Abstract
Background: Acute cervical spinal cord injury (cSCI) disrupts supraspinal autonomic cardiovascular control. Heart rate variability (HRV), assessed using linear and nonlinear methods, provides a sensitive index of cardiovascular autonomic regulation.
Objective: To evaluate linear and nonlinear HRV parameters associated with autonomic cardiovascular dysregulation in acute cSCI.
Methods: This monocentric prospective study included 500 patients with traumatic or non-traumatic acute cSCI and 200 age- and sex-matched healthy controls. All participants underwent 24-hour Holter ECG monitoring. Time- and frequency-domain indices and nonlinear metrics, including Poincaré plot analysis, entropy, detrended fluctuation analysis, and recurrence quantification analysis, were calculated. Predictive modeling used logistic regression, random forest, gradient boosting, and support vector machine algorithms with cross-validation.
Results: Both cSCI groups showed significantly reduced HRV amplitude and altered autonomic cardiovascular regulation. Nonlinear indices demonstrated reduced entropy, altered fractal scaling, and decreased recurrence dynamics. Female patients showed relatively preserved nonlinear HRV parameters. Models combining linear and nonlinear features discriminated autonomic dysfunction better than models using conventional HRV metrics alone.
Conclusion: Combined linear and nonlinear HRV analysis provides complementary characterization of autonomic cardiovascular dysregulation following acute cSCI.