The goal of resuscitation is to achieve adequate tissue perfusion. The patient's history, ongoing bleeding and clinical signs of hypovolemia are the first trigger for volume substitution. Tissue perfusion is conventionally monitored using arterial base deficit and serum lactate, but these laboratory parameters are not always immediately available at the point-of-care. Additionally, relevant blood loss may not be detected by conventional hemodynamic monitoring 1 .

Therefore, in the trauma setting, there is growing interest in non-invasive techniques measuring tissue perfusion. Near infrared spectroscopy (NIRS) has become a widely used method for tissue hemoglobin oxygen saturation (StO₂) measurement in muscle and has been validated in animals 2 3 4 and in humans 5 6 7 8 9 10 11 .

NIRS uses light with a wavelength of 650-950 nm and easily crosses biological tissues. Near infrared light is mainly absorbed by hemoglobin and only monitors vessels with a diameter <1 mm, because the blood levels in larger vessels are too high to reflect enough light 12 . In this setting, NIRS measurements primarily indicate the venous oxyhemoglobin concentration, as only 20% of blood volume is arterial and the NIRS device does not differentiate between systole and diastole 13 .

NIRS can be used to observe the microcirculatory reactivity of peripheral tissue after a standardized ischemic event. It has been shown that StO₂ recovery after transient occlusion of the brachial artery differs in hemodynamically unstable trauma patients in comparison to healthy volunteers 14 . The StO₂ response after a vascular occlusion test is an example of functional hemodynamic monitoring, in which the response of the specific system to a pre-determined stress is the monitored variable. After induction of an artificial ischemic stress, local metabolic demand and reperfusion reserve can be assessed from the changes in parameters 14 .

The aim of this study was to investigate the impact of a controlled and clinically relevant blood loss (500 ml) on resting StO₂ values, as well as microcirculatory reactivity triggered by a vascular occlusion test in healthy blood donors. 500 ml of blood donation correspond to a loss of 10% of the whole blood volume. 10% of blood loss is considered to be the cut off for clinical relevant blood loss according to the current literature and may become relevant if pre- existing morbidity exists. For example, postpartum hemorrhage, defined as blood loss more than 500 ml after a vaginal delivery, is a major cause of maternal morbidity and mortality 15 . Other authors have been able to detect 500 ml of blood loss with NIRS in various settings 16 17 .

We have decided to investigate the 500 ml blood loss because of the controlled and safe study setting and because higher blood volume losses are anyway related with detectable changes in physiological parameters (e.g. blood pressure, pulse rate). For the clinician, "borderline" patients without obvious signs of blood loss present the greater challenge.

We hypothesized that NIRS after a vascular occlusion test would be able to detect this amount of blood loss, because of the industries promise of "earliest" detection of volume depletion.

There was a markedly decrease in systolic blood pressure in both groups before and after blood donation. Systolic blood pressure decreased from 118 mmHg median (interquartile range: 118 - 123) to 110 mmHg (106 - 115) (p < 0.001) in the donor group and from 112 mmHg (103 - 118) to 106 mmHg (100 - 111) (p = 0.02) in the control group. Systolic blood pressure decreased markedly more in the donor group than in the control group (p = 0.02). (Fig. 4)

For variables measured with NIRS, the differences from the first to the second vascular occlusion test did not differ between the two groups. The only exception was the differences in O₂Hb 2 min median, which was slightly larger in the control group -0.4 (-1.0 - -0.3) than in the blood donor group -0.1 (-0.4 - 0.3) (p = 0.03). Therefore the two groups are considered to be similar and comparable.

StO₂ and O₂Hb 2 min median did not change substantially in the blood donor group before and after blood donation, but decreased in the control group after the rest period: StO₂ 2 min median decreased from 82.0% (79.5 - 85.0) to 80.5% (76.8 - 83.8) (p = 0.02), O₂Hb 2 min median decreased from 10.9 (9.6 - 12.4) to 10.2 (9.4 - 11.8) (p = 0.03). HHb 2 min median did not change in the blood donor group, but increased in the control group markedly from 2.35 (2.08 - 2.56) to 2.45 (2.20 - 2.92) (p = 0.01) (Table 1).

Variables measured during the vascular occlusion test did not change in the blood donor group: The rate of decrease, rate of recovery and delta for StO₂ and O₂Hb did not show any substantial difference before and after blood donation. In the control group, StO₂ variables did not change markedly - in contrast to the O₂Hb values. O₂Hb rate of decrease was less steep after the rest period (before: -0.039 (-0.051 - -0.026); after: -0.031 (-0.046 - -0.020); p < 0.01), whereas O₂Hb rate of recovery (before: 0.49 (0.39 - 0.63); after: 0.54 (0.39 - 0.69); p = 0.05) and O₂Hb delta (before: 3.88 (2.43 - 4.20); after: 4.0 (3.32 - 5.72); p = 0.01) increased after the rest period (Table 2).

The principal finding of this study is, that although systolic blood pressure decreased substantially, blood loss had no effect on superficial thenar StO₂ before and after blood donation measured by NIRS. However other experimental NIRS devices have been successfully used to monitor blood loss after blood donation. Torella et al. showed that even 2% blood volume loss reduced peripheral hemoglobin oxygen saturation (PsO₂, calf) significantly 16 . Soller et al. simulated hypovolemia using progressive lower body negative pressure and showed that hypovolemia corresponding to 400 - 500 ml of blood loss resulted in a decrease in forearm muscle oxygen saturation 17 . The same group could also show, in the same setting, that deep muscle oxygen saturation is more sensitive in detecting hypovolemia (experimental NIRS device) than superficial thenar StO₂ (Hutchinson) 21 . Soller et al. discussed, that the location of the NIRS sensor as well as the tissue depth might be responsible for variations in results, although the technique used by the two devices is in principle the same. We did use exactly the same superficial thenar StO₂ probe from Hutchinson as Soller in her study, which may explain why we did not detect any changes in NIRS parameters following blood loss.

A second finding of this study is that there might be an intrinsic effect of repeated vascular occlusion tests on O₂Hb. This hypothesis is based on the increase in delta O₂Hb and O₂Hb rate of recovery and a decrease in oxygen consumption (O₂Hb rate of decrease), which was measured in the control group but not in the blood donor group. This could be explained if there is a time effect of repeated vascular occlusion tests, designated as ischemic preconditioning 22 . This improves the ischemic tolerance of various tissues, including skeletal muscle, after one or more brief periods of ischemia, each followed by a short reperfusion phase. The underlying mechanisms are still not clear. One reason may be the reduction in the resting rate of oxygen consumption 23 . Although blood loss does not result in different StO₂ or O₂Hb variables after the vascular occlusion test, it seems to mask the effect of ischemic preconditioning in the blood donor group.

StO₂ measured at the thenar eminence seems to be insensitive to blood loss of 500 ml in this setting. Probably blood loss greater than this might lead to detectable changes guiding the treating physician. The exact cut off for detectable changes and the time effect on repeated vascular occlusion tests should be explored further. Until now no such data exist. Finally, our findings suggest that StO₂ measured by NIRS is not able to detect acute blood loss of 500 ml and therefore, it might not be useful to apply this device in the acute assessment of a bleeding trauma patient. However, the device could help to identify hemodynamic worsening during the ICU stay but at the moment not enough evidence exists to support this hypothesis.

The authors declare that they have no competing interests.

VJ, SF and AE conducted the study. All authors contributed to the design of the study, the analysis of the data and the writing of the manuscript. All authors have seen the original study data, reviewed the analysis of the data, and approved the final manuscript. VJ is the author responsible for archiving the study files.