Thomas Lodise, PharmD, PhD, discusses the need for appropriate initial therapy, how to identify patients at increased risk of multidrug resistance, and the mechanisms that lead to multidrug resistance.
Thomas Lodise, PharmD, PhD: So, when we think about multidrug-resistant pathogens, certain groups rise to the top of who’s at greatest risk. And when I think about patients at risk of multidrug-resistant infections, I consider patients who come to the hospital with an infection and those who develop their infection during their hospitalization. When we think about people coming to the hospital with multidrug-resistant infections, really, the patient population I think about most is our health care frequent flyers—those coming from long-term care facilities, dialysis patients, immunosuppressed patients, patients with extensive prior antibiotic histories. It’s in those patients that we have a heightened awareness for a likelihood of resistant pathogen and also ensure our initial antibiotics have high likelihood of covering those pathogens.
Similarly, when we think about hospital-onset infections, the group at greatest risk of multidrug-resistant infections is those who have accumulated a number of prior antibiotic exposures. And what we find is that as people accumulate more and more antibiotics, they also increase their likelihood of developing a subsequent multidrug-resistant infection. So, again, on both sides of the equation, coming to the hospital and developing during a hospital setting, we really think about time at health care facilities and time at risk, which is exacerbated by the receipt of prior antimicrobial drugs.
When thinking about antibiotic-resistant pathogens, it’s important to consider different classifications. Terms are used interchangeably within the literature, but the 3 most common terms are “pan-resistant,” “multidrug-resistant,” and “extensively drug-resistant.” When we consider the term “pan-resistant,” what we’re talking about is bacteria that’s resistant to all commonly used agents. A good example of this would be Pseudomonas, where all our commercial available β-lactams, aminoglycosides, and fluoroquinolones have no activity against that bug. “Multidrug-resistant,” depending on the pathogen and your organization defining it, usually defines bacteria that are resistant to 3 or more commonly used classes.
So, again, if we would think about a multidrug-resistant Pseudomonas aeruginosa, here we’re talking about resistance to the fluoroquinolones, aminoglycosides, and perhaps advanced-generation cephalosporins. Another term that’s often used is “extensively drug-resistant.” Here, we’re talking about resistance, not quite pan-resistant but a level above multidrug-resistant. So, here, when we think about Pseudomonas aeruginosas, a working example, we’re talking about resistance to 4 or 5 commonly used classes. “Extensively drug-resistant” is often applied to Acinetobacter, given the high levels of resistance observed within that pathogen.
When we think about identifying multidrug-resistant pathogens, I think the first place we should look for data is from surveillance programs. So, surveillance programs—these are national programs where they have sentinel hospitals throughout the United States, and what these sentinel sites do is provide a central laboratory a certain number of isolates for susceptibility testing. So, here, this is a good benchmark as to what are the resistant pathogens within your area and what changes have occurred from prior years. A good example of the importance of large-scale surveillance programs is what we’re seeing with carbapenem-resistant Klebsiella pneumoniae. Overall, we have a low rate within the United States, but we do see pockets that do have higher rates of carbapenem-resistant Klebsiella.
In addition to looking at national surveillance data, it’s always important to see what’s going on at your institution. A lot of these surveillance programs typically lag a few years behind real-time data. So, for probably one of the best glimpses of multidrug-resistant pathogens that may be problematic in your institution, one needs to look no further than what’s going on at their own hospital with their local antibiogram. As an example of this, this enabled us to identify CRE early at our institutions years prior. It just became reported within the national literature, and also, we’ve been able to recently identify some resistant phenotypes within Acinetobacter, as well. So, again, it is not only important to look at surveillance data but also to look at data within one’s institution.
When we think about multidrug-resistant bacteria, often what we’re talking about is the presence of an organism that has different resistance mechanisms. And these different resistance mechanisms confer resistance at different classes of antimicrobials. Probably the ones that we’re most concerned about are β-lactamases. So, when I think about β-lactamases, although there are different classification schemes, I really put them in 3 different categories: your broad-spectrum β-lactamases; your extended-spectrum β-lactamases, or ESBLS; and your carbapenem-resistant β-lactamases. So, when we think about broad-spectrum β-lactamases, what we find here is that these β-lactamases hydrolyze first- and second-generation cephalosporins and some of your commonly used penicillins. To date or in current times, most Enterobacteriaceae and other problematic gram-negative bacteria possess a broad-spectrum β-lactamase.
ESBLs are extended-spectrum β-lactamases. What we have here are β-lactamases that actually hydrolyze third-generation cephalosporins, aztreonams, and they may or may not be inhibited by clavulanic acid or other commonly used β-lactamase inhibitors. So, here, these are very problematic organisms. Depending where you are, about 10% of all Enterobacteriaceae express in ESBL. We find the highest rates in Klebsiella pneumoniae, followed by E. coli, but they can be present in other organisms like Enterobacteria, as well. For your carbapenem-resistant β-lactamases, these are the most problematic and really come in 2 varieties: your CRE- or KPC-producing β-lactamases, which are commonly found in Klebsiella pneumoniae and really the 1 that’s problematic within the United States.
Another emergent phenotype that is seen worldwide but becoming more prevalent in the United States is your metallo-β-lactamases (MBLs). This is the NDM I think we’re all familiar with, this is the New Delhi β-lactamase‑very concerning, because a lot of our new β-lactamase inhibitors brought to market have no activity against gram-negative bacteria expressing MBL-producing β-lactamases.
An important part of identifying multidrug-resistant pathogens is to look at the data that are brought forth from our clinical microbiology labs. Initially, what often comes up is, are we accurately capturing resistance within our institutions? So, in the past, there was a sense that we were probably underestimating CREs due to the susceptibility of break points being too high. Since then, they have been revised. I think we’re at a point now where we’re identifying CRE at an appropriate level.
One resistant pathogen I don’t think we’re identifying well is when we observe colistin resistance. You may or may not be aware that with the Enterobacteriaceae, we do not have a defined breakpoint. We infer it from other pathogens. And also, too, when we do colistin testing, there are very specific methods that are now recommended, and not all institutions perform this. So, the 1 antibiotic where I feel a bit hesitant relying on susceptible data is colistin. So, again, I think we all need to look carefully, ask our microbiology lab how they’re doing susceptibility testing, and make sure that they’re following the best methods.