Abstract
Heart rate variability (HRV), a noninvasive technique used to quantify fluctuations in the interval between normal heart beats (NN), is a predictor of mortality in some patient groups. The aim of this study was to assess HRV in burn trauma patients as a predictor of mortality. The authors prospectively performed 24-hour Holter monitoring on burn patients and collected demographic information, burn injury details, and in-hospital clinical events. Analysis of HRV in the time and frequency domains was performed. A total of 40 burn patients with a mean age of 44 ± 15 years were enrolled. Mean %TBSA burn was 27 ± 22% for the overall population and was significantly higher in those who died compared with those who survived (55 ± 23% vs 19 ± 13%; P < .0001). There was a statistically significant inverse linear correlation between SD of NN intervals and %TBSA (r = -.337, R = 0.113, 95% CI = -0.587 to -0.028, two-tailed P = .034), as well as with ultra low frequency power and %TBSA burn (r = -0.351, R = 0.123, 95% CI = -0.152 to -0.009; P = .027). The receiver-operator characteristic showed the area under the curve for %TBSA as a predictor of death was 0.82 (P < .001), for SDANN was 0.94 (P < .0001), and for ultra low frequency power was 0.96 (P < .0001). Deranged HRV in the early postburn period is a strong predictor of death.