Drs. Jonathan Ryder and Nico Cortes-Penfield take us through the evidence and their approach to prosthetic joint infections!

Please consider filling out this survey to help us better understand utilization and satisfaction with the Febrile podcast and learning resource!  Your input will help guide future programming and only takes 5-10 min!!

https://hms.az1.qualtrics.com/jfe/form/SV_6o2iYLc1Sq7PMQC

We are conducting a research survey to better understand how Febrile is used and what can be improved!  Survey is voluntary, and all responses are anonymous.  If you have questions regarding this survey, please email swdong@bidmc.harvard.edu.

Link for registration for IDSA Fellows Peer Mentorship Program: https://docs.google.com/forms/d/e/1FAIpQLSdSBny4BGbo3-LTcv0mBd5IIJ7nHQXvTJhoEY0FAuMO3BrSZA/viewform

Episodes | Consult Notes | Subscribe | Twitter | febrilepodcast@gmail.com

Hi, everyone. Welcome to Febrile - a cultured podcast about all things infectious disease. We use consult questions to dive into ID clinical reasoning, diagnostics and antimicrobial management. I'm your host, Sara Dong. I'm a combined Med-Peds ID Fellow. Here on Febrile, we use patient cases and consult questions to learn about high yield ID topics. We will present pieces of the story of a patient's case and pause along the way to hear from our guest consultant.

My co-host today is Jonathan Ryder. Y'all remember Jonathan from our ID Week recap episode #23. He is an ID Fellow at the University of Nebraska Medical Center, which is in Omaha, Nebraska. His academic interests include antimicrobial stewardship, endocarditis and medical education.

Our guest discussant today is Dr. Nico Cortes-Penfield. Nico is an assistant professor in the Division of Infectious Diseases at the University of Nebraska Medical Center. He serves as the medical director of the Outpatient Parenteral Antibiotic Therapy Program, or OPAT, and is an associate medical director of Antimicrobial Stewardship. He completed his medical school, residency, and fellowship training at Baylor College of Medicine. His clinical and research interests are complicated bone and joint infections, and he's part of the core faculty of the Ortho ID service line at UNMC.

Welcome to the show, guys.

Hey.

How are you?

I'm doing good. So before we jump into the case, we're going to do our usual nonmedical start and just ask if either of you have a piece of culture that you would like to share, something you've enjoyed recently.

Sure. So my pitch is going to be for a website. It's called www.archive.org. It is a free Internet library where you can go have a free account and you can read thousands of books scanned in from public libraries across the country.

Oh, cool. I don't know this website.

Oh, that is so lovely. And then, Jonathan, did you have anything you wanted to share today?

Well, I don't have a piece of culture per se, but I have something to plug. Today, the IDSA is launching a fellows peer mentorship program for infectious diseases fellows and trainees interested in ID to pair with senior fellows and early career faculty from around the world. There will be opportunities for mentorship, coaching, support and networking. There's a link that I can provide where people can sign up until the deadline in mid March.

I love it. The more mentors, the better. I have signed up, and I'm looking forward to seeing who I get paired with. But I will not delay the episode any longer. We will jump into the case.

So today's consult question is from Orthopedic Surgery. They're calling us about an older male with a red, swollen and painful joint. They would like us to evaluate and provide antibiotic recommendations, so I will hand it over to Jonathan.

Thanks, Sara.

So today, our patient is a 54 year old male. He has a past medical history significant for well controlled diabetes mellitus, rheumatoid arthritis on infliximab, hypertension, and chronic kidney disease, stage three. His surgical history is notable for a right total knee arthroplasty for osteoarthritis two months prior to his presentation, and today he presents with right knee pain and swelling. Postoperatively, he had persistent swelling thought to be due to a hematoma at that time, and while that swelling initially improved, it gradually worsened. Over the past two weeks, his knee became painful with surrounding redness, and he hadn't really noticed any fevers or shaking chills, and there was no drainage from his incision site.

His social history is notable for no tobacco use or substance use. He's a fairly active guy and considers himself an amateur paleontologist, but he has not traveled the world doing so. He likes to walk with his dog named Neil and his two tarantulas named ELla and BOWie.

For imaging, he has a right knee X ray that shows a right total knee arthroplasty without hardware complications and moderate soft tissue swelling with joint effusion. A joint aspiration is performed, which is cloudy in appearance with a yellow color. The white blood cell count is 22,000 with 91% neutrophils. So at this time, Nico, what is your approach to this patient?

Sure. So to sum all that up, we have a 50 year old guy with diabetes and rheumatoid arthritis on a TNF Alpha inhibitor as well as obesity, and he's coming in with right knee swelling and pain two months after a total knee arthroplasty. He's got a number of risk factors for prosthetic joint infection. In general, knee arthroplasties are the most likely to become infected and most difficult to treat. Shoulder arthroplasties rarely get infected and usually are pretty easy to treat, and hips are somewhere in the middle. Beyond that, we can divide the risk factors for prosthetic joint infection to three main buckets. The first is any kind of immunocompromising state- rheumatoid arthritis, diabetes, chronic kidney disease, treatment with steroids or TNF alpha inhibitors, things that make the surgery or the wound healing more difficult, like obesity, having had a prior prosthetic joint infection at the same site, or this being a revision arthroplasty for a non infectious reason of failure and then having a disseminated infection, which primarily there we're talking about Staph aureus bacteremia, or occasionally invasive strep infections. I want to point out specifically or call out that recent dental work frequently misbelieved to be a risk factor for prosthetic joint infection. For a while, both the orthopedic and dental societies have said you don't routinely need to offer dental prophylaxis because it's not a true association. And then just recently there was an observational study of like 9000 patients in England where no one gets antibiotic prophylaxis with dental work. And that found again, there was no temporal association between having a dental procedure and developing PJI. So dental work frequently misbelieved to be a risk factor, but not actually risk factor.

PJI can present either acutely or chronically. In the acute form, these patients often have rapid onset of pain and local inflammation of the joint. They may or may not have fever, chills, or other systemic symptoms. You can also have a subacute form of onset, in which case patients may have more vague or subtle physical exam findings, but will often complain of either sort of indolent pain or instability, which is essentially the bone that the metal part of the implant is seated in is becoming soft and being eaten away. And so that implant is rocking, and they perceive that as the leg about to give out. You may also see a delayed wound healing or ongoing drainage after a revision of arthroplasty. That's a frequent, sort of immediate, post operative sign of prosthetic joint infection.

There are standard diagnostic criteria for PJI. They're the 2018 MSIS criteria. They can be used clinically, although honestly, I think of them mainly as a consensus definition for PJI research. But they're kind of like the Duke's criteria. They have major and minor criteria, and there's a scoring system that you can use to classify the diagnosis as proven, refuted or inconclusive. What I think is most worth remembering about the MSIS scoring system is that there are two major criteria that confirm PJI, either a sinus tract that communicates from the outside world to the joint, or multiple synovial fluid or tissue cultures that are growing the same bug. If you have either of those, case closed, PJI. There are a whole laundry list of the minor criteria. I'm not going to go through them. Even as an Ortho ID doc, that's necessarily worth carrying around in my brain. But each one of them is assigned one to three points. Each at least six points proves PJI. At least two points is at least inconclusive for PJI. And these are things like the elevated inflammatory markers, abnormalities of the white blood cell count or PMNs in some synovial fluid, some synovial specific assays like you can actually do synovial CRP or alpha defensin and then interoperative findings like pus, histopathology and a single positive culture.

Next, we can talk about the microbiology of PJI. So the classic teaching you've probably gotten is that what causes PJI varies by the chronicity, whether or not it's early postoperative infection, delayed onset postoperative infection, or a late infection. And the idea here is that both early and delayed infections come from bacteria that were introduced during surgery. The early (that's within the first three months of surgery) is due to more pathogenic organisms that grow really quickly, like Staph aureus, Gram negative rods. And then the later or delayed onset infections that occur between three months and one year are more indolent organisms like coagulase negative Staph and Cutibacterium. And then after one year, these infections are usually things that have seeded the joint from the blood, which, again, is going to be Staph aureus, Strep, Gram negatives. Honestly, I think this framework makes sense. I just don't find it all that practically useful. If you'll notice, I mentioned Staph in every single one of those categories. And if you put together coagulase negative Staph and Staph aureus, that's the majority of PJI in all settings -- so basically always have to think about Staph.

What I find it more useful for me is to think about when I need to think about something that isn't Staph. And there are a couple of major scenarios there. The biggest one is shoulder prosthetic joint infection. So for human microbiome reasons, total shoulders and reverse total shoulders have a lot of Cutibacterium acne infections. It's actually one of the most common pathogens. So if I'm seeing your shoulder, I'm thinking about C.acnes, and then if I'm thinking about recurrent prosthetic joint infection and specifically sort of early recurrent prosthetic joint infection after surgery, when someone has just received a bunch of broad spectrum antibiotics, then I'm thinking about more drug resistant things like VRE and some of the unusual Corynebacterium. Corynebacterium striatum, for some reason, is the one I see the most because those things are resistant to gosh darn near everything. And so if you've gotten bought spectrum antibiotics, you start to select for them.

Thank you, Nico. That's very helpful. So there was a joint aspirate fluid culture, and on Grah stain, it showed gram positive cocci in clusters, and those cultures grew MRSA. Surprise, Staph is the most common. So with this information, what next steps do you take? And specifically, what kind of conversations do you have with the orthopedic surgeons?

So first, let's talk about the two things we could do. The first approach is called debridement and implant retention, or DAIR, D-A-I-R. This is also known as I&D with poly exchange or just poly exchange, or sometimes exchange of the mobile components. And the idea is that the prosthetic joint has really two major components. There's the metal part that's hammered and cemented into the bone and then there's a polyethylene liner or cap, that goes over the articulating component, the actual joint, and that can be popped off. And so in a DAIR procedure, which is way less invasive than actually taking out the entire implant, what you're doing is opening the joint, popping off that polyethylene component, scrubbing everything really well, trying to get the bacteria off, putting on a new polyethylene liner, and then closing everything up. As I just mentioned, the upside of this is it's much less invasive, much shorter period of time before your patient's ambulating, so lower risk of deconditioning. The downside is that it's harder to be sure you've gotten all the infected material out. And these patients, particularly if they've got a lot of residual bacteria and biofilm, a big bioburden left, are at higher risk of relapse and need longer courses of antibiotics. So hopefully, knowing that, it makes sense that we basically like to use DAIR in the first few weeks of surgery, and there hasn't been a lot of biofilm build up yet, or at least when we've caught the patient shortly after the onset of their symptoms, if they're not in the early post operative period. And then we never do it when we've got either a sinus tract communicating to the joint or we've got advanced infection like infection to the point that we can see the hardware loosening on imaging, or we've got really difficult to treat organisms like fungi, like Candida, I think is probably the major contraindication. Or if you cannot do DAIR, you really want to do anything else. And then like a relative contraindication, I think would be Staph aureus.

So in those folks, we want to go for some sort of exchange arthroplasty, which means taking the whole thing out and then putting in a new prosthesis. You can do that either in a single stage or a two stage procedure. In a single stage procedure, the surgeon, immediately after taking out the original joint, is going to put a new one in, whereas in two stage, they're first going to place a cement spacer traditionally mixed up with some antibiotics, like vancomycin, tobramycin, gentamycin, and then they'll come back in a couple of months later to take out the spacer and put in a new prosthesis. And the idea here is we're not only removing all of that infected material, but we're putting in some local antibiotics that leech into the joint space and kill off any straggler bacteria or anything that might be stuck to leftover cement. And that's going to give us a much better likelihood of having durable cure once we put a new prosthesis back in. The downside, it's a bigger surgery than a DAIR. It requires often multiple months in which the patient's non weight bearing because they've got a spacer. What it's worth, there are some new, what they're called functional spacers that actually allow the patient to have some partial or total weight bearing to avoid the deconditioning of that.

I'm going to refer to both one stage and two stage procedures as exchange arthroplasty because there's a big regional variation in how they're used. So specifically in Europe, the main approach is single stage exchange. Two stage exchange is really less common and they mostly use it in the highest risk cases or folks who've had prior treatment failures. So that's like kind of an extreme thing to get a two stage. Versus in here, and I'm not sure what historical or reimbursement reasons drive that, but here in the US, we primarily use two stage exchange. To go back to our specific patient. At this point, my impression is that this guy's pretty high risk for about outcome, and I think his best bet is probably to go with exchange arthroplasty, like a two stage exchange. I say this because he's got a bunch of independent risk factors PJI treatment failure. So rheumatoid arthritis; major medical comorbidities, he has diabetes and CKD, obesity, which is going to make his surgery more difficult; Infection of a knee, which I know is likely to fail; an infection with MRSA, which is specifically a risk factor for treatment failure. He's also more than 30 days out from his initial surgery. So in terms of prognosis, honestly, the only thing this guy has going for him is that we don't have hardware loosening yet and we can't see his prosthesis from the outside world. As to what I'd say to my surgical colleagues, I really try not to dictate the type of surgery that I think the surgeon should do on mutual patients because I didn't do an orthopedic residency. Instead, I try to focus on the ID prognosis because that's where I can really quote data to them to back up my argument. So I'll say, "Hey, I'm pretty worried this guy is going to fail a DAIR because of all these reasons. Is there anything else we can do?" I find that gets a more productive response than "you really need to do a two stage".

And really quickly before we move on. I was wondering if you could talk just a little bit about interoperative cultures and the techniques that we use, particularly in the setting of PJI, because I think there are a couple of things that we do, like sonication cultures, that sometimes people aren't quite as familiar with.

The first thing to say is you want to ask your surgeon for multiple operative samples, not only because the MSIS criteria have that two samples of the same bug as a major criterion, but also just because you're going to sleep much better at night if you know that coagulase negative Staph is growing from one out of six cultures versus one out of one or one out of two. The second thing is there are a couple of different ways that your surgeons can help optimize the yield of those tissue cultures. One is we can do what's called sonication, where we basically remove the implant or orthopedic hardware, and we put it in a bucket and we shake the heck out of it, and we dislodge all the little bio films and bacteria that might be stuck on there and then try and grow bacteria from that fluid. An alternative approach, which some folks may find a little bit easier, is you can take that tissue or synovial fluid and inoculate it directly into blood culture bottles. And that has also been shown to significantly improve the yield of cultures. There probably will be a PCR based like multiplex assay approved by the FDA in the next couple of months. Look for data on that. I won't say anything more about that. And then, of course, you can also do broad range PCRs. You can send them to Mayo Clinic, University of Washington, on tissue or synovial fluid. I haven't had great results with that from prosthetic joints. I had one or two where we got something unexpected that changed management, but I'm still waiting for that PCR based test that revolutionizes my diagnosis of PJI.

That's really helpful. So at this point in time, because there is MRSA, blood cultures are obtained, but there is no growth at five days. The orthopedic surgeons take the patient to the operating room and intraoperative findings were consistent with infection, and multiple cultures were obtained. They excised the prosthesis and cement, debrided the soft tissue and bone, and placed an antibiotic impregnated spacer containing vancomycin and gentamicin. And he has started on IV vancomycin post operatively. The interoperative cultures grow MRSA. What is the medical management of MRSA prosthetic joint infection for a two stage exchange? And then, what is the deal with these antibiotic spacers?

Okay, so let's talk about what drugs we can use, and then we'll talk about how long we're going to use those drugs. And then I'll talk a little bit about what's up with antibiotic spacers. So in 2019, there's a landmark randomized control trial published in the New England Journal [of Medicine] called OVIVA, and that compared the long term treatment outcomes of oral versus IV antibiotics for bone & joint infections. So, in OVIVA, everyone got a week of IV therapy, and then they either got more IV therapy or oral antibiotics with the remainder of the treatment course. And the context of this trial is that oral antibiotics have been used for Ortho infections in Europe and across the world for decades, with great results. And there have been multiple comparative studies of oral and IV antibiotics for osteomyelitis showing equivalent outcomes. What OVIVA adds was being a really well designed, multicenter RCT with long term follow up in a bunch of PJI and other ortho hardware infections. Up to about a quarter of folks had infected hardware, and another 10% or so had hardware placed an infected site. The other thing OVIVA brought was having, just sample size, having enough sample to be powered for non inferiority, to show that non inferiority of oral antibiotics at a clinically meaningful threshold. And so what the OVIVA group found was that their treatment failure rates were exactly the same between the oral and IV groups, like 13% versus 14%. And there was no subgroup that benefited clearly from IV versus PO antibiotics. Now there were some subgroups that trended towards better outcomes with IV. I think the most important of those were culture negative infections and infections in which the oral agent was a penicillin. One of the downsides of a OVIVA's design is that they didn't record what specific drugs or doses were being used or what drugs were being used for what bugs. So in personal communication with the OVIVA investigators, some centers were using things like amoxicillin-clavulanate for Gram negatives and polymicrobial infections. And so the way I've interpreted this trial in my practice is that I'm happy to use high dose amoxicillin for Strep or Cutibacterium because I know the penicillin MICs are super low, but I'm a little more cautious about using them for things like Enterococcus or Gram negatives because I know the MICs are much higher. Otherwise, I think OVIVA definitively answered the question of whether oral antibiotics can be used routinely and whether we can switch to oral antibiotics early in bone and joint infections. That answer is yes.

So if we talk about what are the specific oral antibiotic options for Staph aureus? The best studied combination is a fluoroquinolone plus rifampin, neither of which I just feel compelled to say you would ever use as monotherapy in Staph aureus because of rapid resistance. However, quinolone + rifampin has been used and published on Europe for bone and joint infections for years, with great success. Now, I personally don't use this combination a ton, mostly because of the high incidence of medication interactions and specifically rifampin with direct acting anti coagulation. For what it's worth, there's some new data coming out on rifabutin for Staph aureus infections, and if that's effective, that may make drug interactions easier to handle. I also had several patients not tolerate this regimen due to tenosynovitis and other quinolone side effects. And there's a recent paper showing that having to stop your quinolone during treatment of a musculoskeletal infection isn't that rare. Trimethoprim-sulfa is another option. Most experts recommend about 10 mg/kg/day of trimethoprim, that comes out to about two double strength tablets twice a day or more. These high doses don't actually have comparative data versus the standard dose, so I have used one double strength tab twice a day with anecdotally good results, but that's a bit out of step with, I would say, consensus practice. And honestly, I don't use a ton of trim-sulfa because it's hard to get weekly labs on patients who are going home without a PICC and a nurse to come draw labs every week. And there's risk for acute kidney injury or even fatal hyperkalemia with Bactrim, especially when you're combining it with things like ACE inhibitors, ARBs, or spironolactone, which a lot of my patients are on. So honestly, what I use the most for Staph aureus bone and joint infections is doxycycline, or sometimes minocycline. Doxy actually doesn't have a ton of data for it in the literature, although there are several dozen patients treated with doxy in OVIVA and it looks just as good as any other drug. But we found it works really well. It's tolerated in long term use as long as you warn patients about photosensitivity. The main annoyance of doxycycline is that rifampin can decrease its serum levels. So if you're going to lose your sleep, lose sleep over your patient not being on rifampin, maybe doxy isn't your best choice. Other things you could go with would be like clindamycin, which our pediatric ID colleagues have used for quite a long time, again with good success. And possibly if you just need something to mop up the last couple of weeks, linezolid -- although I wouldn't use that in long term use because of the neurotoxicity and bone marrow suppression and I'm-killing-your-mitochondria-related issues. The last thing I'll shout out is if this was MSSA instead of MRSA, we could use cefadroxil, which is a cousin of cephalexin, and it's another oral first generation cephalosporin, ID doctors on Twitter really like it. One of those... No, the reason I call this drug out specifically is because it has a slightly longer halflife than cephalexin, which means that it can be dosed BID instead of TID or QID. And honestly, I don't think most patients are able to take more than a BID dose medication. Anecdotally I've had pretty good success with it.

Okay, so that was which drugs. Now let's talk about how long. So in patients who get a two stage exchange, we're going to start with six weeks of antibiotic therapy, and then we're going to go off antibiotics for some number of weeks, usually two to six, which we call an antibiotic holiday. The idea here is we're trying to convince ourselves that the infection is gone. Most centers will trend CRP and ESR, checking the values at the start of the six weeks of antibiotics, at the end of antibiotics, and then after that antibiotic holiday. The idea is you want to see those inflammatory markers normalize on the antibiotics and then not go back up again once we stop them. For the record, this is one of those generally accepted practices that isn't based on much. Actually, there was a 2011 study out of the UK that found the CRP trend doesn't really predict outcome of PJI, but it's something everyone does. I personally haven't found CRP catching eventual treatment failures that we weren't clinically concerned about. And I would say a very common question I get from patients at the end of that initial six weeks is- how do we know it's gone? And so what I tell them is- the most valuable test for me is to see you after your antibiotic holiday, have your surgeon see you after your antibiotic holiday, listen to your symptoms, examine you, and go by what you're feeling and what we see. After the antibiotic holiday, there is again some regional variation from center to center. Some places routinely will do a diagnostic arthrocentesis to again prove that that joint isn't infected. Some places will only do that if the CRP has been trending up after off antibiotics. Some places, if they see the CRP go up, are just going to go ahead straight away and do an antibiotic spacer exchange instead of reimplantation knowing that that's a better tolerated surgery. And then finally, when our patient gets their second stage revision, we're going to get some repeat tissue cultures to prove that infection has been eradicated and in some centers will give some secondary antibiotic prophylaxis, which we can talk about later.

Last but not least, what's the deal with these antibiotic spacers? So it's actually really interesting. You guys asked me this and I had to go look it up, but it's really cool. So material scientist had been showing all the way back in the 1980s that you could put antibiotics into acrylic and then you could put that into a person or an animal and it would reach out tissue antibiotics in meaningful levels. And then as far as I can tell, in the 1990s, our Ortho colleagues just started adding them into bone cement and spacers and reporting how that went for their patients. And it turned out the patients mostly did really well, not a lot of major side effects and very few reinfections. And so this practice rapidly became sort of standard of care. And one result of this is that it's really hard to find actual comparative studies that look at spacers with antibiotics versus with no antibiotics. There is one meta analysis from 2008. It's cited in the UpToDate article on PJI, and it pulls together the data on antibiotic bone cement. Not spacers, just cement. But it did show, it showed "reduction in reinfection". And when I say that, it's with air quotes, because it's a forest plot of one positive study and like six negative studies, and that comes out as evidence. So the data is pretty limited. It is worth expressing the skepticism because although antibiotic spacers are usually benign, you can get detectable serum levels of vancomycin and aminoglycosides just from a spacer. You can get acute kidney injury from antibiotics just in a spacer, especially when we start doing dumb stuff like putting multiple aminoglycosides in the same bone cement. And you can get allergic reactions up to and including anaphylaxis from just the local spacer antibiotics. So they're not completely benign.

Great. So our patient actually ended up treated with intravenous vancomycin for six weeks, despite data to the contrary. And his clinical signs of infection have resolved, his CRP normalized, and he completed a two week antibiotic holiday. And so he re-presents and undergoes a successful re-implantation to complete his two stage exchange. And repeat cultures do not grow any bacteria. So at this time, what is the role for secondary prophylaxis after a two stage exchange?

Okay, so the guidelines based answer to this question is nothing, because the IDSA PJI guidelines are from 2012, and the data that's supporting this practice is more recent. However, our group and centers around the country have started giving some secondary antibiotic prophylaxis after the second stage of a two stage exchange. This is based on a randomized control trial by Yang and colleagues that was published in 2020. They recruited patients who had just completed their two stage exchange and randomized them to either get three months of oral antibiotics directed towards their prior cultures or no antibiotics. And they looked at reinfection in folks who were followed for at least two years. What they found was that the antibiotic prophylaxis group had less than half the rate of reinfection of the new antibiotics group of like 13% versus 29%, which translates to a number needed treat of about six. Now, the study has some design flaws. There are two letters to the editor about the paper that point this out at greater length. But briefly, it was underpowered, and it had a lot of patients excluded from the primary analysis due to things like death, loss to follow up, yada yada. So I wouldn't say there's widespread consensus in the ID community at least that this is practice changing data. The other thing that's worth saying is our colleague Dr. Laura Certain, out at the University of Utah, just published their experience with extended antibiotic prophylaxis, and what they found was that although it's maybe a slightly lower incidence of treatment failure and not statistically significant, of the folks who did every infection, two thirds had a drug resistant bug in the prophylaxis group versus none of the patients in the no antibiotics group. So whether or not this actually reduces reinfection, it seems like it definitely drives resistance in the folks who do have a reinfection. What we do here at the University of Nebraska is a little bit different than the approach that was actually used in this study. And I'll explain that briefly. When you look at the reinfections in the trial, they're mostly with new bacteria rather than recurrence of the prior bacteria. And so we decided that basing prophylaxis on the prior cultures didn't make a ton of sense. So around here, both coag negative Staph and Staph aureus are pretty universally susceptible to doxycycline. So, instead of culture directed antibiotics, we just give doxycycline. The other difference is that we felt that three months of extended oral antibiotic prophylaxis was a little bit unreasonable from the stewardship perspective, particularly because the authors didn't have a clear rationale for picking that duration and just didn't make sense to us that if you can sterilize bone with four weeks antibiotics, you would need to give prophylaxis for three times that period of time. So we give it for two weeks, and we figure that the bulk of benefits, probably during that early period when the wound is healing up. We haven't published our outcomes with this. Maybe this isn't the right thing to do, but this is our particular practice. And as I mentioned, there's a wide diversity of practice around the country. There are folks who aren't doing this at all. There are folks who are doing it exactly as the study detailed. People are still trying to figure out what to do with this.

So our patient is treated with oral doxycycline for two weeks after his reimplantation operation. However, a week after completing his doxycycline, he has wound dehiscence and he undergoes debridement with implant retention or DAIR, and those cultures grow MRSA again, and he started on IV vancomycin. So now, after this failure, what are your next steps in medical management?

So in patients who are managed with DAIR, we give prolonged antibiotic therapy, typically between three and six months for Staph, with six months for Staph and infected knee arthroplasty, three months for Staph and infected not knee (so hip or shoulder.) The guidelines again from 2012 would say that you could do as little as six weeks after DAIR for non Staph bugs.

But last year, there was a big randomized control trial called DATIPO that compared six to twelve weeks of antibiotics and prosthetic joint infection, and they found a large benefit to long term care within twelve weeks in the DAIR group.

And when I say large, I mean 16% absolute risk reduction.

So again, like number needed to treat of about six.

And so in the context of that trial, I think probably most patients who receive DAIR, regardless of organism, should receive at least three months of antibiotic therapy.

Next, let's talk about rifampin.

So patients who have retained orthopedic hardware infections with Staph are also recommended to receive a regimen containing rifampin, "for biofilm" if the organism is susceptible.

And I'll kind of briefly explain the thinking behind this.

If you remember, most antibiotics really target actively dividing cells, for example, by messing up cell synthesis or protein synthesis, which means bacteria that are just kind of chilling in biofilm are intrinsically more resistant to most antibiotics versus when they're not actively multiplying.

rifampin is unique in that it goes after RNA polymerase.

RNA is not a very stable molecule, so even if you're not doing anything, you need to be making more RNA if you're a cell and you want to survive.

And the idea, therefore, behind adding rifampin is that it's going to be able to eradicate these metabolically quiescent bacteria in the biofilm and prevent relapse that way.

All that sounds great on paper, and it works great in in vitro and animal models, but the clinical data that drove the adoption of this practice is pretty problematic.

So it really traces back to this randomized control trial by Zimmerli, et al. published in JAMA in 1998.

And boy, if you think that OVIVA or DATIPO or even the trial by Yang are problematic,

I can't wait to hear what you think about this.

So let me tell you about it.

They enrolled 33 patients with retained infected hardware infections due to Staph, including only 15 patients with prosthetic joint infection.

And the comparison groups were ciprofloxacin + rifampin versus ciprofloxacin monotherapy, which I think today most of us would agree is not really appropriate for invasive Staph aureus infection.

And then a third of patients in the rifampin arm, which again only had 18 patients to begin with, dropped out of the study and weren't included in the primary analysis, which reported 100% cure in the quinolone + rifampin arm versus just 58% cure in the twelve patients in the cipro monotherapy arm.

That is how we get to a PJI guideline that says three to six months of combination, including rifampin, A-1 recommendation

Now, in fairness, there have been multiple retrospective studies since 1998, several of which have suggested a benefit of rifampin-based therapy.

However, these papers all suffer from selection bias, residual confounding, and other retrospective paper problems.

Also, several of them don't stratify the outcomes based on the antibiotic regimen, or they don't include patients that get regimens besides quinolone and rifampin, or they do do both of those things, and then they find that there seems to be a specific benefit with quinolone + rifampin

Moreover, there was another randomized control trial in 2020 by Carlson, et al. that partnered rifampin or placebo with more standard drugs for PJI like vancomycin or beta-lactam, and they failed to show any benefit.

Now, there was a lot of criticism of this trial.

It was small, it was underpowered, it was 99 patients, it was several times larger than the trial that got us started doing this in the first place.

And then finally in 2021, there was a meta analysis by Aydin et al and they found that some of the evidence didn't really suggest a benefit for rifampin based regimens.

So to sum all that up, it's not really clear to me whether or not the evidence actually shows that rifampin is useful in retained ortho hardware infections, but it might be.

But if it is, I'm also not sure whether or not it's generally useful as an adjunctive agent or maybe just

fluoroquinolone plus rifampin is a particularly good regimen for Staph PJI.

So how do I put this into practice? Well, I do use a lot of rifampin in retained orthopedic hardware infections if I don't have a good reason not to.

However, if the patient has, for example, a serious drug drug interaction between rifampin and some other medication that they have, like a more evidence based reason to need, then I don't add the rifampin.

And if they develop elevated transaminases or nausea and vomiting or some other side effect, I stop it and I don't lose sleep about it.

I think the next question in terms of antibiotic duration for this patient is whether to give chronic antibiotic suppression.

When we say suppressive antibiotic therapy, typically we're talking about any antibiotics after the three to six months that would be indicated for definitive treatment.

The data on suppressive antibiotic therapy is entirely retrospective and there's not very much of it.

There are like three or four studies that compared suppression to no suppression, and they indicate maybe there's a reduction of long term reinfection risk.

The downsides are, of course, pill burden, cost, long term risk of being colonized with drug resistant organisms, and the adverse events of the chronic antibiotics themselves.

So interestingly, there was a recent systematic review of folks on suppressive antibiotics that suggest that adverse events that cause treatment discontinuation are pretty rare, like 4-5%, probably because most of the folks had been on that suppressive antibiotic for the tail end of the three to six months in the first place, and so we knew that they were going to do well.

The other thing I'll say is that one of the most interesting things in that six versus twelve week trial (DATIPO) is that none of those folks got suppressive antibiotics.

And so you can actually look at the success rate in the group who received DAIR and then twelve weeks and it's 85%.

And so that tells you that most patients who have a DAIR don't need antibiotic expression, couldn't possibly benefit because they're going to do well.

So the way I approach talking about and deciding whether or not to start suppressing antibiotic therapy after DAIR is shared decision making with my patient. If they don't have at least two or three independent risk factors for treatment failure.

Again, things like Staph aureus, polymicrobial infection, prior treatment failure, multiple surgeries on the joint, major immunocompromise.

I probably start the discussion by saying this is not something I think you need. If they do have multiple risk factors,

The next question I'm asking myself is- would they be a candidate for future surgery if they had a reinfection? I should say future surgery that will preserve the joint; or is the next step amputation and major loss of functional status?

And then the last questions I ask are about the individual's adverse event risk profile and that is basically a composite of life expectancy,

(is this just a little old lady I'm trying to keep out of the OR until she dies of something else) and the individual drugs safety profile (So can I use something I know is safe and long term use like doxy or Bactrim or do I only have something awful like once daily linezolid?) I have this discussion, I would say, with most of my patients just because I think the data on risk benefit calculus of suppressive antibiotics is pretty murky.

And so I don't know, I just think I owe it to my patients to kind of walk through my reasoning process.

I will say, despite having this discussion very often, very few patients actually opt for the chronic suppressive therapy.

And it's mostly these high risk folks in whom I kind of wanted to do that in the first place.

The last thing I would say about suppressive therapy is I try,

I know I just said lifelong,

I try not to use that word out loud or say antibiotics forever or anything like that because there's really no evidence that you get benefit from antibiotic suppression forever.

And I've had plenty of patients who had a late adverse event or just decided they were tired to have taken antibiotics and wanted to stop.

And I give them my blessing.

And so far have all done great.

Awesome.

So upon hospital discharge, he has transitioned to oral levofloxacin with rifampin for a three month course and then switched to oral doxycycline for prolonged suppression given his immunosuppression and recurrent infections with MRSA.

I wanted to say thank you to you both for coming today. And this is one of the topics that I get emailed about all the time as a specific request. And I wanted us to save it up and have a really thoughtful episode on it. So I'm very grateful to both of you for creating this. I always end by offering, are there other take home points that you either want to really emphasize and point out or anything that you think we missed, that you want to make sure that the listeners hear.

Thanks for having me on.

I really enjoyed this talk.

I don't think there's anything you missed.

If I could call it a few specific points, I would say if you're still routinely prescribing, like six weeks plus of IV antibiotics, I mean, yes,

Hey, listen, you do you, you're also an infectious disease expert.

But also go check out OVIVA

again, that was a really well designed trial.

And again, it's in the context of all the data we had prior to that

also saying the same thing.

Patients truly don't like having their PICCs.

They're not only expensive, but they're burdensome.

It's tough to take those to work.

They can cause life threatening adverse events.

You guys already know this, but please get on board the OVIVA train.

It works really well.

And I guess the other thing I'll say is, hey, we should have some new PJI guidelines! Someone should write those,

It's been ten years!