All right, everyone. Thanks for being here today. For those of you that don't know me, I'm Walid, one of the chief internal medicine residents. Thank you all for being here. This is our second grand rounds now in this new format. We were hoping to kind of take advantage of the new, nice location to mix up a little bit how grand rounds works. And so for the talk we have for you today, hopefully, it will be really interesting. It's bringing together a very multi-disciplinary team to talk about a topic that is of importance to a lot of people that provide care in the hospital. I'll give you a little bit of a very, very brief intro on everyone, and then just a little bit about the format of what the discussion is going to be. I'm sure all of you kind of saw the topic of the talk. It's going to be a discussion between a representative from the Infectious Disease Department, a representative from the Pulmonary and Critical Care Medicine Department, and then Dr. Douvas is here as a bit of a mediator role. So kind of-- [LAUGHTER] So like I said, Dr. Douvas has had the fortune of being our peacekeeper. He's representing the perspective of the Hem/Onc Department, a service for whom obviously today's discussion is very, very pertinent. He's going to start us off with our talk. He's going to give just a few minutes, maybe five or so minutes, just to introduce the topic, introduce the other two speakers a little bit, and give us some background on what we're going to be talking about today. And then we have our two speakers, Dr. Donowitz and Dr. Davis, each of which is going to give about 15 minutes of time discussing their perspective on this topic. And then after you get a chance to hear from each of them, we'll bring it back to the whole group, and all three of them will get together and talk and lay out a little bit for you all what the big picture looks like. And then hopefully, we'll have a lot of time at the end to spare for questions. We have until 1:15 today. So that's the time we'll plan to go for. But before I bring up Dr. Douvas, I'll just mention very, very quickly, for those of you that don't know Dr. Davis, obviously, like I said, he's here representing the Pulmonary and Critical Care Department. He's been with us here at UVA. He's been a fixture in our Sleep Medicine Program since he's been here. He's been an excellent educator to many of our residents that have rotated into the sleep clinic, myself included. I've had the fortune of working with him. And outside of his clinical duties, he's extensively published in the area of sleep medicine, and specifically studies the cardiovascular outcomes of sleep disorders like sleep apnea. And then representing the infectious disease side, we have Dr. Donowitz here coming to give us the ID perspective. As you all know, he's played a very significant role here in our department as prior program director, as well as current vice chair of medical education, has worked closely with a lot of our residents, and played a major role, obviously, in my own education here, and is extensively published in the area of immunocompromised patients, specifically patients with neutropenia and the infections that they face. So without further ado, I'll yield the floor to Dr. Douvas here to introduce the topic to you all. And hopefully it'll be fun-- and civil. [LAUGHTER, APPLAUSE] So one thing I didn't get was a primer on how to bring up the-- there we go-- There it is. There it is. This guy right here. Perfect. There we go. So Dr. Donowitz actually told me I had eight minutes. So I might go a little bit longer. He was very specific, so I'll watch my time. So I would say that first off, none of the three of us have any relevant personal or professional financial relationships to disclose today. I think my role is to frame the conversation, to-- I'm going to present a few cases of patients of mine in various settings who have had pulmonary issues in the setting of immunosuppression, and whet your appetite in regards to thinking about how to figure out what's going on with these folks, and then give a brief overview of the approach, and then turn it over to each of them. These are the learning objectives that hopefully, we will meet. So for those of you who don't know me, I've been here for close to 13 years. I trained in internal medicine and pediatrics [INAUDIBLE] Monmouth University of North Carolina Chapel Hill. And what I spend most of my time doing here is taking care of adolescents and young adults with cancer. And that's primarily with leukemias, lymphomas, and sarcomas and rare, unusual things. Each of these cases is one of my patients. They're in vastly different contexts. But I think we'll highlight, in each of them-- we had a diagnostic approach. I think I might be the only person in the room who knows what actually-- I know what two of the three of them have. One is actually still a bit of an unknown, and it has played a critical role in his care and ongoing care to try to figure out what's going on with him in regards to his pulmonary process. So the first is a really sweet young woman who has lived in the States for many years but originally immigrated from El Salvador as a child. And she had early stage Hodgkin's disease, for which she completed six cycles of ABDD. She really did wonderfully well with her treatment. And not too far after it completed, she called me, because she had a cough. And I thought, common things being common, it's going to be something that's very simple. And we tried a variety of simple things like a chest X-ray, which was unrevealing, empiric antibiotics, which didn't help, a proton pump inhibitor, and anti-histamines, which didn't help. And then ultimately, she had advanced imaging, which showed that she had a pulmonary process, as can be seen in this nodule that's at her left hilum. And so we went off trying to figure out, in someone who's recently completed treatment and immunocompromised, what could be causing this? The second case is also another young woman who I met from the Danville area, presented to us very sick with malignant hypercalcemia, and an absolutely gigantic mediastinal mass, with a diagnosis of primary mediastinals lymphoma/ diffuse large B-cell lymphoma. She had central nervous system involvement. And she actually got through her treatment very well, and it seemed to be in remission. And again, it was a little bit longer afterwards. But in routine restaging follow up, she had absolutely no symptoms, had absolutely no findings that would otherwise suggest of a recurrence of her lymphoma, but had this extremely interesting chest CT with all these sort of micro nodules with cavitation throughout her lungs. And again, we were charged with the task of trying to figure out what is going on within her lungs and with her. And the third case is the one that's the most active. He's unfortunately not doing very well in the hospital currently. But he's a 29-year-old man who I've known for many years, and had T cell lymphoma for which he went into a sustained remission, but had a recent recurrence, and as a part of that recurrence, had fevers, and had this pulmonary process discovered with symptoms of a cough and ongoing high fevers . So three very different looking chest CTs, three different presentations, but all of them, patients who are immunocompromised with a pulmonary process-- so initially, when we were asked to give this grand rounds, I think we were specifically asked to talk about the work up of patients in the context of febrile neutropenia, but we decided to broaden it to make it the workup of pulmonary processes in patients who are immunocompromised. So we're really going to be talking about patients with solid tumors, humans hematologic malignancies, stem cell transplants, who are immunosuppressed in a variety of ways, including sometimes in the context of neutropenia. But we feel that our conversation will be relevant to patients who have solid organ transplants, who have autoimmune processes, and are immunosuppressed sometimes as a consequence of that, of patients who have acquired immunodeficiencies such as HIV, and people who are born with congenital immunodeficiencies of all types. The differential diagnosis for these folks is very broad. And this is not meant to be a comprehensive list, but to start getting you thinking about all the things that it could be when you start hearing Dr. Donowitz and Davis talk about the tests that we do. Maybe it's my own self-defense mechanisms, but oftentimes, whenever any of my patients have a pulmonary process, I sort of interpret that everyone thinks that it's always their cancer that's in their lungs. And I don't know if that's what everybody thinks. It's what I perceive many people think. And sometimes, I always think they must be right. It's always their cancer is the answer. And certainly, neoplasm and recurrence are on the list or progression of disease are on the list of things that can look like an infectious process. Infections can be very broad. They can include primary bacterial pneumonias. They can include septic emboli from vascular sources. Again, some of the patients that I presented to you, it was thought that maybe some of their pulmonary process was a manifestation of that, making a viral processes, fungal processes, microbacterial processes. They can have a variety of inflammatory conditions, including drug induced. A lot of the medicines that we give to treat them for their cancers, as well as other medicines that we give them in supportive care can trigger lung reactions that can look and behave as if they're infectious. And they can have a variety of other inflammatory reactions. And I just list a few, like pulmonary histiocytosis can sometimes masquerade in this setting. We have a variety of diagnostic tests that we can go to. And again, this is not a comprehensive list. This will be left to my esteemed colleagues to go through. But we can do serologic tests-- things like fungitell, fungal immunodiffusion, quantiferon gold-- that look for indirect evidence of pulmonary processes. We can likewise look in urine and things like histoplasma antigen. We can proceed to try to sample the lungs with a lavage via bronchoscopy. And then ultimately, we have at our disposal a variety of possible biopsies, including CT-guided percutaneous biopsies, biopsies that are obtained by different ways through bronchoscopy, and through surgical procedures, like VATS. Some people who know me better, particularly maybe people in the Hem/Onc Division, know that sometimes I unfortunately can get a little nihilistic and pessimistic, although it's very dichotomous, because some people think that I'm the ultimate optimist on the other side. And there was a time when I thought that-- I'm not sure that even I've seen a bronchoscopy affect the management of a patient with febrile neutropenia. This was a few years ago. Emily Hughes, who some of you may know, was a resident here who's now a fellow at Lake Forest in hem/onc. And she was looking for a research project, and we were talking one day. And we talked about my very incorrect opinion on this. And so she launched a little project to look at in patients who have leukemia, acute leukemia and febrile neutropenia, and had bronchoscopies-- what were the outcomes? And again, much to my surprise and much to my education, her conclusion was that management was changed about a third of the time through either positive or negative findings. Either we got a diagnosis through the bronchoscopy or we eliminated certain things that we thought it was through the bronchoscopy. And in a good way, as well, there were very few complications of the BAL. So I say this as a preface to them and also as a pitch to residents out there, who see things within the hospital or in the clinic that strike them as interesting, if you have an idea of a question you want to answer, this is literally how this project started. And I was very much educated by this project. With that, I will turn it over to Dr. Donowitz, who will talk about the ID perspective of this evaluation. So the way this session got put together was that Dr. Davis and I have co-consulted on a number of patients on 8 West. And I can't say that we have agreed on what to do most of the time. We don't. And so when grand rounds was changed and they said, well, [INAUDIBLE] from different types of grand rounds, I said, what about a little debate? We could have a debate. And that's the way we started off. So there have been rumors flying about what sort of debate this was going to be. So there are two rumors that I just have to put down right now. Dr. [INAUDIBLE] and I will not be playing banjo or guitar for this grand rounds. We're not going to do that. And we are not going to have hand wrestling, arm wrestling at the end of this. Dr. Davis and I will not. What we are going to do, though, is have a chat between ID, pulmonary, and hem/onc about ways to approach this sort of very challenging group of patients. So for my part of this chat, there are going to be really three objectives I'd like to go over. Define the nature of the problem of pulmonary infiltrates in the compromised host-- talk about that a little bit, as Dr. Douvas has. Review the invasive procedures that are available to make a diagnosis. And then begin to develop a plan of what to do and when to do it. And I say it's a plan. I can't say that everyone will agree with it. But it is a plan. So we'll go from there. So pulmonary infiltrates in the compromised host-- if anyone has walked through 8 West who have been there for any period of time, you know that about 25% of patients with profound-- that is, less than 300 to 500 neutrophils per millimeter cubed-- prolonged neutropenia greater than 10 days, will develop pulmonary infiltrates. That's sort of been in the literature for a long time. 30% to 60% of people who are getting hematopoietic stem cell transplants of some type will have pulmonary complications of some sort-- of some sort. It doesn't necessarily mean infection. If you think of the etiologies, though, they're myriad. So there's underlying disease just by itself. There's disease-related. And that could be diffuse ovular hemorrhage. It could be reconstitution syndrome, idiopathic pneumonia syndrome. It could be a lot. It could be drug effect. And as there are an increasing amount of monoclonal antibodies and drugs that we use in this group, we're becoming aware that there are drug-related pulmonary issues that throw themselves into the differential. And you have to be aware of them, whether it's a rituximab, gemcitabine, or a host of others. Finally, there is infection-- near and dear to my heart. And the challenge of infection is that it's not one type of infection. It's one of several. So this is from a review. If you have to read one review, this is the one to look at. It's from the pulmonary group at M.D. Anderson. And they just said, here are the infectious etiologies of pulmonary infiltrates in the compromised host. And they were very organized. They put it in groups of gram positive-- sorry. Gram positive-- well, that's sort of scary. Gram positive-- [LAUGHTER] --gram negative, viruses, and fungi-- and the little asterisk by some of these agents are the ones that are associated with drug resistance and difficulty to treat. And you should be struck by the fact that there are a lot of them that could have resistance and are difficult to treat. So there are a lot of possibilities. The other thing to keep in mind is that about 15% of all infections in the compromised host may be polymicrobial. This was work done by Ken Ralston and Jerry Bodhi from M.D. Anderson. So that's not one thing, but maybe several things. OK. So I think cases are always important. This case was from last week when I was on the ID service. 47-year-old female, stage one breast cancer, recently completing four cycles of cytoxan and adriamycin-- on September 16th, she develops headache, fever, productive cough. She's not neutropenic. ANC's over 3,000. Lymphocytes-- she's lymphopenic, and the CT is as shown. So diffuse upper lobe-- sort of interesting-- multiple pulmonary opacities, consolidation, and they were saying, yeah, maybe there's some interstitial edema as well. So the differential should have been, was, community acquired pneumonia. Pneumocystis, or beta-D-glucan, was over 300 picograms per mL. Normal here is 80. Legionella-- could be fungi-- histo, blasto, aspergillus, go back to beta-D-glucan, see why we're thinking about that. Or it could have been the cytoxan. Bronchoscopy occurred after five days of cefepime therapy. I'll come back to the boldface part of that. She was still febrile. She was bronch, had no growth. Because of that, amphotericin was started. And this is her fever curve. And what you'll note is that after amphotericin was started, she became afebrile. She felt like a million bucks, and we looked like heroes. We had no clue what was causing her pneumonia. And at this point, the question for us, the question for pulmonary, the question for hem/onc, is what do we do with her antibiotics? Is this community-acquired pneumonia? I'm going to treat that for eight days. Is this actually a fungal pneumonia that we haven't diagnosed, and she'll be on amphotericin for months? What do we do with this? And that's really the crux of my part of this case. So the clinical challenge of infiltrating the compromised host-- differential is large. We already said that. Infections are only one part of the possibilities. Dr. Douvas has brought it up, that if it's infectious it, could be fungi, it could be bacteria, it could be viruses, or a combination of both. It could be polymicrobial. There is no empiric therapy that covers all the possibilities. And we know from fungal pneumonia and bacterial pneumonia, if you pick the wrong antibiotics upfront in the first 48, 72 hours, your mortality rate goes up. So the challenge is pretty high, pretty great. So what you do noninvasively has been outlined by Dr. Douvas. And I don't want to repeat that. The first thing is one of those fancy tests that you should do, that physical exam test. See what you see. You find physical findings. If they're severely neutropenic, you may not. On the other hand, if you see skin manifestations and they weren't there yesterday, all of a sudden, you're thinking of certain etiologies, like fungi. If you're not finding that, you're saying, OK, I've done the blood cultures. I've gotten another white count. I'm going to do galactomannan, beta-D-glucan. Well, look, most of the time-- not all the time, but most of the time, it's not going to lead you to where you want to be. So then you have to say, if I've done the noninvasive, what do I do next? Which brings us to the invasive procedures. The one that we all agree with-- we're not going to really challenge each other-- is BAL, bronchoscopy with BAL. The problem is-- not the problem-- but the fact is that there are not great studies on how good that procedure is. It's good, but the studies are not wonderful. So I picked a couple just to talk about. This is a prospective study, again, out of M.D. Anderson. They, took people 284 patients, who had malignancy, over half, or solid tumors or miscellaneous. Everybody had platelets over 20,000. And everybody had a PO2 of greater than 90 with less than 5 liters per minute of supplemental oxygen. So they were pretty good. With BAL, if you look to the key malignancies, over on my left, 29.4% yield. This non-heme malignancy, a little higher, at 42.3%-- so not wonderful. This is from New Zealand. And this was sort of a much smaller study, but really represented the patients that we worry about the most. These are the febrile neutropenic patients with pulmonary lesions. BAL, 23%-- now, if you look at the results of this paper, and you said, well, wait a minute. Wait a minute. Candida albicans hands from a bronch is not a microbiologic diagnosis, because Candida pneumonia occurs rarely. And to prove it, you need histologic diagnosis. If you take that away, you're down to 17% yield. If you take all those studies, the ones I thought that were reasonable, you'll see that the yield goes from mid-20s to 41%-- very different patient populations. Some were stem cell transplants. Some were hematologic versus solid malignancies. The range is very high. So it's not a great literature-- very varied. But I thought there were some lessons that we all should take away from this. And this is going to be part of our reasoning. So we talked about the yield. Bronchoscopy can be safely performed with platelets of 20,000 to 40,000. There was no major bleeding problem in a study out of Kansas-- it is Kansas-- where 60% of the people had platelet counts less than 60,000, 16% had less than 16,000 platelets. There were no major bleeding problems. Diagnostic yield can be related to location in the lung, nature of the infiltrate, and symptomatology, which is interesting. This, again, is from the study from Kansas. If you notice, consolidated lesions, round blast lesions, and tree-in-bud lesions gave you much higher diagnostic yields than reticular-nodular patterns, which should guide us. If you look at the lower lobes and left upper lobe-- so bilateral low lobes and left upper-- they give you a much better yield than right upper and right middle. I have no idea why. I'll ask the pulmonologist, because I have no explanation. That's what they see. And this should not surprise anyone. If you have fever and chest symptoms, that's better than having no symptoms for diagnostic yield. But remember, pulmonary symptoms in neutropenic may not exist. They may not have fever, cough or anything else other than, you know, I'm not feeling well. Your physical exam may lead to scant rounds. And it's only on an X-ray or CT that you'll see what you need to see. So keep that in mind. So lesson number two-- there are high yields-- tree-in-bud, consolidation, ground glass-- and there are lower diagnostic yields-- reticular-nodular infiltrates. I've left out biomarkers, which, when these studies were done, were not being used. But as Dr. Douvas mentioned, aspergillus antigen has a varying sensitivity, depending on what your cutoff is-- somewhere between 49% and 90%-- needs standardization, but if you're doing a BAL and you see it, not to be ignored in a neutropenic. I think we'll hear a little bit more about. Lesson three-- be wary of the microbiology. And I think if you just look at some of the results and you say, does that make sense, well, CMV can't be diagnosed just by seeing it in your BAL. You have to see histopathologic evidence of infection. Rotavirus is important probably because of co-infected agents, not rhinovirus itself. And herpes simplex is coming from the oral pharynx, most likely. That shouldn't make a diagnosis. Candida of any type, whether it's albicans, or glabrata, or any of the other strains of Candida, you have to prove histology. Candida pneumonia is rare. That happens. You can prove it. You need some sort of histology showing that there is invasion. Merely getting it on the BAL is not going to help you, not going help the patient. BAL, as far as what do you believe-- you see mycobacteria, believe it. Neumocystis, absolutely; legionella, absolutely-- bacteria, most of the time it's very hopeful. Enterococci rarely, if ever, causes pneumonia. It doesn't. Coagulase-negative staph, I think not. Candida, I just talked about, and CMV I just talked about. So when you look at these studies, look at the microbiology, and just be aware that some of things are isolated and have nothing to do with the disease that you're worried about. OK. Timing of the bronchoscopy-- this is probably the most important lesson, number four. This was a study, again, out of M.D. Anderson. And they looked at stem cell transplants. And they looked at bronchs that were done before-- less than or equal to four days of antibiotics or greater. The yield was 73% if you're doing the bronchoscopy early, 31% if you're doing it later. The yield was 75% within 24 hours of bronchoscopy, 24 hours after the presentation, 40% if you waited five days, 14% if you waited greater than 10 days with a patient on antibiotic therapy. That's pretty important. So that's bronchoscopy. So some of the yields are so-so, and then you have to say, what's left to do? And the choices are transbronchial, CT guided, or open lung-- slash VATS. Sorry As far as transbronchial biopsy, if the literature was not wonderful for bronchoscopy with BAL, it's actually a little less wonderful for transbronchial biopsies. Most Of the studies that you're going to see will have 100 patients getting bronchoscopy, 17 of which had a transbronchial biopsy, and they'll compare the results. If you look at a review that Dr. Davis was nice enough to give me, the range of people who got transbronchial biopsies, as well as BALs, went from 2.9% to 60%. So the selection bias and who are you doing a transbronchial on and how is that meaningful really comes to play. This is probably the single best study looking at transbronchials and BALs. This is out of Verona, Italy. And they did transbronchials and BALs on all the patients they reported on-- so 55% yield with a transbronchial, 20% with a BAL. The transbronchial biopsy added 14 diagnoses to what would have been looking only at the BAL. The BAL added no extra diagnoses to the transbronchial. Side effects were 2.5%, none of which were serious. So if you look at transbronchial biopsy. There are some benefits, and there are some risks. The benefits are, while you may make a diagnosis in 50% to 60%, in some studies, it's 7.6 increase over the BAL, diagnostic yield, and it may be the only source of your diagnoses at 18%, sometimes, versus studies that say, 86% of the time with a transbronchial, you're going to get a non-diagnostic yield. The risk-- bleeding and pneumothoraces. The incidents of those risks are 2.5% to about 31%. 31% was with pneumothoraces, 15% of which needed a small bore chest tube for 24 hours. So nobody died-- just saying. [LAUGHTER] Just saying-- the next procedure would be CT-guided biopsy. And, again, this is a study which was retrospective-- 213 patients, CT-guided So 60% got specific diagnoses. 53% of those had a negative BAL and transbronchial. Most of those-- two thirds were malignancy, one third were infections. So helpful-- a lot of the patients with lymphoma had spread of their disease. People who were neutropenic and leukemic, most of those were infections. So it depends on the patient population you're dealing with. So again, with CT-guided, there were some high yield things and low yield things. High yield-- lesions greater than a sonometer, cavitary lesions, lung masses; lower yield-- lesions less than a somometer, lung nodules and not masses, and just consolidation alone. Complications-- pneumothorax requiring a chest tube, 15%, no deaths. If you look at open lung biopsy and VATS, this is a retrospective review. 63 patients, 67 procedures-- you just have to stop and say, some patients got multiple open lung biopsies. That's a little gutsy. 62% yield-- sort of an uneven distribution of inflammatory lesions, infection, and malignancy across the board. Complication rate-- 13%, major 3%-- prolonged chest tubes, mechanical ventilation, 7%-- one death from the biopsy, fewer complications if your platelet count was greater than 50,000. The importance of lung biopsy, though, has some subtleties that you need to be aware of. There's one study, 31 cases, with CT suggesting aspergillus. Halo sign, nodules-- it looked like aspergillus, didn't mention galactomannan. Lung biopsy proved to aspergillus 53% of the time. The other diagnoses-- one new core, one CMV, one TB, I think two bronchiolitis obliterans. So all that glitters is not gold. It may look like aspergillus, but it may not be. In another study, there were patients with neutropenic and abnormal CT. Most of them had nodules. 24% had the halo sign. Bronch gave you a yield of 12.8%, perc lung biopsy 35%, open lung biopsy 100% yield. So if you think you're dealing with aspergillus and you're not sure, that may be because you can't be sure. And therefore, getting lung tissue is extremely important. So again, the lessons that I hope you take away is that when you do bronchoscopy as a first procedure do is extremely important. Timing is everything. If it's not working for you, and you're not getting the diagnosis, then you have to sort of weigh, what's the radiographic manifestation of the disease you're looking at? What's the nature of the patient's underlying disease? And what would be the logical next invasive procedure to do? And when should you do it? So you would think that that would be the source of debate, and it has been in the past between Dr. Davis and myself. But stay tuned. Dr. Davis is going to have a plan. Thank you. [APPLAUSE] So that was [INAUDIBLE] for the record. So that's a tough act to follow. And I'm a little embarrassed. I thought we were going to talk about CPAP in neutropenic patients. So to that end, know what is a sleep pulmonologist doing up here having this debate? And I've been here about four years. And when I do pulmonary consults, Dr. Donowitz and I have often gotten into this discussion. And I think it's a really good thing to debate, because there's maybe not really one clear right answer. And so we've both gotten to know the literature well enough to have conversations in real time. But I think, hopefully, as you'll see, that as we all put our heads together with the oncologists, we're all trying to do what's best for the patient. So I'm going to show you my version of the data interpretation. And a lot of it's very similar to what Gerry already presented. But before I show that, I wanted to just give you some quick videos of the procedures that we're talking about for those who haven't been in the bronchoscopy suite. So hopefully this works. Good. So this is a bronchoscopy with a BAL, a bronchal alveolar lavage. And the idea is that we get our scope down and we wedge it into the area where the infiltrate is. And we instill about 150s cc's fluid and 350 cc aliquots. And of that 150 cc's, we get a varied amount of back that we suction back into this collection chamber. Last week, John Watson and I got 100 cc's back, which is my record. We also got 16 cc's back on a different patient out of 150. So there's a big variety. And there's a lot of reasons why that is. But when we do a bronch with a BAL-- this is a very safe procedure. But if we leave a lot of fluid in there and your patients are sicker after the procedure, there are reasons why. And sometimes, they translocate their bacteria. We had someone a couple weeks ago who needed to be in the unit following the procedure for 24 hours. So those are some known complications. But it's usually very well tolerated. Here's another diagnostic procedure that we can do. So in this case, we have the forceps out. We're looking-- I believe this is the main carina between where the right and the left lobes break off. And we can advance the scope and grab hold of the tissue and pull it right out. So this looks pretty safe. And in fact, it's very safe. That is not a transbronchial biopsy. That is an endobronchial biopsy. And when I was a first year fellow and I did my first transbronchial biopsy, I thought we were going to do that. [LAUGHTER] And my attending said turn on the fluoro, and I said, for what? So I think it's important to know what we're being asked to do. And here's what a transbronchial biopsy is. So this is where we go into the area that we're concerned about. And we wedge our scope as tightly as we can. And we advance the forceps beyond the sight of view, as close to pleura as possible, which is explaining the pneumothorax, right, which i can't believe is not higher. And we take a bite. And in order to help guide us, this is an example-- these are from the ATS-- of us navigating down into the lower lobes here, wedging the scope. And you'll continue to see the field of view down here in the bottom right hand corner. This is what we're looking at. And we turn on the fluoro and advance the forceps down here, fairly close to the diaphragm. And we try to go to an area where we know there's an infiltrate. And sometimes, we can see that on fluoroscopy. But oftentimes, we can't, even if it's viewable on chest X-ray and CT. We open up the forceps, and we advance until we almost hit resistance. And then we pull it out. And we're done with that. And generally, people do between six and 12 [INAUDIBLE] depends on the indication. And this is a good way to-- certainly, a lot of lung transplant patients get this procedure to evaluate for rejection. So that's what a transbronchial biopsy is. There's a number of other procedures, including endobronchial ultrasound or navigational CT-guided biopsy. Those are things I'm not going to talk about today. But I do think certainly CT-guided biopsing through bronchoscopy will play a role in this discussion in the future. So this is my most important slide. And this is what we see when things don't go well. And this is looking down the bronchoscope at blood. And the risk is rather low. But this is partly why, largely why, we're concerned. And when we see this, we try not to panic. And we're really not allowed to move out of the area, because part of our goal is to contain the bleeding. And so our ability to mitigate this or assess other parts of the lung are impaired, because we can't move. And cold saline is one of our best strategies to deal with this, but can be a concerning treatment modality. And usually, you just wait it out. And so I often ask myself this question. Perhaps I'm just overly concerned and I should stick with CPAP. This is one of my favorite CPAP masks, by the way. Nobody really likes it but me. But I think it's really effective for claustrophobia, actually. But you know, I ask myself this question, and I try to keep that in mind when-- I am probably one of the more conservative pulmonologists. So it's just something to keep in mind as we think about the risk-benefit ratio. But I'm not the only one. So Paul Suratt actually did one of the original surveys looking at deaths and complications associated with fiberoptic bronchoscopy. And he ran the bronchoscopy suite, too, full disclosure, at the time. And so he sent out a number of questionnaires and got some responses, as you'll see, and reported upon 12 deaths, 27 life threatening cardiovascular complications, and 52 life threatening airway complications-- and really making the case early on in bronchoscopy that this is a safe procedure, which it is. And in general, the complication rates for transbronchial biopsies are really rather low. This is the number that we quote patients when they come in-- all comers, about 2.8% risk of hemorrhage-- so really quite low-- an overall complication rate of 6.8%. And this is based on a study that was published in "Chest" a while ago. And all of these patients had platelet counts of more than 60,000 in normal coagulation parameters. And pneumothorax, although sounds like it's more concerning, it's easier to put it a test tube in than to deal with massive pulmonary hemorrhage. So this is the concern we have, is about 2.8%. In terms of the high risk populations that we're, at least in part, talking about today. One of the early studies looking at risks was published in 1985 here in "Chest." And this included 24 patients with a platelet kind of less than 60,000. And these are patients that all had, as you can see, AML, aplastic anemia, and other heme malignancies. And this is about a 20% significant bleeding complication rate. One of those was a neck hematoma. So I'm not exactly sure how that happened. [LAUGHTER] But you know, there's some concerning amounts of blood. And one of these patients died. So in some regard, the patients we're talking about today really don't meet the overall complication rate that we cite to all comers. And we are cognizant that there are some risks with not doing that. So I really don't want it either leaves saying that we're always going to say no to transbronchial biopsies. And I really just want to highlight that there is a need to pause and think, what will we do with that information? And what are the chances that we'll change our management? And I want to highlight just a couple of papers here, and then allow some time for some questions. But one of the early studies really did point to the use of transbronchial biopsies in febrile neutropenics. And this was at a time where they surveyed-- there were 43 patients at the Brigham, 14 of whom had transbronchial biopsies. And nine of those led to a diagnosis. And this was before BAL was invented, I guess. So just highlighting here that this was of value, with about a 64% positive diagnostic yield-- so pretty high yield. Again, another paper that can show that there's benefit here. This is a prospective study, 2004. And this was non-HIV immunocompromised patients. And as Dr. Donowitz pointed out, there's a lot of issues with these studies, because their sampling was really determined by the operator. So to highlight that, 95% of the patients underwent a BAL, whereas 43% underwent a transbronchial biopsy. And what they demonstrate is that your diagnostic yield at this procedure is 38% with the BAL, and up to 70% when you combine BAL with transbronchial biopsies. This table is, I think, important to review, because when they looked at what diagnosis was made exclusively by transbronchial biopsy, it was predominantly non-infectious causes. So radiation or chemotherapy induced pneumonitis, Wagener's, diffuse alveolar damage, other inflammatory conditions. And so transbronchial biopsies certainly has been demonstrated here, and more recently, in a paper that Andrew Mihalik sent me looking at stem cell transplant recipients that, in a noninfectious work up, there's value in considering transbronchial biopsy, because it certainly adds a lot. Complication rates, you can see here, and consistent with what I was saying. Now, here's a paper that I like to cite when I'm trying to make my argument. And this was done in the Mayo Clinic Proceedings. And they looked, again, retrospectively, looking at bronchoscopies that included BALs, and in some cases, transbronchial biopsies. And they demonstrated that the diagnostic yield of looking at sputum and BAL was around 60%, although I often do quote around 20% to 60%, as Gerry had indicated, based on number of trials, and that the additional use of transbronchial biopsy-- of the nine patients who underwent that procedure, one of them added value in terms of the diagnostic yield. And so it's about a-- they say between 2% and 4% increased diagnostic yield. And so I often try to ask, do we want to do nine transbronchial biopsies to help one patient and expose those other eight to a pretty substantial increased risk of bleeding and other complications? And so that's the way our thought process often is. It had concluded that if you think about the use of transbronchs, you can increase the complication rate 2% to 10%, and that it only changed the management in one patient. So as we think about the noninfectious work up, tissue is important. And as Gerry pointed out, fungal organisms are better to detect when you have tissue. And the use of BAL galactomannan, I think, kind of makes us question a lot of the data that we're looking at, because this test has really been added in the last 10 years or so. And a lot of the papers that we both have been citing even preceded that. But BAL galactomannan is something that we should and frankly do send on all the patients that undergo a BAL. And we know that the BAL galactomannan is often less affected by anti-fungal treatment, neutrophil count, more so than blood samples. And you can see here that the sensitivities vary, depending on which study we want to cite. But in general, it's two to three times greater in BAL than it is in blood samples. And looking at a recent meta analysis-- I guess that was a few years ago-- depending on what cutoff we use for positivity of the BAL galactomannan or the aspergill antigen, you can see that the sensitivity is very high. The specificity also remained quite high. And if I plotted the receiver operating curve, it looks like a good test. So it's something that I think we can consider as increasing our diagnostic yield, compared to a number of studies that we've been citing. So I don't know how much clarity we've provided. And I think this is why it's a good topic to debate. And one thing I wanted to do to prepare was to just survey some faculty here, and some fellows here, and also some people that Meg and I train with who are around the country. And in general, we are all very supportive of empiric anti-microbial therapy, and a bronchoscopy with BAL as first line interventions when we're consulted. And to highlight some of the data about four days or less, earlier BAL is better. This is certainly one of the procedures that we will be happy to do over the weekend if we're [INAUDIBLE] on a Friday night. This is a very good reason to call us so we can do it on Saturday instead of coming in, waiting till Monday. And few of us answer that they would consider transbronchial biopsy during the initial bronch, just based on my very informal survey. And we tend to try to deflect towards CT-guided biopsy as a preferred next way to get tissue if the initial bronch is negative and we still have some questions or concerns. So we do not have guidelines that are published by the American Society of Hematology/Oncology that I could find. The German Society of Hematology and Medical Oncology does have published guidelines. And really, they outline how we tend to practice. You all see this way before we see it as consultants. But when you're on the oncology service, fever, pulmonary symptoms, X-ray and CT-- you can sometimes get MRI of the chest apparently, . And you can do a bronchoscopy with a BAL early on-- that's what we would recommend-- and start some preventive therapy. You target it at 20% to 60% at a time that the BAL helps you. And if you don't have a targeted option, then this is the question that we're here to debate. And how do we proceed? And we want to propose the UVA version of this in a moment as we think through it. They do specifically warn about the risk-benefit ratio of transbronchial biopsies. But this is not based on particularly strong data. I don't want to really focus on this, because I want to open it up for discussion. But part of the reason that we often recommend consideration of CT-guided biopsy is that the yields are generally very high. They can get more of a tissue sample. Instead of using fluoroscopy, they're using CT in real time. And the pneumothorax rate is rather high, as you can imagine. But you could argue that the risk-benefit ratio is a little bit higher. And this is what a CT-guided biopsy looks like, as an example. So do we have a recommended pathway here at UVA? I don't know if you all want to come up for this portion, or if I can present this, and then we can all get up here together and take some questions for the last 10 minutes. But before I show this, I'd say that the three of us really have had a lot of fun putting this talk together over the last few weeks and having this discussion over the last several years. And I think we did the best we could to try to review things for you all to give you some guidance. And our plan, which is open for discussion, is really getting a bronchoscopy with BAL potentially early in the course adds tremendous value in the diagnostic field. And so it's generally pretty achievable to get patients to have platelet counts greater than 20,000. And if that BAL helps us, then we all feel good about it. We're done. If it's negative, and they're not getting better, or there's lack of clarity, then that's the point where tissue really is important. And the, really, decision tree comes down to do we want to pursue a transbronchial biopsy, a CT-guided biopsy, or a VATS? And to try to give you clear guidance based on imaging or patient factors-- we really can't do that. It depends on so many things that are really patient specific, regarding what the lesion looks like on CT, how the patient looks, what the platelet counts are, what meds they're on. But that's where, I think, there's tremendous value in our three teams coming together to have these discussions at the bedside with you all and trying to navigate this. So I'll leave this up. And I hope this was a good and helpful. And we'll take some questions as a group. Thank you. [APPLAUSE] Thanks. [INAUDIBLE] a question? [INAUDIBLE] So a question for the audience, of let's do a show of hands of people who [AUDIO OUT] [INAUDIBLE] [AUDIO OUT] [LAUGHTER] Just to highlight, I think that Eric's point, I think, about [AUDIO OUT] what to do with the [INAUDIBLE] value-added by BAL [INAUDIBLE]-- [AUDIO OUT] really don't know [INAUDIBLE] as we're collecting the data now what [INAUDIBLE] [AUDIO OUT] and for diagnostical [AUDIO OUT]. In addition, where [AUDIO OUT] Also, sort of [INAUDIBLE]-- [AUDIO OUT] I think that's right. And as people live longer, as we do more to prevent the infections that we're worried about, there's no question that we're going to start to see more and different infections that we hadn't thought about before. And then the question you're raising is, so how good is all this data for those infections? Yeah. There's no question. And on that point, what you saw in the previous slide that's showing that algorithm [INAUDIBLE] [AUDIO OUT] --question that we see all the time with empiric therapy. As the moderator of the discussion, I actually would like to chime in at the end, because, again, I thought it was a really good discussion from both of you. Those three patients, again, I think are very illustrative in terms of what happens to them, in terms of having the answers. The first one had both serologic test and a transbronchial biopsy that showed she had histoplasmosis. And she was treated with posaconazole and the proof is in the pudding, I guess, because she got better and is well, both radiographically and clinically. The second patient developed some symptoms after not having them at first. And one thing I didn't tell you-- that even though she was fairly young, she actually was a heavy smoker, and ultimately, went a VATS procedure, and had some of those lesions biopsied. And she had pulmonary histiocytosis-- neither an infection nor a recurrence of her lymphoma. And she actually, again, was advised to quit smoking, and actually, has worked on that, and is still doing OK. The third patient I alluded to is more of an active patient. He had a bronchoscopy with a BAL, which yielded a positive aspergillus antigen on two samples. So it was felt that he likely had aspergillus. Despite that, he actually had a CT-guided biopsy, because again, even in my own mind, I actually did doubt that diagnosis. He was treated. The CT-guided biopsy did result in a pneumothorax, which he still has and wasn't, I say, clinically all that significant. It did not yield anything, meaning that biopsy was pretty worthless. It didn't show aspergillus. It didn't show his lymphoma, which is actually what, if you asked me well I believe, that that's actually what's in his lungs. But I have absolutely no proof of that. And despite treatment for his aspergillus, he continued to worsen and had many subsequent tests, and you to this day, still is a mystery. So again, in terms of that, again, I actually feel I learned a lot about BAL, all the different types of biopsies. But I actually feel that we probably under-utilize VATS in a lot of circumstances, because I think it would be very helpful. Of course, we have a group of doctors who I don't think are out in the room, which is the CT surgeons, and their view of how to use that test [INAUDIBLE]. You're asking whether doing a CT should be done earlier than you're doing them now? Yeah. Yeah, because now [INAUDIBLE]. Right. But if the chest X-ray is not perfect-- that is, well, there may be something-- then actually, a chest CT is extremely helpful. And I want to say-- I remember that paper-- I want to say 60%, roughly-- it's high in picking up abnormalities. If the chest X-ray is absolutely perfect, then that rate of picking something up by CT went down. The tough part of that study was it wasn't really clear with the abnormalities were that they were picking up on CT and whether they were meaningful or not. The patient has respiratory symptoms. And they may not, if they're neutropenic. Chest X-ray, and unless it's absolutely perfect, yeah, CT would be the next thing. But that's four hours later. That's not 48 hours later. Again, tough questions, because again, some of them will be back to [INAUDIBLE]. And so I think one of the arguments are at least give culture a little time to see if the-- [INTERPOSING VOICES] And sometimes, they're related. Sometimes, they're not. Well, if I could comment, too, I think the first thing we look at after you call us is the CT. And I try to encourage us to go to the bedside and do an exam. But we're guilty of that. And so it is helpful to get that information. And to what extent getting it sooner may help us do that bronchoscopy within four days versus be unnecessary because the blood cultures come back in a day or two-- it's a tough balance that you all manage. But you don't have to wait through the weekend to call us or to get that CT. We can get moving on the sooner side after that's determined necessary. All right. Thank you all. Thank you very much. Thank you. Good questions. [APPLAUSE] Perfect timing. The timing was great.