Ep 35. Trainee Publishing Programme 2023: Dr Pallavi Shamdasani

How can we best diagnose septic arthritis in the acute care setting?

Dr Pallavi Shamdasani joins Prof Rene Westhovens to discuss the utility of synovial aspiration and intra-operative synovial fluid and tissue culture for the accurate diagnosis of septic arthritis.

Dr Shamdasani is a dual trained rheumatologist and obstetric medicine physician based in Melbourne, Australia. Her paper was a winning entry to the 2023 Rheumatology Advances in Practice Trainee Publishing Programme.

Read the winning paper: Diagnosis of septic arthritis in the acute care setting: the value of routine intra-operative sample culture

Thanks for listening to Talking Rheumatology Research! Join the conversation on X using #TalkingRheumResearch, tweet us @RheumJnl, or find us on Instagram and LinkedIn. 

Want to read more rheumatology research? Explore Rheumatology and Rheumatology Advances in Practice. 

Transcript

Copy Transcript Copied

Voiceover
You're listening to the Talking Rheumatology Research podcast, brought to you by the British Society for Rheumatology.

Rene Westhovens
Hello everyone, I'm Rene Westhovens. I am an Emeritus Professor at the KU Leuven in Belgium, and I'm a Co-Editor of Rheumatology Advances in Practice. We are here to discuss a paper about the diagnosis of septic arthritis in the acute care setting by Dr Pallavi Shamdasani, if I pronounce well, and this is a paper that was submitted for the special junior researcher programme that we have in our journal. May I ask you to present yourself?

Pallavi Shamdasani
Yes, thank you so much. Hello everyone, my name is Pallavi Shamdasani. I'm the first author on this paper looking at the diagnosis of septic arthritis in the acute care setting and the value of intra-operative sample culture. I'm a dual trained rheumatologist and obstetric medicine physician based in Melbourne, Australia. I'm currently working at the Northern Hospital, a large public hospital in Melbourne, as well as doing private consulting. I went to medical school here at Monash in Melbourne and I've done my basic and advanced training in rheumatology through the Austin Northern and Royal Melbourne Hospitals. I then did a fellowship in medical obstetrics at the Royal Women's Hospital. I enjoy treating a wide variety of rheumatic diseases, but my particular interests are in managing rheumatic diseases in young women and especially through the pre-conception, antenatal and postnatal periods.

Rene Westhovens
Interesting to hear, it's a lot of interest and enthusiasm, I would say. First of all, the subject of your study. I was surprised by that because you know, septic arthritis is not the most preferred topic by rheumatologists. And so I was wondering, what was the real reason and the motivation to start your study or to design your study?

Pallavi Shamdasani
I think that's a great question. I agree. We definitely do not publish a lot of research on septic arthritis as rheumatologists. And that's actually something that drew me to this topic. As rheumatology trainees, especially in the inpatient or hospital setting, we are often called to see undifferentiated patients who are unwell. And for some of these patients, septic arthritis is an important differential and we're often working together with our orthopaedic colleagues here, when it comes to these patients. And I actually began this project as an intern at the Austin Hospital in Melbourne, almost eight years ago now. And I always knew I was interested in pursuing rheumatology, but the first term I had as an intern was on orthopaedics. And because the Austin Hospital is very large tertiary referral service in a large metropolitan hospital with a big orthopaedic unit, we often saw a lot of patients with suspected septic arthritis and they were also referred to the rheumatology team. So that was where that question always arose - was this septic arthritis or was this an another inflammatory or crystal arthritis? And that's really how this project began.

Rene Westhovens
Well, it all starts in the clinic, isn't it? Well, I had actually 25 years ago, I had the same experience. I was also working as a junior doctor in a university hospital and the orthopaedic ward was very close, so I had the same feeling and so that was that I had the immediate good feeling with your work, actually, when I got it as a Co-Editor. Are there a lot of referrals for septic arthritis to your department? Is this common in the region? How is this hospital linked to the referral doctors?

Pallavi Shamdasani
So this is a large metropolitan hospital, it's a tertiary referral centre. So for our study particularly, we looked at patients who came through the emergency department, so presented directly or were referred in by their general practitioner, so their primary care physician, for suspected septic arthritis or, during their admission, it was suspected that they had septic arthritis by the emergency department clinician. That's the primary way we picked up patients, which is not uncommon in large metropolitan centres.

Rene Westhovens
It was the same actually at my place. We had the largest tertiary hospital in Belgium, so I had the same feelings.

Pallavi Shamdasani
Exactly, yeah.

Rene Westhovens
Yeah. Well, this study is a retrospective file study of patients that were, as you said, referred to the emergency department. Is there a very close correlation between orthopaedics and rheumatology? Are you systematically asked by orthopaedics when there is a diagnostic problem or a diagnostic uncertainty?

Pallavi Shamdasani
So I think at our hospital, rheumatology and orthopaedics work very collaboratively together. I think we can see that in this study as well. Professor Andrew Hardidge is the Director of Orthopaedic Surgery at the Austin and he was one of the supervising authors on this paper. So we work very collaboratively not only in clinical settings, but also in research and also with our infectious disease colleagues when it comes to these patients and Dr Marcel Leroi, also on this paper, helped from a microbiology point of view - he was very helpful with the synovial biopsy samples. So there is a lot of collaboration in our clinical care. Patients are often cross referred to multiple teams when they are unwell with suspected septic arthritis. So it's not uncommon for patients to be referred to orthopaedics, rheumatology and infectious diseases in that setting. And we all very much communicate with each other for patient care. And we also have multidisciplinary meetings together to discuss these complex patients as well.

Rene Westhovens
Oh that's good to hear, but I think we should... we have to realise that it's not always the case in other centres, but it should be like that I fully agree. Well, now coming to your study, can you describe the methodology and the ultimate goal of your study?

Pallavi Shamdasani 
So, in brief, basically all patients who are referred to the orthopaedic service at our hospital by the emergency department are entered prospectively into an orthopaedic database by the orthopaedic team. So that database collects information about their demographics, their presenting complaint and their preliminary investigations at the time of referral. So these patients are thought to have suspected septic arthritis by their referring physician, which in this case is the emergency department clinician, on the basis of their clinical features and their risk factors and their preliminary investigation results such as a high C reactive protein. And then, unless contra-indicated those with the highest suspicion then proceed to the operating theatre for a joint arthrotomy or arthroscopic washout, during which they generally take synovial fluid and tissue biopsies for culture. So what we did was we interrogated this database, the orthopaedic database, between 2014 and 2019 for patients with suspected septic arthritis who were referred to orthopaedics. We then went back and retrospectively had a look at their electronic charts or their presenting symptoms, the joints that were affected, the lab parameters, their blood culture results, if they had any bedside or ultrasound guided joint aspirations, pre-intra-operative specimen collection, and then the results of that intra-operative synovial fluid and tissue biopsy cultures. So we wanted to know who gets septic arthritis, what are the clues in their history and lab parameters that can help us, and what do their bedside synovial aspirates tell us, and does proceeding to an intra-operative specimen add anything on top of the synovial aspirate taken when we want to reliably diagnose septic arthritis?

Rene Westhovens
So just one question - so not everyone, I suppose, got a biopsy and the culture of the biopsy. So there's always a kind of selection bias possible?

Pallavi Shamdasani
Absolutely. So when we closely examined the data, we kind of divided it into those subgroups: those who only had an aspirate; those who had both an aspirate and an intra-operative collection; those who had neither. So there is an inherent bias in terms of we're looking at patients who've had both done so we are able to compare the utility of that intra-operative culture, but these are often the patients that you want to know what to do with. You don't have enough of a suspicion and it's a really low probability and you've got another reason, you often don't proceed to theatre anyway. Or if you have a really high suspicion and they're really clinically unwell you often take them straight to theatre without waiting for the aspirate. So it's these grey zone, middle zone patients that are often kind of causing the most distress for the clinician in terms of what's going on. And it's these patients we concentrated on.

Rene Westhovens
Okay, well understood and I think this is very reasonable and very close to practice.

Pallavi Shamdasani
Absolutely.

Rene Westhovens
Yeah. Can you summarise the main - say the key - results of your study?

Pallavi Shamdasani
I guess firstly, we should say for the purpose of the study how we actually diagnosed septic arthritis. So as you know, a gold standard is really hard to define. So, for our study, we defined true septic arthritis as those who had an organism grown on either the preoperative aspirate or on the intra-operative specimen, or both. And if patients had discordant results, for example a positive aspect and then a negative intra-operative specimen, we asked an independent microbiologist and infectious disease physician for their expert opinion as to whether it was true septic arthritis or not. So that's how we diagnosed septic arthritis in our study. So over this study, we had 268 patients with 274 emergency department presentations, and 143 of these patients had both the aspirate and then the intra-operative specimen collected. And of this subgroup of 143 patients, 85 had true septic arthritis. So I guess there are two main stories that this paper tells. The first part is comparing the septic arthritis group and the non septic arthritis group, so the 85 that had septic arthritis and the 58 who didn't. In terms of "what do the septic arthritis patients look like?" - well, we know that the knee joint was the most commonly affected in the majority of patients, and the majority of patients had at least one risk factor, whether that's pre-existing joint disease in that joint, joint trauma or intravenous drug use. These patients were also more likely to then proceed on to get an operating theatre specimen. In terms of helpful lab parameters, the median CRP, ESR and white cell count were all higher in the septic arthritis group, but they were very broadly distributed. So this is something we've seen before and I guess highlights that preliminary lab parameters can be unhelpful. And interestingly, blood cultures were present about 48% of the time that were positive in the septic arthritis group, whereas only about 5% of the time in the non septic arthritis group. So taking blood cultures ideally before antibiotics are commenced is a worthwhile investigation. And once again, the synovial white cell count was not a useful differentiator, which we've seen before as well.

Rene Westhovens
It was classical?

Pallavi Shamdasani
Exactly. The second part of the story is the utility of the intra-operative tissue biopsy. So 143 patients had both of these done, 31 only had the aspirate, and 88 only had the intra-operative specimen. So the majority had both done and that's helpful because that can help us examine the utility of the intra-operative biopsy. So in terms of the added benefit of taking an intra-operative specimen, of the 143 patients were both, there was 63 patients who initially had a negative aspirate. Six of these patients then went on to have a positive intra-operative specimen that was deemed to be significant by our infectious disease technician. So that's the number needed to treat a 10.5 to find a positive intra-operative sample in someone with a negative aspirate. And all of these were on tissue biopsy, not on intra-operative fluid culture, so that's a really interesting result that for those who have a negative aspirate, there is a value in proceeding to a intra-operative specimen.

Rene Westhovens
I think this is a very important message for rheumatologists because many times rheumatologists just think that the synovial fluid investigation can give the answer and you showed that this is actually not sufficient. So I'm really pleased with that, and therefore I think it's an important message specifically for rheumatologists. Do you have any idea about the, let's say, the type of infection that means for instance, you know, you have infections with, for instance, TB that are very difficult to culture and is biopsy there specifically useful or are there other infections where the biopsy is specifically useful?

Pallavi Shamdasani
I think that's a very good question. We didn't present this data specifically in the paper but the most common organisms that were cultured were Staph aureus and E coli. So it wasn't necessarily that it has to be TB for it to be biopsy proven only. Even in other organisms, there is that additional utility of tissue biopsy.

Rene Westhovens
Do we have any idea about specifically risk factors where, let's say, a biopsy was more often pursued, like say, people with immune compromising situations. Do you have any idea about this?

Pallavi Shamdasani
So I think because we looked at just that subgroup who had both the aspirate and the intra-operative specimen collected, majority of those patients did have a risk factor. So whether that was having joint trauma to that joint, having a pre existing joint condition, so whether that's crystal arthropathy, degenerative disease or inflammatory arthritis in that joint, or a history of intravenous drug use. In terms of teasing apart which risk factors correlate that are highest in terms of then getting a positive tissue biopsy, we just didn't have the numbers to be able to really make a clear distinction of those things. But it does kind of highlight the importance of taking a very detailed history because those risk factors are significant.

Rene Westhovens
Just one other question about, let's say, risk factors. Elderly people are, let's say, specifically at risk nowadays, certainly in Belgium. Orthopaedics proceed more often to joint prosthesis in also in elderly people while they are vulnerable. Any idea about risk factors there and the possibility also for getting tissue?

Pallavi Shamdasani
Yeah, I think that is a really good question. I guess in our study, the median age was 65 in the septic arthritis group and 64 in the non septic arthritis group. So make of that what you will - doesn't seem like a significant risk factor. And the knee was also the most commonly affected joint in both. I think the joint prosthesis question is a very important question and unfortunately, we weren't able to answer that in this study, but it's something I'd really like to explore in the future. Because in most tertiary centres, people are a bit nervous about putting a needle to a joint prosthesis and often you need expertise to be able to do that. So actually the number of people in that subgroup of 143 that had a joint prosthesis and then had both the aspirate done and the intra-operative specimen taken, it was only about 15 patients.

Rene Westhovens
Okay.

Pallavi Shamdasani
So it was a very small number that had both done. The vast majority of them went straight to theatre, which is what we would expect in, kind of, clinical practice as well.

Rene Westhovens
Certainly a point of interest for the future and I fully agree. Well, you know, we had a very nice talk. Can you now summarise really very shortly with some key messages for the clinicians the importance of your study?

Pallavi Shamdasani
I think the key messages for me would be that there are no foolproof lab parameters or aspirate cell counts that can reliably distinguish septic arthritis from other pathology. When you're suspicious for septic arthritis, culture from an intra-operative synovial tissue biopsy, and that's tissue not fluid, is a useful investigation but because it will increase your certainty of making the diagnosis of septic arthritis. We can also improve our accuracy by taking a detailed history of risk factors, taking peripheral blood cultures, performing timely aspirations before antibiotic exposure, and taking that tissue biopsy. And I think in the future, I'd really like to look at things like this joint prosthesis group in the immunosuppressed subgroup so there's definitely room for further exploration of that in the future.

Rene Westhovens
Very nice. Well, you know, I think we can conclude this is a very important study - that's why we asked you to do a podcast about this. Any idea, I think you know, now after this amazing results in your centre, do you think that they are going to start to implement this even more so, this kind of tissue biopsy?

Pallavi Shamdasani
So I think because the study has been such a long time in the making and we've been collecting this data and analysing it for a long time, it's definitely changed our practice in our department. We're definitely more forthright in asking our orthopaedic colleagues to help us in organising that intra-operative sample and, more specifically, asking for a tissue biopsy just to make us more confident with our diagnosis. Not only so we can guide our antibiotic therapy and the duration of therapy but also because we don't want to miss other important diagnoses and delay management for those as well. So I think this is definitely practice changing - it has been in our department and hopefully it will be wider as well.

Rene Westhovens
Yeah, I fully agree. Well, you know, I think we have discussed all important points. I would like to congratulate you with your study. I wish you good luck in your future career, and it was really nice to talk with you. I talked with a real expert, I think. Thank you.

Pallavi Shamdasani
Thank you so much. Thank you for your time.

Voiceover
Thank you for listening to Talking Rheumatology Research, brought to you by BSR. Please do rate, share and subscribe through your favourite podcast app.