Abstract
Background
Unlike warfarin direct oral anticoagulants (DOACs) are administered in fixed doses, which raises concerns of its effectiveness on larger patients. Data from randomized trials are limited on the safety and efficacy of DOACs in morbidly obese individuals with atrial fibrillation (AF).
Methods
We analyzed a cohort of obese (≥ 120 kg) and morbidly obese (BMI > 40 kg/m2) patients from the Veterans Health Administration system with AF who initiated apixaban, rivaroxaban, dabigatran, or warfarin between years 2012 and 2018. We used inverse probability of treatment weighting (IPTW) and Cox proportional hazards regression models to evaluate the relative hazard of death, myocardial infarction (MI), ischemic stroke, heart failure (HF), and bleeding events between oral anticoagulant (OAC) groups while censoring for medication cessation.
Results
We identified 6052 obese patients on apixaban, 4233 on dabigatran, 4309 on rivaroxaban, and 13,417 on warfarin (mean age 66.7 years, 91% males, 80.4% whites). At baseline patients on apixaban had the lowest glomerular filtration rate and highest rates of previous stroke and MI compared to other OACs. Among patients with weight ≥ 120 kg and those with BMI > 40 kg/m2, all DOACs were associated with lower risk of any hemorrhage, hemorrhagic stroke, and gastrointestinal (GI) bleeding. Patients with BMI > 40 kg/m2 treated with DOACs had similar ischemic stroke risk with those on warfarin.
Conclusions
In this large cohort of obese Veterans Health Administration system patients, the use of DOACs resulted in lower hemorrhagic complications than warfarin while maintaining efficacy on ischemic stroke prevention.
Keywords: Atrial fibrillation, Obesity, Stroke, Bleeding, Direct oral anticoagulants
Introduction
Oral anticoagulants, either vitamin K antagonists (VKA) such as warfarin, or direct oral anticoagulants (DOACs) reduce that thromboembolic risk in atrial fibrillation (AF) patients by about two-thirds irrespective of baseline risk [1]. Randomized controlled trials (RCT) of DOACs (dabigatran, rivaroxaban, apixaban, and edoxaban) have demonstrated similar protection against ischemic stroke but lower rates of ICH compared with VKAs [2–5]. Obese patients with AF have high risk for stroke or systemic embolism, death, and disability [6]. Obese individuals (body mass index (BMI) > 30 kg/m2) are significantly more likely to develop atrial fibrillation (AF) than those with BMI of < 25 kg/m2 [7, 8]. Furthermore, analysis of the Framingham Heart Study demonstrated a 5% increase in risk of AF with every unit increase in BMI [9], and data from the ARIC (Atherosclerosis Risk In Communities) study suggested that one in five cases of AF can be attributed to obesity [10]. Part of the risk for AF among obese patients is attributed to commonly encountered conditions such as hypertension, diabetes, and obstructive sleep apnea. Moreover, recent evidence suggests that patients with a BMI more than 40 kg/m2 have significantly higher warfarin requirements [11]. Therefore, the use of DOACs for thromboembolism prophylaxis would obviate need for frequent INR monitoring and dose adjustments in these patients. However, there is a paucity of large-scale clinical trial data or pharmacokinetic analyses in obese patients of high BMI. The International Society of Hemostasis and Thrombosis (ISTH) recommends avoidance of DOACs in individuals with a body mass index (BMI) > 40 kg/m2, or weight ≥ 120 kg, based on a review of available literature [12]. Nonetheless, retrospective studies have demonstrated a low incidence of stroke at 30days after direct current cardioversion for AF or atrial flutter among patients with BMI > 40 kg/m2 on DOACs or warfarin [13]. Also, a pharmacokinetic study of healthy volunteers with a weight over 120 kg who were taking rivaroxaban suggested limited impact of weight on pharmacokinetics and pharmacodynamics or rivaroxaban [14].
Currently, no randomized controlled trials of DOACs administered specifically to morbidly obese patients exist. In this context, we sought to study obese patients that initiated DOACs and are at risk for decreased exposure to DOAC drug concentrations and compare the risk of mortality, ischemic stroke, bleeding events, myocardial infarction, and heart failure of these patients to patients receiving warfarin in a community-based sample from the Veterans Health Administration (VHA) system.
Methods
Data Source
For this retrospective study, we utilized data from the Veterans Health Administration (VHA) Corporate Data Warehouse (CDW) for the period January 1, 2010, through December 31, 2018, available through the VA Informatics and Computing Infrastructure (VINCI). We extracted patient demographics, details of inpatient and outpatient physician visits, laboratory results, and pharmacy claims with detailed prescription fill information from the integrated electronic health record and administrative files of the VHA CDW. We used the Vital Status File, which includes information from multiple VA and non-VA sources including VA Beneficiary Death File, VA Medicare Status File, and Social Security Administration Death Master File, and provides information on date of death for deceased patients. Also, we identified fee-based admissions (admissions outside VA paid by VA). The study was approved by the University of Iowa and Iowa City VA Medical Center Institutional Review Board, with a waiver of consent due to the large number of patients involved.
Patient Population
Our primary cohort included obese veterans with weight ≥ 120 kg who initiated a DOAC or warfarin between 2013 and 2018. We chose 120 kg as the cutoff, as this is the weight above which there are concerns about lower drug level exposure according to ISTH [12]. Patients were included if they had a diagnosis of AF prior to initiating the DOAC or warfarin. AF was defined as International Classification of Diseases–Ninth Revision–Clinical Modification (ICD-9CM) code 427.31 or ICD-Tenth Revision (ICD-10) code I48.0, I48.1, I48.2, and I48.91, as primary or secondary diagnosis. We identified patients who initiated apixaban 5 mg or 2.5 mg twice daily, dabigatran 150 mg twice daily, or rivaroxaban 20 mg or 15 mg once daily. We did not include patients receiving edoxaban due to relatively low use of this drug during the study period. Patients were categorized into mutually exclusive groups according to the first DOAC received.
Covariates
Patient-level characteristics such as demographics, comorbid conditions, concurrent medication use, and prior VA health services use were extracted from the VHA CDW. Comorbid diseases were identified by ICD-9-CM/ICD-10 diagnoses on encounters during the 12 months preceding the date of first DOAC or warfarin fill. History of major bleeding was defined as any prior major bleeding, gastrointestinal bleeding, intracranial hemorrhage, or prior receipt of transfusion. Additional conditions included liver disease, alcohol abuse, obesity, chronic obstructive pulmonary disease, peripheral vascular disease, pulmonary circulation disease, heart valve disease, history of coronary revascularization, history of implantable cardiac device, and prior pulmonary embolism or deep vein thrombosis. We also extracted data on other prescription medication use during the 90 days before the initial DOAC or warfarin fill, including insulin, statins, beta blockers, angiotensin-converting enzyme (ACE) inhibitors, angiotensin-receptor blockers (ARBs), calcium channel blockers, diuretics, antiarrhythmics, prednisone, and non-steroidal anti-inflammatory drugs.
Endpoints
We identified the following clinical endpoints based on the primary ICD-9-CM/ICD-10 diagnosis on inpatient claims for acute care stays: (i) ischemic stroke, (ii) any major bleeding, (iii) gastrointestinal bleeding (GI), (iv) intracranial bleeding, (v) acute myocardial infarction, and (vi) heart failure. Diagnosis codes used to identify relevant hospital admissions are provided in Supplemental materials. All-cause mortality was identified from VA Vital Status Files. Patients were followed from the date of the initial DOAC or warfarin prescription until death, cessation of initial DOAC or warfarin, or December 31, 2018.
Statistical Analysis
We compared demographic characteristics, comorbid diseases, and concurrent medication use among patients on different DOACs and warfarin using the chi-square test for categorical variables and ANOVA or Kruskal-Wallis test (as appropriate) for continuous variables. We subsequently estimated inverse probability of treatment weighting (IPTW) for the treatment groups using multinomial logit models to predict the probability of receiving each DOAC or warfarin. We used n-way weighting with IPTW [15] to compare rates of all-cause mortality, ischemic stroke, any major bleeding, GI hemorrhage, intracranial hemorrhage, myocardial infarction, and heart failure per patient-year of follow-up across DOACs and warfarin. Furthermore, we used IPTW weights in multivariable Cox proportional hazards regression models with dependent variables being time from medication initiation to specific event to evaluate the relative hazard of each event while further controlling for patient characteristics at the time of DOAC or warfarin initiation. The results of regression analyses were reported as hazard ratios (HRs) with 95% confidence intervals (CIs) for apixaban versus warfarin, dabigatran versus warfarin, rivaroxaban versus warfarin, apixaban versus dabigatran, apixaban versus rivaroxaban, and dabigatran versus rivaroxaban.
Finally, we identified a second cohort of obese individuals with BMI > 40 kg/m2 and repeated all analyses described above on the second cohort. Analyses were conducted with the use of SAS, with 2-tailed level of significance set at 0.05.
Results
Patient Characteristics
We identified 28,011 veterans with weight ≥ 120 kg at the time of DOAC or warfarin initiation, including 6052 on apixaban, 3246 on dabigatran, 3299 on rivaroxaban, and 10,338 on warfarin. The characteristics of patients on DOACs and warfarin before and after applying IPTW weights are presented in Table 1. At baseline prior to applying IPTW weights, patients on apixaban had a longer time since initial AF diagnosis were older with higher rates of renal failure, peripheral vascular disease, previous stroke and MI compared with dabigatran and rivaroxaban. Dabigatran patients had higher rates of diabetes mellitus than apixaban and rivaroxaban. Patients on warfarin had higher rates of prior revascularization, myocardial infarction, heart failure, peripheral vascular disease, and renal failure than patients on DOACs. Also, the rates of prior gastrointestinal and any major bleeding were higher among patients on warfarin compared with DOACs. After applying IPTW weights, all standardized differences in demographics, comorbid conditions, and medications use among oral anticoagulant groups were < 10% (Fig. 1).
Table 1.
Baseline characteristics of obese patients on direct oral anticoagulants and warfarin
| Unweighted characteristics (before IPTW) | Characteristics after weighting for IPTW | |||||||
|---|---|---|---|---|---|---|---|---|
|
|
|
|||||||
| Variables | Apixaban (n = 4471) | Dabigatran (n = 3246) | Rivaroxaban (n = 3299) | Warfarin (n = 10,338) | Apixaban (n = 4471) | Dabigatran (n = 3246) | Rivaroxaban (n = 3299) | Warfarin (n = 10,338) |
|
| ||||||||
| Age, mean (years) | 69.9 | 65.7 | 66.7 | 66.5 | 67.3 | 67.2 | 67.3 | 67.2 |
| Female sex, % | 1% | 1% | 1% | 1.1% | 8% | 7% | 10% | 11% |
| Years | ||||||||
| 2013–2014 (%) | 1.4% | 26.2% | 9.9% | 45% | 22.6% | 29% | 26.9% | 28.9% |
| 2015–2016 (%) | 47.3% | 50.1% | 52.5% | 39.8% | 46.2% | 46.8% | 45.2% | 46.3% |
| 2017–2018 (%) | 51.3% | 23.7% | 37.6% | 38.6% | 31.1% | 24.2% | 27.9% | 24.7% |
| Race, % | ||||||||
| White | 81.2% | 81.5% | 80.3% | 80.8% | 79.5% | 81% | 80.3% | 81% |
| Black | 9.9% | 9.6% | 11.0% | 10.8% | 11.6% | 10.1% | 10.3% | 10.6% |
| Hispanic | 1.8% | 2.1% | 2% | 1.9% | 2.6% | 1.9% | 2.1% | 1.9% |
| Other | 1.6% | 1.8% | 1.8% | 1.6% | 1.6% | 2.0% | 1.9% | 1.6% |
| Comorbid conditions, % | ||||||||
| Hypertension | 84.9% | 84.5% | 83.2% | 86.8% | 85.4% | 86.2% | 85.2% | 85.2% |
| Diabetes mellitus | 22% | 29.1% | 25.9% | 31.8% | 26.7% | 28.7% | 27. 5% | 28.7% |
| Obstructive sleep apnea | 43.1% | 43.9% | 43.5% | 46.3% | 45.2% | 45.4% | 45.1% | 45.2% |
| Cardiac implantable devices | 7.3% | 8.3% | 6.5% | 7.2% | 7.8% | 7.3% | 6.9% | 7.4% |
| Ischemic stroke | 7% | 5% | 4.5% | 7.3% | 6.1% | 7.2% | 6.3% | 6.5% |
| Prior revascularization | 3.4% | 4.7% | 3.2% | 7.2% | 5.4% | 6.1% | 5.3% | 5.5% |
| Prior myocardial infarction | 3.5% | 1.7% | 2.4% | 4% | 5.1% | 3.4% | 3.1% | 3.4% |
| Congestive heart failure | 31.1% | 26.2% | 27.7% | 35.8% | 34.4% | 31.2% | 32% | 31.8% |
| Valvular disease | 6.9% | 6.3% | 5% | 10.6% | 8.6% | 8.8% | 8.6% | 8.5% |
| Peripheral vascular disease | 11.4% | 8.4% | 8.8% | 12% | 10.4% | 10.5% | 10.4% | 11.2% |
| Chronic obstructive pulmonary disease | 26.8% | 24.8% | 24.6% | 29% | 28.6% | 27.6% | 27.9% | 27.3% |
| Hypothyroidism | 11.7% | 10.5% | 10.8% | 10.7% | 12.2% | 11.6% | 10.9% | 10.8% |
| Renal failure | 17.2% | 8% | 10% | 19.7% | 20.7% | 15.7% | 16.7% | 16.2% |
| Liver failure | 4.1% | 3.7% | 4% | 4.6% | 5.2% | 4.3% | 4.0% | 4.6% |
| Alcohol abuse | 3.9% | 5.1% | 5.2% | 6% | 5.2% | 4.9% | 5.6% | 5.2% |
| Drug abuse | 2.3% | 2.7% | 3% | 2.8 | 4.1% | 3.0% | 3.3% | 2.5% |
| Depression | 19% | 17.7% | 19.5% | 20% | 19.5% | 19.4% | 19.3% | 19.5% |
| History of bleeding (%) | ||||||||
| Gastrointestinal bleeding | 3.7% | 3.3% | 4% | 6.2% | 5.8% | 4.8% | 5.2% | 5.1% |
| Intracranial bleeding | 0.2% | 0.1% | 0.2% | 0.3% | 0.2% | 0.2% | 0.2% | 0.2% |
| Any major bleeding | 8.7% | 7.2% | 7.8% | 11.6% | 11.2% | 9.5% | 10.5% | 10.2% |
| Days since AF diagnosis | 484 | 446 | 465 | 371 | 420.3 | 419.2 | 426.7 | 422.5 |
| Medications (%) | ||||||||
| Beta blocker | 65.3% | 64.6% | 63.8% | 72.9% | 69.7% | 69.3% | 68.4% | 69.0% |
| Angiotensin receptor blocker | 17.6% | 15.6% | 17.7% | 15.8% | 17.1% | 16.5% | 17.4% | 16.1% |
| ACE inhibitor | 38% | 42.6% | 40.1% | 46.2% | 42.8% | 44.3% | 44.1% | 43.8% |
| Calcium channel blocker | 15.2% | 17.2% | 15.8% | 16.5% | 17.1% | 15.8% | 15.1% | 16.7% |
| Corticosteroid, oral | 5.3% | 4.5% | 5.5% | 6.7% | 5.8% | 5.3% | 6.5% | 6.3% |
| NSAID | 11.6% | 12.9% | 11.9% | 12.1% | 11.5% | 12.4% | 11.3% | 12.5% |
| Statin | 62.4% | 60.6% | 59.7% | 65.2% | 64.3% | 63.1% | 63.4% | 63.7% |
| SSRI/SNRI | 20.3% | 19.7% | 20.2% | 21.5% | 18.7% | 19.6% | 20.5% | 20.9% |
| Antiarrhythmic | 18.7% | 19.9% | 19.4% | 16.7% | 18.4% | 18.1% | 18.0% | 18.2% |
| Insulin | 24.7% | 20.3% | 22.7% | 27.7% | 28.9% | 25.3% | 24.9% | 25.5% |
| Diuretic | 42.5% | 37.3% | 38.7% | 49.5% | 45.5% | 45.2% | 46.0% | 45.3% |
| Opioid | 31.3% | 32.2% | 31.2% | 38.4% | 35.1% | 35.4% | 35.7% | 35.3% |
Abbreviations: AF atrial fibrillation, ACE angiotensin-converting enzyme, ASD absolute standardized difference, Apix apixaban, Dabi dabigatran, Riva rivaroxaban, SSRI selective serotonin reuptake inhibitor, NSAID non-steroidal anti-inflammatory drugs
Fig. 1.
Standardized differences between oral anticoagulant groups at the time of treatment initiation, after adjusting for inverse probability of treatment weighting (IPTW)
Outcomes: Comparisons of Different DOACs
The mean follow-up time to death or medication cessation over all patients was 19 months. Prior to weighting, there were 211 total ischemic strokes and 837 major bleeding events in all treatment groups. Table 2 shows unadjusted numbers of events and event rates per 100 patient-years, after adjusting for IPTW.
Table 2.
Number of events and event rates per 100 patient-years among obese patients (≥ 120 kg) on direct oral anticoagulants and warfarin
| Apixaban (n = 6052) | Dabigatran (n = 4233) | Rivaroxaban (n = 4309) | Warfarin (n = 13,417) | |
|---|---|---|---|---|
|
| ||||
| Median follow up, years | 1.27 | 1.74 | 1.50 | 1.95 |
| Outcomes | ||||
| Ischemic Stroke | ||||
| Events before IPTW | 32 | 29 | 26 | 124 |
| Event rate (95%CI)/patient years after IPTW | 0.63 (0.57–0.71) | 0.48 (0.42–0.55) | 0.35 (0.30–0.41) | 0.42 (0.37–0.49) |
| Gastrointestinal hemorrhage | ||||
| Events before IPTW | 68 | 50 | 59 | 381 |
| Event rate (95% CI)/patient-years after IPTW | 0.83 (0.75–0.92) | 1.02 (0.93–1.1) | 1.14 (1.05–1.24) | 1.39 (1.29–1.5) |
| Hemorrhagic Stroke | ||||
| Events before IPTW | 7 | 2 | 7 | 53 |
| Event rate (95% CI)/patient-years after IPTW | 0.04 (0.2–0.06) | 0.03 (0.01–0.05) | 0.07 (0.05–0.09) | 0.18 (0.15–0.23) |
| Any major bleeding | ||||
| Events before IPTW | 99 | 64 | 91 | 583 |
| Event rate (95%CI)/patient-years after IPTW | 1.1 (1.02–1.21) | 1.21 (1.11–1.31) | 1.72 (1.6–1.84) | 2.1 (2–2.2) |
| Acute myocardial infarction | ||||
| Events before IPTW | 40 | 42 | 50 | 208 |
| Event rate (95%CI)/patient-years after IPTW | 0.78 (0.7–0.86) | 0.59 (0.52–0.66) | 0.85 (0.77–0.94) | 0.75 (0.67–0.83) |
| Heart failure admission | ||||
| Events before IPTW | 429 | 305 | 330 | 1788 |
| Event rate (95%CI)/patient-years after IPTW | 6.2 (5.98–6.44) | 5.6 (5.4–5.84) | 6.14 (5.92–6.37) | 7.2 (6.9–7.42) |
| All-cause mortality | ||||
| Events before IPTW | 328 | 183 | 177 | 1047 |
| Event rate (95%CI)/patient-years after IPTW | 4.44 (4.26–4.64) | 3.12 (2.96–3.28) | 2.83 (2.68–2.98) | 3.84 (3.67–4.02) |
Abbreviations: IPTW inverse probability of treatment weighting, CI confidence intervals
In Cox regression models that incorporated IPTW weights (Table 3, Fig. 2), we found significantly higher all-cause mortality among obese patients ≥ 120 kg, on apixaban compared with dabigatran (p < 0.001) and rivaroxaban (p < 0.001) and dabigatran compared with rivaroxaban (p = 0.009). Furthermore, we found significantly higher ischemic stroke risk with apixaban compared with rivaroxaban (p < 0.001) and dabigatran (p < 0.001) as well as dabigatran compared with rivaroxaban (p = 0.004). Analysis of any major bleeding events demonstrated significantly lower major bleeding risk with apixaban than rivaroxaban (p < 0.001), similar major bleeding risk between apixaban and dabigatran (p = 0.23), and lower major bleeding risk with dabigatran compared to rivaroxaban (p < 0.001). Apixaban was associated with significantly lower gastrointestinal bleeding than rivaroxaban (p < 0.001) and dabigatran (p = 0.005), while dabigatran was associated with significantly lower risk than rivaroxaban (p = 0.048). Finally, the hemorrhagic stroke risk was significantly lower with apixaban than rivaroxaban (p = 0.049) and dabigatran than rivaroxaban (p = 0.006), while there was no difference in risk between apixaban and dabigatran (p = 0.4). Myocardial infarction risk was significantly lower with dabigatran versus rivaroxaban (p < 0.001) and higher with apixaban compared with dabigatran (p < 0.001) while similar in risk between apixaban and rivaroxaban (p = 0.3). The risk of heart failure was significantly higher in the apixaban group (p < 0.001) compared with dabigatran and lower in the dabigatran group compared with rivaroxaban (p < 0.001) but similar between apixaban and rivaroxaban.
Table 3.
Outcomes of obese patients (> 120 kg) with AF treated with different direct oral anticoagulants or warfarin
| Direct oral anticoagulants (DOACs) compared to each other | Direct oral anticoagulants (DOACs) compared to warfarin | |||||
|---|---|---|---|---|---|---|
|
|
|
|||||
| Apixaban versus rivaroxaban | Apixaban versus dabigatran | Dabigatran versus rivaroxaban | Apixaban versus warfarin | Dabigatran versus warfarin | Rivaroxaban versus warfarin | |
|
| ||||||
| All-cause mortality | 1.57 (1.46–1.68) | 1.42 (1.33–1.52) | 1.1 (1.03–1.19) | 1.15 (1.08–1.23) | 0.81 (0.76–0.87) | 0.74 (0.69–0.79) |
| Ischemic stroke | 1.83 (1.52–2.21) | 1.37 (1.15–1.63) | 1.34 (1.09–1.63) | 1.56 (1.3–1.87) | 1.14 (0.94–1.38) | 0.85 (0.69–1.04) |
| Bleeding | ||||||
| Any major | 0.65 (0.59–0.72) | 0.93 (0.83–1.05) | 0.7 (0.63–0.78) | 0.53 (0.47–0.59) | 0.57 (0.51–0.63) | 0.81 (0.74–0.89) |
| GI hemorrhage | 0.73 (0.64–0.83) | 0.82 (0.72–0.94) | 0.88 (0.78–0.99) | 0.6 (0.53–0.68) | 0.73 (0.65–0.83) | 0.83 (0.74–0.93) |
| Hemorrhagic stroke | 0.56 (0.32–0.99) | 1.35 (0.67–2.7) | 0.42 (0.23–0.78) | 0.23 (0.14–0.38) | 0.17 (0.1–0.3) | 0.4 (0.27–0.6) |
| Acute myocardial infarction | 0.93 (0.8–1.07) | 1.34 (1.14–1.56) | 0.69 (0.6–0.81) | 1.06 (0.9–1.2) | 0.79 (0.67–0.93) | 1.14 (0.99–1.31) |
| Acute heart failure | 1 (0.95–1.05) | 1.1 (1.04–1.16) | 0.91 (0.86–0.96) | 0.87 (0.83–0.92) | 0.79 (0.75–0.83) | 0.87 (0.83–0.92) |
Fig. 2.
Forest plot presenting the hazard ratios with 95% confidence intervals of patients weighing ≥ 120 kg
Outcomes: Comparisons of DOACs vs Warfarin
Both dabigatran (p < 0.001) and rivaroxaban (p < 0.001) had lower rates of all-cause mortality than warfarin. Apixaban had higher all-cause mortality risk than warfarin (p < 0.001) (Table 3). The risk of ischemic stroke was similar between dabigatran, rivaroxaban, and warfarin but higher in the apixaban group compared with warfarin (p < 0.001). However, all DOACs had lower risk of bleeding than warfarin. Specifically, any major bleeding risk was significantly lower with apixaban, dabigatran, and rivaroxaban compared with warfarin. Gastrointestinal bleeding was significantly lower with all three DOACs compared with warfarin (p < 0.001 for all comparisons). Similarly, hemorrhagic stroke risk was significantly lower with apixaban, dabigatran, and rivaroxaban than warfarin (p < 0.001 for all comparisons). Myocardial infarction rates were significantly lower with dabigatran compared with warfarin (p = 0.004) without significant differences between apixaban, rivaroxaban, and warfarin. Finally, admissions for heart failure were significantly lower with each DOAC compared to warfarin (p < 0.001 for all comparisons) (Table 3).
Subgroup Analysis of Morbidly Obese Patients
Among patients with BMI > 40 kg/m2, we identified 3414 on apixaban, 2405 on dabigatran, 2340 on rivaroxaban, and 8267 on warfarin. Associations between anticoagulant type and outcomes in IPTW analysis in patients with BMI > 40 kg/m2 are shown on Table 4 and Fig. 3. Dabigatran was associated with significantly lower all-cause mortality compared with rivaroxaban in the subset of patients with morbid obesity (p = 0.001), whereas apixaban was associated with higher mortality than dabigatran (p < 0.001) and rivaroxaban (p = 0.013). There were no differences in the risk of ischemic stroke across DOAC groups. Any major bleeding risk was similar between apixaban and rivaroxaban (p = 0.25), higher with apixaban than dabigatran (p < 0.001) and lower with dabigatran than rivaroxaban (p = 0.005). With regard to the risk of gastrointestinal bleeding, no significant differences among DOAC groups were found. The risk of hemorrhagic stroke was significantly lower with apixaban than rivaroxaban (p = 0.01) and dabigatran than rivaroxaban (p = 0.02). Regarding the risk of myocardial infarction, apixaban was associated with significantly lower risk than rivaroxaban (p < 0.001) and similar risk with dabigatran (p = 0.09), whereas dabigatran was associated with significantly lower risk than rivaroxaban (p < 0.001) (Table 4). The risk of heart failure was higher with apixaban than dabigatran (p < 0.001) and rivaroxaban (p = 0.011), whereas dabigatran was associated with significantly lower risk than rivaroxaban (p < 0.001). Comparisons of each DOAC to warfarin were generally similar to these of the primary analysis, with minor differences. Apixaban (p = 0.048), dabigatran (p < 0.001), and rivaroxaban (p < 0.001) had lower rates of all-cause mortality than warfarin but similar risk of stroke (Table 4, Fig. 3). The rates of any major bleeding, gastrointestinal bleeding, and intracranial bleeding were significantly lower with apixaban, dabigatran, and rivaroxaban compared with warfarin (p < 0.01 for all comparisons) (Table 4, Fig. 3). The risk of myocardial infarction was significantly lower with apixaban (p = 0.03) and dabigatran (p < 0.001) compared with warfarin but was higher in the rivaroxaban group compared with warfarin (p < 0.001). Finally, heart failure risk was similar between apixaban and warfarin (p = 0.14) but significantly lower with dabigatran and rivaroxaban compared with warfarin (p < 0.001 for both comparisons) (Table 4, Fig. 3).
Table 4.
Outcomes of obese patients (BMI > 40 kg/m2) with AF treated with different direct oral anticoagulants or warfarin
| Direct oral anticoagulants (DOACs) compared to each other | Direct oral anticoagulants (DOACs) compared to warfarin | |||||
|---|---|---|---|---|---|---|
|
|
|
|||||
| Apixaban versus rivaroxaban | Apixaban versus dabigatran | Dabigatran versus rivaroxaban | Apixaban versus warfarin | Dabigatran versus warfarin | Rivaroxaban versus warfarin | |
|
| ||||||
| All-cause mortality | 1.12 (1.02–1.23) | 1.3 (1.20–1.44) | 0.85 (0.78–0.94) | 0.92 (0.84–0.99) | 0.7 (0.64–0.76) | 0.82 (0.75–0.89) |
| Ischemic stroke | 0.85 (0.65–1.11) | 1 (0.76–1.32) | 0.85 (0.65–1.1) | 0.93 (0.71–1.22) | 0.92 (0.71–1.2) | 1.1 (0.84–1.41) |
| Bleeding | ||||||
| Any major | 1.08 (0.94–1.25) | 1.35 (1.17–1.57) | 0.8 (0.69–0.93) | 0.72 (0.63–0.82) | 0.53 (0.46–0.61) | 0.66 (0.58–0.75) |
| GI hemorrhage | 1.13 (0.95–1.34) | 1.08 (0.94–1.25) | 1.03 (0.87–1.23) | 0.78 (0.67–0.92) | 0.71 (0.61–0.84) | 0.7 (0.59–0.82) |
| hemorrhagic stroke | 0.27 (0.1–0.74) | 0.77 (0.25–2.39) | 0.36 (0.15–0.84) | 0.13 (0.05–0.32) | 0.17 (0.08–0.36) | 0.47 (0.27–0.81) |
| Acute myocardial infarction | 0.55 (0.45–0.67) | 1.22 (0.97–1.53) | 0.45 (0.37–0.56) | 0.79 (0.64–0.98) | 0.65 (0.52–0.80) | 1.44 (1.2–1.72) |
| Acute heart failure | 1.09 (1.02–1.16) | 1.32 (1.23–1.42) | 0.82 (0.77–0.8) | 0.95 (0.9–1.01) | 0.72 (0.67–0.77) | 0.87 (0.82–0.93) |
Fig. 3.
Forest plot presenting the hazard ratios with 95% confidence intervals of the main study outcomes among morbidly obese patients (BMI > 40)
Discussion
The salient findings of this analysis of a large sample of obese veterans with AF treated with DOACs or warfarin can be summarized as follows: (i) among morbidly obese patients, ischemic stroke risk did not differ significantly among apixaban, dabigatran, rivaroxaban, and warfarin, whereas in the group of patients with weight ≥ 120 kg, apixaban was associated with higher risk of stroke than warfarin; (ii) the hemorrhagic stroke risk was similar among the three DOACs and significantly lower compared with warfarin; (iii) all three DOACs had significantly lower bleeding risk compared to warfarin, while rivaroxaban had higher hemorrhagic stroke risk compared with apixaban and dabigatran in morbidly obese patients and in the group of patients with weight ≥ 120 kg; (iv) dabigatran and rivaroxaban was associated with lower mortality risk compared to apixaban and warfarin; and (v) all-cause mortality was higher with apixaban compared with dabigatran and rivaroxaban in morbidly obese patients and those with weight ≥ 120 kg. It is important to note that differences in all-cause mortality among DOACs may represent heterogeneous populations and variable comorbidities not captured by our analysis rather than differential effects on thromboembolic and bleeding risk. For example, at baseline before IPTW, the rate of renal failure was higher among apixaban and warfarin recipients, and this rate remained numerically higher but with standardized difference < 0.1 after IPTW. Therefore, it is possible that unmeasured differences in comorbidity burden among different DOAC groups may account for the observed differences in outcomes. However, when focusing on thromboembolic and bleeding outcomes in morbidly obese patients, ischemic stroke risk was similar among DOACs and warfarin, whereas DOACs had lower bleeding risk than warfarin with apixaban and dabigatran having overall better safety profile in terms of bleeding, compared with rivaroxaban.
The most recent update of the American Heart Association/American College of Cardiology/Heart Rhythm Society (AHA/ACC/HRS) guidelines in 2019 on the management of AF recommended that DOACs are the preferred agents for anticoagulation rather than warfarin. The guidelines also acknowledge the limited evidence of efficacy and safety of DOACs in severely obese patients. Thus, the recommendation is to monitor serum levels of these drugs in patients with BMI > 40 kg/m2 or weight > 120 kg [16]. In the three landmark trials comparing DOACs to warfarin in AF patients, the average BMI for included individuals was < 30 kg/m2, and prevalence of patients with BMI > 35 kg/m2 was 10–15% [2–4]. Several studies aimed to evaluate safety and efficacy of DOACs in severely obese patients; however, their results exhibited limitations including single-center source of data, small sample sizes, or pooled analysis of DOACs in a single group [17–19]. Given the increasing epidemic of obesity, with estimated prevalence of obesity in the USA of 40% [20], the lack of evidence of comparative efficacy and safety of anticoagulation agents in obese patients is concerning. Furthermore, pharmacokinetic data suggest that body weight may affect the pharmacokinetics of apixaban, rivaroxaban, and dabigatran following administration of fixed doses [21]. Following a single dose of apixaban in healthy subjects, weight ≥ 120 kg had 30% higher clearance and 24% higher volume of distribution [22], which translated into a 31% lower peak concentration and 23% lower area under the curve compared with the normal-weight group. In a similar study conducted with rivaroxaban, weight > 120 kg was not associated with significantly altered rivaroxaban exposure or volume of distribution [14]. Weight ≥ 100 kg is associated with 21% lower dabigatran trough concentration compared with patients weighing 50–100 kg [23]. Despite the effects of body weight extremes on pharmacologic properties of DOACs, our analysis suggests that the use of these agents is safe and effective in obese and morbidly obese patients compared with warfarin, with similar ischemic stroke risk but lower risk of bleeding, mortality, and heart failure.
In ARISTOTLE and RE-LY trials, apixaban and dabigatran were associated with lower rates of stroke and systemic embolism compared to warfarin [2–4]. While in ROCKET AF trial, rivaroxaban had similar rates of stroke and systemic embolism compared to warfarin [3]. In our study, DOACs and warfarin had similar comparative efficacy in AF patients with BMI > 40 kg/m2, as there was no difference between the four anticoagulation agents in stroke rates. However, in the analysis of patients with weight ≥ 120 kg, patients on apixaban had higher risk of ischemic stroke than dabigatran and rivaroxaban. Baseline differences in comorbidities but also administration of certain DOACs at doses lower than recommended may contribute to these differences [24]. Regarding rates of hemorrhagic stroke, in our study, all DOACs were associated with lower risk compared to warfarin in obese patients with AF. Furthermore, all three DOACs were associated with lower rates of major bleeding compared to warfarin. These findings are similar to the three landmark trials comparing apixaban, rivaroxaban, and dabigatran to warfarin in AF patients. Collectively, these findings suggest that in obese and morbidly obese patients, DOACs in general are as effective as warfarin, and they offer the advantage of greater safety in major bleeding and hemorrhagic stroke risks.
Interestingly, we detected differences among DOACs on the risks of death, thromboembolism, and bleeding. Although, these differences may be explained by heterogeneity in the burden of comorbidities among DOAC groups, which were not corrected by the IPTW, there are other potential explanations for this finding. First, dabigatran mechanism of action is different than the other two DOACs. It acts as a direct thrombin inhibitor, while both apixaban and rivaroxaban are factor Xa inhibitors. Use of enoxaparin in severely obese patients, which is also and indirect factor Xa inhibitor through antithrombin, is associated with unreliable factor Xa inhibition and requires continuous monitoring of factor Xa levels [25]. Second, dabigatran has the highest volume of distribution (60–70 l) [26], compared to rivaroxaban (50 l) and apixaban (21 l) [27, 28]. Dabigatran also is metabolized through hepatic glucuronidation, while rivaroxaban and apixaban are metabolized through the cytochrome P450 system [26–28]. Obesity was shown to significantly affect volume of distribution of different drugs and to affect cytochrome P450 activity [29]. These differences in pharmacokinetics and pharmacodynamics, in addition to the findings of our study, suggest that subtle differences exist among DOACs, but they all appear safer options than warfarin in morbidly obese patients with AF.
To our knowledge, our study is the first to date to report extensive comparative safety and effectiveness analysis between three different DOACs and warfarin in a large sample (> 20,000) of patients. Furthermore, in our analysis, we adjusted for numerous important variables including important medications use and laboratory measurements as GFR. However, our study has several limitations. First, we lack information on INR levels of warfarin users and time in therapeutic range (TTR). Second, we do not have access to outcomes of admissions of VA patients to health care facilities outside of the VA. However, given the large sample size, the relative risk of admission outside the VA should not differ by drug type. Third, there is still possibility of residual confounding from unmeasured comorbidities and other confounders and off-label dosing of DOACs. Fourth, our study included only veterans, so findings might not be generalizable to other populations. Last, our study included small proportion of females.
In conclusion, dabigatran, apixaban, and rivaroxaban are as effective as warfarin in preventing stroke in severely obese patients with AF. Furthermore, these drugs offer better safety with lower bleeding and hemorrhagic stroke rates compared with warfarin.
Supplementary Material
Footnotes
Supplementary Information The online version contains supplementary material available at https://doi.org/10.1007/s10557-020-07126-2.
Compliance with Ethical Standards
Conflict of Interest The authors declare that they have no conflict of interest.
Ethical Approval This article does not contain any studies with human participants or animals performed by any of the authors.
Informed Consent Informed consent was not required as the data were de-identified and the study was deemed IRB-exempt by the University of Iowa IRB.
References
- 1.Friberg L, Rosenqvist M, Lip GY. Net clinical benefit of warfarin in patients with atrial fibrillation: a report from the Swedish atrial fibrillation cohort study. Circulation. 2012;125:2298–307. [DOI] [PubMed] [Google Scholar]
- 2.Connolly SJ, Ezekowitz MD, Yusuf S, Eikelboom J, Oldgren J, Parekh A, et al. Dabigatran versus warfarin in patients with atrial fibrillation. N Engl J Med. 2009;361:1139–51. [DOI] [PubMed] [Google Scholar]
- 3.Patel MR, Mahaffey KW, Garg J, Pan G, Singer DE, Hacke W, et al. Rivaroxaban versus warfarin in nonvalvular atrial fibrillation. N Engl J Med. 2011;365:883–91. [DOI] [PubMed] [Google Scholar]
- 4.Granger CB, Alexander JH, McMurray JJ, Lopes RD, Hylek EM, Hanna M, et al. Apixaban versus warfarin in patients with atrial fibrillation. N Engl J Med. 2011;365:981–92. [DOI] [PubMed] [Google Scholar]
- 5.Giugliano RP, Ruff CT, Braunwald E, Murphy SA, Wiviott SD, Halperin JL, et al. Edoxaban versus warfarin in patients with atrial fibrillation. N Engl J Med. 2013;369:2093–104. [DOI] [PubMed] [Google Scholar]
- 6.Wolf PA, Abbott RD, Kannel WB. Atrial fibrillation as an independent risk factor for stroke: the Framingham study. Stroke. 1991;22: 983–8. [DOI] [PubMed] [Google Scholar]
- 7.Mahajan R, Lau DH, Brooks AG, Shipp NJ, Manavis J, Wood JP, et al. Electrophysiological, electroanatomical, and structural remodeling of the atria as consequences of sustained obesity. J Am Coll Cardiol. 2015;66(1):1–11. [DOI] [PubMed] [Google Scholar]
- 8.Feng T, Vegard M, Strand LB, Laugsand LE, Mørkedal B, Aune D, et al. Weight and weight change and risk of atrial fibrillation: the HUNT study. Eur Heart J. 2019;40(34):2859–72. [DOI] [PubMed] [Google Scholar]
- 9.Wang TJ, Parise H, Levy D, D’Agostino RB Sr, Wolf PA, Vasan RS, et al. Obesity and the risk of new-onset atrial fibrillation. JAMA. 2004;292(20):2471–9. [DOI] [PubMed] [Google Scholar]
- 10.Huxley RR, Lopez FL, Folsom AR, Agarwal SK, Loehr LR, Soliman EZ, et al. Absolute and attributable risks of atrial fibrillation in relation to optimal and borderline risk factors: the Atherosclerosis Risk in Communities (ARIC) study. Circulation. 2011;123(14):1501–14. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Tellor KB, Nguyen SN, Bultas AC, Armbruster AL, Greenwald NA, Yancey AM. Evaluation of the impact of body mass index on warfarin requirements in hospitalized patients. Ther Adv Cardiovasc Dis. 2018;12(8):207–16. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Martin K, Beyer-Westendorf J, Davidson BL, Huisman MV, Sandset PM, Moll S. Use of the direct oral anticoagulants in obese patients: guidance from the SSC of the ISTH. J Thromb Haemost. 2016;14(6):1308–13. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Kaplan RM, Diaz CL, Strzelczyk T, You C, Saour B, Fine M, et al. Outcomes with novel Oral anticoagulants in obese patients who underwent electrical cardioversion for atrial tachyarrhythmias. Am J Cardiol. 2018;122(7):1175–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Kubitza D, Becka M, Zuehlsdorf M, Mueck W. Body weight has limited influence on the safety, tolerability, pharmacokinetics, or pharmacodynamics of rivaroxaban (BAY 59–7939) in healthy subjects. J Clin Pharmacol. 2007;47(2):218–26. [DOI] [PubMed] [Google Scholar]
- 15.Rosenbaum PR. Model-based direct adjustment. J Am Stat Assoc. 1987;82:387–94. [Google Scholar]
- 16.January CT, Wann LS, Calkins H, Chen LY, Cigarroa JE, Cleveland JC Jr, et al. 2019 AHA/ACC/HRS Focused Update of the 2014 AHA/ACC/HRS guideline for the management of patients with atrial fibrillation: a report of the American College of Cardiology/American Heart Association Task Force on Clinical Practice Guidelines and the Heart Rhythm Society. J Am Coll Cardiol. 2019;74:104–32. [DOI] [PubMed] [Google Scholar]
- 17.Kido K, Ngorsuraches S. Comparing the efficacy and safety of direct oral anticoagulants with warfarin in the morbidly obese population with atrial fibrillation. Ann Pharmacother. 2019;53:165–70. [DOI] [PubMed] [Google Scholar]
- 18.Peterson ED, Ashton V, Chen YW, Wu B, Spyropoulos AC. Comparative effectiveness, safety, and costs of rivaroxaban and warfarin among morbidly obese patients with atrial fibrillation. Am Heart J. 2019;212:113–9. [DOI] [PubMed] [Google Scholar]
- 19.Kushnir M, Choi Y, Eisenberg R, Rao D, Tolu S, Gao J, et al. Efficacy and safety of direct oral factor Xa inhibitors compared with warfarin in patients with morbid obesity: a single-centre, retrospective analysis of chart data. Lancet Haematol. 2019;6:e359–65. [DOI] [PubMed] [Google Scholar]
- 20.Hales CM, Carroll MD, Fryar CD, Ogden CL. Prevalence of obesity among adults and youth: United States, 2015–2016. NCHS Data Brief. 2017:1–8. [PubMed] [Google Scholar]
- 21.Patel JP, Roberts LN, Arya R. Anticoagulating obese patients in the modern era. Br J Haematol. 2011;155:137–49. [DOI] [PubMed] [Google Scholar]
- 22.Upreti VV, Wang J, Barrett YC, Byon W, Boyd RA, Pursley J, et al. Effect of extremes of body weight on the pharmacokinetics, pharmacodynamics, safety and tolerability of apixaban in healthy subjects. Br J Clin Pharmacol. 2013;76:908–16. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Reilly PA, Lehr T, Haertter S, Connolly SJ, Yusuf S, Eikelboom JW, et al. The effect of dabigatran plasma concentrations and patient characteristics on the frequency of ischemic stroke and major bleeding in atrial fibrillation patients: the RE-LY trial (Randomized Evaluation of Long-Term Anticoagulation Therapy). J Am Coll Cardiol. 2014;63:321–8. [DOI] [PubMed] [Google Scholar]
- 24.Briasoulis A, Gao Y, Inampudi C, Alvarez P, Asleh R, Chrischilles E, et al. Characteristics and outcomes in patients with atrial fibrillation receiving direct oral anticoagulants in off-label doses. BMC Cardiovasc Disord. 2020;20(1):42. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Tahaineh L, Edaily SM, Gharaibeh SF. Anti-factor Xa levels in obese patients receiving enoxaparin for treatment and prophylaxis indications. Clin Pharmacol. 2018;10:63–70. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Stangier J, Clemens A. Pharmacology, pharmacokinetics, and pharmacodynamics of dabigatran etexilate, an oral direct thrombin inhibitor. Clin Appl Thromb Hemost. 2009;15(Suppl 1):9S–16S. [DOI] [PubMed] [Google Scholar]
- 27.Piccini JP, Patel MR, Mahaffey KW, Fox KA, Califf RM. Rivaroxaban, an oral direct factor Xa inhibitor. Expert Opin Investig Drugs. 2008;17:925–37. [DOI] [PubMed] [Google Scholar]
- 28.Raghavan N, Frost CE, Yu Z, He K, Zhang H, Humphreys WG, et al. Apixaban metabolism and pharmacokinetics after oral administration to humans. Drug Metab Dispos. 2009;37:74–81. [DOI] [PubMed] [Google Scholar]
- 29.Hanley MJ, Abernethy DR, Greenblatt DJ. Effect of obesity on the pharmacokinetics of drugs in humans. Clin Pharmacokinet. 2010;49:71–87. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.