International Stroke Conference 2021 Update
Ravi Menon, MD
May 14, 2021
Swedish Cerebrovascular Symposium
Disclosures
• No relevant financial disclosures
Objectives
• Review some of the considerations in CRYSTAL AF/monitoring for atrial fibrillation
• Gain a greater understanding of the approach/findings of STROKE AF
• Discuss considerations related to STROKE AF
STROKE AF
• International Stroke Conference (ISC) 2021
Late-breaking abstract 6. Presented March 18, 2021
• Schwamm LH, et.al. “STROKE AF: atrial fibrillation in non-cardioembolic stroke of presumed known origin”
• Industry funded
CRYSTAL AF - Cryptogenic Stroke and Underlying AF (CRYSTAL AF)• Cryptogenic Stroke and Underlying AF (CRYSTAL AF)
• ICM detected AF in 12.4% of patients during a 12-month period
• Approximately 23% of patients who have had a first stroke have recurrent stroke, for which AF is a risk factor
• Increased use of Implantable Cardiac Monitor to detect afib in patients with stroke of unknown cause (cryptogenic stroke) has become common.
Considerations
• Guidelines for the prevention of secondary stroke do not recommend long-term monitoring for AF for patients with ischemic stroke attributed to large-artery atherosclerosis or small-vessel occlusion
• Goal: Determine the rate of AF in this population
• Goal: Analyze whether ICMs detect AF > SOCare
Considerations
• Shift in mindset - Small vessel stroke "looks like this" on imaging; Large vessel stroke "looks like that" on imaging
• Shift mindset from purely mechanistic etiology for stroke
• Account for CV risk – at-risk population/stroke diagnosis
• STROKE AF –evaluates defined population for atrial fibrillation
STROKE AF SUMMARY
• In patients with recent ischemic stroke with presumed etiology of small-vessel or large-vessel atherosclerosis, atrial fibrillation is detected more often with Implantable Cardiac Monitor (ICM) vs Standard of Care (SOC)
• ICMs may detect AF in as many as 1 in 8 patients with stroke of this etiology
STROKE AF Trial
• Prospective, randomized, controlled study 33 US centers
• Eligible no history of AF diagnosis
• Index stroke 10 or fewer days before ICM insertion
• No contraindication to long-term oral anticoagulation
• Age: >= 60 or 50 to 59 years + 1 additional stroke RF (HTN/CHF/DM/2nd stroke >90 days, vascular disease)
STROKE AF Design
• Randomly assigned: ICM or site-specific SOCare for detection of AF
• ICM- detects episodes of AF lasting at least 2 minutes
• Episodes of AF (at least 30 sec) adjudicated by a clinical events committee
• Primary goal: compare the AF incidence rate at 12 months ICM vs SOC
• Subgroup analysis: incidence rates of AF in the two study arms with respect to stroke subtypes. Secondary analyses at 3 years
Population
• 496 participants enrolled: Random assignment 242 ICM vs 250 SOC
• Similar baseline characteristics between the study arms
• Mean age 67 years
• 62% of the population was male
• Median CHA2DS2-VASc score was 5: significant stroke risk among patients for whom AF was detected
Results
• At 12 months, AF detected:
• 12.1% ICM pts vs 1.8% SOC pts (hazard ratio [HR], 7.41; P < .001)
• Median time to first detection of AF in the ICM arm was 99 days
• Rate of AF detected w/in 30 days: ICM pts 2.6% vs SOC 0.4%
• Oral anticoagulation prescribed: ICM pts (7.4%) vs SOC (1.2%)
• Fewer recurrent strokes in the ICM pts (15 vs 23; HR, 0.67; P = .23)
Results
• ICM pts: similar rate of AF detection large-artery atherosclerosis (11.7%) vs small-vessel occlusion (12.6%)
• ICM arm 55.5% of patients in whom AF was detected experienced an episode that lasted more than 1 hour
• About 44% of patients experienced an episode that lasted more than 4 hours
• About 96% of first AF episodes = Asymptomatic
• Occurrence of undiagnosed AF is more frequent than presumed among stroke patients with large-artery atherosclerosis and small-vessel disease
• Previous research has indicated that AF occurs more frequently among patients with cryptogenic stroke, especially when a possible embolic mechanism is suspected
Provocative, needs more study
• Current state = ~30 days ambulatory cardiac monitoring misses the majority of afib episodes
• ICM in this population may be beneficial to detect poststroke AF and to inform optimal stroke prevention
• More data from 36 month study
Study Limitations
• ICM may undercount events of afib: miss AF episodes that last less than 2 minutes
• Lack of a control group
• Underpowered
Cost/Resource/System considerations
• Who should be monitored? How long? With what?
• Age/disease states?
• Older age => likely vascular risk factors =large vessel vs small vessel
• 2014 data - cost of ICM insertion was estimated at ~$6500 (operator/acquisition); monthly monitoring costs were estimated at $58
Take home
• In patients with recent ischemic stroke with presumed etiology of small-vessel or large-vessel atherosclerosis afib detected more often Insertable Cardiac Monitor (ICM) vs Standard of Care (SOC)
• 78% missed with 30-day monitoring (median time to first Afib 99 days)
• NNMonitor to detect one case of Afib: ICM 8 vs SOC 56
• 1 year AF detected:
12.1% ICM vs 1.8% SOC (hazard ratio 7.41; P < .001)
• Fewer recurrent strokes ICM at 12 months: ICM 15 vs SOC 23 (HR 0.67; P = .23)
STROKE AF SUMMARY
• ICMs may detect AF in as many as 1 in 8 patients with stroke of this etiology