Identification of Viable Myocardium
1996; Lippincott Williams & Wilkins; Volume: 94; Issue: 11 Linguagem: Inglês
10.1161/01.cir.94.11.2674
ISSN1524-4539
Autores Tópico(s)Cardiac electrophysiology and arrhythmias
ResumoHomeCirculationVol. 94, No. 11Identification of Viable Myocardium Free AccessResearch ArticleDownload EPUBAboutView EPUBSections ToolsAdd to favoritesDownload citationsTrack citationsPermissions ShareShare onFacebookTwitterLinked InMendeleyReddit Jump toFree AccessResearch ArticleDownload EPUBIdentification of Viable Myocardium Robert O. Bonow Robert O. BonowRobert O. Bonow the Division of Cardiology, Northwestern University Medical School, Chicago, Ill. Originally published1 Dec 1996https://doi.org/10.1161/01.CIR.94.11.2674Circulation. 1996;94:2674–2680In recent years, diagnostic testing to evaluate the presence and extent of viable but dysfunctional myocardium has become an important component of the clinical assessment of patients with chronic CAD and LV dysfunction. It is well established that impaired LV function in such patients is not always an irreversible process related to previous myocardial infarction, because LV function may improve considerably after myocardial revascularization procedures.123456The mechanism for this improvement in systolic function remains a matter of uncertainty and debate because the underlying processes responsible for reversible contractile dysfunction are often difficult to ascertain in patients, and the development of animal models of chronic reversible dysfunction has been disappointing to date. Restoration of blood flow to chronically underperfused myocardium may lead to the functional recovery of hibernating myocardium,135 whereas revascularization of myocardium with adequate perfusion at rest but with recurrent ischemic episodes during stress may successfully reverse persistent contractile dysfunction caused by repetitive stunning.78910 Although the terms "hibernation" and "stunning" represent uniquely different pathophysiological processes with distinct definitions, in clinical circumstances the boundaries between stunning and hibernation are often indistinct. It is likely that both hibernation and repetitive stunning do occur clinically and contribute to ischemic LV dysfunction. Moreover, both processes may occur in the same patient and even coexist in the same myocardial region. Given the critical balance between reduced perfusion, reduced function, and reduced coronary flow reserve in the hibernating myocardium, some myocardial regions that are hibernating at rest may develop ischemia during exercise with a subsequent process of postischemic stunning superimposed on the baseline hibernating state.Thus, the clarity with which chronic LV dysfunction can be defined as hibernation, repetitive stunning, or a combination of the two processes is limited. On the other hand, regardless of the definition, it is clear from multiple clinical series that there is an important subset of patients with chronic CAD and LV dysfunction who manifest substantial improvement in LV function after myocardial revascularization. It has been estimated that between 25% and 40% of patients with chronic CAD and global LV dysfunction have the potential for significant improvement in LV ejection fraction after revascularization.6111213 The differentiation of viable from nonviable myocardium is a relevant diagnostic issue in patients being considered for coronary revascularization, because these procedures are often accompanied by high operative morbidity and mortality in patients with LV dysfunction, many of whom have already undergone a previous bypass operation. However, this is the same population that ultimately may benefit the most from revascularization.Several noninvasive imaging methods have evolved during the past decade to identify physiological markers of myocardial viability in regions with contractile dysfunction.141516 These include, among others, PET imaging to assess myocardial metabolic activity, 201Tl imaging to assess myocardial perfusion and membrane integrity, and dobutamine echocardiography to assess myocardial contractile reserve. In the absence of definitive trials comparing all three methods in a large series of patients undergoing revascularization, uncertainty persists regarding the relative accuracies of each method in predicting recovery of LV function and whether some patient subsets are better evaluated by a particular test. However, there is a rapidly expanding literature addressing the role of these methods in identifying viable myocardium. The study reported by Perrone-Filardi et al17 in this issue of Circulation contributes importantly to this body of data and provides new insights into the relative strengths and limitations of thallium SPECT imaging and dobutamine echocardiography for predicting functional recovery after myocardial revascularization.In the 1980s, at a time when the limitations of standard thallium imaging and two-dimensional echocardiography for viability assessment were quite apparent, PET became established as an exceptional method for demonstrating viable myocardium in patients with impaired LV function, by demonstrating preserved metabolic activity in regions with contractile dysfunction.18192021 The most extensive experience thus far with PET has been achieved using 18F-fluorodeoxyglucose as a marker of myocardial glucose utilization. On the basis of the cumulative experience provided by six studies of patients with LV dysfunction undergoing myocardial revascularization,192122232425 this metabolic marker has a positive predictive accuracy of 82% and a negative predictive accuracy of 83% for predicting recovery of regional function after revascularization (Figure).The requirements for cellular viability, in addition to metabolic processes to generate high-energy phosphates, include intact sarcolemmal function to maintain electrochemical gradients across the cell membrane and adequate blood flow to deliver substrates and wash out metabolites. Because the retention of 201Tl is an active process that is a function of cell membrane integrity and cell viability as well as blood flow, 201Tl in theory should be taken up and retained by myocardial regions that also extract and retain fluorodeoxyglucose and other metabolic tracers. Two thallium imaging methods, one using a stress-redistribution-reinjection protocol262728 and the other a rest-redistribution protocol,2930 have been explored in the past few years with promising results.To date, nine studies of stress-redistribution-reinjection SPECT imaging,263132333435363738 involving a total of 295 patients, have demonstrated a cumulative positive and negative predictive accuracy of 69% and 89%, respectively, regarding improvement in regional function after revascularization. Including the current study by Perrone-Filardi et al, four studies of rest-redistribution SPECT imaging,17394041 involving a total of 83 patients, have shown comparable positive and negative predictive accuracies of 69% and 92%, respectively. The similarity of these two thallium protocols in predicting functional recovery is not surprising given the excellent concordance in regional thallium activity between these two thallium methods when studied in the same patients.30 Considering this concordance, the overall results of these two methods can be combined, as indicated in the Figure. This experience indicates that thallium SPECT imaging yields a higher negative predictive value, owing to its higher sensitivity, and a lower positive predictive value, owing to its lower specificity, compared with metabolic PET imaging.Twelve of the above 13 studies analyzed the thallium data quantitatively on the basis of regional thallium activity, and the criterion for viability in most cases was based on a threshold level of thallium activity, such that thallium levels >50% or 60% of the activity in normal myocardial segments were considered viable. This in essence treats the thallium data as a binary function that provides positive or negative information. One of the greatest strengths of thallium imaging for viability assessment is that the level of regional tracer activity can be approached as a continuum rather than in a binary manner. It has been shown that there is an inverse relationship between thallium activity in irreversible defects and the extent of myocardial fibrosis.41A In the current study, Perrone-Filardi et al17 demonstrate the nearly linear relation between regional thallium activity and the likelihood of recovery of regional function after revascularization. Hence, although all myocardial segments with thallium activity >50% of normal activity were considered "viable," only 56% of segments with thallium activities in the range of 50% to 60% improved after revascularization, whereas 83% of segments with thallium activities >80% showed functional improvement after revascularization. These findings support previous observations by the same investigators42 and by Udelson et al.39 Thus, one important factor explaining the wide range of positive predictive values of thallium imaging reported thus far in individual studies (Figure) may be differences in relative thallium activity in regions considered "viable" among these studies.Perrone-Filardi et al17 also provide insights into the relative value of dobutamine echocardiography in predicting functional recovery after myocardial revascularization. A characteristic feature of viable myocardium that is stunned and/or hibernating is the presence of residual inotropic reserve that may be elicited by catecholamine stimulation. It has been shown in several experimental situations that stunned myocardium exhibits contractile reserve during adrenergic stimulation.43444546 The available models indicate that hibernating myocardium also exhibits contractile reserve with catecholamine stimulation,474849 although this appears to be an unstable process that can be maintained only temporarily,47495051 in keeping with the lack of normal coronary flow reserve. These experimental observations are underscored by studies 2 decades ago that demonstrated contractile reserve in patients with chronic CAD using epinephrine administration or postextrasystolic potentiation during left ventriculography.52535455In keeping with these early reports, low-dose dobutamine echocardiography has emerged in the 1990s as a means of eliciting contractile reserve in patients with acute and chronic LV dysfunction. In the last 3 years, 15 studies of 402 patients with chronic CAD and LV dysfunction undergoing dobutamine echocardiography before revascularization, including the current study by Perrone-Filardi et al,101734353637384042565758596061 have demonstrated that the predictive accuracy of this method regarding recovery of LV function after revascularization is equivalent to that achieved with use of PET or thallium SPECT protocols. The cumulative positive predictive accuracy of dobutamine echocardiography is 83%, with a negative predictive accuracy of 81% (Figure).Thus, the cumulative experience from multiple studies suggests that thallium SPECT imaging provides more sensitive results than dobutamine echocardiography, yielding a 9% greater negative predictive value, but less specific results, yielding a 14% lower positive predictive value. This conclusion is somewhat tenuous considering the wide range of predictive values reported in the individual studies and the lack of standardization of the methods among the investigators. However, the higher sensitivity and lower specificity of thallium imaging compared with dobutamine echocardiography in identifying viable tissue and predicting functional recovery is supported by the current data of Perrone-Filardi et al17 as well as eight other studies3435363738405762 in which thallium SPECT imaging and dobutamine echocardiography were compared directly in the same patients with LV dysfunction.Thus, it is apparent that many myocardial segments with baseline systolic dysfunction will manifest thallium uptake but lack inotropic reserve during dobutamine administration. This concept is supported by recent contrast echocardiographic findings of preserved perfusion but absent contractile reserve in akinetic myocardial segments61 and by the limited number of studies comparing dobutamine echocardiography and PET, which report dysfunctional segments with preserved metabolic activity but without contractile reserve.636465 Thus, similar to thallium imaging, PET identifies viability in a greater number of myocardial segments with contractile dysfunction than does dobutamine echocardiography. There is excellent agreement between PET and dobutamine echocardiography in identifying myocardial regions that have preserved blood flow and are presumably stunned. In contrast, the regions with discordant findings between the two techniques tend to be those that are presumably hibernating, in which blood flow is reduced at rest.6465The lower sensitivity of dobutamine echocardiography in detecting viable myocardium, especially in regions with reduced perfusion, may indicate that some regions of hibernating myocardium are so delicately balanced between the reductions in flow and function, with exhausted coronary flow reserve, that any catecholamine stimulation to increase oxygen demands will merely result in ischemia and inability to elicit enhanced contractile function. Moreover, in the later stages of hibernation, there may be cellular dedifferentiation at the ultrastructural level with dropout of myofibrillar units and accumulation of intracellular glycogen,8666768 resulting in reduced or absent responsiveness to catecholamine stimulation.The increased sensitivity of thallium imaging compared with dobutamine echocardiography occurs at the expense of lower specificity regarding the potential for recovery of function. Although uptake of thallium can occur only if there is viable tissue with intact membrane function, it is important to distinguish between the mere presence of tissue viability and the likelihood that this viable tissue will have functional recovery after revascularization. The identification of viable myocardium is an academic exercise unless it translates into the potential for recovery of function, from which a meaningful decision regarding the advisability of revascularization can be made in the patient with poor LV function. Although thallium imaging may identify more viable segments than does dobutamine echocardiography, it appears to overestimate the potential for recovery of wall motion after revascularization. The recruitment of contractile reserve by dobutamine may indicate which of these viable segments with contractile dysfunction have the potential for improved function after revascularization. This is certainly the case in myocardial regions in which thallium uptake is at the lower end of the viability range, such as thallium activity measuring 50% to 60% of the activity in normal zones, in which the likelihood of functional recovery may depend on whether the magnitude and distribution of viable cells is sufficient to maintain contractile responsiveness.It should be emphasized that there are several issues of both a technical and a physiological nature that cloud the interpretation of the existing comparisons of thallium imaging and dobutamine echocardiography for assessing viable myocardium. Each of these issues warrants further investigation. First, all comparative studies have the potential for anatomic misalignment between the SPECT study and the echocardiogram, because the orientation of the heart is inherently different between the techniques. This is especially problematic because the standard for functional recovery in most studies is the echocardiographic determination of improved wall motion. For the dobutamine echocardiographic procedures, correct registration of the segments with the follow-up echocardiogram is virtually guaranteed, whereas with a PET or SPECT study, substantial assumptions must be made regarding the location of perfusion abnormalities relative to echocardiographic landmarks. Thus, the finding that a dobutamine echocardiogram predicts recovery of echocardiographic wall motion with greater accuracy than either PET or SPECT data is somewhat of a self-fulfilling prophecy.Second, although quantitative analyses of regional thallium activity have been performed in the majority of comparative studies, these data have been interpreted in a binary fashion as positive or negative rather than as a continuum in virtually all previous studies. According to the earlier work of Panza et al,62 confirmed by the current data of Perrone-Filardi et al,17 the magnitude of regional thallium uptake is related to the likelihood that dobutamine will elicit contractile reserve. Segments with a higher level of thallium activity have a greater likelihood of responding to dobutamine stimulation.17426269 Thus, differences in the concordance between thallium imaging and dobutamine echocardiography that have been reported in previous studies may reflect differences among studies in the relative thallium activities in the segments with contractile dysfunction.Third, the severity of regional and global LV dysfunction is an important factor that is likely to alter the diagnostic accuracies of both thallium imaging and dobutamine echocardiography. For example, in the current study by Perrone-Filardi et al,17 the concordance between the two methods was 82% in hypokinetic segments but fell to 43% in akinetic segments. Hypokinetic regions are more likely to manifest inotropic reserve than are regions with akinesia or dyskinesia. In akinetic segments, the sensitivity of dobutamine echocardiography for predicting improvement in function after revascularization has ranged from 74% to as low as 26%.7071 This affects the thallium data as well. Perrone-Filardi et al17 report that the positive predictive value of thallium imaging decreases precipitously from 72% for all dysfunctional segments to 51% for akinetic segments. Regions with severe wall motion abnormalities will have either more fibrosis, more cellular dedifferentiation in hibernating segments, or both, and it appears that the limitations of both thallium imaging and dobutamine echocardiography in identifying viable tissue are accentuated in akinetic compared with hypokinetic segments. Moreover, the mean ejection fraction in the patients studied by Perrone-Filardi et al was 45%, and the average ejection fraction in the patients studied by dobutamine echocardiography in the Figure (reported in 372 of the 402 patients) was 36%. Considering the impact of severity of regional function on the diagnostic accuracy of these tests, it is uncertain whether the predictive accuracy of dobutamine echocardiography and thallium imaging will be maintained in the subset of patients with severe rather than mild LV dysfunction. It is this subgroup of patients in whom viability information is most critical in management decision making.Fourth, with the exception of the previous study of La Canna et al,58 the postrevascularization assessment of regional and global LV function has been based on a single echocardiographic study performed a short time after revascularization. In the current study, Perrone-Filardi et al17 reassessed LV function at a median of 1 month after revascularization. This single look at LV function shortly after revascularization may underestimate the true degree of functional recovery, because many segments (particularly those with cellular dedifferentiation and loss of contractile units) may require a longer period of time to manifest recovery of contractile function. It is conceivable that segments manifesting inotropic responsiveness to dobutamine, reflecting an intact contractile apparatus, will manifest early recovery of systolic function, whereas those segments with thallium evidence of viability but without contractile reserve will also show functional recovery but only after a lengthier recovery period necessary for regeneration of myofibrillar units.Fifth, in the majority of investigations, inotropic reserve during dobutamine administration has been classified as present or absent. Recently, the more complex but commonly encountered situation of biphasic responses to dobutamine has been studied.6072 Because dobutamine causes progressive increases in oxygen demand at higher doses, it often precipitates myocardial ischemia in regions served by a critical stenosis. Hence, wall motion in some dysfunctional regions will improve at low doses of dobutamine but then deteriorate as ischemia is produced at higher doses.6072 It has been demonstrated recently that a biphasic contractile response to dobutamine is the most accurate echocardiographic predictor of functional recovery after revascularization. In a study by Afridi et al,60 a biphasic response had the highest predictive value (72%) for recovery of function whereas the lowest predictive value was observed in segments with either no change (13%) or sustained improvement (15%) during dobutamine. None of the studies thus far that have compared thallium imaging with dobutamine echocardiography, including the most recent series, have examined the biphasic response to dobutamine relative to thallium data.Sixth, there are limitations of dobutamine echocardiography and thallium imaging in general that are particularly important in assessing myocardial viability in patients with moderate to severe LV dysfunction. The qualitative detection of changes in regional wall motion with dobutamine in segments with baseline contractile dysfunction is often difficult. Whether the accuracy and reproducibility of this method in this group of patients can be transported easily from academic centers to the community is uncertain. There are also issues with regard to echocardiographic image quality in a sizeable subset of patients. Up to 37% of patients may have poor-quality studies with failure to visualize all myocardial segments.737475 This may be overcome in the future with contrast echocardiography, which will enhance endocardial border definition in addition to providing evidence of microvascular integrity.6176 Limitations of thallium imaging are also apparent. Attenuation artifacts may mimic perfusion defects in normally perfused regions or may make mild defects appear to be severe, thereby confounding viability assessment. Such photon attenuation is more likely in the severely dilated ventricles encountered in patients with moderate to severe LV systolic dysfunction. Although quantitative analysis of regional thallium activity adds objectivity to the analysis, as noted previously, certainty regarding viability or nonviability occurs only at either extreme of the thallium activity spectrum, while uncertainty exists in defects of moderate severity. Whether the use of 99mTc-based perfusion tracers,3977 iodinated fatty acids,78 or SPECT imaging of 18F-fluorodeoxyglucose with high-energy collimators3879 will overcome these issues and provide greater accuracy than thallium imaging for assessing viable myocardium awaits further study.Finally, the apparently low specificity and positive predictive value of thallium imaging should be reexamined against end points other than improvement in LV function alone. Recovery of regional LV function after revascularization, long the gold standard against which noninvasive imaging techniques have been compared, may not be the only or even the most important benefit of revascularization of viable but dysfunctional myocardium. Even in the absence of improved LV systolic function, revascularization of viable myocardium downstream from a critical coronary artery stenosis may provide clinical benefit by attenuating LV dilatation and remodeling, reducing ventricular arrhythmias, and reducing the risk of subsequent fatal ischemic events. If this proves to be the case, then the relatively low specificity and positive predictive accuracy of thallium imaging regarding recovery of systolic function would be less important than the higher specificity for predicting beneficial effects of revascularization.Two other important points raised by Perrone-Filardi et al17 are worth addressing. Regional thallium activity on the delayed redistribution images in their study was a more important determinant of functional recovery after revascularization and a stronger determinant of dobutamine responsiveness than was the change in regional activity between the initial resting images and the redistribution study. This observation regarding functional recovery supports the earlier observations of Ragosta et al,29 who used a quantitative planar thallium analysis, whereas the observation regarding dobutamine responsiveness is unique. Perrone-Filardi et al17 also point out that late 24-hour redistribution imaging as part of a thallium rest-redistribution protocol results in defect reversibility in a large number of defects that persist at 4 hours. Thus, 21% of irreversible defects at 4 hours and 18% of defects with partial redistribution showed reversibility at 24 hours. The investigators downplay the importance of this finding, because >90% of defects showing these late changes were in the mild-to-moderate category and thus were already considered viable, and imaging at 24 hours changed the classification from nonviable to viable (shifting from 50% of normal activity) in only 2% of defects. Viewed from a different perspective, however, these data are potentially very important. As noted previously, the thallium data attain greatest potential when treated as a continuum rather than as a binary function. It matters little that a segment achieves the status of "viable" by achieving a thallium level of 50% to 60%; only half of such segments improve after revascularization. However, increasing relative thallium activity within the viable range (for example, from 55% to 75%) could have a major impact on the likelihood of functional recovery after revascularization. The finding that roughly 20% of mild to moderate persistent defects at 4 hours, although considered "viable," achieve a higher level of thallium activity at 24 hours should provide greater confidence regarding the potential for functional recovery in such segments. This clearly is a point worthy of further investigation.Above all, the clinical relevance of viability assessment by these and other imaging modalities requires further extensive study. The specific patients likely to benefit clinically from this information are not fully delineated. Several studies have addressed the prognostic implications of medical versus surgical management of patients with LV dysfunction and an extensive amount of viable myocardium, indicating better survival in those patients treated with revascularization.808182 Although suggestive, these studies have been small, nonrandomized, retrospective analyses. At present, the identification of viable myocardium is only one factor that enters into the equation to recommend or not recommend revascularization in the patient with impaired LV function. As in any other patient with CAD, this decision should also be based on clinical presentation, coronary anatomy, LV function, and evidence of inducible ischemia. Increasingly, however, determination of the viability of myocardial territories to be revascularized plays a pivotal role in this decision-making process. Definitive, accurate, and cost-effective methods are essential to make this determination, and nuclear cardiology and echocardiographic techniques will be called on for this purpose with increasing frequency in the future.Selected Abbreviations and AcronymsCAD=coronary artery diseaseLV=left ventricularPET=positron emission tomographySPECT=single photon emission computed tomographyThe opinions expressed in this editorial are not necessarily those of the editors or of the American Heart Association.Download figureDownload PowerPoint Figure 1. Likelihood of improved regional LV function after revascularization based on noninvasive methods to detect viable myocardium. Data are summarized from 6 studies using PET,192122232425 involving 146 patients, in which enhanced 18F-fluorodeoxyglucose (FDG) uptake relative to myocardial blood flow (MBF) was used as a marker of myocardial viability. These PET data are compared with 13 studies using thallium SPECT,17263132333435363738394041 involving 378 patients, and 15 studies using dobutamine echocardiography,101734353637384042565758596061 involving 402 patients. The range of values reported from the individual studies is indicated by the horizontal bars connected by vertical lines. The shaded bars represent the positive predictive value, and the open bars represent the inverse of the negative predictive value. CABG indicates coronary artery bypass graft surgery; PTCA, percutaneous transluminal coronary angioplasty.FootnotesCorrespondence to Robert O. Bonow, MD, Division of Cardiology, Northwestern University Medical School, 250 E Superior St, Suite 524, Chicago, IL 60611. References 1 Rahimtoola SH. A perspective on the three large multicenter randomized clinical trials of coronary bypass surgery for chronic stable angina. Circulation. 1985;72(suppl V):V-123-V-135.Google Scholar2 Braunwald E, Rutherford JD. Reversible ischemic left ventricular dysfunction: evidence for 'hibernating' myocardium. J Am Coll Cardiol..1986; 8:1467-1470.CrossrefMedlineGoogle Scholar3 Rahimtoola SH. The hibernating myocardium. Am Heart J..1989; 117:211-213.CrossrefMedlineGoogle Scholar4 Dilsizian V, Bonow RO, Cannon RO, Tracy CM, Vitale DF, McIntosch CL, Clark RE, Bacharach SL, Green MV. The effect of coronary artery bypass grafting on left ventricular systolic function at rest: evidence for preoperative subclinical myocardial ischemia. Am J Cardiol..1988; 61:1248-1254.CrossrefMedlineGoogle Scholar5 Ross J Jr. Myocardial perfusion-contraction matching: implications for coronary artery disease and hibernation. Circulation..
Referência(s)