Protecting the brain and spinal cord
1987; Elsevier BV; Volume: 5; Issue: 3 Linguagem: Inglês
10.1016/0741-5214(87)90087-5
ISSN1097-6809
Autores Tópico(s)Aortic aneurysm repair treatments
ResumoNeural tissue, specifically the brain and spinal cord, is particularly sensitive to ischemia. Although brief periods of ischemia might result in only a transient neurologic dysfunction, more prolonged lack of oxygen will result in irreversible neuronal damage and permanent neurologic deficit. Cerebral ischemia, such as that occurring during some cases of carotid endarterectomy, can be readily prevented by the use of intraluminal shunts during the time of arterial occlusion. However, other causes of cerebral ischemia, such as those occurring during cardiac arrest, hypoxemia, or severe hypotension, are not readily prevented. Similarly, spinal cord ischemia, which can occur during repair of thoracoabdominal aneurysms or dissecting aortic aneurysms, can result in paraplegia. Thus far, no single mechanism has been identified to explain all cases of paraplegia during these procedures nor has there been any technique proposed as an absolute prevention of this devastating problem. The purpose of this article is to review briefly the physiologic and etiologic aspects of neural ischemia and to discuss experimental and clinical methods to prevent these neurologic complications. Neuronal function is dependent on the cerebral metabolism, by aerobic oxidation, of glucose and oxygen perfused to the brain. Because of inadequate storage of these primary substrates within the neurons, the brain is highly dependent on continuous vascular perfusion for delivery of these substrates of oxygen and glucose. In most situations, cerebral glucose stores exceed cerebral oxygen stores; thus, loss of neuronal function is generally related to an oxygen deficit when cerebral circulation is suddenly reduced. Neurologic function, as determined by electroencephalographic activity, appears to cease in humans when hemispheric flow rates fall below 16 to 17 ml/ 100 gm of tissue/min.1Sundt TM Sharbrough FW Piepgras DG Kearns TP Messick Jr, JM O'Fallon WM. Correlation of cerebral blood flow and electroencephalographic changes duuring carotid endarterectomy: with results of surgery and hemodynamics of cerebral ischemia.Mayo Clin Proc. 1981; 56: 533-543PubMed Google Scholar Different pathologic states can result in different levels of cerebral ischemia. Carotid cross-clamping during endarterectomy may result in a local area of cerebral ischemia but rarely results in complete ischemia since collateral blood flow from the contralateral carotid artery and from the vertebral arteries will generally provide collateral flow to some extent. On the other hand, cardiac arrest results in complete ischemia of the brain since flow through all cerebral vessels ceases. Such as seen with cardiac arrest, complete ischemia produces severe neuronal injury by interrupting the oxygen supply and the metabolic substrates. After recovery from complete ischemia, the brain undergoes a transient (10 to 20 minutes) hyperperfusion because of vasomotor paralysis. Weakening of the blood-brain barrier occurs with resultant edema formation. Afterward, a prolonged (6 to 18 hours) hypoperfusion occurs because of reactive vasoconstriction and a severe worsening of the neurologic damage ensues. Hypothermia, because it reduces the metabolic rate and the oxygen requirement, is able to prolong safe ischemia time in patients who are to undergo complete ischemia. Reduction of core temperatures in humans to 12° to 20° C can extend safe ischemia time to 30 to 60 minutes.2Crawford ES Snyder DM. Treatment of aneurysm of the aortic arch.J Thorac Cardiovasc Surg. 1983; 85: 237-246PubMed Google Scholar, 3Crawford ES Saleh SA. Transverse aortic arch aneurysm, improved results of treatment employing new modifications of aortic reconstruction and hypothermic cerebral circulatory arrest.Ann Surg. 1981; 194: 180-188Crossref PubMed Scopus (109) Google Scholar, 4Griepp RB Stinson EB Hollingsworth JF Buehler D. Prosthetic replacement of the aortic arch.J Thorac Cardiovasc Surg. 1975; 70: 1051-1063PubMed Google Scholar This degree of hypothermia will generally produce fully dilated pupils and an isoelectric electroencephalogram. Obviously hypothermia must be started before the ischemic event. Calcium entry blockers, such as nimodipine, which modify postischemic metabolic processes, have also been shown to be effective in cases of complete global ischemia when given before the ischemic event.5Newberg LA. Cerebral resuscitation: advances and controversies.Ann Emerg Med. 1984; 13: 853-856Abstract Full Text PDF PubMed Scopus (13) Google Scholar, 6Stein PA Mitchenfelder JD. Cerebral protection with barbiturates: relation to anesthetic effect.Stroke. 1978; 9: 140-142Crossref PubMed Scopus (74) Google Scholar Ischemia that is incomplete (e.g., hypoxemia, carotid clamping, or shock) also reduces the oxygen and glucose delivery to the brain but the oxygen deprivation is clearly the major detrimental factor. In fact, with hyperglycemia, incomplete ischemia is actually more detrimental to the brain than complete ischemia because glucose can be translocated from blood to tissue with continued metabolic activity in this hypoxic state, thereby giving rise to excessive acidosis. This results in more detrimental alterations in metabolism than had complete cessation of oxygen and glucose flow occurred. On the other hand, incomplete ischemia in the face of hypoglycemia has better recovery and less neurologic and metabolic derangement than occurs in complete ischemia. Generally, the secondary hypoperfusion does not occur. Thus, complete ischemia produces a more severe insult to neurologic function than does incomplete ischemia except in the presence of hyperglycemia. Barbiturates have been shown to provide cerebral protection when given before focal or incomplete ischemia. In 1962, Arnfred and Secher7Arnfred I Secher O. Anoxia and barbiturates: tolerance to anoxia in mice influenced by barbiturates.Arch Int Pharmacodyn Ther. 1962; 139: 67-74Google Scholar demonstrated that mice anesthetized with thiopental survived longer than did control mice when exposed to gas mixtures containing 0% to 5% oxygen. It has been demonstrated that the major mechanism of action of barbiturates is to reduce cerebral oxygen metabolism to 40% of normal because they reduce or suppress neuronal electrical function as measured by electroencephalography (EEG). Barbiturates are unable to effect the cerebral metabolism required for cellular function, such as biosynthesis or maintenance of ion gradients. This mechanism explains protection in instances of incomplete global ischemia in which some basal nutrient supply to cerebral tissue is still maintained. However, since oxygen delivery, although inadequate, does continue during incomplete global ischemia, a reduction in oxygen metabolism is beneficial. Barbiturates cannot be demonstrated to provide any protection after complete global ischemia and thus do not provide any protection after cardiac arrest. Therefore, hypothermia offers cerebral protection in cases of both incomplete and complete ischemia. Barbiturates offer protection when given before incomplete cerebral ischemia and calcium channel blockers may provide benefit when given before episodes of complete global ischemia. However, in peripheral vascular surgery the most common cause of ischemia is carotid cross-clamping during endarterectomy. Since the use of an intraluminal shunt can prevent this ischemia, shunting is more beneficial than any of the pharmacologic methods available for cerebral protection. In those rare cases where severe reduction of flow is anticipated (such as complex arch vessel reconstructions), high-dose thiopental (Pentothal) may offer some prolongation of safe ischemia time. Paraplegia is a dreaded complication of thoracic and thoracoabdominal aortic repair. Prevention of this neurologic deficit has been the object of many research efforts in the past. Some of the techniques and pharmacologic intervention that have demonstrated usefulness experimentally are induced hypothermia of the spinal cord, systemic infusion of magnesium ions, opiate antagonists (e.g., naloxone), steroids, and reduction in spinal fluid pressure. Since hypothermia has been shown to be effective in preventing ischemic injury to the brain, one might expect it to be similarly beneficial during periods of spinal cord ischemia. There does appear to be a linear relationship between temperature and duration of ischemia tolerated in the spinal cord; each 1° C reduction in temperature, within some limits, appears to increase safe ischemia time in the cord by about 5½ minutes.8Vacanti FX Ames III., A Mild hypothermia and magensium protect against irreversible damage during CNS ischemia.Stroke. 1984; 15: 695-698Crossref PubMed Scopus (171) Google Scholar, 9Coles JG Wilson GJ Sima AF et al.Intraoperative management of thoracic aortic aneurysm, experimental evaluation of perfusion cooling of the spinal cord.J Thorac Cardiovasc Surg. 1983; 85: 292-299PubMed Google Scholar The magnesium ion, possibly because of its ability to increase plasma membrane stability, to compete with calcium for entry into cells, to inhibit neurotransmitter release, and to inhibit contraction of smooth muscle, has also been shown to be advantageous in preventing ischemic cord damage and also appears to act synergistically with cooling. Cooling to 34° C resulted in doubling of the safe ischemia time in the same model, whereas the addition of magnesium plus cooling to 30° C tripled safe ischemia time in this model. Experimental work by Faden et al.10Faden Al Jacobs TP Smith MP Zivin JA. Naloxone in experimental spinal cord ischemia: dose response studies.Eur J Pharmacol. 1984; 103: 115-120Crossref PubMed Scopus (64) Google Scholar has shown that naloxone improved neurologic outcome after spinal cord injury. They postulated that after a period of spinal cord injury, endogenous opiates (endorphins) are secreted, which reduce spinal cord blood flow. Opiate antagonists, such as naloxone, might be beneficial by blocking this response. Laschinger et al.11Laschinger JC Cunningham Jr, JN Cooper MM Krieger K Nathan IM Spencer FC. Prevention of ischemic spinal cord injury following aortic crossclamping: use of corticosteroids.Ann Thorac Surg. 1984; 38: 500-507Abstract Full Text PDF PubMed Scopus (114) Google Scholar also demonstrated some spinal cord protection provided by methyl prednisolone in dogs if given before aortic cross-clamping as well as after operation. The protective effect might in some way be related to a reduction in spinal cord edema or reduction in cerebrospinal fluid pressure as a result of steroid administration. Perhaps the most intriguing experimental studies regarding spinal cord protection relates to the interaction of arterial blood pressure and cerebrospinal fluid pressure within the spinal canal. Experimental cross-clamping of the descending thoracic aorta in dogs has been reported to result in a 3 to 4 mm Hg increase in cerebrospinal fluid pressure and one study on humans has reported an average increase of 7 mm Hg in spinal fluid pressure in eight patients.12Berendes JN Bredee JJ Schipperheyn JJ Mashhour YAS. Mechanisms of spinal cord injury after cross-clamping of the descending thoracic aorta.Circulation. 1982; 66: 1112-1116Google Scholar Nugent, Kaye, and McGoon13Nugent M Kaye MP McGoon DC. Effects of nitroprusside on aortic and intraspinal pressures during thoracic aortic crossclamping.Anesthesiology. 1984; 61 (Abstract): 68Crossref Google Scholar demonstrated an increase in cerebrospinal fluid pressure in four of seven animals, with one animal that had increased spinal fluid pressure above the arterial pressure measured below the cross-clamp. Nitroprusside further increased elevations in cerebrospinal fluid pressure in these animals. These findings suggest that marked elevations in spinal fluid pressure could effectively reduce arterial perfusion of the spinal cord. As depicted in Fig. 1, if aortic cross-clamping causes a reduction in anterior spinal artery pressure to 40 mm Hg and cerebrospinal fluid pressure increases to 40 mm Hg, one anticipates a loss of effective perfusion pressure to that segment of the spinal cord.This mechanism could have caused many otherwise unexplained cases of paraplegia despite reimplantation of intercostal arteries. Aortic cross-clamping will reduce aortic blood pressure below the cross-clamp, thereby reducing the arterial pressure to the spinal cord; at the same time, the increase in pressure above the clamp can increase cerebrospinal fluid pressure, resulting in greater pressure surrounding the spinal cord. If the spinal fluid pressure exceeds the arterial blood pressure delivering blood flow to the spinal cord, a relative decrease in cord perfusion can occur, resulting in cord ischemia. An equation could be written: spinal cord perfusion = spinal artery blood pressure − cerebrospinal fluid pressure. In these situations, an increase in pressure in the distal aorta (such as that achievable with a Gott shunt or femorofemoral bypass) may reduce the incidence of paraplegia. Similarly, one might expect a reduction in paraplegia in these patients by reducing the cerebrospinal fluid pressure. Miyamoto et al.14Miyamoto K Ueno A Wada T Kimoto S. A new and simple method of preventing spinal cord damage following temporary occlusion of the thoracic aorta by draining the cerebrospinal fluid.J Cardiovasc Surg. 1960; 16: 188-197Google Scholar in 1960 and Blaisdell and Cooley15Blaisdell FW Cooley DA. The mechanism of paraplegia after temporary thoracic aortic occlusion in its relationship to spinal fluid pressure.Surgery. 1962; 51: 351-355PubMed Google Scholar in 1962 showed that cerebrospinal fluid drainage markedly decreased paraplegia rates in dogs undergoing aortic cross-clamping. Oka and Miyamoto16Oka Y Miyamoto T. Prevention of spinal cord injury after crossclamping of the thoracic aorta.Jpn J Surg. 1984; 14: 159-162Crossref PubMed Scopus (26) Google Scholar recently reconfirmed this earlier work and found that paraplegia routinely occurred in animals that did not have adequate relative perfusion pressure, but no paraplegia occurred when a gradient pressure greater than 15 mm Hg was maintained between the aorta and the cerebrospinal fluid pressure. Experimental data thus heavily support the concept that spinal cord perfusion is dependent not only on absolute arterial pressure below the cross-clamp but also on the relationship of spinal fluid pressure and arterial blood pressure. Since other factors also have direct impact on spinal fluid pressure, such as osmotic diuretics and vasodilators, the clinical implications remain complex and are yet to be clarified in humans. In attempts to ensure adequate spinal cord revascularization during thoracoabdominal aneurysm repair, various methods have been advocated. Some authors have advocated routine reimplantation of all prominent intercostal arteries. Others have advocated preoperative identification of major cord blood supply by the use of selective angiography of intercostal vessels. Still others have proposed the routine use of somatosensory evoked potentials (SSEP) in an effort to determine which patients need intercostal artery reimplantation. Since many of these patients have coexistent renal dysfunction with elevated creatinine values, one might be hesitant to recommend selective intercostal angiography because a large angiographic contrast load may worsen renal function. In addition, paraplegia as a result of direct intercostal artery angiography has been reported. Use of SSEP has been poorly accepted because of inadequate sensitivity and specificity. In many patients, the peripheral nerve ischemia in the lower extremities associated with aortic clamping during thoracoabdominal aneurysm repair interferes with nerve conduction when the inciting signal originates at the ankle level. Thus, false positive findings can develop because of peripheral nerve ischemia rather than spinal cord ischemia. Some authors have prevented this problem by having the stimulae originate in the epidural space, thereby avoiding the question of distal ischemia. However, even this technique has been inaccurate in some cases. Since SSEP are mediated by posterior columns of the spinal cord, anterior column ischemia, specifically the motor pathways primarily supplied by the anterior spinal artery, can develop and yet remain undetected by SSEP. More recent use of motor evoked potentials appear promising and may be more reliable for the determination of spinal cord ischemia. However, because of these various pitfalls, we have felt more comfortable with routine reimplantation of all intercostal arteries identified during the course of thoracoabdominal aneurysm repair. The reimplantation of one or more aortic cuffs, containing multiple pairs of intercostal arteries, has facilitated spinal cord revascularization in these cases. Many surgeons advocate the routine use of the Gott shunt, femorofemoral bypass, or left atrial—to—femoral bypass with use of the centrifugal pump. In some series of thoracic aortic aneurysm repair routine use of these devices has suggested a beneficial effect in reducing the incidence of paraplegia. Livesay et al.,17Livesay JJ Cooley DA Ventimiglia RA et al.Surgical experience in descending thoracic aneurysmectomy with and without adjuncts to avoid ischemia.Ann Thorac Surg. 1985; 39: 37-46Abstract Full Text PDF PubMed Scopus (240) Google Scholar in a review of thoracic aortic aneurysm repair, noted that prolonged aortic clamp time was associated with an increased incidence of paraplegia unless a shunt was used. Jex et al.18Jex RK Schaff HV Piehler JM et al.Early and late results following repair of dissection of the descending thoracic aorta.J Vasc Surg. 1986; 3: 226-237Abstract Full Text Full Text PDF PubMed Scopus (110) Google Scholar noted similar findings after repair of dissecting aneurysms of the descending thoracic aorta. However, use of shunts or bypass in repair of extensive thoracoabdominal aortic aneurysms does not appear as beneficial probably because the distal perfusion in these cases may be so far distal that it cannot increase flow to the spinal cord, since the collateral vessels arise above the level of lower perfusion. Moreover, use of the Gott shunt or bypass pump is not without complications and has resulted in increased bleeding as well as increased morbidity and deaths from embolization or aortic disruption at the site of shunt implantation in the proximal or distal aorta. In general, it appears that a shunt is unnecessary for spinal cord protection if aortic cross-clamp time is kept below 30 minutes. However, if one anticipates prolonged cross-clamp time, use of a shunt or a temporary bypass may be advantageous if anatomic and technical features would allow perfusion of the distal aorta at least up to the level of L-3. Some thoracoabdominal aneurysms, particularly the very low and the very high, may allow technical modifications of graft placement that reduce the time required for anastomosis and thus a reduction in overall aortic clamp time. These technical modifications have previously been reported by us.19Hollier LH. Technical modifications in the repair of thoracoabdominal aortic aneurysm.in: Vascular surgical techniques. Butterworth, Kent1984: 99-106Google Scholar Currently, my own approach to thoracoabdominal aneurysm repair consists of the routine use of nitroprusside for reduction of blood pressure and improvement in collateral flow distally. I prefer simple aortic clamping, rapid reanastomosis, and restoration of flow without the use of shunts or bypass and with the technical modifications previously reported. All intercostal vessels are reimplanted; we have abandoned the use of SSEP. With these techniques, the incidence of neurologic deficit, including both paraplegia and paraparesis, has been 6.3% in the last 64 patients in whom thoracoabdominal aneurysms were repaired without the use of shunt or bypass (Table I).Table IDemographic dataNo. of patients64 Type I (mainly thoracic)6 Type II (Diffuse thoracic and abdominal)44 Type III (mainly abdominal)14Age Mean (yr)72.3 Range (yr)31-89Ruptured (%)10.9Symptomatic (%)60.9NOTE: These 64 patients underwent thoracoabdominal aneurysm repair without use of a shunt or bypass. Open table in a new tab The overall mortality rate was similarly 6.3%, whereas 4.7% of these patients required temporary dialysis and 4.7% required reoperation for postoperative complications, primarily bleeding (Table II).Table IIDeaths and complications occurring after thoracoabdominal repair (N = 64)No. of patients%Mortality46.3Neurologic deficit46.3Renal failure (dialysis)34.7Reoperation (bleeding)34.7NOTE: Three of the four patients who sustained paraplegia did not have reimplantation of all intercostal arteries. Open table in a new tab This compares favorably with the 11% neurologic deficit reported by Crawford et al.20Crawford ES Crawford JL Safi HJ et al.Thoracoabdominal aortic aneurysms: preoperative and intraoperative factors determining immediate and long-term results of operations in 605 patients.J Vasc Surg. 1986; 3: 389-404Crossref PubMed Google Scholar NOTE: These 64 patients underwent thoracoabdominal aneurysm repair without use of a shunt or bypass. NOTE: Three of the four patients who sustained paraplegia did not have reimplantation of all intercostal arteries. In light of the more recent understanding of the relationship of spinal fluid pressure to spinal cord perfusion, our current practice also includes the routine monitoring of spinal fluid pressure during thoracoabdominal aortic aneurysm repair. It is still too early in our experience to report whether this will provide additional information and lead to further intervention to reduce spinal cord ischemia. However, it is anticipated that by a better understanding of the entire spectrum of physiologic changes that occur in these patients, we can continue to develop improvements in technique that can ultimately abolish paraplegia as a serious risk of thoracoabdominal aneurysm repair.
Referência(s)