The Acute Effects of Hemorrhagic Shock on Cerebral Blood Flow,Brain Tissue Oxygen Tension,and Spreading Depolarization following Penetrating Ballistic-Like Brain Injury

Introduction

T raumatic brain injury (TBI) causes immediate and complex physiological changes, including cerebral hemodynamic instability, ionic balance disturbance, and increased intracranial pressure (ICP). These changes contribute to secondary insults and worsen outcomes in TBI patients. Following brain trauma, normal cerebrovascular autoregulation may be disrupted. As a result of heterogeneity of trauma, cerebral blood flow (CBF) irregularities vary from abnormally low to abnormally high. ¹ Low CBF leads to depressed brain metabolism and ischemia, whereas augmented CBF (hyperemia) exacerbates brain swelling and high ICP. Both events impair local delivery of oxygen and glucose, leading to low brain tissue oxygen tension (P_btO₂), ionic imbalance, and disturbance of mitochondrial functions and brain functionality. Sustained breakdown of ion homeostasis results in cortical spreading depolarization (CSD), characterized as near-complete depolarization of neurons and glia recurring in gray matter that triggers intracellular signaling pathways, leading to cell death. ² In TBI, the cascades of injury processes make the compromised tissues more vulnerable to additional insults, such as hypotension resulting from blood loss, systemic inflammatory responses resulting from bone fracture, and sepsis (i.e., polytrauma). Hemorrhagic hypotension is commonly associated with worsening morbidity and mortality in TBI patients. ³ Animal studies combining TBI models and hemorrhagic hypotension demonstrated a delayed recovery of cognitive function, ⁴ increased contusion volume and reduction of CBF, ⁵ increased blood–brain barrier (BBB) disruption and acute neuronal damage, ⁶ as well as down-regulation of neuroprotective genes in hippocampal neurons. ⁷ In addition, combined hypobaric hypotension and cortical cold lesion triggered an additional CSD coupled with augmented CBF, which enhanced lesion expansion. ⁸

Spontaneous CSD has been demonstrated in TBI patients ^9,10 and in experimental animal models of brain injury. ^8,11 Under pathological conditions, CSD causes vasoconstriction as a result of disrupted neurovascular response, resulting in spreading ischemia. ² The mismatch between energy demand and supply prolongs neuronal depolarization, which triggers intracellular signaling pathways leading to cell death. ¹² Whether CSDs contribute to post-traumatic secondary brain damage remains open to debate. On one hand, a study that induced additional CSDs by potassium chloride (KCl) injections into the injured hemisphere of TBI-injured mice failed to show a significant relationship between the number of CSDs and lesion size. ¹³ In contrast, Rogatsky and colleagues suggested that an increasing number of spontaneous CSDs accompanied by high ICP (≥20 mmHg) exacerbated brain damage. ¹⁴

Previous work investigating CSD occurrence in a rat model of penetrating ballistic-like brain injury (PBBI) demonstrated evidence of CSD from 5 to 72 h postinjury. ¹⁵ The acute incidence of depolarization (within 2 h postinjury) and the changes in CBF and P_btO₂ have not been examined. Moreover, the effect of hemorrhagic hypotension on these acute physiological changes after brain trauma is unclear. In the present study, we investigated the effect of hypotension on spreading depolarization, CBF, and P_btO₂ after PBBI. In particular, these changes were examined in core lesion and perilesional areas simultaneously, thus providing a comprehensive physiological profile acutely subsequent to the injury.

Methods

A total of 60 male adult Sprague-Dawley rats (280–320 g; Charles River Labs, Raleigh, VA) were used in these experiments. Animals were randomized between the following four groups: sham (n=16); PBBI (n=17); hemorrhagic shock (HS; n=12) or PBBI+HS (n=15). All procedures involving animal use were reviewed and approved by the institutional animal care and use committee of Walter Reed Army Institute of Research (Silver Springs, MD). Research was conducted in compliance with the animal welfare act, Guide for the Care and Use of Laboratory Animals (National Research Council, Washington, DC), and other federal statutes and regulations. Animals were housed individually under a 12-h light/dark cycle in a facility accredited by the International Association for Assessment and Accreditation of Laboratory Animal Care.

General procedures

Anesthesia was induced with 3.5% isoflurane delivered in air/oxygen mixture (F_iO₂=0.26) and maintained at 1.5% throughout the surgery and recording. Core body temperature was maintained at 37.0°C by a homeothermic heating system (Harvard Apparatus, Holliston, MA). In all animals, the right femoral artery and vein were cannulated for mean arterial blood pressure (MAP) monitoring and fluid resuscitation respectively. In addition, the tail artery was cannulated for inducing HS by withdrawing blood and for blood gas analysis. Partial pressures of oxygen and carbon dioxide as well as pH were measured with a blood gas analyzer (ABL5; Radiometer America Inc., Westlake, OH) before and after injury. To detect the cortical spreading depolarization and its associated changes of regional CBF (rCBF) and P_btO₂, electrodes (epidural) and probes (dorsoventral 2.5 mm from dura) were placed at the frontal and parietal cortices representing the core lesion and perilesional areas, as shown in Figure 1A. Signals were allowed to stabilize (∼30 min). After 20 min of baseline recording, rats were subjected to either PBBI or HS. In the PBBI+HS group, HS started at 5 min after PBBI. Sham control animals received craniotomy only. Signals were recorded continuously for 150 min, at which time rats were euthanized. As a positive control of the experiment, CSD was triggered by topical application of 1 M of KCl solution at the right frontal cortex (PBBI location) in a separate group of animals (n=5). The waveform, velocity of the CSD propagation and its duration, as well as the changes in rCBF and P_btO₂ were examined.

OPEN IN VIEWER

FIG. 1.

(A) Locations and stereotaxic coordinates of PBBI probe insertion, electrodes for DC shift, recording probes for P_btO₂, and laser Doppler probes for CBF. (B) Representative traces from simultaneous P_btO₂, CBF, and DC shift recordings in animals subjected to KCl (left), PBBI (middle), or HS (right). The two DC shift recording sites were 3 mm apart. Latency and duration were defined. Scale bars represent the following: P_btO₂=1 mmHg; CBF=10 flux; DC shift=2 mV; time=10 sec. PBBI, penetrating ballistic-like brain injury; DC, direct current; P_btO₂, brain tissue oxygen tension; CBF, cerebral blood flow; KCl, potassium chloride; HS, hemorrhagic shock.

Results

Acute physiological changes were recorded continuously before and after injury. Figure 1B shows the representative tracings of rCBF, P_btO₂, and DC shifts evoked by KCl, which serves as a positive control. Application of KCl for 6 min triggered a mean of 10 depolarizations per rat, with a propagation speed of approximately 4.08 mm/min on the contralateral cortex and 4.10 mm/min on the ipsilateral cortex. CSD was coupled with slight increases in CBF and P_btO₂ (Fig. 1B).

Reduction in blood pressure and heart rate after PBBI+HS

MAP remained between 85 and 100 mmHg throughout the recording period in the sham control group (Fig. 2A). In animals subjected to PBBI, significant reductions in MAP were detected at the time of injury that recovered within 5 min. In contrast, HS-induced reductions in MAP did not resolve in either the HS or PBBI+HS groups, even after fluid resuscitation (p<0.05). Notably, PBBI also caused a significant, abrupt drop in heart rate that recovered within 30 min post-PBBI (Fig. 2B). During hemorrhagic hypotension, the heart rate decreased significantly from 380 to 260 beats per minute in both groups, showing an initial uncompensated response resulting from blood loss. It remained significantly lower in the PBBI+HS group after fluid resuscitation, compared to that in the PBBI group (p<0.05). Notably, MAP was slightly lower in the HS-alone group than that in the PBBI+HS group during hypotension and after resuscitation. Levels of pH, partial pressure oxygen, and carbon dioxide in arterial blood between groups (Table 1) remained within a normal range after injury.

OPEN IN VIEWER

FIG. 2.

(A) Mean arterial blood pressure and (B) heart rate over time. *p<0.05, compared between sham and PBBI+HS groups; ^‡ p<0.05, compared between PBBI and PBBI+HS groups; ^§ p<0.05, compared between HS and PBBI+HS (Fisher's LSD post-hoc tests). PBBI, penetrating ballistic-like brain injury; HS, hemorrhagic shock; LSD, least significant difference.

OPEN IN VIEWER

Table 1.

Blood Gas Before, During, and After Injury in All Groups

	Preinjury	Injury (PBBI/HS)	Postinjury
pH
Sham	7.44±0.01	7.43±0.01	7.43±0.01
PBBI	7.43±0.01	7.43±0.01	7.43±0.02
HS	7.43±0.01	7.42±0.01	7.43±0.01
PBBI+HS	7.42±0.01	7.42±0.01	7.44±0.01
PaCO2
Sham	41.32±1.28	40.16±2.19	39.25±1.49
PBBI	42.67±1.15	40.26±1.30	39.00±2.18
HS	41.92±1.35	38.36±1.30	39.60±2.36
PBBI+HS	43.40±1.00	41.13±0.77	39.20±1.24
PaO2
Sham	117.68±2.34	120.68±2.55	118.50±2.87
PBBI	111.90±2.87	115.21±2.45	108.22±2.28
HS	112.75±2.88	125.18±4.94	119.20±3.53
PBBI+HS	116.60±2.30	122.40±3.49	122.20±4.79

PBBI, penetrating ballistic-like brain injury; HS, hemorrhagic shock; PBBI+HS, combined PBBI and hemorrhagic shock.

Increased CSD occurrence, reduced propagation speed, and lowered threshold

Spontaneous CSD occurred in 84% (10 of 12) of the rats in the HS group, 79% (13 of 17) in the PBBI group, and 85% (11 of 13) in the PBBI+HS group. No CSD was detected in sham control animals. PBBI caused a massive nonpropagating depolarization at time of injury on both contralateral and ipsilateral cortices (Fig. 1B). However, this was not counted as a CSD event in our analysis. The number of DC shifts occurring in the ipsilateral cortex in the PBBI+HS group (5.92±1.34/rat) was almost six times greater than in the HS-alone (1.00±0.22/rat) and twice the number of occurrences in the PBBI-alone (2.29±0.63/rat) groups (p=0.024; Fig. 3A). The PBBI+HS group displayed a slight increase in CSD in the uninjured (contralateral) hemisphere, compared to the other groups, but the difference was not statistically significant. Patterns of DC shifts varied widely between rats, yet the durations consistently ranged from 18.06 sec to 19.65 min. CSD induced by transient hypotension had a longer duration than that induced by KCl, PBBI, or PBBI+HS on both contralateral and ipsilateral hemispheres. Though the duration appeared to be more prolonged in the PBBI+HS group (contralateral, 150.81±12.82 sec; ipsilateral, 194.71±14.94 sec), compared to the PBBI group (contralateral, 102.34±11.43 sec; ipsilateral, 123.19±11.88 sec), these difference failed to reach statistical significance (Fig. 3B). In some cases, depolarization was detected at only one recording electrode, showing no propagation. In addition, simultaneous or synchronous depolarizations were observed at the two epicortical electrodes (speed, ≥8 mm/min). As such, these cases were excluded from subsequent analysis of CSD propagation speed. HS after PBBI caused a significantly slower propagation (2.97±0.22 mm/min; p<0.01) in the injured cortex, compared to the other groups (KCl, 4.36±0.20; PBBI, 4.84±0.45; HS, 4.62±0.54 mm/min). In addition, CSD induced by HS alone occurred at a MAP level of ∼50 mmHg. Critically, after PBBI, the combined HS insult was able to induce CSD at a significantly higher MAP level (∼65 mmHg; p<0.05), as shown in Figure 3C.

OPEN IN VIEWER

FIG. 3.

Dot density plots of CSD parameters, including (A) the number of CSDs that occurred during the recording period, (B) duration, and (C) threshold with respect to MAP. *p<0.05, compared to sham control; ^‡ p<0.05, compared to the PBBI group; ^§ p<0.05, compared to the HS group (Mann-Whitney's tests with Bonferroni-Holm's correction). CSD, cortical spreading depolarization; MAP, mean arterial pressure; PBBI, penetrating ballistic-like brain injury; HS, hemorrhagic shock; KCl, potassium chloride.

Persistent reduction of P_btO₂ near core lesion

On the ipsilateral side near the core lesion, P_btO₂ reduced significantly after PBBI (70% baseline) and gradually recovered to 80% baseline at the end of the recording. The addition of HS after PBBI further decreased oxygen tension to 40% baseline. After resuscitation with lactated Ringer's solution, oxygen level increased slightly, but was still significantly lower than that in the HS-alone and PBBI-alone groups (p<0.05). In contrast, P_btO₂ in animals subjected to HS only returned to baseline level after resuscitation (Fig. 4A). In the perilesional region (Fig. 4B), P_btO₂ decreased to 80% baseline after PBBI, gradually returning to baseline. HS caused a significant reduction (60% baseline) in P_btO₂ during the hypotensive state that was restored to 80% of baseline in the HS group and 70% in the PBBI+HS group after fluid resuscitation. P_btO₂ increased with time in sham control animals on both contralateral and ipsilateral sides.

OPEN IN VIEWER

FIG. 4.

Temporal changes of P_btO₂ in (A) anterior parietal cortex (near core lesion) and (B) posterior parietal cortex (perilesion area), ipsilateral to the injury. *p<0.05, compared between sham and PBBI+HS groups; ^‡ p<0.05, compared between PBBI and PBBI+HS groups; ^§ p<0.05 compared between HS and PBBI+HS (Fisher's LSD post-hoc tests). P_btO₂, brain tissue oxygen tension; PBBI, penetrating ballistic-like brain injury; HS, hemorrhagic shock; LSD, least significant difference.

Significant decline of rCBF after PBBI in core and perilesion regions

Cerebral blood flow was measured at four locations: contralateral anterior parietal cortex; contralateral posterior parietal cortex; ipsilateral core lesion; and the ipsilateral perilesional area (Fig. 1A). In all regions, rCBF in the sham control group increased over time from 100% baseline to 120%, likely as a result of an isoflurane-induced vasodilation. PBBI caused a significant, persistent reduction of rCBF on the ipsilateral side at the both core (35% baseline; Fig. 5C) and perilesional areas (80% baseline; Fig. 5D), but a gradual increase on the contralateral side (Fig. 5A,B). During the hypotensive state, rCBF dropped to 60–70% of baseline, but was immediately recovered after fluid resuscitation, except in animals subjected to PBBI prior to HS. In this group (i.e., PBBI+HS), fluid resuscitation did not restore rCBF to normal levels in either the core lesion and/or perilesional region. On the uninjured side, the HS and PBBI+HS groups showed similar rCBF changes, in which the rCBF decreased significantly during the hypotensive state and increased immediately after resuscitation (Fig. 5A,B).

OPEN IN VIEWER

FIG. 5.

Temporal changes of CBF in the contralateral hemisphere (A) anterior parietal cortex and (B) posterior parietal cortex and in the ipsilateral hemisphere (C) anterior parietal cortex (near core lesion) and (D) posterior parietal cortex (perilesion area). *p<0.05, compared between sham and PBBI+HS groups; ^‡ p<0.05, compared between PBBI and PBBI+HS groups; ^§ p<0.05, compared between HS and PBBI+HS (Fisher's LSD post-hoc tests). CBF, cerebral blood flow; PBBI, penetrating ballistic-like brain injury; HS, hemorrhagic shock; LSD, least significant difference.

PBBI and HS disrupted autoregulation

Examination of the relationship between rCBF and MAP (Figure 6A–D), which is reflective of cerebral autoregulation, showed that PBBI+HS significantly decreased the threshold for autoregulation (i.e., increased breakpoint), compared to HS alone (Fig. 6E; PBBI+HS, 67.41±1.51 mmHg; HS, 59.41±1.45 mmHg; Mann-Whitney's test, p<0.05). In addition, slope A in both groups were positive, indicating impaired autoregulation when MAP fell below the lower limit of autoregulation. On the contrary, slope B in both groups were negative, implying intact autoregulation when MAP was above the lower limit or the breakpoint. Both slope A and B were significantly greater in the HS-alone group than the PBBI+HS group (Fig. 6F,G), suggesting that after PBBI, the cerebrovascular system might be less responsive to changing MAP, possibly as a result of the severe damage of the vasculature especially on the side of injury (Fig. 6C,D).

OPEN IN VIEWER

FIG. 6.

Piecewise correlations (two segments) between MAP and CBF of representative rats from the HS and PBBI+HS groups at four locations: (A) contralateral anterior parietal cortex; (B) contralateral posterior parietal cortex; (C) ipsilateral anterior parietal cortex; and (D) ipsilateral posterior parietal cortex. The bottom figures showed the dot density plots of (E) breakpoint, (F) slope A, and (G) slope B of the HS and PBBI+HS groups. *p<0.05, compared between the HS and PBBI+HS groups (Mann-Whitney's tests). MAP, mean arterial pressure; CBF, cerebral blood flow; PBBI, penetrating ballistic-like brain injury; HS, hemorrhagic shock.

Discussion

In polytrauma patients, HS-induced hypotension accompanied by TBI is not uncommon. The secondary injury cascade in this combined event is complex, yet poorly understood. Acute physiological changes determine injury outcome and understanding these changes would facilitate the clinical design of neuromonitoring and early therapeutic interventions for polytrauma patients. Therefore, we sought to investigate how HS affects physiological parameters, such as CBF, P_btO₂, and CSD, acutely following PBBI. The main finding of this study was that HS increased CSD incidence after PBBI. CBF and P_btO₂ were significantly reduced, yet they did not change in response to CSD. Additionally, PBBI prior to HS did not adversely affect HS-induced cardiovascular responses.

Our previous study showed that ICP increased abruptly and dramatically in conjunction with the epoch of the PBBI injury. ¹⁸ It is likely that the dynamic pressure caused by the rapid expansion of the elastic tubing on the PBBI probe propagates through the brain parenchyma to the brainstem, causing abnormal cardiovascular and respiratory responses. Because the inflation only lasted for 40 ms, these abnormal responses resolved rapidly and blood pressure returned to normal after the pressure wave dissipated. In the present study, PBBI caused a similar sudden, profound decrease in arterial blood pressure and heart rate. However, the addition of HS led to a more persistent reduction in blood pressure and heart rate. Paradoxical bradycardia was evident in clinical and animal studies of severe blood loss ¹⁹ and was suggested to be a mechanism involving parasympathetic reflex to prevent cardiac collapse. ²⁰ This can be reversed by infusing crystalloid solution, such as lactated Ringer's solution, to restore intravascular volume and thus blood pressure. The combined effect of PBBI and HS on this cardiac reflex was similar to that of HS alone. However, previous animal studies using fluid percussion injury (FPI) have reported negative effects of brain trauma on HS-induced cardiovascular response, where the response was delayed during the initial phase of HS in brain-injured animals, compared to that in animals subjected to HS alone. ^21,22 The previous brain trauma also impeded the efficacy of fluid resuscitation. ²² One plausible explanation of the conflicting result is that the lesion produced by PBBI is mainly the frontal cortex and striatum, whereas FPI affected primarily the parietal cortex and brainstem. ²³ Thus, the effect of brain injury on HS-induced cardiac reflex does not depend solely on the injury severity, as suggested by McMahon and colleagues, ²¹ but is also dependent on the brain regions affected in the TBI model. Interestingly, MAP in the PBBI+HS group was slightly higher than that in the HS-alone group during the hypotensive state and after fluid resuscitation, suggesting that systemic hemodynamics may be augmented by PBBI. Such hyperdynamic response was observed acutely in severe TBI patients and was related to sympathetic nervous hyperactivity. ²⁴ Nevertheless, additional studies specifically designed for cardiovascular homeostasis is required to fully understand this physiological response.

During PBBI, ionic homeostasis was massively disrupted, leading to a near-complete depolarization on both contra- and ipsilateral cortices. A few or zero spontaneous CSDs occurred after this substantial depolarization; however, more recurrent CSDs occurred in animals subjected to HS after PBBI, suggesting that the additional insult worsened the ionic imbalance and energy depletion. In human TBI, a higher incidence of spontaneous CSD has been observed, compared to experimental TBI. ²⁵ Indeed, spreading depolarization has been shown to be less readily elicited in the gyrencephalic brain than in the lissencephalic brain. ²⁶ The high CSD incidence in humans may be explained by the observation that clinical TBI is often accompanied by secondary insults, such as hypoxia and hypotension, and its pathophysiology is more complicated and heterogeneous than occurs in a less-complex TBI animal model. ²⁷

HS alone caused sustained depolarization (i.e., prolonged duration of DC shift). A previous study has demonstrated the prolongation of the CSD duration by reducing MAP to 50 mmHg in rats subjected to systemic hypotension. ²⁸ Unlike KCl-induced CSD, CBF or P_btO₂ were not responsive to HS- or PBBI+HS-induced CSDs in the present study, implicating that the normal hyperemic response was impaired. PBBI rats, however, lacked prolonged DC shift duration. This is consistent with other TBI rodent models, in which the CSDs were of short duration. ¹³ In contrast to the animal studies, a prolonged DC shift has been reported in TBI patients and suggested to be associated with poor prognosis. ²⁹ A possible explanation for this difference between human versus experimental TBI is that the experimental models simply cannot reproduce the complex and heterogeneous nature of the clinical TBI. Notably, the additional HS insult after PBBI in our study appeared to extend CSD duration. Sustained depolarization results in intracellular calcium overload and increased free radical production, triggering apoptotic signals and proteases leading to cell death. ² Apart from extending DC shift duration, combined PBBI and HS also reduced the propagation speed of CSD. Several studies suggested a positive correlation of CSD speed with cortical excitability ³⁰ and MAP ²⁸ as well as inverse correlation with cortical myelination. ³¹ PBBI causes ionic imbalance, damage to white matter, elevated ICP, inflammation, and necrotic and apoptotic cell death. ¹⁵ These might be worsened by the hypotension insult after PBBI, where, in the present study, the PBBI+HS group showed a significant reduction in MAP. The balance between TBI-induced cortical hyperexcitability and hypotension-induced slow propagation determines resultant CSD speed. Further study is required to examine this hypothesis.

PBBI lowered the threshold of spontaneous CSD. Our data indicated that CSD occurred when blood pressure dropped to ∼50 mmHg in rats subjected to HS only, whereas it occurred at higher blood pressure in animals subjected to PBBI prior to HS. Although MAP is an indirect factor, it is associated with cerebral perfusion. Cerebral hypoperfusion depresses electrical activity of the brain, ³² and it has been reported that in TBI patients, CSD incidence increased as a function of reduced MAP and cerebral perfusion pressure. ³³ The threshold of CSD can be modulated by several factors, including reduced extracellular nitric oxide (NO) levels ³⁴ and mutations in the Cacna1A gene that encodes calcium channels ³⁵ and the ATP1A2 gene that encodes the astrocytic sodium pump. ³⁶ Low NO levels as a result of reduced CBF ³⁷ and mutated Cacna1A related to cerebral edema and coma after TBI ³⁸ may account for the lowered threshold (higher susceptibility) to spontaneous CSD after PBBI.

Despite the emergence of CSD, the corresponding changes in rCBF or P_btO₂ during the occurrence of CSD in the injury groups were not observed. Under normal physiological conditions, spreading depolarization is often associated with spreading hyperemia to meet the energy demand. ² Brain tissue in the pathological condition, however, exhibits an inverse hemodynamic response, leading to spreading ischemia, ² that was evident in patients with subarachnoid hemorrhage. ³⁹ Spreading ischemia was not detected in the present study, probably because of the severely damaged cerebral vasculature and the extensive lesion where CBF in the injured hemisphere remained low throughout the recording period after PBBI or PBBI+HS. Interestingly, HS did not further reduce CBF after PBBI, possibly because of the vascular dysfunction and intracerebral hemorrhage. On the other hand, P_btO₂ level in the PBBI+HS group was slightly lower than that in the PBBI group before and after fluid resuscitation, whereas in the PBBI group it decreased shortly after injury, followed by a gradual increase. This gradual increase in P_btO₂ may be a result of the prolonged isoflurane anesthesia in spite of the presence of a lesion. ⁴⁰ In addition to the anesthesia effect, the primary insult causes substantial necrotic cell death ¹⁶ and reduces oxygen demand, resulting in excess oxygen supply. Levels of CBF and P_btO₂ may also have been affected by the fluid resuscitation strategy. In the present study, lactated Ringer's solution was infused intravenously to resuscitate the animal after HS. This is a standard resuscitation regimen for acute hemorrhage, according to The American College of Surgeons Advanced Trauma Life Support guidelines. However, it only temporarily restored the systemic volume and electrolytes, but not CBF or P_btO₂, on the side of injury. The solution rapidly equilibrates with the interstitial space and is not confined to the vascular space. This explains the reduction in MAP shortly after fluid resuscitation, and the reduced MAP might account for the decreased P_btO₂ level in both the HS and PBBI+HS groups after resuscitation. Although infusion of lactated Ringer's solution caused a decrease in MAP, it was capable of restoring CBF to normal levels in the HS group as well as in the contralateral side in the PBBI+HS group. In the core lesion and perilesional regions, CBF remained steadily at a low level. In addition to MAP (beyond the autoregulatory range), ICP and cerebrovascular resistance (CVR) are important determinants of CBF. ⁴¹ Data on ICP and CVR are required to elaborate on the effect of the fluid resuscitation strategy on CBF after combined PBBI and HS.

Autoregulation represents the ability to maintain a stable CBF in response to changes in systemic blood pressure. When blood pressure drops below the autoregulatory range (50–150 mmHg), global hypoperfusion occurs, making the brain more susceptible to injury. ⁴² In the present study, the threshold of cerebral autoregulation impairment was lowered (increased lower limit of autoregulation) after PBBI. Our finding is consistent with a previous study showing an increased lower limit of autoregulation (LLA) after TBI (weight-drop model). The loss of cerebral autoregulation was triggered when MAP was 62.2±20.8 mmHg at 24 h after closed head injury in rats, whereas without brain injury it was triggered at 46.9±12.7 mmHg. This increased LLA may be related to hypotension-induced sympathetic activation. ¹⁷ In addition, it appeared that the cerebral autoregulatory response was not lateralized after a unilateral brain injury. A recent study using a neonatal swine model of severe CCI reported similar findings. ⁴³ In spite of the intact cerebrovascular structure, impaired cerebrovascular response was found in the contralateral hemisphere in animals subjected to PBBI+HS. However, the effect on the contralateral hemisphere may be transient and indeed resolved within hours after injury, which may explain the discrepancy between the clinical and pre-clinical observations where asymmetric cerebral autoregulation has been reported in TBI patients 3 days postinjury. ⁴⁴

The present study was designed to examine the acute changes in CBF, P_btO₂, and CSD as well as the general physiology, including MAP, heart rate, and blood gas. Detailed histopathological analysis was not performed in these animals because of the acute nature of the study and the additional trauma caused by the invasive recording technique and possible neuroprotective effects of prolonged isoflurane anesthesia. Another limitation of our study is the Ag/AgCl ball electrodes used for recording DC shift. These electrodes record the electrical activity on the surface of the cortex by an intact dura where the amplitude of depolarization cannot be accurately measured and was therefore not reported. Glass micropipette electrodes are capable of measuring amplitude in the cortex with minimal tissue disruption. However, this technique is not suitable in the PBBI model because the extremely fine tip of the glass electrode could be broken during the PBBI injury. Nevertheless, duration and number of CSDs are of clinical relevance. Along these lines, our results were consistent with the CSD findings from reported clinical studies in which prolonged duration and increased CSD incidence were observed in TBI patients. ²⁹

In conclusion, our data show that HS after PBBI increased the occurrence of CSD that appeared to have prolonged duration and slower propagation speed. In addition, ipsilateral CBF and P_btO₂ remained low and were not responsive to the depolarizations. These detrimental events, together with the lowered threshold of cerebral autoregulation, are likely to make compromised brain tissue more susceptible to secondary injury, resulting in more cell death and, very possibly, enhanced metabolic dysregulation.