IODINE-131 IS commonly administered as part of a diagnosis or therapy treatment plan for patients with differentiated thyroid cancer (DTC) and hyperthyroidism. The criteria for protection of members of the public from radionuclide therapy patients have been established by the US Nuclear Regulatory Commission (NRC) Regulatory Guide 8.39 (US NRC 1997). In prior work, Han et al. (2013) studied the effective dose to family members of an adult patient following 131I treatment, and Han et al. (2014) gave a practical guideline for the release of patients treated by 131I based on calculated dose to family members. However, patients administered 131I as a part of medical therapy will frequently stay in a hotel during their isolation period to avoid exposing family members (Vogel 2010). Dewji et al. (2015a) examined the case of an 131I patient staying in a hotel room and estimated the external dose rates to a guest staying in an adjacent room. However, dosimetric estimates to hotel workers handling potentially contaminated bed linens due to 131I radionuclide therapy has not been considered.
In the scope of this work, calculations of external dose rate coefficients were conducted for hotel housekeepers handling bed linens used by potentially contaminated by patients administered 131I. Detailed simulations consider both the tissue and material attenuation of radiation through the use of anthropomorphic phantoms, as well as considering the biokinetic body burden to determine the time-dependent excretion of 131I from the patient’s body. The effective dose rate coefficients to the housekeeper handling contaminated linens were calculated using International Commission on Radiological Protection (ICRP) Publication 103 methodology (ICRP 2007). The external dose rate estimates were derived using Monte Carlo methods and the phantom with moving arms and legs (PIMAL), previously developed by Oak Ridge National Laboratory (ORNL) and the US NRC (Akkurt and Eckerman 2011; Dewji and Hiller 2016; Dewji and Hertel 2017), to model the housekeeper handling a pile of bed linens. The time-dependent distribution of activity in the housekeeper’s body was predicted using a physiologically realistic biokinetic model for iodine developed at ORNL (Leggett 2010; Dewji et al. 2015a and b). Patient release criteria (Dewji et al. 2015a and b; Dewji and Hertel 2017) and linen contamination level postadministration (Nishizawa et al. 1980) were also used to estimate effective dose rate to the housekeeper. Three cases with selected distributions of contamination in the linens were modeled in the scope of this study.
Cases in which 131I patients with DTC and hyperthyroidism stay in a hotel after being administered radionuclide therapy were modeled to estimate the organ and effective dose rate coefficients to a housekeeper handling contaminated bed linens from 1 to 5 d postadministration. Hertel et al. (2014) report a study of a similar scenario but examine the internal dose rate to a housekeeper who inhales a puff of 131I upon handling the bed linens. The scope of this study focuses on the external dose rate estimation resulting from a housekeeper handling a pile of bed linens contaminated by deposition of perspiration and urinary excretion.
To model a housekeeper holding a pile of bed linens with an embedded 131I distributed source, an articulated stylized phantom was generated using the PIMAL 4.1.0 software (Dewji and Hertel 2017) with coupled radiation transport conducted with the Monte Carlo N-Particle (MCNP) radiation transport code version 6.2 (Werner 2017). PIMAL is a stylized computational phantom that enables articulation of the arms at the shoulder and elbow joints, and of the legs at the hip and knee joints. This software includes both male and female mathematical (stylized) phantoms that are based on modifications of the ORNL phantom models (Han et al. 2006). The male and female mathematical phantoms were both employed during this study and were modeled in a standing upright position, with arms extended holding a spherical model of bed linens with a radius of 19 cm to fit directly between the arms of each phantom. A total of six cases were studied during this investigation: three cases using a male mathematical phantom and three cases using a female mathematical phantom. The geometries of the phantoms themselves do not change during the study, but the size/distribution of the 131I embedded contamination in the bed linens changes from case to case. Fig. 1 depicts the simulation models generated using VisEd (Schwarz et al. 2007) from MCNP6.2 (Werner 2017) and PIMAL.
The male and female mathematical phantom models used during this study incorporate tissue compositions and densities from ICRP Publication 89 (ICRP 2002). The spherical pile of bed linens used during this investigation consists of two linens and a coverlet, all composed of cotton, which is 99% cellulose, has a density of 1.5 g cm−3 (NIH 2012), and has a volume of 2.87 × 104 cm3.
Source and tally definition
The effective dose rate calculations due to external exposure to contaminated bed linens were conducted by creating a homogeneously distributed 131I source within a pile of linens for each of the simulation cases. For each calculation, a unit activity of 131I was assigned to the contaminated region, and the 131I photon spectrum was taken from ICRP Publication 107 (2009). These cases, as depicted in Fig. 2, are as follows.
- Case A: the contaminated region is an embedded spherical source with a radius of 8 cm in the middle of the pile of linens with contamination uniformly distributed in the embedded source.
- Case B: the entire interior of the pile of linens is contaminated as a uniformly distributed volumetric source.
- Case C: the entire exterior surface of the pile of linens is uniformly contaminated (i.e., interior contamination is negligible).
The organ doses were computed using the kerma approximation (MeV g−1) in MCNP (F6 tally) due to the photon source. Beta emissions of 131I were not considered during this study. Doses to active marrow and bone surface were estimated by applying the skeletal response functions reported by Cristy and Eckerman (1987) as cell-averaged flux (particles cm−2) in MCNP (F4 tally) in the skeletal regions outside of PIMAL.
Effective dose rate calculations
Time-dependent effective dose rate coefficients were determined for the exposed housekeeper for each contamination scenario as a function of time postadministration. Monte Carlo simulations were run for each case until convergence for a relative error below 5% for the absorbed dose tally for the adrenals, corresponding to a particle history of 1 × 107 particles. Effective dose computations were employed using the methodology recommended in ICRP Publication 103 (2007). Sex-averaged tissue-weighting factors given in ICRP Publication 103 are summarized in Table 1, which were then used to calculate effective dose rate coefficients from the organ absorbed dose rate coefficients modeled from the MCNP 131I photon exposure simulations.
The computation of the effective dose coefficient E employed the methodology recommended in ICRP Publication 103 (2007). The effective dose coefficient is defined as follows:
is the equivalent dose coefficient in organ/tissue T for the male phantom,
is the equivalent dose coefficient in organ/tissue T for the female phantom, and wT is the tissue-weighting factor for organ/tissue T (from Table 1).
The male and female equivalent dose coefficients were determined using the following equation:
where wR is the radiation-weighting factor and DT,R is the absorbed dose from radiation R to organ/tissue T. The scope of this work was focused on photon sources from 131I, so a radiation-weighting factor of 1 was employed, as recommended in ICRP Publication 103 (2007).
The biokinetic model for systemic iodine developed by Leggett (2010) and modified for use in the study by Dewji et al. (2015a) was used to predict the activity distribution in the 131I patient’s body following radionuclide administration. This model was used to estimate the activity that would be excreted by urination on bed linens from 1 to 5 d after dose administration, in order to calculate external dose rate estimates to a hotel housekeeper handling the bed linen. The model used for 131I distribution was simplified into three regions: thyroid contents, urinary bladder contents, and a combination of all other soft tissue.
Two levels of 131I uptake were considered for each scenario during this work:
- (a) DTC: peak uptake of 5% of administered activity.
- (b) Hyperthyroidism: peak uptake of 80% of administered activity.
The model used for this work was produced under the assumption that the 131I patient will void urinary bladder contents every 4 h. Fig. 3 depicts the retention of iodine in the patient’s body for the first 120 h post-131I administration (Dewji et al. 2015a). The excretion model is driven by bladder excretion for DTC (in which thyroid metabolism is minor due to the removal of the gland), whereas hyperthyroidism metabolic excretion is dominated by thyroid metabolism rather than bladder excretion.
In an alternative case, maximum excretion rates and maximum recoverable activity values due to perspiration on human skin reported by Nishizawa et al. (1980) were adopted in this study to estimate effective dose rates to the housekeeper for DTC and hyperthyroidism scenarios. In the original study by Nishizawa et al., activity values were determined by performing a smear test on nine patients: six with hyperthyroidism and three with differentiated thyroid cancer. This test involved collecting activity with filter paper from the patient’s skin 1 h before and after an initial smear and then summing the total activity. This was done for each of the first 5 d postadministration. The total activity recovered for each patient was divided by the administration dose and then multiplied by an average perspiration value of 37.5 g h−1 to give a maximum fraction of administration dose recoverable in perspiration per hour. This value was averaged among the nine patients tested. Nishizawa and collegues’ data employed in effective dose computation in this study are summarized in Table 2.
Assumptions employed during this study include:
- (a) Dose from internal exposure is negligible.
- (b) The 131I patient goes to sleep on bed linens 10 h after administration.
- (c) The patient sleeps on linens for 9 h.
- (d) The 131I decays normally on the pile of linens between the time the patient is no longer in contact with it and the time the housekeeper handles it.
Results were tabulated during this investigation for six different cases: three using a female phantom and another three using a male phantom. All six cases consisted of the phantom standing in an upright position with arms extended holding a spherical pile of cotton bed linens with a radius of 19 cm. The three different cases for each phantom consist of changing only the source size/distribution.
Bed linen activity determination
Biokinetic excretion data for a patient voiding urinary bladder contents every 4 h was used to determine the maximum fraction of administration activity remaining on the bed linens after the patient sleeps on them. It is assumed that the patient first contacts the linens 10 h postadministration and sleeps for 9 h. It is assumed that the incontinent patient will void onto the bed linens twice overnight, which represents an upper bound for the study. Table 3 shows the maximum fraction of administered 131I excreted during the 9-h sleeping period and compared to the fraction of excreted activity recoverable as perspiration on the patient’s skin during the 9-h sleeping period. The perspiration value was determined using data reported by Nishizawa et al. (1980), which gives fractional administration activity recoverable in perspiration on the patient’s skin surface for the first 5 d postadministration. To determine the activity from the exposure source to the housekeeper, a physical decay constant of 3.6 × 10−3 h−1 for 131I was used for the 5 h interval between the time the patient leaves the bed and the time the housekeeper contacts the linens. This decay constant was employed in all excretion models.
Case A: localized volume contamination results
The first case investigated during this study examines the scenario in which the contamination to the bed linens is a localized stain from an 131I patient, and the stain is represented as a sphere with a radius of 8 cm located on the interior of the pile of bed linens. Computed sex-averaged dose rate coefficients to organs of interest are given in Table 4 from the MCNP simulations. The effective dose rate coefficients due to removable perspiration from the patient and excretion from the incontinent patient to the housekeeper as a function of time postadministration for the DTC and hyperthyroidism cases are summarized in Table 5.
Case B: full volume contamination results
The second case investigated during this study examines a case in which the contamination on the bed linens is the entire interior of the pile of linens due to excretion from an 131I patient. Computed sex-averaged dose rate coefficients to core organs of interest resulting from this scenario are displayed in Table 6 from the MNCP simulations. The contamination is represented as a sphere with a radius infinitesimally smaller than the outer radius of 19 cm of the pile of bed linens. The effective dose rate coefficients due to removable perspiration from the patient and excretion from the incontinent patient to the housekeeper as a function of time postadministration for the DTC and hyperthyroidism cases are summarized in Table 7.
Case C: surface contamination results
The final case of this investigation examines the distribution of contamination over the entire surface of the pile of linens contaminated due to excretion from the 131I patient. Calculated dose rate coefficients to core organs from this scenario are given in Table 8. Table 9 summarizes the effective dose rate coefficients to the housekeeper as a function of time postadministration for each 131I patient case.
Each scenario examined during this study yielded effective dose rate coefficients that were lower for the hyperthyroid case than the DTC case on the first day, within 1% on the second day, and greater for the hyperthyroid case than the DTC case for the following 3 d postadministration. This is assumed to be a result of the DTC patient having a higher excretion rate driven by active urinary bladder voiding (instead of thyroid metabolism) over the first day after administration (when the patient was awake and voiding) and as confirmed by the biokinetic models in Fig. 3. The case that provided the highest effective dose rate coefficient to the housekeeper was the surface contamination case, followed by the full volume contamination case. If administration activities corresponding to 1.1 GBq for hyperthyroidism and 7 GBq for DTC are assumed, aligned with the administered activities from Nishizawa et al. (1980), DTC patients will yield a higher effective dose to the housekeeper, which was consistent with the exposure studies by Dewji et al. (2015a).
Table 10 displays effective dose rates to the hotel housekeeper based on typical administration doses for 131I patients from the bed linen surface contamination scenario. This table shows that for the first 2 d following radionuclide administration, patients treated with 5.5 GBq and 7 GBq will yield effective dose rates to the housekeeper that are above 80 μSv h−1, assuming full excretion by the patient. Thompson (2001) reports that any member of a radiation safety office staff who receives more than 50 μSv h−1 while cleaning the bed sheets of an 131I patient needs to be monitored closely, meaning the housekeeper in the upper-bound cases of this study would need close monitoring. The effective dose rate coefficients calculated during this study are conservative, assuming full excretion onto the bed linens by the 131I patient. As perspiration is the most likely form of excretion onto the bed linens, effective dose rate per unit administered activity to the housekeeper due to contamination from perspiration were calculated and were an average of 2–3 orders of magnitude lower than if full excretion by urination is assumed. Recoverable activity values from this study are consistent with values obtained in Nishizawa et al. (1980), who report a maximum contamination of 0.141 MBq on the bed linens on the first day after administration for a patient who was administered a 1.11 GBq dose. If a 1.11 GBq administration dose is assumed in this study, the maximum contamination on the bed linens on the first day after administration is 0.235 MBq. This value represents the conservative nature of this study, which assumed a 9-h sleeping period and 100% removal of skin contamination as perspiration onto the bed linens. The highest effective dose rate coefficient to the housekeeper calculated during this study was on the first day of the surface contamination case and was 1.12 × 10−1 μSv (MBq h)−1 for an incontinent patient with differentiated thyroid cancer.
Patients administered 131I as part of a medical treatment plan may stay in a hotel following the administration to avoid contact with family members and household items. During this study, tissue attenuation and biokinetics along with biological removal of the radionuclide reported by Nishizawa (1980) were used to determine a maximum amount of activity that could remain on bed linens postadministration due to excretion from the patient. This activity was then used along with computed dose rate coefficients to estimate effective dose rates per unit administered activity to a hotel housekeeper handling the bed linens.
The case that yields the highest effective dose rate per unit activity to the housekeeper is the case in which the entire surface of the pile of linens is contaminated with urine from the 131I patient. The dose rate from the DTC patient sleeping on the linens is greater or equal to the dose rate from the hyperthyroid patient for the first 2 d and lower for the following 3 d postadministration, but if average administration activities of 1.1 GBq for hyperthyroidism and 7 GBq for DTC are assumed, the dose rate will be higher in the DTC patient case than the hyperthyroidism patient case for all 5 d. The scenario examined during this study that yielded the lowest effective dose rate per unit activity to the housekeeper was the case in which the contamination was modeled as a localized volume embedded in the bed linens.
Prior to this study, limited work has been conducted to approximate the external dosimetric risk using computational phantoms and biokinetic models, and no studies have addressed the application of external exposures to hotel housekeepers from 131I patients. Effective dose rates per unit activity to the hotel housekeeper from the DTC patient are greater or equal to the dose rates from the hyperthyroid patient. This is assumed to be a result of the DTC patient having a higher excretion due to urinary bladder voiding over the first day after administration, and the patient is assumed to void urinary bladder contents twice overnight. Potential scenarios of interest that will be considered in future work include computation of removable contamination transferred directly to the housekeeper’s skin or reabsorption onto the patient’s skin from excretion on the linens, in addition to urine excretion in a contaminated bathroom; these scenarios are being considered in the next phase of this project, as is expanding upon the list of radionuclides of interest for nuclear medicine patient release. Finally, evaluating occupancy factors in the context of long-term proximity to potentially contaminated environments would provide data that would permit evaluating dosimetric risk to hotel workers.
Akkurt HW, Eckerman K. Recent updates to radiation dose estimation software: PIMAL. Trans Am Nucl Soc 104:635–636; 2011.
Cristy M, Eckerman K. Specific absorbed fractions of energy at various ages from internal photon sources; I. Methods. Oak Ridge, TN: Oak Ridge National Laboratory; ORNL-8381 V/1; 1987.
Dewji S, Bellamy MB, Hertel N, Leggett R, Sherbini S, Saba M, Eckerman E. Estimated dose rates to members of the public from external exposure to patients with I-131 thyroid treatment. Med Phys 42:1851–1857; 2015a.
Dewji SA, Bellamy M, Hertel N, Leggett R, Sherbini S, Saba M, Eckerman K. Assessment of the point-source method for estimating dose rates to members of the public from exposure to patients with 131
I thyroid treatment. Health Phys 109:233–241; 2015b.
Dewji S, Hertel NE. Patient release after radiodine therapy: a review of the technical literature, dose calculations and recommendations. Rockville, MD: US Nuclear Regulatory Commission; 2017. Available at https://www.nrc.gov/docs/ML1726/ML17262A909.pdf
. Accessed 20 December 2018.
Dewji SA, Hiller MM. PIMAL: phantom with moving arms and legs, version 4.1.0. Rockville, MD: US Nuclear Regulatory Commission; NUREG/CR-7243; 2016.
Han EY, Bolch WE, Eckerman KE. Revisions to the ORNL series of adult and pediatric computational phantoms for use with the MIRD schema. Health Phys 90:337–356; 2006.
Han EY, Lee C, Mcguire L, Bolch WE. A practical guideline for the release of patients treated by I-131 based on Monte Carlo dose calculations for family members. J Radiol Prot 34:N7; 2014.
Han EY, Lee C, Mcguire L, Brown TLY, Bolch WE. Organ S values and effective doses for family members exposed to adult patients following I-131 treatment: a Monte Carlo simulation study. Med Phys 40:083901; 2013.
Hertel NE, Bellamy MB, Dewji S, Leggett R, Eckerman K, Sherbini S, Saba M. Estimates of internal dose resulting from I-131 patients in hotels. In: Proceedings of the 18th topical meeting of the Radiation Protection & Shielding Division of the American Nuclear Society. La Grange Park, IL: ANS; 2014: 38–45.
International Commission on Radiological Protection. Basic anatomical and physiological data for use in radiological protection: reference values. Oxford: Pergamon Press; ICRP Publication 89; 2002.
International Commission on Radiological Protection. Nuclear decay data for dosimetric calculations. Oxford: Pergamon Press; ICRP Publication 107; 2009.
International Commission on Radiological Protection. The 2007 recommendations of the International Commission on Radiological Protection. Oxford: Pergamon Press; ICRP Publication 103; 2007: 64–65.
Leggett RW. A physiological systems model for iodine for use in radiation protection. Radiat Res 174:496–516; 2010.
National Institutes of Health. PubChem: open chemistry database [online]. 2012. National Center for Biotechnology Information, US National Library of Medicine. Available at https://pubchem.ncbi.nlm.nih.gov
. Accessed 20 December 2018.
Nishizawa KKO, Ohshima M, Maekoshi H, Orito T, Watanabe T. Monitoring of I-131 excretions and used materials of patients treated with I-131. Health Phys 38:467–481; 1980.
Schwarz R, Carter LL, Schwarz A. Modification to the Monte Carlo N-Particle (MCNP) Visual Editor (MCNPVisEd). Los Alamos, NM: Los Alamos National Laboratory; 2007.
Thompson M. Radiation safety precautions in the management of the hospitalized 131
I therapy patient. J Nucl Med Technol 29:61–66; 2001.
US Nuclear Regulatory Commission. Release of patients administered radioactive material: regulatory guide 8.39. Rockville, MD: US Nuclear Regulatory Commission; 1997.
Vogel L. Reducing the health risks posed by “hot” patients. CMAJ 193:E41–42; 2010.
Werner CJ, ed. MCNP user’s manual, code version 6.2. Los Alamos, NM: Los Alamos National Laboratory; LA-UR-29981; 2017.