Comparison of Major Haemorrhagic Events in Atrial Fibrillation Patients Treated with Warfarin, Rivaroxaban, and Apixaban: A Randomized Controlled Trial

Atrial fibrillation (AF) is the most common cardiac arrhythmia, affecting approximately 3 to 6 million in the USA.¹,² The prevalence of AF rises in the elderly population. AF is associated with a 4- to 5-fold increased risk of ischemic stroke, and the incidence of strokes associated with AF increases with advancing age, ranging from 10% overall to 24% in those aged 80–89 years.³ The burden of stroke is substantial as this leads to functional impairment and effective prevention is best approach to limit this burden. Oral anticoagulation therapy with direct-acting oral anticoagulants (DOACs) and warfarin is the current standard of care for stroke prevention in patients with nonvalvular AF (NVAF).⁴ Patients with AF suffer from high stroke-related disease burden as they are 3 to 4 times more likely to suffer severe strokes. Among patients with AF, stroke-related risks of 1-year disability and 1-year mortality are twice those of non-AF-related strokes.⁵

 Rivaroxaban, a factor Xa inhibitor used in clinical practice since 2011, has increasing use in patients with existing and newly diagnosed AF. ⁶ Results from clinical and observational studies support the efficacy and safety of rivaroxaban in this population, including better effectiveness than warfarin for preventing stroke or systemic embolism. In the noninferiority clinical trial, ROCKET-AF, secondary end points of all-cause mortality and stroke severity were favorably impacted by rivaroxaban versus warfarin. Apixaban is the only DOAC to show risk reduction in both stroke/systemic embolism (SE) and major bleeding compared to warfarin in its Phase 3 trial (ARISTOTLE).⁷ Apixaban is also used to treat NVAF. Compared to warfarin, no anticoagulation monitoring is required for DOAC and fewer drug and food interactions are evident.⁸ Warfarin has narrow therapeutic range and broad dose-response variability. The food and drug interactions also pose challenges to anticoagulation maintenance. Patients with NVAF and INR <2 have an increased risk of death or cardiovascular-related events as acute coronary syndrome, ischemic stroke, transient ischemic attack, and systemic embolism compared with those with INR of 2 to 3.⁹ The need for constant INR monitoring and the potential consequences of poor anticoagulation in warfarin-treated patients pose a considerable burden.

 A study was conducted on warfarin, rivaroxaban and apixaban in 2023.¹⁰ New onset of bleeding was observed in 23.9% of patients treated with warfarin, 40.1% of the patients on rivaroxaban and 23.9% patients on apixaban. However, no other recent data was found comparing haemorrhagic outcomes of these drugs. Current study will be performed to compare the outcomes data in patients with AF being treated with warfarin, rivaroxaban and apixaban.

Keeping in view the above context this study was conducted to compare the incidence of major haemorrhagic events among patients receiving warfarin, rivaroxaban and apixaban for atrial fibrillation

 

The mean age of patients ranged from 58.98 ± 6.18 years in the Warfarin group to 60.28 ± 6.42 years in the Rivaroxaban group, with no statistically significant difference among the three groups (p = 0.337).

 Similarly, the mean body mass index (BMI) was comparable across all groups (Warfarin = 29.2 ± 4.1 kg/m², Rivaroxaban = 29.3 ± 3.9 kg/m², Apixaban = 29.5 ± 4.0 kg/m²; p = 0.300).

 In terms of gender distribution, males constituted approximately 58–60% of each group (Warfarin: 59 males, Rivaroxaban: 61, Apixaban: 57; p = 0.812), indicating no gender imbalance among groups.

The proportion of smokers was also similar, observed in 37.3% of Warfarin users, 35.3% of Rivaroxaban users, and 32.4% of Apixaban users (p = 0.742).

 Likewise, the prevalence of diabetes mellitus did not differ significantly between groups (Warfarin = 40.2%, Rivaroxaban = 38.2%, Apixaban = 39.2%; p = 0.961).

 Overall, no statistically significant differences were found among the three treatment groups for any baseline characteristic (all p > 0.05), confirming that the study populations were well matched and comparable at baseline Given in Table.1.a

Table 1. Baseline Characteristics of Study Participants (n = 306)

Characteristic		Warfarin (n = 102)	Rivaroxaban (n = 102)	Apixaban (n = 102)	p-value
Age (years), Mean ± SD		58.98 ± 6.18	60.28 ± 6.42	60.03 ± 7.28	0.337
Gender	Male n (%)	59 (57.8 %)	61 (59.8 %)	57 (55.9 %)	0.812
	Female n (%)	43 (42.2 %)	41 (40.2 %)	45 (44.1 %)
BMI (kg/m²), Mean ± SD		29.2 ± 4.1	29.3 ± 3.9	29.5 ± 4.0	0.3
Smoker n (%)		38 (37.3 %)	36 (35.3 %)	33 (32.4 %)	0.742
Diabetes Mellitus n (%)		41 (40.2 %)	39 (38.2 %)	40 (39.2 %)	0.961

 

Gastrointestinal (GI) bleeding was significantly more frequent in patients treated with Warfarin, occurring in 25 (24.5%) cases, compared to 10 (9.8%) in the Rivaroxaban group and 7 (6.9%) in the Apixaban group (p < 0.001).

 Intracranial hemorrhage (ICH) was observed in 6 (5.9%) Warfarin users, 1 (1.0%) Rivaroxaban user, and 4 (3.9%) Apixaban users, with the difference not reaching statistical significance (p = 0.167).

 When considering any major bleeding event (either GI bleed or ICH), 30 (29.4%) patients on Warfarin experienced bleeding compared with 11 (10.8%) on Rivaroxaban and 11 (10.8%) on Apixaban (p < 0.001), indicating a significantly lower overall bleeding risk with direct oral anticoagulants (DOACs).

 Death within 30 days of a bleeding event, evaluated as a secondary outcome, occurred in 5 (4.9%) patients in the Warfarin group, and in 2 (2.0%) patients each in the Rivaroxaban and Apixaban groups (p > 0.05). Although the difference was not statistically significant, mortality was numerically higher in Warfarin-treated patients.

 Overall, the results demonstrate that patients on DOACs, particularly Rivaroxaban and Apixaban, had a significantly lower incidence of major haemorrhagic complications compared with those on Warfarin, while intracranial haemorrhage and short-term mortality showed no significant intergroup difference given in Table 2.

 Table 2. Comparison of Haemorrhagic Events by Treatment Group

Outcome	Event	Warfarin (n = 102)	Rivaroxaban (n = 102)	Apixaban (n = 102)	Total (n = 306)	p-value
GI Bleed	No	77 (75.5 %)	92 (90.2 %)	95 (93.1 %)	264 (86.3 %)	< 0.001
	Yes	25 (24.5 %)	10 (9.8 %)	7 (6.9 %)	42 (13.7 %)
ICH	No	96 (94.1 %)	101 (99.0 %)	98 (96.1 %)	295 (96.4 %)	0.167
	Yes	6 (5.9 %)	1 (1.0 %)	4 (3.9 %)	11 (3.6 %)
Any Bleed	No	72 (70.6 %)	91 (89.2 %)	91 (89.2 %)	254 (83.0 %)	< 0.001
	Yes	30 (29.4 %)	11 (10.8 %)	11 (10.8 %)	52 (17.0 %)
Death within 30 days(Secondary Outcome)	No	97 (95.1 %)	100 (98.0 %)	100 (98.0 %)	297 (97.1 %)	> 0.05
	Yes	5 (4.9 %)	2 (2.0 %)	2 (2.0 %)	9 (2.9 %)

 

The present study compared the incidence of major hemorrhagic events among patients with atrial fibrillation (AF) treated with Warfarin, Rivaroxaban, and Apixaban. Our findings demonstrate that patients receiving direct oral anticoagulants (DOACs) experienced significantly fewer overall and gastrointestinal bleeding events than those treated with Warfarin, whereas intracranial hemorrhage (ICH) and short-term mortality did not differ significantly. These results are consistent with global evidence showing that DOACs offer superior safety while maintaining comparable or improved efficacy in stroke prevention among patients with non-valvular AF.

 Atrial fibrillation is the most common sustained cardiac arrhythmia, affecting over 33 million individuals worldwide and contributing substantially to ischemic stroke and mortality risk¹˒². Effective anticoagulation remains the cornerstone of stroke prevention in AF, but traditional Vitamin-K antagonists such as Warfarin have several limitations, including narrow therapeutic range, variable metabolism, and multiple food and drug interactions³˒⁷. The advent of DOACs including Rivaroxaban and Apixaban has transformed anticoagulation practice by offering predictable pharmacokinetics, rapid onset, and fewer interactions, without the need for routine monitoring⁵˒⁶.

 In our study, Warfarin users had nearly three times the frequency of major bleeding compared with patients receiving DOACs. This aligns with prior large randomized trials and real-world analyses. The ROCKET-AF and ARISTOTLE trials reported reduced rates of major or fatal bleeding with Rivaroxaban and Apixaban respectively, compared with Warfarin⁵˒⁶. Similarly, contemporary observational cohorts and meta-analyses have confirmed that Apixaban and Rivaroxaban are associated with lower risks of gastrointestinal or intracranial bleeding than Warfarin⁹˒¹⁰˒¹¹˒¹². Recent comparative studies further indicate that among DOACs, Apixaban may provide the most favorable bleeding profile, particularly in elderly and high-risk populations¹³˒¹⁴˒¹⁵.

 The lower incidence of GI and overall bleeding in our DOAC groups may be explained by their more targeted inhibition of coagulation factors (Xa inhibition) without suppression of vitamin-K–dependent proteins, resulting in stable anticoagulant effects and reduced variability in plasma concentration⁷˒¹⁰. Moreover, DOACs eliminate the need for frequent INR monitoring, reducing the risk of sub- or supra-therapeutic anticoagulation, a well-recognized cause of bleeding in Warfarin users⁸.

 Our results are supported by prior real-world studies that validated the safety and effectiveness of Rivaroxaban in clinical practice⁵ and demonstrated consistent efficacy of Apixaban across different risk profiles⁶˒⁹. The pooled data from multiple meta-analyses confirm that DOACs not only reduce major bleeding events but also lower the risk of intracranial haemorrhage compared with Warfarin⁹˒¹³˒¹⁵. Nevertheless, as in our findings, minor bleeding and gastrointestinal events remain the most frequent adverse outcomes requiring clinical attention.

 The present findings reinforce international guidelines from the American Heart Association and European societies, which recommend DOACs as the first-line therapy for stroke prevention in non-valvular AF³˒¹¹. Given the comparable efficacy with a superior safety profile, particularly in terms of major and life-threatening bleeding, DOACs represent a clinically advantageous alternative to Warfarin in routine practice.

 However, this study has certain limitations. It was conducted in a single tertiary-care centre with a moderate sample size and relatively short follow-up period. Laboratory confirmation of anticoagulation intensity (e.g., time in therapeutic range for Warfarin) was not included, which may underestimate Warfarin’s potential if managed optimally. Future multicentre, longer-term studies are warranted to evaluate the consistency of these findings in broader and higher-risk populations

In conclusion, our study confirms that direct oral anticoagulants Rivaroxaban and Apixaban are associated with significantly fewer major haemorrhagic events compared with Warfarin in patients with atrial fibrillation. These results support the growing body of international evidence favouring DOACs as safer, effective, and more convenient options for long-term anticoagulation¹⁰–¹⁵.

 Conflict Of interest: None

Funding Source: None