accordingly, perfusion scores with the two techniques were compared in 91 ventricular regions in 21 patients with previous {>6 weeks old) myocardial infarction. Complete concordance between the two techniques was found in 63 {69%) regions; 25 (27%) reg!ons were discordant by only 1 grade, and complete discordance (2 grades) was found in only 3 (3%) regions. A kappa statistic of 0.65 indicated good concordance between the two techniques. Although the scores on both techniques demonstrated a relation with the wall motion score, the correlation between the myocardial contrast echocardiography and wall motion scores was closer {r = -0.63 vs r = -0.50, p = 0.05). It is concluded that myocardial contrast echocardiography provides similar information regarding myocardial viability as myocardial perfusion scintigraphy in patients with coronary artery disease and previous myocardial infarction. (Am Heart J 1997;134:835-40.)
The extent of myocardial viability in patients with chronic coronary artery disease, previous myocardial infarction (biD, and reduced left ventricular systolic function has both prognostic and therapeutic significance) Its assessment is therefore important in the clinical treatment of such patients, especially w h e n a revascularization procedure is being considered. Myocardial perfusion scintigraphy (MPS) is an established method for the assessment of myocardial viability in the clinical setting. 2,3 It is usually performed by measuring the uptake of thallium-201 (which requires functional myocardial cell membranes 4) or, less frequently, of 99nrrc-labeled agents such as sestamibi (which requires intact myocardial mitochondrial functionS). Recently, myocardial contrast echocardiography (MCE) with intracoronary injecFrom the Cardiovascular Division, University of Virginia. *Supported in part by a grant from the National Institutes of Health, Bethesda, Maryland IRO1-HL48890), and an Established Investigator Award from the American Heart Association, Dallas, Texas. Presented in part at the Seventh Annual Scientific Session of the American Society of Echocardlography, Chicago, Ill., June10-12, 1996. Submitted Feb. 16, 1997; accepted Aug. 15, 199Z Reprint requests: Michael Ragosta, MD, Cardiovascular Divisbn, Box 158, Medical Center, University of Virginia, Charlottesville, VA 22908. Copyright 9 by Mosby- Yeor Book, Inc. 0002-8703/97/$5.00 + 0 411186009
tion of microbubbles has b e e n demonstrated to be useful for the assessment of myocardial viability in patients with both recent MI6~ and those with p o o r left ventricular systolic function associated with chronic coronary artery disease. 9 The aim of this study was to compare perfusion patterns on MCE with those on MPS for the assessment of myocardial viability in patients with chronic coronary artery disease and previous MI.
Methods Patient population The study was approved by the Human Investigation Committee at the University of Virginia, and all patients gave written informed consent. Patients with previous MI (>6 weeks old) who had a corresponding wall motion abnormality on left ventriculography underwent MCE at the time of diagnostic catheterization. Twenty-one of these patients were also referred by their physicians for MPS within 4 weeks of catheterization. These 21 patients form the basis of this report, and their clinical characteristics are provided in Table I.
MCE We lmve previously described the method of performing MCE in the cardiac catheterization laboratory.1~ In brief, sonicated Renografin-76 (Squibb), which contains 500,000 +200,000 microbubbles of air with a mean diameter of 6 lam,
836
Vernon et al.
Table I. Clinical characteristics of the patient population No.
Age (yr) Male Time from MI (weeks) Indications for catheterization Unstable angina Exertional angina Congestive heart failure Extentof coronary artery disease Single-vessel disease Muhivessel disease
33-75
i Figure I .
.
.
.
.
.
.
.
.
.
21 (median = 63) 61% 41 (median)
kappa=0.65
43% 19% 38% 14% 86%
3O
0ocl
# of regions S'
was injected into the left main (1.5 ml) and right coronary (1.0 ml) arteries during simultaneously performed transthoracic two-dimensional echocardiography in multiple views (mid-papillary muscle short-axis and apical four- and twochamber views).6 In patients with a totally occluded artery, contrast patte~'ns within the occluded bed were noted during injection of We nonoccluded arteries, which result from collateral flow as described by us previously.7
MPS MPS was performed with planar 2~ imaging in 17 patients and '~mTc-sestamibi single photon emission computed tomography in the remaining 4. For 17 patients 5 mCi of z~ was injected during exercise 5 minutes before the initial images were obtained. The delayed images were obtained 2 hours later. Images were obtained in the anterior, 45-degree, and 70-degree left anterior oblique projections and were analyzed with a computer-assisted approach. 2 For four patients 8 mCi of 99mTc-sestamibiwas injected at rest, and images were acquired 1 hour later. Exercise images were acquired later in the day 1 hour after injection of 25 to 30 mCi of 99mTc-sestamibi during peak exercise. Data were processed with Ramp and low-pass filters and back-projection, after which tomographic images were created in the horizontal and vertical long-axis views. Quantitative measurements were then performed in different myocardial regions. Only the delayed planar 2~ and the rest 99mTc-sestamibiimages were analyzed for this study.
Image interpretation The left ventricle was divided into five regions for both forms of imaging: apex and interventricular septum and lateral, inferior, and anterior walls. Each region was evaluated by two observers on MCE and MPS. Similar scores were used for both methods of imaging: 0 = no opacification on MCE and severe defect (<50% of maximal counts) on MPS; 0.5 = partial or patchy opacification (including that seen only in the epicardial rim) on MCE and mild to moderate defect (25% to 50% of maximal counts) on MPS; and 1 = homogeneous opacification on MCE and normal uptake on MPS.6 The interobserver and intraobserver errors for MCE and b,IPS are small and have been previously reported. 7,n42 Each region was
0 vov
good
w
ID
poor
MCE score Concordance between MCE and MPS scores.
also graded for wall motion as follows: 1 = normal, 2 = mild hypokinesia, 3 = severe hypokinesia, 4 = akinesia, and 5 = dyskinesia.13 Our interobserver and intmobserver errors for wall motion score calculation are also small and have been previously reported. 13 The wall motion scores were assigned while the observers were blinded to the MCE and MPS data.
Statistical methods All data were analyzed with RS/1 (Bolt, Bemnek, and Newman). A weighted kappa statistic was used to assess concordance between MCE and MPS. The weight favored fewer and penalized greater differences in scores between the methods. Correlations between scores were performed with Spearman's rank statistic.
Results Ninety-one of the 105 ventricular regions were visualized by both techniques. O f the 14 regions not visualized, 10 were because of inadequate MCE images where the interventricular septum was not seen in 4, the lateral wall in 3, the anterior wall in 2, and the apex in 1. Four regions were not well seen o n MPS and included the interventricular septum in two and the lateral and anterior walls each in one. Fig. 1 illustrates the concordance between the two tests. In the 91 regions visualized by both techniques, complete concordance was noted in 63 (69%) between the two methods; 25 regions (27%) were discordant by only 1 grade, and complete discordance (2 grades) was found in only three (3%) regions. A kappa statistic of 0.65 indicated good concordance between the two techniques.
AmericanHearl JournaT November 1997
Vernon etal.
Figure 3
Figure 2
J
Ae
AO
Apical 2C View no
Apical 4C V i e w no
6
.! IIll
Anterior View
45 LAO View
Example of concordance between MCE (A) and MPS (B) in patient with previous inferior MI. Arrows in A indicate endocardium. Opacification is localized in epicardial rim on MCE, whereas mild perfusion defect involving entire inferior wall is seen on planar 2mTI imaging.
Example of discordance between MCE (A) and MPS {B) in patient with previous anteroapical MI. Clear apical defect (arrows) is seen on MCE, whereas MPS shows normal 2~ uptake in apex. Other views also showed normal 2~ uptake in apex.
Fig. 2 is an example of concordance b e t w e e n the two techniques a n d demonstrates an inferoapical defect in a patient with previous inferior MI. Although planar 2~ shows overall low counts in the inferior wall, the MCE image shows opacification limited only to the epicardial
rim, with n o n e seen in the rest of the myocardium. The inferior wall received a score of 0.5 by both techniques. In 16 regions the MCE score was higher than the score on hIPS, whereas in 12 regions the converse occurred. Most o f the discordance occurred by 1
American Heart Journal Volume 134, Number 5, Parl ]
837
838
Vernon etal.
Discussion
: Asur,4
5
3
r h o = 0.61 p<0.~l
Our study demonstrates that in patients with previous MI, blCE provides similar information reghrding viability as MPS, a more established method. These findings potentially broaden the role of this technique for the assessment of myocardial viability in the cardiac catheterization laboratory.
n=17
n=~
1
~ 1
03
good
0 ~ poor
M C E score B
rho = 0.S p < 0.001
t~
~ o
n=16 n=33
3: n=42 2
_| 1
i
O.5
good
0
_
b- poor
M C E score
Relations between wall motion score on echocardiography (yaxis) and perfusion scores {x-axis) on MCE (A) and MPS (B).
grade. The three regions with complete discordance (>1 grade) included a lateral wall defect seen on MPS but not on MCE and two apical defects noted on blCE but not on MPS. Thus discordance was noted in only 3 of the 21 patients. An example of complete discordance between the two techniques is depicted in Fig. 3 in a patient with a previous anteroapical MI. A large perfusion defect is seen in the apex on MCE, and no such defect is seen in the apex on planar 2~ imaging. Although the 2~ data are shown only in the 45-degree left anterior oblique projection, other projections also did not s h o w a perfusion defect in the apex. Although the scores on both MCE and MPS showed a correlation with wall motion score (Fig. 4), the correlation between the MCE and wall motion scores was closer (p = 0.63 vs p = 0.50, p = 0.05).
Reasons for concordance The delayed 201tl image reflects the amount of 201'I] sequestered in the myocardium. Because 2~ enters the myocytes mostly by w a y of the Na+/K+ pump, 4 its presence within the myocardium indicates that the cell membrane is intact. Similarly, at rest, 99mTc-sestamibi diffuses into the extravascular space and passively enters the myocyte before binding to the negatively charged mitochondrial membrane. 5 Consequently, its retention within the myocyte is dependent on intact mitochondrial function. The relative activity of both tracers therefore provides an approximate estimate of the number of myocytes that are viable within a region. In comparison, microbubbles used for MCE reside entirely within the vascular space. 14 They do not enter the extravascular space, nor are they extracted by myocytes. They do not enter regions where microvessels are absent such as in scar tissue. 6,8 Microbubbles are also not seen in regions with other forms of microvascular disruption, plugging, and obliteration, which is frequently noted in areas with infarction. The high degree of concordance between MCE and MPS should therefore not be surprising and has also been found in studies with venous injections of microbubbles. n blild differences in score (1 grade) noted in our study can be explained on the different methods of image representation (tomographic vs planar). Reasons for discordance Although complete discordfince between the two techniques was exceptional, it could be most dramatic as in the example shown in Fig. 3. There are several possible reasons for this discordance. Although no clinically documented events occurred between cardiac catheterization and hIPS, there is a small chance that a new blI could have occurred. The likelihood of an artifact on MCE is small, particularly if it encompasses the entire apex. It is possible that the microvasculature within the apex was so sparse as not to be detected on MCE. The low level of myocardial perfusion may, however, still have allowed accumulation of 2~ over several hours within viable myocardium. Thus unless very low levels of flow are detectable on MCE, an
American Heart Journal November 1997
Vernonelal. 839
extractable tracer may have an advantage in terms of assessing myocardial viability. It is likely, however, that more sensitive methods of detecting the presence of microbubbles in tissue such as intermittent harmonic imaging will make it possible to measure low levels of flow o n M C E . 11 Microbubbles are destroyed on ultrasound exposure. 15 During the process of destruction, they produce an "acoustic noise," which contains many frequencies including the frequency to which they were exposed (fundamental frequency). Selective acquisition of the nonfundamental frequencies (including harmonics of the fundamental frequency) results in increased signal-to-noise ratio and better imaging of the microbubbles. 15,16 The signal-to-noise ratio is further increased when microbubble destruction is minimized by obtaining images intermittently rather than continuously "i15,16 Experimental data from our laboratory indicate that this method can be used to measure very low levels of flow. 17
AdvantagesOf contrast echocardiography In the cardiac catheterization laboratory MCE offers a practical advantage over other techniques in that it can provide more immediate viability assessment, which can then be used to guide clinical decision making and obviate the need for a noninvasive viability test done on another day, which can increase hospital stay and cost. Because of its superior spatial resolution, blCE can define the transmural location of viable tissue, which can have prognostic implications. 1 For instance, in Fig. 2 both techniques demonstrated partial viability. However, on blCE it is clear that opacification is limited only in the epicardial rim. Because wall thickening is dependent on endocardial viability,1 this patient is unlikely to demonstrate improvement in regional function after revascularization. The 2~ image does not provide this potentially important information because it cannot distinguish between reduced perfusion caused by low flow from that caused by partial MI.
Limitations of the study Except for four patients, the MPS data were not tomographic. Data registration between the two techniques was therefore not optimal. It is for this reason that we reduced the data to regions of the left ventricular myocardium instead of comparing segments to each other. This approach, however, did not detract from the way we would have read each study clinically. Despite the use of two different isotopes, no differences were seen in the results, although the number of patients studied was small.
American Heart Journal Volume 134, Number 5, Part }
We did not perfom~ a revascularization procedure to determine which method predicts recovery in function. Revascularization was not clinically indicated in many of these patients, and follow-up studies were not performed in those who received such a procedure. We and others have demonstrated the predictive value of both techniques for improvement of regional function after revascularization in similar kinds of patients. 2,3,9 This is the first study, however, that compares the results of the two techniques in the same patients. This study was performed in the cardiac catheterization laboratory with selective coronary injections. Recent experimental data indicate that similar results can be obtained with aortic root injections when intermittent harmonic imaging is used, 18 which will make the technique easier and simpler to perform.
Conclusions MCE provides similar information as MPS for the putative assessment of myocardial viability in patients with coronary artery disease and old MI. \X~ and others have previously demonstrated similar results in patients with recent blI.68 These findings potentially broaden the role of this technique for the assessment of myocardial viability in the cardiac catheterization laboratory.
References 1. Kaul S. There may be more to myocardial viability than meets the eye! Circulation 1995;92:2790-3. 2. Ragosta M, Beller GA, Watson DD, Kaul S, Gimple I.W. Quantitative planar rest-redistribution2~ imaging in detection of myocardial viability and prediction of improvement in left ventricular function after coronary bypass surgery in patients with severely depressed left ventricular function. Circulation 1993;87:1630-41. 3. Udelson ]E, Coleman PS, Melherall ], Pandian NG, Gomez AR, Griffith JL, el al. Predicting recovery of severe regional ventricular dysfunction: comparison of resting scintigraphy with 2~ and 99mTc-sestamibi. Circulation 1994;89:2552-61. 4. Goldhaber SZ, Newell JB, Ingwall JS, Pohosl GM, AIpert NM, Fossel El'. Effectsof reduced coronary flow on thallium-201 accumulation and release in an in vitro rat heart preparation. Am J Cardiol 1983;51:891-6. 5. Piwnica-Worms D, Kronauge JF, Chiu ML Uptake and retention of hexakis (2 methoxyisobutyl isonitrile) technetium (I) in cultured chick myocardial cells. Mitochondrial and plasma membrane potential dependance. Circulation 1990;82:1826-38. 6. Ragosta M, Camarano GP, Kaul S, Powers E, Sarembock U, Gimple LW. Microvascular integrity indicates myocellular viability in patients with recent myocardial infarction: new insights using myocardial contrast echocardiography. Circulation 1994;14;89:2562-9. Z Sabia PJ, Powers ER, Ragosta M, Sarembock IJ, Burwell tR, Kaul S. An association belween collateral blood flow and myocardial viability in patients with recent myocardial infarction. N Engl J Med 1992;12;372:1825-31. 8. Ito H, Tomooka 1",Sakai N, Yu H, Higashino Y, Fujii K, el al. lack of
840
Vernon et al.
myocardial perfusion immediately after successful thromboIysis:a predictor of poor recovery of left ventricular function in anterior myocardial infarction. Circulation 1992;85:1699-705. 9. deFilippi CR, Willett DWL, Irani WN, Eichhorn EJ,Velasco CE, Grayburn PA. Comparison of myocardial contrast echocardiography and low
ment using two-dimensional echocordiography. J Am Coil Cordial 1989;13:1506-13. 14. Keller MW, Segal SS, Koul S, Duling B. The behavior of sonicated albumin mlcrobubbles within the microcirculation:a basis for their use during myocardial contrast echocardiography. Circ Res 1989;65:458-67 15. Wei K, Skyba DM, FirschkeC, LindnerJR,Jayaweera AR, Kaul S. Interaction between microbubbtes and ultrasound. In vitro and in viva observatiot!s. J Am Coil Cardio11997;29:1081-8. 16. FirschkeC, I.indnerJR,Wei K, Skyba D, Goodman NC, Kaul S. Myocardial perfusion imaging in the setting of coronary artery stenosis and acute myocardial infarction using venous iniection of FS-069, a second generation echocardiographic contrast agent. Circulation 1997;96:95967. 1Z Wei K, Firoozan S, Jayaweera AR, I.inka A, Skyba DM, Koul S. Ultrasound-induced microbubble destruction during venous infusion: a novel method for the quantification of myocardial blood flow with echocardiography. Circulation 1997. In press. 18. FirschkeC, I.indnerJR, Goodman NC, Skyba DM, Wei K, Kaul S. Myocardial contrast echocardiography in acute myocardial infarction using aortic root injections of microbubbles: potential application in the cardiac catheterization laboratory. J Am Call Cardio11997;29:207-16.
American Heart.loumal November 1997