A retrospective observational cohort study was conducted including consecutive patients admitted to the Cardiovascular Critical Care Unit (CCCU) of a tertiary care center from December 2011 to May 2021 with ST-segment elevation myocardial infarction (STEMI) or non-ST-segment elevation myocardial infarction (NSTEMI) diagnosis who developed a first episode of paroxysmal AF during hospitalization.

Atrial fibrillation was defined according to current clinical practice guidelines1 as documented by a standard 12-lead electrocardiogram (ECG) or a single-lead ECG tracing more than 30 seconds showing a cardiac rhythm with indiscernible P waves followed by irregular RR intervals. Paroxysmal AF was defined as AF that reverted spontaneously or required intervention within the first seven days after diagnosis.1 Myocardial infarction was diagnosed by the attending physician according to current guidelines,21,22 based on evidence of myocardial damage (elevated cardiac troponin levels >99th percentile of the upper reference limit), with evidence of necrosis in a clinical context consistent with myocardial ischemia. No restrictions were made regarding the type of AMI including types 1–5 as defined by the Universal Definition of Myocardial Infarction.23 Coronary anatomy was assessed using diagnostic coronary angiography, and the number of diseased vessels and treated vessels was recorded. Complete revascularization was defined as the treatment of all vessels with a diameter >2 mm and stenosis >70%, or >50% in the case of the left main coronary artery.

Exclusion criteria included patients with a history of prior AF (paroxysmal, permanent, or persistent, even if they presented with sinus rhythm at admission), those who remained in AF at the time of discharge, and those for whom one-year follow-up was not possible. This included patients who died during hospitalization and those who were treated in another healthcare system, as their medical records could not be reliably followed.

Demographic, laboratory, and echocardiographic data, along with interventional and pharmacological treatments during hospitalization and at discharge, were collected. All variables are defined in the supplementary material (see S2. Variable definitions). A minimum follow-up of one year from the date of hospital discharge was conducted by reviewing patients’ medical records through the healthcare information system of the Castilla-La Mancha Health Service (SESCAM).

Categorical variables were expressed as numbers and percentages. Quantitative variables were expressed as median and interquartile range (IQR 25–75%). Normal distribution of quantitative variables was assessed using the Shapiro–Wilk test. Categorical variables were compared between groups using the χ2 test or Fisher's exact test, as appropriate. For normally distributed continuous variables, comparisons were made using Student's t-test, while the Mann-Whitney U test was used for non-normally distributed variables. Comparisons of continuous variables across independent groups were conducted using analysis of variance with Bonferroni correction for multiple comparisons. A two-sided p-value <0.05 was considered statistically significant. Survival outcomes were analyzed using Kaplan–Meier survival curves and Cox proportional hazards regression.

Multivariate regression analysis was performed, including variables related to the index event as reported in the literature,6,24 as well as those variables that showed p<0.2 in the bivariate analysis. An additional regression analysis powered by bootstrap technique on 1000 samples was performed ad hoc. All statistical analyses were performed using the Statistical Package for the Social Sciences software (SPSS v. 23.0 for Windows, SPSS Inc., Chicago, IL, USA). This article follows the STROBE recommendations25 for reporting observational studies.

The study was conducted in accordance with the principles of the Declaration of Helsinki and good clinical practice guidelines. It was approved by the hospital's Ethics Committee, which granted an exemption from the requirement to obtain informed consent from patients.

The relationship between AF and AMI has been widely studied due to the high prevalence of both conditions and their significant clinical and socioeconomic impact.8,26,27 However, data on NOAF are more limited. The prevalence of NOAF in the context of AMI in our cohort was 10.5% (354 patients who developed a first episode of AF during hospitalization out of 3382 patients without a prior history of AF). This finding is similar to previous studies such as the RIMA registry (8.8%) and Batra et al. (7.6%).9,10 A key distinction in our study is the inclusion of both STEMI and NSTEMI patients, whereas previous studies often focused solely on STEMI. Furthermore, we excluded other supraventricular arrhythmias, ensuring that our analysis was specific to NOAF. Additionally, we provided information on patients’ heart rhythm at discharge, finding that most were cardioverted during hospitalization (spontaneously or actively), with 24.6% remaining in AF at discharge. These findings are consistent with other studies, while also offering additional insights into rhythm management during hospitalization.

To our knowledge, this is the first study specifically analyzing the recurrence of paroxysmal AF in the context of AMI. We observed a recurrence rate of 9.1% during the first year. Although continuous electrocardiographic monitoring was not used –which could lead to missed self-limiting arrhythmic episodes– our results provide valuable insights into the clinical course of NOAF following AMI. It is well established that acute ischemic events can act as triggers for AF through mechanisms such as atrial ischemia, increased left ventricular filling pressures, and subsequent atrial dilation.6 These factors may resolve after the acute phase, potentially preventing long-term structural damage, which could explain why many patients remain in sinus rhythm during follow-up. This recurrence raises important questions about the need for long-term anticoagulation in patients who maintain sinus rhythm, especially since most of these patients receive dual antiplatelet therapy for 6–12 months post-AMI. Identifying predictors of recurrence is crucial to guiding early anticoagulation strategies. In our study, multivariate logistic regression suggested a protective effect of ACEis on AF recurrence. Given the limitations of our sample size, a secondary analysis using bootstrapping was performed, which further suggested an association between diabetes and AF recurrence and confirmed the protective effect of ACEis and ARBs.

It is worth noting that there were no significant differences in baseline characteristics or treatment between patients with and without AF recurrence, which suggests that the perceived cardiovascular risk by the treating physicians was similar across groups. However, changes in clinical practice guidelines over the decade-long study period may have introduced some bias, particularly regarding newer treatments such as neprilysin inhibitors and SGLT2 inhibitors, which were not widely available at the start of the study.

A prior history of AF has been historically associated with worse outcomes in patients with AMI, including higher mortality, hospitalizations, and reduced quality of life.8,9,20,28 However, NOAF which occurs in response to acute ischemic events may have a different prognosis. Studies have reported conflicting results regarding the impact of NOAF recurrence on outcomes.4,9 The CREDO-Kyoto study4 reported that both NOAF and chronic AF were associated with increased long-term mortality in AMI patients at five years, regardless of whether sinus rhythm was restored. However, in the RIMA registry,9 NOAF was associated with worse cardiovascular outcomes only in the short-term, particularly in the first year following AMI (HR 2.49; 95% CI 1.32–4.67; p=0.004).

In our study, AF recurrence during the first year was associated with worse clinical outcomes, especially HF hospitalizations, compared to patients who maintained sinus rhythm. Although we did not find significant differences in overall mortality, likely due to the small sample size, there was a trend toward higher mortality in the AF group. This highlights the need for closer monitoring and management of HF in patients with AF recurrence following AMI.

Larger studies are necessary to confirm the long-term impact of AF recurrence and to determine whether early rhythm control or anticoagulation strategies could improve outcomes in this high-risk group.

Our study has certain limitations inherent to its design. Firstly, as a retrospective observational study, the data were limited to the information available in the electronic medical records, which may introduce bias or incomplete data. Additionally, as it was a single-center study, the sample size was restricted, which could limit the ability to achieve statistical significance in some analyses. As such, these results should be interpreted as hypothesis-generating. A prospective multicenter study would be valuable to further investigate the questions raised in this analysis. Moreover, the study spans a ten-year period, during which updates to clinical practice guidelines were made as new evidence emerged. This may have introduced variability in the management of patients over time, potentially affecting the consistency of treatments and outcomes.