Eur J Cardiothorac Surg 2007;31:637-642. doi:10.1016/j.ejcts.2007.01.007
Copyright © 2007, European Association for Cardio-Thoracic Surgery. Published by Elsevier B.V. All rights reserved
Multilevel somatosensory evoked potentials and cerebrospinal proteins: indicators of spinal cord injury in thoracoabdominal aortic aneurysm surgery
Anders Winnerkvista,b,*,
Russell E. Andersona,b,
Lars-Olof Hanssonc,
Lars Rosengrend,
Anthony E. Estrerae,
Tam T.T. Huynhe,
Eyal E. Porate,
Hazim J. Safie
a Department of Cardiothoracic Surgery and Anaesthesiology, Karolinska University Hospital, SE-171 76 Stockholm, Sweden
b Department of Molecular Medicine and Surgery, Karolinska Institutet, Stockholm, Sweden
c Department of Clinical Chemistry, Karolinska University Hospital, Stockholm, Sweden
d Department of Neurology, Sahlgrenska University Hospital, Gothenburg, Sweden
e Department of Cardiothoracic and Vascular Surgery, The University of Texas at Houston Medical School, Memorial Hermann Hospital, Houston, TX, United States
Received 5 September 2006;
received in revised form 22 December 2006;
accepted 8 January 2007.
* Corresponding author. Address: Department of Cardiothoracic Surgery and Anaesthesiology, Karolinska University Hospital, SE-171 76 Stockholm, Sweden. Tel.: +46 8 51770823; fax: +46 8 322701. (Email: anders.winnerkvist{at}ki.se).
 |
Abstract
|
|---|
Objective: Multilevel somatosensory evoked potentials (SSEP) and the release of biochemical markers in cerebrospinal fluid (CSF) were investigated to identify patients with spinal cord ischemia during thoracoabdominal aortic repair and/or a vulnerable spinal cord during the postoperative period. Methods: Thirty-nine consecutive patients undergoing elective aneurysm repair using distal aortic perfusion and cerebrospinal fluid drainage were studied. Continuous SSEP were monitored using nerve stimulation of the right and left posterior tibial nerves with signal recording at the level of both common peroneal nerves, the cervical cord and at the cortical level. CSF concentrations of the markers glial fibrillary acidic protein (GFAp), the light subunit of neurofilament triplet protein (NFL), and S100B were determined at different time points from before surgery until 3 days postoperatively. Results: SSEP indicated spinal cord ischemia in two patients leading to additional intercostal artery reattachments. In one of them the signal loss was permanent and the patient woke up with paraplegia. In the other the signal returned but the patient later developed delayed paraplegia. Three patients without SSEP indications of spinal cord ischemia during surgery later developed delayed paraplegia. The patients with spinal cord symptoms had significant increases, during the postoperative period of CSF biomarkers GFAp (571-fold), NFL (14-fold) and S100B (18-fold) compared to asymptomatic patients. GFAp increased before or in parallel to onset of symptoms in the patients with delayed paraplegia. Conclusions: Peroperative multilevel SSEP has a high specificity in detecting spinal cord ischemia but does not identify all patients with a postoperative vulnerable spinal cord. Biochemical markers in CSF increase too late for intraoperative monitoring but GFAp is promising for identifying patients at risk for postoperative delayed paraplegia.
Key Words: Delayed paraplegia • Glial fibrillary acidic protein • Neurofilament triplet protein • Somatosensory evoked potentials • Spinal cord ischemia • Thoracoabdominal aortic aneurysm
|
1. Introduction
|
|---|
The incidence of neurological complications after repair of thoracoabdominal aortic aneurysms (TAAA) has decreased with use of techniques such as distal aortic perfusion (DAP) and cerebrospinal fluid (CSF) drainage [1]. Spinal cord injury with paraplegia remains, however, a dreaded complication especially in extensive type II TAAA.
Paraplegia following TAAA repair can be categorized by the time of onset. Immediate refers to paraplegia at awakening from anesthesia. Delayed occurs after a patient has recovered from anesthesia and has been evaluated as neurologically intact yet develops paraplegia hours or days – even many days – later. The distinction between immediate and delayed onset is important because the prognosis for immediate paraplegia is poor with unlikely neurologic recovery and an increased risk of early death. The better prognosis associated with delayed paraplegia depends largely on early, aggressive treatment with increased systemic perfusion pressure and CSF drainage.
Controversy remains whether intraoperative neurological monitoring can give the surgeon real-time information on the integrity of spinal function with the possibility of altering surgical procedure when there are indications of spinal cord ischemia. Most studies dealing with neurological monitoring have examined an array of electrophysiological methods [2–4].
A few studies have exploited the indwelling spinal drain as a real-time source of potential biochemical markers [5–7]. Such markers will generally reflect release from any CNS tissue. A pilot study demonstrated an increase in several biochemical markers with both cerebral and spinal cord injury [8].
The aim of the current study was to investigate the concurrent use of multilevel somatosensory evoked potentials (SSEP) and biochemical markers of injury in a larger cohort undergoing thoracoabdominal aortic aneurysm repair. The goal was to find methods of monitoring which could warn the surgeon during surgery and also identify patients with increased risk for delayed paraplegia who require prolonged intensive care and optimization of hemodynamic parameters and CSF pressures.
|
2. Materials and methods
|
|---|
Thirty-nine consecutive patients undergoing elective repair of an aneurysm of the descending or thoracoabdominal aorta were studied after approval from the local ethics committee and written informed consent. No patient had any preoperative neurological symptoms or deficit. Patient characteristics are summarized in Table 1
including the aneurysm extent.
View this table:
[in this window]
[in a new window]
|
Table 1 Demographic and surgical data (mean (range)) comparing patients who did vs those who did not develop spinal cord dysfunction after repair of thoracoabdominal aortic aneurysms
|
|
Anesthesia was induced with fentanyl (10–15 µg/kg), midazolam (0.05 mg/kg), pancuronium (0.1 mg/kg) and propofol (40–70 mg). Maintenance anesthesia was isoflurane 0.5–1.0% in oxygen/air. Aprotinin was administered as a 1 million-unit load followed by an infusion of 250,000 U/h.
The patients were by routine continuously monitored with electroencephalogram (EEG) and multilevel SSEP during surgery using a Nicolet Viking IV electrodiagnostic system (Nicolet Biomedical Inc, WI USA). Using a modified 10/20 system a longitudinal bipolar 8-channel EEG montage was employed. Changes in EEG frequency or amplitude were noted. Closing frequencies and amplitudes were compared to initial values. Multilevel SSEP monitoring was performed employing a second screen on the Nicolet Viking IV. The left and right posterior tibial nerves were stimulated at the ankles using surface electrodes at a rate of 4.7 Hz, with stimulus duration of 0.5 ms and 100 milliamp stimulus intensity. Peripheral responses were recorded via surface or needle electrodes placed at the popliteal fossa for recording the peroneal responses bilaterally. This enabled differentiating the loss of SSEP signal due to leg ischemia from loss of SSEP signal due to spinal cord ischemia. Another needle electrode placed in the posterior neck recorded dorsal column signal as they entered the foramen magnum. Cortical electrodes placed over the somatosensory cortex at the midline recorded cortical waveforms bilaterally. SSEP amplitudes and latencies were assessed. Changes in SSEP were characterized as transient or permanent depending on if changes were present or not at closing. They were further defined as presumably due to peripheral ischemia if the peroneal response was absent or likely due to spinal cord ischemia if the peroneal signals presented with absence of the cervical and cortical responses.
A lumbar intrathecal catheter was introduced at level L4-5 and connected to a pressure transducer and a MoniTorrTM ICP External Drainage and Monitoring System (Integra Neuroscience Implants S.A., Sophia Antipolis, France). CSF samples were drawn from the indwelling catheter and CSF pressure was maintained <10 mmHg by spontaneous drainage. The CSF drainage catheter was maintained for 3 days after surgery or longer when the patients developed neurological deficit.
In all but two patients, DAP with left pulmonary vein to femoral artery bypass was performed using a Biomedicus centrifugal pump (Minneapolis, MN, USA). Arterial outflow cannulation was via the left common femoral artery or distal aorta depending on the individual vascular pathology. One patient had a type IV aneurysm not suitable for DAP. Patients received 100 U kg–1 heparin IV. A cell saving device (Shiley Therapeutic Autotransfusion System; Dideco Shiley SpA; Modena, Italy) was used in all patients. Patient temperature was permitted to drift downward to approximately 34 °C but held above 32 °C (rectal/nasopharyngeal).
The aorta was cross-clamped sequentially for graft replacement depending on aneurysm extent. Patent intercostal arteries in the level from T8 to T12 were reattached to the graft. The visceral arteries (celiac, superior mesenteric, and both renal arteries) were perfused via #9 or #12 F Pruitt catheters. Active cooling permitted lowering kidney temperature to 15 °C measured in the left kidney.
Sampling and analysis: CSF samples were drawn at start of surgery, before aortic cross-clamping, 20 min after cross-clamp, just before release of last cross-clamp, after 15 min reperfusion, at end of case, 1 h postoperatively, 3 h postoperatively, and early morning on postoperative days 1–3 (POD 1–3). All samples were immediately spun down and stored at –80 °C until blinded batchwise assay, for glial fibrillary acidic protein (GFAp), the light subunit of neurofilament triplet protein (NFL), S100B and albumin. For serum S100B and serum-albumin analysis, arterial blood was sampled from an indwelling radial catheter at the same times as CSF sampling. GFAp was measured using an in-house sandwich ELISA based on polyclonal antibodies raised in rabbits and hens [9] and the light subunit of the NFL was measured using a similar ELISA technique adapted to NFL [10]. S100B from CSF and serum were assayed using chemiluminiscent immunometric techniques (Sangtec Medical, Bromma, Sweden). S100B was measured using monoclonal antibodies directed against three of the different epitopes of the ß-subunit of bovine S100B (Sangtec 100 LIA). The S100B assay measures the summed concentrations of S100A1B and S100BB. CSF-albumin was measured using an immuno-nephelometric assay (Image, Coulter-Beckman Inc, CA USA).
Statistical group comparison was performed using nonparametric ANOVA.
|
3. Results
|
|---|
Patient demographic data, level of aneurysm repair and surgical information in Table 1 are grouped according to presence of spinal cord symptoms. There were no differences in age, mean DAP pressure, cross-clamp time or amount of CSF drained.
EEG was continuously measured in all but one patient. Three patients had a period of mild EEG slowing during surgery with no clinical correlate. SSEP findings are summarized in Table 2
. SSEP were indicative of spinal cord ischemia in two patients leading to additional intercostal artery reattachments. In one the signal loss was permanent and the patient woke up with paraplegia. In the other the signal returned but the patient later developed delayed paraplegia. Six patients had SSEP changes attributable to interrupted flow to the lower extremities during aortic repair: The signals returned promptly after reestablishment of flow but one of these patients later developed delayed paraplegia. In nine patients there were changes in signals from the left leg attributable to temporary leg ischemia due to femoral cannulation for distal aortic perfusion and the signals returned at the end of surgery. The signals were completely normal in 22 patients. Two of these later developed delayed paraplegia.
View this table:
[in this window]
[in a new window]
|
Table 2 Continuous intraoperative measurement of multilevel somatosensory evoked potentials (SSEP) in 39 patients undergoing repair of thoracoabdominal aortic aneurysms
|
|
Biochemical markers from the five patients with spinal symptoms are shown individually in Fig. 1
. In addition their average values are compared to the 33 patients without clinical postoperative neurological symptoms in Fig. 2
. Data is normalized to starting values and presented on a logarithmic scale. In the symptomatic group, the mean peak-increase in CSF was 571-fold for GFAp, 16-fold for NFL and 18-fold for S100B. In the group without symptoms the mean concentrations were not increased but some patients showed moderate elevations. The group with spinal symptoms had significantly elevated values the morning after surgery (POD 1) for all three markers when compared to the asymptomatic group but the peak increases came during POD 2 or 3 (Fig. 2).
View larger version (11K):
[in this window]
[in a new window]
|
Fig. 1. CSF concentrations of biomarkers glial fibrillary acidic protein (GFAp), the light subunit of neurofilament triplet protein (NFL), and S100B normalized to their initial values and plotted vs time during/after repair of thoracoabdominal aortic aneurysms. Data for five patients with spinal cord dysfunction are shown (log scale) individually and averaged (±SEM) for 33 non-symptomatic patients.
|
|
View larger version (8K):
[in this window]
[in a new window]
|
Fig. 2. Average CSF concentrations of biomarkers glial fibrillary acidic protein (GFAp), the light subunit of neurofilament triplet protein (NFL), and S100B normalized to their initial values and plotted vs time during/after repair of thoracoabdominal aortic aneurysms. Data averaged (±SEM) for five patients with spinal cord dysfunction is compared to 33 non-symptomatic patients for each time point (*
P
< 0.05, **
P
< 0.01, ***
P
< 0.001).
|
|
Values with ranges for the markers are shown in Table 3
. One patient's biochemical data were excluded from comparison regarding spinal cord injury as he developed cerebral symptoms with severe confusion and the markers will reflect release from any ischemic CNS tissue.
View this table:
[in this window]
[in a new window]
|
Table 3 Initial CSF-values in all patients for glial fibrillary acidic protein (GFAp), the light subunit of neurofilament triplet protein (NFL), and S100B together with peak values for these parameters in patients with and without neurological symptoms after thoracoabdominal aortic aneurysm surgery
|
|
One patient developed paraplegia 10 h postoperatively, two the morning of POD 1 and one on the morning of POD 3. GFAp did not increase during or up to 3 h after surgery (Fig. 1). Unfortunately the data 3 h after surgery is missing in the patient with immediate paraplegia. GFAp was markedly elevated on POD 1–3, especially on day 2. This elevation preceded the onset of symptoms in the two patients with delayed paraplegia after POD 1 and came parallel with symptom onset in the patients with delayed paraplegia during the first postoperative day. Corresponding numbers for NFL and S-100B are also shown.
The data for each patient were not available during the study since the samples for all patients were analyzed together batchwise. Therefore, the treatment of all four patients with delayed paraplegia, were established after occurrence of symptoms. In three of the patients no postoperative triggering event could be recognized (two with normal SSEP and one with SSEP indicative of peroperative injury). In the fourth patient the weakness in the legs came after a hypotensive episode due to cardiac arrhythmia. All four patients had CSF drain in place and CSF could without delay be allowed to drain freely to lower intrathecal pressure. Their blood pressure and hemoglobin levels were kept high to optimize oxygen delivery. Two patients recovered normal leg movement. One patient remained paraplegic and one partially recovered with a remaining weakness in one leg.
Serum S100B and CSF/serum albumin ratio was somewhat elevated during the intra and postoperative period with no difference between patients with and without spinal symptoms (data not shown). The intraoperative increase in serum S100B was not paralleled with any increase in S100B-CSF.
|
4. Discussion
|
|---|
The goal of this study was to investigate the role of multilevel SSEP and the release of biochemical markers into CSF to detect spinal cord injury during thoracoabdominal aortic repair and/or to identify patients with a vulnerable spinal cord during the postoperative period. The principal finding of this study is that peroperative multilevel SSEP has a high specificity in detecting spinal cord ischemia but does not identify all patients with a postoperative vulnerable spinal cord. Biochemical markers in CSF increase too late for intraoperative monitoring but GFAp is promising for identifying patients at risk for postoperative delayed paraplegia.
The quest for methods to minimize spinal cord injury has led to the development of multimodal protective techniques, and search for methods to detect the ischemic damage in which neurons are non-functional but may still be salvageable by reperfusion. Monitoring of evoked potentials is thought to be effective but not completely sufficient in detecting all patients with a vulnerable spinal cord. Motor evoked potentials has been suggested to be more specific and sensitive than SSEP due to the specific monitoring of the anterior horn grey matter [3,4] but multilevel SSEP are on the other hand suggested to more easily pinpoint precisely the type of spinal cord ischemia [2]. In this study SSEP appropriately signaled spinal cord ischemia in two patients leading to a change in the surgical plan. In the other patients with signal changes, the SSEP changes could be attributed to other mechanisms due to the ability of multilevel recordings to distinguish alterations originating from spinal cord ischemia from those resulting from peripheral nerve ischemia [2]. Three patients that developed delayed paraplegia, however, were not identified with intraoperative SSEP monitoring.
Earlier data show that the use of the adjuncts (distal aortic perfusion and CSF drainage) reduces the overall incidence of neurological deficit [11,12] with a proportional increase of delayed [13]. A patient who would have developed immediate neurological deficit previous to the use of adjuncts might now recover from repair with intact but vulnerable neurological function. The exact mechanisms for delayed paraplegia are not yet fully understood but it is reasonable to argue that it is due to partial/temporary spinal cord ischemia making the cord vulnerable. Delayed paraplegia may be reversed if identified early and treated promptly with therapeutic measures such as CSF drainage and optimization of blood pressure and oxygen delivery [13].
Postoperative factors that may incite delayed paraplegia include hypotension, systemic inflammatory response syndrome, sepsis, cardiac dysrhythmias or cardiac failure, and diminished oxygen delivery caused by anemia, hypoxia, and low cardiac output.
Matsumoto et al. [14] have shown in rabbits that astrocyte activity with increased GFAp production in the grey substance of the spinal cord correlates with delayed paraplegia. In their model histological evaluation of spinal cord sections were made 2, 4, 8, 12, 24, 48 h after 25 min spinal cord ischemia. They saw in animals with high risk of delayed paraplegia an increase in GFAp immunoreactivity in the gray matter which became intense after 24 h. They speculate that ischemia initiates the functional activation of astrocytes leading to increased GFAp synthesis which may have a protective role in neurons. This may relate to our patients with a post-ischemic vulnerable spinal cord who may have activated astrocytes with release of GFAp that if caught in the CSF could warn for delayed paraplegia. The activation and induced synthesis being intense first after 24 h correlate well with the findings in this study.
Continued evoked potential monitoring during the postoperative phase can be cumbersome and staff demanding. Biochemical markers of CNS damage are, on the other hand, readily accessible via the spinal drain which is routinely used for monitoring and reducing intrathecal pressure by CSF drainage. Each of the markers used in this study has a unique biochemical background that has been described earlier [8]. Neurofilament protein, which we did not analyze earlier, is a major structural element of neurons. Its principal role is to maintain axon caliber, neuronal size and shape [15]. The neurofilament is composed of a triplet protein of which the light subunit (NFL) is the essential component of the neurofilament core. NFL is a component of the axonal cytoskeleton. It is suggested that the concentration of the NFL in CSF might reflect axonal damage or the extent of white matter damage [10].
Spinal cord injury in patients undergoing repair of aneurysms of the descending thoracic and thoraco-abdominal aorta is reflected by increase of biochemical CSF markers GFAp, NFL and S100B. GFAp increased more dramatically than NFL and S100B and seems to be the most discriminating parameter although this study demonstrated no temporal differences. This increase was evident first in the morning after surgery, thus coming too late to be used for intraoperative monitoring even if samples were analyzed directly. Interestingly GFAp in patients with delayed paraplegia increased before or in parallel with onset of symptoms. GFAp increased 24 h before symptom debut in the patient developing paraplegia on POD 3 as did one patient with delayed paraplegia in our earlier study [8]. GFAp increased in the morning after surgery in all patients developing delayed paraplegia. This suggests that GFAp may have a clinical potential to identify patients prone to developing delayed paraplegia. Matsumoto's study discussed above provides theoretical support for its use [14]. Neither S100B nor NFL provided any additional information over GFAp.
To identify patients at risk is important since delayed paraplegia may be reversed if identified early and if immediate countermeasures are undertaken according to above. Earlier data has demonstrated that 75% of patients can recover function when the CSF drain was in place at the time of the delayed paraplegia compared to 43% when the CSF drain had to be reinserted [13]. Knowing which patients require prolonged/augmented CSF drainage is also important since the drainage itself may induce complications.
|
5. Limitations of the study
|
|---|
Both our studies indicate that monitoring of cerebrospinal proteins in CSF may also be used as general markers of CNS injury such as stroke. While this can be a confounding parameter in diagnosing the vulnerable spinal cord, it would not be a clinical limitation. The dramatic increase of all the markers studied identifies a weakness of the study protocol by not sampling CSF more frequently during the first postoperative day. Confirming the temporal relation between CSF-GFAp and symptoms is crucial, in particular if its increase precedes symptoms. This together with the fact that the patients with ‘event’ are few makes the data preliminary and do not allow defined conclusions on the impact of biochemical markers on the development of paraplegia after thoracoabdominal repair.
None of the presently used protein markers are currently assayed ‘on-line’ and we were therefore during the study not aware of the increase in GFAp in the patients who were to develop delayed paraplegia. However, faster techniques might be developed if they prove to be an attractive complement in detecting patients prone to develop delayed paraplegia. When increased levels are found the patient would retain his intrathecal catheter and remain in the intensive care unit for optimizing blood pressure, improving oxygen delivery and lowering CSF pressure by augmented CSF drainage.
It has been argued that the concentration of CNS marker in CSF could be falsely lowered in a complex relation to the amount of CSF drained [16]. Since CSF drainage was only performed when the intrathecal pressure was above 10 mmHg, and/or when the patients were symptomatic the amount of CSF drained varied between patients potentially influencing the results. However in this study we found no significant difference in the total amount CSF drained in the group of patients with spinal cord symptoms and the patients without symptoms (Table 1) and our results should not be affected.
|
6. Conclusions
|
|---|
Peroperative multilevel cervical SSEP has a high specificity in detecting spinal cord ischemia but does not identify all patients with a postoperative vulnerable spinal cord. GFAp, NFL and S100B increase in CSF with peroperative spinal cord injury but the increase is evident too late to change the surgical procedure even if samples were analyzed directly. GFAp is, however, promising in identifying the patients with a vulnerable spinal cord and the risk of delayed paraplegia. It should be measured during the postoperative period and in cases with evident increase, hemodynamic parameters should be optimized by increasing the blood pressure and hemoglobin level and retaining the intrathecal drain for increased CSF drainage.
|
References
|
|---|
- Safi HJ, Miller III CC, Huynh TT, Estrera AL, Porat EE, Winnerkvist AN, Allen BS, Hassoun HT, Moore FA. Distal aortic perfusion and cerebrospinal fluid drainage for thoracoabdominal and descending thoracic aortic repair: ten years of organ protection. Ann Surg 2003;238(3):372-380.[Medline]
- Guerit JM, Dion RA. State-of-the-art of neuromonitoring for prevention of immediate and delayed paraplegia in thoracic and thoracoabdominal aorta surgery. Ann Thorac Surg 2002;74(5):S1867-S1869.[Abstract/Free Full Text]
- Dong CC, MacDonald DB, Janusz MT. Intraoperative spinal cord monitoring during descending thoracic and thoracoabdominal aneurysm surgery. Ann Thorac Surg 2002;74:S1873-S1876.[Abstract/Free Full Text]
- Jacobs MJ, Mess W, Mochtar B, Nijenhuis RJ, Statius van Eps RG, Schurink GW. The value of motor evoked potentials in reducing paraplegia during thoracoabdominal aneurysm repair. J Vasc Surg 2006;43:239-246.[CrossRef][Medline]
- van Dongen EP, ter Beek HT, Schepens MA, Morshuis WJ, Haas FJ, de Boer A, Boezeman EH, Aarts LP. The relationship between evoked potentials and measurements of S-100 protein in cerebrospinal fluid during and after thoracoabdominal aortic aneurysm surgery. J Vasc Surg 1999;30:293-300.[CrossRef][Medline]
- Kunihara T, Shiiya N, Yasuda K. Changes in S100 beta protein levels in cerebrospinal fluid after thoracoabdominal aortic operations. J Thorac Cardiovasc Surg 2001;122:1019-1020.[Free Full Text]
- Lases EC, Schepens MA, Haas FJ, Aarts LP, ter Beek HT, van Dongen EP, Siegers HP, van der Tweel I, Boezeman EH. Clinical prospective study of biochemical markers and evoked potentials for identifying adverse neurological outcome after thoracic and thoracoabdominal aortic aneurysm surgery. Br J Anaesth 2005;95(5):651-661.[Abstract/Free Full Text]
- Anderson RE, Winnerkvist A, Hansson LO, Nilsson O, Rosengren L, Settergren G, Vaage J. Biochemical markers of cerebrospinal ischemia after repair of aneurysms of the descending and thoracoabdominal aorta. J Cardiothorac Vasc Anesth 2003;17(5):598-603.[CrossRef][Medline]
- Rosengren LE, Wikkelso C, Hagberg L. A sensitive ELISA for glial fibrillary acidic protein: application in CSF of adults. J Neurosci Methods 1994;51:197-204.[CrossRef][Medline]
- Rosengren LE, Karlsson J-E, Karlsson J-O, Persson LI, Wikkelsø C. Patients with amyotrophic lateral sclerosis and other neurodegenerative diseases have increased levels of neurofilament protein in CSF. J Neurochem 1996;67:2013-2018.[Medline]
- Estrera AL, Miller III CC, Huynh TT, Porat E, Safi HJ. Neurologic outcome after thoracic and thoracoabdominal aortic aneurysm repair. Ann Thorac Surg 2001;72:1225-1230.[Abstract/Free Full Text]
- Coselli JS, Lemaire SA, Koksoy C, Schmittling ZC, Curling PE. Cerebrospinal fluid drainage reduces paraplegia after thoracoabdominal aortic aneurysm repair: results of a randomized clinical trial. J Vasc Surg 2002;35:631-639.[CrossRef][Medline]
- Estrera AL, Miller III CC, Huynh TT, Azizzadeh A, Porat EE, Vinnerkvist A, Ignacio C, Sheinbaum R, Safi HJ. Preoperative and operative predictors of delayed neurologic deficit following repair of thoracoabdominal aortic aneurysm. J Thorac Cardiovasc Surg 2003;126(5):1288-1294.[Abstract/Free Full Text]
- Matsumoto S, Matsumoto M, Yamashita A, Ohtake K, Ishida K, Morimoto Y, Sakabe T. The temporal profile of the reaction of microglia, astrocytes, and macrophages in the delayed onset paraplegia after transient spinal cord ischemia in rabbits. Anesth Analg 2003;96(6):1777-1784.[Abstract/Free Full Text]
- Lasek RJ. Studying the intrinsic determinants of neuronal form and function. In: Lasek RJ, Black MM, editors. Intrinsic determinants of neuronal form and function. New York: Alan R. Liss; 1988. pp. 1-60.
- Shore PM, Thomas NJ, Clark RS, Adelson PD, Wisniewski SR, Janesko KL, Bayir H, Jackson EK, Kochanek PM. Continuous versus intermittent cerebrospinal fluid drainage after severe traumatic brain injury in children: effect on biochemical markers. J Neurotrauma 2004;21:1113-1122.[CrossRef][Medline]
Top
Abstract
1. Introduction
