Functional outcome of routine versus on-demand removal of the syndesmotic screw

Listen to Andrew Duckworth and Tim Schepers discuss the paper 'Functional outcome of routine versus on-demand removal of the syndesmotic screw: a multicentre randomized controlled trial' published in the November 2021 issue of The Bone & Joint Journal.

Click here to read the paper

Transcript

 [00:00:00] Welcome everyone to our BJJ podcast for the month of November. I am Andrew Duckworth and a warm welcome from your team here at The Bone & Joint Journal. As always, we'd like to start by thanking all of you for your continued comments and support as well as a big thank you to our many authors and colleagues who've taken part over the past. We hope that you are continuing to enjoy our podcast and all our knowledge translation work delivered from the journal. This including focus on papers published each month here at the BJJ as we will be doing today, as well as our special edition podcast series that has included our insights from the US series, along with our series with our invaluable specialty editors here at the journal.

So today I have the pleasure of being joined by Dr. Tim Schepers, who is a trauma surgeon at the department of surgery at Amsterdam UMC, and the University of Amsterdam to discuss their study entitled Functional outcome of routine versus on-demand removal of the syndesmotic screw: a multicentre randomized clinical trial, which has been published in the November edition of the BJJ. Welcome Tim and a big thank you for taking the time to join us today. 

Thank you very much, Andrew. Thank you for the invitation. Really appreciate the opportunity to be able to [00:01:00] discuss our latest paper being published in The BJJ. 

No, thank you Tim. 

So the aim of your trial, which was the rodeo trial, which is a, a multicenter randomized trial across 17 centers in Europe, was to investigate whether in surgically managed ankle fractures, if on demand removal of the syndesmosis screw is non-inferior to routine removal with regards functional outcome.

So Tim, if you could just give us a brief introduction to the paper and some background to what the literature currently says regarding the pros and cons or whether the screw removal should be performed or not in these cases. 

Yes, of course. Thank you. Well, during my training many, many years ago, I was taught that the syndesmotic screw needed routine removal. And at that time the preferred moment of removal was even at six weeks. However, after seeing some infectious complications and a few recurrent *inaudible* I started looking into this topic and a survey performed [00:02:00] some 10 years ago showed that almost 90% of the surgeons in the Netherlands remove the syndesmotic screw routinely. And the same was seen in other papers from around the world. 

Complication rates have been shown to be quite high, even up to about 9%, including surgical site infections, recurrent diastasis, and unnecessary removal of broken screws. A more recent study published also in The Bone & Joint Journal in 2016 collected 11 studies looking at the need for routine removal of which were nine K series, mainly small retrospective. No clear justification for routine removal was identified in these studies. And at this time we decided to start a prospective randomized control trial. There was one already by Boil also in [00:03:00] 2016 which was somewhat smaller. But we felt that there was a need for a larger randomized control trial on the need for routine removal versus on demand removal. And that's actually how we got started. 

That's very interesting to me, but I think a lot of people have gone through that same process of, you know, you go through your training and it was standard and you just took the screw out. And actually a lot of people were still doing that up until recently. It's interesting, based on limited, should we say literature, that's out there as you, as you've nicely described so far. Do you have a feel, I mean, that was your, in the Netherlands it was 90%. Has that changed at all? Do you think, or is it different around the world at all. Do you have any feel for that at all? 

Yeah. We're actually looking into that right now. I'm doing a recurrent survey and the data is just coming in and it looks like that because we gave it much attention and really appreciate that 17 hospitals in the Netherlands participated in this study. And without [00:04:00] their help, we wouldn't have been able to do this. So creating the awareness probably already led to some decline in routine removal. And as I said, the literature doesn't really show a great benefit. It does show some complications and it varies in the literature from zero to 9%. On average, I believe it's somewhere around 4% wound complications, which in itself, of course is also not only a big burden for the patient, but also a cost problem for the healthcare and with the government healthcare scarcity, I think we should be really concerned with whom we operate on and who we should probably not operate on.

So I think that's probably the reason why we are currently already seeing some decline in routine removal. However, there are still fellow [00:05:00] surgeons who are very strict in their teaching that the screws need removal because of a fear of breakage and the possible limited range of motion after some time. However I feel that that is not per se the biggest problem in this topic. 

No, absolutely. That's very nicely put and very nicely how it is. I think you're right. There is a slight frame shift in this, but there is still some people who are very adamant that the screws should still be removed. And until there's data like the trial we're going to discuss I think it's hard to convince them. 

So if we move on to the trial itself and the study design, it was a pragmatic multicentre randomized controlled trial at the 17 European centres that we said of which 14 were teaching hospitals and three were academic level on trauma centre.

So Tim, first off, I suppose, very key to any study of this nature. What were the inclusion and exclusion criteria and why?

So we [00:06:00] decided to include only adult patients for one reason, of course, that it's easier to have adult patients in your study for ethical acute syndesmotic injuries. So no chronic syndesmotic injuries. They needed to be treated surgically with a metallic screw. Absorbable screws are not encountered in the Netherlands anymore but we decided to add that as well. They should have been operated within two weeks to make sure that it was not a overlooked injury. Or even being a increase in delay to make sure that there's a bit of homogenity in the treatment of these patients. And we included both the isolated injuries without fracture of the fibula and ankle fractures with syndesmotic injuries. And the exclusion criteria were poly trauma patients. Usually [00:07:00] because they have well, much more problems with their rehabilitation including neurological injuries where you can lose the focus of the ankle. And that includes as well also the patients with additional injuries at the lower limb where it is a bit difficult to see whether or not the outcome is related to the other injury or to the ankle fracture with syndesmotic injury. And of course they needed to be able to fill out the questionnaires and they needed to be able to go through surgery at least once but also the second time they have to be fit for surgery for a second time to remove the screws if they were in the routine removal. 

Yeah, no, very nice. They're very clear. I think they're all very, very logical in terms of the inclusion exclusion criteria. And we'll go on to how many screens and recruited. But before we do that, sort of in terms of the [00:08:00] pathways they followed, other than the screw removal, was that quite prescribed in terms of the treatment and their rehab or were you quite pragmatic in the rest of their care.

Yeah. We tended to leave the study as pragmatic as possible. There are two reasons: one surgeons don't like to be told what to do and to make sure that we got as many people on board on the study, we decided to let them do their thing as they do best. And I think that worked out quite well.

And the second reason is that so they were free in the way that they treated their patients during the primary fracture surgery. The only thing that we said is it has to be a metallic screw, one or two and it needs to be removed in the routine removal group, between eight to 12 weeks. We decided to add that to make sure that the rate of recurrent diastasis would be as low as possible. [00:09:00] I think that the current literature shifts away from the six weeks removal. So that's actually the 

only one thing that we said that that we would like to see that it was removed between eight and 12 weeks. Yeah, I know.

And that's it, like I say, very pragmatic and so of representative of what we all do day to day. So in terms of your, your primary outcome measure, what did, what was chosen for that at what time points and, and what did you need in terms of your samples? So, yes, the primary outcome was the Olerud-Molander score, which is of course a relatively old score, but it's still the most used score out to date. So the primary outcome was Olerud-Molander at 12 months. And the score was chosen to be able to compare the results of this study with most other studies on the same topic. And also to use the data from the study by Boyle as mentioned before to calculate the effect [00:10:00] size and the the needed power. And because we decided to do a non-inferiority study we estimated that both groups would probably have similar outcome as based on the study by Boyle. 

Yeah. No, absolutely. Absolutely. And like you say, your study sort of determining the total of 196 would have to be included that was accounting for followup and various other things. But, so it was 152 required for non-inferiority, which we can obviously come on to in a minute.

In terms of what about secondary outcome measures though? And what other things did you look at? 

Yeah, so the Olerud-Molander score at three and six months was one of the secondary outcome scores. The other one is the American orthopedic foot and ankle society score which is debated in the literature we know, but it is the second most used outcome score in ankle fractures. So again, for comparison purposes, we thought it would be interesting to add this score as well. In addition, range of motion at the ankle [00:11:00] and complications during the followup, which included also the complications of the surgery, but also complications of the primary reduction and all these secondary outcome scores were taken at three, six and 12 months as well. 

Yeah. Yeah, absolutely. And so in terms of, before we move on to the results just very briefly, what type of analysis did you perform? Was it intention to treat, or did you do anything else as well as that? 

We did both intention to treat per protocol. So we tested the outcome for non-inferiority. We hypothesised that the on-demand removal would be similar as routine removal. And the primary outcome score was analyzed both endangered to treat where patients were analyzed according to the group they were initially initially randomized in and per protocol so many of them were analyzed as well in the group they were actually treated in, including the crossovers. 

[00:12:00] Yeah. Yeah. Okay, great. So if we move on to the results then the 409 patients assess relative eligibility. 197 were eventually randomized with 93 allegated to routine removal and 104 were allocated to on-demand removal. And just under a quarter of those actually had the screw removed within the first year. 

The baseline demographic data was very similar in both two rounds as you'd expect. Participants had a mean age of 47 years and just under 60% were male. But before we move on to the sort of key findings, can you tell us about any crossovers between the groups, Tim, and also what the follow-up rates were like?

Yeah. So in both groups, we had five people crossing over and that was quite similar as in the study by Boyle as well, where they saw two crossovers in the routine removal and five in which wanted a on-demand removal follow-up was available for three quarters of the patient leaving 73 patients in the routine removal [00:13:00] and 79 people in the on-demand removal group. 

Complications of fixation, including those like mal reduction, revision infection, and other minor wound complications, like wound adhesions were 22% in the routine removal at 50% in the on-demand removal.

Right. And so in terms of the primary and secondary outcomes, there were no differences found on either analysis. Is that right? 

Yes, exactly. So regarding the Olerud-Molander score at 12 months, we saw a score of 85 points out of a hundred in the routine removal group and 84 in the on-demand removal. And the non-inferiority test showed that both for the intention to treat any per protocol analysis on demand for removal was not inferior to routine removal.

Absolutely. And there was no other differences at the other time points either where there? 

No, exactly. And also [00:14:00] for the secondary outcome which showed that the Olerud-Molander score was comparable in both groups at three and six months. American orthopedic foot and ankle society also didn't show any difference. And at 12 months, the AOFAS was 87 for routine removal and 85 for on demand removal. Range of motion and base scores were comparable. And there were significantly more complications in the routine removal, 16% uh regarding implant removal in the routine removal group, and 1.3 in the on-demand removal group.

So the 16% in the routine removal group included five wound adhesions, two superficial infections and two deep infections. So those are quite big numbers. 

Absolutely. And significant complications aren't they as well. Indeed. 

Yeah. Yeah. [00:15:00] 

And so finally, before we move on to the implications of the paper, you also did a sort of analysis to look at the predictions of function outcome, which was your primary outcome, the 12 month. And what did you sort of find when you performed regression analysis to look at the factors that were associated with this. 

Yes. Well, let's skip the univariate analysis because we found numerous factors associated with outcome, but the multivariate analysis identified the following factors as predicting worse outcome, which were female gender, a longer duration of the primary procedure complications of the initial fracture treatment. And of course the complications of the implant removal. All of these showed a negative impact on the Olerud-Molander score even at 12 months. 

Even at 12%, actually, that, that was what was interesting to me. It's still maintaining it, even if those complications happened early, which often they do. 

Absolutely absolutely several studies in the literature [00:16:00] show that wound complications, especially deep wound complications, mainly through lower outcome, even after several years after our initial treat, but often can ankle injuries.

 Yeah, absolutely. Absolutely. So if we move on to the implications of this study, Tim, you know, the strengths of it without question in terms of the design, the methodology used, the numbers included and I think it's clearly shown that on-demand removal of the syndesmotic screw is not inferior to receive removal with regards to functional outcome. Pain, range of motion, all those time points over that one year. And obviously, as you've highlighted very nicely, there were significantly more complications with routine removal and in particular wound problems. And only just under a quarter of patients in the on demand group had removal within the first year. So, Tim, what, what do you feel are the real key take home clinical messages of the study, I suppose, considering an end caveating with any potential limitations of the data you want to highlight?

Well, the key take-home message, in my opinion, is that even [00:17:00] though syndesmotic screw removal might be considered minor surgery, it is associated with a higher than expected complication rate in this prospective randomized trial. And in addition, there appears to be no benefit from removing this screw when comparing routine removal versus on-demand removal. And again, only a relatively low percentage actually needs screw removal. 

The main limitation is because we let the surgeons do their own thing follow up, following our local protocols, and we do not have both operative CT scans, for example, to look at any mal reduction which might've influenced outcome. However, it is likely that this may have occurred in both groups equally. There was a one fifth loss of followup. As most patients resumed work and daily living between three and six months, [00:18:00] they were less motivated to visit the clinic again at 12 months. However, even though the loss of follow-up was higher than the 10% we expected, we still reach the power of over 80%. Th e conclusions are still valid.

No, absolutely. I think your comment about the CT scan I would agree I would defend the paper. I think it would have been, it would have been one of these things that would be nice to have there, but I don't think it, I think you're right. I think the, the occurrence that would probably be equal between the groups and actually you would expect the outcome measures otherwise to probably pick up whatever they were detecting. No, I agree. 

And in terms of what we didn't touch upon in the results, but you obviously did a subgroup analysis, which you had powered for before looking at patients over 60 years of age and under 60 years of age. And, you know, what did you find with that? And what do you feel the implications are for that?

Looking at the age group below 60 and above 60 perhaps needs a little bit explanation. Then the reason we chose for this subgroup analysis was to see whether the [00:19:00] usually more active, younger population would be bothered more by retained screw. And it is of, so of course, conceivable that younger patients being more likely to be involved in sporting activities might be hampered by a intact screw at some point in time versus the assumption that a slightly mobile ankle would be less problematic in a elderly population. 

However, the in the group below the age of 60 was 85 versus 85. So no, absolutely no difference there. And then the group above the age of 60 the outcome score was 78 in the routine removal versus 80 in the on-demand removal and both were non-inferior.

I think that adding this subgroup analysis clearly shows that [00:20:00] the fear that younger patients might be bothered more by a syndesmotic screw that is retained is very likely unfounded.

Absolutely. And as you know, I mean, it would be in terms of numbers, it would be interesting, but actually in the older, older age, be surely your risk of complications of routine removal going up, you're giving an older person, an extra operation and all these things, and they're more prone to infection. So no, I completely agree. I think it is a really interesting finding of the paper.

I suppose, to sort of finish off then, you know, we've alluded to it already. And, and sort of alluded to it all the way through Tim is, you know, what are your thoughts on the potential cost implications of this? Not only for the patients, but also for healthcare services, like you say, in our really difficult times that we're going through at the moment.

Yeah. Great question Andrew. We're currently working out the cost-effectiveness analysis which is quite a big task. One can of course, imagine that if only one quarter of the patients actually do need [00:21:00] removal after syndesmotic screw, that this implies a significant reduction in costs. So I can say that probably the paper coming out on cost-effectiveness is going to show a reduction in costs. 

And in addition, I think it improves the counseling of our patients that after syndesmotic screw placement counseling our patients, that routine removal is not necessary and they can wait out and see what their outcome is. And if they're even bothered by the screw after one year, it might even drop the number of needed removal even further. 

Yes. 

In addition about 20% of the patients in this study because of the pragmatic design receive a four cortical screw. And it might be interesting to look into that as well in the future, seeing that three cortical [00:22:00] screws loose or break more often than a four cortical screw, it might be worthwhile to an aggregate three cortical screws to lower the need for implant removal even further. 

Absolutely. I think like you said there, I think the big thing for me is it's our ability now counsel, the patients and say, there's good evidence now that there is there's, there's really no benefit to this. And you may be having a bit discomfort in those three six months, but if we wait this out, it may all settle down and be fine. I think that it just adds such valuable data. 

So Tim I'm afraid. That's all we have time for today. So thank you so much for taking the time to join us. And, and it was great to talk to you and congratulations on a really excellent trial that has without doubt, added to the literature considerably in the area. It was great to have you with us today and thanks. 

Thank you very much, Andrew. And again, thank you for the invitation. 

Thank you, Tim. And to our listeners, we do hope you've enjoyed joining us and we encourage you to share your thoughts and comments through social media and a like. Feel free to tweet or post about anything we've discussed here [00:23:00] today.

And thanks again for joining us. Everyone take care.