Thank you very much for the opportunity to present. It's my pleasure to be discussing non-perfusion imaging viability through a case review. My name is Serge Harp. I'm one of the cardiologists at the Cleveland Clinic in Cleveland, Ohio. I have nothing to disclose. What are the learning objectives of my presentation? One, understand the imaging principle behind FDG PET imaging. Two, differentiate scar tissue from hibernating tissue, which is the main goal of FDG PET imaging. Three, discuss patient preparation for the test, which is critical. And then finally, we're going to discuss PET imaging applications through a series of cases. I have six cases to present. As you all know, FDG PET imaging is the only FDA-approved technique to assess, to image myocardial viability. It's relying on FDG, which stands for fluorodeoxyglucose. It has a similar chemical morphology to glucose, so it competes with it for transport and phosphorylation. We all know that under fasting and aerobic conditions, fatty acids are the preferred fuel in the heart. However, in postprandial states, glucose becomes a preferred energy substrate. And we're going to use this when we discuss the principles behind it. What is the goal of FDG PET imaging? It's mainly to enter the following question. Is the reduction of myocardial blood flow at rest due to scarred, meaning non-viable, or hibernating, meaning viable myocardium? So this is the main question that we're trying to answer. It is important because it has direct clinical application. Hibernating myocardium should be revascularized because it may improve with revascularization, while scarred myocardium typically reflects dead myocardium, and then probably not worth revascularization. What is hibernation? Hibernation is a state of chronic reduction of myocardial blood flow. So the myocardium becomes highly dependent on glucose, and we take advantage of this because it will uptake FDG so we can image it. However, scarred myocardium means myocardial cell death, so there will be no FDG uptake. And this is the main differentiation that we rely on in PET FDG imaging. So let's take this rest static image showing a decreased perfusion. When we give the FDG, if there is uptake, we call it hibernating myocardium, so there is mismatch between the decreased perfusion on the rest static image and the FDG image. However, if both perfusion defects match, meaning there is a resting perfusion defect, and then on FDG there is no uptake, this is reflective of a scarred myocardium. When we do not need to image with FDG, if you have normal myocardium, so myocardium with normal resting perfusion is viable, so no need for FDG imaging. And then if you have ischemic myocardium, it's also viable and no need for FDG, as you can see in the polar plots above, on the resting image we have normal perfusion at rest in the infrared segments, which decreases with stress, so ischemia. Ischemic myocardium is viable, so no need for FDG. So the only application for FDG is when you have a resting perfusion defect, and then we can, to help us define if it's hibernating or viable myocardium. What about patient preparation? It's critical for the test, and the goal is to favor the metabolism of glucose over fatty acids, because then we will maximize the FDG uptake in the living myocardium. It differs between non-diabetic and diabetic patients. I'm not going to go through the details of the protocols, because I will show them in different, in other slides, but typically for non-diabetics we rely on the patient's own insulin response, and we would like this insulin response because it reduces fatty acids and it favors glucose uptake, so we can take advantage of it and image FDG. For diabetics, we're going to use exogenous insulin, along with close monitoring of blood glucose. For all patients, we're going to recommend NPO after midnight the day prior, including insulin and oral anti-diabetic medications. These are the protocols that, according to the ASNIC imaging guidelines, again I'm not going to go through this, you can review it separately, and this is an example of our own protocol at Cleveland Clinic, which is a minor variation to the guidelines. Again, we distinguish non-diabetic from diabetic patients, and based on the blood glucose, we decide on the amount of dextrose and insulin to give before FDG. This is the protocol for diabetic patients. Now let's move on to the applications, which is the key topic that I would like to discuss, and I will present the following six cases. Let's start with case number one. This was a 61-year-old gentleman with an occluded proximal LAD and severe LV dysfunction. His EF by echo is 10 to 15%. He was symptomatic, and an attempt was performed to open up the LAD, however unfortunately it failed. So he was being reconsidered for another high-risk PCI to the LAD, given his continued symptoms. Because of the high stakes of the procedure, we decided to refer him for PET-REST viability to see if there is enough viability to justify this high-risk procedure. So let's look at his images. So these are the REST-FTG images. In the upper half of the screen, you can see the short axis images, in the lower half, the horizontal long axis, and then the vertical long axis images. And in each of these panels, the lower part is FDG and the upper part is RESTing. So RESTing and FDG, RESTing and FDG in horizontal long axis, RESTing and FDG in vertical long axis. And what we can click and notice is what is pointed out by the yellow arrows, that there is a large perfusion defect at REST, predominantly in the LAD territory, in the anterior septal and anterior septal segments. When we look at the FDG, there is a corresponding no uptake of FDG. So they are matching. So it's scarred. You can see this on the vertical long axis and the horizontal long axis and the vertical long axis. In addition, we see this dark structure within the MV myocardium, which is pointed out by the red arrows, which I will be discussing shortly afterwards. When we look at the gated images, you can see that this is a severely dysfunctional left ventricle with one motion abnormalities predominantly in a large LAD territory. Looking at this echo, similar, so severe LV dysfunction with LAD one motion abnormalities and the dense structure that we saw on the PET images corresponded to this layer of thrombus on echo. So if we want to summarize the case using the polar plots, what we can see is that there is a severe reduction in counts on the rest and the perfusion images, resting images, and with a corresponding no uptake of FDG on PET FDG imaging. So a matching defect, so a large scar in the territory of the LAD. The left ventricle is severely dilated with severely reduced function. The measured EF was 15% and the large layer thrombus. Because of the large amount of scar and the lack of evidence of significant viable myocardium, the patient was not offered revascularization, alternatively he was referred for advanced heart failure management. Let's now move on to case number two. This was a 60 year old man with a history of three vessel coronary artery disease and ischemic cardiomyopathy with left ventricular dysfunction and EF of 35% by echo. He underwent a cardiac catheterization, which showed 100% middle LAD stenosis and 60% proximal RCA with 70% proximal left circumflex, so three vessel disease. For proceeding with revascularization, particularly given his low EF, we decided to proceed with a PET rest-rest viability imaging to see if it's worth proceeding with revascularization. So this is the imaging test. On the upper panel, you see the short axis, on the lower panel to the left, the vertical long axis, and to the right, the horizontal long axis. Now in these images, you have three rows, and then the lowest one is the FDG, the middle one is the rest, and the upper one is the stress. And this is for the short axis, the vertical long axis, and the horizontal long axis. As we can see at rest, the infralateral segment has normal perfusion, however with stress there is decreased perfusion. So ischemia, and very interestingly, you can see that on FDG there is uptake in that same region. This is what we call ischemic memory, and I will be discussing it shortly. When we look at the inferior segments, similar, normal perfusion at rest, severe defect with stress, so ischemia, and then on FDG corresponding uptake. Now when we look at the septal and apical segments, particularly the LAD territory, we can see that there is a resting perfusion defect. It doesn't significantly change with stress, so no significant ischemia. When we look at this resting perfusion reflex is hibernating or scar tissue, we can see that it takes FDG, so there is mismatch, so there is hibernation in the LAD territory. We look at the gated images, the LV function, as you can see, is moderately reduced, and there are wall motion abnormalities predominantly in the LAD territory. Using the polar plots to summarize the case, we can see that at rest there is a defect predominantly in the LAD territory. It takes up FDG, so this is hibernating myocardium, and then as you remember, there is ischemia in the RCA, and the less circumflex territories, with ischemic memory, meaning that there is FDG uptake in these segments that are ischemic, and we're going to discuss this. So RCA and LCX ischemia, large hibernation in the LAD, with a left ventricle that's moderately dilated and moderately reduced in systolic function. Because of the large amount of hibernation and ischemia, we will see later that the patient underwent revascularization. However, I would like to stop and discuss ischemic memory. It is a metabolic switch from fatty acids to glucose, which seems pivotal in preserving myocardial viability, and is one of the earliest adaptive response to myocardial ischemia. So how can we see this on PET imaging? Again, so when there is myocardial ischemia, the myocardium switches to glucose, so it will take FDG. Let's look at this example. So you can see that there is normal perfusion at rest in the inferolateral segments, with stress, there is decreased perfusion, so ischemia, so the myocardial ischemic is going to switch to FDG. That's why when you give FDG, it takes it up. This is what's called ischemic memory, and which was highlighted in this case. So again, as you may remember, large area of hibernation, ischemia, so the patient was referred for coronary revascularization, he underwent coronary artery bypass grafting, with multiple bypasses, as you can see, LIMA-12AD, vein graft to NOM, another one to APL branch, vein graft to the ramus, and then vein graft to an acute margin. So he underwent complete revascularization, he felt significantly better clinically, and then echoperformed a few months afterwards, showed an improvement in his LV function, with an LVEF of 40%. Let's now move to case number three. This was a 70-year-old man, who had prior aortic valve replacement, and one vessel coronary artery bypass graft, with a vein graft to the diagonal, this was many years ago, and he was presenting with severe multivessel coronary artery disease, with LV dysfunction and EF of 30%, he also had severe ischemic mitral regurgitation. He underwent left heart catheterization, which showed severe LAD and RCA disease. The CERP was occluded, and the vein graft to the diagonal was occluded also, unfortunately. So the main question was, given his LV dysfunction and severe multivessel disease, whether we should proceed with revascularization, so he was referred for PET rest viability. Let's move now to look at the images. In the upper half of the screen shows you the short axis images, then you have the horizontal long axis, and the vertical long axis. We have, for each, we have the upper panel shows the resting images, and then the lower shows the FTG uptake, for the short axis, the horizontal long axis, and the vertical long axis. As you can see, and as pointed by the arrow, there is severe decreased perfusion at rest in the inferior infralateral segment, so predominantly in the circumflex territory, which correspond to pointing no uptake in the FTG, with FTG imaging. So a matching defect, so a scar. However, as you can see, the remaining myocardium has normal perfusion at rest, so is therefore viable. So we have probably a medium scar in the circumflex territory, but the remaining myocardium despite all its extensive disease, had normal perfusion at rest, and thus is viable. Let's go back and look at his gated images. So here you can see that there are one-motion abnormalities in the predominantly in the subterritory, inferior lateral and inferior segments, and the remaining myocardium had relatively preserved function. Overall, the LP function was moderately reduced, to severely reduced. So in summary, if we want to summarize the case, there is a resting perfusion defect in the inferior infralateral segments, with a matching lack of FTG uptake, so a medium scar. However, you can see that the remaining myocardium, all the entire segments have normal perfusion at rest, and are thus viable. Because of this, he underwent reduced anatomy, and coronary artery bypass grafting, with V-metolyd, vein graft to the OM, and a vein graft to the PDA, along with aortic valve replacement, and mitral valve repair for his ischemic mitral regurgitation. He felt significantly better on follow-up, and had his EF improved with revascularization. Now you may think, why did we revascularize the circumflex territory, because it had a scar? Well, he went for bypass, so we opted for full revascularization, we're putting a V-metolyd, a vein graft to the PDA, so we also opted to do a vein graft to the OM. Case number four is a 74-year-old lady with type 1 diabetes and peripheral artery disease, who was admitted to an outside hospital with non-STEMI, she had a complicated course, but underwent a left heart catalyzation, which showed severe 3-vessel disease, with an occluded RCA, and severe LAD and circumflex disease. And the main, her ECHO showed an EF of 40%, so she was referred for PET rest viability, with the main question is, should we, should she undergo high-risk coronary artery bypass grafting, because she had a complicated course. So we proceeded with the rest viability PET, short axis in the upper half, horizontal long axis and vertical long axis in the lower half, and you can, upper panel is the resting images and below is the FTG, in each of the views. So we can see that in the lateral segments, there is a decreased perfusion at rest, with a corresponding FTG uptake, so a mismatch, so hibernation, this is a large area of hibernation as you can see on these images. On the gated images, the LV function was down, predominantly with one motion abnormalities in the circumflex territory, in the lateral segments. So no scar, the resting perfusion defects had FTG uptake, so it was hibernating, so large area of hibernation in the circumflex territory, the LVF was down to 30%. These are the portal plots, because the, due to the lack of scar and the vast area of hibernation with normal perfusion in the other segments, the patient underwent 3-vessel coronary artery bypass grafting, revascularizing the LAD, the CERC, and the RCA territories, clinical tissue improved, and the repeat echo showed an improvement in LV function with an EF of 50%. Let us now move to case number 5. This was a 49-year-old man with severe multivessel coronary artery disease and ischemic cardiomyopathy with an EF of 25% who presented with ventricular tachycardia. And there were two key questions for his care. One is the VT related to ischemia, does he have ischemia that's driving the VT, or is it a scar-mediated VT? And the second question was, should revascularization be pursued? So PET can answer both of these questions, so a rest-stress viability scan was ordered. Let's now look at the images. So again, the upper panel is, upper half of the screen is short axis, and the left lower is vertical long axis, while the right lower is a horizontal long axis. And now we have three rows. So the rest is the middle, on top of it is the stress, and below it is the FTG uptake. So this is true for all these images. So let's look first at the LAD territory, so along the anterior and septal segments, you can see that there is a decreased perfusion at rest, so a perfusion defect at rest. With stress, there is no significant change, and on FTG there is no uptake. So if we compare rest and stress, we know that there is no significant ischemia, and then if we compare rest and FTG, we know that there is no significant viability. So predominantly an LAD scar, and you can see this on the horizontal and the vertical long axis. In addition, when we look at the inferior segments, very similar, decreased perfusion at rest, no change with stress, so no significant ischemia, and on FTG imaging, there was no FTG uptake. So also an RCA scar, and you can also see it here clearly on the vertical long axis images. When we looked at the gated images, you can see that the LV is severely down, with warm motion abnormalities predominantly in the LAD and RCA territories. So in summary, if we look at the polar plots, you can see that in the middle you have the resting polar plot, and then to this side is the stress, and to the other side is the FTG. When we compare rest and stress, there is no significant difference, so no significant ischemia. When we compare the rest and FTG, there is a matching defect, so a large scar, both in the LAD and RCA territories. The left ventricle was severely dilated, with a severely reduced systolic function, the AEF was measured as 20%. Because of the large scars, the lack of hibernation, the lack of ischemia, we didn't feel that the patient would benefit from revascularization, so he was referred for advanced heart failure management. Now we come to our last case, case number six. This was a 62-year-old man with a complicated coronary history. He actually had coronary artery bypass grafting, followed by stenting of two of the vein grafts, the vein graft of the RCA and the vein graft of the OM. His LIMA to LAD remained patent. He was symptomatic, though his symptoms were a bit atypical, not related to exertion and different from the symptoms that he had prior to the coronary artery bypass grafting. His AEF was around 45%. He underwent cardiac catheterization, which revealed severe disease in the vein graft of the RCA. Remember, this was stented before, so it was a recurrent disease. Fortunately for him, his LIMA was patent. However, the vein graft to the OM was occluded. And the main question was, should further revascularization be pursued? In this case, probably, given that the LIMA is patent and one of the vein grafts is completely occluded, the main question was, should we intervene on the vein graft of the RCA? So we decided to proceed with a PET-REST test to understand if his symptoms are related to ischemia, given that they were atypical, and also to look at the viability to see if it's worth revascularization. So on the PET-REST test, again, upper half of the screen is the short axis, left lower is horizontal long axis, and right lower is vertical long axis. And for each of these images, you have three panels, three rows. In the middle is the resting image, and on top is the stress, and below is the FQG. You can see that in the entorhateral segments, as pointed out by the yellow arrows, there is normal perfusion at rest. With stress, there is decreased perfusion, so there is a marked difference. So there is significant entorhateral ischemia, and this is also depicted on the horizontal long axis images. Now, if we look at the inferior segment, at rest there is a decreased perfusion. With stress, this perfusion defect becomes worse, so there is ischemia. And very interestingly, if you look at this and you compare the rest images to the FQG, you can see that there is FQG uptake, so this area is hibernating. So in the RCA territory, there is a mixture between hibernation and ischemia, and there is entorhateral ischemia also. So when we look at the gated images, we can see that his LV function was not terribly down, and he had predominantly warmush abnormalities in the inferior segments. So if we want to summarize this case, again, the polar plot in the middle is the rest, to the left is the stress, and to the right is the FQG. You can see that he has entorhateral ischemia, with decreased perfusion on stress compared to rest, and then when we compare the rest to the FQG, there is decreased perfusion in the inferior segments, but corresponding FQG uptake, so mismatch, so hibernation. Also very interestingly, if you look at this segment, the inferior segments, and you look at stress, there is ischemia, so it gets worse. So we have moderate entorhateral ischemia, a medium area of hibernation in the RCA with mild associated ischemia, and the ventricle had only mildly reduced the systolic function with an EF of 50%. For these reasons, he underwent successful stenting to the vein graft to the RCA, and had significant improvement in his symptoms. Of course, we opted to first start with a stenting to the vein graft to the RCA, because as you remember, he had a patient pre-med with LID, and the other vein graft was completely occluded, and fortunately, his symptoms improved with this strategy. Let me summarize the teaching points that we discussed. Differentiating between normal, ischemic, hibernating, and scarred myocardial mousse pad provides important information that is critical for clinical decisions, particularly regarding revascularization. Please remember that patient preparation is critical for optimal pad viability FTG imaging. It relies on the fact that FTG competes with glucose. The goal for patient preparation is to favor cardiac cellular metabolism of glucose over fatty acids, so we want the myocardium to use glucose, so we can give FTG and it will take it. And as we discussed in the protocols, briefly that particular attention for diabetic patients in terms of glucose monitoring and insulin use. So in the setting of a resting perfusion defect, this is where FTG imaging is helpful. FTG imaging showing a matching defect, so meaning that there is no FTG update, so decreased perfusion at rest and no FTG update, a matching defect, there is no viability, it's scarred. However, if there is a mismatch, meaning decreased perfusion at rest with FTG uptake, there is viability and it's worth revascularization. FTG imaging is unnecessary if there is no resting perfusion defect, we discussed this. So normal resting perfusion, or if we see ischemia, or if we see hibernation, all of these reflect viable myocardium. We remember the case where there was, you know, a small to moderate scar, but the remaining myocardium, the entire remaining myocardium had normal resting perfusion and we proceeded with revascularization and the patient improved. So normal perfusion, ischemia, and hibernation all reflect myocardium that's viable and is likely to benefit from revascularization. However, scarred myocardium is non-viable and unlikely to benefit from revascularization. Thank you very much for your attention, it's been my pleasure to present this talk to you. Thank you.

non-perfusion imaging

myocardial viability

FDG PET imaging

scar tissue

hibernating tissue

patient preparation