EFFECTS OF ANESTHETIC PROTOCOL ON NORMAL CANINE BRAIN

UPTAKE OF 18F-FDG ASSESSED BY PET/CT

MIN SU LEE, JEFF KO, AH RA LEE, IN HYE LEE, MI AE JUNG, BRENDA AUSTIN, HYUNWOO CHUNG,SANGSOEP NAHM, KIDONG EOM

The purpose of this study was to assess the effects of four anesthetic protocols on normal canine brain uptake of

2-deoxy-2-[18F]fluoro-D-glucose (FDG) using positron emission tomography/computed tomography (PET/CT).

Five clinically normal beagle dogs were anesthetized with (1) propofol/isoflurane, (2) medetomidine/pentobar-

bital, (3) xylazine/ketamine, and (4) medetomidine/tiletamine–zolazepam in a randomized cross-over design.

The standard uptake value (SUV) of FDG was obtained in the frontal, parietal, temporal and occipital lobes,

cerebellum, brainstem and whole brain, and compared within and between anesthetic protocols using the

Friedman test with significance set at Po0.05. Significant differences in SUVs were observed in various part of

the brain associated with each anesthetic protocol. The SUV for the frontal and occipital lobes was significantly

higher than in the brainstem in all dogs. Dogs receiving medetomidine/tiletamine–zolazepam also had signifi-

cantly higher whole brain SUVs than the propofol/isoflurane group. We concluded that each anesthetic protocol

exerted a different regional brain glucose uptake pattern. As a result, when comparing brain glucose uptake

using PET/CT, one should consider the effects of anesthetic protocols on different regions of the glucose uptake

in the dog’s brain. r 2010 Veterinary Radiology & Ultrasound, Vol. 51, No. 2, 2010, pp 130–135.

Key words: 18F-FDG, anesthesia, brain glucose uptake, dog, PET/CT, SUV.

Introduction

POSITRON EMISSION TOMOGRAPHY (PET) is useful for de-

tection of functional brain abnormalities before ana-

tomic changes develop.1–3 When PET is coupled with

computed tomography (CT) it provides more precise an-

atomic information than PET alone, especially when

knowing the spatial relationship between anatomic and

functional information is important.4–6 The use of 2-deoxy-

2-[18F]fluoro-D-glucose (FDG) PET/CT is being adapted in

the laboratory for brain function and tumor imaging tech-

niques,7–12 and also in veterinary practice for evaluation of

brain disease.3,13,14 When performing FDG PET/CT in

animals, movement of the animal during the scanning

process leads to artifacts and spurious results.4,5 Therefore,

to obtain accurate FDG PET/CT images of the brain or

other areas, animals must be anesthetized for imaging. In

humans, brain glucose metabolism is decreased under an-

esthesia compared with the conscious state.13–15 Similar

results have been found for different anesthetic protocols in

PET/CT studies with regard to body uptake of FDG in

rats and mice.7–10,16

Previously we described brain FDG PET/CT in two

dogs with necrotizing menegioencephalitis.3 We found

FDG PET/CT is a useful tool for detecting abnormal brain

glucose metabolism in dogs and relating this to regional

morphology. We also realized that anesthetic agents may

have influenced FDG uptake in different brain regions.

To assess this, we hypothesized that there was different

regional brain glucose metabolism in dogs using the same

anesthetic protocol. Furthermore, we also hypothesized

that different anesthetic protocols could result in different

glucose metabolism either within the whole brain or within

different brain regions. We attempted to quantify such

differences using several common anesthetic combinations

in normal canine brain using FDG PET/CT.

Materials and Methods

Five healthy Beagle dogs (two males and three females)

weighing between 5.6 and 6.8 kg and aged between 12 and

15 months were used. All dogs were in good health based

on clinical examination, clinical hematology, and skull ra-

diographs, computed tomography and magnetic resonance

imaging. Blood glucose measured before FDG adminis-

tration (mean¼ 97.4� 9.6mg/dl) was within the normal

reference range (75–128mg/dl) in all dogs.

Dogs were housed in a University Animal and Care

Committee approved facility and fed a standard diet. These

Address correspondence and reprint requests to Ki Dong Eom, at theabove address. E-mail: eomkd@konkuk.ac.krReceived July 7, 2009; accepted for publication October 27, 2009.doi: 10.1111/j.1740-8261.2009.01636.x

From the Department of Veterinary Diagnostic Imaging (MS Lee, ARLee, IH Lee, Jung, Eom) and the Department of Veterinary Anatomy(Nahm), the College of Veterinary Medicine, Konkuk University, Seoul,Korea; the Department of Nuclear Medicine of Konkuk UniversityMedical Center, Seoul, Korea (Chung); the Department of VeterinaryClinical Sciences, School of Veterinary Medicine, Purdue University, 625Harrison Street, West Lafayette, IN (Ko, Austin).

130

dogs were acclimated in the facility for 3 weeks before the

study. This study protocol was approved by the University

of Konkuk Institutional Animal Care and Use Committee.

This study was a cross-over design with each dog sub-

jected to four anesthetic protocols with a 2-week period

between protocols. The four anesthetic protocols were:

(1) propofol� (6mg/kg, IV) for orotracheal intubation

followed by isoflurane maintenancew administered at 1.5–

2% with 100% oxygen, (2) medetomidine premedication

(20mg/kg, IM), followed by pentobarbitalz (7mg/kg, IV)

15min later for induction and maintenance of anesthesia,

(3) xylazine hydrochloridey (1mg/kg, IM) premedication

with ketamine hydrochloridez (12mg/kg, IV) 15min later

for induction, and maintenance of anesthesia (4) mede-

tomidinek (20mg/kg, IM) premedication with tiletamine/

zolazepam�� (8mg/kg, IV) 15min later for induction and

maintenance of anesthesia.

These protocols and dosages were selected based on the

BSAVA Manual of Canine and Feline Anaethesia and

Analgesia.17 Dogs were fasted for 12 h before anesthesia

with free access to drinking water. A cephalic venous cath-

eter was placed for anesthetic, radioactive tracer and fluid

administration. Dogs received 10ml/kg/h of Lactated

Ringer’s solution during the scanning procedure.

FDG (185MBq, 5mci) was administered intravenously

before anesthetic drug administration and imaging. Follow-

ing FDG administration, the dogs were kept in a dim room

for an hour to minimize excitement and allow for FDG

uptake. After an hour, anesthesia was induced and PET/CT

performed with a Philips Gemini PET/CT system.ww There

were 28 detectors with an array of 29 � 22 gadolinuim

oxyorthosilicate crystals (4 � 6 � 20mm), providing an

18 cm field of view. Transverse and axial resolutions at

1 cm radius were 5.1 and 5.9mm full-width at half-maxi-

mum (FWHM). Transverse radial, transverse tangential,

and axial resolution at 10 cm were 5.4, 5.4, and 6.4mm

FWHM,18 respectively. The CT scanner was operated at

200mA and 120kV and a scan speed of 2 rotations/s.

The total duration of scanning for the PET/CT was

20min. The images were reconstructed through 3D Row

Action Maximum Likelihood Algorithms. The whole-body

protocol was used to measure SUV because specific Gemini

regional brain SUV software was not available. The PET

images were derived from brain sections with a slice thick-

ness of 4mm and acquired 170–200 section images. The CT

images were 3.2mm and acquired 370–400 section images.

The Syntegra ver 2.1F software was used for PET and CT

fusion images.

The FDG SUV was quantified by dividing the brain into

six regions: frontal, parietal, temporal and occipital lobes,

cerebellum, and brainstem. The region of interest (ROI) of

each brain region was determined by two veterinarians

and one person experienced in human nuclear imaging and

drawn manually on transverse (parietal and temporal), mid-

sagittal (frontal, occipital, cerebellum, brain stem), and

dorsal (parietal) images (Fig. 1). The SUV value with a ROI

around the focal FDG uptake zones was calculated with the

whole body protocol using formulation shown as follows19:

SUV ¼decay corrected activity ðkBqÞ

tissue ðmlÞinjected FDG dose ðkBqÞ

body weight ðgÞ

Mean values and 95% confidence intervals were calculated

for the appropriate SUVs. The mean SUV in each of the six

brain region was compared within and between each an-

esthetic protocol. In addition, the whole brain SUVs was

compared between anesthetic protocols. The data were an-

alyzed using the Friedman test with significance set at a

P value of 0.05. If a significant difference was detected,

Dunn’s multiple comparisons were performed.

Results

The results of SUV for each of the six brain region as

well as whole brain SUV is presented in Table 1 and Fig. 2.

The SUV for the frontal and occipital lobes was signifi-

cantly higher than in the brainstem in all dogs with each

Fig. 1. 18F FDG-PET/CT images. Regions of interest were manually drawn at the (A) transverse (the parietal lobe), (B) mid-saggital (the frontal and theoccipital lobes, the cerebellum and the brainstem), and (C) dorsal (the temporal lobe) images.

�Anepols

, Hana Pharm, Hwasung, Korea.wIfran

s

, Hana Pharm, Seoul, Korea.zEntobals, Hanlim, Youngin, Korea.yRompun

s

, Bayer Korea, Ansan, Korea.zKetamine

s

, Yuhan Co., Gunpo, Korea.kDomitor

s

, Pfizer, Korea, Seoul, Korea.��Zoletils, Virbac, Carros, France.wwPhilips, Eindhoven, the Netherlands.

131CANINE ANESTHETIC PROTOCOL ON BRAIN18

F-FDGUPTAKEVol. 51, No. 2

anesthetic protocol. In addition, in the medetomidine/

pentobarbital dogs, the occipital lobe had a significantly

higher SUV when compared with the temporal lobe (Fig.

2). In dogs receiving medetomidine/tiletamine/zolazepam,

the occipital lobe had a significantly higher SUV when

compared with the temporal lobe and the cerebellum. In

general, the SUV in the subcortical regions was signifi-

cantly lower than in the cerebral cortex regions (Fig. 3 and

Table 1). The whole brain SUV with medetomidine/tileta-

mine/zolazepam was significantly higher than in dogs re-

ceiving propofol/isoflurane (Fig. 2).

Discussion

The results of this study confirmed our first hypothesis

that there was a significant difference in different brain re-

gional SUV using the same anesthetic protocol. The frontal

and occipital lobes had a consistently higher SUV compared

with the cerebellum and the brainstem in all dogs with each

anesthetic protocol (Fig. 2). The precise reason for this

substantial regional high SUV in the frontal and occipital

lobes in dogs is unknown. Glucose metabolism is increased

in the cerebral cortex because of the distribution of as-

trocytes, which consume glucose strongly.20 In humans, the

frontal lobe is related to reasoning, planning, speech, move-

ment, and emotions. These behavioral tasks may not be

applicable in dogs and is difficult use to explain such ob-

served SUV differences. The occipital lobe is related to vi-

sual processing, and could partially explain the increased

glucose utilization due to increased neuronal activity when

compared with the cerebellum and brainstem in these dogs.

Besides the aforementioned differences in glucose metab-

olism between the frontal and occipital lobes with other brain

regions, data obtained from dogs anesthetized with mede-

tomidine/pentobarbital and medetomidine/tiletamine-zolaze-

pam also showed a significantly higher glucose metabolism

in the occipital lobe than the temporal lobe and, both tem-

poral and cerebellum, respectively (Fig. 2). These observed

regional SUV differences could be medetomidine specific

because this was not observed in other anesthetic protocols.

Our second hypothesis was that different anesthetic pro-

tocols could result in different whole brain glucose metab-

olism was supported by the fact that the dogs receiving

medetomidine/tiletamine/zolazepam had a significantly in-

creased glucose metabolism in the entire brain compared

with dogs receiving propofol/isoflurane. The effect of

propofol and isoflurane anesthesia on glucose metabolism

has been well demonstrated in humans. Glucose metabo-

lism is depressed as much as 55% with propofol and 46%

with isoflurane, respectively, when compared with the

awake state in humans.13 The depressed glucose uptake is

due to metabolic reductions occurring during anesthesia.21

Gamma aminobutyric acid type A (GABAA) receptors are

one of the most important inhibitory neurotransmitter

Table1.ResultsofSUVsin

Six

Brain

Lobes

andWhole

Brain

DependonDifferentAnesthetic

Protocols

FrontalLobe

Parietal

Lobe

Tem

poralLobe

OccipitalLobe

Cerebellum

Brainstem

P-value

ANOVA�

Whole

Brain

Mean

CI

Mean

CI

Mean

CI

Mean

CI

Mean

CI

Mean

CI

Mean

CI

Pro–Iso(n¼5)

4.21w

(3.96–4.46)

3.82

(3.44–4.20)

3.46

(2.97–3.95)

4.78z

(4.07–5.48)

3.53

(3.09–3.96)

2.99

(2.67–3.31)

o0.001

3.79

(3.13–4.45)

M—

P(n¼5)

4.52w

(4.22–4.82)

4.04

(3.71–4.37)

3.64

(3.37–3.91)

4.99z

,y(4.18–5.84)

3.85

(3.52–4.19)

2.85

(2.44–3.26)

o0.001

3.98

(3.20–4.75)

X—

K(n¼5)

4.75w

(3.97–5.54)

4.45

(3.28–5.62)

3.91

(3.09–4.74)

5.10z

(4.51–5.70)

3.55

(3.03–4.07)

3.14

(2.72–3.55)

o0.001

4.15

(3.36–4.93)

MZT(n¼5)

4.94w

(3.68–6.19)

4.64

(3.38–5.09)

3.94

(2.83–5.05)

5.93z

,y,z

(4.26–7.59)

3.76

(2.91–4.61)

3.83

(3.09–4.56)

o0.001

4.50k

(3.62–5.39)

P-valueANOVA�

0.52

0.56

0.94

0.77

0.65

0.10

o0.001

Data

arepresentedasthemean(95%

CIs)in

parentheses.� F

riedmantestsrevealedsignificantdifference

insixbrain

lobes

witheach

fouranestheticprotocols(Po0.05).

w Significance

difference

in

thefrontallobevsthebrainstem

.zSignificance

difference

intheoccipitallobevsthebrainstem

.y Significance

difference

intheoccipitallobevsthetemporallobe.

zSignificance

difference

inthe

occipitallobevsthecerebellum.k Significance

difference

inPro-Iso

vsMZT.Pro–Iso,propofol–isoflurane;

M–P,medetomidine–pentabarbital;X–K,xylazine–ketam

ine;

MZT,medetomidine–

zolazepam–tiletamine;

SUV,standard

uptakevalue.

132 LEE ET AL. 2010

systems in the brain and spinal cord and considered to be

the major molecular target of general anesthetics.22,23

Propofol and isoflurane involve modulation of inhibitory

function through GABAA receptors. Propofol appears to

suppress glucose metabolism more in the cortical region

than the subcortical region in humans, while isoflurane

appears to decrease the glucose metabolism uniformly.13,24

The difference seen in glucose metabolism with propofol is

Fig. 2. Comparison of features with 18F FDG-PET/CT images on transverse, mid-sagittal, and dorsal views between propofol/isoflurane (A), medetomidine/pentabarbital (B), xylazine/ketamine (C), and zolazepam/tiletamine (D). Red or yellow regions correspond to hypermetabolism, meaning higher glucose uptake,while blue or black regions are related with hypometabolism, meaning lower glucose uptake. The cerebral cortical regions are characterized by hypermetabolismwith yellowish and reddish color compared with the cerebellum and the brainstem regions with greenish or blue color meaning hypometabolism.

8.00

6.00

5.00

7.00Frontal lobe

Parietal lobe

11

22,3 2

2,3,4

5

3.00

4.00

Temporal lobe

Occipital lobe

Cerebellum

SU

Vs 1

1

1.00

2.00Brainstem

Whole brain0.00

Pro-Iso

Anesthetic protocols

M-P X-K MZT

Fig. 3. Standard uptake value (SUV) of 2-Deoxy-2-[18F]fluoro-D-glucose (FDG) obtained in the frontal, parietal, temporal, and occipital lobes, cerebellum,brainstem, and whole brain of normal dogs (n¼ 5) with different anesthetic protocols. Statistical significance (Po0.05) in SUVs associate with differentanesthetic protocol between brain regions are shown as follows: 1significance difference in the frontal lobe vs. the brainstem; 2significance difference in theoccipital lobe vs. the brainstem; 3significance difference in the occipital lobe vs. the temporal lobe; 4significance difference in the occipital lobe vs. the cerebellum;5 significance difference in propofol–isoflurane vs. medetomidine–zolazepam–tiletamine anesthetic protocol. (Pro–Iso: propofol–isoflurane; M–P: medetomi-dine–pentabarbital; X–K: xylazine–ketamine; MZT: medetomidine–zolazepam–tiletamine).

133CANINE ANESTHETIC PROTOCOL ON BRAIN18

F-FDGUPTAKEVol. 51, No. 2

likely related to the disparity of GABAA receptor distri-

bution in the central nervous system.25,26 Propofol induced

metabolic changes in the rat have been described as a ho-

mogenous depression within the brain parenchyma.27

Medetomidine stimulates a-2 adrenoreceptors in the

pancreas, which suppresses insulin release and in turn

induces a hyperglycemic effect.28,29 Medetomidine compet-

itively inhibits FDG uptake of by brain.20 Tiletamine–

zolazepam (Zoletil or Telazol) is composed of two

anesthetic agents, zolazepam as a benzodiazepine tranquil-

izer and tiletamine as a dissociative anesthetic that has a

similar mechanism of action as ketamine.17 Tiletamine acts

as a noncompetitive N-methyl D-aspartate antagonist and

results in increased or normal brain glucose metabolism

compared with the conscious state.10,30 The duration of

tiletamine is longer than ketamine.17 In rats, cerebral glu-

cose metabolism was increased with ketamine alone, com-

pared with the conscious state.31 The results of our study

on whole brain SUVs were similar to those previous studies

reported in rats that medetomidine/tiletamine/zolazepam

increased in glucose metabolism, whereas propofol/isoflu-

rane tends to decreased in glucose metabolism.

The technique used in the current study provided some

advantages when compared with traditional cerebral glu-

cose metabolism study. For example, arterial blood sam-

pling is commonly required to evaluate cerebral glucose

metabolism with 18F-FDG PET.13,24,27,32 Arterial blood

sampling is inconvenient and invasive and cumbersome. In

the currently, we chose to calculate the SUVs which do not

require frequent arterial blood sampling and therefore

provide a relative convenient way to estimate SUV as

cerebral glucose metabolism in dogs.

There are many factors, which can influence the results of

SUVs. Technical factors such as attenuation correction, ob-

ject size and weight, partial volume effect, ROI, dose ad-

ministrated and route of tracer administration can all play a

role in influencing SUV outcome.33 The differences in SUV

observed in different brain regions in this study could be

partially explained by two main factors. First, anesthetic ac-

tions activated different portions of the brain and resulted in

different regional cerebral glucose metabolisms, therefore

different SUV values. Second, anesthetic effects resulted in

different levels of oxygenation and CO2 in cerebral paren-

chymal tissues. Both cerebral oxygenation and CO2 mediated

cerebral blood flow changes could affect the cerebral tissue

metabolism and result in the different SUV values observed.

The main weakness of this study was that cardiorespira-

tory variables of the anesthetized dogs were not strictly con-

trolled. For example, in the isoflurane/propofol group, 100%

oxygen was provided due to the need of 100% oxygen as an

isoflurane carrier. In contrast, for the rest of the treatment

groups the dogs were allowed to breathe room air. Never-

theless, the results of our study demonstrated that (1) SUV

coupled with PET–CT technique can be used in the anesthe-

tized dogs for evaluating of cerebral glucose metabolism; (2)

there was evidence of different regional brain SUVs in dogs

with the same anesthetic protocol, (3) different anesthetic

protocols could result in different whole brain SUVs in dogs.

In conclusion, measuring SUVs using 18F-FDG cou-

pling with PET/CT technique can be used as a semi-quan-

titative method to evaluate cerebral glucose metabolism in

the anesthetized dogs. Furthermore, when evaluating cere-

bral glucose metabolism in the anesthetized dogs, anesthe-

sia protocols play a key role in influencing SUV result.

ACKNOWLEDGMENT

This work was supported by National Research Foundation of KoreaGrant funded by the Korean Government. KRF-2008-A003-0061.

REFERENCES

1. Chen W. Clinical application of PET in brain tumors. J Nucl Med2007;48:1468–1481.

2. Weber WA, Avril N, Schwaiger M. Relevance of positron emissiontomography (PET) in oncology. Strahlenther Onkol 1999;175:356–373.

3. Eom KD, Lim CY, Gu SH, et al. Positron emission tomographyfeatures of canine necrotizing meningoencephalitis. Vet Radiol Ultrasound2008;49:595–599.

4. Blodgett TM, McCook BM, Federle MP. Positron emission tomog-raphy/computed tomography: protocol issues and options. Semin Nucl Med2006;36:157–158.

5. Cohade C, Wahl RL. Applications of positron emission tomography/computed tomography image fusion in clinical positron emission tomogra-phy-clinical use, interpretation methods, diagnostic improvements. SeminNucl Med 2003;33:228–237.

6. Blodget T, Meltzer CC, Townsend DW. PET/CT: form and function.Radiology 2007;242:360–385.

7. Toyama H, Ichise M, Liow JS, et al. Evalution of anethesia effect on[18F]FDG uptake in mouse brain and heart using small animal PET. NuclMed Biol 2004;31:251–256.

8. Woo SK, Lee TS, Kim KM, et al. Anesthesia condition for 18F-FDGimaging of lung metastasis tumors using small animal PET. Nucl Med Biol2008;35:143–150.

9. Matsumura A, Mizokawa S, Tanaka M, et al. Assessment of micro-PET performance in analyzing the rat brain under different types of anes-thesia: comparison between quantitative data obtained with microPET andex vivo autoradiography. Neuroimage 2003;20:2040–2050.

10. Momosaki S, Hatano K, Kawasumi Y, et al. Rat-PET study withoutanesthesia: anesthetics modify the dopamine D1 receptor binding in ratbrain. Synapse 2004;54:207–213.

11. Ballegeer EA, Forrest LJ, Jeraj R, et al. PET/CT following intensity-modulated radiation therapy for primary lung tumor in a dog. Vet RadiolUltrasound 2006;47:228–233.

12. LeBlanc AK, Jakoby BW, Townsend DW, et al. 18FDG-PETimaging in canine lymphoma and cutaneous mast cell tumor. Vet RadiolUltrasound 2009;50:215–23.

13. Alkire MT, Haier RJ, Shah NK, et al. Positron emission tomogra-phy study of regional cerebral metabolism in humans during isofluraneanasthesia. Anesthesiology 1997;86:549–557.

14. Alkire MT, Haier RJ, Baker SJ, et al. Cerebral metabolism duringpropofol anesthesia in humans studied with positron emission tomography.Anesthesiology 1995;82:393–403.

15. Alkire MT. Quantitative EEG correlations with brain glucosemetabolic rate during anesthesia in volunteers. Anesthesiology 1998;89:323–333.

134 LEE ET AL. 2010

16. Akira M, Shigekazu M, Masaaki T, et al. Assessment of micro PETperformance in analyzing the rat brain under different types of anesthesia:comparisom between quantitative data obtained with microPET and ex vivoautoradiorraphy. Neuroimage 2003;20:2040–2050.

17. Kastner SBR. Intravenous anaesthetics. In: Seymour C, NovakovskiTD (eds): BSAVA manual of canine and feline anaesthesia and analgesia,2nd ed. Gloucester: British Small Animal Veterinary Association, 2007;133–149.

18. Kim JS, Lee JS, Lee BI, et al. Performance Characteristics of 3DGSO PET/CT Scanner (Philips GEMINI PET/CT). Korean J Nucl Med2004;38:318–324.

19. Keyes JW Jr. SUV: standard uptake or silly useless value? J NuclMed 1995;36:1836–1839.

20. Magistretti PJ, Pellerin L. Cellular mechanisms of brain energymetabolism and their relevance to functional brain imaging. Philos Trans RSoc Lond B Biol sci 1999;354:1155–1163.

21. Alkire MT, Haier RJ. Correlating in vivo anaesthetic effects withex vivo receptor density data supports a GABAergic mechanism of actionfor propofol, but not for isoflurane. Br J Anaesth 2001;86:618–626.

22. Franks NP, Lieb WR. Molecular and cellular mechanisms of generalanaesthesia. Nature 1994;367:607–614.

23. Campagna JA, Miller KW, Forman SA. Mechanisms of actions ofinhaled anesthetics. N Engl J Med 2003;348:2110–2124.

24. Kaisti KK, Liisa M, Mika T, et al. Effect of surgical levels ofpropofol and sevoflurane anesthesia on cerebral blood flow in healthy sub-

jects studied with positron emission tomography. Anesthesiology 2002;96:1358–1370.

25. McKernan RM, Whiting PJ. Which GABAA-receptor subtypesreally occur in the brain? Trends Neurosci 1996;19:139–143.

26. Velly LJ, Rey MF, Bruder NJ, et al. Differential dynamic of actionon cortical and subcortical structures of anesthetic agents during inductionof anesthesia. Anesthesiology 2007;107:202–212.

27. Dam M, Ori C, Pizzolato G, et al. The effect of propofol anesthesiaon local cerebral glucose utilization in the rat. Anethesiology 1990;73:499–505.

28. Lee KH, Ko BH, Paik JY, et al. Effects of anesthetic agents andfasting duration on 18F-FDG biodistribution and insulin levels in tumor-bearing mice. J Nucl Med 2005;46:1531–1536.

29. Ambrisko TD, Hikasa Y. Neurohormonal and metabolic effects ofmedetomidine compared with xylazine in beagle dogs. Can J Vet Res 2002;66:42–49.

30. Itoh T, Wakahara S, Suzuki K, et al. Effects of anesthesia upon 18F-FDG uptake in rhesus monkey brains. Ann Nucl Med 2005;19:373–377.

31. Eintrei C, Sokoloff L, Smith CB. Effect of diazepam and ketamineadministered individually or in combination on regional rates of glucoseutilization in rat brain. Br J Anaesth 1999;82:596–602.

32. Heinke W, Schwarzbauer C. In vivo imaging of anaesthetic action inhumens: approaches with positron emission tomography (PET) and func-tional magnetic resonance imaging (fMRI). Br J Anaesth 2002;89:112–122.

33. Huang SC. Anatomy of SUV. Nucl Med Biol 2000;27:643–646.

135CANINE ANESTHETIC PROTOCOL ON BRAIN18

F-FDGUPTAKEVol. 51, No. 2