A prospective, multi-site, observational study assessing the treatment thresholds of new onset atrial fibrillation in the critically ill

Objective To define haemodynamic thresholds for pharmacological treatment of new-onset atrial fibrillation (NOAF) in critically ill patients and examine associations with short-term outcomes including reversion to sinus rhythm and in-hospital mortality. Design, setting, and participants prospective multicentre observational cohort study conducted across six Australian intensive care units (ICUs) from December 2024 to September 2025. Adult patients (≥18 years) who developed NOAF during ICU admission were included. A Cox proportional hazards model treating time to treatment as a time-dependent covariate was used to assess associations between pharmacological therapy and secondary outcomes. Main outcome measures The primary outcome was heart rate at initiation of the first pharmacological therapy for NOAF. Secondary outcomes included mean arterial pressure at time of treatment, reversion to sinus rhythm, haemodynamic deterioration, ICU length of stay, and in-hospital mortality. Results Four hundred eighty-two patients were screened for eligibility and 210 met the inclusion criteria. One hundred fifty-four (73.3%) received pharmacological treatment for NOAF. The median heart rate at treatment initiation was 128 beats per minute (interquartile range [IQR]: 113-147) and median mean arterial pressure 76 mmHg (IQR: 70-87). Pharmacological treatment was not independently associated with sustained reversion to sinus rhythm (adjusted hazard ratio: 1.44, 95% confidence interval [CI]: 0.94-2.20) or in-hospital mortality (adjusted odds ratio: 0.96, 95% CI: 0.36-2.58). Pharmacological treatment was associated with prolonged ICU stay (multiplicative adjusted effect: 1.39, 95% CI: 1.06-1.82). Conclusions Pharmacological treatment for NOAF is typically initiated at heart rates approximating 130 bpm. After accounting for time-dependent exposure, pharmacological treatment was not independently associated with reversion to sustained sinus rhythm. Given the methodological limitations encountered, these findings warrant validation in randomised trials to better define optimal treatment strategies for NOAF in critical illness.