Good afternoon everyone. I hope you're having a great meeting. It's so nice to be presenting in front of live faces instead of a screen. So, many thanks to the program committee for inviting me today. And, um, like our wonderful moderator said, I'll be talking about HAP and VAP. I'm going to speak mostly about hospital-acquired bacterial pneumonia and ventilator-associated bacterial pneumonia as we go. Thank you. All right. I have no disclosures for this talk, and this graffiti wall represents what we try and problem-solve every day. So, 54 percent of our patients that are in the ICU are treated for suspected or proven infection. And as we work our way down from that, 44 percent are community-acquired, 34 percent are nosocomial, either hospital or healthcare-acquired, and 21 percent are acquired in the ICU. And we're dealing with pneumonia all the time. The 60 percent of those are the site of infection, which is respiratory in origin. So, as we work our way through this talk, what we're going to do is focus on sort of the literature. As our previous speaker said, there's so much literature out there, some of the things that we're doing from start to finish in the treatment of HAP and VAP. So, we'll touch on rapid diagnostics and, quite honestly, spend about a third of the talk on this new technology. Then we'll get into empiric therapy. We're all well-versed in the treatment guidelines and what we use for anti-pseudomonials or anti-MRSA agents at our hospital. Let's talk about some of these newer antibiotics and the potential for some of this newer data to lead us in changes in what the clinical practice guidelines might say. Likewise, let's take a quick look at the adjunctive inhaled antibiotics and the randomized control trials that were published there. I think probably if I asked you, by and large, we're a community of antimicrobial stewardship believers and for the treatment of these infections, we believe in short courses. Let's look at some of that data and then some of the tools that might help us in terms of biomarkers and then, you know, the philosophy we all practice in de-escalation. So, let's just do a quick whirlwind of all this literature in the past five or six years since the guidelines have been published. So, we have two sets of guidelines. The IDSA, ATS guidelines that were published in 2016, we have the North American version and then a year later, the European societies published their guidelines for HAP and VAP. And from a microbiologic diagnosis process, pretty much the same. You know, get a respiratory specimen, generally by a non-invasive route via sputum or an endotracheal aspirate. There's no difference in outcomes when you look at that compared to an invasive strategy. So, let's just get a specimen. The Europeans do say for ventilator associated pneumonia though, if you can perform an invasive approach such as a bronchoalveolar lavage and really the idea behind that is to improve our accuracy and then to guide subsequent antibiotic decision making. You can see the thresholds which are recommended there, but more recently, we have at our disposal rapid diagnostics in the form of two different pneumonia panels. One by Univero, a lower respiratory tract, and then BioFire, the film arrayed by BioMeriU. And you can see here, I've put in the different columns, the organisms that are identified by PCR methods and then also some resistance genes that are identified. So, all of these tools from those respiratory specimens can produce these results almost instantaneously compared to what we're used to from a traditional submit that for culture, can wait for results over 24 to 48 to 72 hours. So, you can see here the gram-negative organisms that can be identified, gram-positive organisms, some of the atypical pathogens, pneumocystis with one. You can compare that to the BioFire, some slight differences there. Detection for carbapenem aces, for third generation cephalosporin aces, different ways to detect for the MRSA genes. So, we get a wealth of information at our fingertips almost instantaneously these days. Let's take a look at how these things perform in the clinical setting. So, this is a multicenter study evaluating the BioFire film array and essentially looking at bronchoscopy samples and sputum samples. And what are we seeing when we're getting these results? When you look at total positive specimens, I think what this bar graph on the left tells us is we're getting a lot more return than what we may be used to with traditional culture. So, you can see that 48% of the BAL specimens return something for us to evaluate and perhaps make antibiotic decisions based upon. If we look at the sputum specimens there in the red bar graph, it's even higher. Nearly three quarters of the sputum specimens sent will have some sort of result. And then you can see in the middle bar graph and in the bar graph on the right, is that one organism that's targeted or is that one resistance mechanism that we identify? You can see that of those percentages, what's seen there, and then two or more animals. So, we're getting a lot of information. And what are the pathogens that we're finding here is on the right, so I've just listed the top five bacteria in hospitalized patients. No surprise, Staph aureus being at the top, Haemophilus and Pseudomonas, either two or three. And then you can see Klebsiella and then some other Enterobacteriaceae are there. These rapid diagnostics have incredible sensitivity, much more so than the cultures, right? Because our cultures are affected by our post-immune response, perhaps previous antibiotic therapy. So, we're getting much more information, as I've stated. How do we utilize this? That's what's important. Well, the same group that gave us this evaluation took this a step further and wanted to compare the rapid diagnostic panel to our traditional culture. And they performed this assessment in bronchoalveolar lavage samples, 260 patients roughly. And you can see there in the arrow on the left, what they found were that there were an increase by 63% in positive BALs. And then when you look at the targets reported, whether that's a bacteria or resistance gene, increased 95%. And that may have been a little bit of an underestimate compared to culture because not all of these are reported based on the quantification of 10 to the fourth, which, you know, you may have some growth, but it's not necessarily reported from the probe. But I think what's important about this evaluation is what can this do for us from an antimicrobial stewardship standpoint? So this group looked at patients with concordant findings, meaning they had a positive panel and then a positive culture or a negative and a negative. So essentially the gold standard there. And what they found was that antibiotic modifications could be made in about 70% of patients. It's overwhelming. And remember, this can be done within hours of initiating empiric therapy. The majority of those changes would be de-escalation or discontinuation in 80% of those modifications, which equated to about 50% of the patients. And then their extrapolation of this data was, I thought, something that was unique. Over 18,000 hours of antibiotic time spared, which averaged to be about four days per antibiotic. Pretty impressive if this is true and something that we can apply in our practice. About 4% of patients were escalated. And then, of course, there are false positives. So about 17% of patients were, perhaps, over-treated. Well, we have the Flagship 2 trial, which was published in Landsat Respiratory Medicine just last year, which gives us some idea of, in a randomized fashion, how we can utilize these approaches. So one of the rapid diagnostics, patients were randomized to that strategy or standard culture. And then in the rapid diagnostic arm, protocolized antibiotic recommendations were made within five hours. And their primary outcome in the study was the time on inappropriate antibiotics. We probably scanned through the list on the right already and what those criteria were for inappropriate antibiotics. I want to highlight number five there. I think we're all used to, if we don't identify MRSA, maybe it's by nasal PCR, nasal pharyngeal PCR, or maybe it's pneumonia. We have no hesitancy in stopping anti-MRSA therapy. But oftentimes we'll continue our beta-lactam backbone. This particular, if that was done, it was considered inappropriate there in bullet five. So what we find here, comparing inappropriate antibiotic therapy duration in this box plot on the left with PCR in the red box plot and standard culture in the blue box plot, is that inappropriate time was significantly reduced, maybe intuitively, in those patients that were randomized to the PCR arm. And the mean difference was roughly 40 hours reduction in inappropriate therapy. And then on the right here, you have a hazards plot of freedom from inappropriate therapy. So the probability of being free of inappropriate therapy, again I'll say that, probability of being free from inappropriate therapy was increased threefold in these patients. And you can see that plotted over the course of seven days. So I think, you know, for all of us who practice antimicrobial stewardship, utilizing these tools can be something that can be very, very beneficial. I want to give you two screenshots of what we see on our end at Barnes-Jewish Hospital when we send off one of the pneumonia panels. So this is a particularly tracheal aspirate. You can see here that right away, within hours, we had 10 to the seventh copies of serratia and 10 to the fifth copies of Klebsiella pneumoniae. Then over the course of time, that culture percolates and you can see here we had a quantified number of those. And this was a concordant result. And of course, we get our susceptibilities and we can appropriately deescalate if, depending on our therapy. So that is what we hope happens the majority of times. But quite honestly, what I've found in practice is something like this, where here's a sputum sample from a patient that comes in. On our molecular analysis, we get a whole host of things, acinetobacter, protease, pseudomonas, serratia. We have ESBL production. We have carbapenemase production here. So we broaden out our therapy early on. And we're on that for several days and then you can see down here at the bottom, nothing isolated in that culture. So what do we do at that point? Well, were they on antibiotics previously? Did it suppress that growth? How sick is the patient? So this is a real conundrum, for my money, of where the guidelines are going to go in terms of this microbiologic diagnosis. It's all easy. Yes, do it. Make decisions early on based on that. But what if we have discordant results? What if we fall into that category of potential antibiotic excess? So I think that's something as a group that the clinical practice guidelines authors are really going to have to, you know, sit down and think about. Right now, it would be at that expert opinion level. You do at your institution or I do at my institution. Probably not wrong. So we need some confirmatory trials there before we probably get strong recommendations for utilizing this technology. All right. Let's now move on to the empiric therapy. We all know what we utilize for an anti-pseudomonal, whether it's cephalosporin or beta-lactam, beta-lactamase inhibitor. Maybe it's a carbapenem. We probably generally reserve those in our empiric therapy. Plus an anti-MRSA agent. Well, we have a whole host of new antibiotics that have FDA approval for the treatment of hospital-acquired bacterial pneumonia and ventilator-associated bacterial pneumonia. And these are the four trials that have been published since the last guidelines. Both guidelines were there. And what I have for you are the intervention in the control arm, the dose, the frequency. Also interestingly, the infusion time. We now need to... We're so used to dose, you know, route, frequency. We also got to talk about infusion time when we talk about antibiotics now. I think that's an important thing. What the comparator is. And you can think about, you know, almost all of this is miripenem. Again, probably not necessarily what we do, but from a randomized controlled trial standpoint, essentially the gold standard, right? Where they're published and then what the adjunctive therapy was. So that sets the stage for that. Probably not surprisingly, here are the results. And all of these were non-inferiority studies. I have here the reprove it on the left, which was a clinical cure primary outcome. So that's why the bar graph looks different. So you're looking at higher percentages of clinical cure between the intervention and the control aspect in P, restore ME2, 8XNP. All those were mortality outcomes and all of them were shown to be non-inferior. So in the next guideline iteration, perhaps we have new alternatives to what they recommended previously in these four different antibiotics. I don't think any of us are ready to take the leap as some of them expand our spectrums, right, to cover KPCs or the like, to utilize those up front. But at least we have options, particularly in the light of these new rapid diagnostics, which gives us information within that first five, six, seven hours. But the other thing about this that I want to bring your attention to that may impact the next guidelines are all the potential patient populations that we have larger samples than what we're accustomed to. So for example, just looking at a randomized control trial, if we combine all this from an intervention and a control aspect and ventilator associated pneumonia, roughly 500 in both groups combining these four studies. But what about half? You go look for hospital acquired pneumonia literature and it's really hard to find from a randomized control trial standpoint. So now we have all of these samples. So just, you know, go back to your site. Think about this. You know, if you want to do some projects, we could really do that. Same thing with immunoglycosides, with and without. How many of you use immunoglycosides up front for pneumonia? I'm guessing not many, right? But now we have some literature. Maybe it'll support that utilization. Intermittent versus extended infusion. How about augmented renal clearance? I know us as a group of pharmacists certainly think about this all the time in terms of therapeutic optimization. And even pseudomonas, big patient populations. All right, what about the adjunctive inhaled antibiotics? I don't need to spend a lot of time on this. I think we've all probably been a part of teams that utilize this therapy, particularly for patients that have histories of resistant pathogens and along those lines. Well, what do the randomized control trials tell us? There's been two, the IASIS and the INHALE. And you're probably familiar with these, so let's just quickly go over them. These are combinations of amikacin, fosfamycin in the IASIS and amikacin by itself in the second one. And you can see here two different endpoints. So the IASIS on the left was an improvement in clinical pulmonary infection score. And you can see between these two lines, which were the amikacin in the blue diamond and placebo in the circles, essentially superimposed lines over the 14-day period. So no difference in that study. What about the randomized control trial, the amikacin INHALE versus placebo with a mortality endpoint? I've showed you the hazards plot here. And again, superimposed for the cumulative proportion of mortality over 30 days. So really no additional benefit. You know, I don't want to rule it out, but in the guidelines, these aren't going to be something that we utilize in most of our patients. What about treatment duration recommendations? You know, here's, again, the IDSA, ATS on the left and the European recommendations on the right. We're sort of locked into that week-long duration of therapy. I know in community-acquired pneumonia, we're thinking about shorter durations. No new literature on that yet in nosocomial pneumonias, but something that we need to start considering. I give you the strength of recommendation and the quality of evidence for these because I think even though the same studies were considered and they're systematic reviews and meta-analysis and they both use the Gregg criteria, they came up with different, you know, strengths of recommendation. So very interesting. One other caveat, you know, the asterisk from the European guidelines that seven to eight days may not be long enough in patients, particularly those who are immunocompromised, those with cystic fibrosis, and so on down the line. You can read those things there. And that's probably because the thought that pathogens like Pseudomonas or non-fermenting gram-negatives, you know, maybe treating for a week isn't long enough. So we're all well aware of the Shastra study published back in the early 2000s comparing that. Well, there's been a recent publication of a follow-up. The French, you know, attempted to do a wonderful study here where they had patients with Pseudomonas aeruginosa, ventilator-associated pneumonia, and randomized them to either that eight versus 15 days. And, you know, by their upfront estimates, they wanted to enroll 600 patients to show differences in death and VAP recurrence at 90 days. Well, what an extraordinary task that was going to be. And they had many centers. But in the end, really just couldn't recruit that many patients to get a definitive answer. So I share with you, you know, essentially a snapshot of this. You can see the number of patients, the intention to treat, and per protocol arms. And the results here. And for their primary outcome of death and recurrence, both the intention to treat and per protocol, they were not able to demonstrate non-inferiority. Probably not a surprise given their sample size. There was a significant difference in recurrence, higher in the eight-day group, similar to what we've seen before in the larger trial. No difference in 90-day mortality. And reductions in mechanical ventilation and antibiotics. So, you know, again, I think this just leaves a question. In the next guidelines, you know, what are we going to say for those non-fermenters? I think what is important is this study. The longer that we treat these anti-pseudomotiles in particular, cefepime, meropenem, peptazo, this comes from our institution. We're putting our patients at risk at increasing resistance over the course of the next 60 days. So you can see here the odds ratio of any anti-pseudomotile beta-lactam. And the three ones broken up that for each additional day of exposure, we're putting our patients at increased risk. And the plot on the right shows that there's really no ceiling effect. So the longer we go, we increase that risk over time. So this is just wonderful data published by my colleague, Basu Tesho. What about biomarkers to help us? You know, if I had you raise your hands, probably some of you use PCT or procalcitonin, some of you don't. We saw for C. diff, you know, there's IL-1 or there's IL-8. The VAP rapid study's been, or VAP rapid study's been published. Let's just go over procalcitonin, because I think it's probably the most readily available. The study on the left is a multi-center randomized open-label study. It focused on patients with severe sepsis, septic shock, not necessarily pneumonia. You can see the percentages of those that had pneumonia. And in this particular study of roughly 800 patients in both groups, you know, the daily defined doses of antibiotics were significantly reduced in the PCT group. Antibiotic duration to antibiotic free days all showed that. Well, what about the other trials? We're probably all well aware of the PROACT study that was published in New England a few years back. These were primarily patients with community acquired pneumonia. And the bottom line here, there was no difference in total antibiotic exposure. The PROGRESS study, which looked at hospitalized patients with sepsis, again randomized them to the two published in the Blue Journal, very small numbers of hospital acquired pneumonia. But in the end, they did show a significantly reduced rate of infection associated adverse effects, which was the diagnosis of C. diff or the diagnosis of a multidrug-resistant organism. All right then, to finish up, de-escalation. These are just a couple of my favorite studies from observational studies that have been published most recently in open form and infectious disease and using a spectrum score. I think we're used to de-escalating meropenem to cefepime to ceftriaxone. You know, that's a de-escalation. This puts a little bit of objectivity to it. And essentially what this showed in patients with HAP or VAP is that, you know, similar rates of treatment failure, which was death or recurrence, but we had fewer antibiotic days, fewer episodes of C. diff. So it's a good thing. It's a good thing. We haven't shown that it necessarily decreased, it's been associated with decreased resistance, but it's a good thing. And then two studies, one in infection control and hospital epidemiology that, you know, positive versus negative cultures, it would seem that we're able to de-escalate more frequently, maybe have shorter courses with positive data, but that's not necessarily the case. I think we're at a place as a community of critical care practitioners from an antimicrobial stewardship slant that we're squeezing that balloon as much as we can. So let's utilize these new technologies if our hospitals allow. Let's utilize these new antibiotics. And thanks for your attention. Thank you.

hospital-acquired bacterial pneumonia

ventilator-associated bacterial pneumonia

rapid diagnostics

pneumonia panels

antibiotic decision making

antimicrobial stewardship