Cerebrovascular response to continuous cold perfusion and hypothermic circulatory arrest

OBJECTIVE

Clinical and laboratory studies have documented changes in cerebrovascular resistance after hypothermic circulatory arrest, both with and without adjunctive cerebral perfusion modalities. This study was designed to clarify whether these changes are due to cerebral edema, resistance vessel abnormalities, or alterations in the cerebral microcirculation.

METHODS

Four mature swine underwent hypothermic circulatory arrest for 60 minutes, and 7 mature swine underwent cold cerebral perfusion for 60 minutes to simulate antegrade selective perfusion. All were rewarmed and weaned from cardiopulmonary bypass. Pial vascular diameter and reactivity were measured in vivo through a cranial window and ex vivo in an organ chamber; cerebral microvascular endothelium was studied in culture for release of vasoactive mediators. Cerebral water content was recorded.

RESULTS

Cold perfusion caused pial arteriole and venule constriction, whereas hypothermic circulatory arrest alone caused pial arteriole and venule dilatation. Cold perfusion caused a temporal loss of endothelium-dependent vasodilatation, most notably to bradykinin. Hypothermic circulatory arrest caused a loss of nitric oxide-mediated endothelium-dependent vasodilatation. Endothelium-independent vasoreactivity remained intact in both groups. Endothelial cells from the cold group had a vasoconstrictive secretory phenotype, whereas endothelial cells from the hypothermic circulatory arrest group had a vasodilatory phenotype. Cerebral water content was the same in both groups.

CONCLUSION

The increase in cerebrovascular resistance observed after cold cerebral perfusion is caused by resistance vessel constriction and may be promoted by an altered microcirculation. Hypothermic circulatory arrest alone is associated with endothelium-dependent vasoparesis. Both could contribute to cerebral injury in the early hours after operation.

Matrix metalloproteinase and tissue inhibitor expression in atherosclerotic and nonatherosclerotic thoracic aortic aneurysms

OBJECTIVES

The altered expression of matrix metalloproteinases and their inhibitors influences the formation of atherosclerotic abdominal aortic aneurysms. Their association with thoracic aneurysms is less clear. This study describes the expression of metalloproteinases and their inhibitors in atherosclerotic and nonatherosclerotic thoracic aneurysms, and compares these with age-matched controls.

METHODS

Matrix metalloproteinase-2 and 9 activity were measured by antibody capture, and tissue inhibitor-1 and 2 levels were measured by enzyme-linked immunosorbent assay in 24 patients with atherosclerotic aneurysms and in 63 patients with nonatherosclerotic aneurysms. Gene expression was assessed with reverse transcriptase polymerase chain reaction. The results were compared with 17 controls.

RESULTS

Data are in nanograms per milligram of protein. Matrix metalloproteinase-2 activity was greater in controls than in the atherosclerotic and nonatherosclerotic groups (80 +/- 67 vs 49 +/- 50 and 35 +/- 44, P = .002). Matrix metalloproteinase-9 activity was greater in the atherosclerotic group than in the nonatherosclerotic group and controls (11.7 +/- 15.7 vs 2.5 +/- 2.2 and 1.7 +/- 1.9, P = .001). Tissue inhibitor-1 and 2 levels were greater in controls than in either aneurysm group (tissue inhibitor of metalloproteinase-1: 376 +/- 192 vs 234 +/- 233 and 174 +/- 148, P = .003; tissue inhibitor of metalloproteinase-2: 143 +/- 74 vs 14 +/- 13 and 27 +/- 43, P < .001). Atherosclerotic aneurysms expressed more matrix metalloproteinase mRNA than controls.

CONCLUSIONS

The metalloproteinase/tissue inhibitor phenotype of atherosclerotic thoracic aneurysms is similar to that of abdominal aneurysms. The diminished expression of metalloproteinases and tissue inhibitors in nonatherosclerotic thoracic aneurysms relative to aged controls may represent a loss of smooth muscle cells.

Matrix metalloproteinase activity in thoracic aortic aneurysms associated with bicuspid and tricuspid aortic valves

OBJECTIVE

Matrix metalloproteinases are endopeptidases that function in cell matrix turnover. Abnormal matrix metalloproteinase activity has been implicated in the formation of atherosclerotic abdominal aortic aneurysms. Recent studies suggest that abnormal matrix metalloproteinase activity may also be associated with the formation of atherosclerotic and nonatherosclerotic thoracic aortic aneurysms. Bicuspid aortic valves are associated with an intrinsic aortic pathology that predisposes to formation of proximal thoracic aneurysms while tricuspid aortic valves are not. The objective of this study was to compare the activities of matrix metalloproteinases and levels of their inhibitors in thoracic aneurysms of patients with bicuspid and tricuspid aortic valves.

METHODS

Endogenous and total activity of matrix metalloproteinase-2 and matrix metalloproteinase-9 were measured in proximal nonatherosclerotic thoracic aortic aneurysms of 16 patients with bicuspid aortic valves and 12 patients with tricuspid aortic valves. Levels of tissue inhibitor metalloproteinase-1 and -2 were also measured. Results were standardized to total protein (mg).

RESULTS

Total matrix metalloproteinase-2 activity was greater in aneurysms associated with bicuspid valves when compared with those from tricuspid valves (43 +/- 11 ng/mg vs 14 +/- 2 ng/mg, P =.02). Total matrix metalloproteinase-9 activity was also greater in aneurysms associated with bicuspid aortic valves (4.0 +/- 0.9 vs 1.5 +/- 0.3, P =.02). There was no meaningful difference between groups in levels of tissue inhibitor-1 and -2.

CONCLUSION

The increased activity of matrix metalloproteinases in the walls of aneurysms associated with bicuspid aortic valves may partly explain the predilection to aneurysm formation in these patients.

A simple trick for repairing coronary pseudoaneurysm complicating a Bentall operation

Coronary pseudoaneurysms are a known complication of the Bentall wrap-inclusion method of composite valve grafting. We describe two cases to illustrate a straightforward technique for repair and prevention of coronary pseudoaneurysm formation.

Mycotic arch aneurysm and aortoesophageal fistula in a patient with melioidosis

Aortoesophageal fistula due to an aortic arch aneurysm is a rare entity with an extremely high mortality. There are few reports of successfully managed cases and even fewer of long term survival. We report a case of an aortoesophageal fistula resulting from a mycotic pseudoaneurysm of the distal aortic arch in a patient with melioidosis, its surgical management, and outcome.

The effect of lesion volume on cerebral vasomotor tone after focal brain injury and shock

We have previously shown that the volume of a focal brain injury influences cerebral blood flow. We hypothesized that the cerebral vasomotor tone after traumatic brain injury and shock is related to lesion volume and that the size of the lesion would affect vasomotor reactivity. Swine were randomized to receive either a large or small cryogenic injury followed by shock, and were studied for 5 h postresuscitation. A small brain injury and shock produced a significant and sustained increase in bihemispheric pial arteriolar diameter when compared to the large lesion group and controls. A large brain injury and shock resulted in a significant decrease in the pial arteriolar diameter in the injured hemisphere. We also noted significant differences between and within groups in interhemispheric pial arteriolar diameter and pial arteriolar reactivity to acetylcholine and hypocarbia. These data suggest that the volume of injured tissue influences cerebral blood flow by a vascular mechanism, which may be due in part to an alteration in cerebral endothelial cell function.

Colloid infusion after brain injury: effect on intracranial pressure, cerebral blood flow, and oxygen delivery

OBJECTIVES

We sought to determine the effects of colloid osmotic pressure on cerebral edema formation after brain injury. We hypothesized that an increase in plasma oncotic pressure accompanying a colloid infusion would be associated with a decrease in intracranial pressure and increases in cerebral blood flow and oxygen delivery when compared with isotonic crystalloid.

DESIGN

Prospective, laboratory study.

SETTING

University surgical research laboratory.

SUBJECTS

Adult swine, both genders.

INTERVENTIONS

Cryogenic brain injury with intravenous fluid infusion of either lactated Ringer's solution or 6% dextran-70 in normal saline. The effect of this intervention was monitored for 24 hrs.

MEASUREMENTS

Mean arterial pressure, central venous pressure, intracranial pressure, hemoglobin concentration, plasma oncotic pressure, serum osmolality, cerebral blood flow, and specific gravity of cortical biopsies.

RESULTS

Cryogenic injury significantly increased the cortical water content and the intracranial pressure and significantly decreased the cerebral blood flow and oxygen delivery (p < .05). Dextran infusion significantly increased the colloid oncotic pressure. There were no differences between the lactated Ringer's solution and dextran groups in intracranial pressure, cerebral oxygen delivery, or cortical water content after 24 hrs.

CONCLUSIONS

Colloid infusion after a focal cryogenic injury does not increase cerebral oxygen delivery or reduce either cerebral edema formation or intracranial pressure when compared with lactated Ringer's solution. Colloid is not superior to isotonic crystalloid in the management of isolated brain injury.

The effect of hypertonic resuscitation on pial arteriolar tone after brain injury and shock

Acute brain injury followed by hemorrhagic shock (HEM) causes prohibitive mortality in trauma patients because these combined events lead to low cerebral blood flow (CBF) and cerebral oxygen delivery (co2del). Proper treatment therefore requires rapid correction of cerebral perfusion deficits. Previous studies have shown that hypertonic crystalloid resuscitation significantly improves CBF and co2del in a model of brain injury and HEM when compared to lactated Ringer's (LR) solution. The mechanism or mechanisms for this advantage, however, are not well understood. We hypothesized that hypertonic fluid resuscitation would reduce pial arteriolar tone after brain injury and HEM resulting in an increase in CBF when compared to LR resuscitation. We measured cerebral and systemic variables in a porcine model of focal cryogenic brain injury and hemorrhagic shock over a 5-hour period. Swine were randomized to receive either hypertonic sodium lactate (HSL) or LR fluid resuscitation. The HSL resuscitation produced a significant and sustained elevation in cerebral perfusion pressure and pial arteriole diameter (p < 0.05), and a sustained elevation in CBF after brain injury and HEM when compared with LR. These data suggest that hypertonic fluid resuscitation following brain injury and HEM improves CBF, at least in part, by causing vasodilation of cerebral resistance vessels.

The relationship of fluid balance and sodium administration to cerebral edema formation and intracranial pressure in a porcine model of brain injury

BACKGROUND

Fluid and sodium restriction have been advocated after brain injury as a means of controlling intracranial pressure (ICP). Recent clinical data showing no significant relationships between the amount of fluid infused or sodium administered (Na) and ICP question this practice.

OBJECTIVE

To analyze the relationship of amount of fluid, Na, and fluid balance to cerebral edema formation and ICP.

METHODS

A cryogenic brain injury with and without hemorrhagic shock was studied after 24 hours in swine (n = 35) randomized to receive either lactated Ringer's solution (LR) or hypertonic sodium lactate (HSL). Cerebral edema formation as indicated by cortical water content (CWC) was determined by measurement of specific gravity.

RESULTS

There was a significant positive correlation between the following variables: (1) amount of fluid and ICP (r = 0.598; p < 0.01); (2) fluid balance and ICP (r = 0.684; p < 0.01); and (3) free water and ICP (r = 0.614; p < 0.01). There was a significant negative correlation between serum osmolarity and ICP (r = -0.654; p < 0.01). The study failed to demonstrate a significant correlation between Na and ICP, amount of fluid and CWC, or fluid balance and CWC.

CONCLUSIONS

These data suggest that both the volume of fluid infused and the fluid balance do affect the ICP, but the amount of Na infused does not. The lack of a significant correlation between any of the independent variables and CWC suggests that their effect on ICP is not related to cerebral edema formation. These findings, combined with the observed significant positive correlation between free H2O infused and ICP, and the significant negative correlation between serum osmolarity and ICP, suggest that HSL resuscitation increases intracranial compliance after brain injury while LR decreases it. The data also suggest that free water restriction is warranted in patients with head injuries.

Brain edema formation after brain injury, shock, and resuscitation: effects of venous and arterial pressure

BACKGROUND

Recent work suggests that increased intracranial pressure (ICP) following brain injury and shock is related to increased central venous pressure (CVP) following resuscitation.

OBJECTIVE

To analyze the relationship of intravascular pressures to edema formation and ICP in an experimental model.

METHODS

In a porcine model of cryogenic brain injury and hemorrhagic shock, we studied CVP, mean arterial pressure (MAP), ICP, and cortical water content (CWC, as cortical specific gravity) at baseline (BL), 45 minutes after shock (H45), and 1, 3, 6, 12, and 24 hours (H) after resuscitation. Group 1 was the control group, group 2 brain injury only, group 3 shock only, and group 4 brain injury and shock.

RESULTS

Brain injury significantly increased ICP and CWC. Mean arterial pressure significantly correlated with ICP (r = 0.54, p = 0.02) and with CWC (r = -0.48, p = 0.03) in group 4 at 24H but not in the other groups at any time period. There was no significant correlation between CVP and ICP or CWC in any group at any time interval.

CONCLUSIONS

These data suggest that brain edema formation in the injured hemisphere is related to MAP and not CVP, but variability in MAP accounts for only 29% of the variability in CWC and ICP, suggesting the importance of factors other than hydrostatic pressure in determining the amount of edema and the ICP after brain injury. Previous work demonstrating the significant correlation of polymorphonuclear leukocyte infiltration with ICP (r = 0.71, p < 0.001) and with CWC (r = -0.63, p < 0.001) suggests that inflammation may be one of these factors.

Effect of lesion volume on cerebral hemodynamics after focal brain injury and shock

Cerebral blood flow (CBF) varies unpredictably in patients after head injury and hemorrhagic shock. Proper treatment requires knowledge of ischemic versus hyperemic flow. The degree to which the size or severity of the injury may contribute to CBF abnormalities is unknown. We hypothesized that lesion size is a determinant of postinjury CBF. We measured cerebral and systemic variables in a porcine model of focal cryogenic brain injury and hemorrhagic shock over a 5-hour period. Swine were randomized to receive either a large or small lesion followed by hemorrhage. In the small lesion group traumatic brain injury, followed by shock and resuscitation, produced a significant and sustained elevation in bihemispheric regional CBF and cerebral oxygen delivery that was significantly greater than that observed in either the large lesion group or the controls (p < 0.05). There were no significant differences between the experimental groups in volume of hemorrhage, intracranial pressure, cerebral perfusion pressure, arterial oxygen content, or PaCO2. These data suggest that the volume of injured tissue may determine post-resuscitation CBF, and that interventions to reduce cerebral blood volume (i.e., hyperventilation) may not be universally applicable in all head injured patients.

The association of leukocytes with secondary brain injury

Shock increases mortality from brain injuries, but the mechanism is poorly understood. We hypothesized that brain injury followed by shock and resuscitation leads to a secondary reperfusion injury mediated in part by polymorphonuclear leukocytes (PMNs). To validate this hypothesis, we studied cerebral perfusion pressure (CPP), intracranial pressure (ICP), cerebral blood flow (CBF), cortical water content (CWC), and hemodynamic variables in a porcine model of focal cryogenic brain injury and hemorrhagic shock. Cerebral PMN accumulation (CPMN) in the injured and uninjured hemispheres was determined histologically from the total PMNs in five high-power fields (400x). Twenty-nine mature swine were randomized to four groups. Group 1, the control group, was instrumented only. Group 2 animals had a brain injury alone and were studied for 24 hours. Group 3 animals had a brain injury and hemorrhagic shock. Group 4 animals had hemorrhagic shock alone. Brain injury followed by shock caused a significantly greater ICP and a significantly lower CBF than brain injury or shock alone. There was no significant difference in CPP between groups after resuscitation. The CWC of the lesioned area was similar in both brain-injured groups but was significantly increased when compared with the controls and the shock-only group. The CWC of the nonlesioned hemisphere was higher in group 3 than in group 2. The CPMN in both hemispheres in group 3 was significantly greater than in either group 2 or group 4. There was a significant positive correlation between CPMN and both ICP and CWC, and a significant negative correlation between CPMN and CBF. These data suggest an association between CPMN accumulation and secondary brain injury.

The deleterious effects of intraoperative hypotension on outcome in patients with severe head injuries

Prehospital or admission hypotension doubles the mortality for patients with severe head injury (SHI = Glasgow Coma Scale score less than or equal to 8). To our knowledge no study to date has determined the effects of intraoperative hypotension [IH: systolic blood pressure (SBP) less than 90 mm Hg] on outcome in patients with SHI. This study examined 53 patients who had SHI and required early surgical intervention (surgery within 72 hours of injury). All patients were initially normotensive on arrival. There were 17 patients (32%) who developed IH and 36 (68%) who remained normotensive throughout surgery. The mortality rate was 82% in the IH group and 25% in the normotensive group (p less than 0.001). The duration of IH was inversely correlated with Glasgow Outcome Scale using linear regression (R = -0.30; p = 0.02). Despite vigorous fluid resuscitation in the IH group, additional pharmacologic support was used in only 32%. These data suggest that IH is not uncommon after SHI (32%) and that it does have a significant effect on patient outcome.

An analysis of the relationship between fluid and sodium administration and intracranial pressure after head injury

Severe head injury is the leading cause of traumatic death. When a severe head injury is combined with hypotension the mortality doubles. The use of asanguineous salt solutions to maintain blood pressure, however, may contribute to cerebral swelling and intracranial hypertension. For this reason, restrictions of fluids (FLD) and sodium (Na) have been advocated. To our knowledge, however, there are no clinical data to support this recommendation. We hypothesized that in adult patients sustaining severe head injuries (Glasgow Coma Scale score less than or equal to 8) with or without associated injuries: (1) FLD balance and total Na administered during the initial 72 hours of hospital admission are positively and significantly correlated with each other, and (2) total FLD, FLD balance, and total Na administration during the initial 72 hours are significantly and positively correlated with changes in ICP and adverse outcome. We retrospectively studied 40 adult trauma patients with severe head injuries. We found a significant correlation between total Na and FLD balance (R2 = 0.54; p less than 0.05). However, we found no significant correlation between total FLD and maximum ICP (R2 = 0.081), ICP score (R2 = 0.01), or outcome (R2 = 0.066), no significant correlation between FLD balance and maximum ICP (R2 = 0.000), ICP score (R2 = 0.000), or outcome (R2 = 0.01), and no significant correlation between total Na and maximum ICP (R2 = 0.000), ICP score (R2 = 0.001), or outcome (R2 = 0.02). We conclude that Na and FLD administration are not independent determinants of ICP during the initial 72 hours after brain injury.

Focal brain injury results in severe cerebral ischemia despite maintenance of cerebral perfusion pressure

Severe head injury often causes an increase in intracranial pressure (ICP) and decreases in cerebral blood flow (CBF) and cerebral oxygen delivery (CO2del). To determine if this reduction in CBF and CO2del would produce cerebral ischemia and if this reduction would be abrogated by maintaining global cerebral perfusion pressure (CPP), we studied CPP, ICP, CBF, CO2del, cerebral oxygen extraction ratio (CO2ER), and cortical water content (CWC) in a porcine model of focal cryogenic brain injury. Fifteen mature swine were randomized to two groups. The experimental group (n = 7) had a brain lesion and was studied for 24 hours. The control group (n = 8) was instrumented only. Cryogenic injury significantly increased ICP and decreased CBF and CO2del compared with controls. There were no significant differences in CPP between the groups for the entire experiment, and the CPP was well above the ischemic threshold. The CO2ER significantly increased in the first three hours after brain injury. However, CO2ER in experimental animals tended to decrease 12 hours after brain injury and was not significantly different from that in controls. Cryogenic injury significantly increased the CWC in the lesioned hemisphere. These data indicate that focal brain injury results in persistent ischemia despite the normalization of CPP, suggesting that a significant increase in cerebral vascular resistance (CVR) occurs after brain injury. We conclude that in addition to maintenance of CPP, intervention to reduce CVR may be important in the management of brain injury.

Hemorrhagic hypotension after brain injury causes an early and sustained reduction in cerebral oxygen delivery despite normalization of systemic oxygen delivery

Morbidity and mortality are doubled when hemorrhagic hypotension (HEM) accompanies a traumatic brain injury (TBI). Hemorrhagic hypotension initiates a "secondary" injury (SI) that has been attributed to ischemia, but this has not been confirmed in the laboratory. All previous studies have been of relatively short duration (less than 6 hours), allowing insufficient time to study the pathophysiology of SI, since maximal intracranial pressure (ICP) elevations may occur 16 to 20 hours after injury. We hypothesized that HEM combined with TBI would reduce cerebral oxygen delivery (cO2del) and cerebral metabolic rate for oxygen (cMRO2) to a greater degree than would occur with TBI alone. In a porcine model of TBI and HEM we recorded systemic oxygen delivery (sO2del), ICP, cerebral blood flow (CBF), cO2del, cMRO2, brain oxygen extraction ratio (cO2ER), and cortical water content (CWC) over a 24-hour study period. Controls (n = 7) were instrumented only, group 1 (n = 14) received a focal cryogenic lesion only, group 2 (n = 21) received a cryogenic lesion plus hemorrhage to 50 mm Hg for 45 minutes. Animals were resuscitated with crystalloid solutions; shed blood in group 2 animals was returned after one hour. Hemorrhagic hypotension following TBI produced a significant and sustained reduction in cO2del associated with a lower cMRO2 and cO2ER, and higher ICP and CWC, than seen with lesion alone. This occurred despite adequate early restoration of sO2del. This confirms that cerebral ischemia is ongoing despite restoration of systemic hemodynamics.

Hypertonic fluid resuscitation improves cerebral oxygen delivery and reduces intracranial pressure after hemorrhagic shock

Prospective clinical studies have shown that hypotension from hemorrhage contributes to increased morbidity and mortality in patients with traumatic brain injury. It is implied that poorer outcome is the result of secondary brain injury from impaired cerebral oxygen delivery (cO2del). We studied the early and late effects of hypertonic sodium lactate (HSL: 500 mOsm/L) resuscitation on mean arterial pressure (MAP), cardiac output (CO), systemic oxygen delivery (sO2del), cerebral perfusion pressure (CPP), intracranial pressure (ICP), cO2del, cerebral blood flow (CBF), serum osmolality, and cortical water content (CWC) in a porcine model of hemorrhagic shock. Swine were randomized to receive a bolus (4 mL/kg) of either lactated Ringer's solution (LR: 274 mOsm/L) or HSL after shock, followed by either LR or HSL to return MAP to baseline levels. Shed blood was returned 1 hour after resuscitation, and all animals were studied for 24 hours. Control animals were instrumented only. The HSL resuscitation significantly increased cO2del and CBF for 24 hours postresuscitation when compared with LR. The ICP in the HSL-treated animals was significantly lower throughout the postresuscitation phase when compared with the LR-treated animals (p less than 0.05). The CWC was significantly lower in the HSL-treated animals (p less than 0.05). We attribute these effects to hypertonic dehydration of both the brain parenchyma and the cerebrovascular endothelium. These data suggest that by decreasing ICP and improving cO2del after shock, HSL could decrease secondary brain injury when brain injury and shock occur together.