ABSTRACT
BACKGROUND: Heart rate complexity (HRC), commonly described as a "new vital sign," has shown promise in predicting injury severity, but its use in clinical practice has been precluded by the absence of real-time data. This study was conducted to evaluate the utility of real-time, automated, instantaneous, hand-held heart rate entropy analysis in predicting the need for lifesaving interventions (LSIs). We hypothesized that real-time HRC would predict LSIs. METHODS: Prospective enrollment of patients who met criteria for trauma team activation was conducted at a Level I trauma center (September 2011 to February 2012). A novel, hand-held, portable device was used to measure HRC (by sample entropy) and time-domain heart rate variability continuously in real time for 2 hours after the moment of presentation. Electric impedance cardiography was used to determine cardiac output. Patients who received an LSI were compared with patients without any intervention (non-LSI). Multivariable analysis was performed to control for differences between the groups. RESULTS: Of 82 patients enrolled, 21 (26%) received 67 LSIs within 24 hours of hospital arrival. Initial systolic blood pressure was similar in both groups. LSI patients had a lower Glasgow Coma Scale (GCS) score (9.2 [5.1] vs. 14.9 [0.2], p < 0.0001). The mean (SD) HRC value on presentation was 0.8 (0.6) in the LSI group compared with 1.5 (0.6) in the non-LSI group (p < 0.0001). With the use of logistic regression, initial HRC was the only significant predictor of LSI. A cutoff value for HRC of 1.1 yields sensitivity, specificity, negative predictive value, and positive predictive value of 86%, 74%, 94%, and 53%, respectively, with an accuracy of 77% for predicting an LSI. CONCLUSION: Decreased HRC on hospital arrival is an independent predictor of the need for LSI in trauma activation patients. Real-time HRC may be a useful adjunct to standard vital signs monitoring and predicts LSIs. LEVEL OF EVIDENCE: Prognostic and diagnostic study, level III.
Subject(s)
Advanced Cardiac Life Support , Heart Rate , Wounds and Injuries/physiopathology , Adult , Cardiac Output/physiology , Female , Glasgow Coma Scale , Heart Rate/physiology , Humans , Injury Severity Score , Logistic Models , Male , Multivariate Analysis , Prospective Studies , Sensitivity and Specificity , Wounds and Injuries/therapyABSTRACT
INTRODUCTION: The usefulness of heart rate variability (HRV) and heart rate complexity (HRC) analysis as a potential triage tool has been limited by the inability to perform real-time analysis on a portable, handheld monitoring platform. Through a multidisciplinary effort of academia and industry, we report on the development of a rugged, handheld and noninvasive device that provides HRV and HRC analysis in real-time in critically ill patients. METHODS: After extensive re-engineering, real-time HRV and HRC analyses were incorporated into an existing, rugged, handheld monitoring platform. Following IRB approval, the prototype device was used to monitor 20 critically ill patients and 20 healthy controls to demonstrate real-world discriminatory potential. Patients were compared to healthy controls using a Student?s t test as well as repeated measures analysis. Receiver operator characteristic (ROC) curves were generated for HRV and HRC. RESULTS: Critically ill patients had a mean APACHE-2 score of 15, and over 50% were mechanically ventilated and requiring vasopressor support. HRV and HRC were both lower in the critically ill patients compared to healthy controls (p < 0.0001) and remained so after repeated measures analysis. The area under the ROC for HRV and HRC was 0.95 and 0.93, respectively. CONCLUSIONS: This is the first demonstration of real-time, handheld HRV and HRC analysis. This prototype device successfully discriminates critically ill patients from healthy controls. This may open up possibilities for real-world use as a trauma triage tool, particularly on the battlefield.
