false
Catalog
EP Fellows Curriculum: Cardiac Implantable Electro ...
EP Fellows Curriculum: Cardiac Implantable Electro ...
EP Fellows Curriculum: Cardiac Implantable Electronic Device Infections
Back to course
[Please upgrade your browser to play this video content]
Video Transcription
Thank you. Good morning. It's a great pleasure to be here. There are only a few bright spots in this whole pandemic, and programs like this are one of them, and so I really want to thank Dr. Verma for including me in the program. So I'm going to be talking about CIED infections today. This is maybe not as glamorous a topic as things like intracardiac EGMs and maneuvers for diagnosing SVT, but unfortunately it is a really important topic. Device infections have been a problem for years. They are continuing to be a problem, and unfortunately for all of the trainees, they are going to continue to be a problem for you guys in the future. Those are my disclosures. So what's the scope of the problem? So unfortunately, there are a lot of patients who do get infections every year in the United States and around the world. So right now, over a quarter of a million devices are implanted annually in the United States. In some years, if you look at the data, it's probably closer to half a million. The infection rates are a little bit hard to drill down, and I'm going to talk more in detail about this, but in general, we think that for new implants, the infection rate is probably somewhere around one to two percent. The rate of infection goes up for subsequent procedures, and it could be possibly as high as five percent in those patients. When we say subsequent procedures, we're talking about things like generator changes, early re-accessing of pockets for lead dislodgements or micro-perforations, or of course upgrades, particularly upgrades to CRT devices. One thing that seems pretty clear is that the infection rates are rising out of proportion to increase in implants. The figures on this slide are from a paper that was published back in 2011, so it's getting a little bit of age on it now, but the authors of this study looked at the National Inpatient Sample Database to try to get at the epidemiology of device infections. So the top figure is showing the rate of infections, and you can see that they started tracking in the early 1990s, and by around 2004, we see this sudden uptick in the rate of CIED infections, and by 2008, the rate of infection was upwards of 2.4 percent. So why is this? Well, it's probably multifactorial, but a lot of the reason is probably shown on the bottom figure, and this is showing the incidence of four very common comorbidities, renal failure, heart failure, diabetes, and respiratory failure, and not surprisingly, you can see somewhere around 2004-2005, we have a sudden uptick in those as well, and so it's pretty clear that we are putting pacers and defibrillators in patients who are older and sicker, and because of that, that is part of the reason why we're probably seeing this increased rate of infection. This is from a subsequent paper that was also looking at that same database and sort of expanded that data out to 2012, and concerningly, what the authors found was that by 2012, in this same database, the infection rate had climbed to 3.41 percent, so even higher. They also found what's shown on this figure, which is that if you just look at total device procedures, the percentage of the procedures that were performed specifically because the patient had a device infection was climbing. So on this graph, the blue line is the total device procedures, and the red line is the percentage of them that were being performed because specifically the patient had an infection. So again, we're putting in a lot of these devices, and the rates of infection seem to be climbing. So this is a paper that was sort of hot off the presses. This was just published this year in HeartRhythm, and this was specifically looking at the epidemiology of device infections in the U.S., and they used the Healthcare Cost and Utilization Project database, and they specifically looked at the year 2016, and what they found was over 191,000 implants, but there were 8,060 device infections, so that's 4.2 percent, so even higher. And one thing that's sort of interesting when you try to look at these numbers is there's a disparity in the rate of infection between these sorts of studies that are looking at a real-world population and are randomized controlled trials. So if you look at things like the RAPID trial, which was the Tyrex pouch trial, their control arm had an infection rate of 1.2 percent, and so when we talk about device infections, we all tend to sort of throw out those sorts of numbers, right? The rate of infection is, you know, 1 to 2 percent, and if you look again in these kinds of randomized controlled trials, that's what you see. But if you look at these more real-world samples, we start to see these higher numbers, and so the question is in the randomized controlled trials, is this sort of a rarefied population where high-risk patients are excluded and there's, you know, special attention to detail, and so the rate looks lower than what it is in reality? Of course, these sorts of database studies are subject to error as well, but again, you know, I think we have to accept that probably the rate of infection is a little bit higher than what we would like it to be. In this study, the average length of stay of patients got infections was 14 days, and the average hospital cost incrementally was over $50,000. I think it would come as a surprise to no one that more comorbidities were associated with an increased length of stay and worsened outcomes, and in this study, they had an interesting gender finding, and that was that women accounted for 43 percent of implants but only 30.8 percent of infections, and of course, this sort of work isn't able to tell us why that is, whether women are actually less likely to get infected or if there's some sort of, you know, failure to diagnose. So CIED infections are really, really expensive, so this is, again, from the original study that I was showing you. This is the in-hospital charges with adjusted to inflation for the year 2009, and so they're very expensive, and then also the cost in dealing with them was climbing throughout the study period. This is from another study that was trying to determine the incremental additional cost for a patient who received a device and then it got infected, and again, what you see is whether it's an initial implant or, you know, a generator change, you know, upwards of $50,000 incremental cost is added to caring for the patient. So I apologize for this super busy slide. I promise I won't read it to you. This is from a really nice, again, very recently published paper. This is the European Heart Rhythm Association International Consensus Document on How to Prevent, Diagnose, and Treat CIED Infections, and this was endorsed by HRS and a number of other societies, and if you haven't had a chance to look at it, it was simultaneously published in several journals earlier this year, and I would recommend that you check it out, but this is an interesting table. They basically took all of the studies they could find and tried to pool them to determine how important various risk factors were for the chance of patients getting infected, and they broke it down into patient related factors, procedure related factors, and then device related factors. I don't think any of the patient related factors would surprise anyone. Obviously, renal failure, history of prior infections, you know, diabetes, those are all pretty powerful markers for a patient's risk of becoming infected, but I put a little star next to the procedure related factors because I think particularly for this group who are, you know, mostly trainees, you know, we think about sterile technique, but there's some other procedure related factors which matter a great deal for a patient's risk of getting infected. Procedure duration is actually a pretty powerful marker. Hematoma formation, bacteria, you know, love old blood, and so hematoma formation is a pretty profound marker. Patients needing lead repositioning and inexperienced operators. Now, when we say inexperienced operators, we don't mean there's trainees. We're talking about the attending physician, but the point of the matter is, you know, being able to perform these procedures efficiently and skillfully really does matter as far as patients outcomes and risk of having infections, and so as you're going through your training, you really want to not just learn to put in pacemakers but become a really, you know, skilled operator. So, you know, when we talk about the actual microbes involved in device infections, what are we talking about? In the vast majority of cases, we're really talking about Staphylococcal species. They account for 60 to 80 percent of CIED infections, so we're talking about methicillin-sensitive Staph aureus, methicillin-resistant Staph aureus, and then coag-negative Staphylococci, and coag-negative Staph are also highly resistant to many antibiotics, and so if you're in a situation where you're having to empirically treat a patient for an infection, you don't know what the bacteria is, you're kind of obligated to include vancomycin simply because there's a high likelihood that whatever it is is going to be resistant. Other important organisms, Enterococcus, Streptococcus, Corynebacterium, and of course the dreaded fungal infection. So, when we talk about device infections, they sort of come in two, you know, obviously related but slightly different flavors, so we have pocket infections and then we have lead infections. So, pocket infections are usually due to direct procedural contamination, and so it's more likely to occur in relation to a device procedure. That is not to say that they necessarily have to, you know, develop the inflamed pocket, the hot pocket, immediately after the device procedure or even within a few weeks to months. Many of the bacteria that cause these pocket infections are skin flora, so they're kind of indolent, and so you can see pocket infections blow up a year or even more after a device procedure. Now, lead infections, in contrast, are usually due to hematogenous seeding. So, what happens is the patient gets septic for some other reason and then because, you know, many of these bacteria do like hardware, it will basically seed the lead. If you think about it, leads represent a huge amount of hardware in the blood pool, and they don't completely endothelialize the way things like, you know, cardiac stents do, and because of that, really, even if these devices have been in for years and years and years, the patient can get a lead infection at any point if they become bacteremic. So, even 10 years later, after a device implant, if they become septic, those leads are at risk for getting infected. So, part of the issue in treating device infections, or really any hardware infection, is that many of these bacteria end up in what we call the biofilm. We don't talk about the biofilm very much, but it's critically important because it defines how we have to manage these infections. So, the biofilm is an extracellular matrix that helps bacteria adhere to devices and then protects them from antibiotics and patient immune factors. And so, the classic scenario is a patient gets a pocket infection, they go to see their primary care physician, they get put on two weeks of Keflex, everything starts to look better, they think it's all healing up, they complete the course of antibiotics, and then a week after that, the pocket starts looking hot again. And then again, that's because of this biofilm, there's still bacteria that are existing on the device. In addition, Staphylococci have adhesive polysaccharide capsules, and so all of these things, again, make it very, very difficult to clear infections if it's actually on the hardware. So, if you look through the literature, there are tons of studies, mostly small ones, that show failure of conservative management of device infections. If you look in the literature, you can find case reports, small case series of patients who were coaxed through with conservative management. But in general, the data clearly shows the conservative management of device infections just doesn't work. So, this is just one of many studies that are showing this. This was a group of 48 patients who either were thought to not be candidates for extraction of their system, or the patients declined it. And, you know, this is showing survival curve. But overall, basically, 18% of the patients treated with chronic suppressive antibiotics had a relapse of sepsis within a year. And then a third of those, they ended up having to try to take the device out, you know, even if they weren't thought to be good candidates, because they simply couldn't clear the cultures. So, conservative management does not really work, and that's certainly reflected in the guidelines. So, diagnosis. So, how do we diagnose these? So, pocket infections tend to be very straightforward. The patients have erythema, warmth, and fluctuance. Erosions, including dry erosions, are by definition infected. So, even if it, you know, there's no pus coming out, even if it just looks like the, you know, lead or the generator eroded through, if it's communicating with the wider world, it is considered to be infected. And if the generator's infected, the leads are assumed to be as well, right? It's all connected. And so, occasionally, people will try to do things like take out a generator and do a generator change for a pocket infection, and this almost always fails. So, if, you know, even if it's just a pocket infection, ideally, the leads are going to need to come out as well. Now, lead infections can be a little bit less straightforward. Sometimes, there'll be vegetations on imaging, which can be helpful in a septic patient. But sometimes, it's just relapsing sepsis without a clear source. And so, just because vegetations are not seen on imaging does not necessarily mean the leads are not infected. And again, this goes back to that whole biofilm issue that you may have just, you know, colonization of that lead. And, you know, patients will be treated with a course of antibiotics and then relapse after the antibiotics are discontinued. As far as ancillary testing for diagnosis, repeat blood cultures can be helpful. If somebody, you know, is not able to clear their cultures for, you know, five days or so, even if the leads were not the original source of infection, they're almost certainly infected now. Obviously, transesophageal echocardiography can be helpful. And one of the emerging technologies, which is really probably going to become, you know, a part of our diagnosing of these cases, is PET imaging. It seems to have a pretty high sensitivity and specificity for diagnosis. And the one thing many of us are struggling right now is it's not a covered indication for most insurance companies. So, these are some nice images from a recent paper looking at device infections. And I think this really shows you how PET imaging can be helpful. So, if you look in figure A, you can see, you know, there's all kinds of areas lighting up on this PET scan. So, you can see the pocket area is lighting up, the lead is lighting up, and this patient actually has a septic emboli, which is sort of the extremely large area there in the lung. If you look at B, this is a patient who has a lead infection. So, you can see the lead is lighting up. Panel C, you can see, is a pocket infection, and you can see the pocket area lighting up. And then, panel D also shows how PET can be really useful in trying to diagnose and manage these patients. This panel D was a patient who had what was thought to be a vegetation on echo imaging, and there's an inset there where you can see what they were looking at. But this was a patient who, you know, didn't seem to be septic and seemed to be doing okay, and so they did a PET scan. And many of you know from looking at lots and lots of echo images that many patients who have pacemakers and defibrillators do have this fibrinous stranding present on their leads, and it doesn't necessarily mean it's infection. But if there is any question, you know, a PET scan can help delineate this for you. So, I'm not going to read the guidelines to you because, you know, you'll all go to sleep on me, but the most recent Heart Rhythm Society guidelines are from 2017. And just to highlight a couple important points, obviously, it is a class one indication to give a complete course of antibiotics, you know, and ideally tailored to whatever the bacteria is. So, for a pocket infection, we're talking about a two-week antibiotic course. If we have lead infections, particularly if the patient actually has valvular vegetations, you're going to be talking about four to six weeks of antibiotics. Another class one indication, complete device and lead removal is recommended for all patients with definite infection. So, that's either a pocket infection or lead infection. It is a class one indication for that system to come out. I will also point out that in the subtext for this, you can see it says early diagnosis of CIED infection and performing lead extraction within three days of diagnosis is associated with a lower in-hospital mortality. And this is based upon several studies. And here is just one of them. So, here the authors had 52 patients. Some were pocket infections and some were lead infections. And what they found was that lead extraction performed greater than three days after admission was statistically correlated with in-hospital mortality in all 52 patients. And that was, again, statistically significant. Now, when they broke it apart for pocket infections versus lead infections, it certainly looks like that's driven mostly by the lead infections as opposed to the pocket infections. But there have been other studies, as I said, who have showed similar findings. And so, this has certainly made it into the guidelines. I'm not doing a lead extraction talk per se here. I'm talking about device infections. But as somebody who does do extractions, I can tell you if you're doing high-risk extraction and you're trying to, you know, have surgical backup available and blood products and all of that, trying to pull that together for a hospitalized patient within three days, particularly when you start involving weekends and holidays, can be a challenge. But the data is pretty clear that if someone has a device infection, getting it out earlier is probably going to be better for the patients. So, another class one indication from the HRS guidelines, which I think is worth noting, is that complete device and lead removal is recommended for patients with persistent or recurrent bacteremia or fungimia, despite appropriate antibiotic therapy and no other identifiable source for relapsed or continued infection. And when these guidelines came out, this guideline, which again is a class one guideline, caused a little bit of consternation. You know, what is persistent? What is recurrent? You know, how long do blood cultures need to stay positive before we say it's persistent? Now, in general, most people are going to say three days, but not everybody does. Some of my ID colleagues say 48 hours, others say five days. And so, in general, I think people use three days. But, you know, again, what is persistent? And then what is recurrent, right? I mean, sometimes that's obvious, right? If somebody comes back with recurrent staph sepsis with no identifiable source, you know, within a month, you know, I think we would all agree that's recurrent. But, you know, what if it's two years? Do we still think that that's the device? And do we think that it needs to definitely come out? But there is a reason that this guideline made it into the literature, and that is because staph aureus bacteremia, staph aureus really does love hardware. And even if the device isn't the original source of an infection, it's really common for them to become infected if the patient is septic for some other reason. So, this is a small study, and this was done by the Mayo Group. And basically, what they did was they just retrospectively looked for all patients who had staph aureus and also had a device in place. And they found 62 patients. And basically, they found that over a third of those patients did have device involvement, whether or not the device was the original problem or not. Ultimately, over a third of them had device involvement. And when it was the ICD patients, it was closer to two-thirds of those patients had device and clear device involvement. And so, again, staph aureus, you know, the risk of infecting the device if somebody becomes septic for some other reason is just so high that when somebody has staph sepsis, you always have to think about the device and is the device involved. This is, again, I already mentioned that the European Heart Rhythm Association put together a nice consensus document that was just published this year. Again, I'm not going to go through all of that and read it to you. But, you know, one thing to highlight is there's increased emphasis since the Heart Rhythm Society guidelines on these novel imaging techniques. For the European guidelines, they use, you know, a green heart for class 1 indication, a yellow heart for class 2, and a red heart for class 3. And what you can see is they've even brought intracardiac echocardiography into this. So, they give a class 2 indication to using ice if you're concerned that a patient might have a device infection and the TEE was not revealing. And then they give a class 1 indication for using either PET scanning or tagged white blood cell scanning or maybe even contrast CT to try to identify infections if your TEE was not revealing. Another important thing, going back to what I was just talking about, is they actually give a class 1 indication to use a PET scan for a patient who has staph aureus bacteremia. And again, this is just this reflection that patients who get staph sepsis and have these devices in place are just at that higher risk of developing infection. And then they also point out that you can use things like PET and tagged white blood cell scanning for, you know, patients who have, you know, embolization and where would those, you know, septic emboli be located. All right. So, how are we going to manage these infections? And again, there's sort of pocket infections and then there's lead infections. And the management of those two things is going to be slightly different. So, these are some figures that I've taken from a paper from the Mayo Group and I sort of liked how they laid it out. And it's very similar to what you find in the guideline statements as well. So, basically for a pocket infection. So, you've got pocket findings without systemic symptoms. All right. So, you have the patient come in. They've got the hot pocket. They don't seem to be febrile or septic. So, the first question to ask, especially if it is a fresh implant is, is the deep pocket actually infected? So, after an initial implant, people can get some superficial, you know, stitch abscesses and things of that nature. If that's the case, you might not necessarily need to take out the device. And so, what we recommend is that you do a couple weeks of antibiotics and retain the device. I do recommend if you do this that you watch the patient really closely and make sure that there doesn't become evidence of a deep pocket infection. Okay. But if it's pretty clear that this is a deep pocket infection, you're going to get blood cultures, right, because you don't want to miss a lead infection. Now, if your blood cultures become a positive, now we're going down the, you know, lead infection pathway. So, that'll be on the next slide. But if they're negative and you really feel like this is just a deep pocket infection, you're going to take out the device. You're probably going to make a good effort to try to culture it and see if you can get a bacterial diagnosis. One of the challenges for pocket infections is that people tend to put patients on antibiotics pretty quickly after identifying that they have a pocket infection. And of course, if they have negative blood cultures, then you don't have a bacteria to, you know, to tailor your antibiotic therapy for. So, it's important when you take those out to, as much as you can, try to to culture, you know, the pocket and the hardware in order to get that bacterial diagnosis so you can tailor the antibiotic therapy. Once your pocket is controlled, then you're going to re-implant probably on the contralateral side. And then, you know, a couple weeks of antibiotics is generally going to be considered to be sufficient for, you know, an isolated pocket infection. So, lead infections are understandably a little bit more complicated. So, I put in parentheses here management of a bacteremic patient who has a CIED in place because that's actually what we're talking about. Because again, you know, trying to identify if a lead is actually infected can be a little bit more challenging. So, starting at the top of the flow chart, so we're concerned the patient might have a device infection. So, the first question is, what's the bacteremic, you know, what's the bacteria, you know, that's causing the sepsis? So, first of all, if it's a gram-negative bacteria, those don't tend to affect devices. So, if that's the case, then, you know, just try to figure out where the source is and treat the patient. But if it is one of our bacteria that likes hardware, so again, Staph aureus, Coag-negative Staph, Enterococcus fungus, something like that, your next step is going to be a TEE. Now, if your TEE is positive, then we're going to say, okay, you've got vegetations on the leads, it's infected, we're going to have to extract this device. So, we're going to extract the device and then they break it out by is there valvular involvement or not. So, if you have a lead infection and the valves don't seem to be involved, then they say two to four weeks of antibiotics. Personally, I feel like if, you know, there's something in someone's heart that's infected, I generally go with the four to six weeks of antibiotics. But then, it's probably okay to re-implant after several days of negative blood cultures. Then, they say, but if there is valvular involvement, and, you know, certainly if you have tricuspid valve involvement, then we're looking at four to six weeks of antibiotics, and you probably do want to wait a little bit longer to re-implant. So, what if we have our bacteremic patient, but the TEE is negative? Well, then we start to look at the length of time that they've been bacteremic, and if we have an alternative source to account for this. And they say, if you've been bacteremic for less than 72 hours, and there's a clear alternative source, it's probably reasonable to retain the device, treat the alternative source. But again, I would emphasize that you really, really want to try to follow these patients very closely and make sure that they don't become, you know, recurrently bacteremic. But if they've been bacteremic for more than three days, or there's no clear alternative source, then we bring in the PET scan, right, and we try to see if the PET scan can demonstrate any infection. If it does, you're going to want to take it out. If it doesn't, then they say, you know, retain, treat, but then you're probably, again, going to want to be really cautious and do surveillance blood cultures and just make sure that, you know, your patient is safe. So, I mentioned already that it's really important, especially for the pocket infections, to really culture, you know, the hardware well. For the lead infections in the septic patient, you're usually already going to have a bacterial diagnosis. But again, for the pocket infections, the patients often have been started on antibiotics, and we don't have a bacterial diagnosis. And so, you really want to try to get that when you're doing the extraction. So, the most common thing that people do tends to be pocket swaps. So, you know, those little Q-tips that you put in the pocket and then, you know, put in the sterile container and send down to the lab. Unfortunately, the reality of the matter is the sensitivity on those is really pretty poor. The specificity isn't bad, but the sensitivity isn't great. Tissue culture is better. So, if you can actually get a piece of the, you know, fibrous capsule, particularly in an area that hasn't been, you know, you know, subject to electrocautery or something like that, that's going to have a better sensitivity. And then, you also can send sonicated fluid. So, sonicated fluid, basically what you're doing is you're taking usually the generator, putting it into some sterile fluid, and then you put it in a sonicator. And they do make medical sonicators, but you could also essentially use a jewelry cleaner, which are all sonicating devices. And basically, the idea is the sonication then dislodges the biofilm. And so, you know, basically in this sense, we're using that biofilm, which is sheltering the bacteria, and hopefully some of them are still alive, even if the patient has gotten some antibiotic therapy. And if you can dislodge that biofilm, then you can use that sonicated fluid to get the diagnosis. So, again, you put the generator in a sterile container with some sterile fluid, put it in the sonicator, and then send that fluid to the lab. And the sensitivity is a good bit better, although, you know, the specificity is a little bit less, not surprisingly. So, as far as post-extraction management, you know, considerations. So, is the patient pacer-dependent, right? Because if they're dependent, you know, they're going to need some sort of temporary system, and you have to think about where you're going to put that, especially because generally we want to reserve the contralateral side for the future device implant. You know, is the patient going to need a PICC line? Where is that going to be? Hemodialysis access can be an issue as well. You certainly want to clear cultures prior to any reimplant. As I already mentioned, probably if there's valvular involvement, particularly tricuspid valve involvement, you're going to want to wait longer for re-implant. And, of course, a novel site should always be used for the new device. So, what about when there's other hardware in the patient, particularly other hardware in their heart? So, when I was a fellow, which was, you know, 100 years ago, we would actually, if patients had prosthetic valves in place, and they had, you know, sepsis, and they had, you know, a CIED in place, we would often say, well, if the surgeons aren't going to do anything about the prosthetic valve, particularly if it's a metal valve, why are we going to do anything about their pacemaker or defibrillator, which is likely infected? There's a few studies that actually do suggest that it's probably still worth taking out the device. So, this is outcomes of transvenous lead extraction for, you know, device infections in patients with prosthetic heart valves. And so, you can see that some of them were localized pocket infection. Some of them were, you know, device-related endocarditis. Some of the patients clearly had valve, prosthetic valve involvement. But there were 61 patients overall, 127 leads. They extracted the devices and did comprehensive antibiotic therapy, not necessarily long-term suppressive, but they, you know, had a full four-to-six-week course. They did not devalvular surgery, and they found that 95 percent of the survivors were infection-free at one year of follow-up. So, even if the patients do have a prosthetic valve in place, and, you know, the surgeons don't want to do anything about that, whether it's involved or not, we do have some evidence that it's probably still reasonable to take out their pacemaker or defibrillator if we think it's involved. The same may be true for ventricular assist devices. So, this is transvenous lead extraction in patients with LVAD, device-associated cardiovascular implantable electronic device infection. You know, again, small numbers of patients. There were 68 leads extracted from 27 patients. They had good success in getting the device out. These patients were, of course, put on suppressive oral antibiotics because of their LVAD. Fifteen percent of them did have persistent infection, but the 12-month survival was upwards of 80 percent. So, you know, again, you know, this is a sick population. Many of them are waiting for transplant, but, you know, available evidence would suggest that it's probably still reasonable to try to take infected pacemakers or defibrillators out, even if the patient is thought to have LVAD infection as well. So, when are device infections surgical? Well, I mean, the obvious answer there is when the patient has another indication for an open procedure. So, if your patient has a valvular abscess, valvular regurgitation that needs to be dealt with acutely, or a fungal infection, which is likely to be surgical, it might be reasonable to have the surgeons take the device out if practical. I will say on a few occasions I've gone into the OR and done a, you know, a, you know, laser or rotational extraction with the surgeons if they weren't planning on opening the heart entirely. So, you can do those sorts of hybrid procedures. One issue that has come up is vegetation size. If you look in some of the old literature, people suggested that if your vegetation was greater than a certain size, so a very large vegetation, it might be safer for the patient to have a surgical removal simply to decrease the risk of embolization. There's no question that when we do lead extractions in patients, whether or not they're infected or being done for another reason, there's no question that we embolize some material, and that material, of course, is going to end up in the lungs. Unfortunately, most patients do just fine with that, but the concern is that if a patient has a very large vegetation and you swipe that off during, you know, a lead extraction, is this going to cause a septic emboli, which is going to make the patient have a worsened outcome. I think in general many of us have trended towards thinking that, you know, having an open procedure with a sternotomy is a pretty invasive thing to do if we can get, if we were able to get the leads out and get the patient through. You can do angiovac suction of large vegetations, and I will say in Europe this is a really common thing that they do, probably a little bit less common in the United States, again, partially, you know, due to coverage issues and whatnot, but certainly that can be done and probably is safer for patients. Other reasons why patients might need to go to the OR for device infections, epicardial patches that are involved, and then, of course, you know, not that it ever happens, but if there would be failure of a percutaneous extraction, then, you know, if the patients have recurrent bacteremia, then we're talking about an open procedure to get all the hardware out. So I've talked about diagnosing and treating infections, but, of course, what we really, really would like to do is prevent infections, and so things that we do to prevent infection, perioperative pharmacotherapy, antibiotic irrigation of the pocket and any incisions, preoperative chlorhexidine scrub, this is a pretty standard thing that's really done for, you know, most surgeries, whether it's orthopedic or cardiac, you know, chlorhexidine scrub is very inexpensive, and decolonizing the skin has been shown in a few studies to help reduce long-term infection, and then also we have some antibiotic envelopes, so there is the Medtronic Tyrex antibiotic envelope, and then more recently the Kangaroo envelope, which is sort of a biocellular matrix, which is designed to promote, you know, ingrowth of the native tissue. So this is the prevention of arrhythmia device infection trial, or the PADET trial, and this was an interesting trial, very, very large. It was a randomized crossover trial at 28 centers, upwards of 20,000 patients, and they had two groups. There was the conventional care group, and the conventional care group was cefazolin, so basically the patient received cefazolin around the time of the procedure and then afterwards, and then they had what they called the incremental care group, and they really sort of threw the kitchen sink at this group. They got the cefazolin, they got vancomycin, vasotracin was put in the wash, and then they got two days of cefalexin orally, post-operatively. The bottom line was there was no benefit to incremental care in the total group or any subgroup, and the figure here is showing, you know, the actual risk of hospital admission for infection at one year, and what the figure is showing is there was a trend towards the incremental group doing a little bit better. The trend being, if you looked at the CRT group, which was the highest rate of infection, the conventional therapy group had a rate of about 1.9 percent, and the incremental group had a rate of about 1.5 percent, and again, this was not statistically significant, and it was an extremely large group of patients, so certainly you want to give your patient perioperative antibiotic therapy, but, you know, going overboard and using multiple antibiotics probably is not going to be necessary. This is the RAPIT trial. This was the trial looking at the Tyrex pouch, and so basically it was sponsored by Medtronic, and they had upwards of 7,000 patients. The duration of follow-up was a little bit over a year and a half, and basically their inclusion criteria was patients having placement of a device pocket revision generator placement, system upgrade, or initial implant of a cardiac resynchronization defibrillator, so they were trying to pick some patients with, you know, repeat procedures who might be a little higher risk. I will say they did exclude some higher risk groups like renal failure, and basically it was a positive trial. The p-value was 0.04. The control group had an infection rate of 1.2 percent, and the Tyrex envelope group had a 12-month rate of infection of 0.7 percent, and so certainly the Medtronic folks were touting this as, you know, nearly a 50 percent reduction in infections, but again our control group had a really pretty low infection rate, and again, if you go back to what I was talking about way back at the beginning, you know, this is an extremely low infection rate in the control group. It may not be representative of what we're seeing in a real population, but again it was a positive trial, and so, you know, certainly, you know, it has indication for use for preventing infection. One question many of us had when the Tyrex envelope was released is, is it worth it? You know, I think most of us don't think our infection rates are that high. The Tyrex envelope runs about a thousand dollars. You know, is this something that you're really going to do in every patient? Is this really a cost-efficient thing to do? The problem is, as I also discussed at the beginning, device infections are really, really, expensive, right? As I presented, you know, there's probably an additional $50,000 price tag on a patient getting a device who then becomes infected, so you can put in a lot of Tyrex pouches, you know, before you're going to get to that kind of cost, so this is from the Journal of Medical Economics, and they were trying to look at the cost-effectiveness of the Tyrex envelope in preventing device infections, and basically, after their analysis, what they concluded was, if your rate of infection for a CRTD is greater than 1.65%, for a CRTP greater than 1.95%, greater than 1.87% for a standard pacemaker, or 1.38% for an ICD, it probably is cost-efficient to put them in your patients. So, you know, a lot of this depends on your own rate of infection. I think what many of us are doing is using them in patients that we think are high-risk, but there's not a lot of clear data to tell us exactly who to use these things in. So, the kangaroo envelope is also now available. This is an extracellular matrix biomaterial, and basically, this has been approved for use in other realms as well. Basically, the idea is that it's an envelope. It is soaked in fluid. It's soaked in antibiotic-containing fluid, and then the device is put in this, and then put in the pocket, and the idea is that, first of all, you know, there's antibiotics in it, which will help prevent infection, and then also it's going to promote in-growth of native material, so you don't get the thick fibrous capsule. I think it is important to note that this has not really been studied, you know, rigorously in any human clinical trial. The data for this is in vitro, in vivo, and in animal trials. So, you know, I think this should be treated a little bit with caution. I think people like this product because, compared to the Tyrex pouch, which is kind of a heavy mesh and has sort of poor handling characteristics, this is kind of like a heavy rice paper, is what it feels like, and so it has really nice handling characteristics, but I think we should bear in mind that the data isn't exactly there. So, as far as preventing recurrent infections, we do know that having a prior device in the infection is a risk factor for future infections, so we do want to prevent recurrent infections as much as we can, and one thing that is worth noting is we do have several new technologies that, you know, are blowing up that standard paradigm of transvenous leads. So, we now have the subcutaneous ICD, so nothing is in the blood pool, everything is just under the skin, everything is just under the skin, and we do have the leadless pacemaker, which is implanted directly in the heart and is very small and we think completely endothelializes, and so we think these things might mean that these two devices are less likely to get infected than transvenous systems. There is some small data to support this notion. This was infection and mortality after implantation of a subcutaneous ICD after transvenous ICD extraction. So, basically, they looked at patients who got an SICD rather than a transvenous system and compared them to patients who had transvenous systems, and what they found is that the SICD patients did seem to do just fine. So, you know, certainly, you know, that's good. You know, the other thing that's important to note is even if a subcutaneous ICD does get infected, there's no question that it's lower risk to extract compared to transvenous systems. So, that's another reason, you know, even if it's not less likely to get infected, which I think it probably is, it's certainly lower risk to get it out if it becomes infected. So, of course, subcutaneous ICDs cannot, you know, cannot act as pacemakers, but for your patients who just need primary prevention defibrillation, you know, this is certainly an appealing option. We also do have some evidence that leadless pacing might be lower risk for infections. This is from the MICRA post-study. Major complications in 105 patients with prior device infection and extraction who then got the MICRA. And the bottom line is they didn't really see issues with device infections. You know, there were some procedural complications, but the MICRA getting infected did not seem to be an issue. So, I'm just going to conclude with the things that, you know, we really don't know about dealing with device infections. And again, this is from the European recent consensus statement. And they just laid out a nice table of things that we really don't know and need more information on. So, as far as gaps in evidence. So, first, one thing that's worth pointing out is that the mortality for patients with device infections long-term is really high. It's 2.5 times the mortality rate of non-infected patients. And we don't really know why. You know, I think for those of us who do lead extraction, you know, we have our sick infected patient, we extract the device, and the staff all breathes a sigh of relief when the extraction is done safely. But the truth of the matter is those patients still have a really poor prognosis, both at 30 days and long-term. And we don't know exactly why this is. Is it related to the infection? Is it related to the fact that, you know, whatever caused them to get infected as far as comorbidities is, you know, meaning that, you know, they have a poor outcome? We really don't know. Secondly, the optimal time for re-intervention for device complications in order to reduce the risk of infection is unclear. And what I mean is, let's say you've put in a pacemaker, you have a lead dislodgement. I think most of us, you know, would trend towards saying let's get this fixed as soon as possible. But it's also possible that, you know, waiting a bit, giving a little bit more time would reduce risk of infection. And we really don't know. We don't know, you know, best drapes to use to help prevent infection. We don't have, you know, risk stratification strategies or risk calculators to minimize, you know, risk for CIED. And again, the Tyrex antibiotic envelope, you know, is a great example of, you know, we don't have, you know, clear definitions of who would be the best patients to put those in. They point out that the role of PET scanning and tagged white blood cell scanning, the evidence is growing, but we still need more to really define how to use this. The duration of antibiotic therapy, you know, is not necessarily completely clear. Another important point is the most optimal salvage strategy for patients who we can't extract. We really don't know what that is. Appropriate timing of reimplantation after CID infections is unclear. And this ends up being, I think this is true for many institutions, you know, when I'm, you know, extracting a patient for a device infection, we often have the infectious disease, our colleagues, they're involved. And, you know, there's a lot of different recommendations. You know, some people are fine with three days. You know, I've had colleagues want us to wait a full four to six weeks before reimplanting. In pacemaker-dependent patients, what's the best strategy to deal with temporary pacing and then, you know, long-term permanent pacing? You know, do we do externalized to temporary wires, which I think many of us do do? You know, what's the role of leadless pacing in that? When do patients get epicardial leads? And then finally, you know, if you have a pocket infection, so, you know, you don't have a clear bacteremic patient, but you have a pocket infection and they have a contralateral abandoned lead, does that need to come out or not? And again, we don't really have that answer. And so while we do know a lot about lead extraction and for, I'm sorry, we do know a lot about device infections, the reality of the matter is there are some gaps in the data that we do need to fill in. So I'm going to end there, you know, in conclusion, you know, device infections are increasing in the United States and worldwide. In the setting of infection, complete device removal is really indicated. Most common organisms are gram-positive cocci and impaired coverage should include vancomycin. Lead infections compared to pocket infections can be difficult to diagnose and they really require vigilance. And, you know, of course, in the end, you know, these cause significant morbidity and mortality for our patients and impose a significant cost burden on the medical system. And I will end there. Okay. Thank you. That was awesome. Excellent review and very timely topic. So I was just going to ask you a few questions that have come through the chat function here. The first one I'm going to combine a few, it's on non-traditional devices. So subcutaneous ICD, leadless pacemakers and epicardial systems. Are you any more forgiving in your strategy prior to removing those devices? Is your threshold for extraction higher? How do you manage those? So the question was things like subcutaneous ICDs. Do I have a higher? Yeah. So the thing about subcutaneous ICDs is, you know, I don't know as far as pocket infection that they're less likely to get infected, you know, certainly because they're not in the blood pool. There's less risk if a patient becomes septic of it becoming involved. And certainly they're low risk to take out. We also think if you do get a pocket infection because the lead isn't in the vasculature, waiting a little bit longer, you know, trying to see if it's just a superficial infection as opposed to a deep infection, you know, is probably safer than that would be in a transvenous system because even if it is infected, you know, the lead isn't in the vasculature. So there's probably less likelihood that the patient's going to get septic. So I think I'm a little, for the SICDs, I'm a little bit more willing to wait and observe. But in the end, if you have a deep pocket infection with an SICD, it's going to have to come out. You're not really going to be able to clear that infection. Okay. And then I guess the other device, which was specifically asked about was an epicardial system. Do you have the same sort of strategy for that? Yeah. I mean, for an epicardial leads, obviously, you know, it's a really invasive surgery for patients to get epicardial leads removed. And so I tend to really want to be sure that those are infected before you put a patient through that. And again, this is the role where things like PET, CTE can be really helpful to try to define if they're involved. Okay. All right. And then I guess there's a question about who you routinely are using antibiotic pouches in. Is there a patient population that you have found that you always use it in? So I will freely admit that I don't have, you know, some sort of rigorous protocol that I use or some sort of scoring system that I use. Patients that I would definitely use those in dialysis patients, for sure. I tend to use it in patients who are, you know, are hospitalized for other reason who then end up needing to get a device because they're, you know, they're in the hospital environment. Very prolonged procedures. I will use them in. So if you have a, you know, a CRT device, that's just taken a really, really long time. I'll consider it for those. And then, of course, anybody who's immunosuppressed. Great. And then the last thing was any role for penicillin allergy testing in these patients who carry penicillin allergies in their chart so that they can get cephalosporins as opposed to vancomycin? So, you know, the cross reactivity between penicillins and cephalosporins is actually not that high. So at my institution, you know, if patients report a penicillin allergy, oftentimes if you track through their chart and really look in their history, many of them have already had cephalosporins without any issues. But frankly, we usually just give them the cephalosporins and, you know, we really haven't had any issues.
Video Summary
In the video, the speaker discusses the increasing problem of device infections, particularly in patients with cardiac implantable electronic devices (CIEDs). Device infections are a significant issue due to the high number of patients receiving implants each year and the rising infection rates. There is a need to improve the understanding and management of these infections to reduce their impact on patients.<br /><br />The speaker highlights that most device infections are caused by Staphylococcal species, particularly Staphylococcus aureus and coagulase-negative Staphylococci. These infections are difficult to treat as the bacteria form biofilms on the devices, making them resistant to antibiotics. The current treatment approach involves complete removal of the infected device and leads, followed by a course of antibiotics. However, more research is needed to determine optimal treatment strategies and prevent recurrent infections. <br /><br />The speaker also mentions new technologies such as subcutaneous ICDs and leadless pacemakers, which may have a lower risk of infection. Preventive measures including perioperative pharmacotherapy and antibiotic irrigation of the pocket are discussed, as well as the use of antibiotic envelopes. The speaker emphasizes the importance of accurate diagnosis, particularly in cases of lead infections, and the need for further research to fill gaps in knowledge and improve management strategies.
Keywords
device infections
CIEDs
infection rates
Staphylococcal species
biofilms
antibiotics
treatment strategies
subcutaneous ICDs
leadless pacemakers
Heart Rhythm Society
1325 G Street NW, Suite 500
Washington, DC 20005
P: 202-464-3400 F: 202-464-3401
E: questions@heartrhythm365.org
© Heart Rhythm Society
Privacy Policy
|
Cookie Declaration
|
Linking Policy
|
Patient Education Disclaimer
|
State Nonprofit Disclosures
|
FAQ
×
Please select your language
1
English