Abstract
STUDY DESIGN: Cross-sectional. OBJECTIVES: To investigate test-retest reliability of heart rate variability (HRV) metrics in SCI without restriction of activity over long (24-h) and shorter durations (5-min, 10-min, 1-h, 3-h and 6-h). SETTINGS: University hospital in Khon Kaen, Thailand. METHODS: Forty-five participants (11 with tetraplegia and 34 with paraplegia) underwent two 24-h recordings of RR-intervals to derive time and frequency HRV metrics. Relative reliability was assessed by intraclass correlation coefficient (ICC) and absolute reliability by coefficient of variation (CV) and Bland-Altman limits of agreement (LoA). RESULTS: For 5- and 10-min durations, eight of eleven HRV metrics had moderate to excellent reliability (ICC 0.40-0.76); the remaining three were poor (ICC < 0.4). HRV values from 1-h and 3-h durations showed moderate to excellent reliability (ICC of 0.46-0.81), except for 1-h reliability of ULF and TP (ICC of 0.06 and 0.30, respectively). Relative reliability was excellent (ICC of 0.77-0.92) for 6-h and 24-h durations in all HRV metrics. Absolute reliability improved as recording duration increased (lower CVs and narrower LoAs). Participants with high AD risk (SCI level at or above T6) showed lower test-retest reliability of HF and LF values than participants with low AD risk. CONCLUSION: Relative reliability of HRV was excellent for 6-h and 24-h. The best absolute reliability values were for 24-h duration. Time-domain outcomes were more reliable than frequency domain outcomes. Participants with high risk of AD, particularly those with tetraplegia, showed lower reliability, especially for HF and LF.