Prevention of Brain Damage During Profound Hypothermia and Circulatory Arrest

Acquired ocular motor apraxia after aortic surgery

PURPOSE

To describe an unusual form of acquired ocular motor apraxia.

METHODS

Case reports with electronic eye movement recordings.

RESULTS

Three patients had surgery to repair aortic root or arch dissections or aneurysms. A few days after surgery, all had ophthalmoplegia. Neuro-ophthalmic examination found complete absence of horizontal and vertical volitional and reflex saccades in 1 patient and slow, hypometric saccades in 2 others. However, smooth pursuit, slow phases of optokinetic nystagmus, and the vestibulo-ocular response (VOR) were intact. Fast phases of the VOR were absent in 2 patients but were intact in the other. Video and electronic eye movement recordings documented the findings. Magnetic resonance imaging (MRI) in 1 patient showed small infarcts in a cerebellar hemisphere, pons, and cerebral hemispheres. The other patients' MRIs showed no significant lesions.

CONCLUSIONS

Acquired ocular motor apraxia with profoundly impaired volitional saccades after aortic surgery is a distinctive syndrome, but its pathophysiology is unclear. Studies of neurologic damage in animals and patients undergoing similar surgical procedures provide conflicting data. However, knowledge about the complex neural pathways generating saccades from animal and human studies, and detailed clinical observations, as in the patients described here, can help to determine the location of lesions. Based on the 3 cases reported here, we propose that this syndrome might be due to damage to excitatory burst and/or omnipause neurons in the brainstem or by damage to pathways from the frontal eye fields to the brainstem.

Comparing dual-stream and standard cardiopulmonary bypass in pigs

BACKGROUND

Dual-stream (DS) and standard cardiopulmonary bypass (CPB) were compared.

METHODS

A DS catheter inserted into the distal ascending aorta across the arch pumps blood through an upper lumen (maximum 2.25 L/min) directed by a bloodstreaming baffle toward the arch vessels. A separate lower lumen pumps blood (maximum 3.75 L/min) into the aorta caudad to the inflated baffle. The baffle is flat and horizontal along the catheter. When the baffle is collapsed the heart or both lumens may perfuse all organs. For 30 minutes 8 randomized CPB pigs had corporeal cooling to 32 degrees C and for 30 minutes had rewarming to 36 degrees C. Eight randomized DS pigs had 25 degrees C upper lumen cooling for 60 minutes. Lower lumen blood flow was streamed at 32 degrees C for 30 minutes, then rewarmed to 36 degrees C for 30 minutes.

RESULTS

The change in relative lower lumen to brain blood flow as determined by brain-counted microspheres (15 micron) injected into the ascending aorta was less for DS brains than controls during full flow (DS 63.4+/-129.5 versus CPB 2,585.4+/-250.8, p < 0.001), and when injected into the ejecting-heart left atrium just after weaning off only lower lumen blood flow (DS 250.8+/-297.3 versus CPB 1,159.1+/-782.3, p < 0.001). DS brain temperatures were lower at an equal pump-off core temperature of 36 degrees C+/-0.5 degrees C (DS 31.6 degrees C+/-3.2 degrees C versus CPB 36.5 degrees C+/-1.7 degrees C, p < 0.025). Jugular O2 saturations were not different.

CONCLUSIONS

DS-CPB prioritizes pump-filtered separate cold blood flow to the brain over a blood-streaming baffle to wash away potentially surgery related air and particulate matter arising from the heart or ascending aorta.

Higher hematocrit improves cerebral outcome after deep hypothermic circulatory arrest

BACKGROUND

Various degrees of hemodilution are currently in clinical use during deep hypothermic circulatory arrest to counteract deleterious rheologic effects linked with brain injury by previous reports.

MATERIAL AND METHODS

Seventeen piglets were randomly assigned to three groups. Group I piglets (n = 7) received colloid and crystalloid prime (hematocrit < 10%), group II piglets (n = 5) received blood and crystalloid prime (hematocrit 20%), group III piglets (n = 5) received blood prime (hematocrit 30%). All groups underwent 60 minutes of deep hypothermic circulatory arrest at 15 degrees C with continuous magnetic resonance spectroscopy and near-infrared spectroscopy Neurologic recovery was evaluated for 4 days (neurologic deficit score 0, normal, to 500, brain death; overall performance category 1, normal, to 5, brain death). Neurohistologic score (0, normal, to 5+, necrosis) was assessed after the animals were euthanized on day 4.

RESULTS

Group I had significant loss of phosphocreatine and intracellular acidosis during early cooling (phosphocreatine in group I, 86.3% +/- 26.8%; group II, 117.3% +/- 8.6%; group III, 110.9% +/- 2.68%; p = 0.0008; intracellular pH in group I, 6.95 +/- 0.18; group II, 7.28 +/- 0.04; group III, 7.49 +/- 0.04; p = 0.0048). Final recovery was the same for all groups. Cytochrome aa3 was more reduced in group I during deep hypothermic circulatory arrest than in either of the other groups (group I, -43.6 +/- 2.6; group II, -16.0 +/- 5.2; group III, 1.3 +/= 3.1; p < 0.0001). Neurologic deficit score was best preserved in group III (p < 0.05 group II vs group III) on the first postoperative day, although this difference diminished with time and all animals were neurologically normal after 4 days. Histologic assessment was worst among group I in neocortex area (group I, 1.33 +/- 0.3; group II, 0.22 +/- 0.1; group III, 0.40 +/- 0.2, p < 0.05, group I vs group II; p = 0.0287, group I vs group III).

CONCLUSION

Extreme hemodilution during cardiopulmonary bypass may cause inadequate oxygen delivery during early cooling. The higher hematocrit with a blood prime is associated with improved cerebral recovery after deep hypothermic circulatory arrest.

University of Wisconsin cerebroplegia in a piglet survival model of circulatory arrest

BACKGROUND

Previous acute studies in immature piglets at our institution have demonstrated improved recovery of cerebral blood flow, intracellular pH, and high-energy phosphates with the administration of multidose University of Wisconsin solution as cerebroplegia during a period of deep hypothermic circulatory arrest (HCA). In an effort to define further the clinical applicability of this technique, we have developed a survival model of swine cardiopulmonary bypass (CPB) and HCA.

METHODS

12 Yorkshire pigs (age 4 to 5 weeks) were placed on CPB via the right femoral artery and right atrium. Animals were cooled to a rectal temperature of 15 degrees C and submitted to 90 minutes of HCA. Group UW (n = 6) received a single infusion of 50 mL/kg of 4 degrees C University of Wisconsin solution delivered antegrade to the cerebral circulation. The control group (n = 6) received no intervention. Animals were reperfused, rewarmed to 35 degrees C, and weaned from CPB. Neurologic assessments using neurologic deficit scoring (0 = normal, 500 = brain death) and overall performance categories (1 = normal, 5 = brain death) were performed at 24-hour intervals for 5 days. On the 5th postoperative day all brains were perfusion-fixed and examined for histologic evidence of neuronal injury (0 = normal, 5 = severe injury).

RESULTS

All animals were extubated 18 to 20 hours postoperatively. There was no significant difference between the mean neurologic score of the two groups. The mean day 5 neurologic deficit score was 108 for the UW group and 68 for the control group (p > 0.05). The day 5 overall performance category was 2.8 for the UW group and 2.0 for the control group (p > 0.05). Three of the UW animals but none of the control animals experienced generalized seizures. Histologic examination revealed more severe damage in UW animals, primarily in the cerebral cortex. Injury was more widespread in UW animals, involving cerebellum and hippocampus. The mean histologic injury score was 3.8 for UW animals and 2.4 for the control group (p = 0.06).

CONCLUSIONS

A clinically relevant survival model of CPB with HCA in immature swine is feasible. Cold UW solution as single-dose cerebroplegia is not beneficial, and may be injurious to the immature swine brain subjected to CPB and HCA. Further studies are indicated to determine optimal composition and administration of cerebroplegic solutions.

A new solution for life without blood. Asanguineous low-flow perfusion of a whole-body perfusate during 3 hours of cardiac arrest and profound hypothermia

BACKGROUND

The benefits of hypothermia for preventing ischemic injury are well known, but its application in surgery to protect the whole body during procedures requiring circulatory arrest is currently limited to < 1 hour at 15 degrees C using 50% hemodilution. In a significant departure from previous methods, we have developed a technique of asanguineous blood substitution with low-flow perfusion and cardiac arrest at < 10 degrees C in a canine model. Our approach has been to design a hypothermic blood substitute that would protect the brain and visceral organs during several hours of bloodless perfusion. Two different solutions have been designed to fulfill separate requirements in the procedure.

METHODS AND RESULTS

With the use of extracorporeal cardiac bypass, 14 adult dogs were exsanguinated during cooling; 11 dogs were blood substituted using in combination the "purge" and "maintenance" solutions (group 1), and 3 dogs were perfused throughout with the "purge" solution alone as controls (group 2). After cardiac arrest, the solutions were continuously circulated for 3 1/2 hours by the extracorporeal pump (flow rate, 40 to 85 mL.kg-1.min-1; mean arterial blood pressure, 25 to 40 mm Hg). The temperature was maintained at < 10 degrees C (nadir, 6.6 +/- 0.1 degrees C) for 3 hours, and the hematocrit was kept at < 1% before controlled rewarming and autotransfusion. In the experimental group, the heart always started spontaneously in the temperature range of 11 degrees C to 27 degrees C, and 8 animals have survived long-term (current range, 14 to 110 weeks) without any detectable neurological deficit. In contrast, two control animals survived after extensive and aggressive cardiac resuscitation efforts; after surgery they exhibited transient motor and sensory deficits for approximately 1 week. Evaluation of biochemical and hematological parameters showed only a transient and inconsequential elevation in enzymes (eg, brain, liver, cardiac) in group 1 compared with the markedly greater elevations in group 2. For example, immediate postoperative values (mean +/- SEM) for lactate dehydrogenase were 114 +/- 10 for group 1 versus 490 +/- 210 for group 2 (P < .03); for SGOT, values were 93 +/- 18 for group 1 versus 734 +/- 540 for group 2 (P < .05). On day 1 for creatine kinase (CK), the group 1 value was 7841 +/- 2307 versus 71,550 +/- 2658 for group 2 (P = .03), and for CK-BB, the group 1 value was 108 +/- 22 versus 617 +/- 154 for group 2 (P = .03). Neurological evaluation using deficit scores (NDS) was based on a modification of the Glasgow Coma Scale score: 0, normal; 1, minimal abnormality; 2, weakness; 3, paralysis; 4, coma; and 5, death. At days 1 and 2 after surgery, NDS (mean +/- SEM) were 0 +/- 0 for the experimental group versus 1.5 +/- 0.5 for the control group. At days 3 and 7 after surgery, NDS were 0 +/- 0 for group 1 versus 1.0 +/- 1.0 for group 2.

CONCLUSIONS

The faster neurological recovery of dogs treated with the "intracellular-type" maintenance solution supports the biochemical data showing the benefits of this type of blood substitute for extending the safe limits of hypothermic cardiac arrest procedures to > 3 hours.

Hypothermia, metabolic stress, and NMDA-mediated excitotoxicity

Isolated embryonic retinas were metabolically stressed by inhibition of glycolysis either with iodoacetate (IOA) or by glucose withdrawal plus 10 mM 2-deoxy-D-glucose, and the effects of hypothermia were examined. Incubation at 30 versus 37 degrees C during 30 min of hypoglycemia with IOA completely reduced the rapid swelling-related GABA release [6 +/- 2 vs. 68 +/- 10 nmol/100 mg of protein (mean +/- SEM) for 30 and 37 degrees C, respectively]. Histology of the retina immediately following 30 min of metabolic stress at 30 degrees C appeared normal, whereas that at 37 degrees C showed a pattern of acute edema, characteristic of NMDA-mediated acute excitotoxicity. Coincubation with a competitive or noncompetitive NMDA antagonist, respectively, CGS-19755 (10 microM) or MK-801 (1 microM), during 30 min of hypoglycemia at 37 degrees C completely prevented tissue swelling, whereas extracellular GABA content remained at basal levels, indicating that the cytotoxic effects of IOA treatment for 30 min at 37 degrees C were NMDA receptor mediated. Longer periods of hypoglycemia at 37 degrees C produced acute toxicity that was only partially NMDA receptor mediated. Hypothermia delayed the onset of NMDA-mediated toxicity by 30-60 min. At 30 degrees C, the rate of loss of ATP was slowed during the first several minutes of hypoglycemia (82 and 58% of maximal tissue levels at 30 and 37 degrees C, respectively, at 5 min, but by 10 min, ATP levels were comparably reduced. After a transient exposure of retina to 50 microM NMDA in Mg(2+)-free medium, hypothermia significantly attenuated acute GABA release by 30%.(ABSTRACT TRUNCATED AT 250 WORDS)

Brain protection during circulatory arrest

Previous nuclear magnetic resonance studies in this laboratory have shown a beneficial biochemical effect of antegrade cerebroplegia (CP-A) during hypothermic circulatory arrest. This study compared CP-A with other methods of cerebral protection during hypothermic circulatory arrest to assess the clinical utility of this technique. Twenty-three sheep were divided into four groups: systemic hypothermia alone (SYST) and systemic hypothermia combined with external cranial cooling (EXTNL), retrograde cerebroplegia (CP-R), or CP-A. Cardiopulmonary bypass was started, and the sheep were cooled to 15 degrees C and subjected to 2 hours of circulatory arrest. Cardiopulmonary bypass was restarted, and the animals were rewarmed and weaned from cardiopulmonary bypass. Serial neurological examinations were performed and hourly scores assigned until the animals were extubated. Postanesthetic neurological scores improved in all groups throughout the 6-hour recovery period except the CP-R group. The improvement over time for these scores was similar for the EXTNL and CP-A groups and significantly better than for the SYST or CP-R groups (p = 0.004). The CP-A group had 5 of 7 animals with deficit-free survival despite the similarity in recovery of baseline brainstem function. We conclude that both antegrade infusion of cerebroplegia and external cranial cooling confer distinct cerebroprotective effects after a protracted period of hypothermic circulatory arrest when compared with the other methods studied.

Protection from cerebral ischemia by brain cooling without reduced lactate accumulation in dogs

Hypothermia protects tissue function in ischemia. This study determined if selective brain cooling inhibits cerebral cortical lactate accumulation and thus accounts for imporved neurologic outcome after complete cerebral ischemia in dogs. The brain was selectively cooled (hippocampal temperature 33 degrees C) by nasal lavage with water at 5 degrees C. Control dogs received nasal lavage with water at 39 degrees C. Mean +/- SEM rectal temperature in both groups was 39 +/- 1 degree C prior to ischemia. Selective brain cooling before and during 10 minutes of cardiac arrest was associated with significantly improved neurologic function and 100% survival, whereas normothermic cardiac arrest produced marked neurologic dysfunction and 100% mortality. Cerebral cortical lactate accumulation was measured in a complementary series of dogs exposed to the same two treatments but with the addition of six cerebral cortical brain biopsies taken before, during, and immediately after cardiac arrest. Brain and rectal temperatures of dogs in the brain biopsy protocol were similar to those of dogs in the recovery protocol. There was no difference detected in cerebral lactate accumulation during ischemia between brain-cooled and control dogs. Thus, reduction in cortical brain lactate during ischemia cannot account for the postischemic functional protection afforded by preischemic selective brain cooling.

Brain damage after deep hypothermia in dogs

Histopathological and succinic acid dehydrogenase (SAD) activity changes, and the electron spin resonance (ESR) were examined in the brains of dogs exposed to various hypothermic conditions. A simple hypothermic cardiac arrest of 5 min duration occurred in the group subjected to 28 degrees C, and normal SAD activity was noted in dogs exposed 1-hour extracorporeal circulation under normothermia. In cases of a hypothermia below 28 degrees C, a decrease in enzyme activity, marked histopathological changes and the disappearance of the ESR signal were observed. In dogs killed immediately after induction of 20 degrees C hypothermia, the histopathological changes were primarily ischemic in nature, while in dogs killed at 1-12 month intervals after the experiments, irreversible gliosis was dominant. Our findings suggest that, to avoid brain damage under conditions of deep hypothermic circulatory arrest, the critical range is between 26-28 degrees C for rectal, and approximately 28-30 degrees C for brain temperature.

Aortic arch aneurysm. Resection and replacement: protection of the nervous system

A patient with an aortic arch aneurysm, who was submitted to resection and replacement by a prosthetic graft during a 36-minute period of cerebral arterial occlusion at 20° C, and who subsequently died of ischaemic brain damage, is reported.

Current methods for protection of the central nervous system from ischaemic damage during operations of this type are reviewed. It is considered that total body perfusion, including all arteries which supply the circle of Willis, with hypothermia as low as 20° C, will provide the safest background to a successful operation.

Profound hypothermia with prolonged circulatory arrest. An experimental study

Images