CLINICALSCIENCES

In 1973 Russ et al. (1,2) developeda steady-statetechnique for imaging the distribution of gases labeledwith oxygen-I 5 (Tip 122 sec). Jones et al. (3), andlater in more detail Subramanyam et al. (4), suggesteda theoretical treatment, applicable to the brain, for obtaming regional oxygen extraction from data obtainedby this technique. Their treatment did not include oxygenutilization capability. Another study used the steadystate technique to monitor changes in tumor and surrounding tissues caused by radiation therapy (5,6). Thisstudy gave empirical evidence for an oxygen-utilizationinterpretation of the observed changes based upon thesuggestion that tissues exchanging water slowly may notrequire correction for either water clearance or recirculation.

* Present address: Data General Corporation, MS/A 1 21 , 4400

Computer Drive, Westboro,MA 01580.ReceivedJuly I 5, 1980;revisionacceptedJune29, 1981.For reprintscontact:R. E. Bigler,BiophysicsLaboratory,Memorial

Sloan-Kettering Cancer Center, I 275 York Ave., New York, NYI0021.

A difficulty for analyses aimed at deriving oxygenutilization information from the steady-state approachis the need to account for the circulation of labeled waterof metabolism (H215O) from and to the tissue underinvestigation following its production in, and clearancefrom, all body tissue. The purpose of this work is toprovide an analysis of oxygen utilization, including recirculation of labeled water of metabolism, based on amulticompartment model (7). The modeling was carriedout with the SAAM (simulation, analysis, and modeling)program of Berman et al. (8,9).

Modelfortotal-bodywater.We beginwithananalysisof water distribution in the whole body. We first establisha three-compartment model consisting of central pool(includingthe water containedin the arterial bloodsystem), lumped slowly exchanging tissues, and lumpedrapidly exchanging tissues. This model is subsequentlyextended to include cerebral white and gray matter. Toestablish the parameters for the three-compartmentmodel, we use the data of Edelman (10). His experimentalstudyusedD2O asa tracerfor water,andpro

Volume 22, Number 11 959

INVESTIGATIVE NUCLEAR MEDICINE

Compartmental Analysis of the Steady-State Distribution of 1502 and H2150 in

Total Body

Rodney E. Bigler, Jeffrey A. Kostick,@ and John A. Gillespie

Memorial Sloan-Kettering Cancer Center and Lehman College, City University of New York, New York, New York

it hasbeensuggestedthat regionaloxygenmetabolismmay be measuredquantitativelyby analysisof the steady-statedistributionof 0—15(T112= 122 sec). Forthisanalysiswe havedevelopeda compartmentalmodelthat incorporatescorrectionsdue to clearance and recirculationof water of metabolism.The oxygenutilizationrate is simplyproportionalto the local 0-15 activity if water of metabolismisnotrecirculatedfromothertissuesandisnotlostto the circulationfora time longcomparedwiththe half-lifeof0-15. We evaluatedthe magnitudeof biologicalmetabolic water lossanduptakein the steadystate. Ouranalysisindicatesthat themagnitudeof these effects for rapidlyexchangingtissues(such as cerebral grayand white mafter), may precludea simple,noninvasive,and quantitativedeterminationof regionaloxygenmetabolism.Slowlyexchangingcompartments,however(such as skeletal muscleandperhapssometumors),appearamenableto correctionforclearanceandrecirculationeffectswithsufficientaccuracyto make determinationsof regionaloxygenmetabolismfeasible.

J Ned Med 22: 959—965,1981

Characteristic Mean±standarddeviation

TABLE2. ThREE-COMPARTMENTWATER-MODELPARAMETERS

Male Tracer rate constants (min@) Compartmeotvolumes (liters)subjects k12 k21 k13 k31 1 2 3

MC 0.159 0.162 0.0273 0.0486 11.8 12.0 21.0CB 0.191 0.146 0.0240 0.0605 9.9 7.6 25.0RHC 0.180 0.250 0.0289 0.0433 10.0 13.9 15.0NR 0.160 0.174 0.0275 0.0718 8.7 9.5 22.7PT 0.190 0.177 0.0276 0.0571 11.2 10.3 23.2

Mean 0.176 0.182 0.0271 0.0563 10.3 10.7 21.4

BIGLER,KOSTICK.AND GILLESPIE

TABLE1.FIVEMALEHUMANSUBJECTSINDEUTERIUM OXIDE STUDY OF EDELMAN

Weight 70.8 ±5.8 kg23.0 ±4.1 yr42.0 ±2.6 I59.4 ±2.710.3 ±1.2 I14.6 ±1.7

AgeTotal-bodywater% bodyweightZero-time volume of dilution% body weight

. Reference 10.

vided arterial blood clearance data as well as total-bodyvolume of distribution from equilibrium measurements.The characteristics of Edelman's subject population aregiveninTable 1.

The three-compartment model is illustrated in Fig.I . The tracer input to Compartment 1 may be either bya bolus injection into the patient's arterial system or byconstant administration, e.g., inhalation. The physicalradioactive decay constant for the tracer is indicated byA. The intercompartmentaltracerrateconstants@(these can be derived from clearance half-times: k@= (ln2)/t1@)aredeterminedusingBerman'sSAAM program(8) and the arterial disappearance parameters given in

Ref. 10. (For the deuterium experiments A = 0, sincedeuterium is stable.) The results of the rate-constantcalculationsaregiveninTable2.Therapidlyexchangingtissues (Compartment 2) have a clearance half-time of3.9 ±0.4 mm; the slowlyexchangingtissues(Compartment 3) 25.6 ±1.7 mm. According to Edelman (10)the viscera form the bulk of the rapidly exchangingtissues,while skeletalmuscleand supportingstructures(probably skin and connective tissue as well as bone)constitute the slowly exchanging tissues.

Once the parameters k1@have thus been fixed by thebolus experiments, we use the model to determine theH2@@Odistributionsduring steady-stateadministrationof H215Ointo Compartment I . In this case it is essentialto take intoaccountthe radioactivedecayconstantX,which for 0-15 is 0.341 min1.

Thesemodelcalculationsprovidea determinationof

960 THE JOURNAL OF NUCLEAR MEDICINE

Input

k21 J@2 k12@@ @k13 ] 3

@ iFIG.1. Three-compartmentmodelforanalysisofbody-waterdistribution: (1)central pool, (2) rapIdlyexchangingtissues, (3) slowlyexchanging tissues. Input is by bolos injection or by constant InfuSiOnforsteady-statestudies;A isthephysicalredlOaCtiVedecayconstant,andtracer rateconstantsk, interconnectingthehoe compartments,are shown.

the activity concentrations (as well as the mean deviations for our patient sample) in the three compartmentsfor any rate of administration of H215O activity intoCompartment 1. The arterial blood concentrationachieved in Compartment 1 during steady-state administration is an essential parameter for analysis ofregional cerebral blood flow (11,12). Figure 2 shows theactivity concentrations (in @zCi/ccwater) at steady statein the three compartments when the administration rateof H215Ointo Compartment 1 is I mCi/mm.

We notethat thedramaticdifferencein the two tissuecompartments (2 and 3) excludes any simple modelwhere the tissue types are lumped as one. At steady state,the rapidly exchanging compartment shows 34% of theactivity concentration of Compartment I , and the slowlyexchanging compartment shows 7%.

For all compartments the standard deviation as apercent of the mean (across Edelman's patient sample)is approximately 12%. In Fig. 2 and all other figures, thisdeviation is indicated by the error bars.

Cerebral tissue: steady-state H2@Odistribution.Tostudy exchange effects in cerebral white and gray matter,whichturnwateroververyrapidly,weusetheclearancedata for 0-1 5-labeled oxyhemoglobin given by TerPogossian et al. (13,14), shown in Table 3, together withwater volumes for white and gray matter of 0.58 and 0.511,respectively(15—17).Theclearanceparameterswereobtained by averaging data from the frontal, parietal,and occipital brain regions of each patient, then calcu

TABLE3. OXYGENCEREBRALBLOOD-FLOWPARAMETERSMatterClearance

half-times(sec)Low

High Mean ±s.d.rCBF±s.d.

(ml/min/100g)@‘ay

White30.938.0 34.4 ±3.6

127.4 142.0 134.7 ±7.4126±13

27.2 ±1.5.

Reference 13. Twomales aged 41 and 42 and two females aged 68 yr with brain disease.

CLINICAL SCIENCESINVESTIGATIVE NUCLEAR MEDICINE

0.19

0.066

FIG. 2. Steady-stateactivityconcentrations,In @Ciper cc ofbody H20, for constant H21@Oadministration. N 5(10).

lating a mean and deviation over the group of patients.

Water clearance can be investigated either by administration of 0-1 5-labeled oxyhemoglobin whose released oxygen is subsequently converted to water ofmetabolism, or by administration of H215O. The subsequent egress rate of water is identical for the twomethods within experimental error (13,14). We used theoxyhemoglobin data since they are more directly relatedto the metabolicprocess.

Our five-compartmentmodel is shownin Fig. 3 foranalysis of the steady-state administration of H2' @Oatthe rate R1 to the water compartment (1), with Cornpartments 4 and 5 representing, respectively, cerebralwhiteandgraymatter.Compartment2 wasreducedbyan amount equal to the water volumes of white and graymatter, and k12adjusted so that the product ofk21 withthe volume of Compartment 1 (see Table 2) remainedequal to the product of the new volume of Compartment2 with 2(conservationof mass).Compartment 3 isasin Fig. I . The steady-stateactivityconcentrationlevels(in zCi/cc water) are shown in Fig. 4.

The deviations across the patient sample are againindicated by the error bars. Note that these do not indudevariationsin thewatervolumeor in theclearanceparameters of white and gray matter, which were fixedin thesecalculations.

Clearly H215Odoes not distribute in direct proportionto the water volume in each tissue (the activity concentration is not equal in all compartments). The modelwould not allow an estimate of the water content of the

FIG.3. Five-compartmentmodel, with cerebral white (4) and gray(5) compartments,foranalysisof body-waterdistribution.

various tissues more accurate than that indicated by theerror bars in Fig. 4, even in a group of individuals whoclosely approximate the Edelman sample. Although themodel may be further elaborated, it is also evident thatcompartmental activity is not directly proportional toperfusion (18). This confirms similar conclusions in otherstudies (11,12).

Steady-state ‘@02distribution (water of metabolism).Once our model for body-water distribution has beendeveloped, it can be used to investigate the concentrationof H215Oof metabolism derived from ‘@O2utilization.In this case, the H2150is producedin Compartments 2and 3 by metabolismof ‘@O2,rather than being introduced initially into Compartment 1 as H2150. Figure 5shows the compartment model used to evaluate the dynamics of water of metabolism.

On the plausibleassumptionthat the oxygenmetabolism of brain, heart, kidney, liver, and lung can be described as components of Compartment 2 (rapidly exchanging water) with 79% of the total-body oxygenmetabolism, the rest of the body's oxygen metabolism(21%) is assigned to the slowly exchanging Compartment 3 (/9). This is equivalent to setting R2 = 0.79R1and R3 = 0.21R1, where the R1 are the constant rates atwhichoxygenissuppliedtothevarioustissues,i, bythearterial blood. Total oxygen utilization used for thesecalculations is 22. 13 1oxygen/hr in the resting state (20).The specific activity of the delivered oxygen is taken to

Volume 22, Number I I 961

BIGLER,KOSTICK.AND GILLESPIE

FIG.4. Steady-stateH2150activity, in sClper cc of bodyH20, forfive-compartment model. N 5 ( 10)plus 4 ( 13)humansubjects.

be I mCi/l ofoxygen gas.The resultant distribution of activity (in mCi) for the

model of Fig. 5 is shown in Fig. 6. The shaded distribution is that due to metabolism, assuming it is not redistributed within the body after its formation (the centralcompartment, I , would have no activity without recirculation). Error bars indicate the variation over the patient sample, which is approximately ±3%.

Figure 7 shows a breakdown of fast and slow cornpartments giving percentage gain and loss relative to100%produced—e.g.,8.4%gainand34.1%lossfor thefast compartment. Much smaller changes are seen forthe slowly exchanging tissues, with the difference between produced and observed activities being only + 6.6±3%.

Oxygen metabolism with cerebral tissue. Our fivecompartment model for oxygen of metabolism, includingbrain white and gray matter (Compartments 4 and 5,respectively), is shown in Fig. 8. Combined white andgray cerebral tissue accounts for 21% of the body'soxygen metabolism (19,21 ). Both in-vivo and in-vitrostudies indicate that the rate of metabolism in graymatter is about twice that in white matter (22,23). Thisratio is confirmed in recent studies of cerebral glucosemetabolism (24,25). Hence R4 = 0.07R1 and R5 =

I . The rapid body compartment is then reduced to

58%(R2= 0.58R1).Assuming that the activity is delivered at 1 mCi/l

oxygen, the resultant millicurie contents for Compartments 1-5 are 0.260 ±0.015, 0.487 ±0.021, 0.252 ±0.008,0.0460±0.0006,and0.0449±0.0011,respectively.

Figure 9 shows the predicted contents for white andgray matter, normalized to 100%, together with recirculation gains and losses. White matter shows a 47.7%loss and 8.1% gain; gray matter shows a 77.8% loss and7.2% gain.

When the parameters of Table 3 for brain waterclearance are varied ±1 s.d. around the mean values

c... Without...... Recirculatson

Slow2

Fast

FIG.6. Steady-stateactivitycontents(inmCi)forthree-compartment waterof metabolismmodelof Fig.5, with recirculation(clear)and without (shaded).N 5(10).

962 THE JOURNAL OF NUCLEAR MEDICINE

F2 ______ 73Lk21 I 1 k3@J

2@ k124 1 f@@,ki3j 3

@% @‘ I

FIG.5. Ttwee-compartmentmodel for water of metabolism.

(with the other model parameters held at their averagevalues) the observed values of Fig. 9 vary by ±4% and±10%for white and gray matter, respectively. Obviouslythe loss component of the correction is most sensitive tovariations in the clearance parameters.

DISCUSSION

The ability to perform low-hazard, noninvasive, andreasonably accurate regional oxygen utilization measurements in vivopromises great value for the assessmentof changes in oxygen metabolism caused by diseasee.g., infarcts, necrosis, inflammation, tumor growth, andtumor destruction produced by treatment. It is generallybelieved that functional disturbances occur earlier thanstructural changes caused by disease and disease treatment, and that in most cases earlier appreciation offunctional abnormalities would improve the success oftreatment. The initial suggestion that steady-state administration of ‘@O2might provide such a method (1)was followed by preliminary reports that local functionalchanges could apparently be monitored in the brain(26)—and in locations possibly involving tumorswhere water is exchanged slowly (5).

The results of the multicompartmental analysis provided by this study allow a quantitative determinationof the accuracy with which the oxygen- I5 steady-statemethod (1,2) can be applied to measure regional oxygenutilization in tissues that exchange oxygen- 15-labeledwater of metabolism at a variety of rates, and in which

rrLLCentral

Comportment

CLINICAL SCIENCESINVESTIGATIVE NUCLEAR MEDICINE

White Gray

FiG.9. Steady-statedistributionfor white andgraycerebral tissue,obtainedfrom five-compartmentmodelof waterof metabolism,withpercentages of locally produced water of metabolism gained andlostfromrecirculation.

is approximately ±3%. If a H215O or deuterium oxidebolus infusion study, with arterial disappearance measurement, were carried out on each patient, this variationcould be further reduced.

The water-exchange situation for gray and whitematter can be evaluated from Fig. 9. Without the model,the error in predicting the amount of water of metabolism produced from that observed in white matter wouldbe —40%,and in gray matter it would be —71%.

The magnitudes of these recirculation correctionswere seen to be sensitive to the cerebral water-clearanceparameters. Evaluation of the cerebral water kineticsexperimentally in each patient would require a carotidartery administration of the tracer (H2150) and measurement of regional brain-activity disappearance (13).It is possible that the cerebral clearance parameterscould be determined less invasively by using a singleinhalation of C15O2with subsequent arterial sampling.Note that if oxygen utilization is to be measured in aregion of the brain (such as a tumor site), the rate forclearance of water from the region must be determinedfrom the brain disappearance data. This is not likely tobe possible with the probe system in current use (13), atleast for small tumors.

If white and gray matter cannot be resolved quantitatively so as to determine the fraction of each tissue typein a region of interest, our model-derived correctionwould be between that of all white and all gray matter(—40%to —71%).Currentlyavailableimagingprocedures cannot resolve gray and white matter quantitatively unless the object size is known (28). A high-resolution transaxial transmission computed tomographicimage of the region could provide the data on object size,thus reducing this error substantially.

The effects of clearance and recirculation of water ofmetabolism can be reduced by using a tracer for oxygenwith a shorter half-life. Oxygen-I4 (T112 71 sec) issuitable for this purpose, and it has been prepared in

FIG. 7. Steady-state distribution of water of metabolism, withpercentagesof activityproduced,gained,andlost from recirculation,for fastandslowbodycompartments.N 5(10).

recirculation of oxygen-i 5-labeled water of metabolismis evaluated.

Qualitatively we can expect that tissues with low bloodflow would exhibit slow loss to the bloodstream of locallyproduced water of metabolism. In contrast, tissues suchas brain, with high blood flow, will show rapid removalof water of metabolism from the site. Although one cananticipate large recirculation corrections from tissueswith high blood flow, it is important to recognize thatneither metabolism nor water exchange is simply proportional to blood flow (11,13,14). Cerebral circulationis demand-limited, regulated by the metabolism itself(27), whereas other tissues may be limited by cellular

exchange rates rather than by circulation (10).The oxygen-utilization results for slowly exchanging

tissues can be appreciated from the right half of Fig. 7.As expected (5), a relatively small fraction (7.3%) of thewater of metabolism produced in these tissues is clearedfrom them. Recirculation provides a 13.9% increase.These results show that a model without clearance orrecirculation (one compartment) would require a correction of + 7 ±3%. In either case, for Edelman's (10)patient sample (Table 1) the variation across the group

FIG.8. Five-compartmentmodel, includingcerebralwhite andgraytissue, for analysis of distributiOn of water of metabolism. N 5(10)plus4 (13).

Volume 22, Number I I 963

BIGLER,KOSTICK,AND GILLESPIE

5. BIGLERRE, KOSTICKJA, DAVIS DC, Ctal: The use of C'50and @@O2steady-stateimaging to monitor radiation treatmenteffects. IEEE Trans Nuc! Sci NS-25: I 74- 179, 1978

6. KosTicK JA: Steady-state imaging with C'50 and ‘@O2to

monitor effects of radiation treatment. Master's Report,Brooklyn, New York, Polytechnic Institute of New York,1978

7. BIGLERRE, K0sTICK JA, LAUGHLINiS, Ctal:Compartmental analysis of the steady-state distribution of oxygen-ISlabeledoxygen and water. J Nuci Med 20:606, 1979 (abst)

8. BERMAN M, WEISS MF: SAAM Manual. Washington, DC,Department of Health Education and Welfare Publ. No.(NIH)78-180, 1978

9. BERMAN M: Compartmental modeling. In Advances in

MedicalPhysics.LaughliniS, WebsterEW,Eds.Boston,The SecondInternationalConferenceon Medical Physics,Inc.,1971,pp 279-296

10. EDELMAN IS: Exchange of water between blood and tissues:Characteristics of deuterium oxide equilibration in bodywater. Am J Physiol I 71:279-296, 1952

11. HUANG SC, PHELPSME, HOFFMANEJ, et al: A theoreticalstudy of quantitative flow measurementswith constant infusion of short-lived isotopes.Phys Med Biol 24:1151—1161,I979

12. JONESSC, REIVICHM, GREENBERGJH: Error propagationin the determinationofcerebral bloodflow with the inhalationofC'5O2. J Nuci Med 20:606, 1979(abst)

13. TER-POGOSSIAN MM, EICHLING JO, DAVIS DO, et al: Thedetermination of regional cerebral blood flow by meansofwater labeled with radioactive oxygen-I 5. Radiology 93:31-40, 1969

14. TER-POGOSSIAN MM, EICHLING JO, DAVIS DO, Ct al: Themeasure in vivo of regional cerebral oxygen utilization bymeansofoxyhemoglobin labeledwith radioactive oxygen-IS.J C!inInvest49:381—391,1970

15. RASMUSSEN KH, SKINHOJ E: In vivo measurements of therelative weightsof gray and white matter in the human brain.Neurology16:515-520,1966

16. ALTMAN PL, DITTMER DS, EDs: Biology Data Book, VolI, 2nd Ed. Bethesda, Federation of American Societies forExperimental Biology, 1972,p 392

17. SNYDERWS, CooK MJ, NASSETES:Report ofthe TaskGroup on Reference Man. International Commission onRadiological Protection. New York, PergamonPress,I 975,pp211—215

18. GUYTON AC: Textbook ofMedica! Physiology. Philadelphiaand London, W. B. SaundersCo., 1966,p 279

19. 1-IOLLIDAYMA, POTTERD, JARRAH A, et al: The relationof metabolic rate to body weight and organ size.Pediatr Res1:185—195,1967

20. ALTMAN PL, DITTMER DS, EDs: Biology Data Book, VolIII, 2nd Ed. Bethesda, Federation of American Societies forExperimental Biology, I 974, p 527

21. KETY 55, SCHMIDT CF: The nitrous oxide method for thequantitative determination of cerebral blood flow in man:theory, procedureand normal values.J Clin Invest 27:476-483,1948

22. MCILWAIN H, BACHELARDHS: Biochemistryand theCentralNervousSystem.Edinburghand London,ChurchillLivingston,1971,p 68

23. KOREY SR. ORCHEN M: Relative respiration of neuronaland glial cells. J Neurochem 3:277—285,1959

24. HUANG S-C, PHELPSME, HOFFMAN EJ,et al: Noninvasivedetermination of local cerebral metabolic rate of glucoseinman.Am J Physiol238:E69-E82,1980

25. PHELPSME, HUANG SC, HOFFMANEl, et al:Tomographicmeasurementof local cerebral glucosemetabolic rate in hu

amounts adequate for medical imaging (29). Itssteady-state imaging characteristics have been comparedwith those of oxygen-I 5 (T112 122 sec) (30). Theprimary concern with the use of oxygen-14 is the factthat in 100% ofthe decays it emits a 2.3-MeV gammaphoton in coincidence with the positron emission.

CONCLUSIONS

From the results of our model calculations, we condude that corrections and errors will render absolutemeasurement of oxygen metabolism in the brain by thesteady-state method more difficult than has been generally appreciated. This is especially true for cases ofbrain disease, since the kinetics are likely to be modifiedin an unpredictable manner. Furthermore, if the resolution of any quantitative imaging device were insufficient to resolvewhite and gray matter, additionaluncertainties would be introduced.

The absolute regional measurement of oxygen metabolism in tissues that exchange water slowly may bemeasured in vivo in humans, noninvasively and withreasonable accuracy, without corrections for eitherclearance or recirculation of water of metabolism withinthe body.

We are currently using the oxygen-I5 steady-statetechnique in conjunction with positron tomography (31)to extend to humans the earlier evaluation study of tumortherapy in animals (5). We expect to gain new insightinto the responses of neoplastic and normal tissues toradiation and chemotherapy, and this information shouldresult in improved therapeutic procedures.

ACKNOWLEDGMENTS

We express our gratitude to Dr. John S. Laughlin for his supportandencouragementof this work.

This researchwassupportedin part by the U.S. DepartmentofEnergyContract #EE-77-S-02-4258;by the NIH-National Heart,Lungand BloodInstitute #HL-22371; andby NIH-National CancerInstitute CoreGrant CA-08748.

REFERENCES

I. RUss GA, BIGLER RE, TILBURY RS, Ct al: Whole bodyscanning and organ imaging with oxygen-I 5 at the steadystate. World Federation of Nuclear Medicine and Biology,First World Congressof Nuclear Medicine, Tokyo, Japan,Sept.30—Oct.5,1974,pp 904-906

2. RUss GA, BIGLERRE, MCDONALD JM, et al:Oxygen-I5scanningand gamma camera imaging in the steady-state.JNuc/Med 15:529-530,1974(abst)

3. JONES T, CHESLER DA, TER-POGOSSIAN MM: The continuous inhalation of oxygen-I 5 for assessingregionaloxygenextraction in the brain of man. Br J Radio! 49:339-343,I976

4. SUBRAMANYAM R, AI.PERT NM, Hoop B, et al: A modelfor regional cerebral oxygen distribution during continuousinhalation of â€@̃O2,C' @O,and C'502. J Nuci Med 19:48-53,I 978

964 THE JOURNAL OF NUCLEAR MEDICINE

CLINICALSCIENCESINVESTIGATIVENUCLEAR MEDICINE

mans with (F-18)2-fluoro-2-deoxy-D-glucose: Validation ofmethod. Ann Neurol 6:371—388,1979

26. JONES T, MCKENZIE CG, Moss S, et al: The noninvasiveuse of oxygen-iS for studying regional brain function in patients with cerebral tumors. In Radioactive Isotope in Klinikund Forschung 12. Bank, Vienna, Austria, EgermannDruckereigeSellschaftm.b.H. andCo., 1976,pp 517—523

27. LASSEN NA: Cerebral blood flow and oxygen consumptioninman. PhysiolRev 39:183—238,1959

28. HOFFMAN EJ, HUANG S-C, PHELPSME: Quantitation inpositron emission computed tomography: 1. Effect of object

size.J Comput Assist Tomogr 3:299-308, 197929. DAHL JR, TILBURY RS, Russ GA, Ct al: The preparation

ofoxygen-14for biomedicaluse.RadiochemRadioanalLett26:107—116,1976

30. Russ GA, BIGLERRE, DAHLJR, et al: Regionalimagingwith oxygen-14. IEEE Trans NucI Sd NS-23:573-576,1976

31. BROWNELLGL, CORREIAJA, ZAMENHOFRG: Positroninstrumentation. In Recent Advances in Nuclear Medicine.Lawrence JH, Budinger TF, Eds. New York, Grune andStratton, 1978,Vol 5, Chap 1, pp 1—49

Volume 22, Number 11 965

ANNOUNCEMENTThe Education and Research Foundation of the Society of Nuclear Medicine welcomes applications for Student Fellowships and Pilot Research grants. These awards are made possible through donations from SNM members as wellas from variouscommercial firmswhoseproductsare usedinthe practiceof Nuclear Medicine.Applicationsreceivedprior to December 15ofany year will be evaluated bythe ERFBoard on a competitive basis. Awards will be announcedon or about February 15of the following year.

STUDENT FELLOWSHIP GRANTSThese awards are designed to stimulate interest among students in the United States and Canada in the field of Nuclear Medicine.The awardsare intendedto providean opportunitytospendelectivequartersand/orsummersinactivedepartmentsworking and associatingwith experts in the field. Maximum grant:$1.500. Lettersof applicationshouldbe submitted in duplicate and shouldcontain the following:applicant'sname, address,birth date, period for whichsupport is requested, name and institutionof sponsor, previouseducation, previous research,and brief summaryof the proposed project, including an appropriate bibliography.

PILOT RESEARCH GRANTSThe goal ofthis research support is to provide money to young scientists working in Nuclear Medicine who desire support for a research project. Priority will be given to those proposals that are of a pilot nature in either clinical or basicresearch.The grants are not intendedto supportsalaries,purchasemajorequipment,or for travel,but are designedto provideessentialmaterialsso that innovativeideascan be quickly tested.Maximum grant:$3,000.

SPECIAL ANNOUNCEMENT: SECOND TETALMAN MEMORIAL AWARDA fund hasbeen establishedin the ERF by friendsof Marc Tetelman, M.D., who wasa tragichomicidevictimwhileattending the SNM meeting in Atlanta in June, 1979.Thisfund will permit an award of $3,000to be made in June, 1982toa young investigator(35 yearsof ageoryounger) who ispursuinga career inNuclear Medicine.Thisaward isto be repeated annually. It is possible that additional contributions to ourfund will permit the stipend to be increased in futureyears. Applicantsshould submit prior to March 1, 1982a curriculum vitae together with data supportingcurrent researchefforts.

All letters and applications should be addressed to:Merle K. Loken,M.D.

President,E & A Foundationdo Societyof Nuclear Medicine

475 ParkAvenueSouthNewYork, NY 10016

ANNOUNCEMENTOFBERSON—YALOWAWARDThe Society of Nuclear Medicine invitesmanuscriptsfor considerationfor the Fifth Annual Berson—YalowAward.Workwill bejudged on originality and contributionto the fieldsof basicorclinical radioassay.The manuscriptwillbepresentedat the 29th Annual Meetingof the Societyof Nuclear Medicine in Miami Beach,FL,June 15-18, 1982,andasuitablyengravedplaquewill be awardedto the authorsby the Educationand ResearchFoundationof the SocietyofNuclear Medicine.

The manuscriptshould be approximately ten pages in length (typed,double-spaced). A letter requestingconsideration for the award, includingthe author'sfull mailingaddressand telephone number,shouldaccompanythe manuscript. Original manuscript and eight copies must be received by January 18, l982atthe Societyof Nuclear Medicineoffice,475ParkAve.So.,NewYork,NY10016,Attn:Mr.DennisL Park.

DEADLINEFORRECEIPTOFMANUSCRIPTS:January18,1982