+++
Single-Photon Emission Computed Tomography
++
Myocardium viability imaging can be performed with 201-thallium (201Tl) or 99m-technetium (99mTc) and relies on the integrity of the sarcolemma and mitochondria, respectively.29–31 Comparisons between 201Tl SPECT, 99mTc SPECT, and 18FDG PET show that scintigraphy with 99mTc-based imaging may underestimate the amount of viable myocardium,32–36 while one direct comparison study observed that 201Tl provided comparable information with 18FDG PET.37 A meta-analysis of 40 studies (1119 patients) using 201Tl SPECT showed a mean sensitivity, specificity, predictive positive value (PPV), and negative predictive value (NPV) of 87%, 54%, 67%, and 79%, respectively.4 A meta-analysis of 25 studies (721 patients) that used 99mTc SPECT to assess viability showed sensitivity, specificity, PPV, and NPV of 83%, 65%, 74%, and 76%, respectively.4 Both were not as sensitive as 18FDG PET.4 However, other studies showed similar values between 201Tl and 99mTc-based imaging.38,39 Overall, both approaches are considered reasonable means to assess myocardial viability.8
+++
Imaging Protocols for 201-Thallium SPECT
++
Thallium is a potassium analog and its uptake is both passive and also active via a process requiring the normal function of the sodium-potassium ATPase pump and cellular membrane integrity.40 Since membrane integrity is a requisite for cell viability, thallium-201 uptake visualized by SPECT images is indicative of myocardium viability.
++
Different protocols are described to assess viability with 201Tl SPECT. The American Society of Nuclear Cardiology (ASNC) guidelines41 describe a didactic format (Figs. 21-2 and 21-3). Although the rest-redistribution protocol can be performed (see Fig. 21-2), the most common protocol starts with a stress phase and injection of 2.5 to 3.5 mCi of 201Tl at peak stress (Fig. 21.3). After 10 to 15 minutes, the stress images are acquired with redistribution (rest) imaging acquired 2.5 to 4 hours later. Up to this point, a regular stress/rest 201 Tl protocol is described. When a persistent (fixed) defect is present and viability needs to be assessed, a late-redistribution imaging can be performed at 18 to 24 hours. 201Tl redistribution is a continual process that requires blood supply to the viable tissue and thus its uptake is also related to perfusion and the severity of coronary artery stenosis.42 Studies have shown that in late images (8–72 hours) the viable myocardium segments show thallium redistribution (reversible defects) while truly nonviable myocardium appears as a persistent (fixed) defects on the late perfusion images (no thallium uptake).42–45 To maximize the protocol, addition 1 to 2 mCi of 201Tl can be reinjected (Fig. 21-3).41
++
++
++
Subanalysis of a pooled meta-analysis compared 201Tl SPECT rest-redistribution to the 201Tl SPECT reinjection protocol and showed comparable sensitivities (87% for both protocols) but with higher specificity and PPV for 201Tl SPECT rest-redistribution (56% vs 50% and 71% vs. 58%, respectively) (p < 0.05 for both).4 On the other hand, head to head comparison between 24-hour redistribution with the reinjection protocol suggested that reinjection may provide better diagnostic information with a significantly greater ability to identify hibernating myocardium.46 When comparing the reinjection protocol to 18FDG PET, there was very good correlation, although 201Tl may have inferior sensitivity compared to glucose metabolism assessment with PET.4,37,47,48 The difference in sensitivity may be because perfusion/18FDG PET already considers perfusion. 18FDG then adds metabolic information when there is mismatch. Figure 21-4 shows an example of 201Tl SPECT rest-redistribution imaging.
++
+++
Imaging Protocols for 99m-Technetium SPECT
++
99mTechnetium-sestamibi and 99mTc-tetrofosmin have lipophilic proprieties and enter cells passively. However, their retention by the mitochondria is an active process and depends on mitochondrial membrane integrity.49 99mTc-based radiotracers (sestamibi and tetrofosmin) have almost no redistribution when compared to 201Tl,50 and different approaches are suggested to increase sensitivity for viability detection.
++
While resting 99mTc-based perfusion imaging can be used alone to define viability, several studies have shown that the use of nitrates improves viability detection when compared to rest 99mTc-based imaging and correlates with improvement after revascularization.39,51–55 Although the exact mechanism is not well understood, it is proposed that it may be related to improvement in blood flow secondary to vasodilation improving blood supply to the hibernating myocardium and therefor tracer uptake.54 Studies have also demonstrated that abnormal contractility can lead to perfusion defects with SPECT perfusion imaging.56,57 Incremental value can be achieved by adding electrocardiogram (ECG) gating and dobutamine to 99mTc-based imaging, enabling the assessment of both perfusion and contractile reserve in a single study.58
++
Dual-isotope imaging with 99mTc at stress and 201Tl at rest is also possible. The 201Tl rest portion is used for viability interpretation with the addition of late redistribution and reinjection to increase the test sensitivity. However, this protocol is associated with a significantly higher radiation dose and may not be the ideal primary test when other diagnostic options are available.41
+++
Imaging Interpretation of SPECT Viability Images
++
After choosing the image acquisition protocol according to patient characteristics and institutional availability, imaging interpretation should be performed carefully.
++
While resting 99mTc-based perfusion imaging can be used alone to define viability, 99mTc-based SPECT viability imaging interpretation can be enhanced by including both rest and post nitrate images read simultaneously.50 Persistent (fixed) defects with absence of tracer uptake noticed on both rest and post nitrate images are unlikely to recovery with revascularization, and represent the so-called nonviable myocardium (scar). Dysfunctional segments with tracer uptake similar to other segments with normal wall thickening may recover, and segments that improve tracer uptake with nitroglycerin administration are likely to recover (viable hibernating myocardium).50,59,60 For 99mTc-based SPECT, 50% to 55% of peak activity is usually defined as the cutoff to predict functional recovery and, therefore, to determine viability61 (Table 21-2). It is important to remember that 99mTc-based SPECT can also be performed to assess ischemia. In this case, images would be acquired in rest and post stress (exercise or pharmacological stress). A moderate to severe reversible defect in this case represents ischemia and may benefit from revascularization—viability imaging would not be needed. (Stress perfusion imaging is considered in a separate chapter.) In general, viability testing should be reserved for patients where there are persistent defects on stress perfusion imaging with mild or no ischemia or in patients where stress imaging may be considered to have increased risk such as severe multivessel disease or very severe LV dysfunction.
++
++
Similar approaches may be used when 201Tl SPECT imaging is performed. In the first step of the 201Tl protocol, images are acquired at stress and after rest and then, a third optional delayed image (redistribution) is acquired if a persistent defect is present on stress and rest images.50 After acquisition of the first set of images (stress/rest), images should be reviewed. If the defect is reversible, the interpretation is that myocardium is ischemic and therefore viable (and thus may benefit from revascularization). If the defect is persistent (fixed), then a redistribution image is acquired and interpreted.50 A persistent (fixed) severe defect (<50% of peak tracer uptake) present in all three images suggests a myocardium which is unlikely to recovery after revascularization (nonviable). A reversible defect on the third set of images (late redistribution) or delayed images of a rest-redistribution protocol, with tracer uptake 55% to 60% of peak activity,50,61 indicates the presence of viable hibernating myocardium which is likely to benefit from revascularization (Table 21-3).
++
+++
Positron Emission Tomography
++
The rationale for 18FDG PET use in viability assessment is based on the fundamentals of myocardial metabolism. The cardiomyocytes can use free fatty acid (FFA), glucose, lactate, pyruvate, and ketone as energy substrates, with the first two being the most important sources of energy.62–65 Fasting, adrenergic stimulation, ischemia, or insulin can shift the preferred energy substrate toward either FFA (in the case of fasting) or glucose.66–69
++
During fasting, FFAs are the preferred energy substrate. FFA plasma levels are increased as a consequence of low insulin levels and increased peripheral lipolysis in adipose tissue.65,69 However, during hyperglycemic states (postprandial), the higher insulin plasma level suppresses lipolysis 70 and stimulates myocardial glucose uptake, with glucose becoming the primary substrate.71 This process is primarily mediated by the glucose transporters 1 and 4 (GLUT 1 and 4). During ischemia and high insulin metabolic state, GLUT 1 and GLUT 4 are transported from intracellular storages to the plasma membrane to increase glucose uptake by the myocyte.72,73 Similarly, under adrenergic stimulation and ischemic conditions (myocardial ischemia), the FFA oxidation process decreases or may cease and glycolysis (anaerobic glycolysis) facilitates the use of glucose as the main source of myocardial energy.66,68,71 Glucose is the primary substrate in hibernating myocardium and is the basis for the utility of 18FDG PET for viability imaging.
++
18FDG is a glucose analog and is used clinically to assess and quantify glucose utilization. Although 18FDG is thought of as a marker of glucose metabolism, technically speaking, it is more specifically a direct measure of exogenous glucose uptake. After 18FDG is transported across cellular membranes, it is converted to FDG-6-phosphate and becomes trapped in the myocyte where it is trapped and does not continue down the metabolic pathway.74
+++
Imaging Protocols for FDG
++
Viability imaging protocols using PET are usually composed of two portions: rest perfusion imaging with 13-ammonia (13N) or 82-rubidium (82Rb) and metabolic imaging with 18FDG. However, 99mTc-based SPECT and 201Tl SPECT rest perfusion images can also be used to compare with metabolic 18FDG PET images in cases where PET perfusion with 13N or 82Rb are not available (preferably with attenuation correction of the SPECT MPI).75
++
Patient preparation is an important component of the 18FDG PET viability protocol. In a region of reduced perfusion, areas of 18FDG uptake indicate viable myocardium, while the absence of uptake indicates nonviable myocardium.50,76–79 Thus patient preparation is very important for the success of the viability study. Options include:
++
Fasting: Fasting is a simple approach but can lead to inferior image quality when compared to glucose loaded.80 The rationale behind this approach is that the normal myocardium will consume FFA while ischemic myocardium (viable) will preferentially use 18FDG and appear as a "hot spot." This approach is generally not recommended as an isolated technique.50,76 However, fasting is routinely needed as a part of the preparation before administration of glucose +/– insulin used in the three current methods (Fig. 21-5).
Glucose loading: Following a fasting period of at least 6 hours (preferable 12 hours), a glucose load is administered intravenously or orally.76 Due to its simplicity, this protocol is commonly used at many centers. By increasing glucose plasma levels, insulin release is stimulated which decreases FFA plasma levels and shifts the myocytes toward glucose (and therefore 18FDG) utilization. However, 20% to 25% of the patients with CAD may have poor image quality.50 This is most commonly problematic in diabetics who have impaired insulin release or insulin resistance. In such cases, intravenous (IV) insulin may need to be administered according to a sliding scale81 (45–60 minutes after glucose loading) repeating every 15 minutes targeting a glucose serum level between 100 and 140 mg/dL.76 FDG is administered approximately 1 hour after glucose load.
Hyperinsulinemic/euglycemic clamp: The hyperinsulinemic/euglycemic clamp protocol uses the intravenous infusion of both glucose and insulin. Blood glucose levels need to be monitored and the amount of glucose/insulin is tailored to each individual patient.50,76 Although this is a more time-consuming protocol and requires an experienced team, the image quality is superior to the standard glucose loading protocol.78,82 Some centers use this method routinely in all patients; others reserve it for all patients with diabetes.
Acipimox: Acipimox is a nicotinic acid derivative that inhibits peripheral lipolysis thus reducing FFA availability. This process indirectly forces myocardial utilization of glucose as preferable substrate by reducing circulating FFAs. Acipimox administration provides comparable image quality as the hyperinsulinemic/euglycemic IV clamp77,79 although nicotinic acid itself (i.e., Niacin) does not.78,82
++
++
After adequate patient preparation, 18FDG (5–10 mCi/185–370 MBq) is administered and image acquisition begins 40 to 60 minutes after 18FDG injection. Imaging can be performed using 2D or 3D scanners, in static, ECG-gated or list-mode and lasts 10 to 30 minutes. Attenuation correction should be applied and images reconstructed using an iterative statistical method. Preferable reconstructed pixel size is 2 to 3 mm but 3 to 4 mm is also acceptable.50,76
++
It is important to carefully review and align CT used for attenuation correction and emission images in all views (transaxial, coronal, and sagittal). Misaligned images can cause artifacts and misinterpretation. As for SPECT, 18FDG images are reoriented and displayed in short-axis (SAO), horizontal long-axis (HLA) and vertical long-axis (VLA) and metabolism images are normalized according to rest perfusion images. For interpretation, a combined assessment is performed and 18FDG scan metabolism images are compared with rest perfusion images in the same way we compare SPECT rest/stress images.76
+++
Imaging Interpretation of FDG
++
18FDG and rest (13N or 82Rb) myocardial PET perfusion images are reviewed simultaneously. 18FDG PET images can be also interpreted in conjunction with resting SPECT MPI.83 Care is needed with the later approach, especially when comparing a non–attenuation-corrected SPECT myocardial perfusion images with attenuation-corrected 18FDG PET images.74
++
Four patterns of flow/metabolism can be observed (Table 21-4)84: (1) normal myocardial perfusion and glucose metabolism (viable, not ischemic myocardium at rest), (2) mismatch with reduced perfusion and preserved (or partly preserved) metabolism (hibernating [viable] myocardium), (3) matched reduction in perfusion and metabolism (nonviable myocardium = scar), and (4) normal perfusion with reduced metabolism (reverse mismatch). Figures 21-6 and 21-7 show examples of clinical cases and their interpretation. Further details are also noted in the next section below.
++
++
++
++
Reverse mismatch is described in patients with left bundle branch block (LBBB) with altered septal metabolism in ischemic and nonischemic cardiomyopathy, in patients with diabetes where glucose uptake is impaired in nonischemic tissue,85 in repetitive stunning and early postrevascularization following acute MI.3,20,86
+++
Evidence for the Clinical Role of Viability Imaging in Ischemic Heart Disease
++
Nuclear, dobutamine echo and MRI viability imaging modalities have demonstrated ability to predict LV function recovery and outcome benefit in observational trials. In a previous comparative systematic review meta-analysis, 18FDG PET emerged as the most sensitive imaging technique to predict function recovery after revascularization,4 with a pooled analysis of 24 studies (756 patients) showing a mean sensitivity of 92%, specificity of 63%, and positive and negative predictive values of 74% and 87%, respectively.4 Differences in sensitivity and specificity between 18FDG PET, dobutamine echocardiography, SPECT with thallium (201Tl-SPECT) and SPECT with technetium (99mTc SPECT) are illustrated in Figure 21-8. While 18FDG PET is recognized as a sensitivity modality, dobutamine ECHO was the most specific for predicting the recovery of function.4 However, cardiac MRI was underrepresented in this prior analysis. More recent meta-analysis that considered cardiac MRI alone, showed delayed-enhancement (DE-MRI) to also be highly sensitive while low-dose dobutamine cardiac MRI was more specific.87
++
++
Providing prospective data for viability assessment is very important, but fewer studies are available. One randomized trial evaluated 18FDG PET-guided management in patients with severe LV dysfunction. In the PARR-2 (Positron emission tomography And Recovery following Revascularization Phase 2), 430 patients from 9 centers were randomized to undergo 18FDG PET or standard care before revascularization decisions.5 Patients were followed for cardiac death, MI and cardiac hospitalization at 1 year. A trend of benefit was noticed when using PET to assist with management decisions.5 The composite event was 30% in the PET arm and 36% in the standard care arm (relative risk 0.82; p = 0.16 and hazard ratio (HR) 0.78; p = 0.15).5 However, it was observed that clinicians did not uniformly adhere to the advice from imaging results. Therefore, a post-hoc analysis was performed in patients whose revascularization decisions were based on imaging results. In this analysis, a significant reduction in adverse outcome was observed in the 18FDG PET arm (HR 0.62; p = 0.019).5
++
Further post-hoc sub-group analyses of PARR-2 showed that not only is adherence important, but also the amount of hibernating myocardium.6 Among 182 patients in the PET arm, increasing amounts of mismatch (hibernating myocardium) was associated with increasing likelihood of benefit from revascularization. A threshold of 7% LV mismatch appeared to differentiate between those who benefit from revascularization versus optimal medical therapy (Fig. 21-9).6 A subsequent PARR-2 substudy focusing on an experienced cardiac PET center (the Ottawa-FIVE study) included 111 patients from a center with experience, ready access to 18FDG PET and integration between imaging, HF and revascularization teams, supporting that PET-assisted management improved outcome.7 The cumulative proportion of events in the standard care group was 41% versus 19% in the PET-assisted management group.7 In a multivariable Cox proportional hazards regression, the 18FDG PET–assisted strategy showed benefit (HR 0.34; confidence interval, 0.16–0.72; p = 0.005). The 5-year follow-up of PARR-2, demonstrated similar findings as the 1-year study regarding the use of FDG PET to direct clinical management (revascularization vs. medical therapy). When PET recommendations were followed, primary outcome (cardiac death, MI, or cardiac hospitalization) improved, with an HR of 0.73 (95% confidence interval 0.54–0.99; p = 0.042).10
++
++
Revascularization decisions based on the amount of hibernating myocardium found by PARR-2 were similar to the findings of previous work by Di Carli et al. (5%)12 and Lee et al. (7.6%).13 Ling et al. also described the relationship between increasing amounts of hibernating myocardium and increasing likelihood of revascularization benefit for reducing all-cause death and found that the threshold was approximately 10% (Fig. 21-10).11
++
++
Despite all the current supporting literature regarding the benefit of viability imaging-guided revascularization5–8,88–91 and its ability to cost-effectively select patients who would likely improve HF symptoms,9,92,93 the STICH trial suggested the opposite.94 Among the 1212 patients enrolled in the main STICH trial94 to compare optimal medical therapy (OMT) with revascularization plus OMT, 601 underwent myocardial viability assessment with either SPECT or dobutamine echocardiography.94 Viability imaging was not part of the randomization. Of this group, 303 received optimal medical therapy alone and 298 received optimal medical therapy plus coronary artery bypass graft (CABG). Although HR for death in the group with viable myocardium was 0.64 (p = 0.003), after adjustment for patients' baseline variables this association was no longer significant (p = 0.21), suggesting no benefit of myocardial viability assessment in this population.94
++
PARR-2 and STICH trials appear to yield conflicting interpretation, but their results should be interpreted with caution.90 The trial design and populations differed between the two studies. PARR-2 was a randomized trial that enrolled subjects with uncertain revascularization plans, and therefore viability assessment was clinically needed.5 Conversely, in the STICH study, the allocation of patients for viability assessment was not randomized and all subjects were already accepted for revascularization (and thus had suitable coronary anatomy). There were also differences in comorbidities (specifically renal dysfunction and prior CABG).90,94
++
A smaller randomized trial by Siebelink et al.95 compared 13N-ammonia/18FDG PET with 99mTc-sestamibi SPECT viability to guide therapy. They enrolled a total of 112 patients and observed no difference in cardiac event-free survival (cardiac death, MI, and revascularization) between the PET versus SPECT (11 vs. 13, p = NS, respectively).95 However, this study also suffered from patient selection bias since only 35% of the population had severe LV dysfunction. Since there was a low rate of total events, the study was likely underpowered to detect a difference between groups.96
++
Although currently the literature has these mixed results, current recommendations support the use of viability imaging assessment prior to revascularization in patients with CAD and LV dysfunction.97–99 The level of evidence, class of recommendation, and appropriate use criteria for radionuclide imaging are outlined in Table 21-5.
++
++
Despite the recognition of the importance of viability assessment, updated practical guidelines are needed. A multicenter trial100 and registry101 are currently being conducted and will hopefully address many unanswered questions that still exist. Alternative Imaging Modalities Ischemic Heart Failure (AIMI-HF) IMAGE HF includes centers from North America, Europe, and Latin America and is a large trial comparing standard investigations with SPECT to advanced imaging modalities (PET and cardiac magnetic resonance [CMR]).100 Currently, most centers rely on their local experience to guide therapy which may be subjective and biased by limited personal experiences of the imaging physician, cardiologist, and cardiac surgeon. The standardized algorithm for reporting and guiding recommendations being used in the AIMI-HF trial of the IMAGE HF program is shown in Figure 21-11.
++
++
Recently published, the STICH Extension Study (STICHES),102 a 10-year follow-up of the original trial,94 showed benefit of revascularization for all cause of death, cardiovascular death, and cardiovascular hospitalization. The STICHE study did not assess the impact of viability imaging but the knowledge of greater benefit of revascularization against medical treatment in a long-term follow-up highlights the importance of proper assessment of each patient with ischemic cardiomyopathy weighing long-term benefit versus procedural risk. Viability assessment may have value in helping to define the extent of hibernating myocardium, and thus may be an important parameter to consider with other factors in decision making for revascularization.6,11,99
+++
When Should Viability Imaging Be Used?
++
Although the importance of viability imaging is recognized, in clinical practice it can be difficult to determine the best approach according to each clinical scenario. The first step is to understand patient symptoms, clinical history, and other imaging results.
++
As exemplified in a previous review by Di Carli et al.,103 defining treatment strategy may be challenging for some patients. Consider the contrast, for example, whereby in (i) a 50-year-old patient with ischemic cardiomyopathy, three-vessel CAD, LV ejection fraction of 30%, angina class III, no other comorbidities, and good target vessels, the decision to revascularization would generally be straightforward for most physicians. As opposed to a clinical scenario whereby (ii) an elderly patient with NYHA class III, no angina, multiple comorbidities (including renal impairment, diabetes and COPD), severe LV dysfunction, LV dilatation, history of previous CABG, and mediocre target vessels; the risks versus benefit of revascularization may be less clear.103 Viability assessment is a great tool to assist the decision making in such patients.90 Figure 21-12 suggests an algorithm for use of viability imaging adapted from prior work.104,105 A pragmatic approach at least for FDG PET imaging is considered below.106
++
++
In our hands we consider FDG PET imaging for patients with:
++
Known or strongly suspected IHD*
NYHA II*
Moderate–severe LV dysfunction (LVEF <40%)*
Moderate-to-large persistent perfusion defects–no significant ischemia*
+/– Significant comorbidities +/or poor distal targets
+/or Equivocal viability results on another test
++
In our hands, FDG PET viability imaging is often not considered in patients with:
++
Predominantly Angina CCS >II
Normal or mild LV dysfunction
Critical LMCA (left main coronary artery) disease
Good targets
Documented moderate or severe ischemia
Minimal or no comorbidities
++
After deciding on the need for viability imaging, it is necessary to choose the most appropriate test for each patient. Although some studies report superior sensitivity for 18FDG PET4 and specificity for dobutamine echocardiogram and MRI,4,87 there is lack of head-to-head study to help us decide the "right test for the right patient." All techniques (18FDG PET, dobutamine echocardiography, 201Tl-SPECT, 99mTc-SPECT, and cardiac MRI) can be used for viability assessment and guidance in decision making. Contraindications for modalities (e.g., pacemakers and implantable cardiac defibrillators for MRI), technique availability, and center expertise are key factors to take into account when deciding about the right test for the right patient. A practical approach to consider regarding which test to use in which patient circumstances (used by the author) is as follows:
++
Normal or mild LV dysfunction—viability imaging is rarely needed.
Moderate LV dysfunction—any method can be considered depending on availability and local expertise.
Very severe LV dysfunction, consider nuclear methods (SPECT, FDG PET) or MRI which is more sensitive than contractile reserve.4,87,107
Renal failure (GFR <30) or implanted devices—avoid MRI.
LMCA disease or severe proximal three-vessel disease—avoid dobutamine.
Equivocal results on another viability test or negative results on another viability test where certainty is needed to completely rule of viability—consider FDG PET or MRI as they are highly sensitive methods.4,87,107
+
++