S7 E9: Best DOAC for VTE; Muscle loss with GLP1s; CDSSs - why don't we use them?

Welcome to the Hot Topics podcast from NB Medical with Dr Neal Tucker. In this episode, we have three new pieces of research that affect us in general practice. 

First: What interim anti-coagulation would you give your patient with suspected VTE? One common option has 5 times the rate of major bleeding.

Second: How much muscle mass do people lose with weight loss on GLP1s? Is this different from lifestyle-based interventions, and is there anything patients can do about it?

Third: Why don't we use clinical decision support tools for cancer? This research looks at the barriers and then I over-share my own thoughts on the subject...

References

NEJM Apixaban or rivaroxaban bleeding risk with acute VTE

Diabetes, Obesity & Metabolism Lean mass loss with incretin therapies vs lifestyle interventions

BJGP: Why don't we use clinical decision tools

www.nbmedical.com/podcast

Transcript

Intro

Neal Tucker 0:12

Sat at the computer, watching problems roll in, thinking where do I put them? Click, click on the screen. Did we say yes to this? I don't think that we did. Is it what I imagined? Being a GP ment. When I was a young Doc. I wanted to be a GP. Now I just triage total strangers. I just click, click, click on the screen. No time to see my patients. No continuity. No human contact. I just click, click, click on the screen. Click, click, click, click, click. Click, click, click, click, click. Click, click, click, click, click. Click, click, click, click. Keep your list open. Even if you feel broken, if it's overwhelming you, it's what you gotta do, is there empathy, for words on the screen. If we can't see what does it all mean? When I was a young Doc. I want ed to be a GP. Now I just triage total strangers. I just click, click, click on the screen. No time to see my patients, no continuity. No human contact. I just click click click on the screen. Click, click, click, click. Click, click, click, click. Click click click click. Click click ..... Click. It's Friday, the 17th of April, and this is the Hot Topics Podcast. Welcome to the Hot Topics Podcast from NB Medical with me, your host, as usual, Neil Tucker. I hope you had a good Easter. I hope you've been enjoying some of the nicer weather, and I hope you're ready for a bit more bang up to the minute medical research relevant to us in general practice. As usual, we've got three different pieces of research to have a look at today in the New England Journal of Medicine, a paper on bleeding risk with different DOACs for the treatment of acute venous thromboembolism. What do you choose for your patients? If you've got to give them some interim anticoagulant before they get seen in the DVT clinic, are you giving them the safest option?

Neal Tucker 3:02

Second, we've got a paper that's just published in a journal that I don't think we've had on the podcast before: Diabetes, Obesity and Metabolism Journal. This one caught my eye because it's a really topical subject. Lean body mass changes within critine therapies. So RGLP1s and dual therapy agonists, yes, you're losing fat, but how much of the good stuff are you losing as well?

NB Medical News

Neal Tucker 3:28

And finally, a paper in the BJGP, and I, like many of us, have been thinking a lot about AI recently. And this paper touches on where the future might be going, but how we might react to it in general practice. In fact, this isn't about AI, it's about something much simpler clinical decision support tools for early cancer detection, asking the question, why aren't they used? So, before we start off with the research, quick bit of NB News. So our spring hot topics course has been running for the last month or so now. Hope you've joined us for that. But if you haven't, our next live webinar is on Saturday, the 25th of April. So next weekend, join us for that. If you haven't been on the course in the past year, loads more courses coming up as well. Live webinars, do check out the nbmedical.com website for those.

What interim anti-coagulation would you give your patient with suspected VTE?

Neal Tucker 4:17

Also, if you haven't subscribed to NB Plus, now's the perfect opportunity. We've got our spring offer, which saves you £25 off an annual subscription, just £320 for year. You can come on all of our live online courses, more than a dozen of them now. Should you wish to get that much CPD in one year? Access to all of our online books as well, and much, much more. It is an absolute bargain. So do check out the website for that as well. Okay, let's get on with the research, shall we? And we'll kick off with this paper in the New England Journal of Medicine. Picture a scenario where you have a patient who comes in and you're pretty convinced it's a DVT. It's also Friday afternoon, it's too late to get them to the DVT clinic, maybe the clinic doesn't work on the weekends, and your patient's gonna have to survive until Monday, and you're gonna have to give them some anticoagulation.

Neal Tucker 5:13

The good news is, of course, as we all know, things got a lot easier when the DOACs came along. No longer do we have to scrabble around trying to find low molecular weight heparin, teach the patient how to administer it themselves, or heaven forbid, try and get the district nurses last minute to go in for the whole weekend to administer it themselves. All of a sudden, we could just do a simple prescription, give your patient a few tablets, pop over a referral form, and 'jobs a good un'. But which DOAC would you choose now? In my local guidelines now, I can prescribe either Ap ixaban or Riva roxaban. They're both options. I don't know why they give us two options. Why not just make it simple and give us one?

Neal Tucker 6:01

Well, perhaps the reason is it goes back to the research when they were looking to see whether we could use DOACs in this context, they had separate trials or Apixaban compared to vitamin K antagonists, separate trials for Rivaroxaban comparing to vitamin K antagonists, never comparing the two of them. The data showed they weren't inferior in terms of efficacy, and in terms of bleeding risk, Apixaban, 4% of patients had a significant bleed compared with almost 10% who had vitamin K antagonists, and for Rivaroxaban, it was higher at 8% compared to 8% on their vitamin K agonists. So we've got a slightly different figure here, and on the face of it, Apixaban looks like it's coming out on top. However, there was heterogeneity between the trials, different patient populations, differences in the trial design, and of course, that leads to the debate about whether one is genuinely better than another, and they concluded that no one could be certain whether there was a genuine difference in risk between Apixaban and Rivaroxaban. So that is what this randomized controlled trial is trying to address.

Neal Tucker 7:15

So the title is Bleeding Risk with Apixaban versus Rivaroxaban. Now, just as an aside, I don't want to derail this podcast too much, I've changed the way over the years about how I approach this. And currently, what I tend to do is I read through all the research, find what I think is useful and important, I go through the studies in detail, and then I make notes. And recently I've been using I use Word and I use the dictation tool on Word. This is like my I haven't used digital scribes in general practice yet, but this is like a little window into that world. Possibly they're better than this, they've got to be geared towards medical terms, right? But what does Word's dictation software come up with? When I say Rivaroxaban, it says river rocks abandon. So, as I'm reading through my kind of like dictated notes of what I think's useful to talk about, I've got to do some kind of like on the fly translation as well. So if I suddenly start going off piste, like you know, one of those news presenters who's teleprompted to say something really, really peculiar, you'll know exactly why.

Neal Tucker 8:34

Anyway, back to the title The bleeding risk is Apixaban versus river rocks abandoned, Rivaroxaban and acute venous thromboembolism. So it does what it says on the tin. They recruited patients with acute symptomatic PE or proximal deep vein thrombosis, and they randomised them to one of those two options for three months. As I'm sure you'll be familiar with, initially you get a higher treatment dose for a short period of time, and then that reduces down to a maintenance dose. For Apixaban, it's 10 milligrams twice a day for seven days, then five milligrams twice daily thereafter. For River Rox of Van, so slightly different this time, Rivaroxaban is 15 milligrams twice a day for 21 days. So that's an important discrepancy to keep in mind right there, the duration that you're on that higher treatment dose. In defense of the discrepancy in this research, that is the licensing for Rivaroxaban. It's what you'll find in the BNF, it's also what I find in my local guidance as well. So 21 days at that higher dose, followed by 20 milligrams once daily thereafter. The primary outcome they were examining was clinically relevant bleeding, which was a composite of major bleeding or clinically relevant non-major bleeding, and then they also looked at secondary outcomes such as death from any cause.

Neal Tucker 10:04

This was a pretty large study, 3,000 patients or near enough underwent randomisation, and they demonstrated a big difference between the two groups. So the primary outcome event, so that composite of major and clinically significant non-major bleeding occurred in 3.3% of patients on Apixaban, but 7.1% on river rox are banned, Rivaroxaban, so slightly less than half the relative risk if you're taking Apixaban, which is a statistically significant result, perhaps unsurprisingly. If you just considered major bleeding, then that occurred in 0.4% of the Apixaban group and 2.4% of the river rocks a bang group. How is it different every single time? It's like it's playing with me. Anyway, no one died because of bleeding, no one died because of recurrent venous thromboembolism either. As the linked editorial points out, the major point of discussion here is between the duration of those loading dose phases for each of these drugs, Rivaroxaban, this time spelt correctly by the dictation, by the way, is three times longer than Apixaban. And that does seem to have a significant effect on the rate of major bleeds.

Neal Tucker 11:28

Most of these are happening within the first 21 days rather than later throughout the three months of follow-up when they're on the maintenance dose. What doesn't appear to be happening as a consequence of reducing the loading phase to just seven days for Apixaban is a higher rate of recurrent venous thromboembolism. You seem to be pretty safe from that, although it is worth noting that this trial wasn't specifically powered to be able to definitively prove a difference here because that is obviously a relatively rare event, and you're going to need a much larger study population, which might make it quite impractical doing randomised controlled trials in that sense. Nevertheless, there was nothing in the original research on Apixaban, which was a larger trial, to suggest an increase in the rate of VTE. Indeed, the rate was about 15% lower than in the conventional therapy group when they were initially examining whether Apixaban could be used for acute treatment of VTE.

How much muscle mass do people lose with weight loss on GLP1s?

Neal Tucker 12:27

Now, of course, no study is perfect. The editorial points out just a few things to be aware of. So this was a predominantly white population, 90% of the study group were white. They didn't look at patients with cancer-related thrombosis or if they had significant renal or hepatic dysfunction or a body weight of over 120 kilos. Nevertheless, these findings I think are compelling, with the paper concluding amongst patients with acute venous thromboembolism, the risk of clinically relevant bleeding was significantly lower with Apixaban than with river rock suban. It's even underlined suban, saying that's not a word. No kidding. So Rivaroxaban during the three-month treatment. What does this mean for your and my patients? Well, I've got two options on my local guideline. This data's definitely pushing me towards using Apixaban for the small payoff of needing to take this medication twice a day in the longer term. And that is something that does require some consideration for some of our patients. But for that little payoff of using it a medicine twice a day, we can more than halve their risk of having a significant bleed in the first few days of treatment. Right. On to talking about GLP1s. We haven't done that since at least the last podcast, so probably better squeeze some more research in on these medications. What's interesting about this research is that for the past few years, most of it's concentrated on showing what these medicines can do. Weight loss, diabetes management, CVD protection. We'll now be prescribing them in England for exactly that after cardiovascular events if they're overweight. Liver disease, renal disease, PCOS, you name it, but what we haven't covered much and what you don't get as much in the literature, this is a common problem with evidence-based medicine, is you don't necessarily in the early stages of using medications get much on the potential downsides.

Neal Tucker 14:33

We touch on one aspect of that in the current hot topics course. What else are patients losing other than fat when they're rapidly losing weight? In a recent BMJ editorial noted that up to 40% of weight loss may be lean muscle mass, but it didn't go into a lot more detail. This, of course, is rather worrying. If you're losing muscle, is that going to be a problem for you? Are we driving sarcopenia? If you lose a significant amount of weight, of course, perhaps you don't need as much muscle, but it still sounds disproportionately high, right? And we have the data that shows most people regain the weight if they stop the jabs, which will mostly be fat, not muscle. And if people are buying this privately, as lots of people are at the moment, if they're doing the whole start-stop thing where they get to their target weight and then stop the meds because they're expensive and they put the weight back on and they do them again, you get this yo-yoing weight, which could then lead to multiple opportunities to lose more muscle mass and never regain it, which overall will make things worse for them.

Neal Tucker 15:42

So this is a really important area then. This new paper titled Lean Mass Change with Incretin Therapy versus Lifestyle Intervention, a systematic review and meta-analysis of randomised controlled trials, published in Diabetes, Obesity and Metabolism Journal. It explores this area in more detail. Although, as the title suggests, this is not just looking at GLP1 induced weight loss and the effects on lean muscle mass, which are of note, I should say, is not just about muscle, it includes your bones and organs, but hopefully you don't see those shrinking too much, but also looking beyond GLP1s to ask, does the same thing happen if people use intensive lifestyle interventions? So diet and exercise? Because if it does, then we can perhaps switch the focus from saying the jabs are bad because of this to more, well, what can we do about it in general?

Neal Tucker 16:41

The authors set the scene. So with decreasing lean body mass, you get decreased basal metabolism, so you can't eat as much, or you'll put weight back on, reduce strength and reduce functional capacity. It risks the development of sarcopenia, especially in older people. These, of course, are the ones who we may be more likely to be prescribing GLP1s for now, with large expansions of their use for people with diabetes and existing cardiovascular disease. And that could lead to sarcopenic obesity. So this is linked with poorer metabolic and poorer functional outcomes than either sarcopenia or obesity alone.

Neal Tucker 17:23

How bad then are incretine therapies for this and are lifestyle interventions any better? Inclusion criteria for studies into this systematic review, then. So number one, it needed to be a randomised controlled trial. Number two, it had to be an adult that was overweight, so BMI of 27 plus one or more weight-related comorbidities or a BMI of over 30. Number three, it needed to use an incretine or an intensive lifestyle-based intervention, and that needed to be compared to placebo, usual care, or an active comparator. Number four, they had to have body composition assessed with DEXA or MRI. Number five, it had to be at least 12 weeks' follow-up. And number six, yeah, there's a lot of there's a lot of entry criteria here. And it needed to produce enough data that they could actually use that and draw some conclusions upon it. 20 studies met all of these. 15 of them were incretin-based studies, five of them lifestyle studies, 15,000 plus patients included in total.

Neal Tucker 18:25

For incretin-based therapies, they showed that lean mass accounted for 25 to 39%, depending on which studies you were looking at, of the total weight lost. That sounds like quite a lot to me. You're thinking you're losing fat, but you're losing up to 40% of your lean mass when you're losing that weight. For lifestyle interventions, actually, it came out at 26%. So, generally speaking, kind of comparable. As usual, the devil is in the detail. You typically see much greater weight loss on the incretine therapies rather than the lifestyle therapies, at least in the trials that they've included here. The absolute lean mass loss rather than simply the relative lean mass loss, which is what those percentages were just now, the absolute lean mass loss is much higher with incretine therapies than with lifestyle measures, about one and a half times more, which is significant.

Neal Tucker 19:29

I don't know if that was statistically significant, but it seems clinically significant. As the paper points out, in one of the surmount trials, they showed the relative lean mass percentage actually went up. But on average, people had lost seven kilograms of lean mass. Seven kilos of muscle, that sounds like a lot to me. One really important observation in this study was that when resistance training, e.g., going to the gym and doing some weights, was incorporated into the weight loss program, this led to much more favorable lean mass preservation. So that's true for both lifestyle approaches and one study that used Liraglutide. The license for incretin meds does, of course, suggest that people need to increase the levels of exercise they do. But this specifically needs to be resistance training, not just cardio and stretching. That isn't sufficient. You need to do the resistance training if you want to benefit that muscle mass.

Why don't we use clinical decision support tools for cancer?

Neal Tucker 20:32

So the authors conclude that lean mass lost during significant weight reduction is substantial, and that the proportion of weight lost as lean mass is broadly comparable between incretin-based pharmacotherapies and lifestyle interventions. They suggest that muscle mass can be significantly preserved by integrating resistance training, adequate protein intake, and body composition monitoring into weight loss treatment programs. My big worry with this is that most of our patients who are on these medications are, of course, receiving them privately because they have to be. They're not going to be able to get them on the NHS, not for years and years and years. But I worry that they're not getting adequate counseling about this issue of muscle mass loss and the measures that they need to or can do to try and mitigate that. That's perhaps something that when we're having a chat with our patients, maybe we can just give them some clearer information about. Okay, let's look at the BJGP. This is our final paper titled Why Aren't They Used? Systematic Review of Barriers to Implementation of Clinical Decision Support Systems for Early Cancer Detection in Primary Care. Now I know what you're thinking talking about clinical decision support tools on the podcast, it's not gonna set the world on fire, right? But I thought this was particularly interesting because we're in this phase where we're seeing big push for big data in the NHS, particularly with artificial intelligence companies. Don't even get me started on that. They've got these big contracts to work within the NHS. The future is algorithm-based, and you can see how it could be used to identify patients at high risk of various medical problems, and this could link to an investigation pathway, for example.

Neal Tucker 22:33

But AI isn't magic, it's just algorithms, and for years we've had clinical decision support tools, mostly which are just algorithms, to help us. For me, what's really interesting is why we've chosen to accept certain of these support tools but not others. For example, we all seem to have a fundamental belief in Q risk for cardiovascular disease. This is very reasonable because the data shows close correlation between predicted and real-world outcomes. And off the back of this, we've then prescribed millions of slightly higher risk patients who have never had any cardiovascular problems, medication to try and reduce their cardiovascular risk.

Neal Tucker 23:19

But the Q Research team have produced multiple other tools like the Q-Cancer tool. It works out an individual's 10-year risk of having various types of cancer, yet it never caught on. How many of us are actually using this or one of the various other cancel tools that are embedded in our computer systems? I don't think there's widespread use or even interest in using them. You can't fault the maths behind them. The Q Research, the Q-Cancer tool is as well researched as the Q-Risk tool for cardiovascular disease, but something is stopping us from getting behind them.

Neal Tucker 23:57

This paper then is a systematic review of published research looking at the barriers to implementation. The authors found 29 cancer-specific C DSSs, 15 focusing on screening, 14 on symptomatic presentation, most of which were from the last 10 years, with the screening data mostly coming from the US and the symptomatic data from the UK and Australia.

Neal Tucker 24:23

The key barriers they found will not come as a surprise to most of us. Time constraints, lack of integration into systems and workflows, low usability, cognitive overload, fears around how it might affect patients, confusion about the benefits or even the purpose of these tools, lack of trust in the results, need for training, frustration caused by technology, and negative impact on the doctor-patient relationship. I think these will be familiar themes to many of us. I'm very busy during my normal working day just trying to make sure the patients I see are dealt with to the best of my ability, that the scripts in the bloods are done, that I've read and actioned my letters and so forth. Now you tell me I've got to learn a new tool with a new template. And if this happens, I have to click here, click there, click here again twice, and then write a letter to the admin team. Give me a break. We've got 20 different conditions with 20 different clinical support tools vying for our attention, 20 different pathways to learn. Couple that with the idea that if a computer says a patient has a 4% risk of cancer, I need to call the patient and explain that we might need to investigate them or refer them. They're stressing about it, even though 24 out of 25 of these people won't have anything wrong with them at all. It's stressing because these tools have flagged up a risk, but it doesn't meet any of the urgent cancer referral criteria. If they did, I would have referred them already. It's not straightforward.

Neal Tucker 25:56

Then Marjorie from admin gives me a printout of 40 patients that the computer's identified with a hemoglobin of 119, and I have to go through all of them to decide which ones normally just run at 119 and which ones are genuinely worrying. And then don't even get me started on pop-up fatigue. Every time you click on a patient, every time you try and issue a prescription, every time you put a code in. Honestly, I think it's got to the point where it's dangerous. Our computer systems may give us really useful information about cautions with a medicine that I'm about to prescribe in my patient with their specific comorbidities. Very useful information about interactions with other medicines. But it's often buried in so much other stuff that I just click through, I don't really look at it. I'm never going to see it. I've got pop-up fatigue. Our systems have made it simultaneously less likely for me to think for myself, whilst making it medico-legally indefensible to have prescribed, I don't know, nitrofurantoin in a patient with mild CPD or slightly wonky kidney function. What was the phrase they used? Cognitive overload. Is this just me?

Final Thoughts

Neal Tucker 27:08

Okay, rant over. So perhaps the key here is rather than forcing all these tools on us, which we never asked for, perhaps engage with clinicians better and ask them what they want, what they might find genuinely useful, and what we might actually use. Which, to be fair, is the conclusion of the paper. Phew, I think I had some kind of cathartic release there. I hope that hasn't stressed you out as you've been listening to this. I think we're done with the medicine for one day. Thanks for joining us on the Hot Topics podcast as ever. We'll be back in three or four weeks or so. In the meantime, don't forget to check out the nbmedical.com website. Don't forget to check out the NB Plus offer running for the next few days. Do join us on our upcoming live webinar courses, and remember you can always get in touch. Email hottopics @ nbmedical.com. Find us on Facebook and X and Instagram as well, I think these days. And remember to take some time for yourself as well.

Neal Tucker 28:17

I am off to the allotment this weekend. We've just got an allotment. Time to reduce my cognitive overload, I think. I hope you get to do something fun as well. Look after yourselves. Bye bye.