Evaluating the efficacy of wireless near infrared spectroscopy sensors for detecting central hypovolemia during simulated hemorrhage in humans
Date
ORCID
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
Background: Early identification of blood loss is essential to decrease mortality from hemorrhage, a major cause of death in the military and civilian trauma settings. An industry partner has created noninvasive and wireless near infrared spectroscopy (NIRS) sensors to measure somatic tissue oxygenation (StO₂) for early detection of blood loss. In this study, we investigated the efficacy of these sensors for tracking the reduction in central blood volume (indexed by stroke volume) in humans undergoing simulated hemorrhage. We hypothesized that each NIRS sensor will progressively track the reduction in central blood volume during simulated hemorrhage in humans.
Methods: Eight healthy humans (3 F, 5M; 25.3 ± 2.0 y) participated in a simulated hemorrhage protocol induced via application of lower body negative pressure (LBNP) to presyncope. Following baseline, the LBNP chamber pressure was decreased every 5-min to -15, -30, -45, -60, -70, -80, -90 and -100 mmHg, or until the onset of presyncopal symptoms (defined as a systolic arterial pressure <80 mmHg or subjective symptoms). Heart rate (via lead II ECG) and arterial pressure (via finger photoplethysmography) were monitored continuously. Stroke volume was estimated from pulse contour analysis of the finger photoplethysmography waveform. A total of five NIRS sensors measured StO2 at different anatomical locations including the sternum, forearm, deltoid, thigh, and calf. Data were analyzed over the final 3-min of each LBNP stage and the 1-min immediately prior to the onset of presyncope. Correlations between the relative changes in StO₂ of each sensor and stroke volume were assessed.
Results: Stroke volume decreased by 46.9± 16.3 % at presyncope. StO₂ decreased by 1.5 ± 7.8 % at the sternum, 9.6 ± 7.4 % at the forearm, 1.5 ± 3.6 % at the deltoid, 21.3 ± 15.8 % at the thigh, and 30.4 ± 27.5 % at the calf. Of all the sites, the strongest relationship between decreases in StO₂ and stroke volume was at the calf (R-value range: 0.70-0.99, R-value mean: 0.89 ± 0.11). The sensors located at each of the other sites tracked stroke volume with high inter-participant variability (sternum, R-value range: -0.71-0.99, R-value mean: 0.24 ± 0.79; forearm, R-value range: -0.09-0.99, R-value mean: 0.61 ± 0.40; deltoid, R-value range: -0.97-0.96, R-value mean: 0.26 ± 0.83; thigh, R-value range: -0.75-0.99; R-value mean: 0.68 ± 0.64).
Conclusion: Unexpectedly, the NIRS sensor on the calf, which was inside the LBNP chamber, performed the best out of the five sites in tracking the progressive reduction in central blood volume in healthy human participants. This may be due to the pooling of blood volume in the lower limbs with the LBNP stimulus, which increased deoxygenated hemoglobin, resulting in an overall lower measurement of tissue oxygen saturation. This finding is interesting, and further modifications and testing of these sensors are required to reliably track blood volume loss in patient populations.