← Back

NAVIGATE ESUS LV Dysfunction Subgroup

Left Ventricular Dysfunction Among Patients With Embolic Stroke of Undetermined Source and the Effect of Rivaroxaban vs Aspirin: A Subgroup Analysis of the NAVIGATE ESUS Randomized Clinical Trial

Share Share Copied! Compare

Year of Publication: 2021

Authors: Alexander E. Merkler, Lesly A. Pearce, Scott E. Kasner, ..., Richa Sharma

Journal: JAMA Neurology

Citation: JAMA Neurol. 2021;78(12):1454-1460. doi:10.1001/jamaneurol.2021.3828

Link: https://jamanetwork.com/journals/jamaneu...article/2785321

Clinical Question

Is anticoagulation superior to aspirin in reducing recurrent stroke in patients with recent embolic stroke of undetermined source (ESUS) and left ventricular dysfunction?

Bottom Line

Rivaroxaban was superior to aspirin in reducing recurrent stroke or systemic embolism among NAVIGATE ESUS participants with left ventricular dysfunction, but not in those without LV dysfunction.

        Major Points
        Post hoc exploratory analysis of 7,107 NAVIGATE ESUS participants with documented LV function assessment
LV dysfunction was present in 502 participants (7.1%) defined as moderate-severe global contractility impairment and/or regional wall motion abnormality
Significant treatment interaction by LV dysfunction status (P = 0.03 for interaction)
Among patients with LV dysfunction: rivaroxaban 2.4%/year vs aspirin 6.5%/year (HR 0.36, 95% CI 0.14-0.93)
Among patients without LV dysfunction: rivaroxaban 5.3%/year vs aspirin 4.5%/year (HR 1.16, 95% CI 0.93-1.46)
Results consistent for secondary composite outcome
Major bleeding events were infrequent in both LV dysfunction groups

      

Design

Study Type: Post hoc exploratory subgroup analysis of randomized controlled trial

Randomization: 1

Blinding: Double-blind (participants and investigators)

Enrollment Period: December 2014 to September 2017

Follow-up Duration: Median 10.4 months (IQR 5.2-17.0)

Centers: 459

Countries: 31 countries including Canada, US, Latin America, Western Europe, Eastern Europe, East Asia

Sample Size: 7107

Analysis: Intention-to-treat analysis using Cox proportional hazards model, Kaplan-Meier survival curves, treatment interaction testing, sensitivity analysis adjusting for diabetes. Software: SPSS version 27.0.1 and MedCalc Statistical Software version 19.6.4

Inclusion Criteria

Participants from original NAVIGATE ESUS trial
Age ≥50 years at time of stroke (if 50-59 years, ≥1 additional stroke risk factor required)
Neuroimaging-confirmed ischemic stroke within 7 days to 6 months of symptom onset
ESUS criteria: not lacunar, <50% luminal stenosis of supplying artery, no high-risk cardiac sources, no other identified stroke source
Documented assessment of LV function at study entry via echocardiography

Exclusion Criteria

High-risk sources of cardiac embolism (atrial fibrillation, LV thrombus, mechanical cardiac valve, severe mitral stenosis)
Lacunar stroke
More than 50% luminal stenosis of artery supplying area of ischemia
Other identified source of stroke by time of randomization
No documented LV function assessment

Arms

Field	Rivaroxaban	Control
Intervention	Rivaroxaban 15 mg once daily	Aspirin 100 mg once daily
Duration	Until study completion	Until study completion

Outcomes

Outcome	Type	Control	Intervention	HR / OR / RR	P-value
Composite of recurrent stroke or systemic embolism	Primary	LV dysfunction: 6.5%/year (95% CI 4.0-11.0); No LV dysfunction: 4.5%/year (95% CI 3.8-5.3)	LV dysfunction: 2.4%/year (95% CI 1.1-5.4); No LV dysfunction: 5.3%/year (95% CI 4.5-6.2)	LV dysfunction: 0.36; No LV dysfunction: 1.16	P for interaction = 0.03
Composite of recurrent stroke, systemic embolism, myocardial infarction, or cardiovascular death	Secondary	LV dysfunction: 9.5%/year (95% CI 6.3-14); No LV dysfunction: 5.4%/year (95% CI 4.7-6.3)	LV dysfunction: 4.9%/year (95% CI 2.8-8.6); No LV dysfunction: 6.2%/year (95% CI 5.4-7.1)	LV dysfunction: 0.51; No LV dysfunction: 1.1	P for interaction = 0.05
Major bleeding	Adverse	Overall: 23 events; LV dysfunction: 0 events	Overall: 55 events; LV dysfunction: 5 events
Hemorrhagic stroke	Adverse	2 events in no LV dysfunction group	12 events in no LV dysfunction group; 1 event in LV dysfunction group

Subgroup Analysis

The primary analysis was a subgroup analysis by LV dysfunction status. LV dysfunction was defined as moderate to severely impaired global LV contractility and/or regional wall motion abnormality. Results showed significant treatment interaction with rivaroxaban being superior only in patients with LV dysfunction.

Criticisms

Post hoc exploratory analysis of negative overall trial - results should be considered hypothesis-generating
LV dysfunction assessment was not standardized across participating sites
No requirement to document specific echocardiographic parameters
Relatively small number of participants with LV dysfunction (7.1%)
Lack of granular data on presence or extent of myocardial scar
Limited power to detect associations due to small LV dysfunction subgroup
Non-prespecified combination of LV dysfunction categories due to low event rates

Funding

American Heart Association, National Institutes of Health, Michael Goldberg Research Fund

Based on: NAVIGATE ESUS LV Dysfunction Subgroup (JAMA Neurology, 2021)