A prospective randomized controlled trial comparing CT-based planning with conventional total hip arthroplasty versus robotic arm-assisted total hip arthroplasty

Listen to Andrew Duckworth and Fares Haddad discuss the paper 'A prospective randomized controlled trial comparing CT-based planning with conventional total hip arthroplasty versus robotic arm-assisted total hip arthroplasty' published in the April 2024 issue of The Bone & Joint Journal.

Click here to read the paper.

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

Transcript

[00:00:00] Welcome everyone to one of our BJJ podcasts for the month of April. I'm Andrew Duckworth and a warm welcome back to you all from your team here at The Bone & Joint Journal. I'd like to start by thanking all of you for your continued comments and support as well as a big thanks from myself and our team to our many authors and colleagues who have taken part in the series which highlights just some of the great work published each month here in the journal.

So today I have the pleasure of welcoming back our Editor-in-Chief here at the BJJ to discuss their paper from University College London entitled 'A prospective randomized controlled trial comparing CT-based planning with conventional total hip arthroplasty versus robotic arm-assisted total hip arthroplasty', which has been published in the April edition of the BJJ.

Welcome Prof, it's great to have you back with us as always. Andrew, thank you and it's great to be with you as ever. 

It's a really interesting paper, you know, that was published this month from your centre. You know, the primary objective of this study was to compare the reproducibility of the planned preoperative centre of hip rotation or the COR, in patients undergoing robotic arm-assisted THA versus conventional THA. And you looked at a variety of other outcomes as well, sort of, acetabular component [00:01:00] orientation, restoration of offset and leg length and functional outcomes and PROMs, which I know we'll, we'll come on to. But just as a start Prof, for our listeners, can you give us sort of a, a brief overview, sort of setting the scene really, if you will, for this RCT on the current robotic hip literature and how this sort of trial came about for you?

Yeah, no, no, thank you. And this is, I should say at the outset, you know, I'm here talking to you, but a trial like this takes a lot of people and a lot of work. And particularly, Jenni Tahmassebi in the background and Andreas Fontalis and Babar Kayani working hard at the front-end, have all delivered what I think is a very useful study.

I mean, in reality, there's two aspects to this worth thinking about. The first is that robotics in hip surgery have existed for some time, but the previous generation of robots really didn't deliver good outcomes and very much went into abeyance and suddenly we've got new technology that seems to be promising and has baseline data that was [00:02:00] attractive to us.

It's the ability to plan on this on a CT scan, the three dimensions, which has become more and more attractive, both in hip preservation and in hip arthroplasty and in hip revision is now with us. And the ability to execute with haptics and with a robotic arm again, seemed to have very good preliminary data.

So we were at a stage where we had access to a technology that we could really start to evaluate against the standard and start to study. And beyond that, there is this misconception that hip replacement is sorted out and it is a perfect operation and everybody does well. Well, the reality is that not everybody does that brilliantly.

We know there's a significant proportion of our patients who have ongoing pain or are not completely satisfied with their intervention. We know that from PROMS data, we know that from registry data and from a number of studies that we've described previously in the journal. We also know that our demographic for hip replacement is changing.

We're getting a younger and younger, more demanding population, who [00:03:00] really want to be active. They don't just want pain relief. They want to do sport. They need a perfect or as perfect as what can be hip replacement. And even our older patients now want to be active. So in reality, we face an expanding need in a more demanding set of patients.

And in order to deliver that, we need to probably up our game in terms of restoring anatomy and restoring biomechanics. And bear in mind, for all those who think that hip replacement is solved, we're still hearing and seeing about patients with instability, leg length inequality, all those related problems that are essentially solved by improved surgery.

So I think there was a, there's a lot to attract us to doing studies to try and to see whether this technology will help us. Yeah, no, absolutely Prof. I think that's really nice summary of how it's laid out in your study in terms of that sort of multifactorial reasons of why this is important. I think that one of the key ones there is like you say, is that we presume it's it's a perfect operation and it is a very good [00:04:00] operation, but there's a lot to be gained from.

And I think, as you say, with caveating with that is that, you know, our patients do demand more now and not in a bad way. It's a, it's a good thing, you know, and, and that makes us strive for better, doesn't it as well. And, you know, it's looking at the, you know, the, the exponential increase you quote in the paper from The American Joint Replacement Registry of the, you know, the 18.3% annual increase in procedural volume in 2020. I mean, that's, that's, that's amazing. You know, that's a huge increase in, and that's just increases the demand, but also it means that we need, we need to do better, don't we? And we should always strive to do better with, with all our technologies, really.

No, absolutely. I mean, I don't know whether the anti-obesity drugs are going to change the game. I, I doubt it. I suspect they'll bring more patients into the realm, but only time will tell. 

Absolutely. Absolutely. So if we, if we move on to the sort of study design, it's a prospective randomized controlled trial, which we've alluded to, that included 60 patients, 30 in each arm with symptomatic hip OA undergoing conventional THA versus robotic arm-assisted. Very briefly, just the inclusion exclusion criteria for the study, obviously very important for any RCT like this. What, [00:05:00] what are those? 

So in reality, what we wanted is a group of patients who are relatively homogeneous. So they had to have osteoarthritis. They had to be adults between the age of 18 and 80. Although kind of we landed in the 60s as an average, which is what we see in our, in our population. And above all, we wanted them to be suitable for the implants that we used with this technology. In fact, with a lot of the robotic platforms in the hip and the knee are closed platforms. With this particular design, you can use either a, a cementless or a cemented stem.

But for uniformity, we wanted patients that were suitable for a cementless stem based on our, on our templating, in our planning. So, so those patients were selected. One of the, you'll face this yourself as well in studies. One of the big issues is making sure you can actually study your patient population.

We wanted to almost man-mark this group of patients and not, not lose them. So we selected patients who are local, not tertiary referrals, not ones from far away, and [00:06:00] ones that could easily access the hospital, were mobile enough. And obviously we're happy to comply and consent to the study and join us on this journey. So I think it's pretty tight. It's just a fairly homogeneous group. 

Absolutely. And so you alluded to there, what, what, what about the preoperative imaging and planning that, and as well as the surgical protocol, that was fairly standardized as well, wasn't it? 

Absolutely. So I think I mean, the first thing is what we really want to do is remove a lot of the variables here, because if you go from conventional to robotic surgery, you're changing quite a few things.

So, so what we've done here is everybody had the standard x-rays. Everybody had the spinal pelvic planning, which we now do routinely, and everybody had the CT scan. And the reason for that is we wanted everybody to start the intervention with a predefined, very clear plan. So there wasn't one advantage to one group over the other.

And then we did the intervention and it was the, the, the surgery was either robotic or conventional, same [00:07:00] group of surgeons within the same institution doing it according to one pathway, there's only one way things get done on our team. And then essentially followed up in a standardized way with standardized milestones and metrics.

And the key thing here is they had a CT at six weeks for us to, you know, did we achieve what we set out to achieve? Because ultimately this is a mechanistic study. This is a study to measure whether we can deliver the target we aimed to deliver. Because actually, if you take a step away from this, one of the big things we're going towards here is both in the hip and the knee is being able to define our target for our joint arthroplasty.

You know, we, we were just very happy that we could do joint replacements, but now we've got personalized joint replacements. And, and if you can then define a target, the ability to hit that target is really important. Yeah, absolutely. And that, that really does take us nicely onto your outcomes. Cause I think that's, it's quite an interesting thing about your primary outcome, the other outcomes you collected and when these were retained Prof, wasn't it? In [00:08:00] terms of what you've powered the study to.

Yeah, absolutely. So, so, you know, this, as I said, this is a kind of mechanistic study. This is a single centre study, really trying to look at, can we hit that target? And so we looked at centre of rotation because we'd done our pilot data. In fact, we published our pilot data, as you know, and we had some metrics to go on, and as per a number of case series before, there was a signal that you could actually reproduce centre of rotation more easily if you had access to this technology. So now we had everyone had access to the CT scan. Everyone had access to the plan. Could you, by using the robotic arm, and execute according to this protocol, actually hit that centre of rotation better.

So we, we essentially powered this study to our existing data on centre of rotation. And that, that, you know, that allowed us to study it for that metric. 

Absolutely. And obviously you collected various other outcomes, which we've already mentioned, as well as a variety of PROMs and sort of including the WOMAC. So the HOOS [00:09:00] score wasn't at the Oxford HIP score. There's a range of, and even the UCL Activity Score as well. And they were collected preoperatively at one year. That's right. Isn't it? Is that right? Absolutely. So we've got, so I think we've published the one-year data. The patients are still being followed up, but this is the, this is the, the, the minimum one-year follow up data.

And this, this is not the study that's designed to look at PROMs, you know, there are bigger studies like RACER, which you and I have discussed before on, on BJJ podcast that are multicentre studies that are, you know, powered to look at PROMs. This is much more about whether we're hitting the target or not.

Yeah. I mean, interesting, interesting. My view in the hip is that we're not going to achieve a huge difference in PROMs cause we do pretty well anyway. It's the reality is the difference may be a reduction in complications, a reduction in instability and leg length inequality or the ability to do sport a little bit better.

They're more subtle things that standard PROMs may not pick up and they certainly won't pick up as to the [00:10:00] extent of an MCID difference. To tell you the truth Prof, I think that was what I picked up from this and I liked is that you're sort of actually, if you can hit that target and if you can just squeeze those outliers more into the middle, it's actually because you won't like, so you won't pick that up on average of a PROM, will you at all?

No, that's true. And everybody has, good days and bad days. And the whole idea is to get rid of those outliers for everybody. Absolutely. So if we move on to obviously briefly what you found, because I want to spend a bit of time on obviously what the, the implication of it. So a total of 60 patients completed the study, as we said, the median age of study cohort was 64 years and a roughly equal number of males and females. And as you'd expect, the sort of overall baseline characteristics demographic data between the two groups was comparable in the study. So Prof, in terms of the primary outcome, restoration of the native horizontal COR, what did you find?

Essentially, we found that we were significantly more likely to hit that centre of rotation using the robotic arm than we were without. Yeah. And that, that clearly reached statistical significance. We're talking about a few [00:11:00] millimeters. So I think we could question whether there is a clinical significance to this, but if you compound that with the fact that then you look at your vertical central rotation, and again, we're better at getting vertical central rotation.

We're better at getting offset, right? We're better at getting acetabular version, right? So I think if you add one to the other, there eventually will reach something that may lead to consequences. And that's right. That's how I actually interpret it, Prof, is that actually you're getting marginal gains and all these things, which eventually could add up to a much bigger, a bigger gain in terms of like you say, about a second component orientation, combined offset, leg lengths, all these things seem to improve.

And, and I know we've mentioned already, it's not powered for it, but in terms of sort of surgical characteristics, PROMs, safety profile complications, how did they compare?

Pretty, pretty similar. So essentially, you know, as a patient population coming through, when you look at their complications, when you look at the safety profile, they were the same. We haven't seen an increase in problems [00:12:00] or because we've used the, the, the, the robotic arm and put pins, pins in the pelvis to mount the arrays and so on. We haven't seen an increase in infections. We haven't seen an increase in fractures or, or, or anything like that from that point of view. Yeah, absolutely.

So if we go on to the implications, Prof, you know, I mean, I think the strengths of the RCT, you know, the study are really clear, you know, and obviously, you know, it's, as you say in the paper, it's been the first RCT to explore the impact of robotics on clinical and radiological outcomes in total hip arthroplasty, you know, a comprehensive assessment of various technical objectives, as well as the functional outcomes and the PROMs that you've looked at. But how do you, what sort of your take home message or how do you interpret these findings? And I think I'm quite interested to know, is it what you expected with your experience with robotics?

So, I mean, I think this is what we felt would happen because you suddenly find, as do your physios, as do your nurses, that you're sort of doing something different and it feels a little bit easier and a [00:13:00] little bit better and a little bit more reproducible.

And then that's, often you blame yourself for that. You think actually I'm biased because I like new technologies. I, I like the fact that I have access to, to this, but actually it kind of, even though that bias pre-existed, this is a way of looking at it objectively and objectively, we seem to be hitting, you know, the target we aim for a lot better.

And I think that that does translate to patients who are slightly more comfortable, get mobile a little bit more quickly. And for example, in this study, even though the milestones and needs for physiotherapy were defined completely objectively based on set criteria, there was less physiotherapy, outpatient usage from the robotic arm group.

So there are, there are little signals here that are starting to creep in. And I think only reproducing these studies, doing them on a larger scale, seeing if they translate to other centres will eventually determine whether that leads to cost effectiveness, whether that leads to really true added value on a population level.[00:14:00] 

But certainly for us, for me as a surgeon, This makes me a better surgeon. It makes me a more comfortable surgeon. And it reassures me that I'm putting the hip where I want to put it. And I'm minimizing the risks of things like instability and leg length inequality for my patients. 

Yeah, no, absolutely. And you had a point there, Prof, I was interested in is, once you've done robotic surgery and you've got your, through the learning curve and things, you've got a handle on it. Does it make you better when you do a conventional joint replacement? Because you have a feel of where it should be in inverted commas, if you see what I mean.

I think it does help. I think it's a little bit like having that sense check, particularly things like leg length. If you're training, if you're training someone and you say, look, look, look at this number, feel where it is. And then you come back to it. Look at this offset, look at the soft tissue envelope, see what the soft tissue balance looks like. It's, it's great.

Yeah, absolutely. I think that's really interesting. And you mentioned it as well there about cost. I know we've discussed this before, but obviously you have not done a cost analysis part of this study, which is clear. [00:15:00] Do you think, well, do you think, how do you think that's going to pan out in the future?

So I think, look, the cost analysis is coming and as, as, as other studies will repeat as well, there's no doubt there is an added front-end cost to getting the technology. There's added imaging with the technology and, and there are disposables, but it will all come down to in the long run to the healthcare costs that come thereafter.

So length of stay, outpatient visits, physio, and then ultimately revisions. Cause actually, if we look, if we take a wider lens, this RCT, and in fact, probably even studies like RACER are unlikely to capture what the big data studies are showing, which is the instability risk is dramatically decreased if you're using this technology.

So two big North American institutions looking at all their data have seen a fourfold reduction in hip dislocation when they're using CT-based planning and robotics. So, it's, it's when you look at cost overall, including revisions, [00:16:00] including reoperations, that's the bit that really will pan out.

Right now, there's no doubt there's an added expense. Yeah. And that's, those are the analysis. So we are, we're currently exploring the health economic side of all the, you know, the, a number of RCTs that we've done. But you know, I think the front-end costs will decrease over time. Yeah. And we will start to learn what the true implications over time in terms of the longevity of the implants, complications, re-intervention.

So we still don't know. No. The bottom line is we really don't know. No, but that is interesting. But that, that, that, that big data work of fourfold. I mean, if you, I mean, revisions, as we know, costs so much that actually that could have massive implications overall, couldn't it? Yeah. And I think we need to go over time and, and pick that up.

And there, therein lie a few other problems in that, you know, does the data from HSS or from our centre translate to everyone else because everyone uses technology [00:17:00] slightly differently. So that, you know, that makes it slightly trickier in terms of measuring these things. And then the other piece, which we're going to struggle with so far is, is our own registry in this country going to be able to pick up these cases.

Yeah, absolutely. Yeah, that makes sense. So I think we hope they will, but. Yeah, absolutely. And so you sort of made me touch on this, but just to finish up, you know, what do you feel the obviously the, you know, we've discussed the strengths of this, of your trial, do you feel there's any sort of limitations or anything you'd want to highlight, you know, sort of as sort of caveats with regards at all?

Absolutely. I mean, so I think, you know, the first thing say it's a small study. This is a small single centre mechanistic study. I think I was like this was an investigator initiated study, but it is funded by industry. So I think people have to always there are always conflicts of interest. I think the reality is if it's an RCT, it's transparent. The data is there. People can look at it, but nevertheless, it will be accused of being conflicted on that basis. But above all, [00:18:00] it's a small study in a specialist centre. And whether that will be replicated elsewhere you know, only time will tell. I think it's important that we now look at bigger numbers and see that.

Probably the other thing to say is that we are now repeating this study, but using the functional positioning software. Cause I think one of the most interesting things about robotics is the ability to adjust based on the spinal pelvic relationship to decrease impingement. And that wasn't available when we started this particular study.

So this study was aiming for a target to get coverage and accurate positioning. And the next study will be functional positioning. So that would be really interesting. That is really interesting. And you know, and just to finish off, you know, you've already alluded to it. What other data do you think we need in this area? Where's the next sort of steps forward, do you think? 

I think we need, we need the multicentre studies, if only to see about the translation, but ultimately I think the real thing we need is the registry data. We need to, we need to start seeing what the impact of [00:19:00] these technologies are on registry data. And I think we've got to remember here again, it's less of a problem right now in the hip than in the knee, is there are multiple devices, multiple technologies. And so we're going to have to pick up the impact of each one. So we're going to have to be careful on how we look at the data, but ultimately we've got to see that there is an impact at population level.

Absolutely. Absolutely. Well Prof, that's great. I think that's a nice place for us to, to, to finish up and, you know, congratulations on, I think a really great study and outstanding studies, clearly a very important addition to the literature in this area. And I really, as always it's great to have you with us and thank you for taking the time to join us.

Well, no, thank you very much. Really enjoyed talking to you about it. 

And to our listeners, we do hope you've enjoyed joining us and we encourage you to share your thoughts and comments about our podcast and the paper on social media and the like, and feel free to post about anything we've discussed here today.

And thanks again for joining us. Take care, everyone.