470. Blood-induced, but volume-independent changes of brain edema, tissue death and glucose metabolism after acute subdural hematoma
B. Alessandri1, H. Bächli2, A. Heimann1, M. Behzad1, M. Schreckenberger3, H.-G. Buchholz3 and O. Kempski1
1Institute for Neurosurgical Pathophysiology, University Medicine, Johannes Gutenberg-University of Mainz, Mainz, Germany; 2Department of Neurosurgery, University Clinics of Basel, Basel, Switzerland; 3Department of Nuclear Medicine, University Medicine, Johannes Gutenberg-University of Mainz, Mainz, Germany
Objective: Traumatic brain injury is often accompanied by acute subdural hemorrhage (ASDH). Underneath the ASDH ischemic and metabolic changes are induced which worsen outcome of patients dramatically. Despite early surgical intervention, i.e. evacuation of the blood volume, many patients die or remain severely disabled. Besides the blood volume, additional factors from extravasated blood may play a role in lesion development. In order to separate blood from volume induced effects:, we compare the effects of a subdurally infused blood volume with the same volume of paraffin oil (no blood components) on their effects on acute and chronic pathophysiological parameters.
Method: Sprague-Dawley rats were anesthetized with chloral hydrate and a catheter was placed in the tail artery and jugular vein. A craniotomy was prepared for the insertion of a L-shaped, blunted needle into the subdural space and closed with acrylic glue and dental cement. ASDH was produced by infusion of 300 μL unheparanized autologous venous blood or paraffin oil (50 μL/min) into the subdural space. Three different experiments were carried out:
Effects on MAP, ICP, CPP during 40 min and on histological damage 48 h after subdural infusion of blood or paraffin oil (n = 8/group).
Acute and chronic effects of blood or paraffin oil on brain edema: Brains were removed either 2, 24 or 48 h after subdural infusion. Brain water content was measured by the wet-dry weight method (n = 6-8/group).
Acute glucose metabolism by means of PET using [18F]-deoxyglucose: FDG activity was measured 2 h after blood or paraffin oil infusion.
Values were normalized for injected FDG activity and body weight. Values were analyzed either by calculating an ipsi-to-contralateral ratio of cortical and subcortical regions of interest (ROI) or by categorizing all voxels into low-high normalized FDG activity.
Results: Blood and paraffin oil produced almost identical changes of MAP, ICP, CPP and CBf during the 40 min monitoring after subdural infusion. However, 2 h after injury glucose metabolism, was significantly reduced by blood in the cortical, but not in the subcortical ROI. Voxel-by-voxel analysis also showed significantly more voxels with lower activity in the blood infusion group than in the paraffin oil group. Ipsilateral brain water content was not affected at 2 h, but differred significantly between groups at 24 h ((80.3%±1.2%) versus 79.0%±0.2%, P<0.05) and at 48 h (80.7%±1.2% vs. 78.9%±0.5%), P<0.05 post-injury. Blood produced also a significantly larger brain damage (48.1±23.0 mm3 vs. 21.1±11.8 mm3, P<0.02).
Conclusion: Despite the same volume of blood and paraffin oil produced similar acute changes in ICP, CPP and CBF glucose metabolism underneath the infusion site was more reduced with blood than with paraffin oil at 2 h post-injury followed by larger brain swelling and lesion volume 48 h after injury. Thus, lesion development after ASDH is not only determined by the blood volume, but in part also by blood constituents.
132. Stable isotope comparison of cerebral lactate and glucose metabolism in resting normal control cohorts and traumatically brain injured patients
T. Glenn1, M. Horning2, G. Brooks2 and N. Martin1
1Neurosurgery, University of California, Los Angeles, Los Angeles; 2Integrative Biology, University of California, Berkeley, Berekely, California, USA
Objectives: Traditional clinical hallmarks of severe traumatic brain injury (TBI) include reduced cerebral blood flow and cerebral metabolic rates for oxygen. Recent clinical studies of cerebral metabolism based on arterial-jugular venous difference measurements have shown that the brain takes up lactate during the acute period after traumatic brain injury (Glenn, 2003). Additionally, initial studies in TBI patients using [3–13C]lactate infusion showed that the brain oxidized the lactate (Glenn, 2007). To further elucidate the role of post-traumatic cerebral metabolism a normal volunteer cohort was studied in addition to TBI patients. We hypothesized that the injured brain will take up and oxidize more lactate than in the normal cohort.
Methods: Four severely injured TBI (age = 37.9±11.2, all male) and 3 normal resting subjects (age = 31.3±5.6, 2 male, 1 female) were consented for infusion studies. All TBI patients were simultaneously infused on days 5.86±2.0 with [6,6-D2]glucose and [3–13C]lactate followed the next day by [6,6-D2]glucose [1–13C]glucose. Normal subjects were infused with [6,6-D2]glucose and [3–13C]lactate. On all subjects, arterial and jugular venous blood samples, and expired breath were collected every 30 min for 3 h. Isotopic enrichments for plasma samples were analyzed by GC/MS and breath IE's were analyzed by GC/IRMS. Cerebral blood flow was measured by the bedside 133Xenon clearance technique.
Results: Cerebral blood flow was significantly lower in TBI patients than in Normals (33.0±8.6 and 46.0±7.8 ml/100g/min, respectively, P = 0.03). Mean arterial and jugular isotopic enrichments for lactate were significantly lower (P = 0.047 and P = 0.045 respectively) in TBI (0.024%±0.013% and 0.020%±0.011%) versus Normals (0.044%±0.011% and 0.038%±0.010%, respectively). Thus, whole body lactate turnover (Ra and Rd) showed a 100% increase as opposed to whole-body glucose turnover rates that increased only 20% following TBI compared to Normals. Cerebral glucose fractional extraction was not different from zero, but lactate fractional extraction ((Ea*Ca)—(Ev*Cv)/(Ea*Ca)) was significant and similar between TBI (11.7±5.2) and Normal (14.3±4.1). As well, during days 4–6 the CMR for lactate was similar between TBI (−0.03±0.2) and Normal (−0.7±0.18 mg/100g/min). However, following TBI, cerebral lactate oxidation (= [(13CO2v)(CVCO2)]—[(13CO2a)(CaCO2)*(CBF)]/Ea) was 3-fold higher following TBI (0.9 mg/100 g/min vs 0.25 mg/100g/min in Normals).
Conclusions: The normal resting human brain can take up and oxidize lactate. Following TBI, whole-body lactate production and removal were greatly increased as is tracer-measured cerebral lactate oxidation. In conclusion, this study shows that the injured brain utilizes fuels other than glucose and suggests that a metabolic fuel based therapy may be an effective aid to recovery.
319. Mitochondrial uncoupling Protein-4 polymorphisms are associated with depth of coma after traumatic brain injury
1University of Pittsburgh School of Medicine; 2School of Nursing, University of Pittsburgh; 3Critical Care Medicine, Safar Center for Resuscitation Research U. of Pittsburgh; 4Neurological Surgery, University of Pittsburgh School of Medicine, Pittsburgh, Pennsylvania, USA
Background and aims: Mitochondrial uncoupling proteins (UCP) are a family of anion-carrier proteins that allow controlled proton leak into the mitochondrial matrix, subsequently reducing membrane potential. UCP1 is found mainly in brown adipose tissue, whereas UCP2 and UCP4 are found in mitochondria within the central nervous system. UCP are thought to serve as key mediators in mitochondrial homeostasis and possibly neuronal function and synaptic transmission,1 and may protect against oxidative stress and calcium overload after brain injury.1–4 Recently it has been demonstrated that single nucleotide polymorphisms (SNP) of UCP genes are associated with schizophrenia and multiple sclerosis.5–7 However, there are no data examining the relationship between UCP polymorphisms and traumatic brain injury (TBI), the most common cause of death in children and young adults.
Methods: DNA samples collected from 177 adult patients (mean age 34±15 y; 83% male) with severe TBI were assayed for tagging SNP of UCP2 (rs660339) and UCP4 (rs3757241 and rs9381469). For each SNP, patients were dichotomized based on allele frequencies for multivariate logistic regression analysis to determine independent associations with initial Glasgow Coma Scale score (GCS), and Glasgow Outcome Score (GOS) assigned at 3 to 12 months (the last available GOS was used), adjusting for age and sex.
Purpose: To determine whether SNP of UCP2 and UCP4 genes are associated with relevant clinical variables after TBI.
Results: Frequencies for UCP2 at rs660339 were AA (16%), AG (50%), and GG (34%); UCP4 at rs3757241 were CC (61%), CT (36%), and TT (3%); and UCP4 at rs9381469 were AA (26%), AG (48%), and GG (27%). These frequencies were similar to published data available through the International HapMap Project, Public release #26, 2008-11-26. Both UCP4 polymorphisms, but not the UCP2 polymorphism, were independently associated with initial GCS. There was no association between any SNP and GOS.
Conclusions: UCP4 is the most prevalent mitochondrial UCP in the brain. These data suggest that UCP4 polymorphisms are associated with initial depth of coma (as assessed by the GCS) after severe TBI, and are consistent with a role for UCP4 in neuronal activity1 after brain injury. Further study is warranted. Support: NINDS NS38620/NS30318.
UCP4 Polymorphism and glasgow coma scale score
UCP4 polymorphism
Inital GCS (Median [range])
Odds ratio [5–95% confidence limits]
P
rs3757241
CC, 6 [3–13] versus CT/TT, 5 [3–10]
1.243 [1.021–1.512]
0.03
rs9381469
AG/GG, 5 [3–10] versus AA, 6 [3–13]
0.818 [0.674–0.994
0.04
729. Diffuse optical monitoring of cerebral hemodynamics in piglet with traumatic brain injury
C. Zhou1, S.A. Eucker2, T. Durduran1,3, G. Yu1,4, J. Ralston2, S.H. Friess2, R.N. Ichord5, S.S. Margulies2 and A.G. Yodh1
1Physics and Astronomy; 2Bioengineering; 3Radiology, University of Pennsylvania, Philadelphia, Pennsylvania; 4Wenner-Gren Research Lab, Center for Biomedical Engineering, University of Kentucky, Lexington, Kentucky; 5Neurology, Children's Hospital of Philadelphia, Philadelphia, Pennsylvania, USA
Objectives: Traumatic brain injury (TBI) is a leading cause of childhood morbidity and mortality in the United States, with an annual rate of approximately 200 per 100,000 children requiring hospitalization.1 An understanding of the conditions of oxygen supply and consumption, and measurement of blood flow after traumatic brain injury is therefore important for clinical applications. In this study, we explored the feasibility of two diffuse optical techniques, diffuse reflectance spectroscopy (DRS2,3) and diffuse correlation spectroscopy (DCS4,5), for continuous, non-invasive measurement of cerebral blood flow and blood oxygenation through the intact scalp and skull in a newborn piglet traumatic brain injury model6 at the bedside.
Methods: Eighteen anesthetized 3–5 day old female piglets (n = 18) were used for the study. Cerebral hemodynamic changes, including relative oxy-, deoxy- and total hemoglobin concentrations (rHbO2, rHb and rTHC respectively), blood oxygen saturation (StO2) and relative cerebral blood flow (rCBF), were quantified using DRS and DCS continuously and non-invasively before injury and up to 6 h after the injury. Fluorescent microspheres (FM) were used to validate against rCBF measurements with DCS at discrete time points.
Results: A strong linear correlation was observed between the rCBF measurements with DCS and FM techniques (R = 0.89, P<0.00001). Compared to pre-injury levels, rHb, rHbO2 and rTHC increased 67%±13% (P = 0.0006), 24%±13% (P = 0.1) and 41%±11% (P = 0.005) respectively immediately after injury and continued increasing over the next six hours (Figure 1A–C). StO2 dropped significantly immediately post-injury (52%±3%, P = 0.02), maintained this reduction at 30 min post-injury (54%±2%, P = 0.009), but gradually normalized to pre-injury levels (Figure 1D). rCBF decreased significantly to 50%±6% of pre-injury level (P = 0.00003) immediately after injury and continued at that level over the six hours (Figure 1E). Diffuse optical techniques were robust for the entire six hours measured after injury, were sensitive to spontaneous, transient physiological events, such as apnea, cardiac arrest and hypertonic saline infusion and were consistent with vital physiological recordings, including mean arterial blood pressure, arterial oxygen saturation and heart rate.
Conclusions: The investigation corroborates potential of the optical methods for bedside monitoring of pediatric and adult human patients in the neuro-intensive care unit.
Average injury effects.
747. Effect of the anesthetic regimen during experimental brain trauma on histological outcome and cerebral inflammation
S. Thal, C. Ricken, K. Gierth, R. Timaru-Kast, C. Werner and K. Engelhard
Department of Anesthesiology, Johannes Gutenberg-University, Mainz, Germany
Objectives: Application of different anesthetics for 1 h following traumatic brain injury (TBI) alters the extent of neuronal injury and influences functional outcome.1 The question therefore arises, if already the use of anesthetics for a short period during induction of an experimental TBI influences secondary pathophysiological mechanisms. Knowledge about the effect of anesthetics on experimental data is essential for the selection of an appropriate anesthetic agent. The present study therefore compares the effect of two volatile and one intraperitoneal anesthesia on histological damage, functional outcome and expression of inflammatory marker genes after experimental brain trauma.
Methods: Male C57Bl/6 mice were randomly assigned to an anesthesia with 1.4 vol% isoflurane (iso), 3.5 vol% sevoflurane (sevo) or an i.p. injection (comb) of midazolam (5 mg/kg), fentanyl (0.05 mg/kg) and medetomidine (0.5 mg/kg). A pneumatic brain trauma was induced on the right parietal cortex (controlled cortical impact, CCI). 15 mins and 24 h after CCI brain contusion volume was determined in cresyl violet stained sections. mRNA expression of TNFα, IL1β, IL6, COX2 and iNOS was determined in contused brain tissue and normalized against Cyclophilin A as control gene. Neurological function was evaluated with a 10-point-neurological severity score. Statistics: ANOVA on ranks, P<0.05.
Results: 15 mins after CCI brain contusion volume was not different between the anesthesia regimens. Within 24 h histological brain damage increased significantly in all groups (15 mins: 19.4 mm3±4.5; 24 h: sevo = 45.3 mm3±9.0; iso = 31.5 mm3±4.0; comb = 44.2 mm3±6.2). Sevo and comb anesthesia resulted in a significantly larger contusion volume compared to iso. After 24 h neurological function was significantly better in animals anesthetized with iso in comparison to comb (sevo: 4.6 pts.±1.3; iso: 3.9 pts.±0.8; comb: 5.1 pts.±1.6). The expression of inflammatory marker genes increased significantly from 15 min to 24 h after CCI, but without differences between the anesthesia protocols.
Conclusion: The data demonstrate that a short anesthesia during the induction of a brain trauma has a profound impact on extent of secondary brain damage. Despite differences in histological brain damage, cerebral inflammation was similar between the anesthetic regimens. Differences in histological brain damage are therefore not mediated by variation in cerebral inflammation. Differences in post-traumatic cerebral perfusion or influence on apoptotic processes might account of the observed effects. In conclusion, in order to investigate the pathophysiological mechanisms of brain trauma selection of the anesthetic agent may influence results and interpretation of a study.
762. Prolonged cerebral hypometabolism following mild traumatic brain injury detected by 1H magnetic resonance spectroscopic imaging at 3T
B. Bartnik-Olson1, K. Tong1, S. Uffindell2, V. Wong3, S. Ashwal4 and B. Holshouser1
1Radiology; 2Neurology, Loma Linda University, Loma Linda; 3Redlands Pediatric and Adult Medicine, Redlands; 4Pediatrics, Loma Linda University, Loma Linda, California, USA
Objectives: Approximately 80% of traumatic brain injuries (TBI) are classified as mild.1 Neurocognitive deficits are estimated to occur in 50–80% of patients, which may persist for several years after injury2 even though conventional imaging findings may be normal. 1H magnetic resonance spectroscopic imaging (MRSI) offers an alternative approach to non-invasively measure cerebral metabolism; specifically, N-acetlyaspartate (NAA), representing neuronal metabolism; total creatine (Cr), indicative of energy metabolism and total choline (Cho), representing membrane metabolism. The study objective was to compare cerebral metabolism in mild TBI subjects (mTBI) with neurocognitive deficits to controls using 3D 1H MRSI. We hypothesized that mTBI subjects would have areas of hypometabolism, as defined by decreased Cr levels, which may explain cognitive deficits.
Methods: Five adult mTBI (90–1050 days post-injury) and 3 control subjects were retrospectively identified for this study. Using a Siemens Tim Trio 3T scanner, 3D MRSI (PRESS TR/TE = 1700/144 msec, voxel size 1 cm3) of a 9 cm A-P × 8 cm R-L × 6 cm S-I volume of interest (VOI) covering the centrum semiovale through the midbrain was obtained. 3D fast spin echo T2-weighted MRI (SPACE, TR/TE = 3200/458 msec, pixel size = 1 mm3) was used for VOI positioning. LCmodel was used to semi-quantitatively measure NAA, Cr, and Cho peak integrals with NAA/Cr, NAA/Cho, and Cho/Cr ratios calculated for each voxel. MRSI voxels were manually selected in normal appearing brain in the frontal grey (FG), frontal white (FW), parieto-occipital grey (POG) and parieto-occipital white (POW) matter. Voxels containing >10% CSF were excluded. Spectral findings were classified as hypometabolic if ratios were 2 standard deviations above the mean control ratio for that region due to decreased Cr but normal NAA. Pooled regional ratios were calculated for each subject. Statistical significance was measured using independent samples t-test with a significance = P<0.05.
Results: Pooled regional ratios did not differ between mTBI and controls. In contrast, 58% of FW and 49% of FG voxels in mTBI subjects showed evidence of hypometabolism as seen by increased NAA/Cr (P<0.001) and Cho/Cr (31% and 39%, respectively; voxel P<0.001). In addition, 56% of FW and 37% of FG voxels showed increased NAA/Cho (P<0.001) compared to controls. Less than 5% of voxels in the POG or POW showed an increase in any metabolite ratio.
Conclusions: To our knowledge this is the first study to report decreased Cr and Cho in mTBI adults. Decreased Cr indicates a prolonged reduction in energy metabolism and may be related to decreased perfusion reported in SPECT studies following mTBI.3 Decreased Cho may suggest altered membrane biosynthesis, resulting in decreased cholinergic function and memory impairment.4 These findings show metabolic abnormalities in normal appearing brain that may help explain cognitive deficits.
References
1.
GlennTCKellyDFBoscardinWJMcArthurDLVespaPMOertelMFHovdaDABergsneiderMHilleredLMartinN. Energy dysfunction as a predictor of outcome after moderate or severe head injury: Indices of oxygen, glucose and lactate metabolism. J Cereb Blood Flow Metab Oct; 2003; 23(10):1239–50.
2.
GlennTCWallisGHirtDVespaPDusickJBrooksGHovdaDMartinN. Lactate is oxidized by the human brain following traumatic brain injury. Brain '07, Osaka Japan BP46–3U, 2007.
3.
AndrewsZB. Nat Rev Neurosci2005;6:829–40.
4.
SullivanPG. J Bioenerg Biomembr2004;36:353–56.
5.
LiuD. Neuromolecular Med2006;8:389–414.
6.
MattiassonG. Nat Med2003;9:1062–68.
7.
YasunoK. Am J Med Genet B Neuropsychiatr Genet2007;144:250–53.
StatlerKDAlexanderHVagniVDixonCEClarkRSBJenkinsLKochanekPM. Comparison of seven anesthetic agents on outcome after experimental traumatic brain injury in adult, male rats. J Neurotrauma2006;23:97–108.
17.
EsselmanPCUmotoJM. Brain Inj1995;9:417–24.
18.
GovindarajuV. AJNR Am J Neuroradiol2005; 25:730–7.
