Robotic-assisted surgery and functional alignment in total knee arthroplasty: the RASKAL registry-nested 2 × 2 factorial randomized trial

Listen to Andrew Duckworth, Samuel MacDessi and Fares Haddad discuss the paper 'Robotic-assisted surgery and functional alignment in total knee arthroplasty: the RASKAL registry-nested 2 × 2 factorial randomized trial' published in the May 2026 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, Instagram, LinkedIn, Tik Tok or Facebook!

Transcript

[00:00:00] Welcome, everyone, to our BJJ podcast series. I'm Andrew Duckworth, and a warm welcome back to you all from your team here at The Bone & Joint Journal. As always, we'd like to start by thanking you all for your continued comments and support, as well as expressing a big gratitude to our many authors and colleagues who take part in the series that highlights just some of the excellent work published in the journal each month.

For today's podcast, we have the pleasure of being joined by one of the authors of an RCT published in the May edition of the BJJ, reporting on the results of the RASKAL study, which was a registry-nested randomised trial from the Australian Orthopaedic Association Clinical Studies Unit that looked at robotic-assisted surgery and functional alignment in total knee arthroplasty.

So firstly, I have the pleasure of being joined by one of my editorial board colleagues here at the journal and the lead author of the study, Professor Sam MacDessi, who is the director of research and training for the Sydney Knee Specialist Group. Sam, great to have you with us. Thanks for having me, Andrew.

Joining Sam, we are delighted to welcome back our awesome editor-in-chief at the BJJ, Professor Fares Haddad. Prof, great to have you back with us. Thanks, Andrew. Delighted to be here. So guys, today we're talking about the RASKAL trial, which looked at two aspects of knee replacement surgery, robotic-assisted versus computer-assisted surgery, [00:01:00] and also looking at the concept of functional versus mechanical alignment.

So Prof, maybe if you could kick us off. Prior to this study, can you give us a quick overview of where the literature was at that stage with regards these topics and where the sort of debate currently sit or sat at that stage? Yeah no. These are-- This study is really impressive because it brings together two really critical topics in knee arthroplasty surgery.

The first one is alignment or positioning of implants. The-- when knee replacement first came on the scene the need to be able to do perpendicular cuts, the initial thinking, the implants that were available meant that everything was put at right angles, and that mechanical alignment was the rule for absolutely everybody.

Some great work from Bellemans and others quickly showed us, that Hungerford and other people as well, that it-- not everybody is the same shape. Yeah. And, Sam will speak a little bit further, but Sam's contributed greatly to this with his work on the CPAK classification as an example.

[00:02:00] So we've worked out that there is a broad variability in our patient base. We focused on the coronal plane, but actually there's also variability across the whole sort of morphotype, if you like. And so everybody thinking has drifted from mechanical alignment to a variety of alternative, more modern alignments, of which the extreme is kinematic.

But within that there is, restricted kinematic, inverse kinematic, and the new kid on the block, if you like, functional alignment or functional positioning, as some people like to call it, which basically allows us to try and fit the bone optimally and cut the bone rather than releasing soft tissues in order to put a patient's knee replacement in their soft-tissue envelope.

And our view on this and, my group has published on this of course as well, is that we're aiming for joint-line obliquity like the native joint line and MCL isometry. So [00:03:00] there's been a whole load of work shifting us in this direction in a very positive sense to the point where I think a significant proportion of surgeons no longer put in mechanical alignment unless you've got a mechanically-aligned patient, the CPAK V type patient.

They're generally trying to alter the alignment to suit that patient. Within that, we can debate what happens with a varus knee or a valgus knee. In parallel, we've spent the last two or three decades embracing technologies to help us with knee arthroplasty. You know- Yeah ... we all have good days and bad days.

There are always outliers, more in some cases than others. And so navigation initially came along as a tool that was helpful to surgeons and actually gave surgeons probably m- a closer achievement of the alignment they desired, particularly in the coronal plane, than they did otherwise, but did not change outcomes.

Yeah. And so more recently, with the advent of robotics, albeit a whole variety of [00:04:00] different effector arms, some image-based, some imageless, some with haptics, some without. The perception in many ways was that the robotics were finally going to deliver a patient benefit from

having this technology so that, the precision that you get, allied to our understanding of alignment that has improved, was going to deliver an improved outcome for our patients. Yeah. Yeah. And hence the importance of s- of studying both those things. And I think one last thing from me, the really interesting thing, people have been busy comparing robots to conventional instruments, and it's great that Sam and his colleagues have embraced the fact, hang on, we've had technology for a while.

Why don't we compare to navigation, which is the technology we were using for those who were, technophilic... until robotics came along. Yeah. No, and I think that's an important point, Prof, isn't it, actually, as you say, because one of my-- part of my background reading for this, I looked at those other studies that I have looked at very much, it's been robotic versus manual, hasn't it?

And this sort of takes it that step further, as you say, [00:05:00] in trying to advance that. And Sam, if I come to you, and you know what? That robotic concept, it really is growing, isn't it? And as I n- I thought it was really interesting in your paper when you said robotic-assisted surgery is rapidly being adopted worldwide with a predicted usage of 70% of all total knees in the USA by 2030, which is a remarkable thing, isn't it?

And, what sort of, based on what, Fares just said, what made you set up this study, and what made you develop this sort of concept the way you did? Yes. So back in 2018, our hospital decided to acquire its first Mako robot. And the technology was relatively new at the time, and there was tremendous enthusiasm about the technology.

At the time, surgeons, patients, and even the industry w- were increasingly embracing the idea that it would deliver faster recovery times, soft-tissue balance, and ultimately superior outcomes. And we were seeing that hype in marketing by surgeons. Yeah. Yet despite all that enthusiasm, as, as we see a l- a lot of the time in orthopaedics, when these things jump on the market, there's often very little high-quality [00:06:00] evidence- to support those claims. So Ian Harris and I at the time felt the timing was right to ask some simple questions. Does robotic technology actually improve outcomes for patients? And if so, what are those benefits? And importantly as Fares did mention, w- we thought it was important not to compare it to manual instruments, but to the current gold standard at the time of computer-assisted surgery.

Because it's been around for 20 years, it's cheap technology, and we know it's precise. It may not improve the outcomes of the patients, but it is a precise technology to achieving the alignment that you want to attain. Sure. But as mentioned, the other thing that's huge is the alignment space at the moment.

Every meeting you go to worldwide, the thing that predominates is discussions on alignment philosophy. And we thought that we couldn't do that study in isolation without bringing in that alignment concept. If we studied robotics alone, any differences in outcome could simply reflect- ... the alignment philosophy rather than the technology itself.

So that's why we designed the RASKAL 2 x 2 [00:07:00] factorial trial to try to answer those questions, simultaneously. Absolutely, Sam. Yeah, that, that's really interesting. And I think that brings us nicely onto the study itself, so as you say, Sam, RASKAL's a registry-nested, multicentre, randomized, blinded as well 2 x 2 factorial trial, and it was in just over 300, 303 total knee arthroplasty patients.

And so for our listeners, Sam, y- you've l- you've gone onto this a little bit already, why the design was chosen, but what were the inclusion/exclusion criteria? Who was actually in the study and who wasn't? Yeah, so i- in the study w- we tried to include a general typical primary total knee replacement patient.

So we included patients aged 40 to 80 years of age who were undergoing a u- a unilateral knee replacement for osteoarthritis. We excluded patients with severe deformities. Varus more than 15 degrees or valgus more than 10 degrees, with major collateral or PCL deficiencies of the knee, those who had undergone osteotomies around the knee, or those who needed increased constraint, a hinged [00:08:00] prosthesis or a constrained condylar prosthesis.

So basically what we were trying to do is find a population who represented the majority of patients undergoing a contemporary primary total knee replacement. Yeah, absolutely. And in terms of the intervention, Sam, is there anything in particular for the, for our listeners that you want to highlight about the groups?

Yeah, if it's okay, I might just step back on what a factorial trial is, 'cause there's a little bit of- Of course, yeah. Yeah. So look because it's not a common study design. Classically in orthopaedics when we randomize people in a parallel group design where they get a, where they get an intervention or maybe a second intervention and then a control or a placebo.

But in this study, there were two questions we're trying to answer simultaneously. We're trying to work out whether robotic technology achieve what we were trying to achieve, and whether alignment philosophy, functional alignment achieved a better outcome as well. And so it allows us to capture or perform two trials in one- to answer both questions simultaneously. So we randomized people to one of four groups. Yeah. They either got the robot or the computer-assisted [00:09:00] surgery, and they either got functional alignment or mechanical alignment. Yeah. And this allowed us to assess both the individual and combined effects of those treatments to work out whether the the sum of the parts of those two combinations w- were better than in isolation.

So the interventions firstly we tried to s- be f- we tried to standardize care amongst all patients. So- Yeah ... so we truly tested, w- what the interventions were trying to achieve. So all patients got a cruciate retaining cemented implant- ... with the Triathlon prosthesis. And in terms of the interventions, w- we used the Mako robotic cutting arm from Stryker.

Yeah. A- and we compared it to the precision optical navigation by Stryker as well. So we're using one company's- ... technology with one company's implant design. Yeah. And the reason we chose the Mako at the time of the trial commencement was it had the greatest clinical uptake at the time.

Yeah. It was really the first to the market. A- and really it's maintained its position as the benchmark for robotic platforms worldwide. It's still the [00:10:00] number one robot. It has CT-based planning. And it has this called haptic technology, which allows you the saw t- to be controlled into a certain position, so it minimizes penetration into the soft tissues.

So we thought if we're gonna test one robot, it's gotta be one that's widely used. But it is challenging trying to- Yeah ... trying to test, when there are so many robots on the market In regards to functional alignment, the question was h- how were we gonna position the prosthesis? Because there's so many different ways you can do this operation.

You can set it, you can do functional alignment and just position the implants based off m- mechanical alignment position, or you can do functional alignment and set the implants based off the native joint line, and that's how we did this operation. We set it off the native joint line, but we had boundaries from six degrees of varus to three degrees of valgus for the m- mechanical HKA, and also for the tibial coronal angle.

And for the femur, the reverse of that, six six degrees of valgus to three degrees of varus. Yeah. And the studies have shown that it captures most patients in a population if you use those boundaries. But [00:11:00] at the same time, you're reducing extreme li- outliers- Wow. ... Which reduce the risk of having potentially an implant failure.

And in the sagittal and rotational plane as well, we were quite broad. We allowed up to seven degrees of tibial slope and femoral flexion, and quite a lot of rotation, up to six degrees rotation either way on the femur- ... to try to achieve a balanced knee, and as Fares said, whilst trying to maintain, that native joint line obliquity. And in the mechanically aligned group, they were strictly mechanically aligned. We aimed for a zero mechanical hip-knee ankle angle and perpendicular cuts. And we asked the surgeons to validate all their cuts. Once they performed those cuts, they had to validate them- to ensure that they were hitting the targets that w- that we wanted them to hit. Yeah. Yeah, absolutely. So I think that's really clear in terms of how you broke down those four groups and what those criteria were for those interventions. And before we come into what you found, some of the results just very briefly about, the outcome measures.

So what were you looking at? What was your primary outcome measure, and when did you follow up the patients? Yeah so it's always tricky trying to in- to work out what your primary endpoint's going to be in a [00:12:00] study where you're looking at clinical outcomes. Because there's always the question, what is the best PROM?

So we spent quite a bit of time trying to choose one, and the one we ended up sitting on was the KOOS 12- ... questionnaire, which takes about three to five minutes to ad- administer. It ... Really what it does, it has subscales for pain, improvement in function and quality of life, and they're things that we think matter to patients the most.

So it is quite a robust tool. It's similar in terms of its accuracy to the full KOOS questionnaire. It has minimal ceiling effects. So that's what we use. But we really recognize that all these PROMs, there's no single instrument that, that does what you want it to do. And so we included a lot of other secondary PROMs.

We included the Forgotten Joint score for joint awareness, the Oxford Knee Score. Satisfaction, joint change, quality of life, which was using the EQ-5D-5L. And the other thing we were looking at was there are tr- knee replacement's getting so good, it's very hard to discern what's going on at, i- in that very upper echelon of patients who do quite well.

So we included the concept of the patient [00:13:00] acceptable symptoms state thresholds... for the Oxford Knee Score and KOOS, and also the OMERACT and OARSI criteria, looking at how patients respond to that intervention. Yeah. So overall we had 334 patients we recruited over two and a half years, and 303 patients were finally in- included in the analysis.

Yeah, absolutely. And so Sam, if you could then summarize for them, so of those, as you followed them up over that, that two-year period, what would be the key sort of findings you'd want to highlight for our listeners? Yeah, s- so the headline findings are really quite simple. At three months, at six months, at one year, and at two years, there was no difference in the primary endpoint of the KOOS-12 score

when we compared robotics to computer-assisted surgery, and also when we independently compared functional alignment to mechanical alignment. Yeah. And that was reflected across all the secondary PROMs as well. Yeah. Every other PROM we tested, n- no [00:14:00] difference comparing the robot to the computer and comparing functional to mechanical.

There were no difference in patients who achieved that acceptable state on their PASS thresholds or the OMERACT-OARSI criteria and their responder status. Yeah. We also took the time to, to look at early outcomes because one of the, perceived benefits r- of robotics is that if you're creating less soft-tissue trauma, potentially the early recovery would be better.

Yeah. Once again, there was no difference in VAS pain at three weeks and six weeks. There was no difference in time to discharge. And there was no difference in the amount of analgesics these patients consumed in any of the groups. But there were some notable changes in the operating room that we found.

The robot was about 11 times 11 minutes faster than computer-assisted surgery, which kind of rev- reverses what the older studies used to show about robots. Yeah. The other thing the robot did achieve was it actually did preserve the PCL significantly better- ... than the computer-assisted surgery [00:15:00] did, which really aligns with what Fares' work has shown early on, that those haptic boundaries do preserve that PCL footprint beautifully, and that's something I see in the operating room.

Yeah. And that the functional alignment philosophy required far fewer soft-tissue releases, significantly far fewer. In the mechanical alignment group, 45% of patients needed a release. Yeah. Remember these surgeons were using the optical technology to drive that release. Whereas the other needed releases in 8% with functional alignment, and that odds ratio was about nine.

Yeah. But I think the other interesting thing was surgeons overwhelmingly when we... preferred the use of the robot versus computer-assisted surgery. Yeah. And they overwhelmingly preferred the use of functional alignment. So, maybe no difference in outcomes at the end of the day but the operations were easier with the robot.

They were easier with functional alignment. Yeah ... Despite patients receiving still experiencing similar excellent outcomes. Yeah. I thought that was really interesting as well, Sam. I l- I love that sentence at the end of the results bit in the abstract saying, "Most cases, surgeons just preferred the [00:16:00] robotics and functional alignment."

And it does ... w- well, I'll come back to Prof in a second about, people vote with their feet, don't they, a little bit here, and I think this is maybe what we've seen. But we'll come onto that in a second. But so Sam, just if we move on to, the implications of it in terms of, you, you've gone over the take home message but, what you feel they are.

But what limitations of the study would you want the r- listeners to know about, to put it into context, yeah. So look I think some of the limitations are, we did use those restricted boundaries which means potentially some of the knees may have been incompletely balanced.

Yeah. But we want to include those extreme outliers in constitutional phenotypes because of the risk of failure. A- and we are following up with secondary analyses now where we put pressure sensors in the knees at the time of surgery. Yeah ... that the surgeons were blinded to, and we'll be presenting those results soon.

 Secondly I think one of the limitations is we didn't compare it to manual instrumentation. We compared it to computer-assisted surgery, so it would be nice to know... Yeah ... how these robots do sh- you know, match up to manual instruments. There, there is also [00:17:00] this concept, as Fares said, called unrestricted kinematic alignment using caliper, and that's been broadly based broadly adopted worldwide where surgeons use manual instruments to execute this operation to try to reproduce the patient's anatomy.

And I think it's now incumbent on those adoptees to actually study this in a similar way to, to test that technique versus manual instruments as well in a head-to-head trial. Yeah. And lastly, this is only one robot, one implant design, so we're gonna have to see how, what bears out in the future with other studies

testing other robots on the market. No, absolutely, Sam. Yeah, I think that's a very fair assessment. And so Prof, if I can maybe come back to you now, how do you interpret these results? And I suppose it, was it what you expected?

Yeah, no, it's-- I think it's a really good study, and congratulations to Sam and the team for doing it. I, I think the study is at variance with some of the data, including our own early data. So it's often, the reality bites when you spread something to multiple surgeons.

And, and.. Yeah ... that is interesting. So I think... Yeah ... there, there are some facets here worth [00:18:00] exploring further or at least thinking about further. So for example, in cavitary surgery, what the robots have allowed is surgeons to do the more complex cases... Yeah ... better or, to deliver better outcomes in those cases.

In hip and knee surgery, we've done it the other way around. We've gone for the straightforward cases, and this is actually what this study has looked at. Yeah. Also, Australia's got a big adoption of technology in computer-assisted surgery, and these will be, and Sam can confirm this, high volume, good surgeons who are used to what they're doing.

So actually their baseline is, Yeah ... is very good. Yeah. So actually understanding, some of the sensitivity analyses that need to be done here is, quite what proportion of patients were in that group where mechanical alignment and functional alignment aren't that different.

Yeah. What were the surgeons doing in mechanically aligned knees? Because a good surgeon doing the releases versus a less good surgeon may be completely different. So- ... I think there are some facets [00:19:00] here that are worth exploring further. We shouldn't throw the baby out with the bathwater.

But actually it's really interesting. The fact that I take away from it the most is the fact that surgeons don't wanna let go of the robot. Yeah, absolutely. And, if you try and take it away, you're gonna hurt the surgeons. And so that tells you there is something there that we need to keep looking at, and it's... Yeah...

it's fascinating. If you look at our my MDT, which runs, all the x-rays are shared, and then you look at the MDT, and the output of the MDT, in 2026 compared to the output with the MDT 10 years ago, completely different in terms of the variability of the outliers. Yeah ... what people are doing. So there is something there that surgeons like.

It's maybe how we're measuring it. Yeah. There is also, there are lots of things to think about here in terms of do we know the target? You know... Yeah ... do we really know what we're aiming for? So when we're calling something functional alignment, is that, as Sam alluded to is that actually the correct target or is [00:20:00] it different?

And that's mainly focused on the tibiofemoral joint. We're not looking in that much detail at the patellofemoral joint, and so that questions to a certain extent the implants that we're using. So you may need to be looking at a small number of patients, and in those patients you make a difference rather than across the totality.

We've already seen this in Simon Young's study. Looking, which, you know, l- looking at that in the mechanical versus functional alignment. If you look at the totality of his study, you don't see a difference. If you suddenly look at CPAK I and II patients, then there is a signal. So I think those things are really important.

And then there's another facet here we need to think about in terms of the ergonomics of using robotics, the comfort... Yeah ... the sort of physiological and psychological stress to the surgeon of having these tools... probably helps. At the end of the day, we want you and Sam and everybody else involved to be operating for a very long time.

Because the need for our services is gonna continue to increase. So if robotics helps from that point of [00:21:00] view I think that, that will be a significant significant benefit, not just to surgeons, but ultimately to patients. And of course if the implants are potentially in inverted commas, because we don't know that they are in better positions, and you're removing the outliers, there may be a signal that we see further down the line, which is very nice. Sam may wish to come back to this. It's registry-nested. They're gonna follow these cases up. Yeah. Let's see if there is a change. We may not see a change later on... Yeah... but if there is a change late-later on, it would be really nice from that point of view.

I think what surprised me, if you like, is with a haptic-controlled robot, I would have expected more safety, if you like, less secondary injuries... less problems. I haven't seen that. Sam may have more detail. They, they're doing all the other analyses, and I think it'll be interesting to see when they've drilled and drilled.

There's a massive amount of data here. Data, yeah. Yeah ... what what they actually come out with. Yeah. So I d- I don't think this should be taken negatively. I think it's really important that the [00:22:00] orthopaedic community is doing these studies. This is one signal here, and it just means that's something we need to continue to study and continue to explore, and we definitely should not dismiss precision.

If, if someone has given us a tool that allows us to put things in more precisely, and it is not showing a benefit, that's not the tool's fault. Yeah. That's either the fault of the surgeon guiding the tool, ... or the fault of the implant that we're putting in, or the fault of our measurement tool.

Now that, that may not be enough for our pay masters, but actually at the end of the day, we as surgeons who want to improve knee arthroplasty for our patients need to remember that and remember actually that if we got precision, we gotta live up to that precision. And it may be that the technology's good, and remember the technology's continuing to evolve.

So from when Sam started this study to what is available now is completely different to what Sam used before. But, ultimately we may be [00:23:00] failing the technology rather than technology failing us. Yeah. No, I think that's a really good summary, Prof. And actually, I think, like you say, I think I would encourage our readers to not only read obviously Sam's study, which is fantastic, but also your editorial accompanying it and highlighting these points really nicely.

And I think there's something for me, I, I'm a trauma surgeon. I don't do knees, and I have no skin in the game. But what is fascinating to me is how my colleagues all rave about robotics and the benefits that it's getting and they talk about it. And that, that figure that Sam quotes in his paper about the number of robotic surgeries now gonna be going on, the surgeons can't all be wrong.

Actually, everybody's s- is hitting it. And so I don't think I've ever seen anything in my career where actually there's been such a frame shift of people wanting to use this. So I'm, I very much am leaning more towards the fact that we're maybe just measuring the wrong thing or we're using the wrong thing to measure those other those points.

And I think also, as you said very nicely, if you asked our patients, "Would you like a more precise joint replacement?" I think most of them would say yes. And so I think we... it's something that is very much evolving, isn't it? And I [00:24:00] think we're just using all this data to try and build that picture more and more, and actually probably build a picture of how we better assess it more than anything.

So I think that's a really great summary. So well, both, I'm afraid that's all we have time for. Thank you so much to y- both of you joining us. Sam, congrats on a fantastic study, and it was really great to chat to you both about this topic. And to our listeners, we do hope you have enjoyed joining us, and we do encourage you all to share your thoughts and comments on the various platforms.

Feel free to post about anything we have discussed here today, and thanks again for joining us. Take care, everyone.