Abstract
Background
For patients with atrial fibrillation seen in the emergency department (ED) following a transient ischemic attack (TIA) or minor stroke, the impact of initiating oral anticoagulation immediately rather than deferring the decision to outpatient follow‐up is unknown.
Methods and Results
We conducted a planned secondary data analysis of a prospective cohort of 11 507 adults in 13 Canadian EDs between 2006 and 2018. Patients were eligible if they were aged 18 years or older, with a final diagnosis of TIA or minor stroke with previously documented or newly diagnosed atrial fibrillation. The primary outcome was subsequent stroke, recurrent TIA, or all‐cause mortality within 90 days of the index TIA diagnosis. Secondary outcomes included stroke, recurrent TIA, or death and rates of major bleeding. Of 11 507 subjects with TIA/minor stroke, atrial fibrillation was identified in 11.2% (1286, mean age, 77.3 [SD 11.1] years, 52.4% male). Over half (699; 54.4%) were already taking anticoagulation, 89 (6.9%) were newly prescribed anticoagulation in the ED. By 90 days, 4.0% of the atrial fibrillation cohort had experienced a subsequent stroke, 6.5% subsequent TIA, and 2.6% died. Results of a multivariable logistic regression indicate no association between prescribed anticoagulation in the ED and these 90‐day outcomes (composite odds ratio, 1.37 [95% CI, 0.74–2.52]). Major bleeding was found in 5 patients, none of whom were in the ED‐initiated anticoagulation group.
Conclusions
Initiating oral anticoagulation in the ED following new TIA was not associated with lower recurrence rates of neurovascular events or all‐cause mortality in patients with atrial fibrillation.
Keywords: anticoagulation, emergency medicine, stroke, transient ischemic attack
Subject Categories: Cerebrovascular Disease/Stroke, Ischemic Stroke, Transient Ischemic Attack (TIA)
Nonstandard Abbreviations and Acronyms
- DOACs
direct oral anticoagulants
Clinical Perspective.
What Is New?
For patients with atrial fibrillation seen in the emergency department following a transient ischemic attack (TIA) or minor stroke, the impact of initiating oral anticoagulation immediately rather than deferring the decision to outpatient follow‐up is unknown.
We sought to determine the association of starting anticoagulation in the emergency department and the outcome of stroke, TIA, or death at 90 days.
What Are the Clinical Implications?
We found that emergency department‐initiated anticoagulation did not lower recurrence rates of TIA, stroke, or all‐cause mortality at 90 days for patients with atrial fibrillation or a history of atrial fibrillation who presented with a new TIA or minor stroke.
The role and timing of emergency department‐prescribed anticoagulation in patients with atrial fibrillation and a new TIA who are not already on anticoagulation needs further investigation.
Transient ischemic attacks (TIA) are common and may herald a future stroke, with an estimated risk of 3.5% within 48 hours, 4% to 10% within 7 days, and 8% to 12% by 90 days of the index event. 1 , 2 , 3 , 4 TIAs are caused by ischemia from vascular disease or emboli, of which a significant proportion is attributable to cardioembolic stroke from atrial fibrillation. It is well established that patients with atrial fibrillation have nearly a 5‐fold higher risk of stroke, higher recurrence rates, greater disability, longer hospital stays, higher mortality rates, and greater medical costs than those in sinus rhythm. 5 , 6 , 7 , 8 , 9 A patient with atrial fibrillation and a prior stroke or TIA has a 2‐fold higher risk of recurrence in comparison to patients without, with 1 study reporting a 5‐year recurrence rate after atrial fibrillation‐related stroke as high as 21.5%. 10 , 11 , 12
This elevated baseline risk for stroke or TIA in patients with atrial fibrillation in addition to the high risk of recurrence is well recognized and provides the rationale for early primary and secondary stroke interventions. Oral anticoagulants taken long term prevent ischemic stroke and systemic embolism in patients with atrial fibrillation. The Canadian Best Practice Guidelines recommend that patients with nonvalvular atrial fibrillation receive oral anticoagulation with direct thrombin inhibitors (eg, dabigatran) or Factor Xa inhibitors (eg, apixaban, rivaroxaban) following ischemic stroke or TIA. 13 These interventions have been shown to be as effective or superior to the previous standard therapy with warfarin in reducing the risk for stroke, embolic events, and all‐cause death. 13 , 14 , 15 , 16 , 17 , 18 , 19 , 20
However, the optimal timing for initiating anticoagulation in patients following TIA remains controversial, in part because the risks and benefits of anticoagulation likely differ in the short term versus long term. Experts often recommend anticoagulation begin on the same day or 1 day post event, arguing efficacy and medication adherence are best emphasized at the index encounter. Others wonder about drug safety, dose titration, hemorrhagic transformation, and continuity of care following emergency prescribing. 21 As a result, there is disagreement regarding whether oral anticoagulation is best initiated in the emergency department (ED) at the time of an index TIA or minor stroke. However, the early risks and benefits of initiating oral anticoagulation in the ED in patients with atrial fibrillation and new TIA are not known.
The objective of this study is to investigate whether patients with atrial fibrillation and a new TIA or minor stroke have lower rates of stroke, TIA, or all‐cause mortality 90 days after ED‐prescribed anticoagulation.
Methods
Design
The data that support the findings of this study are available from Dr. Jeff Perry upon reasonable request. This is a research ethics board approved planned secondary data analysis of a prospective multicentered cohort study of 11 507 adults who presented to 13 Canadian EDs over a 12‐year period with TIA or minor stroke as their final ED diagnosis. Consent was waived for patients; however, they provided verbal consent at the time of follow‐up telephone calls.
Setting
We combined subjects from the derivation (3900 patients from 8 EDs from 2006 to 2011) and validation (7607 patients from 13 EDs from 2012 to 2017) phases used to develop the Canadian TIA Score (Figure 1). 22 , 23 Data were extracted from data collection forms completed at the time of the index ED visit by the treating emergency physician or neurologist and by trained research personnel from the hospital medical records, outpatient consultations, and diagnostic imaging reports.
Figure 1. Consort diagram of patients from 13 Canadian EDs over a 12‐year period with TIA or minor stroke as their final ED diagnosis.
AF indicates atrial fibrillation; ED, emergency department; and TIA, transient ischemic attack.
Population
Adult patients with a final ED diagnosis of TIA or minor stroke were eligible for the parent studies. All those with >24‐hour deficit, decreased level of consciousness, presenting >7 days after event, or given fibrinolytics were excluded. Only those with previously documented or newly diagnosed atrial fibrillation at the time of the index ED visit were retained for this current study.
Outcome Measures
The primary outcome was the composite of subsequent stroke (ischemic or hemorrhagic), subsequent TIA, or death within 90 days of the index diagnosis of TIA in the ED. Secondary outcomes included stroke, TIA, or death and rates of major bleeding (major gastrointestinal, external, or intracerebral) at days 2, 7, 30, and 90.
Trained research personnel scrutinized hospital medical records and conducted scripted telephone follow‐up on days 7 and 90 to identify possible outcomes, which were then reviewed by blinded local adjudication committees composed of 2 stroke neurologists and 1 experienced ED physician. A positive outcome was confirmed when 2 of 3 members independently agreed.
Statistical Analysis
Descriptive statistics are presented using means with SDs for continuous variables and frequencies with percentages for categorical variables. A multivariable logistic regression model was used to evaluate the association between stroke, TIA, and all‐cause mortality within 90 days of index visit with variables deemed important. These included the Canadian TIA Score, sex, CHA2DS2‐VASc score, international normalized ratio, patient already taking anticoagulant medication, and ED‐prescribed anticoagulant. Multiple imputations were performed to handle missing data. The results of the model are presented as odds ratios (ORs) with 95% CIs. We conducted all statistical analyses using SAS version 9.4 (SAS Institute Inc., Cary, NC).
Results
Baseline Characteristics
Of 11 507 patients enrolled in the initial studies, 1286 patients with atrial fibrillation were selected for this substudy. Table 1 shows the demographics of this cohort. The mean age of the patients was 77.3 years (SD 11.1) and 52.4% were male. The mean CHA2DS2‐VASc was 4.0 (4.6%–6.7% risk of stroke/TIA/systemic embolism). In the ED, 18.6% and 6.9% were started new prescriptions of antiplatelets and anticoagulants as shown in Table 2. Patients were most commonly on warfarin for anticoagulation at 37.9%, with 23.3% being treated with a direct oral anticoagulants (DOACs). Warfarin was started in the ED for 3.8% of patients versus 3.1% for DOACs between 2006 and 2017.
Table 1.
Clinical Characteristics of Patients With Atrial Fibrillation and New TIA
| Characteristic, n (%) | No. =1286 |
|---|---|
| Demographics | |
| Age, y, mean (SD) | 77.3 (11.1) |
| Male sex | 674 (52.4) |
| Medical history | |
| Congestive heart failure | 128 (10.0) |
| Hypertension | 945 (73.5) |
| Known prior stroke or TIA | 569 (44.2) |
| Diabetes | 276 (21.5) |
| Peripheral vascular disease | 79 (6.1) |
| CHADS265, mean (SD) | 2.8 (1.4) |
| CHA2D2S‐VASc, mean (SD) | 4.0 (1.7) |
| Antiplatelet medications | |
| Already taken | 507 (39.4) |
| Started in ED | 239 (18.6) |
| Dual | |
| Aspirin+clopidogrel | |
| Already taken | 36 (2.8) |
| Started in ED | 3 (0.2) |
| Aspirin+dipyridamole | |
| Already taken | 12 (0.9) |
| Started in ED | 31 (2.4) |
| Aspirin+prasugrel | |
| Already taken | 1 (0.1) |
| Started in ED | 0 (0.0) |
| Aspirin+ticlopidine | |
| Already taken | 1 (0.1) |
| Started in ED | 1 (0.1) |
| Aspirin | |
| Already taken | 415 (32.3) |
| Started in ED | 149 (11.6) |
| Clopidogrel | |
| Already taken | 40 (3.1) |
| Started in ED | 52 (4.0) |
| Ticlopidine | |
| Already taken | 2 (0.2) |
| Started in ED | 3 (0.2) |
| Anticoagulant medications | |
| Already taken | 699 (54.4) |
| Started in ED | 89 (6.9) |
| Warfarin | |
| Already taken | 439 (34.1) |
| Started in ED | 49 (3.8) |
| Direct oral anticoagulant | |
| Apixaban | |
| Already taken | 88 (6.8) |
| Started in ED | 15 (1.2) |
| Dabigatran | |
| Already taken | 86 (6.7) |
| Started in ED | 10 (0.8) |
| Rivaroxaban | |
| Already taken | 86 (6.7) |
| Started in ED | 15 (1.2) |
CHADS265 indicates congestive heart failure, hypertension, diabetes, stroke, sex (female), age>65; CHA2DS2‐VASc, congestive heart failure, hypertension, age≥75 (doubled), diabetes, stroke (doubled), vascular disease, age 65 to 74, and sex category (female); ED, emergency department; and TIA, transient ischemic attack.
Table 2.
Clinical Characteristics of Patients With Atrial Fibrillation and New TIA on ED‐Prescribed Anticoagulant or No Anticoagulant
| Characteristic, n (%) | ED‐prescribed anticoagulant | No anticoagulant |
|---|---|---|
| N=89 | N=499 | |
| Demographics | ||
| Age, y, mean (SD) | 76 (10.6) | 75.3 (13.1) |
| Male sex | 44 (49.4) | 265 (53.1) |
| Medical history | ||
| Congestive heart failure | 4 (4.5) | 38 (7.6) |
| Hypertension | 61 (68.5) | 352 (70.5) |
| Known prior stroke or TIA | 24 (27.0) | 178 (35.7) |
| Diabetes | 24 (27.0) | 96 (19.2) |
| Peripheral vascular disease | 5 (5.6) | 32 (6.4) |
| CHADS265, mean (SD) | 7.7 (2.4) | 7.2 (2.6) |
| CHA2DS2‐VASc, mean (SD) | 2.4 (1.4) | 2.5 (1.4) |
| Antiplatelet medications | ||
| Already taken | 42 (47.2) | 336 (67.3) |
| Started in ED | 10 (11.2) | 164 (32.9) |
| Dual | ||
| Aspirin+clopidogrel | ||
| Already taken | 0 (0.0) | 32 (6.4) |
| Started in ED | 0 (0.0) | 3 (0.6) |
| Aspirin+dipyridamole | ||
| Already taken | 0 (0.0) | 11 (2.2) |
| Started in ED | 1 (1.1) | 28 (5.6) |
| Aspirin+prasugrel | ||
| Already taken | 0 (0.0) | 1 (0.2) |
| Started in ED | 0 (0.0) | 0 (0.0) |
| Aspirin+ticlopidine | ||
| Already taken | 0 (0.0) | 1 (0.2) |
| Started in ED | 0 (0.0) | 0 (0.0) |
| Aspirin | ||
| Already taken | 40 (44.9) | 259 (51.9) |
| Started in ED | 9 (10.1) | 82 (16.4) |
| Clopidogrel | ||
| Already taken | 2 (2.2) | 30 (6) |
| Started in ED | 0 (0.0) | 49 (9.8) |
| Ticlopidine | ||
| Already taken | 0 (0.0) | 2 (0.4) |
| Started in ED | 0 (0.0) | 2 (0.4) |
| Anticoagulant medications | ||
| Already taken | 1 (1.1) | 0 (0.0) |
| Started in ED | 89 (100.0) | 0 (0.0) |
| Warfarin | ||
| Already taken | 1 (1.1) | 0 (0.0) |
| Started in ED | 49 (55.1) | 0 (0.0) |
| Direct oral anticoagulant | ||
| Apixaban | ||
| Already taken | 0 (0.0) | 0 (0.0) |
| Started in ED | 15 (16.9) | 0 (0.0) |
| Dabigatran | ||
| Already taken | 0 (0.0) | 0 (0.0) |
| Started in ED | 10 (11.2) | 0 (0.0) |
| Rivaroxaban | ||
| Already taken | 0 (0.0) | 0 (0.0) |
| Started in ED | 15 (16.9) | 0 (0.0) |
CHADS265 indicates congestive heart failure, hypertension, diabetes, stroke, sex (female), age>65; CHA2DS2‐VASc, congestive heart failure, hypertension, age≥75 (doubled), diabetes, stroke (doubled), vascular disease, age 65 to 74, and sex category (female); ED, emergency department; and TIA, transient ischemic attack.
Table 3 shows that at 90 days, 52 (4.0%) patients had a subsequent stroke, 83 (6.5%) TIA, and 34 (2.6%) died. Five (0.4%) patients had major bleeds all in the form of an intracerebral hemorrhage.
Table 3.
Primary and Secondary Outcomes of Patients With Atrial Fibrillation and New TIA
| Outcomes | No. =1286 |
|---|---|
| Ischemic stroke | |
| Cumulative stroke ≤2 d from index visit | 18 (1.4) |
| Cumulative stroke ≤7 d from index visit | 28 (2.2) |
| Cumulative stroke ≤30 d from index visit | 41 (3.2) |
| Cumulative stroke ≤90 d from index visit | 52 (4.0) |
| TIA | |
| Cumulative repeat TIA ≤2 d from index visit | 21 (1.6) |
| Cumulative repeat TIA ≤7 d from index visit | 29 (2.3) |
| Cumulative repeat TIA ≤30 d from index visit | 57 (4.4) |
| Cumulative repeat TIA ≤90 d from index visit | 83 (6.5) |
| Bleeding | |
| Intracerebral hemorrhage ≤90 d from index visit | 5 (0.4) |
| Subarachnoid hemorrhage ≤90 d from index visit | 0 (0.0) |
| Mortality | |
| Cumulative mortality at ≤2 d from index visit | 2 (0.2) |
| Cumulative mortality at ≤7 d from index visit | 3 (0.2) |
| Cumulative mortality at ≤30 d from index visit | 10 (0.8) |
| Cumulative mortality at ≤90 d from index visit | 34 (2.6) |
| Stroke, TIA, or death at ≤90 d from index visit | 150 (11.7) |
TIA indicates transient ischemic attack.
The composite rates of stroke, TIA, or death within 90 days of the index event were 16.9% for ED‐prescribed anticoagulation compared to 10.2% if they were already taking an anticoagulant and 12.8% if they were on no anticoagulant with no immediate prescription for one from the ED (Table 4). Between‐group comparisons in those with and without subsequent stroke, TIA, or death within 90 days revealed no statistical difference in demographics, risk scores, and antiplatelet or anticoagulant choice (Table 5). There was no statistical difference in stroke, TIA, and mortality if the patient was on an anticoagulant before the index event (49.2% versus 54.9%). There was no statistical difference in stroke, TIA, and mortality between DOAC and warfarin (25.0% versus 10.2%).
Table 4.
Primary and Secondary Outcomes in Atrial Fibrillation Patients With ED‐Prescribed Anticoagulant Versus Those Already on an Anticoagulant Versus No Anticoagulant
| ED‐prescribed anticoagulant | Already taken anticoagulant | No anticoagulant | |
|---|---|---|---|
| N=89 | N=699 | N=499 | |
| Ischemic stroke | |||
| Cumulative stroke ≤2 d from index visit | 4 (4.5) | 7 (1.0) | 7 (1.4) |
| Cumulative stroke ≤7 d from index visit | 7 (7.9) | 11 (1.6) | 10 (2.0) |
| Cumulative stroke ≤30 d from index visit | 8 (9.0) | 16 (2.3) | 17 (3.4) |
| Cumulative stroke ≤90 d from index visit | 9 (10.1) | 23 (3.3) | 20 (4.0) |
| TIA | |||
| Cumulative repeat TIA ≤2 d from index visit | 4 (4.5) | 8 (1.1) | 9 (1.8) |
| Cumulative repeat TIA ≤7 d from index visit | 4 (4.5) | 15 (2.1) | 10 (2.0) |
| Cumulative repeat TIA ≤30 d from index visit | 5 (5.6) | 32 (4.6) | 20 (4.0) |
| Cumulative repeat TIA ≤90 d from index visit | 7 (7.9) | 47 (6.7) | 29 (5.8) |
| Bleeding | |||
| Intracerebral hemorrhage ≤90 d from index visit | 0 (0.0) | 2 (0.3) | 3 (0.6) |
| Subarachnoid hemorrhage ≤90 d from index visit | 0 (0.0) | 0 (0.0) | 0 (0.0) |
| Mortality | |||
| Cumulative mortality at ≤2 d from index visit | 0 (0.0) | 1 (0.1) | 1 (0.2) |
| Cumulative mortality at ≤7 d from index visit | 1 (1.1) | 1 (0.1) | 1 (0.2) |
| Cumulative mortality at ≤30 d from index visit | 1 (1.1) | 5 (0.7) | 4 (0.8) |
| Cumulative mortality at ≤90 d from index visit | 3 (3.4) | 14 (2.0) | 17 (3.4) |
| Stroke, TIA, or death at ≤90 d from index visit | 15 (16.9) | 71 (10.2) | 64 (12.8) |
ED indicates emergency department; and TIA, transient ischemic attack.
Table 5.
Comparison of Clinical Characteristics of Patients With and Without Recurrent TIA/Stroke/Death Within 90 Days of Index Event
| Clinical feature, n (%) | TIA/stroke/death recurrence | No TIA, stroke, or death |
|---|---|---|
| N=132 | N=1154 | |
| Demographics | ||
| Age, y, mean, SD | 78.7 (9.4) | 77.2 (11.3) |
| Male sex | 68 (51.5) | 606 (52.5) |
| Medical history | ||
| Congestive heart failure | 16 (12.1) | 112 (9.7) |
| Hypertension | 57 (43.2) | 512 (44.4) |
| Known prior stroke or TIA | 28 (21.2) | 248 (21.5) |
| Diabetes | 28 (21.2) | 248 (21.5) |
| Peripheral vascular disease | 8 (6.1) | 71 (6.2) |
| CTS, mean (SD) | 7.4 (2.7) | 6.5 (2.5) |
| CHADS265, mean (SD) | 2.8 (1.4) | 2.8 (1.4) |
| CHA2DS2‐VASc, mean (SD) | 4.1 (1.7) | 4.0 (1.8) |
| Diagnostic tests in ED | ||
| Computed tomography of head | 128 (97) | 1109 (96.1) |
| Acute infarct | 4 (3.0) | 60 (5.2) |
| Old infarct | 60 (45.5) | 422 (36.6) |
| International normalized ratio (patients on warfarin n=439) | ||
| <2 | 19 (39.6) | 127 (32.5) |
| 2–3 | 18 (37.5) | 196 (50.1) |
| >3 | 7 (14.6) | 41 (10.5) |
| Antiplatelet medications | ||
| Already taken | 59 (44.7) | 448 (38.8) |
| Started in ED | 23 (17.4) | 216 (18.7) |
| Dual | ||
| Aspirin+clopidogrel | ||
| Already taken | 1 (0.8) | 35 (3.0) |
| Started in ED | 0 (0.0) | 3 (0.3) |
| Aspirin+dipyridamole | ||
| Already taken | 1 (0.8) | 11 (1.0) |
| Started in ED | 2 (1.5) | 29 (2.5) |
| Aspirin+prasugrel | ||
| Already taken | 1 (0.8) | 0 (0.0) |
| Started in ED | 0 (0.0) | 0 (0.0) |
| Aspirin+ticlopidine | ||
| Already taken | 0 (0.0) | 1 (0.1) |
| Started in ED | 0 (0.0) | 1 (0.1) |
| Aspirin | ||
| Already taken | 51 (38.6) | 364 (31.5) |
| Started in ED | 15 (11.4) | 133 (11.5) |
| Clopidogrel | ||
| Already taken | 5 (3.8) | 35 (3.0) |
| Started in ED | 5 (3.8) | 47 (4.1) |
| Ticlopidine | ||
| Already taken | 0 (0.0) | 2 (0.2) |
| Started in ED | 0 (0.0) | 3 (0.3) |
| Anticoagulant medications | ||
| Already taken | 65 (49.2) | 634 (54.9) |
| Started in ED | 15 (11.4) | 74 (6.4) |
| Warfarin | ||
| Already taken | 48 (36.4) | 391 (33.9) |
| Started in ED | 5 (3.8) | 44 (3.8) |
| Direct oral anticoagulants | ||
| Apixaban | ||
| Already taken | 7 (5.3) | 81 (7.0) |
| Started in ED | 5 (3.8) | 10 (0.9) |
| Dabigatran | ||
| Already taken | 6 (4.5) | 80 (6.9) |
| Started in ED | 3 (2.3) | 7 (0.6) |
| Rivaroxaban | ||
| Already taken | 4 (3.0) | 82 (7.1) |
| Started in ED | 3 (2.3) | 12 (1.0) |
| Outcomes | ||
| Ischemic stroke | ||
| Cumulative stroke ≤2 d from index visit | 18 (13.6) | 0 (0.0) |
| Cumulative stroke ≤7 d from index visit | 28 (21.2) | 0 (0.0) |
| Cumulative stroke ≤30 d from index visit | 41 (31.1) | 0 (0.0) |
| Cumulative stroke ≤90 d from index visit | 52 (39.4) | 0 (0.0) |
| TIA | ||
| Cumulative repeat TIA ≤2 d from index visit | 21 (15.9) | 0 (0.0) |
| Cumulative repeat TIA ≤7 d from index visit | 29 (22.0) | 0 (0.0) |
| Cumulative repeat TIA ≤30 d from index visit | 57 (43.2) | 0 (0.0) |
| Cumulative repeat TIA ≤90 d from index visit | 83 (62.9) | 0 (0.0) |
| Bleeding | ||
| Intracerebral hemorrhage ≤90 d from index visit | 5 (3.8) | 0 (0.0) |
| Subarachnoid hemorrhage ≤90 d from index visit | 0 (0.0) | 0 (0.0) |
| Mortality | ||
| Cumulative mortality at ≤2 d from index visit | 1 (0.8) | 1 (0.1) |
| Cumulative mortality at ≤7 d from index visit | 2 (1.5) | 1 (0.1) |
| Cumulative mortality at ≤30 d from index visit | 6 (4.5) | 4 (0.3) |
| Cumulative mortality at ≤90 d from index visit | 16 (12.1) | 18 (1.6) |
| Stroke, TIA, or death at ≤90 d from index visit | 132 (100) | 18 (1.6) |
CHADS265 indicates congestive heart failure, hypertension, diabetes, stroke, sex (female), age>65); CHA2DS2‐VASc, congestive heart failure, hypertension, age≥75 (doubled), diabetes, stroke (doubled), vascular disease, age 65 to 74, and sex category (female); CTS, Canadian TIA Score; ED, emergency department; and TIA, transient ischemic attack.
The rates of anticoagulation prescribed in the ED did not increase over the time course of this study. Over this time period the rates of stroke and TIA within 90 days did not show a reduction in events (Figure 2).
Figure 2. Rates of anticoagulant prescription and outcomes by year.
ED indicates emergency department; and TIA, transient ischemic attack.
Multivariable logistic regression in Table 6 (area under the curve, 0.64 [95% CI, 0.59–0.68], likelihood ratio P value=0.004) found an OR of stroke, TIA, or death within 90 days to be 0.74 (95% CI, 0.44–1.24) if these patients were already receiving anticoagulation as compared to an OR of 1.37 (95% CI, 0.74–2.52) with ED‐prescribed anticoagulation.
Table 6.
Regression Model – Stroke, TIA, or Death ≤90 Days
| Variable | Odds ratio | 95% CI | |
|---|---|---|---|
| Risk as per Canadian TIA Score* | |||
| Low | ref | ||
| Medium | 3.18 | 1.47 | 6.85 |
| High | 1.81 | 0.88 | 3.73 |
| Male sex | 1.08 | 0.76 | 1.52 |
| Age | 1.02 | 1.00 | 1.04 |
| Congestive heart failure | 1.45 | 0.86 | 2.46 |
| Hypertension | 0.91 | 0.61 | 1.35 |
| Peripheral vascular disease | 1.08 | 0.56 | 2.11 |
| Diabetes | 0.91 | 0.60 | 1.39 |
| INR | |||
| On warfarin and 2<INR<3 | ref | ||
| Not on warfarin | 1.01 | 0.54 | 1.88 |
| On warfarin and INR<2 | 1.47 | 0.78 | 2.79 |
| On warfarin and INR>3 | 1.60 | 0.65 | 3.95 |
| Already taking anticoagulant medications | 0.74 | 0.44 | 1.24 |
| Emergency department‐prescribed anticoagulant | 1.37 | 0.74 | 2.52 |
Area under the curve: 0.64 (95% CI, 0.59–0.68); likelihood ratio P value=0.004. INR indicates international normalized ratio; and TIA, transient ischemic attack.
Canadian TIA Score: low=from −3 to 3; medium=4 to 8; high=9 to 14.
Discussion
This planned secondary data analysis of a prospective multicentered cohort study of 11 507 adults identified 1286 with atrial fibrillation or a history of atrial fibrillation who would be recommended for anticoagulation by the Canadian Best Practice Guidelines. 18 These patients at our 13 Canadian EDs were diagnosed with TIA or minor stroke as their final ED diagnosis. We found that ED‐prescribed anticoagulation, compared to patients who did not receive anticoagulation by an ED physician, did not lower recurrence rates of TIA, stroke, or all‐cause mortality at 90 days. The overall rates of anticoagulation for these patients did not change between 2008 and 2017, nor was there a large uptake in the use of DOACs.
It is unclear why in this study ED‐prescribed anticoagulation did not have a more favorable outcome but potentially related to the increased baseline risk of this cohort and the caution emergency physicians may have when prescribing or altering a patient's anticoagulant in these complex clinical scenarios. This is reflected in a 2016 survey by Perry et al, where respondents indicated that patients with subsequent risk of stroke above 5% may benefit from more comprehensive investigations and admission at the time of the index ED visit. 24
ED prescribing rates over time were reviewed as this study was started in 2006 when the evidence for DOACs were evolving. A subgroup analysis comparing ED‐prescribed DOACs versus warfarin was completed. With the limitations associated with warfarin's unpredictable pharmacodynamics and pharmacokinetics, prescription rates by emergency physicians were anticipated to have varied across time. This trend in the underuse of oral anticoagulation by physicians despite societal guidelines is in keeping with prior studies. 7 , 24 , 25 , 26
A majority of our cohort with TIA were also being treated with warfarin instead of a DOAC, as recommended by societal guidelines, which may explain the increased OR with ED‐prescribed anticoagulants. 15 , 16 , 17 , 18 The bulk of our ischemic strokes occurred in <30 days, and given it can take up to 14 days to reach steady state with warfarin, 27 this OR likely reflects anticoagulant choice and less the ED‐prescribed anticoagulation. Nevertheless, our subgroup analysis did not find a benefit in our DOAC group either.
Limitations
This was a large multicentered study from many sites across Canada, including both academic and community hospitals. Nevertheless, a limitation in our study is the small number of patients with TIA taking or subsequently prescribed anticoagulants.
Another limitation is we cannot be certain that patients were compliant with taking their prescribed anticoagulation. During the 90‐day follow‐up, it was never clarified if the patient started the anticoagulant in the ED, the next day, a week later, only after consulting the neurologist, or even if the prescription was filled. We do not know how many patients were started on anticoagulants post ED discharge but within the 90 days. Lastly, it is possible that several other factors, including patient preference, could affect an emergency physician's decision whether or not to give a patient anticoagulants in the ED. The lack of evidence during the early stages of DOAC implementation and comfort level using or not using warfarin, especially in places with limited access to bloodwork, are all factors that were not captured in this study and could potentially serve as serious confounders that are not equally distributed among our study groups.
Policy and Research Implications
The results of this study provide insufficient evidence to adjust or change the current guidelines, and we reinforce that all patients with nonvalvular atrial fibrillation and ischemic stroke or TIA should continue to receive oral anticoagulation with direct thrombin inhibitors or Factor Xa inhibitors. 13 The optimal timing for initiating anticoagulation in patients with TIA remains unknown. Given the devastating effects of stroke and the evidence supporting early and aggressive primary and secondary prevention, further research is needed to better clarify the timing of anticoagulation in patients with atrial fibrillation and a new TIA or minor stroke who are not already on anticoagulation. This high‐risk population will become ever more common in the ED with our aging population. Not only does the prevalence of atrial fibrillation increase with age but so does the risk of stroke secondary to this arrhythmia, increasing from 1.5% in those 50 to 59 years, to 23.5% in patients between 80 and 95 years of age. 5
Conclusions
We found that ED‐initiated anticoagulation did not lower recurrence rates of TIA, stroke, or all‐cause mortality at 90 days for patients with atrial fibrillation or a history of atrial fibrillation who presented with a new TIA or minor stroke who were not previously prescribed anticoagulants. The role and timing of ED‐prescribed anticoagulation in patients with atrial fibrillation and a new TIA who are not already on anticoagulation need further investigation.
Sources of Funding
This work was supported by the Canadian Institutes of Health Research. The funder had no role in the study design; in the collection, analysis, and interpretation of data; in the writing of the report; and in the decision to submit the article for publication. Researchers are independent from funders and all authors, external and internal, had full access to all of the data (including statistical reports and tables) in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.
Disclosures
Dr Perry is supported by a midcareer award from the Heart and Stroke Foundation of Ontario. The remaining authors have no disclosures to report.
Acknowledgments
The authors thank Angela Marcantonio and Gabriel Sandino‐Gold.
For Sources of Funding and Disclosures, see page 9.
References
- 1. Hill MD, Yiannakoulias N, Jeerakathil T, Tu JV, Svenson LW, Schopflocher DP. The high risk of stroke immediately after transient ischemic attack. Neurology. 2004;62:2015–2020. doi: 10.1212/01.WNL.0000129482.70315.2F [DOI] [PubMed] [Google Scholar]
- 2. Wu CM, McLaughlin K, Lorenzetti DL, Hill MD, Manns BJ, Ghali WA. Early risk of stroke after transient ischemic attack: a systematic review and meta‐analysis. Arch Intern Med. 2007;167:2417–2422. doi: 10.1001/archinte.167.22.2417 [DOI] [PubMed] [Google Scholar]
- 3. Giles MF, Rothwell PM. Risk of stroke early after transient ischaemic attack: a systematic review and meta‐analysis. Lancet. 2007;6:1063–1072. doi: 10.1016/S1474-4422(07)70274-0 [DOI] [PubMed] [Google Scholar]
- 4. Lovett JK, Dennis MS, Sandercock PAG, Bamford J, Warlow CP, Rothwell PM. Very early risk of stroke after a first transient ischemic attack. Stroke. 2003;34:e138–e142. doi: 10.1161/01.STR.0000080935.01264.91 [DOI] [PubMed] [Google Scholar]
- 5. Wolf PA, Abbott RD, Kannel WB. Atrial fibrillation as an independent risk factor for stroke: the Framingham study. Stroke. 1991;22:983–988. doi: 10.1161/01.STR.22.8.983 [DOI] [PubMed] [Google Scholar]
- 6. Mattle HP. Long‐term outcome after stroke due to atrial fibrillation. Cerebrovasc Dis. 2003;16:3–8. doi: 10.1159/000069934 [DOI] [PubMed] [Google Scholar]
- 7. Hannon N, Sheehan O, Kelly L, Marnane M, Merwick A, Moore A, Kyne L, Duggan J, Moroney J, McCormack PME, et al. Stroke associated with atrial fibrillation: incidence and early outcomes in the North Dublin population stroke study. Cerebrovasc Dis. 2010;29:43–49. doi: 10.1159/000255973 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Marini C, De Santis F, Sacco S, Russo T, Olivieri L, Totaro R, Carolei A. Contribution of atrial fibrillation to incidence and outcome of ischemic stroke: results from a population‐based study. Stroke. 2005;36:1115–1119. doi: 10.1161/01.STR.0000166053.83476.4a [DOI] [PubMed] [Google Scholar]
- 9. Candelise L, Pinardi G, Morabito A. Mortality in acute stroke with atrial fibrillation: the Italian Acute Stroke Study Group. Stroke. 1991;22:169–174. doi: 10.1161/01.STR.22.2.169 [DOI] [PubMed] [Google Scholar]
- 10. Klijn CJM, Paciaroni M, Berge E, Korompoki E, Korv J, Lal A, Putaala J, Werring DJ. Antithrombotic treatment for secondary prevention of stroke and other thromboembolic events in patients with stroke or transient ischemic attack and non‐valvular atrial fibrillation: a European Stroke Organisation guideline. Eur Stroke J. 2019;4:198–223. doi: 10.1177/2396987319841187 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Prasad V, Kaplan RM, Passman RS. New frontiers for stroke prevention in atrial fibrillation. Cerebrovasc Dis. 2012;33:199–208. [DOI] [PubMed] [Google Scholar]
- 12. Hayden DT, Hannon N, Callaly E, Ní Chróinín D, Horgan G, Kyne L, Duggan J, Dolan E, O'Rourke K, Williams D, et al. Rates and determinants of 5‐year outcomes after atrial fibrillation‐related stroke: a population study 2015. Stroke. 2015;46:3488–3493. doi: 10.1161/STROKEAHA.115.011139 [DOI] [PubMed] [Google Scholar]
- 13. Wein T, Lindsay MP, Côté R, Foley N, Berlingieri J, Bhogal S, Bourgoin A, Buck BH, Cox J, Davidson D, et al. Canadian Stroke best practice recommendations: secondary prevention of stroke, sixth edition practice guidelines, update 2017. Int J Stroke. 2018;13:420–443. doi: 10.1177/1747493017743062 [DOI] [PubMed] [Google Scholar]
- 14. Connolly SJ, Ezekowitz MD, Yusuf S, Eikelboom J, Oldgren J, Parekh A, Pogue J, Reilly PA, Themeles E, Varrone J, et al. Dabigatran versus warfarin in patients with atrial fibrillation. N Engl J Med. 2009;361:1139–1151. doi: 10.1056/NEJMoa0905561 [DOI] [PubMed] [Google Scholar]
- 15. Patel MR, Mahaffey KW, Garg J, Pan G, Singer DE, Hacke W, Breithardt G, Halperin JL, Hankey GJ, Piccini JP, et al. Rivaroxaban versus warfarin in nonvalvular atrial fibrillation. N Engl J Med. 2011;365:883–891. doi: 10.1056/NEJMoa1009638 [DOI] [PubMed] [Google Scholar]
- 16. Granger CB, Alexander JH, McMurray JJ, Lopes RD, Hylek EM, Hanna M, Al‐Khalidi HR, Ansell J, Atar D, Avezum A, et al. Apixaban versus warfarin in patients with atrial fibrillation. N Engl J Med. 2011;365:981–992. doi: 10.1056/NEJMoa1107039 [DOI] [PubMed] [Google Scholar]
- 17. Giugliano RP, Ruff CT, Braunwald E, Murphy SA, Wiviott SD, Halperin JL, Waldo AL, Ezekowitz MD, Weitz JI, Špinar J, et al. Edoxaban versus warfarin in patients with atrial fibrillation. N Engl J Med. 2013;369:2093–2104. doi: 10.1056/NEJMoa1310907 [DOI] [PubMed] [Google Scholar]
- 18. Ruff CT, Giugliano RP, Braunwald E, Hoffman EB, Deenadayalu N, Ezekowitz MD, Camm AJ, Weitz JI, Lewis BS, Parkhomenko A, et al. Comparison of the efficacy and safety of new oral anticoagulants with warfarin in patients with atrial fibrillation: a meta‐analysis of randomized trials. Lancet. 2014;383:955–962. doi: 10.1016/S0140-6736(13)62343-0 [DOI] [PubMed] [Google Scholar]
- 19. Bruins Slot KM, Berge E. Factor Xa inhibitors versus vitamin K antagonists for preventing cerebral or systemic embolism in patients with atrial fibrillation. Cochrane Database Syst Rev. 2013;3:Cd008980. doi: 10.1002/14651858.CD008980.pub2 [DOI] [PubMed] [Google Scholar]
- 20. Kwong JS, Lam Y‐Y, Yan BP, Yu C‐M. Bleeding of new oral anticoagulants for stroke prevention in atrial fibrillation: a meta‐analysis of randomized controlled trials. Cardiovasc Drugs Ther. 2013;27:23–35. doi: 10.1007/s10557-012-6426-9 [DOI] [PubMed] [Google Scholar]
- 21. Andrade JG, Aguilar M, Atzema C, Bell A, Cairns JA, Cheung CC, Cox JL, Dorian P, Gladstone DJ, Healey JS, et al. The 2020 Canadian Cardiovascular Society/Canadian Heart Rhythm Society comprehensive guidelines for the management of atrial fibrillation. Can J Cardiol. 2020;36:1847–1948. doi: 10.1016/j.cjca.2020.09.001 [DOI] [PubMed] [Google Scholar]
- 22. Perry JJ, Sivilotti MLA, Émond M, Stiell IG, Stotts G, Lee J, Worster A, Morris J, Cheung KW, Jin AY, et al. Prospective validation of Canadian TIA score and comparison with ABCD2 and ABCD2i for subsequent stroke risk after transient ischaemic attack: multicentre prospective cohort study. BMJ. 2021;372:n49. doi: 10.1136/bmj.n49 Erratum in: BMJ. 2021;372:n453. PMID: 33541890; PMCID: PMC7859838 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23. Advisory Council for the National Institute of neurological and communicative disorders and stroke, National Institutes of Health, Bethesda, Maryland 20014 . A classification and outline of cerebrovascular diseases II. Stroke. 1975;6:564–616. doi: 10.1161/01.STR.6.5.564 [DOI] [PubMed] [Google Scholar]
- 24. Perry JJ, Losier JH, Stiell IG, Sharma M, Abdulaziz K. National survey of emergency physicians for transient ischemic attack (TIA) risk stratification consensus and appropriate treatment for a given level of risk. CJEM. 2016;18:10–8. doi: 10.1017/cem.2015.57 pmid:26226855 [DOI] [PubMed] [Google Scholar]
- 25. Aho K, Harmsen P, Hatano S, Marquardsen J, Smirnov VE, Strasser T. Cerebrovascular disease in the community: results of a WHO collaborative study. Bull World Health Organ. 1980;58:113–130. [PMC free article] [PubMed] [Google Scholar]
- 26. Oqab Z, McIntyre WF, Quinn KL, Lamb T, Quadros K, Yazdan‐Ashoori P, van Oosten E, Chu K, Lamba J, Barake W, et al. Resident physicians choices of anticoagulation for stroke prevention in patients with nonvalvular atrial fibrillation. Can J Cardiol. 2016;32:824–828. doi: 10.1016/j.cjca.2015.08.004 [DOI] [PubMed] [Google Scholar]
- 27. UC San Diego Health . (n.d.). Warfarin Initiation, Anticoagulation Clinic | UC san Diego Health. https://Health.Ucsd.Edu/Specialties/Anticoagulation/Providers/Warfarin/Warfarin‐Initiation/Pages/Default.Aspx.