Patients Not Eligible for Warfarin Use

Three studies have specifically looked at effectiveness of therapy in patients who were considered unsuitable for warfarin therapy.^1–3 The ACTIVE-A trial¹ was designed to determine whether the combination of clopidogrel (75mg daily) plus aspirin (75 to 100mg daily) was better than aspirin alone for prevention of stroke and cardiovascular events (non-CNS embolism, MI, or vascular death) in patients with AF and at least one additional risk factor for vascular events who were considered unsuitable for warfarin therapy. A total of 7,554 patients were enrolled in a double-blind fashion from 580 centers in 33 countries, and the median followup was 3.6 years. In the ITT analyses, the combination of clopidogrel plus aspirin compared with aspirin alone significantly reduced the primary outcome by 11 percent, primarily due to a 28 percent reduction in stroke (ischemic or unknown origin) (RR 0.72; 95% CI 0.62 to 0.83; p<0.001). MI occurred in 90 patients in the clopidogrel group (0.7% per year) and in 115 in the placebo group (0.9% per year; RR 0.78; 95% CI 0.59 to 1.03; p=0.08). Importantly, clopidogrel plus aspirin compared with aspirin alone significantly increased the rate of major bleeding, including intracranial and extracranial bleeding, from 1.3 percent to 2.0 percent per year (RR 1.57; 95% CI 1.29 to 1.92; p<0.001). The rates of bleeding in the clopidogrel plus aspirin group were very similar to those observed in the warfarin arm from the ACTIVE-W study. One should also keep in mind that among the reasons for enrolling in this trial, 50 percent of the time this was due to physician assessment that the patient was inappropriate for warfarin and therefore could be in the study, which is a subjective decision. On the other hand, it is known that this subjective decision from physicians is common in clinical practice, and the results of this trial might be applicable to daily practice. In summary, if we treat 1,000 AF patients that “cannot be put on warfarin” during 3 years, clopidogrel plus aspirin would prevent 28 strokes and 6 MIs, but it would cause 20 major bleeding events, 3 of them fatal. Thus, caution is warranted when considering clopidogrel plus aspirin for patients with AF for stroke prevention.

In the light of the ACTIVE-A results, another recent study deserves special attention. In patients with AF who failed, or were unsuitable for VKA treatment, apixaban (5mg orally twice daily) was compared with aspirin (81–324mg daily) in the AVERROES trial, a randomized, double-blind, and multicenter study.³ In a prespecified analysis of the AVERROES trial, results were consistent in the subgroup of patients who tried but failed VKA therapy. Of 5599 patients, 2216 (40%) had previously failed VKA treatment [main reasons: poor international normalized ratio (INR) control 42%, refusal 37%, bleeding on VKA 8%]. Compared with those expected to be unsuitable for VKA therapy, those who had previously failed were older, more often male, had higher body mass index, more likely to have moderate renal impairment and a history of stroke and less likely to have heart failure or to be medically undertreated. The effects of apixaban compared with aspirin were consistent in those who previously failed and those who were expected to be unsuitable, for both SSE (p=0.13 for interaction) and major bleeding (p=0.74 for interaction) and were also consistent among different subgroups of patients who had previously failed VKA therapy defined by reasons for unsuitability, age, sex, renal function, CHADS2 score, aspirin dose, duration, indication, and quality of INR control of prior VKA use.

A subanalysis of the AVERROES trial explored the patterns of bleeding during treatment and defined bleeding risks based on stroke risk with aspirin versus apixaban in patients with atrial fibrillation unsuitable for warfarin. The rate of a bleeding event was 3.8% per year with aspirin and 4.5% per year with apixaban (hazard ratio with apixaban, 1.18; 95% CI 0.92-1.51; P=0.19). The anatomic site of bleeding did not differ between therapies. Risk factors for bleeding common to apixaban and aspirin were use of non-study aspirin>50% of the time and a history of daily/occasional nosebleeds. The rates of both stroke and bleeding increased with higher CHADS2 scores but apixaban compared with aspirin was associated with a similar relative risk of bleeding (p=0.21 for interaction) and a reduced relative risk of stroke (p=0.37 for interaction) irrespective of CHADS2 category.

In a multicenter prospective, nonrandomized trial² the ASAP study evaluated left atrial appendage closure with the Watchman device in patients with a contraindication for oral anticoagulation. The purpose of this study was to assess the safety and efficacy of left atrial appendage (LAA) closure in nonvalvular atrial fibrillation patients ineligible for warfarin therapy. The mean CHADS score and CHADS-VASc (CHADS score plus 2 points for age ≥75 years and 1 point for vascular disease, age 65 to 74 years, or female sex) score were 2.8 ± 1.2 and 4.4 ± 1.7, respectively. History of hemorrhagic/bleeding tendencies (93%) was the most common reason for warfarin ineligibility. Mean duration of followup was 14.4 ± 8.6 months. Serious procedure- or device-related safety events occurred in 8.7% of patients (13 of 150 patients). All-cause stroke or systemic embolism occurred in 4 patients (2.3% per year): ischemic stroke in 3 patients (1.7% per year) and hemorrhagic stroke in 1 patient (0.6% per year). This ischemic stroke rate was less than that expected (7.3% per year) based on the CHADS scores of the patient cohort.

In summary, three studies, evaluating very different interventions, included patients with nonvalvular AF who were deemed unsuitable for oral anticoagulation with warfarin; these studies found that there are alternative treatments for prevention of ischemic events in this patient population. One study found that clopidogrel plus aspirin was superior to aspirin alone for stroke prevention, but was associated with a higher risk of bleeding. One study found that apixaban compared with aspirin was associated with a lower risk of stroke and no difference in risk of bleeding. One single arm study found that use of the Watchman device was associated with a lower risk of stroke compared to the risk predicted by the CHADS scores of the participants in the study.

Patients With AF and Renal Impairment

Seven substudies from five large RCTS evaluated stroke prevention treatment in patients with AF and renal impairment. One substudy⁴ of the ROCKET AF study⁵ analyzed the efficacy results using rivaroxaban compared with warfarin in patients with renal impairment. ITT analysis showed that both medications had similar results with similar rates of stroke or systemic embolism (HR 0.86; 95% CI 0.63 to 1.17). In the per-protocol population, there were 2,950 patients (20.7%) with renal impairment (creatinine clearance 30–49 mL/min) using rivaroxaban 15mg/d (n=1,434) or warfarin (n=1,462). Among those patients, the primary outcome of stroke or systemic embolism occurred in 2.32 per 100 patient-years using rivaroxaban versus 2.77 per 100 patient-years with warfarin (HR 0.84; 95% CI 0.57 to 1.23). Rates of the principal safety outcome in the safety population (major and clinically relevant non-major bleeding: 17.82 vs. 18.28 per 100 patient-years; p=0.76) and intracranial bleeding (0.71 vs. 0.88 per 100 patient-years; p=0.54) were similar with rivaroxaban or warfarin. Fatal bleeding (0.28 vs. 0.74% per 100 patient-years; p=0.047) occurred less often with rivaroxaban. This study suggested that patients with AF and moderate renal insufficiency have higher rates of stroke and bleeding than those with normal renal function. Rivaroxaban preserved the benefit of warfarin in preventing stroke and systemic embolus and produced lower rates while on treatment. Bleeding rates with the reduced dose of rivaroxaban were similar to those on warfarin therapy, and there were fewer fatal bleeds with rivaroxaban.

Another substudy⁶ of the ROCKET AF trial⁵ evaluated outcomes in patients with worsening renal function (WRF), as defined as >20% decline in creatinine clearance (CrCl) measurement at any point in the study. Dose of rivaroxaban was determined based on CrCl during the initial screening visit and despite changes in renal function over time, dose was not changed unless patient had two consecutive measurements of CrCl <25 mL/min at which point the medication was discontinued. Overall, patients treated with Rivaroxaban had similar screening CrCl compared to those randomized to warfarin (68 mL/min (IQR 53 to 87) vs. 68 mL/min (IQR 53 to 88); p=0.36). Patients randomized to warfarin had a larger decline in mean CrCl compared to those taking rivaroxaban (-4.3 vs. -3.5; p<0.0001). Compared to patients with stable renal function (SRF), there was no difference in stroke or systemic embolism among patients with worsening renal function (Adj HR 1.25; 95% CI 0.89 to 1.75; p=0.19). However, patients with worsening renal function had higher rates of all-cause mortality (HR 1.49; 95% CI 1.12 to 1.98; p=0.0067) and the composite outcome of stroke/systemic embolism/vascular death/MI (HR 1.40; 95% CI 1.13-1.73; p=0.0023). Among patients with worsening renal function, those randomized to treatment with rivaroxaban were less likely to have stroke/systemic embolism (WRF HR 0.50; 95% CI 0.27 to 0.93; SRF HR 0.97; 95% CI 0.76 to 1.24; p value for interaction 0.05), more likely to have a hemoglobin decrease (WRF HR 1.98; 95% CI 1.11 to 3.55; SRF HR 1.06; 95% CI 0.85 to 1.32; p value for interaction 0.047) and had no difference in major or NMCR bleeding (HR WRF 1.06; 95% CI 0.80 to 1.39; HR SRF 0.98; 95% CI 0.89 to 1.08; p value for interaction 0.61).

One substudy⁷ of the AVERROES trial³ compared apixaban 5mg twice daily (2.5mg twice daily in selected patients) with aspirin 81–324mg daily in 1,697 patients with stage III chronic kidney disease (CKD). Apixaban significantly reduced primary events (stroke and systemic embolism) by 68 percent (5.6% per year on aspirin vs. 1.8% per year on apixaban; HR 0.32; 95% CI 0.18 to 0.55; p<001) for stage III CKD participants and by 43 percent (2.8% per year on aspirin vs. 1.6% per year on apixaban; HR 0.57; 95% CI 0.37 to 0.87; p=.009) for patients with an estimated glomerular filtration rate (eGFR) ≥60 mL/min per 1.73m² (p value for interaction=0.10) in the ITT population. There was no significant difference in major bleeding in stage III CKD patients by treatment (2.2% per year with aspirin vs. 2.5% per year with apixaban; HR 1.20; 95% CI 0.65 to 2.1).

A substudy⁸ of the ARISTOTLE trial⁹ compared apixaban 5mg twice daily with warfarin (target INR 2·0–3·0) in different levels of GFR. According to baseline Cockcroft–Gault, there were 7,518 patients (42%) with an eGFR >80 mL/min, 7,587 (42%) with an eGFR between 50 and 80 mL/min, and 3,017 (15%) with an eGFR ≤50 mL/min. In the ITT population, rates of cardiovascular events and bleeding were higher at impaired renal function levels (eGFR ≤80 mL/min). Apixaban was more effective than warfarin in preventing stroke or systemic embolism and in reducing mortality irrespective of renal function, with no significant interaction between the treatment effect and the level of renal dysfunction. These results were consistent regardless of methods for GFR estimation, achieving statistical significance on the subgroup ≤50 mL/min by Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) (all-cause mortality and stroke/systemic embolism), subgroup Cockcroft–Gault 50-80 mL/min (stroke/systemic embolism), and subgroup cystatin C >80 mL/min (stroke/systemic embolism). Apixaban was associated with fewer major bleeding events across all ranges of eGFRs. The relative risk reduction in major bleeding was greater in patients with an eGFR ≤50 mL/min using Cockcroft–Gault (HR 0.50; 95% CI 0.38 to 0.66; p value for interaction=0.005) or CKD-EPI equations (HR 0.48; 95% CI 0.37 to 0.64; p value for interaction=0.003]. When cystatin C was used to estimate GFR, apixaban was associated with fewer bleeding events across all ranges of eGFR, but without any significant interaction with the treatment effect on major bleeding (p value for interaction=0.54).

In sensitivity analyses, trial investigators examined whether the reduction in bleeding in patients with impaired renal function was due to the more frequent use of the lower apixaban dose (2.5mg twice daily). In both sensitivity analyses, the interaction between treatment and renal function remained statistically significant for major bleeding.

Another substudy¹⁰ of the ARISTOTLE trial⁹ evaluated outcomes related to change in renal function over time in patients treated with 5mg apixaban twice daily compared to warfarin. In patients with worsening renal function over 12 months of followup, apixaban showed numerically lower relative risk of stroke or systemic embolism (HR 0.80; 95% CI 0.51 to 1.24; p=0.86) as well as major bleeding (HR 0.76; 95% CI 0.54 to 1.07; p=0.73) compared to warfarin, although neither reached statistical significance. These results were similar across levels of renal dysfunction, defined as eGFR >80 mL/min, eGFR 50-80 mL/min and eGFR <50 mL/min.

In the ENGAGE AF study,¹¹ patients randomized to the high dose edoxaban arm received 60mg daily if their CrCl was over 50 ml/min or 30mg daily if their CrCl was between 30mg/min and 50mg/min. In a substudy,¹² no statistically significant interaction was found between treatment (edoxaban vs. warfarin) and CrCl (30-50 ml/min vs. >50 ml/min) on the primary efficacy outcome of stroke or systemic embolic event (p = 0.94 for interaction). In both renal function groups, there was no statistically significant difference between edoxaban and warfarin (HR 0.87; 95% CI 0.65 to 1.18 for CrCl >50ml/min and HR 0.87; 95% CI 0.72 to 1.04 for CrCl 30-50ml/min). There was also no statistically significant interaction between treatment and CrCl on major bleeding (p=0.62 for interaction). In exploratory analyses, there was no statistically significant interaction between CrCl subgroups (30-50 ml/min, >50-95 ml/min, and >95ml/min) and treatment on stroke or systemic embolic event, systemic embolic events, any stroke, ischemic stroke, hemorrhagic stroke, MI, any cause death, cardiovascular death, fatal bleeding, intracranial hemorrhage, or minor bleeding. There was, however, a statistically significant interaction on GI bleeding (p=0.02 for interaction) in which patients with CrCl of >50-95 ml/min had a higher risk with edoxaban vs. warfarin (HR 1.47; 95% CI 1.15 to 1.87) than the other two CrCl subgroups (HR 1.17; 95% CI 0.78 to 1.76 for CrCl 30-50ml/min and HR 0.67; 95% CI 0.40 to 1.10 for CrCl >95ml/min).

A prespecified study of the RE-LY trial¹³ investigated the outcomes of the trial in relation to renal function. Glomerular filtration rate was estimated with the Cockcroft-Gault, Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI), and Modification of Diet in Renal Disease (MDRD) equations in all randomized patients with available creatinine at baseline (n=17 951), and cystatin C-based glomerular filtration rate was estimated in a subpopulation with measurements available (n=6190). A glomerular filtration rate ≥80, 50 to <80, and <50mL/min was estimated in 32.6%, 47.6%, and 19.8% and in 21.6%, 59.6%, and 18.8% of patients based on Cockcroft-Gault and CKD-EPI, respectively. Rates of stroke or systemic embolism, major bleeding, and all-cause mortality increased as renal function decreased. The rates of stroke or systemic embolism were lower with dabigatran 150mg and similar with 110mg twice daily compared with warfarin, without significant heterogeneity in subgroups defined by renal function (interaction P>0.1 for all). For the outcome of major bleeding, there were significant interactions between treatment and renal function according to CKD-EPI and MDRD equations, respectively (P<0.05). The relative reduction in major bleeding with either dabigatran dose compared with warfarin was greater in patients with glomerular filtration rate ≥80 mL/min.

In summary, sub-studies of 5 large RCTs evaluated the effects of DOACs compared to either warfarin or aspirin in patients with some degree of renal disease. These studies demonstrated that compared to participants with normal renal function, participants with renal disease had increased risk of ischemic events, bleeding, and all-cause mortality. In all 5 sub-studies, among participants with renal disease, use of the DOACs were consistently similar to or better than warfarin in the prevention of stroke/SE and bleeding events. One sub-study demonstrated that in patients with stage 3 CKD, compared to aspirin, apixaban was associated with lower risk of stroke and no difference in bleeding.

Patients With Paroxysmal Versus Sustained AF

One substudy¹⁴ of the ACTIVE W RCT¹⁵ analyzed the results in patients with paroxysmal AF (n=1,202) as compared with those who had sustained (persistent or permanent) AF (n=5,495). Patients with paroxysmal AF were younger, had a shorter AF history, more hypertension, and less valvular disease, heart failure, and diabetes mellitus than patients with sustained AF. Irrespective of type of AF, the incidence of stroke and non-CNS embolism was lower for patients treated with oral anticoagulation. There were more bleedings of any type in patients receiving clopidogrel plus aspirin, irrespective of the type of AF, but major bleeding events were similar in all groups (paroxysmal vs. sustained, and oral anticoagulants vs. clopidogrel+aspirin).

A secondary analysis¹⁶ of the ARISTOTLE trial⁹ evaluated treatment with apixaban 5mg twice daily compared to warfarin in patients with paroxysmal or persistent AF. Overall, patients with paroxysmal atrial fibrillation were less likely to have stroke or systemic embolism (HR 0.65; 95% CI 0.48 to 0.87; p=0.003) and all-cause mortality was also significantly less (HR 0.72; 95% CI 0.61 to 0.85; p=0.0002). There was no significant interaction with regard to stroke or systemic embolism by type of AF and treatment type (HR Paroxysmal 0.72; 95% CI 0.41 to 1.25; HR Persistent 0.80; 95% CI 0.66 to 0.97; p value for interaction 0.71), all-cause mortality (HR Paroxysmal 0.99; 95% CI 0.72 to 1.37; HR Persistent 0.88; 95% CI 0.78 to 0.99; p value for interaction 0.50) and major bleeding (HR Paroxysmal 0.73; 95% CI 0.49 to 1.08; HR Persistent 0.68; 95% CI 0.59 to 0.80; p value for interaction 0.75) in patients treated with apixaban compared with warfarin.

In summary, analysis of two large RCTs evaluated for differences in treatment effects (clopidogrel plus aspirin vs warfarin or apixaban vs warfarin) for stroke prevention/bleeding by type of AF (paroxysmal or persistent). In neither study was there a difference in treatment effect by type of AF.

Patients With Recently Diagnosed AF

One substudy¹⁷ of the ARISTOTLE RCT⁹ evaluated patients with AF first diagnosed within 30 days prior to randomization. Regardless of timing of diagnosis, apixaban had similar benefits on prevention of stroke or systemic embolism and major bleeding compared to warfarin (interaction p values 0.94 and 0.78 respectively).

Patients With AF After Stroke

Eight studies explored stroke prevention treatment in patients with AF who had previously suffered a stroke.^18–25

The Heparin in Acute Embolic Stroke Trial (HAEST)²⁰ was a multicenter RCT on the effect of LMWH (dalteparin 100 IU/kg subcutaneously twice a day) or aspirin (160mg every day) for the treatment of 449 patients with acute ischemic stroke and AF. The primary aim was to test whether treatment with LMWH, started within 30 hours of stroke onset, is superior to aspirin for the prevention of recurrent stroke during the first 14 days. The frequency of recurrent ischemic stroke during the first 14 days was 19/244 (8·5%) in dalteparin-allocated patients versus 17/225 (7·5%) in aspirin-allocated patients (OR 1·13; 95% CI 0·57 to 2·24). In the ITT analyses, the OR remained unchanged after adjusting for sex in logistic-regression analysis (1·19; 95% CI 0·60 to 2·36). The secondary events during the first 14 days also revealed no benefit of dalteparin compared with aspirin. There were no significant differences in functional outcome or death at 14 days or 3 months.

A prespecified subgroup analysis²¹ of the ROCKET AF study⁵ investigated whether the efficacy and safety of rivaroxaban compared with warfarin was consistent among patients with and without previous stroke or TIA. A total of 14,264 patients from 1,178 centers in 45 countries were included. Patients with AF who were at increased risk of stroke (CHADS₂ score >2) were randomly assigned (1:1) in a double-blind manner to rivaroxaban 20mg daily or adjusted dose warfarin (to maintain INR 2.0–3.0). Patients and investigators were masked to treatment allocation. The primary outcome was the composite of stroke or non-CNS systemic embolism as a safety outcome. The treatment effects of rivaroxaban and warfarin were compared among patients with and without previous stroke or TIA. The safety analyses were done in the on-treatment population. Efficacy analyses were analyzed by ITT, and 7,468 (52%) patients had a previous stroke (n=4,907) or TIA (n=2,561). The number of events per 100 person-years for the primary outcome in patients treated with rivaroxaban compared with warfarin was consistent among patients with previous stroke or TIA (2.79% rivaroxaban vs. 2.96% warfarin; HR 0.94; 95% CI 0.77 to 1.16) and those without (1.44% vs. 1.88%; HR 0.77; 95% CI 0.58 to 1.01; comparison interaction p=0.23). Similarly, the number of major and non-major clinically relevant bleeding events per 100 person-years in patients treated with rivaroxaban compared with warfarin was consistent among patients with previous stroke or TIA (13.31% rivaroxaban vs. 13.87% warfarin; HR 0.96; 95% CI 0.87 to 1.07) and those without (16.69% vs. 15.19%; HR 1.10; 95% CI 0.99 to 1.21; comparison interaction p=0.08).

One observational study¹⁸ followed a consecutive series of AF patients with first-ever ischemic stroke and evaluated prospectively those with moderate to severe disability (grade 4–5 on the modified Rankin Scale) who were treated during a 5-year followup period with either warfarin or aspirin. Death and recurrent vascular events were documented. Out of a pool of 438 AF patients, 191 were prospectively assessed. During a mean followup of 50.4 months, the cumulative 5-year mortality was 76.7% (95% CI 69.0 to 84.3), and the 5-year recurrence rate was 33.7% (95% CI 23.3 to 44.1). Additionally, two non-cerebral major bleeding events requiring hospital admission and blood transfusion were recorded in the warfarin group. Only one non-cerebral bleeding event was documented in the aspirin group. The annual event rates for all major bleeding complications in aspirin and warfarin groups were 0.7 and 3.3 percent, respectively. Aspirin versus warfarin was an independent predictor of mortality. Prior TIA and aspirin versus warfarin were predictors of vascular recurrence. Anticoagulation was associated with a decreased risk of death (HR 0.44; 95% CI 0.27 to 0.70; p<0.001) and recurrent thromboembolism (HR 0.36; 95% CI 0.17 to 0.77; p<0.01). The results of this observational study suggest that chronic anticoagulation therapy may be effective in lengthening survival and preventing recurrent thromboembolism in AF patients who have suffered a severely disabling ischemic stroke.

An observational study¹⁹ analyzed recurrent cerebral and non-cerebral ischemic vascular events, major intracerebral and extracerebral bleeding, and vascular death in 401 consecutive patients with ischemic stroke or TIA and AF who were discharged with oral anticoagulation, antiplatelet agents, or heparin only in a clinical routine setting. Patients on oral anticoagulation at time of discharge were significantly younger and had suffered a major stroke less often than patients who received antiplatelet agents or heparin at discharge. One year after discharge, adherence to therapy was higher in patients discharged on oral anticoagulation (72%) than in those on antiplatelet agents (46%; p<0.001). The majority of patients discharged on heparin were subsequently treated with oral anticoagulation. During a median followup of 25 months (IQR, 15–38), 103 (26%) patients experienced a complication: 91 (88%) patients an ischemic complication and 12 (12%) a bleeding complication. The rate of ischemic complications and the overall rate of complications were lowest in patients discharged on oral anticoagulation. Patients on antiplatelet agents at discharge suffered from ischemic complications significantly more often during the followup period than patients on oral anticoagulation or heparin at discharge (30% vs. 16% vs. 23%; p=0.031). Patients on antiplatelet agents suffered their first vascular complication significantly sooner after discharge than patients on oral anticoagulation. Safety outcomes showed that three percent of the patients on antiplatelet agents and four percent of those on oral anticoagulation suffered from major bleeding complications during followup (p=0.028). The rate of intracranial bleeding was higher in patients on oral anticoagulation (3% vs. 1%), but the total numbers were too small to allow a valid statistical comparison. Total mortality was lowest in patients discharged on oral anticoagulation, and vascular mortality also seemed somewhat lower in this group but the difference was not significant.

A predefined analysis²² was conducted of the outcomes of the RE-LY trial²⁶ in subgroups of patients with or without previous stroke or transient ischemic attack. The primary efficacy outcome was stroke or systemic embolism, and the primary safety outcome was major hemorrhage. Within the subgroup of patients with previous stroke or TIA, 1,195 patients were from the 110mg dabigatran group, 1,233 from the 150mg dabigatran group, and 1,195 from the warfarin group. Stroke or systemic embolism occurred in 65 patients (2.78% per year) on warfarin compared with 55 (2.32% per year) on 110mg dabigatran (relative risk [RR] 0.84; 95% CI 0.58 to 1.20) and 51 (2.07% per year) on 150mg dabigatran (RR 0.75, 95% CI 0.52 to 1.08). The rate of major bleeding was significantly lower in patients on 110mg dabigatran (RR 0.66; 95% CI 0.48 to 0.90) and similar in those on 150mg dabigatran (RR 1.01; 95% CI 0.77 to 1.34) compared with those on warfarin. The effects of both doses of dabigatran compared with warfarin were not significantly different between patients with previous stroke or TIA and those without for any of the outcomes from RE-LY apart from vascular death (110mg group compared with warfarin group, interaction p=0.038). By these results, the effects of 110mg dabigatran and 150mg dabigatran twice daily in patients with previous stroke or TIA are consistent with those of other patients in RE-LY, for whom, compared with warfarin, 150mg dabigatran reduced stroke or systemic embolism and 110mg dabigatran was noninferior.

A prespecified subgroup analysis²³ of AVERROES³ included 5,599 patients (mean age 70 years) with AF who were at increased risk of stroke and unsuitable for warfarin therapy. These patients were randomly assigned to receive apixaban 5mg twice daily (n=2,808) or aspirin 81–324mg per day (n=2,791). The primary efficacy outcome was stroke or systemic embolism in the ITT population; the primary safety outcome was major bleeding. In this subanalysis of patients with previous stroke or TIA, the effects of apixaban in patients with and without previous stroke or TIA were compared. The cumulative HR for stroke or systemic embolism at 1 year was 5.73% (95% CI 4.10 to 8.02) in patients with previous stroke or TIA and 2.36% (1.93 to 2.89) in those without. In patients with previous stroke or TIA treated with apixaban, the rates of stroke or systemic embolism, ischemic stroke, and disabling or fatal stroke were consistently lower than those in patients treated with aspirin. In patients with previous stroke or TIA, 10 events of stroke or systemic embolism occurred in the apixaban group (n=390), cumulative hazard 2.39% per year) compared with 33 in the aspirin group (n=374). This resulted in a cumulative hazard of 2.39 percent in the apixaban group and 9.16 percent per year in the aspirin group (HR 0.29; 95% CI 0.15 to 0.60). In those without previous stroke or TIA, 41 events (n=2,417, 1.68% per year) and 80 events (n=2,415, 3.06% per year) occurred in the apixaban and aspirin groups, respectively (HR 0.51; 95% CI 0.35 to 0.74). Compared with those treated with aspirin, the 1-year risk of stroke or systemic embolism decreased by 73 percent in patients treated with apixaban and with previous stroke or TIA (1-year absolute risk reduction of 6.4%; 95% CI 2.8 to 10.0) and by 45 percent in patients treated with apixaban and without previous stroke or TIA (1-year absolute risk reduction of 1.4%, 95% CI 0.4 to 2.3). The p values for interaction between history of previous stroke or TIA and treatment were not significant, indicating that the results in the subgroups were consistent with the overall result of the study. Major bleeding, the primary safety outcome, was more frequent in patients with history of previous stroke or TIA than in patients without this history (HR 2.88; 95% CI 1.77 to 4.55), but risk of this event did not differ between treatment groups. The effect of apixaban versus aspirin for bleeding complications was consistent in the two subgroups, with nonsignificant interaction p values.

A prespecified subgroup analysis²⁴ from the ARISTOTLE trial⁹ evaluated the efficacy and safety of apixaban compared with warfarin in subgroups of patients with and without previous stroke or TIA. The primary efficacy outcome was stroke or systemic embolism, analyzed by intention to treat. The primary safety outcome was major bleeding in the on-treatment population. Outcomes in patients with and without previous stroke or TIA were compared. Of the trial population, 3,436 (19%) had a previous stroke or TIA. In the subgroup of patients with previous stroke or TIA, the rate of stroke or systemic embolism was 2.46 per 100 patient-years of followup in the apixaban group and 3.24 in the warfarin group (HR 0.76; 95% CI 0.56 to 1.03); in the subgroup of patients without previous stroke or TIA, the rate of stroke or systemic embolism was 1.01 per 100 patient-years of followup with apixaban and 1.23 with warfarin (HR 0.82; 95% CI 0.65 to 1.03). The relative risk reduction of stroke or systemic embolism with apixaban versus warfarin was similar among patients with and those without previous stroke or TIA (p for interaction=0.71). The reduction in rates of cardiovascular death, disabling or fatal stroke, and all-cause mortality with apixaban versus warfarin was similar in patients with and without previous stroke or TIA (p for interaction=0.53, 0.18, and 0.89, respectively). Compared with patients without previous stroke or TIA, patients with previous stroke or TIA were more likely to have major bleeding (HR 1.37; 95% CI 1.17 to 1.62) and intracranial bleeding (2.15, 95% CI 1.57 to 2.96). The relative risk reductions in major bleeding and total bleeding with apixaban versus warfarin were similar in both groups (p for interaction=0.69 and 0·.0, respectively). Intracranial bleeding was reduced in the apixaban groups from 1.49 per 100 patient-years of followup on warfarin to 0.55 per 100 patient-years on apixaban in those with previous stroke or TIA (HR 0.37; 95% CI 0.21 to 0.67) and from 0.65 per 100 patient-years of followup on warfarin to 0.29 per 100 patient-years on apixaban in those without previous stroke or TIA (0.44, 95% CI 0.30 to 0.66). Based on these results, the effects of apixaban versus warfarin were consistent in patients with AF with and without previous stroke or TIA.

In a substudy of the ENGAGE AF study²⁵, in which with prior ischemic stroke or TIA were compared with patients without prior ischemic stroke or TIA, no statistically significant interaction was found between prior stroke/TIA and treatment (high dose edoxaban vs. warfarin) for stroke or systemic embolic event, any stroke, hemorrhagic stroke, ischemic stroke, any cause death, or cardiovascular death.

Studies were inconsistent in terms of the interventions evaluated and their findings. Three studies compared anticoagulation to aspirin therapy.^18,20,23. Anticoagulation with either apixaban or warfarin was superior to aspirin therapy in preventing recurrent thromboembolism.^18,23 Four studies compared direct oral anticoagulants to warfarin therapy.^21,22,24,25 These studies demonstrated that there was no difference in risk of stroke or systemic embolism when comparing direct oral anticoagulants (edoxaban, rivaroxaban, apixaban, dabigatran 110mg BID) to warfarin therapy. The only exception was the dabigatran 150mg BID dose showed reduced risk of stroke or systemic embolism compared to warfarin therapy.

Patients With AF and Different Thromboembolic Risks

Six studies explored the comparative safety and effectiveness of stroke prevention therapy in patients with different thromboembolic risks.^15,27–31

An observational study²⁷ sought to determine the efficacy and safety of warfarin and aspirin in patients with nonvalvular AF, with separate analyses according to predicted thromboembolic and bleeding risk. Nationwide registries allowed the identification of all patients discharged with nonvalvular AF in Denmark (n=132,372). For every patient, the risk of stroke and bleeding was calculated by CHADS₂, CHA₂DS₂-VASc, and HAS-BLED. In different groups according to thromboembolic risks, warfarin consistently lowered the risk of thromboembolism compared with aspirin; the combination of warfarin+aspirin did not yield any additional benefit. In patients at high thromboembolic risk, HRs (95% CIs) for thromboembolism were (adjusted for all baseline characteristics): CHA₂DS₂-VASc ≥2: HR 1.81 (1.73 to1.90), 1.14 (1.06 to 1.23) for aspirin and warfarin+aspirin, respectively, compared with warfarin; CHADS₂ ≥2: HR 1.73 (1.64 to 1.83), 1.05 (0.96 to 1.15), for aspirin and warfarin+aspirin, respectively, compared with warfarin. The risk of bleeding was increased with warfarin, aspirin, and warfarin+aspirin compared with no treatment; the HRs were 1.0 (warfarin; reference), 0.93 (aspirin; 0.89–0.97), 1.64 (warfarin+aspirin; 1.55–1.74), and 0.84 (no treatment; 0.81–0.88), respectively. This large cohort study corroborates the effectiveness of warfarin and no effect of aspirin treatment on the risk of stroke/thromboembolism. Also, the risk of bleeding was increased with both warfarin and aspirin treatment, but the net clinical benefit was clearly positive, in favor of warfarin in patients with increased risk of stroke/thromboembolism.

A prospective cohort study²⁸ analyzed the effectiveness and safety of oral anticoagulants in 796 outpatients with nonvalvular AF in daily clinical practice, according to embolic risk evaluated by means of CHADS₂ score. Oral anticoagulation was prescribed to 564 (71%) patients. After 2.4 ± 1.9 years of followup, the embolic event (TIA, ischemic stroke, peripheral embolism) rates (per 100 patient-years) for each stratum of the CHADS₂ score for patients with/without oral anticoagulants were: 1/4.1; p=0.23 (CHADS₂=0); 0.6/7.1; p=0.0018 (CHADS₂=1); 0.5/5.1; p=0.0014 (CHADS₂=2); 2.4/12.5; p=0.0017 (CHADS₂=3) and 2.9/20; p=0.013 (CHADS₂≥4). The severe bleeding rates for the same CHADS₂ score strata were 3/0.8, 0.8/0.7, 1.3/0.7, 0.4/0, and 2.9/5 in patients with/without oral anticoagulants (nonsignificant.). This study demonstrated that oral anticoagulants appeared safe and effective in patients with CHADS₂≥1.

In ACTIVE W,¹⁵ oral anticoagulation was more efficacious than combined clopidogrel plus aspirin in preventing vascular events in patients with AF. A subanalysis of ACTIVE W³² evaluated the findings according to risk stratification using the CHADS₂ score. Treatment-specific rates of stroke and major bleeding were calculated for patients with a CHADS₂=1 and compared with those with a CHADS₂ >1. The ACTIVE W primary outcome (stroke, noncentral nervous system systemic embolism, all-cause mortality, and MI) occurred more frequently in patients on clopidogrel+aspirin, both with CHADS₂=1 (3.28% per year versus 1.92% per year, RR=1.72; p=0.01) and with CHADS₂ >1 (7.14% per year versus 5.18% per year, RR 1.40; p=0.0035). CHADS₂ status did not significantly affect the relative benefit of oral anticoagulants for this outcome (P for interaction=0.41). Observed stroke rates for those with a CHADS₂=1 were 1.25 percent per year on clopidogrel+aspirin and 0.43 percent per year on oral anticoagulants (RR 2.96; 95% CI 1.26 to 6.98; p=0.01). Among patients with a CHADS₂>1, the stroke rates were 3.15 percent per year on clopidogrel+aspirin and 2.01 percent per year on oral anticoagulants (RR 1.58; 95% CI 1.11 to 2.24; p=0.01; p for interaction between stroke risk category and efficacy of oral anticoagulants=0.19). The risk of major bleeding during oral anticoagulants was significantly lower among patients with CHADS₂=1 (1.36% per year) compared with CHADS₂>1 (2.75% per year) (RR 0.49; 95% CI 0.30 to 0.79; p=0.003). For patients with CHADS₂=1, the rate of major bleeding was 2.09 percent per year on clopidogrel+aspirin, which was higher than the rate of 1.36 percent per year on oral anticoagulants (RR 1.55; 95% CI 0.91 to 2.64; p=0.11). For patients with CHADS₂>1, major bleeding occurred at a rate of 2.63 percent per year on clopidogrel+aspirin and 2.75 percent per year on oral anticoagulants (RR 0.97; 95% CI 0.69 to 1.35; p=0.84). The relative risk of major bleeding with clopidogrel+aspirin, compared with oral anticoagulants was not significantly different between patients with high and low CHADS₂ scores (p for interaction=0.15); however, the absolute risk of major bleeding on oral anticoagulants was significantly lower among patients with CHADS₂=1 compared with CHADS₂>1 (RR=0.49; 95% CI 0.30 to 0.79; p=0.0003). Based on these results, patients with a CHADS₂=1 had a low risk of stroke, yet still derived a modest (<1% per year) but statistically significant absolute reduction in stroke with oral anticoagulants compared with clopidogrel+aspirin and had low rates of major hemorrhage on oral anticoagulants.

A subgroup analysis²⁹ of the RE-LY trial²⁶ evaluated the prognostic importance of CHADS₂ risk score in patients with AF receiving oral anticoagulants, including warfarin and the direct thrombin inhibitor dabigatran. Of the18,112 patients, the distribution of CHADS₂ scores were as follows: 0–1, 5,775 patients; 2, 6,455 patients; and 3–6, 5,882 patients. Annual rates of the primary outcome of stroke or systemic embolism among all participants were 0.93, 1.22, and 2.24 percent in patients with a CHADS₂ score of 0–1, 2, and 3–6 respectively. Annual rates of other outcomes among all participants with CHADS₂ scores of 0–1, 2, and 3–6, respectively, were 2.26, 3.11, and 4.42 percent (major bleeding); 0.31, 0.40, and 0.61 percent (intracranial bleeding); and 1.35, 2.39, and 3.68 percent (vascular mortality) (p <0.001 for all comparisons). Rates of stroke or systemic embolism, major and intracranial bleeding, and vascular and total mortality each increased in the warfarin and dabigatran groups with increasing CHADS₂ score. The reduction in stroke or systemic embolism with dabigatran 150mg twice daily versus warfarin was consistent across the CHADS₂ risk groups. Across CHADS₂ risk groups, the rates of stroke or systemic embolism were similar with dabigatran 110mg twice daily and warfarin. The rates of intracranial bleeding with dabigatran 150mg or 110mg twice daily were lower than those with warfarin; there was no significant heterogeneity in subgroups defined by CHADS₂ scores.

A fair-quality observational study³⁰ that included 8,962 patients with AF and a CHA₂DS₂-VASc score=0 showed that among untreated patients, the rates of stroke/thromboembolism, major bleeding, and mortality were 0.64 percent, 1.12 percent, and 1.08 percent per year, respectively. Use of oral anticoagulation and/or antiplatelet therapy was not associated with a reduction in stroke/thromboembolism (RR 0.99; 95% CI 0.25 to 3.99; p=0.99) and was not associated with a different prognosis in terms of bleeding events, improved survival, or a composite outcome of stroke/thromboembolism, bleeding, and death (RR 0.80; 95% CI 0.40 to1.61; p=0.53).

Finally, a secondary analysis³¹ of the ARISTOTLE trial⁹ compared apixaban 5mg twice daily versus warfarin (target INR 2·0–3·0) in patients with different levels of risk of stroke and of bleeding in AF, according to patients’ CHADS₂, CHA₂DS₂-VASc, and HAS-BLED scores. Irrespective of CHADS₂ score, patients assigned to apixaban had significantly lower rates of stroke or systemic embolism, mortality, International Society on Thrombosis and Haemostasis (ISTH) major bleeding, intracranial bleeding, and any bleeding than did those assigned warfarin, with no evidence of statistical heterogeneity. The benefits of apixaban compared with warfarin for all outcomes (including events during treatment only) across CHA₂DS₂-VASccategories were similar to those seen across CHADS₂ score categories. No difference was recorded for MI. Irrespective of HAS-BLED score, patients assigned to apixaban had lower rates of stroke or systemic embolism, mortality, ISTH major bleeding, Thrombolysis in Myocardial Infarction (TIMI) major or minor bleeding, Global Use of Strategies to Open Occluded Coronary Arteries (GUSTO) severe or moderate bleeding, and any bleeding, including events during treatment only, than did those assigned to warfarin. The reduction in intracranial bleeding with apixaban compared with warfarin was greater in patients with a HAS-BLED score of 3 or higher (HR 0·22; 95% CI 0·10 to 0·48) than was the reduction seen in those with a HAS-BLED score of 0–1 (HR 0·66; 95% CI 0·39 to 1·12), but not significantly so (p value for interaction=0·0604). Finally, regardless of CHADS₂, CHA₂DS₂-VASc, and HAS-BLED score, patients who received apixaban had fewer events than did patients who received warfarin, with lower rates of the composite of stroke, systemic embolism, ISTH major bleeding, and all-cause mortality.

The studies were inconsistent in terms of the comparisons evaluated and the findings. Two studies showed a decrease in risk of thromboembolism when comparing warfarin therapy to aspirin and clopidogrel regardless of calculated risk.^15,27. When comparing direct oral anticoagulants (apixaban or dabigatran) to warfarin therapy, a decrease in risk of thromboembolism was seen with direct oral anticoagulant agents.^30,31 Lastly, one study looking at only patients with CHA₂DS₂-VASc score=0 showed no different in risk of thromboembolism between those using oral anticoagulation and/or antiplatelet therapy.³⁰

Patients With AF According to INR Control

Four studies evaluated treatment safety and effectiveness according to center-based INR control.^33–36 In the first study,³³ incident ischemic strokes were evaluated in a cohort of 13,559 patients with nonvalvular AF. Of 596 ischemic strokes, 32 percent occurred during warfarin therapy, 27 percent during aspirin therapy, and 42 percent during neither type of therapy. Among patients who were taking warfarin, an INR of <2.0 at admission, as compared with an INR of ≥2.0, independently increased the odds of a severe stroke in a proportional odds logistic-regression model (OR 1.9; 95% CI 1.1 to 3.4) across three severity categories of stroke and the risk of death within 30 days (HR 3.4; 95% CI 1.1 to 10.1). The proportion of patients who had a severe or fatal stroke did not differ significantly between those with an admission INR of 1.5–1.9 and those with an INR of <1.5. After adjustment for potential confounders in the proportional odds model, the medication group remained an independent risk factor for the severity of stroke when patients who had an INR ≥2.0 were compared with those who had an INR of <2.0 or those who were taking neither aspirin nor warfarin. An INR of 1.5–1.9 at admission was associated with a mortality rate similar to that for an INR of <1.5 (18% and 15%, respectively). The 30-day mortality rate among patients who were taking aspirin at the time of the stroke was similar to that among patients who were taking warfarin and who had an INR <2.0. The rate of ischemic stroke was highest at INR values <2.0, especially values <1.5. By contrast, there was no marked absolute increase in the rate of intracranial hemorrhage at INR values <4.0. Based on these results, anticoagulation that results in an INR ≥2.0 in patients with nonvalvular AF reduces the frequency of ischemic stroke, its severity, and the risk of death from stroke.

A second observational study included an analysis of warfarin subgroups according to INR control compared with no therapy.³⁴ Ischemic stroke rate relative risk (RR) was 0.93 (95% CI 0.71 to1.22) in patients below therapeutic range (INR<2), 0.69 (0.57 to0.83) in the group within therapeutic range (INR 2–3), 0.82 (0.57 to 1.20) in patients above therapeutic range (INR >3), and 0.62 (0.56 to 0.69) in the group with unknown therapeutic range. Intracranial hemorrhage RR was 1.16 (95% CI 0.62 to 2.16) in patients below therapeutic range (INR <2), 1.13 (0.74 to 1.72) in the group within therapeutic range (INR 2–3), 3.26 (1.67 to 6.38) in patients above therapeutic range (INR >3), and 1.29 (0.98 to 1.69) in the group of unknown therapeutic range.

A post-hoc analysis³⁵ of the ARISTOTLE trial⁹ evaluated apixaban 5mg twice daily compared to warfarin treatment with differing times in therapeutic range. Overall, apixaban significantly reduced the rate of stroke or systemic embolism compared to warfarin (HR 0.79; 95% CI 0.66–0.95). The treatment benefit of apixaban was similar across the lowest and highest quartiles of individual time in therapeutic range (iTTR) without interaction between quality of INR control and frequency of events (iTTR 24.3-60.5 HR 0.70; 95% CI 0.52 to 0.94; iTTR 71.2-83.2 HR 0.87; 95% CI 0.57 to 1.33; p value for interaction 0.060). There were also similar treatment effects with regards to all cause death in the lowest (HR 0.87; 95% CI 0.71 to 1.06) and highest quartiles of iTTR (HR 0.89; 95% CI 0.67 to 1.16; p value for interaction 0.67). Additionally, the same benefit of apixaban with regards to bleeding outcomes was observed across the lowest and highest quartiles of iTTR.

A substudy³⁶ of the ROCKET AF trial⁵ examined rivaroxaban once daily versus warfarin treatment with differing times in therapeutic range. For all patients randomized to warfarin, the mean time in therapeutic range (TTR) was 55%. Patients treated with rivaroxaban were compared to those treated with warfarin, across four quartiles of TTR: Q1=0 to 50.6%; Q2=50.7 to 58.5%; Q3=58.6 to 65.7%; Q4=65.7 to 100%. There was no significant difference in the primary outcomes of stroke or systemic embolism in patients treated with rivaroxaban across center TTR (cTTR) for warfarin (HR Q1 0.70; 95% CI 0.47 to 1.04; HR Q2 0.90; 95% CI 0.64 to 1.26; HR Q3 0.88; 95% CI 0.62 to 1.25; HR Q4 0.73; 95% CI 0.50 to 1.06; p value for interaction 0.71). However, patients treated with rivaroxaban did have lower risk of major or NMCR bleeding compared to patients in the lowest quartile of warfarin cTTR with a significant interaction between treatment and time in therapeutic range (Q1 HR 0.80; 95% CI 0.66 to 0.98; Q2 HR 0.96; 95% CI 0.81 to 1.14; Q3 HR 1.03; 95% CI 0.87 to 1.22; Q4 HR 1.25; 95% CI 1.10 to 1.41; p value for interaction 0.001).

The first two studies from this group suggest that compared to aspirin or no therapy, an INR ≥ 2 lowers the risk of ischemic stroke. However, INR values above the therapeutic range may lead to higher rates of hemorrhagic stroke. The second two studies compared treatment with warfarin to a factor Xa inhibitor and showed that there was no difference in the treatment effect of rivaroxaban and apixaban across the ranges of INR values examined with regards to stroke or systemic embolism outcomes. There is mixed data regarding the interaction between INR control and treatment with regards to bleeding outcomes.

Elderly Patients With AF

Fourteen studies specifically explored the safety and effectiveness of stroke prevention therapies in the elderly.^37–50 A single-center, retrospective, observational study³⁷ included data from patients aged ≥65 years with chronic nonvalvular AF treated at an urban academic geriatrics practice over a 1-year period. Eligible patients were receiving noninvasive management of AF with warfarin or aspirin. A total of 112 patients (mean age, 82 years) were identified; 106 were included in this analysis (80 women, 26 men). Warfarin was prescribed in 85 percent (90 patients); aspirin in 15 percent (16). The distributions of both the CHADS₂ and Outpatient Bleeding Risk Index scores were not significantly different between the warfarin and aspirin groups. The proportions of patients treated with warfarin were not significantly different between the groups with a high risk for hemorrhage and the groups at lower risk. At 12 months in the 90 patients initially treated with warfarin, the rate of stroke was 2 percent (2 patients); major hemorrhage, 6 percent (5); and death, 20 percent (18). The number of patients who received aspirin was too small to provide sufficient power to detect significant treatment differences.

A prospective clinical study³⁸ of four clinical services of geriatric medicine included 209 inpatients, (mean age 84.7±7 years; women 60.8%) with chronic AF. The patients were distributed into two groups (anticoagulant or aspirin) according to medical decision. The evolution of the patients was recorded after 3 months. One hundred and two patients (48.8%) received anticoagulant and 107 patients received aspirin. Patients in the aspirin group were significantly older (86.5±6.5 vs. 82.9±7.1 years), had more frequent social isolation, had higher systolic blood pressure, and had more important subjective bleeding risk and risk of falls. After 3 months, the two groups did not significantly differ for death, bleeding, or ischemic events.

A prospective RCT³⁹ included 973 patients aged 75 years or over (mean age 81·5 years, SD 4·2) with AF from primary care who were randomly assigned to warfarin (target INR 2–3) or aspirin (75mg per day). The primary outcome was fatal or disabling stroke (ischemic or hemorrhagic), intracranial hemorrhage, or clinically significant arterial embolism. Analysis was by intention to treat. There were 24 primary events (21 strokes, 2 other intracranial hemorrhages, and 1 systemic embolus) in people assigned to warfarin, and 48 primary events (44 strokes, 1 other intracranial hemorrhage, and 3 systemic emboli) in people assigned to aspirin in the ITT population (yearly risk 1.8% vs. 3.8%, relative risk 0.48; 95% CI 0.28 to 0.80; p=0·003). Yearly risk of extracranial hemorrhage was 1.4 percent (warfarin) versus 1.6 percent (aspirin) (relative risk 0.87, 95% CI 0.43 to 1.73).

An RCT⁴⁰ of primary thromboprophylaxis for AF included patients aged >80 and <90 randomized to receive dose-adjusted warfarin (INR 2.0–3.0) or aspirin 300mg. The primary outcome measure was a comparative frequency of combined outcomes comprising death, thromboembolism, serious bleeding, and withdrawal from the study. Seventy-five patients (aspirin 39; warfarin 36) were entered (mean age 83.9, 47% male). Patients on aspirin had significantly more adverse events (13/39; 33%) than patients on warfarin (2/36; 6%; p=0.002). Ten of 13 aspirin adverse events were caused by side effects and serious bleeding; there were three deaths (two aspirin, one warfarin).

Another RCT⁴¹ recruited patients over 75 years of age without previous stroke or systemic embolism. Patients were randomized into three groups, (A) aspirin 100mg/day, (B) fixed-dose warfarin 1mg/day; and (C) adjusted-dose warfarin with a target range of INR between 1.6 and 2.5. The study was discontinued 6 months after the enrollment of the first patient for safety reasons. Over a mean followup period of 3.7 months, two patients from group B (n=14) developed a dangerous prolongation of the INR (7.0 and 4.2), which led to the discontinuation of fixed-dose warfarin. Another patient from the same group experienced a major bleeding event 1 month after enrollment in the study (INR 5.5). The percentage of INR measurements within the target range was significantly lower in group B (48.7%) than in group C (83.7%) (p<0.001).

A prospective observational study⁴² included 207 older people (>75 years) with AF and first ever ischemic stroke. During the followup period (mean 88.4 months, range 3–120), the study population was under either oral anticoagulants (n=72) or aspirin (n=135). The cumulative 10-year mortality and recurrence rates were 92.5 percent (95% CI 85.7 to 99.3) and 66.1 percent (95% CI 43.1 to 89.1), respectively. Increasing age, functional dependency at hospital discharge, and antiplatelet versus anticoagulation therapy were independent determinants of mortality. Antiplatelet versus anticoagulation therapy was the sole determinant of vascular recurrence. Anticoagulation was associated with decreased risk of death (HR 0.47; 95% CI 0.31 to 0.72; p=0.001)) and recurrent thromboembolism (HR 0.31; 95% CI 0.16 to 0.62; p=0.002). These results suggest that the benefits of anticoagulation for secondary stroke prevention in AF patients extend to elderly.

A retrospective cohort analysis⁴³ evaluated persons discharged on warfarin after an AF admission using data from Medicare’s National Stroke Project. It examined antiplatelet therapy among warfarin users and the impact on major bleeding rates. Prediction of concurrent antiplatelet use and hospitalization with a major acute bleed within 90 days after discharge from the index AF admission was assessed. A total of 10,093 warfarin patients met inclusion criteria with a mean age of 77 years; 19.4 percent received antiplatelet therapy. Antiplatelet use was less common among women, older persons, and persons with cancer, terminal diagnoses, dementia, and bleeding history. Persons with coronary disease were more likely to receive an antiplatelet agent. Antiplatelets increased major bleeding rates from 1.3 percent to 1.9 percent (P=0.052). In the multivariate analysis, factors associated with bleeding events included age (OR, 1.03; 95% CI 1.002 to 1.05), anemia (OR, 2.52; 95% CI 1.64 to 3.88), a history of bleeding (OR, 2.40; 95% CI 1.71 to 3.38), and concurrent antiplatelet therapy (OR, 1.53; 95% CI 1.05 to 2.22).

A substudy⁴⁴ of the BAFTA trial³⁹ evaluated 665 patients aged 75 or over with AF based in the community who were randomized within the BAFTA trial and were not taking warfarin throughout or for part of the study period. A total of 54 (8%) patients had an ischemic stroke, four (0.6%) had a systemic embolism, and 13 (2%) had a TIA. Based on this single trial population, current risk stratification schemes in older people with AF have only limited ability to predict the risk of stroke.

Another study⁴⁵ examined the effectiveness of oral anticoagulation on risk of stroke of any nature (fatal and nonfatal ischemic and/or hemorrhagic stroke) in patients with nonvalvular AF or flutter living in the County of North Jutland, Denmark. This study used the Hospital Discharge Registry covering the county (490,000 inhabitants) from 1991 to 1998 to identify 2,699 men and 2,425 women with AF or flutter, aged 60–89 years. The risk of stroke associated with use of oral anticoagulation compared with no use was estimated, after adjustment for age, diabetes and underlying cardiovascular diseases. A total of 838 of 2,699 men (31%) and 552 of 2,425 women (23%) with AF had one or more recorded prescriptions of oral anticoagulation. The incidence rates of stroke were 31 per 1000 person-years of followup in men, and 30 per 1000 person-years of followup in women. The adjusted relative risks of stroke during anticoagulation were 0.6 (95% CI 0.4 to 1.0) in men, and 1.0 (95% CI 0.7 to 1.6) in women compared with nonuse periods. The adjusted relative risks of stroke associated with use of oral anticoagulation compared with no use varied by age in men, but not in women. In men aged 60–74 years the adjusted relative risk associated with use of oral anticoagulation compared with no use was 0.5 (95% CI 0.3 to 0.9), and in men aged 75–89 years the adjusted relative risk of stroke associated with oral anticoagulation compared with no use was 0.9 (95% CI 0.4 to 1.8). The adjusted relative risk of stroke increased with age. In men and women, the risk of stroke amongst patients aged 80–89 years was increased by a factor of 2.0 and 2.9 relative to the stroke risk amongst patients aged 60–69 years.

The RE-LY trial²⁶ randomized 18,113 patients to receive dabigatran 110 or 150mg twice a day or warfarin dose adjusted to an INR of 2.0–3.0 for a median followup of 2.0 years. A substudy of this trial⁴⁶ assessed the impact of age on the findings and found that there was a significant treatment-by-age interaction, such that dabigatran 110mg twice a day compared with warfarin was associated with a lower risk of major bleeding in patients aged <75 years (1.89% vs. 3.04%; p<0.001) and a similar risk in those aged ≥75 years (4.43% vs. 4.37%; p=0.89; p for interaction <0.001), whereas dabigatran 150mg twice a day compared with warfarin was associated with a lower risk of major bleeding in those aged <75 years (2.12% vs. 3.04%; p<0.001) and a trend toward higher risk of major bleeding in those aged ≥75 years (5.10% vs. 4.37%; p=0.07; p for interaction <0.001). The interaction with age was evident for extracranial bleeding, but not for intracranial bleeding, with the risk of the latter being consistently reduced with dabigatran compared with warfarin irrespective of age. Based on these results, patients with AF at risk for stroke, both doses of dabigatran compared with warfarin have lower risks of both intracranial and extracranial bleeding in patients aged <75 years. In those aged ≥75 years, intracranial bleeding risk is lower but extracranial bleeding risk is similar or higher with both doses of dabigatran compared with warfarin.

A subgroup analysis of the RE-LY trial⁵¹, attempted to estimate effects of dabigatran, compared with warfarin, on stroke, bleeding and mortality in patients with AF in the Randomized Evaluation of Long-Term Anticoagulant Therapy (RE-LY) trial according to age and analyzed treatment effects using age as a continuous variable and using age categories. The results showed that the benefits of dabigatran versus warfarin regarding stroke (HR range 0.63 (95% CI 0.46 to 0.86) to 0.70 (0.31 to 1.57) for dabigatran 150 mg twice daily), HR range 0.52 (0.21 to 1.29) to 1.08 (0.73 to 1.60) for dabigatran 110 mg twice daily) and intracranial bleeding were maintained across all age groups (interaction p values all not significant). There was a highly significant interaction (p value interaction <0.001) between age and treatment for extracranial major bleeding, with lower rates with both doses of dabigatran compared with warfarin in younger patients (HR 0.78 (0.62 to 0.97) for 150 mg twice daily, HR 0.72 (0.57 to 0.90) for 110 mg twice daily) but similar (HR 1.50 (1.03 to 2.18) for 110 mg twice daily) or higher rates (HR 1.68 (1.18 to 2.41) for 150 mg twice daily) in older patients (≥80 years).

A subgroup analysis of the AVERROES trial looked at the Efficacy and safety of apixaban compared with aspirin in the elderly. compared with aspirin, apixaban was more efficacious for preventing strokes and systemic embolism in patients ≥85 years (absolute rate [AR] 1% per year on apixaban versus 7.5% per year on aspirin; hazard ratio [HR] 0.14, 95% confidence interval [CI] 0.02-0.48) compared with younger patients (AR 1.7% per year on apixaban versus 3.4% per year on aspirin; HR 0.50, 95% CI 0.35-0.69) (P-value for interaction = 0.05). Major hemorrhage was higher in patients ≥85 years compared with younger patients but similar with apixaban versus aspirin in both young and older individuals (4.9% per year versus 1.0% per year on aspirin and 4.7% per year versus 1.2% per year on apixaban) with no significant treatment-by-age interaction (P-value = 0.65).

Two substudies^48,49 of the ARISTOTLE RCT⁹ examined the treatment effects of apixaban 5mg twice daily versus warfarin in elderly patients. In the study by Halvorsen, older patients were at higher overall risk for all cardiovascular events. Risk for events increased in a step-wise manner with age (age <65 vs. age 65-74 vs. age ≥ 75) for stroke or systemic embolism (adj HR age 65-74 1.47; 95% CI 1.11 to 1.94; Adj HR age ≥ 75 1.62; 95% CI 1.18 to 2.22; adjusted p=0.10), all-cause mortality (adj HR age 65-74 1.01; CI 0.84 to 1.21; adj HR age ≥ 75 1.53; 95% CI 1.26 to 1.85; adjusted p<0.0001) and major bleeding (adj HR age 65-74 1.52; 95% CI 1.20 to 1.92; adj HR age ≥75 2.18; 95% CI 1.69 to 2.81; adjusted p<0.0001). Across older age groups, patients treated with apixaban had lower rates of stroke or systemic embolism (HR age 65-74 0.72; 95% CI 0.54 to 0.96; HR age ≥ 75 0.71; 95% CI 0.53 to 0.95; interaction with continuous age p=0.11). Similarly, apixaban reduced the risk of major bleeding compared to treatment with warfarin, across older age groups (HR age 65-74 0.71; 95% CI 0.56 to 0.89; HR age ≥75 0.64; 95% CI 0.52 to 0.79; interaction with continuous age p=0.63). There was no significant difference between treatment groups in stroke or systemic embolism or major bleed in patients <65. Further analysis of patients ≥ 75 years old showed a trend toward increasing benefit of apixaban compared to warfarin therapy with regards to bleeding in patients as renal function worsened (HR eGFR >80 0.60; 95% CI 0.28 to 1.32; HR eGFR >50-80 0.79; 95% CI 0.37 to 1.06; HR eGFR >30-50 0.53; 95% CI 0.37 to 0.76; HR eGFR ≤ 30 0.35; 95% CI 0.14 to 0.86; interaction p value 0.16).

The study by Alexander evaluated patients with one criteria for dose reduction (at least two were required to reduce dose to 2.5mg twice daily): 80 years or older, weight ≤ 60 kg and creatinine level of at least 1.5mg/dL. Among patients with weight ≤ 60 kg, those receiving apixaban had a statistically significant decreased risk of major bleeding event (HR 0.6; 95% CI 0.4 to 0.9). Patients 80 years or older and those with creatinine level of at least 1.5mg/dL, were numerically less likely to have a major bleeding event with apixaban, although this did not reach statistical significance (HR 0.7; 95% CI 0.5-1.1 and HR 0.7; 95% CI 0.5 to 1.2 respectively).

A retrospective study of 233 patients aged 80 years or older with AF evaluated the efficacy and safety of oral anticoagulation therapy with low (2.0) versus standard (2.5) INR targets. Hemorrhages and thromboses occurred only in the group with standard INR.⁴⁷.

Finally, a substudy⁵⁰ of the ROCKET AF⁵ RCT evaluated once daily rivaroxaban versus warfarin in elderly patients with AF. Outcomes in patients <75 were compared with those in patients ≥75. Patients 75 or older had lower BMI (27.3 vs. 29.0; p<0.0001), had higher mean CHADS₂ score (3.69 vs. 3.30; p<0.0001) and lower rates of congestive heart failure (58.6% vs. 65.5%; p<0.0001) and diabetes (33.8% vs. 45.1%; p<0.0001). Compared to patients treated with warfarin, those randomized to treatment with rivaroxaban had similar rates of stroke/systemic embolism (HR Age ≥75 0.80; 95% CI 0.63 to 1.02; HR Age<75 0.95; 95% CI 0.76 to 1.19; p value for interaction 0.31) and major bleeding (HR Age ≥75 1.11; 95% CI 0.92 to 1.34; HR Age<75 0.96; CI 0.78 to 1.19; p value for interaction 0.34), regardless of age. The only significant observed difference between treatment groups was in risk of hemorrhagic stroke for patients <75 years old (HR 0.47; 95% CI 0.25 to 0.88).

Fourteen studies including observational, small RCTs, and sub-studies of large RCTs compared the effect of different strategies to prevent stroke and bleeding in elderly participants with AF. Of 7 studies comparing the effects of warfarin vs aspirin in older adults, compared to aspirin, warfarin was generally found to be associated with lower risk of stroke/SE/bleeding for both primary and secondary prevention. In studies comparing the effects of DOACs vs warfarin, the DOACs were generally found to be associated with similar or decreased risk of stroke/SE/bleeding compared with warfarin among older adults.

Patients With AF and Myocardial Infarction

One substudy of the RE-LY trial²⁶ evaluated the use of therapies for stroke prevention in AF patients with MI.⁵² In this analysis, the relative effects of dabigatran versus warfarin on myocardial ischemic events were consistent in patients with or without a baseline history of MI or coronary artery disease. Patients with a baseline history of coronary artery disease (CAD) or previous MI are at risk for recurrent ischemic events. There were 1,886 (31%) CAD/MI patients in the dabigatran 110mg group, 1,915 (31%) in the dabigatran 150mg group, and 1,849 (31%) in the warfarin group. The relative effects of dabigatran compared with warfarin were highly consistent between patients with prior CAD/MI compared with those without (all probability values for interaction were nonsignificant).

Elderly Patients With AF and Myocardial Infarction

One observational study⁵³ evaluated the effects of a combination of antithrombotics in 7,619 NSTEMI patients aged ≥65 years with AF. Relative to aspirin alone, antithrombotics were associated with increased bleeding risk (adj HR 1.22; 95% CI 1.03 to 1.46 for aspirin+clopidogrel vs. aspirin alone; adj HR 1.46; 95% CI 1.21 to 1.80 for warfarin+aspirin vs. aspirin alone). Patients treated with triple therapy of aspirin+clopidogrel+warfarin had the greatest bleeding risk (HR 1.65; 95% CI 1.30 to2.10). The rates of major cardiac outcomes (death, readmission for MI, or stroke) were similar between groups, although relative to aspirin alone, there was a trend toward lower risk for the warfarin+aspirin group (HR 0.88; 95% CI 0.78 to 1.00).

Patients With AF and Carotid Artery Disease

A single secondary analysis⁵⁴ of the ROCKET AF trial⁵ evaluated outcomes in patients with AF and carotid artery disease, treated with either warfarin or rivaroxaban. After adjustment, there was no significant difference in rates of stroke or systemic embolism in patients with carotid artery disease compared to those without. Similarly, there was no significant difference in the primary safety endpoint of major/NMCR bleeding between the patients with or without carotid artery disease. Compared to those without carotid artery disease, patients with carotid artery disease had similar prevention of stroke and systemic embolism with apixaban versus warfarin (interaction p value 0.96). Similarly, there was no significant interaction between treatment and presence of carotid artery disease with major or NMCR bleeding (interaction p value 0.62). This single study suggests no difference in the treatment effects of rivaroxaban and warfarin in patients with carotid artery disease.

Patients With AF and Peripheral Arterial Disease

One secondary analysis⁵⁵ of the ARISTOTLE trial⁹ evaluated outcomes in patients with AF and peripheral arterial disease (PAD), treated with apixaban versus warfarin. Compared to those without PAD, patients with PAD had similar prevention of stroke and systemic embolism with apixaban versus warfarin (PAD HR 0.63; 95% CI 0.32 to 1.25; No PAD HR 0.80; 95% CI 0.66 to 0.96; interaction p value for PAD versus no PAD 0.52). There was similarly no significant interaction between presence of PAD and treatment group on major bleeding (interaction p value 0.58). While data is only available from one study, this suggests that patients with PAD had similar benefit from treatment with apixaban as compared to those without.

Patients With AF and Underlying Anemia

One analysis⁵⁶ of the ARISTOTLE RCT⁹ examined patients with anemia treated with apixaban versus warfarin. There was no difference in the benefits of reduced stroke or systemic embolization events (Anemia HR 0.56; 95% CI 0.34 to 0.95; No Anemia HR 0.84; 95% CI 0.68 to 1.01; interaction p value for anemia versus no anemia 0.17) with apixaban in patients with anemia. The incidence of new anemia during treatment was lower in patients with apixaban (HR 0.91; 95% CI 0.84 to 0.98; p=0.037) and there was no significant interaction between underlying anemia and treatment group on any of the bleeding outcomes. This single analysis suggests that the same benefits of apixaban, including decreased risk of stroke or systemic embolism, extend to patients with underlying anemia without differential change in bleeding risk.

Patients With AF and History of Bleeding

A secondary analysis⁵⁷ of the ARISTOTLE RCT⁹ evaluated clinical outcomes in patients with history of bleeding treated with 5mg twice daily of apixaban versus warfarin. Patients treated with apixaban had consistently lower rates of bleeding overall and this extended to patients with prior history of bleeding. The only p value for interaction that was significant for apixaban versus warfarin was for major or clinically relevant non-major bleeding (History of bleeding HR 0.82; 95% CI 0.66-1.00; No History of Bleeding HR 0.64; 95% CI 0.57 to 0.72; p value for interaction 0.046). While only informed by one study, this suggests that the lower rates of bleeding observed with treatment with apixaban compared to warfarin are generally similar for patients with a history of bleeding. This benefit may not include lower rates of major or clinically relevant non-major bleeding; further data is necessary to clarify this borderline result.

Patients With AF and Chronic Obstructive Pulmonary Disease

Another analysis⁵⁸ of the ARISTOTLE trial⁹ evaluated the treatment effects of apixaban versus warfarin in patients with chronic obstructive pulmonary disease (COPD). Overall, all-cause mortality was higher in patients with a diagnosis of COPD (adj HR 1.60; 95% CI 1.36 to 1.88; p<0.001) while there was no significant difference in major bleeding. There was no significant difference in the effect of apixaban on all-cause mortality (COPD HR 0.80, 95% CI 0.62 to 1.04; No COPD HR 0.92; 95% CI 0.82 to 1.04; p value for interaction 0.35), stroke or systemic embolism (COPD HR 0.92; 95% CI 0.52 to 1.63; No COPD HR 0.78; 95% CI 0.65 to 0.95; p value for interaction 0.62), or major bleeding (COPD HR 0.83; 95% CI 0.57 to 1.02; No COPD HR 0.67; 95% CI 0.60 to 0.75; p value for interaction 0.42) in patients with and without COPD. This single analysis from the ARISTOTLE trial gives data to suggest that there is no treatment difference in the benefits observed with apixaban in patients with or without COPD.

Patients With AF by Sex

One secondary analysis⁵⁹ of the ARISTOTLE trial⁹ evaluated the treatment of men versus women with apixaban 5mg twice daily or warfarin. After adjustment, there was no difference between women and men with regard to stroke or systemic embolism (Adj HR 0.91; 95% CI 0.74 to 1.12; p=0.38) but women had significantly less all-cause mortality and cardiovascular death (adjusted HR 0.63; 95% CI 0.55 to 0.73; p<0.001). When evaluated by treatment, there was no significant interaction with sex (women HR 0.73; 95% CI 0.54 to 0.97; men HR 0.84; 95% CI 0.66 to 1.05; p value for interaction 0.45), and major bleeding (women HR 0.56; 95% CI 0.44 to 0.72; men HR 0.88; 95% CI 0.64 to 0.90; p value for interaction 0.06).

In a secondary analysis of the AVERROES study⁶⁰ the effect of treatment with aspirin compared with apixaban on ischemic stroke and major bleeding was assessed in women compared with men. Female patients with atrial fibrillation are at increased stroke risk compared with male patients, and the underlying reasons for higher risk are uncertain. Women compared with men tended to be older (aspirin, 71.8 versus 68.8 years; apixaban, 71.4 versus 68.6 years), with a higher proportion of those aged ≥75 years. Also, women had less peripheral artery disease (aspirin, 2.4% versus 3.7%; apixaban, 1.4% versus 3.0%), more heart failure, and higher mean CHADS2 (congestive heart failure, hypertension, age of 75 years or older, diabetes [1 point each], stroke or transient ischemic attack [2 points]) scores (aspirin, 2.2 versus 2.0; apixaban, 2.1 versus 2.0). Women compared with men had higher ischemic stroke rates (aspirin, 3.99% versus 2.28%; apixaban, 1.55% versus 0.82%) but similar bleeding rates (aspirin, 1.29% versus 1.22%; apixaban, 1.15% versus 1.36%). The relative effect of apixaban compared with aspirin was similar in men and women for both ischemic stroke (women, 1.55 % versus 3.99%; hazard ratio, 0.39; 95% confidence interval, 0.23-0.64; men, 0.82 % versus 2.28%; hazard ratio, 0.36; 95% confidence interval, 0.19-0.63; p value for interaction 0.84) and major bleeding (women, 1.15 % versus 1.29%; hazard ratio, 1.15; 95% confidence interval, 0.59-2.23; men, 1.36% versus 1.22%; hazard ratio, 1.13; 95% confidence interval, 0.64-2.02; p value for interaction 0.97).

In only two studies assessing potentially differences in treatment effect by sex both included apixaban but the comparators were different – one was warfarin and one was aspirin. No interaction between sex and treatment was found for major bleeding (for either comparator, warfarin or aspirin) or for ischemic stroke (as compared to aspirin).

Patients With AF and Diabetes

A substudy⁶¹ of the ARISTOTLE RCT⁹, analyzed the treatment effect of apixaban 5mg twice daily versus warfarin in patients with and without diabetes. Overall, patients with diabetes were younger, had higher weights, were more likely to have hypertension and prior stroke or systemic embolism, and had higher CHA₂DS₂-VASc Scores. Compared with warfarin, patients with diabetes and who received apixaban were numerically less likely to have stroke or systemic embolism (HR 0.75; 95% CI 0.53 to 1.05) or death from any cause (HR 0.89; 95% CI 0.66 to 1.20). There were no significant interactions related to diabetes for the efficacy endpoints. All-cause bleeding was significantly lower in patients with diabetes who received apixaban (HR 0.73; 95% CI 0.66 to 0.81). While ISTH major bleeding was not significantly lower in patients with diabetes who were treated with apixaban, it was significantly lower in those without diabetes (diabetes HR 0.96; 95% CI 0.74 to 1.25; no diabetes HR 0.60; 95% CI 0.51 to 0.52; p value for interaction 0.0034). This interaction remained after adjustment.

A substudy⁶² of the ROCKET AF Trial⁵ evaluated treatment effect of rivaroxaban daily versus warfarin in patients with and without diabetes. Overall, 5,695 (39.9%) of patients enrolled in the ROCKET AF trial had diabetes. Patients with diabetes had higher rates of vascular death (3.24 vs. 2.63; p=0.0001) and myocardial infarction (1.35 vs. 0.75; p<0.0001). There was not significant interaction between treatment and diabetes status for the outcomes of stroke/SE (HR diabetes 0.82; 95% CI 0.63 to 1.08; HR no diabetes 0.92; 95% CI 0.75 to 1.13; p value for interaction 0.53) and major/NMCR bleeding (HR diabetes 0.98; 95% CI 0.88 to 1.10; HR no diabetes 1.09; 95% CI 0.99 to 1.20; p value for interaction 0.17). However, in a composite endpoint of stroke/systemic embolism/vascular death/MI, patients with diabetes who were treated with rivaroxaban had slightly lower risk (HR diabetes 0.84; 95% CI 0.72 to 0.99; HR no diabetes 1.01; 95% CI 0.88 to 1.17; p value for interaction 0.097), although the interaction was not significant.

In a supplemental analysis of RE-LY trial.⁶³ Of 18,113 patients in RE-LY, 4221 patients (23.3%) had DM. Patients with DM were younger (70.9 vs. 71.7 years), more likely to have hypertension (86.6% vs. 76.5%), coronary artery disease (37.4% vs. 24.9%) and peripheral vascular disease (5.6% vs. 3.2%); (all p < 0.01). Time in therapeutic range for warfarin-treated patients was 65% for diabetic versus 68% for non-diabetic patients (p < 0.001). Regardless of assigned treatment, stroke or systemic embolism was more common among patients with DM (1.9% per year vs. 1.3% per year; p<0.001). DM was also associated with an increased risk of death (5.1% per year vs. 3.5% per year; p<0.001) and major bleeding (4.2% per year vs. 3.0% per year; p<0.001). The absolute reduction in stroke or systemic embolism with dabigatran compared to warfarin was greater among patients with DM than those without DM (dabigatran 110mg: 0.59% per year vs. 0.05% per year; dabigatran 150mg: 0.89% per year vs. 0.51% per year). There was however, no statistically significant interaction between treatment (dabigatran 110mg or dabigatran 150 mg vs. warfarin) and diabetes for stroke or systemic embolism, ischemic stroke, hemorrhagic stroke, death, major bleeding, or intracranial bleeding.

The results from three studies assessing the potential impact of diabetes on treatment effect were inconsistent; no impact on treatment effect was seen between dabigatran and warfarin on any of the included efficacy or safety outcomes; a statistically significant interaction between treatment (apixaban vs warfarin) was found only for major bleeding (diabetics did not have the same statistically significant reduction in major bleeding as non-diabetics); and a statistically significant interaction between treatment (rivaroxaban vs warfarin) was found only for a composite endpoint of stroke/systemic embolism/vascular death/MI (diabetics had a statistically significant reduction that was not seen in non-diabetics).

Patients With AF and Aspirin Treatment

A secondary analysis⁶⁴ of the ARISTOTLE trial,⁹, evaluated the use of apixaban 5mg twice daily compared to warfarin in patients with concomitant aspirin therapy. Overall, patients treated with aspirin were more likely to be male, have a history of MI, PCI, CABG or PAD and to have diabetes or hypertension. After adjustment for baseline confounders and variables associated with aspirin use, patients treated with aspirin had higher rates of thromboembolic events (stroke or systemic embolism, ischemic stroke, myocardial infarction) and higher rates of bleeding. Apixaban treatment led to similar reductions in stroke or systemic embolism (Aspirin HR 0.58; 95% CI 0.39 to 0.85; No Aspirin HR 0.84; 95% CI 0.66 to 1.07; p value for interaction 0.10) and consistent reductions in major bleeding (aspirin HR 0.77; 95% CI 0.60 to 0.99; no aspirin HR 0.65; 95% CI 0.55 to 0.78; p value for interaction 0.29) in patients treated with and without aspirin.

One study⁶⁵ also evaluated the use of aspirin by treatment group in the ROCKET-AF trial.⁵ Overall, 5,205 (46.5%) of patients had chronic aspirin use at baseline. Patients on aspirin were younger (median age 72 versus 73 years old) and had slightly higher CHADS₂ scores (mean 3.5 versus 3.4). Among all patients, those with baseline aspirin use had higher risk of all-cause death (HR 1.27; 95% CI 1.13 to 1.42; p<0.0001) and vascular death (HR 1.29; 95% CI 1.11 to 1.49; p=0.0006) as well as major or NMCR bleeding (HR 1.32; 95% CI 1.21 to 1.43; p<0.0001) or major bleeding (HR 1.46, 95% CI 1.25 to 1.71; p<0.0001). There was no significant interaction between treatment and use of aspirin versus none on any of the efficacy or safety outcomes (stroke/SE, stroke/SE/vascular death, all-cause death, vascular death, stroke, SE, MI, major/NMCR bleeding, major bleeding, ICH, fatal major bleeding, hemorrhagic stroke).

In an ENGAGE AF substudy,⁶⁶ patients who received a single antiplatelet drug during the study at the discretion of their physician were compared to those who did not receive a single antiplatelet drug during the study. A total of 4,912 patients received a single antiplatelet drug during the study of which 92.5% were aspirin. In the high dose edoxaban vs. warfarin comparisons, there were no statistically significant interactions between treatment and use of single antiplatelet drug vs. none on stroke or systemic embolic events, ischemic stroke, hemorrhagic stroke, MI, cardiovascular death, major bleeding, intracranial bleeding, or any bleeding. Similar results were seen for the low dose edoxaban vs. warfarin comparisons and for the large subset of aspirin only users.

From a total of three studies, no impact on treatment effect between apixaban, rivaroxaban, low dose edoxaban or high dose edoxaban vs warfarin was seen in patients with concomitant aspirin administration.

Patients With AF and Hypertension

One secondary analysis⁶⁷ of the ROCKET AF⁵ RCT evaluated outcomes based on screening systolic blood pressure and hypertension. At baseline, 12,902 patients had a history of controlled or uncontrolled hypertension (HTN). Compared to patients without hypertension, those with hypertension had a trend toward higher risk for stroke or systemic embolism (HTN HR 1.22; 95% CI 0.89 to 1.66; uncontrolled HTN HR 1.42; 95% CI 1.03 to 1.95; p value 0.06). There was no significant interaction between treatment and HTN status (no HTN versus controlled hypertension versus uncontrolled hypertension) on all ischemic/thrombotic or bleeding outcomes. While there is only data from one study available, this suggests that there is no difference in the observed treatment effects of rivaroxaban and warfarin among patients with varying degrees of HTN.

Patients With AF and Heart Failure

In an ENGAGE AF substudy,⁶⁸ the 8145 patients in the ENGAGE AF study in either the warfarin or high dose edoxaban treatment groups who had heart failure (6344 with NYHA I-11 and 1801 with NYHA III-IV) were compared to the 5926 who did not have heart failure. There was no statistically significant interaction between heart failure groups (no heart failure, NYHA I-II, and NYHA III-IV) and treatment for stroke or systemic embolic events, ischemic stroke, hemorrhagic stroke, any cause death, cardiovascular death, cardiovascular hospitalization, major bleeding, intracranial hemorrhage, or GI bleeding.

A secondary analysis⁶⁹ of the ROCKET AF RCT⁵ evaluated treatment with rivaroxaban once daily versus warfarin in patients with heart failure. Overall, 9033 (63.7%) of patients in the ROCKET AF trial had heart failure diagnosis (clinical HF or EF <40%) at the time of randomization. Patients with heart failure were significantly more likely to have stroke/systemic embolism/vascular death (HR 1.28; 95% CI 1.11 to 1.47; p=0.0006) as well as all-cause death (HR 1.34; 95% CI 1.37 to 1.98; p<0.0001) and vascular death (HR 1.65; 95% CI 1.37 to 1.98; p<0.0001). There was no significant interaction with regards to heart failure status for efficacy or safety outcome between treatment groups. However, patients with heart failure who were treated with rivaroxaban were significantly less likely to experience hemorrhagic stroke (HR 0.38; 95% CI 0.19 to 0.76).

Data from these two studies give similar findings and suggest that patients had similar ischemic and bleeding outcomes based on the treatment received regardless of heart failure status.

Patients With AF and Left Ventricular Hypertrophy

In a post-hoc analysis of the Randomized Evaluation of Long-term anticoagulation therapY (RE-LY) Study⁷⁰ the hypothesis that left ventricular hypertrophy (LVH) interferes with the antithrombotic effects of dabigatran and warfarin in patients with atrial fibrillation (AF) was tested. LVH was defined by electrocardiography (ECG) and included patients with AF on the ECG tracing at entry. LVH was present in 2353 (22.7%) out of 10 372 patients. In patients without LVH, the rates of primary outcome (composite of stroke and systemic embolism) were 1.59% per year with warfarin, 1.60% with dabigatran 110 mg (HR vs. warfarin 1.01, 95% confidence interval (CI) 0.75-1.36) and 1.08% with dabigatran 150 mg (HR vs. warfarin 0.68, 95% CI 0.49-0.95). In patients with LVH, the rates of primary outcome were 3.21% per year with warfarin, 1.69% with dabigatran 110 mg (HR vs. warfarin 0.52, 95% CI 0.32-0.84) and 1.55% with 150 mg (HR vs. warfarin 0.48, 95% CI 0.29-0.78). The interaction between LVH status and dabigatran 110 mg vs. warfarin was significant for the primary outcome (P = 0.021) and stroke (P = 0.016), but not for major bleeding (p=0.235). However, there was no statistically significant interaction between LVH status and dabigatran 150 mg vs. warfarin for the primary outcome (p=0.244), any stroke (P=0.147) or major bleeding (p=0.888).

In this single study, the treatment effect (reduced risk of stroke or systemic embolism, reduced risk of any stroke and no difference in major bleeding) between the FDA approved 150 mg dose of dabigatran and warfarin was not statistically significantly impacted by LVH.

Patients With AF and History of Falls

A single substudy⁷¹ of the ARISTOTLE trial⁹ evaluated the comparison of treatment with apixaban versus warfarin in patients with a history of falling. Overall, patients with a history of falling had similar risk of stroke or systemic embolism (adj HR 1.12; 95% CI 0.72 to 1.72; p=0.618) after adjustment compared to those without a history of falls. However, there was an increase in the risk of major or NMCR bleeding (adj HR 1.27; 95% CI 1.03 to 1.58; p=0.028), any bleeding (adj HR 1.19; 95% CI 1.05 to 1.34; p=0.005) and intracranial bleeding (HR 1.96; 95% CI 1.06 to 3.61; p=0.032) in patients with prior history of falling. When outcomes were evaluated based on treatment group, no significant interaction was found between a history of falls and treatment with apixaban versus warfarin for any of the ischemic or bleeding endpoints. This single study suggests that while patients with a history of falls have increased risk of bleeding overall, there was no significant difference in outcomes based on treatment with apixaban compared to warfarin.

Patients With AF and a History of Cancer

One substudy⁷² of the ARISTOTLE trial⁹ examined the treatment of patients with atrial fibrillation and a history of cancer with apixaban compared to warfarin. Overall, three was no difference in the rates of stroke or systemic embolism in patients with a history of cancer compared to those without (HR 0.93; 95% CI 0.63 to 1.37; p=0.710). After adjustment, there was no relationship between cancer history and risk of major or NMCR bleeding (HR 1.25; 95% CI 1.04 to 1.151; p=0.0181). Similarly, evaluation of outcomes based on treatment group showed no significant interaction between a history of cancer and treatment with apixaban or warfarin on either ischemic of bleeding endpoints. There was a trend toward a significant interaction between cancer status and treatment effect only for death from any cause, although this did not reach statistical significance. This single study suggests that there is no difference in the treatment effect observed with apixaban in patients with a history of cancer compared to those without.

References to Appendix G

1.

Connolly SJ, Pogue J, Hart RG, et al. Effect of clopidogrel added to aspirin in patients with atrial fibrillation. N Engl J Med. 2009 May 14;360(20):2066–78. doi: 10.1056/NEJMoa0901301. PMID: 19336502. [PubMed: 19336502] [CrossRef]

2.

Reddy VY, Mobius-Winkler S, Miller MA, et al. Left atrial appendage closure with the Watchman device in patients with a contraindication for oral anticoagulation: the ASAP study (ASA Plavix Feasibility Study With Watchman Left Atrial Appendage Closure Technology). J Am Coll Cardiol. 2013 Jun 25;61(25):2551–6. doi: 10.1016/j.jacc.2013.03.035. PMID: 23583249. [PubMed: 23583249] [CrossRef]

3.

Connolly SJ, Eikelboom J, Joyner C, et al. Apixaban in patients with atrial fibrillation. N Engl J Med. 2011 Mar 3;364(9):806–17. doi: 10.1056/NEJMoa1007432. PMID: 21309657. [PubMed: 21309657] [CrossRef]

4.

Fox KA, Piccini JP, Wojdyla D, et al. Prevention of stroke and systemic embolism with rivaroxaban compared with warfarin in patients with non-valvular atrial fibrillation and moderate renal impairment. Eur Heart J. 2011 Oct;32(19):2387–94. doi: 10.1093/eurheartj/ehr342. PMID: 21873708. [PubMed: 21873708] [CrossRef]

5.

Patel MR, Mahaffey KW, Garg J, et al. Rivaroxaban versus warfarin in nonvalvular atrial fibrillation. N Engl J Med. 2011 Sep 8;365(10):883–91. doi: 10.1056/NEJMoa1009638. PMID: 21830957. [PubMed: 21830957] [CrossRef]

6.

Fordyce CB, Hellkamp AS, Lokhnygina Y, et al. On-Treatment Outcomes in Patients With Worsening Renal Function With Rivaroxaban Compared With Warfarin: Insights From ROCKET AF. Circulation. 2016 Jul 05;134(1):37–47. doi: 10.1161/circulationaha.116.021890. PMID: 27358435. [PubMed: 27358435] [CrossRef]

7.

Eikelboom JW, Connolly SJ, Gao P, et al. Stroke risk and efficacy of apixaban in atrial fibrillation patients with moderate chronic kidney disease. J Stroke Cerebrovasc Dis. 2012 Aug;21(6):429–35. doi: 10.1016/j.jstrokecerebrovasdis.2012.05.007. PMID: 22818021. [PubMed: 22818021] [CrossRef]

8.

Hohnloser SH, Hijazi Z, Thomas L, et al. Efficacy of apixaban when compared with warfarin in relation to renal function in patients with atrial fibrillation: insights from the ARISTOTLE trial. Eur Heart J. 2012 Aug 29 doi: 10.1093/eurheartj/ehs274. PMID: 22933567. [PubMed: 22933567] [CrossRef]

9.

Granger CB, Alexander JH, McMurray JJ, et al. Apixaban versus warfarin in patients with atrial fibrillation. N Engl J Med. 2011 Sep 15;365(11):981–92. doi: 10.1056/NEJMoa1107039. PMID: 21870978. [PubMed: 21870978] [CrossRef]

10.

Hijazi Z, Hohnloser SH, Andersson U, et al. Efficacy and Safety of Apixaban Compared With Warfarin in Patients With Atrial Fibrillation in Relation to Renal Function Over Time: Insights From the ARISTOTLE Randomized Clinical Trial. JAMA Cardiol. 2016 Jul 01;1(4):451–60. doi: 10.1001/jamacardio.2016.1170. PMID: 27438322. [PubMed: 27438322] [CrossRef]

11.

Giugliano RP, Ruff CT, Braunwald E, et al. Edoxaban versus warfarin in patients with atrial fibrillation. N Engl J Med. 2013 Nov 28;369(22):2093–104. doi: 10.1056/NEJMoa1310907. PMID: 24251359. [PubMed: 24251359] [CrossRef]

12.

Bohula EA, Giugliano RP, Ruff CT, et al. Impact of Renal Function on Outcomes With Edoxaban in the ENGAGE AF-TIMI 48 Trial. Circulation. 2016 Jul 05;134(1):24–36. doi: 10.1161/circulationaha.116.022361. PMID: 27358434. [PubMed: 27358434] [CrossRef]

13.

Hijazi Z, Hohnloser SH, Oldgren J, et al. Efficacy and safety of dabigatran compared with warfarin in relation to baseline renal function in patients with atrial fibrillation: a RE-LY (Randomized Evaluation of Long-term Anticoagulation Therapy) trial analysis. Circulation. 2014 Mar 04;129(9):961–70. doi: 10.1161/circulationaha.113.003628. PMID: 24323795. [PubMed: 24323795] [CrossRef]

14.

Hohnloser SH, Pajitnev D, Pogue J, et al. Incidence of stroke in paroxysmal versus sustained atrial fibrillation in patients taking oral anticoagulation or combined antiplatelet therapy: an ACTIVE W Substudy. J Am Coll Cardiol. 2007 Nov 27;50(22):2156–61. doi: 10.1016/j.jacc.2007.07.076. PMID: 18036454. [PubMed: 18036454] [CrossRef]

15.

Connolly S, Pogue J, Hart R, et al. Clopidogrel plus aspirin versus oral anticoagulation for atrial fibrillation in the Atrial fibrillation Clopidogrel Trial with Irbesartan for prevention of Vascular Events (ACTIVE W): a randomised controlled trial. Lancet. 2006 Jun 10;367(9526):1903–12. doi: 10.1016/s0140-6736(06)68845-4. PMID: 16765759. [PubMed: 16765759] [CrossRef]

16.

Al-Khatib SM, Thomas L, Wallentin L, et al. Outcomes of apixaban vs. warfarin by type and duration of atrial fibrillation: results from the ARISTOTLE trial. Eur Heart J. 2013 Aug;34(31):2464–71. doi: 10.1093/eurheartj/eht135. PMID: 23594592. [PubMed: 23594592] [CrossRef]

17.

Guimaraes PO, Wojdyla DM, Alexander JH, et al. Anticoagulation therapy and clinical outcomes in patients with recently diagnosed atrial fibrillation: Insights from the ARISTOTLE trial. Int J Cardiol. 2017 Jan 15;227:443–9. doi: 10.1016/j.ijcard.2016.11.014. PMID: 27852444. [PubMed: 27852444] [CrossRef]

18.

Vemmos KN, Tsivgoulis G, Spengos K, et al. Anticoagulation influences long-term outcome in patients with nonvalvular atrial fibrillation and severe ischemic stroke. Am J Geriatr Pharmacother. 2004 Dec;2(4):265–73. PMID: 15903285. [PubMed: 15903285]

19.

Tentschert S, Parigger S, Dorda V, et al. Recurrent vascular events in patients with ischemic stroke/TIA and atrial fibrillation in relation to secondary prevention at hospital discharge. Wien Klin Wochenschr. 2004 Dec 30;116(24):834–8. PMID: 15690967. [PubMed: 15690967]

20.

Berge E, Abdelnoor M, Nakstad PH, et al. Low molecular-weight heparin versus aspirin in patients with acute ischaemic stroke and atrial fibrillation: a double-blind randomised study. HAEST Study Group. Heparin in Acute Embolic Stroke Trial. Lancet. 2000 Apr 8;355(9211):1205–10. PMID: 10770301. [PubMed: 10770301]

21.

Hankey GJ, Patel MR, Stevens SR, et al. Rivaroxaban compared with warfarin in patients with atrial fibrillation and previous stroke or transient ischaemic attack: a subgroup analysis of ROCKET AF. Lancet Neurol. 2012 Apr;11(4):315–22. doi: 10.1016/s1474-4422(12)70042-x. PMID: 22402056. [PubMed: 22402056] [CrossRef]

22.

Diener HC, Connolly SJ, Ezekowitz MD, et al. Dabigatran compared with warfarin in patients with atrial fibrillation and previous transient ischaemic attack or stroke: a subgroup analysis of the RE-LY trial. Lancet Neurol. 2010 Dec;9(12):1157–63. doi: 10.1016/s1474-4422(10)70274-x. PMID: 21059484. [PubMed: 21059484] [CrossRef]

23.

Lawrence J, Pogue J, Synhorst D, et al. Apixaban versus aspirin in patients with atrial fibrillation and previous stroke or transient ischaemic attack: a predefined subgroup analysis from AVERROES, a randomised trial. Lancet Neurol. 2012 Mar;11(3):225–31. doi: 10.1016/s1474-4422(12)70017-0. PMID: 22305462. [PubMed: 22305462] [CrossRef]

24.

Easton JD, Lopes RD, Bahit MC, et al. Apixaban compared with warfarin in patients with atrial fibrillation and previous stroke or transient ischaemic attack: a subgroup analysis of the ARISTOTLE trial. Lancet Neurol. 2012 Jun;11(6):503–11. doi: 10.1016/s1474-4422(12)70092-3. PMID: 22572202. [PubMed: 22572202] [CrossRef]

25.

Rost NS, Giugliano RP, Ruff CT, et al. Outcomes With Edoxaban Versus Warfarin in Patients With Previous Cerebrovascular Events: Findings From ENGAGE AF-TIMI 48 (Effective Anticoagulation With Factor Xa Next Generation in Atrial Fibrillation-Thrombolysis in Myocardial Infarction 48). Stroke. 2016 Aug;47(8):2075–82. doi: 10.1161/strokeaha.116.013540. PMID: 27387994. [PubMed: 27387994] [CrossRef]

26.

Connolly SJ, Ezekowitz MD, Yusuf S, et al. Dabigatran versus warfarin in patients with atrial fibrillation. N Engl J Med. 2009 Sep 17;361(12):1139–51. doi: 10.1056/NEJMoa0905561. PMID: 19717844. [PubMed: 19717844] [CrossRef]

27.

Olesen JB, Lip GY, Lindhardsen J, et al. Risks of thromboembolism and bleeding with thromboprophylaxis in patients with atrial fibrillation: A net clinical benefit analysis using a ‘real world’ nationwide cohort study. Thromb Haemost. 2011 Oct;106(4):739–49. doi: 10.1160/th11-05-0364. PMID: 21789337. [PubMed: 21789337] [CrossRef]

28.

Ruiz Ortiz M, Romo E, Mesa D, et al. Oral anticoagulation in nonvalvular atrial fibrillation in clinical practice: impact of CHADS(2) score on outcome. Cardiology. 2010;115(3):200–4. doi: 10.1159/000284450. PMID: 20160440. [PubMed: 20160440] [CrossRef]

29.

Oldgren J, Alings M, Darius H, et al. Risks for Stroke, Bleeding, and Death in Patients With Atrial Fibrillation Receiving Dabigatran or Warfarin in Relation to the CHADS2 Score: A Subgroup Analysis of the RE-LY Trial. Ann Intern Med. 2011 Nov 15;155(10):660–7. doi: 10.1059/0003-4819-155-10-201111150-00004. PMID: 22084332. [PubMed: 22084332] [CrossRef]

30.

Taillandier S, Olesen JB, Clementy N, et al. Prognosis in Patients with Atrial Fibrillation and CHA(2) DS(2) -VASc Score = 0 in a Community-Based Cohort Study. J Cardiovasc Electrophysiol. 2012 Jul;23(7):708–13. doi: 10.1111/j.1540-8167.2011.02257.x. PMID: 22268375. [PubMed: 22268375] [CrossRef]

31.

Lopes RD, Al-Khatib SM, Wallentin L, et al. Efficacy and safety of apixaban compared with warfarin according to patient risk of stroke and of bleeding in atrial fibrillation: a secondary analysis of a randomised controlled trial. Lancet. 2012 Oct 1 doi: 10.1016/s0140-6736(12)60986-6. PMID: 23036896. [PubMed: 23036896] [CrossRef]

32.

Healey JS, Hart RG, Pogue J, et al. Risks and benefits of oral anticoagulation compared with clopidogrel plus aspirin in patients with atrial fibrillation according to stroke risk: the atrial fibrillation clopidogrel trial with irbesartan for prevention of vascular events (ACTIVE-W). Stroke. 2008 May;39(5):1482–6. doi: 10.1161/strokeaha.107.500199. PMID: 18323500. [PubMed: 18323500] [CrossRef]

33.

Hylek EM, Go AS, Chang Y, et al. Effect of intensity of oral anticoagulation on stroke severity and mortality in atrial fibrillation. N Engl J Med. 2003 Sep 11;349(11):1019–26. doi: 10.1056/NEJMoa022913. PMID: 12968085. [PubMed: 12968085] [CrossRef]

34.

Azoulay L, Dell’aniello S, Simon TA, et al. A net clinical benefit analysis of warfarin and aspirin on stroke in patients with atrial fibrillation: a nested case-control study. BMC Cardiovasc Disord. 2012 Jun 26;12(1):49. doi: 10.1186/1471-2261-12-49. PMID: 22734842. [PMC free article: PMC3444325] [PubMed: 22734842] [CrossRef]

35.

Wallentin L, Lopes RD, Hanna M, et al. Efficacy and safety of apixaban compared with warfarin at different levels of predicted international normalized ratio control for stroke prevention in atrial fibrillation. Circulation. 2013 Jun 04;127(22):2166–76. doi: 10.1161/circulationaha.112.142158. PMID: 23640971. [PubMed: 23640971] [CrossRef]

36.

Piccini JP, Hellkamp AS, Lokhnygina Y, et al. Relationship between time in therapeutic range and comparative treatment effect of rivaroxaban and warfarin: results from the ROCKET AF trial. J Am Heart Assoc. 2014 Apr 22;3(2):e000521. doi: 10.1161/jaha.113.000521. PMID: 24755148. [PMC free article: PMC4187517] [PubMed: 24755148] [CrossRef]

37.

Jacobs LG, Billett HH, Freeman K, et al. Anticoagulation for stroke prevention in elderly patients with atrial fibrillation, including those with falls and/or early-stage dementia: a single-center, retrospective, observational study. Am J Geriatr Pharmacother. 2009 Jun;7(3):159–66. doi: 10.1016/j.amjopharm.2009.06.002. PMID: 19616184. [PubMed: 19616184] [CrossRef]

38.

Doucet J, Greboval-Furstenfeld E, Tavildari A, et al. Which parameters differ in very old patients with chronic atrial fibrillation treated by anticoagulant or aspirin? Antithrombotic treatment of atrial fibrillation in the elderly. Fundam Clin Pharmacol. 2008 Oct;22(5):569–74. doi: 10.1111/j.1472-8206.2008.00629.x. PMID: 18844728. [PubMed: 18844728] [CrossRef]

39.

Mant J, Hobbs FD, Fletcher K, et al. Warfarin versus aspirin for stroke prevention in an elderly community population with atrial fibrillation (the Birmingham Atrial Fibrillation Treatment of the Aged Study, BAFTA): a randomised controlled trial. Lancet. 2007 Aug 11;370(9586):493–503. doi: 10.1016/s0140-6736(07)61233-1. PMID: 17693178. [PubMed: 17693178] [CrossRef]

40.

Rash A, Downes T, Portner R, et al. A randomised controlled trial of warfarin versus aspirin for stroke prevention in octogenarians with atrial fibrillation (WASPO). Age Ageing. 2007 Mar;36(2):151–6. doi: 10.1093/ageing/afl129. PMID: 17175564. [PubMed: 17175564] [CrossRef]

41.

Vemmos KN, Tsivgoulis G, Spengos K, et al. Primary prevention of arterial thromboembolism in the oldest old with atrial fibrillation-a randomized pilot trial comparing adjusted-dose and fixed low-dose coumadin with aspirin. Eur J Intern Med. 2006 Jan;17(1):48–52. doi: 10.1016/j.ejim.2005.08.005. PMID: 16378886. [PubMed: 16378886] [CrossRef]

42.

Tsivgoulis G, Spengos K, Zakopoulos N, et al. Efficacy of anticoagulation for secondary stroke prevention in older people with non-valvular atrial fibrillation: a prospective case series study. Age Ageing. 2005 Jan;34(1):35–40. doi: 10.1093/ageing/afi004. PMID: 15591481. [PubMed: 15591481] [CrossRef]

43.

Shireman TI, Howard PA, Kresowik TF, et al. Combined anticoagulant-antiplatelet use and major bleeding events in elderly atrial fibrillation patients. Stroke. 2004 Oct;35(10):2362–7. doi: 10.1161/01.STR.0000141933.75462.c2. PMID: 15331796. [PubMed: 15331796] [CrossRef]

44.

Hobbs FD, Roalfe AK, Lip GY, et al. Performance of stroke risk scores in older people with atrial fibrillation not taking warfarin: comparative cohort study from BAFTA trial. BMJ. 2011;342:d3653. PMID: 21700651. [PMC free article: PMC3121229] [PubMed: 21700651]

45.

Frost L, Johnsen SP, Pedersen L, et al. Atrial fibrillation or flutter and stroke: a Danish population-based study of the effectiveness of oral anticoagulation in clinical practice. J Intern Med. 2002 Jul;252(1):64–9. PMID: 12074740. [PubMed: 12074740]

46.

Eikelboom JW, Wallentin L, Connolly SJ, et al. Risk of bleeding with 2 doses of dabigatran compared with warfarin in older and younger patients with atrial fibrillation: an analysis of the randomized evaluation of long-term anticoagulant therapy (RE-LY) trial. Circulation. 2011 May 31;123(21):2363–72. doi: 10.1161/circulationaha.110.004747. PMID: 21576658. [PubMed: 21576658] [CrossRef]

47.

Cafolla A, Campanelli M, Baldacci E, et al. Oral anticoagulant therapy in Italian patients 80 yr of age or older with atrial fibrillation: A pilot study of low vs. standard PT/INR targets. Eur J Haematol. 2012 July;89(1):81–6. PMID: 2012352519. [PubMed: 22519759]

48.

Halvorsen S, Atar D, Yang H, et al. Efficacy and safety of apixaban compared with warfarin according to age for stroke prevention in atrial fibrillation: observations from the ARISTOTLE trial. Eur Heart J. 2014 Jul 21;35(28):1864–72. doi: 10.1093/eurheartj/ehu046. PMID: 24561548. [PMC free article: PMC4104493] [PubMed: 24561548] [CrossRef]

49.

Alexander JH, Andersson U, Lopes RD, et al. Apixaban 5 mg Twice Daily and Clinical Outcomes in Patients With Atrial Fibrillation and Advanced Age, Low Body Weight, or High Creatinine: A Secondary Analysis of a Randomized Clinical Trial. JAMA Cardiol. 2016 Sep 01;1(6):673–81. doi: 10.1001/jamacardio.2016.1829. PMID: 27463942. [PubMed: 27463942] [CrossRef]

50.

Halperin JL, Hankey GJ, Wojdyla DM, et al. Efficacy and safety of rivaroxaban compared with warfarin among elderly patients with nonvalvular atrial fibrillation in the Rivaroxaban Once Daily, Oral, Direct Factor Xa Inhibition Compared With Vitamin K Antagonism for Prevention of Stroke and Embolism Trial in Atrial Fibrillation (ROCKET AF). Circulation. 2014 Jul 08;130(2):138–46. doi: 10.1161/circulationaha.113.005008. PMID: 24895454. [PubMed: 24895454] [CrossRef]

51.

Lauw MN, Eikelboom JW, Coppens M, et al. Effects of dabigatran according to age in atrial fibrillation. Heart. 2017 Jul;103(13):1015–23. doi: 10.1136/heartjnl-2016-310358. PMID: 28213368. [PubMed: 28213368] [CrossRef]

52.

Hohnloser SH, Oldgren J, Yang S, et al. Myocardial ischemic events in patients with atrial fibrillation treated with dabigatran or warfarin in the RE-LY (Randomized Evaluation of Long-Term Anticoagulation Therapy) trial. Circulation. 2012 Feb 7;125(5):669–76. doi: 10.1161/circulationaha.111.055970. PMID: 22215856. [PubMed: 22215856] [CrossRef]

53.

Fosbol EL, Wang TY, Li S, et al. Safety and effectiveness of antithrombotic strategies in older adult patients with atrial fibrillation and non-ST elevation myocardial infarction. Am Heart J. 2012 Apr;163(4):720–8. doi: 10.1016/j.ahj.2012.01.017. PMID: 22520540. [PubMed: 22520540] [CrossRef]

54.

Kochar A, Hellkamp AS, Lokhnygina Y, et al. Efficacy and safety of rivaroxaban compared with warfarin in patients with carotid artery disease and nonvalvular atrial fibrillation: Insights from the ROCKET AF trial. Clin Cardiol. 2018 Jan;41(1):39–45. doi: 10.1002/clc.22846. PMID: 29389037. [PMC free article: PMC6489762] [PubMed: 29389037] [CrossRef]

55.

Hu PT, Lopes RD, Stevens SR, et al. Efficacy and Safety of Apixaban Compared With Warfarin in Patients With Atrial Fibrillation and Peripheral Artery Disease: Insights From the ARISTOTLE Trial. J Am Heart Assoc. 2017 Jan 17;6(1) doi: 10.1161/jaha.116.004699. PMID: 28096100. [PMC free article: PMC5523640] [PubMed: 28096100] [CrossRef]

56.

Westenbrink BD, Alings M, Granger CB, et al. Anemia is associated with bleeding and mortality, but not stroke, in patients with atrial fibrillation: Insights from the Apixaban for Reduction in Stroke and Other Thromboembolic Events in Atrial Fibrillation (ARISTOTLE) trial. Am Heart J. 2017 Mar;185:140–9. doi: 10.1016/j.ahj.2016.12.008. PMID: 28267467. [PubMed: 28267467] [CrossRef]

57.

De Caterina R, Andersson U, Alexander JH, et al. History of bleeding and outcomes with apixaban versus warfarin in patients with atrial fibrillation in the Apixaban for Reduction in Stroke and Other Thromboembolic Events in Atrial Fibrillation trial. Am Heart J. 2016 May;175:175–83. doi: 10.1016/j.ahj.2016.01.005. PMID: 27179738. [PubMed: 27179738] [CrossRef]

58.

Durheim MT, Cyr DD, Lopes RD, et al. Chronic obstructive pulmonary disease in patients with atrial fibrillation: Insights from the ARISTOTLE trial. Int J Cardiol. 2016 Jan 01;202:589–94. doi: 10.1016/j.ijcard.2015.09.062. PMID: 26447668. [PubMed: 26447668] [CrossRef]

59.

Vinereanu D, Stevens SR, Alexander JH, et al. Clinical outcomes in patients with atrial fibrillation according to sex during anticoagulation with apixaban or warfarin: a secondary analysis of a randomized controlled trial. Eur Heart J. 2015 Dec 07;36(46):3268–75. doi: 10.1093/eurheartj/ehv447. PMID: 26371113. [PubMed: 26371113] [CrossRef]

60.

Lip GY, Eikelboom J, Yusuf S, et al. Modification of outcomes with aspirin or apixaban in relation to female and male sex in patients with atrial fibrillation: a secondary analysis of the AVERROES study. Stroke. 2014 Jul;45(7):2127–30. doi: 10.1161/strokeaha.114.005746. PMID: 24916911. [PubMed: 24916911] [CrossRef]

61.

Ezekowitz JA, Lewis BS, Lopes RD, et al. Clinical outcomes of patients with diabetes and atrial fibrillation treated with apixaban: results from the ARISTOTLE trial. Eur Heart J Cardiovasc Pharmacother. 2015 Apr;1(2):86–94. doi: 10.1093/ehjcvp/pvu024. PMID: 27533976. [PubMed: 27533976] [CrossRef]

62.

Bansilal S, Bloomgarden Z, Halperin JL, et al. Efficacy and safety of rivaroxaban in patients with diabetes and nonvalvular atrial fibrillation: the Rivaroxaban Once-daily, Oral, Direct Factor Xa Inhibition Compared with Vitamin K Antagonism for Prevention of Stroke and Embolism Trial in Atrial Fibrillation (ROCKET AF Trial). Am Heart J. 2015 Oct;170(4):675–82.e8. doi: 10.1016/j.ahj.2015.07.006. PMID: 26386791. [PubMed: 26386791] [CrossRef]

63.

Brambatti M, Darius H, Oldgren J, et al. Comparison of dabigatran versus warfarin in diabetic patients with atrial fibrillation: Results from the RE-LY trial. Int J Cardiol. 2015 Oct 01;196:127–31. doi: 10.1016/j.ijcard.2015.05.141. PMID: 26093161. [PubMed: 26093161] [CrossRef]

64.

Alexander JH, Lopes RD, Thomas L, et al. Apixaban vs. warfarin with concomitant aspirin in patients with atrial fibrillation: insights from the ARISTOTLE trial. Eur Heart J. 2014 Jan;35(4):224–32. doi: 10.1093/eurheartj/eht445. PMID: 24144788. [PubMed: 24144788] [CrossRef]

65.

Shah R, Hellkamp A, Lokhnygina Y, et al. Use of concomitant aspirin in patients with atrial fibrillation: Findings from the ROCKET AF trial. Am Heart J. 2016 Sep;179:77–86. doi: 10.1016/j.ahj.2016.05.019. PMID: 27595682. [PubMed: 27595682] [CrossRef]

66.

Xu H, Ruff CT, Giugliano RP, et al. Concomitant Use of Single Antiplatelet Therapy With Edoxaban or Warfarin in Patients With Atrial Fibrillation: Analysis From the ENGAGE AF-TIMI48 Trial. J Am Heart Assoc. 2016 Feb 23;5(2) doi: 10.1161/jaha.115.002587. PMID: 26908401. [PMC free article: PMC4802446] [PubMed: 26908401] [CrossRef]

67.

Vemulapalli S, Hellkamp AS, Jones WS, et al. Blood pressure control and stroke or bleeding risk in anticoagulated patients with atrial fibrillation: Results from the ROCKET AF Trial. Am Heart J. 2016 Aug;178:74–84. doi: 10.1016/j.ahj.2016.05.001. PMID: 27502854. [PubMed: 27502854] [CrossRef]

68.

Magnani G, Giugliano RP, Ruff CT, et al. Efficacy and safety of edoxaban compared with warfarin in patients with atrial fibrillation and heart failure: insights from ENGAGE AF-TIMI 48. Eur J Heart Fail. 2016 Sep;18(9):1153–61. doi: 10.1002/ejhf.595. PMID: 27349698. [PubMed: 27349698] [CrossRef]

69.

van Diepen S, Hellkamp AS, Patel MR, et al. Efficacy and safety of rivaroxaban in patients with heart failure and nonvalvular atrial fibrillation: insights from ROCKET AF. Circ Heart Fail. 2013 Jul;6(4):740–7. doi: 10.1161/circheartfailure.113.000212. PMID: 23723250. [PubMed: 23723250] [CrossRef]

70.

Verdecchia P, Reboldi G, Angeli F, et al. Dabigatran vs. warfarin in relation to the presence of left ventricular hypertrophy in patients with atrial fibrillation- the Randomized Evaluation of Long-term anticoagulation therapY (RE-LY) study. Europace. 2017 May 17 doi: 10.1093/europace/eux022. PMID: 28520924. [PMC free article: PMC5834147] [PubMed: 28520924] [CrossRef]

71.

Rao MP, Vinereanu D, Wojdyla DM, et al. Clinical Outcomes and History of Fall in Patients with Atrial Fibrillation Treated with Oral Anticoagulation: Insights From the ARISTOTLE Trial. Am J Med. 2017 Nov 6 doi: 10.1016/j.amjmed.2017.10.036. PMID: 29122636. [PubMed: 29122636] [CrossRef]

72.

Melloni C, Dunning A, Granger CB, et al. Efficacy and Safety of Apixaban Versus Warfarin in Patients with Atrial Fibrillation and a History of Cancer: Insights from the ARISTOTLE Trial. Am J Med. 2017 Dec;130(12):1440–8.e1. doi: 10.1016/j.amjmed.2017.06.026. PMID: 28739198. [PubMed: 28739198] [CrossRef]