Identifying patients at low risk for atrial fibrillation after cryptogenic stroke: the PAVA score.
Maille B, Resseguier N, et al. • Europace : European pacing, arrhythmias, and cardiac electrophysiology : journal of the working groups on cardiac pacing, arrhythmias, and cardiac cellular electrophysiology of the European Society of Cardiology • 2026
The PAVA score, comprising premature atrial contractions, age, significant valvular heart disease, and left atrial dilatation, demonstrates good discrimination for atrial fibrillation detection after cryptogenic stroke, with a score ≤2 yielding high negative predictive values of 89.2% and 100% in development and validation cohorts respectively.
Key Findings
Results
Cumulative incidences of AF or atrial tachycardia detected by ILR after cryptogenic stroke were substantial across follow-up timepoints.
Among 668 patients in the development cohort, cumulative incidences of AF/AT at 1, 2, and 3 years were 20.3%, 28.9%, and 33.0%, respectively.
The total study population comprised 1002 consecutive patients with mean age 66.0 ± 12.3 years and 57.9% male.
This was a multicentre study of patients who received an ILR after cryptogenic stroke.
The primary outcome was a first episode of AF or atrial tachycardia (AT).
Results
Four independent predictors of AF/AT detection were identified and incorporated into the PAVA score.
The four predictors were: premature atrial contractions (PAC) >200/24h on Holter-ECG (2 points), age >58–78 years (3 points) or >78 years (4 points), significant valvular heart disease (1 point), and left atrial dilatation (2 points).
The score was developed in a randomly split development cohort using hazard ratio-based weighting.
Analysis accounted for competing risks, missing data imputation, and hazard ratio-based weighting.
The acronym PAVA stands for Premature atrial contractions, Age, significant Valvular heart disease, and left Atrial dilatation.
Results
The PAVA score demonstrated good discrimination in both development and validation cohorts.
C-index was 0.73 in the development cohort and 0.79 in the validation cohort.
The 3-year areas under the curve (AUC) were 0.800 and 0.905 in the development and validation cohorts, respectively.
ROC-curve analysis confirmed superior clinical utility of the PAVA score over existing scores including Brown ESUS-AF and HAVOC.
Results
A PAVA score ≤2 identified a low-risk subgroup with high negative predictive values for AF/AT detection.
A PAVA score ≤2 was observed in 23.6% of patients.
Negative predictive values for AF/AT on ILR were 89.2% in the development cohort and 100% in the validation cohort for a PAVA score ≤2.
This low-risk cutoff offers a practical tool to optimize patient prioritization for ILR implantation.
Results
The PAVA score demonstrated superior clinical utility compared to existing scores for predicting AF after cryptogenic stroke.
ROC-curve analysis confirmed superior performance over the Brown ESUS-AF (Brown embolic stroke of undetermined source-atrial fibrillation) score and the HAVOC score.
The PAVA score was specifically developed and validated for patients undergoing ILR monitoring after cryptogenic stroke, as opposed to more general existing scores.
Internal validation was performed on the held-out validation cohort separate from the development cohort.
What This Means
This research suggests that a new scoring system called the PAVA score can help identify patients who are unlikely to develop atrial fibrillation (AF) — an irregular heart rhythm — after a cryptogenic stroke (a stroke with no identified cause). The study followed over 1,000 patients who received implantable loop recorders (small heart monitors placed under the skin) to continuously track heart rhythm after their stroke. The researchers found that roughly 1 in 5 patients developed AF or a related rhythm problem within the first year, rising to about 1 in 3 by three years, highlighting how common hidden AF is in this population.
The PAVA score assigns points based on four factors detectable before implanting the monitor: frequent premature heartbeats (extra beats) seen on a standard heart tracing test, the patient's age, the presence of significant heart valve disease, and enlargement of the heart's left upper chamber. Patients scoring 2 or fewer points — about 24% of the study population — had a very low likelihood of being found to have AF on monitoring, with negative predictive values of 89% and 100% in the two study groups tested. The score also outperformed two previously existing prediction tools (Brown ESUS-AF and HAVOC scores).
This research suggests that the PAVA score could help doctors make more informed decisions about which patients most need an implantable heart monitor after a cryptogenic stroke. Since these devices are expensive and require a minor surgical procedure, a reliable way to identify patients at low risk for AF could help prioritize those who would benefit most from monitoring, while potentially sparing low-risk patients from an unnecessary procedure. However, further external validation in independent patient populations would strengthen confidence in these findings before widespread clinical adoption.
Maille B, Resseguier N, Marijon E, Behar N, Badoz M, Vuillier F, et al.. (2026). Identifying patients at low risk for atrial fibrillation after cryptogenic stroke: the PAVA score.. Europace : European pacing, arrhythmias, and cardiac electrophysiology : journal of the working groups on cardiac pacing, arrhythmias, and cardiac cellular electrophysiology of the European Society of Cardiology. https://doi.org/10.1093/europace/euag183