New Oral Anticoagulants v. Warfin

By JASON R. PELLETIER | UMKC | Pharm.D. Candidate

Since it first appeared on the market in the early 1950s, warfarin has dominated the oral anticoagulation realm and kept many patients from premature thrombotic events. However, warfarin is also associated with a great number of drug and food interactions, trouble with genomic variations, and a narrow therapeutic window that make routine blood tests mandatory. These tests, along with faithful adherence to diet and drug dose (which may change each day) are not always achievable in some patients, and may result in a failure to maintain a therapeutic INR of 2-3. The blood tests and strict diet may also be considered a nuisance even in highly adherent patients. Fortunately for both groups, there are three new oral anticoagulants on the horizon: Eliquis™ (apixaban), Xarelto® (rivaroxaban), and Pradaxa® (dabigatran). These drugs are heralded as the new anticoagulants of the future that do not require monitoring and while this may be ideal in some patients it is not in others. A deeper comparative look between the new drugs and warfarin is necessary to decide which agent to use on a patient-by-patient basis.

Dabigatran is a direct thrombin inhibitor that inhibits factor IIa and is approved for prevention of stroke and systemic embolism in non-valvular atrial fibrillation. Unlike the other two new oral agents, dabigatran is predominately eliminated unchanged and does not interact with P450 enzymes. Dabigatran is, however, a substrate of the p-glycoprotein enzyme, so other drugs affected by this enzyme may cause increased or decreased plasma levels. Dabigatran is also very poorly protein bound, which means dialysis to remove the agent is reasonably possible in emergency situations. Unfortunately, short of a blood transfusion, there are no other agents currently available to reverse dabigatran.

In the RE-LY trial, two different doses of dabigatran (150mg and 110mg both given twice daily) were compared to warfarin; however the FDA has so far approved 75mg and 150mg. Both doses were shown to be non-inferior to warfarin, and the 150mg dose was superior. The 110mg dose of dabigatran showed significantly fewer rates of bleeding when compared to warfarin and higher dosed dabigatran. This would suggest that the 110mg dose of dabigatran may be associated with roughly the same level of anticoagulation as warfarin but has an even better safety profile. The 150mg dose of dabigatran demonstrated significantly lower rates of hemorrhagic stroke but higher rates of gastrointestinal bleeds when compared to warfarin.

Further analysis of the trial demonstrates that there was a trend for fewer strokes (non-hemorrhagic) in patients with warfarin when compared to 150mg of dabigatran as the quality of INR control improved. There were actually lower rates of major bleeding, major gastrointestinal bleeding, and mortality in the highest quartile (patients who spent the most time in a therapeutic INR of 2-3) of the warfarin population compared to 150mg of dabigatran. Dabigatran became statistically safer than warfarin as the time spent in an INR of 2-3 declined, however. Various lab tests such as activated partial thromboplastin time (aPTT), thrombin time (TT), and ecarin clotting time (ECT) may be elevated at therapeutic doses of dabigatran, however, there are no standards in the literature that define the appropriate therapeutic ranges of these lab tests yet. The current approximations for a reference range in those lab values are taken directly from the results seen in the RE-LY trial.

Rivaroxaban is an inhibitor of factor Xa and is approved for post-operative thromboprophylaxis in patients who have undergone hip or knee replacement surgery at an approved dose of 10mg daily (20mg if on concurrent CYP3A4 inducers). Rivaroxaban is also used off-label for the prevention of stroke and systemic embolism in patients with non-valvular atrial fibrillation. Rivaroxaban is a substrate of both p-glycoprotein and metabolized by CYP3A4 so while it may avoid some of the drug and food interactions with warfarin, it will also have additional interactions with substrates, inhibitors, or inducers of either enzyme.

The ROCKET-AF trial demonstrated a once daily dose of 20mg of rivaroxaban was non-inferior to warfarin and caused fewer major bleeding episodes in the prevention of stroke and systemic embolism in non-valvular atrial fibrillation. However, in the same trial, rivaroxaban was also shown to have been associated with a greater number of clinically relevant non-major bleeding episodes that resulted in either transfusion requirements or hemoglobin drops of 2 g/dL or greater. It may be prudent to take some of the results of this trial with a grain of salt, as the time spent in a therapeutic INR (2-3) for the warfarin patients in this trial was only 57.8%.

Like dabigatran, there is no antidote or reversal agent for rivaroxaban. Major clinical trials have not really addressed laboratory monitoring for rivaroxaban. However, PT changes have been shown to correlate well with rivaroxaban concentrations so this may be used to approximate a patient’s level of anticoagulation even though there is no standard therapeutic level in the literature yet. Lastly, even though it’s dosed once daily, rivaroxaban’s half-life is only 5-13 hours which may create a sub-therapeutic problem much faster (when compared to warfarin) in patients who have trouble with adherence.

Apixaban is also an inhibitor of factor Xa, but does not yet have any FDA approved indication and is consequently not on the market. Like rivaroxaban, apixaban is also a substrate of p-glycoprotein and metabolized by CYP3A4. Lastly, like the rest of the new oral agents, there is no antidote or reversal agent available for apixaban and monitoring is not required. The lack of monitoring may also be considered a downfall in certain situations. Very few of the trials out so far address patients with renal or hepatic insufficiency in great detail for any of the new anticoagulants. Treating these patients without under or overdose may be difficult without standardized monitoring parameters.

A recent trial for the drug in atrial fibrillation, ARISTOTLE, was published in the New England Journal of Medicine on September 15th, 2011. In this trial, participants were given 5mg dosed twice daily of apixaban or warfarin with a target INR of 2-3. Patients in the trial had atrial fibrillation and at least one additional risk factor for stroke. The results of this trial demonstrated that apixaban was superior to warfarin in the prevention of stroke or systemic embolism in patients with atrial fibrillation. The trial also demonstrated a lower risk of major and minor bleeding when compared to warfarin.

Like all new drugs, information regarding interactions, safety and efficacy is constantly coming in for the new oral anticoagulants. To date, these agents do not appear to have any major food interactions like warfarin does. However, a fed state in general may increase maximum plasma concentrations as well as the time to reach those concentrations, but does not appear to affect the elimination half-life of rivaroxaban. The manufacturer further recommends taking only the 20mg dose of rivaroxaban with food.

Cost is also a major concern, especially when a new drug comes on the market. In a cost-effectiveness article published by the American Heart Association in 2010, dabigatran 150mg was estimated to be more cost-effective for stroke and systemic embolism prevention in atrial fibrillation patients as compared to warfarin when the time spent in a therapeutic INR (2-3 with warfarin) was < 57.1%. Warfarin, on the other hand, was more cost effective than dabigatran when the time spent at the therapeutic INR was > 72.6%. The current information from the manufacturer states that an opened bottle must be used within 30 days due to drug degradation from moisture, although the manufacturer is currently evaluating stability at 60 days and beyond. The FDA currently recommends using any opened bottle within 60 days. This impaired stability in the presence of moisture has the potential to increase direct drug costs if stored inappropriately and healthcare costs if unaware patients continue to use a product that has been compromised and are no longer therapeutically anticoagulated.

Information presented at the International Society on Thrombosis and Haemostasis in 2009 evaluated the cost-effectiveness of using rivaroxaban over enoxaparin for post-operative DVT prophylaxis in hip and knee replacement patients. The results of this comparison evaluated four different studies and found that rivaroxaban was more cost-effective than enoxaparin (indirect healthcare costs were included). The studies looked at prophylaxis for 10-35 days and within that time frame rivaroxaban was roughly $100-400 cheaper depending slightly on who administered the enoxaparin.

Due largely to the lack of necessary therapeutic monitoring and food interactions, these drugs may be better choices for certain patients that cannot otherwise maintain an INR of 2-3 on warfarin. The drugs may directly cost more than warfarin at this point in time, but when considering the increased indirect healthcare costs due to stroke or systemic embolism for a poor warfarin candidate the new anticoagulants may come out ahead. However, if the only reason a specific patient is unable to maintain a therapeutic INR is due to trouble with adherence to a daily dose of warfarin then dabigatran and abixaban may not immediately counter the problem as they are dosed twice daily. Rivaroxaban also has a half-life significantly shorter than warfarin so intermittent non-adherence will lead to a sub-therapeutic state of anticoagulation and potential thrombotic event much quicker.

While less monitoring may be significantly more convenient for some patients and practitioners alike, it also means the drugs may be more dangerous to use in patient populations that are not as well studied yet, such as those with more severe renal or hepatic insufficiency or patients that are extremely over or under weight. Drug-drug interactions may also become problematic, as some can increase or decrease plasma levels of the anticoagulants and leave a patient sub- or supra-therapeutic for longer periods of time that regular warfarin monitoring would have caught. The risks of exposure to sub- or supra-therapeutic doses carry serious consequences that can lead to death and irreversible bleeding. The inability to reverse any of the three new agents without blood transfusions may also pose additional complications and unforeseen increases in healthcare expenditures when compared to patients that may have been kept out of the hospital if monitored and kept at a therapeutic dose of warfarin or given vitamin K to help reverse a supra-therapeutic warfarin bleed risk.

In summary, while these new agents may be great anticoagulation tools for some patients, they also carry with them potential problems. In general, patients who are poor warfarin candidates and do not spend sufficient time at the therapeutic INR range of 2-3 due to food or drug interactions or possibly adherence (to regular monitoring or different dose adjustments) and have healthy renal and hepatic function are the ideal candidates for the new oral anticoagulants. On the other hand, patients who are able to maintain a therapeutic INR without major complications may find that the new oral anticoagulants are not quite as safe and cost-effective as simply remaining on warfarin at this point in time. As with any new drug, some of this information may change as more and more data from post-marketing studies continues to come in.


Shah, Shimoli, MD and Brian Gage, MD, MSc. “Cost-Effectiveness of Dabigatran for Stroke Prophylaxis in Atrial Fibrillation.” American Heart Association. 123. (2011): 2562-2570. Web. 29 Sep. 2011.

Kwong, Louis, Aurea Duran, Alexi Diamantopoulos, Nishan Sengupta and Michael Lees. “Cost-Effectiveness of Rivaroxaban for Prevention of Deep Vein Thrombosis and Pulmonary Embolism after Total Hip or Knee Replacement in US Patients.” International Society on Thrombosis and Haemostasis (ISTH) XXII Congress. Bayer Schering Pharma AG and Johnson & Johnson Pharmaceutical Research & Development, L.L.C. Boston, MA. 11-16 July 2009. Poster Presentation.

Granger, Christopher, et al. “Apixaban versus Warfarin in Patients with Atrial Fibrillation.” New England Journal of Medicine. 365.11 (2011): 981-992. Web. 29 Sep. 2011.

Connolly, Stuart, et al. “Dabigatran versus Warfarin in Patients with Atrial Fibrillation.” New England Journal of Medicine. 361.12 (2009): 1139-1151. Web. 29 Sep. 2011.

Patel, Manesh, et al. “Rivaroxaban versus Warfarin in Nonvalvular Atrial Fibrillation.” New England Journal of Medicine. 365.10 (2011): 883-891. Web. 29 Sep. 2011.

Dabigatran. Lexicomp Online. Updated 9/22/11. Accessed 9/29/11.

Rivaroxaban. Lexicomp Online. Updated 9/22/11. Accessed 9/29/11.

Dabigatran. Micromedex 2.0. Accessed 9/29/11.

Rivaroxaban. Micromedex 2.0. Accessed 9/29/11.

Apixaban. Micromedex 2.0. Accessed 9/29/11.

Tweet with @TheMPA