Background: Reduced heart rate variability (HRV) reflects autonomic dysfunction and can triage patients better than routine trauma criteria or vital signs. However, there is questionable specificity and no consensus measurement technique. The purpose of this study was to analyze whether factors that alter autonomic function affect the specificity of HRV for assessing traumatic injury.
Methods: We evaluated 216 hemodynamically stable adults (3:1 M:F; 97:3 blunt:penetrating; age 49 years ± 1 year, mean ± standard error) undergoing computed axial tomography (CT) scan to rule out traumatic brain injury (TBI). All were prospectively instrumented with a Mars Holter system (GE Healthcare, Milwaukee, WI). HRV was determined offline using time domain (standard deviation of normal–normal intervals, root-mean-square successive difference) and frequency domain (very low frequency [VLF], LF, wideband frequency, high frequency [HF], low to HF index ratio) calculations from 15-minute electrocardiogram and correlated with routine vital signs, mortality, TBI, morbidity, length of stay (LOS), and comorbidities. Significance (p ≤ 0.05) was determined using nonparametric analysis, Student's t test, analysis of variance, or multiple logistic regression.
Results: VLF alone predicted survival, severity of TBI, intensive care unit LOS, and hospital LOS (all p < 0.05). Beta-blockers or diabetes had no effect, whereas age, sedation, mechanical ventilation, spinal cord injury, and intoxication influenced one or more of the variables with age being the most powerful confounder (all p < 0.05). Except for the Glasgow Coma Scale, no other routine trauma or hemodynamic criteria correlated with any of these outcomes.
Conclusions: Decreased VLF is an independent predictor of mortality and morbidity in hemodynamically stable trauma patients. Other time and other frequency domain variables correlated with some, but not all, outcomes. All were heavily influenced by factors that alter autonomic function, especially patient age.
From the Dewitt-Daughtry Family Department of Surgery (M.L.R., M.P.O., B.M.T.P., J.C.G.-R., R.J.M., P.A.V., K.G.P.) and Department of Epidemiology and Public Health (R.C.D.), University of Miami Miller School of Medicine, Miami, Florida.
Submitted for publication September 15, 2010.
Accepted for publication March 2, 2011.
Supported, in part, by Grant No. N140610670 from the Office of Naval Research.
Presented at the 69th AAST Annual Meeting on September 22–25, 2010, Boston, Massachusetts.
Address for reprints: Kenneth G. Proctor, PhD, Divisions of Trauma and Surgical Critical Care, Dewitt Daughtry Family Department of Surgery, University of Miami Miller School of Medicine, Ryder Trauma Center, 1800 NW 10th Ave., Miami, FL 33136; email: email@example.com.