Metabolic crisis without brain ischemia is common after traumatic brain injury: a combined microdialysis and positron emission tomography study

Metabolic crisis without brain ischemia iscommon after traumatic brain injury:a combined microdialysis and positronemission tomography study

Paul Vespa, Marvin Bergsneider, Nayoa Hattori, Hsiao-Ming Wu, Sung-Cheng Huang,Neil A Martin, Thomas C Glenn, David L McArthur and David A Hovda

David Geffen School of Medicine at UCLA, UCLA Medical Center, University of California, Los Angeles,California, USA

Brain trauma is accompanied by regional alterations of brain metabolism, reduction in metabolicrates and possible energy crisis. We hypothesize that microdialysis markers of energy crisis arepresent during the critical period of intensive care despite the absence of brain ischemia. In all, 19brain injury patients (mean GCS 6) underwent combined positron emission tomography (PET) formetabolism of glucose (CMRglu) and oxygen (CMRO2) and cerebral microdialysis (MD) at a meantime of 36 h after injury. Microdialysis values were compared with the regional mean PET valuesadjacent to the probe. Longitudinal MD data revealed a 25% incidence rate of metabolic crisis(elevated lactate/pyruvate ratio (LPR)440) but only a 2.4% incidence rate of ischemia. Positronemission tomography imaging revealed a 1% incidence of ischemia across all voxels as measuredby oxygen extraction fraction (OEF) and cerebral venous oxygen content (CvO2). In the region of theMD probe, PET imaging revealed ischemia in a single patient despite increased LPR in otherpatients. Lactate/pyruvate ratio correlated negatively with CMRO2 (Po0.001), but not with OEF orCvO2. Traumatic brain injury leads to a state of persistent metabolic crisis as reflected by abnormalcerebral microdialysis LPR that is not related to ischemia.Journal of Cerebral Blood Flow & Metabolism advance online publication,16 February 2005; doi:10.1038/sj.jcbfm.9600073

Keywords: brain injury; ischemia; lactate; lactate/pyruvate ratio; microdialysis; positron emission tomography;pyruvate

Introduction

Traumatic brain injury (TBI) results in primarycellular death in a limited region of the braindirectly involved in the insult, while creating amore widespread state of metabolic dysfunction inremote areas of the brain (Feeney and Baron, 1986;Vink et al, 1988). This metabolic dysfunction is bestcharacterized as a reduction in oxidative metabo-lism (CMRO2) (Vink et al, 1988; Hovda et al, 1991)with alteration in glucose metabolism (Yoshino et al,1991, 1992; Kawamata et al, 2000) and has beenshown both in experimental (Hayes et al, 1988;Yoshino et al, 1992; Hovda et al, 1991; Andersen

and Maramarou, 1989) and in human brain injury(Bergsneider et al, 1997, 2000, 2001). Specifically,CMRO2 is reduced by at least 50% after TBI in theacute period. This reduction is thought to be due toa predominantly calcium-mediated impairment ofmitochondrial respiratory function as well as earlycell death. In the immediate postinjury period, thereis a compensatory increase in glucose utilization tosupply energy to membrane ionic pumps to facil-itate restoration of ionic gradients. This increase inglucose utilization has been coined hyperglycolysisand is short-lived and followed rapidly by areduction in glucose utilization. These changes inbrain metabolism occur acutely during periods ofincreased energy demand and proceed throughphasic changes initially with increased glucosemetabolism and then evolve to a subacute periodof apparent metabolic depression. The metabolicdepression is distinct from that of ischemia, sinceoxygen extraction appears to be low or normal, andlasts for days to weeks in humans. During this acute

Received 7 June 2004; revised 20 September 2004; accepted 22November 2004

Correspondence: Dr Paul Vespa, David Geffen School of Medicineat UCLA, CHS 18-218, UCLA Medical Center, University ofCalifornia, Los Angeles, CA 90095, USA.E-mail: [email protected]

Journal of Cerebral Blood Flow & Metabolism (2005), 1–12& 2005 ISCBFM All rights reserved 0271-678X/05 $30.00

www.jcbfm.com

period, secondary events such as seizures orelevated intracranial pressure can place demandson the tissue to supply additional energy. In such asetting of increased energy demand, the metabolicreserve of the tissue might be impaired.

Numerous clinical studies have documentedischemia early after TBI. The evidence of brainischemia occurring after TBI comes from patho-logical autopsy series of fatal cases of brain injuryand from hyperacute studies of cerebral blood flowand jugular venous oximetry (Bouma et al, 1991;De Deyne et al, 1996; Vigue et al, 1999). Earlypathological studies of fatal brain injuries such asthose by Graham and Adams (1971), Graham et al(1989) documented ischemic damage in 91% ofcases. This ischemic injury was severe in 27% andmoderate in 43%. Most of this ischemic damage wasnoted to occur in the hippocampus, basal ganglia,and cerebellum. The stability of these findings hasbeen shown over the last two decades (Jennett et al,2001; Kotapka et al, 1991). However, the overallprevalence of ischemia after the initial 24 h has beendifficult to document despite persistent markers ofdisturbed brain metabolism seen in microdialysismonitoring (Vespa et al, 2003).

It is in this context that the present study attemptsto determine if there is a metabolic crisis due toischemia or due to mechanisms other than ischemiaby examining a representative region of brain tissueremote from the primary injury site through the useof combined positron emission tomography andbrain microdialysis. We define metabolic crisis asthe elevation of the lactate/pyruvate ratio (LPR)440and we define ischemia by the PET criteria ofOEF40.75 or by the combined microdialysis criteriaof LPR440 and microdialysis glucose o0.2 mmol/L.In this region of interest (ROI), we prospectivelymeasure the metabolic state of the tissue andattempt to validate the usefulness of microdialysisas an indicator of ischemia in TBI.

Materials and methods

The University of California at Los Angeles institutionalreview board for human research approved of this study.This study was conducted as an integral part of the UCLABrain Injury Program on patients with severe TBI witheither GCSr8 or evidence of traumatic mass lesion oncomputerized tomographic scan and GCSr12. Subjectswere identified in the emergency room, consented byproxy and enrolled into the study as soon as possible. Themanagement of patients has been previously described(Vespa et al, 1999, 2003). Acute-phase studies wereperformed during the initial 10 days after hospitaladmission, subject to constraints including catheterremoval, the patient’s graduation from intensive care, ordeath. Determination of the injury severity score (ISS)was performed using the conventional assessment tool(Baker et al, 1974).

Cerebral microdialysis was performed using the CMA70probe (10 cm flexible shaft, 10 mm membrane length, 20 kDacutoff, CMA, Stockholm, Sweden) inserted via a twist drillburr hole adjacent to an existing ventriculostomy.The microdialysis catheter was inserted into a depth of

1.5 to 2 cm below the dura at an angle 301 lateral to thetrajectory of the ventriculostomy, to place the catheter tipinto the white matter (which was confirmed by computer-ized tomography of the hyperdense tip located in whitematter). The location was in the nondominant frontal lobe.The probe was tunneled 3 cm under the skin and secured tothe scalp with a flat profile, and then attached to theCMA103 perfusion pump. Normal saline was perfusedthrough the catheter at 2mL/min, and fluid was collected in60-min samples and then placed into dry ice or directly intothe CMA600 instrument. The initial 60-min sample was notused for analysis because this was the time allowed forstabilization of the probe. Microdialysis was not interruptedfor transport or bedside testing. In five most recent patients,subcutaneous probes were placed in the skin overlying theright lower abdominal quadrant to control for hourlysystemic changes in glucose, lactate and urea. In onepatient, two microdialysis probes were placed into thebrain, with the second probe located in a pericontusionalwhite matter location.

Positron emission tomography was performed using aquantitative method previously described (Bergsneideret al, 2001; Wu et al, 2004). Patients were placed into thescanner and physiological monitoring of ICP, end-tidalCO2, arterial blood gases, arterial blood pressure, contin-uous electroencephalography, core temperature, and heartrate wes monitored using a portable intensive care unit(ICU) monitoring system. These physiological parameterswere kept similar to those that were present in the ICU.ICP was kept under 20 mm Hg, CO2 was kept at 30 to34 mm Hg, and temperature was kept 37 to 37.61C.Patients underwent serial O-15 PET scans (C15O, O15O,H2

15O) using dynamic blood sampling to determineregional cerebral blood flow (CBF, mL/(100 g/min�1)),oxygen extraction fraction (OEF, %), and the cerebralmetabolic rate of oxygen (CMRO2, mg/(100 g/min�1)).Images of oxygen extraction coefficient (OEF) weregenerated using the initial 5 min of the 15O2 study and amethod that is based on a compartmental model foroxygen that accounts for recirculated H2

15O (Mintun et al,1984; Ohta et al, 1992). The 15O2 raw images (voxelsize: 1.471 mm� 1.471 mm� 2.45 mm) were smoothedwith a 3D filter (in plane fwhm¼ 2.942 mm, axialfwhm¼ 2.45 mm) before OEF images were generated on avoxel basis. The extracranial tissue and CSF wereexcluded in addition to contusions and hemorrhage asdescribed by Coles et al (2004). This was followed by anFDG-PET scan obtained using a quantitative technique forthe calculation of the regional cerebral metabolic rate ofglucose (CMRglc, mg/(100 g/min�1)). Using CMRO2 andCMRglc values derived from ROI analysis, the oxygen-glucose ratio (OGR, mg O2/mg glucose) was calculated.For the purpose of the present study, regional glucosemetabolic rates were determined in a 2 cm3 region of thecerebral microdialysis probe. Microdialysis probe locationwas confirmed by computerized tomography (CT) or

Metabolic crisis after brain injuryP Vespa et al

2

Journal of Cerebral Blood Flow & Metabolism (2005), 1–12

magnetic resonance image (MRI), and this location wasthen coregistered to the PET. With the MD membrane atthe center, a 1.5-cm diameter ROI was selected foranalysis. To evaluate the influence of heterogeneity ofPET measures within the area of brain containing themicrodialysis probe, we mapped the distribution of themetabolic parameter in the 2 cm3 ROI along with con-centric ROIs from 1 to 5 cm3 that used the microdialysisprobe as the epicenter. Both these procedures showedlittle variation of PET-derived mean and standard devia-tion values in the vicinity of the probe and o5%difference in the PET signal in progressively larger ROIs.Thus, 2 cm3 was chosen as the operational area forcomparison.

A detailed patient event log was kept by the bedsidenurse and research team to identify important events,and to record times of vial sampling. In addition,automated computerized capture of all physio-logical monitoring data was achieved. Sampling of thephysiology occurs every 2 min and a 1-h mean isgenerated. An experienced nurse then confirms thehourly mean values. The following data on an hourlybasis were recorded: intracranial pressure (ICP), meanarterial blood pressure (MAP), cerebral perfusion pressure(CPP), heart rate, arterial oxygen saturation, core temp-erature (jugular), jugular venous oxygen saturation(SjvO2), regional brain oxygen partial pressure (PTiO2),electroencephalography, and Glasgow coma score. Inaddition, serial measurements of dose of sedatives,mannitol, and other neurologically active medicationwere recorded for each hour.

Daily testing was performed using the radioactiveXenon-133 Kety–Schmidt technique for global measure-ment of CBF, and glucose and oxygen metabolic rates werecalculated using methods previously defined (Lee et al,2001). Matched samples of arterial and jugular bulbvenous blood samples were taken on a daily basisand matched to the corresponding hourly microdialysisglucose sample. Blood glucose levels were deter-mined using the glucose oxidase method. Intravenousinjection of radioactive Xenon-133 was performed and theglobal CBF-15 was determined. Thereafter, global ratesof glucose and oxidative metabolism were obtained.Determination of global glucose metabolism and CBFwas performed for each patient and compared with amatched cerebral microdialysis sample taken during thehour of study.

Frozen samples were thawed, then briefly centrifuged,and then analyzed on the CMA600 in batch analysis withstandard CMA600 reagents. These samples were analyzedfor glucose, lactate, pyruvate, glutamate, glycerol, andurea. These hourly samples were run twice each for eachanalyte and the mean final value was used. Qualitycontrol measurements using normal saline and waterblank samples, as well as standardized solutions across arange of concentrations (0.025 to 3.0 mmol/L) mimickingthose of the human samples, were run weekly, with anadditional internal control sample for each subject.Acceptable values of coefficient of variation (3% to 5%)and accuracy were obtained to validate very lowsample concentrations of selected analytes. Samples with

extremely low values (o0.05 mmol/L) underwent repeattesting for confirmation.

The incidence of ischemia in various brain regions onPET was determined by looking at all voxels contained inthe individual slices across the entire brain and applyingone of two sets of criteria for ischemia. For this analysis,contusions were excluded but pericontusional tissue wasincluded. In the first instance, we used the criterion ofOEF40.75 and in the second we used the criteria ofcritical oligemia proposed by Coles et al (2004). Criticaloligemia uses the concept that the lowest cerebral venousoxygen content (CvO2) in an infarction is below thethreshold of 3.5 mL/100 mL (Yundt and Diringer, 1997;Powers et al, 1985. Hence the formula for critical oligemiaused was OEFcrit¼ (CaO2–3.5)/CaO2. Using this technique,we examined on average approximately 98,300 voxels perpatient. With regard to the molar ratio of oxygen to glucosemetabolism, OGR was considered to be high if it wasgreater than the expected stoichiometric ratio of 5.8 anddecreased if r4.9.

Subjects were closely followed for recovery of functionover the next 6 months by home and outpatient officevisits. At 6 months, patients underwent in-person follow-up testing consisting of the extended Glasgow Outcomescore (GOSe) (Wilson et al, 2000). The GOSe was deemedto be stable at 6 months after injury and was used to assessthe profile of microdialysis based in each major categoryof GOSe.

General Management Protocol

All patients underwent continuous EEG (cEEG) monitor-ing in the neurosurgical ICU as part of the standardizedcare protocol. Patients requiring an initial emergencyoperation for a mass lesion were taken to operating roomwithin 1 h of admission. Intracranial pressure (ICP)monitoring, starting within 2 h of injury, was performedusing a ventriculostomy. ICP was kept below 20 mm Hgusing a stepwise management strategy (i.e. cerebrospinalfluid drainage, hyperventilation to PCO2 of 30 to 34, andhypertonic saline). Ventriculostomy drainage was used forpersistent elevation of the ICP420 mm Hg for 45 min.CSF was drained for 10 min and then the ventriculostomywas set to monitor. Cerebral perfusion pressure (CPP) waskept equal to or greater than 60 mm Hg using norepi-nephrine when required. Jugular venous oximetry (SJO2)was performed to monitor for jugular venous desaturationand for hyperemia. Blood pressure was adjusted to keepthe SJO2 between 60% to 70%. Core temperature from thejugular vein was used and kept between 371C to 37.61Cthough the use of medications (acetominophen) andsurface-cooling devices. Systemic glucose was controlledusing subcutaneous insulin to achieve a serum glucosebetween 110 and 200 mg/dL. All patients received aloading dose of phenytoin (18 mg/kg) in the emergencyroom and were continued on phenytoin 300 mg/day forseven days.

Data analyses included Pearson product–momentcorrelations, analyses of proportions, analyses ofvariance, computation of odds ratios with 95% confidence

Metabolic crisis after brain injuryP Vespa et al

3

Journal of Cerebral Blood Flow & Metabolism (2005), 1–12

intervals, and linear regression. Data acquisition washandled in Access 97 (Microsoft Corp., Redmond WA,USA), while statistical procedures were conducted withinStatistica 5.5 (StatSoft, Inc., Tulsa, OK, USA).

Results

In all, 19 subjects underwent combined microdia-lysis and PET imaging after TBI. The main diag-nostic and epidemiological indicators for eachsubject are outlined in Table 1. The patients hadsevere TBI with prolonged ICU stays. A total of 2614samples of hourly microdialysis samples wereobtained. The start time of cerebral microdialysisaveraged 12 h after injury. During the period ofmonitoring microdialysis, longitudinal continuousmonitoring for brain ischemia in the ICU wasperformed using jugular venous oximetry and dailyglobal xenon-133 cerebral blood flow measures. Theincidence of any single marker of global brainischemia was low, namely o2% (SJVO2 o50% orCBF o20 cm3/(100 g/min�1)) (Glenn et al, 2003). Inthe present cohort, ischemia occurred in one of twosituations: transient events lasting less than 60 min(6 patients) and terminal events occurring duringthe final few hours of monitoring (1 patient). Table 1summarizes the per patient incidence of increasedICP, reduced CPP, and reduced SJVO2 in this cohortof subjects. The PaCO2 ranged from 28 to 41 mm Hg,with most patients in the range from 30 to 34 mmHg. There was poor correlation between PaCO2 andCBF, OEF, CMRO2, and OGR (pearson correlationcoefficients ranging from �0.18 to 0.07).

Probe Location

MRI was performed to locate the MD probe and tocoregister the probe with the PET scan. Theterminus of the MD probes was confirmed to be inthe white matter u fibers directly adjacent to thecortical ribbon.

Longitudinal Microdialysis Study

In the longitudinal study of microdialysis markers ofischemia, the mean LPR day by day is shown inFigure 1. Overall, 25% of all LPR values exceededthe predefined critical value of 40. Elevation ofLPR440 occurred in 12/18 patients, with a meanduration of each episode of 2.470.3 h. The highestincidence of LPR440 occurred on postinjury dayzero (31% of all values). An example patient inwhom the LPR was persistently elevated despitehaving ICP, SJVO2, and CPP actively treated tomaintain values within the normal treatment rangeis illustrated in Figure 2. Based on the availableliterature of microdialysis markers of ischemia(Landolt et al, 1994; Langemann et al, 2001; Hlatkyet al, 2004) and the single patient who showedischemia during PET in Figure 3, we utilized a morestringent pattern of microdialysis to determine theincidence of ischemia. The criteria for ischemiawere simultaneous glucose o0.2 mmol andLPR440. When this combination of microdialysismarkers was applied to this cohort of patients, theincidence of ischemia in the longitudinal data set is2.4%, with a mean duration of 0.570.4 h.

Table 1 Demographic data

Subject Age(years)

iGCS Lesiontype

Surgery PETtime

SJO2 atPET

PaCO2 atPET

% timeICP420

% timeCPPo60

% timeSJO2o50

6GOSe

1 24 12 Multi Ctx N 36 56 34 28 1 0 72 27 5 tSAH n 96 56 28 0 1 0 33 15 14 SDH, Evac y 28 78 34 0 40 15 44 38 8 EDH, Multi Ctx n 60 80 41 0 6 0 85 24 3 Multi Ctx n 98 65 32 14 14 0 16 47 7 Multi Ctx n 96 75 34 0 0 0 37 29 3 Multi Ctx n 70 69 29 8 17 1 18 53 14 Multi Ctx y 20 88 33 11 1 1 59 44 7 EDH, Multi Ctx y 30 68 36 0 3 0 6

10 35 14 t SAH n 24 84 34 0 13 0 611 20 12 Multi Ctx y 72 73 33 1 1 0 712 31 8 SDH, Evac y 48 59 34 3 58 1 513 57 4 Multi Ctx y 88 81 30 1 19 1 314 41 7 Multi Ctx n 41 66 31 0 0 0 815 38 3 EDH, SDH y 39 59 39 0 0 0 316 18 3 t SAH n 115 66 31 0 0 0 617 16 3 Multi Ctx n 12 92 31 2 4 0 718 25 8 Multi Ctx y 98 54 30 11 21 1 319 31 3 Multi Ctx y 77 77 33 15 5 1 5

ICP420—the percent time within subject of elevated intracranial pressure over 20 mm Hg, CPP o60—the percent time within subject of decreased cerebralperfusion pressure, SLJO o50%—the percent time within subject of decreased jugular venous oxygen below the global ischemic level, GOSe—the 6-monthextended Glasgow outcome score for each patient. PET time—the postinjury hour during which PET was completed.

Metabolic crisis after brain injuryP Vespa et al

4

Journal of Cerebral Blood Flow & Metabolism (2005), 1–12

Evidence of Regional Ischemia Based on PositronEmission Tomography

The incidence of ischemia in various brain regionson PET was determined by looking at all voxelscontained in the individual slices across the entirebrain and applying one of two sets of criteria forischemia. The latency from injury to PET was60730 h after injury. For this analysis, CT and MRI

coregistration was performed to determine thelocation and size of hemorrhagic contusions, andareas of hemorrhage, including parenchymal andCSF within the core of these lesions, were excluded.These structural images were obtained on the day ofthe PET. However, pericontusional tissue, defined ashypodense edematous tissue, was included. In thefirst instance we used the criterion of OEF40.75,and in the second we used the criteria of criticaloligemia proposed by Coles et al (2004). Criticaloligemia uses the concept that the lowest cerebralvenous oxygen content (CvO2) in an infarction isbelow the threshold of 3.5 mL/100 mL (Yundt andDiringer, 1997; Powers et al, 1985). Hence, theformula for critical oligemia used was OEFcrit¼(CaO2–3.5)/CaO2. Using this technique, we examinedon average approximately 98,300 voxels per patient.Using the OEF 40.75 threshold, we found the meanincidence rate of ischemic voxels to be 0.1170.18%,with a maximum of 1% of voxels in a single patient.Using the critical oligemia threshold as defined byColes et al (2004), the incidence of ischemic voxelsaveraged 0.1470.28% of voxels. This translated intoa calculated mean ischemic brain volume of 1.573.1 cm3 (range 0.00 to 11.0 cm3). Separate analysiswas performed that incorporated contusions andthese data did not differ significantly from theabove analysis (data not shown). Only a singleROI containing the microdialysis probe showedischemia based on the above definitions (Table 2).In this patient, the microdialysis markers of lowglucose (glucose¼ 0.07) and LPR440 (LPR¼ 57)were found to be present.

Regional Metabolism

Microdialysis metabolites including LPR werecompared with each of the PET-derived measuresof metabolism including CBF, CMRO2, CMRglucose,and OGR. In all, 20 ROIs were examined in 19subjects, with one subject having two MD probes.These results are presented in Table 2, whichshows a wide range of microdialysis LPR values(12 to 100) and corresponding range of CMRO2

values (0.7 to 3.8). Overall only one patientshowed PET evidence of ischemia (Figure 3). Thispatient shows increased OEF without a decrease inCMRO2, suggesting viable yet ischemic tissue.Commensurate with this increase in OEF is anincrease in OGR. Correlation analysis was madebetween each metabolite and each PET parameter.LPR was negatively correlated with CMRO2

(r¼�0.44, Po0.001) (Figure 4), but not correlatedwith other PET metabolic parameters. Of note,there were strong correlations between severalPET measures and microdialysis metabolite levels:lactate vesus OGR (r¼ 0.48), lactate versus OEF(0.54) (Table 3).

In examining the PET measures, the oxygen-to-glucose ratio (OGR) was found to be abnormal in 15

0.0

20.0

40.0

60.0

80.0

100.0

120.0

140.0

160.0

PID 0 PID 1 PID 2 PID 3 PID 4 PID 5 PID 6 PID 7

LP

R o

r %

LP

R >

40

LPR% LPR> 40

Figure 1 Mean daily LPRs over the first 8 days after injury.Mean LPR values day by day are shown in solid bars. Thepercentage of LPR 440 on each postinjury day are shown inhatched bars.

MD LPR

0102030405060708090

100

13 20 27 34 41 48 55 62 69 76 83 90Post injury hour

LP

R

0

20

40

60

80

100

120

140

160

14 20 26 32 38 44 50 56 62 68 74 80 86 92Post injury hour

mm

Hg

or

sats

ICPCPP MAPSJVO2

Figure 2 An example patient in whom the LPR was elevatedabove 40 for several hours. During this time, the ICP was beingactively treated to maintain the ICP o20 mm Hg. The jugularvenous oximetry (sjvo2) % saturation, arterial carbon dioxidepressure in mm Hg (PaCO2), the CPP, and MAP were all withinnormal limits and show no evidence of brain ischemia.

Metabolic crisis after brain injuryP Vespa et al

5

Journal of Cerebral Blood Flow & Metabolism (2005), 1–12

of the 19 patients evaluated. OGR was consideredto be high if it was greater than the expected ratio of5.8 and decreased if r4.9. OGR was increased infour patients and decreased in 11 patients. Giventhat low OGR may indicate a disproportionateincrease in anaerobic glucose metabolism, weendeavored to evaluate the microdialysis metabo-lites as a function of the OGR. The microdialysislevels of glucose, lactate, glutamate, glycerol, andLPR are similar in these regions, despite statisticallysignificant differences in OGR. Hence, microdialysis

levels did not differentiate between reductions orincreases in OGR. Four ROIs showed normal OGRwith no distinguishing features seen in the micro-dialysate. In four subjects, the OGR is increasedwithout a significant difference in the levels ofcerebral microdialysis levels compared with thosesubjects with normal or low OGR. Similar compar-isons were made by segregating the microdialysisvalues according to normal, low or high CMRO2,OEF and CMRgluc; however, no significant relation-ships were forthcoming.

Example ROI with increased LPR and OEF

OGR = 8.9OEF = 0.82CBF = 21.5CMRG = 3.7

Extracellular Analytes at PET:

Glucose 0.07

Lactate 1.8

Lactate/Pyruvvate 53

Probe

CMRO2 = 3.8

MRI MD probe location

Figure 3 An example patient who showed ischemic values on OEF on PET scan. There was a corresponding increase in LPR andreduction in glucose. This patient was used to define the ischemic pattern of microdialysis in the entire cohort. Values within theregion of interest are shown below each PET image. Abbreviations: CMRgluc—metabolic rate of glucose, CMRO2—metabolic rate ofoxygen, CBF—cerebral blood flow, OGR—oxygen to glucose ratio, OEF—oxygen extraction fraction.

Table 2 Microdialysis and PET values by region of interest

ROI Glucose Glutamate Lactate Glycerol LPR LGR CMRG CBF CMRO2 OEF OGR

1 1.27 30.42 1.61 191.1 34.5 1.3 3.5 38.0 1.6 0.5 6.92 0.75 2.23 7.61 488.0 100.0 10.2 2.1 29.0 1.2 0.4 4.63 0.32 1.40 0.56 18.0 12.4 1.8 2.4 46.0 1.6 0.1 2.74 0.10 1.90 0.49 7.4 28.7 4.9 3.8 31.0 1.3 0.4 2.85 0.07 1.60 0.90 7.8 53.0 12.0 3.7 21.5 3.8 0.8 8.96 1.72 2.00 0.58 28.7 10.4 0.3 4.3 41.0 2.0 0.3 3.57 0.23 1.50 0.51 27.6 24.0 2.2 3.3 30.9 1.8 0.3 3.98 0.34 1.14 0.36 23.4 30.0 1.1 3.5 28.0 2.2 0.6 5.39 2.19 0.58 1.41 13.4 11.2 0.6 4.1 36.8 2.6 0.5 5.0

10 1.20 1.90 0.85 14.8 46.0 0.7 3.8 86.0 1.4 0.2 4.911 0.36 4.00 0.30 2.4 22.0 0.8 2.7 24.0 1.4 0.5 4.012 0.25 5.00 0.96 15.0 38.1 3.9 2.9 31.0 1.7 0.4 4.813 0.05 59.00 4.85 40.0 68.4 107.9 3.3 33.0 0.8 0.5 5.414 0.41 1.62 2.12 95.7 64.2 5.2 2.0 34.0 0.8 0.1 2.415 0.40 0.29 0.42 37.5 40.0 1.0 3.7 84.0 1.4 0.1 7.016 0.32 1.02 0.55 18.1 80.6 1.7 3.1 31.0 0.7 0.4 4.217 1.62 1.35 0.78 8.9 12.2 0.5 2.9 51.0 2.8 0.4 7.918 0.60 1.97 0.30 8.5 20.6 0.5 4.7 48.0 2.5 0.4 5.419 1.13 1.08 0.80 12.1 18.2 0.7 3.2 23.0 2.0 0.4 3.420 0.13 2.50 1.20 29.3 26.3 9.2 3.0 29.0 2.5 0.4 5.4

Microdialysis metabolites of glucose, glutamate, lactate, glycerol, LPR are shown along with the PET metabolic rates of glucose (CMRgluc) and oxygen(CMRO2), CBF, OEF and OGR.

Metabolic crisis after brain injuryP Vespa et al

6

Journal of Cerebral Blood Flow & Metabolism (2005), 1–12

Significance of LPR

LPR440 was found to be a sensitive but nonspecificindicator of ischemia in the present cohort. Sevenindividuals showed LPR440, but only a singlepatient showed regional ischemia (Figure 3). In thispatient, the OEF in the region of the microdialysiswas increased and CBF reduced, suggesting anischemic response with compensatory increase inoxidative metabolism. Examination of other patientswith increased LPR showes a trend towards lowerglucose and higher glutamate and glycerol levels,however, given the small sample size and standarddeviation, these differences were not statisticallysignificant. Cluster analysis using a composite ofabnormal microdialysis parameters including thecombination of glucose o0.1 mmol and LPR425was performed to determine if this combinationmore accurately reflected compromised metabolism.

Only three subjects showed this pattern. Thecorresponding mean PET parameters of these threepatients were CMRO2 2.171.8, OEF 0.5570.24,5.773.1, and CBF 28.576.1. Sequential analysis ofvarious combinations of abnormal microdialysisparameters was performed, but no other specificcluster was better able to reflect the abnormalitiesseen on the PET images. Hence, LPR 440 alone wasnot a specific indicator of ischemia in this cohort.

Discussion

Traumatic brain injury results in a reduction inoxidative metabolism and metabolic crisis that islong lasting. The current study used PET combinedwith cerebral microdialysis to study the regionaland temporal course of this disturbance. Theprincipal findings of our study were that (1) theincidence of posttraumatic regional and globalischemia was low, accounting for a minority of themetabolic disturbance, (2) there was ongoing non-ischemic metabolic crisis indicated by elevatedmicrodialysis lactate/pyruvate ratio (LPR) that can-not be accounted for by brain ischemia; (3) LPRcorrelated negatively with CMRO2; (4) that LPRwas a nonspecific indicator of posttraumatic brainischemia. Our findings are restricted temporally tothe period of observation in the ICU, generally frompostinjury hour 18 to postinjury day 10, and cannotaddress the incidence of ultra-early brain ischemiaor the usefulness of LPR during the ultra-earlyperiod.

The Incidence of Ischemia in the ICU after TBI

Several lines of evidence from clinical brain injuryresearch suggest that ischemia occurs frequently inTBI (Graham and Adams, 1971; Bouma et al, 1991).These data are most robust for the initial 12 h afterinjury in which brain-imaging studies as well aswhole brain oxygenation monitoring has shown thatischemia occurs in 30% of the population. More-over, autopsy series have shown that necroticcellular changes are frequent in fatal TBI and thesechanges are thought to be due to ischemia, ratherthan other mechanisms of cell death. In contra-distinction to these studies, recent PET studiesperformed in TBI patients both early and late afterthe injury have failed to show an elevated incidenceof brain ischemia. In similarly conducted PETstudies of oxidative metabolism conducted byDiringer and colleagues, PET studies were per-formed at a mean of 12 h after injury. In the Diringerstudy, the OEF, CBF, CMRO2 values are similar tothose in the current study at a baseline arterialcarbon dioxide (PaCO2) concentration of 40 mm Hg.With induced hyperventilation, to PaCO2 of 30 or25 mm Hg, Diringer found global and regionalincreases in OEF without reduction in CMRO2. In

Figure 4 Region of interest cerebral metabolic rate of oxygen(CMRO2) and the microdialysis LPR. Increasing LPR correlatedwith reducing CMRO2.

Table 3 Correlation matrix of PET parameters and microdialysismetabolites

CMRgluc CBF CMRO2 OEF OGR

Glucose 0.3 0.25 0.23 �0.05 0.09P 0.19 0.27 0.32 0.83 0.69Glutamate 0.01 �0.09 �0.31 0.22 0.17P 0.96 0.19 0.19 0.37 0.46Lactate �0.42 �0.18 �0.32 0.06 �0.01P 0.06 0.44 0.17 0.79 0.99Glycerol �0.43 �0.12 �0.27 �0.01 0P 0.06 0.65 0.26 0.96 0LPR �0.41 �0.09 �0.49 0.02 0.04P 0.07 0.7 0.03 0.92 0.86LGR �0.04 �0.14 �0.27 0.19 0.1P 0.83 0.57 0.27 0.43 0.69

The bold characters highlight the only statistically significant findings.

Metabolic crisis after brain injuryP Vespa et al

7

Journal of Cerebral Blood Flow & Metabolism (2005), 1–12

ROIs with initial CBF o10 mL/(100 g/min�1), OEFincreased from 0.3 to 0.71, without a reduction inCMRO2. These results indicate that, under basalconditions, ischemia was not present even in areasof low CBF, but those areas could become ischemicwithout energy failure during hyperventilation.Similarly, Coles and colleagues used oxidative PETstudies in TBI patients and found that at baselineCO2 of 30 to 34 mm Hg, the mean ischemic brainvolume was 67 cm3. This translates into ischemiabeing present in approximately 6% of the totalbrain volume. However, the volume of potentiallyischemic tissue ranged up to 224 cm3 and in six of 15patients the ischemic volume averaged 50 cm3 T.

In the current study, we found a lower meanischemic brain volume, 1.5 cm3, which translatesinto 1.5% of total brain volume. In addition, usingan ROI of 1.5 cm3 that theoretically covers the regionthat is sampled by the microdialysis probe, wefound only a single instance of ischemia in theregion of the probe. Hence, of the three contem-porary studies, using different volumes of tissuesampling by PET, the incidence of ischemia acrossall voxels is low and constitutes between 1.5% and6% of the total brain. In contrast to previous studies,we performed PET at later time points after injury,and during the delivery of intensive care. Our studyis limited in part because the timing of PET was laterthan that in the Diringer study (mean 60 h afterinjury) and ischemia might have been presentduring at earlier times. However, among the fourpatients studied within 28 h within our study, wedid not see a difference in the volume of tissue inthe ischemic range. We did not perform within-subjects baseline and hyperventilation studies,however, we did not find increased volume ofpotentially ischemic brain in those subjects whohad PaCo2 in the 28 to 31 mm Hg range comparedwith those who had higher PaCo2 values. Thecohorts may differ in ways that are not known.Thus, several factors could explain the differences,yet the main result is the same, namely thatischemia is not the common physiological state inthe brain after brain injury. However, early PETimaging performed by Diringer and colleagues(2002) also showd that the presence of ischemiawas low, as indicated on oxygen extraction fractionvalues in multiple ROIs. Thus, the incidence ofischemia after TBI on PET appears to be uncommon.

LPR as a Marker of Metabolic Crisis

There is a robust and growing literature on cerebralmicrodialysis in patients with TBI and otherneurocritical care illnesses such as subarachnoidhemorrhage (Vespa et al, 2004). Microdialysis is asafe and effective monitoring technique that enablessampling of brain neurochemistry and inferentialassessment of brain metabolism. Despite this grow-ing body of knowledge (see also Hutchinson et al,

2002), the exact relationship between microdialysismarkers of metabolism and independent measuresof metabolism has yet to be clearly defined invarious disease states. Hence, the validation ofmicrodialysis markers of cellular metabolism inhuman subjects is of paramount priority to clin-icians and researchers alike. The current papercontributes to this validation process.

The ratio of lactate to pyruvate has been one ofthe commonly reported features of microdialysismonitoring since proposed by Hillered (Persson andHillered, 1992; Valtysson et al, 1998). LPR rationormally ranges o20 under conditions of uncom-plicated metabolism. The LPR is considered to be amarker of the relative redox state of the tissue, withincreases in LPR indicating ischemia and a conse-quent shift in the NAD/NADH ratio. Thus, LPR hasbeen proposed to be a reliable marker of ischemia.Indeed, the combination of reduction in extracellu-lar glucose and elevation in LPR has been robustlyshown to reflect energy crisis (Landolt et al, 1994;Langemann et al, 2001) and these have been shownto be preterminal events when glucose is reduced toundetectable levels. These preterminal events occurin TBI patients under conditions of cerebral circu-latory arrest and loss of all brain function. Underexperimental conditions, LPR has been documentedto indicate severely impaired oxidative respiration(Enblad et al, 2001) and the impaired redox statethat occurs with ATP depletion. Under conditions ofpermanent ischemia, the LPR increases to above 40and often plateaus in the range of 80 to 120. Underconditions of impending brain death, LPR has beendocumented in the range of 500 to 1000. Withreversible ischemia, the LPR does normalize within60 to 90 min of restoration of the CBF. The increasein the ratio is comprised primarily of a reduction inthe pyruvate concentration by 10- to 100-foldcompared with increases in lactate by 2- to 5-fold(Vespa et al, 2003). Reversible increases of the LPRfollowed by restoration to normal values have beenseen with specific reversible mitochondrial poison-ing (Clausen et al, 2001. Thus, the LPR appears tobe a sensitive marker of reversible mitochondrialdysfunction.

In the present study, we found that increasedLPR most tightly corresponds to nonischemicreduction in the CMRO2. Since CMRO2 is a measureof mitochondrial oxidative function, our findingscorrespond to the previous experimental obser-vations. The PET-derived measures of CMRO2

potentially correspond to reversible mitochondrialdysfunction. In our current PET cohort, areasadjacent to contusions or within the center ofcontusions shown very low CMRO2 values whichwere consistent with irreversible cell loss. However,these areas were not included in the ROIs used forthe comparison with microdialysis. In contrast, theregions containing the microdialysis probes did notshow anatomic injury on CT and MRI, and hencethe tissue is not considered irreversibly injured.

Metabolic crisis after brain injuryP Vespa et al

8

Journal of Cerebral Blood Flow & Metabolism (2005), 1–12

Therefore, the relationship between LPR andCMRO2, which we found in our cohort, is mostlikely related to reversible metabolic dysfunctionrather than cell loss. As such, it appears that LPR isable to reflect this metabolic dysfunction even in theabsence of tissue damage or brain ischemia.

The Nature of Posttraumatic Metabolic Crisis

In our data, we found that ischemia is uncommon inPET and microdialysis measurements taken duringthe period of early intensive care. Yet, the LPR andother microdialysis measurements were abnormal,suggesting that even in the absence of brainischemia, the brain is in metabolic crisis. Indeed,in a recent microdialysis paper (Vespa et al, 2003),persistent and long-lasting evidence of metaboliccrisis lasting several days despite the absence ofbrain ischemia was documented. The exact nature ofthis metabolic crisis remains to be determined.However, we did find that LPR elevations werecorrelated with reduced oxidative metabolism,suggesting that LPR is an indicator of mitochondrialdysfunction.

Alterations in other PET parameters such as OGRsupport the concept of nonischemic metaboliccrisis, but also point to the heterogeneity ofmetabolism after TBI. Both reductions and increasesin the OGR have been shown to occur after TBI inthe current study. This alteration of OGR waspreviously characterized by Wu et al (2004). OGRis the stochiometric relationship between oxidativeto glucose metabolism, which is normally in a ratioof approximately 6. Reduction in OGR indicates adisproportionate reduction in oxidative metabolism,thought to be due to impaired mitochondrialfunction. Reduction in OGR has been seen primarilyin white matter structures reflecting the widespreadmetabolic dysfunction after TBI. In some circum-stances, reduction in OGR is related to an observedabsolute increase in glucose metabolism. However,this is not frequently seen; instead, the absoluterates of glucose metabolism are decreased as well.Hence, there is a combined reduction in bothoxidative and glucose metabolism, but the degreeof reduction in each is not equivalent on a molarbasis. This implies a relative increase in glu-cose metabolism, compared with that of oxidativemetabolism.

However, increases in OGR have been seen insome circumstances. Increased OGR indicates anincrease in oxidative metabolism that is not depen-dent on an increase in glucose utilization. Theincrease in OGR would indirectly indicate utiliza-tion of a fuel other than glucose to generate theoxidative rates seen. The nature of this fuel is notknown at the present time, but experimentalevidence suggests that the brain is capable ofutilizing ketones, lactate, and pyruvate as alterna-tive fuels under selected conditions. Thus, the

nature of the metabolic crisis might be predicatedon impaired oxidative metabolism. Hence, cerebralmicrodialysis monitoring of brain metabolismshould be interpreted in the context that changesin metabolites may reflect important nonischemicchanges in metabolism.

Defining Ischemia after Traumatic Brain Injury

The definition of ischemia and determining theincidence of ischemia after TBI remain elusive. Inthis paper, we have used classic thresholds of PETOEF 40.75 as well as microdialysis metabolitevalues to define ischemia. The PET OEF 40.75threshold is based in part on the work of Senda et al(1989) and Marchal et al (1999). In the former study(Senda et al, 1989), mean OEF in acute ischemicterritories was 0.78 within 2 days of the ischemicstroke. In the latter study, Marchal et al report thatthe upper limit of OEF beyond which acutelyischemic tissue was irreversibly injured was 40.73.Experimentally, a similar OEF 40.75 threshold hasbeen reported (Sakoh et al, 2000). Arguably, definingischemia at a different threshold, such as OEF 40.5,potentially would result in an alternative interpreta-tion. For example, if we use an OEF 40.50, then 2 of19 patients would have regional ischemia. Indeed, asingle valid OEF threshold for ischemia has not beenagreed on in the literature. Alternatively, a reductionin CMRO2 o1.4 mL/(100 mL/min�1) has been shownto be a threshold below which tissue viability is lostunder conditions of ischemia (Powers et al, 1985;Ackerman et al, 1989). If we use this CMRO2

definition of irreversibly injured tissue, then in thethree subjects who displayed CMRO2 r1.3, 2/3 hadan LPR 440 and might be considered ischemic,while one did not. Given that CMRO2 is reversiblyreduced in TBI, the CMRO2 o1.3 definition may notbe valid. However, it is intriguing that LPR negativelycorrelates with CMRO2, and hence LPR may be amarker of cellular viability rather than of ischemiaper se. Presently, we are exploring the long-termtissue viability as a function of LPR to determine if infact elevated LPR predicts long-term tissue atrophy.

To be independent of a single OEF threshold, weused an alternative definition of ischemia, namelyCvO2 o3/5 mL/100 mL, proposed by Powers et al(1985), Sutton et al (1990), and recently by Coles et al(2004). The CvO2-based analysis reveals that only1% of all regions imaged were in the ischemic range.Hence, the present study shows that ischemia is rareat either PET threshold for ischemia. Similarly, if weuse an alternative microdialysis threshold for ische-mia, LPR alone, then we would determine that theincidence of ischemia is between 30% and 40% ofall hourly values. This incidence would be incon-sistent with other continuous measures of brainoxygenation in our study, and in all others (Gopinathet al, 1994, 1999; Coles et al, 2004). The micro-dialysis definition of ischemia that we used, namely

Metabolic crisis after brain injuryP Vespa et al

9

Journal of Cerebral Blood Flow & Metabolism (2005), 1–12

a reduction in glucose o0.2 mmol/L and LPR 440,was initially proposed by Landolt et al (1994) andLangemann et al (2001) in the setting of terminalherniation leading to ischemia and cell death afterTBI. Indeed, Hlatky et al (2004) recently confirm thisdefinition of ischemia, in which microdialysisglucose values decrease to to o0.2 mmol/L andLPR 440. In the same study (Hlatky et al, 2004),elevated LPR 440 alone did not correlate withischemia, as measured by a brain tissue oxygenprobe. Our study differs from that of Hlatky, since wemake use of a direct comparison of OEF in the regionof the probe rather than a comparison with thepartial pressure of oxygen. In addition, there arelimitations to these microdialysis criteria for ische-mia. However, the finding that LPR alone is not asensitive measure of brain ischemia in the TBIpatient is similar to that of the Hlatky study. Hence,we consider our PET and microdialysis definitions ofischemia to be robust and, even with the use ofalternative, more conservative definitions of ische-mia, the incidence of ischemia is low despite thepersistent abnormal LPR.

Summary

The primary findings of the current study were thatthe injured brain has persistent impairments inmetabolism that can be reflected by cerebral micro-dialysis. Specifically, the LPR best reflects impairedoxidative metabolism, but is not specific for brainischemia. Moreover, the metabolic crisis was notprimarily a result on persistent brain ischemia, andhence elevations in LPR are not specific for brainischemia. Given that most clinical monitors such asthe brain parenchymal oxygen monitor and thejugular venous oximeter are designed to look forischemia, the finding of a low incidence of brainischemia is important. Moreover, these findingssuggest that the use of microdialysis monitoring ofvarious metabolites to determine the overall state ofenergy balance between supply and demand mightbe the most appropriate monitoring tool in TBI.

Acknowledgements

This research was supported by NINDS 03039,NS02089, and the State of California NeurotraumaInitiative grant.

References

Ackerman RH, Lev MH, Mackay BC, Katz PM, BabikianVL, Alpert NM (1989) PET studies in acute strokefindings and relevance to therapy (abstract). J CerebBlood Flow Metab 9(S1):S359

Andersen BJ, Maramarou A (1989) Isolated stimulationof glycolysis following traumatic brain injury. In

Intracranial pressure VII Hoff JT, Betz AL (eds), Berlin:Springer pp. 575–80

Baker SP, O’Neill B, Haddon Jr W, Long WB (1974) Theinjury severity score: a method for describing patientswith multiple injuries and evaluating emergency care.J Trauma 14:187–96

Bergsneider MA, Hovda DA, Shalmon E, Kelly DF, VespaPM, Martin NA, Phelps ME, McArthur DL, Caron MJ,Kraus JF, Becker DP (1997) Cerebral hyperglycolysisfollowing severe human traumatic brain injury: apositron emission tomography study. J Neurosurg86:241–51

Bergsneider M, Hovda DA, Lee SM, Kelly DF, McArthurDL, Vespa PM, Lee JH, Huang SC, Martin NA,Phelps ME, Becker DP (2000) Dissociation of cerebralglucose metabolism and level of consciousnessduring the period of metabolic depression followinghuman traumatic brain injury. J Neurotrauma 17:389–401

Bergsneider M, Hovda DA, McArthur DL, Etchepare M,Huang SC, Sehati N, Satz P, Phelps ME, Becker DP(2001) Metabolic recovery following human traumaticbrain injury based on FDG-PET: time course andrelationship to neurological disability. J Head TraumaRehabil 16:135–48

Bouma GJ, Muizelaar JP, Choi SC, Newlon PG, Young HF(1991) Cerebral circulation and metabolism after severetraumatic brain injury: the elusive role of ischemia.J Neurosurg. 75:685–93

Clausen T, Zauner A, Levasseur JE, Rice AC, Bullock R(2001) Induced mitochondrial failure in the felinebrain: implications for understanding acute post-traumatic metabolic events. Brain Res 20:358

Coles JP, Fryer TD, Smielewski P, Rice K, Clark JC, PickardJD, Menon DK (2004) Defining ischemic burden aftertraumatic brain injury using 15O PET imaging ofcerebral physiology. J Cereb Blood Flow Metab24:191–201

de Deyne C, Vandekerckhove T, Decruyenaere J, ColardynF (1996) Analysis of abnormal jugular bulb oxygensaturation data in patients with severe head injury.Acta Neurochir (Wien) 138:1409–15

Diringer MN, Videen TO, Yundt K, Zazulia AR, Aiyagari V,Dacey Jr RG, Grubb RL, Powers WJ (2002) Regionalcerebrovascular and metabolic effects of hyperventila-tion after severe traumatic brain injury. J Neurosurg96:103–8

Enblad P, Frykholm P, Valtysson J, Silander HC, Anders-son J, Fasth KJ, Watanabe Y, Langstrom B, Hillered L,Persson L (2001) Middle cerebral artery occlusion andreperfusion in primates monitored by microdialysisand sequential positron emission tomography. Stroke32:1574–80

Feeney DM, Baron JC (1986) Diaschisis. Stroke 17:817–30

Glenn TC, Kelly DF, Boscardin WJ, McArthur DL, Vespa PM,Oertel M, Hovda DA, Bergsneider M, Hillered L, MartinNA (2003) Energy dysfunction as a predictor of outcomeafter moderate or severe head injury: indices of oxygen,glucose and lactate metabolism. J Cereb Blood FlowMetab 23:1239–50

Gopinath SP, Robertson CS, Contant CF, Hayes C, FeldmanZ, Narayan RK, Grossman RG (1994) Jugular venousdesaturation and outcome after head injury. J NeurolNeurosurg Psychiatry 57:717–23

Gopinath SP, Valadka AB, Uzura M, Robertson CS (1999)Comparison of jugular venous oxygen saturation and

Metabolic crisis after brain injuryP Vespa et al

10

Journal of Cerebral Blood Flow & Metabolism (2005), 1–12

brain tissue Po2 as monitors of cerebral ischemia afterhead injury. Crit Care Med 27:2337–45

Graham DI, Adams JH (1971) Ischaemic brain damage infatal head injuries. Lancet 6:265–6

Graham DI, Ford I, Adams JH, Doyle D, Teasdale GM,Lawrence AE, McLellan DR (1989) Ischaemic braindamage is still common in fatal non-missile headinjury. J Neurol Neurosurg Psychiatry 52:346–50

Hayes RL, Katayama Y, Jenkins LW, Lyeth BG, Clifton GL,Gunter J, Povlishock JT, Young HF (1988) Regional ratesof glucose utilization in the cat following concussivehead injury. J Neurotrauma 5:121–37

Hlatky R, Valadka AB, Goodman JC, Contant CF, RobertsonCS (2004) Patterns of energy substrates during ischemiameasured in the brain by microdialysis. J Neurotrauma21:894–906

Hovda DA, Yoshino A, Kawamata T, Katayama Y, BeckerDP (1991) Diffuse prolonged depression of cerebraloxidative metabolism following concussive braininjury in the rat: a cytochrome oxidase histochemistrystudy. Brain Res 3:1–10

Hillered L, Persson L (1999) Neurochemical monitoringof the acutely injured human brain. Scand J Clin LabInvest Suppl 229:9–18

Hutchinson PJ, Gupta AK, Fryer TF, Al-Rawi PG, ChatfieldDA, Coles JP, O’Connell MT, Kett-White R, Minhas PS,Aigbirhio FI, Clark JC, Kirkpatrick PJ, Menon DK,Pickard JD (2002) Correlation between cerebral bloodflow, substrate delivery, and metabolism in head injury:a combined microdialysis and triple oxygen positronemission tomography study. J Cereb Blood Flow Metab22:735–45

Jennett B, Adams JH, Murray LS, Graham DI (2001)Neuropathology in vegetative and severely disabledpatients after head injury. Neurology 56:486–90

Kawamata T, Katayama Y, Aoyama N, Mori T (2000)Heterogeneous mechanisms of early edema formationin cerebral contusion: diffusion MRI and ADC mappingstudy. Acta Neurochir Suppl 76:9–12

Kotapka MJ, Gennarelli TA, Graham DI, Adams JH,Thibault LE, Ross DT, Ford I (1991) Selective vulner-ability of hippocampal neurons in acceleration-induced experimental head injury. J Neurotrauma8:247–58

Landolt H, Langemann H, Mendelowitsch A, Gratzl O(1994) Neurochemical monitoring and on-line pHmeasruements using brain microdialysis in patients inintensive care. Acta Neurochir 60:47578

Langemann H, Alessandri B, Mendelowitsch A, Feuer-stein T, Landolt H, Gratzl O (2001) Extracellular levelsof glucose and lactate measured by quantitative micro-dialysis in the human brain. Neurol Res 23:531–6

Lee JH, Kelly DF, Oertel M, McArthur DL, Glenn TC, VespaP, Boscardin WJ, Martin NA (2001) Carbon dioxidereactivity, pressure autoregulation, and metabolic sup-pression reactivity after head injury: a transcranialDoppler study. J Neurosurg 95:222–32

Marchal G, Benali K, Iglesias S, Viader F, Derlon JM,Baron JC (1999) Voxel-based mapping of irreversibleischaemic damage with PET in acute stroke. Brain122:2387–400

Mintun MA, Raichle ME, Martin WR, Herscovitch P(1984) Brain oxygen utilization measured with O-15radiotracers and positron emission tomography. J NuclMed 25:177–87

Ohta S, Meyer E, Thompson CJ, Gjedde A (1992) Oxygenconsumption of the living human brain measured after

a single inhalation of positron emitting oxygen. J CerebBlood Flow Metab 12:179–92

Persson L, Hillered L (1992) Chemical monitoring ofneurosurgical intensive care patients using intra-cerebral microdialysis. J Neurosurg 76:72–80

Powers WJ, Grubb Jr RL, Darriet D, Raichle ME (1985)Cerebral blood flow and cerebral metabolic rate ofoxygen requirements for cerebral function and viabilityin humans. J Cereb Blood Flow Metab 5:600–8

Sakoh M, Ostergaard L, Rohl L, Smith DF, Simonsen CZ,Sorensen JC, Poulsen PV, Gyldensted C, Sakaki S,Gjedde A (2000) Relationship between residual cere-bral blood flow and oxygen metabolism as predictive ofischemic tissue viability: sequential multitracer posi-tron emission tomography scanning of middle cerebralartery occlusion during the critical first 6 hours afterstroke in pigs. J Neurosurg 93:647–57

Senda M, Alpert NM, Mackay BC, Buxton RB, Correia JA,Weise SB, Ackerman RH, Dorer D, Buonanno FS(1989) Evaluation of the 11CO2 positron emissiontomographic method for measuring brain pH. II.Quantitative pH mapping in patients with ischemiccerebrovascular diseases. J Cereb Blood Flow Metab9:859–73

Sutton LN, McLaughlin AC, Dante S, Kotapka M, SinwellT, Mills E (1990) Cerebral venous oxygen content as ameasure of brain energy metabolism with increasedintracranial pressure and hyperventilation. J Neurosurg73:927–32

Valtysson J, Persson L, Hillered L (1998) Extracellularischaemia markers in repeated global ischaemia andsecondary hypoxaemia monitored by microdialysis inrat brain. Acta Neurochir (Wien) 140:387–95

Vespa PM, Nuwer MR, Nenov V, Ronne-Engstrom E,Hovda DA, Bergsneider M, Kelly DF, Martin NA,Becker DP (1999) Increased incidence and impact ofnonconvulsive and convulsive seizures after traumaticbrain injury as detected by continuous electroen-cephalographic monitoring. J Neurosurg 91:750–60

Vespa P, McArthur D, Glenn T, O’Phelan K, Etchepare M,Kelly D, Bergsneider M, Martin NA, Hovda DA,Neurotrauma J (2003) Persistently reduced levels ofextracellular glucose early after traumatic braininjury correlate with poor outcome at six months: Amicrodialysis study. J Cereb Blood Flow Metab 23:865–877

Vespa P, McArthur D, Alger J, O’Phelan K, Glenn T,Bergsneider B, Martin NA, Hovda DA (2004) Regionalheterogeneity of brain metabolism using cerebralmicrodialysis: concordance with magnetic resonancespectroscopy and positron emission tomography. BrainPathol 14:210–4

Vigue B, Ract C, Benayed M, Zlotine N, Leblanc PE,Samii K, Bissonnette B (1999) Early SjvO2 monitoringin patients with severe brain trauma. Intens Care Med25:445–51

Vink R, Faden AI, McIntosh TK (1988) Changes incellular bioenergetic state following graded traumaticbrain injury in rats: determination by phosphorus 31magnetic resonance spectroscopy. J Neurotrauma5:315–30

Wilson JT, Pettigrew LE, Teasdale GM (2000) Emotionaland cognitive consequences of head injury in relationto the glasgow outcome scale. J Neurol NeurosurgPsychiatry 69:204–9

Wu HM, Huang SC, Hattori N, Glenn TC, Vespa PM, YuCL, Hovda DA, Phelps ME, Bergsneider M (2004)

Metabolic crisis after brain injuryP Vespa et al

11

Journal of Cerebral Blood Flow & Metabolism (2005), 1–12

Selective metabolic reduction in gray matter acutelyfollowing human traumatic brain injury. J Neurotrauma21:149–61

Yoshino A, Hovda DA, Kawamata T, Katayama Y, BeckerDP (1991) Dynamic changes in local cerebral glucoseutilization following cerebral conclusion in rats: evi-dence of a hyper- and subsequent hypometabolic state.Brain Res 4:106–19

Yoshino A, Hovda DA, Katayama Y, Kawamata T, BeckerDP (1992) Hippocampal CA3 lesion prevents postcon-cussive metabolic dysfunction in CA1. J Cereb BloodFlow Metab 12:996–1006

Yundt KD, Diringer MN (1997) The use of hyperventila-tion and its impact on cerebral ischemia in thetreatment of traumatic brain injury. Crit Care Clin13:163–84

Metabolic crisis after brain injuryP Vespa et al

12

Journal of Cerebral Blood Flow & Metabolism (2005), 1–12