Thanks, guys, for taking the time after lunch. It's a nice intimate little group here, so I'll be quick. Today I'm hoping to talk about the pulmonary embolism response team at our center at the University of Colorado and what we found. This is my brief cheesy get to know you slide. We can skip it, but you can find me on Twitter at August Longino, or if you're no longer a Twitter person, perfused and enthused at quickcare.social on Mastodon. No disclosures. But what I hope to do in the next 15 minutes or so, so I want to talk to you about the trends in the treatment of intermediate and high-risk PE in Colorado and elsewhere, emerging evidence for pulmonary embolism response teams, a growing force in the treatment of this disease. And I want to talk about the results of PERT implementation versus PERT consultation at our center, and why it's important that those are two different exposures for our patients. Discussion, and then kind of next steps in what we're doing at Colorado. So where we stand, you guys have all treated PE, so you're aware that they are common. These are national-level statistics, and they can be quite deadly. So the range of mortality varies between massive and intermediate-risk PEs, from 22 down to 7.7 or so. And you probably also know that the management options for PE are complex and getting more so every year. So you can see here that starting back in the 1970s when we developed IVC filters, since that time, there's just been a dizzying array of different catheters and suction devices and all kinds of different gadgets and gizmos that we're able to deploy to help save our patients. And with that increasing complexity, we've needed more and more people to help manage that complexity. So this is where the PE response team was born. This is a multidisciplinary team composed of a variety of different practitioners, from pulmonologists to interventional radiologists to surgeons, that come up with treatment plans that can rapidly be applied to patients with PE. So the first of these started in 2012 at Harvard. A national PERT consortium was founded in 2015. And then more recently, international guidelines have endorsed the use of a PERT at major tertiary centers. And as a result, these have been widely adopted. There's now well over 100 in the United States and more internationally. But there's a question, which is, do they actually help? As in so much of medicine, there's a lot of things that we do without knowing whether or not these are really helpful interventions. And so far, the truth is that evidence is mixed. So there's a recent meta-analysis of 16 studies. And what they found was that if you look back at PERT implementation over the years, there's not been a huge mortality benefit. There's not been a lot of change in bleeding complications. It does seem that with PERT implementation, we've seen more procedural interventions, so more things like catheter-directed lysis or suction catheterization to remove clots. And then there's been a decrease of IVC filter placement. Now, and this is just to summarize those points. But I do want to highlight an important limitation of studies like these is that most of them are retrospective. And basically, what they do is they look at a pre-PERT time point and they compare it to a post-PERT time point. And they say, what's the difference between A and B? And you could imagine that that kind of study could be easily confounded by changes in global practice. So you might remember back in 2015, this study came out, which is a negative trial for IVC filters. And then just a few years later, in 2019, they repeated a similar study in trauma patients, which was likewise negative. And you could imagine that if you were comparing simply a pre- to post-PERT analysis, that might account for a graph that looks like this one. People's global practices change regardless of whether or not a PERT was involved in their care. Similarly, you'll recall from that meta-analysis that there was an increase in those advanced procedural interventions. And I'll remind you of our first slide, where you saw that basically more tools have become available to us over time. So as the toolbox has grown, you can imagine that more procedural interventions are performed. So what did we do? So we wanted to look at two exposures. First, the presence of a PERT at our institution, and then consultation, which is different. And so we first replicated some of those older studies, and then we tried to account for that temporality component. So we looked at a lot of the same outcomes, mortality, length of stay, anticoagulation, bleeding, and the number of procedures that patients underwent. We did this via retrospective chart review. You can see I got 684 patients that came in with intermediate or high-risk PE. And then that first exposure that I talked about was just a pre- and post-analysis. But then within the patients in the post-PERT era, we looked specifically at those who did and did not receive a PERT consult, so really nailing down the intervention of interest. And so when we looked at just the pre-post-PERT era, our PERT started April 1st, 2019. We saw that in the pre-PERT era, you're a little bit more likely to have cancer, and of course, much less likely to have COVID-19, something else that we accounted for. We did see that in the pre-PERT era, you were much more likely to undergo catheter-directed lysis. You're much more likely to have an IVC filter placed and less likely to undergo thromboaspiration. There was no difference in time to therapeutic anticoagulation, nor 30-day mortality, nor the hospital length of stay. So really, our conclusions here at Colorado, the only thing that really distinguished us from prior years was this change in procedures. So as before, other literature had suggested there was more procedures, we saw fewer. We also saw that pre-PERT patients had more cancer and less COVID, and that we did fewer procedures overall. We did note a trend towards an increase in thromboaspiration. I agree, a fraught word. But no changes in 30-day mortality, length of stay, time to anticoagulation, largely similar to things before. We did do multivariable models accounting for common comorbidities, like CHF, cancer, and cirrhosis, as well as age and sex, and still saw the same outcomes. So a little bit of change in procedures. Otherwise, similar to what's gone before. But then we narrowed down and looked at PERT consultation itself. So this is just in that post-PERT era. And we saw a couple of things that interested us. So first, the patients that received a PERT consult were more likely to have COPD. But patients with cancer and cirrhosis were less likely to receive a PERT consult. And those with thrombophilia were more likely to receive a PERT consult as well. Another interesting finding, of the 369 patients who were eligible, only about half of them received PERT consultation. About half of those that could have. And we saw the main difference in terms of procedures was that the vast majority of people who received thromboaspiration or Flotriever received a PERT consult. And that was probably because you were required to have a PERT consult in order to undergo that procedure. But this is where it gets interesting. So what we found was that there was a significant difference in 30-day mortality if you received a PERT consult. So 90% versus 74% of patients were alive at 30 days if you received PERT consultation. Another remarkable finding, we thought, was that you had a dramatically shorter hospital length of stay. And in order to elucidate why these disparities might be there, we did look at whether or not a patient's goals of care might be affecting this decision. Whether very gravely ill patients might not be getting this option. And we found that only in 7% of cases did patients' goals of care specifically preclude PERT consultation. We did find one expected finding is that if you got a PERT consult, you were more likely to receive inoxaparin as your initial anticoagulant as opposed to unfiltered heparin. And accordingly, you had a shorter time to therapeutic anticoagulation. Again, a trend towards significance. We did the same multivariable modeling that we did for that pre-post analysis. And here we see strikingly different results. So an odds ratio of mortality at 30 days of 0.35% for PERT recipients. And a decrease in your hospital length of stay by a little over five days. So briefly, this is a miraculous thing. A mortality benefit. Shorter length of stay. Shorter time to anticoagulation. Greater use of low molecular weight heparin or inoxaparin. And greater use of thromboaspiration. But the highlight, and I think the most concerning finding to us, was that only 54% of patients were receiving these consults. And there are, of course, potential confounders here. When we think about these results, there are important caveats to mention. So one important one was that non-PERT patients were more likely to have cancer and cirrhosis. Which introduces the possibility of something like a futility bias. You can imagine that a provider with a patient who has terrible cirrhosis or advanced cancer might say my patient is simply too ill and they would not benefit from PERT consultation. And that might very well account for these results that we see. But bear in mind that the goals of care of the patient only would have affected that decision in about 7% of cases. So if this bias exists on the part of providers, we know it does not exist on the part of patients. And then, of course, there are changes in global practice patterns that would certainly affect that first analysis, as we discussed earlier. And there are, of course, as has been discussed in terms of all of our other propensity matching caveats, that there are always untold other effects that you can't account for in retrospective studies. But things that we're working towards at Colorado, we're working to increase the awareness of the PERT team. And we're looking at doing a QI now and trying to figure out which specialties and which locations within the hospital are doing the most PERT consultation and who's not. We're trying to characterize the barriers to PERT consultation. And then I would really encourage everyone that's interested in this kind of research to always remember to study the exposure of interest, which in our case was PERT consultation itself, rather than simply the existence of a PERT at a given institution. Thank you so much.

Pulmonary Embolism Response Team

multidisciplinary team

mortality

hospital stays

PERT consultation