Recently,severalstudieshavetriedtoclarifycerebralhemodynamics and metabolism during the acute phase ofbrain injuryusing various methods (1—4). It is generally recognized thatPET is the most useful method of quantitatively examiningcerebral hemodynamics and metabolism in vivo. There havebeen, however, surprisingly few PET studies of head injury(5—13)and most ofthese studies have been presented as part ofa general review of PET. Previous systematic PET studies ofhead injury ( 7,12, 13 ) examined patients with focal brain injury,but there are no reports of PET findings during the acute phaseof severe diffuse brain injury (SDBI). In the present study, wemeasured cerebral hemodynamics and metabolism in threepatients with SDBI 10 days after onset using PET and report thechanges in cerebral hemodynamics and metabolism in 5DB! inthe early stage.

CASE REPORTS

Patients and MethodsThe three patients included an 18-yr-old woman, a 25-yr-old

woman and a 30-yr-old man who had head injuries caused bytraffic accidents. All patients were transferred to the emergencyhospital within 15 mm after onset of SDBI. All patients were in adeep comatose state with Glasgow Coma Scale (GCS) scores (14)on admission of 4, 5 and 5, respectively. However, the GCS scoreimproved to 8 within 3 hr in Patient 2 and 6 within 5 hr in Patient3 after admission. The GCS of Patient 1 did not improve prior to

death. For all patients, CT obtained on admission, revealed typicalfindings ofDiffuse Injury II (Fig. 1) according to the classificationsproposed by Marshall et al. (15). On admission, respirationpatterns for each patient were normal and data from arterial gas

Received Mar. 29, 1995; revision accepted Oct. 8, 1995.For correspondence@ reprints contact: Tarumi Yamaki, MD, Department of Neuro

surgery,KyotoPrefecturalUniversityof Medicine,KawaramachiHirokoji,Kamigyo-ku,Kyoto, 602, Japan.

FFIGURE1. @A)AdmissionCTimageof Patient1 depicts traumaticsubarachnold hemorrhageinthe amb@ntcistern,quadngeminalcistern and cerebralcontusion in the right temporal lobe. (B)AdmissionCT scan of Patient 2demonstrates intraventricularhemorrhageinthe leftlateralventricle(arrows).(C)Mmission CT scan of Patient3 shows smallhemorrhagiclesionsin thebkemporal lobesand smallepkluralhematomainthetemporal regkn (arrow).

analysis showed no remarkable hypoxia. The patients were treatedconservatively. PET studies were performed 10 days after injuryand follow-up studies were performed 3 mo after injury for Patients

1166 THEJOURNALOFNUCLEARMEDICINE•Vol. 37 . No. 7 •July 1996

Cerebral Hemodynamics and Metabolism of SevereDiffuse Brain Injury Measured by PETTarumi Yamaki, Yoshio Imahori, Yoshio Ohmori, Eiji Yoshino, Takashi Hohri, Toshihiko Ebisu and Satoshi UedaDepartment ofNeurosurgery, Kyoto Prefectural University ofMedicine, Kyoto, Japan

Cerebral hemodynamics and metabolism in three patients withsevere diffuse brain injury were measured 10 days after onset usingPET. In this study, regional cerebral blood flow (rCBF), oxygenextraction fraction (rOEF), cerebral blood volume (rCBV),cerebralmetabolic rate for oxygen (rCMRO2), cerebral metabolic rate forglucose (rCMRglc)and cerebral metabolic ratio (rCMRO2/rCMRglc)were measured. The Glasgow Coma Scale scores on admissionwere 4, 5 and 5, respectively, and CT on admission showed typicalfindings of diffuse brain injury. As a result, PET revealed miseryperfusion and hyperglycotysis in Patient 1 and matching low perfusion and low glucose metabolism in Patients 2 and 3. AlthoughPatient 1 died, Patients 2 and 3 had good recoveries. We speculatethat a long-lasting anaerobic state, indicated by a high OEF valueand low metabolic ratio, is an important undesirable factor related tothe outcome.

Key Words: diffuse brain injury cerebral blood flowand metabolism;PET

J NuciMed1996;37:1166-1170

by on January 26, 2019. For personal use only. jnm.snmjournals.org Downloaded from

181.042.07.5396.97.92Study

191.139.57.4197.012.0Study2104.938.27.3997.811.3387.339.07.4296.811.0

rCBF rOEF rCMRO2 rCMRgIcMetabolicratio

(rCMRO@/rCMRglc)Normal

valueof43 ±7 mV 0.42 ±0.08 3.3 ±0.5 mV 7.0 ±0.8mg/graymatter100 g/min 100 g/min 100g/minAt.

frontal33 0.78 2.58.40.30Lt.frontal34 0.62 2.48.60.28Rt.temporal35 0.71 2.68.90.29Lt.temporal37 0.81 2.39.20.25Rt.occipital34 0.77 2.58.30.30Lt.occipital37 0.65 2.58.10.31Rt.parietal30 0.78 2.58.30.31Lt.parietal31 0.74 2.48.50.28Average34

0.73 2.58.50.29Normalvalue25 ±4 mV 0.44 ±0.06 2.0 ±0.3 mV 4.9 ±0.8mg/ofwhitematterbOg/mm

bOg/mmbOg/mmAt.white matter20 0.58 1.26.70.18Lt.white matter22 0.58 1.46.30.22Average21

0.58 1.36.50.20TABLE

3Regional Data of PET Parameters 10 Days after Injury in Patient2rCBF

rOEF rCMRO2 rCMRgICMetabolicratio

(rCMRO@rCMRglc)Normal

valueof43 ±7 mV 0.42 ±0.08 3.3 ±0.5 mV 7.0 ±0.8mg/graymatter100 g/mmn 100 g/mmn 100g/minRt.

frontal41 0.40 2.76.90.39Lt.frontal40 0.40 2.56.70.37At.temporal45 0.39 2.86.50.43Lt.temporal41 0.41 2.76.50.42Rt.occipital40 0.42 2.66.50.40Lt.occipital36 0.45 2.66.60.39Rt.panetal37 0.45 2.76.30.43Lt.parietal35 0.44 2.66.30.43Average39

0.42 2.76.50.41Normalvalue25 ±4 mV 0.44 ±0.06 2.0 ±0.3 mV 4.9 ±0.8mg/of

white matter100 g/mmn 100 g/min 100g/minAt.white matter22 0.44 1.54.70.32Lt.white matter22 0.39 1.44.50.31Average22

0.42 1.5 4.60.32

TABLE IResults of ArterialGas Analysison PETStudies

Patient no. Pa02 (mmHg) PacO2 (mmHg) pH Sa02 (%) Hb (g/dl)

Regional cerebral blood flow (rCBF), oxygen extractionfraction (rOEF) and regional metabolic rate of oxygen (rCMRO2) were measured using a ‘5O-labeledgas steady-statetechnique as described by Frackowiak et al. (17). Regionalcerebral blood volume was measured after bolus inhalation of‘5O-labeledcarbon monoxide gas as described by Phelps et al.(18). Steady state was obtained by continuous inhalation of

0.1—0.4GBq/min C'5O2 and ‘@O2gas, and dose of C'5O bolusinhalation was 0.8—4 GBq. In each study, we corrected foroverestimation of rOEF and rCMRO2 by CBV according to themethod by Lammertsma et al. (18). In each patient, regionalmetabolic rates of glucose (rCMRglc) were also measured afterbolus injection of ‘8F-labeled2-deoxy-D-glucose (‘8FDG)(3.7MBqfkg) immediately after the ‘50-gasstudies. rCMRglc wasestimated by the formula of Hutchins et al. (20) using thelumped constant and rate constants described by Phelps et al.(21 ). We calculated the metabolic ratio (rCMRO2/rCMRglc) ineach patient as the index of the balance between oxygenmetabolism and glucose metabolism. Circular regions of interest(ROIs) in the gray matter contained 188 mm2 (47 pixels) andwere 16 mm in diameter, whereas the white matter ROIscontained 116 mm2 (29 pixels) and were 12 mm in diameter.

2 and 3. GCS scores on the day of the first PET study were 4, 14and 10, respectively, and GCS scores on the day of the follow-upPET studies were 15 (Patient 2) and 14 (Patient 3).

PET scans were obtained with a scanner that has an imageresolution of 8 mm FWHM and a slice thickness of 11—13mmFWHM. Tomographic planes were established 1.5 cm apart, 3.0,4.5, 6.0 cm above and parallel to the orbitomeatal line. Theseplanes were set parallel those in the CT studies. Arterial oxygensaturation (Sa02) was derived from measured arterial oxygentension (Pa02) using the equation of Hill (16). Total arterialoxygen content (Ca02) was then calculated as 1.39 X hemoglobin(Hb) concentration X SaO2 + 0.003 1 X Pa02.

TABLE 2Regional Data of PET Parameters 10 Days after Injury in Patient 1

PET INDIFFUSEBRAININJURY•Yamaki et al. 1167

by on January 26, 2019. For personal use only. jnm.snmjournals.org Downloaded from

rCBFrOEFI'CMRO2rCMRglcMetabolicratio

(rCMRO@/rCMRglc)Normal

valueof43 ±7 mV0.42 ±0.083.3 ±0.5 mV7.0 ±0.8mg/graymatter100 g/mmn100 g/min100g/minRt.

frontal390.382.34.30.54Lt.frontal390.402.34.20.55Rt.temporal480.382.74.60.59Lttemporal360.432.34.60.50Rt.occipital380.392.24.10.54Lt.occipital370.382.14.20.50Rt.parietal370.392.24.30.51Lt.

pailetal380.402.14.00.53Average390.402.34.30.53Normal

value25 ±4 mV0.44 ±0.062.0 ±0.3 mV4.9 ±0.8mg/ofwhitematter100 g/mmn100 g/min100g/rnmnRt.

whitematter240.371.33.20.41Lt.white matter240.371.33.10.42Average240.371

.33.20.42

TABLE 4Regional Data of PET Parameters 10 Days after Injury in Patient 3

As a control, the corresponding brain regions were investigatedby the same method in seven healthy adult volunteers, aged 20—64yr (37.4 ± 15.6), who had been previously studied in ourdepartment (22).

RESULTSThe arterial gas and Hb values in each patient's PET study

are shown in Table 1. In all patients, the raw rCBF, rOEF,rCMRO2, rCMRglc values and metabolic ratios were similarthroughout the gray and white matter. This was illustrated bythe average values of the bilateral frontal, temporal, occipitaland parietal gray matter as well as the white matter of thecentrum semiovale shown in Tables 2, 3 and 4. The averagevalues of the seven normal volunteers are also shown.

As shown in Table 2 and Figure 2, the rOEF in Patient 1 wasextremely high and the metabolic ratio was 0.29 ml 02/mg glcin the gray matter and 0.20 ml 02/mg glc in the white matter.These findings indicated anaerobic glycolysis. The patient's

FIGURE2@PETimagesshowrCMRO2(upper)andrCMRglc(lower)inPatient 1. Three tomographic planes are shown. Images show remarkabledecrease in rCMRO2and an increase in rCMRglc.The absence of theradioactive portion (arrows) seen on the rCMRO2 and rCMRglc imagesrepresents normalleftSyManfissure.

clinical condition did not improve after admission and she died12 days after the injury.

In Patients 2 and 3, PET findings demonstrated matching lowperfusion and low metabolism as shown in Tables 3 and 4. Themetabolic ratios were 0.40 and 0.53 ml 02/mg gic in the cortexand 0.32 and 0.41 ml 02/mg glc in the white matter, respectively. The outcome in each case was good recovery accordingto Glasgow Outcome Scale (23) 6 mo after injury. Follow-upPET studies performed 3 mo after injury showed normalfindings on all parameters in Patient 2 (Table 5). In Patient 3,only rCMRglc was studied and the results were within normallimits (Table 6). CT scans obtained at the same time as the PETstudy did not reveal any abnormal findings.

DISCUSSIONAnaerobic glycolysis in the brain is a well-known phenom

enon in cases of ischemic brain damage (24) and in malignantbrain tumors (25). It is also well known that the lactate level incerebrospinal fluid (CSF) increases after head injury, and thelevel of CSF lactate may indicate the severity of cerebral injuryand therefore have some prognostic value (26,27). Recently,some studies demonstrated anaerobic glycolysis in severetraumatic brain injury by showing that the intracranial temperature is higher than the bladder temperature (4,28). Thesereports, however, suggest that anaerobic glycolysis after headinjury could be determined indirectly.

In their PET studies, Baron et al. (24) found that a depressedmetabolic ratio was associated with increased OEF pointing toprotracted tissue hypoxia as the trigger for anaerobic glycolysis.In the present study, the results ofarterial gas analysis in Patient1 suggest that tissue hypoxia might not be a cause of anaerobicglycolysis as previously suggested (3,29).

In experimentalstudies of cerebralconcussion, Duckrow etal. (30) and Yang et al. (31 ) noted that an elevated lactate levelin the cortex was immediately shown after cortical concussion.They speculated that elevation of the lactate level is not aconsequence of ischemia and hypoxia but may depend onderangement of brain energy metabolism in the absence ofsubstrate limitation. Their findings support the observation thatanaerobic glycolysis, as seen in our Patient 1, might not dependon the hypoxic state but on the magnitude of cerebral celldamage. Recently, elevation of local rCMRglc as well aselevated lactate levels after experimental cerebral concussion in

@0

41

1168 THEJOURNALOFNUCLEARMEDICINE•Vol. 37 •No. 7 •July 1996

rCMRO2, 5.1 mi/iNg/min

I 0!:rCMRgIc 12 mg/iNg/min

I

S

by on January 26, 2019. For personal use only. jnm.snmjournals.org Downloaded from

rCBFrOEFrCMRO2rCMRglcMetabolicratio

(rCMRO@/rCMRgIC)Normal

value of43 ±7 mV0.42 ±0.083.3 ±0.5 mV7.0 ±0.8mg/graymatter100 g/min100 g/mmn100g/mmnRt.

frontal460.402.97.10.41Lt.frontal440.402.96.90.42At.temporal480.433.26.90.46Lt.temporal480.453.26.60.49Rt.occipital430.463.07.60.40Lt.occipital400.482.97.50.39Rt.parietal450.483.47.10.48Lt.

parietal410.493.17.00.44Average440.453.17.10.44Normal

value of25 ±4 mV0.44 ±0.062.0 ±0.3 mV4.9 ±0.8mg/whitematter100 g/min100 g/mmn100g/mmnRt.

white matter250.401.65.00.32Lt.white matter240.401.64.90.33Average250.401

.65.00.33

TABLE 5Regional Data of PET Parameters 3 Months after Injury in Patient 2

TABLE 6Regional Data of PET Parameters 3 Months after

Injuryin Patient 3

REFERENCESI. Shigemori M, Monyama T, Harada K, Kikuchi T, Kuramoto K. Intracranial haemo

dynamics in diffuse and focal injuries. Evaluation with transcranial Doppler (lCD)ultrasound. Acta Neurochir 1990;107:5—lO.

2. Asahi H, Ito K, Yamagiwa 0, et al. SPECT as neuroimaging in severe head injury.Neurotraumatology l992;15:59—65.

3. Bouma Gi, Muizelaar JP, Stringer WA, Choi SC, Fatouros P. Young HF. Ultra-earlyevaluation of regional cerebral blood flow in severely head injured patients usingxenon-enhanced computerized tomography. J Neurosurg 1992;77:360—368.

4. Hirayama T, Kinoshita K, Katayama Y, Tsubokawa I. Relative ischemic conditionduring ultra-acute phase in patients with traumatic diffuse damage. Neurotraumatology1993; 16:93—99.

5. Rao N, Turski PA, Polcyn RE, Nickels Ri, Matthews CG, Flynn MM. Fluorine-l8positron emission computed tomography in closed head injury. Arch Phys Med Re/tabI984;64:780—785.

6. Alavi A, Langfitt T, Fazekas 5, Dunaime T, Zimmerman R, Reivich M. Correlativestudies of head trauma with PET, MRI and XCT [Abstract]. J NucI Med 1986;27(suppl):9 19.

7. Langfitt 1W, Obrist WD, Alavi A, et al. Computerized tomography, magneticresonance imaging and positron emission tomography in study of brain trauma.J Neurosurg1986;64:760—767.

8. Alavi A, Fazekas F, Alves W, et at. Positron emission tomography in evaluation ofhead injury [Abstract]. J Cereb Blood Flow Metab I987;7(suppl I):S646.

9. Jamieson D, Alavi A, Jolles P. Chawluk JB, Reivich M. Positron emission tomographyin the investigation of central nervous system disorders. Radiol Clin North Am1988;26: 1075—1088.

10. Humayun MS. Presty 5K, Lafrance ND, Long DM, Wagner HN, Gordon B. Localcerebral glucose abnormalities in mild closed head injured patients with cognitiveimpairments. NucI Med Commun l989;I0:335—344.

I I. Jolles PR, Chapman PR, Alavi A. PET, CT and MRI in the evaluation of neuropsychiatric disorders: current applications. J NucI Med 1989;30:1589—1606.

12. Starkstein SE, Mayberg HS, Berthir ML, et at. Mania after brain injury: neuroradiological and metabolic findings. Ann Neurol 1990;27:652—659.

13. Tenjin H, Ueda 5, Mizukawa N, et al. Positron emission tomographic studies oncerebral hemodynamics in patients with cerebral contusion. Neurosurge@ l990;26:971—979.

14. Teasdale G, Jenneti B. Assessment of outcome coma and impaired consciousness: apractical scale. Lancet I974;2:81—84.

15. Marshall LF, Marshall SB, Klauber MR. et al. A new classification of head injurybased on computerized tomography. J Neurosurg 199l;75:S14—S20.

16. Hill DW. Methods ofmeasuring oxygen content ofblood. In: Payne JP, Hill DW, eds.A symposium on oxygen measurements in blood and tissues and their sign jflcance.London: Churchill; 1966:63.

17. Frackowiak RSJ, Lenzi GL, Jones I. Quantitative measurement of regional cerebralblood flow and oxygen metabolism in man using ISO and positron emissiontomography: theory, procedure and normal values. J Comput Assist Tomogr l980;4:727—736.

18. Phelps ME, Huang S-C, Hoffman EJ, Kuhi DE. Validation of tomographic measurement of cerebral blood volume with ‘‘C-labeledcarboxyhemoglobin. J NucI MedI979;20:328 —334.

19. Lammertsma AA, Jones I. Correction for the presence of intravascular oxygen-15 insteady-state technique for measuring regional oxygen extraction ratio in the brain: I.Description of the method. J Cereb Blood Flow Metab I983;3:4I6—424.

20. Hutchins GD, Holden JE, Koeppe Si, et at. Alternative approach to single-scanestimation of cerebral glucose metabolic rate using glucose analogs with particularapplication ischemia. J Cereb Blood Flow Metab I984;4:35—40.

21. Phelps ME, Huang S-C, Hoffman EJ, Selin C, Sokoloff L, KuhI DE. Tomographicmeasurement of local cerebral glucose metabolic rate in humans with “F-2-fluoro-2-deoxy-D-glucose: validationof method.Ann Neurol 1979;6:371—388.

rCMRglc

Normal valueof gray matter

Rt. frontalLt. frontalRt. temporalLt. temporalAt.occipitalLt. occipitalRt. parietalLt. parietalAverageNormalvalue

of white matterFit.white matterLt. white matterAverage

7.0 ±0.8 mg/bOOg/min

7.16.96.96.67.67.57.17.07.14.9 ±0.8 mg/bOOg/min

5.04.95.0

the rat using [14C]deoxyglucose autoradiography have beenreported (32,33). Kawamata et al. (33) speculated that excitatory amino acid-activated ion channels were involved in thepost-traumatic increase in glucose utilization, which reflects theenergy demand of cells required to drive pumping mechanismsagainst an ionic perturbation seen immediately after the concussive injury. These reports suggest that anaerobic glycolysiswould be observed not only in Patient 1 but also in Patients 2and 3 in the early phase after the injury. Such a state, however,was not maintained for a long period in Patients 2 and 3.

It was reported that the poor outcome of SDBI patients is wellcorrelated to low rCMRO2 values during the acute phase(34,35). According to the present data, a low rCMRO2 value isthought to be a necessary condition but not a necessary andsufficient indicator of poor outcome in these patients becausePatients 2 and 3 in this present report had low rCMRO2 valuessimilar to that in Patient 1. Significant findings from PETstudies in Patient 1 were that rCBF was the lowest, while rOEFand rCMRglc were the highest, compared with those values inthe other two patients. In the present study, although the numberof patients was limited to three, we speculate that a long-lastingstate of anaerobic glycolysis is an important factor in the pooroutcome of 5DB! patients.

PET IN DIFFUSEBRAININJURY•Yamaki et al. 1169

by on January 26, 2019. For personal use only. jnm.snmjournals.org Downloaded from

22. Tenjin H, Ueda 5, Mizukawa N, et at. The changes in hemodynamics with aging. JKyoto PrefUnis' Med 1989;98:1087—I093.

23. Jenntt B, Bond M. Assessment ofoutcome after severe brain damage: a practical scale.Lancet 1975; 1:480—484.

24. Baron JC, Rougemont D, Soussaline F, et al. Local interrelationships of cerebraloxygen consumption and glucose utilization in normal subjects and in ischemic strokepatients: a positron tomography study. J Cereb Blood Flow Metab 1984;4:l40—l49.

25. Luyten PR, Marien AJH, Heidel W, et al. Metabolic imaging of patients withintracranialtumors: ‘HMR spectroscopic imaging and PET. Radiology 1990;176:791—799.

26. King LR, McLaurin RL. Knowles HC Jr. Acid-base balance and arterial and CSFlactate levels following human head injury. J Neurosurg l974;40:617—625.

27. Enevoldsen EM, Cold G. Jensen FT. Malmros R. Dynamic changes in regional CBF,intraventricular pressure. CSF pH and lactate levels during the acute phase of headinjury. J Neurosurg l976;44:19l—2l4.

28. Stemau L, Thompson C. Dietrich WD. et al. Intracranial temperature—observations inhumanbrain[Abstract].J CerebBlood Flow Metab l99l;l l(suppl 2):5123.

29. William AP, Arthur JLC. Metabolic encephalopathies and coma. In: Siegal G,

Agranoff B, Albes RW, Molinoff P. eds. Basic Neurochemistry, 4th ed. New York:RavenPress; 1989:765—781.

30. Duckrow RB, LaManna JC, Rosenthal M, Levasseur JE, Patterson JL Jr. Oxidativemetabolic activity of cerebral cortex after fluid-percussion head injury in the cat.J Neurosurg1981;54:607—614.

31. Yang MS. DeWitt DS. Becker DP, Hayes RL. Regional brain metabolite levelsfollowing mild experimental head injury in the cat. J Neurosurg l985;63:617—621.

32. Yoshino A, Hovda DA, Kawamata T, Katayama Y, Becker DP. Dynamic changes inlocal cerebral glucose utilization following cerebral concussion in rats: evidence ofhyper- and subsequent hypometabolic state. Brain Res l991;56l : 106—I19.

33. Kawamata I, Katayama V. Hovda DA, Yoshino A, Becker DP. Administration ofexcitatatory amino acid antagonists via microdialysis attenuates the increase in glucoseutilization seen following concussive brain injury. J Cereb Blood Flow Metab1992; 12:12—24.

34. Jaggi JL, Obrist WD, Gennarelli TA, Langfitt TW. Relationship ofearly cerebral bloodflow and metabolism to outcome in acute head injury. J Neurosurg 1990;72: 176—I82.

35. Tomita H, Ito U, Tone 0, Masaoka H, Tominaga B, Horikawa N. CBF and CMRO2in early phase of head injury and outcome of patients with traumatic diffuse braininjury. Neurotraumatology l993;l6:l0l—105.

Brain injuryfollowingheadtraumaisaserious central nervous system disorderthat affects several hundred thousand individuals in the United States every year.Appropriate assessments of the degree ofdamage caused is essential in initiatingthe required therapies and determiningprognosis.

Over the past two decades, we havewitnessed many significant developmentsin the diagnosis and management of patients with head injury. The introductionof a single and reproducible grading system, the Glascow Coma Scale (GCS),was a major step toward categorizingpatients with regard to the severity ofinjury sustained (1 ). The utilization ofx-ray computed tomography in the early1970s and MRI in the l980s also madeimmeasurable contribution to the detection and characterization of a variety oflesions caused by head trauma (2,3).Despite such important advances, however, much needs to be accomplished tounderstand the underlying pathophysiologic and metabolic alteration that accompany two major injuries to the brain:ischemic cell damage and diffuse axonalinjury (DAI). Ischemic cell damage cccurs in over 90% of patients who succumb to head injury (4). In many victimsof head injury with postmortem ischemicor hypoxic lesions, no predisposing disorders are found prior to death. Diffuseaxional injury is probably the basicpathological damage in head injury (5).

Received Dec. 12, 1995; revision accepted Dec. 28,1995.

For correspondence or reprints contact Abass Alavi,MD,DM@onof NudearMedicine,117DonnerBldg.,Hospft@of the University of Philadelphia, 3400 SpruceSt.. Philadelphia. PA 19104.

This injury occurs as a result ofstretching and tearing of axons in thewhite matter of the cerebral hemisphere and the brainstem.

In addition, it has been demonstrated that brain injury results in aseries of molecular events that lead toaccumulation of toxic products whicheventually result in ischemia—reperfusion type of injury (6—8). This isconsidered the underlying mechanismfor ischemia noted with various insultsto the brain, including trauma or subarachnoid hemorrhage and stroke (9).Although postmortem studies have revealed clear evidence for focal andglobal ischemia in patients with headtrauma, no convincing clinical evidence for such a complication following head injury has been demonstratedby researchers (10—15). Most ischemic changes have been observed inthe “frontoparietal―watershed areas.These changes were noted in patientswho were severely injured (16). Mosthead trauma investigators believe thatischemia plays a distinct role in theclinical outcome of patient with headtrauma.

A powerful methodology which hasallowed investigation of hemodynamic and metabolic changes in thebrain is the noninvasive measurementof absolute cerebral blood flow (CBF)and some metabolic parameters following the intravenous administrationor inhalation of ‘33Xe(1 7). Utilizingthis powerful technique, the relationship among CBF, cerebral metabolicrate for oxygen (02) utilization(CMRO2) and the level of consciousness have been elucidated (14, 15).

With this technique, a measurement of onlylimited brain tissue sample is made utilizingdetectors which are placed over the skull.Metabolic rates for oxygen (CMRO2) can becalculated by measuring arterio-jugular 02difference and multiplying by the averageCBF estimates from the detectors selectedfor this purpose (15). Obviously, this technique provides information about superficialstructures of predetermined sites in thebrain. In spite of these shortcomings, muchknowledge has been gained by studyingpatients with acute head injury. It has beennoted that CMRO2 is consistently depressedin head-injured comatose patients andwhose magnitude is correlated well withGcS (15). However, an interesting observation has been made when CBF and CMRO2were compared in such patients. In abouthalf of the patients, CBF and CMRO2 arecoupled, as seen in normal states. In theremainder, an uncoupling of blood flow andmetabolism is clearly demonstrated. In thesepatients, relative hyperemia is detected despite significantly reduced CMRO2. In otherwords, CBF exceeds the metabolic requirements ofthe tissue perfused. In patients withhyperemia, there is a high incidence ofincreased intracranial hypertension. In spiteof its major contributions to the understanding of hemodynamic and metabolic consequences of head injury, the ‘33Xe CBFtechnique cannot demonstrate evidence forcerebral ischemia in this disorder.

The use of regional functional imagingtechniques such as SPECT, PET (18), nuclear magnetic resonance (NMR) (19,20)spectroscopy and stable xenon x-ray CT(21,22) has provided a unique opportunityto visualize and quantify several importantphysiologic and metabolic parameters thatare considered important in head injury.

1170 THEJOURNALOFNUCLEARMEDICINE•Vol. 37 •No. 7 •July 1996

EDITORIAL

Metabolic Consequences of Acute Brain Trauma:IsTherea Rolefor PET?

by on January 26, 2019. For personal use only. jnm.snmjournals.org Downloaded from

1996;37:1166-1170.J Nucl Med.   Tarumi Yamaki, Yoshio Imahori, Yoshio Ohmori, Eiji Yoshino, Takashi Hohri, Toshihiko Ebisu and Satoshi Ueda  PETCerebral Hemodynamics and Metabolism of Severe Diffuse Brain Injury Measured by

http://jnm.snmjournals.org/content/37/7/1166This article and updated information are available at:

  http://jnm.snmjournals.org/site/subscriptions/online.xhtml

Information about subscriptions to JNM can be found at:  

http://jnm.snmjournals.org/site/misc/permission.xhtmlInformation about reproducing figures, tables, or other portions of this article can be found online at:

(Print ISSN: 0161-5505, Online ISSN: 2159-662X)1850 Samuel Morse Drive, Reston, VA 20190.SNMMI | Society of Nuclear Medicine and Molecular Imaging

is published monthly.The Journal of Nuclear Medicine

© Copyright 1996 SNMMI; all rights reserved.

by on January 26, 2019. For personal use only. jnm.snmjournals.org Downloaded from