Postoperative opioid cessation rates based on preoperative opioid use: analysis of common orthopaedic procedures

Listen to Mr Andrew Duckworth interview Dr Azeem T. Malik  about his paper 'Postoperative opioid cessation rates based on preoperative opioid use: analysis of common orthopaedic procedures', published in the December 2019 issue of The Bone and Joint Journal.

Click here to read the article

Transcript

[00:00:00] Welcome everyone to this month's BJJ Podcast. I am Andrew Duckworth and a warm welcome to our last podcast of the year and of the decade from your team here at The Bone & Joint Journal. It is now every year since we commenced our new podcast series and from all the team we would like to thank our readers and listeners for the comments and support we've received so far as well as to our many authors and guests interviews who have taken part.

In the past year, we've covered a range of topics and we've heard from a range of authors from across the globe. We've spoken to guests all the way from the US to India, including a fascinating dialogue with Mr. David Beverland on the role of follow-up post total hip replacement, as well as last months very informative discussion with Dr. Matt Abdul from the Mayo clinic on the survivorship of primary total knee replacement for osteonecrosis of the knee. Through this, along with a series of podcasts that accompanies our supplements from the American Hip and Knee Society closed meetings, we do hope these are improving the accessibility and visibility of the studies we published for both you as our readers, as well as for our many authors.

So moving on to this month study, as you know, over the next 15 to 20 minutes or so we'll be covering [00:01:00] a range of aspects from the chosen work, emphasizing the important points of how the study has been designed, as well as the key findings from the data and how these potentially fit into each of your day-to-day clinical practice. We also hope to give you a behind the scenes insight, should we say, into how the authors have developed the study and give them an opportunity to put forward the key findings of their work. 

So today I have the pleasure of being joined by Dr. Azeem Malik from the Ohio state university Wexner medical center in Columbus, Ohio to discuss their study entitled Postoperative opioid cessation rates based on preoperative, opioid use and analysis of common orthopedic procedures, which we published in the December edition of the BJJ. Welcome. Azeem and a big thank you for taking the time to join us today. 

Thank you for having me. 

So if we move straight onto the paper,  the aim of the study was to characterize the relationships between pre and postoperative opioid use among patients undergoing common elective orthopedic procedures, which is obviously a hugely topical area of interest, not just in medicine, but globally as in recent years, the negative consequences of widespread opioid use is becoming [00:02:00] increasingly clear. So if you could just give us a brief introduction to the paper and where you currently feel the literature stands in this area with regards to the opioid crisis and MSK pathology. 

So definitely an excellent question to start off with. So simply stating the opiod crisis in the United States is currently a major public health concern. I mean our rates of opiod usage are pretty much the highest in the world right now. Canada is a close second and obviously Europe and the UK are nowhere as close as we are. Now coupled with the fact that we have been using opioids continuously over the past two decades, and we have seen improvements in access to care for our patients and more individuals are now getting elective surgeries, which were obviously not of the norm to that get to go. It's not exactly surprising that we're giving these medications more often at this point but the more concerning fact is now that we know that this is not a great drug to begin with. It's not a great pain medication, as you said, given those negative outcomes. [00:03:00] We're trying our best to somehow build up evidence so that we can launch some former protocols or some form of control measures in which we can try to limit the use of this medication over the coming years. 

That's a really good, summary. I think like you say it is a major public health concern across the globe now. Isn't it really? But what do we know Azeem already about preoperative opioid use and cessation following orthopedic procedures, is the much out there in the literature already? 

So I think over the past five years or so thanks to all these big data sets which the US has produced we are seeing a surge of literature surrounding opioid usage, particularly in orthopedic procedures. And we already know that most orthopedic conditions particularly are degenerative elective surgeries and I mean, they're pretty much a debilitating condition. So obviously patients do end up taking these pain medications before they ever come to a doctor to seek care for their surgical opinion, essentially.

[00:04:00] So for now I can tell you that cessation rates are being studied but we're still building up the ground foundation at this point. So we really need more evidence and more evidence that we can have the better knowledge we have to build up to understand on how we can control this. 

Yeah. Yeah. So it's very much in its infancy, isn't it really, even though it's been a problem for many years. So, sort of given, given all that, Azeem, what was the sort of the aim and I suppose the hypothesis of your study and how did you aim to address those limitations?

So the hypothesis of the study was essentially to characterize what is the rate of chronic preoperative opioid usage in patients who are undergoing elective, orthopedic surgeries? And then we also aimed to look at chronic post-op opiod usage in these individuals. And then we tried to see the relation to see whether chronic preoperative opioid use has impacted post-op opiod use.

Okay. Great, great. That's a great summary of what the paper aims to do. So if we move on [00:05:00] to, I suppose, how the study was put together. It was a large retrospective study using large data, big data from the human and national claims database in the US and it covered a period just under 10 years from 2007 to the third quarter of 2016. So just for our listeners some of who won't be really very familiar with that database. Could you brief a brief overview of it and what data it routinely collects and I suppose how accurate it is? 

Ah, so I can tell you that it's a private peered commercial data set, which means that obviously there are so many private insurance companies present that the US so this is just one of the many right now. And it's based off billing instances so anytime an insurance company reimburses a provider or a hospital for anything that the enrollee seeks healthcare  utilization for it gets recorded into that data set and it is made available to researchers to actually access it, identify patients, do analysis, look at medication users and all these kinds of things.

Okay. And I've [00:06:00] seen it obviously before in many other papers - it's quite commonly used in the literature, isn't it as a data source?

Exactly.

So moving on to the actual crux of the study - so if you give us basically a brief overview of the inclusion, exclusion criteria for the study, which you'd lay out very nicely in the flow diagram, in the paper.

So briefly summarizing we identified patients with billing instances or primary indexed arthoplasties, carpal tunnel surgery with either *inaudible* and spinal  fusions. We then went on to exclude patients who were undergoing a non index or non-primary surgery revision case, patients who are undergoing surgery for fractured or trauma indications, and those who were undergoing surgery for scoliosis. So we were essentially trying to capture a perfect elective, degenerative, patient population. And then we also aimed to ensure that we were only including individuals who maintained the coverage with that specific insurance company over the one-year period so as to prevent any dropouts. 

Yeah. So it's very robust inclusion-exclusion criteria really, and very much looking at those core active procedures.

[00:07:00] Exactly. Yeah. 

Okay. And what sort of data was collected with regards particularly, I suppose, general data and also with regards opioid use and how you group those for the study itself. 

So I think the table itself in the paper is a bit tricky to understand, which is why I'll try explaining this, but I would still urge listeners to actually go and read up the paper to get a better gist of how we characterized these preoperative opioid usage. But essentially preoperative opiod uses was categorized basically into two groups - those who had more than six months of continuous use prior to undergoing surgery, which was defined as four or more prescriptions, and those who had history of at least more than six months of use, but then stopped within three months prior to surgery. So essentially that defined an opiod free or weening period. 

And then post-op opioid usage was characterized into chronic, which was essentially eight or more prescriptions within the whole year after the surgery. And then we also had two initial sub definitions within this chronic opiod use definition, which are [00:08:00] basically those individuals, who stopped taking chronic opioid use or those who had only short term opioid use, which was four most prescriptions within a period of three months.

Yeah, and I think that's a great summary. It's quite complex, but I think it's a good way to group them. And that's a really nice explanation of it. So in terms of the analysis you performed mostly there, again, they're very nicely laid out in the manuscript, but for our listeners could just give us a brief, basic, simple overview of what was performed?

So in a really brief summary, what we were trying to do is we essentially took hold of a multivariate logistic regression model. And then we used that to assess the independent impact of pre-op opiod usage on post-op opioid usage, while controlling multiple covariates, which included obviously baseline with comorbidity burden, age of patients,  other comorbidity measures, essentially. So we're trying to control for any possible confounders that might impact post-op opiod usage. 

Perfect. Yeah. I mean, I think that's pretty, one of the important things with this type of study is trying to control for those factors. Isn't it? I agree.

So moving on to the results [00:09:00] of study, just to recap for our listeners in terms of the patients included, there was just under a hundred thousand - 98,769 so large datasets. And of these are the range of procedures performed.  Just under a third of those were total knee replacement and a range of hip arthroplasty and spinal fusions. And over 10,000 carpal tunnel releases and almost 20,000 rotator cuff repairs. And for these, just to reiterate for our listeners, you looked at chronic opioid use after surgery among opioid naive  patients and also opioid cessation rates amongst chronic preoperative users, and the effects of stopping before surgery amongst the chronic users in particular. So could you just detail the key findings in relation to the pre and postoperative opioid use, particularly in relation to the opioid naive and chronic users?

So I  think, some of these answers are going to overlap with the rest of the questions too, but so well firstly we looked at the proportion of chronic pre-op and post-op opiod users and this pretty much *inaudible* all across the board. For the most part as expected the patients who are undergoing surgery for more [00:10:00] debilitating conditions, which include a total joint arthroplasties and spinal fusions. They obviously had high rates of chronic pre-op opiod usage. And then obviously also had lower proportion of opiod naive individuals. 

 How do these figures vary between the ranges of procedures you looked at in the study?

So, again, procedures which were done for more debilitating conditions, which were spinal fusions, total joint arthroplasties again, they had typically higher rates of chronic pre-op  opiod usage and a lower rate of opiate native individuals. These individuals were also more likely to continue using opioids chronically in the post-operative period specifically.

Yeah. No, absolutely. And like you say, there's just so much interesting data there. I remember looking at the third  table and the chronic postoperative opioid use and the total hip replacement was almost similar to the carpal tunnel releases. It's just fascinating these figures and seeing them like that, it's just sometimes not what you would expect, really.

So what did you find with regards to the risk factors for the chronic postoperative opioid use? 

[00:11:00] So I think most of the risk factors, which we found not well detailed at literature, for the most part. I think the most important key finding with regards to risk factors was that one, the biggest risk factor is always going to be those patients who are taking opioids, preoperatively they're always going to end up turning into chronic post-op opiod users.

And I think moving onto the next sort of part is that patients, even among these chronic opioid users, those who actually had a three month opiod free period before surgery, they actually had a better cessation rate. So they had the lower rates of continued opioid use. And that really means that patients who had that opioid weaning protocol in place, or you had an opiate free period, they actually ended up doing much better than chronic opiod users. And this is a finding which is really helpful to surgeons primarily because it really gives supported the fact that we really need to look at possibly launching a broad opioid weaning protocol across [00:12:00] institutions all across the US. 

Absolutely. Absolutely no. So that sort of moves us nicely along to the greater implications and the other study.

So the strengths are without questions, very large data set, very robust analysis performed. It's clearly gone a long way into characterizing the preoperative and postoperative opioid use patterns among patients undergoing a range of common elective procedures, orthopedic procedures. So just to summarize again, what do you feel the key findings of the work are, but I suppose with a caveat of any potential limitations of the data.

Okay. So I'll start over with the key finding, which is essentially the fact that in chronic if you have a chronic pre-op opiod user, if you give them a three month opioid free period, they're less likely to actually end up taking opioids continuously in the post-op period.

And with regards to the limitations, I think one of the drawbacks of using these insurance claims data sets are that they actually rely on billing instances, which means that you have a bunch of coders in the hospital who are sitting somewhere down in the [00:13:00] basement, putting in codes, sending this information to an insurance company. And obviously human error is something which we cannot remove from this source of data. There are always going to be some miscoding they're always going to be missed patients. But for the most part, I think the power of the study primarily relies on the fact that it's such a large sample size that even if there are some miscoding, we're pretty much accounting for that.

No. Yeah. I totally agree. And I think given, like you say you can't account for those errors, but what you can account for it is by having such a large data set like you have. And then in terms of the previous  literature, how do you feel this fits in? Or how does it compare in particularly maybe to the type of procedure performed? What do you think it tells us regards to that? 

So I can tell you that most of the prior literature has largely focused on identifying risk factors for post-op opioid usage. They haven't exactly looked at identifying whether an opioid free period may actually impact post-op opioid usage. So that was essentially the main step of our first study [00:14:00] was to somehow prove to everyone that, Hey, these patients are actually not taking opioids within the last three months. And they actually have a better cessation rate as compared to patients who  continually take opioids for the most part. 

Yeah, no, absolutely. And I think one of the other, I mean, just moving on finally to the other interesting findings of the study and something you talk about in the discussion is that this study has suggested that orthopedic surgery, our intervention decreases chronic opioid use on a population level. I mean, how does that, I suppose, how does that fit into the literature and where do we go forward with that? How do we take that forward? 

So I think this is pretty much an intuitive finding, right? Because most musculoskeletal conditions are really debilitating to begin with. And like I said patients do end up taking pain medications. So if they do have the option of undergoing surgery, I mean, if you think about it, total joint arthoplasty technically can be declared the surgery of the decade, right. Or not of the decade, but the surgery of the century right? Because you do a total joint arthoplasty and an [00:15:00] individual can essentially regain that functionality, which they did not have before. So obviously they will not have to resort to pain medication. So obviously on a societal level, you're going to see patients not taking opioids to begin with.

But at the same time, I mean, it's sort of a conundrum. It doesn't exactly mean that orthopedic surgery would lead to reduced opiod levels. We're just saying that due to improved access of care now that patients can get surgery for these degenerative elective conditions, you would likely see less people taking opioids for simply managing these pathologies. 

Yeah, no, absolutely. And you've also already alluded to it, but in terms of the implications of studying moving forward, what do you think the key thing is from this study?

 Well, I think the next step, we want to sort of identify what will be the right opioid weaning protocol? How long do we make patients wait? And if we do make patients wait, what other alternative pain medications we can place them on because obviously we cannot just take them off opioids and ask [00:16:00] them to wait for that six week or three month period for the most part, because that is obviously not going to be great for their functionality.

And I think the last thing would also be to look at the new methods of multimodal analgesia, which are kind of coming out. So that will be also effective in curbing the postoperative opioid burden too. No. I agree. That's really good. Well, Azeem that's probably all we have time for today, but thank you so much for joining us and congratulations on a really excellent study that is without doubt, an invaluable addition to the literature in the air and given us all much food for thought. So thanks for joining us. 

Thank you so much for having me.

 And to our listeners we do hope you've enjoyed joining us, and we encourage you to share your thoughts and comments through Twitter, Facebook, and a like. Feel free to post a tweet about anything we have discussed here today. And thanks again for joining us.