Pulsatile Versus Nonpulsatile Flow : No Difference in Cerebral Blood Flow or Metabolism during Normothermic Cardiopulmonary Bypass in Rabbits

Experimental protocols were approved by the Animal Care and Use Committee of the University of Iowa in accordance with the Guide for the Care and Use of Laboratory Animals.* The following procedures were common to both experimental protocols.

Anesthesia was induced in New Zealand white rabbits (weight, 4.0–4.8 kg) by intravenous administration of 2 mg/kg diazepam and 10–15 micro gram/kg fentanyl via a 22-G ear vein catheter. After local infiltration with 1% lidocaine, a tracheotomy was performed and the trachea intubated with a 3.0 cuffed endotracheal tube. Thereafter, the animals' lungs were mechanically ventilated to achieve normocarbia, and anesthesia was maintained with 2% isoflurane in oxygen for the remainder of pre-CPB preparation. Animals were paralyzed with a succinylcholine/lactated Ringer's infusion (4 ml *symbol* kg¹*symbol* h sup -1) and placed prone. After a midline sagittal scalp incision, a 2-mm burr hole was drilled over the right frontoparietal cortex, and a 1-mm thermocouple (K-type, L-08419–02, Cole Parmer, Chicago, IL) was introduced under the cranium to rest on the dural surface. A posterior midline craniectomy was performed, exposing the confluens sinuum. Heparin was administered as a bolus (200 U/kg intravenously) and was added to the succinylcholine/lactated Ringer's infusion to give a maintenance dose of 200 U *symbol* kg¹*symbol* h¹. The tip of a saline-filled polyethylene catheter (PE-90, Intramedic, Parsippany, NJ) was placed in the confluens sinuum, permitting collection of cerebral venous blood. The cortical thermocouple and cerebral venous catheter were secured with bone wax and fast-drying cyanoacrylate cement, and the animals were placed supine.

The tip of a saline-filled catheter (PE-90), introduced via the right external jugular vein, was advanced to the superior vena cava to measure central venous pressure. An incision was made 2–3 mm inferior to the midportion of the mandibular ramus, and the facial artery was isolated. The carotid sinus and internal carotid artery were not manipulated. The facial artery was cannulated in retrograde fashion with an “end-view” 4-Fr solid-state pressure transducer (model 110–4G, Camino Laboratories, San Diego, CA) such that the tip of the transducer was approximately 8–11 mm distal to the internal/external carotid bifurcation. At selected intervals during CPB (see below), the signal from this catheter was digitized and recorded on a computer hard disk. Both brachial arteries were cannulated (saline-filled PE-160 tubing) for microsphere reference blood sampling. The left brachial arterial catheter also was used for arterial pressure monitoring and collection of arterial blood.

A midline abdominal incision was made. Viscera were covered with a transparent plastic sheet (SaranWrap, Dow Brands, Indianapolis, IN), and the distal abdominal aorta was isolated from surrounding tissue. The sternum was divided in midline, the thymus was retracted, and a Teflon-pledgeted 4-O silk purse-string suture was placed in the right atrium. After systemic anticoagulation with heparin (300 U/kg, intravenously), the distal aorta was ligated and cannulated in retrograde fashion with a 10-Fr pediatric arterial perfusion cannula (Biomedicus, Eden Prairie, MN) 7–10 mm superior to the distal aortic bifurcation. (Our prior technique of bilateral femoral artery cannulation [11] was abandoned because of marked damping of pulsatile waveforms.) A 21-Fr venous cannula (Polystan, Ballerup, Denmark) was placed in the right atrium. The aortic and right atrial cannulas were connected to the perfusion circuit, and normothermic CPB was initiated as described below. Approximately 30 min before CPB, isoflurane, maintenance fluids, and the succinylcholine/heparin infusion were discontinued. Anesthesia was maintained for the rest of the experiment with fentanyl (100-micro gram/kg bolus, 150-micro gram *symbol* kg sup -1 *symbol* h sup -1 infusion) and diazepam (2-mg/kg bolus, 3-mg *symbol* kg sup -1 *symbol* h sup -1 infusion). Muscle relaxation was achieved with pancuronium (0.2 mg/kg).

The CPB circuit consisted of a venous reservoir, a membrane oxygenator/heat exchanger (Capiox 308, Terumo, Piscataway, NJ), a variable-temperature water pump (VWR Scientific, San Francisco, CA), and either a pulsatile perfusion pump (Medical Engineering Consultants, Los Angeles, CA) or a nonpulsatile centrifugal pump (model 540, Biomedicus, BP-50 pump head), depending on the experiment (see below). With the pulsatile pump, the duration of systolic ejection (ejection period, ms), pulse rate (pulse/min), and stroke volume (ml) were each independently variable. Circuit priming fluid consisted of 300 ml 6.5%(weight/vol) high molecular weight hydroxyethyl starch (McGaw, Irvine, CA) in 0.72 N sodium chloride, 18 mEq sodium bicarbonate, 250 mg CaCl², and 1,000 U heparin. The priming fluid was circulated through a 40-micro meter filter for 15–20 min before addition of approximately 150 ml fresh, filtered, packed rabbit erythrocytes, achieving a priming hemoglobin concentration of 6–10 g/dL (OSM3 (rabbit absorption coefficients), Radiometer, Copenhagen, Denmark).

CPB was initiated and maintained throughout the experiment at a systemic flow rate of 100 ml *symbol* kg¹*symbol* min¹, monitored with a calibrated in-line electromagnetic flow meter (Biomedicus, TX-40P). The pulmonary artery was clamped to ensure complete venous outflow to the CPB circuit. To prevent left ventricular ejection and/or distension, the tip of a 14-G catheter was placed transapically in the left ventricle to permit drainage to the venous reservoir. Water temperature to the heat exchanger was maintained at 37 degrees Celsius. The oxygenator was ventilated with a variable mixture of oxygen and nitrogen to maintain Pa^CO2near 40 mmHg and Pa sub O²near 250 mmHg when measured at an electrode temperature of 37 degrees Celsius (IL1304, Instrumentation Laboratory, Lexington, MA). Shed blood from the surgical field was returned to the venous reservoir after passing through a 40-micro meter filter. Sodium bicarbonate was given to maintain a base excess greater than -4 mEq/L, calculated at 37 degrees Celsius. Rabbit erythrocytes were given to maintain hemoglobin concentration between 6.7–8.5 g/dL. No pharmacologic or mechanical means were used to control arterial pressure. At experiment completion, animals were killed by discontinuation of CPB and intracardiac administration of KCl solution.

CBF was measured by the radioactive microsphere technique. Isotopes used included⁴⁶Scandium,⁸⁵Strontium,⁹⁵Niobium,¹⁴¹Cerium (New England Nuclear, Boston, MA), although only two isotopes were used in each experiment. Stock microspheres (200 micro liter, approximately 0.9 million microspheres), vigorously mixed for 5 min before withdrawal, were diluted in 1.5 ml suspending solution (10% dextran-40 in normal saline with 0.5%(vol/vol) Tween-80) and mixed an additional 60 s. Microspheres were injected over 30 s into the arterial perfusion line approximately 25 cm proximal to the distal tip of the aortic cannula. Starting 15 s before microsphere injection and continuing 2 min thereafter, blood was simultaneously withdrawn from each brachial arterial catheter via a calibrated withdrawal pump (1.96 ml/min). After the experiment, the brain was removed and dissected into the following regions: right and left cerebral hemispheres, cerebellum, midbrain, and medulla. Fresh tissue samples were weighed, placed in counting tubes and, with reference blood samples, each counted for 5 min in a sodium iodide well-type gamma counter (Minaxi gamma Auto-Gamma 5000, Packard Instruments, Meriden, CT). Isotope separation, background, overlap corrections, and organ blood flow calculations (ml *symbol* 100 g¹*symbol* min¹) were performed by standard techniques. [12–14] Weight-averaged values for right and left cerebral hemispheric blood flow were used to calculate mean hemispheric CBF.

Oxygen content (ml Oxygen²/dL) was calculated as (1.39 x percent saturation x hemoglobin concentration (g/dL))+(Pa^O2x 0.003). Cerebral oxygen extraction ratio (OER) was calculated as the arterial-cerebral venous oxygen content difference, divided by arterial oxygen content. CMR^O2(ml Oxygen²*symbol* 100 g¹*symbol* min¹) was calculated as the product of mean hemispheric CBF (ml *symbol* 100 g¹*symbol* min¹) and the arterial-cerebral venous oxygen content difference.

The purpose of this experiment was to determine whether CBF or CMR^O2, depended on arterial pressure waveform configuration. We did this to ensure that, before comparing pulsatile versus nonpulsatile CPB (experiment B), artificial pulsatile waveforms adequately reproduced the essential features of native pressure waveforms.

Sixteen animals were anesthetized and prepared for CPB as described. For the first 30 min of CPB, all animals underwent pulsatile perfusion with an ejection period of 140 ms and a pulse rate of 250 pulse/min (pump stroke volume adjusted to maintain systemic flow at 100 ml *symbol* kg¹*symbol* min¹). Animals were randomly assigned to one of two groups: A1 (n = 8, 100-ms ejection period) and A2 (n = 8, 140-ms ejection period). In each animal, the ejection period of the pulsatile pump was adjusted according to assignment and kept constant for the remainder of the experiment. Thereafter, perfusion was maintained for 30 min at each of two different pulse rates: 150 and 250 pulse/min. The order of determination was randomized (150, 250 vs. 250, 150). In each instance, pump stroke volume was adjusted to maintain systemic flow at 100 ml *symbol* kg¹*symbol* min¹. Thus, four arterial pressure waveforms were created (two ejection periods x two pulse rates). At the end of each 30 min perfusion period (60 and 90 min of CPB), the following were recorded: 5 s of arterial pressure waveform (facial artery), arterial pressure from the brachial artery, central venous pressure, systemic flow rate, brain (epidural) temperature, and arterial hemoglobin concentration. Concurrent with these measurements, CBF determinations were made, and arterial and cerebral venous blood was collected for blood gas analysis and measurement of oxyhemoglobin saturation (OSM3, Radiometer).

Finding CBF and CMR^O2to be equivalent among the four tested waveforms (see Results, Experiment A), the aim of the second experiment was to compare CBF and CMR^O2between pulsatile and nonpulsatile normothermic CPB. Twenty-two additional animals were prepared for and underwent CPB as described. Animals were randomly assigned to one of two groups: B1 (n = 11, pulsatile perfusion) and B2 (n = 11, nonpulsatile perfusion). In the pulsatile group, CPB was maintained with a constant ejection period of 100 ms and pulse rate of 250 pulse/min. In animals undergoing nonpulsatile perfusion, CPB was maintained with a centrifugal pump. As in experiment A, no pharmacologic or mechanical means were used to influence arterial blood pressure.

At 30 and 60 min of CPB, the following were recorded in each animal: 5 s of arterial pressure waveform (facial artery), arterial pressure from the brachial artery, central venous pressure, systemic flow rate, brain (epidural) temperature, and arterial hemoglobin concentration. As before, CBF determinations made at this time, and arterial and cerebral venous blood was collected for blood gas analysis and measurement of oxyhemoglobin saturation.

The main objective of waveform analysis was to provide a visual representation of arterial pressure waveforms, giving equal weight to each animal at a given combination of ejection period and pulse rate. In addition, we wished to provide numeric estimates of arterial pressure waveform characteristics (e.g., maximal rate of change of pressure (dP/dt) and pulse pressure (difference between systolic and diastolic pressure)). Because these estimates were subject to large errors, no statistical analysis was undertaken. Analog outputs from the facial artery catheter (see above) were taken from the monitoring module (model 420, Camino), amplified, and digitized (model DT2801A, Data Translation, Marlboro, MA; 350 Hz, 12-bit resolution). The first complete trough-to-trough waveform in each animal was used to represent the entire 5-s sample. We estimated dP/dt for each pressure wave by visually selecting data points at the start and end of the pressure upstroke. Pulse pressure was calculated as the difference between peak systolic and trough diastolic pressures. Waveforms were grouped according to group assignment and either pulse rate (experiment A) or CPB duration (experiment B). Waveforms were visually phase aligned using the slope of the pressure wave at each data point. Waveforms were pressure-aligned by subtracting the mean value of the pressure wave from each data point, to create a “corrected waveform,” with a mean of zero. A median waveform was generated by finding the median pressure at each time across all corrected waveforms. The median of the mean pressures was added to each data point of this waveform, creating an overall representative (median) waveform. To provide upper and lower boundaries about median waveforms, maximum and minimum values of corrected waveforms were added to the median of the mean pressures from appropriate data sets.

Right and left microsphere counts appeared to be normally distributed, permitting linear regression analysis to test adequacy of microsphere mixing and distribution. In contrast, box and whisker plots suggested that many physiologic variables were not normally distributed. Consequently, all physiologic variables were summarized using their median plus/minus quartile deviation, the latter equaling half the difference between the first and third quartiles. Changes in physiologic variables were assessed qualitatively to preserve statistical power to detect differences in CBF and CMR^O2.

Analyses were performed using Systat statistical software. [15] CBF appeared to follow a normal distribution, whereas CMR^Osub 2 appeared normally distributed after log transformation. Thus, CMR sub O²data were log-transformed before analysis. In experiment A, CBF and CMR^O2were compared between ejection period groups and pulse rates using univariate repeated measures analysis of variance. In experiment B, CBF and CMR^O2were compared between groups and over time using univariate repeated measures analysis of variance. No interaction terms were statistically significant (P > 0.15). Post hoc power to detect differences between groups was estimated using the mean of the two measurements in each animal. The estimate of the pooled variance between groups used to calculate power [16] incorporated linear correlation between the two measurements. [17].

Paired right and left microsphere reference counts were well matched (r²= 0.995, slope = 1.02, intercept (-37 cpm) not significantly different than zero), indicating adequate microsphere mixing and uniform distribution. There were no right-left CBF asymmetries between the cerebral hemispheres.

Systemic physiologic variables during CPB for groups A1 and A2 are summarized in Table 1. Figure 1shows the corresponding representative (median) arterial pressure waveforms and the upper and lower boundaries about the medians (i.e., the worst cases). Figure 2shows a native (non-CPB) rabbit facial arterial pressure waveform for comparison to artificial waveforms. There were no differences between or within groups A1 and A2 with respect to the following: mean arterial pressure, systemic flow, central venous pressure, CPB duration, arterial pH, P^CO2, P^O2, hemoglobin concentration, or oxygen content. Although mean arterial pressure did not differ between groups (approximately 78 mmHg), as expected, arterial systolic pressure decreased and diastolic pressure increased with increasing pulse rate. Consequently, pulse pressure decreased with increasing pulse rate. Likewise, dP/dt decreased with increasing pulse rate. Mean arterial pressures recorded from the facial artery were 8 plus/minus 4 mmHg (median plus/minus quartile deviation) less than those recorded from the brachial artery.

Cerebral physiologic variables are summarized in Table 2. There were no differences between or within ejection period groups with respect to the following: brain temperature, cerebral venous oxygen content, arterial-cerebral venous oxygen content difference, and cerebral oxygen extraction ratio. Hemispheric CBF (approximately 85 ml *symbol* 100 g sup -1 *symbol* min sup -1) was independent of both ejection period group (P = 0.92) and pulse rate (P = 0.56). Similarly, CMR^O2(approximately 4.3 ml O²*symbol* 100 g sup -1 *symbol* min sup -1) was independent of both ejection period group (P = 0.36) and pulse rate (P = 0.99).

Paired right and left microsphere reference counts were well matched (r²= 0.99, slope = 1.04, intercept (163 cpm) not significantly different than zero), indicating adequate micro-sphere mixing and uniform distribution. There were no right-left CBF asymmetries between the cerebral hemispheres.

Systemic physiologic variables for groups B1 (pulsatile CPB) and B2 (nonpulsatile CPB) are summarized in Table 3. Figure 3shows the corresponding representative (median) arterial pressure waveforms and the upper and lower boundaries about the medians (i.e., the worst cases). There were no differences between groups or within groups (over time) with respect to the following: mean arterial pressure, systemic flow, central venous pressure, arterial pH, P^CO2, P^O2, hemoglobin concentration, or oxygen content. Mean arterial pressures recorded from the facial artery were 11 plus/minus 5 mmHg less than those recorded from the brachial artery.

Cerebral physiologic variables are summarized in Table 4. There were no differences in brain temperature (approximately 37.0 degrees Celsius) between pulsatile and nonpulsatile CPB or over time. Hemispheric CBF did not differ between pulsatile and nonpulsatile perfusion (P = 0.95) at either 30 min (73 plus/minus 5 vs. 79 plus/minus 8 ml *symbol* 100 g sup -1 *symbol* min sup -1, respectively), or 60 min (72 plus/minus 8 vs. 74 plus/minus 9 ml *symbol* 100 g sup -1 *symbol* min sup -1, respectively). There was no significant change in hemispheric CBF in the 30-min interval between measurements (P = 0.08). CMR^O2did not differ between pulsatile and nonpulsatile CPB (P = 0.33) at either 30 min (4.7 plus/minus 0.5 vs. 4.3 plus/minus 0.5 ml O sub 2 *symbol* 100 g sup -1 *symbol* min sup -1, respectively) or 60 min (4.2 plus/minus 0.6 vs. 4.0 plus/minus 0.6 ml O²*symbol* 100 g sup -1 *symbol* min sup -1, respectively). There was a significant (P = 0.014) decrease in CMR^O2between measurements (0.4 plus/minus 0.4 ml O²*symbol* 100 g sup -1 *symbol* min sup -1).

We found no significant difference in either CBF or CMR^Osub 2 between pulsatile and nonpulsatile CPB at normothermia. The rabbit was chosen as the experimental species because, like humans but unlike dogs, cats, sheep, and pigs, rabbits do not possess rete mirabilae—internal/external carotid arterial communications. The rabbit brain is supplied exclusively by paired internal carotid and vertebral arteries, with a fully developed circle of Willis. [18] Also, many physiologic characteristics of rabbits closely approximate those of humans, including:(1) arterial blood pressure, (2) cerebrovascular responses to changing Pa^CO2[19] and blood pressure (autoregulation), [20] and (3) resting CBF and CMR^O2. [21] Besides species differences, this study differs from prior animal studies of pulsatile CPB because, in addition to CBF, CMR^O2was measured. Combined CBF and CMR^O2measurements provide for a better assessment of the relationship between cerebral oxygen supply and demand than CBF measurements alone.

A criticism sometimes made of studies that find no difference between pulsatile and nonpulsatile CPB is that the artificial pressure waveform may not have adequately reproduced the essential features of the native waveform. [22] This criticism is easy to make and difficult to refute because the critical features of pulsatile flow (e.g., dP/dt, pulse rate, pulse pressure, systolic/diastolic ratio) are not known. We addressed this issue in experiment A, by first determining whether CBF or CMR^O2depended on arterial pressure waveform configuration. By altering pump ejection period and pulse rate, four different arterial pressure waveforms were created. Within the range of waveforms tested, we found neither CBF nor CMR^O2depended on arterial pressure waveform configuration. Thus, all four pulsatile waveforms were physiologically equivalent with respect to CBF and CMR sub O²and apparently adequately represented a “true” pulsatile waveform.

We selected one of the four waveforms to serve as our test waveform in the subsequent experiment. In experiment B, we found no difference in CBF between pulsatile and nonpulsatile CPB. This is in contrast to the work of Dernevik et al. [3] and Tranmer et al., [4] in which, in normothermic dogs, pulsatile CPB increased CBF to values 16–19% greater than achieved during nonpulsatile CPB. Post hoc analysis indicates we had sufficient power (alpha = 0.05, beta = 0.80) to detect hemispheric CBF differences between pulsatile and nonpulsatile CPB of 20% of the overall mean. Thus, our group sizes were sufficient to detect previously reported CBF differences. What can explain this discrepancy other than differences in cerebrovascular anatomy between dogs and rabbits/humans?

In the cited dog studies, all animals first underwent nonpulsatile CPB. After CBF was determined, pulsation was initiated, and then the second (pulsatile) CBF determination was made. During nonpulsatile perfusion, carotid sinus-mediated autonomic reflexes increase sympathetic nervous system activity. [23,24] Increases in sympathetic nervous system activity result in cerebral vasoconstriction and modest (approximately 20%) reductions in CBF. [25–27] However, over 3–20 min (depending on species), compensatory vasodilation in the pial and parenchymal vasculature [25,28,29] restores CBF to baseline. [25,29,30] When changing from nonpulsatile to pulsatile CPB (as was done in the cited dog studies), sympathetic nervous system activity should decrease, at least transiently. This acute decrease in sympathetic tone should result in a modest temporary decrease in cerebrovascular resistance [25] and a transient increase (approximately 20%) in CBF. Again, however, this change in sympathetic nervous system activity will not have sustained effects on CBF because of counter-regulatory mechanisms. Thus, it seems possible that, when going from one mode to another, temporary CBF differences, mediated by alterations in sympathetic tone, could exist between pulsatile and nonpulsatile CPB. However, when CPB techniques are maintained at steady-state, as they were in this experiment, CBF differences between pulsatile and nonpulsatile CPB are negligible.

The absence of CBF differences in this study suggests other mechanisms, thought to distinguish pulsatile from nonpulsatile flow, may not influence CBF during CPB. Nitric oxide is a potent cerebral vasodilator [31,32] and plays a major role in maintaining basal cerebrovascular tone. [33] Both Hecker et al. [34] and Lamontagne et al. [35] have shown nitric oxide release increases with increasing endothelial shear stress. Likewise, Pohl et al. [36] and Rubanyi et al. [37] found endothelial nitric oxide release increased when flow was changed from a nonpulsatile to a pulsatile mode. Therefore, one might expect pulsatile CPB to have resulted in greater endothelial nitric oxide release and, consequently, greater CBF as compared to nonpulsatile CPB. However, it has been shown that nitric oxide exerts negative feedback regulation of its own synthesis. [38] Were nonpulsatile flow to decrease nitric oxide release, a feedback-mediated increase of nitric oxide synthesis might restore cerebrovascular tone to baseline. In such a circumstance, sustained CBF differences between pulsatile and nonpulsatile CPB would not exist. Yet another explanation for the lack of an effect of pulsation upon CBF in this experiment may relate to hemodilution. Todd et al. found hemodilution to a hematocrit of approximately 14% completely eliminated CBF responses to nitric oxide synthase inhibition, i.e., no decrease in CBF.* Although a much less profound degree of hemodilution was used in the current experiment (hematocrit approximately 24%), perhaps the hemodilution of CPB decreases basal endothelial nitric oxide release. In such a circumstance, pulse-mediated effects on nitric oxide release and CBF could be rendered inconsequential.

To our knowledge, this is the first study to assess whether CMR^O2differs between pulsatile and nonpulsatile CPB at normothermia. We found no significant difference. Post hoc analysis indicates we had sufficient power (alpha = 0.05, beta = 0.80) to detect CMR^O2differences between pulsatile and nonpulsatile CPB of approximately 25% of the overall mean. Therefore, CMR^O2differences less than 25% could have gone undetected. Some elements of our data suggest there might be a small CMR^O2difference between pulsatile and nonpulsatile CPB. The OER was approximately 10% greater with pulsatile CPB (Table 4). Analysis of the OER data (repeated-measures analysis of variance) suggests OER might have been greater in the pulsatile group (P = 0.03). Given CBF equivalence under pulsatile and nonpulsatile conditions, this implies CMR^O2should be approximately 10% greater with pulsatile flow. Inspection of Table 4shows median CMR^O2values were approximately 10% greater in the pulsatile group. However, several considerations must be kept in mind. First, inspection of Table 4shows OER differences between pulsatile and nonpulsatile CPB were neither large nor uniformly present and, given the number of statistical comparisons made, a P value of 0.03 cannot be considered significant. Second, unlike all other comparisons made in this study, OER data violated several assumptions inherent in the statistical methodology. Third, one should not perform independent analyses of CMR^O2and OER because they share common factors in their calculation. Fourth, power analysis indicates, given our CMR^O2variance, that more than 70 animals per group would be required to demonstrate a CMR^O2difference of 10% between groups. Finally, in a study of normothermic CPB in dogs, there was no difference in systemic OER between pulsatile and nonpulsatile flow, even under conditions of markedly reduced perfusion. [39] Thus, greater oxygen extraction with pulsatile flow has not been demonstrated in any other system. Therefore, if CMR^O2differences do exist between steady-state pulsatile and nonpulsatile CPB (and we emphasize that our data do not indicate they do), they are exceedingly small and would require an unacceptably large number of animals to demonstrate.

In a prior study with this preparation, we did not detect a change in either CBF or CMR^O2during 60 min of normothermic nonpulsatile CPB. [11] Consequently, we were surprised to find that, in the current experiment, CMR^O2decreased slightly over the 30-min measurement interval and that CBF may have done the same. In our prior study, power analysis indicated we had sufficient power to detect a change in CBF of 19 ml *symbol* 100 g sup -1 *symbol* min sup -1 and CMR^O2of 1.4 ml *symbol* 100 g sup -1 *symbol* min sup -1. [11] Thus, the decreases in CBF and CMR^O2observed in the current experiment (4 plus/minus 3 ml *symbol* 100 g sup -1 *symbol* min sup -1 and 0.4 plus/minus 0.4 ml Oxygen²*symbol* 100 g sup -1 *symbol* min sup -1, respectively) are quite small and were well below the limits of resolution of our prior study. In the current experiment, OER remained constant over time. This finding indicates changes in CBF and CMR^O2, whatever their cause, were coupled. Although the current experiment cannot provide insight into the mechanisms responsible for decreases in CMR^O2and CBF during CPB, one thing is clear. Because decreases in CMR^O2and CBF were equivalent during both pulsatile and nonpulsatile CPB, lack of pulsatile flow per se was not responsible for the observed changes. Other factors, such as cerebral embolization, must be responsible.

In summary, this study indicates that, at normothermia, nonpulsatile CPB does not appear disadvantageous to the rabbit brain in terms of either bulk blood flow or oxygen metabolism. Although CMR^Osub 2 decreased slightly over time during CPB, lack of arterial pulsation did not appear to be the factor responsible for these changes. This study cannot address potential differences in regional CBF, microvascular perfusion, or cerebrovascular responses to ischemia between pulsatile and nonpulsatile CPB. Therefore, additional studies will be necessary before the impact of pulsatile CPB upon neurologic outcome after cardiac surgery can be definitively established.

The authors thank Michelle Elkins, of Medical Engineering Consultants, for her help in setting up and maintaining the pulsatile perfusion pump.

*Todd MM, Wu B, Warner DS: The roles of adenosine and nitric oxide in the cerebral blood flow effects of hypoxia and hemodilution (abstract). Soc Neurosci Abstr 19:1225, 1993.