Radiation Dose from Radiopharmaceuticals Contaminated with Molybdenum

Document Sample
Radiation Dose from Radiopharmaceuticals Contaminated with Molybdenum Powered By Docstoc
					Radiation Dose from Radiopharmaceuticals
Contaminated with Molybdenum-99
Douglas R. Shearer, John C. Pezzullo,         Marleen   M. Moore, Patricia Coleman,
and Stephen I. Frater
Rhode Island Hospital,   Providence, Rhode Island

Sixteen patients undergoing routine nuclear imaging procedures were injected with @“Tc
labeled radiopharmaceuticals   contalning      which exceeded the recommended limit of 1 @@Ci
      per mCi
of @“Mo of               The                            distributionin 14 of these
                         @Tc. kinetics of the resulting @Mo
patients were studied over a period of several weeks. The mean biologic half-life [T½b]
from about 19.3 days to 11.2 days depending on the model used. Similarly,the mean
radiation dose to the liver ranged from         rad/@Ci of
                                            @0.02                 to           of
                                                                @Mo 0.05 rad/@@Ci      @Mo.

J Nucl Med 29:695—700,1988

      he range of values in the literature for biologic half      of Standards.The two assaysagreed with each other on the
life and organ uptake leads to dose estimates for admin           quantity of the @Mo administered to all patients to within
istered molybdenum-99      (@Mo) which differ widely (1-          18% in the worst case and to within 10% in most cases.
16). There is also some disagreement as to the critical              The amount of@Mo delivered to the patients for each mCi
organs. The following paper presents clinical data on                                                             f
                                                                  of@mTc administered was 170.4 ±14.2 @tCi/mCior the first
                                                                  elution and 134.0 ±12.5 MCi/mCi for the second elution,
the kinetics and dosimetry of 99Mo injected intrave
                                                                  where the errors are the standard deviations of the results
nously in association with technetium-99m (99mTc)as
                                                                  from the independent assays.
Tc04 and in association with various radiopharmaceu
ticals used in clinical nuclear medicine procedures. This
is compared with dosimetric information available in             Calibration of Gamma Camera
the literature.                                                     Preliminary images established that the primary site of@Mo
                                                                 localization was the liver. Four liver phantoms were made up
                                                                 to accommodate    thevariabilityin patientsize.The phantoms
METhOD                                                           consisted of plastic jugs of 150, 5 10, 980, and 1,800 ml,
                                                                 respectively, which were filled with water containing a known
  Sixteen patients were injected with @mTc@labeled     radio     amount of @Mo      eluted from a generator column using a
pharmaceuticals containing excessive quantities of @Moe   b      sodium hydroxide solution and then left for several days to
cause ofa problem discovered after the injection. Fourteen of    reach equilibrium.
these patients were available for follow-upstudies. Table 1         Severalthicknessesoflucite (density= 1.05)simulatedthe
summarizes the patient data, the radiopharmaceuticals and        attenuation of overlying tissue. A gamma camera fitted with
the quantity of 99mTcadministered. These patient doses were      a high sensitivity collimator and pulse height analyzer (PHA)
prepared from two separate elutions on consecutive days, of      set on a 20% window for the 140 keV emission from @mTc
a generatorconsisting of fission product @Moadsorbedon an        was used to acquire the images. (It should be noted that with
alumina column. The occurrence of this problem highlights        a collimated gamma camera there may be a problem in
the need for careful checking of @Moontamination from            measuring the 140 keY gamma rays from @mTc the presin
generatorsbefore administration of radiopharmaceuticalsto        ence of high-energy @Mo      gamma rays. If an auto-ranging
patients.                                                        multi-channel analyzer (MCA) is used to determine the loca
   The activity of @Mo dministered to each patient was
                        a                                        tion of the 99mTcpeak, it may not be visible. This type of
determined by two independent assays of the elution vials 7      MCA scales its display to the largestsignal in memory. High
to 11days after the administration.Theseassayswerecarried        energy @Mo     gamma rays are transmitted with little attenua
out on a dose calibrator (Capintec CRC-l7, Montvale, NJ),        tion by the lead septa of the collimator, whereas the @mTc
which had a @Mo
              calibration traceable to the National Bureau       gamma rays are greatlyattenuated. Accordingly, with a colli
                                                                 mator, the spectrum is dominated by the @Moomponentc
                                                                 and the 99mTcpeak is suppressed. Without a collimator, the
  ReceivedJan. 20, 1986;revision accepted Nov. 24, 1987.                                                   p
                                                                 99mTcpeak is visible but not the @Moeak, due, in part, to
  For reprints contact: Douglas R. Shearer, PhD, Rhode Island    the greater interaction probability ofthe 140-keYgamma rays.
Hospital (APC 1072),593 Eddy St., Providence, RI 02902.          The PHA window should therefore be adjusted with a pure

Volume29 •
           Number5 •
                     May1988                                                                                           695
                                        TABLE I                                          contaminant (or contaminants) with an apparent half-life of
                                       Patient Data                                       10.2 ±11.7 days. This contamination was not otherwise
                                     Dose                                                identified. The average difference between calculated and
                                     ofTc-99m                                                                      1.8%
                                                                                         measured readings was @‘ with a maximum deviation of
                                           ActivityEstimated                             the measured from the theoretical data of6.6%. Thus, count
     Admin.Patient                                                                       ing errors due to instrument drift are assumed to be at most
              Age                           (mCi)(mCi)                                   7% and usually <2%.
                      Technetium-99mmedronate-bone imaging                               Method   of Determining   Patient   Activity
           A               41               20.0                3.52 ±0.63                 Data were acquired on 14 of the 16 patients with each
           B               24               19.9                3.52±0.63               patient providing between two and six measurements. These
           C               68               19.0                2.22±0.09                                                                  days
                                                                                         data, however, could not be acquired until 10—14 after
           D               38               20.0                2.48±0.10               administration due to the delay in recognizing the nature of
           E               46               19.9                2.63±0.11               the problem and locating the patients for further testing.
           F               84               19.9                2.62±0.11               During this period, considerable redistribution of @Mo aym
           G               68               19.8                2.74±0.11
                                                                                         have taken place as discussed below. Each patient was given
           H               10                6.0                0.94±0.06
                                                                                         a preliminary whole body survey to determine the localization
                  Technetium-99msulfur colloid-livermaging                                             AP
                                                                                         of the @Mo. and PA images demonstrated that the liver
           I               75                4.0                0.47 ±0.05              was the only organ containingobservablequantitiesof @Mo.
           J               75                4.0                0.53 ±0.04              Initially, the time taken to collect a 300,000 count image of
           K               75                4.0                      0.04
                                                                0.59 ±                  the liver was noted. Later in the study, as the activity declined,
           L               48                4.0                      0.04
                                                                0.59 ±
                                                                                         the acquisition times became unacceptably long and 100,000-
                       Technetium-99m DTPA-kidneyimaging                                 and 50,000- count images were collected. All data were cor
           M              1          3.0           0.40±0.01                            rected for background, and the net count-rate (counts/sec.)
                                            imaging                                      the determined. By use ofthe calibration data, this count rate
           N         59          4.0             0.74 ±0.09                                                    of       in
                                                                                         was corrected to @tCi @Mo the liver for each patient.
                Technetium-99mdisofenin-gallbladderimaging                               Methods of Data Analysis and Dose Estimation
           0               38                5.0                0.74 ±0.03                 The biologic data for each patient were initially fitted by a
                            Technetium-99m-pertechnetate                                 single exponential least squares fit which assumed that the
           P               63         20.0            3.31 ±
                                                            0.21                            was
                                                                                          @Mo originallytaken up and excretedby the liver.
                                                                                            For six ofthese patients (denoted by an asterisk in Table 2)
                                                                                         the extrapolated activity at time of administration was deter
   . The       dose     calibrator    was   also   responding   to   the    @Mo   corn

ponent and thus would give an overestimate of the                          @Tc
                                                                             actually    mined to be greater than that assayed by factors which in
administered.                                                                            general cannot be accounted for by experimental or calcula
                                                                                         tional errors. Thus, the single exponential single compartment
                                                                                         model is not appropriate.
                                                                                           A two-compartment model was derived in which the           @Mo
99mTcsource before the study to ensure that it is correctly
                                                                                         was assumed to be in two fractions, one of which was bound
                                                                                         to the pharmaceutical and delivered to the radiopharmaceu
   All clinical measurements were performed using the 20%
                                                                                         tical target organ [compartment A]; the remaining free 99Mo
99mTc  peak to acquire acceptable images of kinetics data. This
                                                                                         being taken up directly by the liver [compartment B] (1 7).
technique permitted the @Moctivity to be determined, as
                                                                                         The experimental basis for the model is the finding that @Mo
   and   w
 @Mo 99mTc ere in equilibrium at the time of measure
                                                                                         may tag directly to the radiopharmaceutical. (@@-70% MDP
ment because the originally injected @mTc decayed leaving
                                                                                         and @20% sulfur colloid as determined by thin layer
only the 99mTcproduced by transmutation of the @Mo.     The
                                                                                         chromatography). A nonlinear least squares fit to the data for
results in cps/@Ci of @Mo the four phantoms and the
                                                                                         each patient was made using this model with the known
thicknesses of overlying material were used to translate the
                                                                                         administered dose and values for the free molybdenum frac
                                       of       l
counting rate for each patient into @iCi @Moocalized in
                                                                                         tion [1]0. 1, 0.2, 0.5, and 0.8. Dose estimates have been made
the liver with some correction for patient size and overlying                            for bone in the case of MDP, assuming that the radiophar
tissue.                                                                                  maceutical is equally distributed between cortical and trabec
Standardization                                                                          ular bone (18). For sulfur colloid, it has been assumed that
   In order to ensure that the data werebeingcollectedunder                              the radiopharmaceutical target organ is the liver since the
the same conditions from day to day, an old @Mo    generator                             radiopharmaceutical is released from the Kupifer cells only to
core, giving approximately the same counting rate as the                                 be taken up by the polygonalcells. Thus, only liver dose is
patients, was used as the standard. The time taken to collect                            calculated for this agent.
300,000 counts was noted using the same 20% @mTc     window                                 Doses have been estimated using the MIRD tables (19)
as for the patient counts. Backgroundwas subtracted.As the                                           S
                                                                                         with @Mo -factors which have been adjusted for pediatric
study progressed, it became apparent that the core was not                               cases (Cristy M., personal communication) where required.
pure @Mo. core decay data may be adequately described                                    The dose from 99mTcin equilibrium with the @Mo        (again
as the sum of two exponentials,one of which is assumed to                                using age adjusted S-factors) has been added to these dose
be due to the decay of @Mo = 2.75 days) and the other a
                         (T½                                                                                                               S
                                                                                         estimates as it is not included in the published @Mo-factors.

696                         Pezzullo,
                      Shearer,         e
                                   Moore tal                                                                          The Journal of Nuclear Medicine
                                                                                                 TABLE 2
                                                       Results Obtained from One and Two Compartment Models

     B             78            46         0                        N/A                         68          36           17               43          35         36       36              34             31
     C             57            52         0                        N/A                         50          45           12               32          43         18       27              42             20
     D*           262            34         0             70         42              4           54          40           11               40          38         16       36              37             18
     E             94            55         0             92         52              5           70          SI           13               50          50         19       42              49             21
     F            125            45         0             81         46              5           60          44           13               42          42         13       37              41             25
     G             81            61         0                        N/A                         72          55           15               50          53         15       42              53             25
     H*           250            49         0
     I             13            64         0                        N/A                         22          57                0           22          55            0     22              55               0
     J*            75            42         0             23         55               0          25          56                0           24          49            0     25              52               0
     L*            43            48         0             26         55               0          28          57                0           27          48            0     29              56               0
     M*           357            41         0             85         56               0          58          59                0           32          57            0     25              54               0
     0             14            55         0             23         46               0          17          44                0           12    42      0                 10              41               0
     P             79            62         0                        N/A                                     N/A                           —N/A--——                                  N/A

  . The   dose      calibrator        was       also    responding        to   the       °Mo   component         and   thus       would        give   an   overestimate        of   the        @“1c actually


The dose from the original @mTc administered in conjunction                                                 DISCUSSION
with the radiopharmaceutical has not been included.
                                                                                                               Sorenson (1—3) nd Hennig (4) have examined the
RESULTS                                                                                                                                     a
                                                                                                            internal distribution of @Mond agree that @@70-80%
                                                                                                            of the administered molybdenum is taken up by the
  The relationship of the data to the single and double                                                     liver. They are also in reasonable agreement with Rosoff
exponential models is shown in Figure 1 for typical                                                         and Spencer (5) concerning biologic retention giving
cases. The effect ofapplying a single exponential model                                                                                              days. An excel
                                                                                                            biologic half-lives ofthe order of 10—40
to determine the initial activity results in an extrapo                                                     lent review of human data is given by Cifka and Vesely
lated activity at the time of administration which is                                                       (6).
anomalously             greater than the total activity adminis                                                In spite ofthis, reference publications(7—16)tabulate
tered.                                                                                                      data which are in marked disagreement                                     concerning                  the
   Table 2 summarizes the dosimetry results for the                                                         biological half-life, the target organ and the fraction of
one- and two-compartment models. Both the single                                                             @Motaken up in this organ.
exponential and the double exponential model (with                                                           The Critical organs given in ICRP #2 (7) (kidney and
different assumed initial activities in the liver) give                                                     01 tract) are the results of early studies (e.g., (20,21))
adequate         fits to the data. This is true for all patients                                            which were concerned with the effects of molybdenum
who were administered MDP with the exception of                                                             in the diet of grazing animals. This information could
Patient H. In all cases, however, the single exponential                                                    also be applied to ingestion ofradionucides                                     by workers.
fit gives a higher estimate ofthe administered                                       dose than              Thus, great attention was given to oral ingestion and
the double exponential model—particularly for patients                                                    absorption in the 01 tract with the major route of
D, F, and H.                                                                                                excretion being through the kidneys. The Maximum
  Similar fits were obtained for the patients adminis                                                       Permissible Body Burden (MPBB) was derived on the
tered sulfur colloid. An example (Patient 1), is given in                                                   basis that internal organs such as the liver, kidneys, etc.
Figure lB. For Patients J & L, however, the single                                                          should not receive a dose rate in excess of 300 mrem/
exponential model severely overestimates                                       the adminis                  week as a result of this burden being continuously
tered activity.                                                                                             present in the whole body. Thus, the MPBB was criti
  The results for the patients administered DTPA and                                                        caily dependent on the values which were assumed for
DISIDA           exhibit         the same              behavior,      with the initial                      the fraction absorbed from the GI tract, blood, etc.
activity for Patient M being overestimated by a factor                                                      These values for man were in general not known and
of -‘@5. results for Patient P who was administered                                                       were therefore, inferred on the basis of work with ani
pertechnetate are shown in Figure 2 with a single cx                                                        mals. The MPBB can also be derived on the basis of
ponential fit.                                                                                              the fraction ofradionucide present in the organ relative

Volume 29 •
            Number 5 •
                       May1988                                                                                                                                                                              697
                                                                                                             PATIENT F EMDP]

                                             C.,    1.



                                                                                                 0.1 0.0168
                                                                                                 0.2 0.0165
                                                                                                             Li            L2
                                                                                                                         0.0 168
                                             C.)                          C      DE    058950    0.5 0.0157              0.0 157
                                                                          d      DE    058950    0.8 0.0150              0.01@0
                                                                          e      SE    080351    — 0.0154


                                             A                                                                            HOURS
                                                                                                             PATIENT I [Sc]

    FIGURE 1                                       10.000           d
    Single exponential (SE) and double
@   exponential (DE)least squares fits to
@   the data obtained. The single expo-
@   nential model gives a least squares             1.000           a
@   fit to the initial activity A, and the
@   effective decay constant X@. The
    double exponential model gives least     @2       100
@                            and us-
    squares fit values to X,@ AB,,,,
                                                                        Curve     Eq      Ao             f    Li           L2
@   ing the known administered activity
                                                                          a       DE 0010764 0.10.01580.0127
    A0and assumed values of f, the frac                    10             b       DE 0010764         0.20.0156           0.0126
@   lion of contaminant taken up directly                                              001@;:@                           0.0123
                  and      a
    by organ B. AA•,, XB,, re the ef
    fective decay constants for compart                     1                    1             _1@__@               I                I         I          I         @-1
    ments A and B. A: Patient F; B: Pa                          0               100            200                 300             400        500       600       700
    tientl.                                  B                                                                           HOURS

    to that in the total body. The values for the dose (rem/                           in muscle, really refers to xenon-i33. It also appears to
    XCi) given for the ICRP#2     (7) data have been derived                           be in error by a factor of 2.) The low estimate in ICRP
    using a biologic halflife of45 days and recalculated for                           #30 (14) is a result ofthe assumption that only 30% of
    a single administration using a single compartment                                                           is
                                                                                       the administered @Mo taken up by the liver. The
    model with exponential elimination. From these data                                information in NCRP #65 (15) is taken directiy from
    the dose to the liver (0.006 rem/@iCiof administered                               ICRP #2 (7) reworked                         to give a kidney dose of 0.012
    99Mo)was calculated for the case ofintravenous admin                               rem/@iCi.Cifka and Vesely (6) synthesized the relevant
    istration.   The low value is due to the small fraction                            human data to derive a dose of 0.03 rem/MCi.
    (0. 1) of 99Moassumed to be absorbed from the blood.                                 For the single exponential model, the present work
       The major discrepancy in biologic half-life can be                                                                                days
                                                                                       has assumed that the biologic data gathered 10—14
    ascribed to a typographical     error by Kaui et al. (10)                          after administration                       reflects liver excretion.   Thus, all
    where the 20 day biologic half-life referenced from                                patients have been included for this estimation of bio
    Sorenson (1-3) has been stated as 20 hr. This leads to                             logic and effective half-life (19.3 ±7.4 and 2.1 ±0.1
    an effective half-life of 15.4 hr giving a low estimate of                         days, respectively). The estimate ofliver dose (0.026 ±
    liver dose (0.0089 rem/@iCi) in NCRP 70 (6). The                                                does not include the data from the six
                                                                                       0.003 rem/@&Ci)
    highest tabulated estimate of liver dose (0.05 rem/@Ci)                            patients whose 99Mo dose was clearly overestimated by
    is given by ICRP 17 (9) because of the adoption of                                 this model. Similarly, only 11 patients have been in
    Sorenson's upper limit which assumes no biologic dim                               cluded for the estimate of effective and biologic half
    ination at all. (It should be noted that the entry for                             life using the two-compartment    model (1 1.2 ±2.0 and
      @Mo this reference under the heading of blood flow
          in                                                                                  0.
                                                                                       2.2 ± 1 days, respectively). (One was excluded because

    698                Pezzullo,
                 Shearer,         e
                              Moore tal                                                                                              The Journal of Nuclear Medicine
                                       PATIENT P [TcOfl



LU       100

z                Curve   Eq     Ao       f    xaeff      XA@ff
I-                 a     SE   036838    —    0.011        —
                                                                                               FIGURE 2
0         10                                                                                   A   single     exponential   (SE)   least
                                                                                               squares fit to the data obtained from
                                                                                               patient P. The single exponential
                                                                                               model gives a least squares fit to the
                                                                                               initial activity A0and the effective de
                                                HOURS                                          cay constant XB,,.

of a poor fit to the data; the other was administered                     mined and the kinetics of well-defined 99Mo-tagged
Tc04      and therefore the application         of the two-com            radiopharmaceuticals should be studied in an animal
partment model is not immediately obvious). The dose                      model.
in rem/MCiof @Mo          hasbeenderivedfor                                 Clinically, in cases where molybdenum is inadvert
each group of patients. This dose was determined to be                    ently administered to patients in excess of allowable
0.024 ±0.003 rem/MCi for all except the sulfur colloid                   limits, not only should the amount of administered
patients, which have a relatively high estimate of 0.046                  99Mo be determined,      but      the radiopharmaceutical
±0.003 as a result of the assumption that all the                        should also be evaluated chromatographically to deter
administered radionuclide was taken up directly by the                    mine the fraction of bound 99Mo. Biologic measure
liver.                                                                    ments should be initiated as soon as possible, preferably
  Although the two compartmental          model gives ade                 within 24 hr of administration to determine the way in
quate fits for all but one case, (Patient H), it is unlikely                          i
                                                                          which @Mos taken up and excreted by the target organ
that       is
         @Mo excreted by the simple exponential mech                      and total body with both          and
                                                                                                         @mTc            s
                                                                                                                       @Moettings on
anism described. It is more likely that there are several                 the PHA. If this is not done the dose initially localized
components to the release from both the radiophar                                                      by
                                                                          in theliver maybeoverestimated usingan extrapo
maceutical target organ and the liver. A simple model                     lated single exponential. Similarly, information taken
of this type (1 7) assumes a slow degradation of the                      at random from the literature may cause dose estimates
radiopharmaceutical      in the target organ after which free             to be considerably in error.
  @Mos cleared from the organ to be taken up by the                          For prospective dose evaluation a value of 0.02—0.04
liver, i.e., a slow component followed by a fast compo                             o
                                                                          rem/@zCi fadmiistered       @Mo(depending on the radio
nent. There is, however, little information in the liter                  pharmaceutical   involved), with a biologic half-life of
attire on the long-term fate of bone-seeking agents or                    10-40 days, would seem appropriate. For accurate in
sulfur colloid as these agents have mostly been studied                   dividual results, the procedures described above should
using 99mTcwhich decays too quickly for useful data to                    be followed in order to take account ofthe considerable
be obtained.    Consequently,    doses have not been calcu                variations in biology and radiopharmaceutical       tagging.
lated using this model.
   Although not directly comparable with the data pre
viously given in the literature, the results of this work                 REFERENCES
point to the liver as the principal target organ for 99Mo
and provide dose estimates and biologic and effective                      1. Sorenson LB. Archanbault       M. Visualization of the
                                                                              liver by scanning with Mo-99 (Molybdate) as tracer. J
half-lives in good agreement with other human data.                           Lab Clin Med 1963;62, 330.
The range of dose estimates available in the literature 2. Sorenson LB. Molybdenum-99, a new isotope for
can also be understood.                                    scintillation scanning of the liver. Strahientherapie,
  The problem of dose estimation in the case of @Mo        Vol. 60, Supplement, 1963, pp. 126—135.
contamination     of a radiopharmaceutical             still requires a    3. Sorenson LB. Molybdenum-99, a new isotope for
                                                                              scintillation scanning of the liver. International Sym
great deal of work. For example, the degree of 99Mo                           posium ofRadioactive Isotopes in Clinical Medicine
tagging to clinically used                  radiopharma
                                 @mTc@labeled                                 andResearch, p. 498.
ceuticals prepared as used clinically should be deter                      4. Hennig K, Woller P. Strietzel M, et al. Mo-99-mar

Volume 29 •
            Number 5 •
                       May1988                                                                                                     699
   kiertes natriummolybdat—herstellung    und eigen        12. Colombetti LG, Husak V. Study of the purity of Tc
   schaften sowie verhalten im menschlichen korper. J              99m sublimed from fission Mo-99 and the radiation
   NukiearMedizin ol.5,#1,1965,
                      V               88—96.                     dose from the impurities. Intern J App! Radiat Isot
5. RosoffB, Spencer H. Fate ofmolybdenum-99 in man.                1974;  25:35—41.
   Nature1964;   202,410—411                               13.   Colombetti LG. Rapid estimation of the radiation
6. Cifka J, Vesely P. Review of biological behavior of             dose to patients from radionucidic impurities in
   pertechnetate Tc-99m in some organs and calculation             eluates from Generators, International Journal of Ap
   of radiation doses to those organs and calculation of           plied Radiation and Isotopes. 1975; Vol. 26: 667—670.
   radiation doses to those organs from Tc-99m and its       14.   Limits for intakes of radionuclides for workers. ICRP
   possible radionuclide impurities. Report UJV, 2696-             Publication #30, 1978.
   D., Rez (1971). (Obtainable from Radiopharmaceuti         15.   Management of persons accidentally contaminated
   cal Internal Dose Center, Oak Ridge Associated Urn              with radionuclides, NCRP #65, 1980.
   versities, Oak Ridge, TN 37831.                           16.   Nuclear medicine—factors influencing the choice in
7. Recommendations      of the International   Commission          use of radionucides in diagnosis and therapy, NCRP
   ofRadiological Protection. Report ofCommittee 2 on              #70, 1982.
   Permissibledose for internal radiation. Publication #2,   17. Shearer DR. Pezzullo JC. A kinetic model for the
   ICRP.                                                         dosimetry of radiopharmaceuticals contaminated by
8. Maximum Permissible Body Burdens and Maximum                  Mo-99. In: Proceedings offourth international radio
   Permissible Concentrations of Radionuclides in Air            pharmaceutical dosimetry symposium. Oak Ridge,
   and in Water for Occupational Exposure, 1959.                 Tennessee, 1985; 319—330.
9. Protection of the Patient in Radionuclide Investiga       18. Subramanian     JG, McAfee G, Blair RJ, et al. I Nucl
   tions, ICRP Publication #1 7, 1969.                           Med 1975; 16:8: 744—755.
10. Kaul A, OeffK,   Roedler HD, et al. Radiopharmaceu       19. New York, Society ofNuclear Medicine, MIRD Com
   ticals—biokinetic data and results of radiation dose                P
                                                                 mitteeamphlet     #11,1975.
    calculation. Berlin, 1973.                               20. Comar CL. Radioisotopes in nutritional trace element
11. Colombetti LG, Barnes WE. Absorbed radiation dose            studies. I. Nucleonics 1949; 3: 32—45.
    from radionuclide impurities in Tc-99m obtained by       21. Comar CL, Singer L, Davis GK. Molybdenum metab
   different procedures. Intern J App! Radiat Isot 1974;         olism and interrelationships with copper and phospho
   25: 455—461.                                                     J
                                                                   n's. BiolChem 1974;180:913—922.

700              Pezzullo,
           Shearer,         e
                        Moore tal                                                       The Journal of Nuclear Medicine