An assessment of early functional rehabilitation and hospital discharge in conventional versus robotic-arm assisted unicompartmental knee arthroplasty

Listen to Mr Andrew Duckworth interviewing Professor Fares S. Haddad (Editor-in-Chief of the BJJ) about his paper "An assessment of early functional rehabilitation and hospital discharge in conventional versus robotic-arm assisted unicompartmental knee arthroplasty" published in the January 2019 issue of The Bone and Joint Journal.

Click here to read the article

Transcript

[00:00:00] Hello, BJJ listeners. I am Andrew Duckworth and a warm welcome to  our second podcast from your team here at The Bone & Joint Journal. As many of you know, the aim and hopes of our podcasts are that we will improve the accessibility and visibility of the studies we publish for both you as our readers, as well as for our authors. 

During the next 15 minutes or so we hope to discuss a range of aspects of the chosen study emphasizing the important points of how the work has been designed as well as the key findings from the study and how these potentially fit into your everyday clinical practices. We also hope to give you a behind the scenes insight into how the authors have developed the study and how it's gone through the peer-review process and giving them an opportunity to put forward the key findings of the study while also, hopefully, adding to your reader experience. 

Today I have the pleasure and honour of being joined by our editor-in-chief here at the BJJ, Professor Fares Haddad, to discuss their study entitled An assessment of early functional rehabilitation and hospital discharge in conventional versus robotic-arm assisted unicompartmental knee arthroplasty, which will be published in the upcoming edition of the BJJ. [00:01:00] Welcome Prof and a big thank you for taking your time to join us.

Thank you, Andrew. Thanks for doing this. It is great to have the opportunity to discuss this paper. 

Great. So Prof your study is obviously looking at unicompartmental knee replacement, which for isolated medial compartment OA is known to be an effective procedure that's reported in the literature to have good patient satisfaction, sound preservation of the native knee kinematics and good long-term functional outcomes. But as you nicely put in your paper with an increased balance of, with an increased risk of implant failure, and shorter time to revision surgery, when compared to with total knee replacement. 

Prof to get us started, can you give us a brief background to the paper in relation to the state of the current literature on UKA and what are felt to be the perceived benefits of robotic-arm assisted surgery?

Absolutely. So I think the reality is many of us believe that UKA is a smaller intervention than total knee arthroplasty. It gives us the opportunity to restore more normal knee kinematics, to allow patients to recover with less morbidity with [00:02:00] a better functional profile, if you like, the more normal gait pattern or speed. And some data suggests there's less mortality from that intervention. 

I think there's lots out there that has been in the BJJ and elsewhere that suggests that unicompartmental arthroplasty performed well in the right patient is an excellent operation and that we should be performing that in a proportion of our knee arthroplasties and I would say, in my practice, that's about 15 or 20%, but certainly that goes up to, as you know, some authors were talking as high as 40 or 50%.

Yeah. 

I think we have to face the reality that although all of us see excellent outcomes in individual patients with this procedure, we're also aware that some are failing early requiring revisions. And when we've looked at this before, it's tended to be related either to surgical [00:03:00] technique and fixation or related to progression of arthritis, which is often malalignment or over correction of the deformity.

Yeah. 

And so we certainly saw an opportunity in a very established uni practice, over the last couple of decades, to see a new enhanced technology and see whether we can use this technology to replicate, if you like, a more native alignment for the patient, more reproducibly, hopefully limit surgical error, hopefully improve fixation and balance. And see where we could go with that. And this study really, we shouldn't over egg it because it's really pilot data. This is the introduction of this technology in our practice and in comparison with the prior data that we had from our existing technique that we had evolved over the past 20 years. 

Yeah.

And it would be fair to say, the literature is still very much in its infancy about the robotic arm assisted surgery, would you agree with that? We are still learning as [00:04:00] we go. 

Absolutely, no doubt about that. There is very little that's written in this area. This is a relatively novel technology. When you look at the literature, there are a few studies out there, but they're really case series generally often from innovators or early adopters. So this is an area that requires scrutiny. And requires much bigger studies. 

Yeah. Yeah. So as you alluded to Prof, it was obviously, you know, like you say, it's a series of comparing the two different types of uni knee. So you're obviously your primary objection of the study was to compare the difference in the post-operative pain. That was your primary outcome between the conventional jig-based mobile bearing uni knees and a robotic-arm fixed bearing uni knees. And there was obviously a range of other secondary outcome measures, which obviously we will come on too.  

In the study, you had 146 patients and they all had symptomatic medial compartment OA. And you had 73 consecutive patients who underwent the conventional surgery then followed by 73 consecutive patients who underwent robotic-arm [00:05:00] surgery. So just to give the readers an idea about the type of patient involved, could you just go through the inclusion, exclusion criteria for us Prof and a bit about the, obviously the surgical techniques which are well described in the paper and the perioperative care.

So these are all patients that would fit in the category of what you would call anteromedial or various osteoarthritis. They have isolated medial compartment disease, the ACL is  intact,  they have not all had cross-sectional imaging in terms of MRI scan. They've had CT if they've had a robotic procedure, but as far as we know, they have correctable varus with a well-preserved lateral compartment. We've accepted those with patellofemoral disease on the medial side or centrally in the trochlear, but we've excluded the patients who had a patella subluxation or grade three or four damage laterally. We've also been careful to fit fixed flection about 15 degrees was the extreme we would go to and 10 degrees of varus. Anything beyond that we tended to rule out [00:06:00] on the basis those patients ended up with a total knee replacement. 

And so in terms, Prof, all patients underwent a general anaesthetic, was this part of your practice? Because obviously it's a single surgeon series and how would you sort of relate that to the, you know, the standard of, you know, it's quite variable, isn't it? The practice regarding whether it's a GA or a spinal or just a GA and a block? 

No, it's a great, it's a great question because it's just been our practice as a historical practice based on the two anaesthetists who were involved in this series. So all these patients were anaesthetised by one of two anaesthetists who just happened to be comfortable with GAs. There's no question that the enhanced recovery pathways that we're looking at now increasingly and the push towards day-case arthoplasty, increasingly people are using single shot spinals and regional blocks are becoming more and more common. But our practice, and we are very careful here not to change our practice and surgical technique when we introduced the robotic arm, [00:07:00] has been to give a general anaesthetic. So we've continued doing it. 

Yeah. And I suppose in some ways, you know, because it was GA if anything, you could have underestimated the results really of the robotic arm. Would that be fair to say in some ways? 

That's true. That's, that's absolutely true.

In terms of the outcome measures, Prof, what sort of made you pick the, you know, the primary outcome measure as it was in terms of the pain and then the second, what was your thinking behind that? What drove those decisions?

 A number of things. I mean, the first is that although there are very few studies on robotic surgery, there was a randomized control study, 15 or more years ago from Justin Cob that suggested a similar signal with decreased early pain and Mark Blythe study in Glasgow also suggested decreased early pain. So we thought they might be a signal there. We wanted a validated score, which this pain score is. It's been more used than the shoulder and the knee, but there was some preexisting data we could do a [00:08:00] power analysis on. And we wanted something that the patients could score rather than us being able to bias it. So this is a patient derived score. And it's measuring the one variable we think is key to focus on here, particularly as we saw the opportunity to try, if we reduce pain, we're likely to be able to reduce length of stay and improve recovery.

Yeah. I mean, I think it's a interesting point that isn't, I think obviously pain is the primary outcome, but a lot of your secondary outcome measures are sort of associated with that aren't they, in terms of opiate use, how quickly they can get to a straight leg raise, the number of physiotherapy sessions. They're all sort of markers of that really aren't they? 

Absolutely. I mean, I think what this study really shows is for maybe one of a number of reasons, and we can discuss those, the patients in this study who had the robotic-arm used, had less pain and hence were able to get moving earlier, needed less analgesics and were ultimately able [00:09:00] to hit their physio milestones and get out of hospital more quickly. 

Yeah. Yeah, absolutely. So that moves us nicely on to the key results to the paper, Prof. There were 46 patients. They were obviously very well-balanced in terms of their demographics which you sort of expect with the way I suppose you've recruited them and the nature of the practice and they seem to fit what you'd say as a standard sort of uni knee group. And obviously just, if you could summarize what you think the key findings are of the outcomes that you looked at, including pain and then all the secondary outcome measures. 

I mean, I think the biggest difference we saw was that these patients had less pain, particularly during the first couple of days. As a surrogate of that, they were able to mobilize more quickly. They required fewer physio-therapy interventions before they hit their discharge milestones and they required less total *inaudible*.

And in terms of the physiotherapy sessions was that that was very standardized? There [00:10:00] was the median number in the conventional group was nine and then in the robotic arm it was five. But that was just, that was just purely indicating in terms of what they needed before they fit the discharge criteria. Is that right? 

Yeah. That's one of the nice things about the pathway we had in place in that first of all, it's important to say it's the same group of physios, the same group of nurses. This is the same one hospital, one surgeon, one operating theatre suite all shifting in the same direction. So there was no big change here. The standardized pathway of physios seeing patients in the morning and the afternoon, third time, if needed that day and doing the same length session each time with a view to hitting those milestones before the physios ticked a box saying milestones met, ready to go home. 

Yeah. Yeah. And in terms of your postoperative complications, they were followed up for 90 days or roughly three months. That's right. And there was no difference between the two groups either at that stage was there? 

No, not in terms of the overall postoperative complications. It really was the early recovery and the [00:11:00] improvement in discharge status.

And are you continuing to follow them up Prof, or is it sort of working towards a different type of study? What was your sort of plan moving forward? 

No, so great questions.  I mean, we stopped at 73 because we were worried that by then there was probably a cohort effect. In that the nurses, the physios, the junior doctors, everybody was kind of kind of aware that this was looking different. So there was a point. So all these patients have been followed up. So we're currently looking at the one year and two year PROMS data and functional outcomes. But the most important thing, this really was the preliminary study to look at the effect and see what we could measure, because now we've got a randomized study that is well into recruitment.

Looking into this, it's a slightly more elegant way in that the patients now can be blinded. In the randomized study the patients can't tell whether they've had the robotic arm or whether they've had the standard procedure with navigation. 

Excellent. Yeah. I mean, I think that's key. Obviously [00:12:00] its been very topical at the moment is the placebo effect isn't it? And I think automatically when you tell a patient they're going to get a robotic knee the assumption is they're getting something potentially better, isn't it? I think that's inevitable really. 

Absolutely. And if you think of 73 cases, you know, that's over several months. It is certainly possible here that there was a perception by the team that, you know, these patients can be pushed harder. They can move more quickly because we started from a very poor starting point in terms of length of stay for this operation. So massive opportunity. So some of this opportunity could have arisen just as a result of that. I think we have to accept there are quite significant limitations here.

Yeah, I agree. I think though, in terms of, you know, we're always striving for that level one evidence, which is obviously the best thing to do, but I think in terms of, you know, the way that you balanced two groups the best you can. I dont think you can really find the sort of case cohort series, you know, better in terms of, you know, they are well-matched and you've used the same surgeon on the same [00:13:00] sort of protocols. But in terms of sort of moving forward then, and, you know, obviously we've mentioned the randomized controlled trial, but how do you see us moving forward in terms of, you know, first of all, can we find a patient reported, you know the patient reported outcomes obviously seen as the gold standard now, do you think we'll ever find a difference in those either in the short or longer term? And do you feel that we can prove this to be  cost-effective in the NHS? 

So those are really important questions because what this paper looks at is the immediate peri-operative outcome. That's a lovely signal, but actually it's a small part of the whole story. What we've really got to see there is an economic benefit of course, to leaving hospital earlier, to potentially being able to do this as day surgery, there's a benefit to requiring less physiotherapy as an inpatient or an outpatient, there's a benefit to potentially requiring less outpatient support, but what we really need to see is good function beyond the first couple of years and ultimately the really important thing is this brings down the revision rate and that requires longer term follow-up.

[00:14:00] So... I think RCTs are great here because we can then see if the signal we've picked up here is real, but we're then going to have to go beyond the sort of single centre RCT to see if this is generalizable beyond surgeons who are doing a high volume because the very interesting thing about navigation and robotics is you get so much information back, that if you're into that, if you like that, that's tremendous and you can probably improve what you do much better, but actually it can also confuse. So it's really important to see that this is generalizable, but I think the registries are going to have an important role to play here. What we really need to be doing is looking at two year data, five-year data and the registries and saying that when this becomes disseminated, when this is out in many hospitals, are we seeing a lower intervention rates? 

Absolutely. I think like you just alluded to there. I think the one thing whenever, you know, we do get positive data with new innovations is that [00:15:00] the tendency is to, everybody, wants to try and pick it up. And I think we have to be obviously, and the tendency then is to expand your inclusion criteria and then it's like bushfire, isn't it? But actually what you want is, you know, the right patients and then the data will speak for itself really. I think that's the key to it isn't it really? 

It will be. And I mean, I think you mentioned whether these patients are going to function better. I don't think the PROMs are going to pick that up necessarily. We can do, and we do do in fact performance-based studies. So if you get patients up on treadmills, walking uphill or downhill at speed, if you look at their gate, I suspect if we get their biomechanics absolutely spot on, then we may do better. But I think in terms of the bigger picture, in terms of the population, the really key thing will be reducing revision rates and getting good implant longevity. That will really deliver the cost effectiveness. 

Yeah, that's what it is. Isn't it? Cause I mean, there's, I think it was recently published in the BMJ Open saying that the probably even some of the data from [00:16:00] the NJR says the uni knee replacement might be more cost effective than the total. But if we can get the revision rate down, that will be the key really to it moving forward. 

Yeah, no, that would be the big win. We're obviously a long way from that with this study, but I think this is reassuring because this study captured our learning curve. So we've not really seen anything adverse in the learning curve. These patients are doing better than the technique that I certainly have been doing for the last 20 years and the other interesting thing here, which is not necessarily part of this paper, but it comes from this cohort is that you get the implant alignment right straight away with this technology. There doesn't  seem to be a learning curve to getting the implant alignment right. There is an increase in time when you're learning the robotic technique. Undoubtedly takes the surgeon and the operating team a bit of time to get used to it. So having the robotic arm in the theatre to adjusting to the extra information. I think there is a learning effect there.

Okay. Excellent. Yeah. Well, thank [00:17:00] you so much for joining us for our BJJ Podcast and congratulations on an excellent study. And to our listeners, we do hope you've enjoyed listening in, and we encourage you to share your thoughts and comments through Twitter, Facebook, and a like, and feel free to post a tweet about anything we've discussed here today. And thanks again for joining us.