(Circulation. 1996;94:2685-2688.)
© 1996 American Heart Association, Inc.
Articles |
Searching for Hibernating Myocardium
Time to Reevaluate Investigative Strategies?
the Departments of Internal Medicine (A.I.M.), University of Texas–Houston and Cardiac Unit (A.W.), Massachusetts General Hospital, Boston.
Key Words: Editorials • hibernation, myocardial • myocardial viability • imaging
 |
Introduction |
|---|
 Top Introduction What Needs to Be...References |
|---|
He wakes, he lives, 'tis death is dead not he.—Percy Bysshe Shelly
In the early 1980s, development of clinical methods for coronary revascularization (CABG/PTCA) was rapidly followed by presentation of data that indicated that dysfunctional myocardial segments supplied by vessels with reduced or seemingly absent flow could often recover function after flow was restored. The term "hibernating" has been used to describe myocytes that reside in regions that receive blood flow sufficient to support the low energy–requiring functions needed to maintain structural integrity but inadequate to sustain the high-energy requirements of contraction.1 The recognition that such regions exist and that clinical recovery is possible is particularly important in patients with poor ventricular function and has driven research for reliable, noninvasive methods to detect hibernation. A variety of such techniques is available and can be broadly divided into two groups2 3 : (1) radioactive tracers of perfusion that depend on the integrity of the sarcolemmal membrane for myocardial uptake and retention (eg, 201Tl or 82Rb) or on preservation of myocardial metabolism (eg, [18F] fluorodeoxyglucose [FDG]) to indicate viability and (2) stimulants of inotropic reserve, of which the most widely used is low-dose dobutamine stress echocardiography (LDDSE). Thus the nuclear techniques, with which there is greater experience, evaluate structural viability, whereas LDDSE defines viability on the basis of functional recovery. When assessing these techniques, it is important to remember that inherent in the definition of hibernation is ultimate recovery in function. Therefore, it is only after function has been restored that the effects of ischemia can be shown to have been transient and that the dysfunctional myocardium can be confirmed to have been hibernating. LDDSE, which predicts that function can recover, would appear to be a more direct predictor of functional recovery, but there has been no uniform association of either approach with outcome. Hence, the ideal method remains to be defined.
In this issue, Perrone-Filardi et al4 present intriguing data that compare the predictive accuracies of 201Tl single-photon emission computed tomography (SPECT) and LDDSE at rest and at 4 and 24 hours. They studied 40 patients with stable coronary artery disease; 30 (75%) had a history of previous myocardial infarction, and 20 (50%) had single-vessel disease. Estimates of left ventricular function were available in 35 patients with a mean left ventricular ejection fraction of 45±10%. Only 18 patients had an ejection fraction 
45%, and none had a recent history of cardiac failure. The authors found good concordance between the two techniques for evaluation of hypokinetic myocardial segments. Of 105 hypokinetic segments evaluated, 99 (94%) were viable by 201Tl SPECT, and 88 (84%) showed contractile reserve. However, in evaluation of akinetic segments, agreement between the techniques often differed. In 155 akinetic segments, 119 (77%) were viable by 201Tl SPECT, but only 34 (22%) had contractile reserve. A subgroup of patients underwent revascularization (n=18) in which 109 segments were revascularized. The positive predictive accuracy for functional recovery assessed by echocardiography after a mean of 43±39 days was 72% for 201Tl SPECT and 92% for LDDSE. The negative predictive accuracy for functional recovery was 100% and 65% for 201Tl SPECT and LDDSE, respectively.
In their original paper on stress-redistribution-reinjection 201Tl imaging in patients with left ventricular dysfunction, Dilsizian et al5 reported on a subgroup of patients (n=20) who underwent revascularization. Using this technique, these researchers observed positive and negative predictive values of 87% and 100%, respectively, for functional improvement after revascularization. Ohtani et al6 reported similar findings, with positive and negative predictive accuracies of 80% and 72%, respectively. Ragosta et al7 studied 21 patients with coronary artery disease and severely depressed left ventricular function undergoing bypass surgery using rest-redistribution planar 201Tl imaging. They observed positive and negative predictive accuracies of 57% and 77%, respectively, for predicting functional improvement after revascularization. Interestingly, when the authors included only patients who showed improvement of 201Tl uptake after surgery (which suggests adequate revascularization) the positive predictive accuracy increased to 73%. Udelson et al8 reported positive and negative predictive values of 75% and 80%, respectively, for predicting functional improvement with the use of rest-redistribution 201Tl SPECT imaging in 18 patients undergoing revascularization. Therefore the low positive predictive accuracy of 201Tl imaging reported by Perrone-Filardi et al4 is lower than expected but not outside the range of previously reported values. At present, data are limited that compare the performance of LDDSE to that of 201Tl SPECT imaging for evaluation of myocardial viability. Panza et al9 compared stress-redistribution-reinjection 201Tl SPECT imaging with low-dose dobutamine transesophageal echocardiography in patients with chronic coronary artery disease and left ventricular dysfunction. They observed that although concordance was generally good in myocardial segments that were considered nonviable, only 64% of segments considered viable with the use of 201Tl imaging showed contractile improvement with the use of LDDSE. A similar study that compared dobutamine transesophageal echocardiography with FDG positron emission tomography reported positive and negative predictive accuracies of 81% and 97%, respectively.10 Unfortunately, no revascularization data were reported in either of these studies. Arnese et al11 compared LDDSE with dobutamine 201Tl imaging for evaluation of myocardial viability in patients with coronary artery disease and depressed left ventricular dysfunction. They reported positive and negative predictive accuracies for functional improvement after revascularization of 85% and 93%, respectively, for LDDSE. The corresponding positive and negative predictive accuracies for 201Tl imaging were 33% and 94%, respectively. However, details regarding the 201Tl methodology used were suboptimal and probably had a negative effect on the accuracy of the 201Tl results.12 When viewed as a whole, however, these data consistently suggest that 201Tl imaging is more sensitive, whereas LDDSE is more specific for detecting hibernation. The explanation for this discordance between 201Tl imaging and LDDSE in this and other studies seems to be based on a combination of physiological and technical factors and data registration.
There are several potential reasons why myocardium that demonstrates uptake of 201Tl may not show functional improvement after revascularization. A fixed 201Tl perfusion defect of mild-to-moderate severity shown through rest-redistribution imaging may result from either resting ischemia or nontransmural scarring. In the former, revascularization should result in return of function. In the latter, the transmural extent of subendocardial scarring may be sufficient to prevent effective contraction despite viable subepicardial tissue. The existing literature suggests that the predictive value is higher if stress imaging is performed in addition to rest-redistribution imaging.2 If reversibility is detected in this context it implies the presence of an area of viable myocardium with reduced flow reserve, which is likely to be associated with functional improvement after revascularization.
Timing of follow-up assessment of functional recovery after revascularization was extremely variable in the study of Perrone-Filardi et al,4 and data are available in only 18 of 40 patients. The time of echocardiographic follow-up ranged from 20 to 196 days, with a median of 32 days. Some existing data suggest that it may take several months before full morphological, metabolic, and functional recovery is restored to dysfunctional myocardium after revascularization.13 14 15 The positive predictive value of 201Tl imaging may have been higher and the concordance between 201Tl and LDDSE findings closer if follow-up functional data had been obtained at a single time point later after revascularization. The more-dysfunctional, akinetic myocardial segments with absent contractile reserve may have been those areas exhibiting protracted recovery after revascularization.
Adequacy and completeness of revascularization also has a significant effect on the amount of functional recovery observed. Inadequate revascularization may result from poor runoff distal to graft insertion, stenosis at the insertion of the bypass graft, acute graft closure, perioperative infarction, diffuse atherosclerosis, early and late restenosis, endothelial dysfunction, etc. Failure to take into consideration the adequacy of revascularization may have negatively influenced the positive predictive value of 201Tl imaging in the study of Perrone-Filardi et al.4 The importance of this was well illustrated in the study of Ragosta et al7 by the fact that a significant improvement in the positive predictive accuracy (57% versus 73%) of 201Tl imaging for predicting functional improvement was shown when segments with inadequate revascularization were excluded from analysis. Unlike the original study by Perrone-Filardi et al,16 their study included in this issue of Circulation4 contains no data concerning the adequacy of revascularization.
It is likely that a variety of nonphysiological factors plays an important role in the discrepancies between the two techniques, including nuclear imaging artifacts (eg motion artifact and soft-tissue attenuation) and subjective interpretation and suboptimal delineation of endocardium in the echocardiograms. An intrinsic technical limitation of the article by Perrone-Filardi et al4 relates to the receiver-operator curve used for analysis of the 201Tl images. They found a very close linear correlation between 201Tl uptake at 4 hours and functional recovery, r=.95. Although the negative predictive value of rest-redistribution 201Tl imaging was 100%, the positive predictive accuracy was only 72%. This very high negative predictive accuracy suggests that the authors were operating on a suboptimal receiver-operator curve. They used a cutoff value of 50% for 201Tl uptake, which was too low for this study. Udelson et al8 used a value of 60% in their study. If Perrone-Filardi et al4 had chosen an appropriately higher value, they could have optimized the receiver-operator curve for 201Tl imaging and improved the overall predictive accuracy of the technique.
In the current study by Perrone-Filardi et al4 and in virtually all clinical studies of this type, the techniques are compared by use of a segmental construct. Segmental systems are attractive because of their simple concept, can be anatomically defined, and are well suited to visual determinations of function or perfusion. Unfortunately, they are also the least precise systems for such comparisons. In this study, 201Tl and LDDSE are compared by use of a 13-segment construct of the ventricle. Segments were grouped based on myocardial perfusion and function and were analyzed for correspondence. This approach has a number of important limitations. First, the resolution of all segmental constructs is limited by the number of segments used. At best, a 13-segment model gives a resolution of 
8% of the ventricle. As a result, any functional or metabolic value reflects the mean of a large region in a disorder that can have great local variability. Second, although segmental systems are anatomically defined, the relative contribution of a segment may be artificially altered by the construct. For example, anatomically, the apical third of the ventricle normally represents 20% of the ventricular muscle mass. However, in the study by Perrone-Filardi et al,4 the apex is represented by a single segment (8%) and thus is underrepresented in the overall evaluation. Third, once defined, segments cannot account for ventricular expansion or contraction. Thus, while changes in function can be assessed, changes in morphology cannot. Fourth, when segments are analyzed in isolation, all spatial data are lost so that questions such as the effect of tethering cannot be addressed and the relative contribution of individual segments to global function cannot be assessed. Fifth, although segmental systems are conceptually simple, the three-dimensional interaction of segments and tomographic imaging planes is often confused. For example, in the article by Perrone-Filardi et al4 ". . . four standard views of the left ventricle were obtained for each acquisition: parasternal long-axis; short-axis; and apical two and four chamber views and regional systolic dysfunction was defined when a score 
2 was assigned to a myocardial segment in at least two different echocardiographic views." Unfortunately, with the use of these tomographic planes only 7 of the 13 segments can be recorded in two views, and it is unclear how such comparisons can be made. Finally, large segments of fixed arc rarely correspond to the margins of dysfunctional or hypoperfused regions. This results in multiple segments containing both normal and dysfunctional regions, with the result that small rotational misalignments can have major effects on correspondence. In studies based on qualitative assessment, both the degree and distribution of a parameter are subject to individual interpretation. Given these difficulties, a noise floor to the resulting data is to be expected and although not defined as suggested previously this may be nontrivial.
|
What Needs to Be Done in the Future? |
|---|
|
TopIntroductionWhat Needs to Be...References |
|---|
Perrone-Filardi et al4 end by suggesting the need for larger postrevascularization studies designed to compare the accuracy of the two techniques in predicting recovery of function after revascularization in patients with severely impaired systolic function. There is little question that such studies should and will be undertaken. It is unlikely, however, that there will be major differences in the results or illumination of mechanism if the same methods are used. Even if quantified, studies of 201Tl imaging are subject to variation and echocardiographic results are totally subjective. Quantitative methods are needed to compare variables that are both ordered and reproducible. The segmental systems used are subject to all of the limitations cited above, and although efforts at precision in registration may decrease noise, it will be virtually impossible to eliminate. New approaches to registration are required, using either planar mapping or 3D reconstruction, which should provide more precise registration of data sets, define the presence and effects of tethering and relate flow to function, and spatially relate flow to function. Higher resolution must be achieved, because segments encompassing 60° of the ventricular circumference can contain any and all of the components of the ischemic milieu in any proportion and geographic distribution. Additional approaches to define the presence and extent of infarcted muscle within regions need to be used to understand how the mixture of contractile, viable but noncontractile, and dead tissue affects resultant function so that the effects of these variables can be demonstrated rather than simply hypothesized.
What are the possible implications of the study by Perrone-Filardi et al.4 To the clinician, the study suggests that rest-redistribution 201Tl imaging is an excellent tool for identifying viable and nonviable myocardia. However, in this study at least, viability assessed by rest-redistribution 201Tl imaging did not equate with functional recovery; for this, LDDSE appeared to be more accurate, although without the degree of precision one might hope for. A reasonable strategy might be to first perform stress-redistribution and follow if necessary with reinjection 201Tl imaging to ascertain whether regions of viable myocardium exist. If 201Tl imaging does reveal significant amounts of viable myocardium that are amenable to revascularization, then an LDDSE image could be obtained to determine how much of the myocardium that can recover will recover. If the results are similar, then the likelihood of function recovery should be high. If the area of functional reserve is small relative to the extent of 201Tl viability, then one can only hope that the difference represents dobutamine false negatives. In revascularized patients, this represents 7% of all dysfunctional segments but only 6% of akinetic segments in revascularized patients with ejection fractions 45%. In patients without gold standard data, the differences are greater but the significance can only be inferred. Therefore, the study of Perrone-Filardi et al,4 as with most studies in areas of controversy and ignorance, raises more questions than it answers. Clinical research studies should be designed to identify methods that best predict recovery of function and should focus on patients with poor left ventricular function in whom revascularization has important clinical and prognostic implications. The procedures that assess both function and perfusion should be performed both before and after revascularization. Such studies are not common, and the numbers of patients in them typically are small. Unfortunately many of the basic questions raised by this study4 cannot be readily answered by clinical studies. These questions include such fundamental issues as the following: Why do so many metabolically viable aggregates of cells not recover function? Is this primary to the cells or secondary to their surrounding environment? If it is primary to the cells, can it be overcome, and if so, how? If not, are there better ways to identify those cells? Most of these questions must be answered in experimental models of hibernation, and it is essential that such models be developed.
|
Footnotes |
|---|
Reprint requests to A. Iain McGhie, MD, Assistant Professor of Medicine, University of Texas–Houston, 6431 Fannin, MSB 1.228, Houston, TX 77030.(Circulation. 1996;94:2685-2688.)
The opinions expressed in this editorial are not necessarily those of the editors or of the American Heart Association.
|
References |
|---|
|
TopIntroductionWhat Needs to Be...References |
|---|
1. Braunwald E, Rutherford JD. Reversible ischemic left ventricular dysfunction: evidence of the hibernating myocardium. J Am Coll Cardiol.. 1986;8:1467-1470.[Medline] [Order article via Infotrieve]
2.
Dilsizian V, Bonow RO. Current diagnostic techniques of assessing myocardial viability in patients with hibernating and stunned myocardium. Circulation.. 1993;87:1-20.
3.
Cigarroa CG, de Filippi CR, Brickner ME, Alvarez LG, Wait MA, Grayburn PA. Dobutamine stress echocardiography identifies hibernating myocardium and predicts recovery of left ventricular function after coronary revascularization. Circulation.. 1993;88:430-436.
4.
Perrone-Filardi P, Pace L, Prastaro M, Squame F, Betocchi S, Soricelli A, Piscione F, Indolfi C, Crisci T, Salvatore M, Chiariello M. Assessment of myocardial viability in patients with chronic coronary artery disease: rest–4-hour–24-hour 201Tl tomography versus dobutamine echocardiography. Circulation.. 1996;94:2712-2719.
5. Dilsizian V, Rocco TP, Freedman NMT, Leon MB, Bonow RO. Enhanced detection of ischemic but viable myocardium by the reinjection of thallium after stress-redistribution imaging. N Engl J Med.. 1990;323:141-146.[Abstract]
6. Ohtani H, Tamaki N, Yonekura Y, Mohiuddin IH, Hirati K, Ban T, Konishi J. Value of Tl-201 reinjection after SPECT imaging for predicting reversible ischemia after coronary artery bypass grafting. Am J Cardiol.. 1990;66:394-399.[Medline] [Order article via Infotrieve]
7.
Ragosta M, Beller GA, Watson D, Kaul S, Gimple LW. Quantitative planar rest-redistribution Tl-201 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;87:1630-1641.
8.
Udelson JE, Coleman PS, Metherall J, Pandian NG, Gomez AR, Griffith JL, Shea NL, Oates E, Konstam MA. Predicting recovery of severe regional dysfunction: comparison of resting scintigraphy with Tl-201 and Tc-99m sestamibi. Circulation.. 1994;89:2552-2561.
9.
Panza JA, Dilsizian V, Laurienzo JM, Curiel RV, Katsiyiannis PT. Relation between thallium uptake and contractile response to dobutamine: implications regarding myocardial viability in patients with chronic coronary artery disease and left ventricular dysfunction. Circulation.. 1995;91:990-998.
10. Baer FM, Voth E, Deutsch HJ, Schneider CA, Schicha H, Sechtem U. Assessment of viable myocardium by dobutamine transesophageal echocardiography and comparison with fluorine-18 fluorodeoxyglucose positron emission tomography. J Am Coll Cardiol.. 1994;24:343-353.[Abstract]
11.
Arnese M, Cornel JH, Salustri A, Maat APWM, Elhendy A, Reijs AEM, ten Cate FJ, Keane D, Balk AHMM, Roelandt JRTC, Fioretti PM. Prediction of improvement of regional left ventricular function after surgical revascularization: a comparison of low-dose dobutamine echocardiography with Tl-201 single-photon emission tomography. Circulation.. 1995;91:2748-2752.
12.
Dilsizian V, Bonow RO. Differential uptake and apparent thallium washout after thallium reinjection: options regarding early redistribution imaging before reinjection or late redistribution imaging after reinjection. Circulation.. 1992;85:1032-1038.
13. Luu M, Stevenson LW, Brunken RC, Drinkwater DM, Schelbert HR, Tillisch JH. Delayed recovery of revascularized myocardium after referral for cardiac transplantation. Am Heart J.. 1990;119:668-670.[Medline] [Order article via Infotrieve]
14.
Marwick TH, MacIntyre WJ, Lafont A, Nemec JJ, Salcedo EE. Metabolic responses of hibernating and infarcted myocardium to revascularization: a follow-up study of regional perfusion, function, and metabolism. Circulation.. 1992;85:1347-1353.
15. Bashour TT, Mason DT. Myocardial hibernation and embalment. Am Heart J.. 1990;119:706-708. Editorial.[Medline] [Order article via Infotrieve]
16.
Perrone-Filardi P, Pace L, Prastaro M, Piscione F, Betocchi S, Squame F, Vezzuto P, Soricelli A, Indolfi C, Salvatore M, Chiariello M. Dobutamine echocardiography predicts functional improvement of hypoperfused dysfunctional myocardium following revascularization in patients with coronary artery disease. Circulation.. 1995;91:2556-2565.
This article has been cited by other articles:
 |
|
 |
|
  A. Keck, K. Hertting, Y. Schwartz, R. Kitzing, M. Weber, B. Leisner, C. Franke, E. Bahlmann, C. Schneider, T. Twisselmann, et al. Electromechanical mapping for determination of myocardial contractility and viability: A comparison with echocardiography, myocardial single-photon emission computed tomography, and positron emission tomography J. Am. Coll. Cardiol., September 18, 2002; 40(6): 1067 - 1074. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Tawakol, H. A. Skopicki, S. A. Abraham, N. M. Alpert, A. J. Fischman, M. H. Picard, and H. Gewirtz Evidence of reduced resting blood flow in viable myocardial regions with chronic asynergy J. Am. Coll. Cardiol., December 1, 2000; 36(7): 2146 - 2153. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. Holmvang, S. Fry, H. A. Skopicki, S. A. Abraham, N. M. Alpert, A. J. Fischman, M. H. Picard, and H. Gewirtz Relation Between Coronary "Steal" and Contractile Function at Rest in Collateral-Dependent Myocardium of Humans With Ischemic Heart Disease Circulation, May 18, 1999; 99(19): 2510 - 2516. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. Gepstein, A. Goldin, J. Lessick, G. Hayam, S. Shpun, Y. Schwartz, G. Hakim, R. Shofty, A. Turgeman, D. Kirshenbaum, et al. Electromechanical Characterization of Chronic Myocardial Infarction in the Canine Coronary Occlusion Model Circulation, November 10, 1998; 98(19): 2055 - 2064. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. A. Kloner, R. Bolli, E. Marban, L. Reinlib, and E. Braunwald Medical and Cellular Implications of Stunning, Hibernation, and Preconditioning : An NHLBI Workshop Circulation, May 19, 1998; 97(18): 1848 - 1867. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||