Asian Cardiovasc Thorac Ann 2006;14:273-278
© 2006 Asia Publishing EXchange Ltd
Cannulation and Cardiopulmonary Bypass Produce Selective Brain Lesions in Pigs
Zsolt Tóth, PhD,
Iván Györimolnár, MD1,
Hajnalka Ábrahám, MD2,
Ákos Hevesi, DVM3
Division of Cardiac Surgery
1 Cardiac Care Unit, Heart Institute
2 Central Electron Microscopic Laboratory, Faculty of Medicine, University of Pécs
3 Institute of Diagnostic Imaging and Radiation Oncology Faculty of Animal Science, University of Kaposvár Hungary
For reprint information contact: Zsolt Tóth, PhD Tel: 36 72 536 390 Fax: 36 72 536 399 Email: zsolt.toth{at}aok.pte.hu, Division of Cardiac Surgery, Heart Institute, Faculty of Medicine, University of Pécs, 7624 Pécs, Ifjuság út 13, Hungary.
 |
ABSTRACT
|
|---|
Whether cardiopulmonary bypass alone or together with manipulation of the aorta produces neurologic complications remains controversial. Using a pig model, the immediate effects of aortic cannulation and cardiopulmonary bypass on neural injury in different brain regions were investigated in 3 experimental groups: non-operated controls; operated controls with aortic cannulation without cardiopulmonary bypass; and operated animals undergoing cardiopulmonary bypass. Immunohistochemistry using a monoclonal antibody against calretinin was used to show possible ischemic damage in the hippocampal formation which is one of the most vulnerable regions to ischemia. Both cannulation of the aorta alone and cardiopulmonary bypass resulted in numerous argyrophilic neurons in discrete regions of the prefrontal and cerebellar cortex. Decreased calretinin immunoreaction and a reduced number of calretinin-positive neurons were observed following aortic cannulation or cardiopulmonary bypass compared to the non-operated controls. This suggests that both cannulation of the aorta alone and cardiopulmonary bypass affect a selected population of neurons. Therefore, off-pump, aorta no-touch technique may prevent neurologic complications.
|
INTRODUCTION
|
|---|
Neurologic injury can be a devastating complication of cardiac surgery, resulting in longer hospitalization, increased costs, and higher mortality rates. The manifestations of such injury range from neurocognitive dysfunction to stroke.1 Strategies aimed at reducing neurologic injury have focused mainly on the technical aspects of cardiopulmonary bypass (CPB).2 Potential mechanisms include macroembolization of air or particulate matter; microembolization of gas, fat, aggregates of blood cells, platelets or fibrin, particles of silicone or polyvinyl chloride tubing; and inadequate cerebral perfusion pressure.2,3 Embolism causes transient ischemia, but it may also lead to infarction. Neurologic disturbances after CPB are associated with an ischemic period of hyperexcitability that is the result of glutamate accumulation as well as production of the possible neurotoxin nitric oxide.4,5 The resultant neuronal hyperactivity induces a cascade of cellular events causing necrosis and apoptosis, leading to acute or delayed cell death in selected areas of the brain, including the hippocampus, basal ganglia, and cerebellum.4–6 A previous study showed that off-pump surgery reduced the inflammatory response and perioperative release of markers of neuronal damage.7 Thus, off-pump operations reduce adverse neurologic outcomes compared with on-pump procedures.7–14 It has been reported that aortic manipulation does not significantly influence the neurologic outcome in off-pump patients.11 This disagrees with a retrospective study of a large number of patients in whom aortic manipulation was found to be an independent risk factor for cerebrovascular accident.15
Using an animal model, we investigated the acute effects of cannulation of the aorta alone and CPB on possible neuronal injury in the prefrontal neocortex, the cortical layers of the cerebellum, and the hippocampal formation. Prefrontal cortex was chosen because earlier studies suggested that neuronal damage of this region may be responsible for the cognitive deficits in patients after CPB.1,4 More than 90% of the clinical spectrum of cerebral complications are psychological or cognitive deficits and not major neurological complications.1 The cerebellum and the hippocampal formation are the regions most sensitive to ischemia, and damage to both causes deficits in memory.16,17 The Purkinje cells of the cerebellum and the pyramidal cells of the CA1 area of the hippocampus are selectively vulnerable to decreased cerebral blood flow, both in human pathology and in experimental animals. In addition to death of the CA1 pyramidal cells, acute degeneration of calretinin-immunoreactive neurons of the hilus of the hippocampal dentate gyrus was observed in the rodent model of global ischemia.18 The goals of this study were: to investigate whether acute neuronal injury was induced by cannulation of the aorta alone and combined with CPB; to define the distribution of vulnerable neurons in the neo- and archicortex following CPB; and to determine whether neuronal injury or acute cell death were among the consequences of ischemia that may occur during the operation itself.
|
MATERIALS AND METHODS
|
|---|
All animals received humane care in compliance with the guidelines ‘Principles of Laboratory Animal Care’ formulated by the National Society for Medical Research and the ‘Guide for the Care and Use of Laboratory Animals’ published by the National Institute of Health (NIH Publication No. 88-23, revised 1985). An animal paradigm for CPB was used as described previously.19 Briefly, 18 male and female neurologically mature young domestic pigs (commercial farm, 20 to 30 kg, 89–100 days old) were used. Pre-anesthesia was induced with xylazine 2.2 mg·kg–1, ketamine 20 mg·kg–1, and atropine 0.03 mg·kg–1 intramuscularly. After endotracheal intubation, the animal was ventilated mechanically with 60% oxygen and 40% nitrogen. The ventilator rate and tidal volume were adjusted to maintain the arterial carbon dioxide level. Anesthesia was maintained with 1.0% to 2.0% isoflurane. A temperature probe was placed in the esophagus to monitor the core temperature. Catheters were placed in the left femoral artery and vein for withdrawal of blood samples and measurement of blood pressure. An ear vein was used for infusion and drug support.
The chest was opened via a median sternotomy. Heparin was given at 300 IU·kg–1 intravenously. A cannula (size 18) was inserted into the ascending aorta for arterial blood return to the body during CPB. Venous cannulas were placed into both the superior (size 26) and inferior (size 28) venae cavae. The CPB circuit consisted of Pemco roller pumps (Pemco, Inc., Cleveland, OH, USA), a cardiotomy reservoir (Minimax 1316, filtered hardshell reservoir), arterial filter (Capiox), water bath (Hemotherm), and a membrane oxygenator (Minimax Plus 3381, hollow fiber oxygenator with plasma resistant fiber) with integral heat exchanger. The circuit was primed with 500 mL lactated Ringer’s solution, 1,000 mL homologous blood, and 5,000 IU heparin. Sodium bicarbonate was given to maintain the pH at 7.40. Blood electrolytes and osmolality were monitored with a NOVA analyzer (Biomedica Hungary, Budapest, Hungary) and maintained within the normal range. The hematocrit was kept at 23% to 30% (the normal pig hematocrit is approximately 30%) during CPB. The overall strategy was to compare neuronal injury in 3 experimental groups: non-operated controls (n = 3), operated controls with cannulation of the great vessels without CPB (n = 5), and CPB for 30 min (n = 10).
For all histological studies, control and experimental animals were processed in a blinded fashion. At the end of the experiment, animals under deep anesthesia were perfused transcardially with saline followed by 4% paraformaldehyde in 0.1 M phosphate buffer (pH 7.4). Brains were removed and post-fixed for a week in the same fixative. The following regions were chosen for histological experiments: the cerebellum, prefrontal cortex, and the archicortical hippocampal formation. From the cerebellum, tissue slices were cut at the middle parasagittal level of the right hemisphere at the primary fissure. From the hippocampal formation a complete cross-section was cut at the mid septotemporal level, thus the dentate gyrus, all subregions of Ammon’s horn, subiculum, and parasubiculum were in the section. From the prefrontal cortex, coronal tissue slices were cut from the right hemisphere across the superior arcuate sulcus. The sections were cut from the prefrontal brain region that approximately corresponds to the superior frontal gyrus in humans. The tissue slices were approximately 10 mm thick.
The tissue blocks were sectioned with a Vibratome (Technical Products, Inc., St Louis, MO, USA) at 60 µm for silver staining and calretinin immunohistochemistry, and at 10 µm for Nissl staining. Gallyas silver staining was carried out for "dark" neurons. After dehydration in 50%, 75%, and 100% propanol, sections were esterified with propanol containing 0.8% sulfuric acid. Sections were rehydrated with 50% and 25% isopropyl and distilled water, treated with 3% acetic acid for 5 min, and developed for about 10 min. The fresh developing solution consisted of equal volumes of 10% Na2CO3 and a solution composed of 0.2% AgNO3, 0.25% NH4NO3, 2% tungstosilicic acid, and 0.4% formaldehyde.20 When developed, sections were dehydrated and cover-slipped. The number of pyknotic cells was counted in all 3 regions. In each animal and each region 1,000 cells, including the pyknotic cells, were counted. For immunohistochemistry, sections were processed immediately after cutting. After washing 3 times for 15 min each in 0.05 M Tris buffer (pH 7.6), the free-floating sections were incubated in normal horse serum (1%) in Tris buffer for 1 hr at room temperature. The primary monoclonal antibody against calretinin (1:2500) containing 0.4% Triton X-100 (Swant, Bellinzona, Switzerland) was then added.
After incubation with the primary antibody at 4°C on a shaker for 3 days, sections were washed 3 times for 15 min each with Tris buffer and then incubated with a biotinylated pan-specific universal secondary antiserum (Vector Laboratories, Burlingame, CA, USA) diluted 1:50 in Tris buffer, for 2 hr at room temperature. The sections were washed with Tris buffer 3 times for 15 min and incubated with avidin-biotin-peroxidase complex (Vector Laboratories; 1:50) for 2 hr at room temperature. After further washing in Tris buffer, the reaction product was visualized with 3,3'-diaminobenzidine (DAB) as the chromogen (10 mg DAB in 25 mL Tris buffer (pH 7.4) with 25 µL 3% H2O2 added just before use). Progress of the reaction was monitored using a microscope, and the reaction was stopped with Tris buffer (pH 7.6). After a final wash in Tris buffer, sections were mounted on slides and air-dried overnight. One third of the sections were counterstained with 1% cresyl violet according to the Nissl method, to reveal the general cytoarchitecture. All sections were dehydrated in alcohol, cleared in xylene, and covered with DPX (Mountant for Histology, Fluka, Switzerland). Specificity of the antisera was controlled by the producers. In addition, in sections processed with omission of the primary antiserum, no staining was detected.
|
RESULTS
|
|---|
Aortic cannulation alone and CPB lead to little immediate cell death, but there was neuronal injury in discrete prefrontal, cerebellar, and hippocampal regions. In all histological preparations the number of pyknotic cell nuclei was counted in the examined areas. There were very few pyknotic cells (< 0.1%) both in the neo- and archicortex in the experimental and control groups. Silver-stained neurons were found in sections from animals sacrificed 1 hr after aortic cannulation alone and those undergoing CPB. In the prefrontal cortex, argyrophilic neurons occurred in all layers but were numerous in the uppermost layers (Figure 1B
). In the cerebellar cortex (Figure 1D), mainly Purkinje cells appeared to be affected. In the hippocampus, the dentate gyrus and the CA1-3 areas of Ammon’s horn did not contain silver-stained neurons. However, in animals undergoing CPB, the granule cell layer contained argyrophilic neurons (Figure 1F). No silver-stained neurons were visible (Figure 1A, 1C, 1E) in sections from the non-operated control group. In cresyl violet stained sections, a few neurons appeared hyperbasophilic and shrunken and they were scattered among normal-appearing neurons. Darkly stained and normal-appearing neurons were often found adjacent to each other. In the non-operated control group the distribution and number of calretinin-immunoreactive (ir) cells and fibers in the dentate gyrus (Figure 2A, 2B) were similar to that described in a separate study. In the outer one-third of the molecular layer of the dentate gyrus, and in the stratum lacunosum-moleculare of Ammon’s horn, large numbers of bipolar calretinin-ir neurons were identified as Cajal-Retzius cells. The inner half of the molecular layer and the hilus of the dentate gyrus contained calretinin-positive local circuit neurons, while in the supragranular layer, a dense axonal plexus was observed (Figure 2A, 2B). Similarly to other species, the origin of these axons is probably the neurons of the supramammillary nucleus. Following cannulation of the aorta alone, the number of calretinin-ir interneurons in the hilus of the dentate gyrus decreased, and only a few cells per section could be found (Figure 2C, 2D). Similarly, the calretinin-ir axonal band from the supramammillary nucleus was much thinner and stained weaker (Figure 2C, 2D). In contrast, the calretinin-positive Cajal-Retzius cells were preserved in the outer molecular layer of the dentate gyrus and in the stratum lacunosum-moleculare of Ammon’s horn (Figure 2C, 2D). Cardiopulmonary bypass resulted in complete disappearance of the calretinin-ir hilar cells and the supragranular axonal plexus (Figure 2E, 2F). The calretinin-ir Cajal-Retzius cells in the outer molecular layer of the dentate gyrus and in the stratum lacunosum-moleculare of Ammon’s horn were still preserved (Figure 2E).
View larger version (123K):
[in this window]
[in a new window]
|
Figure 1. Photomicrographs showing Gallyas silver stained neurons in sections of the prefrontal cortex, cerebellum, and hippocampus of pigs. (A) There are no argyrophilic neurons evident in the prefrontal cortex of non-operated controls. (B) Silver-stained neurons, probably pyramidal cells, in layer II of the prefrontal cortex in an animal sacrificed 1 hr after aorta cannulation alone. (C) No silver-stained neurons can be seen in the cerebellum of a non-operated control animal, but silver-stained Purkinje cells somata (arrow) and their dendrites (arrowheads) in the Purkinje cell layer (P) and molecular layer (ml) of the cerebellum are found in a pig subjected to CPB. (D) Similar staining was observed in the cerebellum of animals sacrificed 1 hr after aorta cannulation alone. No silver-stained neurons were observed in the granule cell layer (gl) of the cerebellum after either aorta cannulation alone or CPB. (E) In the hippocampus, silver-stained neurons were not found in the non-operated controls, but a large number were evident in the granule cell layer (g) of the dentate gyrus in animals subjected to CPB. (F) The arrow points to a soma of a granule cell, and arrowheads point to the dendrites extending to the molecular layer (m). Histological results were similar in the hippocampus of animals sacrificed 1 hr after aortic cannulation alone. p = pyramidal cell layer, r = stratum radiatum, l-m = lacunosum-moleculare, h = hilus. Bar = 50 µm in A, B, D, F; 200 µm in C and E.
|
|
View larger version (201K):
[in this window]
[in a new window]
|
Figure 2. Photomicrographs showing the distribution of calretinin-ir neurons and fibers in the hippocampal dentate gyrus of non-operated controls (A); boxed area is shown at higher magnification (B). Arrows point to calretinin-ir interneurons in the hilus (h), and arrowheads show calretinin-ir interneurons in the molecular layer (m) of the dentate gyrus. A large population of calretinin-positive bipolar Cajal-Retzius cells can be observed in the outer half of the molecular layer. On the outer surface of the granule cell layer (g) of the dentate gyrus, a dense calretinin-ir axonal plexus can be seen. (C) Low-magnification photomicrograph showing the calretinin-ir cells and fibers after cannulation of the aorta alone; boxed area is shown with high power (D). Arrows point to the calretinin-ir cells in the hilus (h), arrowheads show calretinin-positive interneurons in the molecular layer (m) of the dentate gyrus. Note the pale staining of the calretinin-ir axonal plexus above the granule cell layer (g). (E) Photomicrograph showing calretinin-ir cells in the dentate gyrus after CPB; boxed area is shown at higher magnification (F). Note that the hilus (h) lacks calretinin-ir cells, while the arrowhead points to a calretinin-ir interneuron in the molecular layer (m) of the dentate gyrus. No calretinin-positive fiber bundle is seen above the granule cell layer (g).
|
|
|
DISCUSSION
|
|---|
The principal findings of this study are that both cannulation of aorta alone and CPB cause injury to selected neuronal populations in the hippocampus, prefrontal, and cerebellar cortex. This is the first report showing the locations and distribution of injured neurons following CPB. Ischemia-induced neuronal injury following CPB was more severe than after cannulation of the aorta alone. The pattern of distribution of the argyrophilic neurons, taken together with results of previous studies, gives information on the size and location of neuronal injury induced by CPB. The distribution of injured neurons may provide information about brain regions where pyramidal neurons may be vulnerable to CPB in this model. The distribution pattern of argyrophilic neurons following CPB may reveal selective vulnerability in distinct brain areas that needs further study.
Our finding of a fast disappearance of calretinin-immunoreactivity following CPB correlates with other results showing early generation of argyrophilic "dark" neurons in different brain regions. The vulnerability of the hilar calretinin-containing interneurons to hypoxia or ischemia is known.18 In addition, calretinin- immunoreactivity rapidly disappears in the hilus following the termination of cerebral blood circulation caused by either death of the animal or occlusion of the blood vessels supplying the hippocampal formation. Similarly, surgically removed, but not immediately fixed, human hippocampi also show a significant decrease in calretinin-immunoreactivity. Calretinin staining of hippocampal neurons of domestic pig brains harvested from the slaughter house shows that hypoxia and postmortem delay markedly decreases the immunoreactivity of calretinin-positive hilar neurons and that of the supragranular axonal plexus. However, the quality of calretinin-staining of the Cajal-Retzius-type cells along the hippocampal fissure was not changed. The markedly decreased calretinin-immunostaining following cannulation of the aorta alone may be caused by transient ischemia. Transient ischemia might also be caused by debris released from the wall of the aorta, but the aortas of our experimental animals lacked macroscopically visible sclerotic plaques.1,8,9
In our experiments, CPB resulted in more severe and extended neuronal damage. This correlates with previously published data showing that off-pump surgery reduces the inflammatory response and perioperative release of markers of neuronal damage.7 We demonstrated that in pigs subjected to CPB, the calretinin-ir hilar cells and the supragranular calretinin-positive fiber bundle completely disappeared. The resistance to ischemia/hypoxia of the bipolar calretinin-ir Cajal-Retzius cells may be explained by their more immature morphology and the distinct connection, neurochemical content, and function of this cell type, as demonstrated in both rats and domestic pigs.
Our study provides a histological analysis of neuronal injury induced by CPB in an animal model of extracorporeal perfusion. Cannulation of the aorta alone produced similar although more moderate effects on cortical neurons. Therefore, strategies aimed at reducing emboli in the cerebral circulation, maintaining cerebral oxygenation, and minimization of the whole-body inflammatory response to the bypass circuit should reduce the unwanted side effects. On the other hand, it can be supposed that off-pump surgery, avoiding CPB and applying the "aorta no-touch" technique, may greatly improve the neurological outcome of patients undergoing cardiac surgery.8,9,12,13,15
The widespread ischemia-induced selective neuronal injury in the prefrontal cortex may lead to significant alterations in function. Since it is known that the prefrontal cortex plays an important role in cognitive function and mood disorders, our findings may correlate with the cognitive changes often observed after cardiac operations in patients. In clinical practice, neuropsychological deficits occur in 60% to 80% of patients in the first week, and in 20% to 40% at 8 weeks postoperatively.1 Therefore, further experimental studies are needed to describe the immediate and long-lasting pathological effects that may occur after cardiac operations. Most importantly, clinical trials are needed to find evidence that connect cortical areas with the observed neurocognitive disturbances in patients subjected to cardiac operations with or without CPB.
|
ACKNOWLEDGMENTS
|
|---|
The authors wish to thank Prof. László Seress for critically reading and correcting the manuscript and for his help in the histological procedures. The technical assistance of György Wéber, Borbála Szabó, Áron Sztaniszláv, and János Fülöp is appreciated.
This work was supported by the Hungarian Science Fund (OTKA) T035255 and FKFP 0026/2001 to Zsolt Tóth.
|
REFERENCES
|
|---|
- Taylor KM. Brain damage during cardiopulmonary bypass. Ann Thorac Surg 1998;65(4 Suppl):S20–6.[Medline]
- Mackensen GB, Sato Y, Nellgard B, Pineda J, Newman MF, Warner DS, et al. Cardiopulmonary bypass induces neurologic and neurocognitive dysfunction in the rat. Anesthesiology 2001;95:1485–91.[Medline]
- Taylor KM. Central nervous system effects of cardiopulmonary bypass. Ann Thorac Surg 1998;66(5 Suppl):S20–4.[Abstract/Free Full Text]
- Baumgartner WA, Walinsky PL, Salazar JD, Tseng EE, Brock MV, Doty JR, et al. Assessing the impact of cerebral injury after cardiac surgery: will determining the mechanism reduce this injury? Ann Thorac Surg 1999;67:1871–3.[Abstract/Free Full Text]
- Tseng EE, Brock MV, Lange MS, Blue ME, Troncoso JC, Kwon CC, et al. Neuronal nitric oxide synthase inhibition reduces neuronal apoptosis after hypothermic circulatory arrest. Ann Thorac Surg 1997;64:1639–47.[Abstract/Free Full Text]
- Hagl C, Tatton NA, Khaladj N, Zhang N, Nandor S, Insolia S, et al. Involvement of apoptosis in neurological injury after hypothermic circulatory arrest: a new target for therapeutic intervention? Ann Thorac Surg 2001;72:1457–64.[Abstract/Free Full Text]
- Mazzone A, Gianetti J, Picano E, Bevilacqua S, Zucchelli G, Biagini A, et al. Correlation between inflammatory response and markers of neuronal damage in coronary revascularization with and without cardiopulmonary bypass. Perfusion 2003;18:3–8.[Free Full Text]
- Kobayashi J, Sasako Y, Bando K, Niwaya K, Tagusari O, Nakajima H, et al. Multiple off-pump coronary revascularization with "aorta no-touch" technique using composite and sequential methods. Heart Surg Forum 2002;5:114–8.[Medline]
- Bittner HB, Savitt MA, Ching PV, Ward HB. Off-pump coronary artery revascularization: ideal indication for patients with porcelain aorta and calcification of great vessels. J Cardiovasc Surg (Torino) 2003;44:217–21.[Medline]
- Chavanon O, Durand M, Hacini R, Bouvaist H, Noirclerc M, Ayad T, et al. Coronary artery bypass grafting with left internal mammary artery and right gastroepiploic artery, with and without bypass. Ann Thorac Surg 2002;73:499–504.[Abstract/Free Full Text]
- Patel NC, Deodhar AP, Grayson AD, Pullan DM, Keenan DJ, Hasan R, et al. Neurological outcomes in coronary surgery: independent effect of avoiding cardiopulmonary bypass. Ann Thorac Surg 2002;74:400–6.[Abstract/Free Full Text]
- Patel NC, Pullan DM, Fabri BM. Does off-pump total arterial revascularization without aortic manipulation influence neurological outcome? A study of 226 consecutive, unselected cases. Heart Surg Forum 2002;5:28–32.[Medline]
- Al-Ruzzeh S, Ambler G, Asimakopoulos G, Omar RZ, Hasan R, Fabri B, et al. Off-Pump Coronary Artery Bypass (OPCAB) surgery reduces risk-stratified morbidity and mortality: a United Kingdom Multi-Center Comparative Analysis of Early Clinical Outcome. Circulation 2003;108 Suppl I):II1–8.
- Yacoub M. Off-pump coronary bypass surgery: in search of an identity. Circulation 2001;104:1743–5.[Free Full Text]
- Calafiore AM, Di Mauro M, Teodori G, Di Giammarco G, Cirmeni S, Contini M, et al. Impact of aortic manipulation on incidence of cerebrovascular accidents after surgical myocardial revascularization. Ann Thorac Surg 2002;73:1387–93.[Abstract/Free Full Text]
- Cervos-Navarro J, Diemer NH. Selective vulnerability in brain hypoxia. Crit Rev Neurobiol 1991;6:149–82.[Medline]
- Schmidt-Kastner R, Freund TF. Selective vulnerability of the hippocampus in brain ischemia. Neuroscience 1991;40:599–636.[Medline]
- Freund TF, Magloczky Z. Early degeneration of calretinin-containing neurons in the rat hippocampus after ischemia. Neuroscience 1993;56:581–96.[Medline]
- Ye J, Yang L, Del Bigio MR, Filgueiras CL, Ede M, Summers R, et al. Neuronal damage after hypothermic circulatory arrest and retrograde cerebral perfusion in the pig. Ann Thorac Surg 1996;61:1316–22.[Abstract/Free Full Text]
- Gallyas F, Zoltay G, Dames W. Formation of "dark" (argyrophilic) neurons of various origins proceeds with a common mechanism of biophysical nature (a novel hypothesis). Acta Neuropathol (Berl) 1992;83:504–9.[Medline]
TOP
ABSTRACT
INTRODUCTION
