Anticoagulants: The Definitive Guide
A professor of mine used to say that evolution hasn't caught up to our current lifestyles.
What he meant was that our blood evolved in a setting where it needed to clot. At any moment, we could be hit with a spear or mauled by some pre historic beast such as a Mega-Tiger.
Our blood needed to react to trauma.
In order to keep us alive, it had to clot (and clot well).
In today's world, we're much less likely to get mauled by a Mega-Tiger. In fact, in the latest reports by the WHO, I can't find any reported incidents of death by Mega-Tiger (it looks like our vigilance has paid off...for now). But now we're seeing a shift to the opposite end of the spectrum. Now our blood clots so well that it can clot when we don't need it to.
This is problematic because it cuts off the flow of blood (read: oxygen) to some of our important inside parts like the brain, heart, lungs, and other internal organs.
In order to keep us alive when this happens (and to prevent it from happening in the first place), we've come up with some nifty drugs called anticoagulants. These all inhibit some part of blood clot formation, and keep our blood thin and flowing.
Anticoagulation is an area of high impact for pharmacists.
And so that means we've gotta spend some time learning about it.
By the time you account for renal function, pharmacokinetics, and drug interactions, you realize that anticoagulants are kind of a double-edged sword.
They save lives by preventing and treating strokes, pulmonary embolisms, and DVTs…
But if they're not dosed correctly, anticoagulants can cause the minor inconvenience of massive internal hemorrhaging. Go just a little too far with the dosing and anticoagulants can kill. It turns out that blood pooling around the brain can cut off the oxygen supply just as well as a blood clot can.
In this guide, I'll break down the pharmacology of all things anticoagulation. I'll also give you clinical pearls (i.e. test questions) that will help you pass the NAPLEX and your next therapeutics exam. For the most part, I'm going to avoid getting into the specifics of dosing. There are so many subtle dosing differences depending on the indication that this article will get unwieldy in a hurry if I try to include them all.
However, for your convenience, we have created an incredibly handy cheat sheet that covers the common doses (and a whole lot more) for anticoagulants and antiplatelets. If you've grabbed any of our cheat sheets before, you already know how much info we cram into them. Our anticoagulant cheat sheet is easily the most detailed one yet.
Anyway, moving on...let's get started with our “road map” for the lesson…
Not confusing at all, right? I'll be referring back to this behemoth throughout the guide. So get good and comfy with it. Be friends with it. Get to know it. See if you two have anything in common.
You can plug in visually where each anticoagulant fits in the clotting cascade. That covers most of the pharmacology right there (Welp...my job here is done...).
Background and Terminology
Normally, your blood flows all loosey-goosey through your veins and arteries. It goes from the heart, to the lungs, back to the heart, and then it's pumped all over the body before returning to the heart to start the whole trip over again.
But sometimes a clot can happen and muck everything up.
In medicine, a clot is called a "thrombus." And the resulting block in blood flow is called a "thrombosis."
Sometimes a stationary clot can break free and travel through the circulatory system. At this point, it's called an "embolus" (and it results in an "embolism."). Usually, an embolism travels until it gets stuck in a smaller blood vessel where it proceeds to take up permanent residence.
The result of a thrombus or an embolus is bad:
- Myocardial infarction (MI) (i.e. heart attack)
- Pulmonary embolism (PE)
- Deep vein thrombosis (DVT)
- And still others...
Collectively, these disorders are the most common cause of death in most developed countries.
So what actually makes up a clot?
If you refer to that monstrosity of a diagram we've made above, you'll see that a clot is made up of a lot of things in your blood. But there are two things that are the most important in terms of clot stability and structural integrity:
Fibrin and platelets basically stop a clot from falling apart once it has formed (and they have a role in actually forming the clot in the first place).
Platelets notice a hole in the blood vessel, and they plug it up. They first adhere to the surface (adhesion), and then they send off a chemical signal flare to tell all their friends to join them (activation). Once their friends arrive, they all join hands in an impressive showing of unity (aggregation). All of these platelets joined together effectively form a nice plug that stops the bleeding.
At some point during the platelet aggregation process, Fibrin is created via the clotting cascade (again, see the above diagram). Fibrin lays a cross-layered sheath over the clot which basically stops it from falling apart and keeps it structurally safe and sound.
There is, of course, some overlap between the processes of platelet aggregation and the creation of the fibrin sheath. And both processes are important whether the clot is being made in an artery or a vein.
You could make an argument that fibrin is more important to clots formed in a vein (red clots). These are the DVTs and PEs that you know and love. Likewise, you might say that platelets are more important to clots formed in an artery (white clots). These are the strokes and MIs of the clotting world.
Again, I'm not saying they are singularly important. Fibrin and platelets are both important whether a clot is arterial or venous. But the medical management of clots is different depending on where the clot formed. We tend to use antiplatelet medications for strokes and MIs, but we use anticoagulants for DVTs and PEs. That's because strokes and MIs are usually arterial clots, while DVTs and PEs are usually venous clots.
Alright, now that we're good and comfy with how clots form, let's move on to fixing those clots. To do this, we'll use anticoagulants and antiplatelet medications.
This article will focus on anticoagulants. We'll cover antiplatelet medications in a future article.
A Quick Note on Anticoagulants versus Thrombolytics
There's an important bit of trivia that I need to mention before we dig in any further.
None of the anticoagulants we're about to review actually break up clots.
This seems counter intuitive, doesn't it? Why are we giving heparin or rivaroxaban to a patient with a DVT if it doesn't actually break up the clot?
Look again at that monster clotting diagram above. Notice that it's only moving in one direction. Every step of it proceeds to the final process of cross-linking fibrin. There is no reverse.
All of the anticoagulants that we'll cover in this article stop one or more steps of the above diagram. So they stop the clotting process from moving forward, but they don't do anything to reverse it. Instead, they stop an existing clot from growing or spreading. And they stop new clots from forming (having an existing clot is a major risk factor for developing a secondary clot).
So how do you get rid of a clot?
Your body makes something called plasmin. This is your own natural clot buster (aka: thrombolytic). While anticoagulants prevent new clots and stabilize existing ones, your body converts plasminogen into plasmin (via an enzyme named plasminogen activator [PA]) and gets to work actually destroying the clot.
Do we have a drug that does that too?
Absolutely. We've got several thrombolytics, in fact. The most commonly used is recombinant tissue plasminogen activator (tPA). This is also known as alteplase or by it's brand name, Activase. Just like the name implies, it's a synthetic version of the enzyme that activates your body's own plasminogen into plasmin. So if you give tPA, your body will make more plasmin, and existing clots will go bye-bye.
Some other versions (besides alteplase) of tPA that you may come across are reteplase and tenecteplase. They are pretty similar, but there are key differences in FDA approved indications between the 3 formulations of tPA.
- Alteplase [Activase] - Indicated for Acute Ischemic Stroke, Acute Myocardial Infarction (AMI), and the lysis of Acute Massive Pulmonary Embolism
- Reteplase [Retavase] - Indicated for AMI
- Tenecteplase [TNKase] - Indicated for AMI
It's unlikely you'll see anything else used to dissolve clots (at least in the US), but for the sake of completeness, streptokinase and urokinase also exist. To the best of my knowledge, urokinase isn't even available in the US anymore. Streptokinase is still available, but it's side effect profile make it much less preferred compared to any formulation of tPA (streptokinase is derived from streptococci bacteria, and so is antigenic in many patients). Urokinase is non-antigenic, but also considerably more expensive than streptokinase.
And additionally, the clot-busting power of streptokinase and urokinase are much less predictable than tPA. So suffice it to say, you're probably going to be dealing primarily with tPA in your practice career (and it's probably going to be alteplase).
Here's a super handy chart (taken from this super handy book chapter) that highlights the critical differences between the various thrombolytics available in the US:
Where do we use thrombolytics clinically?
Most commonly, they are used in tiny doses to clear clogged up IV lines and ports. There's a handy brand of alteplase called [Cathflo] that comes in a 2 mg vial and is quite useful at clearing up clogged lines.
Otherwise, thrombolytics (I'm speaking specifically to alteplase now) are primarily used in ischemic stroke and acute MI. You can also see alteplase used for PE (but generally, it has to be a pretty massive PE for the risk/benefit ratio of using alteplase to payoff). Occasionally, you may see alteplase used during vascular procedures or as an (off label) adjunctive treatment for frost bite in cold parts of the world.
It's critical to differentiate an ischemic versus a hemorrhagic stroke before giving alteplase. Thrombolytics work very well at destroying clots and there is a very significant bleed risk associated with their use.
In the setting of stroke, an MRI (preferred) or a CT of the head must be performed to rule out hemorrhagic stroke before giving alteplase. There is also a very long checklist of bleeding risk factors that must be evaluated beforehand.
You can see an example of one such checklist here, but in general you're looking for any history of clinically significant bleeding or recent use of anticoagulants. There's also a time factor involved. If the onset of symptoms was greater than 3 hours ago (or 4.5 hours in some special cases), the risk/benefit ratio of using alteplase isn't worth it. The clinical benefit decreases as time progresses; but the risk of bleeding remains the same.
If you ever work in a certified stroke center, you'll see what an organized effort it takes to get a patient to the hospital, with an MRI or CT to rule out hemorrhagic stroke, with the tPA checklist completed, and to administer the drug within the 3 hour window.
Risk Factors for Developing a Blood Clot
We're almost ready to get into the drug part, I promise! But first, we should briefly mention what sort of patients develop clots in the first place. That way we can be on the lookout ahead of time. In some high risk cases (which we'll discuss below), we may elect to prophylactically give an anticoagulant.
Anyway, the following is not an exhaustive list. But here are some common risk factors associated with developing a thrombosis:
- Inherited disorders such as Antiphospholipid Syndrome (APS) or Factor V Leiden
- Certain cardiac arrhythmias such as atrial fibrillation
- Presence of a mechanical heart valve
- Prolonged bed rest/immobility or paralysis
- Oral contraceptive pills
- Traumatic blood vessel injury or surgery
- Having an existing clot (or a history of clotting)
It should be noted, also, that these risk factors can be synergistic. If your patient is an overweight female who smokes and takes an oral contraceptive, the relative risk of developing a DVT is higher.
Prophylaxis versus Treatment
I hinted above that we occasionally give anticoagulants preemptively to high risk patients to prevent a clot from developing in the first place. What sort of patients do we do that with? Who actually needs prophylactic anticoagulation?
It's pretty much impossible for me to give you a clear answer. Like most other things in medicine, it's a gray area with wiggle room for clinical judgement.
There's also room for common sense here. If you have a cancer patient with a history of DVTs that presents to the ER with a GI bleed, prophylactic anticoagulation probably shouldn't be on your immediate "to do" list.
In some areas, it's common to perform a thromboprophylaxis risk assessment to help guide the decision on prophylaxis or not. Here's a helpful sample calculator to give you an idea of what criteria goes into the decision.
Another point to mention is that the doses we use for prophylaxis are usually (but not always) much lower than the doses we use for treatment. This changes the need for monitoring (we almost never monitor prophylactic dosing).
In some of the cases above, patients will get full therapeutic dosing of anticoagulants as prophylaxis. Often times, this is used as secondary prevention after a clot has developed. But not always.
As an example, cancer patients with any history of DVT or PE usually require lifelong prophylaxis with therapeutic enoxaparin. Another example is any patient on warfarin therapy (there isn't really a "prophylactic" dose of warfarin). So here you'll see patients with a-fib, APS, mechanical heart valves, and the like.
So just think of prophylaxis on a case by case basis. Assess the risk factors for clotting, as well as the bleed risk for the patient.
Anticoagulant Drugs (Finally)
As a quick disclaimer, before moving on to the reason you're reading this article. All of the following medications have a potential side effect of bleeding. I’m not going to repeat that each and every time.
They're anticoagulants. It's just what they do. They don't wanna hurt nobody but a bee has got to sting.
I'm going to cover the most important pharmacology and therapeutic info below. But again, I'll stop just shy of providing dosing (which in some cases is going to be institution-specific anyway).
If you find yourself wanting some guidance on dosing info, check out our Anticoagulant/Antiplatelet Cheat Sheet. It's loaded with common doses, dosing adjustments, clinical pearls, and a whole lot more.
Unfractionated Heparin (UFH) is one of the oldest (and most widely used) anticoagulants. It binds to and activates antithrombin III.
So what is antithrombin III? It’s one of your body’s natural anticoagulants. It floats around your blood and inhibits thrombin (Factor IIa) and Factor Xa from propagating the coagulation cascade. That's pretty easy to remember, right? Antithrombin inhibits thrombin.
So in essence, heparin binds to antithrombin III (AT-III) and makes it work more efficiently. It's like a task master that pushes AT-III productivity to new heights.
I sort of envision it like this:
Heparin primarily impacts Factors IIa and Xa (via AT-III). And in a nutshell, that’s its mechanism of action.
So here's the thing with heparin. Its kinetics are horribly unpredictable. It seems to affect everyone a little differently, and the only way for us to get around this is by monitoring.
In my experience, most institutions monitor the activated Partial Thromboplastin Time (aPTT). But you may come across some that monitor the anti-Xa level. Whether aPTT or anti-Xa is more accurate is actually the subject of ongoing debate. It seems like you can safely use either.
However, at least at the time of this writing, aPTT is used more commonly. So use that as your launching point for test questions while you're in pharmacy school.
When used for treatment (usually dosed IV), heparin is very closely monitored. It is usually given as a continuous infusion and the aPTT (or anti-Xa) levels are monitored every 6 hours.
If it’s being used for prophylaxis (dosed subcutaneously), it doesn’t require monitoring. I should point out here that heparin does have a subcutaneous treatment dose listed in it's labeling (250 units/kg q12h). I have never seen this in practice, and I strongly suspect you won't run into either. But at least you know there's labeling for it now.
Practicing Pharmacists: If you have used the subcutaneous treatment dosing of heparin in your practice, please email me at email@example.com with any wisdom on patient population, clinical scenario, etc... I'll update this article with the knowledge you drop my way. Thanks!
Despite the erratic kinetic profile, heparin is still a drug of choice in a lot of situations (for treatment or prophylaxis). It has a very short half life, and heparin can be used in any stage of renal failure (including dialysis).
That's an important fact, because many of our other anticoagulants cannot be used in the later stages of renal failure.
Another benefit of heparin is that it has an antidote; protamine sulfate. 1 mg of protamine will neutralize 100 units of heparin (I'd remember that for the NAPLEX if I were you). Protamine has a max dose of 50 mg. It's also derived from fish, so it should be used with caution or avoided altogether in patients with a fish allergy.
Adverse Effects of Heparin
Heparin has a number of "weird" side effects that you wouldn't necessarily think about for an anticoagulant.
For starters, it can actually block the synthesis of aldosterone, which basically makes it an aldosterone antagonist. As you might suspect, this can lead to hyperkalemia.
Through an unknown mechanism, heparin also seems to increase osteoclast activity, and decrease osteoblast activity. So you can end up with osteoporosis after prolonged use.
Heparin is porcine in origin. So if your patient has a pork allergy (or even if they prefer to avoid pork products for cultural or religious reasons) then you need to look elsewhere. In the case of pork allergy, administering heparin can lead to anaphylaxis.
The final clinical pearl I’ll offer on heparin is a potential (but rare) side effect. It’s called Heparin Induced Thrombocytopenia (HIT).
As the name implies, HIT is a situation where your platelet count drops in response to heparin therapy. There are two types of HIT (Type I and Type II).
HIT Type I, for all intents and purposes, is the "better" one to have if you're the patient. It's also known as heparin "associated" thrombocytopenia. It's a transient, usually mild drop in platelet count that happens within the first couple of days of initiating heparin. Platelets rarely drop below 100k, and the effect is reversed (within 3 or 4 days) after stopping heparin. There's actually a group of clinicians proposing to change the name from HIT Type I to "non-immune heparin associated thrombocytopenia" to reduce the confusion between Type I and Type II.
HIT Type II is a much worse clinical picture. This time, there's an immune response associated with the drop in platelets. Effectively, heparin binds to your platelet and undergoes a conformational change that makes it immunogenic. Your immune system doesn't like this, so it goes to the scene to bust up the offending platelet like a bouncer at a night club. Your platelet counts will drop accordingly. But in a weird twist, your immune system actually activates the platelets before it clears them out. So, ironically, even though platelet counts are dropping, patients with HIT Type II are at risk for developing clots (pretty much any and everywhere in the body). This article does a nice job at summing up the pathophysiology.
To diagnose HIT, we often use a test informally called the 4 Ts of HIT. One of the primary things to look at are platelet count (specifically, a 50% reduction from baseline). This reduction in count should also coincide with heparin administration. So if the patients platelets were 300k on Monday, then heparin was started and the patients platelets were 150k on Wednesday, that's the beginning of a HIT diagnosis.
It's not the full HIT diagnosis of course. You also want to evaluate if there are any thromboses or any other reason for thrombocytopenia (maybe the patient just received chemotherapy?). Since HIT Type II is immune mediated, we would also run an antigen test to confirm the presence of heparin antibodies.
Obviously, you don't wait around for the results of HIT antigen testing to intervene clinically here. The patient needs anticoagulation (and is at additional risk now due to the platelet activation caused by HIT Type II). LMWH also has an association with HIT (albeit, much lower than unfractionated heparin), so you cannot transition the patient to LMWH.
In the acute setting, you'll usually transition the patient to either fondaparinux or argatroban (I'll have more on each of them below). Once you're gearing up for a transition to the outpatient world, you can use pretty much any oral anticoagulant on the market. Warfarin, or any of the NOACs are a fine choice here (though it is recommended to wait until the platelet count recovers before initiating).
If the HIT antigen test comes back negative, you could consider rechallenging with heparin (depending on your institution's protocol and everyone's comfort level with re-challenging). If the HIT antigen is positive, you can pretty much remove heparin and LMWH as possible anticoagulants for this patient in the future.
Low Molecular Weight Heparin (LMWH)
Next up, we have heparin’s little brother, LMWH. These all end in the suffix "-parin." Far and away, the most common LMWH heparin is enoxaparin [Lovenox]. But for the sake of completeness, there is also dalteparin and tinzaparin.
As the name implies, these are low molecular weight versions of heparin. They're literally a heparin molecule that’s been cut down to a fraction of the original size.
LMWH still binds to AT-III like heparin. But something weird happens with the smaller size of LMWH. Antithrombin III preferentially starts inhibiting only Factor Xa (and stops inhibiting Factor IIa).
To be clear, LMWH still slightly inhibits IIa. But it inhibits Xa a lot more.
So, Xa >>> IIa.
What does that mean clinically?
For starters, the kinetics are much more predictable. Unlike heparin, LMWH is pretty consistent with its anticoagulation effect. The upshot of this is that we don't routinely need to monitor levels (even with therapeutic dosing).
I italicized "routinely" above...that should clue you in that there are some exceptions to the "You don't need to monitor LMWH" rule. Namely, we might monitor in renal disease (like heparin, LMWH is renally cleared), extremes of body weight (because it's not extensively studied in very under or overweight patients), and pregnancy (because you just have to be careful in pregnancy). If someone experienced a treatment failure or a bleed while on therapeutic LMWH, you'd probably also want to monitor levels going forward.
To recap that (in bullet form!), here are some patient populations where you would consider monitoring LMWH levels:
- Renal disease
- Extremely under or overweight
- Prior bleeds or treatment failures
As for what to monitor with LMWH, you usually will monitor anti-Xa levels. This makes sense because remember that LMWH is much more specific to Xa than to IIa. I'd consider that the 'gold standard' of monitoring that should be your starting point on any test question.
That being said, you may come across literature or an occasional institution that uses the aPTT for LMWH. You may even have some institutions with a special LMWH assay. Just be aware that these exist, so you're not completely thrown off if you see them in practice.
And remember, for most patients, we can give therapeutic doses LMWH without monitoring. Those therapeutic doses are given as subcutaneous injections twice daily (and in some cases only once daily). Compare that to the IV drip of heparin with monitoring every 6 hours and LMWH looks pretty attractive in the right patient population.
Even better, enoxaparin syringes are pre-filled in a variety of strengths. Patients can self-administer therapeutic doses at home and don't need to stay in the hospital, providing another benefit of LMWH to heparin.
Adverse Effects of LMWH
We run into an issue with LMWH in renal disease. Remember from above that we can give heparin no matter what your kidney function is. Even in dialysis.
That's not true with LWMH.
In general, we start getting cautious (and monitoring) when LMWH is used in patients with CrCl < 30. And it’s absolutely contraindicated in dialysis patients. Unlike heparin, LWMH has a pretty long half life (that's why we can get away with dosing it once or twice a day). It will accumulate in renal disease (and predispose the patient to bleeding).
So, if your patient has bad (or even just unstable) renal function, I'd recommend heparin.
As for other side effects, remember that LMWH is like a little cousin to heparin. The potential side effects are similar, but the likelihood of those side effects is less frequent with LMWH.
LMWH does have a risk of hyperkalemia. According to this small trial with n=60, it's got the same risk as heparin. However, this older (and smaller) trial says there's no risk at all. When you run into contradictory situations like this, my recommendation is to go with a "guilty until proven innocent" policy. I'm not saying you should empirically put a LMWH patient on kayexalate or anything. Just that you should be mindful of potassium levels and watch out for synergistic drug interactions.
As for Osteoporosis...it's also debated whether there is an association with LMWH. Some think we're chasing a red herring just because LMWH is similar to heparin. The jury is still out, but again I'd consider it a possibility worthy of monitoring until proven otherwise.
LMWH is also associated with HIT, but to a much lesser degree than heparin. So it's much less likely. That being said, if you have a patient with confirmed HIT, you will not be switching them to LMWH.
LMWH is also contraindicated in patients with neuraxial anesthesia (i.e. an epidural). It can cause a spinal hematoma which can lead to paralysis. So, umm. Don't do that.
As for a reversal agent. There is conflicting data (are you noticing a trend here?). Some sources that say you can use protamine as a reversal agent for LMWH. The reversal is considered "incomplete," but if you have a patient with a serious bleed I think most clinicians would take that over nothing. Just keep in mind that not everyone is on board with the using protamine for LMWH train. It's not a clearly defined choice like it is with heparin.
The dose is 1 mg of protamine to neutralize 1 mg of LMWH. Just like with heparin, the max dose of protamine is still 50 mg (and don't forget about that fish allergy thing).
Last in the AT-III family of anticoagulants we have fondaparinux. You will often (incorrectly) see fondaparinux grouped with the direct factor Xa inhibitors (we'll cover these in a bit). That's because fondaparinux only inhibits factor Xa. But, fondaparinux works via AT-III...it doesn't bind to Xa at all (so it's not "direct"). So it belongs in a list with the other AT-III inhibitors.
I mentioned above that LMWH is just a trimmed up version of heparin. Think of fondaparinux as being trimmed up even further. It is literally the precise 5 pentasaccharide sequence that binds to AT-III (so you could say that it's binding is pretty specific).
Clinically, this means that it only inhibits factor Xa. There is no significant IIa inhibition here. I think this is why it often gets grouped with the direct factor Xa inhibitors.
In general, monitoring is not required for fondaparinux. The special patient populations in which you would monitor are less clearly defined than they are with LMWH. However, what you should monitor is pretty clear: anti-Xa.
Personally, I'd monitor fondaparinux in the same situations that I would for LMWH. So that's renal disease, extremes of body weight, pregnancy, and history of bleeds or treatment failures. The package insert even goes so far as to recommend monitoring in patients who are less than 50kg, as fondaparinux is associated with a higher risk of bleeding in these patients.
Adverse Effects of Fondaparinux
The neat thing with fondaparinux is that it's not really associated with those "weird" side effects from above like osteoporisis and hyperkalmia.
Additionally, (and this is important), it has no association with HIT.
Remember from above that fondaparinux is one of the two preferred options in the acute setting for patients who develop HIT.
In my experience, that's been the most frequent reason for using fondaparinux. So why don't we use it more?
Like most decisions in modern medicine, cost is certainly a factor. Fondaparinux is considerably more expensive than LMWH, and even more so than heparin.
Its other big strike is in renal failure. Fondaparinux has a solid 17 - 21 hour half life (even longer than LMWH). So its risk of accumulating (and causing a subsequent bleed) in renal failure is even more probable. In fact, fondaparinux is contraindicated with a CrCl of < 30ml/min.
So here's a quick summary our AT-III anticoagulants in renal failure:
- Heparin – Can use in HD
- LMWH – Contraindicated in HD…consider monitoring if CrCl < 30
- Fondaparinux – Contraindicated CrCl < 30
Making matters worse for renal accumulation, there is no specific reversal agent for fondaparinux. Protamine is not recommended. So you really have to use caution when selecting patients for fondaparinux.
That pretty much wraps up our AT-III anticoagulants. Let’s move on to some other agents.
Argatroban is a direct thrombin inhibitor (DTI). Refer back to the clotting cascade earlier from this post (there's also one included in our anticoagulant cheat sheet). You'll see that Thrombin (Factor IIa) is the thing that activates fibrinogen to fibrin (which is then the material that cross-links the clot). So you inhibit thrombin, you inhibit fibrin (and cross-linking). Make sense?
The earlier DTIs (hirudin, lepirudin, and bivalirudin) are used so rarely anymore that I'm not going cover them in this article. You can read a neat timeline on the history of DTIs here if you're interested.
You can probably surmise this from the name "Direct Thrombin Inhibitor," but argatroban directly binds to thrombin (Factor IIa) and inhibits it. Clever, no?
It is not used very often, but it does have its niche in therapy.
As previously mentioned, it is the second of the two "go to" drugs for patients that develop HIT. Far and away, this is the most common reason that argatroban is used.
Should you use argatroban or fondaparinux for HIT? It depends on why the patient was on heparin in the first place. Argatroban has a super short half life (about 45 min) compared to the 17 - 21 hour ordeal for fondaparinux. So if the patient is getting surgical procedures or has renal failure, you'd lean towards argatroban.
Obviously, a drug with a 45 minute half life is going to have the need for an IV infusion, and it's usually going to be titrated to therapeutic effect (which means monitoring). There are two options for monitoring argatroban: aPTT and the Activated Clotting Time (ACT).
Argatroban and bivalirudin are the only two drugs in somewhat regular use that are monitored with ACT that I'm aware of...so I'd at least be aware of that fact as a possible test question.
Outside of HIT, you'll usually see argatroban used during invasive surgical procedures, such as PCI. Again, that glorious 45 minute half life makes argatroban pretty useful for short-term anticoagulation during a procedure.
There is no reversal agent for argatroban, but this usually doesn't matter. Its half life is so small that it’s effect will wear off quickly on its own by just stopping the infusion.
It should also be noted that argatroban has to be used carefully (if at all) in patients with liver failure. Everything we've talked about up to this point has been renally cleared, so this is an exception worth remembering.
As a final clinical pearl, I offer you a common test (and NAPLEX) question. Argatroban interferes with the assay we use to monitor warfarin therapy (the INR). Specifically, it falsely elevates the INR by several points.
So if you were transitioning a patient from argatroban to warfarin (which is exactly what you’d do in a patient with HIT who needed long-term anticoagulation), you'd actually shoot for an initial target INR of > 4.0 before stopping the argatroban.
That seems like crazy talk, but remember, it's a falsely elevated INR. Argatroban's effect will wear off fairly soon after stopping the infusion, at which point you'd adjust the INR to the "normal" range for the patient's indication.
New Oral Anticoagulants (NOACs)
Next up, we've got a new classification of drugs called Novel Oral Anticoagulants (NOACs). NOACs are not tied to a specific mechanism. They are just “newer” than warfarin (which is the original oral anticoagulant...in fact, warfarin was the only oral anticoagulant for about 60 years).
Currently, our NOACS are either Direct Thrombin Inhibitors (dabigatran) or Direct Factor Xa Inhibitors (rivaroxaban, apixaban, edoxaban).
I'm going to discuss the NOACs all as one group (even though they have different mechanisms of action). The reason I'm doing this is because they're pretty much all approved for the same indications (with some minor variances). It'll be easier to talk about the pros and cons of using one over another in a single section instead of making you scroll back and forth.
You're welcome, by the way.
I'll also make sure to highlight some relevant clinical pearls for each of the NOACs individually.
The big "draw" of the NOACs is that none of them require the cumbersome monitoring that warfarin requires (although that does keep pharmacists employed). And although all of them have drug interactions via the P-gp pathway (and some have interactions with CYP as well), NOACs have considerably less drug interactions than warfarin. And they have basically no interactions with food.
On the down side, in many ways the frequent monitoring associated with warfarin can be considered a plus in many cases. Monitoring is the only way to be sure you're not over or under anticoagulating your patient. Additionally, in the case of a bleed, warfarin can be reversed with Vitamin K, fresh frozen plasma (FFP), or 4PCC (which is just a concentrated version of FFP that goes by the brand name Kcentra). As of right now, the only NOAC with a reversal agent is dabigatran (although a new drug called andexanet alfa is in clinical trials as a reversal agent for Xa inhibitors). And several NOACs carry a higher bleeding risk than warfarin in certain patients or for certain indications.
All of that is to say that while NOACs have a definite place in therapy, they do not yet spell the end of warfarin. Warfarin is cheaper, has a longer history of safety/efficacy, and is still the only oral anticoagulant approved for some indications (such as mechanical heart valves).
If you want more info, here is an excellent read that gives a more detailed comparison of the NOACs. If you're happy taking my word for it, I'm going to do my "tl;dr thing" with the NOACs.
Alright, so let's get into the nitty gritty of this. Again, when I say "NOAC" I'm referring to the following:
- Dabigatran [Pradaxa] - DTI
- Rivaroxaban [Xarelto] - Xa Inhibitor
- Apixaban [Eliquis] - Xa Inhibitor
- Edoxaban [Savaysa] - Xa Inhibitor
Notice how all of the Xa Inhibitors end in "-xaban." Think of this as "Xa Ban," and it'll make it easy for you to remember their pharmacology.
In general, NOACs are approved for the following:
- Stroke prevention in non-valvular afib
- DVT/PE treatment
I said "in general" because the above is a pretty big generalization. You may see a couple of them used for DVT prophylaxis in patients who just had hip or knee surgery. You can also see them used as a sort of secondary prophylaxis in patients who've had recurrent DVT/PEs. Rivaroxaban, for example, is approved for both of those indications.
Also notice that NOACs are specifically approved for non-valvular afib. If the patient has valvular disease and/or a mechanical or bioprosthetic heart valve, warfarin is your only PO choice.
And then, literally as I was writing this article, the FDA had to go all fancy on me and approve betrixaban [Bevyxxa]. So now we've got another Xa inhibitor in the mix. I'll probably write up a new FDA approval post to cover it in more detail (and I'll update this post later). But here's where we're at right now with betrixaban. It's approved for VTE prophylaxis in hospitalized patients at risk for developing VTE.
Its trials compared it to prophylactic enoxaparin at VTE prevention. The benefit of betrixaban is that it can be given orally. In terms of preventing VTE in patients that do not have a history of VTE, this is the only oral option we have (unless you believe in the silly orthopedic recommendation of aspirin 325 mg BID. Spoiler alert: I do not).
Betrixaban has the same P-gp interactions as other Xa inhibitors. And it's also got a renal adjustment (you halve the dose at CrCl < 30 ml/min and avoid use entirely if CrCl < 15 ml/min). It's also recommended to avoid use in patients with reduced liver function. Finally, it has not yet been studied in pregnant patients or in patients with mechanical heart valves (once again, your options there are warfarin, warfarin, and warfarin).
I suspect that the makers of betrixaban (Portola Pharmaceuticals) will be going after more indications (because right now it's only approved for prophylaxis, not treatment). Again, I'll dig in more with a new FDA approval post and will update this article (and our anticoagulant cheat sheet) as more info becomes available.
Alright, now that that's out of the way. Let's continue with our "painting NOACs with a broad brush" generalization...
How do you know which NOAC to use?
I'll assume you've already determined that the NOAC(s) you're considering can be used for the indication you're treating (again, it's usually DVT/PE or non-valvular afib).
From there, a good place to start is renal function. Check to see if the NOAC has a dose adjustment or contraindication based on your patient's renal function. The dose recommendations change by indication and renal function, so I won't list them out comprehensively here. Instead, I'll give you some general cutoffs:
- Dabigatran - Dose adjustment for afib at CrCl < 30 ml/min. Contraindicated for DVT/PE at CrCl < 30 ml/min. If patient is taking a strong P-gp inhibitor, the dose adjustments start at CrCl < 50 ml/min.
- Rivaroxaban - Dose adjustments start at CrCl < 50 ml/min
- Apixaban - No renal dose adjustment necessary, including in dialysis (although I've previously written about my skepticism with this recommendation)
- Edoxaban - Dose adjustments start at CrCl < 50 ml/min. Amazingly, there is actually an upper CrCl limit for afib. You must avoid edoxaban in patients with non-valvular afib and a CrCl > 95 ml/min. (I can almost guarantee you that will show up on a test question sometime in your life)
Alright, we've got renal function accounted for. I also want to point out that there are also potential dose adjustments for hepatic function with the NOACs, but they are much less clear cut (so I'm going to take the easy way out and just not talk about it :))
Let's move on to dosing schedule and enteral status (there's some weird restrictions with feeding tubes).
- Dabigatran - Dosed twice daily. Capsule CANNOT be opened and administered via tube.
- Rivaroxaban - Dosed once daily. Tablet CAN be opened and administered via tube. However, rivaroxaban must go through the stomach to be properly absorbed. So if it is given via tube, that tube must end in the stomach. NG, OG, PEG, or G tubes are all ok for rivaroxaban. You CANNOT use tubes that bypass the stomach such as NJ, J, or GJ.
- Apixaban - Dosed twice daily. Tablet CAN be crushed and administered via tube.
- Edoxaban - Dose once daily. Tablet CANNOT be crushed and administered via tube.
Moving on, what about drug interactions?
- Dabigatran - P-gp interactions
- Rivaroxaban - P-gp and CYP 3A4 interactions
- Apixaban - P-gp and CYP 3A4 interactions
- Edoxaban - P-gp interactions
The recommendation for how to handle the drug interaction varies depending on the indication and on how strong of an inhibitor/inducer you're working with. In general, it's recommended to avoid any NOAC with a strong P-gp inhibitor. There's enough of an overlap between P-gp inhibitors and CYP 3A4 inhibitors that you'll run into a lot of problems here. If your patient has HIV and is on a protease inhibitor, you're on a one-way trip to warfarin land.
What about if the patient develops a bleed? Is there a reversal agent available?
- Dabigatran - Reverse with idarucizumab [Praxbind]
- Rivaroxaban - No reversal agent
- Apixaban - No reversal agent
- Edoxaban - No reversal agent
I'd certainly hone in the fact that dabigatran has a reversal agent. If your patient has a history of bleeding or is high risk, that definitely can help to tip the scales of your clinical decision towards dabigatran.
Now, let's talk about efficacy...
Like everything else on this site, the following are my opinions. So please take them with a grain of salt.
I'll start with DVT/PE treatment, because that's an easy starting point.
In my mind, all NOACs are equally effective for DVT/PE treatment. You can select which one fits your patient best based on the above differences (renal function, dosing schedule, etc...) and on insurance coverage, formulary choices, and patient preference.
The only thing to keep in mind with NOACs and DVT/PE treatment is that they must be "bridged." I don't mean in exactly the same way that you bridge someone to warfarin, but the concept is sort of the same. With all NOACs, you either start with a parenteral anticoagulant (such as heparin or LMWH), or you start with a higher dose of the NOAC itself for a set period. Here are the specifics:
- Dabigatran - Start with 5 - 10 days of a parenteral anticoagulant. Then begin normal dosing of dabigatran (150 mg BID)
- Rivaroxaban - Start with 15 mg BID x 21 days. Afterwards, the dose is 20 mg Daily
- Apixaban - Start with 10 mg BID x 7 days. Afterwards, the dose is 5 mg BID
- Edoxaban - Start with 5 - 10 days of a parenteral anticoagulant. Then begin normal dosing of edoxaban (60 mg Daily)
That pretty much covers DVT/PE treatment. And again, you may also see some of the NOACs used for prophylaxis following hip or knee surgeries or in the setting of recurrent DVT/PE. Edoxaban is the only NOAC that does not have an FDA approval for this indication.
Let's move on now to non-valvular afib.
All of the NOACs were compared to the standard of care (warfarin) in their phase III trials. And all of the trials were non-inferiority trials (meaning they were just trying to prove that the NOAC was equal in effect to warfarin, not "better").
Because anticoagulation is such a high impact area for pharmacists, it's probably a good idea to look more closely at the individual trials. I'll give you my tl;dr take home points here, but I encourage you to read more closely into the studies for more detail. This post is already over 7000 words long, so if I were to try to add a journal club to it I might break the internet.
Here are the phase III trials for each of the NOACs:
And again, this article does a pretty great job at comparing and contrasting the NOACs (and their phase III trials). For some context, when we talk about stroke prevention in afib, we often calculate a CHADS2 score (or CHADS2VASC, or any of several other variations).
All these scores are doing is looking at patient risk factors to calculate their risk of having a stroke sometime within the next 10 years (it's kind of like the ASCVD score, but specifically for strokes). A higher CHADS2 score means a greater risk for a stroke.
Here is my "off the cuff" way of how I think about NOACs non-valvular afib for stroke prevention (at least in terms of the results of the above trials).
- Dabigatran - Non-inferior to warfarin, but higher risk of GI bleed. Patients in the trial had an average CHADS2 score of >/= 1
- Rivaroxaban - Non-inferior to warfarin. Patients had a higher average CHADS2 score (>/= 2)
- Apixaban - Superior to warfarin. Less risk of bleeding compared to warfarin. Patients had an average CHADS2 score of >/= 1
- Edoxaban - Non-inferior to warfarin. Average CHADS2 score was >/= 2
So what does all of that mean?
If your patient has a higher CHADS2 score, you could consider leaning towards rivaroxaban or edoxaban (because the patients in those trials had a higher CHADS2 score). But remember, edoxaban has that silly upper CrCl limit of 95 ml/min, so make sure your patient doesn't fit that profile.
All things being equal, apixaban is probably my favorite NOAC. It can be used in renal failure (again, with a caveat for dialysis patients), and I like the safety profile. Plus it can be crushed and administered via tube easily in the inpatient setting.
I don't like that dabigatran has an increased risk of bleeds. And unless the patient uses the blister packs (and doesn't open them), the drug has a ridiculously short shelf life. If a bottle of dabigatran is opened, it's only considered stable for 4 months. After that, you have to discard the medication.
So again, those are only my opinions. But that's how I think about the NOACs in afib.
What about Warfarin?!
I don't want to leave you hanging, but there's so much to say about warfarin that it really needs it's own article. I've written some short blurbs on it here and here, but warfarin deserves its own definitive guide. Eventually, we'll also write one up for antiplatelet medications. So stay tuned for those.
Get Your Anticoagulant/Antiplatelet Cheat Sheet
We've covered a lot in this article. If this is new for you, you might be a little overwhelmed. Or, maybe you want information on dosing (which we've included in our cheat sheet). Either way, the anticoagulant cheat sheet is the cure for what ails you.
This is easily our most comprehensive cheat sheet to date. It covers basically everything you need to know about anticoagulants and antiplatelet medications. You'll get common dosing (by indication), dosage adjustments for renal and hepatic failure, drug interactions, reversal agents, hold times for procedures, pregnancy category, basic pharmacology, relevant clinical pearls, and a whole lot more.
Honestly, you'll be amazed at how much information fits onto 3 sheets of paper when designed properly.
Use this sheet to ace your next exam. Or hang it up on the wall at your practice site as a quick and handy reference. You can even add it to your peripheral brain. Trust us, this cheat sheet is exactly what you've been looking for if you're someone who works with anticoagulants and antiplatelet medications.
And (like all of our other products), if you buy this cheat sheet, every revision or update we make to it is yours free for life. We are wayyyy to Type-A not to update this thing if a new anticoagulant gets approved by the FDA in 2 years. For literally $7, you can have your anticoagulant bases covered for the rest of your practice career.