Listen to Tim Coughlin, Dr. William Geerts, Ms. Sue Deakin and Dr John Murnaghan in a round table discussion of thromboembolism and VTE prophylaxis in orthopaedics.
Papers discussed:
A double-blind, randomized controlled trial of the prevention of clinically important venous thromboembolism after isolated lower leg fractures
Aspirin or Rivaroxaban for VTE Prophylaxis after Hip or Knee Arthroplasty
Listen to Tim Coughlin, Dr. William Geerts, Ms. Sue Deakin and Dr John Murnaghan in a round table discussion of thromboembolism and VTE prophylaxis in orthopaedics.
Papers discussed:
A double-blind, randomized controlled trial of the prevention of clinically important venous thromboembolism after isolated lower leg fractures
Aspirin or Rivaroxaban for VTE Prophylaxis after Hip or Knee Arthroplasty
[00:00:00] Welcome to this month's Bone & Joint 360 Podcast. My name is Tim Coughlin and this month we're going to be looking at venous thromboembolism. Deep vein thrombosis as a common occurrence with around one per thousand people per year recurring in the general population. It is a significant risk for patients in hospital and prior to the introduction of thromboprophylaxis was historically identified as a common cause of death for patients, particularly those undergoing surgery.
Before we look at where we're currently and the future of thromboprophylaxis, we will take a quick look at the historical events that brought us to this point.
The first name description of a deep vein thrombosis was an a Norman cobbler called Raoul. The understanding of the disease and its treatment was limited in 1271. Raouls treating physician suggested a wait and watch approach, which unfortunately led to the development of paced thrombotic syndrome and [00:01:00] ulceration.
Raoul had a number of unsuccessful treatments applied with his final treatment being the application of dust from underneath the tomb of King St. Louis. Contemporaneous records surprisingly report this to be a successful treatment with complete resolution of his symptoms. And he went on to live over a decade more.
Tomb dust didn't catch on as a treatment and skipping forward 500 years, deep vein thrombosis had become a well-recognized condition. In the 16th century. Venous ligation above a clot was the treatment of choice. By the mid 19th century, Rudolph Virchow had described his famous triad of venous stasis hypercoagulability and endothelial injury.
Chemical treatment, however, remained elusive until heparin was first isolated in 1911. The first case series of its use as a [00:02:00] thromboprophylactic agent was in 1937 and the results were dramatic. Warfarin was to come to the fore shortly after in the 1950s with the two being commonly used in combination.
Low molecular weight heparin came onto the scene in the 1980s. And by the 1990s was a safe and common alternative, which no longer required closed patient monitoring or hospital treatment.
The most recently developed group of drugs started life as novel oral anticoagulants, or no acts such as dabigatran and rivaroxaban. There can be confusion as to the difference between NOACs and DOACs the direct oral anticoagulants. But this is mainly because they're the same thing. After a few years, it seems the novelty wears off. And while both terms remain in use DOAC is probably the most commonly used now.
We know that deep vein thrombosis can lead to pulmonary [00:03:00] embolism and post thrombotic syndrome. And it is generally accepted that trauma, surgery and immobilization are all risk factors. But who exactly is most at risk, how much they're at risk and what we should do to reduce this risk remains actually quite unclear.
We're going to hear today from three internationally renowned surgeons. First, we're going to hear about thromboprophylaxis in trauma patients. So I'm here with, Dr. William Geerts from Sunnybrook Health Sciences Center in Toronto. And we're here to discuss the paper entitled A double blind randomized control trial of the prevention of clinically important venous thromboembolism after isolated lower leg fractures, which was published in the Journal of Orthopedic Trauma in May, 2015. This was a multicenter double-blind randomized control trial, the patient group were those patients [00:04:00] presenting with fractures of the tibia, fibula, or ankle undergoing surgical repair in the absence of polytrauma and the intervention to which patients were randomized were either 5,000 units of dalteparin, or a placebo. The followup was 90 days and there were 242 patients randomized. Welcome Dr. Geerts.
Afternoon.
Could you define for us, cause it's one of the keys in the paper, what act clinically important venous thromboembolism certainly meant in this study and whether you think this is a devolving definition?
Thank you. So when we were designing the study, we obviously reviewed the literature and realized that there were a number of other prior trials that used surrogate outcomes, including venography and ultrasound screening of people who had received prophylaxis. We were not impressed that that outcome of asymptomatic DVG would [00:05:00] lead to a change in practice. So when we designed our trial, we chose an outcome that we thought would influence clinical practice. And so the clinically important VTE is we defined as a symptomatic, objectively proven DVG or PE. And, asymptomatic proximal DVT based on a single ultrasound, at two weeks after randomization, because we had the feeling and discussions with orthopedic surgeons that preventing proximal DVG even if it was asymptomatic would be a relevant outcome.
I guess it's interesting, based on the previous decisions using asymptomatic imaging in sort of Innography for everyone gives a much higher rate. What I was just wondering what your thoughts are on what happens to those thrombi and potentially would happen to these asymptomatic proximal ones if they didn't go on to become.
In the days when low molecular [00:06:00] heparins and DOACs were new and we had no idea whether they were effective and safe, it was reasonable to start off with studies that looked for asymptomatic events to look for the proof of concept. Do these agents, can they reduce venous thrombosis? But once it became obvious that these agents were effective at preventing asymptomatic DVG then the trials should go on to the studies where more clinically relevant outcomes are ascertained. And so either symptomatic events or symptomatic events, plus proximal DVG to me seems like the appropriate outcome measure that might influence clinical practice.
Do you think we over concern ourselves with the risk of bleeding using low molecular weight heparins or otherwise?
In short, yes. I mean, I think that bleeding should always be a concern when added claims are given to patients, but our [00:07:00] practice now for a long time, our experience for a long time for 30 years almost is that if prophylactic low molecular heparin or prophylactic DOACs are given to surgical or orthopedic surgery patients, the risk of clinically important bleeding is incredibly low, as long as patients are selected such that they are not actively bleeding at the time the anticoagulant is started. And that is why you may recall that in the early days of low molecular heparin use one of the practices, particularly in Europe was to start preoperative low-molecular weight heparin. In North America we were concerned about that and we started postoperatively and doesn't seem to matter very much, whether you start pre-op or post-op . starting post-op makes anesthesia decisions much more sensible, any bleeding that occurs intraoperatively, you cannot blame on a preoperative dose because they didn't get any.
So I really think that [00:08:00] well, bleeding is always an issue in my practice and I work at the largest trauma center in Canada, we virtually never see bleeding that is due to prophylaxis. Occasionally sick patients will have a bleed. But I don't believe that it's due to the prophylaxis. So with currently available agents and used sensibly, I think bleeding is incredibly uncommon. And I don't think for the average patient that it's a serious consideration anymore.
Now I thought it was interesting that the choice of 14 days in this study. In the UK, we get the guidance for a number of things from the National Institute for Care Excellence or NICE, and although our guidance on lower limb immobilization, isn't specific to trauma surgery, I think it encompasses elective ankle procedures. The recommendation is up to 42 days. So I was interested in what your thoughts were in the context of trauma, whether you thought 14 days shorter or longer or what the [00:09:00] guidance potentially should be for that?
Great question. So nobody in the world knows for any specific patient group, what is the optimal duration of prophylaxis, but it's a balance between effectiveness, giving it long enough that we reduce the main burden of the disease versus patient compliance and cost. And so for each patient group and perhaps even for each patient that equation differs a bit . I'm a very practical practitioner and guidance giver. And so I like things to be fairly simple. And so we felt that two weeks took care of two issues. Number one, we know that the highest risk for developing thrombosis is early on, or at least initiating the clot is early on. And so we wanted to cover that period. And then the two week period was chosen as [00:10:00] a reasonably acceptable duration that patients might tolerate doing self-injection. The guidance for prophylaxis in total hip replacement has been for up to about a month of prophylaxis. When we started our DOAC trials in arthroplasty, we said, why does it have to be a month? Has anyone compared different periods of time? And we said, why don't we choose a 15 day period of time. And if we have outcomes, we'll go longer. But it turns out that the event rates are incredibly low if you give prophylaxis in arthroplasty patients for that period of time. So while nobody knows exactly what the optimal time is I think going for long periods of time is not reasonable and I'm totally aware that in some centers in Europe, patients are given prophylaxis as long as they have a plaster cast on. To me, that is an outrageously long period of time and unnecessary period of time [00:11:00] for prophylaxis.
Because do you think it's the injury? Do you think it's the surgery or do you think it's the cast or the weight-bearing status or is it a combination of these things that brings us into the risk or the not risk group of VTE?
I think is likely combined, you know, that in animal models and in humans, it is incredibly difficult to cause thrombosis with immobility alone. So if you take a healthy person and you put a cast on their leg, you would be really surprised if any of those patients developed thrombosis. If you injure the leg before you put the cast on, or you have a fracture or you do surgery to the leg, then you may experience events. And I think the immobility modifies the risk due to the primary insult, which is the fracture and or the surgery. And so patients who are less mobile are [00:12:00] probably more likely to manifest a symptomatic event than patients who are partially mobile.
That's really interesting. What do you think the future direction for research should be then when it comes to trauma patients?
I presume you mean isolated lower extremity?
Yes, a similar population to this study.
Yeah. Yeah. Well, I would really like to know what are the risk factors for thrombosis in those patients? And we really don't have good data. If you look at the entire burden of the entire body of literature in that area, you would include patients who are older, patients who have surgery, patients who are less mobile, patients who have cancer, patients who have previous VTE. Those are all patient groups, additional comorbidities. They are all patient groups that in some study has been shown to be a risk factor for thrombosis.
Yeah.
But the weight of each of those risk factors, we don't know. The only [00:13:00] one that I personally take into account when it comes to clinical practice is somebody with a previous venous thrombosis who now has a fracture and and is being discharged from hospital. That's a patient group where I would consider post discharge prophylaxis with the DOACs in that selected group. For the others age, mobility, cancer, obesity, those are all factors I don't know the weight of, and I need you to know that before making a guidance decision. So I'd like to know in large populations of patients, and certainly these are patients now in big databases around the world, you could look at it - symptomatic events and look at what were the key risk factors for those to help identify a subgroup of patients who you might offer prophylaxis to.
Now, [00:14:00] we're going to go on to hear about integrating thromboprophylaxis into a virtual fracture clinic environment. So I'm here with Ms. Sue Deakin who's a consultant trauma and orthopedic surgeon in West Suffolk. Thank you very much for taking the time to speak to us.
So you were involved in setting up one of the first virtual fracture clinic services in England. And it's been running now for over six years. So it'd be really interesting to hear your perspectives on how thromboprophylaxis integrates into a virtual fracture clinic environment, because a lot of us don't have that experience. How did you go about integrating thromboprophylaxis into your virtual practice clinic environment?
Okay. So we recognize that there is a risk of DVT and PE following lower limb injuries that we might be treating only in the outpatient department or even not meeting those patients at all, because some of our patients are discharged by leaflet, for instance, patients with a fracture of [00:15:00] the base of first metatarsal. And so as a group of orthopedic surgeons in combination with our thrombo prophylactic committee at the hospital, we agreed that we would need, according to NICE guidelines to consider using prophylaxis in all patients who are in below knee casts or walking boots. We use quite a lot of walking boots instead of casts. Which often has the advantage of being able to be removed so that the calf muscle can be moved.
So we didn't feel confident that the orthopedic surgeons would reliably remember to do VTE assessment in the emergency department or via the virtual clinic. So we discussed this with our emergency department. We'd managed to get some really good bonds between our orthopaedic team and emergency nurse practitioners who see the majority of these patients in the emergency department. [00:16:00] And emergency practitioners are much more likely to follow a protocol, I think, than orthopedic surgeons. So all patients who come through the emergency department who are placed into a walker boot or a cast get a VTE assessment. And the patient is assessed. The offer of tinzaparin is given via shared decision-making for six weeks. But we do reassess that situation if the patient then returns to a face-to-face clinic at some point.
Okay. And tinzaparin is obviously your agent of choice, I think probably because of the NICE guidelines. Have you considered whether aspirin could be an option in the future if NICE guidance were to be updated?
Yes. In fact, many years ago , I've been here for 18 years, we used to use aspirin for almost everything. And it was because of NICE Guidelines [00:17:00] and from the thrombo prophylaxis committee that it was decided to change to tinzaparin for our patients. So, absolutely yes. Even though we don't have absolute evidence anywhere that compliance with injectable heparin is any less than with aspirin, common sense tells us that aspirin would probably be more convenient for the patient, cheaper and more convenient for us. Possibly, but not definitely patients would be more compliant with taking it. And if the evidence is there, if somebody can find the evidence to assure us that it is equivalent or better than tinzaparin as well, then that's a win-win.
I think that's really interesting to hear. Cause we heard earlier on in the podcast, the experience in Canada was there was quite a lot of resistance. They are using injectable low molecular weight heparin as thromboprophylaxis. But it sounds like you've not had the same problems. In general, do your patients seem to be taking it as prescribed them?
[00:18:00] They certainly tell us they do. They don't complain about it. They're relieved when they have to stop when they're allowed to stop. Interestingly we're near Suffolk in East of England with the nearest airport being Stansted. So we get a lot of people from the continent to who nip over here or nip over there and get an injury and come back already on tinzaparin from being in France or Holland. So I think, I don't know whether culturally we're any different from there, from the team in North America. Because my impression is that they're much more used to be expecting to be on some form of DVT prophylaxis after any orthopedic injury.
That's interesting. And I think the the NICE guidance isn't specific to trauma patients . It more generally refers to lower limb immobilization. Do you tend to use the full 42 days of [00:19:00] prophylaxis when you've got patients in a below knee splint or cast? Again, I ask this because we heard the experience in Canada was that they were using much shorter durations of around two weeks rather than a week.
Yeah. So I think we take it on a case-by-case basis here. So if we have somebody who has been in a walker boot who has been weightbearing on or has been in a cast and then becomes weight bearing, then we'll have a discussion with the patients again in the clinic when we see them face-to-face and we may reduce the length of time, depending on how well they're weight bearing and how well they're moving their ankle when they remove their boot. And again, we talk to the patient about whether we think it's risky or not remembering that none of us know exactly what the risks are.
Finally, we're going to go on to hear about thromboprophylaxis in elective hip and knee arthroplasty patients. I'm joined here by Dr. John Murnaghan who's kindly joined us from Sunnybrook Health Sciences [00:20:00] Center in Toronto. And he's here to speak with us about the the research paper Aspirin or rivaroxaban for VTE prophylaxis after hip or knee arthroplasty, which was published in the New England Journal of Medicine in February, 2018.
So the summary of the paper is it's a multicenter double-blind RCT with a non-inferiority design and intention to treat analysis. The patient group being studied were undergoing elective hip and knee arthroplasty. The intervention was that all patients received five days of rivaroxaban at 10 milligrams orally, and then there were randomized at that point for the total hip replacement group to either 30 days of continued rivaroxaban or 81 milligrams of aspirin. And for the knee replacement group, randomized to nine days of continued rivaroxaban or 81 milligrams of aspirin.
They overall followup for us for 90 days. And there were [00:21:00] 3,424 patients included with almost no loss to follow up with the primary outcome being symptomatic VTE and the primary safety outcome being major bleeding or localized bleeding event.
And I think the key major finding for the study was that aspirin and rivaroxaban showed no significant difference in the rate of VTE or bleeding complications. When given after an initial period of five days of rivaroxaban. Does that sound reasonable Dr. Murnaghan?
Yes.
I think there was a few really interesting discussion points I wanted to bring out from this study. And my first question really was why not design the study from the outset as a direct aspirin versus rivaroxaban study rather than giving the initial five days of rivaroxaban?
Well, at the time of the design study, there wasn't really strong evidence to show that aspirin was effective enough in this high risk population after joint [00:22:00] surgery. So as part of a programme of research, we decided to start with five days of treatment and then go to aspirin. And if we could show it was non-inferior, the next step in that programme research was to go head-to-head aspirin versus five days of rivaroxaban and aspirin.
And is that something that's in the works now?
Well, yes. And, you know, good old COVID we actually had our site initiation visit ready to go last February and boom, just all of our research kind of collapsed due to risk, but we're back up operating at about 85% now and are just about ready to go. So I hope within the next four to six weeks, we'll be recruiting patients into the aspirin immediately after surgery versus five days of rivaroxaban and go from there.
That sounds fantastic. I think we're all very much looking forward to the results of that. And I guess we're looking at what, four or five years time [00:23:00] for results?
Well, we've got 16 centers across the country, so we actually got our EPCAT II study done eight months shorter than we planned. So the push is on to try and get it done within the next two years.
That sounds great.
Now, I guess my second question was when designing the protocol, we've obviously specifically said there is five days of rivaroxaban, how did we choose five days versus maybe three or seven or any other amounts?
Yes. Well, that's quite interesting. We had done a prior study with almost 4,000 patients that were followed up just using rivaroxaban and it was a phase four trial. So we actually looked at what the time course was of DVT and PE events after surgery and certainly in the knees the vast majority of PE events were within those first five days. So we felt with safety in mind for our [00:24:00] patients, that if we cover them for the first five days, then we could see how things went after that.
Okay. That's very interesting. And actually looking at the data from that study, the DVT is a much more graduated throughout the first three months really.
It's fascinating isn't it? We hadn't seen data like this before and there's just kind of a slow progression with the DVTs, but the PEs, if they're going to happen, tend to happen very early.
Do you think the clots todof those PEs then are happening pretty much peri-surgery, at time of surgery and it's that clot that's causing that early burst of pulmonary emboli?
I think that's a very plausible argument. Can't be sure, but that's a very plausible argument.
Very interesting. Now I think one of the things that people often worry about is bleeding with any kind of anticoagulant in this sort of peri-operative time. So I think in this study, the majority of the bleeding complications occurred within the first 10 days, but there was less data on when the VTE events occurred, which [00:25:00] obviously we've looked at in this study. Did you have a chance to look and see whether that these results we've just discussed were replicated in this data?
The bleeding or the timing?
For the timing of the VTE events, because I think that the data from your study shows that the majority of the bleeding events are all early. I was just wondering if you'd replicated the results of pulmonary embolize and DVT on this study.
No, unfortunately I couldn't convince my co-investigators that it was an important enough endpoint to keep track of that closely.
Okay.
We don't have confirmation data from the second study.
I think that the first study is so big that I think it seems likely that's going to be replicated. Is that fair to say?
Yes.
Very good.
Now when thinking about the 20% of patients who were on long-term aspirin in this study effectively, that meant because obviously everyone was blinded, they got a double dose of aspirin if they were randomized to the aspirin group. And there was, I mean, accepting the numbers are very low for bleeding, so [00:26:00] there's no significance to it. I think you mentioned in the conclusion, there was maybe a trend toward more bleeding in those receiving a double dose of aspirin. Do you think we know enough on the correct dose of aspirin to use in the context of VTE prophylaxis, whether it should be 81 higher or lower doses, or is that something that still needs to be elucidated?
Well, I guess the evidence is accumulating. We chose 81 because it had been shown to be effective in a hip fracture population. So we thought it would be a reasonable thing to try here. So that's the way we've been going. There was a stratification that included that people using longer-term aspirin. So we kept track of those and that's how we could analyze for that to affect it. But it was small and not significant. So we think anything in the 80 to 200 milligram range should be effective.
You don't think there's any [00:27:00] reasons to suggest that it could be less than 81 might be effective. I guess that's the only other consideration. It's difficult isn't it?
It is because I don't think there's anything marketed under 81.
No. Yeah. I guess that that's one of those things we may never know. Just as long as me know that 81 is effective.
Yes.
Very good.
Now, the last thing that personally interests me, just purely, because I'm a bit of an MCID nerd because my thesis was based around the MCID calculation. I thought this was interesting because when we're looking at something that is relatively infrequent as VTE is, that occurs around or in the under 1% range, it was interesting that your finding from an expert group was that the minimum clinically important difference in terms of the shifting of the needle for VTE events was about 1%, which actually represents quite a large increase over and above the rate that you think. And in this study that wasn't so relevant because [00:28:00] non-inferiority was shown. But do you think that seems like quite a large number and I just wanted to get your thoughts on that figure.
Well, that's very perceptive. And the practical thing is in the design of a study, if you're going for a 0.5% change, it's got huge implications in terms of your number of patients and the cost of doing the whole study. And then the other thing is the confidence with , you know, there was a 0.5%. Is that really going to be enough for clinicians to switch? 1% as a group after some prolonged discussion, we decided, yes, I would think about changing if you could show me, there was a 1% difference. And fortunately the way the research worked out, it came out just about even so it was not inferior. So that worked out fortunately.
Well, look, thank you very much for taking the time to speak to us today. I've really enjoyed our conversation. Is there anything else you want to add that you think we should know about the paper or future work?
Well [00:29:00] I'm cautiously optimistic about this role of aspirin. The question is in my mind, is it going to be enough early after total knee? And I guess time is going to tell that for us, I hope fairly soon up till now we've been using a rivaroxaban dose post-op day one, usually at 10. O'clock just the way our processes work. And the question is if we go to aspirin, will an earlier dose of aspirin help and that's I guess something we've got to sort out locally.
I think we've had some fascinating insights into the future direction of thromboprophylaxis in orthopedics. I hope you've taken away some insights from this month's Bone & Joint 360 Podcast. Look out for our next episode, discussing the current issue of the journal in the next few weeks. And if you want to subscribe to the podcast, find us on your usual podcasting app.