Paper
COMPARISON OF MEASURED AND CALCULATED DOSE RATES NEAR
NUCLEAR MEDICINE PATIENTS
Y. Yi,* M.G. Stabin,† M.H. McKaskle,† M.D. Shone,† and A.B. Johnson†
AbstractVWidely used release criteria for patients receiving
radiopharmaceuticals (NUREG-1556, Vol. 9, Rev.1, Appendix
U) are known to be overly conservative. The authors measured
external exposure rates near patients treated with 131I, 99mTc,
and 18F and compared the measurements to calculated values
using point and line source models. The external exposure dose
rates for 231, 11, and 52 patients scanned or treated with 131I,
99m
Tc, and 18F, respectively, were measured at 0.3 m and 1.0 m
shortly after radiopharmaceutical administration. Calculated
values were always higher than measured values and suggested
the application of ‘‘self-shielding factors,’’ as suggested by Siegel
et al. in 2002. The self-shielding factors of point and line source
models for 131I at 1 m were 0.60 T 0.16 and 0.73 T 0.20, re-
spectively. For 99mTc patients, the self-shielding factors for point
and line source models were 0.44 T 0.19 and 0.55 T 0.23, and the
values were 0.50 T 0.09 and 0.60 T 0.12, respectively, for 18F (all
FDG) patients. Treating patients as unshielded point sources of
radiation is clearly inappropriate. In reality, they are volume
sources, but treatment of their exposures using a line source model
with appropriate self-shielding factors produces a more realistic,
but still conservative, approach for managing patient release.
Health Phys. 105(2):187Y191; 2013
Key words: exposure, population; exposure, radiation; radiation
dose; radiation, medical
INTRODUCTION
PATIENTS WHO have received radiopharmaceuticals for di-
agnostic or therapeutic purposes are potential sources of
radiation exposure for persons who may be near the pa-
tients; e.g., other family members or persons in cars or
public transportation. It is of general interest to know
*Institute of Radiation Medicine, Fudan University, No. 2094,
Xietu Road, Shanghai, 200032, P.R. China; †Department of Radiology
and Radiological Sciences, Vanderbilt University, 1161 21st Avenue
South, Nashville, TN 37232-2675.
The authors declare no conflicts of interest.
For correspondence contact: Yanling Yi, Institute of Radiation
Medicine, Fudan University, No.2094, Xietu Road, Shanghai 200032,
P.R. China, or email at ylyi@fudan.edu.cn.
(Manuscript accepted 6 March 2013)
0017-9078/13/0
Copyright * 2013 Health Physics Society
DOI: 10.1097/HP.0b013e318290cc0e
www.health-physics.com
Copyright © 2013 Health Physics Society. Unauthorized reproduction of this article is prohibited.
when to discharge the patients to assess potential ex-
posures to others. As the exposure rates for therapeutic
nuclear medicine procedures are higher than those for
diagnostic procedures, the former is of higher interest.
Current federal regulations require an estimation of dose
to be received by family members or other members of
the public before the patients can be released from med-
ical care facilities. Specifically, U.S. Nuclear Regulatory
Commission (NRC) regulations for the release of patients
administered radioactive material, as given in 10 CFR
35.75, authorize patient release according to a dose-based
limit; i.e., the total effective dose equivalent (TEDE) to
other individuals exposed to the patient must be shown by
reasonable calculation to be less than 5 mSv (500 mrem)
(USNRC 1997). The NRC has also issued a non-binding
guidance document NUREG-1556 (UNRC 2005) showing
one method to assess these doses, but the methods put
forth in this guidance document are overly conservative
due to modeling patients as unshielded point sources of
radiation and not accounting for biological removal of ac-
tivity from the patients’ bodies, among other factors. This
may prescribe overly restrictive methods for patient re-
lease, including delays in patient release, resulting in un-
necessary inconvenience to patients and their families and
higher costs to medical institutions. Nonetheless, many li-
censees employ this method, although alternative calcula-
tional methods have been presented, and measurements
have shown that the methods in NUREG-1556 significantly
overestimate the actual exposures (Siegel et al. 2007).
Cormack and Shearer (1998) reported a methodol-
ogy to calculate dose to others from patients to whom
radioactive nuclides have been administered based on the
radiation exposure, half-life of each clearance, and inter-
val contact time. For patients after Tositumomab and
131
IYTositumomab therapy, Siegel et al. (2002a) reported a
methodology for patient release and showed that mean
measured dose rates were about 60% of those from theo-
retical calculations based on a point source in air model.
Siegel et al. (2007) reviewed calculational methods for
calculating dose to family members and suggested alter-
native activity levels for patient release that would provide
187
188 Health Physics August 2013, Volume 105, Number 2

compliance with 10 CFR 35.75. Several authors also have scintillation survey meter was used. All dose rate
proposed the use of more realistic theoretical models for measurements were made at the anterior midline of the
the calculation of dose from radioactive patients, including patients at about 15 min after radiopharmaceutical ad-
line and volume sources instead of unshielded point sources ministration. Patient height and weight were recorded.
(Sparks et al. 1998; Siegel et al. 2002b; de Carvalho et al.
2011). All of these works demonstrate that, for several Model calculations
types of nuclear medicine therapy, an unshielded point Several models from other reporters were used here to
source approach to dosimetry is considerably conservative evaluate the collective dose to other members from patients.
and should be replaced with more realistic, but still con-
servative, approaches. Willegaignon et al. (2011) mea- Unshielded point source (NUREG-1556 method).
sured doses to individuals associated with 131I nuclear The exposure rate near an individual receiving Na131I
medicine outpatient procedures and found a maximum for thyroid cancer therapy soon after administration was
absorbed dose of 1.6 mSv, far below the 5 mSv limit, as estimated as:
well as very low levels of radioactive contamination in AG
Ẋ ¼ 2 ; ð1Þ
the patients’ homes., They concluded that ‘‘the treatment d
of thyroid cancer by applying radioiodine activities up to where A = activity administered to the subject (MBq);
7.4 GBq, on an outpatient basis, is a safe procedure.’’ Some
G = exposure rate constant for an unshielded point source
authors have made limited measurements on exposure
(for 131I, G = 0.595 mSv cm2 MBqj1 hj1 (assuming that
dose to family members from patients with radionuclide 1 mSv equivalent dose may result from an exposure of
therapy (Zanzonico et al. 1997; Grigsby et al. 2000; Rutar
0.01 mR); and d = distance from subject (cm).
et al 2001; Siegel et al. 2002a) and have confirmed that the
NUREG-1556 method considerably overestimates doses Line source model. The exposure can be shown as:
actually received by others. Very little data has been 


published regarding the exposure rates near patients re- A
G j1 l 1 j1 l 2
Ẋ ¼ tan þ tan ; ð2Þ
ceiving diagnostic levels of radiopharmaceuticals, as con- a
d d d
cerns are generally lower.
where A is the source activity (MBq), d is the distance from
The authors measured dose rates near a number of
the source (m), and a = l1 + l2 is the source length (in m);
nuclear medicine patients who had received 131I NaI for
i.e., the patient height.
treatment of thyroid cancer and compared the measured
As measurements were made at the patient midline,
dose rates near the subjects to point, line, and shielded line
l1 , l2, yielding:
source calculations. Exposure rates near subjects receiving
various pharmaceuticals for diagnostic purposes were also A
Gh  a i
Ẋ ¼ 2
tanj1 : ð3Þ
measured. At times, health care and other professionals a
d 2
d
have raised concerns about being exposed to radioactive Shielded line source model. The same line source
patients, and a better characterization of the actual expo- equation was used and was multiplied by a factor of 0.6,
sure and dose rates near these patients would help answer
these concerns. Other sample calculations of theoretical
doses to medical professionals and others involved with
radioactive patients are shown that can help to allay such
concerns in most cases.

MATERIALS AND METHODS

Measurements
The authors made exposure rate measurements at
0.3 m and 1 m from over 230 thyroid carcinoma patients
treated in Vanderbilt University Hospital from 2009Y2012.
They also made measurements and on over 50 patients
receiving diagnostic studies (mostly 18FDG and various
99m
Tc radiopharmaceuticals), as regulatory bodies in two
NRC Agreement States proposed estimating doses from
patients receiving diagnostic, as well as therapeutic, radio- Fig. 1. Activity levels administered to thyroid cancer patients re-
pharmaceuticals. A regularly calibrated sodium iodide ceiving 131I.
www.health-physics.com

Copyright © 2013 Health Physics Society. Unauthorized reproduction of this article is prohibited.
 Dose rates near nuclear medicine patients c Y. YI ET AL. 189

Table 1. Ratios of calculated/measured exposure rates near thyroid Table 2. Ratios of calculated/measured exposure rates near patients
cancer patients receiving 131I. receiving 18FDG.
Unshielded point Unshielded line Unshielded point Unshielded line
source source source source
0.3 m 1m 0.3 m 1m 0.3 m 1m 0.3 m 1m
Average 2.98 1.79 1.31 1.48 Average 5.11 2.10 2.26 1.74
Standard deviation 1.33 0.54 0.58 0.45 Standard deviation 1.29 1.29 0.60 0.33
Minimum 1.33 0.82 0.59 0.69 Minimum 2.79 1.18 1.18 0.97
Maximum 8.89 4.13 3.91 3.42 Maximum 8.13 3.35 3.88 2.76

as established by Siegel et al. for patients receiving 131I
Bexxar (Siegel et al. 2002a):
0:6
A
G h  a i
Ẋ ¼ 2
tanj1 : ð4Þ
a
d 2
d
The ratio of the unshielded line source values was com-
pared to measured values to determine approximate self-
shielding values for the other radionuclides.
RESULTS
The average administered activity for thyroid cancer
patients was 4,750 MBq (Fig. 1). The average, standard
deviation, and minimum and maximum ratios of the cal-
culated/measured dose rates for these subjects are shown
in Table 1. Plots of the data at the 1 m distance are shown
in Fig. 2.
Mean activities for 99mTc (n = 11) and 18F (n = 52)
were 562 MBq and 494 MBq, respectively. All of the 18F
studies were with 18FDG; there were three Myoview
studies, five MDP studies, two Mebrofenin studies, and one
MAA study with 99mTc. The average, standard deviation,
and minimum and maximum ratios of the calculated/
measured dose rates for these subjects are shown in
Tables 2 and 3. Plots of the data at the 1 m distance for
18
F and 99mTc are shown in Fig. 3 and Fig. 4.
DISCUSSION
These findings confirm those of others, namely that
the use of an unshielded point source method to model
exposure rates near nuclear medicine patients produces
significant overestimates of the real dose rates. For 131I
cancer patients, who typically have very small thyroid
remnant uptakes and thus activity mostly distributed uni-
formly throughout the body, the exposure rates are
overestimated by a factor of about three at 0.3 m and about
2 at 1 m (Table 1). The minimum of all patients was about
0.8, but the maximum was a factor of nearly 9. The
unshielded line source method did better, overestimating
exposure only by factors of about 1.3Y1.5 at 0.3 or 1 m. The
authors got the same self-shielding factor of 0.6 suggested
by Siegel et al. for 131I (2002a), which brings the point
source model calculation/measurement ratios very close

Fig. 2. Ratios of calculated/measured exposure rates at 1 m from thyroid cancer patients receiving 131I.
www.health-physics.com

Copyright © 2013 Health Physics Society. Unauthorized reproduction of this article is prohibited.
190 Health Physics August 2013, Volume 105, Number 2

Table 3. Ratios of calculated/measured exposure rates near patients

receiving 99mTc radiopharmaceuticals.
Unshielded point Unshielded line
source source
0.3 m 1m 0.3 m 1m
Average 6.47 3.05 2.71 2.47
Standard deviation 2.69 2.23 1.15 1.78
Minimum 3.08 1.47 1.13 1.13
Maximum 12.66 8.86 5.39 7.04

to 1.0 on average, certainly within the uncertainty of the

data, so these data basically confirm the adequacy of this
approach. Uptake in the thyroid is probably well modeled
as an unshielded point source, but the activity in the rest of
the body should be treated with this self-shielding value.
For example, consider a cumulative dose calculation for
Fig. 4. Ratios of calculated/measured exposure rates at 1 m from
a thyroid cancer patient receiving a 3.7 GBq Na131I activity patients given 99mTc radiopharmaceuticals.
treatment prescription. Using the NUREG-1556 method,
treating all activity as an unshielded point source, one
would obtain: fraction); T2eff = 7.3 d (thyroid effective half-time); TP =
34:6GA0
8.04 d; 0.33 d ‘‘non-void’’ period; 0.75 occupancy
DðVÞ¼ fE 1 T p ð0:8Þð1jej0:693ð0:33Þ=T p Þþ ej0:693ð0:33Þ=T p E2 F 1 T 1eff þ ej0:693ð0:33Þ=T p E 2 F 2 T 2eff g
ð100cmÞ2 during ‘‘non-void;’’ and 0.25 occupancy post ‘‘non-void.’’
Treating activity in the body and thyroid as point
þ0:25
0:95
0:32d
ej0:693ð0:33dÞ=8:04d þ0:25
0:05
7:3d
ej0:693ð0:33dÞ=8:04d g sources but with a self-shielding factor of 0.6 for activity in
the body with a uniform occupancy factor of 0.25, one
DDEðVÞ ¼ 2:27mSv ð5Þ would obtain:


where A0=Initial activity at the start of the time interval; h mSvjcm2 3700 MBq
DðVÞ ¼ 1:443
24 0:595

d MBqjh ð100 cmÞ2
0:25
f0:05
7:3 d þ 0:6
0:95
0:32 dg
F1 = 0.95 (extrathyroidal fraction); T1eff = 0.32 d (effective
half-time for extrathyroidal fraction); F2 = 0.05 (thyroid DðVÞ ¼ 0:70 þ 0:34 ¼ 1:04 mSv ð6Þ

18
Fig. 3. Ratios of calculated/measured exposure rates at 1 m from patients given FDG.
www.health-physics.com

Copyright © 2013 Health Physics Society. Unauthorized reproduction of this article is prohibited.
 Dose rates near nuclear medicine patients c Y. YI
For the line source model, the authors found a self-
shielding factor of 0.73 for 131I. Treating activity in the
body as a self-shielding line source, one would obtain:


h mSvjcm2
DðVÞ ¼ 1:443
24 0:595
0:25

d MBqjh
" REFERENCES
3700 MBq

ð0:05
7:3 dÞ þ 0:73
0:95
0:32 d

ð100 cmÞ2


170 cm
3700 MBq
170 cm


2
a tan
2
100 cm
100 cm
 son of point, line and volume dose calculations for expo-
DðVÞ ¼ 0:70 þ 0:35 ¼ 1:05 mSv:
This result is not notably different than the second calcu-
lation, but note that the 5% in the thyroid modeled as an
unshielded point source accounts for the majority of the
dose. If there is no significant thyroid remnant, the line
source dose estimate is about 80% of the point source
calculation. So the NUREG method is more than a factor
of two higher than a point source calculation employing
the reduction factor of 0.6, and as noted above, a line
source calculation with a self-shielding factor of 0.73
would be the most realistic.
These data for 99mTc and 18FDG suggest similar self-
shielding factors of point and line source model calcula-
tions of about 0.4 and 0.5, 0.5 and 0.6, respectively (i.e.,
60% of the 99mTc and 50% of the 18F photons may be
absorbed within the body). In all of these studies, the ac-
tivity in the body is fairly well distributed to many organs,
so a line source approximation with a self-attenuation
factor represents a reasonable, conservative approach.
CONCLUSION
This study shows that theoretical calculations using
point and line source models both overestimate actual
exposure rates near patients. Either model, when used
with an appropriate self-shielding factor, provides rea-
sonable data for determining patient release criteria. The
authors found self-shielding factors using point and line
www.health-physics.com
Copyright © 2013 Health Physics Society. Unauthorized reproduction of this article is prohibited.
ET AL. 191
source model calculations for 131I, 18FDG and 99mTc from
231, 52 and 11 patients of 0.6 and 0.73, 0.4 and 0.5, and
0.5 and 0.6 (point and line source models for the three
radionuclides), respectively.
Cormack J, Shearer J. Calculation of radiation exposures from
patients to whom radioactive materials have been adminis-
tered. Phys Med Biol 43:501Y516; 1998.
de Carvalho AB Jr, Stabin MG, Siegel JA, Hunt J. Compari-
sure to nuclear medicine therapy patients. Health Phys 100:
185Y190; 2011.
Grigsby PW, Siegel BA, Baker S. Radiation exposure from
ð7Þ outpatient radioactive iodine (131I) therapy for thyroid car-
cinoma. JAMA 283:2272Y2274; 2000.
Rutar FJ, Augustine SC, Colcher D, Siegel JA, Jacobson DA,
Tempero MA, Dukat VJ, Hohenstein MA, Gobar LS, Vose JM.
Outpatient treatment with 131I-Anti-B1 antibody: radiation
exposure to family members. J Nucl Med 42:907Y915; 2001.
Siegel JA, Kroll S, Regan D, Kaminski MS, Wahl RL. A practical
methodology for patient release after Tositumomab and 131I-
Tositumomab therapy. J Nucl Med 43:354Y363; 2002a.
Siegel JA, Marcus CS, Sparks RB. Calculating the absorbed dose
to others from the radioactive patient: line source model versus
point source model. J Nucl Med 43:1241Y1244; 2002b.
Siegel JA, Marcus CS, Stabin MG. Licensee over-reliance on
conservatisms in NRC Guidance regarding the release of
patients treated with 131I. Health Phys 93:667Y677; 2007.
Sparks RB, Siegel JA, Wahl RL. The need for better methods to
determine release criteria for patients administered radioac-
tive material. Health Phys 75:385Y388; 1998.
U.S. Nuclear Regulatory Commission. Criteria for the release of
individuals administered radioactive material. Washington,
DC: U.S. Nuclear Regulatory Commission; 10 CFR Parts 20
and 35: 62 FR 4120; 1997.
U.S. Nuclear Regulatory Commission. Model procedure for
release of patients or human research subjects administered
radioactive materials. Washington, DC: U.S. Nuclear Regu-
latory Commission; NUREG-1556, Vol. 9, Rev. 1, Appendix
U; 2005.
Willegaignon J, Sapienza M, Ono C, Watanabe T, Guimaraes MI,
Gutterres R, Marechal MH, Buchpiguel C. Outpatient ra-
dioiodine therapy for thyroid cancer a safe nuclear medicine
procedure. Clin Nucl Med 36:440Y445; 2011.
Zanzonico PB. Radiation dose to patients and relatives incident
to 131I therapy. Thyroid 7:199Y204; 1997.
¡¡