Hypertension with hemodilution prevents multifocal cerebral hypoperfusion after cardiac arrest in dogs

Asphyxiation versus ventricular fibrillation cardiac arrest in dogs. Differences in cerebral resuscitation effects--a preliminary study

We explored the hypothesis that brain damage after cardiac arrest caused by ventricular fibrillation (VF) needs different therapies than that after asphyxiation, which has been studied less thoroughly. In 67 healthy mongrel dogs of both sexes cardiac arrest (at normothermia) by ventricular fibrillation (no blood flow lasting 10 min) or asphyxiation (no blood flow lasting 7 min) was reversed by normothermic external cardiopulmonary resuscitation, followed by intermittent positive-pressure ventilation for 20 h, and intensive care to 96 h. To ameliorate ischemic brain damage, the calcium entry blocker lidoflazine or a solution of free radical scavengers (mannitol and L-methionine in dextran 40) plus magnesium sulphate, was given intravenously immediately upon restoration of spontaneous circulation. Outcome was evaluated as functional deficit, brain creatine kinase (CK) leakage into the cerebrospinal fluid (CSF) and brain morphologic changes. Lidoflazine seemed to improve cerebral outcome after VF but not after asphyxiation. Free radical scavengers plus magnesium sulphate seemed to improve cerebral outcome after asphyxiation, but not after VF. After VF, scattered ischemic neuronal changes in multiple brain regions dominated, and total brain histopathologic damage scores correlated with final neurologic deficit scores at 96 h (r = 0.66) and with peak CK levels in CSF (r = 0.81). After asphyxiation, in addition to the same ischemic neuronal changes, microinfarcts occurred, and there was no correlation between total brain histopathologic damage scores and neurologic deficit scores or CK levels in CSF.

CONCLUSIONS

Different mechanisms of cardiac arrest, which cause different morphologic patterns of brain damage, may need different cerebral resuscitation treatments.

Delayed hyperemia causing intracranial hypertension after cardiopulmonary resuscitation

OBJECTIVE

To clarify whether early or delayed failure of cerebral perfusion after cardiopulmonary resuscitation (CPR) occurs in humans and contributes to secondary brain damage.

DESIGN

Prospective, repeated-measures study.

SETTING

Intensive care unit of Hiroshima University School of Medicine.

PATIENTS

Eight comatose patients who had undergone successful resuscitation from cardiac arrest.

INTERVENTIONS

All patients underwent transcranial Doppler sonography examination. The intracranial cerebral pressure (ICP) and jugular venous oxygen saturation (SO2) also were continuously monitored in five patients and three patients, respectively.

MEASUREMENTS AND MAIN RESULTS

In each patient, we measured the mean flow velocity of the middle cerebral artery transcranially and the mean flow velocity of the internal carotid artery, high in the neck, using transcranial Doppler sonography. The pulsatility index for each measurement was also calculated. The first examinations were performed within 4 to 12 hrs of CPR, and repeat examinations were performed approximately every 12 hrs. The initial mean flow velocities of the middle cerebral artery and the initial mean flow velocities of the internal carotid artery were relatively low, with relatively high pulsatility indices. The mean flow velocities of the middle cerebral artery began to increase at 12 to 24 hrs after CPR and peaked 24 to 120 hrs after CPR. A simultaneous increase in mean flow velocities of the internal carotid artery was observed during this period. The pulsatility index in both arteries dropped significantly during peak mean flow velocity of the middle cerebral artery. In six of seven patients with an abnormal increase (> 100 cm/ sec) in peak mean flow velocity of the middle cerebral artery, the ratio of mean flow velocity of the middle cerebral artery to mean flow velocity of the internal carotid artery was < 3. This value tended to be lower in patients with poor outcomes. An increased mean flow velocity of the middle cerebral artery, with a ratio of < 3 for mean flow velocity of the middle cerebral artery to mean flow velocity of the internal carotid artery, was defined as hyperemia. Although the mean flow velocity of the internal carotid artery was not measured, another patient with an abnormal increase in mean flow velocity of the middle cerebral artery revealed a high jugular venous SO2 value of 83.5%, also representing hyperemia. All ICP values were within the normal range 4 to 12 hrs after CPR and tended to increase before peak mean flow velocity of the middle cerebral artery. The two patients with the lowest ratios of mean flow velocity of the middle cerebral artery to mean flow velocity of the internal carotid artery showed significant increases in ICP after the peak mean flow velocity of the middle cerebral artery. These two patients subsequently developed brain death.

CONCLUSIONS

Delayed hyperemia occurs in humans after resuscitation from cardiac arrest. Our data suggest that this delayed hyperemia can lead to intracranial hypertension and occasionally acute brain swelling, contributing to a poor outcome. A high mean flow velocity of the middle cerebral artery with a low ratio of mean flow velocity of the middle cerebral artery to mean flow velocity of the internal carotid artery may be predictive of critical hyperemia. As an indirect method of measuring cerebral blood flow transcranial Doppler sonography can be used to adjust treatment for failure of cerebral perfusion after resuscitation.

Effect of an endothelin-1 antagonist, BQ-485, on cerebral oxygen metabolism after complete global cerebral ischemia in dogs

Endothelin-1 (ET-1) plays an important role in the physiologic or pathophysiologic regulation of cerebral circulation. To evaluate the effects of the newly synthesized ETA receptor-selective antagonist, BQ-485 (N-perhydroazepin-l-ylcarbonyl-Leu-D-Trp-D-Trp-OH), on the cerebral metabolism of oxygen during the delayed cerebral hypoperfusion that follows global cerebral ischemia, we occluded the ascending aorta and caval veins of 10 beagle dogs for 12.5 min. The animals were randomized into two groups. BQ-485 was given directly into the carotid artery at 0.03 mg/kg per min for 30 min, starting 15 min after reperfusion in the treatment group (n = 5). Isotonic saline was infused in the control group (n = 5). A fiberoptic catheter was inserted into the superior sagittal sinus to monitor its oxygen saturation (SssO2) continuously. Arterial O2 content (CaO2), and sagittal sinus O2 content (CssO2) were monitored before and at 0.5, 1, 2, 4, 6 and 8 h after the ischemic insult. BQ-485 significantly prevented the expected decrease in SssO2 and increase in the cerebral O2 utilization coefficient at 4, 6 and 8 h after the ischemic insult (P < 0.05). Thus, BQ-485 ameliorated the mismatch between O2 supply and demand in the delayed hypoperfusion phase. We conclude that ET may be involved in the pathogenesis of delayed cerebral hypoperfusion after cardiac arrest.

Effect of mild hypothermia on ischemia-induced release of endothelin-1 in dog brain

Endothelin-1 (ET-1) plays an important role in the physiologic or pathophysiologic regulation of cerebral circulation. To evaluate the effect of mild hypothermia on the cerebral concentration of ET-1 and on the cerebral metabolism of oxygen after complete global cerebral ischemia, we occluded the ascending aorta and caval veins of 9 dogs for 15 min. A fiberoptic catheter was inserted into the sagittal sinus to monitor venous oxygen saturation (S(SO)2) continuously. Blood samples were collected 30 min before and 30 min, 1 h, 2 h, 4 h and 6 h after the ischemic insult. Concentrations of ET-1 were assayed in the blood of the sagittal sinus and abdominal aorta. Before, during and after the aortic occlusion, we compared findings in a normothermic control Group 1 (pulmonary artery temperature 38.5 degrees C) (n = 4) with those in the mildly hypothermic Group 2 (pulmonary artery temperature 34.0 degrees C) (n = 5) by surface cooling induced before and maintained during and after ischemia for 6 h. Following ischemia, the plasma concentration difference of ET-1 (sagittal sinus--arterial) was significantly decreased in Group 2 (P < 0.05). Differences in S(SO)2 between the two groups were not statistically significant. Mild hypothermia reduced the ET-1 release in the cerebral circulation but did not improve cerebral oxygen metabolism after complete cerebral ischemia. Findings indicated that the decrease in ET-1 induced by mild hypothermia contributes to the improvement of the cerebral microcirculation after ischemia.

The determinant of severe cerebral dysfunction in patients undergoing emergency extracorporeal life support following cardiopulmonary resuscitation

We investigated the factors associated with cerebral dysfunction in patients undergoing extracorporeal life support (ECLS) following conventional advanced cardiac life support (ACLS). The subjects were 9 patients in whom ECLS was started following ACLS because of intractable cardiac arrest. We investigated whether the irreversibility of cerebral dysfunction during ECLS was related to the cardiopulmonary resuscitation (CPR) time, arterial pH and blood gases, hemoglobin concentration (Hb), peak arterial pressure (PAP) before the start of ECLS and total doses of epinephrine and sodium bicarbonate administered during CPR. Two of the 3 patients who recovered consciousness were weaned from ECLS and survived, while all 6 patients who did not recover from coma were not weaned and died. There was no difference in the CPR time, Hb and PAP before the start of ECLS along with total doses of epinephrine and sodium bicarbonate administered during CPR between the patients who recovered consciousness and those who did not. In addition, there was no difference in arterial pH and blood gases except the arterial oxygen tension (PaO2) between the groups. The PaO2 values before the start of ECLS in the patients who remained in coma ranged from 34 to 58 mmHg, whereas those in the patients who recovered consciousness ranged from 132 to 442 mmHg. The PaO2 values before the start of ECLS in the patients who remained in coma were less than 60 mmHg, whereas those in the patients who recovered consciousness were over 60 mmHg. The present study suggests that hypoxemia during CPR may play a major role in severe cerebral dysfunction in patients undergoing ECLS and PaO2 during CPR.

Early jugular bulb oxygenation monitoring in comatose patients after an out-of-hospital cardiac arrest

OBJECTIVE

To determine the role of early jugular bulb oxygenation monitoring in comatose patients after cardiac arrest.

DESIGN

Prospective sequential study.

SETTING

Medical intensive care unit in a university hospital.

PATIENTS

Thirteen patients comatose after out-of-hospital cardiac arrest.

INTERVENTIONS

A standard hemodynamic protocol.

MEASUREMENTS AND RESULTS

Jugular bulb oxygen saturation levels and oxygen extraction ratios could not discriminate between patients with good (6) and poor (7) cerebral outcome. This was also true for the jugular bulb-arterial lactate difference. Survivors had significantly higher overall oxygen transport values than non-survivors.

CONCLUSIONS

Jugular bulb oxygenation monitoring during the first few hours after cardiac arrest cannot reliably discriminate between comatose patients with a good and poor cerebral outcome. Further studies with an extended monitoring period are thus required.

[The importance of clinical hemorheology in the study of cerebral blood flow in normal conditions and in cerebrovascular ischemia]

A critical review on the importance of hemorheology for establishing clinical management of acute cerebrovascular insufficiency is presented. With this purpose a revision is made on cerebral blood flow, acute cerebrovascular insufficiency, and clinical hemorheology. Data support an evaluation on main drugs presently used in the management of stroke, and on general principles adopted for medical treatment and prevention of stroke.

Norepinephrine-induced hypertension following cardiac arrest: effects on myocardial oxygen use in a swine model

STUDY OBJECTIVE

Recent studies suggest that norepinephrine-induced hypertension early after cardiac arrest ameliorates cerebral hypoperfusion and improves neurologic outcome. The purpose of this study was to evaluate the effects of early norepinephrine-induced hypertension on postresuscitation myocardial blood flow and oxygen use.

DESIGN

Prospective, controlled laboratory study.

PARTICIPANTS

Ten swine.

INTERVENTIONS

All animals underwent 10 minutes of ventricular fibrillation cardiac arrest followed by 5 minutes of low-flow cardiopulmonary bypass (10 mL/kg.min), norepinephrine (0.12 mg/kg), and defibrillation. Animals then were assigned to a hypertension group (mean aortic pressure, 95 mm Hg) or a control group (mean aortic pressure, 75 mm Hg) by titrating a norepinephrine infusion to attain the prescribed aortic pressure.

RESULTS

Myocardial blood flow, perfusion pressure, and oxygen metabolism were compared between groups at different times using analysis of variance with a post-hoc Tukey test. Groups had similar myocardial blood flow during ventricular fibrillation, total defibrillation energy, and time to restoration of spontaneous circulation. Fifteen minutes after restoration of spontaneous circulation, the hypertension group had significantly elevated myocardial blood flow, 965 +/- 314 mL/min.100 g versus 325 +/- 67 mL/min.100 g in the control group (P < .001), myocardial oxygen consumption of 51.2 +/- 26.9 mL O2/min.100 g versus 6.4 +/- 3.4 mL O2/min.100 g (P < .001), and myocardial oxygen extraction of 46% +/- 20% versus 14% +/- 4% (P < .01).

CONCLUSION

In the early resuscitation period, increasing the norepinephrine dose to induce mild hypertension significantly increases oxygen use in the postischemic myocardium.

Hemoconcentration during cardiac arrest and CPR

The objective of this study was to determine if hemoconcentration occurs during cardiac arrest and cardiopulmonary resuscitation (CPR). The design was an animal model of cardiac arrest and CPR performed at a research institute using six mongrel dogs. After the induction of cardiac arrest, animals were subjected to 4 minutes of ventricular fibrillation followed by 20 minutes of CPR. Resuscitation was then achieved using countershocks, drugs, and intravenous fluids. Hemoglobin concentrations were obtained before arrest and every 5 minutes during CPR. An average peak increase in hemoglobin concentration of 21% was observed during CPR. Hemoconcentration occurs during cardiac arrest and CPR, and this may be a result of a shift in volume from the intravascular to the extravascular space.

Organ blood flow following cardiac arrest in a swine low-flow cardiopulmonary bypass model

STUDY OBJECTIVE

To determine organ blood flow changes, relative to baseline, following cardiac arrest and resuscitation in a closed-chest cardiac arrest swine model using cardiopulmonary bypass to achieve reproducible return of spontaneous circulation (ROSC).

INTERVENTIONS

Following 10 min of ventricular fibrillation (VF), animals (n = 10) received low-flow cardiopulmonary bypass at 10 ml/kg/min from 10-15 min. At 15 min of VF, norepinephrine (0.12 mg/kg) was given and bypass flow increased to 50 ml/kg/min, followed by countershocks at 16 min. Following ROSC, cardiopulmonary bypass was immediately weaned off with norepinephrine support. Organ blood flows were determined during normal sinus rhythm, during reperfusion of VF and during the early post-ROSC period while off cardiopulmonary bypass support. Organ blood flows during the early ROSC period were compared with organ blood flow at baseline and during VF.

RESULTS

During early reperfusion of VF prior to any drug therapy, myocardial, cerebral and abdominal organ blood flows were all low. All animals achieved ROSC at 16.9 +/- 0.7 min and were weaned from bypass in < 5 min following ROSC. During the early post-ROSC period, blood flow to the myocardial, cerebral and adrenal vascular beds was significantly elevated relative to baseline. Simultaneously, blood flow to the kidneys, liver, spleen and lungs was reduced relative to baseline.

CONCLUSIONS

This low-flow bypass model produces reproducible high resuscitation rates and ROSC times. Early post-resuscitation organ blood flow is characterized by a selective hyperemia involving the cerebral, myocardial and adrenal vascular beds, in contrast to hypoperfusion of the pulmonary and mesenteric vascular beds.

Mild hypothermia after cardiac arrest in dogs does not affect postarrest cerebral oxygen uptake/delivery mismatching

PURPOSE

To compare measurements of cerebral arteriovenous oxygen content differences (oxygen extraction ratios, oxygen utilization coefficients) in dogs after cardiac arrest, resuscitated under normothermia vs. mild hypothermia for 1-2 h or 12 h.

METHODS

In 20 dogs, we used our model of ventricular fibrillation (no blood flow) of 12.5 min, reperfusion with brief cardiopulmonary bypass, and controlled ventilation, normotension, normoxemia, and mild hypocapnia to 24 h. We compared a normothermic control Group I (37.5 degrees C) (n = 8); with brief mild hypothermia in Group II (core and tympanic membrane temperature about 34 degrees C during the first hour after arrest) (n = 6); and with prolonged mild hypothermia in Group III (34 degrees C during the first 12 h after arrest) (n = 6).

RESULTS

In Group I, the cerebral arteriovenous O2 content difference was 5.6 +/- 1.6 ml/dl before arrest; was low during reperfusion (transient hyperemia) and increased (worsened) significantly to 8.8 +/- 2.8 ml/dl at 1 h, remained increased until 18 h, and returned to baseline levels at 24 h after reperfusion. These values were not significantly different in hypothermic Groups II and III. The cerebral venous (saggital sinus) PO2 (PssO2) was about 40 mmHg (range 29-53) in all three groups before arrest and decreased significantly below baseline values, between 1 h and 18 h after arrest; the lowest mean values were 19 +/- 19 mmHg in Group I, 15 +/- 8 in Group II (NS), and 21 +/- 3 in Group III (NS). Postarrest PssO2 values of < or = 20 mmHg were found in 6/8 dogs in Group I, 5/6 in Group II and 4/6 in Group III. Among the 120 values of PssO2 measured between 1 h and 18 h after arrest, 32 were below the critical value of 20 mmHg.

CONCLUSIONS

After prolonged cardiac arrest, critically low cerebral venous O2 values suggest inadequate cerebral O2 delivery. Brief or prolonged mild hypothermia after arrest does not mitigate the postarrest cerebral O2 uptake/delivery mismatching.

Cerebral and systemic arteriovenous oxygen monitoring after cardiac arrest. Inadequate cerebral oxygen delivery

BACKGROUND

After prolonged cardiac arrest, under controlled normotension, cardiac output and cerebral blood flow are reduced for several hours. This dog study documents for the first time the postarrest reduction in oxygen (O2) delivery in relation to O2 uptake for brain and entire organism.

METHODS

In eight dogs we used our model of ventricular fibrillation (VF) cardiac arrest of 12.5 min, reperfusion with brief cardiopulmonary bypass, and controlled normotension, normoxemia, and mild hypocapnia to 24 h.

RESULTS

Between 4 and 24 h after cardiac arrest, cardiac output decreased by about 25% and the systemic arteriovenous O2 content difference doubled, while the calculated systemic O2 utilization coefficient (O2 UC) increased and the systemic venous PO2 decreased, both not to critical levels. The cerebral arteriovenous O2 content difference however, which was 5.6 +/- 1.7 ml/dl before arrest, increased between 1 and 18 h, to 10.8 +/- 3.2 ml/dl at 4 h. The cerebral O2 UC increased and the cerebral venous PO2 decreased, both to critical levels.

CONCLUSIONS

After prolonged cardiac arrest in dogs with previously fit hearts, the reduction of O2 transport to the brain is worse than its reduction to the whole organism. Monitoring these values might help in titrating life-support therapies.

[Transfusional strategy in neurosurgery: preoperative hemodilution by erythrocytapheresis]

Autologous blood transfusion techniques have been devised in order to decrease the risk of homologous transfusion-related complications. In neurosurgery, preoperative autologous blood collection is difficult because of the rather short time interval before surgery, as well as the risk of increasing cerebral oedema or intracranial hypertension. Therefore erythrocytapheresis has been performed the day before surgery as a preoperative haemodilution in 33 patients, using a discontinuous flow cell separator (PCS + Heamonetics). Patients with anaemia, unstable cardiovascular condition, infections, malignant tumor with a bad prognosis, or a poor peripheral venous status were not included. The mean volume of collected red cells was 526 +/- 176 mL, allowing a minimal colloid perfusion adjusted on this volume, with a simultaneous restitution of plasma and platelets. For a mean peroperative estimated blood loss of 1,040 +/- 52 mL, a homologous blood transfusion was avoided in 29 patients (88%). Four patients who underwent meningioma surgery received homologous red cells units in addition to their autologous blood. Two patients did not require any transfusion. Finally, 88% of autologous red cells units were readministered and 8 units were not retransfused. Preoperative erythrocytapheresis has proven to be a very simple and well tolerated technique. It can be considered for elective neurosurgery, when the time delay before surgery is short and when the blood loss is anticipated as to be moderate. It may also be associated with iterative autologous blood donation programme or the peroperative use of a cell saver.

Cerebral resuscitation after cardiac arrest: research initiatives and future directions

At present, fewer than 10% of cardiopulmonary resuscitation (CPR) attempts prehospital or in hospitals outside special care units result in survival without brain damage. Minimizing response times and optimizing CPR performance would improve results. A breakthrough, however, can be expected to occur only when cerebral resuscitation research has achieved consistent conscious survival after normothermic cardiac arrest (no flow) times of not only five minutes but up to ten minutes. Most cerebral neurons and cardiac myocytes tolerate normothermic ischemic anoxia of up to 20 minutes. Particularly vulnerable neurons die, in part, because of the complex secondary post-reflow derangements in vital organs (the postresuscitation syndrome) which can be mitigated. Brain-orientation of CPR led to the cardiopulmonary-cerebral resuscitation (CPCR) system of basic, advanced, and prolonged life support. In large animal models with cardiac arrest of 10 to 15 minutes, external CPR, life support of at least three days, and outcome evaluation, the numbers of conscious survivors (although not with normal brain histology) have been increased with more effective reperfusion by open-chest CPR or emergency cardiopulmonary bypass, an early hypertensive bout, early post-arrest calcium entry blocker therapy, or mild cerebral hypothermia (34 C) immediately following cardiac arrest. More than ten drug treatments evaluated have not reproducibly mitigated brain damage in such animal models. Controlled clinical trials of novel CPCR treatments reveal feasibility and side effects but, in the absence of a breakthrough effect, may not discriminate between a treatment's ability to mitigate brain damage in selected cases and the absence of any treatment effect. More intensified, coordinated, multicenter cerebral resuscitation research is justified.

Systematic development of cerebral resuscitation after cardiac arrest. Three promising treatments: cardiopulmonary bypass, hypertensive hemodilution, and mild hypothermia

Since 1970 we have investigated postischemic anoxic encephalopathy and potential treatments for cerebral resuscitation after cardiac arrest by cardiopulmonary-cerebral resuscitation (CPCR). The post-resuscitation syndrome has been studied at the levels of cell, organ, organism and community. Short-term and long-term models in rats, dogs, and monkeys have been developed, and an international multicenter randomized clinical trial mechanism was established. Clinical studies disproved the 5-min limit of reversible cardiac arrest and yielded other valuable data on treatments and prognostication. Thiopental loading or calcium entry blocker therapy (lidoflazine) gave no significant improvement in patients. Free radical scavengers are under investigation in the laboratory. We hypothesize that post-arrest perfusion failure and necrotizing cascades require etiology-specific combination treatments. Standard (control) therapy in a current dog model of cardiac arrest (no flow) of 12.5-20 min, reperfusion with cardiopulmonary bypass, and intensive care for 72-96 h has consistently resulted in survival with brain damage. After ventricular-fibrillation (VF) arrest of 17 min, moderate hypothermia (28-32 degrees C) inconsistently improved cerebral outcome. After VF arrest of 12.5 min, hypertension plus hemodilution normalized the local (multifocal) cerebral hypoperfusion post-arrest and, again, inconsistently improved cerebral outcome. Additional mild hypothermia (34-36 degrees C), however, consistently improved cerebral outcome, whether induced before or during and after arrest.

Multifocal cerebral blood flow by Xe-CT and global cerebral metabolism after prolonged cardiac arrest in dogs. Reperfusion with open-chest CPR or cardiopulmonary bypass

Using the stable xenon-enhanced computed tomography (Xe-CT) method in dogs, we studied local, regional and global cerebral blood flow (LCBF, rCBF and gCBF) in two sham experiments and nine cardiac arrest experiments. Within the same experiments without arrest, gCBF and rCBF values were reproducible and stable. LCBF values varied over time. In group I (n = 4), ventricular fibrillation cardiac arrest (no blood flow) of 10 min was reversed by open-chest cardiopulmonary resuscitation (CPR). In group II (n = 5), ventricular fibrillation cardiac arrest of 12.5 min was reversed by brief closed-chest cardiopulmonary bypass. This was followed by controlled ventilation, normotension, normoxia, normocarbia and normothermia to 4 h (n = 7) or 20 h (n = 2) postarrest. The postarrest CBF patterns were similar in both groups. Open-chest CPR during ventricular fibrillation generated near-baseline gCBF and lower LCBF ranges. During postarrest spontaneous circulation, transient diffuse hyperemia was without low-flow regions, longer in brain stem and basal ganglia than in neocortex. During delayed hypoperfusion at 1-4 h postarrest (n = 9), mean gCBF was 44-60% baseline, rCBF in primarily gray matter regions was 15-49 ml/100 cm3 per min and LCBF voxels with trickle-flow and low-flow values, in percent of CT cut area, were increased over baseline. Global CMRO2 (n = 3 of group II) recovered to near baseline values between 1 and 4 h postarrest, while gCBF and O2 delivery were about 50% baseline (mismatching of O2 uptake and O2 delivery).