| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Ann Thorac Surg 2004;78:188-196
© 2004 The Society of Thoracic Surgeons
a Duke University Medical Center, Division of Thoracic Surgery, Durham, North Carolina, USA, Papworth Hospital, Cambridge, United Kingdom
b Division of Pediatric Cardiac Surgery, Doernbecher Children's Hospital, Oregon Health and Science University, Portland, Oregon, USA
Accepted for publication November 7, 2003.
* Address reprint requests to Dr Ungerleider, Pediatric Cardiac Surgery, Doernbecher Children's Hospital, Oregon Health & Science University, 3181 SW Sam Jackson Park Rd, DC8S, Portland, OR, USA 97201-3098
e-mail: ungerlei{at}ohsu.edu
Presented at the Thirty-ninth Annual Meeting of The Society of Thoracic Surgeons, San Diego, CA, Jan 31–Feb 2, 2003.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterial and methodsResultsCommentDiscussionReferences |
|---|
METHODS: Eighteen neonatal piglets were randomly assigned to three groups. The control group was ventilated; the cardiopulmonary bypass group underwent 60 minutes of normothermic cardiopulmonary bypass, and the DHCA group underwent cardiopulmonary bypass and 60 minutes of DHCA (16° to 18°C) followed by rewarming. Hemodynamic and cerebral perfusion data were measured at an arterial partial pressure of oxygen (PaO2) of 150 to 250 mm Hg, and then at moderate hypoxemia (PaO2, 50 to 60 mm Hg) and severe hypoxemia (PaO2, 25 to 35 mm Hg).
RESULTS: Cerebral oxygen delivery decreased by 44% from PaO2 150 to 250 mm Hg to severe hypoxemia (p < 0.001). Cerebral oxygen extraction increased from moderate hypoxemia to severe hypoxemia in the control (57.9% ± 3.7% to 71.8% ± 3.8%; p = 0.002) and cardiopulmonary bypass groups (61.2% ± 2.6% to 70.6% ± 1.2%; p = 0.035); however, the cerebral oxygen extraction of the DHCA group did not increase under these conditions (82.8% ± 1.8% to 77.9% ± 4.3%; p = 0.32). The cerebral metabolic rate of oxygen consumption of the DHCA group decreased from PaO2 150 to 250 mm Hg to severe hypoxemia (1.86 ± 0.20 to 0.99 ± 0.24 mL O2 · 100 g–1 · min–1; p = 0.02), whereas the cerebral metabolic rate of oxygen consumption did not change under these conditions in the control and cardiopulmonary bypass groups.
CONCLUSIONS: Under hypoxemic conditions cerebral metabolic rate of oxygen consumption deteriorates after DHCA. Infants exposed to DHCA may be at greater risk of brain injury when postoperative hypoxemia is present. Because maximal cerebral oxygen extraction after DHCA occurs at moderate hypoxemia, techniques that increase cerebral oxygen delivery may reduce the risk of hypoxic brain injury.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsCommentDiscussionReferences |
|---|
Deep hypothermic circulatory arrest (DHCA) is commonly used during the surgical treatment of complex congenital heart disease, including stage I of the Norwood procedure. Prolonged DHCA increases the risk of anoxic brain injury as demonstrated by histology [4]. Shorter intervals of DHCA are known to be followed by a period of decreased cerebral perfusion and reduced cerebral metabolic rate of oxygen consumption (CMRO2) [5, 6]. It is believed that post-DHCA reductions in CMRO2 are a sign of neuronal injury [7, 8]. It is unknown whether DHCA interferes with the homeostatic mechanisms that maintain adequate cerebral oxygen delivery (CDO2). If the cerebral circulation were unable to regulate oxygen delivery after DHCA, the brain would be at increased risk of hypoxic injury during episodes of hypoxemia in the perioperative period.
Pediatric patients who have undergone DHCA have a higher risk of neurologic deficits and delayed motor development [9, 10]. An inverse relationship has been demonstrated between intelligence quotient and the duration of the DHCA period [11]. It is possible these neurologic deficits result from a combination of brain injury during the DHCA interval and post-DHCA brain hypoxia resulting from inadequate CDO2. This study is designed to investigate the impact of hypoxemia on CDO2 and cerebral metabolism after exposure to DHCA in a neonatal pig model.
|
Material and methods |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsCommentDiscussionReferences |
|---|
Sedation was induced with intramuscular ketamine (50 mg/kg) and acepromazine (15 mg/kg), and then paralysis was obtained with intravenous pancuronium (300 mg/kg). Anesthesia was obtained with a bolus of intravenous fentanyl (100 mg/kg) and maintained with an intravenous fentanyl citrate infusion (25 mg · kg–1 · h–1). The piglets were then given intravenous solumedrol (30 mg/kg). After orotracheal intubation, ventilation was provided by an infant pressure-cycled ventilator (model IV-100B, Sechrist, Industries, Anaheim, CA). A rectal temperature probe was inserted. A femoral artery catheter was used for blood pressure and blood gas monitoring. The heart was exposed through a median sternotomy. An 8-mm pulmonary artery perivascular flow probe (Transonic Systems, Ithaca, NY) was used to measure cardiac output before initiation of CPB in all animals and also provided all cardiac output measurements for the control group. A micromanometer (Millar Instruments Inc, Houston, TX) was introduced into the right atrium to monitor right atrial pressure. A silicone elastomer catheter was inserted into the left atrium for microsphere infusion. Each animal was anticoagulated with 500 IU/kg of intravenous heparin. The scalp over the vertex of the animals' skull was raised, and access to the superior sagittal sinus was obtained through two 2-mm burr holes. A micromanometer (Millar Instruments Inc) was introduced through the posterior burr hole to monitor sagittal sinus pressure. The anterior burr hole was used for sagittal sinus venous blood gas sampling.
The reference sample, radiolabeled microsphere technique was used to measure cerebral blood flow (CBF) [12]. Microspheres, 15.5 + 0.1 µm (Du Pont de Nemours & Co, Wilmington, DE), were suspended in 10% dextran and 0.01% Tween 80 at a concentration of 106 microspheres per milliliter. Five different radioisotopes (gadolinium 153, tin 113, niobium 95, ruthenium 103, and scandium 46) were used in a random sequence. At baseline for all animals, and at all data collection points in the control group, the radioactive microspheres were injected through the left atrial catheter. For later data collection points in the CPB and DHCA groups, the radioactive microspheres were injected using the side port of the aortic cannula. The technique of microsphere injection used by our laboratory has been previously reported [13]. At the end of the experiment the brain was removed, divided into right and left cerebral hemispheres, and weighed fresh. The tissue was then dissolved in 2 mol/L potassium hydroxide solution. A gamma counter (Auto-gamma 5530, Packard Instruments Co, Meridian, CT) was used to measure the quantity of each type of microsphere present in the brain tissue and blood samples. Regional blood flow was calculated using the ratio of counts of the tissue per counts of the blood specimen, and the withdrawal rate of the reference blood sample. The fresh tissue weight was then used to express this value in milliliters of blood flow per 100 g of tissue per minute. Cerebral blood flow was defined as the average of the right and left cerebral blood flow measurements.
Cardiopulmonary bypass
In the CPB and DHCA groups, a Cobe VP-CML membrane oxygenator (Cobe Inc, Arvada, CO) without an arterial filter was used for CPB. The oxygenator was primed with heparinized, fresh, donor pig blood. Ringer's lactate solution, 400 µg of fentanyl citrate, and bicarbonate were added to the prime to obtain a hematocrit of 25% and a pH of 7.4 at 37°C. The temperature of the perfusate was controlled with the integral heat exchanger of the oxygenator reservoir and a water bath system (Bio-Cal 370, Biomedicus, Minneapolis, MN). After instrumentation and baseline data acquisition, the aortic root and right atrial appendage were cannulated. Normothermic, nonpulsatile CPB using a Stockert-Shiley roller-pump was established at a rate of 150 mL · kg–1 · min–1 and mean systemic arterial pressure of 50 to 60 mm Hg. In the CPB and DHCA animals, after initiation of CPB, the cardiac output was defined as the pump flow rate divided by the animal's weight. A gas blender (Sechrist Industries Inc) was used to control the sweep gas rate and fraction of inspired oxygen.
In the DHCA group, the animals were perfusion cooled to 18°C for more than 20 minutes. Arterial blood gases were managed using 
-stat strategy [14]. The piglets were exsanguinated into the oxygenator, and DHCA was initiated. After 60 minutes of DHCA, CPB was reestablished at 150 mL · kg–1 · min–1. The piglets were rewarmed to 36°C during 60 minutes to ensure all animals were fully rewarmed and the temperature was stabilized. Any metabolic acidosis was corrected with the addition of sodium bicarbonate solution to the reservoir.
Experimental conditions
Blood gas analyses were performed using a Gem-Stat Blood Gas/Electrolyte Monitor (Mallinckrodt Sensor Systems Inc, Ann Arbor, MI), and co-oximeter (Instrumentation Laboratory Corp, Lexington, MA). At baseline the PaO2 was 150 to 250 mm Hg. In the control group, decreasing the fraction of inspired oxygen of the ventilator produced moderate hypoxia (PaO2 of 50 to 60 mm Hg), and later severe hypoxia (PaO2 of 25 to 35 mm Hg). For the CPB and DHCA groups, adjustment of the percent oxygen delivered to the oxygenator produced a PaO2 of 150 to 250 mm Hg, followed sequentially by moderate hypoxia and severe hypoxia. The arterial partial pressure of carbon dioxide was maintained between 35 and 45 mm Hg, and the hematocrit between 23% and 28% throughout the study period. The mean arterial pressure was allowed to drift.
At each data collection point, hemodynamic recordings and blood gas analyses were performed immediately before microsphere injection. Baseline CBF measurements were made at a PaO2 between 150 and 250 mm Hg. The PaO2 150 to 250 mm Hg data point of the CPB group was made after 60 minutes of normothermic CPB. The PaO2 150 to 250 mm Hg data point of the DHCA group was obtained at normothermia, 60 minutes after the DHCA interval. Total CPB time for each animal in the DHCA group, before data collection at a PaO2 of 150 to 250 mm Hg, was 80 minutes. The percent oxygen was then reduced to obtain moderate hypoxemia and severe hypoxemia. At all data points, the animals were allowed to stabilize for 10 minutes before data acquisition. Cerebral perfusion pressure, in millimeters of mercury, was defined as the difference between the mean arterial pressure and the mean sagittal sinus pressure. The cerebral vascular resistance (CVR) was calculated as the ratio of cerebral perfusion pressure to CBF (in mm Hg · min · 100 g · mL–1). The oxygen content, in milliliters of oxygen per milliliter of blood, of arterial and venous blood samples was calculated using the following equation:
 |
 |
Statistical analysis
All data are expressed as a mean ± standard error of the mean. Significant differences in all data of this experiment were analyzed with a two-way repeated measures analysis of variance, using the Holm-Sidak method, by applying the Sigma-Stat program, version 3.0 (SPSS Inc, Chicago, IL). Multiple unpaired Student's t tests were used to define statistically significant differences among the variables of the control group against the variables of the CPB group and DHCA group. A paired Student's t test was used to detect differences within each group among a PaO2 150 to 250 mm Hg, moderate hypoxemia, and severe hypoxemia.
|
Results |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsCommentDiscussionReferences |
|---|
|
|
|
|
|
|
|
|
|
Comment |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsCommentDiscussionReferences |
|---|
This study did corroborate previous reports of impaired CBF and cerebral metabolism after DHCA at a PaO2 of 150 to 250 mm Hg [7, 8, 13, 15]. After exposure to DHCA, during PaO2 150 to 250 mm Hg, both the CBF and CMRO2 demonstrated significant decreases, when compared with baseline. Additionally, the DHCA group was not able to increase CBF in response to decreasing arterial oxygen content during moderate hypoxemia and severe hypoxemia. One cause for the decreased CBF seen in the DHCA group could be reduced mean arterial and cerebral perfusion pressures. It is possible that the differences in cerebral perfusion pressure between the DHCA group and the control and CPB groups explains why CBF was low after exposure to DHCA. However, decreased cerebral perfusion pressure does not provide the entire explanation for the reduction in CBF after DHCA, because during severe hypoxemia when the differences in CBF are most pronounced, the mean arterial and cerebral perfusion pressures of the control and DHCA groups are similar. Increasing the cerebral perfusion pressure after exposure to DHCA may be one method to improve CBF, but this intervention was not investigated with this protocol.
The diminished CBF seen after DHCA could also be explained by altered vascular responsiveness. The control and CPB groups were able to progressively decrease CVR in reaction to decreases in arterial oxygen content. The DHCA group was not able to reduce CVR under the same conditions. Deep hypothermic circulatory arrest appears to result in an altered cerebrovascular response to hypoxemia, which limits the ability of the brain to increase CBF during episodes of hypoxemic stress. Deep hypothermic circulatory arrest has been demonstrated to cause endothelial dysfunction in cerebral microvessels [16]. One effect of this endothelial dysfunction may be an inability to respond to alterations in arterial oxygen content, resulting in the findings presented in this study.
However, the results of this study are limited by some of the experimental conditions. First, all animals progressed from a PaO2 of 150 to 250 mm Hg to moderate and later severe hypoxemia. This protocol was chosen because of the concern for hypoxic brain injury, which can occur at severe hypoxemia. Although the sequential reduction in PaO2 may have resulted in some the changes in CEO2 and CMRO2 detailed in this study, the effect of early hypoxic brain injury would likely have a much larger impact if the PaO2 of the data collection points were randomized, and some animals exposed to severe hypoxemia during the early data collection points. Second, the effects of hypothermia are not clearly defined by this protocol. An additional group undergoing hypothermic CPB might have helped separate the effects of circulatory arrest from the effects of hypothermia. It has been demonstrated that the combination of hypothermia and CPB does result in reductions in CBF and CMRO2 during periods of hypothermia [8]. The reductions in CBF and CMRO2 in these patients are not as pronounced as the decreases seen in patients who have undergone DHCA [6, 8]. Additionally, hypothermia does not result in altered CMRO2 or CBF once the patients are restored to normothermia [6]. Third, because of the cooling interval, the DHCA group was exposed to CPB for a longer time than the CPB group. This extended exposure could have resulted in some of the differences between the CPB and DHCA groups. However, no studies have reported the effect of additional exposure to CPB on either CBF or CMRO2.
Another limitation of this study is the method of cerebral perfusion. Although hemodynamic and arterial blood gas values were similar among the three groups, the CBF measurements for both the DHCA and CPB animals were made under conditions of nonpulsatile CPB. The use of nonpulsatile perfusion may be the source of some of the differences seen between these two groups and the control group. Pulsatile CPB has been demonstrated to result in higher CBF than nonpulsatile bypass after exposure to DHCA [17]. However, the use of nonpulsatile flow does not completely explain the inability of the DHCA group to increase CBF in response to hypoxemia because the CPB group was able to increase CBF in response to decreased arterial oxygen content.
In the DHCA group, reductions in CBF were combined with reduced arterial oxygen content, which resulted in severely decreased CDO2. When exposed to moderate hypoxemia, the DHCA animals relied completely on increasing CEO2, in the face of falling CDO2, to maintain the oxygen supply to the brain. With further reductions in arterial oxygen content, seen between moderate and severe hypoxemia, the CEO2 could not be increased. It appeared the maximum efficiency of CEO2 after DHCA was approximately 80%. At a PaO2 of 50 to 60 mm Hg, this degree of CEO2 was able to maintain the CMRO2 at a level comparable to that seen at a PaO2 of 150 to 250 mm Hg. However, once CEO2 is maximized, further decreases in CDO2 result in reductions of oxygen available to the cerebrum, which can be calculated as product of the CDO2 and the CEO2 for each animal. After exposure to DHCA, under conditions of severe hypoxemia decreases in the oxygen available to the cerebrum are reflected by parallel decreases in CMRO2 (Fig 6). The reduction in CMRO2 between moderate and severe hypoxemia may indicate that anoxic brain injury is occurring as a result of inadequate CDO2. Although this study did not investigate histologic or gross evidence of cerebral injury, the findings are disconcerting because infants exposed to postoperative hypoxemia, who have undergone DHCA, may be exposed to an increased risk of perioperative brain injury. Until the homeostatic responses of the brain to hypoxemia return to normal, an interval that may be as long as 8 hours [15], this risk will be present.
|
-stat cooling. Improved early brain recovery may correlate with appropriate responses to hypoxemia in the early perioperative period. The effect of hemodilution on autoregulation of CBF after DHCA is another area to explore. Although the degree of hemodilution in this experiment is mild, using a higher hematocrit appears to result in less cerebral injury [20] and may also help to prevent the loss of responsiveness to hypoxemia seen after DHCA. The results of this experiment imply that the brains of infants may be more vulnerable to hypoxemic brain injury after exposure to DHCA. For those infants who have undergone DHCA, the avoidance of postoperative management strategies that result in hypoxemia, such as using hypoxic ventilation to balance the pulmonary and systemic circulations in single ventricle physiology, should reduce the threat of brain injury. Additional options to prevent brain injury after DHCA might include techniques that have been shown to improve CDO2 in the perioperative period [21].
|
Discussion |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsCommentDiscussionReferences |
|---|
I rise to agree with you in some areas and disagree slightly in others. We have similar results to yours, as you know, in puppies rather than pigs, done in the era when we could afford to use puppies, and one of our most consistent observations in different experiments was that in the post-bypass interval, after a period of deep hypothermic circulatory arrest, there is an elevation in cerebral vascular resistance, so that blood flow is reduced, and this persists from anywhere from about 2 to 8 hours post-bypass. In our model we used somewhat lower temperatures, but our cerebral metabolic rates were normal postoperatively in the deep hypothermic circulatory arrest group, but still there was this cerebral vasospasm or reduction in oxygen delivery.
I have two comments and perhaps you would like to respond to them. One is that I think your data lend additional support to the idea that deep hypothermic circulatory arrest at 18 degrees or so for an hour is not a benign technique, that there are changes that occur in cerebral metabolic rate postoperatively and basically the brain doesn't return to normal. I think, based on our studies in the dog, that if you take it about 5 degrees colder you may get away with an hour. Of course we have clinical studies in adults that suggest that longer than half an hour even at deep temperatures is not entirely safe. But maybe you would like to comment on what you feel about the safety of an hour's hypothermic circulatory arrest at 18 degrees.
One final point is that you mentioned briefly in your discussion the issue of hematocrit or oxygen-carrying capacity, and I think that's a very important one, because for the child who has just had a Norwood repair, you don't have a lot of options sometimes in terms of increasing cerebral oxygen delivery. Cardiac ourput may be increased modestly with catecholamines, but arterial oxygen tension is limited. Increasing oxygen carrying capacity by increasing hematocrit is a very important option. What was the hematocrit in these animals postoperatively, and how high do you think you could push the hematocrit in order to increase oxygen-carrying capacity if there is cerebral vasospasm?
DR SCHULTZ: The hematocrit in this study averaged 25 in all animals in all groups. Currently we're taking a look at a repeat of this study under some different conditions where we have increased the hematocrit to 30. There are a few different ways that we could possibly increase the cerebral delivery of oxygen without simply increasing hematocrit, one of which is using perhaps a mechanical assist device, which could supplement the cardiac output without having to rely upon inotropes.
Regarding your question of how low we decrease the temperature during our circulatory arrest interval, this study is built upon a number of studies that have been done at Duke in the past and that is why we have used this consistent protocol, because it has given us a consistent understanding, or at least an attempt at understanding what the processes are. It is possible if we were to use a lower temperature that we may be able to see not such an adverse effect upon the cerebral metabolic oxygen consumption. I will concur with you, however, that we did note a substantial increase in our DHCA group in cerebral vascular resistance as compared to the other two groups, and not only that, but there was a very dramatic increase in cerebral vascular resistance at both moderate hypoxia and severe hypoxia, which was not seen in the other two groups.
DR PEDRO J. DEL NIDO (Boston, MA): I guess that leads me to the question of what exactly is the defect, because you're implying that if you could only deliver more oxygen, either by more red cells or more flow, then perhaps you would not have injury. If you have increased cerebral vascular resistance, how are you going to overcome the high resistance simply by those maneuvers?
DR SCHULTZ: A number of studies out there have demonstrated that it appears to be a defect in cerebral vascular resistance in the ability to modulate cerebral vascular tone. I guess we overspoke ourselves by saying simply by increasing cerebral delivery of oxygen we will get around this problem. Perhaps by increasing cardiac output, we can therefore simply by doing that increase the cerebral delivery of oxygen. Specifically by improving systemic circulation, we may be able to increase our mixed venous saturation, which may allow us to also increase the amount of arterial oxygenation that is going to the brain without needing to necessarily increase the cerebral blood flow. In addition, it's possible, I guess, with a fixed cerebral vascular resistance, if we were to increase the blood flow to the brain specifically, we could do that, although it would be at the detriment of increasing the cerebral perfusion pressure, which if it got too high could in and of itself cause damage.
DR DEL NIDO: I was thinking more in terms of a pharmacologic intervention. Nitroglycerin, for example, is a very potent cerebral vasodilator.
DR JACOBS: Carbon dioxide.
DR DEL NIDO: Or carbon dioxide.
DR SCHULTZ: Right.
DR MARSHALL L. JACOBS (Philadelphia, PA): Finally, I would like to ask you, as noted, Dr. Griepp and others pointed out many years ago the decrease in cerebral blood flow in the early period after circulatory arrest, and Dr. Ungerleider, your mentor, and his group for years have demonstrated the decreased cerebral metabolic rate for oxygen during that same period. Now, you've shown that a certain markedly decreased level of oxygen delivery reaches a critical point of a disadvantageous mismatch. Is it your feeling and the feeling in Dr. Ungerleider's lab that the decreased cerebral metabolic rate of oxygen during this period after hypothermic circulatory arrest is a signal for potential injury or is it an indication of delayed resumption of normal metabolic activity during a period of decreased blood flow?
DR SCHULTZ: That has been an important question that has been raised time and time again. I think there have been a few studies out there that have correlated the deficiency and the severity of decreased CMRO2 with this histologic severity of neuronal injury. But it all needs to be taken in context. Let's assume for a moment that this amount of decrease in CMRO2 is directly correlated to amount of neurologic injury. The brain is very plastic in these young infants and it's possible that months or years down the road the damage that we're seeing then will not be clinically significant.
|
References |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsCommentDiscussionReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  C. Pizarro, K. A. Murdison, C. D. Derby, and W. Radtke Stage II Reconstruction After Hybrid Palliation for High-Risk Patients With a Single Ventricle Ann. Thorac. Surg., April 1, 2008; 85(4): 1382 - 1388. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. J. Hickey, X. You, V. Kaimaktchiev, and R. M. Ungerleider Hypoxemic reperfusion exacerbates the neurological injury sustained during neonatal deep hypothermic circulatory arrest: a model of cyanotic surgical repair Eur. J. Cardiothorac. Surg., May 1, 2007; 31(5): 906 - 914. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. Alsoufi, J. Bennetts, S. Verma, and C. A. Caldarone New Developments in the Treatment of Hypoplastic Left Heart Syndrome Pediatrics, January 1, 2007; 119(1): 109 - 117. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Ricci, P. Lombardi, A. Galindo, A. Vasquez, J. Zuccarelli, and E. Rosenkranz Distribution of cardiac output and oxygen delivery in an acute animal model of single-ventricle physiology J. Thorac. Cardiovasc. Surg., October 1, 2005; 130(4): 1062 - 1070. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |
—) and cardiopulmonary bypass (—
—) groups increased at hypoxemia (p < 0.04, and p = 0.005, respectively) and at severe hypoxemia (p < 0.001, and p = 0.009, respectively), as compared with PaO2 150 to 250 mm Hg. The cerebral blood flow of the deep hypothermic circulatory arrest (—
—) group was not able to increase during hypoxemia (p = 0.52) or at severe hypoxemia (p = 0.29) when compared with PaO2 150 to 250 mm Hg.
