<i>Pseudomonas</i> Infections: Optimizing Diagnostic Strategies
Peter L. Salgo, MD; Marin Hristos Kollef, MD; Andrew Shorr, MD; Yoav Golan, MD; and Jason Pogue, PharmD, BCPS-AQID, provide considerations for the optimal approaches to diagnosing Pseudomonas infections and identify where imaging and rapid diagnostics fit into the paradigm.
Peter L. Salgo, MD: What are the best imaging and other laboratory studies that you want in terms of knowing what’s going on and prescribing the right antibiotics? What’s the best stuff?
Marin Hristos Kollef, MD: I think the best thing is having a seasoned clinician who can look at the patient and make a decision in regard to whether or not they really think that person’s infected.
Peter L. Salgo, MD: Did I just hear you say that a doctor is a good thing to have?
Marin Hristos Kollef, MD: I think a doctor is a good thing, but it depends in part on the doctor and the pressures that they’re facing. We can talk about tests and how in patients who are on a ventilator if we get a BAL [bronchoalveolar lavage], it might be better than a non-BAL specimen. A quantitative culture might be better in terms of helping with decisions regarding treatment. But I think that too often, what ends up happening in the average ICU [intensive care unit] is that the trigger is a very simple one to pull on antibiotics. It just gets done frequently, and because we have certain drugs that are not restricted, they are the ones that get thrown in there. And then, we run into what Andy was referring to: It may not be the right drug for that patient, they have a drug-resistant organism, and now you’re not treating them correctly and their risk of death is higher.
Peter L. Salgo, MD: Here’s what comes back, right? You have a patient with pneumonia that may or may not be resistant, that may or may not be Pseudomonas, so you send off a BAL and you send off a culture. What is the first thing you’re going to get back? You’re going to get a gram stain. How long does the gram stain take in your hospital?
Andrew Shorr, MD: Too long.
Peter L. Salgo, MD: I know that, but how long is that?
Yoav Golan, MD: Well, I don’t know. We made a little progress with that. We relied on cultures and gram stains for a very long period of time. We realize that real-time diagnostics are really important. We just highlighted the fact that it’s really hard to guess, and if you guess one resistance and get another resistance, the patient may be paying the price. You need more accurate guidance early. And so, we are using more and more PCR [polymerase chain reaction]-based assays and other technologies that do not rely on cultures. But with gram stains, I would take it 1 step backward. First of all, it’s important to have the right specimen. It’s important to have a system: how often you see the patients taking antibiotics without any culture being taken. You talk about being aggressive with antibiotics in a patient who would benefit from early aggressive antibiotic management, but I would say you should be just as aggressive in collecting information that will allow you to correct or de-escalate therapy.
Peter L. Salgo, MD: I’m sorry, did you just say people are on antibiotics without cultures?
Andrew Shorr, MD: Every day in this country.
Peter L. Salgo, MD: Where?
Andrew Shorr, MD: All over.
Marin Hristos Kollef, MD: All over, yes.
Jason Pogue, PharmD, BCPS-AQID: Most of the patients.
Peter L. Salgo, MD: Pardon me, is that just on this side of insane, or am I wrong?
Andrew Shorr, MD: It’s appalling.
Peter L. Salgo, MD: Appalling is on this side of insane; maybe it’s on the other side of insane.
Andrew Shorr, MD: But it has happened. People routinely pull the trigger on antibiotics without thinking about what tools are necessary to actually de-escalate or stop therapy.
Jason Pogue, PharmD, BCPS-AQID: It’s also escalate, too.
Peter L. Salgo, MD: I didn’t realize we’d have to go back to step 1. Number 1, if you’re going to start antibiotics, get a culture first if you can. Number 2, you’re going to get your gram stain, right? Number 3, you want your PCRs. How quickly do the PCRs come back?
Yoav Golan, MD: We use a respiratory panel that mostly has respiratory viruses and a few bacteria, but not the bacteria types of Pseudomonas and others. I guess that we will get into that in the future. We do use PCRs for blood cultures, for example. And for those, we get the results usually a few hours after incubation.
Peter L. Salgo, MD: A few hours?
Yoav Golan, MD: A few hours after the start of incubation. It’s not immediate. But we used to have traditional first-line methods, which were empiric, and then we had culture-guided or gram-stain guided therapy, so I think that this period of time is getting shorter and shorter. The point is that you talk about the fact that many patients don’t get cultures. Many hospitals don’t have molecular diagnostics as well. If you look at the CDC [Centers for Disease Control and Prevention] recommendations for stewardship, one of the points that they make is that if you want to use fewer antibiotics and more respective antibiotics and the right antibiotics, you need to know what to treat and you need to know what to treat early and better use them. But I must say that this is an exciting period of time because for many, many years, since Pasteur or even before him, we’ve been relying on cultures. Now we have better diagnostics and we know that there are downsides to those rapid diagnostics as well because, as you know, PCRs can be too sensitive. As Marin said, at the end of the day it’s the ability of the physician to interpret the results within the context of the patient. It’s always the context of the patient that should determine what cultures and what tests to send and how you interpret them and how aggressive you are.
Peter L. Salgo, MD: So, you get the PCRs back within a few hours. How long does it typically take to get a good culture back? In other words, we sent the cultures and I now know this is Pseudomonas—how long does that take?
Yoav Golan, MD: I must say, again to this point, the time to culture has not changed. When you rely on bacteria to divide, this has not changed in a long period of time. The first thing that you’re going to get is the gram stain. Maybe I shouldn’t say that in this forum.
Peter L. Salgo, MD: Go ahead, nobody’s listening, nobody’s watching.
Andrew Shorr, MD: Did you go to the microbiology laboratory?
Peter L. Salgo, MD: You do that, don’t you?
Yoav Golan, MD: Oh, I go to the microbiology laboratory because the microbiology laboratory says that it is gram negative in the sputum or gram negative in the blood; but when you talk to the microbiology laboratory, they’ll give you a description of the microbe. If you ask them what they think it is, they’ll tell you, “I think it is this,” but because it is a guess, they’re not going to put it down.
Peter L. Salgo, MD: Right. I’ve seen sheets of stuff.
Yoav Golan, MD: Very often, they know early, sometimes a day before the culture results. If the gram-negative shows coliform that should be treated with carbapenem or if it’s Pseudomonas aeruginosa that should be treated with third-generation cephalosporin or what have you, they can give you this information very early. There is less and less communication with the microbiology laboratory. I would say, use your microbiology laboratory. Talk to your microbiologist and even if it’s offsite, give them a call. They’ll be happy to know that they can help you treat the patient. They’ll be happy to know that there’s actually value to what they do.
Andrew Shorr, MD: I think the other point to realize is that knowing the pathogen doesn’t give you a crucial piece of information, which is the actual susceptibility data. One of the flaws or one of the limitations of some of the rapid diagnostics that are coming online right now—not all, but some—is that they can tell you Pseudomonas is present, but they don’t tell you what it’s susceptible to. You’re still guessing. And you can go with your best guess in terms of, “Well, 80% of my Pseudomonas cases are susceptible to this, so I’m going to go with this.” But that’s still a wild guess. Conversely, with the rapid diagnostics that will eventually get to susceptibility, I’d rather know what the organism is susceptible to, not necessarily what it’s called: Pseudomonas, Enterobacteriaceae, or whatever.
But the other issue with the numbers of rapid diagnostics that we’ve alluded to is if all you do is send cultures when they’re not indicated on colonized and uninfected patients and they find something, you still have to interpret whether it’s a colonized or an actual pathogen. There’s a lot of hope that rapid diagnostics or this technology will engineer us out of the clinical judgment piece. I actually think it’s going to make the clinical judgment piece more challenging.
