Spectral analysis of heart rate variability for trauma outcome prediction: an analysis of 210 ICU multiple trauma patients

Purpose This study aimed to test and compare short-term spectral HRV indices with most used trauma scorings in outcome prediction of multiple trauma, and then to explore the efficacy of their combined application. Methods A prospective study was conducted for patients with blunt multiple trauma admitted to an emergency intensive care unit (ICU) between January 2016 and December 2017. Short-term spectral HRV indices on admission were measured, including normalized low-frequency power (nLF), normalized high-frequency power (nHF), and the nLF/nHF ratio. Injury severity score (ISS), new injury severity score (NISS), and revised trauma score (RTS) were evaluated for each patient, as well as probability of survival (Ps) by trauma and injury severity score (TRISS) model. The primary outcome was 30-day mortality and secondary outcomes were incidence of multiple organ dysfunction syndrome (MODS) and length of ICU stay. Results Two hundred and ten patients were recruited. The nLF/nHF ratio, RTS, and Ps((TRISS)) were independent predictors of 30-day mortality, while nLF/nHF, NISS and RTS were independent predictors of MODS. The area under the receiver operating characteristic (ROC) curve (AUC) of nLF/nHF for 30-day mortality prediction was 0.924, comparable to RTS (0.951) and Ps((TRISS)) (0.892). AUC of nLF/nHF-RTS combination was 0.979, significantly greater than that of each alone. Combination of nLF/nHF and Ps((TRISS)) showed an increased AUC (0.984) compared to each of them. The nLF/nHF ratio presented a similar AUC (0.826) to NISS (0.818) or RTS (0.850) for MODS prediction. AUC of nLF/nHF-RTS combination was 0.884, significantly greater than that of nLF/nHF. Combination of nLF/nHF and NISS showed a greater AUC (0.868) than each alone. The nLF/nHF ratio, NISS, RTS, and Ps((TRISS)) were correlated with length of ICU stay for survivors, with correlation coefficients 0.476, 0.617, - 0.588, and - 0.539. Conclusions These findings suggest that the short-term spectral analysis of HRV might be a potential early tool to assess injury severity and predict outcome of multiple trauma. Combination of nLF/nHF and conventional trauma scores can provide more accuracy in outcome prediction of multiple trauma.