Akhilesh K. Sista, MD, FSIR, and James Horowitz, MD, are members of the NYU Langone pulmonary embolism response team (PERT), a model that brings interventional radiology, pulmonology and cardiac surgery together to improve outcomes for patients with pulmonary embolisms. With Dr. Sista’s IR skill set and Dr. Horowitz’s cardiology knowledge, the two have been part of a successful collaboration model for over a decade.
What is it about PE that makes an interdisciplinary approach like PERT so necessary?
Akhilesh K. Sista, MD, FSIR: The beautiful thing about PE is that it has multiple treatments that span different disciplines. There aren’t many single-entity diseases that have a medical, surgical and interventional treatment option. So it makes each stakeholder in the PERT very important in terms of decision-making.
Obviously, every PE patient who can be anticoagulated should be. That’s not a multidisciplinary decision. But when you have someone who cannot be anticoagulated, a PE can be disabling or even fatal. That is when coming together and deciding if escalation of therapy—be it catheter-directed therapy or surgery—is salient to the patient’s outcome. Having all stakeholders in the room is crucial and makes PE treatment so ripe for collaboration.
James Horowitz, MD: Because PE treatment involves so many specialties, PE patients often experience serial consults. It takes time to call Physician A, who suggests you call Physician B, who wants input from Physician C, who disagrees with the model of treatment or requests additional imaging. It’s like playing telephone, and you lose time—and when you have a sick PE patient who can decompensate quickly, waiting hours for consults is dangerous. When everyone is in the room, those conversations happen faster and the treatment model can be customized.
It is also important to consider that there aren’t many guidelines that say when to do a thrombectomy, for example, in a complex case. With cardiology patients, I have association guidelines, and even stroke now has guideline-driven interventions. But PE is still a bit of a grey zone, so the benefit of the interdisciplinary approach is that you can leverage every specialty’s knowledge and experience to have a thoughtful discussion on whether or not to intervene.
With so many specialties in the room, is there difficulty reaching consensus?
AS: This is where relationships start to matter. When you work in a model like this, you start to develop trust in your colleagues’ institutional knowledge and skills. When one of the pulmonologists says, “I’m worried about this patient,” I listen to that more than anything else. We see a lot of PEs together, and so our decision-making has evolved as a group. There may have been disagreements at the start, but that’s rare now. We all understand the merits of each other’s therapies and can step back to recognize what is best for the patient.
JH: It’s a bit like a jury: You stay until you have a decision. Not everyone is happy with every aspect of the treatment plans, but they still recognize it as the best option. The patient outcome comes first.
At what point does your team get called in?
AS: We instruct the hospital and our referrers to call us for any PE they’re concerned about. We don’t ask them to do risk stratifications—we want people to feel comfortable calling us, so we don’t put them in the position of filtering something they may not be an expert in.
There are two categories of PE that we may be activated immediately for. The first is a massive PE in a patient who can’t receive systemic thrombolytics and isn’t a good surgical candidate. We would activate and most likely do a mechanical thrombectomy on that patient, because rates of mortality for massive PE, even with anticoagulation, are very high. The other category of patient we may activate for is someone who isn’t yet hypotensive but is trending that way. Their oxygen requirements are going up, their blood pressure is getting soft, they’re tachycardic and not going in the right direction despite anticoagulation.
But there is a category of patient we treat who has a submassive PE and is a next-morning candidate. Those patients are the ones who, 10–12 hours after anticoagulation, do not see improvement in their heart rate and oxygenation. Our preferred intervention in those patients is a catheter-controlled thrombolysis because we have the time to do so. They aren’t actively crumping, so the longer infusion—and there’s debate in the literature about how long that should be—usually gets to these patients quickly.
Do any case examples highlight the value of this model?
JH: Over a decade ago, Dr. Sista and I had a case at Cornell, where a 35-year-old woman came in 34 weeks pregnant with a submassive PE. We put her in the medical intensive care unit, but her condition didn’t improve. She had more chest pain and shortness of breath, and her repeat echocardiogram showed she was getting worse. It was a difficult situation, because surgery didn’t want to intervene until the baby was born, but the OBGYN didn’t want to deliver the baby until the clot was resolved.
We ended up putting everyone physically together in one room—IR, cardiac surgery, pulmonology, OBGYN, pediatrics and cardiac anesthesia, and we made a plan. We took the patient to the hybrid OR and provided cardiac anesthesia. The IR team was on one side of her bed and the cardiac team was on the other, and at the foot of the bed there was OBGYN and a high-risk pediatrics team, ready to go. The hope was to intubate her and do catheter-directed thrombolysis, but if she crashed, we had back up. We called it a triple set-up.
It worked amazingly well. She remained stable, the clot was removed and she had an IVC filter placed. She gave birth vaginally a week later and IR took out the filter, and she and baby were fine. I love this example because if we had asked everyone serially, we may not have had that outcome.
What is your advice to other physicians looking to join or create a PERT?
AS: You need a very good partner in pulmonology, and someone who will be the point person for all medical care for your patient. You also need a good process for getting the data expeditiously, whether by collaboration with cardiology to get certain echocardiograms or building a process with other collaborators. It is also important not to approach PERT like a proceduralist. Approach as a physician who is keen to get the best outcome for the patient given the state of the data. You cannot work in a PERT model if you don’t respect the value of other specialties and their treatment options. If you’re able to be objective and work toward a consensus decision, you’ll gain the respect of your collaborators and be more likely to be involved in the decision-making process.
JH: There is no one way to form a PERT. It’s a good idea to work with pulmonology and add hospitalists, because they will comprise most of your consults. But some institutions have all IR or all cardiology teams, and each person is given ownership of a certain aspect of care, be it the imaging or the procedures. The key is having a team who can work together and is comfortable with different skills and has a passion for the work. The makeup doesn’t matter—it’s a coalition of the willing, as Ken Rosenfield, MD, from Massachusetts General Hospital says.
