← Back

RE-SPECT ESUS

Dabigatran for Prevention of Stroke after Embolic Stroke of Undetermined Source

Share Share Copied! Compare

Year of Publication: 2019

Authors: H.-C. Diener, R.L. Sacco, J.D. Easton, ..., and K. Toyoda

Journal: The New England Journal of Medicine

Citation: N Engl J Med 2019;380:1906-17.

Link: https://doi.org/10.1056/NEJMoa1813959

PDF: https://www.nejm.org/doi/pdf/10.1056/NEJMoa1813959

Clinical Question

In patients with a recent embolic stroke of undetermined source (ESUS), is dabigatran superior to aspirin for the prevention of recurrent stroke?

Study Overview

Objective

Compare dabigatran versus aspirin for secondary stroke prevention in patients with embolic stroke of undetermined source (ESUS).

Study Summary

Dabigatran was not superior to aspirin in preventing recurrent stroke in ESUS patients. Stroke recurrence and major bleeding rates were similar between groups, but dabigatran had higher rates of clinically relevant nonmajor bleeding.

Intervention

Dabigatran 150 mg BID (or 110 mg BID for patients ≥75 years or CrCl 30–50 mL/min) vs. aspirin 100 mg daily. Double-blind, randomized trial in 5,390 patients with recent ESUS. Median follow-up: 19 months.

Bottom Line

In patients with a recent embolic stroke of undetermined source, dabigatran was not found to be superior to aspirin in preventing recurrent stroke. The incidence of major bleeding was similar between the two groups, but clinically relevant nonmajor bleeding events were more frequent with dabigatran.

        Major Points
        Second major ESUS anticoagulation trial (after NAVIGATE ESUS). Both trials tested whether anticoagulation is superior to aspirin for ESUS — both were negative, effectively ending the ESUS anticoagulation hypothesis for the general ESUS population.
Largest ESUS trial: 5,390 patients across 564 centers in 42 countries. Double-blind design (unlike NAVIGATE ESUS which was open-label) — the gold-standard design for drug comparison.
Primary outcome: recurrent stroke 4.1%/yr dabigatran vs 4.8%/yr aspirin (HR 0.85, 95% CI 0.69–1.03, P=0.10). The trend favoring dabigatran did not reach statistical significance; the paper reports the hazard ratio only and did not report a relative risk reduction.
Notably, disabling stroke showed a nominally lower rate with dabigatran (HR 0.59, 95% CI 0.36–0.96). However, per the prespecified hierarchical analysis plan, no adjustment for multiple comparisons was made and P values were not computed for any secondary outcome after the primary outcome was non-significant — only unadjusted confidence intervals are reported.
Major bleeding was similar (1.7% vs 1.4%/yr, HR 1.19, 95% CI 0.85–1.66) — a key difference from NAVIGATE ESUS where rivaroxaban showed significantly more major bleeding. Dabigatran appeared safer than rivaroxaban in the ESUS context.
Clinically relevant non-major bleeding was higher with dabigatran (HR 1.73, 95% CI 1.17–2.54), but intracranial hemorrhage rates were identical (0.7%/yr in both groups).
Dose-reduced dabigatran (110 mg BID) was used in patients ≥75 years or with CrCl 30–50 mL/min — a protocol feature not available in NAVIGATE ESUS (fixed-dose rivaroxaban).
ESUS definition per Hart criteria: non-lacunar stroke on imaging, no ≥50% stenosis, no AF lasting >6 minutes on cardiac rhythm monitoring of 20 hours or longer, no intracardiac thrombus, no other specific cause.
12.6% of patients had a PFO — these patients might have benefited from PFO closure rather than anticoagulation, potentially diluting the treatment effect.
Together with NAVIGATE ESUS, established that empiric anticoagulation for ESUS is not beneficial — shifting focus to identifying specific ESUS subtypes (AF, PFO, cancer, aortic arch disease) for targeted therapy.

      

Design

Study Type: International, multicenter, randomized, double-blind trial.

Randomization: 1

Blinding: Double-blind.

Enrollment Period: December 2014 through January 2018.

Follow-up Duration: Median of 19 months (IQR 13-27).

Centers: 564

Countries: 42 countries

Sample Size: 5390

Analysis: Intention-to-treat.

Inclusion Criteria

Age ≥60 years with ESUS within the previous 3 months (or 6 months if with a vascular risk factor).
Age 18-59 years with ESUS within the previous 3 months and at least one additional vascular risk factor.
ESUS was defined as a nonlacunar ischemic stroke on imaging with no significant (≥50%) stenosis in arteries supplying the area of the stroke, no atrial fibrillation lasting longer than 6 minutes as shown by cardiac rhythm monitoring for 20 hours or longer, no intracardiac thrombus, and no other specific cause of stroke.

Exclusion Criteria

Full exclusion criteria are provided in Table S1 of the Supplementary Appendix and are not enumerated in the main paper.
By the ESUS definition itself (main text): lacunar stroke on imaging; ≥50% extracranial or intracranial stenosis of an artery supplying the area of the stroke; atrial fibrillation lasting >6 minutes on cardiac rhythm monitoring of ≥20 hours; intracardiac thrombus on transthoracic or transesophageal echocardiography; any other identified specific cause of stroke.
Severe renal impairment is implied by the protocol dose-reduction rule (110 mg BID was used only down to CrCl 30 mL/min).

Baseline Characteristics

Characteristic	Control	Active
Mean age - yr	63.9±11.4	64.5±11.4
Female sex no. (%)	986 (36.6)	1001 (37.1)
Race (White) - no. (%)	1966 (72.9)	1926 (71.5)
Mean body-mass index	27.3±5.0	27.2±5.0
Creatinine clearance <50 ml per minute - no. (%)	203 (7.5)	227 (8.4)
Median time from index stroke to randomization (IQR) - days	43.0 (20.0-78.0)	46.0 (21.0-82.0)
Median score on modified Rankin Scale (IQR)	1 (0-2)	1 (0-2)
Previous TIA or stroke - no. (%)	500 (18.6)	475 (17.6)
Hypertension - no. (%)	1985 (73.7)	1996 (74.1)
Diabetes mellitus - no. (%)	639 (23.7)	585 (21.7)
Patent foramen ovale - no. (%)	361 (13.4)	319 (11.8)

Arms

Field	Control	Dabigatran
Intervention	Aspirin at a dose of 100 mg once daily, plus a dabigatran placebo.	Dabigatran at a dose of 150 mg twice daily, or 110 mg twice daily for patients ≥75 years or with a creatinine clearance of 30 to 50 ml/min, plus an aspirin placebo.
Duration	Median of 19 months	Median of 19 months

Outcomes

Outcome	Type	Control	Intervention	HR / OR / RR	P-value
First recurrent stroke of ischemic, hemorrhagic, or unspecified type.	Primary	7.7% (4.8% per year)	6.6% (4.1% per year)	0.85	0.10
Ischemic stroke	Secondary	7.5% (4.7% per year)	6.4% (4.0% per year)	HR 0.84 (95% CI, 0.68 to 1.03)	Not computed per hierarchical testing
Composite of nonfatal stroke, nonfatal myocardial infarction, or cardiovascular death	Secondary	8.6% (5.4% per year)	7.7% (4.8% per year)	HR 0.88 (95% CI, 0.73 to 1.06)	Not computed per hierarchical testing
Disabling stroke	Secondary	1.6% (0.9% per year)	0.9% (0.6% per year)	HR 0.59 (95% CI, 0.36 to 0.96)	Not computed per hierarchical testing
Major bleeding	Adverse	2.4% (1.4% per year)	2.9% (1.7% per year)	HR 1.19 (95% CI, 0.85 to 1.66)
Clinically relevant nonmajor bleeding	Adverse	1.5% (0.9% per year)	2.6% (1.6% per year)	HR 1.73 (95% CI, 1.17 to 2.54)
Intracranial hemorrhage	Adverse	1.2% (0.7% per year)	1.2% (0.7% per year)	HR 0.98 (95% CI, 0.60 to 1.60)

Subgroup Analysis

The absence of a treatment effect on the primary outcome was consistent across prespecified subgroups (age, sex, geographic region, PPI use, days from index stroke to randomization, baseline creatinine clearance, add-on aspirin at baseline, prior stroke/TIA, cardiac monitoring duration at baseline, CHA2DS2-VASc score, and PFO status). Dose assignment (110 vs 150 mg) was analyzed post hoc and was not prespecified — the paper explicitly states that all reported subgroups were prespecified except this one. Patients with PFO showed a treatment effect consistent with the overall trial results (HR 0.88, 95% CI 0.45-1.71 for PFO present [16/319 dab vs 19/361 asa]; HR 0.83, 95% CI 0.68-1.03 for no PFO). Exploratory/post hoc analyses were reported for PPI use and days from index stroke to randomization (from which inferences cannot be made), and separately for first recurrent strokes before vs after 1 year (post hoc analysis suggested dabigatran may have had an effect on stroke recurrence after 1 year, but no inferences can be made because of the post hoc nature).

Criticisms

ESUS is a heterogeneous entity grouping PFO-related, occult AF, cancer-related, and aortic arch atheroma — treating all with one anticoagulant is likely to dilute any benefit in a specific subgroup.
Only 20 hours or longer of cardiac monitoring was required to exclude AF — modern standards (30-day monitoring or implantable loop recorders) detect AF in 12–30% of cryptogenic stroke patients vs ~5% with brief Holter monitoring. In this trial, extended ECG monitoring was performed in only 14% of patients.
12.6% of patients had a documented PFO — these patients may have benefited more from PFO closure than anticoagulation, confounding the intent-to-treat analysis.
Industry-sponsored (Boehringer Ingelheim) — financial incentive to demonstrate superiority for dabigatran.
Median 19-month follow-up was relatively short — longer follow-up might have allowed the trend (HR 0.85) to reach significance, as seen in RESPECT (initially negative at 2.6 years, positive at 5.9 years).
No aortic arch imaging was mandated — aortic arch atheromas are a known but often undetected source of embolism in ESUS.
The ESUS concept itself has been questioned after both NAVIGATE ESUS and RE-SPECT ESUS failed — some argue the entity should be abandoned in favor of more specific etiological classifications.
Dose reduction criteria (age ≥75 or CrCl 30–50) may have undertreated some patients who could have tolerated full-dose dabigatran.

Funding

Boehringer Ingelheim.

Based on: RE-SPECT ESUS (The New England Journal of Medicine, 2019)