 entitled, “Battling Against Resistant 

 Infections in the ICU.” Multidrug-resistant strains of 

 are increasingly found in hospital intensive care units, putting patients who are already very sick at risk for developing hard-to-treat infections. Timely diagnostics and effective empiric strategies are essential to the successful treatment of these infections. This 

 panel of experts in infectious disease will discuss current diagnostic and treatment strategies and best practices for antibiotic selection for resistant 

I am Dr. Peter Salgo. I’m a professor of medicine and anesthesiology at Columbia University College of Physicians and Surgeons. I’m associate director of Surgical Intensive Care at New York-Presbyterian Hospital. Joining me for this panel discussion are Dr. Yoav Golan, an attending physician at Tufts Medical Center in Boston, Massachusetts; Dr. Marin H. Kollef, professor of medicine and the Virginia E. and Samuel J. Golman Chair in Respiratory Intensive Care Medicine at Washington University School of Medicine, and director of both Clinical Care Research and Respiratory Care Services at Barnes-Jewish Hospital of St. Louis, Missouri; Dr. Jason Pogue, an infectious diseases clinical pharmacist at Sinai-Grace Hospital of Detroit Medical Center and clinical assistant professor of medicine at Wayne State University School of Medicine in Detroit, Michigan; and Dr. Andrew Shorr, section head of Pulmonary and Critical Care Medicine at MedStar Washington Hospital Center, and professor of medicine at Georgetown University in Washington, DC. I want to thank all of you for being here. We have a lot to cover, so let’s begin.

The obvious question is, who are the patients in the ICU who are most susceptible to 

, right? We don’t have to give the full name. Where do we start on this?

 Well Peter, the sickest patients are usually the ones who get 

 infection. In the intensive care unit, there are certain categories of patients who seem to be at higher risk. Certainly, patients who are immunosuppressed fall into one of those categories. At our institution, we happen to have separate ICUs, including an ICU for patients who have bone marrow transplants and certain malignancies. We certainly see a lot of 

 within that population. There are other critically ill patient groups where we tend to see 

 as well. We run a very large lung transplant program, and we have a large cystic fibrosis program. Patients who have structural lung disease, in particular, seem to be at very high risk for developing Pseudomonal infections. Obviously, other patients who have certain disruptions of skin—burn patients, etc&mdash;fall into that category as well. One thing that is important to keep in mind is that many of these patients have 1 common factor in place, and that’s having received prior antibiotics. These critically ill patients are often getting antibiotics, and for that reason, it can predispose them to infections, particularly with resistant bacterial like 

But I was always taught that it isn’t the 

 are actually a marker for really serious disease. The 

, per se, are not all that virulent. Is that fair? Is it 

 that are causing the illnesses, or is it the illness that’s leading to 

 If you look at just the virulence factors. But you’re absolutely right to add in the fact that it hits our most vulnerable patients and it has the highest likelihood of having drug resistance, so we’re most likely to get it wrong up-front. Sometimes we’re forced to use suboptimal therapies on the back end as well. So, it all plays hand-in-hand, but it certainly has its virulence factors as well.

Certainly it does, but what I’m saying is, as you point out, the sick people—the weak, the immunocompromised, the previously treated, the ones who can least afford to get a virulent infection&mdash;they get the 

 In many cases, as Jason says, getting infected with 

 would be the turning point. You mentioned cystic fibrosis. Many patients actually do well until they get colonized with 

, but once they do, they start getting infected. So, you’re right, you need a predisposing factor sometimes. But then, 

 may be the factor that would actually result in a patient’s decline.

 The other point I think is important is that we’re using a paradigm right now, this part of the conversation where we’re presuming the 

 are an ICU-acquired phenomenon. The vast majority of ugly 

 is brought to me from the floor, where antibiotic therapy has been misused, such as 20 days of piperacillin/tazobactam because a patient has an ingrown toenail.

You don’t do that? That’s not a good idea?

 Actually, in our place, it’s actually in the Jell-O on the tray for patients.

But we were told to never stop the antibiotic too soon.

 With that, there’s a whole host of evidence now, and we’re going to drag pseudomonal therapy into the evidence-based literature paradigm at some point. But the other point is, I see plenty of patients with classically health care-associated syndromes where they come from nursing homes, they’ve probably been in the hospital recently, they’ve probably seen prior antibiotics, and they bring their multiresistant 

 with them to the hospital. So, I think it’s important to realize that although there’s plenty of ICU-acquired 

 infection that’s brought to the ICU. You have to be vigilant about it, not only as an ICU-acquired pathogen but also just as a pathogen in sick patients.

Pseudomonas Infections in the ICU: Who's Susceptible?

I heard something else that I think we need to at least mention. I think it was you who said that if you have 

 infection and invasive disease. What’s the difference?

 The clear thing is that a lot of patients are colonized with all kinds of microbiology, but we’re only concerned with patients who actually develop clinical infections from the same place. Colonization will often precede infection occurring, but certainly we’re concerned about the patients once they develop the pneumonia from them having that longstanding colonization of the organism.

I want to bring it back full circle then, which goes to what I was saying. You can be colonized with 

, but in order to get an invasive infection with 

, don’t you need to be sicker? Don’t you need to be one of those patients who’s already, in some way, really ill? Or can you just get 

 Generally, healthy patients in the community don’t develop pseudomonal infection for the most part. There may be a few exceptions to that rule, but to develop a true pseudomonal infection, it really requires certain host factors. I think that’s what we’ve been commenting on.

 In addition to the host factors, I think it is important to recognize that it’s a very virulent organism, and I think the audience really needs to be clear on that particular point. Because it’s a virulent organism and it has certain virulent mechanisms in place, antibiotic therapy does become important. We don’t have mechanisms for treating the virulence of the pathogen, so we have to really err on the side of treating it with an appropriate antibiotic regimen.

, but focusing on the antimicrobial aspect of it, since we have limited options for tackling the virulence factors.

 More importantly, focusing on prevention. There are things that we can do to prevent severe infection in our critically ill patients, whether it’s early ambulation, early liberation from the ventilator, less sedation, or chlorhexidine bathing. One place where we take infection prevention very seriously in the hospital is actually the ICU, where we have bundles of protocols that are evidence based and have been shown in randomized trials to effectively prevent some severe infections. And so, if you’re going to work in an environment where you know it’s enriched for the risk and bad outcomes associated with the bug you’re describing, then it’s incumbent upon all of us to actually start this conversation at, what are we doing for prevention?

You want to back off. Before we even talk about antibiotics, if we could prevent people from becoming infected—or if they’re colonized, we could prevent it from becoming invasive&mdash;that’s the ideal.

We have techniques for that. Whether it’s a VAP (ventilator-associated pneumonia) bundle or something else—as you said, chlorhexidine&mdash;that would be ideal, right?

The data for chlorhexidine are more controversial than they used to be. We’ve got a positive trial, but we’ve got some follow-up on negative trials.

I heard a chuckle on the right side of me.

 Your patient smells better if they’re bathed in chlorhexidine every day, but I think it speaks to the fact that people are striving to find good, effective, inexpensive strategies, some of which have abundant evidence in support, some of which are controversial when you make a decision about the risk-benefit tradeoff.

 I think with some of the most invasive things that we do, there’s a conscious effort in the ICU to shorten the duration. So, ventilation might be an option for that.

 Pulling in the central line sooner on these patients. Those are known risk factors for hospital-acquired infections, chlorhexidine in particular. And so, I think a lot of effort to prevention comes along those lines.

By the way, I think the correct phrase now is health care facility associated infection. It’s morphed again.

 Obviously, another component that is very important in our efforts to prevent infection is our use of antibiotics. What antibiotics are we choosing as a matter of protocol? We’re going to discuss this later, but a lot of that is actually in our hands.

 as if it affects 1 organ system. People always think about 

 pneumonia, which is common, but there are lots of other 

 is a pathogen that, once it sets up shop in your hospital, can cause any kind of infection because it’s a bug-host interaction. It can certainly be a leading pathogen in a ventilator-associated or hospital-acquired pneumonia, but it can still be a leading pathogen in bloodstream infection. My colleagues in the surgical ICU see surgical-site infections in postoperative mediastinitis after CABG (coronary artery bypass graft). If 

 will find a place to get in, and it’s very much the same way with 

. If it has set up shop, it’s not specific for 1 infection or 1 organ infection. It causes infections in any of those syndromes, even including urinary tract infection. You need to be vigilant.

Is there one infection that’s worse than others?

 Before that, just to add to what Andy says, I think it’s important to understand the biology of 

 when you try to predict the kinds of infections. It’s important to remember that 

 is ubiquitous. It’s a moisture type of bacteria. It can exist in the environment. It tends to colonize people, particularly people who get antibiotics and are in health care institutions, particularly on mucosa surfaces. And anything you do to violate the integrity of the mucosa—by putting a line that breaks the mucosa going into the vein, by using a urinary catheter that dilates the mucosa of the urinary tract, by using endotracheal tubes, and so forth&mdash;will provide 

 with an opportunity. So, you tell me what kind of violation of the mucosa you had with your patient and I’ll tell you what the patient is going to be vulnerable to develop.

. You can say that that for staph infections. You can say that for anybody, right?

 are a few factors. One is, as Andy said, that once you get colonized with it, it’s really hard to get rid of it. It’ a long-term colonization. If you look at the risk of infections, it’s right into colonization and how long you are going to be colonized with the bacteria. What is the ability of this bacteria to survive the antibiotics in the environment, as 

 is not unique in principle, but it’s more quantitative. That makes it more likely to infect those patients.

What scares you guys more? Is it pseudomonas pneumonitis? Pseudomonas cystitis? Pseudomonas something else? Is there 1 organ system that, when infected with 

 What scares me the most is once it infects my sickest patients. Because they are the least able to tolerate infection and they’re the ones who require the most aggressive approach.

Aren’t they the ones who are most likely to get it? That’s what makes this bug so nasty, among other things.

 Right, but it is in some way a tautology, right?

 I think about Yoav’s point about it being a function of the patient, not a function of the bug. If you’ve had a kidney transplant, I’m worried about an ascending tract infection from 

 If you’ve been on a ventilator with acute lung injury from flu or on ECMO (extracorporeal membrane oxygenation) for 2 weeks, I’m worried about pneumonia. It is a pathogen that goes where the host offense is weakest. Can I say that pneumonia is associated with the highest accrued mortality rate 

 versus urinary tract infections? Yes, sure, but that’s because it’s easy to eradicate pus in the bladder and it’s hard to eradicate pus in the lung. So, that’s also true for every other pathogen.

Pseudomonas Infections: Matching Causes and Prevention

 infections, as you pointed out, they’re everywhere. They’re in the hospital. Some patients come to your ICU with 

. How common are they, all-comers into the ICU versus those developing 

 in situ in the ICU de novo? That’s a lot of Latin.

 I think it depends on the infection that you’re looking at. At least within our ICUs, pneumonia is quite common when we’re talking about 

. On the other hand, when you look at all our urinary tract infections in the intensive care unit, 

 probably would make up the vast majority.

 I would like to add is that it also is very proficient at forming a biofilm. It can stay hidden for a period of time. That echoes some of the other comments that people have made. If you’ve got a device in place in an endotracheal tube—or even within abnormal mucosa within the lung, within the airways&mdash;it can essentially hide there for long periods of time in patients who are placed on antibiotics, who may have other stress factors, and who are not getting adequate nutrition. It just allows that infection to come forward.

You mean a biofilm is a very gentle way of saying it forms this yuck.

It’s concrete, right, and human antibodies have trouble getting through it.

 And it does quorum sensing, so it doesn’t even reproduce. If an antibiotic penetrates, it’s hiding from it.

 There’s actually very good literature, particularly for 

, showing that in bathrooms it almost develops in a community sense.

 There is communication within this community. And as you know, when bacteria is self-dividing, they become far less susceptible to antibiotics and can actually send messages to each other: “There may be an antibiotic in the area.”

 on social media: “Watch out, they’re coming this way.”

 was still doing cartoons, this would be ripe material.

Ripe material. But that does bring up something that you pointed out, which is the institution. Does every institution need a biogram? This is our 

, this is what we see, and this is how we need to treat it.

 Absolutely. You can argue that you don’t have a multidrug-resistant 

 problem, but hospitals that do that, when I speak with them, have what I call a WNL issue. And WNL doesn’t stand for “within normal limits.” It stands for “we never look.” They don’t look. They don’t think they have a problem. When we look and break out our ICU antibiogram, especially comparing it to our ward or non-ICU antibiogram, that information is usually insightful. There is a lot of MDR 

 in the ICU among the isolates we collect.

There’s a lot on the floor as well, and that process and that conclusion allowed people to realize that we needed a system-wide approach and paradigms in terms of protocols as opposed to saying, “Well, let’s just target the ICU because it’s geographically constrained.” That makes you miss a lot of what’s underneath the tip of the iceberg.

 Yes, I think that floors/ICU observation is absolutely true. But I also think we tend to culture those patients.

 The ICU patients get cultured more, so we see it more. It’s the same thing as looking and seeing it from that standpoint. But I totally agree, and I totally believe them. I see specific antibiograms. They’ve been shown to improve on empiric therapy. But they come with a lot of asterisks that people need to be aware of from that standpoint.

 You need to know patient-specific risk factors. I can look at day 1. If I look at my ICU antibiogram on day 1 of admission and it’s still pretty nasty, it’s because they’re coming from long-term care facilities. They have history of all these previous antibiotic exposures and ventilation. If that patient goes to the floor, they have the same risk of having 

. Actually, they’re just not as sick. But I think that patient-specific factors, as we learn more and more, are even a bigger determinant than ICU versus non-ICU.

I want to take your point about patients not being cultured, which is to say that if you culture enough patients, you can develop an antibiogram anywhere, right? The lab’s got the data; the lab’s got the sensitivities. Are you saying that we need to culture more on the floor?

 I would, as a stewardship pharmacist. I would love the floor patients to be cultured more because then I can actually do antibiotic modifications. When they’re not cultured, they get started on empiric therapy and then we’re stuck with those therapies. But yes, we would understand that microbiology better and we’d also have the ability to modify therapy accordingly.

Pseudomonas Infections in the ICU Versus on the Floor

I’m always worried and nervous when people use the term “culturing” a lot because that often drives antibiotic use. If it’s culturing in response to a syndrome that looks like an infection, I’m all for it. The other comment I’d make, and I use this analogy all the time with the residents on rounds, is that the ICU is basically like a funnel. We’ve got, at the bottom of the funnel, the intensive care unit, and it’s a collecting bin for resistance genes. At the top, we’ve got long-term care, nursing homes, the floor, and other hospitals, especially in a system like ours. And so, all those resistance genes make their way into the intensive care unit.

Interestingly, one of the things we haven’t brought up yet is this issue of health care-associated infections that are community acquired. All that means is that those are infections that occur outside of the hospital that come in, but they’re still due to antibiotic-resistant pathogens.

 Because they got the antibiotics somewhere else.

Or they had contact with the health care system. They were back out there, and then they came back. And that makes up a large significant pool, even more so in the pneumonia patients that we see than ventilator-associated pneumonia. Those are now hospital-acquired pneumonias and health care-associated pneumonias in our intensive care units.

We go back to something that I heard when I was a medical student. This is not new. Patients demand antibiotics from their primary care physicians. “I’ve got a cold, I’ve got an earache, I stubbed my toe. Can’t you give me a tertiary antibiotic of some sort?” The doctor says, “Sure, here’s a script,” and the next thing you know, send us your tired, your poor, your 

 I think it’s a little more nuanced than that. I think the point about antibiotic abuse in the community has a lot to do with resistance to traditionally utilized community-focused antimicrobials, like levofloxacin or what have you. I think what Marin is really alluding to, and what he’s published extensively on, are these patients who are in the iron triangle of health care. They go from the hospital ICU to the floor, to the LTAC (long-term acute care), and then back, and while they’re at the LTAC, someone is treating them with antibiotics because they can do something.

There’s this cognitive bias that doing is better than watching in medicine, which is actually clearly harmful. What happens is that this person is never actually in the community, and so they’re continuing to be exposed to health care. They’re probably in an environment where there’s poor infection control, they’re getting antibiotics, and when they get sick, they show up to the emergency room and the original prescribers in the emergency room are saying, “Oh, well they came from the community, so it must be 

.” They’re missing the fact that this patient has actually never been in the community.

 This goes to your point. You’re worried that people culture for the sake of culturing. But if they were to culture in the LTAC and treat specifically with some laser-focused antibiotic, that might be better.

I think it would be better if clinicians were focused on treating infections. But too often, the patient has a mild temperature spike and the trigger reaction is just to go right to the antibiotics as opposed to really thinking it through and maybe just watching that patient for a period of time.

 Just to connect the adequate use of antibiotics or not using antibiotics when not needed to the culture question. First of all, I think it’s really important to differentiate surveillance cultures. You really want to know what the patient carries for when the patient gets sick, and we don’t do a lot of that. I think they do much more of that in Europe and so forth. I think we have to talk about, how do you know what kind of bacteria is causing this severe infection? How do you use the right antibiotics? I think that we’re going to be using more surveillance cultures, because we find that these are helpful.

I think it’s also important to understand that people have to, not for surveillance’s sake, diagnose an infection for guidance and management. People have to make sure that patients qualify for an infection from a clinical perspective first. In the ICU, we see a lot of patients who are suspected of having pneumonia, for example, who don’t really meet the clinical criteria of actually having pneumonia. And so, any culture that will drive treatment may be driving the unnecessary use of antibiotics. Those patients are at high risk for infections, so a future infection where they actually do need the antibiotic may be harder to treat.

Pseudomonas Infections: Nuancing the Use of Antibiotics

 infections differ from other patients? Are they more expensive? Do they do better? Do they do worse?

 I think that’s our biggest challenge. They don’t have a sign that tells us they have 

 compared with some other resistant gram-negative organism. If you look at risk factors, studies, or scoring systems, everything overlaps. It gets back to the same concept that we’ve been talking about so far: understanding your local epidemiology and knowing your patient-specific microbiology history. What are they? Get surveillance cultures, those types of things, or previous antibiotic exposures. That’s going to drive it because they’re completely overlapping risk factors. Pick your favorite drug or just an organism; they have the exact same risk factors.

Not to put a grenade on the table, but we have a fair number of leukopenic patients who come in with only the fairest wisps of hints that something’s going on. They’re developing pus, they’re not getting big infiltrates until their white cell count comes back, and they need antibiotics. How do you do that? This is all a challenge, isn’t it?

 I do think there’s something unique about 

. I say that because we did a study where we looked at patients who had bacteremic pneumonia at Barnes-Jewish Hospital in St. Louis. What we found was that the patients who had bacteremic pneumonia due to Enterobacteriaceae were more likely to get treated with an antibiotic that didn’t cover the bug. But the mortality was highest in those who had pseudomonal infection. And so, there’s something about the host—bug interaction in that setting that’s really key. It really makes me pay more attention to patients who I think are at risk for having a 

I think that’s where we started, which is not only that 

 is a nasty bug but also that the patients who get it tend to be the most vulnerable. And so, they’re more likely to die of all causes. 

 is maybe what tips them over, but it’s a nasty combination.

 But there’s still a lot to achieve with adequate therapies.

 They are the ones who lose the most from inadequate early therapy.

They have the most to gain and the most to lose. Nobody is sitting back and saying, “Oh, look at Mr. Jones. He’s so old; he’s so immunocompromised; he’s so sick; leave him alone and let him die.” No, that’s the fun part, treating him. Right? Making him better.

 But it’s not just Mr. Jones. Without getting into specific details, we recently had a couple of patients—they both had cystic fibrosis&mdash;in our intensive care unit, and both had multidrug resistances. In fact, it was extremely drug-resistant 

 that led to fatalities in both cases because we couldn’t treat the organism. So, it’s not just older patients who really have nothing left. These are younger people who potentially have years or decades of life ahead of them.

We have to go on and discuss the treatment options that we’ve got because nobody wants to let these people die if we can at all help it. So, what’s out there? What are the antibiotic classes? What are the agents we’ve got that are effective against resistant 

? Why don’t we just run through some of the advantages and disadvantages. Do you want to start us off?

 I would have to go with our own experience, and I would have to say that carbapenems are really the backbone. We’ve gotten to a situation now—and this is even true among our house staff&mdash;where for anyone who comes in with any type of a risk factor, we tend to go to a carbapenem. Now, in our institution, we have other agents that also may be effective, but the carbapenems seem to have the greatest overall activity, at least locally. We’ve become more of a carbapenem-driven hospital. I’ll let some of the other people comment on other things.

 I would just emphasize here that it will be institution specific. You need to know what you’re looking at because we’ve actually flipped our institution’s carbapenems. Cefepime, piperacillin/tazobactam, and meropenem are the least likely to be active against 

 empirically, and it’s just because of the complexity of resistance mechanisms.

 Yes, but I’m talking about in the ICU.

 We have the same thing in our ICU. About 20% of our 

 infections are carbapenem resistant. Piperacillin/tazobactam continues to have the highest activity, and we don’t actually have an ESBL problem in our institution. We’re very fortunate thus far. And so when we do the game and the numbers, we say that theoretically, I want to get to a 95% empiric hit rate from the gram-negatives I might encounter. What do I need to start with? We generally start with piperacillin and tazobactam plus aminoglycoside.

I was going to say, am I hearing this right? You’re using monotherapy against 

 There’s a difference. There’s monotherapy against 

 when you know the pathogen and you know the susceptibilities, which I think we would all call the de-escalation phase or the definitive therapy phase, versus multidrug therapy as an initial pathway to ensure that the patient’s getting at least 1 drug that’s in vitro active against the culprit pathogen. That’s an important paradigm and an important lesson that I just realized: What you start with ought not actually be what you finish with.

Fair enough. Sorry, you wanted to say something?

 Just to build on that a little bit too. It’s not just 

. Andy suggested that before. But if you happen to have a lot of carbapenem-resistant 

 in your ICU, that all has to go into that empiric therapy decision. So, you’re looking at that cocktail, that combination, that gives you the best ability to give it up front.

Now you see, I anger all my colleagues. Maybe not anger, but I make them uneasy. When I see somebody who I really suspect has resistant 

, I take the grenade, I pull the pin, and I throw everything I’ve got against the wall. I’m perfectly happy to calm down later when we get the culture results. But our antibiogram is that these are bad bugs, and we often need 2 or 3 drugs. The aminoglycosides are always part of it, no?

 Well, every patient and every infection is different, and I think that always being very aggressive or never being very aggressive is probably not the way to go. The patient tells you how aggressive it is to be. If the patient tells you, “Look at me and how sick I am; look at me and how quickly I deteriorate.” The patient tells you that you only have 1 chance. In those circumstances, you want to be very aggressive. If the patient tells you, “Listen, I may have a UTI; I may have pneumonia or not; I’m relatively stable; I’m not that sick to begin with,” maybe you’ll take a different approach. Remember that the antibiotics sound like any other medication.

To use overuse them, use too many of them, will affect your ability to treat your patient. You have to balance them. When you talk about the antibiograms, when you talk about personalizing risk factors, I think it’s important to do that. I think that’s the infectious disease doctor’s approach. Very often you say, “I want to do everything because my prospective patient is everything I have on my table right now.” That’s an OK approach, as well, but I think that you have to balance it.

We talk about antibiograms as guidance for treatment. In hospitals that have a lot of those ESBL-producing bacteria, the physician uses a lot of carbapenems. Hospitals that use a lot of carbapenems push their 

 to become resistant. Hospitals that have less of that may have more susceptible 

. So, the antibiogram gives you some guidance, just knowing the type of infections treated in your ICU.

Fighting Pseudomonas: Impact On Patient Prognosis & Approaches to Combat Infection

I haven’t heard 2 classes of bug juice at this table. I haven’t heard anybody really talking about the fluoroquinolones. What about—here’s a really old drug&mdash;polymyxin? Do they have a role here? Does anybody want to step up on this one?

 I’ll take on fluoroquinolones. There’s no role.

 That’s pretty easy. In our hospital, fewer than 30% of our 

 infections are susceptible to fluoroquinolones. Beyond the fact that most people don’t know how to dose the fluoroquinolones properly if they’re going to use them, there are all the issues associated with fluoroquinolone safety, particularly 

 for us. We’re really trying to take fluoroquinolones out of our institution completely. So, they’re not really an option.

As for colistin polymyxin and all those agents, this is an area where we’re actually clearly getting into strong opinions because we don’t have strong data. That’s the nature of medicine: The stronger the opinion, the weaker the data. I think all of us would agree that colistin is probably a laundry detergent masquerading as an antibiotic. Is it sometimes the only choice we have? Yes. Is it ever our first choice? In the United States, no. In other parts of the world, yes, because they’re dealing with a higher prevalence concurrently of 

, CRE (carbapenem-resistant enterobacteriaceae), and 

. And 50% are infections, so they have to start with colistin. But nobody knows how to dose it properly, especially in critically ill patients with changing renal function.

Do I use it sometimes? Yes. Do I use it for MDR 

? Pretty much no because I have other choices before I get there, including not only different classes of antibiotics but also different dosing paradigms for the antibiotics we currently have. One of the lessons that I’ve learned from my ID and pharmacy colleagues, and why I think it’s important to realize this is a multidisciplinary group, is that pharmacokinetics matters. Just because I have an MIC in a test tube doesn’t mean I can’t figure out how to be smarter than the resistance by giving more of the drug over a longer period of time and optimizing these pharmacokinetic principles that we forgot about from 20 years ago.

 I’ll build on both of those things. First with fluoroquinolones, I do think they have a role. First off, in our institution, our susceptibilities have come back a good bit because we restrict them. I think the bigger issue with fluoroquinolones is that the MIC break point might be wrong. To your point, the way that we dose them is probably suboptimal in a lot of those patients. I’m not saying that I like them empirically; I’m not saying they’re my go-to; but they’re potentially a step-down option to these patients. I do think they have a role.

With regards to the polymyxins, yes, they have issues. As Andy said, we have some other agents that we can use. Since you’re talking about 

, you’re often talking about lung infections. Then there’s an emerging body of evidence that suggests those drugs maybe don’t work in the lung, so they would be a suboptimal choice.

 Just to add to that—and correct me if I’m wrong, Jason&mdash;my experience has been that over the past couple of decades, one of the most common combinations of empiric therapy has been Peptazol (pantoprazole) and levofloxacin to cover gram-negatives. And yes, you have to use institution-specific antibiograms, but at our place and at most others that I can find, I can’t find any evidence that there’s any additive effect of combining those 2 drugs. What you end up getting is the side effect of levofloxacin&mdash;

, etc—when in fact what that patient should get is an aminoglycoside.

 I agree wholeheartedly, and I think that brings us into the combination antibiogram piece of the story. If you’re going to choose a second drug, you want that second drug to be active against those that are resistant to the first drug. When you have β-lactam resistance, you often have fluoroquinolone resistance, as well, which makes the aminoglycoside that much of a better empiric drug. I completely agree with that. That said, I can tell you that we’ve long restricted fluoroquinolones at my institution. To Andy’s point, we rarely use them, to be honest with you. We have seen rebounding and susceptibility to a degree, and I think they have a role if used correctly as step-down therapy to complete therapy from an overall standpoint.

Can we just put something on the table? I hear this every day. Are aminoglycosides safe to use in patients with renal failure?

 Everybody would say that 1 dose of aminoglycoside while you are waiting for culture results to come back is fine. It’s the extended duration, and certainly the duration of exposure, associated with the interaction with the current kidney injury and fluid staph infections to the patient that matters. I think none of us use aminoglycosides as our monotherapy definitively for 

, particularly in the lung. In my institution, it’s usually that they’re on something plus amikacin because that’s our best aminoglycoside, for 2 days. If we know what the bug is, we drop to the agent that’s going to kill it. If we have an isolated gram-negative, then it’s gone.

Classes and Agents Available to Treat Pseudomonas Infections

Peter L. Salgo, MD; Jason Pogue, PharmD, BCPS-AQID; and Yoav Golan, MD, discuss 

 infections’ ability to evolve and acquire resistance to fluoroquinolones, polymyxins, aminoglycosides, and other such agents, complicating treatment approaches.

&#8203;&#8203;&#8203;&#8203;&#8203;&#8203;&#8203;

We’ve outlined a whole set of classes of antibiotics. We’ve got the β-lactams; the aminoglycosides; the fluoroquinolones, for better or worse; and then detergent, or polymyxins. How often do we see 

 Resistant to all of them, as in extensively resistant?

By the way, we ought to at least get the definition down. Nothing is 100% resistant necessarily, but the MICs are really high. So, let’s just call it resistance.

 Luckily, we don’t commonly see pandrug resistance to all of those agents, and we actually have some new agents that might play a role in that situation.

 I know. Foreshadowing is what I like to call that. But we do see those patients, I would say infrequently. One of the things that’s important to know with β-lactam, and I think it’s an important lesson to understand, is that we’re used to dealing in the world of carbopenamases and β-lactamases, but for β-lactamase resistance in 

, it’s much more complicated. Sure, there are some β-lactamases, but the different mechanisms of resistance will hit different β-lactams to different degrees. So you could have something that’s carbapenem resistant but also susceptible and so on and so forth. And so it is rare that you completely lose all those options, but it does occur. But multidrug resistance is the norm.

 But there’s a caveat to that. If you see oscillations in ICUs, if there is a bacterium that’s resistant to everything available, which one would it be in your ICU? In many cases, you hear it will be 

. Even though it’s not very common, it exists. But the problem is that if the 

 is susceptible to only aminoglycoside, it’s not treatable by aminoglycoside monotherapy. It’s practically resistant to everything. We now know from several studies that were done recently that when you test as susceptible to the polymyxins—colistin, for example&mdash;at least in 1 study that was published recently, half of the cases were actually resistant and susceptibility was misread. And so you don’t really know. Another factor that’s really important is that we talked about resistance in 

 is one of the most capable bacteria of developing resistance to therapy.

When you start with a high inoculum infection, you hit it with 1 antibiotic or maybe 2, but it’s not susceptible to all antibiotics you have. With high inoculum, bacteria divide and mutate after 3 or 4 days. If you look at different classes of antibiotics that are most vulnerable to this development of resistance on therapy, the product carbapenems are actually the most vulnerable to that. And so you have to take all those and tailor the treatment. If you start the patient on what looks rational—maybe you have the culture result that guides you&mdash;and the patient does not respond the way you would expect the patient to respond, you should think about reculturing. You should think about changing your antibiotic.

Somebody once said to me, and this is almost a direct quote, “If you don’t believe in evolution, take a look at 

.” It’s evolving; it’s changing to accommodate to its environment, in a bad way for us. You mentioned some of the mechanisms of resistance. One of them is this brick wall that develops. What else does it do to become resistant to antibiotics?

 It does many different things, but 1 of the 3 different mechanisms that are perhaps the most substantial, quantitatively, is its ability to produce enzymes that break the antibiotics, mostly β-lactamases but also aminoglycosides. Hydrolyzing or breaking enzymes. At some institutions, this is the main mechanism of resistance. If you are going to discuss the new antibiotics coming to the market with those bodyguards against those enzymes, you’ll see that those antibiotics are very, very good in some institutions and not the best in other institutions depending on that. So, producing an enzyme that will break an antibiotic is 1 approach. And of course, to protect the antibiotic, you can give a drug with a bodyguard that will prevent the enzyme from breaking the antibiotic.

Another mechanism is efflux. It allows the antibiotic to come in, but it pumps it out, and carbapenems are particularly susceptible to that. One of the main mechanisms of resistance is the fact that the 

 has this fluffy additional external membrane that gram-negatives have, and it doesn’t allow antibiotics to go in by downregulating the genes that encode for those channel proteins. The carbapenems are very susceptible to those.

Pseudomonas Infections Continue to Acquire Resistance, Complicating Treatment

Peter Salgo, MD; Yoav Golan, MD; Andrew Shorr, MD; and Marin Hristos Kollef, MD, provide their perspective on the cephalosporin/β-lactamase inhibitor combinations Zerbaxa, Avycaz, and Vabomere, with considerations for mechanisms of action and effectiveness.

We have some new antibiotics. These are cephalosporin/β-lactamase inhibitors. We have 3 new ones. We have Zerbaxa, Avycaz, and Vabomere. Where do we go with these things? Can you run them down?

 All those antibiotics are a combination of some kind of a β-lactam and a β-lactamase inhibitor. The β-lactamase inhibitor will be a bodyguard. So, you give an antibiotic, and the bodyguard, knowing that the bacteria are going to try to break through, is going to take care of that. There are 2 different approaches. One approach is to take an antibiotic that has been very useful for many, many years—like a carbapenem or ceftazidime that we’ve been using&mdash;where a resistance to it has developed over the past several years, and attach to it a bodyguard that will prevent the enzyme and reinstitute the susceptibility to this antibiotic.

Avycaz is such an antibiotic, and studies with Avycaz have shown that by adding avibactam to ceftazidime, you will regain susceptibility against ceftazidime-resistant bacteria that would kill animals and humans as well. It would be susceptible to Avycaz.

Zerbaxa has a different approach, in which the cephalosporin β-lactam component is actually a new component. It’s a new cephalosporin that is particularly good at the ability to truly get through the outer membrane of 

 and resist some of the activity of some of the efflux and so forth. It’s given with an older β-lactamase inhibitor, tazobactam, to allow the antibiotic to resist those β-lactamases.

Vabomere, which is the newest addition, is a combination of meropenem and vaborbactam, vaborbactam being a very, very active β-lactamase inhibitor against those empiric bacteria that produce those ESBL type-A β-lactamases. It’s not that powerful against some of what 

 is producing. So, in many circumstances, Vabomere is actually not going to increase your ability to treat much more than just using ertapenem.

 I think it’s important to realize, though, that besides talking about the in vitro activity of all these different agents, they all have very different clinical data sets that are currently available to us. For example, Avycaz, which has been around the longest, has just completed or just presented its nosocomial pneumonia data set. Marin was one of the lead authors on that, and I think there are issues with that data set because the mortality rate in that population was remarkably low. I think there are a lot of questions about the validity of whether it was actually studying pneumonia.

He could just be remarkably good, you know.

 Oh, he is remarkably good. I don’t deny that. But just not as good as you. Then you have CT, which has some interesting clinical data but no pneumonia data.

 Correct. Zerbaxa is currently undergoing its pneumonia trials. I think it’s important to realize that the dose they’re studying for pneumonia is actually twice the dose that’s currently approved. So, you never want people arguing from one disease state to another without understanding the differences and the pharmacokinetics. And then, when you look at Vabomere, there’s no pneumonia clinical trial but a critically ill population, a highly drug-resistant population, in a study called TANGO-2 on top of their urinary study, also with a unique population. Although the boronic β-lactamase inhibitor doesn’t add anything to its antipseudomonal coverage, the dosing of the meropenem to optimize the pharmacokinetics, which you could do today if you wanted to, doesn’t require that you buy Vabomere. The dose of meropenem is huge and given over a 3-hour infusion 3 times a day to really pick off those higher MDR pseudomonal isolates. Again, it’s optimizing the principle that you could do it today without having to buy their molecule.

Battling Against Resistant Pseudomonas Infections in the ICU

Pseudomonas Infections in the ICU: Who's Susceptible?

Pseudomonas Infections: Matching Causes and Prevention

Pseudomonas Infections in the ICU Versus on the Floor

Pseudomonas Infections: Nuancing the Use of Antibiotics

Fighting Pseudomonas: Impact On Patient Prognosis & Approaches to Combat Infection

Classes and Agents Available to Treat Pseudomonas Infections

Pseudomonas Infections Continue to Acquire Resistance, Complicating Treatment

Exploring Cephalosporin/β-Lactamase Inhibitor Combinations for Treatment of Pseudomonas

Exploring Appropriate Treatment Protocols for Pseudomonas Infections

Core Challenges in Pseudomonas Treatment Protocol

Pseudomonas Infections: Optimizing Diagnostic Strategies

Pseudomonas Infections: Selecting Upfront Antibiotics

Pseudomonas Infections: Secondary Approach to Treatment

Treating Pseudomonas Infections With Zerbaxa or Avycaz

Zerbaxa and Avycaz: Considerations for Treatment

The Future Treatment of Pseudomonas Infections