Long-term prognostic value of a multidomain atrial remodeling score for atrial arrhythmias and cardiovascular events after myocardial infarction
B Couto Viana, L Alves, E Andrade, T Prata-BrancoAbstract
Introduction
Atrial remodeling after acute myocardial infarction (AMI) encompasses structural, electrical, and biochemical changes that may reflect arrhythmic vulnerability and cumulative cardiovascular burden [1,2]. Its long-term prognostic significance remains uncertain.
Purpose
To evaluate whether a multidomain atrial remodeling score (ARS) may predict major adverse cardiovascular events (MACE) and atrial arrhythmias over a ten-year follow-up.
Methods
We included 148 consecutive AMI patients admitted between 2013–2015 to a tertiary care center. Atrial remodeling at 1 year was defined using a multidomain construct integrating structural (LAVI >34 mL/m²), biochemical (BNP in the highest tertile ≥56.1 pg/mL), and electrical domains (≥2 of: P-wave duration ≥120 ms, P-wave terminal force in V1 ≤−4000 μV·ms, or advanced interatrial block). Patients were categorized as having 0, 1, or ≥2 abnormal domains. Outcomes included incident atrial fibrillation or flutter, MACE, and a combined endpoint. Fisher–Freeman–Halton and Mann–Whitney tests were applied.
Results
The population had a mean age of 54.5 ± 10.0 years; 87.2% were male. 74% were smokers; 38.5% had hypertension; 15.6% diabetes; and 64.9% dyslipidemia. STEMI accounted for 69.6% of cases, and the left anterior descending artery was involved in 52.4%. ARS at 1 year demonstrated a graded and statistically significant association with atrial arrhythmias (χ²=10.924, p=0.004), with incidence increasing from 5.0% without remodeling to 33.3% in those with multidomain remodeling. In contrast, atrial remodeling burden was not significantly associated with MACE (p=0.113) or with the combined endpoint (p=0.073), although both outcomes increased numerically across categories. When score components were evaluated individually, higher BNP at one year was the only isolated parameter significantly associated with atrial arrhythmias (p=0.040). No single structural, biochemical, or electrical variable predicted MACE or the combined endpoint.
Conclusions
Atrial remodeling quantified one year after AMI was significantly associated with long-term atrial arrhythmias, whereas its relationship with MACE or the combined endpoint did not reach statistical significance despite numerically higher event rates in higher ARS categories. BNP at one year emerged as the only isolated marker associated with arrhythmias, although with weaker association strength than the ARS. No single one-year atrial parameter predicted MACE. Taken together, these findings suggest that early atrial remodeling may preferentially reflect atrial cardiomyopathy processes predisposing to arrhythmogenesis, while links with broader cardiovascular outcomes appear weaker or may manifest later. This exploratory evidence supports the potential value of integrative atrial phenotyping for long-term arrhythmic risk assessment after AMI.Graphical AbstractFor image description, please refer to the figure legend and surrounding text.