Dosing and Adverse-Effect Management With NOACs

Manesh Patel, MD: Chris, maybe after discharge and then thinking just about management of the agents and drugs that we have, what are things we can do? Discuss what people have been doing with dosing and adverse-effect management. One of the concerns—what I often say is the first bar—was getting people to use the drugs. The second bar is getting them to use the drugs correctly. Unfortunately, there’s some risk we’re seeing there in underdosing or concerns about how we should be dosing. Did you want to walk us through that?

Bernard J. Gersh, MBChb, DPhil, MACC: I might as well, to make a point: the exact same issue plagues heart failure medicine.

Manesh Patel, MD: Absolutely.

Bernard J. Gersh, MBChb, DPhil, MACC: You’ve got a wealth of good trial data, but the problem is that we don’t look at the doses that were given in the trials that were effective. I think 1 of the recent registries pointed out that if you looked at the 3 effective guideline-based drugs that have been shown to be effective in terms of hard end points in heart failure, the likelihood that someone would be on the same doses for all 3 of the drugs was something like under 5%.

Manesh Patel, MD: I think you’re right. It’s not new—obviously with lipid therapy, statins, blood pressure medications, heart failure medications. We can all see that maybe the dosing is not ideal. What’s going on in the NOAC [new oral anticoagulant] world with this?

Christopher Granger, MD: Manesh, let me make 1 additional point about stopping, because this is so important. The single most common site of bleeding is gastrointestinal [GI] bleeding. And it’s an issue for the NOACs because, on average, they do have a little more GI bleeding even than Warfarin. We have 2 issues. One is when we use a NOAC, which we should be doing, how do we reduce the risk of GI bleeding? The most obvious way is to avoid aspirin, avoid nonsteroidals. If somebody is at high risk for GI bleeding, be aware of that and use a proton pump inhibitor. We have some data now that it will reduce the risk of GI bleeding.

If somebody does have GI bleeding, restart the anticoagulant for a couple of weeks on average. That all too often doesn’t happen, especially if you treat the cause, as if it’s an ulcer and you treat the ulcer.

Bernard J. Gersh, MBChb, DPhil, MACC: The interventional field may be 1 of the reasons bleeding is such an adverse risk factor for prognosis when people bleed. They get taken off their drugs and never put back on them again.

Manesh Patel, MD: This is a great point that we focus so much about bleeding and agents when you can do a lot of things for your patients to reduce their risk. I strongly believe there’s a lot of patient risk. You know these studies were different populations; there are patients who are at high risk. What happens in practice? What’s going on in practice with dosing?

Christopher Granger, MD: In practice, all too often what we know is that after GI bleeding, patients don’t get restarted for months. On average, we don’t have exact empirical evidence on how to do this. But on average, a couple of weeks after even a significant GI bleeding episode, anticoagulation should be resumed.

Back to your question, Manesh, about dose. This is especially important, for example, for apixaban, but it’s important for rivaroxaban as well. Dabigatran is less of an issue because in the US the lower dose wasn’t even approved, but now it is available. Nonetheless, the problem is when we look at how the drugs are being used, there’s a high proportion. Somewhere between 10% and 40% of the time, patients are on either too low or too high a dose compared with how they were used in the clinical trials to be shown safe and effective. The biggest issue is underdosing. And the biggest issue with that is with apixaban where the 2.5 mg dose is used in the US probably 20% or 30% of the time, when it should be used 5% in the trial.

The key there for our providers to understand is that the bleeding advantage compared with Warfarin, with the NOACs, including for intracranial hemorrhage, is at the doses they were described. For rivaroxaban, if your creatinine clearance is less than 50 mL/min, use 15 mg—not 20 mg—once a day. For apixaban, if you don’t have 2 of 3 criteria—age greater than 80 years, creatinine greater than or equal to 1.5 mL/s, weight less than or equal to 60 kg—then use the 5-mg dose. If you’re not doing that—and we learned about this from the ENGAGE trial. The lower dose, the 30-mg initial dose for the population, had about a 40% increased risk of ischemic stroke because that dose was never approved. But it shows that if you underdose, you will have preventable ischemic stroke. Don’t underdose these drugs.

Manesh Patel, MD: Especially because I think the trigger for underdosing them is patient comorbidity and a belief about their bleeding risk that you think you’re going to affect. Sean, this is 1 of the things we’ve seen in observational data in Medicare and other places where we’ve seen this, including in several publications.

Sean D. Pokorney, MD: Yeah. There’s very good Clinical registry data looking at this issue. What you find is that when providers use a lower dose, they’re using the lower dose to try to decrease the risk of bleeding but hopefully still provide stroke prevention. What you actually find when you look at the data is that patients who are underdosed have the same rates of bleeding as patients who are appropriately dosed. By decreasing the dose because of a bleeding event, you don’t actually decrease future bleeding events. Their risk of bleeding remains the same, low; it’s a patient risk. What you see is higher rates of stroke in the patients you’re undertreating. So you’re really not accomplishing what you’re trying to accomplish at all, and it reemphasizes how critical it is to use the appropriate dose. If you use the lower dose, you’re going to still have bleeding because it’s a patient risk factor, and you’re going to have higher rates.

Manesh Patel, MD: Very few patients with high bleeding risk don’t have a high stroke rate. Their risk factors seem to go together.

Transcript edited for clarity.