WSO - April 2023 - 403

Diener et al.
403
Figure 2. AF detections rates versus age. Top panel:
CRYSTAL AF study (1 year ICM monitoring).27 Middle panel:
Ziegler et al.30 (6 months ICM monitoring). Bottom panel:
Wachter et al.31 (7 days Holter monitoring).
detection was lower in patients with PFOs found on transthoracic
versus transesophageal echocardiography, suggesting
that the yield of AF monitoring is further reduced
with increasing PFO size. No association between the
monitoring method and the likelihood of AF detection was
noted.
Overall, studies demonstrated superior AF detection by
ICM monitoring compared to non-invasive methods,
including shorter-term monitoring and 30-day external
event recording. Prolonged monitoring showed that AF is a
relatively common finding in mixed age CS patients, with
higher yields after longer monitoring periods.
Predictors for AF detection
Age is a strong predictor for detecting AF in CS patients
(Figure 2). In the CRYSTAL AF study, the likelihood of
ICM-detected AF at 12 months roughly doubled with each
age decade, with AF rarely detected in patients <55 years.27
In other ICM studies, CS patients with no AF detections
were younger than those with detections.28,29
Similar age-dependent effects were reported from other
stroke populations and/or other monitoring methods.31,32
Overall, clinical data suggest that AF is an uncommon finding
in CS patients <55 years, with the likelihood of detecting
AF increasing sharply with older age.
Several other predictors for AF detection in CS patients
have been reported. In the CRYSTAL AF cohort, diabetes,
prolonged PR interval, and frequent premature atrial contractions
were associated with a significantly increased
probability of AF detection.27 The PR interval was a significant
predictor in a multivariate regression model, specifically
in patients without PR interval prolonging medication.
Frequent premature atrial contractions observed by initial
Holter monitoring were also a strong predictor in the
EMBRACE study.33
ischemic event (EMBRACE) study.24 This strategy detected
AF in 16.1% of the patients, compared to 3.2% detected by
conventional 24 h monitoring.
Nevertheless,
the
postembolic
rhythm detection with implantable vs external
monitoring (PER DIEM) trial showed the superiority of
ICM monitoring over prolonged non-invasive ECG monitoring,25
with AF detected in 15.3% and 4.7% by ICM monitoring
and external loop recording, respectively.
The presence of a PFO is associated with a lower AF
detection rate. In a meta-analysis26 of 14 studies (13,245
patients, mean age 61.2 years), the rate of AF detection
was halved in patients with PFO (relative risk (RR) = 0.52,
95% confidence interval (CI) = 0.41-0.63). The rate of AF
Among 196 ESUS patients undergoing 30-day ambulatory
heart rhythm monitoring, AF was associated with
atrial cardiopathy biomarkers of increased left atrial
diameter on echocardiography, P-wave terminal force in
electrocardiogram lead V1, and P-R wave interval on
electrocardiogram.34 Atrial enlargement and interatrial
conduction block (maximum P-wave duration ⩾120 ms)
were reported as predictors for AF in CS populations
monitored by ICM.35 Increased levels of B-type natriuretic
peptide (BNP) and N-terminal pro B-type natriuretic
peptide (NT-proBNP) have also been reported to be associated
with AF detection in CS patients using prolonged
ECG monitoring, with higher specificity achieved by
BNP.36
Prior brain infarctions (i.e. not associated with the qualifying
index event) and leukoaraiosis were both associated
with an almost threefold higher AF detection rate in the
CRYSTAL AF study.37 Similarly, in CS patients monitored
International Journal of Stroke, 18(4)
WSO - April 2023

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