Clinical Chemistry 47: 1458-1462, 2001;
(Clinical Chemistry. 2001;47:1458-1462.)
© 2001 American Association for Clinical Chemistry, Inc.
Excitatory Amino Acids in Cerebrospinal Fluid of Patients with Acute Head Injuries
Zhang Hong1,
Zhang Xinding2,
Zhang Tianlin3 and
Chen Liren1a
1
Chromatographic Center, Lanzhou Institute of Chemical Physics Chinese Academy of Sciences, 342 Tianshui Rd., Lanzhou, China 730000.
2
Department of Neurosurgery, The Second Affiliated Hospital of Lanzhou Medical College, 80 Cuiyingmen, Lanzhou, China 730030.
3
Department of Neurosurgery, The First Affiliated Hospital of Lanzhou Medical College, 1 Donggang West Rd., Lanzhou, China 730000.
aAuthor for correspondence. E-mail lzwzh{at}public.lz.gs.cn
 |
Abstract
|
|---|
Background: The excitatory amino acids (EAAs) glutamate (Glu) and aspartate (Asp) play a role in the pathogenesis of postischemic and posttraumatic brain insult. The changes of EAAs in cerebrospinal fluid (CSF) of patients with traumatic brain injury are incompletely understood.
Methods: We used reversed-phase HPLC with precolumn derivatization with o-phthalaldehyde and fluorescence detection to measure Glu and Asp in CSF of 42 patients with acute head injury and 9 control adults without neurologic diseases. We assessed the Glasgow Coma Scale (GCS) on admission, the main lesion patterns on computed tomography (CT) scan within 24 h post trauma, and the Glasgow Outcome Scale (GOS) 3 months post injury.
Results: The mean concentrations of Glu and Asp in CSF in the brain-injured group were significantly higher than those of the control group (Glu, P <0.001; Asp, P <0.05). In patients admitted within 24 h after severe injury (n = 13), peak Glu values appeared within 48 h in 11 patients (85%), and the mean value remained higher than control values at day 7 (P <0.02). The concentrations of EAAs were higher in patients with severe injuries (GCS 
8) than in those with milder injuries (Glu, P <0.001; Asp, P <0.05). GCS and peak EAAs correlated negatively (Glu, rs = -0.5706, P <0.001; Asp, rs = -0.5503, P <0.001). The patients with focal brain contusion on initial CT scan (n = 8) had a significantly lower peak Glu value than the patients with other patterns (n = 8–15; P <0.02 to 0.001). The peak value of EAAs in the poor-outcome group (including severe disability, vegetative state, and death) was significantly higher than in the good-outcome group (good recovery and moderate disability; Glu, P <0.001; Asp, P <0.01); GOS was closely correlated to the EAA values (Glu, rs = 0.5737, P <0.001; Asp, rs = 0.5470, P <0.001).
Conclusions: The EAA concentrations in CSF increase after acute head injury and remain higher for at least 1 week post injury in severely injured patients. The more severe the trauma, the more obvious the excitotoxicity induced by EAAs and the worse the outcome.
|
Introduction
|
|---|
The excitatory amino acids (EAAs)
1
glutamate (Glu) and aspartate (Asp) usually function as neurotransmitters in the central nervous system, but are known to be neurotoxic at supraphysiologic concentrations (1)(2) and appear to play an important role in experimental traumatic brain injury (TBI) (3)(4)(5). In humans, however, determination of alterations in extracellular EAAs is difficult because conducting microdialysis techniques is ethically problematic. The cerebrospinal fluid (CSF), however, bathes the brain and spinal cord, and variations in the concentrations of EAAs in CSF may reflect the changes occurring in central extracellular fluid (6). In the present study, we measured EAAs in CSF after acute head injury to clarify the relationship of EAAs with clinical features and outcome.
|
Materials and Methods
|
|---|
analytical methods
We used a Perkin-Elmer HPLC system (Model 1022) with a gradient solvent-delivery system (Series-200), a fluorescence detector (Model 240), and a 200 x 4 mm column with 7-µm Nucleosil ODS stationary phase (National Chromatographic Center, Dalian Institute of Chemical Physics, Chinese Academy of Sciences). The excitation and emission wavelengths were 330 and 445 nm, respectively.
The o-phthalaldehyde (OPA; Fluka) was of analytical reagent grade, the mercaptoethanol was chemistry-purity grade (No. 4 Reagent Plant of Shanghai), and the methanol was HPLC grade. "Ultra-pure" water was prepared with an E-pure system (Barnstead Thermolyne). All other reagents were of analytical reagent grade. The amino acid calibration solution was prepared by dissolving 2.46 µmol/L aspartic and 4.16 µmol/L glutamate (No. 3 Reagent Plant of Shanghai) in 0.5 mol/L HCl. Derivatization reagent was prepared by dissolving 50 mg of OPA in 2.5 mL of methanol and adding 50 µL of mercaptoethanol; the mixture was kept at 4 °C in the dark.
All HPLC solvents were filtered through 0.45 µm (pore size) filters and degassed. Mobile phase A was tetrahydrofuran–methanol–0.1 mol/L sodium acetate (pH 7; 5:95:900 by volume); mobile phase B was methanol. The gradient system was: 0 min, 0% mobile phase B, increased to 25% B at 5 min, 35% B at 10 min, 55% B at 15 min, and 95% B at 19 min, and held at 95% B until 22 min. The flow rate was 1.0 mL/min.
Methanol (100 µL) and 100 µL of CSF sample or calibrator were added to a polyethylene microcentrifuge tube and centrifuged at 15 000g for 2 min. A 50-µL aliquot of the supernatant was removed and mixed with 50 µL of OPA and 450 µL of borate buffer (0.4 mol/L; pH 9.25); 15 µL of the mixture was injected into the HPLC 2 min later.
clinical information
The research project was approved by the Ethics Committee. We studied 42 adults (35 men and 7 women; average age, 35 years; age range, 16–70 years) within 24 h (32 cases) or 24–48 h (10 patients) after head injury. All patients denied having a history of cerebrovascular diseases and epilepsy. The main lesion patterns on computed tomography (CT) scan within 24 h post injury were focal contusion (8 cases), diffuse brain injury (8 cases), extradural hematoma (11 cases), and subdural hematoma (15 cases). Twenty-six patients underwent craniotomy, 1 received continuous ventricular drainage, and 15 had nonsurgical treatment. Controls (9 cases) were adults with lumbar anesthesia but without neurologic diseases. Severity of injury was based on the admission Glasgow Coma Scale (GCS) result and the outcome on the Glasgow Outcome Scale (GOS) 3 months post trauma.
collection and storage of csf samples
CSF samples (2–8 mL) were collected every 24 h via a ventricular intracranial pressure monitoring catheter in 27 patients who underwent craniotomy or CSF drainage, and via lumbar puncture in 15 cases with focal contusion or diffuse brain injury, but without significant midline shift or perimesencephalic cistern compression on CT scan. In controls, CSF samples were also collected via lumbar puncture during lumbar anesthesia. All samples were centrifuged (5000g for 20 min) within 30 min and stored at -20 to -25 °C immediately.
statistical analysis
The distributions of Glu and Asp in patients’ CSF were log-normal. The natural logarithm transformation value of the data was analyzed by t-test. The correlation of EAAs with GCS and GOS was evaluated by the Spearman rank correlation. We used analysis of variance to compare differences among groups and between two groups of concentrations of EAAs in CSF stored for various times.
|
Results
|
|---|
determination of EAAs in csf
The chromatograms of Asp and Glu, including glutamine and glycine, are shown in Fig. 1
. The chromatographic peaks in the samples were quantified by peak area measurement and compared with the peak areas of the calibration solution.
View larger version (10K):
[in this window]
[in a new window]
|
Figure 1. Chromatograms of Asp and Glu in the calibration solution (A) and a CSF sample (B).
Peaks: 1, Asp; 2, Glu; 3, glutamine; 4, glycine.
|
|
To evaluate the between-run variation of the method, the same CSF sample was measured seven times during 2 weeks. The CVs for Asp and Glu were 7.3% and 5.3%, respectively. For within-assay reproducibility studies, we assayed the same CSF sample five times in one analytical run. The CVs for ASP and Glu were 5.8% and 5.5%, respectively. The mean analytical recovery was 97.9% (SD, 4.8%; n = 3) for Asp and 98.7% (SD, 4.3%; n = 3) for Glu, as estimated by adding known quantities of amino acids to the CSF samples. The calibration curve for Asp was linear at 0.38–12.3 µmol/L [y = 601 448 + 342 389x (r = 0.9971), where x is the concentration and y is the peak area], whereas for Glu, it was linear at 0.65–20.8 µmol/L [y = 742 829 + 1 262 050x (r = 0.9986)].
In this study, the Glu concentrations were lower than those reported in most previous studies (6)(7)(8). During our work, we found that the chromatogram for the glutamine calibrator prepared immediately in 0.5 mol/L HCl displayed a single glutamine peak (Fig. 2
), but after storage for 6 days at 4 °C, another peak appeared (Fig. 2
). The new peak was identified as Glu by the retention time and ultraviolet spectrum (Fig. 3
). This result implied that glutamine was hydrolyzed under acidic conditions. Previous observations also indicated that Glu concentrations would increase if trichloroacetic or sulfosalicylic acid was used as the CSF deproteinating agent (7). Accordingly, we used the method for storing CSF samples reported by Baker et al. (6). We added 250 µL of 0.3 mol/L perchloric acid to 1.0 mL of CSF and stored the mixture at -20 °C until assay. We collected 100 mL of CSF via drainage from one adult with acute head injury and centrifuged (5000g for 10 min) the fluid immediately. The supernatant was divided into 15 tubes for storage. The EAA concentrations in CSF were measured in five tubes at 4, 8, and 12 weeks, and the results of analysis of variance for the EAA concentrations at different storage times are shown in Table 1
. The among-group difference in Glu concentrations in CSF stored for 4, 8, and 12 weeks in 0.3 mol/L perchloric acid was highly significant [F > F1–0.01 (3, 16); P <0.01]. The differences in Asp concentrations in CSF stored for 4, 8, and 12 weeks in 0.3 mol/L perchloric acid was not significant [F < F1–0.05 (3, 16); P >0.05]. After 4, 8, and 12 weeks in 0.3 mol/L perchloric acid, the mean concentration of Glu in CSF increased by 14.8 µmol/L (207%), 28 µmol/L (302%), and 46 µmol/L (433%), respectively (Table 1
). Thus, 0.3 mol/L perchloric acid can lead to erroneously high measured values for Glu in CSF in vitro, and acidic reagents are not suitable as deproteinating agents in the measurement of EAAs in CSF.
View larger version (6K):
[in this window]
[in a new window]
|
Figure 2. Chromatograms of freshly prepared glutamine calibrator in 0.5 mol/L HCl (A), glutamine calibrator in 0.5 mol/L HCl after storage for 6 days at 4 °C (B), and Glu calibrator (C).
Peaks: 1, glutamine; 2, new peak (tR = 10.41 min); 3, Glu (tR = 10.49 min).
|
|
View larger version (8K):
[in this window]
[in a new window]
|
Figure 3. Ultraviolet spectra of the glutamate calibrator peak (A) and the new peak shown in Fig. 2
(B).
|
|
View this table:
[in this window]
[in a new window]
|
Table 1. Concentrations and analysis of variance of Asp and Glu in CSF1 samples stored in 0.3 mol/L perchloric acid at 20°C.
|
|
EAAs in csf
The concentrations of Glu and Asp (8.0 ± 6.3 and 1.9 ± 0.8 µmol/L; n = 42) in CSF from patients with traumatic brain injury were significantly higher than those in the controls (Glu, 1.9 ± 1.5 µmol/L, P <0.001; Asp, 1.4 ± 0.2 µmol/L, P <0.05; n = 9). Among the 26 patients with severe head injuries, 13 were admitted within 24 h post trauma, and their CSF samples were obtained serially over 3–7 days. Peak Glu values appeared within 48 h post injury in 11 of the 13 patients (85%) and after 48 h in 2. With improvement or stabilization of neurologic status, Glu concentrations decreased progressively in 11 patients (85%; Fig. 4
), but remained higher than in the controls on day 7 (P <0.02). A second increase in Glu and Asp was seen in a patient with acute subdural hematoma with epileptic seizure at day 4 post trauma and in a patient with diffuse brain injury with diffuse brain swelling at day 7.
eaa concentrations and severity of brain injury
The peak EAA concentrations (Glu, 10.3 ± 6.6 µmol/L; Asp, 2.4 ± 1.2 µmol/L; n = 26) were higher in the severely injured patients (GCS 8) than in the mildly injured ones (Glu, 4.3 ± 3.7 µmol/L, P <0.001; Asp, 1.7 ± 0.6 µmol/L, P <0.05; n = 16). GCS and peak EAA values were negatively correlated (Glu, rs = -0.5706, P <0.001; Asp, rs = -0.5503, P <0.001). The patients with focal brain contusion on initial CT scan (n = 8) had a significantly lower peak Glu value (2.8 ± 2.0 µmol/L) than patients with diffuse brain injuries (9.7 ± 7.3 µmol/L; P <0.02; n = 8), extradural hematoma (9.6 ± 4.6 µmol/L; P <0.001; n = 11), and subdural hematoma (8.6 ± 7.3 µmol/L; P <0.001; n = 15).
eaa concentrations and clinical outcome
EAA concentrations were higher in the poor-outcome group than the in good-outcome group (Glu, P <0.001; Asp, P <0.01; Table 2
). GOS was correlated with the EAA concentrations (Glu, rs = 0.57, P <0.001; Asp, rs = 0.55, P <0.001).
|
Discussion
|
|---|
Under physiologic conditions, large amounts of EAAs are stored in intracellular compartments; the EAA concentrations in the cytoplasm of brain cell are conspicuously higher than those in the extracellular space, a gradient maintained by acidic amino acid carriers present in both presynaptic and glial plasma membranes. Because Glu is the major excitatory transmitter in the central nervous system, its uptake is essential for terminating postsynaptic action. The efficacy of these carriers is dependent on the presence of an external Na+ and internal K+ gradient across the plasma membrane (9)(10). However, excessive increases in extracellular Glu and Asp have been demonstrated after ischemia, hypoxia, prolonged seizures, and TBI and are believed to mediate excitotoxicity by acting as agonists at the N-methyl-d-aspartate and/or amino-3-hydroxy-5-methyl-4-isoxazole propionic acid/kainic acid (AMPA-KA) receptor-gated ion channel, leading to cellular edema and accumulation of intracellular Ca2+ and Na+, with subsequent lethal and sublethal excitotoxic effects (1)(2)(4).
It has been suggested that concussive brain injury triggers widespread neuronal depolarization, causing release of Ca2+-dependent Glu from synaptosomes and further depolarization, thus amplifying the toxic effects of EAAs (1). Ischemic brain damage is a common pathologic change in patients with severe head injuries (4)(11). The correlation of extracellular Glu with the extent of cerebral blood flow reduction has been demonstrated in animal TBI models (12). Ischemic Glu release probably originates from either ischemic depolarization or nonpresynaptic vesicles caused by the dysfunction of Glu carriers and an imbalance of transmembrane Na+/K+ gradients (9)(10). Increased extracellular K+ can reverse the uptake of Glu. Correlated increases of extracellular K+ and Glu have also been observed in a TBI model (13).
In the present study, the peak EAA values in CSF were obviously increased in patients with acute head injury, which was consistent with the results in microdialysis experiments after TBI, in agreement with the results of Baker et al. (6) in humans, indicating that there is secondary excitotoxic brain damage induced by EAAs in clinical TBI. Previous studies showed that the increases in EAAs in extracellular fluid in TBI models is transient, lasting only a few hours (5)(10)(12). However, in contrast to experimental TBI models, our study confirmed that the increase of Glu in CSF persists for at least 7 days in patients with severe TBI. Furthermore, second peak values for Glu and Asp in CSF were shown in two patients with neurologic deterioration. These findings indicated that there are either some factors (such as traumatic depolarization, imbalance of ionic gradients, and breakdown of the blood–brain barrier) related to the primary increases in EAAs in extracellular fluid, and others (such as delayed cerebral ischemia, hypoxia, epileptic seizure, hyperthermia, and secondary structural damage) associated with the secondary increase in EAAs. The above results also confirmed the hypothesis that a wide time window exists for the use of antagonists to reduce excitotoxic brain damage induced by EAAs in clinical TBI.
Faden et al. (5) discovered that the concentrations of EAAs in extracellular fluid are related to traumatic severity in a rat TBI model. In our study, Glu and Asp increased significantly in patients with severe head injuries compared with patients with mild head injuries. An obvious negative correlation was found between EAA concentrations and GCS, indicating that the more severe the traumatic status, the higher the EAAs in CSF and the more serious the excitotoxic brain damage. CT scans demonstrated intracranial morphologic changes. In this study, the alteration of Glu in CSF coincided with the main lesion patterns on CT, which suggested that the more severe the structural changes are, the higher is the peak Glu value. The data also showed the consistency between clinical features and biochemical indices as well as neuroradiologic findings. Therefore, measurement of EAAs in CSF might be an index available for judging the severity of the trauma.
In conclusion, our study indicated that EAA concentrations in the CSF of the poor-outcome group were significantly higher than the concentrations in the CSF of the good-outcome group and that GOS was closely correlated with EAA concentrations in this series of patients. Higher EAA values in the CSF correlate with a worse outcome. Because the EAA concentrations in CSF could demonstrate both the severity of TBI and the degree of secondary ischemic brain damage (12) and because EAAs themselves are factors inducing secondary brain insult, the relationship with the outcome is natural.
|
Acknowledgments
|
|---|
We thank Prof. Zhai Suodi and Li Jiaren for expert scientific contributions.
|
Footnotes
|
|---|
1 Nonstandard abbreviations: EAA, excitatory amino acid; Glu, glutamate; Asp, aspartate; TBI, traumatic brain injury; CSF, cerebrospinal fluid; OPA, o-phthalaldehyde; CT, computed tomography; GCS, Glasgow Coma Scale; and GOS, Glasgow Outcome Scale. 
|
References
|
|---|
-
Meldrum B. Amino acids as dietary excitotoxins: a contribution to understanding neurodegenerative disorders. Brain Res Rev 1993;18:293-314.[Medline]
[Order article via Infotrieve]
-
Albin RL, Greenamyre JT. Alternative excitotoxin hypotheses. Neurology 1992;2:733-738.
-
Bullock R, Fujisawa H. The role of glutamate antagonists for the treatment of CNS injury. J Neurotrauma 1992;9(Suppl 2):S443-S462.
-
Chen MH, Bullock R, Graham DI, Miller JD, McCulloch J. Ischemic neuronal damage after acute subdural hematoma in the rat: effects of pretreatment with a glutamate antagonist. J Neurosurg 1991;74:944-950.[ISI][Medline]
[Order article via Infotrieve]
-
Faden AI, Demediuk P, Panter SS, Vink R. The role of excitatory amino acids and NMDA receptors in traumatic brain injury. Science 1989;244:798-800.[Abstract/Free Full Text]
-
Baker AJ, Moulton RJ, MacMillan VH, Shedden PM. Excitatory amino acids in cerebrospinal fluid following traumatic brain injury in humans. J Neurosurg 1993;79:369-372.[ISI][Medline]
[Order article via Infotrieve]
-
Goldsmith RF, Eari JW, Cunningham AM. Determination of
-aminobutyric acid and other amino acids in cerebrospinal fluid of pediatric patients by reversed-phase liquid chromatography. Clin Chem 1987;33:1736-1740.[Abstract/Free Full Text]
-
Ferraro TN, Hare TA. Free and conjugated amino acids in human CSF: Influence of age and sex. Brain Res 1985;338:53-60.[ISI][Medline]
[Order article via Infotrieve]
-
Amara SG. A tale of two families. Nature 1992;360:420-421.[Medline]
[Order article via Infotrieve]
-
Obrenovith TP. Origins of glutamate release in ischaemia. Acta Neurochir (Wien) 1996;66(Suppl):50-55.
-
Graham DI, Ford I, Adams JH, Doyle D, Teasdale GM, Lawrence AE. Ischaemic brain damage is still common in fatal non-missile head injury. J Neurol Neurosurg Psychiatry 1989;52:346-350.[Abstract/Free Full Text]
-
Bullock R, Butcher SP, Chen MH, Kendall L, McCulloch J. Correlation of the extracellular glutamate concentration with extent of blood flow reduction after subdural hematoma in the rat. J Neurosurg 1991;74:794-802.[ISI][Medline]
[Order article via Infotrieve]
-
Katayama Y, Becker DP, Tamura T, Hovda DA. Massive increases in extracellular potassium and the indiscriminate release of glutamate following concussive brain injury. J Neurosurg 1990;73:889-900.[ISI][Medline]
[Order article via Infotrieve]
The following articles in journals at HighWire Press have cited this article:
|
|
|
|
 
R. Sethuraman, T. L. Lee, and S. Tachibana
Simple Quantitative HPLC Method for Measuring Physiologic Amino Acids in Cerebrospinal Fluid without Pretreatment
Clin. Chem.,
March 1, 2004;
50(3):
665 - 669.
[Full Text]
[PDF]
|
|
|
|
|
|
|
G. Franz, R. Beer, A. Kampfl, K. Engelhardt, E. Schmutzhard, H. Ulmer, and F. Deisenhammer
Amyloid beta 1-42 and tau in cerebrospinal fluid after severe traumatic brain injury
Neurology,
May 13, 2003;
60(9):
1457 - 1461.
[Abstract]
[Full Text]
[PDF]
|
|
|
Top
Abstract
Introduction
