Cerebrospinal-fluid drain-related complications in patients undergoing open and endovascular repairs of thoracic and thoraco-abdominal aortic pathologies: a systematic review and meta-analysis.

BACKGROUND: Cerebrospinal-fluid (CSF) drainage is recommended by current guidelines for spinal protection during open and endovascular repairs of thoracic and thoraco-abdominal aortic aneurysms. In the published literature, great variability exists in the rate of CSF-related complications and morbidity. Herein, we perform a systematic review and meta-analysis on the incidence of CSF drainage-related complications, and compare the complication rates between open and endovascular repairs. METHODS: The systematic review was conducted according to the Meta-Analysis of Observational Studies in Epidemiology guidelines. Thirty-four studies (4714 patients) were included in the quantitative analysis. The CSF drainage-related complications were categorised as mild, moderate, and severe. Pooled event rates for each complication category were estimated using a random-effect model. Random-effect uni- and multivariable meta-regression analyses were used to assess the effect of aortic-repair approach (open vs endovascular) and the CSF drainage criteria on CSF drainage-related complications. RESULTS: The pooled event rates were 6.5% [95% confidence interval (CI): 4.3-9.8%] for overall complications, 2% (95% CI: 1.1-3.4%) for minor complications, 3.7% (95% CI: 2.5-5.6%) for moderate complications, and 2.5% (95% CI: 1.6-3.8%) for severe complications. The drainage-related-mortality pooled event rate was 0.9% (95% CI: 0.6-1.4%). The uni- and multivariable meta-regression analyses showed no difference in complication rates between the open and endovascular approaches, or between the different CSF drainage protocols. CONCLUSION: The complication rate for CSF drainage is not negligible. Our results help define a more accurate risk-benefit ratio for CSF drain placement at the time of repair of thoracic and thoraco-abdominal aneurysms.

Automated pressure-controlled cerebrospinal fluid drainage during open thoracoabdominal aortic aneurysm repair.

OBJECTIVE: Perioperative cerebrospinal fluid (CSF) drainage is a well-established technique for spinal cord protection during thoracoabdominal aortic aneurysm (TAAA) open repair and is usually performed using dripping chamber-based systems. A new automated device for controlled and continuous CSF drainage, designed to maintain CSF pressure around the desired set values, thus avoiding unnecessary drainage, is currently available. The aim of our study was to determine whether the use of the new LiquoGuard automated device (Möller Medical GmbH, Fulda, Germany) during TAAA open repair was safe and effective in maintaining the desired CSF pressure values and whether the incidence of complications was reduced compared with a standard catheter connected to a dripping chamber. METHODS: Data of patients who underwent surgical TAAA open repair using perioperative CSF drainage at our institution between October 2012 and October 2014 were recorded. The difference in CSF pressure values between patients who underwent CSF drainage with a conventional dripping chamber-based system (manual group) and patients who underwent CSF drainage with the LiquoGuard (automated group) was measured at the beginning of the intervention (T1), 15 minutes after aortic cross-clamping (T2), just before unclamping (T3), at the end of surgery (T4), and 4 hours after the end of surgery (T5). The choice of the draining systems was randomly alternated with one-to-one rate until the last six patients consecutively treated with LiquoGuard were enrolled. Primary outcomes were occurrence of spinal cord ischemia, intracranial hemorrhage, postdural puncture headache, and in-hospital mortality. RESULTS: The study included 152 patients who underwent open surgical TAAA repair during the study period: 73 patients underwent CSF drainage with the traditional system and 79 with LiquoGuard. The CSF pressure values at T1 and T5 were not considerably different in the two groups. By repeated-measures analysis of variance, a significant upward trend of perioperative CSF pressure was observed in the automated group at T2, T3, and T4 (group × time interaction = F3,66; P < .001). No difference was reported in the occurrence of spinal cord ischemia, intracranial hemorrhage, or mortality. The LiquoGuard group reported significantly reduced postdural puncture headache (3.3% vs 16.9%; P = .01). CONCLUSIONS: Perioperative use of LiquoGuard during TAAA open repair was safe and effective. Despite slightly higher intraoperative CSF pressures, the rate of spinal cord ischemia did not increase in the LiquoGuard group, and postdural puncture headache significantly decreased.

Is cerebrospinal fluid drainage of benefit to neuroprotection in patients undergoing surgery on the descending thoracic aorta or thoracoabdominal aorta?

A best evidence topic in cardiac surgery was written according to a structured protocol. The question addressed was 'Is cerebrospinal fluid (CSF) drainage of benefit in patients undergoing surgery on the descending thoracic aorta or thoracoabdominal aorta?' Altogether 1177 papers were found using the reported search, of which 17 represented the best evidence to answer the clinical question. The authors, journal, date and country of publication, patient group studied, study type, relevant outcomes and results of these papers are tabulated. Ten of 13 studies demonstrate significant neurological protection from CSF drainage (± additional adjuncts), with two further papers showing no significant difference between patients who had or had not had CSF drainage and one study unable to provide any conclusions. For patients having surgery on the thoracic aorta or thoracoabdominal aorta CSF drainage, maintaining pressures <10 mmHg (P < 0.03), in conjunction with other neuroprotective strategies, minimizes the risk of neurological sequelae when compared with patients treated with similar adjuncts but without CSF drainage. The majority of studies used additional neuroprotective strategies, including cooling and reattachment of the intercostal arteries as adjuncts to CSF drainage. Logistic regression curves demonstrated that the longer the ischaemia time, the greater the benefit from CSF drainage (P < 0.04). Four papers observed complications of CSF drainage, of which the main complications were: catheter occlusion or dislodgement, headache, meningitis and subdural haematoma. Overall, CSF drainage does offer a neuroprotective benefit; preventing paraplegia if CSF pressures are maintained <10 mmHg.

Intrathecal lactate concentration and spinal cord injury in thoracoabdominal aortic surgery.

OBJECTIVE: The aim of this study was to evaluate the role of lactate as an early predictor of spinal cord injury during thoracoabdominal aortic aneurysm repair. DESIGN: Observational study. SETTING: University hospital. PARTICIPANTS: Sixteen consecutive patients (10 men and 6 women) scheduled to undergo thoracoabdominal aortic aneurysm repair were enrolled in the study. All patients were affected by atherosclerotic aneurysmal pathology. INTERVENTIONS: None. MEASUREMENTS AND MAIN RESULTS: During surgery, the authors simultaneously withdrew samples of cerebrospinal fluid and arterial blood to evaluate pO(2), pCO(2), pH, and lactate concentration. Samples were collected at 5 fixed times during and after surgery: T1 (before aortic cross-clamping), T2 (15 minutes after clamping), T3 (just before unclamping), T4 (end of surgery), and T5 (4 hours after the end of surgery). Lactate levels in cerebrospinal fluid rose consistently during aortic cross-clamping (T1 = 1.89 mmol/L, T2 = 2.21 mmol/L, T3 = 2.88 mmol/L, T4 = 3.655 mmol/L, and T5 = 3.16 mmol/L). Lactate concentrations in the cerebrospinal fluid were significantly higher in the 4 patients who developed neurologic injury, even at T1 (before surgery), than in those who did not end in spinal cord injury with the 4 highest values belonging to the 4 patients who later developed spinal cord injury. CONCLUSIONS: This study has the potential to elucidate the time course of early lactate level elevation during thoracoabdominal aortic aneurysm repair and its clinical use in predicting the development of postoperative spinal cord injury.

Lumbar cerebrospinal fluid drainage for thoracoabdominal aortic surgery: rationale and practical considerations for management.

Paraplegia remains one of the most devastating complications of thoracoabdominal aortic surgery and is associated with a significant increase in both morbidity and mortality. Modern aortic repair techniques use many modalities aimed at reducing the risk of spinal cord ischemia inherent with surgical management. One of these modalities that acts via optimizing spinal cord blood flow is lumbar cerebrospinal fluid (CSF) drainage. Either alone or in combination with other interventions, CSF drainage remains one of the most frequently used spinal cord protection techniques. Despite no definitive proof of efficacy for reducing spinal cord injury, there are compelling data supporting its use. However, the potential benefit of CSF drainage must be balanced against the risks associated with its use, including nerve injury during insertion, compressive neuraxial hematoma formation, intracranial hemorrhage due to excessive drainage, and infection. The optimal benefit to risk ratio can be achieved by understanding the rationale for its use and following practical management guidelines.

Heat shock proteins HSP70 and HSP27 in the cerebral spinal fluid of patients undergoing thoracic aneurysm repair correlate with the probability of postoperative paralysis.

An understanding of the time course and correlation with injury of heat shock proteins (HSPs) released during brain and/or spinal cord cellular stress (ischemia) is critical in understanding the role of the HSPs in cellular survival, and may provide a clinically useful biomarker of severe cellular stress. We have analyzed the levels of HSPs in the cerebrospinal fluid (CSF) from patients who are undergoing thoracic aneurysm repair. Blood and CSF samples were collected at regular intervals, and CSF was analyzed by enzyme-linked immunosorbent assay for HSP70 and HSP27. These results were correlated with intraoperative somatosensory-evoked potentials measurements and postoperative paralysis. We find that the levels of these proteins in many patients are elevated and that the degree of elevation correlates with the risk of permanent paralysis. We hypothesize that sequential measurement intraoperatively of the levels of the heat shock proteins HSP70 and HSP27 in the CSF can predict those patients who are at greatest risk for paralysis during thoracic aneurysm surgery and will allow us to develop means of preventing or attenuating this severe and often fatal complication.

Measurements of cerebrospinal fluid concentrations of S100beta protein during and after thoracic endovascular stent grafting.

OBJECTIVES: Thoracic endovascular aortic repair is associated with postoperative spinal cord ischemia in approximately 1 to 12.5% of all cases. S100beta is a protein that is released during acute damage of the central nervous system. This study was performed to determine the concentration of S100beta in cerebrospinal fluid during and after stenting of the thoracic aorta in patients at high risk for spinal cord ischemia. DESIGN: Prospective clinical study. MATERIALS AND METHODS: Eight patients who underwent elective thoracic aortic stent grafting underwent lumbar spinal fluid drainage. These patients were at high risk to develop spinal cord ischemia. METHODS: CSF samples for analysis of S100beta protein were drawn after induction of anesthesia, during stenting, once every hour the following six hours and 20 hours after repair. RESULTS: No significant increase in S100beta protein could be detected in CSF and no neurological deficits were detected postoperatively. CONCLUSIONS: The results of this study show us that there is no significant release of S100beta protein in CSF during stenting of the thoracic aorta in this subgroup of patients at high risk for spinal cord ischemia, consistent with clinical exam that there was no significant damage to the central nervous system.

Subdural hematoma induced by cerebrospinal fluid drainage: potentially catastrophic complication of spinal cord protection after thoracoabdominal aneurysm repair.

A 50-year-old man underwent replacement of the descending thoracic aorta for a DeBakey type III-b chronic dissecting aortic aneurysm. During the surgery, lumbar cerebrospinal fluid (CSF) drainage with a 10 cm H2O pop-off pressure was used to protect against spinal cord ischemia. During cardiopulmonary bypass, the patient's pupils were isocoric, but anisocoric at end of the operation. As computed tomography of the brain showed right subdural hematoma, neurosurgical drainage was instituted emergently. Although some neurological deficit remained, the patient recovered well and was discharged. This case provides a strong reminder that CSF drainage for spinal cord protection against ischemia might induce subdural hematoma, which can be catastrophic during an operation for thoracoabdominal aortic aneurysm.

Biochemical markers of cerebrospinal ischemia after repair of aneurysms of the descending and thoracoabdominal aorta.

OBJECTIVE: To investigate the clinical potential of several markers of spinal cord ischemia in cerebrospinal fluid (CSF) and serum during aneurysm repair of the descending thoracic or thoracoabdominal aorta. DESIGN: Observational study of consecutive patients. Nonblinded, nonrandomized. SETTING: University hospital thoracic surgical unit. PARTICIPANTS: Eleven consecutive elective patients. INTERVENTIONS: Distal extracorporeal circulation and maintenance of CSF pressure <10 mmHg until intrathecal catheter removal. MEASUREMENTS AND MAIN RESULTS: CSF and serum levels of S100B (and its isoforms S100A1B and S100BB), neuronal-specific enolase (NSE), and the CSF levels of glial fibrillary acidic protein (GFAp) and lactate were determined. Two patients had postoperative neurologic deficit. One with a stroke showed a 540-fold increased GFAp, a 6-fold NSE, and S100B increase in CSF. One with paraplegia had a 270-fold increase in GFAp, a 2-fold increase in NSE, and 5-fold increased S100B in CSF. One patient without deficit increased GFAp 10-fold, NSE 4-fold, and S100B 23-fold in CSF. CSF lactate increased >50% in 6 of 9 patients without neurologic deficit. Serum S100B increased within 1 hour of surgery in all patients without any concomitant increase in CSF. S100A1B was about 70% of total S100B in both serum and CSF in patients with or without neurologic defects. S100B in CSF increased 3-fold in 3 of 9 asymptomatic patients. CONCLUSIONS: In patients with neurologic deficit, GFAp in CSF showed the most pronounced increase. Biochemical markers in CSF may increase without neurologic symptoms. There is a significant increase in serum S100B from surgical trauma alone without any increase in CSF.

Subdural hematoma after thoracoabdominal aortic aneurysm repair: an underreported complication of spinal fluid drainage?

OBJECTIVE: Cerebrospinal fluid (CSF) drainage is a commonly used adjunct to thoracoabdominal aortic aneurysm (TAAA) repair that improves perioperative spinal cord perfusion and thereby decreases the incidence of paraplegia. To date, little data exist on possible complications, such as subdural hematoma caused by stretching and tearing of dural veins, should CSF drainage be excessive. We reviewed our experience with patients in whom postoperative subdural hematomas were detected. METHODS: The records of 230 patients who underwent TAAA repair at the Johns Hopkins Hospital between January 1992 and February 2001 were reviewed. RESULTS: Eight patients had subdural hematomas (3.5%). The four men and four women had a mean age of 60.6 years; two of these patients had a connective tissue disorder. All patients had lumbar drains placed before surgery, including one patient who underwent an emergency operation for rupture. Drains were set to allow drainage for CSF pressure greater than 5 cm H(2)O in all but one patient set for 10 cm H(2)O; spinal cooling was not performed in any patient. All drains were removed on the third postoperative day. In patients in whom subdural hematomas developed, the mean amount of CSF removed after surgery was 690 +/- 79 mL, which was significantly greater than the amount drained from patients in whom subdural hematomas did not develop (359 +/- 24 mL; P =.0013, Mann-Whitney U test). Six patients had postoperative subdural hematomas detected during hospitalization (mean postoperative day, 9.3; range, 2 to 16), and two patients were seen in delayed fashion after discharge from the hospital at 1.5 and 5 months. Four patients died of the subdural hematoma (50%); only one of these patients had neurosurgical intervention. All four survivors responded to neurosurgical intervention and are neurologically healthy. Two patients, both of whom were seen in delayed fashion, needed a lumbar blood patch. Multivariate logistic regression identified the volume of CSF drained as the only variable predictive of occurrence of subdural hematoma (P =.01). CONCLUSION: Subdural hematoma is an unusual and potentially catastrophic complication after TAAA repair. Prompt recognition and neurosurgical intervention is necessary for survival and recovery after acute presentation. Epidural placement of a blood patch is recommended if a chronic subdural hematoma is detected. Care should be taken to ensure that excessive CSF is not drained perioperatively, and higher (10 cm H(2)O) lumbar drain popoff pressures may be necessary together with meticulous monitoring of patient position and neurologic status.

[New method for the intraoperative biochemical monitoring of cerebrospinal fluid in surgery of thoracoabdominal aortic aneurysms]. / Uj módszer a cerebrospinalis fluidum (CSF) intraoperatív biokémiai monitorozásában thoracoabdominalis aortaaneurysma-mútétek során.

Paraplegia remains to be one of the most dangerous complications following thoracoabdominal aortic surgery with an incidence of 0.5 to 40%. Therefore, intraoperative monitoring of spinal cord function is very important when choosing the appropriate surgical technique. Early detection of spinal cord injury continues to be a crucial problem, moreover, the currently applied electrophysiological methods appear to be inaccurate. The aim of the study was to detect prospective spinal cord injury intraoperatively by monitoring the biochemical parameters of the cerebrospinal fluid (CSF). The authors studied the reversible aerobic/anaerobic metabolic changes by monitoring CSF lactate levels, moreover S-100 protein and neuron-specific enolase (NSE) concentrations--specific for neuroglia and neuronal injury, respectively. One of the important methods to prevent paraplegia is the intraoperative CSF drainage, which may improve spinal cord perfusion. Between 1996-1998 51 patients underwent reconstructive thoracic or thoracoabdominal aortic aneurysm operation. The continuously drained CSF was collected in 10 ml fractions during the preparation, whereas during aortic cross-clamping and de-clamping 10 minute fractions were used. All CSF samples were immediately analysed intraoperatively for pH, pCO2, HCO3, potassium and lactate levels, S-100 protein and NSE were analysed by immunoluminescence. CSF lactate levels increased slightly during aortic clamping and a moderate, but non-significant increase was found in the hyperemic phase (reperfusion) in patients without spinal cord ischemia. Spinal cord injury was detected in 7 cases. These patients exhibited a significant CSF-lactate increase (control vs aortic cross-clamping: 1.9 vs 5.3 mmol/l), moreover CSF-lactate remained elevated throughout the whole operation. Paraplegia did not occur, Tarlov 2 paraparesis developed in four cases and three patients displayed cerebral damage. Intraoperative CSF--especially CSF-lactate--monitoring may help the operating team to detect early anaerobic changes of the metabolism the spinal cord.

The relationship between evoked potentials and measurements of S-100 protein in cerebrospinal fluid during and after thoracoabdominal aortic aneurysm surgery.

OBJECTIVE: This study was performed to correlate the changes in concentration of S-100 protein in the cerebrospinal fluid (CSF) during and after thoracoabdominal aortic aneurysm (TAAA) surgery with the results of somatosensory and motor evoked potential monitoring. METHODS: The study was designed as a prospective study at St Antonius Hospital in Nieuwegein, The Netherlands. The participants were 19 patients who were undergoing elective TAAA surgery. CSF samples for analysis of S-100 protein were drawn after the induction of anesthesia, during the cross-clamp period of the critical aortic segment, after 5 minutes of reperfusion of this segment, during the closure of the skin, and 24 hours after the closure of the skin. In all the patients, continuous intraoperative recording of myogenic motor potentials evoked by transcranial electrical stimulation (tcMEP) and somatosensory potentials evoked by stimulation of the posterior tibial nerve took place to monitor the integrity of the spinal cord. The operative technique consisted of staged or sequential clamping to maximize the beneficial effect of the distal perfusion by the left heart bypass, continuous CSF drainage to keep the CSF pressure below 10 mm Hg, and moderate hypothermia (32 degrees C rectal temperature). We correlated the measured concentrations of S-100 protein in CSF with the results of evoked potential monitoring during surgery and the number of intercostals reimplanted and oversewn. RESULTS: In all the patients, the concentration of S-100 protein was increased in CSF. The highest concentration of S-100 protein was found in the CSF sample taken 5 minutes after reperfusion of the critical aortic segment. There was a good (negative) correlation between the changes in S-100 protein in CSF and the changes in motor evoked potential monitoring during the cross-clamp period. The best (negative) correlation was detected between the S-100 protein elevation in the CSF sample drawn 5 minutes after reperfusion and the tcMEP amplitude reduction during clamping (r = -0.73; P =.007). No relation was found between the S-100 protein dynamics in CSF and somatosensory evoked potential monitoring. A positive (r = 0.58; P =.05) correlation was found between the change in tcMEP amplitude during clamping and the number of reattached intercostals. A moderate to good (r = -0.5 to -0.7; P <.05) correlation between the number of reattached intercostals and the changes in S-100 protein concentration in CSF during TAAA surgery was found. Our data show that transient elevations in S-100 protein after cross clamping are larger in those patients with marked decrease in tcMEP from baseline during the cross-clamp period. CONCLUSION: A correlation is shown between an increasing concentration of S-100 protein in CSF and a reduction in tcMEP amplitude during cross clamping of the aorta. The S-100 protein in CSF seems to be a marker of potential clinical value in the evaluation of the effects of procedures to detect and reduce spinal cord ischemia.

Cerebral spinal fluid drainage and distal aortic perfusion decrease the incidence of neurological deficit: the results of 343 descending and thoracoabdominal aortic aneurysm repairs.

OBJECTIVE: We reviewed our experience of 343 descending and thoracoabdominal aortic aneurysm repairs to determine the impact of the adjuncts distal aortic perfusion and cerebral spinal fluid drainage on neurological deficit and death. MATERIALS AND METHODS: Between January 1991 and March 1996, 104 (30%) patients were operated for thoracoabdominal aortic aneurysm type I, 118 (34%) for type II, 68 (20%) for type III or type IV, and 53 (15%) for descending thoracic type. Before September 1992, simple cross-clamp was used for 94 (27%) patients. After September 1992, adjuncts were used for 186 (54%) patients. RESULTS: Overall neurological deficit was 33/343 (10%). Neurological deficit for simple cross-clamp patients compared to adjunct patients was 15/94 (16%) vs. 12/186 (7%) (O.R. 0.36, p < 0.01). For types I and II the incidence was 11/52 (21%) vs. 12/141 (9%) (O.R. 0.35, p < 0.02) and for type II, nine out of 22 (41%) vs. 11/85 (13%) (O.R. 0.21, p < 0.003). Overall 30-day mortality was 43/343 (13%), including patients presenting with rupture. Excluding these patients, overall 30-day mortality was 33/322 (10%). CONCLUSION: Cerebral spinal fluid drainage and distal aortic perfusion decreased the incidence of neurological deficit and were particularly effective for patients at highest risk with type II thoracoabdominal aortic aneurysm.

Changes in cerebrospinal fluid pressure and lactate concentrations during thoracoabdominal aortic aneurysm surgery.

BACKGROUND: Although ischemic injury to the spinal cord is a well-known complication of aortic surgery, no metabolic markers have been identified as predictors of an adverse outcome. This study evaluated the effect of cerebrospinal fluid (CSF) drainage, with and without distal femoral perfusion or moderate hypothermia on blood and CSF lactate concentrations and CSF pressure during thoracoabdominal aortic aneurysm surgery. METHODS: Three nonconcurrent groups of patients were studied prospectively: patients with normal body temperature (35 degrees C) but without distal femoral bypass (n = 6), patients with normal body temperature with bypass (n = 7), and patients with hypothermia (30 degrees C) and bypass (n = 8). In all patients, CSF pressure was recorded before, during, and after aortic cross-clamping. During the surgical repair, CSF drainage was performed using a 4-Fr intrathecal silicone catheter. Blood and CSF lactate concentrations were measured throughout the operation. RESULTS: Significant increases in blood (490%) and CSF (173%) lactate concentrations were observed during and after thoracic aortic occlusion in patients with normothermia and no bypass (P < 0.02 and 0.05, respectively). Distal perfusion attenuated the increase in both blood and CSF lactate (P < 0.01), and a further reduction was achieved with hypothermia of 30 degrees C (P < 0.001). Patients who became paraplegic showed a greater increase in CSF lactate concentrations after aortic clamp release compared with those who suffered no neurological damage (275% vs. 123% of baseline; P < 0.05). Increased CSF pressure of 42-60% (P < 0.005) was noted soon after thoracic aortic occlusion, both with and without distal femoral bypass. CONCLUSIONS: Incremental reductions in CSF lactate concentrations were achieved using distal femoral bypass and hypothermia. The reduction in CSF lactate correlated with the methods used to protect the spinal cord during thoracoabdominal aortic aneurysm surgery and was associated with better outcome. Decompression by distal bypass of the hemodynamic overload caused by aortic occlusion was insufficient to eliminate the acute increase in CSF pressure. Cerebrospinal fluid lactate measurements during high aortic surgery may accurately represent the spinal cord metabolic balance.

Effects of cerebrospinal fluid drainage in patients undergoing thoracic and thoracoabdominal aortic surgery.

Cerebrospinal fluid (CSF) drainage has been reported to protect the spinal cord during surgical procedures requiring thoracic aortic cross-clamping. In 1986, CSF pressure monitoring and drainage was begun in an attempt to reduce the incidence of paraplegia associated with surgical repair of descending thoracic and thoracoabdominal aortic aneurysms (TAAA). These Group II patients (n = 50) were retrospectively compared to Group I patients (n = 49) who had undergone similar surgical procedures in the previous 3 years before CSF monitoring was introduced into this practice. Group II patients had intrathecal catheters placed for monitoring of CSF pressure and drainage of CSF to maintain the pressure < or = 15 mmHg. Seven patients (four in Group I, three in Group II) died before recovering from the anesthetic. Of the 47 patients in Group II who survived, none had clinically apparent complications such as an epidural hematoma or meningitis from the intrathecal catheter. The mean aortic cross-clamp time was 58.6 +/- 30.5 minutes (mean +/- SD) in Group I versus 65 +/- 42.6 minutes in Group II. Twenty-three patients in Group I and 16 patients in Group II had a shunt to the distal aorta. To maintain a CSF pressure of < or = 15 mmHg in Group II, an average of 46.9 +/- 6.9 mL of CSF was withdrawn. Of the 45 survivors in Group I, 4 developed a spinal cord deficit; the number of patients with spinal cord deficit in Group II was 4 out of the 47 who survived.(ABSTRACT TRUNCATED AT 250 WORDS)