Factor Xa inhibitors versus warfarin in severely obese patients with venous thromboembolism or atrial fibrillation

INTRODUCTION

Obesity represents a significant issue among Americans, as nearly 42% of adults are classified as obese, and 9% of adults are categorized as severely obese.^[1] High total body weight and high body mass index (BMI) are associated with an increased risk of venous thromboembolism (VTE) as well as an increased incidence of atrial fibrillation (AF), although the impact of BMI on stroke in AF is controversial.^[2-4] Current guidelines from the American College of Chest Physicians (CHEST) and the American Society of Hematology for acute VTE management recommend Factor Xa (FXa) inhibitors over warfarin.^[5,6] Similarly, current guidelines from the American College of Cardiology/American Heart Association/Heart Rhythm Society recommend FXa inhibitors over warfarin for the prevention of stroke and systemic embolism in nonvalvular AF.^[7] Compared to warfarin, FXa inhibitors are associated with reduced monitoring requirements, fewer drug/food interactions, and rapid onset of action.^[8]

Clinical data for the safety and efficacy of FXa inhibitors in patients with BMI >40 kg/m² or weight >120 kg are limited.^[9] Phase 3 clinical trials that led to the approval of FXa inhibitors (rivaroxaban, apixaban, and edoxaban) in patients with VTE or AF included relatively few obese subjects ranging from 12% to 34% of the total population. Subgroup analyses suggested weight did not significantly impact FXa inhibitor’s efficacy and safety. However, clinical outcomes in extremely obese patients (BMI >40 kg/m²) were not reported, as these analyses primarily used a BMI cutoff of ≥35kg/m² or weight >100 kg. In addition, results from pharmacokinetic (PK) studies indicated apixaban and dabigatran concentrations were affected by weight raising concerns for possible underdosing in severely obese patients.^[10,11] Alternatively, a PK study with rivaroxaban found weight did not impact maximum blood concentrations, the area under the concentration-time curve or FXa activity.^[12] In 2016, the International Society of Thrombosis and Hemostasis (ISTH) released guidance that suggested direct oral anticoagulants should not be used for AF or VTE treatment in patients with BMI >40 kg/m² or weight >120 kg due to the lack of clinical outcome data.^[9]

Since the ISTH Statement was released, several small, mostly single-center, and retrospective studies have explored the use of FXa inhibitors in severely obese patients with VTE and non-valvular AF.^[13-22] Meta-analyses and systemic reviews conducted with these studies along with post hoc analyses from phase 3 clinical trials indicate FXa inhibitors have similar efficacy outcomes compared to warfarin including similar rates of VTE recurrence in the VTE population and stroke/systemic embolism in the AF population.^[23-25] Major bleeding rates also appear to be similar or reduced in obese patients receiving FXa inhibitors for VTE and AF compared to warfarin. In 2021, the ISTH provided updated recommendations suggesting standard doses of apixaban and rivaroxaban could be used for the treatment of VTE regardless of obesity but did not provide recommendations for FXa inhibitor use in obese patients with AF.^[26] In addition, the authors of the ISTH Statement noted that clinical data was limited for the safety and efficacy of FXa inhibitors in obese patients with BMI >50 kg/m² and weight >150 kg in the treatment of VTE.

As FXa inhibitors continue to be prescribed with increasing frequency over warfarin, more data is needed to further define the risks and benefits of FXa inhibitors in severely obese patients.^[27] The purpose of this retrospective chart review was to determine the safety and efficacy of FXa inhibitors, specifically rivaroxaban and apixaban, compared to warfarin in patients with VTE or non-valvular AF and BMI >40 kg/m² or weight >120 kg. In addition to the evaluation of the safety and efficacy of FXa inhibitors, this study assessed the dosing regimens used. The safety and efficacy of FXa inhibitor’s use in patients with BMI >50 kg/m² or weight >150 kg were also evaluated.

MATERIALS AND METHODS

RESULTS

A total of 840 patients were screened, and 73 patients were included in the final analysis [Figure 1]. The most common reason for exclusion was not meeting inclusion criteria, specifically weight and BMI requirements (i.e., BMI <40 kg/m² or weight <120 kg) followed by receiving warfarin or an FXa inhibitor before admission. Baseline characteristics were generally balanced between the groups [Table 1]. Patients with AF were more likely to receive an FXa inhibitor versus warfarin (53% vs. 23%, P = 0.010). In the AF group, apixaban was the most common anticoagulant, followed by warfarin and rivaroxaban (67% vs. 23% vs. 10%). The most used anticoagulant in the VTE group was warfarin, followed by apixaban and rivaroxaban (54% vs. 29% vs. 17%). Medications that increase bleeding risk were similar between groups, except for aspirin, which was more common in the FXa inhibitor group (49% vs. 23%, P = 0.028). The FXa inhibitor group also reported significantly higher HAS-BLED scores (1 [1–2] vs. 0 [0–1], P = 0.032). Of note, the warfarin group reported higher weight and BMI at anticoagulation initiation, although this difference was not statistically significant.

Close

Figure 1:

Data collection process. DOAC: Direct oral anticoagulants, VTE: Venous thromboembolism, AF: Atrial fibrillation.

Export to PPT

The primary outcome of treatment failure was similar between the FXa inhibitor and warfarin groups (5% vs. 7%, P = 0.814; [Table 2]). Although there was no statistically significant difference in the rates of major bleeding or CRNMB, there was a trend toward decreased bleeding rates in the FXa inhibitor group compared to the warfarin group (9% vs. 23%, P = 0.182). This trend appears to have been driven by major bleeding rates, specifically due to a drop in hemoglobin ≥2 g/dL in the warfarin group (13% vs. 2%, P = 0.152).

Patients receiving apixaban reported higher rates of non-recommended dosing compared to rivaroxaban (3/33 [9%] vs. 0/10 [0%]). All non-recommended dosing in the apixaban group was due to decreased maintenance dose (1/33 [3%]) and missing or shortened lead in phase for VTE treatment (2/33 [6%]). Treatment failure did not occur in any patient who received a non-recommended dosing regimen. In a subgroup analysis of 32 patients with BMI >50 kg/m² or weight >150 kg, there was no difference in treatment failure between the FXa inhibitor and warfarin groups (0/15 [0%] vs. 1/17 [6%], P = 0.611; [Supplemental Table 3]). Similarly, rates of major bleeding or CRNMB did not differ between groups in the patients with BMI >50 kg/m² or weight >150 kg, but the FXa inhibitor group experienced numerically lower bleeding rates compared to the warfarin group (13% vs. 24%, P = 0.596).

DISCUSSION

This retrospective chart review found patients with severe obesity (BMI >40 kg/m² or weight >120 kg) experienced similar rates of treatment failure and bleeding when prescribed an FXa inhibitor (rivaroxaban and apixaban) or warfarin for the treatment of VTE or stroke/systemic embolism prevention in non-valvular AF. Although the rate of non-recommended dosing was higher in the apixaban group, recommended dosing was used in most patients. Despite the ISTH guidance statement in 2016, the FXa inhibitor group was larger than the warfarin group, signaling that FXa inhibitors have continued to be used in the obese population.

Based on the results obtained in this chart review, apixaban or rivaroxaban are appropriate treatment options in severely obese patients (BMI >40 kg/m² or weight >120 kg) at standard doses for VTE or AF. These results correlate with previous studies that compared FXa inhibitors versus warfarin in patients with higher body weights. Kushnir et al. conducted one of the largest single-center retrospective studies in severely obese adults (BMI >40 kg/m²) who were prescribed apixaban, rivaroxaban, or warfarin for VTE or AF (n = 795).^[13] Efficacy outcomes were similar between FXa inhibitors and warfarin specifically VTE recurrence in the VTE group and stroke in the AF group (P = 0.74 and P = 0.71, respectively). Clinically, relevant major and non-major bleeding rates were also similar in both VTE and AF populations (P = 0.45 and P = 0.16, respectively). When analyzed by BMI, a multicenter retrospective cohort study found efficacy and safety of FXa inhibitors were maintained as BMI increased.^[14] In a recent meta-analysis of retrospective studies and post hoc analysis of phase 3 clinical trials, FXa inhibitors were associated with significant reductions in the risk of major bleeding across the spectrum of weight (P = 0.0007), with no significant difference in VTE recurrence and stroke (P = 0.96).^[23] Notably, the dosing regimens in the FXa inhibitor group were often not addressed which could influence clinical outcomes. In a reply to a letter to the editor, Kushnir noted that standard dosing was used in most patients (94% in the rivaroxaban group, 89% in the apixaban group).^[33] Quan et al. found suboptimal dosing was present in 14.8% of obese patients (weight >120 kg) receiving an FXa inhibitor for VTE treatment.^[34] In the present study, recommended dosing was used in over 90% of patients receiving an FXa inhibitor, which is higher than previously reported.^[34]

Studies conducted in patients with solely VTE or AF have largely shown a similar or decreased risk of recurrent VTE, stroke, and bleeding with FXa inhibitors versus warfarin.^[15,16] However, one study found that AF patients with a higher BMI were more likely to experience worse outcomes with direct oral anticoagulants (specifically dabigatran, rivaroxaban, and apixaban) including shorter time to stroke/systemic embolism (P = 0.043) and major bleeding (P < 0.001).^[17] About 55% of the study population received dabigatran, which had previously been shown to be affected by weight in a PK study.^[11,17] On the other hand, another study found no difference in stroke, mortality, or clinically relevant bleeding in patients who weighed >120 kg versus ≤120 kg who received dabigatran.^[18] In meta-analyses of direct oral anticoagulants in patients with VTE or AF alone, efficacy and safety outcomes were similar between direct oral anticoagulants versus warfarin in severely obese patients.^[24,25] Specifically, apixaban and rivaroxaban have been shown to be safe and efficacious in patients with higher body weights in retrospective studies with each individual agent and is further demonstrated in the results of this study.^[19-22]

Although there is emerging data supporting the use of direct oral anticoagulants in patients with BMI >40 kg/m² or weight >120 kg, clinical outcome data in patients with BMI > 50 kg/m² or weight > 150 kg is even more limited in patients with AF or VTE. In fact, some authors recommend using warfarin and avoiding FXa inhibitors in patients with BMI >50 kg/m² or weight > 150 kg in the treatment of AF due to a lack of clinical data.^[35] One single-center retrospective study analyzed apixaban and rivaroxaban use in patients with AF and BMI > 50 kg/m² (n = 595).^[36] In this study, ischemic stroke and bleeding were not significantly different when patients with a BMI > 50 kg/m² were compared to patients with a BMI of 18–30 kg/m² (P = 0.544 for ischemic stroke). Similar results were found in another retrospective chart review in obese patients with weight ≥ 150 kg or BMI ≥ 50 kg/m² who were diagnosed with VTE or AF.^[37] Of note, a recent multidisciplinary expert panel recommended against using weight as a deciding factor for the choice of anticoagulation in patients with VTE but did acknowledge the lack of data in patients with extreme obesity.^[38] Our study provides additional data in this patient population demonstrating that FXa inhibitors continue to be safe and efficacious even with higher weights especially in patients with VTE as shown in the subgroup analysis of patients with BMI > 50 kg/m² or weight > 150 kg.

While this study does have several strengths, it is not without limitations. The retrospective design limited the data to information found in the electronic medical record, and data was unable to be collected if a patient visited an outside institution. Adherence rates for all medications and time in the therapeutic range for warfarin were unable to be assessed. In terms of baseline characteristics, the FXa inhibitor group reported significantly higher rates of aspirin use and higher median HAS-BLED scores, but this did not appear to affect bleeding rates. Weight at anticoagulation initiation was used which may change throughout the patient’s treatment course. Therapeutic drug monitoring of FXa inhibitors was also unavailable, although there is limited data on the utility of anti-FXa levels.^[39,40] A significant number of patients screened were not included due to not meeting weight or BMI requirements. This was most likely due to inaccuracies in ICD coding that were corrected once the individual’s chart was reviewed. Finally, this study included a small number of patients with a low event rate. Thus, the results are potentially underpowered to assess the significance of the primary outcome due to the risk of type II error. Until large, randomized control trial data is available; this study provides additional data supporting the use of FXa inhibitors in severely obese patients which will be used to encourage further study.

CONCLUSION

Our study found that rivaroxaban and apixaban were associated with similar rates of VTE recurrence, stroke, and bleeding compared to warfarin in severely obese patients for the treatment of VTE and stroke or systemic embolism in non-valvular AF. Based on these results, rivaroxaban and apixaban appear to be safe and effective for obese patients, including those with BMI >40 kg/m² or weight >120 kg. However, large prospective studies are needed to confirm these results in severely obese patients.