Acute blood loss linked to severe hypovolemia and hemorrhagic shock is a critical condition in pediatric intensive care. Many conditions can induce hemorrhagic shock or a shock-like state in newly born infants, including trauma, internal bleeding, coagulopathy, surgery or massive maternal bleeding.1,2 In hemorrhagic shock, the blood loss has exceeded the body’s ability to compensate and the body is unable to provide adequate tissue perfusion and oxygenation. The failure of the compensatory mechanisms that causes hemorrhagic shock leads to devastating changes and high mortality. The neurological sequelae, such as impaired cognition, development delay and cerebral palsy, also cannot be neglected as consequences when an infant survives hemorrhagic shock.
Stopping bleeding and providing adequate fluid resuscitation are the major keys to treating hemorrhagic shock. The primary goals are restoration of tissue perfusion and oxygenation, while shortening the time interval during which global and regional ischemia occurs; this minimizes brain injury and prevents neurological sequelae. However, there are few clinical and experimental investigations that have focused on hemorrhagic shock in infants and children.3,4 In addition, brain tissue oxygenation changes during fluid resuscitation have not been well investigated. Conventionally, oxygen delivery and consumption are calculated from the hemoglobin concentration, the oxygen saturation, and the cardiac output. Unfortunately these parameters are indicative of the whole-body oxygen debt and not oxygenation at a local level; measuring the latter requires the placement of an invasive monitoring device, or intermittent blood sampling.
Near-infrared spectroscopy (NIRS) provides a continuous, non-invasive method to measure regional changes in tissue oxygenation.5–7 It relies on the relative transparency of biological tissues to near-infrared light where oxyhemoglobin, deoxyhemoglobin, and cytochrome aa3 have different absorption spectra. As a result, because the hemoglobin monitored by NIRS is located in the tissue circulation (venules, capillaries, and arterioles) and cytochrome aa3 is present in the neurons, when there is an appropriate distance between the near-infrared emitter and detector, the absorption of oxygenated and deoxygenated hemoglobin in the local tissue can be measured noninvasively.
Previous studies using NIRS in newborn piglets at high risk of hypoxic-ischemic brain damage have observed episodes of decreased cerebral hemoglobin oxygenation.8,9 NIRS has also been used to study infant and adult changes in cerebral oxygen utilization during periods of sepsis, transient hypoxia, mental work, surgery, cardiopulmonary bypass surgery, extracorporeal membrane oxygenation and intensive resuscitation.10–13 However, regional cerebral oxygen saturation (rScO2) changes during volume expansion therapy in hemorrhagic shock events have not been well investigated.
We hypothesized that fluid resuscitation with different volume expanders, when used to treat hemorrhagic shock, may influence changes in brain tissue oxygenation. Therefore, we designed this study to investigate the role of different types of fluid resuscitation on cerebral tissue oxygenation employing NIRS monitoring using a piglet model of hemorrhagic shock.
All animals were managed in accordance with the principle of laboratory animal care of the National Institutes of Health. In addition, all procedures were approved by the Institutional Animal Care and Use Committee of Taipei Veterans General Hospital.
2.1. Animal preparation and physiological monitoring
Newborn piglets, less than two weeks of age, were anesthetized with intramuscularly administered atropine (0.1mg/dose) and ketamine (25mg/kg/dose) prior to the surgical procedures. All animals were placed in supine position and given a subcutaneous injection of lidocaine hydrochloride (2%) for local anesthesia. After placing a 3.5-mm inside- diameter uncuffed endotracheal tube (Murphy, Unomedical Sdn. Bhd., Kedah, Malaysia) via a tracheotomy, controlled mechanical ventilation was established using a volume-controlled ventilator (Model 683, Harvard, South Natick, MA, USA). The tidal volume was set at 10mL/kg, the ventilator rate at 30 breaths/min, the inspiratory to expiratory time (I:E) ratio at 1:1 and the positive end-expiratory pressure at 5cmH2O. Finally, the fractional concentration of inspired oxygen was set at 0.21 initially. After the induction of anesthesia, a solution of 0.33% sodium chloride in 5% dextrose was infused continuously at a rate of 5mL/kg/h. Animals were then paralyzed with intravenous pancuronium bromide (0.2mg/kg), sedated with midazolam (0.5mg/kg), and maintained with a continuous infusion of ketamine (5mg/kg/h), midazolam (0.5mg/kg/h), and pancuronium bromide (0.2mg/kg/h). A 3.5-Fr umbilical vessel catheter (Argyle, Sherwood Medical Corp., Chicopee, MA, USA) was placed into the right femoral artery for the continuous monitoring of arterial blood pressure and for arterial blood sampling. Another 3.5-Fr umbilical catheter was inserted into the left jugular vein for administration of fluids, anesthesia and central venous blood sampling. Body temperature was maintained at 38–39°C throughout the experiments via a servo-controlled heating blanket.
Throughout the experiment, electrocardiography, arterial blood pressure, peripheral oxygen saturation and body temperature were continuously monitored (Agilent M1205A, Philips Medical Systems, Andover, MA, USA).
2.2. Near-infrared spectroscopy
A pair of fiberoptic optodes was attached to the scalp of the animal using a probe holder after induction of anesthesia. The optodes were connected to the NIRS device (NIRO-200; Hamamatsu Photonics K.K., Hamamatsu City, Japan). The emitter and receiver were fixed to the probe holder to ensure an interoptode distance of 4cm. The unit uses safe, faint light (wavelength approximately 700–950nm) that passes through the brain tissue to measure rScO2. The readings were continuously monitored until the end of each experiment.
2.3. Experimental protocol
After surgical preparation, the experimental animals underwent a 20-minute equilibrium, followed by an artificial hemorrhage that induced hemorrhagic shock [mean arterial blood pressure (MAP) <45mmHg]. This was done by withdrawing four aliquots of whole blood out from the arterial line during the first 20min (10mL/kg/aliquot at the 4th, 8th and 12thmin and 5mL/kg or 10mL/kg at the 16thmin). After induction of hemorrhagic shock, the animals received treatment with inhaled 100% oxygen, and then were immediately randomized into one of the following experimental groups. These were as follows: (1) control group (n=6): animals did not receive any extra fluid resuscitation; (2) control-normal saline (NS) group (n=6): animals received one dose of standard bolus fluid challenge consisting of 10mL/kg of normal saline (0.9% sodium chloride) over 5min; (3) whole blood (WB) group: the animals received a bolus of NS (10mL/kg) over 5min in a similar manner to the control-NS group, and this was then followed by an intravenous infusion of 15mL/kg of WB over 20min. Additional infusion (15mL/kg/dose) of WB would be given every 30min if MAP did not rise higher than 45mmHg until the end of the experiment; (4) Lactated Ringer’s (LR) group (n=6): the animals received the same protocol as the WB group, but with the infusion fluid replaced by LR solution; (5) NS group (n=6): the animals received the same protocol as the WB group but with the infusion fluid replaced by NS.
During the study period, sodium bicarbonate (1–2mEq/kg/dose) was given to any animal displaying metabolic acidosis, namely a base excess larger than −8mmol/L and a pH less than 7.20. Epinephrine (0.01mg/kg/dose) was given to any animal displaying bradycardia, namely having a heart rate of less than 100beats/min. Cardiac massage was performed if the heart rate dropped to less than 60beats/min and this lasted until the heart rate was equal to or higher than 60beats/min or at most for 30min.
After baseline measurement, the physiological profiles, rScO2, and arterial blood gas levels were recorded every 4min during the first 20-min blood-drawing period, and then at 40min, 60min, 120min, 180min and 240min after the beginning of the experiments. An additional 1mL blood was drawn in order to measure hemoglobin levels at baseline, 20min, 60min, 120min and 240min. At the end of the experiments (240min), animals were sacrificed with a high dose of 15% potassium chloride while they were under deep anesthesia.
Fractional cerebral tissue oxygen extraction (FTOE) was calculated from rScO2 and SaO2 values.10 A ratio of (SaO2-rScO2)/SaO2 was calculated to represent the balance between oxygen delivery and oxygen consumption. An increase in FTOE reflects an increase in oxygen extraction by the brain tissue, and a decrease in FTOE suggests that there is less utilization of oxygen by brain tissue, in relation to the supply of oxygen.10
2.4. Statistical analysis
All data are expressed as mean±standard error of the mean (SEM). Statistical analysis was performed using SigmaStat® 3.1 (Systat Software, Inc., Point Richmond, CA, USA). The basic characteristics, hemorrhagic amounts and infusion amounts among the study groups were compared by one-way analysis of variance (ANOVA) or Kruskal-Wallis one-way ANOVA on ranks as appropriate, which was followed by the post hoc Student-Newman–Keuls’ test for pairwise multiple comparisons. The physiological data between the baseline and after blood loss of the same group were compared by paired t test. The continuous data in terms of cardiopulmonary profiles and cerebral tissue oxygenation at different time points for the five study groups were compared by two-way repeated measures ANOVA, which was followed by the post hoc Student-Newman–Keuls’ test for pairwise multiple comparisons. The survival curves of the study groups were compared by the Gehan–Breslow test followed by the Holm–Sidak method for pairwise multiple comparisons. Significance was accepted at the p<0.05 level.
Mean body weight, age and the blood loss volume needed to induce hemorrhagic shock did not vary significantly among the five study groups (Table 1). At baseline, all cardiovascular profiles, rScO2 and FTOE levels did not vary among the groups (Table 2). Immediately after induction of hemorrhagic shock, MAP, hemoglobin and rScO2 decreased significantly in all animals (p<0.05), but heart rate and FTOE levels increased significantly (p<0.05), shown in Table 2. There was no significant difference in the gas exchange across the study groups from baseline to after blood loss status.
The total fluid transfusion volume was significantly higher in the LR group (p<0.05), and there was no significant difference between WB and NS groups (p>0.05) (Table 1). During the course of the experiments, all animals in the control group died earlier than the end of the expected experimental time period after induction of hemorrhagic shock (Fig. 1), and the survival rate was significantly lower than in the other four groups (p<0.05). The median survival time for the control group was 45min (range 20–125min) after induction of hemorrhagic shock. In addition, the survival rate in the animals receiving only one dose of 10mL/kg NS (control-NS group) was not significantly different from those of the other three treatment groups (WB, LR, and NS) that received additional 15–60mL/kg (mean: 38±6mL/kg) of the various different volume expanders (Table 1 and Fig. 1). The WB group was the only group where all members survived the full experimental period.
The hemoglobin concentrations and MAP were both significantly higher in the WB group compared to all other groups over the experimental period (p<0.05) (Fig. 2). Without any fluid resuscitation, the control group animals presented with a progressively deteriorating MAP and heart rate until death. However, the MAP of all the other treatment groups, including the control-NS group, rose by 10–20mmHg in response to the initial 10mL/kg NS challenge (Fig. 2). On treating the WB, LR, and NS group with the follow-up fluid infusion therapies, only the WB group showed a rising pattern of MAP during the follow-up time (Fig. 2). Although animals from the LR group received the highest fluid infusion volume (Table 1), their MAP did not show any further improvement after the initial saline challenge. Furthermore, the LR group had the lowest MAP and highest HR among the four treatment groups (Fig. 2). The animals from the Control-NS group (total saline infusion volume=10mL/kg) were able to successfully maintain their MAP, and showed no significant difference when compared to the NS group (total saline infusion volume=43±3mL/kg) (Table 1 and Fig. 2). None of the experimental animals were able to increase their MAP back to that at the original baseline within the experimental period.
After the induction of hemorrhagic shock and the use of 100% oxygen, the PaO2 rose in all groups, but the animals in the control group could not maintain oxygenation in their blood or their brain tissue for more than 20min. Thus, the serial values for PaO2, SaO2, and rScO2 of the control group were all significantly lower than those for the other groups (p<0.05), as shown in Fig. 3. Comparing the rScO2 levels across the four treatment groups, those of the WB group were highest (p<0.05). However, their rScO2 levels were not able to reach baseline. When the animals’ cerebral FTOE levels were examined, it was briefly highest in the control group compared to all other groups over the first 40min after induction of hemorrhagic shock (p<0.05) (Fig. 3). However, there was no significant difference between LR and NS groups for rScO2 and cerebral FTOE.
Our study demonstrated that effective fluid resuscitation in addition to 100% oxygen supplementation was able to maintain cardiovascular function and brain tissue oxygenation during the emergent first four hours, even when only one dose of bolus NS challenge was given. WB transfusion was most effective at maintaining blood pressure and brain tissue oxygenation comparing to LR and NS in this piglet model with hemorrhagic shock.
Hypovolemia is the most common cause of circulatory failure in infants and children,3,4 and fluid resuscitation is the major key to restoring intravascular volume. When the tissue perfusion cannot be restored rapidly, critical tissue hypoxia and ischemia may develop, and this leads to multiple organ failure. Different fluid infusion therapies have been discussed to treat hemorrhagic shock and restore intravascular volume, including modified hemoglobin solution, LR solution, 3% hypertonic saline, colloid treatment, other crystalloid treatment, and others.14,15 The design of this present study was based on the most commonly used clinical management protocols used during neonatal and pediatric resuscitation for hemorrhagic shock. Therefore, only the most common volume expanders, namely NS, WB, and LR, were tested in our experiments. The potential effects of other volume expanders will need further investigation to elucidate their benefits in terms of brain tissue oxygenation.
There are several physiological variables used by physicians to detect and treat shock-induced tissue hypoxia.16–18 Some endpoints of resuscitation, such as heart rate, mental status, and blood pressure, are nonspecific, subjective, or both. Other endpoints, such as mixed venous oxygen saturation, arterial lactate, and arterial base deficit, are accurate and objective but are limited in the scope of their applicability because they need invasive monitoring and/or intermittent blood sampling from a central venous catheter, or often even a pulmonary artery catheter. Each of these invasive monitoring modalities adds risk to critically ill patients. A key disadvantage of using these global values is that they provide only a general picture of the patient’s oxygenation and do not examine the variations in oxygen debt seen at the regional tissue level. Pulse oximetry, although noninvasive, is not that easy to read when the patient’s condition is unstable, especially when the blood pressure is low or undetectable during hemorrhagic shock; this is because its signal depends on small artery pulsation. Furthermore, this parameter indicates systemic circulation rather than regional tissue oxygenation.19 When compared with the above, noninvasive NIRS monitoring is able to provide a continuous estimation of regional tissue oxygenation and thus may be a better choice when assessing brain tissue oxygenation in critical patients. Our experiments also demonstrate that it can be used conveniently for the real-time monitoring of a subject’s rScO2 throughout a long experiment without any interruption.
There have been only a few investigations that have evaluated the feasibility of NIRS as a monitoring approach for neonatal and pediatric patients,20,21 and this is also true for animal models undergoing hemorrhagic shock.22–27 Investigations of the relationship between jugular bulb oxygen saturation and regional cerebral oxygenation under conditions of hypoxia were measured by NIRS and found to have good correlation.20,21 Previous hemorrhagic shock studies examining the ability of NIRS to determine adequacy of tissue oxygenation have mostly focused on peripheral muscles, the stomach, the liver, the kidneys, or the legs.22–24 However, in these studies, the investigators have found that regional tissue oxygen saturation is correlated with measurements of systemic oxygen delivery in a linear fashion.22 Furthermore, it has also been found that total body oxygen delivery is correlated with cytochrome aa3 and tissue oxygen saturation during the hemorrhagic phase, and that post-resuscitative NIRS values were low within the gastric and muscular beds, despite normal systemic measures of oxygen delivery.23 Thus, resuscitation would seem to not uniformly restore cellular oxygenation to all tissue beds. Predicting organ dysfunction during traumatic shock resuscitation using NIRS monitoring tissue oxygenation has also been reported.25 Thus, using NIRS to detect rScO2 would seem to be a convenient and reliable technique for monitoring and predicting neurological outcome in hemorrhagic shock cases. Low tissue oxygen saturation is thought to be associated with a worse outcome in critically ill infants, so intervention needs to be started as early as possible in selected high-risk patients, which will result in it being more likely that the intervention will have a demonstrably favorable impact on early survival.28–30 The present study has shown the effects on rScO2 and FTOE of using different fluids to perform resuscitation over the first four hours after hemorrhagic shock. However, long-term effects were not included in our experiments. A more meticulous study design that evaluates the subjects’ long-term neurological outcome is necessary in the future.
One limitation of our investigation is a failure to measure cerebral perfusion and the added inotropic agents in the studied animals. Meybohm et al reported on the effects of rapidly restoring cerebral perfusion pressure in piglets undergoing hemorrhagic shock and showed that brain tissue oxygenation and the tissue oxygen index could not be improved without blood transfusion.26 Using our animal model, our findings are in agreement and demonstrate that WB is the best fluid for raising rScO2 when used with 100% oxygen supplementation. Therefore, blood transfusion to increase total hemoglobin level to restore oxygen carrying capacity is important to brain tissue oxygenation. However, other volume expanders will still have some role in restoring rScO2. Based on the present results, there is no reason why they should not be used in an emergency when WB is not available during the first hours of treatment. Another limitation of our study is that the maintained MAP (between 50–70mmHg) of the treatment groups did not reach the baseline blood pressure (approximately 85–95mmHg) of the study animals, and, similarly, the rScO2 of the treated animals also did not reach the baseline value of the animals. Therefore, the treatments used here are still suboptimal. A more aggressive design using oxygen, fluid resuscitation and inotropic agents to maintain optimal condition may be very useful when evaluating brain tissue oxygenation in clinical cases with acute hemorrhagic shock.
In conclusion, hemorrhagic shock events decrease brain tissue oxygenation and influence oxygen extraction, and such an event results in high mortality in young subjects when there is no effective fluid resuscitation. Inspired oxygen supplementation and effective fluid resuscitation with either one-dose bolus of NS, or this combined with a subsequent transfusion of WB, LR or additional NS would seem to help to stabilize cardiovascular condition and improve cerebral tissue oxygenation during an emergency period. In these circumstances, WB is the best fluid for maintaining an ideal rScO2 when treating hemorrhagic shock.
The authors gratefully acknowledge expert statistical help from Benjamin Ing-Tiau Kuo, M.D., Ph.D. at Taipei Veterans General Hospital, facility support of the Clinical Research Core Laboratory, Taipei Veterans General Hospital, and technical help from Miss Jen-Yu Lin and Miss Shr-Yun Chiou at Taipei Veterans General Hospital.
Support was received from the National Science Council (NSC95-2314-B-010-072) and Taipei Veterans General Hospital (V98C1-059), Taipei, Taiwan, R.O.C.
1. Pua HL, Bissonnette B. Cerebral physiology in paediatric cardiopulmonary bypass. Can J Anaesth
2. Du Plessis AJ. Mechanisms of brain injury during infant cardiac surgery. Semin Pediatr Neurol
3. Carcillo JA, Tasker RC. Fluid resuscitation
of hypovolemic shock: acute medicine’s great triumph for children. Intensive Care Med
4. Boluyt N, Bollen CW, Bos AP, Kok JH, Offringa M. Fluid resuscitation
in neonatal and pediatric hypovolemic shock: a Dutch Pediatric Society evidence-based clinical practice guideline. Intensive Care Med
5. Rose JC, Neill TA, Hemphill JC 3rd. Continuous monitoring of the microcirculation in neurocritical care: an update on brain tissue oxygenation. Curr Opin Crit Care
6. Pellicer A, Bravo MC. Near-infrared spectroscopy: a methodology-focused review. Semin Fetal Neonatal Med
7. Murkin JM, Arango M. Near-infrared spectroscopy as an index of brain and tissue oxygenation. Br J Anaesth
8. Brown DW, Picot PA, Naeini JG, Springett R, Delpy DT, Lee TY. Quantitative near- infrared spectroscopy measurement of cerebral hemodynamics in newborn piglets. Pediatr Res
9. Chien JC, Jeng MJ, Chang HL, Lee YS, Lee PC, Soong WJ, et al. Cerebral oxygenation during hypoxia and resuscitation
by using near-infrared spectroscopy in newborn piglets. J Chin Med Assoc
10. Toet MC, Lemmers PM, van Schelven LJ, van Bel F. Cerebral oxygenation and electrical activity after birth asphyxia: their relation to outcome. Pediatrics
11. Skarda DE, Mulier KE, Myers DE, Taylor JH, Beilman GJ. Dynamic near-infrared spectroscopy measurements in patients with severe sepsis. Shock
12. Edmonds HL Jr, Ganzel BL, Austin EH. Cerebral oximetry for cardiac and vascular surgery. Semin Cardiothorac Vasc Anesth
13. Ward KR, Ivatury RR, Barbee RW, Terner J, Pittman R, Filho IP, et al. Near-infrared spectroscopy for evaluation of the trauma patient: a technology review. Resuscitation
14. Knudson MM, Lee S, Erickson V, Morabito D, Derugin N, Manley GT. Tissue oxygen monitoring during hemorrhagic shock and resuscitation
: a comparison of lactated Ringer’s solution, hypertonic saline dextran, and HBOC-201. J Trauma
15. Pinto FC, Capone-Neto A, Prist R, Silva MRE, Poli-de-Figueiredo LF. Volume replacement with lactated Ringer’s or 3% hypertonic saline solution during combined experimental hemorrhagic shock and traumatic brain injury. J Trauma
16. Wilson M, Davis DP, Coimbra R. Diagnosis and monitoring of hemorrhagic shock during the initial resuscitation
of multiple trauma patients: a review. J Emerg Med
17. Santora RJ, Moore FA. Monitoring trauma and intensive care unit resuscitation
with tissue hemoglobin oxygen saturation. Crit Care
18. Crookes BA, Cohn SM, Bloch S, Amortequi J, Manning R, Li P, et al. Can near-infrared spectroscopy identify the severity of shock in trauma patients? J Trauma
19. Huang L, Ding H, Hou X, Zhou C, Wang G, Tian F. Assessment of the hypoxic-ischemic encephalopathy in neonates using non-invasive near-infrared spectroscopy. Physiol Meas
20. Weiss M, Dullenkopf A, Kolarova A, Schulz G, Frey B, Baenziger O. Near-infrared spectroscopic cerebral oxygenation reading in neonates and infants is associated with central venous oxygen saturation. Paediatr Anaesth
21. Nagdyman N, Fleck T, Schubert S, Ewert P, Peters B, Lange PE, et al. Comparison between cerebral tissue oxygenation index measured by near-infrared spectroscopy and venous jugular bulb saturation in children. Intensive Care Med
22. Beilman GJ, Groehler KE, Lazaron V, Ortner JP. Near-infrared spectroscopy measurement of regional tissue oxyhemoglobin saturation during hemorrhagic shock. Shock
23. Rhee P, Langdale L, Mock C, Gentilello LM. Near-infrared spectroscopy: continuous measurement of cytochrome oxidation during hemorrhagic shock. Crit Care Med
24. Taylor JH, Mulier KE, Myers DE, Beilman GJ. Use of near-infrared spectroscopy in early determination of irreversible hemorrhagic shock. J Trauma
25. Cohn SM, Nathens AB, Moore FA, Rhee P, Puyana JC, Moore EE, et al. Tissue oxygen saturation predicts the development of organ dysfunction during traumatic shock resuscitation
. J Trauma
26. Meybohm P, Renner J, Boening A, Cavus E, Gräsner JT, Grünewald M, et al. Impact of norepinephrine and fluid on cerebral oxygenation in experimental hemorrhagic shock. Pediatr Res
27. Cohn SM, Crookes BA, Proctor KG. Near-infrared spectroscopy in resuscitation
. J Trauma
28. Moore FA, Nelson T, McKinley BA, Moore EE, Nathens AB, Rhee B, et al. Massive transfusion in trauma patients: tissue hemoglobin oxygen saturation predicts poor outcome. J Trauma
29. Smith J, Bricker S, Putnam B. Tissue oxygen saturation predicts the need for early blood transfusion in trauma patients. Am Surg
30. Lima A, van Bommel J, Jansen TC, Ince C, Bakker J. Low tissue oxygen saturation at the end of early goal-directed therapy is associated with worse outcome in critically ill patients. Crit Care