IBCC_EP_105_-_Ventilator_Associated_Pneumonia

Josh and Adam discuss ventilator-associated pneumonia (VAP) on the Interactive Workplace Critical Care Podcast. They highlight the challenges of diagnosing VAP due to the lack of a gold standard test, leading to issues of over-treating or under-treating patients. They touch on the importance of distinguishing VAP from other infections like hospital-acquired pneumonia. The conversation delves into possible, probable, and confirmed VAP diagnoses, with steps to assess inflammation, pulmonary dysfunction, and radiographic evidence. They also mention the differential diagnosis for VAP, including factors like leukocytosis and neutrophil to lymphocyte ratio. Additionally, they discuss the nuances between VAP, ventilator-associated tracheobronchitis, colonization, and infection. Welcome back to the Interactive Workplace Critical Care Podcast. I'm here with Adam Thomas and we're going to talk about ventilator-associated pneumonia or VAP. How much frustration has this given you Josh? Like how much minutes of the day do you think when you're in the unit do you have to commit to this? This is infinitely frustrating. Today we'll cover a stepwise approach as you've already posted, multiple steps and let's just say it all comes down to maybe, maybe not, but Josh has at least tried to form together. But Josh, let's start it off with the riddle of VAP as you call it. This is not a straightforward answer and there's some problems with either over-treating, under-treating, over-investigating, under-investigating or just flipping a coin. So map this nuance out for me. I think the real root of the problem here is that we do not have a gold standard test to determine whether or not a patient has ventilator-associated pneumonia. So if you wanted to say with 100% certainty that a patient has ventilator-associated pneumonia, you would probably need to do like a surgical lung biopsy or an autopsy on them. There's no other test that's 100% sensitive and specific and this leads to these enormous problems of circular logic. So depending on how you define VAP, you can end up defining lots of people with VAP, you can end up defining fewer people with VAP and I'm just always questioning if we're over-treating or under-treating patients and I just, I can't get to the bottom of this. Do you have some solution to this? Give them all peptazole, Josh, it's pretty easy. If they live, you're good. Trials, peptazole, got it, got it. But Twitterster has already been erupting about this. I think you have appropriately answered some things. I think this is the time to unpack this nuance that VAP is difficult. My question for you, do you need an endotracheal tube to have the VAP? What are the considerations here? Yeah, so ventilator-associated pneumonia is defined as pneumonia starting after, you know, over 48 hours of intubation. So you need to be intubated to have a VAP. Theoretically, if you extubate someone and they develop the pneumonia shortly after extubation, they could have an extubated ventilator-associated pneumonia. But I think, practically speaking, most of the patients we're talking about here need to be intubated. And you'll notice that I kind of went out and I didn't really discuss hospital-acquired pneumonia or HAP here, which I think is even more painful. You're kicking it 2016 style if I like it. I guess for me though, why is it important because we already covered community-acquired pneumonia. When I take someone into the ICU, you address this with other infections, we change the flora. So is that why it's really important to figure out hospital-acquired, ventilator-associated? Does that actually change what bugs I'm treating? Yeah, it does. So over time for patients who are in the ICU, they basically will acquire a hospital for a vector. And then there's another shift in the underlying pathophysiology of disease here, which I think is actually helpful for us, which is that typically, when we think about patients coming into the hospital with pneumonia, lots of them have a pneumococcal pneumonia, which is more of a low bar process and it's centered on the alveoli. And that may not always generate a lot of sputum because the infection is really like packed down deep in the lung tissue. Most patients with ventilator-associated pneumonia seem to have more of a broncho-pneumonia, which is more focused in the bronchi and they tend to often have more sputum production. That may make it easier for us to culture them. Now, I can hear the tweets just erupting, what about your tracheobronchitis, Josh? That's a lot of secretions. Yeah. So what about that? I don't know the answer to this. So there is this entity in the universe which is called like ventilator-associated tracheobronchitis or VAT, V-A-T. And presumably, patients can have sputum production and they can kind of have like an infection of their trachea with no radiographic infiltrate in the lungs. Now, the IDSA and other sources recommend not to treat this. And there's not a lot of literature about it. So A, I'm not 100% sure that this is like totally a thing and B, if it is a thing, it's not something that we need to treat. So I'm not 1,000% sure what the difference is between ventilator-associated tracheobronchitis and simply kind of like being colonized with bronchitis. I suppose a patient with VAT could have like a fever and more sputum, but I think there's a little bit of gray area just between ventilator-associated tracheobronchitis, so it is just like colonized with bacteria. I think that's huge to say because let's think about the community-acquired tracheobronchitis. Those patients are super sick, right? They look horrible. Whereas maybe someone is now 10 days into their ventween, they're coming off, they look really good and they have some secretions growing, I don't know, strep, staph, whatever, similar to a trache patient, we might not necessarily treat the colonized patient, will we? Exactly. Exactly. And that's really the crux of this whole topic which is sorting out colonization versus infection because it's very well-known that any patient who's intubated for a duration of time will develop colonization with oftentimes scary bacteria like Pseudomonas and MRSA. So anytime if you just go around and indiscriminately culture patients on a ventilator, you're going to grow really scary things. So the major challenge is really when do we need to culture patients and who actually needs to be treated. So that's a great segue. We're going to walk through the seven steps that you've identified, but basically you can clump those into possible VAP, probable VAP, and then confirmed VAP. So walk me through those three categories before we go into step number one. I think possible VAP is kind of like a pre-contemplative phase where you're starting to consider whether or not you think a patient has ventilator-associated pneumonia. I like how you linked VAP to a use disorder. No, I mean, you have to start thinking about it and then once you kind of make some sort of clinical decision that you think the patient probably has ventilator-associated pneumonia, then that puts you into a probable VAP box. It's really important to realize that patients with probable VAP, only about 40% or 50% of them are truly going to have microbiologically confirmed VAP. So this is truly probable and then those patients are typically going to get antibiotics and cultures and depending on how all of that dust settles and all of the new culture data, they may either, the diagnosis could be refuted or they could progress to a so-called confirmed diagnosis of VAP. Although I should note that without really a tissue diagnosis, this isn't 100%, but we're going to say for practical purposes, it's confirmed. Now, just because the figure is really great, I want people to follow along, but can you walk me through the differential for VAP? There's lots of things in the ICU, especially our more complicated patients who may be having surgical procedures and maybe spending lots of time in the ICU, perhaps they have pre-existing ARDS and so there's lots of things that can cause fever. There's lots of things that can cause a respiratory dysfunction. For example, pulmonary embolism, aspiration pneumonitis, medication-induced pneumonitis, cryptogenic organizing pneumonia. You can only throw a cop in any differential diagnosis, by the way. If you want to sound smart and if you're at a pulmonary conference, a cop is pretty much always a good one. Empyema, fibro-proliferative ARDS and there are just like so many different things. Okay, so step one, we've alluded to this, you're going to consider all your rapidly available information and this is when you're trying to say, is there a possible VAP or not? You're assessing your patient, dear God, maybe someone's making you do pre-rounds. You're going to look for four major things, evidence of inflammation, evidence of pulmonary dysfunction, radiographic evidence and then as Josh just alluded to, the alternative diagnostic. So Josh, what are those evidence of inflammation that I've seen in other instances? Yeah, so here we're talking about some sort of leukocytosis and the nice thing about ventilator-associated pneumonia is these are by definition patients who have been sitting in the ICU for a couple of days. So you will have a baseline, you know, white blood cell count, you'll have a baseline temperature on them. You know, sometimes patients may become hypoglycemic when they're infected. So if you have a patient in the ICU and suddenly the white cell count starts trending up and their fever curve is trending up and they're developing a left shift and their glycemic control is going sideways, these are all things that can suggest inflammation. Now Josh, you snuck it in there but you addressed this in a nice post a few months ago, the neutrophil to lymphocyte ratio. You want to pay homage to that quickly? Oh, this makes me a little sad because I couldn't find any direct evidence on neutrophil to lymphocyte ratio with regards to VAP but I do believe that this is one thing that would probably track in terms of, you know, infection and inflammation especially if you're checking every single day. Nice. Mendevia, left shift, fever, hypothermia, septic shock, they're all on that evidence of inflammation. Now, pulmonary dysfunction, I've got three plus mucopurulent secretions on my morning report. Is that the only thing I need? Typically, the things I think about here are secretions and oxygenation and, you know, I should note that all of these features that we're discussing have really bad performance. So for example, purulent secretions, sensitivity of 77%, specificity of about 40% giving you a positive likelihood ratio of 1.3, negative likelihood ratio of 0.6. So these are all very weak features but I think here with regards to pulmonary dysfunction about secretions and also respiratory dysfunction, so if the patient is doing well in the ventilator and if they're about to wean off and extubate, then probably not really having a severe VAP whereas if everything is getting worse on the ventilator, I think about it more seriously. Now, we're going to wade into the waters of trouble here, radiographic evidence. So it's 1992 and the order set says chest x-ray while intubated every day. This is a bit upsetting, isn't it Josh? So let's talk about the fact that even if I get a little strep pneumonia or low-bar pneumonia right now, it takes a long time for that x-ray to improve, doesn't it? Exactly. And that's one thing that can be helpful and we'll talk about that a bit later is the x-ray changes over time. For example, there are lots of patients in the ICU, they'll get a low-bar infiltrate and it's unclear whether or not it's pneumonia or atelectasis. So if you check another x-ray a day or two later, if it's still there, if it's getting worse, that would support pneumonia diagnosis whereas like atelectasis may improve very rapidly over one or two days. Now, within out the lung ultrasound, we don't really need to get into it because we've talked about it at length but consolidation air bronchograms, perineumonic effusions, these are all helpful, aren't they? These are helpful but lung ultrasonography is probably less useful for ventilator-associated pneumonia compared to community-acquired pneumonia because patients in the ICU very frequently have basilar consolidation simply due to atelectasis and they have dependent B-lines just because they've been on their back for so long. So the performance of ultrasounds here is just not quite as good as it is for community-acquired pneumonia. So I love ultrasounds but I think chest x-ray is probably going to be a bit more helpful here. And as you joked in the community-acquired CT scans for everybody, right? This is another thing I struggle with which is CT scans probably has 100% sensitivity. So if you do a CT scan, there's no infiltrate. You can be very confident the patient does not have a VAP as opposed to chest x-ray which is sensitive about 89% of the time. So it's pretty sensitive but not perfect and CT scan can be useful to try to sort out atelectasis versus consolidation and this is one of these challenges of like how many of these patients do you get a CT scan and I don't know the answer. I think you brought it up because of the logistics challenge. I think my nurses would absolutely kill me if I CT'd everyone just for VAP or not, right? 100%, you'd be assassinated. Okay, we already covered the alternative diagnosis so we're going to move on to step number two. When do I actually pull the trigger on cultures and antibiotics? Yeah, I have no solid answer to this. So there used to be this clinical pulmonary infection score or CPIS which was really hot for a little while but it's largely been disproven and it's no longer recommended. So I think the bottom line here is this is a matter of clinical judgment based on considering all those bits of information that we just talked about plus alternative diagnoses and that's all I got. I don't know. Do you have any like magic wands or like a … No, I think the big thing is it's hard to prescribe an algorithm because it really depends on your patient. If they're doing really well, you have time, maybe a day or two to actually ascertain whether it's a VAP. If they're dying of septic shock, you're just going to cover for it anyway. So it really depends on how the patient's doing, doesn't it? Yeah, 100%. Illness severity is super important. Let's move on to step three then. So in this first part, how do we manage the suspected VAP? So we've talked about the utility of cultures, difficult. We've talked about looking for x-rays or radiographic or signs of inflammation. So what are you going to order here, Josh? If you make a clinical decision that you think the patient has a suspected VAP, then this leads to this bundle of diagnostic tests that need to be done and some of these tests may be indicated anyway for a patient who's febrile and getting worse in the ICU. So this is going to include blood cultures, tracheal aspirate for gram-stained culture, PCR for stuff like influenza, if that's on the table, although practically speaking, most of these patients may have had that testing already. Nasal PCR for MRSA may be super helpful if it hasn't been obtained. I'm going to sneak procalcitonin in here and hope nobody's listening. Repeat chest x-ray after a couple of days. We kind of discussed that briefly. And then in certain situations for patients who are immunocompromised, you might want to do some additional tests or patients who are at risk for invasive aspergillus. We had a whole podcast on that. Space repetition, that's really interesting. So post-ARDS, whatever it is, influenza, COVID, whatever, there's a not insignificant amount with invasive aspergillus, so keep it on the list. All right, Josh. So in summary, we're going to send a whack load of things. I just want to know your opinion. Are you getting these multi-panels for NATPCR, so not just influenza or COVID, but a bunch of fungi, bacteria, viruses, are you using that at all? So I don't have access. Are you referring to like a biofire? Biofire, yeah. Why don't you talk about that a little bit? You have more experience with it. I just do it because my little lab medical microbiologists are great, but basically it's a panel of all the common offenders of viral bacterial pathogens that you can just send off your tracheal aspirate or your nasal pharyngeal, and it'll come back rapid, like within an hour or so. It's nice, especially in these days, to screen of what's actually going on in your respiratory failure patients, but I don't think it actually changes the issue that it could just vary while decolonization. It's just interesting to pick up a new bug, like human metamuma virus that might have been acquired in the unit, and that's why they're getting worse. Yeah. Very interesting. Yeah. I don't have the biofire in my shop. Sometimes we do perform these multiplex PCRs for a variety of viral pathogens, but the interpretation is always really confusing, because as you mentioned, isolating an organism doesn't prove that it's causing infection, and isolating a viral pathogen doesn't prove that there's not also a bacterial pathogen kind of working underneath it. So all these data really need to be interpreted pretty carefully. And I shouldn't complain, because when it takes two days to get a COVID swab back, we shouldn't be talking about biofires. Okay. Let's talk about the management, the second part of step three. We suspected the VAP, we pulled the trigger. Where am I actually going to give them for our antibiotic coverage here? Yeah. This is very challenging as well. So occasionally, for patients with super early onset ventilator-associated pneumonia, if they're like within 5Ds of admission to the hospital, and if they have this panel of characteristics that reduces their likelihood of drug-resistant organisms, then it's possible that the patient was actually kind of like harboring a community-acquired pneumonia that didn't quite really manifest itself until a couple of days into the ICU course. So rarely, it may be possible to treat these patients with a community-acquired pneumonia regimen. That being said, that's pretty uncommon. Typically, for patients who have been in the ICU for longer than five days, we're going to need to get some more broad-spectrum antibiotics. The backbone agent is going to be anti-pseudomonial beta-lactam, typically your cefepime or your piperacillin-tazobactam, because pseudomonas is definitely a player here. Depending on your unit, if you're in an area with really high drug resistance, you could consider meropenem, but I feel like going straight to meropenem would be a little bit of an acceleration. What's your anti-pseudomonial beta-lactam of choice? Depends on what you're on, and if it's lower mainland, especially the east side of the lower mainland, there is a lot of ESBLs and increasing CROs. So not uncommon to go straight to Nero if they're non-colonized with an ESBL or high-risk. But for sure, Piptozo, backbone for us. Very good. The next question is whether or not to cover for MRSA. This is a challenging issue. My philosophy on this is as follows. Among these patients with probable VAP, only about 40% of them are actually going to have a true VAP, maybe even a little bit less than that, depending on how we're defining VAP and what gold standard we're basing this on. Of those 40%, maybe 10% or 15% or 20% may have MRSA, depending on your location. So the likelihood that the patient has a MRSA infection is low, probably about 5%. So I don't think you need to treat for MRSA in every single patient with VAP. I think you can think about this on a patient-by-patient basis, considering various risk factors. And we've listed some risk factors here, including prior IV antibiotics, hospitalization in units with lots of MRSA, patients who are super sick, as you mentioned earlier, for patients with septic shocks, and definitely cover them for MRSA, certainly if they're known colonizers with MRSA, stuff like that. So check out the post for more nuances there. But if they're looking really good and you barely think they have a VAP anyways, reconsider MRSA. So the hallmark here is antisexual and monous coverage in your beta-lactam backbone, plus or minus MRSA. What about atypical coverage? You alluded to some of these might be incompletely treated, can be acquired infections, or more importantly, it might just be an atypical coverage that they acquired. So what's going on here? So occasionally, hospitals will get their water systems infected with Legionella. And it's kind of gross, but I've actually seen this happen at two different hospitals that I've worked at. And they were very good hospitals. Have you ever experienced this? Yes. I'm like, yes. We'll just say yes. Yeah. So if your hospital is having a Legionella issue, then I think you would want to add coverage for that and test for it with azithromycin, for example. But it's typically, hopefully, not going to be happening most of the time. Just to backtrack quickly, one thing that may help with regards to MRSA coverage is a NARES PCR. The evidence for this isn't bulletproof, but if your NARES PCR is negative, that drops your likelihood of MRSA reasonably substantially and potentially could argue against covering for MRSA if you're kind of leaning against it. Nice. Okay. And of course, you do have access to it. Yeah. Yeah. Last up before we move on, I don't think we have to spend a lot of time on this. We talked at length. If the lung is fully aerated, it's very, very unlikely that they will have an anaerobe in there. But if there's an empyema or a lung abscess, totally different. So anaerobic coverage, if you're giving Flagyl to a VAP, you think twice about it. Okay. Step four, data review and diagnostic adjudication. So was I right? Was I wrong? How do I figure this out, Josh? Yeah. So within 24 to 48 hours of pulling the trigger on empiric therapy and getting this flotilla of diagnostic studies, data is going to start coming back. And I think as that data comes back, you need to be continually recalibrating your diagnostic likelihood that the patient actually has ventilator-associated pneumonia and reconsidering your empiric therapies. For example, tracheal aspirate Gram-C culture can be super helpful. For example, if the Gram-C shows no leukocytes, this strongly argues against a ventilator-associated pneumonia. On the flip side, if the Gram-C shows a lot of leukocytes and no squamous cells and a single dominant bacterial morphology, like it's just covered in Gram-positive cocci in clusters, for example, then that could push you toward a specific etiology. Do you ever see this? So I feel like every single Gram-C I get is like GPCs and GMBs and like little this, little that. And then reinterpreted by the microbiologist as normal for after. Yeah, exactly. So it's hard. I think it's the one piece as you keep coming back to that it really is a multifactorial Bayesian analysis almost, right? Exactly. Exactly. One useful bit though is that if the culture is totally negative, so if that culture does come back as normal flora or just like no bacteria, then that really does argue substantially against the ventilator-associated pneumonia. So I think in a lot of cases, that could push you toward potentially stopping therapy or at the very least kind of narrowing therapy and not covering Pseudomonas and MRSA, stuff like that. So they're going down for their CT scan and it's negative, right? Exactly. Okay. We also threw some ProCal and some blood cultures in there. We've addressed that. So let's move on. Let's talk about step five, narrowing our antibiotic regime. So based on our microbiologic studies, say MRSA is negative or positive, we're going to go for that. Say we actually have a detected organism and check out our antibiotic post for that. What am I going to do though if there's no organism detected? How do I start peeling back or focusing coverage? Yeah. So if there's no organism detected, I think you really want to think twice about whether or not the patient truly has ventilator-associated pneumonia. Now one caveat here is that if antibiotics are started before you get your cultures, then the cultures could be falsely negative. So that may not be super helpful. But assuming you were a good doobie and you got your cultures before your antibiotics, then a negative culture really strongly argues against the diagnosis of that. I think in a lot of situations, you might want to consider discontinuing antibiotics. Now on the flip side, if you really deeply believe in your heart of hearts that the patient has a VAP and the culture is negative, I think it could be reasonable to continue antibiotics. But you could probably narrow it down substantially and probably lose your MRSA coverage and your Pseudomonas coverage and just put them on something like Ceftriaxone. And then when antibiotic stewardship comes by, just listen to them and do whatever they want, right? One last thing is MRSA coverage. So I quickly went over it. I said, MRSA, if it's negative, should I discontinue it? MRSA pneumonia typically produces a lot of sputum. It's typically relatively easy to culture. Oftentimes, we can get positive PCR signals on these patients and it doesn't sterilize easily. So one of my preceptors and fellowship pieces say, MRSA isn't shy. So if you go for over 48 hours, go by and you're not culturing MRSA and you're not PCR-ing MRSA and you're not seeing a ton of MRSA on your gram stain, I think you should get rid of your MRSA coverage. And I think one mistake which is sometimes made is people start MRSA coverage like they start vancomycin and then they just continue it for seven days. And I really think that unless there's some evidence documented somewhere that there's MRSA, you should really get rid of it within 48 or at max 72 hours. Agreed. Pull it back to doing less levels, less problems for sure. Okay, duration of therapy, that's a great caveat. So we've stopped our MRSA coverage. How long do I actually, when I prescribe these antibiotics, just because it's missed a lot, should I always put in automatic stop dates or how long will you extend these for? So over time, it seems like our duration of antibiotic therapy for every disease is slowly decreasing. So in the evolution of VAP therapy, we're now down to seven days of antibiotics, even for Pseudomonas. Yay! Yeah, this is awesome. It's based on IDSA, ATS, and ERS guidelines, so all the guidelines seem to agree on this. Except the caveats are what though, Josh? Yeah, there's some rare caveats here for empyema, for lung abscess, for necrotizing pneumonia, for severe immunodeficiency, cystic fibrosis. You may want to extend therapy, but the vast majority of patients, after seven days, you can call it good. Now, this is where Procalcitonin comes in with actual evidence, right? That is hard to argue with. So let's say the duration, as in the Procal's coming down and now negative, is it good for stopping antibiotics now? Yes. So if the Procal goes less than 0.25 nanograms per ml, or if it falls below 20% of its initial value before seven days are up, then you can stop antibiotics before seven days. I want to stress here that our use of the Procal is an antibiotic stopping tool. So we're not using it to make the decision to start antibiotics, we're using it to curtail the duration of antibiotics. And honestly, I find this to be pretty helpful because I think that there are definitely situations where a patient receives a clinical diagnosis of VAP and three days go by and they're afebrile and they're not really behaving like they actually have a VAP. And checking Procalcitonin can kind of help you feel comfortable about the fact that you should stop antibiotics. Josh, quick note about step seven, the last one, treatment failure. So they're not responding like they should. We talked about it in community-acquired pneumonia. Do we increase antibiotics, do we bust out the bronch? What's your approach to this? Yes. So the approach to treatment failure is just to add on more and more and more antibiotics. So you can start out with your Venco, your Peptazo, and you can add on Meropenem, add on Microfungin, add on Ipatursin, and once you hit seven antibiotics, then I think you should consider whether or not it's a non-infectious etiology. That is hyperbole, ladies and gentlemen. Okay, check out the old post around the nuances of question around that. Last up here, Josh, bronchodontics, I think we've addressed this many, many, many, many times. Is it indicated to investigate VAP, yes or no? So this is fascinating. So the IDSA slash American ATS guidelines suggest against routine bronchoscopy, but the European Respiratory Society is more favoring routine bronchoscopy. Now, not being a huge fan of the bronchie-donchie, I'm going to go with the American guidelines on this. And the evidence really doesn't support routine bronchoscopy as having a big role here. So typically, I'm going to avoid doing this unless there's a good reason. For example, if the patient's neocompromised or if we're worried about aspergillus or something like this. Okay, Josh, before you give me those pearls, let's talk about the summary. So at length, we'll do this every morning pretty much. Possible VAP. You're going to look for evidence of inflammation, clinical evidence of pneumonia, and then radiographic evidence. If you have none of those, go looking at your differential and really just always go looking at your differential. Next up, probable VAP, then you're going to send off your more diagnostics. Get more information to say whether you should pull the gun on empiric treatment. And then readdress things whether you have a confirmed diagnosis or not. So that time in the podcast, pearls and pitfalls, when can I screw this up, Josh? I think the most common way this is screwed up is, and I'm just as guilty of this as everyone, but not being kind of systematic about it and kind of focusing down on just a couple things. So you have a patient, they're in ICU, they have a fever, they have infiltrate, you kind of just like anchor on this diagnosis of VAP. You don't really think about the differential and you just like decide, all right, we're going to put them on like potato for seven days. If you really want to do this right, you actually have to do a fair amount of work as far as thinking about it, getting a lot of tests, and then continually going back and reconsidering your diagnosis and realizing that a lot of time you're going to be wrong and that's okay. It's kind of inevitable. Humility always. If COVID has taught us nothing else, right, Josh? Yeah, yeah. And who knows? I mean, since we don't really have a gold standard for VAP, I think it's possible that a lot of the patients that we're treating for VAP don't truly have it. Well, you said we do have a gold standard, we just cut out a bunch of wedges. We're not going to do that, are we? Exactly. Exactly. Josh, Ventilator Associate in Pneumonia, thanks for a helpful post because this is a very difficult burden on us on a daily level, isn't it? Yeah, it's miserable. Thank you, man. So get out there, Linaze, and look to potato for everyone. Exactly. Got it. Thanks, everyone. See you guys later. Bye.