The BOSS Studies: Perthes' disease and SCFE

Listen to Andrew Duckworth and Daniel Perry discuss two papers published in the April 2022 issue of The Bone & Joint Journal.

Papers discussed:

The British Orthopaedic Surgery Surveillance study: Perthes’ disease
The British Orthopaedic Surgery Surveillance study: slipped capital femoral epiphysis

Find out as soon as the next episode is live by following us on Twitter, Instagram, LinkedIn or Facebook!

Transcript

BJJ Podcast April 2022 - DP

[00:00:00] Welcome everyone to our BJJ podcast for the month of April. I'm Andrew Duckworth, and a warm welcome from our 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, thanks to our many authors and colleagues

who've taken part. We hope that you're continuing to enjoy our podcasts. And all our knowledge translation work delivered from your team here at the journal, our podcast continues to focus on papers published each month here at the BJJ. As well as we are special edition podcast series, that includes our Insights from the US along with the Specialty Editor series with our invaluable, Specialty Editors here at the journal, both of which we're continuing into this year.

So today for our monthly podcast, we're actually discussing two papers. And I have the pleasure of being joined by NIH research, professor Dan Perry, from Alderhey Children's Hospital in Liverpool to discuss two studies that have been published in this month's edition of the journal. The first is the British Orthopaedic Surveillance BOSS study on Perthes' disease, the epidemiology and the outcomes from a prospective nationwide UK cohort study.

And the second related study is again, a BOSS study on slipped capital femoral epiphysis, looking at the epidemiology and two-year outcomes from a prospective [00:01:00] nationwide UK cohort with linked hospital administrative data to maximize case assessment. Welcome, Dan. It's great to have you back with us and a big thank you for joining us.

Thank you so much. Thanks for having me. But I thought we would start off with the Perthes' paper that you've published the aim of which was to explore it, inform the epidemiology and treatment of Perthes' disease, looking at disease frequency case mix. The interventions used as well as some surgeon and patient reported outcomes and safety.

So Dan, just, just briefly, if you could, as an introduction to the paper and maybe some background to the current literature on the management of Perthes' disease currently in the UK and any current treatment trends that there are out there. Yeah. Sure. Okay. So, so perhaps, perhaps if we start right at the start about why, why we did what we, where we sat about these two studies together.

So the two studies together were, so when I started this, I wanted to do trials in Perthes' disease. No wants to do trials and slip comes a feminine emphasis, but one of the big challenges in doing trials. Is that we don't know a lot of the fundamental details. So we don't [00:02:00] know much about the incidence, much about a case mix.

And so, so we don't, we didn't have a really solid feel for any of that information. And before, if we're going to write a research grant to, to get a million quid or 2 million pounds, you know, however much trial costs that, then we need that fundamental detail and we'd need it really, really nailed. So the reason we set up both the studies is really to answer all of the feasibility questions that we need in order to inform a trial.

And the best way we thought to do that was to do a big nationwide collaborative approach because then even if a trial wasn't feasible, we've got the most robust, best data possible in order to inform it. And actually we just copied other people. So we copied the obstetricians who already do this through something called you UCOS, which is a big nationwide case finding exercise.

And we modified that, adapted it and brought that to the fruition, which is how we came up with the BOSS studies. So in terms of Perthes' disease, you kind of, you probably remember from, from your registrar days, everyone kind of does something different. And everyone's got really, really solid belief why their [00:03:00] way works.

But actually what we, well, we can kind of come to conclusions in, in a minute, but, but everyone kind of does something different and additionally different for, for really good reasons. But, but there's no really good reasons behind any of it. It's interesting that, that is there. I mean, I think, like you say, I think you taught a certain way in the center that you are trained in and that's it.

And you assume that that's what everybody does throughout the UK. And as we'll see that, that certainly isn't the case. And just before we move on to this study, obviously there's, there's a lot of, as well as the, the, the platforms that you're describing, how the BOSS study was created. There's a lot of setup to these studies.

They're very much prospective studies. Aren't, they're Dan with a lot of patient and public involvement as well. Yeah, no PPI public and patient involvement is, is really, really key to the way that my studies work. And that's because it's so helpful. And also in part it's, because that's what the funders and the National Institute of Health Research, well, that's what they really want us to do.

So, so these are, you know, these are all published open access within the journal because we want, we want. Everyone around the world have access to them, but, [00:04:00] but very much patients have access to so they can, they can read what's going on with absolutely. And as you state in your paper as part of the James and priority setting partnership the treatment of Perth diseases was, it was one of the top research priorities in children, orthopedic surgery.

We're so very much answering that, that question. So if we move on to the study design and there are obviously similarities between the studies and the next one we'll discuss on SCFE, but for, for this study, how were the patients sort of identified briefly and who was included. Sure. So we, we included all new cases of Perthes' disease.

So all cases of, of newly diagnosed, Perthes' disease presenting to someone that was going to be able to definitively manage them. So all kids were between zero and 14 years old, and it was the first time they, they, they went to visited a clinician, able to definitively manage it. Yeah, absolutely. And in terms of sort of, I suppose, sort of checking the robustness of that collection, you crosscheck cases and you did it subtly differently between the two studies.

How did you do it for the Perthes' one. And did you find it, [00:05:00] find it useful? Yeah, so the Perthes' one we did using an independent network of trainees. So, so we got a group of trainees throughout the country and very much took hold of the 20 collaborative groups. And so. So we are, so we asked trainees to, to, to every time they saw a new case of Perthes' disease to log it in and out.

And then what else happened is every month we emailed up to four clinicians every site and said, have you, have you seen any new cases of Perthes' disease? And. And we asked them to confirm whether they'd seen any new cases. And indeed, if they hadn't seen any cases at all, they had to tell us about that as well.

So, so the system the automatic system, we developed emailed every, every site. So 144 sites emailed them once a month. And it kept emailing them in a really pleasant reminder way, until they told us every single month about whether they'd seen any cases or not. So it was quite a robust mechanism to, you know, it was the most robust way we thought we could do the Perthes', one.

So we've got this independent network of trainees. [00:06:00] We've got this this independent network of consultants. And if a trainee spotted one that's that the consultant hadn't told us about that, then he automatically emailed the consultants and said, oh, you haven't told us about this case. Really? And so that was as good as we could make it.

I think. Absolutely. I think I'd like to say for something like this, it's as robust as it can be. So if we move on to sort of the data collected and what was collected, what were the main things you wanted to look at and including, I suppose, sort of outcomes and the problems you looked at. Yeah, sure. Okay.

So, so the, the kind of primary aim was to have a really good, robust idea of incidents and case mix. So, so number of cases and, and, and what those cases look like in terms of that the stage of disease that it present and and how they present. And so that was the kind of fundamental initial data collection.

And we followed all those cases. To two years and we followed them all up to two years with surgeon reports and information. So in Perthes' disease, there was, we asked for details at one year and at two years on how those cases were doing. And [00:07:00] also in some centers, we got them to collect patient reported outcomes.

And so patient reported outcomes were, were linked to that initial data to, to, to those surgeon imported data. Yeah. And when you say certain hospitals that, but just if they sort of agreed or they had the capabilities to collect those problems. Was it Dan? Yeah. So, so it was very much so, so in order to collect prompts from patients, we needed separate approvals from the hospitals.

So we couldn't just open that in all hospitals. If I were to do this again, I'll do it slightly differently. So I wouldn't need that initial permission from the hospitals. But this is all part of the learning process. You know, when we started this children's orthopedic surgeons had never done any research.

I know we're kind of getting better at it. And this is our first really big, getting better at it. Yeah, absolutely. So if we move on to the analysis Dan, before we come onto the results I, wasn't going to go into too much detail here, but just really looking at the analysis you performed and why in particular looking at sort of the, the models you, you looked at.

So, so we try to keep it nice and simple. So we use lots of [00:08:00] descriptive statistics to, to try and just explain, just, just identify what's going on. We then did a regression model to try and understand whether so well mainly w what, what makes a difference. So what makes a difference to the outcome?

And then we also did some kind of interesting stuff that I quite like doing, cause it which is called recursive partitioning. And it sounds really complicated, but basically what you do is you, you, you, you use an algorithm to look at how you can easily most easily describe the data, just using simple flow diagrams.

And so you can kind of try and understand what's going on in surgeons mind. Using loads of data which is kinda cool. And and I thought that was quite useful. No, absolutely. I think we'll come on to that, but obviously there's the figure one in the paper and I think it is really cool. It's a really nice way to explain what potentially will be quite complicated to interpret, but it's a really nice, nice diagram, especially, but we'll come on that.

And also you looked at sort of the, the sort of the there's the multivariate analysis looking at sort of baseline and treatment, in fact factors on that two year radiological outcome come, as you say. So if we move on to those results. There were 371 [00:09:00] children with 396 hits, which are newly effected by Perthes' disease.

And that was from 63 hospitals with a median of two cases, per hospital and the annual incidence was about 2.5 cases per a hundred thousand for the, the, the, the patient population, which was up to 14 years of age. And just over a third, 117 were treated with surgery. So, first of all, what did, in terms of the case mix and the sort of disease characteristics, what did you find and was that suppose what you'd expected?

Yeah. So, so it kind of was what we expected. So, so looking at so looking at, so I've done lots of work in epidemiology of Perthes' disease in the past. And so looking at the, the incidence data, the incidence was perhaps a little bit lower than what I'd expected. So we, we found two and a half cases per hundred thousand previously w previously we've seen about four or five.

So some of my work in the UK, my previous work in UK is looked at as found 4.8 cases per a hundred thousand. So it perhaps was a little bit on the lower side than what I'd expected. But having said that, you know aside from that, it was exactly what, [00:10:00] what we'd expected to find. We find we've had.

W what I find exciting, which is a log normal distribution in terms of age, which is a bit geeky. But, but all that means is that, that you have a kind of peak onset around five years old, and you have so there's that, there's about the same between between three and five years old as there is between five and five and 12 years old.

And in terms of, it's interesting, just going back to the incidence very briefly. I mean, there is some evidence that, that might, that might be correct though, isn't it? It might not be, the cases have been missed just to clarify, there is a feeling that maybe the incidence of Perthes' is declining. Is that correct Dan?

Yeah, so we showed in multiple data sets that the incidence of Perthes' disease isn't declining. So Perthes' disease is very much a disease of deprivation. So disease of of, of lower social conditions. And that's not to say that everyone, you know, th th the all children with Perthes' disease are from poor backgrounds is very much like it's very much like a lot of cancers and in deep death there's you know, anyone can get it, but it's just slightly more common in things in those social groups.

Absolutely. Absolutely. I think in [00:11:00] terms of the treatment decision making, if we come on onto that, I thought that was really interesting in terms of not only if you could sort of describe what the interventions were that people got, but I suppose more importantly, But the treatment, decision-making how this may be evolved, evolved over that two year period for the clinicians.

Yeah, no, that, that was, that was really interesting. So, so what, what were you asked at first is we asked whether, so when people first presented to the 350 or so cases, we asked how many got a plan in place? 249 got a plan in place at the very beginning. And we asked, so we, we, we fitted that into what, what we perceived was what, what was the common that the common.

H how, how we perceive it is commonly done. So, so, so commonly w we know that we, at least we thought we know that surgeons are very reluctant to operate on the kids under six that we thought that there was some some basis on what the x-rays looked like. And some, some basis on whether the kids were stiff or not.

And all of these, we thought fitted into a [00:12:00] kind of decision model. And so, so with a preconceived decision model, we, we fitted all the data together. And, and quantified on how many kids had surgery and how many kids didn't have surgery based on the pre pre-specified decision model and, and said in each group, what proportion were receiving surgery and what proportion

weren't to try and understand. Oh, what the variation was and whether we could predict what the variation was likely to be. And the summary was we couldn't, we were really bad at it. So, so based on previous decision models, at least what I thought was best practice, it kind of just didn't work. It just didn't make sense.

And then we asked the computer to make sense of what was happening and when the computer made sense of what was happening using recursive partitioning. So what it does is instead of taking what we think is the right decision model. It looks through all the data points we've put in at what was the most predictive of, of getting surgery or not getting surgery.

And basically forget about age, forget about x-rays. It's just about stiffness. [00:13:00] So if the kids were stiff and then they were more likely to get surgery, and if they weren't stiff than they didn't get surgery. Yes. I think it's fascinating that Dan, and that figure one, like I say, I would encourage all our listeners to go and look at it.

I think it's really interesting how that breaks down and like you say, it just, it just really comes down to stiffness any more than anything. It's, it's really, really interesting. And if we sort of move on to sort of the outcomes I mean, social reported outcomes, you had that in eight to 9% of patients, which is two years, which is amazing.

But what did you find in particular on regression looking at this sort of independent predictor of a poor, poor, poor radiological outcome then. Yeah. So perhaps the first thing to say is that that stiffness made absolutely no difference in that regression. Absolutely. Which was my favorite thing.

It was so important to surgeons, but it seemed absolutely not to make the blind bit of difference in the actual results. So what we saw, so, so firstly, I think what's important is what we saw. We expected to see to a certain extent. So, so we saw that younger kids do better or the kids do worse. We saw that the more radiographic collapse you have[00:14:00] 

the, the worse your outcome. So kind of knew those things, but that's quite helpful because it tells us, it tells us that what we're seeing is actually kind of makes sense to previous data, but it also tells us that the girls do worse than boys. So we knew all of those things. So then it tells us about stiffness, which we already talked about making absolutely no difference at all and also surgery made absolutely no difference to it.

Didn't seem to show the blindest bit of difference to, to the ultimate outcome, which was about Stulberg grade surgeon reported Stulberg grade. You're right. Yeah. I w I might just maybe pick up on the surgery when we talked about the SCFE one as well. Cause I, I just want it, but we'll come back to that. But in terms of the, you know, just finishing off the results, you know, in terms of complications, as you stated.

With either treatment, thankfully it was, they were relatively rare. I think what was really sort of was quite powerful to me was the, was the problems, you know, in terms of, you know, this is, I believe is the first study that really has reported problems in Perthes' disease of this nature. [00:15:00] And just quite sadly, really the impact it really has on, on children, you know, in terms of actually quantifying it from their point of view.

It's, it's quite sad. Isn't it? The impact it has. I know, I know it's we assume it would be, but actually having that data. I dunno, it hit me quite hard that in terms of how, how these, how these, these kids are impacted by this, by this disease. No, absolutely. You know, this is a disease which really affects their whole life.

It affects not just their life. It affects the life of their family. And I think that kind of quite clearly come out of it. It's not just the physical health. It's, you know, the emotional, social, you know, school health, like it's, it's every part of their life is is affected by this disease. Comparing it to, so, so we used a tool called PedsQL, which is a quality of life tool in kids and comparing their outcomes to kids with other diseases.

You know, this was this affected their life was as much as any rheumatological condition might be. And in fact, you know, even many of the cancers, you know, this, this, this really affects their life in a really big way. I think, [00:16:00] I think that it's important. Isn't it? Isn't it Dan? It actually is a very powerful message to people like funders saying, actually this is important, you know, we can't, it doesn't sound, I know it doesn't seem it's life-threatening or it isn't, but it's a really impactful on their quality of life at a young age.

And it's, that's why it's such an important disease to. To investigate. So if we go, go on to sort of just put it into context and, you know, I mean the strengths of the study, just without question, you know, in terms of the size UK wide study, very robust methodologically and sound set it with a prospective design that showed, I think, you know, sort of interesting, a widespread variation in the treatment strategies for Perthes' disease of the hip.

And as you said that despite the frequency of, of surgical containment surgery, there's no real evidence of improved outcomes. But also. Also, which we'll come on to is that does appear to be a sufficient volume of cases and, and an equipoise amongst those, I suppose, for, for a trial, as you've alluded to, to begin with.

But what do you feel are the key take home messages? And I suppose maybe caveats in that with any, any limitations of the data there. Sure. So, so I think [00:17:00] so firstly, to address the key, take home messages. So, so I think the key take home message for me. And this is talking as a surgeon that does a lot of Perthes' disease surgery cause that that's kind of my normal I think the key take home message for me is that.

The reason to do surgery is certainly a lot less clear than, than perhaps perhaps I thought it was, or perhaps I believed it was. And so for me, I completely believe a trial is, is, is necessary and I believe the trial's possible. And in fact, the rest of my day-to-day is, is writing the Perthes' disease trial.

Cause it's just been advertised by NHR on the back of this, which is really cool. So I think they're the real take ups as well as the fact that, you know, knowing how, how much this affects the children and you know, how, how we should, how we should best address address that in, in, you know, in clinics.

And I know some people involve psychologists in, in managing, managing these children and stuff. And perhaps that is the way to go because you know, this is a big hit on their life. Yeah. In terms of the methodological problems, I think the biggest methodological problem, the one that, that people are likely to [00:18:00] criticize is we only had outcomes to two years.

And so, so we know that Perthes' disease so w we know that if you're going to classify Stulberg properly and correctly. You, you, you, you should be classifying at the end of, at the end of school to maturity. But there is some evidence to say that, you know, you can classify much earlier than that. And so certainly at five years you can classify it reliably.

I think, I think the, the, the really big useful thing that we, we showed and I pointed out before is the fact that the, although we've classified it two years, all of those things that we said, we're going to be important. So female sex, older children more radiological involvement or having worse outcomes.

Well, they all can make them out in the wash. You know, we saw all of those being important. And so even though we were only classifying it two years, I still think this given us really reliable information on what's actually going on, the only drawback is we can only, we can only look at the data on 70% or so of cases because we could only classify those in the late reossification stage.

So they were the will that we, that we classified an outcome on and only. [00:19:00] Only 70 or 70 or so percent of children were in that, that group. Yeah, the news old dataset, I think going forward, if we do a trial, we'll have to have longer follow-up so probably primary outcome at three years. So we get more of those children and then we'll have to make sure we do a long-term followup as well to, you know, to meet everyone that what we're seeing at three years is actually representative.

No. I agree. And I, I appreciate your saying in terms of, you know, like I just only 70% of the hips are sufficiently mature to, to assign that outcome. But I think like you also put it, I think it's like unlikely to introduce bias, but it's something we, we probably should, should acknowledge instead, before we move on to the next study, just, you mentioned the trial without giving maybe too much away.

What do you think, what do you think that will look like? Or what do you think are the key things to consider in the design of that? So, so the trial is going to be of containment versus no containment. So we're going to look at children almost certainly five years and above. We're gonna look at containment surgery versus, no containment surgery, and the primary outcome is going to be a patient reported outcome with with with a [00:20:00] radiological outcomes as a

kind of first secondary outcome, cause that the primary comes important that the patient reported outcomes important to patients and surgeons are really going, gonna focus on how on, how round the hip is for that longevity, which, which I understand. I think one of the key factors, whenever we do trials, one of the key factors is how we're going to get patients on board.

Yeah. And, and the key, the key difficulty is that these patients are really, really sore that they're often in often they, they, they need to go in wheelchairs for short periods. There, there are lots of painkillers. They're not going to school. And we're saying to the parent. I look, we can do nothing or we can do surgery and that's kind of hard.

So, so because whilst patients don't don't want surgery. They also don't want to see their children in pain. If there's any hope they want to be able to do it. So what we're working on at the moment is we're working on, on what best conservative practice looks like. So what best non-operative practice looks like? So we can really give them a package of, of something good compared to surgery to try and compare the two.

And [00:21:00] we've got a PhD student, at the moment, Adam Galloway who's who's, who's working on that to, to, to really try and sell that or to, you know, really, really try and work on that non-operative arm. So, so it's kind of standardized across the, the pitch. W we can we can sell that to patients as a, as a really good option.

And I think looking at that from this and looking from other studies, I think it probably is a good option. All of the previous evidence that said surgery for Perthes' disease, it was all pretty weak. It was all based on postdoc analysis. So I don't think there's any strong evidence to say that the surgeries actually making any difference.

Yeah, no, absolutely. I think like you say the important thing with any sort of non-operative versus operative trial, I suppose, is you really, you don't want the patient thinking that if it's not operative, you're just not going to do anything. It's actually optimizing that non-operative arm, isn't it. So they're actually getting a package that's really, really good.

Like you say, which is why you need the evidence for it. But so just in terms of trying to, well, let me move on to the SCFE paper next. So this is. Again, one of the BOSS using the BOSS platform and similar BOSS methodology and on slipped capital femoral, [00:22:00] epiphysis. And again, the aims are similar to explore and inform

the epidemiology and treatment is disease. And again, just a brief background maybe to the paper Dan and again, where we are in terms of, of, of standard treatment for it and what the evidence is like. Sure. So, so the, the, the real background to this is when I, when I was elect kind of 10 years ago now in Warwick university, the, we were going to set up a trial of Perthes' disease.

So we're going to, so I'm, I'm, I'm on the wrong disease and it's just that I try to put SCFE and it was going to be called the BOSS trial. So it was the British British, orthopedics SCFE surgery trial. And it was really, really complicated trial. And we kind of, we wrote it and pitched it to the funders and they were the ones that came back and said, look, you don't know all of these information.

You can't possibly do a trial. So. Which is where the, the, the British orthopedic SCFE surgery trial became the British orthopedic surgery surveillance study, because we'd already got the logo. So why would we waste the logo? So so it morphed into that and which is where we, so, so the SCFE one was, was kind of where it [00:23:00] started.

And then we chucked in the Perthes' one for free and it was going to answer all those questions, all those feasibility questions for a trial w which is what you know exactly what it's done. Yeah. So, so the, like you say that the setup was, was very much the same as same as the Perthes' disease, one that the really big, important thing for me and the kind of the geeky me really loves this is, is that SCFE is always admitted to hospitals.

So it comes into hospitals, it gets admitted. And in the UK, we, we collect, we collect details of those admissions. And so we collect. So, so, and this was a really, really, really neat bet. And so every time a patient is admitted to hospital the, the, the coders log, the operation code and we collected all of those operation codes every single month from

statistics or from, from the equivalents in Scotland and Wales. And so we got monthly, monthly downloads from their systems and we, we inputted that into our, into our [00:24:00] system. And consequently, rather than relying on trainees, we could rely on rely on this big national independent network. That's routinely collecting this information for payment.

We can cross check what surgeons reported and we could always ping the surgeon in whatever hospital and say, look, I know you operate on SCFE. You know, there's no point in hiding it and please read, tell me about the case. Yeah. So so, so that's kinda how it works. So one of the problems with coding is obviously it identifies cases where but because they're, they're coding a SCFE obviously it also identifies cases where perhaps they're having a revision.

A revision screw exchange, or perhaps someone, a screw removal, or perhaps having some osteotomy, then it'll still collect that SCFE code. So, so there are cases that aren't necessarily fresh SCFE. So we had the option for surgeons to tell us about those. But look, I think this is the most reliable estimate of SCFE incidence

I could possibly give you. Cause I know that I, I know that we got, we cross check that against more than 95% and it has data. Yeah. Yeah. And I think that's one of the key differences, isn't it? Between [00:25:00] the two studies is, is that it's that cross-checking and how effective that was. And in terms of sort of the people included the ages, they were slightly different weren't they Dan in terms of who you were looking at.

Yeah, so we collected six to 18 years old. And that that's because that's the population at risk. You know, it's very unusual for, it's very unusual for six year old to get, get SCFE let alone anyone before six that's the population risk. And it's also the population that most other studies look at us before.

So we try to unify it with, with previous research show. And in terms of the, the followup again, Fairly similar, but you sort of looking particularly at complications, weren't you and principally with development of AVN in particular, is that right? Yeah. So it is in the same way. So we looked at, in this, we look to surgeon reported outcomes at three months and at two years, and again, we had a subset of hospital

patient reported outcomes. Yeah, absolutely. And before we move on to those results, Dan, in terms of the analysis performed, was there anything in [00:26:00] particular you where you wanted to look at and do some models on? So the big thing was about AVN. So. So the big thing was about AVN and trying to, trying to look at AVN, but look at stability at the same time.

And, and I had big battles with my statistician about how we look at this data because AVN is relatively rare and what you'll see in the literature, you'll see lots of papers, try and really drill down on what, you know, you know, what, what could be the cause of AVN in each individual case. But the statistician is quite ruthless and she she said, look, you've only got a small data set, and you can only work.

You know, it was, it's a massive data set. And the biggest SCFE, data set that there is a relatively small number of cases with AVN and therefore we can't dig into it too much because you're just making, making false, false claims, full, you know, false assumptions and conclusions. And so. So I, I, that was really, I mean, for me actually, because, cause I wanted to, you know, you can, you can really try and read more and more into this and the more you [00:27:00] read into it, the less actual, useful information you get.

So, so it it's very much a high level look at what's going on in the UK. What practice looks like. And to try and pick out some of that kind of AVN stuff. Absolutely. Absolutely. And I, along with you have, you know, she, you looked at the contralateral slipped capital femoral epiphysis, but we'll come on to that in the, into the, in the results.

So we look at, look at results. So 486 children with a new diagnosis over the 18 months periods 17 patients with acute due to lack a base based on data that left us with 469 children with 513 hips that made up the study cohort. Annual incidence was just over three cases per a hundred thousand in that age range.

And the rates were broadly similar across the UK. First of all, you've already alluded to it, but obviously with the sensitivity analysis, it was, it was pretty impressive in terms of how robust the data collection was when you, when you had that, that sort of cross-check wasn't it. The, the sensitivity, like, absolutely.

So, so we know that we collect at least 95% of all cases of SCFE in the UK over or in, in yeah. [00:28:00] In England, Scotland and Wales, which is kind of cool. It's really cool. I think that's pretty cool. And it's such a powerful, and it's, it also reinforces actually that when you have the studies like this and you set them up, like this, people contribute to them, you know, it's not the people forget about them.

And I think it's, it reinforces the methodology of the platform really I think so in terms of. It's I should just thank all the surgeons because this isn't just me. This is, this is 250 authors on this paper in the collaborative. So, so, so this is really what kickstarted children's orthopedic research in UK. And, you know, there's so many good people taking part.

So thanks. No, absolutely. Absolutely. So if we move on, so the, so the case mix and disease characteristics, as we did for the previous paper again, was it what you expected to see? Was it, was it sort of consistent with what we already knew? Yeah, I think, I think it was actually, so, so, so so the age and the, the, the kind of the, the age and gender where we're all kind of looking like, what, what has been, what's been written before?

So [00:29:00] in terms of the, the, the, the, whether they're stable or unstable about about 20% were unstable, 80% were stable, which is kind of what is known of the stable, the vast majority were, were on the milder side with. So with about a quarter of the stable been being severe. So, so it was you know, it was broadly, broadly what we're expecting.

No. Absolutely. And I, and as you also highlight that, so the principal co-morbid association was obesity. As, as again, we've set with that, just over a quarter of the, of the patients having, having defined as obese. So again, I suppose, consistent with what we sort of knew it with about, about SCFE in terms of the, the treatment then in terms of what, you know, what the intervention they had.

Again, it was, it was quite interesting, you lay it out really nicely in figure one, but sort of fairly consistent in terms of how people are treating these. Yeah, and I'm fairly consistent. So, so we, again, we tried to lay it out based on what we thought that what we thought people might be doing. So, so what, what we thought what we thought may influence practice.

So we split it up into stable, and [00:30:00] unstable in terms of the unstable weight ones that. About half of people are treating them urgently. So, so treat them as an emergency. About half of people are saying, look, you know, that there's something going on. Let's or there's some sort of inflammatory soup. Let some, let's leave them in bed for a couple of weeks, which is a very, very UK approach.

And if you've got international people listening to this, they'll say, what are they doing in the UK? Cause if ever I talk about that in America, they just think we're bonkers to leave them in bed for two weeks. So so, so there is that w which originally came from, I think it was Mr. Catterall that kind of popularized that from, from Stanmore.

So so, so yeah, that, that, that gets done a lot and we've got no idea if that truly makes a difference. Then in terms of the mild to moderate, we, we. Particularly looking at whether people were whether people just fixed them in situ or whether people were using different types of screws, like growing screws and growing screws, which allow the, which allow the epiphysis to grow.

So the idea is that you put growing screw and the epiphysis grows and then any, any deformity remodels. And despite that being a lovely idea, [00:31:00] hardly anyone's using. Which, which is perhaps understandable and then the severe SCFE if you were the ones where there was quite a lot of quite a lot of variation as well, whether people were fixing in situ, or whether people were doing big osteotomies to try and reduce it back to where it came from.

Yeah. So that would, that severe deformity group is open reduction was about 44%. Wasn't it? And it's very different. It was only 9% in the mild to moderate deformity. So quite a notable difference there. So in terms of the main outcomes you were looking at in particular AVN, what did you sort of see with that and what sort of risk factors did you, did you see.

Sure. So there, so there, so there was AVN actually in everyone, there was AVN in the mild to moderate group who was AVN in the severe group, the, you know, in the stable group and the unstable group, which kind of surprised me a bit because in fact, there was even AVN in the, in the contralateral side, in the opposite side, you know, people pinning things prophylactically, there was even a case of AVN in there.

So there's, there's AVN everywhere. I really kind of bowled, bowled me over. The, the risk factors for AVN are the ones that we know about. So. So [00:32:00] the risk factors are, are, if you've got an unstable slip, then, then our data said that you've got a four and a half times risk of, of, of getting get getting AVN.

So four and a half before 4.4 times the odds. And also if you do an open reduction and so. So, so if you do an open reduction, you've got about seven and a half times the odds again, AVN. So, so they're the big two risk factors for, for getting AVN, reduction or, or having it initially unstable slip. And then just go into that a little bit.

Do you think as well, is that which one comes first? Is it the open reduction which is causing AVN or is it the fact that, that the open reduction is more common in the severe deformity, which has maybe really more associated. How do you sort of break that down? So this is where, this is what I was arguing with statistician a bit to, to try and break that down further.

So in the end, the, the, the, th th the, as far as I got was to include you all in regression model, and that's, as far as she'd, let me go. So, including in a regression model, so statistically, we should be adjusting for the other thing. [00:33:00] So if we adjust for for both things together in the same model, it tells us that that doing an open reduction.

has got a bigger risk of AVN than actually,

 the slip itself. It may not please everyone, but that's, but, but that's where it's at. And then we try to adjust to say, look, you know, if you're a more experienced surgeon doing the, doing the reduction, does that change your risk of AVN? And actually it didn't, we didn't. No, absolutely. And just as soon as you offer sort of results and I mean, I think similarly, you know, in terms of the PROMs that have a huge impact on marked impact on health particularly evident in that physical health domain, but just the final sort of suppose the, on the other controversial topics or the seafood in terms of the contralateral hip, what did you sort of see in terms of what people did and how that affected outcome?

Sure. So in terms of what people did. So, I mean, variations huge on, on, on what people do to the opposite side. I don't, I don't remember the exact number of, of, of [00:34:00] who fixed the opposite side and what I think 120 underwent prophylactic, fixation. I think it was, yeah, I knew you'd be more prepared than me. So 120 underwent prophylactic fixation, which, which left 200 and something not on prophylactic fixation.

So that was really nice actually. Cause because there was, there was big groups either way that let us do some nice statistics and nice stuff. And so what we saw is that, so firstly, to say that in those that underwent prophylactic fixation, It's not a mind. It is a minor procedure, but it's not a procedure without risk.

Like we say, 120 kids, there was kids with osteomyelitis. There is a kids with with avascular necrosis with kids with fracture. Like it's, it's a, it's a risky, like, yeah, it does have risk, which is which I know it does, you know, I'm a surgeon. I kind of know it has risk, but it really highlights it to you when you see all written down on paper in a cohort, which, you know, W which isn't hiding anything, you know, kind of, kind of cool and kind of, you know, kind of [00:35:00] interesting to see it does, it does inform the consent conversation.

You're going to have about what you're going to do on that prophylactic side. Yeah. Yeah. It's not benign. Is that at all? When you read, like you say, when you read that, no, it's not benign and you know, you may think you're the best surgeon in the world, but you're still going to have complications because this isn't an he's like, this is this, isn't a difficult procedure for anyone.

I don't really perceive the, you know, a lot of experience is going to make a huge difference. I agree. I agree. And then in terms of the ones that we fix in terms of the ones that we didn't fix, we, we, we then did w we did some nice Kaplan-Meier plots to look at the risk of the, to look at the SCFE on that side.

So, so in total about about 10% of those had a late SCFE on that side. And we tried to break that down to to, to look at why that may be happening. So, so broadly there was no difference between boys and girls, but there was a difference if you were younger. So. The younger children had got a much higher risk of of, of contralateral SCFE if [00:36:00] it wasn't prophylactically pinned.

Again, we put that into a regression model. So we put all of the different factors that could predict contralateral slip into a regression model. And the only thing that made any difference was age. Yeah, yeah, yeah. And then the cutoff you found was about 12 and a half years, wasn't it? But that's, that's obviously from the modeling.

Yeah, yeah, yeah. So it's, it's really clever. Actually the computer spits out a model of, of what. W what the best, what the best cutoff is and the computer's model of best cutoff was 12 and a half years. Yeah. W which, which I thought was was a really helpful, you know, really helpful little guide to what might be, what might be a useful tool.

I mean, I know it's an extrapolation Dan, but during then, is it reasonable to say that once your patient is, if the patient is over 12 to 13 years of age, it's, it's reasonable than just to leave it certainly is, that, is the evidence strong enough for that you think I. So if you're going to, so if you're going to organize that, if you're going to study that properly, you'd have to, you'd have to try and weight in the relative risks of patients and the relative risk of SCFE.

But, but [00:37:00] if you're going to take at the simplest level, yes. I think, I think a very simple level that that 12 and a half is a, is a good guidance. And certainly, I think I'm probably gonna use in my clinical practice on whether I should put in the other side or not. It's kind of easy to remember. And it makes sense, doesn't it?

Absolutely. Yeah. I agree. So in terms of finally just sort of putting it all into context again, you know, huge, huge study, you know, great methodology prospective design, and it is really, and certainly from my reading around that, you know, the most comprehensive study in the literature on, on, on SCFE and you know, you've found this association between time to diagnosis and severity of the disease and the study along with some of the barriers associated with

making that diagnosis and also the wide variation in treatment, practical practices and the implications that potentially has for these children. So, Dan, what were your sort of key take home messages for you in terms of, for this study? I suppose again, caveat to any limitations, but what, what, what, what do you want the listeners to really take home from it?

So, so I think what I want to take home with that is, is SCFE, SCFE is [00:38:00] rare in all centers. So, so, you know, the, the average number of SCFE you've seen, I think it was two, three month periods. So, so it's. It's a relatively rare disease in, in, in pretty much everywhere. So if there is complicated SCFE then so complicate SCFE, I mean kind of very young SCFE, unstable SCFE, very severe SCFE.

Then I think it's completely reasonable for it to go to a specialty center. You've got more expertise. And I think, I think the kind of GIRFT approach is increasingly adopting. So, so I think that's, I think that's probably fair and I think most people would agree with that. I think the I think there is a risk of AVN and there's, there's an inherent risk of AVN.

the near risk of AVN and the risk factors associated with it. Okay, fine. So there's inherent risk of AVN and there's obviously multiple multiple risk factors associated with it. One of the things for me, actually, one of the things that the, the, that perhaps perhaps doesn't come through in the paper as much as.

As much as I found interesting was [00:39:00] so we asked people to indicate the delays in getting, getting it a diagnosis. And we have wrote about it briefly, but, but, but we looked at all of the different reasons for the delay and it was absolutely astounding. How many reasons for delaying SCFE there were no, there were lots of different reasons.

And, and broadly it was either the family thinking there was a, a pulled muscle or the GP thinking there was a pulled muscle, but there was also loads of miss cases of diagnosis. That that was, you know, an x-ray showing a SCFE, but the radiologist missing it, the, the A&E doc missing it the orthopedic surgeon missing it, you know, people forgetting to take your frog leg lateral.

There was loads of loads of reasons why SCFE gets missed. And I think, I think if you're gonna, you know, you just need to remember to think about SCFE and kids at this age, kids between ten and, you know, especially 10 and 14 years old, but if the kind of peak age, if they got a knee pain, then then, then think about the hip I know we're all orthopedic surgeons and 

I know we kind of know that, but, but I can prove [00:40:00] to you that we don't all know that. Yeah, absolutely. I think, I think it's one of the one thing I took from it, particularly the discussion you talk about that nicely and actually importance of continuing education and getting to these patients and diagnosing these patients early.

And I suppose that's finally Dan as it actually for the previous one or what's next, do you think what, what Well, it's obviously a trial. What would, what would come next for you in this sort of pathway for the management of SCFE? Yeah, so, so a trial is a trial is, is on the cards. In fact, the trial has been been funded.

And and we're setting up at the moment it's called the big BOSS trial. So the big British orthopedics SCFE, SCFE surgery trial, it's about, it's about severe, about severe SCFE. So we're, we're gonna take all the SCFE so the definition of severe we're using is, is any SCFE, which the surgeon believes is going to result in significant impingement.

So we're going to randomise those between pinning in situ, and, and open reduction. Yeah. So that's the big BOSS trial. That's that's been funded and that's coming. Yeah. Aside from that. So, [00:41:00] so there is this question about the unstable ones about bed rest versus not. I think that's a really, really tough trial.

And if I'm honest, I think unless we've got really, really strong international trials group, is this behind that? I think that's going to be a hard one to answer. And we're going to have to leave that to, to kind of BOSS the observation study to kind of park about there I'm afraid. I don't think there is much for questioning mild SCFE.

So I think, I think people are generally pinning in situ with a single screw. And I think that's the, you know, that that's a very well accepted approach. All these newer screws that there's not a lot of appetite for them from looking at the x-rays we saw. So we, we got all of their x-rays.

If I'm honest, so we have qualitative quantitative data, but, but that wasn't a lot of amazing remodeling happening in the majority. Right. So I'm not so convinced about the growing screws. But, but, but, but that's, but Hey, that, that's just my opinion, creeping into a science talk. So I'm sorry. And what about the, especially finally, what about the contralateral hip Dan?

Is it, could you do a trial on that, do you think? Is that, is that [00:42:00] possible? Yeah. No, I think a trial is possible in that. I think a trial is possible. And, and so, so one of my, one of my supervisors, my, my, my mentor on this, who's an obstetrician. And that's because I it's, because we stole the approach from the obstetrician, so, Mariane Knight.

She's a professor of Public health obstetrics. And her main thing is look why you're not doing a trial at the opposite side, because that's what the data is screaming for. And so I think that's I think that's something we will have to address. Yeah. So, yeah. So w we'll we'll get on with that.

Thanks for reminding me. No worries. I'm all done. That's all we have time for today, but thank you so much for taking the time to join us. I really enjoyed that and congratulations on just really two outstanding studies that are real benchmarks. I think for UK nationwide types of research of this nature and, and have without doubt added so much to the literature and given much, much strength to, to really high-level trials being performed, moving forward.

It was great to have you with us Dan. Thanks so much. Thanks for having me 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 like feel free to [00:43:00] tweet or post about anything we've discussed here today. And thanks again for joining us.

Take care of everyone.