Page 8 - BH-1-2

Brain & Heart Post-stroke atrial fibrillation and predictive scores

underlying cause remains undefined even after first-line To address this challenge, tools such as predictive scores
diagnostic work-up. Given that the majority of CS cases have emerged, aiming to tailor the priority for prolonged
are associated with embolic mechanisms, a new term, cardiac monitoring and reducing inappropriate costs.
“embolic stroke of undetermined source” (ESUS), was first In recent years, several studies have reported on clinical
introduced in 2014. scores capable of predicting the probability of detecting

Subclinical AF (SAF) represents the primary underlying SAF during prolonged ECG monitoring. Surprisingly,
cause of CS or ESUS. Despite secondary prevention with there has been a lack of focus on clinical predictive scores
mono or dual antiplatelet therapy, which remains the for detecting post-stroke SAF in CS or ESUS. Hence, the
recommended first therapeutic option, the rate of stroke primary aim of our study was to fill this important gap.
recurrence in CS or ESUS patients is approximately 2. Materials and methods
5% . Notably, randomized clinical trials have shown no
[1]
advantage of direct oral anticoagulants (DOACs) over We sourced data from the PubMed database by searching for
antiplatelets in preventing stroke recurrence in ESUS articles published from January 1, 2000, to January 31, 2023,
patients [2,3] . Conversely, in AF-related strokes, DOACs that reported on scores predicting post-stroke SAF detection.
are recognized as the first-choice therapy for secondary Our search criteria included combining the terms “AF” AND
prevention due to their superior efficacy and safety profile “stroke” in the title along with the term “score” in the title and/
compared to Vitamin K antagonists or antiplatelets . or abstract. To refine our search strategy, we also reviewed
[4]
Therefore, the screening of SAF represents the cornerstone the bibliographies of the retrieved articles. Our search was
of the diagnostic workup for CS or ESUS. It is strongly limited to articles in the English language involving adults
recommended in this context to establish the etiology, (≥18 years) with CS or ESUS, and we focused exclusively
select the appropriate treatment, and prevent recurrence. on papers reporting score derivation studies. Only original
SAF can be detected through electrocardiographic (ECG) studies were included in the study, while meta-analyses,
monitoring, either during hospital stay or after hospital systematic reviews, and review articles were excluded
discharge, using non-implantable or implantable devices. from the study. In addition, internal or external validation
The rate of SAF detection increases proportionally with studies were excluded if they were not included in the article
longer durations of ECG monitoring. For instance, when presenting the derivation study. The first phase of the search
utilizing an implantable loop recorder ECG, the SAF was conducted by LM and subsequently reviewed by EG. In
detection rate is approximately 15% at 6 months, 23% at the first phase, we evaluated the titles and abstracts to identify
1 year, and 43% at the 3-year follow-up in ESUS patients . potentially relevant articles. Following this initial selection,
[5]
In recent years, recommendations for SAF detection in this full-text articles were analyzed, along with their references.
context have been introduced and put into practice. The When necessary for statistical analysis, we employed
AF-SCREEN International Collaboration recommends a MEDCALC statistical software (MedCalc Software Ltd.,
minimum of 72 h of ECG monitoring to detect SAF in cases Acacialaan 22, B-8400 Ostend, Belgium).
of stroke with undetermined origin, utilizing telemetry
during the hospital stay or continuous ambulatory ECG 3. Results
monitoring . In cases where a diagnosis is not achieved In total, we analyzed over 1100 articles. The search process
[6]
with the initial 72-h ECG monitoring and there remains is depicted in Figure 1.
a high SAF risk, the AF-SCREEN collaborators suggest
extending the ECG monitoring period using either non- From this pool, we selected 26 articles that reported
implantable or implantable tools . predictive scores for post-stroke SAF. Fifteen articles were
[6]
excluded because they focused on stroke cases other than
The European Stroke Organization guidelines
recommend cardiac rhythm monitoring lasting longer than CS and ESUS. Finally, we narrowed our selection down to
11 articles, with eight reporting predictive scores derived
48 h for all ESUS patients over 55 years of age. In addition, from CS and three from ESUS. Table 1 summarizes the
when feasible, prolonged monitoring using an implantable characteristics of these retrieved scores.
loop recorder ECG is advised . However, in real-world
[5]
clinical practice, only a minority of ESUS patients undergo Five studies adopted a prospective approach, while four
implantable loop recorder monitoring due to its limited followed a retrospective methodology. In addition, two
availability, invasiveness, and associated costs. Moreover, studies utilized pooled data from three prospective stroke
in many cases, the time between the stroke event and registries or studies for the analysis. In 10 out of the 11
prolonged ECG monitoring can be extensive, increasing studies, multivariate regression analyses were performed to
the risk of stroke recurrence. Thus, it is crucial to pre-select identify predictors of SAF. Conversely, one study employed
CS or ESUS patients with the highest probability of SAF. a univariate binary logistic analysis.

Volume 1 Issue 2 (2023) 2 https://doi.org/10.36922/bh.0955