Hello, this is Gary Heller, and module 22 is how to set up a cardiovascular PET laboratory. This is an important aspect of cardiac PET because there are several differences between running a cardiac PET program and SPECT. Some of these deal with things such as do I have the patient volume, but also build out and training becomes important. So the objective today is to examine all those issues. These are my disclosures. And here are the learning objectives. We want to describe the thought process for establishing a PET lab. The second part of this is describing the patient indications, which is very important before you move into PET. And then some other considerations such as the space for the camera, tracer choices, and what you're going to use for cardiac PET. And then finally, training is extremely important for both technologists and physicians, as well as the community of healthcare providers who will be referring to and getting the PET patients back. So let's think about it this way. What are the considerations for starting a cardiovascular PET program? The first and very important is does your practice have the volume and who are the patients? So I just want to emphasize that because currently you would continue using SPECT in some patients and one will need to choose the number and kind of patients who would go through a cardiac PET study instead of SPECT. The second part of this is to define the goals for the program. I'll get into that in a little bit, as well as considerations for tracers. Once you decide on the goals and the tracers, then space considerations become very important. And that has to do with both line source PET, as well as PET-CT will go into that. As I mentioned previously, the training and education for technologists is a critical part of a successful PET program. So let's look at the first part of that, and that is, do I have enough patients for a cardiac PET program? So let's look at what you need to think about. The first is the driving force of any PET program will be PET perfusion with blood flow. And then as we get into what other things you can do, but if you don't have enough patients for PET perfusion and blood flow, then it's really not wise to begin with programming. A few years ago, Dr. Bateman, Dilsizian, and colleagues established a position statement from both ESSIG and Society of Nuclear Medicine, describing what they felt and the society's feel are the strengths of cardiac PET and why you should be doing a program. And I'll go through these very briefly because you probably know them from other presentations. However, they are very important in considering what sort of patients you want in your laboratory. High diagnostic accuracy, high quality images, low radiation exposure, short acquisition protocols, strong prognostic power, and quantification of myocardial blood flow, which may be quite new to you and your colleagues. So let's look at PET accuracy first. We all think that in obese patients, there's a better specificity. However, the data really confirms that both sensitivity and specificity are an important consideration for this. So think in those lines, I want the best diagnostic accuracy for my patient, both in identifying coronary disease and excluding coronary disease. There are ample data to show that PET is superior, particularly to SPECT in both aspects. This is one of the studies by Parker and others published almost 10 years ago now, showing that the bottom part of this, not only a higher specificity that you're used to, here's PET down here, but a significantly higher sensitivity. So this becomes very important in considering patients that you want to refer to the PET lab. We all think of PET as a higher quality image study, and this is really true. The reasons are better spatial resolution, higher counting efficiencies, and attenuation correction on all studies, which is currently not done in standard SPECT laboratories. So what this means is higher quality images, easier to interpret, greater interpretive confidence. In other words, you're very convinced that this study is either normal or abnormal. Fewer false positives and equivocal studies, which are really a major issue for cardiac SPECT imaging. And what this means is that your health care providers, those referring to your laboratory, will have better confidence in the imaging results, and as a result, you may see your PET program actually grow in subsequent months and years. Radiation exposure is a very important aspect for society, as well as health care providers, patients, and the general public as well. Here's what we know about radiation exposure right now. This is from IEC data, the Intersocietal Accreditation Commission. The estimation is that 95% of laboratories, if you do a rest-stress technetium study, the radiation exposure is 12 to 15 millisievers. We've certainly recommended against using thallium, either thallium alone or dual isotope imaging, but the radiation exposure is 20 to 25 millisievers. And importantly, recent data shows that only 5% of laboratories are using less than 9 millisievers per patient, which is the goal of many societies. Also 10% are above 20% millisievers. So these are the spec data that's very current. Let's turn now to PET data. What is different? Well, the mean radiation exposure is actually, for rubidium, is 3.6 millisievers for ammonia, which is less often used, but still in many laboratories through the country, 3.8 millisievers. And this all is way below the recommended 9 millisievers. So spec higher than 9 millisievers, all PET laboratories very low. This is an illustration of exactly that. So you see in the blue line down here, this is the recommendations of ASTIC. On the left side are spec data from 2012, 2015, and dual isotope 2015, way above recommended standards. All PET studies in that same imaging era are way below. And in fact, many people feel that the magic number is around 3.5 millisievers, which creates no DNA damage. So that's kind of the goal in PET. Using standard technologies is actually there. Using 3D, which we'll discuss in a moment, actually goes below 3 millisievers. The other aspect that these committees have emphasized is the quickness of a PET protocol. I'm focused on rubidium because 95% of laboratories are doing this. Rather than the usual one to three hour protocol, the elapsed time for a PET CT rubidium protocol is 25 minutes, as we discussed before. The radiation exposure is two to six millisievers. This is the protocol itself. I won't go into detail. I'm sure that's been explained in other presentations. I will say that ragadenosine is the primary agent that's being used, although people do use dobutamine, as well as tiperidamol. Adenosine is discouraged because of patient motion. So with all of that in mind, what I skipped over was myocardial blood flow, because we'll talk about that later on. But the question is, who are the patients who would benefit? The information statement of 2016 by Bateman and others I referred to actually took considerations of SPECT versus PET and made some recommendations. They essentially made two categories. One is where cardiac PET is preferred, and the second is recommended. So let's go to preferred. Now that's a pretty bold statement. In other words, if you have SPECT and PET in the same office, PET is actually preferred according to the societal recommendations. And this is in all patients unable to complete exercise stress. In other words, all pharmacologic stress patients, because of the reasons I just outlined, should undergo PET rather than SPECT imaging. That's a pretty important statement, and if you begin a program in your laboratory, you need to consider that shift into the PET laboratory. So preferred is a very important category, but also recommended. Now this is another category. What it's saying is that if you have it available, you really need to think about this. And some of these are pretty obvious. If you had a study that was done by some other modality that was inadequate, then obviously true. We all know body characteristics are very important, and rather than doing a SPECT first, you probably need to be moving into PET as the first choice. Younger patients who have established CAD because of likelihood of overall radiation exposure really should be thought about. High-risk patients in which you don't want to miss coronary disease, such as renal failure, diabetes, hypertension. And then blood flow. Remember, blood flow is done routinely with both rubidium and ammonia in all laboratories now. And if you see the value of blood flow in your practice, that's a consideration to moving towards cardiac PET. So those are the patients that you might think about referring. How does that fit into your practice? The first is the volume depends on multiple factors, such as what can I accommodate? In other words, if you are in a hospital-based setting and you can only do two or three patients a day, that's a different feature if you're in a standalone private practice group that has access basically 24-7, then you might be looking at too much volume. So the question you should ask is, what are the percentage of patients who are far in stress? Generally, it's around 50% of patients. So half of your patients might be candidates for cardiac PET with blood flow imaging. And then there's this other group of recommended I just went through. My experience with other laboratories is that that probably accounts for maybe 20% of the PET patients going in a particular laboratory. So overall, you could be looking at 50% to 70% of patients shifting away from SPECT and even some patients that are new. The other consideration is, what's the percentage of Medicare patients in your practice? And I say this because generally, Medicare allows the health care provider to determine whether the study will be PET or SPECT. We already know the reasons why, but the Medicare is saying you decide as long as they meet appropriate use criteria. It's also important if you're going to initiate a program to begin working with the insurance companies. And this gets easier and easier because there are PET laboratories in every state now and multiple PET laboratories in most states. So they've already been through that. But it's important to begin a conversation with them. And the first thing you would do is bring the position statement to their attention. When a society recommends that PET is preferred, then that resonates with insurance companies. So the bottom line is, and this is sort of a rough figure, but you really should be doing a minimum of three to four patients per day. So this is a thought process that you need to go through or your practice needs to go through when you're beginning a PET program. If you're way below this, then you have to think of other alternatives. The next part of this is, okay, I've got enough volume, but what are the goals of the program? So I just want to run through this with you because there are multiple things you can do with cardiac PET. And the first is, do you want this to be cardiovascular PET? Or do you also want to get involved in CT, at least with calcium scoring? And if that's the case, then you're probably moving in the direction of instrumentation that is PET-CT as opposed to standalone PET cameras. So if you are doing cardiovascular PET alone, remember there's options, perfusion we talked about, perfusion with blood flow we talked about, and I'll go through a little more description. Basically everyone is doing that now, but also remember you can do viability assessment in sarcoid imaging, as well as all of what I just mentioned, as well as viability, sarcoid and infection imaging. Now, generally speaking, that's left for a higher volume laboratory, let's say a practice of say 20 cardiologists who have issues of infection, sarcoid viability. And the other consideration we'll get into in a few minutes is, what about the future? These studies, viability, sarcoid infection, need F18 imaging in the form of FDG, perfusion doesn't. However, Florpiridis, which is in the later stages of a phase three trial, will probably become available in the next few years. And so the consideration is, what am I going to do when an F18 agent becomes available for perfusion? So what are your goals for the program and why does it matter? Several reasons. One is, as I mentioned before, the type of instrumentation depends on whether or not you're going to use an F18 agent or not. The second is floor space, PET-CT tends to have a larger camera situation, which may mean reinforcing the floor or at least a larger space. And again, you need to think about that. Also remember with F18, even with FDG right now, you're probably going to need some sort of a weight room, which is lead line. A patient, particularly with Florpiridis, gets injected, there's a wait time in between this also with F18, FDG glucose, so you're not imaging them right away. So a weight room that's lead line might be an important consideration. And when you're doing a build out, that's the time to think about that. So let's move on to tracers. I've kind of danced around that, but let's give it a little more consideration. Right now, I would say 93, 95% of laboratories are doing rubidium. This is actually the least requirement for lead shielding. So if you say, I'm going to do rubidium only, not do any F18 imaging, then the requirements for lead aligning the walls does become less. But again, think about that in terms of what you're going to do in the future. Ammonia is also a short-lived protocol, and as a result of that, the lining of the room is probably about the same. You can do ammonia with both dedicated and PET-CT. And remember, you can do exercise for ammonia, but we'll talk about that in a moment. The big issue is, am I going to do FDG imaging? And my own thought process is that as fluorocardiaz moves forward and you see the value of PET, you may want to start doing patients that you might have referred for spec because they can exercise. Well, maybe those will be in your PET lab in a few years. Just remember, there are other F18 tracers in development. So the story hasn't ended yet. And if you're going through the time and expense of developing a PET program, you probably want to think about the future in the next five to 10 years. So here's a few more considerations. I'll just run through this briefly with you. But here's the thought process. Rubidium has a very short half-life. As a result, the generator is actually in the room, in the patient, in a lead-lined storage area. So it's right there, but it's only pharmacologic stress. Ammonia has a longer half-life, so you potentially could do both foreign stress and exercise. I must say, my experience in laboratories, particularly in the clinical arena, doing ammonia or actually doing primarily pharmacologic stress exercises is quite a bit trickier to do. Both of these are pretty fast protocols. With rubidium, it's 25 to 35 minutes. Ammonia is a little longer protocol of 60 to 90 minutes. There is continuous availability of rubidium. Ammonia is unit dose, which has some advantages and disadvantages. Patient volume issues are a real issue for rubidium because you got to have those three to five patients per day. If you are sharing a camera, say you're in a hospital-based setting, you actually could do ammonia a couple of days a week, and so that might be a good place to start out using ammonia. For rubidium, there's two providers. I won't go into detail, but either one of those can provide a rubidium for you. Remember with ammonia you need a full scope cyclotron in very close proximity, which is a pretty big deal. Usually it's in a university or large hospital setting, but there is a company now providing a more compact system that when you're beginning the program you might want to think about looking into this, particularly in terms of volume. F-18 tracers, I mentioned this before, we already do have FTG available and is very useful in certain settings. Remember that F-18 has a two-hour half-life as opposed to 75 seconds or 10 minutes. Therefore unit dose is the consideration that's how it's going to be delivered to you, so you don't really need to worry about an on-site generator. With FTG, these are the indications I went through before, so if you have a large group of referring health care providers, you may want to consider this right now in the development of your program. But again, remember F-18 fluoropyridase has already completed one phase 3 trial. There are three-fourths of the way done with the others, so I would say even conservatively speaking, within five years this will be available. So that's an important consideration when you're beginning a PET lab return. What about space? There are two considerations. One is a dedicated PET camera, which generally tends to be older. It certainly does attenuation correction and with most systems you can do myocardial blood flow. But what's going on in the field right now is PET CT. This seems to be the direction people are going on. Either in the secondary market, what's available are 8 to 16 slice systems at a pretty reasonable cost, and I don't want to get into that, but that allows you to do calcium scoring but not CTA. On the other hand, if you really are considering a full-blown PET CT program with CTA and CTFFR, then you really need to go into the more expensive instrumentation that has at least 64 slice and most people you're doing 128. But remember, this is a short, smallest footprint. This is next. The PET CT can be some large instrumentations and that, as well as expense, is an important consideration. So, but there are some things to think about. CMS actually now has a separate billing code for blood flow, so they're certainly encouraging you to get involved in blood flow, which can be done with both dedicated PET as well as PET CT. On the other hand, there is now a separate billing code for PET CT, particularly in the outpatient setting, so that's a consideration where that may give you a little more impetus to move in the PET CT direction. Okay, let's talk a little about space and build-out. I don't want to go into too much detail, but these are all things you got to think about. Are you replacing a spec camera, which means you may have limited space, or is this a new space in which the, both the floor plan as well as the build-out on the floor, in other words, how much weight can it handle, is not a limit. If it is a limit, then moving in the direction of line source seems to make more sense. Also remember that there's configuration things such as a preparation area and a holding area, as I mentioned earlier, particularly as you get into F18 images. Some laboratories that I'm familiar with, if they're doing, let's say, a viability study once or twice a week, will actually leave the patient, they'll do this study at the end of the day and leave the patient in the imaging room itself until they're ready to image. Here's the other considerations. Remember, you're going to need a control room for PET, you need lead shielding, and this differs between rubidium ammonia and F18. My own recommendation is prepare for the future, and the future really is going to involve F18 imaging more so. Remember the weight of the camera systems is an important consideration. This is just a schematic in which you actually could put together a spectrum and a PET CT room with a centralized control room. Again, these are considerations you want to go into as you're developing the program. So just a few comments about dedicated versus line source. Dedicated cameras tend to be older, and in fact there's very few manufacturers doing dedicated cameras. I know of one currently where there's multiple camera systems that are doing PET CT, multiple vendors. So in the standard camera, 2D imaging was pretty commonplace, but right now 3D imaging could be done with most of these cameras, and 3D imaging has some important considerations. A dedicated PET camera, again, this is attenuation correction performed with line source as opposed to CT. The whole thing is easier. Attenuation correction is done very well with these cameras. Here's what you can do. Obviously blood flow, I'm sorry, obviously PET perfusion. Blood flow with most cameras, but not all, so one has to be careful about that. Some people feel that you can do viability. I've certainly done this with a line source. Sarcoid can be done, but it's much more limited. Most people feel a larger field of view camera is a better way to go. So the advantages are as an entry level, it's the least expensive, smallest footprint, lowest weight, and remember there's no radiology ovary. The disadvantage is you really can't, obviously can't do CTA. Some of the uses such as infection imaging, even sarcoid are not done unless you have CT, so that's consideration. And because of this recent decision by CMS to reimburse for CT, actually there's lesser payment by CMS. And as I said before, be careful if you're moving in this direction. If you want to do blood flow, make sure it can be done on that system. The more common cameras right now are PET CT and what you do CT for attenuation correction. The data are, by the way, that whether it's CT or line source, you get very good attenuation correction, probably doesn't affect image quality. But when you move in the direction of CT, you can do everything and that's a consideration, in particular if you're a larger practice that is going to be doing this. Cost on the secondary market is certainly more than dedicated, but the costs are coming down. The new equipment is much higher and again that should be a large practice and particularly if you're moving in the direction of PET and CT. Potential disadvantages for PET CT, one is this higher cost, certainly larger cameras, greater weight requirement. And remember that, and this is a consideration, there is a CT portion even though it's not necessarily diagnostic quality, you still get information. So most societies feel you should be looking at the CT data independent of perfusion and that may be an issue for your practice. The final aspect of this is education and education is very important for both technologists as well as physicians and the surrounding community. Most technologists are very skilled at PET but have virtually no experience with PET and so you've got all these issues of camera systems, protocols, acquisition processing, transmission emission maps. Remember you're going to be doing attenuation correction on everyone and maybe, particularly if you're moving in the direction of CT, your technologists should be trained in CT. Now this is a pretty controversial area and there are actually some programs available to do that. So here's some of the issues and I won't go into great detail, but remember that SPECT is virtually time independent, long acquisitions, filter back projection and very few people are using attenuation correction. It's more likely they're doing ECG gating and prone imaging and motion is pretty independent. In other words, we can motion correct anything. But PET is very different. It's very time dependent. These are very short protocols, you know, 35 to 60 minutes, short wait times, short acquisitions, so everything is very time dependent. Also there's rest stress gating which adds a little more complication, not much, but it's very motion sensitive. In other words, there is no motion correction as a result of that. If the patient moves, the physician is going to have to read through that study. We use additive reconstruction which we think is much better than filter back projection and remember that attenuation correction is done with every PET study. So this is a learning curve for technologists and again, because this is so motion dependent, getting the technologists to understand how important motion is becomes an important part of the training. We think training should begin before the initiation because they need to understand all this. When the site is initiated, remember they're learning a new infusion system, the acquisition and processing is also very important. So it's a pretty hectic couple of days and there's quite an adjustment between PET and SPECT. We think it takes several weeks to get used to that as well as the physicians getting used to it. I can't emphasize again how important motion is, both motion between the attenuation correction study and the perfusion study as well as motion during the perfusion. And as I said before, the injection and data collection are very critical. So all of these things are very important for technologist training. Physician training is also critical. The good news is, as I said before, high quality images, attenuation correction, they're actually easier to read and more definitive. So that's all the good news. The challenges are that you're going to have much more data. And so you need to sort through perfusion, rest and stress gated, TID ratio, perfusion data with attenuation correction. And when you get into it, myocardial blood flow. So you need to synthesize all of that information. The other side is although attenuation correction is less frequent, it does differ. And it's really in the site of motion and misregistration as opposed to attenuation artifact. It's also important that the physician do the interpretation and leading the PET program should understand the differences between referral for PET and SPECT. And as I said before, the transition from SPECT to PET really requires new interpretation skills. So that should be part of the training as well. So here's some of the differences in interpretation between SPECT and PET. We know on SPECT, everything is about attenuation artifact. The tendency is you read for specificity and you sort of discount attenuation artifact. But in SPECT or PET, that's very different because when you have attenuation correction, you've eliminated attenuation artifact. So what you see is most likely coronary disease. So you're switching from specificity rating to sensitivity rating. The data collection for SPECT is drawn out. Remember, you're beginning the acquisition 30 to 60 minutes later. But with PET, the data collection is during hyperamnia. So there's a lot going on such as reversible wall motion, TID, et cetera. There's a learning curve to that. Primary artifact with SPECT is attenuation correction. That's eliminated with SPECT and it's now motion and misregistration. The lower image quality of SPECT means a lot of uncertainty. Higher image quality means greater diagnostic certainty. That's a pleasure. And you'll be very pleased with that. Remember, I emphasized this several times, motion independent versus motion dependent. Blood flow has been promoted, particularly with CZD systems, but it's pretty tough to do. And the tracer, either of the technician tracers are not ideally suited for blood flow. And so very few practices are doing that. On the other hand, we think blood flow is an integral part of a active and up-to-date PET program. And the tracers are much more accurate with regards to blood flow. And in addition, F18 with the higher energy levels are going to be terrific for blood flow. So these are a lot of the things that a PET interpreter needs to get used to. So how does this work? Well, again, as with technologists, you really need to begin before the program, such as patient selection, tracer education, protocol education, quality control, and interpretation. So all of these, the physician needs to start getting used to before the program even starts. It's always done best in conjunction with somebody knowledgeable of PET, at least in the first two to three months. Even as the program matures, there's continued evaluation of protocols, image quality, physician education, and assistant and patient education. PET is a dynamic part of nuclear cardiology, and changes are occurring very frequently. So one has to stay really well-educated to continue with a vibrant PET program. Education of the practice staff and patients is also very important. I won't go into great detail, but it's obvious that they need to understand the advantages as well. Billing and reimbursement, I won't go into great detail about this either, but remember that there are separate billing codes, and particularly because blood flow has now been reimbursed, that's an important thing to consider for your practice. So I want to switch in the last few minutes to blood flow assessment. I skipped over that because my feeling is that blood flow should be initiated after the program has been developed. The reason is that starting a PET program with all of the nuances between both technologists and physicians can be quite daunting. Now, it's all learnable, and within several weeks to several months, you'll be very comfortable with the program, but adding on blood flow at the start can be a little daunting. So my recommendation is do perfusion first, get yourself familiar with it, and then start the blood flow program downstream two to six months later when you're comfortable, both the technologists and the physicians. But the same thing is actually true, and that is you need a whole education process for blood flow because, you know, as I say, it's basically numbers and beginning to appreciate these numbers. There's another whole educational process, and, you know, we're not used to numbers. We're used to looking at images. So this is another important part of the program. So what I see are the necessary components. One is accurate data acquisition. In other words, the technologist really needs to understand how to acquire the data. They need training for the whatever software program you've chosen, and they also need to understand the quality controls, so placement of markers, things like that. Physicians, too, need to understand several things, such as the value it brings, how you're going to incorporate that into the other data. The physician needs to understand the quality control because if the acquisition is incorrect, then the numbers are pretty meaningless, and you need to understand what these numbers mean, particularly in relation to patients with known disease and no known coronary disease, and incorporate that under into appropriate PET data, and then finally into the report. So we think this should be done two to four, six months later separately once you've gotten the program started, but once you begin the blood flow program, you'll see how valuable it is, and it's a further separation from other imaging modalities. So what I've done today is to go through the start of a PET program. These are the goals, and that is do you have the volume? What are you really looking to accomplish? What do you want from a tracer? What kind of space is available? And what you decide on line source PET versus CT, and as I've emphasized towards the last part of this presentation, that the training and education of physicians, technologists, patients, healthcare providers is critical for a successful program. Thank you for your attention.

cardiovascular PET laboratory

cardiac PET program

patient volume

training

space considerations

tracer choices

diagnostic accuracy

prognostic power