A Temporal Reliability and Sustainability Score Framework for Ranking Heart Rate Variability Metrics During Pharmacological Autonomic Blockade in Rats

Background: While Heart Rate Variability (HRV) is a cornerstone of non-invasive autonomic assessment, the sensitivity and sustainability of specific HRV parameters remain debated. This study introduces a "sustainability score" to systematically validate which metrics best reflect autonomic nervous system (ANS) modulation following pharmacological blockade. Methods: Ten male Sprague-Dawley rats were implanted with ECG electrodes. After recovery, rats underwent randomized atropine (1 mg/kg, intraperitoneal) and propranolol (10 mg/kg, intraperitoneal) sessions separated by at least 3 days. Baseline ECG was recorded for 30 min before each session, followed by 90-min post-drug ECG recording. HRV parameters were analyzed in sequential 15-min and 30-min segments. Reliability was assessed using average p-values across post-drug segments and a sustainability score, defined as the number of segments significantly different from baseline. Results: After atropine, HF Power showed the strongest rapid parasympathetic-related response in 15-min segments, with the lowest average p-value (0.1476) and highest sustainability score (4), followed by RSA (p=0.16148; score=4). HFn was less reliable in 15-min segments (p=0.23147; score=2) but became the most reliable parasympathetic-related metric in 30-min segments (p=0.0001; score=3). After propranolol, SI was the most reliable sympathetic-related metric in both 15-min (p=0.18225; score=2) and 30-min segments (p=0.0393; score=2), while SD2/SD1 showed the strongest 30-min sympathovagal-balance response (p=0.0200; score=3). Conclusion: HF Power is recommended for rapid parasympathetic-related assessment in short windows, HFn for longer-window parasympathetic assessment, SI for sympathetic-assessment, and SD2/SD1 for 30-min sympathovagal-balance analysis. Sustainability scoring provides a structured framework for selecting ECG-derived autonomic markers.