Special Podcast Series: Ep 1. Advances in Robotic-Arm Assisted Unicompartmental Knee Replacement: The Edinburgh Experience

Listen to Ben Oliverre, Sam Patton and Nick Clement in the first episode of our special podcast series, kindly sponsored by Stryker. This episode focuses predominantly on total knee replacement and partial knee replacement.

Visit BJ360  here

Chapters

• 4:55 How Sam & Nick got into robotics
• 6:48 How does the robot work?
• 12:27 Markov modelling
• 14:06 Oxford Knee Replacement
• 18:58 Current surgery in the NHS
• 23:31 Training to use the robot
• 28:12 Uni-specific complications
• 31:06 How to learn the skills to operate the robot
• 33:15 Final comments

Transcript

[00:00:00] Hello, my name is Ben Oliverre. I'm the Editor-in-Chief of 360 and I want to welcome you to the first in a series of four podcasts on robotics. This series is kindly sponsored by Stryker. And for today, we're going to focus on predominantly total knee replacement and partial knee replacement. And we're very lucky to have two expert surgeons from Edinburgh. We've got Sam Patton on the line who has been instrumental in the introduction of robotic surgery in the Edinburgh, in Edinburgh of itself, initially in the private sector and now moving across into the NHS. And we're also fortunate to have Nick Clement who is a paper machine in his own right. And I'm sure people have read something written by Nick, even if they don't realize that they have, and he's going to talk to us a little bit about his experience of becoming a surgeon, who's enthusiastic about robotics. 

So I think in the first instance, gentlemen, thank you very much for joining me. For giving up your time this evening. And I think it'd be helpful for you to just sort of talk us through how you ended up in Edinburgh being so keen on robotic surgery. 

Thanks Ben. So I think in Edinburgh, we've had [00:01:00] a very strange experience when it comes to partial knee replacement and we started off in, if we go back to about 2005 I actually wrote a paper in Clinical Orthopaedics that looked at a series of total knee replacements and a series of partial knee replacements. And we essentially found that the two groups functionally were the same, although there were ceiling effects on the scoring system that we were using at that time. But the survivorship of the partial knee replacement was significantly lower or worse than that total knee replacement. 

That was the sort of time when the Oxford group were publishing quite impressive results for uni knees and I remember reading your paper as a trainee, actually. You probably won't believe it as a trauma surgeon now, but I did read your paper as a trainee and discussed it some time ago.  And there was a stark difference between what was coming out of the originating centre and I believe a centre, I think the centre in Belgium or Norway was [00:02:00] included in that series as well if I remembe r,  and what other people were finding. Why do you think that was? 

Yeah. Well, I think if you look back at these sort of survivorship figures that came out of the joint registers, you know, you were seeing knee replacements performing at around about a high nineties survivorship and partial knee replacements coming in significantly below that really in the, in the high eighties. So there was a significant difference and that is still mirrored, in fact, in the National Joint Registry. So the question is why was that? And I think one of the reasons is that a partial knee replacement, particularly, I think has always been quite difficult to put in and there are surgeons and the Oxford group have clearly shown this, that if you put it in very well, you can get very good results. And that is true, but that is not mirrored around the rest of normal practice. And certainly wasn't mirrored, I think, in the studies that we did, which very much [00:03:00] followed the results from the National Joint Registries. 

So I think there are probably several reasons. It's difficult to put in. It's difficult to get it well-aligned. It's difficult to get the size right. It's difficult to make sure that you don't put the knee into valgus or into a poor alignment. And that means that your failures, your early failures from fractures, from difficulties with tension. In other words, spitting out a mobile meniscus, for example, or malalignment from overloading your lateral compartment, for instance, and getting rapid joint wear in other parts of the knee. And I think those are the things that probably made partial knee replacement results inferior to those of total knee replacement. 

And just quickly touching on that, I know we need to move on, but my own experience as a [00:04:00] trainee, it's been a long time since I've been  involved in knee replacement at all, but you know, the startling thing about unis was some of the patients did really very well. And some of them did really not so well. And you know, I know you're talking about the average result being comparable, but actually there seemed to be more of a spread amongst the unis. 

Yeah. I think there was a feeling of a rule of fives and the rule of five was that two were excellent, two were pretty good and one was awful. I think that, you know, that was the feeling about that operation. And so a lot of surgeons didn't take it up because they were actually frightened of it because whilst they knew they could get some great results, there were some that were a bit of a sting in the tail, and many of those to do with patient selection, many of those were to do with technique. But I think a lot of them are due to the inherent difficulties of putting in a partial knee replacement well.

Yeah. And so having sort of established that things were not quite even, how did that lead you to [00:05:00] robotics I guess? I mean, it's not a widespread thing in the UK at the moment. I guess it's probably more commonly used overseas. I know there's a big cluster in Australia like it and large parts of the US are involved, but I've never actually seen a robot for joint replacement, other than a standard day-today show. 

 Well, I suppose my history was I did a fellowship in Grenoble where they were using some of the very early navigation systems. And of course, navigation is one of the key elements of how the robot works. In other words, the computer has to be able to see where the human is in space and so navigation was a key component of that. But we all thought that there are many hospitals with navigation systems sitting, gathering dust in the corner.

Yeah because the operation was the same but took longer.

 The technology step I don't think it was just quite enough. I think it's good. And that's why it took off in Australia because they're big [00:06:00] navigators, but the next step really is to have the precision of cut so that these things fit in well.

And for instance, if you go to Ikea on a Saturday afternoon, you will see the quality of carpentry, the fact that the joints of the tables and the furniture fit absolutely exactly. Whereas if I botched together two bits of wood, I'm lucky if they're straight and I'm luckier if they fit together. 

You should see my Ikea builds.

And I suppose the beautiful thing is with modern robotic systems is they don't come with a difficult Ikea manual, which you have to interpret, you know. 

Just digging back a little bit,  you've talked a little bit Sam about, you know , navigation and triangulation in space and robotics allowing you to do a more complex cut. For people who have not seen it, [00:07:00] or, you know, somebody like myself, who's not used, how does that actually work? So when you get the patient in front of you, you've got the robot in your hand what actually does it do? How does it tie those things together? 

So just as you would in a navigation system, the system that we are using is CT-based. So the patients have a preoperative CT scan, and then you allow the computer to see where it is by putting markers in the bone and registering the bone. So wherever you move the femur or the tibia, the computer can see you. 

  The next step is that, of course, once you've done your approach to the knee and you know, you still need to do a significant approach cause you still need to get the implants in. But I have found you don't need to do nearly as much dissection around the joint because when you take the burr and the surgeon is in control of the burr, the burr is controlled by a boundary. So there is a haptic boundary in which the [00:08:00] drill will work or the saw or burr or whatever. And if you try and go outside that boundary, the robot actually physically won't let you. And if you force it because you're a big, strong man and a standard orthopaedic surgeon, you will find that the burr eventually then switches off. So it is very tightly controlled within parametres and the boundaries set up with the plan that you have for every case.  

Before you even approach the patient, you've actually put the implants in, on a computer screen. So you've actually put the implants in and size them up. And I generally find I can actually open the implant that I'm going to use before I've even started the case. Cause I know exactly what size it's going to be. 

And is that actually accurate? Because again, you know, scaled x-rays, templating, you come to within one size or another, [00:09:00] is it actually accurate of a CT scan?

Yeah. So I think this is the big difference between this technology and navigation is that this is absolutely, you know, when I did my first one and put the implant in, it just clicks in to the holes that you've made. It's perfect. You put the plastic in and the knee runs through a beautiful range of movement and the knee is stable in all positions. The power of that moment when you first put your first one in is quite strong . You know, when you did it, you just thought, well, actually now we're onto something. 

Challenge - all sounds exciting, but isn't it just that it's something new? I mean, you know, every patient's got to have a CT now, time in theatre, press resources, got to buy the expensive robot. And how do you square that away as ... you are clinical director, is that right Sam? 

Yeah. 

How can you square that away with your institution? 

So, I suppose the first [00:10:00] thing is that we brought it into the private sector and you know, obviously you have to make a business case for that. So there has to be a business case that runs behind that. In other words, you can't have a system that's going to take forever, costs a million bucks and patients get no benefit out of. 

And this is really where I brought Nick in and Nick was newly appointed as a consultant in Edinburgh and Nick, I think you can probably take a bit of the story from there cant you.

Yes, thanks Sam. Just to get back to Ben's point about the CT scan and the size, and the calibration, every CT scaner that is used, they're all calibrated by Stryker. So they're accurate within a few millimetres. 

 I sort of meant, to a certain extent, you know, like you guys I work in the really big, regional referral centre in Nottingham, you know, turning around and saying the hundreds of joint replacements we do a year will need a [00:11:00] CT scan, would put an additional resource stress, I guess on the institution. 

What's the cost of a better outcome for the patient though. If it was you and you had the option of a total knee replacement....

Do you want the official answer, Nick, or do you want the answer if it was me.

If it was you and had me doing your knee replacement...

Can I not have Sam? Is that allowed? 

So that's the thing, isn't it? That's patient versus population medicine and as surgeons we should always think about our patients. I completely agree. But unfortunately our managers don't sit in clinic with the patient and you know, so there is that side of things and I guess, you know, you've already touched on the poor outcomes and revision for pain and so on. And you know I examined a PHD relatively recently looking at pain following total knee replacement, and was somewhat surprised in my background meeting cause I was going to take the guy to task about the fact he'd put 15 to 20% of patients have persistent pain after knee replacement. I thought it must've come on in the 20 years since I did [00:12:00] my first knee replacement and it turns out it hasn't.

No. I agree with you.

Just on the cost economics side of things, so I agree, and we've tried to get it into the NHS . That was my first job as a consultant three years ago. Last Friday, we got the robot in the NHS after three years of hard work.We havent even used it on a patient yet. But during that time, we've had  to come up with business cases. So we've been through it all. And first thing we did was a Markov model using published data out there. Markoff model just...

As it's 360, I'm going to take the Liberty, give us two minutes on Markov modeling.

So Markov modeling just uses already published data. So outcomes at one year, two years, three years, EQ-5D or some cost utility, whatever it might well be, SF-60, whatever it might be, then you model that for that patient's lifetime. And then you use published figures for revision. So we used published figures for [00:13:00] revision, which was far better than manual uni-compartmental knees, up until eight years. And then after the eight years we used normal revision rates cause we don't know what it is. 

That is the key to reading a Markov modeling paper, isnt it? It's one of those things that if you have good quality outcome data, you can get some really powerful results and you can work thresholds out, whereas if you have poor case series and the nice thing about knee replacement is there's really good quality data out there.

Yes, but it does depend what you choose. And it does depend if you want to be... you're exactly right. That back table, there'll be one table in the Markov model that tells you how everything was modeled. And if you choose one with the worst outcomes for say a total knee replacement or manual or manual unicompartmental knee versus the best outcome for say a robotic knee, you can certainly skew the results. But anyway we used the mean of the published data, the best we could, at that point in time and due to the better survivorship during the first eight years which was over 50% less , the [00:14:00] cost per quality was around £1,000. So it's way under that 20,000 , that nice quote. 

So a challenge here Nick. You like to ask a challenging  question. So why not just send them all to Oxford or the originating centre and save the money on the robot? 

 I think that's such a poignant bit. If you're doing lots of these, I don't know whether you need a robot, and that's the truth. The Mark Blyth did the kind of the groundbreaking study about 2012 ish. And he showed, even with him, fantastic surgeon. Brilliant. And he does lots of Oxfords, but even with the robot, he was far more accurate with the Mako than he was doing it manually, but he wasn't able to show a functional difference between the two groups. And I think that's probably why - because he is so good. In the average person's hands like me, I think that wouldn't have occurred. However, also in his study because he used the Oxford scores as outcome. He also used the Forgotten Joint score, which doesn't have that seal on effect that Sam was talking [00:15:00] about, and he did show a better outcome from the Forgotten Joint Score. So maybe he was just using the wrong tools to assess the outcome.

But I agree with you. I think if you're doing lots of Oxfords you probably don't need a robot. But I don't know very many people do lots of Oxfords.

 I don't think they do. And you know, it is an interesting question cause that brings us then on to, you know, NJR, National Joint Registry , constraints on surgeons, low volume, high volume, you know, how many do you need to do? So we're going through a process in our place where certain surgeons are being told they don't do enough unis. And so they should all be centralized on one or two surgeons, even in a very big institution. 

You know, I guess the question is that if you're using a better way of inserting the implant and you can prove that that perhaps allows for less populous places to offer these facilities without, or offer these operations, without somebody having to travel miles and miles.

Yeah. I mean, I think the National Joint Registry shows about 11% of patients get a [00:16:00] uni. 

Yeah. 

And you know, in Oxford they're sort of in the sort of 40 to 50% area, I think. And you know, so you're talking essentially about nearly half of knee replacements, perhaps being suitable for a unicompartmental knee replacement.

And if you look at my private practice last year was 50/50. And I regard myself as a pretty conservative surgeon. Now I may have had some excess referrals simply because I'm doing that. But, you know, that was a shock to me that it'd taken over perhaps quite so much. And if it is going to be 50/50, then that is part of standard practice. Is it not? 

It is.

And if we're turning knee replacement, total knee replacement more towards partial knee replacement because we feel that patients recover quicker and have a more normal feeling knee and a better Forgotten Joint [00:17:00] Score. Plus also we can reduce their length of stay. And one of the things that really pushes knee replacement towards a 23 hour, you know, single-day stay, then it has got to be converting your total knee replacement population into a partial knee replacement population. And if you can do that, you know, that instantly dismisses the cost of any CT scans or anything else like that.

Oh, no, I absolutely agree. And I think, I think one of the things that's important to remember with designer series is people often kind of discount them. But it's not the one or two people in Oxford that design the joint, it's the whole group. And I genuinely believe what they publish. And as you say, it's a volume thing, isn't it? You know, if you do a lot of them, you get really good at it, like anything else. And there is that thing about 24 hour stay and, you know, enhanced coverage and so on which you can do with a partial, you can't do with a total so well. No I get it. I get it. I just, I think it is important to, sort of, talk around the topic in [00:18:00] terms of volume and what a better jigging system and a better constraint system can offer. 

  I think the robot itself just makes the average surgeon the same as everybody else. Like we are coming out of lockdown, none of us have done very many operations and all of a sudden with a robot you can do the same knee time and time and time again. 

So that's the point really is that getting rid of that rule of fives. The very first Mako that I put in, it took a while because I was doing it with two of our friends and everything else. We had a chat and it took a while. But of course, now I can put in a Mako uni in 45 minutes, but the point is they've all looked the same. So the very first one was put in well, and they've all been put in well ever since. So, you know, and they all look the same. You sort of get fed up of looking at the post-operative x-ray because they all look the same. There isn't an outlier or there isn't one that's a bit wonky. They all look the same. 

So just [00:19:00] talking about that in terms of practice. So we told you this about how you kind of evolved your practice, what the drivers behind it were and, you know, I get that - it's removing the low outliers and moving the average up. So what does your practice look like now between the two of you? Nick first, I guess. So, you know, in your ideal practice, what proportion of the patients get a total knee, what proportion get a uni and how many of them do you use a robot for?

That's a really simple question for me. 100% get the total knee replacement cause I mean, I only work in the NHS. I don't do any private work and we've only just got the robot on the NHS. So, so all my patients, I discuss with them, I give them the option to my colleagues do manual uni compartmental knees. And if they want to go over to have an Oxford, but I often quote the revision rate is 3%, or revision risk is 3% at 10 years, as opposed to 10 to 12% at 10 years with the Oxford and most people go for total knee replacement. So hopefully that might change in the coming months to years. 

And what was your [00:20:00] aspiration? So, I mean, you're a data-driven guy. Where would you hope to be in three, four years time? 

Well, I must admit when I first  came across the Makolov four years ago in Newcastle with Prof, I didn't really believe so much in the Makolov. I thought it was just a way of making money, publicization and, but the more I see, the more data I see using it in the lab, seeing cases down in Newcastle, total knee replacements. I do think, I definitely think it's the way forward. it's the future isn't it? It has to be like, it's more accurate....

I think it probably is.

 It has to be. Like, there's no way around it. Even me and I hate technology. You've got to embrace it. And even for research, all of a sudden, it's the best research tool in the world because we can put the thing in. We can do all these models in trying to figure out who does well, but actually it's how it's put in and how it's balanced.

And now all of a sudden, that pattern of arthritis gets that knee placement with that gap, with that size implant and that alignment. It's [00:21:00] just, it's going to be, it's going to revolutionize, can you just see 10 years, 20 years time I'm gone, right. You come in with a varus knee with this partner of arthritis, bang, bang, bang. You get this knee replacement in this orientation. Done. Out the door. And I think it'll just be, obviously this is the very beginnings of it, but I think that's where we're going to head.

 Do you think it's going to take on table decision-making entirely to preoperative planning. And I remember being taught, you know, downsize and anterior shift and all those other things that you're taught and, you know, take a little bit off here, gap and all that kind of stuff. Is that out the window? 

It is...

No I dont think...

 I think disagreement on the panel.

No, no, no, not at all. Because of course the preoperative data that you've got is a static CT scan taken with a knee, usually pretty straight...

You have seen right  through me Sam. That was where I was taking you to. 

So of course the beauty of this thing is it's a 3D system, [00:22:00] but it's a 3D system that also moves. So you can assess your ligament balancing throughout the range of movement. And then of course, that's an intraoperative thing. So there are many intraoperative steps and intraoperative data that you are using to finely tune where you're implant goes.

And will it synthesize that for you. So instead of the, you know, the old sage orthopaedic surgeon having a wiggle or passing the knee to the registrar who dare say anything else and saying, isn't this wonderfully balanced? Does it give you a readout? 

It gives you a readout, but of course, like all systems, you know, it depends on to some extent the data you put in. So one of the bits is stressing that medial collateral ligament, and putting it through a range of movement and allowing the computer to assess a physiological tension on that ligament, for instance, if you're doing a medial uni. And of course it interprets that. So if you're a big rugby prop forward and you put a lot of stress in it, you'll do more than somebody who isn't [00:23:00] stressing it so much. You need to have a little bit of surgeon feel in it but it does make it good.

So next question then. 

Just to follow on from that, just because there's no pressure sensors for the uni knees, but there's a recent study by Kyani et al with Professor Haddads patients for total knees. And when he was talking about the robot, the dynamic balancing of the knee, it's perfect using a sensor within the knee. It's within 15 pounds on either side. So this thing was the most expensive gap balancer that we've got and it does it  within a couple of millimeters. 

Yep. No, I get it. I get it. Another question. So you guys have both learned to total knee replacements the same way that I was taught, by feel and  a little bit of intra-operative nails, and getting an idea of what feels right. What do the next lot of trainees get? How do they get that experience? How do they know when the robots got it wrong? 

 It's a bit like doing an arthroscopy, isn't it? Because I was brought up and I remember watching a couple of [00:24:00] open meniscectomies, but I've never done one and it's not part of my vocabulary. Just as the trainees today do not use a pen and paper. They use a computer and a tablet and you know, the world has changed and I don't think they're going to need to use those skills quite so much. 

Now, of course, this is moving into the future, but you know, I do think that this technology is here and it's here to stay. It's not going to go away. It's just going to become more and more and more. So I think that, I think the train is here, you know, we've sort of got to get a board and run with it.

 I completely agree with you. I think if you're halfway through your career, perhaps you can ignore it. But I think these things... these are seismic shifts, aren't they? It's like, you know, I remember being told by very old trainer about the time that they used to do DHS without an II. Can you imagine that now? I mean, you know, that's how it was done. Where it films and you had [00:25:00] another go.

You use IIs in Nottingham, do you?

We do use II in Nottingham.

We use *inaudible* in Edinburgh...

Yeah, I know, but you know it is Edinburgh isnt it? That's advanced for Scotland. 

It is. Yeah. Yeah. 

For the record, I've not done an arthroscopy in over a decade because it's the one operation that drives me really potty, trying to nibble that bit of meniscus at the back of the knee. First thing I gave up as a consultant. 10 years, not put a scope into anything.

But yeah, so yeah, I guess we digressed a little but, one of the things I wanted to kind of tease out, I guess, because again, this is 360 and we need to remember that it is some of the data behind it. So Nick kind of talks a little bit about, you know, the Markov modeling and the stuff that you did  to put it  together, being objective about it, you know, every previous innovation in total joint replacement, because what we're basically trying to say is get rid of those low outliers, get rid of that 15% of patients that have persistent pain for whatever reason. Some of them we know have had them on [00:26:00] operation. Some of them actually have a poorly functioning knee replacement.

What is the state of play with regards to the data that makes this different to say patient-specific jigs, you know , image guidance that we've talked about. You know why for you guys, is this a game changer?

 I think from an evidence side of things, I'm not going to get into the shape much and patient-specific jigs and all of that, but standard practice would probably be a total knee replacement versus some kind of uni-compartmental knee replacement for those that have uni-compartmental disease. And certainly when we've done our study which was retrospective, we showed better outcomes using the unicompartment knee and when we presented that data, everybody said that's known, unicompartmental knees do better than total knees. We know that we know that. Well, actually there's just been a massive randomized trial in the Lancet. Showed no difference. One year, two years, three years, four years and five years. No significant difference in the Oxford Knee score. 

[00:27:00] Some very cross people about that study.

 Yes I am sure there is, but there's a  statement in there that's more cost effective. It might well be, but nonetheless, there's no difference in outcomes, but you've still got that revision burden. You've still got that... people like me, you've still got that 10% revision burden 10 years, then all of a sudden, the Mako allows you to get rid of that 10% revision burden at 10 years cause that's early studies, stuff from Andy Pearl, HSS in New York and Australian Registry data , American Registry data. We know that the revision rates far less. It's approaching the revision rate that you'd see with a manual knee replacement. All of a sudden you've got rid of  that revision burden risk. And now you've got those better functional outcomes, potentially. 

Maybe they use the wrong outcome measure in the top cat, and maybe that's why they didn't show clinical significant difference between the two and we could argue that all night. But certainly there's definite studies out [00:28:00] there that there is less morbidity postoperatively, low infection rates, lower DVT, to early mobilization  , shorter length of stay, less pain postoperatively. So all of a sudden you've got all these benefits. 

What about the uni-specific complications? The things that put people off, the plastic was spat out. We had to do another operation. The cut was a little bit too deep. I mean on a toe fracture. We took more than we expected so our revision... because what's always been said by people who are sold on unis, you know, and I worked for somebody who was very much sold on unis. You know, I remember seeing patients who'd run marathons with their uni-compartmental knee replacement. There's no way you could ever do that with a total. So coming back to the question you asked me what I like. I want that guy's knee replacement. I don't want the average that you normally see, cause I'm an avid marathon runner, but you can't guarantee that because of these very specific complications. Is the effect to do with reducing those complications or actually a general improvement overall, do you think? 

Good [00:29:00] question. It's a great question. Certainly data from the Australian registry would say that was early complications are less wet. This is a fixed bearing as it happens. So  hopefully you don't get the...

If you do, you've got a problem.

And there are certaintly some very early failure rates with the old polytibia that was used, which is no longer used, but setting those early failure rates in the Australian registry were due to infection. And that might have been the early users that might have been a little bit longer or early problems with sterilization around the robot side of things. But certainly, obviously in those early complications and in early complications we are seeing are infection and that's there. It's difficult to get rid of them. Send me the learning curve. Just as you touched on earlier, there's no learning curve for accuracy. Like Sam said, all these are put in identical from day 1 to day 1000. They're all put in the same, but it just takes a little longer. Your first 12 cases take longer. So maybe the early [00:30:00] users might have an infection, slightly greater infection risk because of that. 

So you touched on, in your training, you were fortunate to work in Newcastle where you know actually they're a very enthusiastic robotics centre, full stop. Not just the Mako, they use all sorts of robots as you know. So you've got that background under your belt in training. You know, what about, I never saw one in training. If I wanted to go and do, you know,  say I was a competent knee replacement surgeon, not somebody who'd never touched a bendy bendy joint. And I wanted to  go and do what you've done and start introducing it into the NHS practice. You know, how much, how much do you need? Do you need to go and spend time with somebody? What is the kind of upskill? Cause there will be an upskill. I know you've sold it well, but there will be an upskill cause there is with everything.

Just to clarify Ben, I was trained in Edinburgh, even though I've got this dodgy accent. They didn't give me a job in Newcastle so I had to go up North.

Ah fair enough, apologies.

[00:31:00] Clement, you are fired officially now. 

I'm sorry.

So returning to my question then so you haven't got that experience because you know, Sam's told us he does them  in the private sector. So how are you going to put yourself up to it? 

 I think when I started, I essentially went to Australia and learned how to do it in Australia. I visited some surgeons over there. I spent some time in working with dry bones and then I did a cadaver-based course and certainly the methodology used is that all surgeons wanting to start using this type of technology are put through a series of cadaver-based labs where they actually do these cases and are instructed in how to do it.

Is that enough, is the question, or do you need somebody? Do you need a mentor, somebody who can help you with the planning and that kind of stuff or...

So I suppose the other [00:32:00] aspect of this is that the robot comes with what's called an NPS, which is an operator. And that operator helps with your planning, helps you make decisions and actually can act as a guide to the things that need to be achieved in setting up a programme. 

But you know, so I was scared during my first case. There's no question about that. It was a scary moment, but of course, in fact, I didn't need to be because of course when the bits went in, the thing worked and you know, that's a very, very satisfying place to be. 

Yeah. Because we're almost beyond the early adopter stage now, aren't we? I know in the UK at the sort of innovative stage, but globally, you know, there are people who've done many, many hundreds, if not thousands of these joint replacements. And so you have to just be a little bit careful about how we introduce practice when it's established elsewhere. And I guess it's heartening to hear that there was a well-described route for that. 

[00:33:00] Yeah. 

So I think that that's true. And the other thing that, of course we did in Edinburgh was that we started collecting data and then producing work that actually backed up what we had hoped.

Yeah. Always helpful to hear. 

So just, I guess, final comments from both of you in terms of anything that we haven't covered in terms of the Edinburgh experience and the data supporting that...

 Sam, should I mention a bit about tracker or, or...

Yeah, I think that's actually probably quite a good, good idea.

That is where we're going to end up isn't it? So, to try and justify the robot coming into the NHS we are doing a study  robotic unicomponental Mako, partial knee replacement versus gold standard total knee replacement. So just like the top cat study, really looking at length of stay, functional outcome and all of that. So to try and justify the cost of the robot and hopefully that's going to be recruiting this summer.  One year outcome is [00:34:00] the endpoint. So hopefully in a couple of years time, we'll have some real evidence and some real cost per quality evidence, not made up stuff from a Markov model. 

And is that open only to Edinburgh? Or is that something you're looking for the other centres for? 

No, it was just us.

 When we first put this together, which was a couple of years ago, there wasn't any NHS centres that actually had it. I think Birmingham might've just got theirs, but now there's very few NHS centres wit h a robot again. Then so...  

I hope you've registered the protocol or Fares will never let you publish it. 

Its on Clinicaltrials.gov and now we are nearly there. We're going to submit to trials. Yes.

Which actually, you know, joking aside, is the way to change practices isnt it? You know, we've all been involved in practice change and innovation, and actually you owe it to your patients and yourselves to do it properly. And it sounds like you guys really are, which is really what you'd expect from Edinburgh as a leading light in evidence-based surgery and that kind of stuff. And that's not just because you are on the line. I do genuinely believe that, but don't tell [00:35:00] Duckworth. 

But it's true, you know, you shouldn't...the fact is we are talking about robotics, which is a niche thing. I think you're probably right. I think it is going to become gold-standard. It's going to become the go-to, but there is a journey to get there and how people do that is really important. And it's important that we do it in the right way. And it sounds like you guys are so, you know, thumbs up as it were. 

One of the other things of course, is that the data, that data source, the data that you get from simply from doing a robotic case is enormous. So that, that the power of the data set that you've actually gathered whilst doing the operation is really very, very powerful just in itself. And actually I'm sure we'll, along with various outcome measures, be used as a sort of artificial intelligence algorithm solution for many problems, you know, in the future.

And will your trainees do them? Final question from me, Sam.

I think it is the ideal training tool for trainees [00:36:00] because they can't get it wrong and they get to see what they're supposed to be doing and they get that feel and they're seeing it on the computer screen. They're getting the feel of the case. But you as a boss have the reassurance of knowing they can't actually really get it wrong. So let's just say you're doing, I'm going slightly off beat here, but cup position in a total hip replacement. It lets you, it will do that for you. And it makes sure that you put your cup in the right place. Now that's a great learning resource for trainees because it's making them put it in the right place, even if they then go back and start putting hips in manually again. And that in itself is a very, very powerful educational tool. So I think it's an ideal tool for trainees.

I think you're probably right. And I think it's also that there's some, you know, the programme director's hat on, there are some objective measures there, aren't there, you know, and we're moving much [00:37:00] more towards competency and actually recording what people do, how they do it, how long it takes them, the decision-making process as you move towards and big being forced to kind of break that down, demonstrate that you've done the soft tissue balance or whatever it is, and you've understood it. And haven't overridden or not overddin the software program you started with is crucial actually, because evidence is all in training now, sadly. But it is.

Yeah.

 It removes some of the dark art of knee replacement as well. Cause it's kind of, it's all numbers, it's all there. It's not just the boss who's done it like so many times beforehand  and they just do stuff that you don't know what they're doing. This kind of...you can see everything that's going on, alignments. We talk about measured resection. We don't really do measured  resection. We do something in between. We do mechanical alignment. We're moving towards kinematic alignment or functional line, now it's called. So I think it will advance the knowledge of knee replacement rather than just, this is the way I do it. That's how I balance [00:38:00] it. And you don't know what balance... what's a balanced knee. Nobody knows what a balanced knee is. Is it supposed to be tight all the way through? Is it supposed to be laxing the lack of compartment is it...

That is a discussion for a different day, Nick. That is a very long discussion.

So I just want to thank you both really for a really thorough overview of the Edinburgh experience, I guess, and some real insights into why you've made the decisions that you've made you know, and credit to getting one of the first robots in the country available in the NHS. And, you know, I'd be interested to know how you get on. Cause these things are impossible, or obviously not impossible because you've achieved it, but persuading people to invest extensively in stuff they don't have to buy is actually really difficult, you know. And so well done on that front.

I'd like to thank Stryker as well for supporting us with this podcast this evening. And thanks very much Sam and Nick. Have a good evening. 

Thank you.  Thank you.