HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Add to Personal Folders Download to citation manager Author home page(s): Richard P. Cochran Bruce D. Spiess Permission Requests Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Cochran, R. P. Articles by Spiess, B. D. Search for Related Content PubMed PubMed Citation Articles by Cochran, R. P. Articles by Spiess, B. D.

Ann Thorac Surg 1997;63:1326-1332
© 1997 The Society of Thoracic Surgeons

---

Original Article: Cardiovascular

Perfluorocarbon Emulsion in the Cardiopulmonary Bypass Prime Reduces Neurologic Injury

Division of Cardiothoracic Surgery, Department of Surgery, and Division of Cardiac Anesthesiology, Department of Anesthesiology, University of Washington, Seattle, Washington

Accepted for publication November 27, 1996.

	Abstract

Top Footnotes Abstract Introduction Material and Methods Surgical Preparation Cardiopulmonary Bypass Protocol Methods of Measurement Results Cerebral Infarction Cerebral Blood Flow Electrical Activity of the... Comment Acknowledgments References

 
Background. Perfluorocarbon emulsion has proved beneficial in the prevention and amelioration of experimental air embolism. We examined whether the addition of perfluorocarbon to the prime solution could lead to a reduction in the incidence and severity of neurologic injury after the formation of a massive air embolism during cardiopulmonary bypass.

Methods. Fourteen pigs underwent bypass in which either a crystalloid prime solution or a perfluorocarbon prime solution (10 mL/kg) was used. Ten minutes into bypass a bolus (5 mL/kg) of air or saline (control) was delivered via the carotid artery. The resulting cerebral infarcts were graded on the basis of the findings in triphenyltetrazolium chloride–stained cerebral sections. Colored microspheres were used to measure cerebral blood flow. Bitemporal electroencephalography was used to evaluate cerebral function.

Results. Cerebral infarction was not found in the perfluorocarbon-air group (0 of 5 animals), as compared with its occurrence in 3 of the 5 animals in the crystalloid-air group. Cerebral blood flow was also maintained or increased in the perfluorocarbon-air group (p < 0.05), and the electroencephalogram total power showed less of a decrease and recovered more completely (p < 0.05) than it did in the crystalloid-air group.

Conclusions. The addition of perfluorocarbon emulsion to the cardiopulmonary bypass prime solution leads to a reduction in the incidence and severity of neurologic injury after the formation of a massive air embolism during bypass.

	Introduction

Top Footnotes Abstract Introduction Material and Methods Surgical Preparation Cardiopulmonary Bypass Protocol Methods of Measurement Results Cerebral Infarction Cerebral Blood Flow Electrical Activity of the... Comment Acknowledgments References

 
Cardiopulmonary bypass (CPB) is used in more than 300,000 cardiothoracic operations in the United States each year [1]. One of the most potentially devastating postoperative complications of CPB is neurologic dysfunction. Estimates of the incidence of persistent neurologic dysfunction after CPB range from the 1% to 9% incidence of stroke (defined as a major motor or sensory deficit), to the 3% to 4% incidence of altered mental status, to the 24% to 40% incidence of neuropsychometric deficits [2–4]. Extrapolating from these data, the total number of patients affected may approach 120,000 per year.

Although the etiology or mechanism of neurologic dysfunction after CPB has not been proved, several potential mechanisms have been identified. These include the formation of macroscopic or microscopic embolism (air, atherosclerotic, or thrombotic), alterations in cerebral blood flow, and cellular injury occurring secondary to an abnormal circulatory environment [2–4].

Macroscopic air embolism, one of the causes of brain injury, is a rare and much feared complication of cardiac operations, and it has many potential sources. It may result from oxygenator failure, the introduction of air through "vent" sites or suction cannulas, suboptimal cannulation techniques, inattention to the reservoir level, or retained air, particularly in intracardiac procedures [5]. Over the past decade the incidence of neurologic dysfunction may have decreased as the result of improvements in technology, refined surgical techniques, and better patient management during CPB. However, despite these improvements, the incidence of neurologic complications remains significant and it is therefore essential to further reduce the incidence of these complications because of the devastation resulting from any neurologic loss.

Microscopic air embolism is less dramatic in the manner of its occurrence and its severity than macroscopic air embolism, but it may account for many of the imperfect outcomes seen after "successful" cardiac operations. As techniques of neuropsychometric testing have been used in patients who have undergone cardiac operations, it has been found that short-term function is altered in up to 80% of patients and long-term function in 24% to 40% of patients [3, 4]. Air embolism, whether macroscopic or microscopic, is therefore of utmost concern to the cardiac surgery community. The addition of any technical or pharmacologic method to the armamentarium of cardiac surgery that would reduce the effect of air emboli would therefore be of great benefit.

One possible technique to reduce the incidence and severity of a neurologic injury resulting from the formation of a massive air embolism during CPB is the addition of a perfluorocarbon (PFC) emulsion to the CPB prime solution. Perfluorocarbons are biologically and chemically inert and in an emulsion dissolve and transport gases such as oxygen, carbon dioxide, and nitrogen. Perfluorocarbon emulsions have been shown to absorb venous and arterial air emboli [6] and to reduce the effect of air emboli on the brain [7] and coronary beds [8] in spontaneously perfused animal models. However, the addition of PFC emulsions to the CPB prime solution to protect cerebral tissue from the effects of air embolism and the potential neurologic sequelae needs to be studied.

This study was based on the hypothesis that the addition of a PFC emulsion to a CPB prime solution would lead to a decrease in the effect of a massive cerebral air embolism. Using a pig model, the end-organ effects of cerebral air embolism were evaluated from the standpoint of three variables: the severity of cerebral infarction, cerebral blood flow, and the electrical activity of the brain, as measured by electroencephalography (EEG). Pigs received either a CPB prime solution with a PFC emulsion added or a standard crystalloid prime solution. End points of the study were (1) a reduction in the incidence of or an absence of cerebral infarction after an air "insult," (2) an increase in cerebral blood flow during and after air insult, and (3) an increase in the EEG power during and after an air insult.

	Material and Methods

Top Footnotes Abstract Introduction Material and Methods Surgical Preparation Cardiopulmonary Bypass Protocol Methods of Measurement Results Cerebral Infarction Cerebral Blood Flow Electrical Activity of the... Comment Acknowledgments References

 
Experiments were performed in accordance with the "Principles of Laboratory Animal Care" formulated by the National Society for Medical Research and the "Guide for the Care and Use of Laboratory Animals" prepared by the Institute of Laboratory Animal Resources, National Academy of Sciences (NIH publication 85-23, 1985).

	Surgical Preparation

Top Footnotes Abstract Introduction Material and Methods Surgical Preparation Cardiopulmonary Bypass Protocol Methods of Measurement Results Cerebral Infarction Cerebral Blood Flow Electrical Activity of the... Comment Acknowledgments References

 
Fourteen domestic pigs (age, 2 to 4 months; weight, 17 to 26 kg) were anesthetized with an intramuscular injection of ketamine (22 mg/kg), acepromazine (1.1 mg/kg), and atropine (0.05 mg/kg). After intubation, anesthesia was maintained with halothane (1% to 2%) in 100% oxygen. Respiratory rate and tidal volume of a volume-constant respirator (model 607; Harvard Apparatus, South Natick, MA) were adjusted to provide a physiologic acid-base balance. Four limb leads were placed to monitor the electrocardiogram, and the electrocardiogram and heart rate were monitored continuously. Five silver/silver chloride EEG electrodes (Sentry Medical Products, Irvine, CA) were placed on the animal's skull after the skin was shaved and prepared with skin degreaser. Electrodes were placed in a bitemporal array with a reference electrode in the midline just above the nasion. To acquire right and left temporal recordings, individual negative electrodes were placed on the right and left frontal lobes just above the orbits and the positive electrodes were placed behind the ear over the mastoid processes.

Both femoral arteries and the right femoral vein were isolated surgically and catheterized. The left femoral arterial catheter was used for blood pressure monitoring and the withdrawal of blood samples. The right femoral arterial catheter was reserved for the withdrawal of reference samples for microsphere measurement. The femoral venous catheter was used for drug administration, volume replacement, and the withdrawal of blood samples. Arterial blood samples were withdrawn periodically, and the partial pressures of oxygen and carbon dioxide and the pH were determined by a blood gas analyzer (model 170; Corning, Corning, NY). The ventilatory rate was adjusted to maintain a partial pressure of carbon dioxide of 35 ± 5 mm Hg. In addition, sodium bicarbonate was given as an intravenous bolus when necessary to correct for metabolic acidosis during and after CPB. To prevent a species-specific complement reaction to the PFC, dexamethasone (4 mg/kg) was administered intravenously before sternotomy.

A median sternotomy was performed after the femoral lines were placed. In preparation for CPB, pursestring sutures were placed in the right atrium and the ascending aorta. Next, the right common carotid artery was isolated for subsequent cannulation. Pigs were systemically anticoagulated by an intravenous infusion of heparin sulfate (3 mg/kg). After anticoagulation a 20-gauge arterial catheter was placed into the right common carotid artery and advanced into the internal carotid artery. This catheter was used for infusion of the insult substance (air or saline solution), as described in the section on protocol.

A 28F venous cannula was inserted into the right atrium and a 5-mm arterial cannula into the aorta, then both were connected to the CPB circuit. Before the initiation of CPB, a polyethylene catheter was placed into the left atrium for the administration of colored microspheres at specific time points before and after CPB. During CPB, microspheres were injected through the aortic cannula, because adequate mixing would not occur during CPB if they were injected through the left atrial infusion site.

	Cardiopulmonary Bypass

Top Footnotes Abstract Introduction Material and Methods Surgical Preparation Cardiopulmonary Bypass Protocol Methods of Measurement Results Cerebral Infarction Cerebral Blood Flow Electrical Activity of the... Comment Acknowledgments References

 
The CPB circuit consisted of a hollow-fiber membrane oxygenator (Medtronic Cardiopulmonary, Anaheim, CA), a hard-shell venous reservoir (Sarns 3M, Ann Arbor, MI), a roller pump (Sorin Biomedical, Irvine, CA), and polyvinylchloride tubing. An arterial filter was not used so that a "worst-case scenario" might eventuate. Normothermic CPB (36° to 37°C) was maintained at a flow rate of 80 mL•kg^-1•min^-1, with a fraction of inspired oxygen of 100%. The flow rate is consistent with that used in other studies of pigs on CPB at normothermia [9, 10].

The standard CPB prime solution used in this investigation was crystalloid (Plasmalyte-A; Baxter Healthcare Corp, Chicago, IL). The PFC emulsion used was Oxyfluor (perfluorodichlorooctane emulsified in egg yolk lecithin and safflower oil; HemaGen Inc, St. Louis, MO). This preparation constitutes a 40% PFC volume/volume emulsion. Pigs in the PFC groups received 10 mL/kg of Oxyfluor, which was added to the CPB prime solution before bypass. In all groups the total volume of the CPB prime solution was 1,200 mL.

The emulsifying agent (egg yolk lecithin and safflower oil) for the PFC solution is prepared by a process nearly identical to that used for the preparation of the commercially available fat emulsion Intralipid and is manufactured at the same facility (Pharmacia, Inc, Clayton, NC). Intralipid has been given to many thousands of patients, and very few reactions of any kind have been observed, even in those receiving large amounts [11]. Fluosol-DA also contains egg yolk lecithin, and there have been no reported adverse effects attributable to the phospholipid in this preparation either [11]. Intralipid alone has no demonstrated effect on cerebral blood flow or electrical activity. In addition, infusions of 10% Intralipid in pigs were found to have no effect on any systemic hemodynamic variables, including mean arterial blood pressure, cardiac output, heart rate, stroke volume, systemic vascular resistance, and left ventricular end-diastolic pressure [12].

	Protocol

Top Footnotes Abstract Introduction Material and Methods Surgical Preparation Cardiopulmonary Bypass Protocol Methods of Measurement Results Cerebral Infarction Cerebral Blood Flow Electrical Activity of the... Comment Acknowledgments References

 
Seventeen pigs were randomized to receive either the crystalloid prime solution or the crystalloid prime solution supplemented with PFC. Twelve pigs were entered into the experimental arm of the study and received an air insult. Five pigs served as controls and received only a saline insult. The principal investigators and persons performing the data collection and analysis were blinded to the group assignment.

The pigs' conditions were stabilized on CPB for an initial 10-minute period. Next, in the experimental groups, an insult of room air (5 mL/kg) was delivered via the carotid artery over a 30-second period to simulate the occurrence of a massive air embolism during CPB. The 5 control pigs received either the standard crystalloid prime solution or the PFC-supplemented prime solution; however, instead of room air these pigs received a saline solution bolus (5 mL/kg) after 10 minutes of stabilization on bypass to serve as a sham insult or negative control. Normothermic CPB was maintained for 1 hour after the delivery of air or saline solution. The pigs were then weaned from CPB, and spontaneous perfusion was allowed to occur for 5 more hours while they were still under general anesthesia. All pigs received maintenance intravenous fluids (lactated Ringer's or normal saline solution) throughout the protocol to keep the catheters patent and to clear the lines after the bolus administration of air, saline solution, or microspheres.

The experimental period was extended to 6 hours after the air or saline insult to allow the cerebral infarct to fully evolve. This 6-hour period was chosen on the basis of pilot laboratory studies that determined the time it took for full infarcts to develop, as well as reports in the literature of studies in which the 2,3,5-triphenyltetrazolium chloride (TTC)–immersion technique was used for the examination of infarcted cerebral tissue [13]. At the end of the experimental period, the pigs were euthanized with an injection of concentrated pentobarbital.

The electrocardiogram, EEG, arterial blood pressure, arterial and venous blood gas concentrations, and temperature were monitored throughout the experiment. Pigs were excluded from the study if significant physiologic variations occurred that might have affected cerebral blood flow. Three animals were excluded for such reasons, and the physiologic variations, included (1) hypoxia after weaning from CPB (partial pressure of oxygen of less than 80 mm Hg), (2) hypoxia during CPB and low arterial blood pressure (25 mm Hg) during weaning, and (3) low arterial blood pressure after weaning from CPB and a junctional electrocardiogram. Each of the 3 animals was in a different experimental group, and thus the disturbances were not associated with the protocol used in a particular group. In the final group distribution, 5 pigs were in each experimental group and 2 pigs were in each control group.

	Methods of Measurement

Top Footnotes Abstract Introduction Material and Methods Surgical Preparation Cardiopulmonary Bypass Protocol Methods of Measurement Results Cerebral Infarction Cerebral Blood Flow Electrical Activity of the... Comment Acknowledgments References

 
PHYSIOLOGIC STABILITY.
Physiologic stability was assessed by the continuous monitoring of arterial blood pressure and the intermittent measurement of the partial pressure arterial oxygen and carbon dioxide and the hematocrit. Intermittent measurements were obtained at eleven time points in the study. Five time points corresponded to the times of microsphere delivery (baseline; 10 minutes into CPB; 5, 60, and 360 minutes after insult); the remaining time points were throughout the period of CPB and hourly during the subsequent 5-hour recovery period. Respiratory rate and tidal volume were adjusted (volume constant respirator, model 607; Harvard Apparatus) to provide a physiologic acid-base balance. The pH and minute ventilation were measured at the same eleven time points as the arterial blood gas concentrations and hematocrit. The amount of maintenance intravenous fluids given throughout the study period was also recorded for each animal.

CEREBRAL INFARCTION.
Cerebral infarction was assessed by the macroscopic staining of serial cross-sections of the brain tissue. After the animal was euthanized, the cranium was opened and the whole brain was excised, then weighed, frozen overnight, and sectioned in coronal planes at 7-mm intervals. Seven to nine coronal sections were obtained, depending on the anterior-to-posterior dimension of the brain. Areas of cerebral infarction were identified by immersing the sections in a 1% TTC stain. This stain identifies succinic dehydrogenase, a mitochondrial enzyme found only in viable cells, by staining it bright red. Nonviable cells or areas of infarct do not stain but remain white. The anterior and posterior sides of each serial section were photographed alongside a graded ruler. The photographs were then assessed for regions of infarct using the following criteria: the presence or absence of unstained or white areas, the ability to distinctly define the border of the white areas, and the consistency of the identified areas in adjacent sections. The person assessing the photographs was blinded to the experimental groups.

In specimens in which cerebral infarcts were identified, the volume of the infarcted tissue was quantified by computer digitization. The surface area of each cerebral `section and the area of the infarct were calculated by digitizing the section or infarct border in the photographs. This was done for both the anterior and posterior surface of each section. The area calculation was corrected for any photographic magnification or reduction by digitizing a 5-cm length of the graded ruler in the same photograph and scaling the digitized lengths and areas to the actual size. The volume (V) of each section was then determined from the area measurements, using the following formula:

where A_A is the digitized area on the anterior surface, A_P is the digitized area on the posterior surface, and h is the section thickness (7 mm).

CEREBRAL BLOOD FLOW.
Regional cerebral blood flow was assessed by using colorimetric microspheres. These microspheres were injected into the arterial system of the pig at five time points: baseline; 10 minutes on CPB; and 5, 60, and 360 minutes after insult. At each time point a standardized aliquot (5 x 10⁶) of colored polystyrene microspheres (15 µm; E-Z Trac, Inc, Los Angeles, CA) was injected. The aliquot was injected into the left atrium to ensure adequate mixing when the pig was not on CPB and into the CPB aortic cannula when the pig was on CPB. Reference blood samples were obtained from the right femoral artery catheter at a rate of 10 mL/min, beginning 15 seconds before microsphere injection and continuing until 90 seconds after the injection. After TTC staining of the sections had been completed and the sections photographed 1.5- to 2.5-g samples of brain tissue were excised from each section and the microsphere distribution in each section assessed the hemisphere each section came from (right or left) and the relative position of the section cephalad to caudad were identified. A total of 14 to 18 samples were obtained for each pig, again depending on the anterior-to-posterior dimension of the brain. Blood flow (in milliliters per milligram of tissue per minute) in each section was determined by counting the microspheres (differentiated by color) in each tissue sample and relating the number of microspheres in the tissue to the number of microspheres in the reference blood samples. This analysis was done by an independent laboratory specializing in colorimetric microsphere technology (E-Z Trac, Inc). This laboratory was blinded to the experimental protocol and anticipated outcome.

The mean cerebral blood flow and standard error was calculated for all four groups, after first determining the average cerebral blood flow for each pig (number of samples per pig = 12.1 ± 0.3 [mean ± standard error of the mean]). Differences between the crystalloid-air and PFC-air groups at equivalent time points were assessed by unpaired t tests, and the significance level was set at a p value of less than 0.05.

ELECTRICAL ACTIVITY OF THE BRAIN.
Electrical activity of the brain was determined using a Lifescan monitoring system (Neurometrics/Diatek Inc, San Diego, CA). The EEG recordings were acquired continuously and averaged every 60 seconds. Waveforms with amplitudes of between 13 and 50 mV at frequencies of between 0.5 and 30 Hz were analyzed. The computer displayed an arithmetic summation of millivolts every 60 seconds separately in the alpha, beta, theta, and delta ranges. On-line digital analysis of the original signal was performed using a proprietary software package (Diatek, Inc, San Diego, CA) and a laptop computer (Everex). The power of each frequency band (defined as the arithmetic summation in millivolts that occurred in a particular frequency band over a 1-minute period) in both the right and left hemispheres was recorded, as well as the total power in each hemisphere.

The mean EEG power at each time point was determined for all four groups. Standard errors were determined only for the air insult groups, in which there were 5 animals each. Differences between groups at equivalent time points were assessed by analysis of variance, with the significance level set at a p value of less than 0.05.

	Results

Top Footnotes Abstract Introduction Material and Methods Surgical Preparation Cardiopulmonary Bypass Protocol Methods of Measurement Results Cerebral Infarction Cerebral Blood Flow Electrical Activity of the... Comment Acknowledgments References

 
Physiologic Stability
Arterial blood pressure, arterial partial pressure of oxygen and carbon dioxide, and hematocrit remained within physiologic ranges throughout the protocol in all groups, with minimal intervention. Pigs in all groups were maintained at a physiologic pH throughout the protocol (7.41 ± 0.01 [mean ± standard error of the mean]), including the period of CPB (7.39 ± 0.01 [mean ± standard error of the mean]), with a relatively constant minute ventilation (4.6 ± 0.1 L/min while off CPB [mean ± standard error of the mean]). Maintenance and flush fluids given throughout the protocol were minimal for an experiment lasting more than 6 hours (lactated Ringer's volume, 0.93 ± 0.12 L; saline volume, 0.72 ± 0.12 L [mean ± standard error of the mean]). None of the study pigs required considerable volume resuscitation to maintain physiologic stability.

	Cerebral Infarction

Top Footnotes Abstract Introduction Material and Methods Surgical Preparation Cardiopulmonary Bypass Protocol Methods of Measurement Results Cerebral Infarction Cerebral Blood Flow Electrical Activity of the... Comment Acknowledgments References

 
The incidence and severity of cerebral infarctions (distinctly white areas) was decreased in brain sections from pigs that received PFC-supplemented prime solution and an air insult (PFC-air group), as opposed to those that received the standard crystalloid prime solution and an air insult (crystalloid-air group). In the PFC-air group, there were no areas of infarct in any of the brain sections (Fig 1A). Conversely, in 3 of the 5 pigs in the crystalloid-air group, large areas of infarct were identified in TTC-stained sections from both the left and right hemispheres. The incidence of cerebral infarction (3/5 versus 0/5) approached statistical significance (p = 0.083, Fisher's exact test). The areas of infarct in the crystalloid-air group had distinctly identifiable borders and were consistent in adjacent sections (Fig 1B). Computerized digital analysis of the photographs of the brain sections with cerebral infarcts showed that the volume percents of the infarcted tissue in the 3 pigs with infarcts that received crystalloid solution were 5.9%, 6.9%, and 18.1%, respectively. The average volume of the infarct in the 5 animals in the crystalloid-air group was 3.06 ± 3.36 mL. The difference in the volume between this group and the PFC-air group, in which no infarct occurred, approached statistical significance (p = 0.076, unpaired t test). As anticipated, no areas of infarct were identified in the pig in either control group.

View larger version (95K): [in this window] [in a new window]   Fig 1. . Triphenyltetrazolium chloride–stained cerebral sections. (A) Section from a pig that received perfluorocarbon–supplemented prime solution and an air insult, showing no infarctions. (B) Section from a pig that received crystalloid prime solution and an air insult, showing several large infarct areas.

	Cerebral Blood Flow

Top Footnotes Abstract Introduction Material and Methods Surgical Preparation Cardiopulmonary Bypass Protocol Methods of Measurement Results Cerebral Infarction Cerebral Blood Flow Electrical Activity of the... Comment Acknowledgments References

View larger version (28K): [in this window] [in a new window]   Fig 2. . Improved cerebral blood flow in pigs receiving perfluorocarbon (PFC)–supplemented crystalloid (Cryst) prime solution with initiation of cardiopulmonary bypass (CPB) and immediately after air insult. Values are given and means and standard errors for groups in which there were 5 pigs. (sal = saline solution.)

	Electrical Activity of the Brain

Top Footnotes Abstract Introduction Material and Methods Surgical Preparation Cardiopulmonary Bypass Protocol Methods of Measurement Results Cerebral Infarction Cerebral Blood Flow Electrical Activity of the... Comment Acknowledgments References

The baseline EEG total power did not differ significantly among the groups before CPB (Fig 3). All groups showed a decrease in the EEG total power at 10 minutes after the initiation of CPB, but there were no statistical differences among the groups. The air insult in the PFC-air and crystalloid-air groups resulted in dramatic decreases in the EEG total power within 60 seconds. At 5 minutes after the insult the EEG power had decreased to 19% of the CPB stabilization level in the PFC-air group and to 5% in the crystalloid-air group. The difference between the two groups did not attain statistical significance, however. At 30 minutes after the insult, the EEG power in the PFC-air group had recovered to 45% of the CPB stabilization level, an amount significantly greater than that in the crystalloid-air group, which only recovered to 8% of the CPB stabilization level (p < 0.005 between groups). At 1 hour after the insult, the EEG power in the PFC-air group had recovered to 48% of the CPB stabilization level and that in the crystalloid-air group had recovered to only 18% (p < 0.05 between groups). Control groups (PFC and crystalloid prime) did not show any further reduction in the EEG power after the initial decrease that occurred with the onset of CPB. The group receiving the crystalloid prime and saline insult did show a progressively increasing EEG total power late in the study at 6 hours after the insult. However, this control group only had 2 pigs, and the difference is attributed to the fact that 1 animal did not follow the normal deterioration process typically seen by 6 hours in animals in this particular protocol.

View larger version (27K): [in this window] [in a new window]   Fig 3. . Improved total electroencephalogram (EEG) power in pigs receiving perfluorocarbon (PFC)–supplemented crystalloid (Cryst) prime solution with initiation of cardiopulmonary bypass (CPB), immediately after air insult, and at 30 and 60 minutes after insult. Values are given as means and standard errors are given for groups in which there were 5 pigs. (sal = saline solution.)

	Comment

Top Footnotes Abstract Introduction Material and Methods Surgical Preparation Cardiopulmonary Bypass Protocol Methods of Measurement Results Cerebral Infarction Cerebral Blood Flow Electrical Activity of the... Comment Acknowledgments References

It should be pointed out, however, that the cerebral injuries occurred bilaterally in the crystalloid-air group after a unilateral air insult. This may be explained by the fact that pigs have been reported to have an excellent collateral cerebral circulation [15]. Because of this extensive collateralization, the air introduced into the right-sided circulation would have access to the left-sided circulation through the collateral flow.

The observations in the crystalloid-air group may be explained by the physical nature of an air embolism. The cerebral blood flow in this group was significantly reduced immediately after the delivery of air, and the EEG total power fell correspondingly. The room air bolus (which is 78% nitrogen) is relatively insoluble in blood plasma and is slow to be absorbed or fractionated in the arterial and venous beds of the brain. This results in the formation of large air emboli that in turn become "air locks" in arteries, arterioles, and capillaries. It is these air locks that impede blood flow. This phenomenon may resolve as the air locks are flushed out of the system. However, because cerebral cell death occurs within 3 to 5 minutes of ischemia, by the time the air lock has been dislodged, it is too late for the cerebral cells to recover function. Cerebral cell death is shown in the crystalloid-air group by the continued depression of the EEG activity and the large infarcts shown by TTC staining.

The beneficial effects of PFC observed in this study can be linked to the differences between plasma and PFC and their relative abilities to absorb air. No infarcts were seen in the PFC-air group, and cerebral blood flow and EEG activity were also depressed for less time in this group, and to a lesser extent than that in the crystalloid-air pigs. There are three physical properties of PFC emulsions that account for these observations. First, the solubility of PFC emulsions for gases (ie, oxygen, carbon dioxide, and particularly nitrogen) is 100,000 times greater than that of plasma) [16]. Thus, air introduced into the circulation can be absorbed by the PFC emulsion. The air locks that otherwise prevent cerebral blood flow in the crystalloid-air group are either diminished in number and size or do not occur in the PFC-air group. Second, unlike hemoglobin, PFC emulsions do not actively bind oxygen. Therefore oxygen dissolved in the PFC emulsion does not have to undergo oxyhemoglobin dissociation. In addition, oxygen release is not affected by 2,3-diphosphoglycerate, acid-base balance, or other biochemical factors. The result of all this is that the diffusion capability of gases within the PFC emulsion is approximately 17 times greater than that of gases in blood plasma. Furthermore, the oxygen unloading occurring in nearby erythrocytes may be enhanced by this improved diffusion capability of gases in the PFC emulsion. This enhanced diffusion capability may be of benefit in low-flow states as well, in that the PFC emulsion can release oxygen to cerebral tissue only on the basis of its diffusion gradient. Third, the size of the PFC particles (0.1 to 0.3 µm) is quite small compared with that of red blood cells (7 to 8 µm). Thus, PFC particles can pass through areas of stenosis or obstruction that red blood cells cannot pass through and deliver oxygen to tissues that might otherwise become hypoxic [17]. In addition, the small size of the particles in the PFC emulsion makes for a massive increase in the surface area available for gas exchange, improving embolus absorption as well as oxygen release.

The use of PFC emulsions to reduce the complications of air embolism during CPB has been theoretically appealing [18], but it is only recently that their clinical use has become a real possibility. The first generation of PFC compounds was limited by the toxicity of their emulsifying agents, difficulty with their storage and preparation, and their low concentrations (10% PFC volume/volume) with relatively low gas solubilities [19]. As such, the total contribution of these agents to oxygen delivery may have been less than 1 volume percent. In addition, the first-generation PFC compounds had relatively long half-lives. These limitations have been overcome for Oxyfluor, the second-generation PFC emulsion used in this investigation. Oxyfluor's major emulsifier, lecithin, is nontoxic, and the formulation allows a much greater active PFC concentration (40% PFC volume/volume). This results in a tissue oxygen delivery of 5 to 8 volume percent. Also, Oxyfluor does not undergo hepatic or renal biodegradation but is volatilized through the lungs and cleared within several days of administration. These improvements in the second-generation PFC emulsions have made their clinical application possible.

However, as with any newly proposed clinical intervention, the cost-benefit ratio must be considered. On the basis of United States figures, the average cost of a stroke (hospital, physician, rehabilitation, and equipment charges, as well as lost productivity) is $54,545 [20]. Considering the estimated 1% to 9% incidence of stroke in the 300,000 patients undergoing CPB, the total cost of stroke in this group would range from $164 million to $1.5 billion. Of course, not all strokes related to CPB would be prevented by PFC use, because the mechanisms causing stroke are variable. A conservative estimate of the percentage of CPB-related strokes resulting from air emboli would be 10% to 25%. If the incidence of strokes in this patient population could be reduced by PFC use, there should be an overall cost savings if the commercial PFC product could be made available at a cost of $500 to $1,000 per liter. Additional savings might occur from the use of PFCs to prevent altered mental status (incidence, 3% to 4%) and neuropsychometric deficits (incidence, 24% to 40%), but these savings are much more difficult to quantify. In addition, there may be an overall cost savings if the need for blood products could be reduced through the use of PFC as a blood substitute. To assess whether the cost of PFC pretreatment could be reduced, we have undertaken studies to determine whether lower doses are effective as a prime additive and whether the addition of PFC to the CPB circuit as a treatment after known air embolism would be equally effective.

In summary, we have shown that the use of PFC emulsions as a CPB prime additive led to a profound decrease in the incidence and severity of neurologic injury after massive air embolism in our pig model. It would seem reasonable that this agent, which is effective in preventing the sequelae of massive air embolisms, would be at least as effective in the prevention of microembolisms. Because the surface-to-volume ratio increases with decreasing bubble size, the rate of absorption of smaller bubbles (the case in microembolisms) would be increased. In addition, the physical properties of PFC, including enhanced diffusion and rapid oxygen unloading, make it an appealing additive to the CPB prime solution to provide some protection against other causes of neurologic dysfunction, such as thromboembolism, low-flow states, and other hypoxic injuries. Finally, we believe that there should be a benefit in terms of its preventing the subtle neuropsychometric changes reported after CPB; however, because this cannot be evaluated in a pig model, further experimentation is necessary to define the benefit.

Further work regarding the safety and efficacy of second-generation PFCs in preventing neurologic dysfunction after CPB is ongoing. Phase I safety studies in healthy human volunteers have shown that doses of PFC of up to 1 mL/kg were well tolerated, with the only observed side effects being transient, self-limiting flulike symptoms and mild, transient thrombocytopenia in some volunteers. The safety of Oxyfluor is currently being tested in normal human volunteers at concentrations of 0.25 to 0.5 times the levels used to provide the cerebral organ protection demonstrated in our study, with dexamethasone pretreatment to prevent the flulike side effects. It is hoped that the clinical use of PFC emulsions as an additive to CPB prime solutions will become a testable reality in the near future.

	Acknowledgments

Top Footnotes Abstract Introduction Material and Methods Surgical Preparation Cardiopulmonary Bypass Protocol Methods of Measurement Results Cerebral Infarction Cerebral Blood Flow Electrical Activity of the... Comment Acknowledgments References

 
This study was supported in part by grants from HemaGen, Inc (St. Louis, MO) and the American Society for ExtraCorporeal Technology. We gratefully acknowledge the technical assistance of Christine Rothnie, LVT, and Gina Lindley, BS.

	Footnotes

Top Footnotes Abstract Introduction Material and Methods Surgical Preparation Cardiopulmonary Bypass Protocol Methods of Measurement Results Cerebral Infarction Cerebral Blood Flow Electrical Activity of the... Comment Acknowledgments References

 
Address reprint requests to Dr Cochran, Division of Cardiothoracic Surgery, Box 356310, University of Washington, 1959 NE Pacific St, Seattle, WA 98195-6310 (e-mail: jasmite{at}u.washington.edu).

	References

Top Footnotes Abstract Introduction Material and Methods Surgical Preparation Cardiopulmonary Bypass Protocol Methods of Measurement Results Cerebral Infarction Cerebral Blood Flow Electrical Activity of the... Comment Acknowledgments References

 

1. Mangano DT. Perioperative cardiac morbidity. Anesthesiology 1990;72:153–84.[Medline]
2. Slogoff S, Girgis KZ, Keats AZ. Etiologic factors in neuropsychiatric complications associated with cardiopulmonary bypass. Anesth Analg 1982;61:903–11.[Abstract/Free Full Text]
3. Shaw PH, Bates D, Cartlidge NEF, et al. Neurologic and neuropsychologic morbidity following major surgery. Comparisons of coronary artery bypass and peripheral vascular surgery. Stroke 1987;188:700–7.
4. Smith PL, Treasure T, Neuman SP, et al. Cerebral consequences of cardiopulmonary bypass. Lancet 1986;1:823–5.[Medline]
5. Mills NL, Ochsner JL. Massive air embolism during cardiopulmonary bypass: causes, prevention, and management. J Thorac Cardiovasc Surg 1980;80:208–15.
6. Spiess BD, McCarthy R, Piotrowski D, Ivankovich AD. Protection from venous air embolism with fluorocarbon emulsion FC-43. J Surg Res 1986;41:439–44.[Medline]
7. Spiess BD, Braveman B, Waroniwicz AW, Ivankovich AD. Protection from cerebral air emboli perfluorocarbons in rabbits. Stroke 1986;17:1146–9.[Abstract/Free Full Text]
8. Spiess BD, McCarthy RJ, Tuman KJ, Ivankovich AD. Protection from coronary air embolism by a perfluorocarbon emulsion (FC-43). J Cardiothorac Anesth 1987;1:210–5.[Medline]
9. Gillinov AM, Redmond JM, Zehr KJ, et al. Inhibition of neutrophil adhesion during cardiopulmonary bypass. Ann Thorac Surg 1994;57:126–33.[Abstract]
10. Zehr KH, Poston RS, Lee PC, et al. Platelet activating factor inhibition reduces lung injury after cardiopulmonary bypass. Ann Thorac Surg 1995;59:328–35.[Abstract/Free Full Text]
11. Geyer RP. Perfluorochemicals as oxygen transport vehicles. Biomater Artif Cells Artif Org 1988;16:31–49.
12. VanWoerkens LJ, Van der Giessen WY, Verdouw PD. Cardiovascular profile of 5 novel nitrate-esters: a comparative study with nitroglycerin in pigs with and without left ventricular dysfunction. Br J Pharmacol 1991;104:7–14.[Medline]
13. Dettmers C, Hartmann A, Rommel T, et al. Immersion and perfusion staining with 2,3,5-triphenyltetrazolium chloride (TTC) compared to mitochondrial enzymes 6 hours after MCA occlusion in primates. Neurol Res 1994;16:205–8.[Medline]
14. Blauth C, Griffin S, Harrison M, et al. Neuropsychologic alterations after cardiac operation. J Thorac Cardiovasc Surg 1989;98:454–9.[Medline]
15. Moss G. The adequacy of the cerebral collateral circulation: tolerance of awake experimental animals to acute bilateral common carotid artery occlusion. J Surg Res 1974;16:337–8.[Medline]
16. O'Brien RN. Diffusion coefficients of respiratory gases in a perfluorocarbon liquid. Science 1982;217:153–5.[Abstract/Free Full Text]
17. Lee PA, Sylvia AL, Piantadosi CA. Effect of fluorocarbon for blood exchange as regional cerebral blood flow in rats. Am J Physiol 1988;254:719–26.
18. Menasché P, Pinard E, Desroches AM, et al. Fluorocarbons: a potential treatment of cerebral air embolism in open heart surgery. Ann Thorac Surg 1985;40:494–7.[Abstract]
19. Marchbank A. Fluorocarbon emulsions. Perfusion 1995;10:67–88.[Free Full Text]
20. Dobkin B. The economic impact of stroke. Neurology 1995;45:S6–9.[Abstract]

This article has been cited by other articles:

				B. D. Spiess Perfluorocarbon emulsions as a promising technology: a review of tissue and vascular gas dynamics J Appl Physiol, April 1, 2009; 106(4): 1444 - 1452. [Abstract] [Full Text] [PDF]

				F. de Lange, K. Yoshitani, A. D. Proia, G. B. Mackensen, and H. P. Grocott Perfluorocarbon Administration During Cardiopulmonary Bypass in Rats: An Inflammatory Link to Adverse Outcome? Anesth. Analg., January 1, 2008; 106(1): 24 - 31. [Abstract] [Full Text] [PDF]

				K. Yoshitani, F. de Lange, Q. Ma, H. P. Grocott, and G. B. Mackensen Reduction in Air Bubble Size Using Perfluorocarbons During Cardiopulmonary Bypass in the Rat Anesth. Analg., November 1, 2006; 103(5): 1089 - 1093. [Abstract] [Full Text] [PDF]

				S. E. Hill, H. P. Grocott, B. J. Leone, W. D. White, M. F. Newman, and Neurologic Outcome Research Group of the Duke Hear Cerebral Physiology of Cardiac Surgical Patients Treated with the Perfluorocarbon Emulsion, AF0144 Ann. Thorac. Surg., October 1, 2005; 80(4): 1401 - 1407. [Abstract] [Full Text] [PDF]

				M. Barak and Y. Katz Microbubbles: Pathophysiology and Clinical Implications Chest, October 1, 2005; 128(4): 2918 - 2932. [Abstract] [Full Text] [PDF]

				T. E. Dye, R. P. Cochran, K. S. Kunzelman, C. R. Vocelka, Ccp, H. Akimoto, R. Thomas, L. O. Soltow, and B. D. Spiess Macroscopic Bubbles From Dissolved Nitrogen During CPB Ann. Thorac. Surg., November 1, 1997; 64 (5): 1527 - 1527. [Full Text]

This Article Abstract Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Add to Personal Folders Download to citation manager Author home page(s): Richard P. Cochran Bruce D. Spiess Permission Requests Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Cochran, R. P. Articles by Spiess, B. D. Search for Related Content PubMed PubMed Citation Articles by Cochran, R. P. Articles by Spiess, B. D.