Myocardial ischemia is a clinical syndrome manifesting a variety of tissue effects and global cardiac effects that impair cardiac function. When ischemia is severe and prolonged, it causes myocyte death and results in loss of contractile function and tissue infarction. In cases of less severe ischemia, some myocytes remain viable but have depressed contractile function. Heyndrickx et al described a phenomenon of prolonged depression of regional function after a reversible episode of ischemia in dogs  ; later, this was called myocardial stunning (see the image below). [2, 3, 4, 5]
Heyndrickx et al observed that after 5 minutes of occlusion in the left anterior descending coronary artery, surface electrocardiogram (ECG) findings and regional contraction (with reperfusion) rapidly normalized. When the hyperemic response subsided, regional contractile function was still depressed, and regional contractility recovered only after several hours. Recovery of myocardial contractile function after spontaneous restoration of flow may similarly be protracted after angioplasty or revascularization surgery. 
When ischemia is prolonged, the myocytes have depressed contractile function, but they remain viable. This concept of an adaptive process that shuts down the contractile process and decreases the myocardial oxygen demand in the presence of either chronically or intermittently reduced blood flow has generated considerable interest in clinical and experimental setting.
Diamond et al demonstrated that resting wall-motion abnormalities in patients with coronary artery diseases (CADs) improve after the administration of an inotropic agent (dobutamine or epinephrine) or after coronary revascularization in some vascular territory with depressed contractile function and that such territories eventually improve after revascularization. This finding demonstrates that ischemic but noninfarcted myocardium can exist in a state of hibernation without cell death. Rahimtoola suggested that hibernating myocardium is a state of persistently impaired myocardial and left ventricular (LV) function at rest due to reduced coronary blood flow that can be partially or completely restored to normal either by improving blood flow or by reducing oxygen demand (see the image below). 
Clinically, discerning the etiology of depressed myocardial contractile function is difficult, whether it is due to stunned myocardium, silent ischemia, or hibernating myocardium. Finding all 3 entities in the same patient with chronic myocardial dysfunction is not uncommon (see the image above). The critical difference is that blood flow is normal or near normal with stunning but is reduced in the other 2 conditions. Some authors have suggested that stunning, silent ischemia, and hibernating myocardium are processes by which the myocardium adapts to reduced myocardial oxygen supply to maintain cellular integrity. However, this theory has not yet been confirmed by animal and human experiments.
Mechanisms for myocardial stunning
There are 2 major hypotheses for myocardial stunning: (1) a oxygen-free radical hypothesis and (2) a calcium overload hypothesis. [7, 8, 9, 10] Postischemic dysfunction may be due to cytotoxic oxygen-derived free radicals (ie, hydroxyl radicals, superoxide anions) that are thought to be generated during occlusion or, upon reperfusion. Such radicals cause lipid peroxidation, altering their function and structure.
Normal cardiac contraction depends on the maintenance of calcium cycling and homeostasis across the mitochondrial membrane and sarcoplasmic reticulum during each cardiac cycle. Brief ischemia followed by reperfusion damages Ca2+ pump and ion channels of the sarcoplasmic reticulum. This results in the electromechanical uncoupling of energy generation from contraction that characterizes myocardial stunning. Calcium accumulates in the cell at the time of reperfusion and that is followed by a partial failure of normal beat-to-beat calcium cycling, which perhaps occurs at the level of the sarcoplasmic reticulum. This mechanism is proposed to account for contractile dysfunction. Structurally, myocytes in stunned myocardium appear normal when examined by light microscopy.
The appearance of glycogen vacuoles adjacent to mitochondria and of myofibrillary loss are noted in most cases of hibernating myocardium when examined by electron microscopy. Several controversies exist regarding these histologic changes. [11, 12]
Progression from chronic repetitive stunning to hibernation — a continuum
Several animal models have been proposed to demonstrate the physiologic significance of coronary stenosis, in which a regulation of flow and downregulation of metabolism lead to hibernating myocardium. The key determinant of this adaptive process is a reduction in the coronary perfusion reserve that results from critical coronary stenosis. Conti demonstrated that the production of an acute and critical reduction in coronary flow reserve (CFR) in chronically instrumented pigs leads to an accelerated progression of chronic stunning proceeding to hibernation in less than 2 weeks. [13, 14, 15, 16, 17, 18]
The time frame for the transition from stunning to hibernation can be fairly short, and it is directly related to the degree of flow impairment in a stenotic coronary artery that supplies the dysfunctional segment. As the ischemic threshold decreases with a reduction in the CFR, repetitive stunning results in a delay in the recovery of function that becomes longer than the interval between ischemic episodes. The image below shows a schematic diagram of the potential mechanism of myocardial stunning. [7, 8, 9]
Several other groups have also demonstrated impairment of coronary vasodilator reserve in chronically dysfunctional myocardium. It has been shown that in patients with CAD, flow reserve decreases as the degree of stenosis is increased, and flow reserve is absent with stenoses as great as 80% of the luminal diameter. Progression of coronary stenosis to a critical limit and a loss of the CFR mean that subendocardial flow cannot accommodate the increasing demand, predisposing myocardium to hibernation. Borderline impairment with frequent intermittent ischemia when demand increases may suffice to cause hibernation. Certainly, chronic resting ischemia causes hibernation.
In patients with severe CAD, therefore, the limited flow reserve leads to development of myocardial ischemia, even with a small increase in the oxygen demand, such as in ordinary daily activities and exercise. Thus, intermittent episodes of ischemia and, consequently, postischemic stunning (which should occur frequently in patients with severe CAD) might play a role in the development of chronic reversible LV dysfunction. No animal models are available to support this finding, however. Differentiating between subendocardial ischemia and hibernating myocardium can be difficult.
Collectively, these findings indicate that chronic repetitive ischemia that progresses to hibernating myocardium is associated with regional downregulation of the sarcoplasmic reticulum, with changes in calcium regulation and gene expression. These changes are accompanied by modest increases in myocyte apoptosis and a reduction in the regional myocyte nuclear density. These same structural and functional findings occur in patients with dilated cardiomyopathy of ischemic origin.
With severe CAD, there is a combination of repetitive episodes of stunning superimposed on chronic hibernation with repetitive thrombosis and clot lysis. Over time, the combination results in adaptive downregulation of the contractile machinery to salvage cell integrity (see the image above, panel D). The degree of regional LV dysfunction depends on the severity and repetitive nature of the ischemic episodes. This variability results in a continuum from healthy cardiac tissue with normal contractile function to tissue affected by ischemic episodes, to chronically ischemic tissue, and ultimately to infarcted tissue. Thus, reversible dysfunction of myocardium represents a range of tissue responses to repetitively inadequate supply, intermediate between preserved normal function and infarction.
Hibernating myocardium is characterized by the following:
Episodic and/or chronically reduced blood flow, which is directly responsible for the decrease in the myocardial contractile function.
Tissue ischemia and resultant remodeling without necrosis, which causes prioritization of metabolic process in the myocardial cell relative to contractile function.
Residual contractile reserve in response to inotropic stimulation (in at least half of clinical cases).
Recovery of contractile function after successful revascularization.
Hibernating myocardium is suspected in all patients with CADs and moderate-to-severe chronic LV dysfunction. As many as 50% of patients with previous infarction may have areas of hibernating tissue mixed with scar tissue, even in the presence of Q waves on the electrocardiogram. The extent and severity of ventricular dysfunction can vary considerably. Hibernating myocardium may be limited to a discrete portion of ventricle with normal systolic function, or it may involve global impairment of LV function and result in the clinical syndrome of heart failure. Thus, the clinical dilemma exists regarding attempted coronary artery bypass grafting (CABG) in high-risk patients to restore global ventricular function versus orthotopic cardiac transplantation. [19, 20, 21, 22]
Mechanism of Disease
Pathophysiologic mechanism of hibernating myocardium
Multiple cellular and animal models have been proposed, but most of them fall short of reproducing the clinical syndrome of hibernating myocardium. Arani et al infused soluble hydrogen gas in patients with collateral-dependent myocardium and complete occlusion of the left anterior descending artery and with a distal collateral supply from the right coronary artery. The findings suggested that coronary flow per unit weight is not maintained at usual basal values in densely collateralized myocardium that is totally collateral dependent. The reduction in flow is presumably associated with a marked reduction in local arterial pressure. 
Myocardial perfusion imaging is used to estimate myocardial blood flow (MBF) and metabolic activity. Several factors limit the use of thallium- and technetium-based agents to accurately quantify MBF. Positron emission tomography (PET) overcomes important limitations by providing attenuation correction, enabling quantification of the concentration of the radiotracer in the organ of interest. Initial PET studies revealed that hibernating segments correspond to areas with qualitatively reduced perfusion, as assessed with13 N-labeled ammonia in the presence of preserved fluorodeoxyglucose (FDG) uptake. This resulted in the concept of perfusion and metabolism mismatch in which flow is reduced but metabolic integrity is retained.
Investigators concluded that in dysfunctional segments with a perfusion-metabolism mismatch, MBF was higher than that in dysfunctional, nonviable segments but lower than that in remote, normally contracting myocardium. Review of several studies shows that in most cases, baseline blood flow to hibernating myocardium is within the range of values measured in the myocardium of healthy volunteers by using PET. This discrepancy may be explained by insufficient accuracy of the PET measurement. Using a kinetic model with measurement of freely diffusible tracer in tissue provides a more accurate assessment of flow.
This H2,15 O technique provides the flow per gram of perfused tissue. Because the uptake of H2,15 O in scar tissue is negligible as compared with that of healthy myocardium, in a myocardial region consisting of a mixture of viable and necrotic tissue, this value predominantly reflects flow to residual healthy myocardium.13 NH3 is used to measure the average flow per unit mass of tissue, as with the microsphere technique. Several studies of H2 and15 O revealed that MBF in dysfunctional segments was comparable to that of healthy segments.
In some cases, regional difference in flow is sustained, in part, by higher perfusion in remote myocardium. Elevated perfusion may be triggered as a consequence of higher oxygen consumption. The compensatory changes in flow distribution may be responsible for an apparent reduction of tracer content in the dysfunctional regions, which can be erroneously interpreted as an absolute flow reduction in the same segment. [24, 25, 26, 27]
Histologic specimens obtained at the time of surgery demonstrate profound structural changes in dysfunctional but viable myocardium. These changes are as follows:
A progressive loss of contractile protein (sarcomere) occurs without a loss of cell volume. The depletion of sarcomeres is most prominent near the perinuclear region but may extend to the entire cell.
Numerous small mitochondria can be found in the areas adjacent to the glycogen-rich perinuclear zone.
Nuclear heterochromatin is evenly distributed in the nucleoplasm.
A substantial loss of sarcoplasmic reticulum occurs. The sarcoplasmic reticulum loses T-tubules and becomes disorganized.
Primitive cytoskeleton proteins, including titin and cardiotin, have increased expression in the hibernating myocytes cells.
A direct correlation exists between the severity of these ultrastructural changes and the time course of functional recovery after revascularization. Cell dedifferentiation is reversible after restoration of blood flow with revascularization. Whether cell death occurs by means of apoptosis or necrosis is not clearly known. These concepts need further evaluation.
Identification of Patients With Hibernating Myocardium
Hibernating myocardium is suspected in all patients with CAD and chronic LV dysfunction ranging from regional dysfunction to ischemic cardiomyopathies. The extent and severity of LV dysfunction is not directly related to the extent and severity of CAD. In many patients, preexisting collateral vessels and newly developed vessels in the coronary circulation can result in preservation of normal LV function despite the presence of CAD.
Hibernating myocardium is viable, and this viability can be determined with a variety of imaging techniques that depict either the presence of myocardial tissue that contracts if stimulated appropriately or the persistence of metabolic activity within the region of dysfunctional myocardium. These modalities can be used to distinguish between myocardium with potential for improved function and myocardium with irreversible damage that does not improve after revascularization. Functional recovery after revascularization appears to vary greatly among studies, ranging from 24-84%. Specific findings with various methods are used to determine potential outcomes of revascularization procedures. Several groups have published data regarding predictions of recovery of myocardial dysfunction after revascularization. 
Evaluation of Hibernating Myocardium
Current approaches for identifying viability in hibernating myocardium
The following can be used in the assessment of perfusion, membrane integrity, and metabolism:
Regional perfusion testing - 201 Tl scintigraphy, 99m Tc-sestamibi (MIBI), 99m Tc-tetrafosmin
PET - 82 Rb, 13 NH 3, 18 FDG, 14 C-acetate
MRI for the assessment of regional perfusion and wall function.
The following can be used in the assessment of regional function:
- Nitroglycerin administration
- Post-extrasystolic potentiation (PESP)
- Low-dose catecholamine administration
- Exercise testing
The use of nuclear imaging to detect hibernating myocardium relies on the demonstration of membrane integrity and/or metabolic activity within the hibernating myocardium. Myocardial retention of the potassium analog201 Tl can be detected by using single photon emission computed tomography (SPECT). In healthy myocardium, thallium uptake is initially high but decreases rapidly within hours.
Normal thallium uptake levels with exercise and thallium defects that redistribute on images obtained after a 3-4-hour delay are accurate predictors of viable myocardium. However, the absence or lack of thallium uptake does not necessarily indicate myocardial scarring, because severely ischemic but viable myocardium, as well as a mixture of scar and viable myocardium, may also produce an "irreversible defect."
Gibson and associates demonstrated that 45% of the segments with irreversible defect had improved thallium uptake after CABG.  About 25% of the preoperatively akinetic and dyskinetic segments (as demonstrated by201 Tl defects) showed improved perfusion. This finding is in contrast to the 75% postoperative improvement in healthy and hypokinetic segments.
Segments that are likely to improve have thallium activity that is more than 50% of the activity of healthy myocardium. Thus, standard redistribution images obtained after 3-4 hours may not help in the differentiation of hibernating myocardium from scarred myocardium. From 45-75% of persistent early defects exhibit normal perfusion after revascularization.
Reinjection or 24-hour thallium protocol may reduce the false assessment of irreversible defect. From 31-52% of defects demonstrate improved201 Tl uptake after tracer reinjection following stress using a smaller dose of201 Tl (1 mCi). Some investigators recommend repeating images 24 hours after injection. However, a small number of segments (3-4%) show some degree of reversibility.  (See the image below.)
PET is often considered the gold standard for the detection of viable myocardium. Early studies attempted to identify whether thallium using the standard stress-redistribution protocol overestimates the infarct and/or scar tissue. Study findings clearly indicate that oxidative metabolism is preserved in ischemic and hibernating myocardium. In ischemic and hibernating myocardium, regional substrate utilization shifts from free fatty acids to glucose. [33, 34, 35]
The production of the glucose protein transporter (GLUT) protein is increased, as demonstrated by Schwaiger et al, as is the expression of myocardial glucose transporter messenger RNA protein in patients with advanced CAD.  Glucose utilization is, in turn, influenced by a number of factors, such as coronary perfusion, cardiac work, and insulin and hormonal effects. FDG is a glucose analogue that is intracellularly phosphorylated to FDG-6-phosphate in the myocardium. An increased uptake of18 FDG in relation to perfusion or flow-metabolism mismatch is indicative of hibernating myocardium, whereas matched defects are indicative of scar tissue; values in between these represent healthy myocardial tissue mixed with fibrosis. 
Preoperative measurement of the regional flow and uptake of18 FDG can be used to accurately predict functional recovery after revascularization in patients with depressed ventricular function. The predictive accuracy achieved with PET images is comparable to that of thallium images, with a positive predictive value (PPV) of 80-87% and a negative predictive value (NPV) of 82-100%. Using quantitative analysis, thallium images by stress-redistribution-reinjection protocol is comparable to the standard PET metabolic images, as confirmed by Tamaki et al. [25, 26]
Table 1 shows the potential relationships between these measurements.
Table 1. Potential Relationship Between Regional Wall Motion, Blood Flow, and FDG Uptake (Open Table in a new window)
|Regional Wall Motion||Blood Flow||FDG Uptake||Diagnosis|
|Normal||Normal||Normal||Metabolically acting myocardium|
Three patterns of myocardial viability may occur with a PET perfusion-metabolism mismatch protocol. The perfusion metabolism match pattern demonstrates either a concordantly reduced, absent, or regional myocardial perfusion and FDG uptake. If severe, this implies transmural infarction and irreversible LV function. A pattern in which mild-to-moderate flow is matched to metabolism suggests the presence of both viable and nonviable tissue in a given region of myocardium. On the other hand, when regional myocardial FDG uptake is disproportionately enhanced compared with blood flow, the pattern is termed perfusion-metabolism mismatch. This pattern closely resembles that of hibernating myocardium.
Compared with thallium imaging, FDG-PET imaging provides better results for the differentiation of viable myocardium from scarred myocardium. Brunken et al published data from a comparison of tomographic thallium images with PET images; 47% of the irreversible thallium defects were identified as viable on PET images. 
Tamaki et al subsequently confirmed these findings in 2 comparative studies of SPECT and PET in which 38-42% of the irreversible thallium defects had enhanced FDG uptake suggestive of viable myocardium. [25, 26] Thus, conventional stress-redistribution201 Tl imaging has a low predictive value in the identification of viable myocardium when compared with PET. Brunken et al found comparable results between resting thallium uptake and FDG-PET findings.  (See the image below.)
When these similarities are considered, either thallium protocol yields satisfactory information. PET does provide information about regional blood flow, and metabolic function is assessed independent of flow. Its higher cost and limited availability preclude its wide usage; however, approval of payment by the Health Care Financing Administration (HCFA) has rekindled interest in this imaging technique.
Thallium scintigraphy involves stress early redistribution imaging, late redistribution imaging at 8-24 hours, reinjection imaging, and rest redistribution imaging. 
In hibernating myocardium, the initial uptake is low but then increases gradually; this phenomenon is related to201 Tl redistribution. Regional thallium activity on early (at 3-4 h) or late (at 8-72 h) redistribution imaging obtained after stress has been used to demonstrate the distribution of viable myocardial cells and the extent of myocardial fibrosis. Determining the severity of the defect in201 Tl uptake after the redistribution is important as well.
Several protocols, including reinjection, are used to assess myocardial function and viability. Yang et al and Kiat et al performed late imaging in 118 patients with CAD. [39, 40] In the studies, late distribution was observed in 53% of the patients, but it was seen in only 22% of the segments with 4-hour irreversible defect. A possible explanation for late redistribution may be that the initial uptake of thallium during exercise is sufficiently reduced in certain ischemic myocardial regions so that it continues to mimic the appearance of scarred myocardium on early 3-4 hour images. Therefore, if more time is allowed for redistribution, more viable segments are distinguished from scarred or fibrotic myocardium. Late thallium redistribution, when present, is an accurate indicator of viable myocardium. 
Kiat et al showed that 95% of the segments with late redistribution improved with revascularization.  However, the absence of thallium redistribution on late images remains an inaccurate marker for nonviability; 37% of the segments that remained irreversible on images obtained at both 3-4 and 24 hours also improved after revascularization. This finding suggests that even after the acquisition of late redistribution images, this measure continues to cause overestimation of the frequency and severity of myocardial scar tissue.
On the other hand, reinjection of additional thallium immediately after conventional 3-4 hour imaging significantly improves the detection of viable myocardium in 31-49% of the regions that were interpreted as having an irreversible perfusion defect on conventional redistribution images. Available data suggest that the thallium reinjection in otherwise irreversible defects can be used to predict improvement in regional function after revascularization, with a PPV of 80-87% and an NPV of 82-100%. Thus, thallium reinjection improves the detection of viable myocardium, as shown in several studies, even when regional quantitative analysis is used. [32, 42, 43, 44] (See the image below.)
Some laboratories started using the stress-reinjection protocol instead of stress-redistribution-reinjection and found that 25% of the reversible segments were incorrectly identified as irreversible defects. This result was thought to be due to the phenomenon of differential uptake, a low differential uptake of thallium after reinjection, which causes an appearance of persistent defect. Thus, the data suggest that stress-redistribution-reinjection or stress-reinjection–late redistribution (24 h) techniques may provide comparable information for identifying most of the ischemic but viable myocardium.
Ragosta et al were the first to report that thallium perfusion defects may occur on resting images in patients with CAD in the absence of acute ischemic process or previous myocardial infarction.  Available data suggest that rest-redistribution thallium imaging depicts viable myocardium in most reversible regions, but it may cause underestimation of viable myocardium in as much as two thirds of the irreversible regions.
Quantitative analysis improves the detection of viable myocardium. Findings from comparisons of stress-redistribution-reinjection and stress-reinjection–24-hour imaging may be used to assess the extent and severity of myocardial ischemia and viability. If the clinical question concerns viability, rest-redistribution or late redistribution imaging is a good alternative in distinguishing viable from nonviable myocardium. Late redistribution of201 Tl injected at rest occurs occasionally; only 3% of the segments with initial thallium defects were considered viable with late redistribution images obtained 20-24 hours later. Late redistribution did not significantly improve the prediction of functional recovery after revascularization. Therefore, most of the clinically relevant information can be obtained with conventional rest-early redistribution imaging.
Several groups have investigated the role of dobutamine echocardiography in predicting outcomes after revascularization procedures in patients with severe chronic CADs and hibernating myocardium. [11, 46, 47, 48, 49, 50, 51] This noninvasive technique uses a progressively increasing dose of dobutamine, which first augments regional function and then induces ischemic wall motion abnormalities by increasing the myocardial oxygen demand in the presence of coronary stenosis (see the image below).
However, the response to dobutamine is difficult to interpret because it is dependent on various factors, such as the extent of viable myocardium, the severity of coronary stenosis responsible for the hibernating myocardium, and the collateral circulation. Contractile response of the viable myocardium to dobutamine is, again, dependent on several factors. [7, 52, 53]
Factors affecting contractile reserve in hibernating myocardium include the following:
Amount of interstitial fibrosis (scar tissue) in the myocardium
Sarcoplasmic reticulum function
MBF at rest
Biphasic response with improvement in wall motion at low doses and worsening at high doses (see the image below)Hibernating and stunned myocardium. Positron emission tomography depicts metabolic viability in myocardium with persistent 24-hour thallium-201 defects observed with single photon emission CT. (Adapted from Brunken et al. Circulation. 1992 Nov; 86: 1357-69.)
Sustained improvement in wall motion at low doses, with further improvement at higher doses
Worsening of resting wall motion without any improvement
No change in wall motion during dobutamine echocardiography
Perrone-Filardi et al examined 18 patients with chronic CADs who were undergoing revascularization with 2-dimensional echocardiography and inotropic stimulation (dobutamine). During the dobutamine infusion, none of the 79 hypoperfused dysfunctional segments had a further deterioration of function, whereas 46 (58%) of the dysfunctional segments functionally improved by at least 1 score grade. Functional improvement was observed in 35 of the hypokinetic segments that became normokinetic; of 11 akinetic segments, 4 became hypokinetic and 7 became normokinetic. The remaining 33 dysfunctional hypoperfused segments had no functional recovery. Of 48 hypoperfused dysfunctional segments with improved function after revascularization, 42 (87%) also improved during dobutamine infusion, whereas 27 of 31 segments with unchanged function after revascularization did not improve during the infusion. 
The sensitivity and specificity of the dobutamine infusion technique in identifying dysfunctional segments that are capable of recovering function after revascularization were 88% and 87%, respectively. Of 46 dysfunctional segments that improved during the dobutamine infusion, 42 improved after revascularization, as opposed to 6 (18%) of 33 segments with no changes during the infusion. Therefore, the positive and negative accuracies in predicting functional improvement in dysfunctional hypoperfused segments after revascularization was 91% and 82%, respectively. The data suggest that most of the hypoperfused and dysfunctional segments are capable of improving function with inotropic stimulation. These myocardial segments have a functional and vasodilator reserve despite an increased myocardial oxygen demand during inotropic stimulation. Notably, the dobutamine technique has a lower sensitivity for predicting outcomes in segments that are akinetic preoperatively.
Afridi et al found similar results. They noted that the biphasic response during dobutamine infusion is a best predictor of improvement. Preinterventional identification of improved systolic wall thickening with inotropic stimulation (low-dose dobutamine) at echocardiography has been a preferred method. Viable myocardial segments thicken in response to dobutamine. With increasing doses of dobutamine, deterioration of functional wall thickening occurs, and akinesis and lack of synergy result in the so-called biphasic response. This has a PPV (positive predictive value) of 83% and an NPV of 81%. The finding of a biphasic response had the highest predictive value for recovery. They concluded that low- and high-dose dobutamine infusion are best in evaluating the need for optimal assessment of myocardial viability.
The major finding of all studies is that echocardiographic detection of contractile reserve during low-dose dobutamine infusion is a strong predictor of LV function after coronary revascularization. [45, 46]
Scognamiglio et al demonstrated that post-extrasystolic potentiation (PESP) is another sensitive method for predicting the outcome after revascularization. The specificities of the 2 methods are the same. PESP has an advantage over dobutamine infusion in that it achieves maximal contractility without inducing ischemia. 
Table 2 shows the sensitivities and specificities of the various techniques.
Table 2. Sensitivity and Specificity of Different Methods in the Detection of Hibernating Myocardium in Different Studies (Open Table in a new window)
|Test||Sensitivity, %*||Specificity, %*||No. of Patients|
|MIBI||83 (78-87)||69 (63-74)||207|
|Dobutamine echocardiography||84 (82-86)||81 (79-84)||448|
|201 Tl reinjection||86 (86-89)||47 (43-51)||209|
|FDG PET||88 (84-91)||73 (69-74)||332|
|201 Tl rest-redistribution||90 (87-93)||54 (49-60)||145|
|*Data in parentheses are ranges.|
99m Tc-MIBI is only minimally redistributed, and it is not taken up by necrotic myocardium. Also,99m Tc has a shorter half-life (half-life, 6 hours) than that of thallium (half-life, 2.8 days). Therefore, some have suggested that the use of99m Tc-MIBI causes underestimation of the viable myocardium. However, when resting myocardial perfusion scintigraphy with99m Tc-MIBI is combined with a preinjection administration of nitroglycerin, it may be as effective as redistribution with201 Tl. [51, 55, 56, 57, 58]
The prediction of functional recovery is based on the semiquantitative analysis of residual MIBI uptake in dysfunctional segments compared with remote areas of high uptake. An uptake of 50-60% is used as the threshold for viable tissue. Udelson et al compared the regional activities of201 Tl and99m Tc-MIBI after resting injection. They found that quantitative analysis of the regional activities of both201 Tl and99m Tc-MIBI after resting injections can be used to differentiate viable from nonviable myocardium and that the 2 agents are comparable in predicting the recovery of wall motion abnormalities after revascularization. [59, 60]
Other studies have shown a much lower correlation between the presence of a severe technetium defect and FDG uptake. The addition of ECG gating and the evaluation of regional wall thickening have also been suggested as techniques to enhance the detection of viable myocardium (see the image below). As a result, thallium is preferred to technetium for determining the viability of the myocardium. 
Similarly, low-dose dobutamine infusion with99m Tc-MIBI SPECT has been shown to provide better accuracy for predicting functional recovery than rest SPECT. Thus, an optimal viability SPECT protocol consists of rest-redistribution201 Tl imaging followed by nitroglycerin-augmented low-dose dobutamine99m Tc-MIBI gated SPECT.
In an effort to evaluate the influence of SPECT attenuation correction on the quantification of hibernating myocardium derived from perfusion SPECT and (18)F-FDG PET, investigators studied 20 patients who underwent rest (99m)Tc-tetrofosmin perfusion SPECT/CT and (18)F-FDG PET/CT. Perfusion images were reconstructed without attenuation correction (NC); with attenuation correction based on the CT from the SPECT/CT (AC_SPECT); and with attenuation correction based on the CT from the PET/CT (AC_PET). The authors concluded that AC of SPECT perfusion scans with an attenuation map derived from PET/CT scans is feasible. If AC is unavailable, perfusion scans should be compared to NC normative databases for assessing total perfusion deficit (TPD), hibernation, and mismatch. 
Assessment of regional perfusion with perfusion MRI
Because of the noninvasive nature of MRI, cine MRI (cMRI) of the heart is an excellent method for the assessment of regional wall motion, when it is available. Baer and coworkers reported findings in 35 patients with myocardial infarction and regional akinesis or dyskinesis who underwent rest and dobutamine MRI as well as FDG analysis. In this study, quantitative and functional MRI parameters (end-diastolic wall thickness at rest and dobutamine-induced systolic wall thickening, respectively) were studied as markers of myocardial viability and compared with the corresponding18 FDG uptake, as assessed by PET. [63, 64, 65]
A comparison of segmental MRI and FDG PET results indicated that dobutamine-induced wall thickening was a better predictor of residual metabolic activity (sensitivity, 81%; specificity, 95%; PPV, 96%) than end-diastolic wall thickness (sensitivity, 72%; specificity, 89%; PPV, 91%). When both parameters were taken into account, the overall sensitivity of MRI improved to 88% for metabolic activity assessed at FDG PET, without a major decrease in specificity (87%) or PPV (92%).
Pearlman et al used a porcine model to measure wall motion and wall thickening during the cardiac cycle. They used a serial motion assessment by reference tracking system (SMART) to analyze the cardiac motion. It increased the contrast between the ischemic and normal values for wall motion and thickening change. It is twice as sensitive in detecting abnormalities of wall motion and thickening and is thus useful in differentiating ischemic myocardium from normal myocardium. However further studies are need to validate the system for clinical evaluation (see the image below).
Kim et al reported results of delayed gadolinium enhancement with gated MRI for the diagnosis of hibernating myocardium. The PPV and NPV were 71% and 79%, respectively, for kinetic and dyskinetic segments but, otherwise, 88% and 89%.
Gadolinium-based contrast agents (gadopentetate dimeglumine [Magnevist], gadobenate dimeglumine [MultiHance], gadodiamide [Omniscan], gadoversetamide [OptiMARK], gadoteridol [ProHance]) have been linked to the development of nephrogenic systemic fibrosis (NSF) or nephrogenic fibrosing dermopathy (NFD). For more information, see Nephrogenic Fibrosing Dermopathy. The disease has occurred in patients with moderate to end-stage renal disease after being given a gadolinium-based contrast agent to enhance MRI or MRA scans.
NSF/NFD is a debilitating and sometimes fatal disease. Characteristics include red or dark patches on the skin; burning, itching, swelling, hardening, and tightening of the skin; yellow spots on the whites of the eyes; joint stiffness with trouble moving or straightening the arms, hands, legs, or feet; pain deep in the hip bones or ribs; and muscle weakness. For more information, see the FDA Information on Gadolinium-Based Contrast Agents or Medscape.
Bax et al performed a meta-analysis of various perfusion imaging and echocardiographic techniques and their usefulness in the prediction of myocardial viability (see the image below). The data revealed that the sensitivity in predicting improved regional contractile function after revascularization was high for all techniques analyzed; however, specificity varied greatly and was lowest for201 Tl stress-redistribution and201 Tl rest-redistribution imaging.
Specificity was highest for low-dose dobutamine echocardiography (LDDE). Other data suggested that dobutamine MRI with SMART (serial motion assessment reference tracking) tagging and/or point trajectory assessment is the most accurate because it has better endocardial definition, has higher resolution, and corrects for tethering. It is still evolving and not widely available. The negative predictive value was highest for MRI or FDG PET.
Preliminary data suggested that SMART-MRI may be more accurate. Thus, the available evidence supports the use of LDDE as the first technique of choice in the prediction of regional functional recovery in patients with chronic ischemic LV dysfunction. [30, 66, 67, 68, 69]
Receiver operating characteristics evaluate the performance of different tests in terms of trade-off between sensitivity and specificity. The most useful methods have high area under the curve of true positive fraction (sensitivity) versus false-positive fraction (1-specificity) (eg, .90-1.0). ROC data have documented fairly good ability to identify salvageable myocardium using FDG PET, dobutamine echocardiography, and nitroglycerin-enhanced99m Tc-MIBI imaging.
Revascularization - Effect of Treatment
Patients with CAD are often referred for revascularization because of angina or chronic ventricular dysfunction with viable myocardium. The most important benefit of this procedure is improvement of the global ejection fraction. This improvement is directly related to the number of dysfunctional but viable segments. This is more obvious when functional response to exercise or dobutamine infusion is evaluated after treatment. The magnitude of improvement in the symptoms of heart failure and in exercise capacity is proportional to the mass of revascularized myocardium that was demonstrated to be viable before surgery.
The results of the Coronary Artery Surgery Study (CASS), in which bypass surgery was compared with medical treatment, indicate that patients with multivessel disease and poor ventricular function benefited most from surgery in terms of survival. Revascularization improves the prognosis of patients with hibernating myocardium. Therefore, the identification of patients with hibernating myocardium is vital to overall treatment of patients with CAD.
The evaluation before revascularization in patients with LV dysfunction includes the following:
Assess the presence of available target vessels for bypass or angioplasty.
Consider other risk factors for death after bypass surgery (eg, age, sex, comorbid condition, prior surgery, valvular diseases).
Consider hemodynamic support during angioplasty (use of an intra-aortic balloon pump or percutaneous circulatory assistance).
Evaluate the extent of viable myocardium in the distribution of the target vessels.
If severe heart failure is present, consider use of a device to assist the LV or cardiac transplantation.
Results of studies have led to a new paradigm for the development of chronically reduced regional flow in viable dysfunctional myocardium. In this paradigm, repetitive episodes of ischemia lead to chronic myocardial stunning, in which the resting flow is normal. With time and progression in the physiologic significance of coronary stenosis, a regional downregulation in flow and function occurs. The key determinant of this adaptive response is the extent to which coronary flow during maximal vasodilatation (the CFR) is reduced. Thus, this new conceptual framework links chronic stunning and chronic hibernation on a continuum.
Considering the frequency of ischemia and the prevalence of hibernating myocardium, recognition of hibernation and how it affects treatment are important. The mortality rate is higher in patients with hibernating myocardium who do not undergo revascularization. (See the image below.)
The accurate identification of patients with reversible LV dysfunction is important in clinical decision-making. In clinical practice, several strategies are used for patients with severe LV dysfunction. The current analysis shows that FDG PET has the highest sensitivity, followed by other nuclear techniques, and dobutamine echocardiography has the lowest sensitivity, except for MIBI imaging. The highest NPV was observed with FDG PET, followed by dobutamine echocardiography and other nuclear imaging techniques. Specificity was highest with dobutamine echocardiography, followed by FDG PET. The highest PPV was observed with dobutamine echocardiography, followed by FDG PET. Although not widely available, cardiac MRI offers even greater promise of high accuracy in identifying salvageable myocardium in terms of impaired contractility that can respond to inotropic stimulation.
Issues such as ischemic symptoms, suitability of the coronary anatomy, severity of the LV dysfunction, and the presence or absence of inducible ischemia are determined before clinical decisions are made. Overall, the identification of hibernating myocardium and the prudent use of revascularization procedures have improved treatment outcomes in patients with this condition.