CONTENT

To the Editor,

We read with great interest the study by Yurtseven et al,¹ which evaluated the association between atrial fibrillation (AF) episodes <30 seconds detected on 24-hour Holter monitoring and stroke risk, and we commend the authors for addressing this frequent clinical dilemma and for aligning their message with current European Society of Cardiology (ESC) recommendations based on CHA₂DS₂-VA scores.

That said, several aspects may limit the interpretability of the findings. First, there is an inconsistency in sample size reporting: the Abstract/Methods section states 133 non-sustained AF (NS-AF) cases, whereas the Results mention 163, and¹ again lists n = 133 for NS-AF. This discrepancy should be clarified, and the manuscript’s text and tables made internally consistent. Additionally, because events occurred over a follow-up of approximately 66 ± 6 months, the use of binary logistic regression may not fully capture time-dependent risk. Sensitivity analyses using Cox proportional hazards and, if applicable, competing risk models (e.g., Fine–Gray) could provide a more robust evaluation. Furthermore, even after propensity score matching, age differences remained statistically significant, raising the possibility that the observed association between NS-AF and stroke risk may be partially explained by age. Closer matching or additional adjustment for age would strengthen the findings.

Clarifying these points and incorporating time-dependent statistical approaches would enhance the validity and clinical applicability of the study’s conclusions.

Footnotes

Declaration of Interests: Bülent Özlek serves as an editorial board member for the following journals: Cell Cycle, Muğla Sıtkı Koçman Üniversitesi Tıp Dergisi, Cardiovascular Diabetology – Endocrinology Reports.

Other authors have no conflicts of interest to declare.

Reference

1. Yurtseven E, Ural D, Karaüzüm K. Nonsustained atrial fibrillation in ambulatory ECG recording and thromboembolic events in longterm follow-up. Anatol J Cardiol. 2025;29(8):401-408.