Apixaban, Dabigatran, Edoxaban, vs Rivaroxaban for Patients with Atrial Fibrillation

• • Journal Club Presented by: Dr. Luis A Aguilar Montalvan

Question:

Objective is to do a large-scale comparison between all DOACs (apixaban, dabigatran, edoxaban, and rivaroxaban) in routine clinical practice.

Background:

DOACs are widely used as primary therapy for patient with AF, however, no robust evidence exists to guide the choice of DOACs prescription.

Article Reviewed:

Lau WCY, Torre CO, Man KKC, et al. Comparative Effectiveness and Safety Between Apixaban, Dabigatran, Edoxaban, and Rivaroxaban Among Patients With Atrial Fibrillation : A Multinational Population-Based Cohort Study [published correction appears in Ann Intern Med. 2022 Dec 6;:]. Ann Intern Med. 2022;175(11):1515-1524. doi:10.7326/M22-0511

Methodology:

• Population:
  • Adult patients (older than 18 years) from France, Germany, United Kingdom and the United States with newly diagnosed AF who received a new DOAC prescription (from the DOACs of interest in this study) between 2010 through 2019.
  • Required patients who had at least 1 years worth of recorded medical history in the respective databases, and the new diagnosis happened within 90 days of the start of the index date, or after.

• Outcomes
  • Primary:
    • Rates of ischemic strokes/systemic embolism events, intracranial hemorrhage, gastrointestinal bleeding, and all-cause mortality.
  • Secondary:
    • Not explicitly stated 
    • Similar outcomes but for subgroups often underrepresented in clinical trials, which are adults age 80 yo or older, and patients with chronic kidney disease.
• Design
  • This is a multinational population-based prospective cohort study, that used 5 electronic health databases, standardized for quality-controlled purposes, and that contained information from the primary care, outpatient, and hospital settings.

• Exclusion Criteria:
  • History of mitral stenosis, hyperthyroidism, mechanical valve replacement (DOACs might be contraindicated), transient Atrial Fibrillation (typically secondary to surgery or acute pathology), prescription of Warfarin or other DOACs on or within 180 days before index date, a prescription of another anticoagulant on the index date, and a history of an outcome of interest.

Results:

• Primary Results:
  • Similar age and assigned sex distribution among all 5 databases.
  • Apixaban users tended to be older than other DOAC users in the U.S. Ambulatroy Electronic Medical Records database. (21.1% vs 4-11%, SMD >0.25) as well as Dabigatran users in the U.S. Hospital Charge Data Master. (21% vs 4%)
  • CHA2DS2-VASc score remained between 2-4 across databases. (risk of ischemic stroke 2.2-4.8% or 2.9-6.7% for other embolic events)
  • Most baseline characteristics od DOAC users were similar and remained well balanced during the study.
  • 9530 ischemic stroke or systemic embolism events, 8319 GIB, 1476 deaths, 841 ICH. 
  • Apixaban had lower GIB risk (Vs Dabigatran- HR, 0.81 [95% CI, 0.70 to 0.94]). (vs Rivaroxaban – HR, 0.72 [CI, 0.66 to 0.79]). (vs edoxaban – HR, 0.77 [CI, 0.66 to 0.91]). 
  • Apixaban was associated with lower risk for GIB when compared to rivaroxaban in bith reduced dose (HR, 0.68 [CI, 0.61 to 0.77]) and standard dose (HR, 0.72 [CI, 0.64 to 0.82]) cohort analyses.

• Secondary Results:
  • A total n=71,430 for patients with CKD and total n=101397 for patients 80 years or older
  • Risk of GIB in patients with CKD was lower with apixaban than dabigatran and rivaroxaban in the propensity score-stratified cohorts.
  • Apixaban was associated with lower risk of GIB when compared to other DOACs

Strengths:

• A total of 527,226 new DOAC users met inclusion criteria, seems large enough to notice some significant differences. 
• The comparisons between groups was extensive, and even the methods of analysis seems to have been maximized. 
• The data set used was standardized and the amount of data collected was significantly large. 

Limitations:

• The calculations are based on estimated probability of patients receiving a treatment, only subgroup analysis had explicit dose restrictions in their design. 
• Among the DOACs used, total prescriptions varied as follow: Apixaban, n=  281,320, followed by Rivaroxaban, n= 172,176, Dabigatra, n=61,008, and Edoxaban, n= 12,722, which makes one wonder if the comparison of discrepancies in data volume impact the HRs that they found, either in a positive or negative direction.
• There is no information about the severity of GIB, nor if reversal agent was used.
• Follow-up periods in the US databases was relatively short compared to other databases.
• There is no discussion on whether this new AF needed cardioversion, or warfarin bridge. Although, given exclusion, it seems that the new AF was within 48hr, no explicit data was available to address this. Nor in the supplemental data, which seems to be a limitation of dataset used since it may not include descriptive clinical course.
• Most patients were over the age of 65 y.o, and most, about 70% were on beta-blocker, 30% on CCB, but unsure of onset of these medications.

 Discussion:

• Apixaban and Rivaroxaban were the most commonly prescribed DOACs. 
• The findings in this study are generally consistent with the results from prior trials. Making these findings confirmatory of a known phenomenon, but they also add additional features to not previously address, i.e in patient with CKD and age 80 and older.
• Comparative RCTs are needed to provide further guidance on the superiority of inferiority of specific DOACs, though the data presented favors Apixaban on patients with high risk of GIB, there is still the consideration of not having easily accessible reversal agents in the way Warfarin does. Therefore, shared decision making and individualized risk vs benefit analysis remain mainstay in management, specially in the Emergency Medicine setting where follow-up remains a challenge.