In a prospective single-center study of 6,697 adults undergoing elective cardiac surgery (2023–2025), preoperative atrial fibroblast activation failed to distinguish patients who developed postoperative atrial fibrillation (POAF) from those who didn't. Instead, time-resolved heart rate variability analysis detected progressive autonomic destabilization beginning roughly 20 minutes before POAF onset—specifically, heart-rate-corrected SDNN diverged significantly in the final 10-minute pre-event window (Cohen's d=0.728). Validated in an independent cohort of 3,451 patients, the HRV signal achieved 69.6% sensitivity and 97.5% specificity at a 10-minute prediction horizon.
This finding matters considerably. POAF affects 20–50% of cardiac surgery patients and drives prolonged ICU stays, stroke risk, and readmissions. The dominant clinical assumption has been that structural atrial remodeling—fibrosis—determines susceptibility, guiding risk stratification toward imaging-heavy preoperative workups. These data reframe POAF as a dynamically triggered, state-dependent arrhythmia rather than an inevitable consequence of structural disease, opening a window for real-time autonomic monitoring and potentially targeted pre-emptive intervention. Clinically, this could support continuous HRV surveillance protocols in cardiac ICUs. However, as a preprint not yet peer-reviewed, these results require independent scrutiny—particularly the generalizability of a single-center Chinese cohort and whether the 69.6% sensitivity is clinically actionable before translation to bedside protocols. If confirmed, the paradigm shift from structural determinism to autonomic dynamics could meaningfully redirect both prevention strategy and monitoring infrastructure.