The authors reported a lower short-term incidence of new-onset atrial fibrillation in patients with heart failure with reduced ejection fraction (HFrEF) who were switched from traditional renin–angiotensin system blockade to an angiotensin receptor–neprilysin inhibitor (ARNI). In the propensity score–matched population, new-onset AF occurred at 10.73 versus 12.15 events per 100 person-years in the ARNI and RAS-blockade groups, respectively. The hazard ratio was 0.89 (95% confidence interval 0.81–0.98; P = 0.021). These outcomes reflect the investigators’ primary one-year endpoint and indicate a modest relative reduction in AF incidence associated with ARNI in this retrospective cohort.
Short sentences emphasize the numerical result. Longer sentences place that result into context: the observed reduction applied to a real-world nationwide dataset and was accompanied by a lower risk on a composite endpoint that combined arrhythmia onset with ischaemic stroke and all-cause mortality.
The study team queried the Korean National Health Insurance Service database covering 2017–2021. They identified 39,061 patients with HFrEF who had no prior diagnosis of atrial fibrillation and who received either ARNI (n = 12,263) or traditional RAS blockade (n = 26,798). To balance measured baseline characteristics between treatment groups, investigators performed 1:1 propensity score matching, after which 9,030 patients remained in each arm for the primary analyses.
A short methodological note: the source did not report specific matching covariate calipers or exact lists beyond the variables referenced in sensitivity models. The dataset years, cohort counts, and matching ratio are those provided by the authors.
The pre-specified primary outcome was new-onset AF within one year of treatment assignment. In the matched cohort, ARNI was associated with a lower incidence of that endpoint (HR 0.89; 95% CI 0.81–0.98; P = 0.021). The investigators also evaluated a clinically oriented composite outcome—new-onset AF, ischaemic stroke, or all-cause death—which was reduced in the ARNI group (HR 0.85; 95% CI 0.79–0.92; P < 0.001). These composite results suggest the signal for reduced arrhythmia occurrence occurred alongside benefits detected on a broader outcome that includes hard events.
Where the manuscript presented additional comparisons, the unmatched cohort analyses were more variable. An exploratory model adjusted for a limited set of covariates (age, sex, hypertension, chronic kidney disease, prior ischaemic stroke, socioeconomic status, residential region, and beta-blocker use) returned an HR of 0.85 (95% CI 0.75–0.96). After adjustment for all measured baseline covariates in the unmatched sample, the HR attenuated to 0.94 (95% CI 0.87–1.02), crossing the null.
Investigators performed a competing risk analysis to account for the possibility that death or other events preclude the observation of new-onset AF. In that model, ARNI use was associated with a lower subdistribution hazard for new-onset AF (sHR 0.73; 95% CI 0.58–0.91; P = 0.005). The competing-risk result produced a larger relative effect than the conventional Cox model in the matched cohort. Short sentences state the numeric outcome. The longer clause clarifies its role: the competing-risk model addresses censoring by events that can obviate arrhythmia detection and therefore offers an alternative estimate of association in this observational dataset.
The source did not provide details about dosing regimens, duration of prior RAS blockade before switching, or the specific timing of ARNI initiation relative to heart failure diagnosis. Those elements were not reported in the abstract and therefore are not included here.
This analysis is retrospective and observational; the authors explicitly stated that prospective studies are needed to confirm these findings. Residual confounding cannot be excluded despite propensity score matching and multiple adjusted models. The unmatched analyses showed attenuation of effect after full covariate adjustment, which illustrates sensitivity to model specification. Data needed to assess mechanism—such as echocardiographic changes, biomarkers, or arrhythmia burden on continuous monitoring—were not reported in the abstract. The registry source and design limit causal inference.
The report ends on concrete next steps documented by the authors: further prospective investigations are required to validate whether switching to ARNI yields a true preventive effect against atrial fibrillation in patients with HFrEF, and to define the populations, timing, and mechanistic pathways that might mediate such an effect.