Thank you so much for that kind introduction. My name is Chris Mehta. I'm a cardiac surgeon, not a radiologist, from Northwestern Medicine just down the street. So I appreciate the opportunity to sort of give you a surgeon's perspective on some of the postoperative imaging that we see. If there's one sort of unifying theme of this talk, it's the mutual benefit of collaboration between surgery and radiology, which is in contrast to this video that you're going to see here. Some of you may have seen this video already. I'm sure all of you know surgeons like this. And today I'm going to be doing a sort of case-based review of different expected and unexpected postoperative findings, and I think we'll learn a lot about sort of how surgeons think about some of these things and timing of intervention and that sort of thing. So the first case is a nice evolution of aortic dissection. So this was a 57-year-old woman who presented with chest pain. No surprises here. It's a fairly obvious diagnosis, type aortic dissection. The things I'm looking for on these scans are, you know, where is the entry tear? So I saw the entry tear in the ascending aorta, but I'm also looking for re-entry tears or what I suspect are primary entry tears in the aortic arch because that dictates how much aorta I think I'll need to resect. You know, I'm looking for things like pericardial effusion. Could this patient potentially have tamponade? Malperfusion, I want to see where those branch vessels, especially the viscerals, are coming off because in patients with mesenteric ischemia, sometimes we actually take a malperfusion first approach with an endo-fenestrating and that sort of thing as opposed to doing a proximal ascending repair first. So this lady underwent a repair with my partner. You know, he did an aortic root replacement with a valved conduit and ascending aorta repair in a hemiarch. And as you may have noticed, the, you know, the dissection extends from the hemiarch line and there's false lumen filling and I'm sure you see that quite commonly. I mean, in some studies, it's up to 70% and that's okay. That's not a concerning finding. What that means for me is I have to pay a little bit more attention to that residual aorta because that is an aorta that can develop quicker aneurysmal degeneration. And so, sure enough, you know, the aortic arch and the proximal descending aorta got, you know, aneurysmal. The worry there is, you know, we don't want this patient to rupture. And so, you know, once it met, it was a little bit above 5.0 centimeters, you know, I took her to the operating room for an aortic arch replacement with a frozen elephant trunk. And, you know, I was pretty happy with this repair. You know, you see thrombosis of that proximal descending false lumen and it's redirecting blood flow sort of more distally in the mid-descending aorta. But what the really interesting thing was here, three years after this repair, you really see nice positive aortic remodeling, right? That proximal descending false, thrombosed false lumen is regressing. She still has residual dissection. This is something that we monitor. Again, the primary thing I'm looking for is aneurysmal degeneration in the residual aorta. This next case, this was a 21-year-old man who loved ATVs and he got actually into an ATV accident that resulted in a traumatic aortic transection and underwent a T-VAR. Now, these are tough cases from our standpoint because these are young patients with small aortas. They're not aneurysmal, right? And so that means they tend to have pretty small T-VAR stent grafts placed. And he actually came back, you know, about a year later with symptoms of bilateral leg ischemia. And so, you know, when we scanned him, this is what we found. Zone 2 T-VAR, fine, but you see this interluminal filling defect there. And actually, this particular stent he had had been recalled for this very issue. There were instances of thrombus associated with this particular stent graft. So fortunately, his symptoms resolved before becoming an acute leg ischemia type situation. And we tried him on oral anticoagulation for a few months. Despite that, that filling defect still remained. And so I ended up taking him to surgery for a left-sided thoracotomy incision repair, explant of the stent graft, and then repair with a Dacron graft. And so this is what we found. It was kind of interesting. So, you know, here you see actually what happened is the part of the mechanism of the stent graft actually caught some of the aorta and then it granulated over. So it wasn't thrombus, per se, that we were seeing, but it was actually this big sort of hunk of tissue occluding the distal aspect of the stent graft. And so, you know, he was doing fine for a couple of weeks and, you know, about two weeks into it, we, you know, we saw this X-ray. He was home at this point and had represented to another hospital. And then sure enough, you know, he had a large hemothorax. So this is obviously, you know, something that we need to address. And so, you know, we drained that and his follow-up imaging looked great. So, you know, we're pretty happy after that. This next case is a 45-year-old man with a prior coarctation repair, and he'd had this done in the UK and it was a real challenge to get records from the NHS. And so we didn't entirely know what sort of repair he'd had, but this is what he presented with. And so, you know, I presented this at our multidisciplinary aortic conference where there's radiologists and surgeons, you know, vascular surgeons, cardiac surgeons, and so forth. And part of, you know, what I needed to know for my repair was, is there a flow between this little communication there? Is that just sort of scar tissue? Is that just a little ligamentous type thing? And so, you know, one of my partners in radiology suggested we get a 40MR, and sure enough, you can see there that there is flow in this. And that was tremendously helpful to me because that sort of dictated, you know, what kind of repair I needed to do. And so, again, this patient had a left thoracotomy incision, a thoracobdominal incision, and we replaced that section, entirely excluded it, and then basically sewed Dacron graft from here down to the mid-descending and re-implanted the subclavian artery, and, you know, the post-op 40MR looked a lot better. This unfortunate gentleman was a 73-year-old man who went an elective aortic valve replacement, aortic root ascending, what we call a bent-all procedure, with one of my partners. And you know, this scan is not terribly remarkable. I mean, you know, there's some expected post-operative stuff, I would say, but, you know, beyond that, you know, it looked okay. However, two years later, he presented back with this scan, and here you see the pseudoaneurysm, right? This is a concerning finding, obviously. You know, what you see in the operating room is this tends to be contained by tissue, scar tissue and whatnot. So it's something that needs to be surgically addressed, but it's not like I take these people the very next day to surgery. So I told him, hey, you need surgery. He was kind of upset about having to need surgery a few years later, and so we decided to get a, you know, surveillance imaging. Six months later, he had the scan and again said, I don't want to have surgery. And then the third time, he never showed up. He actually passed away, and it's suspected that it was because of this. So you know, these are things that need to be addressed soon, obviously, but, you know, can't force everyone. This case is a 46-year-old patient who had an incidentally found root aneurysm on an echocardiogram, and then we followed that up with a CTA. So sure enough, large root aneurysm with dissection, you know, I might add. So you know, this was a patient that I said, hey, we need to, you know, operate pretty urgently. And so this is what it looked like before. And incidentally, this is sort of what a classic Marfan root looks like. This patient didn't have Marfan syndrome, but just, you know, for the sake of education, this is very commonly what they look like. And so I did a valve-sparing aortic root replacement on this gentleman, and, you know, so that was the post-op repair. This is the final case. This is a kind of interesting one. Actually, we recently intervened on this lady, but 60-year-old with a complex type aortic dissection, also had a history of COPD and active cocaine use. So I'll show you the 3D recon first so you sort of get a sense of it. But this lady had a type aortic dissection, large entry tear, this pseudoaneurysm proximally, which is sort of creeping into the mediastinum. One of the more unusual findings here is, if you notice, the celiac SMA and left renal are coming off the false lumen, and the right renal is coming off the true, and her legs were perfused by the true. And, you know, if I run through this again, you'll sort of note there's not a lot of fenestration in the septum. I mean, there presumably is some based on this morphology, but, you know, you can't really tell very many fenestrations. And so, again, this was another situation where 4-DMR, we do actually a lot of this at Northwestern, we use 4-DMR to kind of help us identify those fenestrations within the septum of a chronic type B. We didn't see very many. And so, you know, from a surgical perspective, you know, it's a chronic type B with a septum that's, because it's chronic, not terribly movable. We have to exclude, somehow, that proximal pseudoaneurysm while maintaining flow to actually the false lumen, right, for the visceral vessels, but also the true lumen for the right renal on the leg. So it's not as straightforward as, you know, doing a sort of frozen elephant trunk into one lumen and then, you know, thinking that will address everything. She also probably would have been best served by a thoracoabdominal surgery, just a conventional surgery from the left side, but she couldn't tolerate that due to her comorbidities. So we came up with this kind of creative solution here, which was a double barrel frozen elephant trunk. So I replaced the aortic arch and I put a stent graft coming from that into the true lumen and into the false lumen with the idea of excluding that proximal pseudoaneurysm while still maintaining flow. As it scrolls through, as you can see, you know, there's a small endoleak right here, right, which we presumed to be from the distal aspect of the false lumen stent graft. So this is something that, you know, we felt comfortable sitting on for a little bit, you know, sort of let her recover from surgery and all her contrast loads and so forth before we intervene on it. You can see the majority of that pseudo is thrombosing and all her branch vessels are patent. So we brought her back a month later and we extended both lumens with Tvar stent grafts and, you know, we got a nice result. This is an on-table CT angiogram and now that endoleak is excluded and she did really well. So, you know, I hope those cases sort of highlight, you know, some of the postoperative complications we see and, you know, how I think about them. I think these are some kind of key takeaways, what surgeons want to know. First and foremost, is there anything I need to address surgically, right, because that's going to be, you know, a significant difference, you know, for this patient's clinical outcome. So pseudoaneurysm, anastomotic leak, effusions, you know, I think, you know, these are the sort of top things, like if there's something that needs to be done, you know, we need to address these pretty immediately after surgery. Do you think any of these findings are out of ordinary? So what I mean by that is, you know, I think our radiologists do a really good job at Northwestern of saying there are some postoperative findings here. This is more fluid around the graft than is expected, that kind of thing, these sort of qualifiers. I find those tremendously helpful because they give me a sense that, you know, maybe this is something I need to pay a little bit closer attention to, revisit, you know, what we did in surgery and if there's anything that, you know, I need to pay closer attention to. Is there anything you think I need to keep an eye on, right, serial imaging? I mean, I love when you say consider a repeat scan in three months or what have you. And then lastly, I would say call your surgeon if you're worried about something or if you're unsure what you're seeing, right? There's so much, you know, evolution in, especially in the stent graft technology and different repairs we're doing and so forth. So if you don't know what something is, you know, it's really helpful to just talk it through and say, I'm seeing this stent, is this what I'm seeing, that sort of thing. So with that, I thank you for your attention, I'd be happy to take questions. So I'm going to quickly review some of the most significant or key points that the radiologists need to know to do a good interpretation of patient who have had surgical repair of the aorta. So one of the important points to remember is that we do a lot of imaging in the post-operative setting. Cardiovascular diseases is one of the most significant causes of mortality in this country and I will say around the world. 3.3 million Americans died last year of cardiovascular diseases, 20% of them, excuse me, in overall, 20% of them from cardiovascular diseases and roughly about 10,000 deaths annually occur in the U.S. due to aneurysms and dissection. Among these ends in a significant number of open heart surgeries, roughly about half a million surgeries are done in the U.S. every year, 2 million surgeries, open heart surgeries in the world. That means that the U.S. is responsible for 25% of open heart surgeries of the world and 900,000 if you include endovascular procedures. So we do a lot of imaging for the follow-up of all these patients. These are recent guidelines published by the American College of Cardiology and the American Heart Association on how to proceed with a follow-up of patients after having a surgical or endovascular repair. And based on the fact that the evolution of patient with open heart surgery is way more stable than patients who have had endovascular repair, the guidelines clearly differentiate the strategy for follow-up of these two different patients. Patients who have had TAVR need a surveillance CT one month and ideally annually after the endovascular procedure. Different from that, patients who have had open repair tend to be more stable. And the recommendation for these type of patients is to have only more continuous follow-up in those patients that have residual aortopathy, meaning for that patient with Marfan syndrome, megatetor diagnosis, or any underlying conditions. Patients who don't have this condition only need a follow-up one year and every five years after the surgical repair. The state-of-the-art imaging and the ideal imaging strategy for these patients is with CT and doing a volumetric acquisition with three-phased CT. I lost the microphone, I believe, again. That should include non-contra CT. That is critical for the identification of calcifications and any surgical material. The arterial phase, which is critical for the evolution of the lumen. And the delayed phase, which is very important because that's what really allows us to differentiate thrombus versus a slow flow, better see and identify leaks and extravasation. Some suggestions have been made in the literature, doing a second delayed image in five minutes. We don't do it routinely in our center, but occasionally there are cases in which there is very slow filling of leaks or endo leaks in which that may be worth to consider. But in general, the delayed phase that we do is the usual standard, 60 to 120 seconds. Another important point for the radiologist to remember, and this is something that is unfortunately now and then not done routinely, is to remember that measurements are a critical part of our job in the analysis and interpretation of CTs. Measurement is essential. If you do cardiac-gated CT, which we always recommend for patients who have had surgery of the aortic route, remember that there is a slight variation between the measurement done in systole and diastole, roughly about 1.8 millimeters. Measurements should always be done from outer wall to outer wall and perpendicular to the center line. Measurements are critical, in particular when you are using the same modality in the follow-up and give surgeons very important useful information to consider what is the expansion rate of any dilation of the aorta. These are the basic minimal measurements that should be included in a CTA of the aorta at the level of the root, the sino-tubular junction, the mid-descending aorta, the distal-descending aorta, aortic arch, proximal-descending aorta, mid-descending aorta, and aorta at the diaphragm. If there is any particular focal bulge, that should also be included. A couple of important points for the radiologist is to have a basic understanding of the surgical procedures, be familiar with the type of grafts and the type of surgical reconstruction that has been done so you sound intelligent in your report. The most common graft that is used nowadays is polyester or Dacron. Another type of graft is the polytetrafluoroethylene. These materials are very compatible and are very well tolerated by patients. They have no rejection or calcification. They are very durable, and the lifespan of these exceeds the lifespan of the patients. One important point to remember about these grafts is that they are almost invisible. Different from the endovascular stain grafts that have a metallic mesh, these grafts don't have any metallic components, so they are pretty much invisible on CT. The only thing that is very obvious on CT is the areas of the anastomosis and any felt pledges that were used during the surgical procedure. Type of grafts vary significantly. There is a lot of different things available out there. Be familiar with the most common ones. This is, in the picture, I'm showing you what is a composite graft. So, composite graft is when there is a combination of materials, like this one. The most common type is this combination of ascending aortic graft with an aortic valve graft. The other type of graft that you will see in the request is a patient who have an interposition graft. Interposition graft refers to the type of graft that is used when the aorta is resected and is replaced by a synthetic graft. The other type of graft that now and then we see is the inclusion graft. Inclusion graft refers to the type of graft that is put in place, but keeping the native aorta. So, in those cases, the aorta is wrapped around the graft, so there will be really two layers in the surface of the surgical region. Another common combination, just like we saw in the first presentation, is this hybrid technique. Hybrid technique refers to the combination of a surgical graft at the same time that there is an endovascular graft. There are multiple types of surgeries done and multiple variations of the surgeries. Also, the radiologist should be familiar with the basic surgeries. With procedure, the hemi-arch replacement, the total arch replacement, all these are going to look completely different in your CT and are going to be important for the interpretation. Surgeries that affect the aortic root are very important also to be familiar with them, like the bental procedure, the David procedure, the Yakut procedures. Some of them are done specifically to address the aortic root and or to preserve the aortic root and the native coronary artery without touching or affecting them. So, that's important to be familiar with those. Probably one of the techniques that is more confusing to the radiologist who is not routinely doing this type of CT evaluation is the elephant trunk procedure. This can have a very funny appearance on CT. So, important to remember, this is a combination of a graft referrer of the ascending aorta and the aortic arch with a tube graft that is left floating freely in the descending aorta. So, that produces that funny appearance of that multiple layers within the aortic lumen. Be familiar with these. Don't say something funny when it is just the expected surgical procedure. This is how they look on CT. You can see the free-floating graft and also the flap of the descending thoracic aorta dissection. The other surgical procedure that can be confusing is the patient who have had a cabrol procedure. In this case, a graft is attached approximately to the ascending aorta and distally to the coronary arteries. So, that creates this kind of complex appearance of the ascending aorta. And just remember that one is the aortic graft and the other is the coronary artery graft attached to it. Look for combinations in this patient for complications. This is a patient with a cabrol procedure. You can see that one of the arms of the coronary artery graft is patent and the other is thrombose. So, that is a significant finding that has to be immediately informed to the surgical team. The other type of surgery that can be very confusing is the inclusion root technique. In this inclusion root technique, the graft is placed within the native aortic root, which is wrapped around the graft. That creates a potential for flow, creating this, again, double lumen within the surgical territory. Important to be familiar with the appearance of the felt pledges and reinforcement technique that are used at the level of the anastomosis. Surgeons use these surgical flat pledges, which is an unobservable synthetic material that tends to be high density on CT to reinforce the surgical anastomosis. And that creates this high-density material that can have variable shapes and morphologies at the level of the anastomosis and this is one of the reasons we always want to have a CT without contrast before doing your interpretation because after contrast injection, this can look very similar to contrast extravasation and can prompt mistaken interpretations. These are examples of that. The first image you see, the surgical graft of the ascending aorta and the high-density surgical material. And after contrast injection, this can be very confusing if you don't have the non-contrast images obtained before. Another common finding that can be normal or abnormal depending on the clinical context of the size is the presence of perigraft seromas. These are typically low-attenuation perigraft fluid. The majority of perigraft seromas tend to be smaller in size and some studies have shown that when they are less than 50 millimeters in thickness, they tend to resolve and have no clinical significance. Those that are too large have a higher association with complications later, one of which is infection and seroneurisms. They tend to resolve in the first three months. They should not grow. If you have a perigraft seroma and you do follow up, an important consideration is the size, whether it's stable, shrinking, or growing. We don't know exactly what is the exact incidence of these seromas. Different numbers have been cited in the literature, as low as 1%, as high as 30%. Another case illustrating perigraft seromas. So follow-up of this is important to keep in mind the change and evolution in size. Another finding that can be normal in the early postoperative studies is the presence of periortic gas. This periortic gas usually is expected finding for the first two weeks. It should not be seen more than that time, or at least not after the third week, because that can be another suspicion for infection. Complications of the aortic graft repair can be divided in two big families. Those that are complications within the aorta and aortic root and complications outside of the aorta. Of those complications within the aorta and aortic root, seroneurisms, progressing size of the aorta with development of aneurysms in particular patients with aortopathy is an important finding. Dissection, up to 2.5% of the patients can develop this, and the leak in those with hybrid repair is way more common, up to 30% in the literature, and graft infection, about 1% to 5%, 6%. Pulmonary embolism, 3% to 6% of patients will have this complication in the follow-up, so be very mindful of this and check the pulmonary vasculature. Another way of looking to the complications are those that happen early after surgery and those that are delayed after surgery. In the early postoperative days, up to the first month, infection, aneurysms, menostatic hematoma synapses, seroneurysms, and foreign bodies are important things to consider. You can see in this case here the presence of a seroneurysm and a very large menostatic hematoma with a fluid-fluid level. Another case of a very large pericardial hematoma and tamponade two weeks after type A aortic dissection. Important in this case is also to check what is the impact of the hematoma on the cardiovascular function and cardiac chamber. You can see how the aortic root is displaced and deformed, displaced posteriorly, and how the large hematoma is compressing and squeezing the right side heart chambers, which is typical for tamponade physiology. Check for the possibility of unintentional retained surgical items. This is the denomination that Joint Commission gives to surgical foreign bodies, and this is the reason why these surgical sponges are marked with this high-density material so we can easily identify them after surgery. It's something very important from a medical and legal perspective, finding and should be always checked in your post-operative changes. Look here, this is another big textiloma or a gossy fiboma from a retained surgical sponge with the typical right-of-back markers. Late complications after 30 days include infection, also aneurysms, perigraft flow, and progressive enlargement of the aorta, seroneurysms, dissection, thrombosis, and fistulas. A couple of additional examples of seroneurysms, as you can see here and here. Very rarely, decron graft degeneration can happen, like in this case, and you see how the graft is pretty much disappeared, creating this huge seroneurysm. And additional complications include thrombosis and distal dissection, like in these two examples. Okay? I'm going to finish here. In summary, open surgical repair of the thoracic aorta is increasingly complex. Imaging plays an essential role in the post-operative evaluation of these patients. Be familiar, radiologists, with the normal post-operative findings, some of which can mimic abnormalities, and be familiar with the complications. And important is knowledge of the original underlying disease. The repair technique is critical for accurate interpretation of these patients. Thank you. Hi. My name is Jody Shen, and I'll be talking about imaging in ECMO. In this talk, we'll discuss alterations in blood flow with veno-arterial, or VA-ECMO, understand the difficulties of imaging at CTA, and review imaging strategies for optimizing CTA for VA-ECMO. For some background, the use of short-term mechanical circulatory support has increased for select patients with refractory heart failure. And VA-ECMO is a type of mechanical circulatory support used in the ICU setting. ECMO-related complications include bleeding and thrombosis, which often need to be evaluated by CTA. While MRI of ECMO has been shown to be feasible in an animal model, ECMO equipment is not yet MRI-compatible and remains contraindicated for clinical use. And thus, CTA is often the last stop for evaluation of these critically ill patients. It's important for radiologists to be familiar with some of these flow-related artifacts at CTA associated with certain ECMO circuits because they may obscure or even mimic pathology with management implications. So we'll also focus on imaging strategies to mitigate these artifacts. Let's start by familiarizing ourselves with different types of ECMO. In general, ECMO is comprised of an inflow cannula that siphons the patient's deoxygenated blood to a pump and an oxygenator, and then returns the oxygenated blood back to the body via an outflow cannula. Venovenous, or VVECMO, is used in the setting of acute respiratory failure. In this setting, the inflow and outflow cannulas are both in the systemic veins, and blood is removed and returned to the same place. So the rest of the blood flow is physiologic, and because of that, VVECMO is not usually a problem at CTA and can generally be scanned with routine protocols. The problem is venoarterial, or VAECMO, which is used in acute cardiac or combined cardiopulmonary failure. Blood is withdrawn from the systemic veins and returned to the systemic arteries, bypassing the heart and lungs. Here are chest and abdominal radiographs depicting the flow of blood in a VAECMO system. The altered physiology of VAECMO can affect CTA imaging, where standard protocols can be suboptimal or even non-diagnostic. This catheter angiogram is a nice illustration of the alterations of blood flow with VAECMO, and it shows superior displacement of injected contrast material by unopacified blood returning from the ECMO circuit from ephemeral arterial cannula. VAECMO changes the circuit, which presents a challenge for CTA due to the unexpected timing and sequence of contrast enhancement. Cannulation can be established by either a central or peripheral approach, although peripheral is far more common. In either case, the watershed area is typically seen, which is the interface where the two bloodstreams of native cardiac output and blood flow of native cardiac output and ECMO return meet. This watershed area may obscure pathology if contrast material is not in your vessel of interest at the time of the scan. For example, this was an attempted PE protocol that was non-diagnostic because most of the contrast material has been siphoned off by the ECMO circuit. Repeat injection shows that this large pulmonary embolus is in the abdominal aorta. This was an attempted CTA for GI bleed evaluation, and on the first attempt, there's no contrast extravasation, but note that the watershed area is in the abdominal aorta at the level of the superior mesenteric artery. Repeat examination 30 minutes later with the watershed area lower now that contrast material has reached the inferior mesenteric artery. VA ECMO can not only obscure but also simulate pathology. The watershed area in this patient with a left ventricular assist device and VA ECMO was initially mistaken for acute aortoiliac occlusion, but it represents the interface between the two bloodstreams. In this patient, the contrast was injected through a PIC, which just bypassed the ECMO inflow cannula, so most of the contrast was delivered to the native circulation and pumped forward by the LVAD. Here are two patients where there's incomplete opacification of the cardiac chambers at CTA. With the delayed phase, you can see that patient one has thrombus filling the right ventricle, whereas there is no thrombus in patient two. Some other interesting artifacts can be seen, and on this short axis view, the myocardium is diffusely enhancing in the absence of intracavitary contrast material. The three-chambered view and the schematic show that contrast material is coming retrograde into the coronary arteries, which leads to this odd enhancement pattern. It's estimated that up to 50% of patients treated with ECMO have neurological impairments, and may require CTA evaluation. Here's an axial maximum intensity projection, CTA of the head, where we see differential enhancement of the anterior circulation due to VA ECMO effects, with relatively less contrast enhancement of the right anterior circulation. This is because non-opacified blood from the native circulation combines with contrast-enhanced blood returning from ECMO in the aortic arch, and this was the reason for the aortic arch, and this results in mixing artifact within the brachiocephalic artery on this axial CTA image. Now we'll go over some strategies to mitigate some of these flow-related artifacts. Scanning VA ECMO patients can be challenging, since there are many factors at play, including the patient's native cardiac function, flow rate, and contrast injection site. Here are two patients on VA ECMO that were scanned with routine CTA protocols. In the first case, we see that most of the contrast has been siphoned out by the ECMO circuit, and returned to the aorta in a retrograde fashion. There's no contrast in the heart. In the second patient, most of the contrast has been directed through the native circulation, with relatively unopacified ECMO return. You can get lucky if conditions are favorable, but my general advice is to have a plan before attempting a CTA. Here was an overnight cardiac CT for valve-in-valve TAVR planning that was scanned three times, the technologist notes indicating that he spoke frequently with the resident on call regarding injection and timing. The first two attempts were non-diagnostic, and there's a saying that insanity is doing the same thing over and over again and expecting different results, so really think it over before you try the same strategy that failed. I would also suggest prioritizing the most important question when a large scan range is requested. You may not be able to answer all of the questions, but maybe the most important one. Especially for PE evaluation, it's strongly recommended to decrease the VA ECMO flow rate if possible. Some actually recommend decreasing the rate to 500 milliliters per minute for the duration of the scan, but it should only be changed as tolerated by the patient and under the guidance of the critical care team and the perfusionist. You might also want to consider going to the scanner to choose where to place the region of interest for automated bolus tracking, or you may want to manually trigger. You should monitor as close to the vessel of interest as possible because you know if contrast has arrived in that area, you'll have a diagnostic scan. When you're planning your scan, it's also important to consider where your area of interest is with respect to the outflow cannula. This was a patient with a recent aortic root replacement on central VA ECMO, where the area of interest was the right coronary artery. This drawing shows that the outflow cannula bypasses the area of interest. Delayed phase imaging can be helpful because it relies on recirculation and resolves a lot of these flow-related artifacts that we see on first pass imaging. Here's an example where the aortic root and ascending aortic thrombus is obscured by the watershed area, which becomes apparent on delayed phase imaging. High-pitch acquisition modes should be approached with caution since they may cause the tubing to dislodge, which can result in catastrophic hemorrhage, and it's a good idea to test the table excursion first. So, in summary, VA ECMO is challenging at CTA, and the main things to remember are to decrease the ECMO flow rate, if possible, to direct more contrast through the native circulation, monitor as close to the vessel of interest as possible, and add on a delayed phase. All right. Thank you for your time.

postoperative imaging

surgical perspective

aortic surgeries

imaging complications

radiology collaboration

CT imaging techniques

cardiovascular surgery

patient-specific assessment