Heart-rate-derived variability measures and in-hospital mortality in critically ill patients with heart failure and atrial fibrillation: insights from the MIMIC-IV database

The prognostic value of heart-rate-derived variability measures in critically ill patients with coexisting heart failure (HF) and atrial fibrillation (AF) remains unclear. In this high-risk population, conventional electrocardiogram-derived heart rate variability metrics are not directly applicable. We therefore investigated the associations of heart-rate-derived variability measures and early heart rate change patterns with in-hospital mortality in critically ill patients with HF and AF. We conducted a retrospective cohort study using the MIMIC-IV database and included 4,131 adult ICU patients with concurrent HF and AF between 2008 and 2022. Heart-rate-derived variability measures were calculated from charted heart rate time-series data recorded during the first 24 h of ICU admission, including heart rate median absolute deviation (HR MAD), heart rate standard deviation (HR SD), and the root mean square of successive differences in heart rate (HR RMSSD). Early heart rate change patterns were defined according to transitions across a 110 bpm threshold. Multivariable logistic regression models were used to evaluate the independent associations of these measures with in-hospital mortality. Higher HR SD was independently associated with increased in-hospital mortality (OR 1.855, 95% CI 1.211–2.841), whereas higher HR RMSSD was independently associated with lower in-hospital mortality (OR 0.609, 95% CI 0.409–0.906). Patients with early transitions across the 110 bpm threshold had significantly higher mortality risk than those with sustained lower heart rates. Overall, elevated global heart rate dispersion, reduced short-term variability, and early dynamic heart rate changes were all associated with worse in-hospital outcomes in critically ill patients with HF and AF. In critically ill patients with HF and AF, heart-rate-derived variability measures obtained from routine early ICU monitoring are independently associated with in-hospital mortality. Greater overall heart rate dispersion, reduced short-term variability, and early transitions across a clinically relevant rate threshold may provide useful prognostic information in this high-risk population.