All right everyone, we'll go ahead and get started. Welcome to Medical Grand Rounds. I'll go and introduce our speakers for the day. 

Dr. Sharma completed medical school in India. He moved to Ann Arbor, Michigan for his residency training and internal medicine fellowship at St. Joseph Mercy Hospital. He then completed a vascular medicine fellowship at Cleveland Clinic. He came to UVA in 2012, where he directs the anticoagulation clinic and the medical optimization clinic, which helps patients with vascular disorders manage their risk factors with medication. As part of his vascular medicine role he serves on the pulmonary embolism response team. 

Dr. Mihalek completed medical school at UC Davis School of Medicine. He completed residency and a chief residency at Loyola University. After residency he completed a fellowship in pulmonary and critical care medicine at Mass General, and then spent time at the NIH researching the mechanisms of smoking related lung injury. Currently, Dr. Mihalek oversees the pulmonary medicine curriculum for the School of Medicine and leads the development of acute pulmonary embolism response team. He also served as one of the co-directors for the multidisciplinary pulmonary hypertension clinic. 

Today they will co-present on the UVA experience of managing high risk pulmonary emboli. Please join me in welcoming them. 

[APPLAUSE] 

Great. Thank you for that introduction. We're very excited to be here today. This has been a long road for us, and we're happy to share our progress in the program with you all. We do have some small financial disclosures to make for myself. None of these will be discussed today. Dr. Sharma does participate in some trials on the treatment of DVT. 

But in reality, we wanted to also talk to you today a little bit about PEs. And we're going to talk about that very specifically in a very specific patient population. That's going to be a patient population that you know, and love, and fear at the same time. And as a patient population that we actually see quite a bit here at the University of Virginia. 

So we're going to be looking at patients that have high risk for poor outcomes, and we would label those as high or intermediate risk pulmonary emboli. And we'd like to distinguish those between the low risk patients. Along the same lines, we'd like to walk you through a little bit about our experiences in managing those patients through an organized program here, and just go through the data that's generated from that. 

But in reality, before we get started Dr. Sharma I have to make a few personal disclosures. We're not the only people involved in this program. And in fact there are quite a bit of people who are actively involved. And what we're showing you here is a small sampling of the people that are intimately involved in developing an acute clot service here at UVA. 

As you can see, they come from many different service lines. None of them get pay raises for being in our program. All of them work extensively hard, and all of them come to the table with a unique set of skills that makes them qualified to take care of these patients. We meet once a month at 7 o'clock in the morning on a Thursday. And I can remember when we first started the program, Dr. Sharma I used to meet in the glass table conference room in the Echo Lab-- five people just talking about a program that we wanted to start. 

And 2 and 1/2 years later, here we are. Standing room only, usually about 30 to 40 people present at that meeting, attended heavily by trainees as a way of augmenting their education. And to be honest, it is a fantastic conference in that people speak freely, we have high level discussions based on evidence based medicine, we take our problems to the table, we talk about patient outcomes. And it's just a really great place to be. 

And so, in reality what we'd like to take you through for the next 50 minutes is a victory lap of our program. And we want to be very, very proud of it, because of the fact that we have a lot of people working towards it and we're doing really great stuff. 

So before we get going, though, we have to have a little bit of orientation. And I know you guys all just finished your meals. You're in your rest and digest mode right here. So as a group of medicine doctors, I'm going to I'm going to provoke your sympathetic responses just a little bit and show you something really, really scary. 

OK, this is the CT scan. When it comes across your desk, you're kind of dreading. So, this is a patient that has one of the dreaded saddle embolis. Right, so this is anxiety provoking for us. And what happens when we see these patients is we revoke to what we think we know about those patients-- which, as it turns out, may not be rooted in evidence based medicine. 

And what I mean by that is this. Very recently it has come to the attention of the international community that saddle PE-- large clot burden PE, however you like to phrase it-- really isn't the most important aspect of these patients. And so just looking at two studies-- one by Jane in 2017-- what they found was that patients that had non saddle PEs were more likely to present in hypotension than patients that had saddle PEs, which would be stark opposite to what I would think about that. 

Non saddle PEs also, in the same study, had a 90 day mortality that was higher than the patients that had saddle PEs, also contradictory to what I said what I would think about that. And so their conclusion from that study is that clot location is not necessarily associated with patient outcomes. OK, I can buy that. 

But it also appears that saddle PEs may have a place in our annals of anxiety, if you will. And in this study from the Mayo Clinic, they show that saddle PE was more likely to present with hypotension. OK, great, so retrospective data. One study is going to show what I want, and one study is going to show what I don't want. So, what am I going to do with that? 

But it appeared that there was no difference in the length of hospital stay, and no difference in the length of hospital mortality between saddle and non saddle PEs. And so their conclusion for their study was no difference in short term outcomes. OK. 

So, what does that mean for us? Well, it probably means that it's less about the clot, and more about what the clot is doing. And for years we used to categorize these patients along these three lines, whether openly-- as a way we taught our rounds-- or just intuitively from what we knew about the cardiovascular system. 

And what we found is that we thought patients could be classified as massive, we talked about submassive, and we talked about minor PEs. We talked about the big dividing line was whether somebody was hemodynamically stable or not, which separated our massive from our sub-massive. And then our minor basis were patients that had a low risk for going on to death, and maybe not so much scary qualities. 

But the point is here, it was never about the clot. It was what we knew about the patient before us. So intuitively, that always made sense. 

Now I'm going to take you a little bit further to where we look at these patients in 2019. We think about them in the same way that you did. We label them slightly different because of the communication problems that we have when we talk about massive or sub-massive PE. So we try to not use that language anymore. 

Our massive PEs we consider high risk, meaning that they have a higher risk of dying in the next 30 days. A number is as high as 30%. That's a scary number. Our intermediate risk PEs are somewhere a little bit lower than that-- probably about 15%. But still pretty high. And in our low risk PEs basically are our patients who do very, very well-- the 90 days and 180 days. But the point is here is that, once again, we're using clinical information to determine how we're going to risk stratify those patients. So being medicine doctors, we do what medicine doctors do best-- we try to actually come up with a way of predicting which patients will not do well. 

And one of the earliest and best validated scoring systems that we use currently is this concept of the PESI-- so PE risk stratification index-- which takes all the things that you would think would be pertain to something having a poor outcome-- age, male sex-- males do worse than females, as we know, everything that they do, including government-- cancer, chronic heart disease, whether there's a prior evidence of PE, and how they're doing with their vital signs. 

And if we just put that stuff into a computer and it spit out a number, we can get some pretty good hard data. And so I would draw you down here to the bottom where we're talking about real numbers here. So if you're a class one with a little score, very low mortality risk-- less than 1% going up to 2%, 3%, and so on and so forth. So, that may be helpful. 

So very recently there's been a movement to take the evidence based medicine and risk predict our patients, and then use it with what we know clinically does not do well in terms of the long run. And so the Europeans from the cardiology side and pulmonary side came together with a joint consensus statement that housed all that together. And now, in 2019, we use all of that to help predict what will become of these patients. And so we look at patients that have high, early mortality risk, intermittent, and low based on the following qualifications. 

Now the first thing I'll say is that high risk patients are patients that are hemodynamically unstable, and they're going to be the ones who you're still going to be scared about. Our therapies for them are still very limited to systemic lytics, and so nothing has changed from what you know about that group of patients. In terms of our low risk patients, the only thing that's really changed is the fact that those patients should be put on anticoagulation and probably go home from the ED. 

So there's a big move into do PE home programs, one of which we started here, with the idea that these are so low risk that we should not be wasting valuable resources to admit them because we feel uncomfortable. And so what I'll say is that big scary PE patient at the top of the hour, that was a patient who got admitted but went home within 36 hours. So they ended up being an intermittent low risk. OK? 

So they ended up staying for insurance reasons, as we do in Virginia. So we had a hard time getting them their medicine that they needed at home. But the point is that it wasn't the big scary saddle that got them into the hospital. It was more just kind of our resources. 

And so the patients that still, though, occupy a spot of anxiety, and thought provoking, and controversy are these intermediate risk patients. Now, if we stratify them even more into an immediate high and an eerie low, what we're essentially saying is that we have clear evidence that we have a right ventricle that's under distress versus we have just a little bit of evidence that maybe that RVs not doing so well. And that's what you're seeing here in the yellow. 

And so the problem with these patients, though, is that we still don't know what to do with them. And so this is one of the earlier studies that looked at systemic lytics that looked at then those terms-- sub-massive PE. This is a very classic study that every pulmonary fellow has to read and actually present in journal club. 

And what they did was pretty straightforward. It was a randomized placebo controlled trial. They looked at 276 patients. It was in Europe. And they gave some patients lytics with heparin. They gave some patients heparin with placebo. And what they found was it in a pooled set of outcomes-- so that's the probability of event free survival-- it did appear that there was a signal that patients who got lytics did better. 

Now there are 1,000 problems with this study, which is why it makes it journal club fodder. One of the bigger issues is that one of the pooled outcomes that they were looking for is if you failed the therapy that you got, you ended up getting lytics. So in that study you're looking at, basically if you didn't get lytics, you got lytics. And if you got lytics and it did well, you got lytics. And so you can see where the confusion arises there, in terms of the outcomes that we're looking at. 

In 2014 they tried to address this with a little bit of a larger trial-- and this was also from New England Journal of Medicine-- where they looked at a bigger cohort of patients-- so now 1,000-- and basically split them into lytics versus placebo. And there the signal is a little bit more clear, but still problematic. And what they found was that patients seemed to do better with lytic therapy. However, if you got into the weeds of it, it seemed also very clear that patients older than 75-- which ends up being a large majority of patients-- do poorer with worse outcomes, particularly of bleeding. So kind of helpful, but not very helpful. 

And perhaps the major issue here is that major bleeding events here-- defined by basically anything that happened outside the head-- was also very, very significant for patients that still got lytics. So in that regard, you feel a better, because patients do better. But then you feel a little bit worse, because it's clear that the things that you're scared about still happen. So not very helpful in a group of intermediate risk patients. 

Simultaneously to that, we see the advancement of technology come along. And now we see the introduction of non-medicine techniques of dealing with these patients. And the catheter directed therapies that have been offered really since the early 2000s til today are the ones that catch the most notice and the most interest from providers. 

On the left what I'm showing you is an actual fluoroscopy film from a study where basically you access the pulmonary artery, and then you dump thrombolytics directly into the clots. Now, there are a variety of ways of doing that. There's a mechanical aspect of it. Some of these procedures are done with pigtail catheters, and they just basically take them in there and start fiddling around to see if they can break up the clot a little bit. Some of them are on the piggybacks of ultrasound guided therapies, where the thought is that the ultrasound will help break apart fibrin and also help with the delivery of lytics. But they're all basically about the same. 

And I think the one thing I wanted to point out particularly for the trainees in the room-- the one downside of this technique, and the one thing that actually is intrinsic to the technique, is that the clot is not necessarily gotten rid of as much as it's redistributed. Meaning that you're pushing the clot further downstream so that you can potentially negate your V/Q mismatch, but at the same time recognizing that you're creating a showering effect. So in theory if you had somebody that had really bad polyfibrosis or something, where they're V/Q mismatch is really unpredictable, and you put one of those pieces a clot down into the one area along that actually works well, you may actually make things worse for that patient. So just one of the things to consider. 

There is some evidence based medicine behind all this. I'm not going to get into the studies. But basically what else they show is that the procedure is safe. A question about whether it's effective in reducing mortality is a little bit more debatable. So right now we feel that this therapy is safe to offer, may clearly have a place in patients that have a lot of comorbidities-- patients who you don't want to give lytics to, patients that you don't want to just give heparin to, or something-- but it also probably has a place for salvage therapy. So, meaning it might be your best backup if things don't go well the way you think they will, particularly for the first 24 hours. 

So as an example of that, I want to show you two echoes here. And I know not all of us are cardiologists. And so on the left what I'm showing you is a normal parasternal short echo. I'm sorry it froze up on it's there. In the middle, we see our LV, and then our RV on the side of it. And then we see RV strain in our parasternal short. And so what we're seeing here is we're seeing an RV that's very generous with an LV that's suffering the consequences of that generosity. 

So I'm going to change our viewpoint a little bit, though. And I want to tell you this is a patient that we saw two weeks ago. And on the left is a patient who came in to the emergency room with the PE that was hemodynamically stable, with an echo that looked like this. And on the right is what his echo looked like just 20 hours after introduction of catheter directed therapies. 

So when we're providers, and we're looking at pictures like this, that's pretty dramatic. And so it is a therapy that we tend to offer at this institution, and at other academic institutions, because of the outcomes that we see here. 

So that doesn't help you, though. And so in 2019, as a result of all these controversies, we have a lot of different things that we can do for these patients. Gold standard is still anti-coagulation, and we still should know that unfractionated heparin has worse outcomes than fractionated heparin-- and particularly with bleeding. We'll talk about that in a second. 

But in reality, which of these you choose? An even more difficult is the fact that, how do you choose them, because there's no randomized control trials. There's no clear consensus recommendations. This is a big group of patients that go on to poor outcomes. And they're really complex. Throw in some cancer in here, throw in some age problems in here, throw in some heart failure in here. So the question is, what do you do about it? 

Thank you, Andrew. Very eloquently placed. 

So we'll talk about now the concept of pulmonary embolism response team, or PE response team. And so as Dr. Mihalek just spoke about this, scenario is fairly complex that we are coming across, where you have high risk patients who historically we have called as massive PE that should be candidates for systemic lysis. But if you look at real world data, 50% of those patients are not candidates for a systemic lytic therapy, primarily because at least half of them have some kind of contraindication to lytic therapy. 

And so this is sort of a scenario where often-- and that's just high risk PE, but we see the same scenario with intermediate high risk PE as well, where these patients might be stable now, but slowly decompensating. And how do we manage them in that scenario? 

So the cost of that has come up. And this is sort of at a national level that is going on-- and actually now expanding to international level-- is development of PERT, or PE response teams. Pulmonary embolism response team, which is a combination of the rapid response team-- just like how you have STEMI or aortic alerts-- but in combination with a multidisciplinary approach, which often as you see as tumor board or [INAUDIBLE] teams. However, this is being done in real time, and not just scheduled once a week or once a month. 

Often you have a very clear cut clinical core group. This consists of a medical team, consisting of-- depending upon what center you are-- could be vascular medicine, cardiology, pulmonary critical care. At UVA, all three groups are involved. In terms of the international group, it could be interventional radiology, interventional cardiology, or vascular surgery. At UVA it's primarily interventional radiology that is involved with catheter based interventions. And for surgical team at UVA it's cardiac surgery. 

The benefit of this scenario is trying to get a more multidisciplinary sort of medical decision making in these situations, given the fact that these are very complex patients. Most of our patients are scenarios where have had just had a GI bleed, and now have a severe massive PE, and they're slowly decompensating. What do we do in this scenario? Or again, a massive PE who's had a head bleed just about three or four weeks ago. And what could we do for these patients? 

It also helps us really develop more hybrid procedures. We are often now doing in high risk scenarios combination therapies, where we'll put patients who are acutely decompensating on ECMO, stabilize them, and then take them for more definitive therapy, and so on. And it also really helps developing accountability between different service lines, because we have monthly meetings. Every time we have a morbidity or a mortality with a case, we actually review, see what happened. Is there's something better we could have done in our approach to these cases? 

Now, with the PE response teams, or PERT, we talked about the clinical code group. But there are a lot of other members that comprise this group. We have direct members who do direct patient care, emergency medicine, internal medicine folks who are actually initially seeing these patients, evaluating them, and then identifying risk, try to find them, and then calling us. We also have the monitoring group, which is the research aspect of it. We actually have a research database that we collect all these patients on, so we can continue to study them. Of course, the safety reporting aspect, and the quality aspect here. 

Now in terms of calling us the-- PE response team at UVA-- please remember, anytime you see a massive or a submassive PE, the number to call is 42012. This is an acute hotline. This is how your calls STEMI, aortic alerts, everything. You just call them and you say, I would like to speak with the PE response team, and they will get your connected with us immediately. Our goal is to get on that call in less than five minutes. Usually we get on those calls within two minutes or so. 

After we review the case with you, we'll decide if something needs to be done immediately-- as in, we need to sort of gather the troops together, get cardiac surgery, interventional radiology, all of them on the call right now, because something needs to be done right away. Or we have time to come and evaluate the patient, and then have a multidisciplinary discussion with all the groups, and then a decision on how we should proceed in these scenarios. So often we'll do that. We are actually monitoring even how much time it takes us to make decisions, so hopefully we'll be able to create a national standard down the road on how we proceed with this. 

Now, the PE response team has actually gone beyond UVA. We've expanded this to other centers. They are able to call us through the transfer center to provide care for patients at our nearby hospitals. So UVA is considered the center of excellence. For an example, we have places such as Culpeper Regional Hospitals that have an ICU that will be managed by the UVA pulmonary critical care group now. However, they don't have aggressive interventional procedures or surgical procedures they can perform on those patients. Often we'll get calls from those physicians, well give some suggestions in terms of how you could stabilize them, or how we need to gather our forces in terms of having those patients come over here so we can proceed with management. 

We have often institutions like Rockingham Memorial where they have some international capabilities. They have an ICU over there, they do have a cardiac surgeon over there as well. However, they don't have a lot of other resources. So often in those scenarios there might be certain things they can do. They'll have discussion with us about it. If not, then we have the patients come here. In one scenario we the patient was really unstable, so we actually ended up sending our team over there to put the patient on ECMO, stabilize the patient, and get them over here. 

In terms of the entire process, this is just talking about acute management. But when we look at the entire process, in the past PE has often been managed-- as all of you know you know-- once the diagnosis is based the ER physician is trying to figure out who to get the service patient on to, what service it might go on depending on if it's a recent post-op case, might end up on you know surgical service-- otherwise medicine service, ICU. What consults need to be done? Who do we consult? Do we consult cardiac surgery, interventional radiology-- do these folks even need to be involved? Often a lot of these things you know have to be considered. 

With the onset of the response team in the high risk group of patients, as soon as we see these patients, we get this call. We'll triage them often appropriately, in terms of what risk stratification has already been done, what more needs to be done, have a multidisciplinary discussion in terms of what might be the best treatment option, and where the patient should end up-- in the medical ICU, or the cardiac surgery ICU. 

And not only just that, but even as an outpatient we end up following these patients. We actually look at their functional outcomes in terms of six minute walk test at three months, echocardiograms-- see if there's any presence of [INAUDIBLE], and if so then have them being seen by Dr. Mihalek in the pulmonary hypertension clinic, as well. So now we'll process through the sort of the management and outcomes of PE. 

So I'm not going to go through this whole algorithm here. But I put it up to show that there is some method to our madness when you guys make the phone calls. This is the algorithm that we have agreed upon using within our program. I will say, I probably make small changes to this every two months or so. So it's very organic. It's very fluid. But it's the work of a team of individuals that feel that this is the best way to actually treat our patients. 

Well, we also recognize the fact that we don't know if it works. And so the question is, has it changed anything. And so we want to transition a little bit and talk about some of our experiences here at UVA by looking at some of our internal data. 

Now full disclosure, I am presenting other people's data. So this is Dr. [INAUDIBLE] and Dr. [INAUDIBLE]-- you both know and love. You should know these people work very hard in gathering this data, and they've put this all together in a very eloquently written manuscript. It was just submitted to JAMA for publication. We're hoping it gets accepted. But we wanted to share a little bit of that with you. 

So our goal live date was March 22, 2017. So we have just recently crossed our two and 1/2 year anniversary. Taking into account basically a one month lead in for education purposes, from April 2017 to October 2018 we took 120 calls. That's the number there. In intermediate high risk calls, and high risk calls, with the interventions offered here-- or as you see forth. 

Just to kind of take a step back and show you some real numbers, this is what we've looked at for the past six months, concluding on July 31. So we're basically taking about between 8 and 10 calls a month. That's a pretty robust number. I don't know what's in the water in Virginia, but there is a lot of thrombin or something is driving clots here. I mean, I went to a pretty high-throughput fellowship. I never saw the amount of high risk clots that I've seen here. 

And just as a way of sharing this, where we're seeing these coming from our different consulting lines, as you would think, about half of them are coming from the ED. A quarter or coming from general medicine. We're starting to get more calls from the other service lines, including surgery and neuro just to keep our epinephrine up. We also like to take calls from labor and delivery every now and then. And then now we're starting to have more of a component of [INAUDIBLE] transfer calls. 

So looking into these numbers, what [INAUDIBLE] and Alex did here, is they very eloquently took what we knew about our numbers from when PERT was instituted and compared it to basically the year leading up to the PERT institution. So we have a pre-PERT and a post-PERT cohort that we're able to evaluate here. 

And what we found is that, with significance the patients that went through the PERT program actually did better at a pretty good number-- six month mortality here-- which is great. Simultaneously though, on the right hand side when I'm showing you is it those patients that we saw in the PERT era were actually more sick as well. So more patients had a heart failure, more patients were mechanically ventilated, more patients had shock. So sicker patients get better outcomes. So we're happy about that. 

As we move forward though, as it turns out the nice thing about instituting a program is you don't catch everybody in that program, because it takes a while for the program to catch in. So within that same time period, we also have basically an internal control that we can look at for patients that did not get caught by the PERT setup-- which is not a small number for us, by the way, in that first year. 

And so what I'm showing you here now is I'm showing you PERT alerted patients compared to the patients did not get PERT alerted. And although mortality outcomes were this same-- I will argue that this is retrospective data, heavily biased by who gets called and who doesn't. What we do find, though, is that the patients that are called and alerted are much sicker compared to those who don't. Which makes sense, right? So if you're worried about somebody, you're going to call somebody for help. And then somebody is going to tell you, oh you've got PERT alert. Oh, I didn't know that. So we're going to go ahead and do that. 

So that makes sense. But I think the moral of the story is that our outcomes are still the same even though we're handling sicker patients. And I think that's a good thing for the institution. 

If we dive a little bit more into the numbers-- and I'm not going to into this in detail, but I do want to highlight a couple of things. I think one of the things that comes up, well, you know shooter's going to shoot. So if we call this program, you guys are going to do stuff. And as it turns out here, at least compared to national numbers, if we look at patients that get advanced therapies through our thrombolytics program, we're sitting at numbers that are roughly about 14%. Now granted, they are more than the pre-PERT era, and they are more than the patients that did not get PERT alerted. 

But on terms of a national average, which is an outcome that people are looking at, between 20% and 30% of patients on a national level are going for advanced therapies. So we're in that magic 20%, right? That's the same number we look at for excavation success in the ICU, a couple other markers. Probably that's our sweet spot. And I think that we're OK with that. 

The other thing I will highlight, too, is that it looks like that our bleeding complications-- although minor-- are similar across the three groups. Major are less. Now, we could argue that some of that's with the intervention of experts who are guiding patients towards better therapies that are supported by the evidence. 

Fractionated versus unfractionated heparin culture busting is one of the things we take pride in. It should be clear that unfractionated heparin, at least in 2019, is a less than ideal modality for intervening for a patient compared to fractionated heparin. Fractionated heparin has lower bleeding risk, quicker times to anticoagulation, and better outcomes for patients that are admitted hospital than just regular old. Heparin so moral of the story here, make sure you use enoxaparin or some other. And it seems that our numbers match that, in that-- long story short-- if you're going to call the expert, we're going to give you expert evidence based advice, keeping in trends with what's going on across the country. 

There's some bean counters in the room. Survey wants to know, what is this going to cost. And when it all comes down to it-- just at least with this retrospective study, over a short period of time-- what [INAUDIBLE] and Alex have shown is that length of hospital stay is not affected by expert care. More importantly, it doesn't seem that cost is actually affected by expert care, either. So in that regards, we're giving better care for the same amount of money. 

Now, yes we'd like to save you guys money. That's what we're here for. We get it. But I think it's a good sign, and says that we have all of these tools available to us. We have all these practitioners available to us. We have all this expertise. So we've reduced mortality risk, but we have an increased cost along with it. OK, we're a lean program. That's what everybody wants to hear, right? So we're a lean program. So long story short, at least at UVA in the time that we have looked at, what we have found is that there seems to be a benefit of this program in that we take care of sicker patients. And patients that are intermediate risk and high risk tend to do better on our watch without increased costs. 

So there are also benefits that we need to talk about that go extend their power beyond the data. And the first thing is that we get give customer service. And that's the thing that we hear over and over. Now, I have no data to back this up, and yes I'm going to hear what I want to hear. But for the most part, for the people that were involved in taking care of these patients before we had this program and had to rotate through the ER-- I'm looking at some of the fellows here-- I think you remember how was trying to find who's going to take care of this patient. The IR attending is not talking to the ICU attending, he's not talking to the heme attending, he's not talking to the pharmacist, and I have to go see somebody who's got a broken toenail, or whatever they're doing. 

So that's important. We now provide an option for these patients to be taken care of in a seamless manner, that can bridge silos-- which is really what we should be about in 2019. Our patients seem to be better served by this. We have a guaranteed follow up program, assuming everything works. And patients like knowing they're not a disease. They like knowing we're going to be invested in your care at least for the next year. And that's helpful for us as an institution. Outside institutions appreciate having us as an expertise. We're the only institution in the current region that's offering these services, so we have the ability to reach out to places that are seeing these sick patients and are scared. 

And then the other non tangible event is it our ancillary staff has told us that taking care of these patients is a lot easier. Our nursing staff like the fact that we are trying to standardize the care. When somebody comes up on a tPA drip, it's nice to know whether that should be five or 10. It's nice to know whether we're going to be doing that for 24 hours. And now, as opposed to provider preference dictating the care, we now have a program that delivers a care in a way that's consistent and makes sense. 

And then the final thing is-- at least to the hospital leadership-- we're utilizing real resources that already exist. I mentioned the lean part. And then I'll also say that one of the things that we're most proud about is that our mentorship for our trainees is phenomenal. So I cannot tell you how many trainees attend our Thursday 7 o'clock and morning meetings. It's inspirational. We have residents, we have fellows, we have super fellows who have all taken all of this to publication and are using us to advance their careers on a national platform. And we're very proud of that, too. 

So our future directions, though-- in the last couple of minutes we're going to highlight-- is that we want to get better. And so we want to make sure that our systems based approach is appropriate. So we'd like to use the tools of evidence based medicine to guide those therapies further, and push them some so that patient outcomes can be much better. We have a large quality improvement aspect of our research that we're hoping to jettison. 

One of the things that we're most interested is, are the markers that we use to risk stratify patients appropriate? What happens if somebody has LVH, and we're interested in knowing LVH to RV ratio to determine whether there's RV strain. That's an interesting question. Can BNP to be used in obese patients to determine whether that's a reflection of RV strain? It's a good question that we're hoping to answer, We're looking to develop a blood bio repository database, and we'd like to have kind of a similar structure that the other alert two programs have. 

And then the final thing is the therapies that we're offering-- particularly the ones that are coming down the pipeline-- are they ones that are appropriate and are safe? And so one of the things that Dr. Sharma would like to [INAUDIBLE] back up and to share us with is one of the things that's coming into vogue now is the use of ECMO and the use of these patients. Early ECHO. So the question is, is that helpful? 

So one of the key things I think I'll briefly talk about is the use of ECMO for PE. This is sort of coming into a bit of a rise now. The goal of using ECMO so that we can rest the RV, which is under a lot of strain from a intermediate high risk to high risk PE, and see if that stabilization can help improve mortality outcomes, or even morbidity outcomes. And primarily for this, often used is a V-A ECMO. V-B ECMO typically just helps with respiratory insufficiency. 

So these are some of the national trends that we have seen for utilization of ECMO. And you can see there's a massive surge somewhere around 2010, 2011. This was predominantly in urban academic centers, where utilization affect more went up. What's interesting is the HA guidelines came out in 2011, and didn't really mention anything about ECMO because of lack of data. But folks had started using this already. 

This sample database actually showed that in-hospital survival for massive PEs was about 61% with utilization of ECMO. And mortality was lower in massive PEs when ECMO was utilized, in comparison to not using ECMO. Certain institutions have now adopted this in a protocolized fashion, such as University of Maryland Cardiac Surgery Service. What they have done is in the protocol, as soon as they get consulted for a massive PE they first see if the patient is a candidate for ECMO. And if so, rest the RV on ECMO and then decide if anything more is required. 

What they have found out is that their one year survival improved significantly, to 96% in this group of patients compared to 73% prior to that. I do think there's a bias in this group of patient selection, because we don't typically see even 70% mortality improvement, which they have reported in the past. So I think they do select a certain group of patients that perhaps are not as sick. Regardless, there is some benefit in ECMO use in stabilizing PE. 

So this has sort of led to certain initiators. What I want to highlight as one of the things is one of our cardiology fellows, Dr. [INAUDIBLE], has actually worked on doing meta analysis on high risk PE patients and utilization of ECMO. And this is a very detailed meta analysis that has not been done as yet that she's been doing, and it's now ready for submission for publication. And what she's found in about 944 patients that we have data on, literature on this and found out that basically the survival to hospital discharge was about 60% with utilization of ECMO. And if you just specifically look at US based studies, that was actually about 80% survival. 

We actually try to look more, see whether or not in massive PEs if we use ECMO is there any outcome difference. And it actually showed that there was no difference in mortality, whether or not ECMO was used in massive PEs. But when you really look down into the group of patients-- what patients went in these groups-- 34% more patients had pre-ECMO cardiac arrest in the ECMO group. 

So one of the key aspects with massive PE to understand is that when there's cardiac arrest involved-- in the case of massive PE-- then mortality is close to 90% to 95% in general. So clearly a very sick group of massive PE patients were more present in the ECMO category. And so even though there was such high mortality risk patients-- almost 34% more-- in the ECMO category, yet there was no difference in terms of the mortality. So normally we would expect that the mortality would be significantly higher otherwise in that group. 

So we kind of broke down further and said, let's just look at patients who received ECMO and then see massive PE without cardiac arrest versus massive PE with cardiac arrest. And ECMO then much more favorably better and patients who had not yet had a cardiac arrest while you were trying to put them on ECMO, or after cardiac arrest trying to put them on ECMO. Sort of makes us wonder would the outcomes be better if ECMO in massive PE was considered sooner, before your patient ended up getting a cardiac arrest. And what if we considered ECMO as soon as we found somebody had a massive PE. 

So another important aspect I want to talk about is from the PE response team perspective is risk stratification. So this is some data we had collected as Dr. Mihalek talked about-- the pre-PERT data. And what we had seen is, how do we do with risk stratification wise. And so in the blue line you can see here various markers, photo stratification, how they were utilized before both program existed at UVA. Only 45% of patients had received [INAUDIBLE] had been checked with these in an acute setting. 31% with BNP, similarly for our [INAUDIBLE] on all three markers. 

What was interesting is, of course, that patients who had PE response team and sort of alerted, 99% markers were checked. And pretty high significant numbers also had [INAUDIBLE] reported or evaluated immediately. However, what's important is to see the orange bar also. So these are the non PE alerted patients. Even those improved. And a lot of this, I credit this to the PE response team, because physicians learning about the fact that there is a team-- usually they require this photostratification of PE-- actually in turn increased proper risk stratification in general for all PE patients, regardless of not the response team is sort of included eventually in their management. 

So the other aspect of it is risk stratification, based on the European Society of Cardiology guidelines, in terms of stratifying the high, intermediate, high intermediate, or low risk patients. Again the PERT group has all almost all of their patients evaluated for that. But when you look at the non PERT group of patients, even those are better than the historical group. Again, the same fact, I think the involvement of PE response team-- overall education of the PE response team, in terms of risk stratification-- has really helped this aspect, as well. And proper risk stratification and entire system based management of these patients. 

What's also interesting is this a paper that was published by the PE response team at the University of Rochester, who actually showed that impact of their team in the emergency department has gone beyond interventions. What they found out is that now they have significantly reduced the time to patient triaging for PE diagnosis, time to PE diagnosis to heparin infusion, and time from triage to PE admission. So in fact it helps in all these other aspects, beyond what you would think about in terms of just doing more procedures, or hopefully reducing mortality. There's added benefits dubious aspect of it. 

So in conclusion, we've sort of discussed how we would distinguish high risk pulmonary embolism patients from intermediate or low risk, review their treatment options, discuss what we have done at UVA with the PE response team, also. And remember to call us. It's 42012, and if you can get us immediately from that. 

Lastly, I really want the credits-- this is maybe half of the people that we can put on our slides right now that have been involved in a very crucial development of this program. And also, would like to actually give a very large credit to Dr. Brian Annex. I don't know if he's here today, but Dr. Annex is the chief of cardiology. Unfortunately he's leaving us now, but without his support and help we wouldn't be here today discussing this program or how it's being developed. His constant support for the last two, three years have really helped us not only develop this clinical aspect of the program, but also the research aspect, because there's a lot of funding that's involved in this. All of this has been supported a lot through him, and the heart center together. So thank you. 

[APPLAUSE] 

All right, we'll take questions now. Please just raise your hand. I'll bring the microphone to you. 

Really incredible work, and really a great program that brings a lot of distinction to UVA. And you guys really give a great talk. So, one question on your whole be RV strain-- so one of the things that comes up, at least from the echo perspective, is we sometimes have people who don't have a known PE and we're asked to evaluate for PE or PE strain. What's your guys thought about that relative to sorting out if they have a PE first, and then getting the echo. And then secondly, what do you do about the people that can't get a CT? What's really the right pathway for those guys? 

So you can do the first one. 

So I think that's a good question. Obviously this is is sort of beyond the PE response team aspect. We get called in once you've found there's a PE on a CT, or some one way. But I think it's important point-- RV strain can happen for any reason at all. So just because you find RV strain doesn't always definitely ever tell you there's a PE. Even if there's DVT in the legs, it is no way to definitively diagnose a pulmonary embolism. 

The second aspect is again, you can a PE without restraint on echo. So it can be misleading to have that information there. If you don't see RV strain, assume they don't have a PE, they may have a PE. And it may eventually progress, and actually could be life threatening down the road. So I don't think that's a good modality to consider it. 

And so, that was kind of the aspect. I'm sorry, I forgot your second question. 

Yeah. I think you can get a VQ scan if needed, or even an MR. But I think most of the time you'll end up getting a VQ scan. I think that's better. Most of the time, if you have time you can hydrate them. Just put them on heparin if possible, if they're hemodynamically stable. That might be sufficient to wait till you can get a CT, otherwise VQ is the next modality. 

There are scenarios where we'll get called where the patient is decompensating, or it's a massive PE. You don't have time really even go through VQ. In those scenarios, sometimes you might just go straight for a pulmonary angiogram, because you will use very small amount of dye, potentially, and maybe do something at that time. But this is limited to really sick massive PE patients. 

I think the moral of the story as well is that everything we know about PE management-- and we get away from this, because we're now a field of technology, and labs, and things-- everything we know about is based on pre-test probability. And so if your pre-test probability is high enough that I think that you're calling for an echo blindly because you think that it may help you-- which it won't. In that case, you should probably be anticoagulating your patients. 

And I've seen this both ways. So, [INAUDIBLE] and I had a very intense patient that ended up being on [INAUDIBLE] who was a massive PE. He was billed as a PA arrest, but it was clear there was a massive PE. [INAUDIBLE] at cardiology, [INAUDIBLE] cardiology fellow was there. She was like, this is a new right bundle branch block. This patient is tachycardic, hypoxic, and they're hemodynamically unstable. I think that this is a PE. We should be anticoagulating that patient before anything else is done. 

Zack and I also had a patient who couldn't get a CT scan because she was unstable. Her creatinine was terrible. She is a little bit on the heavier side. And we spent a good 10 minutes talking about it, but the moral of the story was that our pre-test probabilities-- we thought there was a PE. So rather than waiting for all that stuff, we anticoagulate. 

The final thing is probably one of the markers for outcomes in managing these patients in the long term that will be looked at-- in terms of quality markers-- is timed to the anticoagulation. And current guidelines support ED staff to start anticoagulation before a CT scan is done. Meaning if you hit the button for a CT scan, you should be hitting the button for the Lovenox now. And then you can always take it back. You can always stop it, et cetera, et cetera. But if the pre-test probability is what's guiding your radiographic study, it should also be getting your therapeutic interventions at the time of that decision. So I think that would be also the moral of the story. Eric? 

So, hi guys. Sorry. The question is, is local tPA given catheter directed into the PA really local, or does that actually become systemic? And one of the struggles I had in dealing with this giving a catheter directed or systemically, is sending them out of the intensive care unit with the potential risk of not being in an intensive care unit. 

Yes, I just want to cover this. The question was is the catheter directed tPA bad? 

Is it really local? It eventually becomes systemic? 

So the problem is we don't have enough large body data to support the answer to that definitively. But what I can tell you is that the bleeding rates for patients that get catheter directed therapy are lower, but only slightly so, and not something that catches my eye compared to systemic. But the argument there is that the total dose of tPA given across a 24 hour time period is about a tenth of what you would get systemically. So if we carried that question out to a larger cohort of patients, there is an argument out there that we should see less complications from our catheter directed therapy patients. 

Now, is that true or not? We certainly don't know. That's why we're studying it, and that's why we're talking about this with you today. But it's that total cumulative dose of tPA-- remember, we're running at about a milligram per hour for 12 hours or so, versus giving a slug of 100 milligrams upfront. So that's the cognitive dissonance that should occur in your head. And it's hard to answer that. 

So if you look at clinical trials data, bleeding from systemic lysis is about 2% to 3%. But catheter directed lysis about 0.35%. But these are clinical trials, so they're not real world. Obviously there's a selection bias in a lot of these. The systemic licensed clinical trials, the majority were massive PEs whereas catheter directed lysis often is a mix of high risk and intermediate risk PEs. 

And of course, high bleeding risk patients often get excluded in these trials. So we still have to really know what truly the real world is. But just like how Andrew said, it's 1/6 to 1/10 the dose. And it's being given in two hours, it's given in a period of 24 hours. So that's the difference. 

Thank you. You said something very interesting at the end about giving anticoagulation straight away. When VQ scans were introduced, one of the effects was you suddenly got an hour delay in the situation when clinically the diagnosis had already been made. And we used to insist on 10,000 of heparin IV stat before they move anywhere. Isn't the danger of a PERT system that when you're really enthusiastic you don't delay things, but if it slips a bit it can end up being a reason for delay in anticoagulation? 

I agree. And the pendulum also may swing the other way, that maybe we start seeing bleeding outcomes worse, or something along those lines. I think housed in that question-- I hate when you go to a conference where people are like, I have a comment, not a question. So I will answer your question with a comment in that one of the things that we worked really hard with our group is ensuring that all of the alternative modalities-- all of the surgical modalities, all the interventional modalities, basically anything that you can think that we can offer-- all the groups that will offer those have agreed that they will offer them on full anticoagulation. 

So we can get away from, oh, I'm going to give them heparin until I find out the surgery is going to take him. Surgery will take them for ECMO. Surgery will take them for a thrombectomy on one milligram per kilogram of enoxaparin. So that relieves you from a lot of having to make those conversations happen in your head. It's just, this is a PE, I'm going to anticoagulate. And then I'll figure out the rest later. 

The interesting aspect of this is the University of Rochester data that I showed-- in fact, after the PERT program was initiated at University of Rochester, they saw that in their ED the time to heparin initiation in a PE patient actually improved. And a lot of this is because now you have folks who are telling them, put them on anticoagulation immediately and we'll come figure it out. 

That's one of the things I often talk to our group, also, and folks who call us. If you think this patient needs aggressive therapy, don't hold off on heparin. Start the anticoagulation, start low molecular weight immediately, and then call us after that. Just because you started that will not change how we approach the patient. And one of the part of our program is when we get called, the first thing we ask, is this patient already on Lovenox or not? And if they say not yet, just go ahead and give it right now. That's sort of what we do. 

We're seeing a lot of folks these days getting started on DOACs rather than on low molecular weight heparin. How does that affect your thinking, and your algorithm if they've gotten that, and then someone thinks to PERT alert them? 

Luckily we haven't had that scenario as yet. I think most of the time if they've been risk stratified, either into intermediate or high risk, they are not getting DOACs right away. But if that happened, we might still proceed with interventions. 

The only problem with that is DOACs will stay in your system for up to 72 hours. So I think it would worry us about complications. It's OK to put catheters in, it's when we start getting them out. That's when the bleeding issue starts. So that will become a little bit of a problem at that time. 

But it may be fine. We don't know. We haven't tried that out as yet. Maybe we'll be OK. But as of now we haven't experimented with that. 

Our institution wise, we're a heavy DOACs institution thanks to Dr. Mason. And so we have to recognize that upfront. But one of the reasons we wanted to get this thing up and off the ground when Dr. Mason was here-- we were doing DOACs for everything, honestly-- is anticipating that it will be the trend. And maybe getting in on the ground floor, particularly on a research side-- to address, is that the right therapy? Because Pradaxa certainly has a reversal agent now. There's some other ones coming down the line for the others. But a little bit more of what you're buying into for this next 72 hours. And so we wanted to make sure that we were part of that conversation earlier and not later. And so we'd like to push the institution as opposed to play catch up with all the other institutions. 

All right. Thank you, everyone. Thank you Dr. Sharma. 

[APPLAUSE]