| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
J Thorac Cardiovasc Surg 1994;107:1030-1035
© 1994 Mosby, Inc.
CARDIOPULMONARY BYPASS, |
Copenhagen, Denmark
Received for publication June 21, 1993. Accepted for publication Sept. 9, 1993. Address for reprints: Jens Waaben, MD, Consultant Anesthesiology, MBC 22, King Faisal Specialist Hospital and Research Centre, P.O. Box 3354, Riyadh 11211, Kingdom of Saudi Arabia.
Abstract
Microemboli in the brain may inhibit brain function during cardiopulmonary bypass, and in a previous study in pigs of normothermic nonpulsatile bypass we reported a significant decrease in cerebral glucose consumption secondary to interruption of the capillary flow, possibly caused by microemboli. In the present study we measured the regional cerebral glucose consumption and the regional capillary diffusion capacity (that is, the number of perfused capillaries) in 10 different brain regions in two separate groups of animals with and without an arterial filter during normothermic cardiopulmonary bypass. Inclusion of a 40µm arterial filter in the bypass circuit increased the regional brain glucose consumption 27% (median; range -12% to 145%) and regional capillary diffusion capacity increased 123% (median; range 36% to 829%). No change in brain histologic features, the cerebrovascular permeability to serum proteins, or cerebral water content was observed. The arterial filter probably protects the cerebral microcirculation and prevents the decrease in cerebral glucose consumption otherwise seen during bypass. (J THORAC CARDIOVASC SURG 1994;107:1030-5)
During cardiopulmonary bypass (CPB) microembolization occurs into the arterial line from minute bubbles of air, thrombi, and platelet aggregates, as well as pieces of plastic and antifoaming agents. 
1-3 Microembolization is believed to be a main cause of the deterioration of brain function seen postoperatively after otherwise uncomplicated CPB procedures. 4-6 It has been suggested that incorporation of an arterial filter in the CPB circuit may itself provide emboli, but it seems sufficiently documented that filters can be used without additional damage and that filters actually remove a large fraction of the microemboli, with considerable benefits to the patient. 7,8
Microemboli in the brain may inhibit brain function during CPB. 9 In a previous study, this laboratory examined brain microvascular function after 2 hours of nonpulsatile CPB in pigs, using a bubble oxygenator but no arterial line filtration. 10 Cerebral metabolic rate for glucose (CMRglc), cerebral blood flow, and the capillary diffusion capacity (that is, the capillary permeability surface area product, or PS product) were measured and a significant decrease in CMRglc was reported. After normothermic CPB this decrease was paralleled by an equal reduction in cerebral blood flow, whereas the PS product was reduced unequivocally, which suggested that capillary flow may have been severely compromised, perhaps as a result of minute air bubbles, although cerebral blood flow was apparently sufficient.
To examine the ability of a filter to improve capillary diffusion properties we investigated in the present study in pigs the effect of a filter in the arterial line of the CPB circuit. We measured the regional CMRglc and the regional PS product in 10 different brain regions in two separate groups of animals, with and without a filter incorporated in the arterial line. Given identical physiologic parameters in the two groups, we predicted an increase of both regional PS product and regional CMRglc, if the filter actually protected the brain microvasculature. In the same pigs, we also examined brain histologic features, the cerebrovascular permeability to serum proteins, and the cerebral water content, because previous investigations of the brain after normothermic CPB without arterial line filtration have shown hydropic degeneration of the brain parenchyma and astrocyte swelling. 11
MATERIALS AND METHODS
Anesthesia, operation, and CPB
All animals were premedicated with azaperone (Sedaperone) and atropine, and anesthesia was maintained during the whole procedure including during CPB with fentanyl 15 µg/kg/hour, midazolam 500 µg/kg/ hour, pancuronium (Pavulon) 200 µg/kg/hour, and halothane 0.5% to 1.0% inspiratory concentration in pure oxygen. The animals were intubated and the lungs mechanically ventilated to maintain carbon dioxide tension 35 to 45 mm Hg and oxygen tension higher than 100 mm Hg. All animals received animal care in compliance with the principles formulated by the Committee on Animal Experimentations of the Department of Justice (Denmark).
Catheters for blood sampling and monitoring of blood pressures were introduced into the femoral artery and vein and threaded into the aorta and inferior vena cava, respectively. For sampling of cerebral venous blood a catheter was placed in the internal jugular vein and advanced to the base of the skull. Aortic and central venous pressures and the electrocardiogram were monitored continuously, and blood samples were drawn from the arterial and venous lines at regular intervals. Blood gases and pH were measured at 37° C with the Radiometer acid-base laboratory ABL-3 (Radiometer A/S, Copenhagen, Denmark).
Fourteen pigs, weighing between 25 and 39 kg, were randomly divided in two groups. Seven pigs had a 40 µm arterial filter (Polystan EP40; Polystan, Copenhagen, Denmark) in the CPB circuit, whereas another seven pigs had CPB without a filter. In both groups the pump flow was 100 ml/kg/min and the blood/gas flow ratio 1:1. In neither group was mean arterial pressure actively supported with vasoactive drugs, and sodium bicarbonate was not used to correct metabolic acidosis. In all animals, rectal temperature was kept at 37° C.
All animals had a median sternotomy. CPB was established with a single cannula in the right atrium and a 6 mm cannula in the aorta. The extracorporeal circuit included a bubble oxygenator (VT 5000, Polystan) and a nonpulsatile pump (Verticlude, Polystan). The system was primed with lactated Ringer's solution and fresh homologous blood, which also was added during CPB to maintain the hematocrit at 22 to 28. The animals were given an initial heparin dose of 3 mg/kg before CPB and 1 mg/kg was added 1 hour later. At the end of 2 hours of CPB, the animals were killed by discontinuation of the CPB. The brain was removed as quickly as possibly, and samples were taken from the same 10 regions of the right hemisphere in all animals.
Measurements of regional CMRglc and regional PS product
The general principle is the determination of the tracer clearance from plasma, equal to the ratio between the amount of tracer accumulated in the organ (brain) and the integral of tracer supplied to the organ by the circulation. Hence the clearance equals the ratio between the radioactivity of the brain tissue and the time concentration integral of tracer in arterial blood.
We measured the regional PS product with a small polar nonelectrolyte (14C) mannitol that slowly crosses the blood-brain barrier in proportion to the capillary surface area. When the endothelial permeability for the tracer is very low and the observation period is extended, the clearance equals the PS product and hence depends on the number and permeability of perfused capillaries. The unidirectional blood-brain clearance of labeled mannitol was determined using the graphic slope-intercept plot for blood-brain barrier transport:
V = V0 + KT
where V is the volume of distribution of tracer-derived radioactivity in brain, T the normalized time integral of tracer circulation in arterial plasma, and V0 the initial volume of distribution, equal to the plasma volume of the brain tissue sample for a tracer of low permeability. 12,13
The regional CMRglc was measured with the glucose analog 2-deoxyglucose (2(3H)-deoxyglucose), which is phosphorylated by brain hexokinase and retained in the tissue. If the duration of the experiment is sufficiently long, the clearance of 2-deoxyglucose equals the glucose phosphorylation rate divided by a lumped constant, defined as the rate of deoxyglucose phosphorylation relative to that of glucose. The deoxyglucose phosphorylation reflects the glycolytic phase of glucose breakdown. Glucose metabolic rate was determined also with the slope-intercept plot modified for estimation of net 2-deoxyglucose clearance 12-14:
V = KT + Vg(1 – e -kT)
where V is the volume of distribution of all tracer-derived radioactivity in the brain and Vg the apparent distribution volume of unphosphorylated deoxyglucose in the brain. According to this equation, V asymptomatically approaches a straight line of slope K and ordinate intercept Vg with a time constant of 1/k. Because the three coefficients K, Vg, and k are all functions of CMRglc and the lumped constant L, CMRglc and L were estimated by nonlinear regression of the operational equation, as described by Kuwabara, Evans, and Gjedde. 15
Procedures
The tracer circulation time differed for the animals within each group (Fig. 1). In each animal, labeled 2-deoxyglucose and mannitol were injected simultaneously, and the tracers were circulated for 5, 10, 15, 20, 25, 30, or 60 minutes, for each respective experiment, before corresponding values of V and T were determined. Total CPB time was approximately 2 hours in all animals. To minimize differences in the length of CPB before isotope injection, the isotopes were injected according to the following scheme: first animal after 117.5 minutes of CPB, second animal after 115 minutes, third animal after 112.5 minutes, fourth animal after 110 minutes, fifth animal after 107.5 minutes, sixth animal after 105 minutes, and seventh animal after 90 minutes of CPB (Fig. 1). After tracer injection arterial blood was sampled every 10 seconds for the first minute and then after 1.5, 2, 3, 4, 5, 7.5, 10, 15, 20, 25, 30, and 60 minutes, depending on the circulation time of the tracer. The blood samples were rapidly centrifuged and the plasma prepared for detection of radioactivity and determination of the arterial plasma glucose concentration.
|
11 The brains were sectioned in the coronal plane. The rostral 5 mm thick coronal block was immediately submerged in kerosene, whereas the caudal block was immersed in paraformaldehyde solution for subsequent histologic and immunocytochemical staining. The 4 µm thick sections were stained with hematoxylin and eosin, van Gieson's, cresyl violet, or Klüver-Barrera stain.
Sections adjacent to those stained for histologic study were processed for immunocytochemical demonstration of serum proteins by a peroxidase-antiperoxidase method 16 with the following steps (in short): (1) deparaffination, (2) 1 hour of incubation in goat serum, (3) overnight incubation in rabbit anti-pig serum, (4) 30 minutes of incubation in goat anti-rabbit immunoglobulin G, (5) incubation for 30 minutes in peroxidase-antiperoxidase-rabbit complex, and (6) immersing the sections in hydrogen peroxide and 3.3-diaminobenzidine tetrahydrochloride.
Dual samples 1 mm 3 in size were dissected from 10 different brain regions, and the specific gravity was determined in a monobromobenzene-kerosene gradient column according to the procedure of Nelson, Mantz, and Maxwell 17 and Fujiwara and associates. 18
Statistics
The physiologic parameters were compared by the unpaired Mann-Whitney significance test, and values of regional CMRglc, regional PS product, and specific gravity were compared by the Wilcoxon matched-pairs signed-rank test. A p value less than 0.05 was considered statistically significant.
RESULTS
Physiologic variables
The physiologic data are summarized in Table I. No significant differences were noted between the two CPB groups. Mean arterial pressure levels were similar in the two experimental groups (74 ± 12 mm Hg versus 71 ± 11 mm Hg).
|
Tables II and III. In the presence of an arterial filter, CMRglc was higher in most regions. The statistical analysis disclosed a significant difference between the two groups (p < 0.01). The use of an arterial filter increased the regional CMRglc 27% (median; range -12% to 145%). The highest values were determined for cortical areas, whereas lower values prevailed in the white matter regions (brain stem and callosal body). The largest increase of regional CMRglc after insertion of an arterial filter occurred in the colliculi superior and inferior, whereas more moderate increases occurred in cortical areas. In the white matter, regional CMRglc values did not change.
|
|
Histologic features, blood-brain barrier, and specific gravity
Macroscopically, the brain surfaces and coronal sections looked normal. Microscopically, swelling of the astrocyte end-feet was noted around blood vessels in discrete areas in both groups, and slight vacuolization was observed also in discrete areas in the white matter of two animals in the nonfilter group and in one animal in the filter group. In both groups tiny perivascular mononuclear infiltrates were seen in a few animals, which probably was not related to the CPB procedure per se, but rather to the general manipulation. Dark neurons caused by the immersion fixation procedure were seen occasionally in all animals. Microembolization, demyelination, widespread cell infiltration, and hemorrhages did not occur. The meninges and ventricles were normal in both groups.
In both CPB groups, the immunocytochemical staining for serum proteins revealed specific staining confined to the pia, choroid plexus, and the regions normally devoid of a blood-brain barrier. In both CPB groups brown peroxidase staining occurred also in the lumen of the vessels except for a very few perivascular effluxes, indicating normal cerebrovascular permeability to serum proteins in both CPB groups.
As seen in Table IV no difference in specific gravity was seen between the two groups in any of the 10 brain regions examined (p < 0.05).
|
Previously we reported that cerebral glucose utilization was decreased during normothermic nonpulsatile CPB and that this decrease may be the result of interruption of blood flow through part of the capillary bed. 10 The present results support this theory and indicatethat inclusion of a 40 µm arterial filter in the CPB circuit prevents the decrease in regional CMRglc. This is probably the result of removal of minute bubbles and platelet microaggregates mainly formed in the bubble oxygenator, which thus reduces the number of microembolic events. 2,4 In the present study, the largest increase in regional CMRglc was observed in the cortical regions and colliculi, that is, the regions that have been reported to sustain the largest decrease in CMR glc after CPB without a filter. 19
When compared with previous results, 10 the regional values of CMRglc in the present study are generally lower, probably because of deeper general anesthesia. In previous studies the animals were anesthetized only with halothane during CPB, 10,19 whereas in the present investigation halothane was supplemented with fentanyl and midazolam. However, the distribution of regional CMRglc values among the 10 brain regions in the present study is consistent with that in previous data. 19
Usually blood does not flow at a continuous rate through the capillaries, but rather flows intermittently. 20 This intermittency has been ascribed to vasomotion, by which is meant that metarterioles and precapillary sphincters constrict and relax in an alternating cycle (5 to 10 times per minute). Blood flow to local tissues is autoregulated by this opening and closing. Adenosine and the local oxygen concentration have been proposed as the main regulatory factors of the intermittency. When the PS product is measured by mannitol, blood in each capillary (containing the slowly diffusing mannitol) is turned over sufficiently to keep a high gradient of mannitol between the capillary and the brain tissue. During CPB, basic homeostasis is altered. Vasomotion may be impaired, that is, some capillaries may be always open, while other capillaries are always closed. Total flow through the capillary bed may be the same, but the calculated PS product, and the number of perfused capillaries, decreases because of a reduced mannitol gradient between the inside and the outside of the capillaries. Oxygen has to travel farther to supply all parts of the brain, and small regions may not have enough oxygen to maintain normal cell function. The present results indicate that this unfavorable effect of CPB may be prevented by the inclusion of an arterial filter. We hypothesize that the filter preserves the microvascular integrity by keeping more capillaries open, thus ensuring an even substrate supply to the specific parts of the brain tissue.
Several studies indicate that microembolization plays a major role in the deterioration of mental function known to accompany CPB. Retinal fluoroscein angiography demonstrated patchy occlusions of the microcirculation during CPB, and retinal histologic study revealed intravascular platelet fibrin microaggregates and focal ischemic changes. 21 The microembolic pattern suggested a dynamic state with new occlusions forming and previous occlusions resolving. The rapid resolution was regarded as equally consistent with platelet microaggregates or gaseous microbubble occlusions, because previous occlusions have been shown to resolve without vessel damage within 20 minutes of experimental embolism with platelet aggregates. 22 Focal extravascular leakage of fluorescein indicative of endothelial cell damage with breakdown of the blood-retinal barrier may occur with experimental embolism of particulate material or gas. 23
The brain may differ from the retina in its susceptibility to embolic ischemia. The cerebrum has arteriolar collaterals, and experimental cerebral microembolism produces less ischemic injury than might be expected. 24 This may be the reason for not finding light microscopic ischemic lesions in dog brains with multiple retinal ischemic lesions. 21 This may also explain why the present study failed to reveal cerebral microembolization or signs of breakdown of the blood-brain barrier, despite the effect of arterial line filtration on regional CMRglc and PS product, which strongly suggested the presence of multiple microemboli during CPB. It is not clear whether histologically apparent cerebral ischemia is a prerequisite for the functional injury detectable postoperatively by psychometric testing in human beings. In the present study we did find hydropic degeneration and perivascular swelling of astrocytic end-feet, but without a significant difference between the two CPB groups.
Åberg and associates 25 reported increased release of the brain enzyme adenylate kinase into the cerebrospinal fluid during CPB when no arterial filtration was used and documented its relationship to intellectual deterioration. They concluded that cerebral ischemic damage after CPB was the result of microembolization. Taylor and associates 26 reported a significant reduction of levels of creatine kinase and creatine kinase B isoenzyme in the cerebrospinal fluid if an arterial filter was used during CPB in dogs.
Because microembolization is a significant cause of cerebral impairment after CPB, the question remains whether arterial line filtration really improves this situation. Carlson and associates 27 reported improved postoperative visual motor test scores when a filter was added to the CPB circuit, and Åberg and associates 25 and Treasure 8 showed that arterial line filtration is useful in decreasing the neuropsychologic dysfunction seen after CPB.
In conclusion, previous investigations have shown a favorable effect of an arterial filter in the CPB circuit. 4,5,8,25-27 The present results support this conclusion by providing the physiologic evidence that inclusion of a 40 µm filter into the arterial line of a CPB circuit, which includes a bubble oxygenator, actually protects the cerebral microcirculation and prevents the decrease in regional CMRglc otherwise seen during bypass. The filter probably reduces the number of microembolic lesions, ensuring a more even substrate supply to the specific parts of the brain tissue.
Footnotes
From the Institute of Experimental Pathology, Rigshospitalet; the Department of Anaesthesia, Rigshospitalet, a the Department of Cardiothoracic Surgery, Rigshospitalet, b the Institute of Neuropathology, Rigshospitalet, c and the Institute of General Physiology and Biophysics, The Panum Institute d; University of Copenhagen, Copenhagen, Denmark. 
Read at the Fifteenth Annual Meeting of the Society of Cardiovascular Anesthesiologists, San Diego, Calif., April 24-28, 1993.
References
This article has been cited by other articles:
|
|
 |
|
  M. Cavaglia, S. G. Seshadri, J. E. Marchand, C. L. Ochocki, R. B. B. Mee, and P. M. Bokesch Increased Transcription Factor Expression and Permeability of the Blood Brain Barrier Associated With Cardiopulmonary Bypass in Lambs Ann. Thorac. Surg., October 1, 2004; 78(4): 1418 - 1425. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Schoenburg, B. Kraus, A. Muehling, U. Taborski, H. Hofmann, G. Erhardt, S. Hein, M. Roth, P. R. Vogt, G. F. Karliczek, et al. The dynamic air bubble trap reduces cerebral microembolism during cardiopulmonary bypass J. Thorac. Cardiovasc. Surg., November 1, 2003; 126(5): 1455 - 1460. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. M. Murkin Central Nervous System Complications in Cardiac Surgery: Retrograde Cerebral Perfusion, Pressure, Pulsatility, Temperature, and pH Management During Cardiopulmonary Bypass Seminars in Cardiothoracic and Vascular Anesthesia, July 1, 2000; 4(2): 65 - 69. [Abstract] [PDF]
|
|
|
|
 |
|
 W. T. Mahle, R. R. Clancy, E. M. Moss, M. Gerdes, D. R. Jobes, and G. Wernovsky Neurodevelopmental Outcome and Lifestyle Assessment in School-Aged and Adolescent Children With Hypoplastic Left Heart Syndrome Pediatrics, May 1, 2000; 105(5): 1082 - 1089. [Abstract] [Full Text]
|
|
|
|
|
|
J. M. Murkin The Role of CPB Management in Neurobehavioral Outcomes After Cardiac Surgery Ann. Thorac. Surg., May 1, 1995; 59(5): 1308 - 1311. [Abstract] [Full Text]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 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 |
