50: Bad to the Bone

Thanks to everyone who listens to the show, to everyone who has supported, created, contributed or joined for prior episodes, and to those who have shared Febrile with a friend. Today, I am joined by my co-host Emily Niehaus. Emily completed her medical school training at the Emory University School Medicine followed by Internal Medicine residency at the University of Utah.

Emily was a resident when she created this episode, but since recording, she has now become a brand new ID fellow at Duke University! As our guest discussant today, we are joined by Laura Certain. She is an Assistant Professor of Medicine in the Division of Infectious Diseases and an Adjunct Assistant Professor of Orthopedics at the University of Utah and Chief of Infectious Diseases at the Salt Lake City VA Medical Center.

She received her MD and PhD degrees from the University of Washington in Seattle and her Internal Medicine residency and Infectious Disease fellowship training at Massachusetts General Hospital in Boston. After fellowship, she completed a post-doctoral research fellowship, studying prosthetic joint infections in an animal model, and then moved to Utah in 2017 to focus more on clinical care of humans with orthopedic infections and a little bit less on mice.

She is currently the Vice President of the Muscoloskeletal Infection Society, a national organization of orthopedic surgeons, infectious disease physicians, and microbiologists with interest in orthopedic infections. Welcome to the show, guys, how are you doing?

He's had some mild intermittent chronic back pain for many years, but four weeks ago, this became persistent and it's progressively worsened since that time. He doesn't describe any lower extremity weakness, numbness, saddle paraesthesia, or radiating pain down his legs. He has some chronic issues with nocturia and sensation of incomplete bladder empty. But no new urinary or bowel incontinence and no recent pain or burning with urination.

He denies chills, uh, night sweats, lethargy, nausea, or weight loss. And his review systems is otherwise negative. So on his exam, uh, he has a temperature of 38.1. Blood pressure of 142/82, and his heart rate's 94 and he's saturating, uh, 96% on room. Complete exam is notable for, uh, spinal tenderness to palpation over the upper lumbar spine and range of motion on both extension and flexion was limited, mostly due to discomfort.

He had normal sensation to light touch over bilateral lower extremities and straight leg tests were negative as well. And his notable workup in labs showed a white count of 12.2. A hemoglobin of 10.5 and platelets of 420 and a creatinine of 2.5 with a BUN of 45, which is around his most recent baseline. And his inflammatory markers, he had an ESR of 45, uh, CRP of 12, uh, imaging that he had initially, um, a chest x-ray with no acute cardiopulmonary process demonstrated.

And then an MRI of his lumbar spine was obtained in the emergency department as well. And this showed marrow edema and enhancement within the L2 and L3 vertebral bodies with no notable loss of, um, body height. And he had abnormal signal and enhancement throughout the L2-L3 disc space. He had no epidural fluid collection and, um, he'd also had edema in enhancement within the paraspinous musculature.

So at this point, we are the infectious disease team and called by the primary team who's admitting this patient due to concern for infectious process. Um, so the first question is how does this presentation compare to the usual presentation you might expect for discitis, um, or osteomylitis.

So it's important to remember that if a, let's say this patient presented with just the back pain, like everything else is exactly the same, but no fever, you would still be worried about, um, vertebral osteo. So it's important to remember that a patient was sort of new onset back pain, that's persistent or steadily worsening over the course of weeks, certainly warrants further workup. And usually that workup involves, um, uh, spinal MRI.

Sometimes people will start with x-rays, but really, if you're worried about vertebral osteo, then a spinal MRI is the imaging modality of choice. We often also get, um, the inflammatory markers, so sed rate (ESR) and CRP, um, they are fairly sensitive, but not that specific obviously. And if it's an infection with kind of a low virulence organism, they may be normal or only mildly elevated.

So again, they can make you sort of more or less worried about a patient , but they almost never tell you what to do for your patient, is how I think about ESR and CRP. Um, so in this case, I would say it's a typical presentation. Um, and we can be fairly suspicious that the MRI findings are infectious in etiology.

Uh, he has BPH and he takes tamulosin for that, but has no, um, acute urinary retention issues and osteoarthritis mostly affecting his knees and he takes Tylenol and has intermittent joint injections for that. Recently, uh, he had a hospitalization for COVID two months ago. And during this hospital stay, he was treated in the ICU for a few days, requiring high flow nasal cannula, um, did receive a few days of antibiotics and, um, has now fully recovered with no oxygen needs.

He has, uh, no medication allergies. And, um, his social history. So he reports that he was born in Mexico. Um, he worked in construction between the United States and Mexico for the last 20 years. Um, but then moved to the US full time five years ago, to be closer to his family. He's lived for extensive periods of time in Northern Mexico, Southern California, Arizona, and Utah. He has no history of injection drug use. He drinks about 10 beers a week. He's a never smoker.

Uh, and he did have periods of homelessness while living in the US in the eighties, but very distantly, um, and no known exposure to a person diagnosed with TB. For other exposures, he did interestingly live on a ranch with livestock in Mexico for the last several years before coming to the us, um, with some intermittent exposure to the animals. Um, and he's unsure if he's ever consumed unpasteurized dairy products.

So now that we have all this information, um, my question is what are this patient's important risk factors for vertebral osteomyelitis and how may these risk factors or other components of its medical or social history impact your differential diagnosis for the microbiology of this possible infection?

That's where we typically have, um, wear and tear in our spine. So those are, that's the most common side of sort of the routine pyogenic vertebral osteo. . Um, and so you can obviously get direct extension from a surgical site after spine surgery, but in the person with no prior trauma or no prior known intervention than the spine, usually it's headed to a site of, um, prior, um, wear and tear damage.

So in, and I will say that this patient's, um, chronic medical conditions probably make him sort of on the mild, mildly immune compromised side with his CKD. And so that may also have put him more at risk for infection. now Staphylococcal species are by far the most common cause. Um, but Strep and Gram negatives make up a sizable portion. And so because of, um, the pathophysiology coming from bacteremia, IV drug use is a significant risk factor though not an issue for this patient.

Um, this patient had a recent hospitalization for COVID, so that probably meant he had some sort of lines put in, IVs probably at some point, maybe for his remdesivir or whatever. So in theory, he could have had a bacterimia associated with that. He could have had a superimposed bacterial pneumonia that might have made him transiently bacteremic. Given his BPH, maybe he had a UTI recently that he just forgot about, or hasn't told us about and had bacteremia from.

So I think any of those things could have led to a transient bacteremia that has now caused for vertebral osteomyelitis. So in the US, we are normally seeing sort of routine run of the mill bacterial, um, vertebral osteomyelitis. In other parts of the world, um, Brucella and TB, uh, make up a large proportion of cases. So it's important to keep those in mind.

And now this patient, for example, he's not sure about his consumption of, uh, unpasteurized dairy products, uh, and has, uh, lived in parts of the world where TB is more common. So certainly you could consider TB, um, vertebral osteo or Potts disease. Traditionally that is thought to involve the thoracic vertebrae more often, but obviously TB can do kind of whatever it wants. So just cuz it's not thoracic doesn't mean that it's not TB.

Of note, I was, um, in preparing for this, uh, podcast, I was looking at studies and there was a study from Turkey, looking at their cases of, um, vertebral osteo and 45% of them were Brucella and 29% were TB. So just to get, which of course is not at all our experience in the US. Right. Um, and I also should note that Brucella, my husband calls it Mediterranean cheese disease because he, uh, helped me study for the boards.

And that's one of the few things he remembers is that Brucella is Mediterranean cheese disease. um, that and tularemia is aerosolized rabbit disease. But in any event, um, so, uh, in this patient, Brucella and TB, you know, possible. Um, so worth considering you could send an interferon gamma release assay. Granted that is not, that's a test for latent TB.

You should not be using it to diagnose active TB, but actually in that study from Turkey, the, a positive PBD had a sensitivity of 0.66, and specificity of 0.97 for diagnosing TB vertebral osteo, which is better than I would've thought. So in a population where it's common, maybe it is helpful. This patient also has risk for endemic fungi, especially Cocci[dioides], um, having spent time in Southern California, Arizona, et cetera.

So that certainly can cause um, vertebral osteo , Cocci can cause bone and joint infections. Um, but it's pretty darn rare. So you could consider sending Cocci serologies or looking for other, you know, uh, serum markers of endemic fungi. But again, most likely this guy is gonna have a routine run of the mill, um, standard bacterial vertebral osteo.

And, um, I just wanna start broadly, what workup and treatment would you first recommend, um, for this patient?

Cause again, IDSA guidelines would say, um, that if you have positive blood cultures with an organism that is known to cause vertebral osteo such as Staph aureus, or Staph lugdunensis, then you can stop there. You can just assume that that is the culprit bug and you do not need to pursue further, um, diagnostic workup in terms of identifying the culprit pathogen. Um, so you definitely would want blood cultures prior to starting antibiotic.

As we discussed above, you might send some other tests depending on risk factors. So test serum markers for, and serologies for fungal infections. If you think it might be that again, you can send an interferon gamma release assay, though. You might get yelled at, by other ID attendings who insist that that is for latent TB testing only, and you should not be sending it in patients where you're considering active TB, but really this person is probably headed for, uh, a bone biopsy.

So probably you're gonna be calling your IR colleagues, uh, to get a sample of the infected tissue to send for culture and path if they can. Sometimes they don't get a ton of tissue. So you're sometimes limited by amounts.

Um, obviously if there's abscess, you want them to drain the abscess and in a pinch, maybe you might end up with a surgical biopsy, but almost certainly the first step is a CT guided biopsy and you should be holding antibiotics in a stable patient until you have a microbiologic diagnosis. That said, I mean, so there's a lot of emphasis on holding antibiotics until you get the biopsy done. A single dose of vancomycin or whatever is unlikely to actually affect the biopsy results that much.

But don't tell anybody that because no, just tell them they keep holding antibiotics until you've got those tissue samples and potentially until you actually have a diagnosis. Because the sensitivity of these bone biopsy is not great. There's a huge range out there in the literature, but maybe 50 50 that you're gonna get your answer from a single biopsy. And so then. if that biopsy is negative, right?

Doesn't give you the answer or grows like one colony of Staph epidermidis, and you're like, well, is that real? Is that not real? You're left, basically trying to convince people to do it again. um, or convincing the surgeons to do an open biopsy or just treating empirically. So I think there are studies that show increased yield with a second needle biopsy, um, or an open biopsy.

I feel like in practice, usually they get their one biopsy, as soon as they're done, they get started on empiric antibiotics, and sometimes they'll get the answer and sometimes they just get treated empirically.

So at that point, his antibiotics, he was, uh, started on vancomycin and now the ortho team signed off because there's no plan for surgical management, just medical management. So the question is what is our medical management and how should we approach this, um, duration of antibiotics and monitoring going forward?

I feel like in the post OVIVA era, you can switch to PO whenever you want, if you have antibiotic that has good oral bioavailability, that will treat the infection. Of note in that Lancet paper looking at duration of treatment, the most common regimen was actually a quinolone plus rifampin so not IV antibiotics. In this patient, um, his kidney disease may make it a little bit challenging to treat, right. So vancomycin is fine, but may be difficult to monitor.

You're gonna wanna be real careful that he doesn't get supratherapeutic on his vanco. You could consider daptomycin instead. High dose Bactrim (trimethoprim-sulfamethaxazole) I'd probably avoid in this patient, that seems like it could be challenging, uh, in somebody with renal disease. If his isolate was susceptible and his drug drug interactions didn't preclude it, levofloxacin + rifampin could be an option for him as well.

Linezolid is pretty hard to tolerate for six weeks and tedizolid is pretty hard to get. So I think for him, he is probably likely to be discharged on vancomycin and then monitored thereafter. Now a lot of people will trend, uh, sed rate and C R P uh, to try to give, to see who's responding and how things are going. And they can give you some sense of who's responding to therapy, but it's not clear what to do if they don't respond.

So there's data that like, if they don't fall, those are patients who are likely to quote unquote, fail therapy or have a recurrence of infection after you stop. But it's not clear that treating them for longer is gonna solve the problem or whether they just need surgical debridement for source control. So we, we look at them, but again, they can make you more or less worried. They never tell you what to do.

I think, and in speaking of recurrence, I mean, you do wanna make sure that the patient didn't have an undrained abscess, right? That's something that might mean make them more likely to fail medical management only if they had an undrained abscess, um, also renal disease and diabetes can make patients more likely to fail. So this and the fact that this patient has MRSA, all of that is a little bit worrisome in terms of his overall prognosis.

So I think in general, for this patient, you would probably put 'em on vanco[mycin] or dapto[mycin],. Check blood work weekly and see how he responds clinically. We typically do not get repeat imaging. A lot of patients will ask like, well, shouldn't I get an MRI at the end of my therapy to make sure that I'm all better and the answer is no, you should not do that.

Uh, especially if the patient is feeling better, if their pain is getting better, their inflammatory markers are coming down, et cetera. Um, patients get better long before their imaging does. And so the imaging will likely, still be abnormal at the end of those six weeks, so if everything else is pointing in a good direction, you just stop therapy. You do not get repeat imaging.

Now, if you are worried about someone like their inflammatory markers aren't coming down, they're having persistent or worsening pain. that's a time when I do get repeat imaging, because maybe they have developed an abscess, maybe they've developed a drainable fluid collection. Uh, so that would be important to know because that you're not gonna solve that problem with antibiotics likely, you need your surgical colleagues to intervene.

Um, he described, uh, new radiating lightning, like pain down his left leg, um, and his back pain just wasn't better. He didn't have any fevers or chills that he was describing and his vital signs were normal in clinic. His exam was similar to when he was first evaluated. It's like spinal tenderness. Um, and then his inflammatory markers kind of throughout his course did not, um, change drastically. And actually at this six week point, his ESR is 40 and that was 45 initially. And his CRP is 15.

So that was 12 initially. And then now, because of concern for an unresolved infection, a repeat MRI was obtained and that showed progressive end plate destruction of L2-L4 and a new epidural fluid collection, um, with a phlegmon or abscess at the level of two to three, and this was contributing to spinal canal stenosis.. So at this point, orthopedic surgery was called and reviewed the imaging and he ended up being admitted to the hospital for a planned decompression infusion of his spine.

Um, the surgery was performed and then he had operative cultures of both the abscess and the bone, again, that grew MRSA. And he was started back on vancomycin after the surgery. Um, but didn't receive any antibiotics beforehand. So now this patient has had, uh, initial failed course of treatment. Um, and now has hardware implanted. So how should we address ongoing as a treatment failure at this point?

Um, and so now he's had a surgical debridement, so that's good that probably gonna help with his source control, but they've also stuck a bunch of new hardware directly into that infected space, which always makes us worried, um, about, uh, creation of biofilm on that on the hardware, making it harder to, um, eradicate the infection. So this patient is looking at another prolonged course of antibiotics.

Exactly how long is not clear, but there was, um, a retrospective study of patients who had had hardware placed into an infected space, who had pyogenic vertebral osteomyelitis, and needed hardware placed for stabilization of their infected spine. And in that study, longer was better. So there were fewer recurrences in the patients who got more than eight or who got at least eight weeks of antibiotics after their surgery. So I usually do at least eight weeks in these patients.

whether or not to add rifampin often comes up and I will say, I have also surveyed my ortho ID colleagues to ask them whether or not they would add rifampin in situations such as these and some do. And some do not.

I think most feel that if there aren't contraindications to it, like they're not on Coumadin or some other, um, difficult to manage drug, drug interactions, and they tolerate the rifampin, it's probably worth adding it because it has well it's controversial and they're definitely the pro rifampin and the anti rifampin teams in ID.

I think there is data out there to suggest that it does help with biofilm related infections and therefore if your patient can tolerate it and there are no contraindications, it's reasonable to try. I do not uniformly add rifampin in this scenario. Um, but I do try to treat for at least eight weeks.

I just kind of wanted to explore additional like hypothetical alternative situation that's a very common reason for consult as well for, or, um, spine infections. Um, so we kind of discussed someone who had surgery and hardware implanted for a primary infectious process, but wanna discuss the alternative situation where someone had hardware implanted for a primary orthopedic indication and then subsequently developed infectious symptoms at that site.

So, how would you approach that scenario differently than what we've already discussed?

Like I keep looking and I feel like maybe I'm U I must be using the wrong search terms or missing something because I keep trying like, surely someone has looked at this in some sort of thorough way, but no, I mean, as far as I can show the evidence out there is mainly, um, retrospective data from single centers and everybody's defined things a little bit differently and what they consider early infection versus late infections, superficial versus deep etc

so it's really challenging because we don't have any great data to guide our management of these patients. Now it is, um, true that some people will distinguish between, um, a superficial infection versus a deep infection. So superficial meaning that it was sort of above the fascia thought to be mainly just a skin and, and superficial soft tissue infection after surgery. And those are usually treated with a relatively short course of antibiotics.

But what we're usually consulted on is when there's concern for deeper infections. So infection of the surgical site that goes below the fascia, where they're concerned that it's an involving the bone and, or the hardware, or at least was really, really close to the bone and the hardware . And so they were worried that, uh, that bacteria are gonna, um, weasel their way in and like again, cause a concern for a chronic infection around the hardware in the bone.

So that's usually what we're talking about is these deep surgical site infections and usually early onset ones. So usually ones that are presenting within the first three months. Um, and usually even sooner than that after their initial fixation surgery. Obviously, yes, people could have an infection at the site that would present further out from surgery, but that is much less common.

So usually what we're talking about is a relatively early, um, surgical site infection after spinal fusion surgery, where there's concern that the infection has progressed to involve the deep space and therefore could be involving the bone and, or the hardware. so in terms of treating them, obviously the surgeons typically take 'em back to the, or open everything up, wash it out as best they can, take samples. And that helps us know how to treat these patients. Cuz we'll know the bug.

In terms of duration, uh, that is anyone's guess . And again, on these, uh, retrospective studies, the range of treatment was quite broad. Um, some people will treat for six weeks, some people three months, some people keep patients on oral suppressive therapy for six months to a year, or maybe even indefinitely.

And I think for me, the decision on how long to treat people often is the shared decision making with the patient and the surgeon, because I don't know when every last bacteria has been eradicated from their spine, there is no test that's gonna tell me that they are cured and some fraction, it may be a small fraction, but some fraction of these patients, especially if they had a Staph infection may relapse after we stop therapy.

. And so I have to have that conversation with the patient and the surgeon about, well, if we stop antibiotics and you have a relapse, what is that gonna mean for you? What surgery are you gonna need to have? And what would that mean for your life? So we'll often have a shared discussion about what the surgery would mean for them and how much they hate being on the antibiotics that they're on. And that will help guide, uh, our management. I also take into consideration the age of the patient.

So for example, if I have a otherwise healthy 35 year old who had spine fixation surgery, and it was a relatively small, um, section of their spine, Not a cervical to pelvis fixation, for example, but like a healthy person who had, you know, maybe three lumbar vertebrae fused, got an early postop MSSA infection got washed out and I put them on levofloxacin and rifampin for three months. And at the end of that time, they're feeling great.

Their inflammatory markers have been normal for eight weeks. Um, and they've had no issues then. Um, I usually stop and see how they do. In someone who was slower to respond to therapy or in whom I am more worried about what another surgery would mean for them medically, like they're more frail, they're elderly, et cetera, or their spine surgery was much more extensive to begin with. I am more likely to keep them on long term oral antibiotics.

I also will sometimes keep patients on oral antibiotics or some kind of suppressive therapy until their spine has fused so that, um, if the infection were to recur, the hardware could be taken out, but spinal fusion can take six to 12 months from the time of their initial surgery. So that's not a particularly short timeframe.

And every time it just kind of feels like you're making it up still. And maybe one day we'll feel a little bit more confident, but I'll think of a different way to say it other than we're making it up. We're making our best guess. But in many ways, that's, that's kind of what we're doing.

I think what I usually say to patients is some variation on, I have no way of knowing whether or not you're cured. If we stop antibiotics and your infection comes back, you will need more surgery. what do you wanna do? yeah. If I have the data to tell them what I think the chances are of their infection coming back, I will share that with them.

So for prosthetic joint infections, for example, we can often say like, well, typically people treated with a two stage exchange have a recurrence risks of X or whatever. Unfortunately, we usually don't really have that data. And so I'm kind of taking my guess of what I think their chances are.

So whether that's neurosurgery or ortho spine, um, especially in a patient who is sort of failing medical therapy or about whom you are worried will be failing medical therapy, it's worth having that, um, conversation, um, with the surgeon because often, you know, the only real indication for surgery or one of the first indications for surgery is that they are having neurologic changes or severe, um, compression of the spinal cord, but sometimes you need surgery for source control.

Sometimes it's important to have that conversation with the surgeons and explain why you're worried that antibiotics alone are not gonna be adequate in this patient or why you're worried that antibiotics alone are not working in this particular patient.