WSO - April 2023 - 402

402
International Journal of Stroke 18(4)
Table 1. Search strategy to identify literature reporting on risk factors/predictors of AF detection in CS patients.
Database
PubMed
Search string
( " cryptogenic stroke " [Title/Abstract] OR " embolic stroke of undetermined source " [Title/Abstract]) AND
" atrial fibrillation " [Title/Abstract] AND ( " detection " [Title/Abstract] OR " monitoring " [Title/Abstract])
Search period 1 January 2014 to 31 March 2021
Filters
English language
Inclusion
criteria
* Reporting on individual risk factors/predictors for detection of AF in patients diagnosed with CS
* Risk factors/predictors based on AF detection by means of a systematic ECG monitoring approach utilizing
at least one of the following:
○ Long-term ambulatory ECG monitoring
○ External event-triggered recording
○ Mobile Cardiac (Outpatient) Telemetry
○ Insertable Cardiac Monitor
* Including ⩾100 patients with initial diagnosis of CS
Exclusion
criteria
* Not reporting on individual studies
* Reporting on mixed stroke population (CS patients and patients with stroke of known etiology)
* Reporting on cardiac rhythm monitoring in other (non-CS) populations or for other purposes than
detection of AF
AF: atrial fibrillation; CS: cryptogenic stroke; ECG: electrocardiography.
Figure 1. PRISMA flowchart of the literature search for
risk factors or predictors of AF detection in CS patients.
A number of PRISMA topics did not apply to this narrative
review (see text).
Findings from AF detection studies
in CS patients
Several studies have reported on detection of AF in CS
patients, generally in those aged 60 years or older
(Supplemental Table S3). Application of outcomes from
these studies to younger patients typically considered for
PFO closure is difficult, given the fact that age is a potent
risk factor for AF.
The cryptogenic stroke and underlying AF (CRYSTAL
AF) study23 randomized 441 CS patients of 40 years or
older (mean age = 61.5 years, 22% with a PFO) to ICM
monitoring or conventional follow-up. At 6 months followup,
ICM monitoring detected AF in significantly more
patients than conventional follow-up (8.9% vs 1.4%,
p < 0.001). In the ICM arm, 75% of first AF episodes were
detected within 84 days after randomization. At 3 years, AF
was detected in 30% of the patients in the ICM arm, while
superiority of ICM monitoring over conventional methods
was maintained.
Other studies on ICM monitoring in CS patients have
randomized trials on PFO closure, approximately 1.2
strokes per 100 patients per year occurred.22 This suggests
that deferring PFO closure to pursue prolonged
cardiac rhythm monitoring carries a small, but non-zero
risk of recurrent stroke during the monitoring interval of
approximately 0.1% per month. However, more data are
required to reliably determine whether this risk is higher
in acute and subacute phases.
reported AF detection rates ranging from approximately
11% at 6 months to 25% or more after longer monitoring
periods. Mostly, patients were older than in the CRYSTAL
AF study. Superior AF detection compared to external cardiac
monitors or standard care was also reported from studies
including patients with known stroke etiology.
Non-invasive long-term monitoring using 30-day external
event-triggered loop recording was evaluated in the 30
day event monitoring belt for recording AF after a cerebral
International Journal of Stroke, 18(4)
WSO - April 2023

Table of Contents for the Digital Edition of WSO - April 2023
