ML20133K620
| ML20133K620 | |
| Person / Time | |
|---|---|
| Site: | 03001786 |
| Issue date: | 11/22/1995 |
| From: | Dwyer J, Donna-Beth Howe, Lodhi A, Shankman S NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION I) |
| To: | |
| Shared Package | |
| ML20133K613 | List: |
| References | |
| 30-01786-95-02, 30-1786-95-2, NUDOCS 9701210284 | |
| Download: ML20133K620 (112) | |
See also: IR 05000630/2011015
Text
. ._ _ ._...__. _._ _. _ . _ . . . . _ _ _ _ _ . . _ ~ _ . _ _ . _ _ _ . . . . _ . - - ~ . _ - - ~
s
4
l
i
'
!
.
,
! U.S. NUCLEAR REGULATORY COMMISSION
j REGION I
2
AUGMENTED INSPECTION TEAM REPORT
j Report No. 030-01786/95-002
l
I Docket No. 030-01786 -
i
s
[ License No. 19-00296-10 Priority 1 Category G1 Program Code 02110
Licensee: Department of Health and Human Services
National Institutes of Health
31 Center Drive. MSC 2260
Bethesda, Maryland 20892-2260
Facility Name: National Institutes of Health -'
Inspection at: Bethesda, Maryland -
Inspection Dates: June 30 through November 15. 1995
Inspectors:
Jam #5'7. Dwyer, Team Leader
//
'Dath
I
~m-
Donna-Beth Howe, Ph.D.
n/ukc
'Dath
A. Sattar Lodhi, Ph.D.
b //
'Dator
I
Approved by: '
Susan F. Shankman, Deputy Director
///Date/
N!
Division of Nuclear Materials Safety
9701210284 970113
PDR ADOCK 03001786
C PDR
_ _ . _ _ . _. -.- _- _
- - . . . ~ . - - _ . . .. - -- .. - -- . - -
i
EXECUTIVE SUMMARY
AREAS INSPECTED:
An Augmented Inspection Team (AIT), consisting of personnel from the Division
I of Nuclear Materials Safety in Region I and the Division of Industrial and
'
Medical Nuclear Safety in the Office of Nuclear Material Safety and :
Safeguards, reviewed the licensee's response to the internal contamination
incident identified by the licensee on the evening of June 29. 1995. The
incident involved the ingestion of phosphorus-32 by a pregnant researcher. !
The specific source of the pregnant researcher's internal contamination is not l
yet known but it is probable that her contamination occurred on June 28, 1995.
The licensee subsequently identified 26 additional individuals who were 1
internally contaminated with lesser quantities of phosphorus-32. A
contaminated water cooler was identified as the source of these additional ;
contaminations on July 14. 1995. Urine bioassay data indicate that the
contamination of the water cooler occurred before July 3,1995.
The AIT was assisted in this effort by a scientific consultant from the Oak
Ridge Institute for Science and Education's Radiation Internal Dose
Information Center and by a medical consultant. The scientific consultant
performed an assessment of the intake and the resultant radiation dose to the
pregnant researcher and to her fetus. The medical consultant provided opinion
on the )otential medical consequences of the exposure to the pregnant
researcler and to her fetus. The NRC contracted with the Lawrence Livermore
National Laboratory to perform a third-party independent assessment of the
intake and dose assessments performed by the licensee and the scientific
consultant. The AIT also included an investigator from the NRC's Office of
Investigations (01). The investigator was charged with assisting the NIH
Police Department and the Federal Bureau of Investigation with the
investigation of the cause of the pregnant researcher's contamination.
RESULTS:
The AIT determined that the ) regnant researcher received an intake of
phosphorus-32 in excess of tie Annual Limit of Intake specified in 10 CFR Part l
20, Appendix B Table 1. Column 1. The resultant dose to the pregnant
researcher from this intake exceeded the occupational dose limit for adults
specified in 10 CFR 20.1201(a)'.1)(i). The radiation exposure received by the
pregnant researcher's fetus exceeds the dose limit for the embryo / fetus l
specified in 10 CFR 20.1208. Five of the 26 additionally contaminated
individuals were determined to be members of the public. One of those five
individuals received a radiation exposure in excess of the dose limits for
individual members of the public, as specified in 10 CFR 20.1301(a)(1). In
all. 27 individuals were unnecessarily exposed to radiation. ;
j
,-
. ~ - - - - . . _ - - - - - - . - - . _ _ - . . _ - -. - - - .
l
l
l
!
The AIT determined that the licensee's initial response to the incident was
l appropriate, however, subsequent follow up inappropriately focused on one
I
ingestion aathway. Radiation surveys performed by the NIH Radiation Safety
l' Branch (RS3) following identification of the incident on June 29, 1995, were
limited in scope and subsequent surveys performed by research staff were not l
adequately reviewed. Had complete surveys of the fifth floor of Building 37 l
been performed, it is likely that the contaminated water cooler would have 1
been identified prior to July 14 and fewer individuals would have received j
l unnecessary exposures. In addition, evidence suggests that the processing of .
I some urine bioassay samples was delayed as long as 3 days. Had these samples
i
been processed earlier, it is likely that the RSB would have been prompted to
survey for additional sources of contamination before July 14 and fewer
individuals would have been affected by the contaminated water cooler.
Although there is currently no evidence to suggest that inadequate security of
radioactive material was the proximate cause of the internal contaminations,
weaknesses in the security of radioactive material at NIH were identified that
may be a contributing factor. Additional problems were identified in the
areas of training, purchase of radioactive material, unauthorized use of
radioactive material, radioactive material inventory, use of external
dosimetry, and program audits.
CAUSES:
The AIT concluded that the contamination of the pregnant researcher and the 26
additional individuals was not accidental. An investigation into the cause(s)
of these coritaminations is ongoing.
!
l
,
j
. - . . _ . . - - . . . .. . .
!
l
TABLE OF CONTENTS
EXECUTIVE SUMMARY
CONTENTS
1. INTRODUCTION
1.1 Initial Notification
1.2 NRC Response
1.3 Augmented Inspection Team Activities
l
l 2. BACKGROUND
2.1 Use of Licensed Radioac'ive Materials at NIH
2.2 Activities in Laboratot.; SD18 Before Internal Contamination Events
3. CHRONOLOGY OF EVENTS
4. DOSE ASSESSMENT
l 4.1 Researcher A Intake Estimates
4.2 Researcher A's Committed Effective Dose Equivalent Estimates
4.3 Dose Estimates for Researcher A's Fetus
4.4 Other Contaminated Individuals
5. SAFETY SIGNIFICANCE I
l
5.1 Effect on the Pregnant Researcher
5.2 Effect on the Fetus of the Pregnant Researcher
5.3 Effect on Other Contaminated Individuals
6. ASSESSMENT OF THE LICENSEE'S RESPONSE TO THE INCIDENT
7. ASSESSMENT OF THE BYPRODUCT MATERIAL PROGRAM AT NIH IN AREAS RELATED TO
THE ACCESS AND CONTROL OF LICENSED RADI0 ACTIVE MATERIAL
7.1 Purchase and Receipt of Radioactive Material
7.2 Radioactive Material Use and Inventory Control
7.3 Waste Disposal i
7.4 Security of Radioactive Materials '
8. CONCLUSIONS
8.1 Cause of the Incident
8.2 Licensee's Initial Response
8.3 Licensee's Follow-up
8.4 Security of Radioactive Material
8.5 Dose to Contaminated Individuals
8.6 Training of Researchers A and B
8.7 Analytical Ca) abilities
8.8 Purchase and Jse of Radioactive Material
! 8.9 Collection and Evaluation of Bioassay Samples
.- - - - -- - -. ._ - - _ - -.
! 8.10 Use of External Dosimetry in Laboratory 5D18
l 8.11 Program Audits '
9. EXIT MEETING
,
10. ABBREVIATIONS AND DEFINITIONS
1
11. APPENDICES l
Appendix A - AIT Charter
Appendix B - CAL-1-95-011 .
Appendix C - CAL-1-95-011, Revision 1 '
Appendix D - Floorplan of NIH Building 37. Fifth Floor
Appendix E - Contaminated Water Cooler Radionuclide Analysis
Appendix F - NIH Ouestionnaire
Appendix G - Ingestion and Committed Effective Dose Equivalent Estimates
Appendix H - Split Bioassay Sample Processing and Data Analysis
Appendix I - Summary of NIH Urine Bioassay Collection and Counting
Chronology
l
l
l
l
4
J
l
l
,
_ _ _ . _ _ _ _ _ _ _ _ . .. _ . _ - _ _ _ . _ _ _ _ . _ _ _ _
,
i
i
'
DETAILS
1 INTRODUCTION l
1.1 Initial Notification f
!
At a) proximately 8:30 a.m. on June 30, 1995. the licensee notified i
an NRC Region I inspector, who was onsite conducting a routine
inspection, that it had identified an incident involving
phosphorus-32 (P-32) internal contamination of a pregnant !
researcher. The contamination initially was discovered by the
pregnant researcher's husband, a coworker, at approximately 5:00 ;
p.m. on June 29, 1995, and was reported to Radiation Safety Branch
, (RSB) personnel at 6:00 ).m. that evening. RSB and National :
l Institutes of Health (NIi) emergency medical personnel responded
to the laboratory.- located on the fifth floor of Building 37. and l
l the pregnant researcher was taken to a local hospital emergency
l room for observation and treatment. Initial urine bioassay :
'
results indicated an intake of approximately 11.1 megabecquerels
(MBq) (300 microcuries) of P-32. The emergency room attending t
physician was concerned that some treatments might. interfere with
the researcher's pregnancy and elected to treat the researcher by ;
simple hydration. The licensee contacted the Oak Ridge Institute l
L for Science and Education (0 RISE) Radiation Emergency Assistance i
Center / Training Site (REAC/TS) and arranged for the attending '
physician to confer with a physician who is a specialist in the
medical management of persons accidentally contaminated with
radioactive materials. The REAC/TS physician stated that he
discussed with the attending physician the )ossibility of
administering sodium phosphate to inhibit a) sorption of P-32 from
the gastrointestinal tract. but recommended against this because
of the amount of time that had passed since the ingestion,
estimated by the RSB at 9 hours1.041667e-4 days <br />0.0025 hours <br />1.488095e-5 weeks <br />3.4245e-6 months <br />
Because the contamination of the pregnant researcher may have been
the result of a malicious act, the licensee notified the NIH
! Police Department on the morning of June 30. 1995.
The inspector notified NRC Region I management about the incident
at approximately 9:00 a.m. on June 30, 1995. The licensee made a ;
formal notification to the NRC Emergency Operations Center at 3:04 i
p.m. on June 30, 1995. !
1.2 NRC Response
!
On June 30, 1995, the NRC Region I Administrator directed the
charter (Appendix A). The AIT was formed because the magnitude
and the cause of the intake were unknown and because the incident
involved a pregnant woman. The AIT included an investigator from ;
,
the NRC's Office of Investigations. The investigator was charged
i
with assisting the NIH Police Department in its investigation of
the cause of the pregnant researcher's contamination. The AIT
.__ _
. .- . - - . - - -
- . . . =- .- -_ . .- - - _. _ . - . . - - . - - -. -
f
also included a medical consultant. The medical consultant stated
that, based on the licensee's initial estimate of P-32 intake by :
the ) regnant researcher, he did not believe that there would be l
any lealth consequences to the researcher or to her fetus. The l
medical consultant concurred with the decision to hydrate the l
pregnant researcher and suggested that a complete hematological l
profile and a nuclear medicine scan.be performed. An NRC l
t
scientific consultant at ORISE's Radiation Internal Dose !
l Information Center (RIDIC) also was consulted. The NRC scientific
consultant recommended that the licensee collect and analyze
i several 24-hour urine samples, in order to more accurately
estimate the pregnant researcher's intake of P-32. perform
)eriodic whole body counts to assess the clearance rate of the
A32 from the pregnant researcher's body, and obtain nuclear
medicine images of the bremsstrahlung radiation from the P-32 to
visualize the distribution of the contaminant. The licensee
separately contacted this same consultant to assist with its dose
estimation.
1.3 Augmented Inspection Team Activities
At 2:00 p.m. on June 30. 1995.-the Alf conducted an entrance
meeting with NIH management and RSB staff to explain the purpose
and scope of the inspection. The AIT conducted inspection
activities at the licensee's facilities in Bethesda, Maryland
between June 30 and July 11. 1995. These activities included
interviews of licensee staff, document and. record review, physical
review of facilities, and follow-up on RSB efforts to quantify the
pregnant researcher's intake of P-32 and identify additional ,
contaminated individuals. !
On July 17. 1995, the AIT returned to the Bethesda campus !
following the licensee's identification, on July 14,1995, of a !
water cooler contaminated with P-32. The water cooler was located
on the fifth floor of Building 37. The licensee discovered the
-contaminated water cooler while conducting radiation surveys
prompted by the identification of additional internally
contaminated individuals.
On July 18. 1995. NRC and NIH management met to discuss the need
for NIH to develop a plan to: (1) substantially reduce the
possibility of further ingestion of radioactive material: and (2)
determine that the full scope of the 3ersonnel contaminations was
known. NRC management also met with VIH management. NIH Police,
and investigators from the Federal Bureau of Investigation (FBI),
to discuss the status of the investigation. On July 21, 1995, the
NRC issued Confirmatory Action Letter (CAL) 1-95-011 (Appendix B)
which described the actions that the licensee would take to reduce
the possibility of further ingestion of radioactive material and
to determine that the full scope of the personnel contaminations
was known. Later on July 21, 1995, to clarify certain points in
the CAL. Revision 1 (Appendix C) was issued. These actions
2
_ - - - . . - -
_ _ . . _ _ _ _ _.
_ ,
i
l
included:
1
(1) completing a radiological survey of all areas of Building 37
by July 24, 1995:
(2) obtaining urine samples from all available and willing l
individuals in Building 37 and completing the analysis of each :
sample by July 28, 1995: )
(3) developing by July 21, 1995, and implementing a statistically i
'
valid plan to collect and analyze urine samples from other NIH
staff:
(4) developing and beginning implementation of an augmented
continuing radiological survey program for the NIH campus by July
28, 1995:
(5) assisting the building safety committee with the development l
and implementation of plans for security of the food storage and I
preparation areas by July 28, 1995: and )
(6) disseminating, by August 1,1995, the enhanced security policy
that was made a permanent policy by the licensee's Radiation
Safety Committee (RSC) during a meeting of that body on July 20,
1995. This policy specified that breaches of security would
result in mandatory suspension of privileges to use radioactive ;
materials. i
The AIT continued its onsite ins]ection through July 28, 1995.
The AIT conducted a technical de3rief with RSB management and RSB
staff on August 3, 1995, and with senior NIH management on August
8, 1995. The AIT conducted a briefing for members of the media on
August 8, 1995.
Further inspection activities, including assessment of radiation
dose to the exposed individuals, and evaluation of a third-party
independent dose assessment, continued through November 15, 1995.
Although sharing safety significant information, the OI
investigator worked independently of the AIT with investigators
from the NIH Police Department and the FBI. An NRC health physics
s)ecialist was assigned to provide technical assistance to the
F 3I .
2 BACKGROUND
2.1 Use of Licensed Radioactive Materials at NIH
NIH has an NRC license of broad scope that authorizes the
possession and use of radioactive material for medical diagnosis,
therapy, and research in humans, as well as non-human research and
development, at facilities in Bethesda, Rockville, Baltimore, and
3
._ ._ . .
i
'
l
! Poolesville. Maryland. The NIH main campus in Bethesda has 21
l buildings housing nearly 3000 biomedical research laboratories. l
l There are more than 800 authorized users and more than 5000 l
l supervised users of radioactive material under NIH's licensed
l program.
l I
The NIH Radiation Safety Committee (RSC) is responsible for
oversight of the licensed program. Members of the RSC are
appointed by the Director of NIH, Although NRC regulations ;
require
monthly. quarterly
Attendance meetings of the RSC.
and participation NIH meetings
at these holds its RSC meetings
are high.
The Radiation Safety Officer (RS0) is res)onsible for the day-to- i
day operation of the licensed program. T1e RSO is also the Chief ;
of the RSB. For radiation safety matters, the RSO reports =to the :
RSC: administratively, the RSO reports to the NIH Director through !
the Division of Safety and the Office of Research Services. The l
RSO is assisted by a RSB staff of 32 individuals, including 23 1
health physicists. The RSB has a similar number of contractors.
i
The RSC must ap3 rove authorized users of radioactive material. l
Approval of autlorized users is based on an individual's training '
and experience. Criteria used by the RSC to approve research by
authorized users are described in the license. Approximately 30
of the more than 800 authorized users are approved to use
" protocol" amounts and types of radioactive material. A protocol
is required for high-quantity use and use of volatile radioactive
materials. The licensee does a detailed review of the training
and experience of these individuals and, in addition, reviews the
applicant's facilities, equipment, and procedures before granting
authorization. This review is comparable to the review performed
by NRC before approving a user on a limited-scope iicense.
The licensee's criteria for approving users of sub-protocol
amounts and types of radioactive material are the successful
completion of the NIH authorized user training course (or
equivalent) and at least 6 weeks of laboratory experience using
radioactive materials. An applicant who has this training and !
experience can be authorized to possess a large scope of I
radioactive material, including up to 370 MBq (10 mci) of
phosphorus-32, 925 MBq (25 mci) of hydrogen-3. 37 MBq (1 mci) of
non-volatile iodine-125. 184 MBq (5 mci) of carbon-14, 370 MBq (10
mci) of sulfur-35. 370 MBq (10 mC1) of chromium-51. I
and 37 MBq (1 mci) of calcium-45. ;
Authorized users are required to successfully complete the
" Radiation Safety for Authorized Users" course. Authorized user
candidates can be exempted from portions of this course if they
have received equivalent training and can demonstrate proficier.cy
by passing a comprehensive examination.
Supervised users are required to successfully complete the
4
_ . - .. - - . _
i " Radiation Safety in the Laboratory" course. This course is
offered each month. Supervised users are permitted to begin
working with radioactive materials before attending this course
3rovided they have been issued appropriate dosimetry and received
l 3asic radiation safety training from the authorized user. This
I basic radiation safety training is entitled, " Radiation Safety
l Orientation for New Personriel Planning to Use Radioactive
Material". Written certification of receipt of this training by
the authorized user is required. Supervised users are required to
attend the first available " Radiation Safety in the Laboratory"
course. Retraining of all research personnel using radioactive
material is required every 2 years.
The NIH license requires, in part, that whole body dosimetry be
i worn by anyone working in a restricted area with beta-emitting
i isotopes having a maximum energy of 200 kiloelectron volts (kev)
or greater. The license also requires that extremity dosimetry be
worn by anyone working with greater than 0.5 millicuries of P-32.
NIH contractors perform periodic audits of each laboratory. The
frequency of audits de) ends upon the level of radioactive material
l use. Most research la) oratories are only audited twice each year.
These audits include detailed radiation surveys. Aside from these
formal audits, the AIT determined that RSB staff and contractors
spend significant time in the laboratories providing consulting
- services, delivering packages, and picking up waste. These
occasions provide opportunities for RSB staff and contractors to
identify problems.
Researchers are required to perform radiation surveys after each
use of radioactive material. This survey is not required to be
documented. Researchers are also required to perform a survey for
removable contamination each month and submit the survey report to
the RSB for review. The survey report also requests information
on the radioactive materials used in the laboratory in the
preceding month and the quantities used. Authorized users usually
delegate the responsibility for conducting this survey to a
supervised user.
2.2 Activities in Laboratory 5D18 Before Internal Contamination Events
Laboratory space for approximately 600 researchers is provided in
Building 37, located on the NIH campus in Bethesda. Maryland. All
of the researchers in Building 37 work for the National Cancer
Institute (NCI). NCI has other laboratory space, both on and off
of the NIH campus, where research using similar types of
radioactive material is performed. However, the research
performed off of the NIH campus is performed under other NRC
licenses.
NCI has four divisions. Within NCI's Division of Cancer Treatment
is the Developmental Therapeutics Program. The Developmental
j 5
l
Therapeutics Program has three laboratory groups: the Laboratory
of Molecular Pharmacology: the Laboratory of Biological Chemistry;
and the Laboratory of Medicinal Chemistry. All three laboratory
groups are housed on the fifth floor of Building 37. The pregnant
researcher (Researcher A) and her husband (Researcher B) worked
within the Laboratory of Molecular Pharmacology under the
direction of an authorized user (Authorized User). Researchers A
and B are research fellows and have 2-year appointments at NCI.
Researchers A and B began work under the Authorized User in August
1994. Laboratory notebooks belonging to Researchers A and B
indicate that they spent most of August and September 1994
planning their research. Researcher B's notebook contained a
reminder that he needed to contact the RSB and schedule training.
The Authorized User and Researcher C, a researcher who worked in
the same laboratory with Researchers A and B through August 1994,
both remembered that Researchers A and B spent time initially
getting ready to do research. Researchers A and B told the AIT
that they were able to begin their research because Researcher C
had P-32 in stock that they were able to use. However, Researcher
C told the AIT that he had not used P-32 for several months prior
to August 1994, but did use sulfur-35 (S-35). Furthermore, a
check of radioactive material receipts in the Authorized User's
laboratory as far back as December 1993 indicated that P-32 was
not received until December 14, 1994. Therefore, there is no ,
indication that Researchers A and B began their research using P- j
32 before December 14, 1994. 1
Thirty-seven MBq (1 millicurie) of S-35 was received on December
13, 1993: June 28, 1994: August 31, 1994; and September 9, 1994.
Records for the December 13, 1993, and June 28, 1994, receipts do
not specifically indicate for whom the S-35 was ordered. However,
records for the August 31 and September 9, 1994, receipts indicate
that the S-35 was ordered for Researcher C. Researcher C stated
that he would not have provided radioactive material to
Researchers A and B at that time because they had not received the
required training and he doubted that anyone else would have ;
provided them with material because they were not known.
The laboratory notebooks belonging to Researchers A and B both
indicated that 37 kilobecquerels (kBq) (1 microcurie) of S-35 was
used in an experiment during the third week of September 1994,
followed by a briefing of the Authorized User on the results of
their experiment on September 24th. As previously stated, the NIH
license permits the use of radioactive material by individuals
under the supervision of an authorized user before recei)t of
formal RSB radiation safety training, as long as the autlorized j
user certifies in writing that he or she has provided the
subordinate users with the required radiation safety training.
The RSB does not have a certification of training for Researchers 2
A and B signed by their Authorized User. RSB records indicate i
that Researchers A and B received formal RSB radiation safety
6
)
. . . =- _ .. . - - . . . - - .
_
I
j training on November 29. 1994.
RSB records indicated that Researchers A and B each were first
issued a film badge to monitor whole body radiation exposure and a ,
thermoluminescent dosimeter (TLD) (ring badge') to monitor -
radiation exposures to their extremities in October 1994. The AIT l
i noted that, although Researchers A and B had not been issued
'
dosimetry at the time of their first use of radioactive material
during the third week of September, the NIH license does not
require that either whole body or extremity dosimetry be worn by
l researchers who use low energy beta-emitting isoto)es such as
S-35. Film badges and TLDs for Researchers A and 3 were sent ,
monthly from the RSB directly to the Authorized User for j
distribution to Researchers A and B. The AIT determined that
dosimetry was not distributed to Researchers A and B by the
Authorized User.
The laboratory notebooks belonging to Researchers A and B
indicated the use of 37 to 74 kBq (1 to 2 microcuries) of
phosphorus-33 (P-33) on November 14, 1994. Receipt records for
the Authorized User indicate that 9.3 MBq (250 microcuries) of
P-33 was received on October 31. 1994. The record of the
October 31 receipt indicates that the P-33 was ordered by, and
for. the Authorized User. The AIT noted that P-33 is a beta-
emitting isotope with a maximum energy of 249 kev and that the NIH
7
__. ._ __ _ _. . _ _ _ ___._ ___._-.______._._m .
licens( requires whole body dosimetry to be worn during use of
this isotope. Whole body dosimetry was not worn by Researchers A
and 8 during their use of this isotope because, as previously -
stated, dosimetry was not distributed to Researchers A and B by
the Authorized User.
'
An RSB contractor performed a routine audit on November 17, 1994.
L of the laboratory used by Researchers A and B. The survey report !
l indicated that only S-35 was used in the laboratory at that time.
.
The AIT noted that this information was incorrect because the ;
i
Authorized User received P-33 on October 31 and the laboratory ;
notebooks belonging to Researchers A and B indicated that P-33 was i
i
!
used in an experiment conducted on November 14. The AIT asked the !
contractor supervisor how the surveyor determined what isotopes
were used in the laboratory. The contractor supervisor said that
the surveyor will sometimes look at container labels and
inventories: other times they will ask the persons present what
i.sotopes are used; sometimes order and receipt records are l
queried; and other times the surveyor will get the information
when the laboratory's monthly survey records are reviewed. The
survey also includes an evaluation of whether personnel external
monitoring is adequate and whether personnel are trained. The
surveyor failed to identify that Researchers A and B were not
wearing the required dosimetry or the fact that they were using
radioactive material and had not been officially trained.
Receipt records for the Authorized User indicate that 9.3 MBq (250
microcuries) of P-33 was received on December 15, 1994. The
record of the December 15, 1994 receipt indicates that the P-33
was ordered for Researchers A and B. No documented use of this
material was identified in the laboratory notebooks belonging to
Researcher A and B.
Laboratory notebooks indicated the use of 37 to 74 kBq (1 to 2
microcuries) of P-32 on more than one occasion between
December 14, 1994, and March 1995. Receipt records for the
Authorized User indicate that 27.8 MBc (750 microcuries) of P-32
was received on December 14,1994, anc 9.3 MBq (250 microcuries)
of P-32 was received on both December 22. 1994, and January 24,
1995. Records indicate that all of these P-32 orders were placed
for Researchers A and B. The AIT noted that the NIH license
requires whole body and extremity dosimetry to be worn during use
of more than 18.5 MBq (0.5 millicuries) of P-32. Neither whole
body nor extremity dosimetry was worn by Researchers A and B
during their use of this isotope because, as previously stated,
dosimetry was not distributed to Researchers A and B by the
Authorized User.
,
8
l
i
l
i
-. . - .
. - _ _ - -- .- - - - . . . __. -- - . . . - .
l
'
Receipt records for the Authorized User also indicate that 37 MBq
(1 millicurie) of S-35 was received on March 3,1995, and that
this material was ordered for Researchers A and B. The AIT did i
not identify any documented use of the S-35 received on March 3 in l
Researcher A's or B's laboratory notebook. ;
!
l Receipt records for the Authorized User indicated that 9.3 MBq ,
l
(250 microcuries) of P-33 was received on March 28 and May 17. '
1995, and 18.5 MBq (500 microcuries) of P-33 was received on
June 15. 1995. All of this material was ordered for Researchers A
and B. No documented use of these radioactive materials was I
identified as Researchers A and B both stopped documenting their
research in the laboratory notebooks in mid-March 1995.
A routine semiannual hall and corridor survey conducted by the i
contractor on April 25, 1995, on corridors C and D. indicated no '
beta or gamma. contamination exceeding the licensee's trigger level
of 220 disintegrations per minute (DPM) per 100 square
centimeters.
On May 19, 1995, an RSB contractor performed a routine audit in
the laboratory used by Researchers A and B. The survey report
indicated that P-33. P-32, and S-35/ carbon-14 (C-14) were used
(Researcher A ordered and received 0.04 to 0.19 MBq (1 to 5
microcurie) C-14 standards during this period of time). The AIT
noted that this information was correct based on the Authorized
User's receipt records. However, the surveyor again failed to l
identify that Researchers A and B were not wearing the dosimetry !
that was issued for them.
The Authorized User delegated responsibility for radiation safety
l oversight of the laboratory to Researcher B in March 1995 because
Researchers A and B were the largest users of radioactive material
'
in the Authorized User's laboratory. Before March 1995, another
researcher had responsibility for radiation safety oversight. The
Authorized User stated that Researcher B was responsible for
performing a monthly radiation survey of the laboratory and
submitting the survey results to the RSB. Researcher B stated
that he performed and submitted the monthly survey of the
laboratory to the RSB and that he performed a survey after each
l
experiment.
Monthly survey records submitted by Researcher B on April 5 and
May 5,1995, indicated that less than 37 MBq (1.0 millicurie) of
S-35. C-14. and P-32 were used in the laboratory within the last
month and that smears were evaluated for the presence of P-32.
l
tritium (H-3), and C-14/S-35. Monthly survey records submitted by
Researcher B on June 1, 1995, indicated that less than 37 MBq (1.0
millicurie) of C-14 and P-33 were used in the laboratory within ,
the last month and that smears were evaluated for the presence of l
i P-32. H-3. C-14/S-35, and P-33. Tne Authorized User's receipt 1
-
records for this period of time indicated that P-32 was received l
9
1
. . .. _ . - - . - = _ _ _ -. . -.. _ -_
l
l on January 24, 1995; S-35 was received on March 3, 1995: and P-33
was received on March 28 and May 17, 1995. As previously stated,
the ins ector noted that Researcher A ordered and received 0.04 to
0.19 MB (1 to 5 microcurie) C-14 standards during this period of
time. he AIT concluded that the radioisotope use described in
monthly survey reports submitted by Researcher B for April, May,
l and June 1995, accurately reflected the Authorized User's records
of radioactive material receipt. Actual use could not be
substantiated by the AIT because, as previously stated, the
laboratory notebooks belonging to Researchers A and B containod no
,
documented use of radioactive materials after the middle of arch
1995.
.
Researchers A and B shared laboratory space with other
researchers. P-32 is used in several other laboratories on the
fifth floor of Building 37.
Researchers A and B said that Researcher B had done all the work
involving the handling of radioactive materials since Researcher
A's 3 regnancy was identified. Researcher A said that Researcher B
had ler leave the laboratory when he used radioactive material.
' Researcher A stated that she was still able to be involved in
other aspects of the research. The Authorized User said that when
Researcher A told him she was pregnant, he called RSB for more
information. The Authorized User said that RSB
provided him with information about declaration of
pregnancy and that he explained it to Researchers A
and B. The Authorized User said that he explained
that declaration of pregnancy was voluntary.
Researchers A and B stated to the AIT that they had
not decided whether to declare the pregnancy at the
time that the contamination event occurred. The AIT
determined that the Authorized User received training
on the 10 CFR Part 20 regulations related to the dose
limits established for the embryo / fetus of a " Declared
Pregnant Woman" on January 31, 1994. The AIT also
determined that this information was provided to ,
Researchers A and B during " Radiation Safety in the !
Laboratory" training received on November 29. 1994.
3 CHRONOLOGY OF EVENTS
Based on interviews and record and documentation review, the following
is a chronology of the internal contamination events at NIH:
l
l
10
- - - - - .- -. . .- --
June 24 Researcher B used 0.74 MBq (20 microcuries) of P-33-labeled
nucleotide in an experiment. _
June 25 Researchers A and B did not use radioactive material on this
day, but instead met in the lab with the Authorized User to
discuss their experiment result.
l June 26 Researchers A and B spent the morning doing literature
.
research in the Building 10 library. Researcher B recalled
l that he may have used some radioactive material in the '
I
afternoon, but that Researcher A was not present during the
use. Researcher B performed a survey and found no
contamination. The Authorized User was out of the
laboratory part of the day.
June 27 Researcher B stated that he believes he used radioactive
material on this day. Researchers A and B both ate food
that had been brought to work that day. This food was
consumed at a table located in the corridor outside of their
laboratory (5D18). Tha left over food was returned to a
brown refrigerator located in the SC conference room (Room
SC25) for storage (floor plan provided in Appendix D).
Researchers A and B frequently used the refrigerator before
Researcher A's pregnancy, but had stopped bringing food to
work because Researcher A had no a) petite. Researchers A
and B stated that just that week t1ey had begun bringing
food to work again because they were working long hours and
Researcher A was feeling better. Researchers A and B worked
late and Researcher B performed a survey that found no
contamination. The Authorized User was out of the
laboratory part of the day.
June 28 The Authorized User was out of the laboratory for most of
the day. Researchers A and B arrived at the lab at
10:00 a.m. Researchers A and B said that the Authorized
User arrived at the laboratory at about 1:00 p.m. and that
they left the laboratory shortly thereafter to go to the
library in Building 10. Researchers A and B returned to the
laboratory at 4:30 to 5:00 p.m. Researchers A and B went to
the SC conference room to get the food that they had placed
in the brown refrigerator the The food
was stored in a plastic bowl. previous evening.The bowl was placed
microwave oven in the conference room. Researchers A and B
left the food in the SC conference room because the I
Authorized User was meeting in the conference room with
another individual and they did not want to disturb him.
Researchers A and B said that the Authorized User called i
them after approximately 30 minutes and told them that they l
should come and get the food. The food was retrieved and
'
Researcher A ate the food at a table located in the corridor l
outside of their laboratory (5D18). Researcher A completed !
her meal between 5:30 and 6:00 p.m. Researcher B said that
11 !
_. . - - .- -- --- - -. .
he did not eat the meal. Researchers A and B left the
laboratory and returned to their home at approximately
6:30 p.m. The bowl containing the food was brought home and
the bowl was washed. Researcher B reportedly told the RSO
that, even though he used no radioactive materials that day,
he still conducted a survey that identified no
contamination. During a conversation with the AIT on
July 20, 1995, Researcher B could not recall whether he
performed a radiation survey on this day.
June 29 The Authorized User spent most of the day in his office.
The Authorized User's office is within laboratory SD18 and
adjacent to the laboratory space used by Researchers A and B
(A)pendix D). Researchers A and B both arrived at the
la) oratory at about 8:30 a.m. Researchers A and B did not
use radioactive material but they used equipment which they
knew to have contamination from their earlier uses.
Researchers A and B both worked until-12:00 or 12:30 p.m.,
then ate lunch at the table in the corridor outside of their
lab. Researchers A and B shared lunch and ate mostl., the
same items. Researcher A's experiment would not be
completed until 10:00 p.m. that evening, so they left the
laboratory at about 2-3:00 p.m. to go home for a rest and a
shower.
Researchers A and B returned to the laboratory at about
4.:30-5:00 p.m. Researcher B had finished his experiment
earlier but, upon return to the laboratory, he performed a
meter survey to check for possible contamination. The meter
was reportedly equip)ed with an end-window Geiger-Mueller
(GM) probe. Researcier B noticed some " signal." which he
identified as coming from Researcher A. Researcher B said
that the signal was initially very weak
and he could have missed it in earlier
surveys.
Researchers A and B tried to find the Authorized User so
that they could inform him of the problem, but could not
locate him. Researcher B said that he questioned the
accuracy of his meter and located a second meter that
confirmed the contamination. Researcher B said that he
called the NIH emergency number (116) and asked for ,
assistance. '
Researchers A and B said that the Authorized User returned
to the laboratory at approximately 5:30 p.m and Researcher B :
told him that Researcher A had " injected" radioactive
material. Later, Researcher B said that he meant that
Researcher A had " ingested" radioactive material. The
Authorized User recalled that he was in his office at l
approximately 5:45 p.m. when Researcher B notified him. The
Authorized User stated that Researcher B demonstrated that
12 ;
i
l
__. . . _
_ _. . . _ _ _ _ - _ _ - _ _-
1
Researcher A had significant contamination. The Authorized
, User said that the ambularice arrived a short time later and
that he contacted the RSB office and notified the Chief of
'
the Radiation Safety Operations Section (RS05) about the
contamination incident. The Authorized User believed the
time to be approximately 6:00 p.m.
.
The NIH Fire Department independently notified the Deputy
"
RSO of a possible radioactive material contamination at 5:55
p.m. The Deputy RSO is at the top of the emergency call
list for response to incidents involving radioactive
materials. An injection of radioactive material was
reported. The Deputy RSO advised the RSO of the report at
, approximately 6:00 p.m. and contacted the NIH Occupational
-
Medical Service (OMS) for information on the incident.
i The Authorized User said that Researcher B told him there
was contamination in the SC conference room refrigerator.
The Authorized User said that he went to the conference room
with a survey meter and identified two bags in the
refrigerator that he believed to be contaminated. The
Authorized User was accompanied at this time by Researchers
E and F.
At 6:15 p.m., a RSB health physicist (Health Physicist A).
l was notified by the RSB receptionist that a second health
)hysicist (Health Physicist B) was on the phone with the
ISOS Chief talking about a possible contamination event in
i Building 37. Health Physicists A and B stated that they
picked up a spill kit and a skin decontamination kit and
4
responded. Health Physicists A and B reportedly met the
Deputy RSO in the RSB (Building 21) parking lot and were
informed that the contaminated researcher (Researcher A) was
being transported to OMS in Building 10. Health Physicists
A and B responded directly to OMS and were advised by the
physician on duty that the contaminated researcher was still
,
in Building 37. Health Physicists A and B responded to the
fifth floor of Building 37, arriving at approximately
6:40 p.m.
Health Physicists A and B interviewed Researcher A.
- determined that she was 4 months pregnant, and verified that
she had not had a recent medical 3rocedure that would
explain the contamination. Healt1 Physicist A performed an
initial survey with a thin end-window sodium iodide crystal
(Nal) probe to verify that there was a contamination
problem. Health Physicist B surveyed the laboratory used by
Researchers A and B. the Authorized User's office, the
corridor outside the laboratory, and all the other
individuals who were present. Health Physicist B used a
3ancake GM (efficiency 24%) and found no contamination.
lealth Physicist A reported that Researchers A and B and the
13
._ _ . _ _ _ __ _ _ __ _ _ _ _ _ _ _ . _ _ _ _ _ _
!
l
'
Authorized User told her that they thought Researcher A had
eaten contaminated food because the refrigerator in the SC
conference room was contaminated. The Authorized User told
Health Physicist A that he had surveyed the refrigerator and
found it to be contaminated.
Health Physicist B went to the conference room and performed
l a survey of the room, the refrigerator, and the contents of
! the refrigerator with a pancake GM. Health Physicist B
l determined that the refrigerator and the food were not
contaminated, but that the floor in front of the
refrigerator had a spot of contamination measuring
approximately 250.000 to 300.000 counts per minute (CPM) as
measured with a pancake GM. Health Physicist B surveyed the
entire conference room and the office leading into the
conference room and took smears of the refrigerator and the
floor for later evaluation.
Health Physicist A performed smears of Researcher A's hands,
face, and neck to determine if any removable contamination
was present. None of the smears measured above background
when held against the NaI probe. The smears were counted ir,
a laboratory liquid scintillation counter in Building 37,
and later at the RSB laboratory, and confirmed no removable
contamination.
Health Physicist A had Researcher A change into a clean 3 air
of surgical scrubs and resurveyed her using both a pancace
GM (30% efficiency for P-32) and the thin end-window NaI
probe (23% efficient for P-32. as determined using a
strontium-90 (Sr-90)/ yttrium-90 (Y-90) source). Health
Physicist A reported the following: head and hair 2500 CPM:
chest 3000 CPM: arms and hands 2000 CPM: front waist
6000 CPM: legs and knees 2000 CPM: and feet 1000 CPM. Health
Physicist A reported that Researcher A's clothes were not
contaminated.
Health Physicist A spoke by telephone with the RSO and
Deputy RSO and, upon their advice, asked Researcher A to
provide a urine sample. Health Physicist A believed that
the time was approximately 7:00 p.m. The urine sample was
provided and Health Physicist A reported that she measured
a) proximately 1500 CPM at about 1 centimeter (0.4 inches)
a]ove the surface of the urine sample with the NaI probe.
The paramedics, who had arrived at approximately 6:00 p.m..
prepared Researcher A for transport to the hospital.
Health Physicist A asked Researcher B when and where he and
Researcher A had eaten. Researcher B reported that he and
his wife had eaten around noon that day at a table in the
corridor outside of their laboratory (5D18) and that trash
was placed in the garbage can located in the corridor.
14
_- . . _ . _ _ _ _ _ _ . - _ _ _ _ _ _ _ . _ __._ ._ _ _._.. __
'
i
,
'
'
.
(
r
Researcher B reported that bowls they.had eaten out of were
still in the refrigerator in the SC conference room and that
.
!
other bowls were at home. Health Physicist A asked
Researcher B to provide the bowls he had at home for survey
the.following day. The initial estimated time of ingestion
.
'
was placed at 12:00pm on June 29th.
Researchers A and B reportedly told Health Physicists A and
B that they brought all of their own food and drink to
Building 37. !
Paramedics transported Researcher A to Holy Cross Hospital, t
Holy Cross was selected over Suburban Hospital which was- ,
much closer, because Suburban did not have an obstetrics ,
,
department. Health Physicist A surveyed the paramedics !
!
before they left the floor and determined that they were not
contaminated. The paramedics were instructed to tell ,
personnel at the hospital emergency room to save Researcher
j. ' A's urine for analysis by NIH. ;
'
, The Authorized User stated that Researcher B was upset that
l they had not taken Researcher A to the hospital earlier.
The Authorized User said that he too was upset that things
were not moving faster. The Authorized User stated that he
thought it was after 8:00 p.m. when the paramedics left to
take Researcher A to the hospital.
Health Physicist A conducted a survey of Researcher A's
desk, the table where Researchers'A and B ate, and the trash I
cans in the corridor, using the thin end-window Nal probe
and found no contamination.
Health Physicist B surveyed the C and D corridors and the
east hallway connecting the two corridors and found no
contamination. Housekeeping was advised not to clean the
floors in these areas until further notice. Health
Physicist B reported that he surveyed all trash containers
and tables in corridors C and D and in the connecting
15
i
, , , I
.. _ ____ _ _._._. _ _.._ ._ _ _ _ _ _ _ _ . - ._
'
i
- -
!
! hallway and found no contamination. Health Physicist B
j
'
re)orted that he specifically monitored the table outside of
la) oratory where Researcher A ate her meals and found no
contamination. *
- ;
- Health Physicist A left Building 37 at approximately
8:30 p.m.. taking the urine sample back to the RSB lab for
.
analysis. . Health Physicist B taped a piece of cardboard
over the contamination found on the floor of the SC ,
conference room, locked the door to the conference room, and
posted'a warning. Health Physicist B said that he covered
the contamination with cardboard he tore off of an empty
computer box that he found in the conference room. Health
Physicist B reportedly resurveyed the area outside of the
conference room door and identified no contamination.
Health Physicist B stated that he resurveyed the hall and
corridors and told housekeeping that they could clean if
they wished. Health Physicist B left Building 37 at
9:00 p.m.
Researder A arrived at Holy Cross Hospital's emergency
room. The attending abysician was concerned that treatment
would interfere with Researcher A's pregnancy and elected to
treat Researcher A by simple hydration. NIH contacted
REAC/TS and had the REAC/TS physician speak directly with
the attending physician. The REAC/TS pnysician stated that
he discussed with the attending physician the possibility of
administering sodium phosphate to inhibit phosphate
absorption from the gastrointestinal tract, but recommended
against this because nine hours had passed since the
estimated time of ingestion and little would be gained from
this treatment. The REAC/TS physician advised the attending
physician of the need to collect 24-hour urine sam)les for
determination of Researcher A's radiation dose. T1e
attending physician does not recall the s)ecifics of his
telephone conversation with the REAC/TS plysician. The
REAC/TS physician faxed Jortions of National Committee for
Radiation Protection (NCRP) publication No. 65, relevant to
the medical management of P-32 contamination, to the
attending-)hysician, but the transmission reportedly was
poor and tie document not readable.
At NIH. Health Physicist A prepared the initial urine sample
and laboratory smears for analysis. Sam)les were run and
the results were provided to the RSO. T1e initial urine
sample taken at 7:00 p.m. measured 16.000 disintegrations
per minute of P-32 Jer milliliter (DPM/ml). The only smear
measuring above baccground was taken from the spot on the
conference room carpet in front of the refrigerator. The
conference room carpet contamination was determined to be
P-32. The RSO left for Holy Cross Hospital.
16
!
1
_ - . _ . _ . _ _. _ .. _ _ _- _ __._ _ _ _ . __ _ _ . _ . _ . _ _ _ . . . _ .
I
l^
Holy Cross collected additional urine samples and a blood- ,
sample. The attending physician recalled that he was given
instruction on the collection of Researcher A's urine
initially from the RSO and later from the Authorized User.
The attending physician believed that the RSO irstructed him
to collect all of Researcher A's arine, although the
Authorized User instructed him to collect only a portion of
each void for analysis along with the total volume of each
void. The attending physician'did not believe that the two
instructions were substantially different, but he elected to
save a sample of each void and placed the remainder of each !
sample into a single " pooled urine" container. 1
The Authorized User called the De)uty RSO at approximately
10:00 p.m. and informed him that le had found a coffee cup
t on the table in the corridor where Researchers A and B had
l reported eating. The cu) contained a 50-cubic-centimeter
j (cc) centrifuge tube wit 1 an orange cap, and the items were
contaminated. Health Physicist A said that she did not
remember seeing these items when she surveyed the area
a) proximately 90 minutes earlier. Health Physicist B said
! tlat these items were not present when he surveyed the table
l a short time earlier. The Authorized User was directed to
store these items in the laboratory and lock the door.
'
The RSO reported that at approximately 10:00 p.m. he
transported Researcher B back to NIH. The RSO also
trans orted a blood sample and two urine samples collected
l
at Ho y Cross Hospital
Blood and urine assays were completed at about 11:00 p.m.
The blood sample measured approximately 4400 DPM of P-32 per
ml of blood. Two urine samples 3rovided less than one hour
apart measured 947 DPM and 3496 JP_M of P-32 per ml.
l June 30 Health Physicist A was sent to the fifth floor of
l Building 37 to make sure everything was secure arriving at
approximately 8:30 a.m. Because the contamine a of
Researcher A may have been the result of a mat .ous act.
the RSO notified the NIH Police Department.
An NRC inspector who was onsite conducting a routine
inspection was advised of the incident u)on arrival at
Building 21 at approximately 8:30 a.m. Early estimates of
P-32 intake were placed at approximately 11.1 MBq (300
microcuries), or 50% of the Annual Limit of Intake (ALI).
i The estimate was based on the licensee's analysis of spot
l urine samples. The licensee stated that 24-hour urine
samples would be collected from Researcher A and "whole body
counts" performed to check clearance of P-32 from the body.
An NRC medical consultant was contacted to get an early
( 17
1
l
l
l
l . . _ .- .- .
- - - - . . - - - . - .. -~ - - ~ .. - - ~ . - . -
f
!
i
assessment of the adequacy of actions taken. The medical
consultant agreed that hydration was appropriate treatment
and suggested that a complete hematological profile be
performed that several nuclear medicine scans be performed
and, unless contraindicated, the woman be offered phosphate
for oral stabilization. As discussed above. this treatment I
was not administered to Researcher A. The medical
consultant also recommended that a stool sample be
collected, to aid in identifying when the intake occurred. 1
'
The medical consultant stated that he did not believe that
there would be any health consequences to the woman or to ;
,
her fetus from this ingestion.
,
! The information 3rovided by the NRC's medical consultant was 1
discussed with t1e RSO and Deputy RSO. The licensee agreed
to perform a complete hematological profile and aerform a
scan in nuclear medicine. The Deputy RSO said tlat he did
.not have the ability to evaluate a stool sample..and that I
NIH's consultant at RIDIC recommended that NIH focus on l
collection of 24-hour urine samples for dose evaluation. ,
The RSO said that he intended to perform a whole body scan '
l as soon as Researcher A arrived at the RSB office.
,
AIT contacted an NRC scientific consultant at RIDIC. NRC's I
scientific consultant was also the licensee's consultant. i
The NRC scientific consultant recommended that the licensee '
collect and analyze several 24-hour urine samples, in order i
to more accurately estimate the aregnant researcher's intake ;
of P-32. perform periodic whole Jody counts to assess the i
clearance rate of the P-32 from the pregnant researcher's
body, and obtain nuclear medicine images of the
bremsstrahlung radiation from the P-32 to visualize the
distribution of the contaminate.
Health Physicist A resurveyed the office floor that leads
into the SC conference room and found three spots of
contamination ranging from 150 CPM to 3000 CPM, with the
highest count rate at the conference room door and the
lowest count rate near the door leading into the SC
corridor. An NIH police officer arrived to secure the scene '
i
at approximately 9:30 a.m. An NIH detective arrived at
ap3roximately 10:15 a.m. Health Physicist A surveyed the
la) oratory used by Researchers A and B for contamination
before the detective entered the room, reporting that the
lab was clear. Health Physicist A then surveyed the j
conference room and identified six more spots of floor i
contamination. Health Physicist A called the RSB and asked ;
for assistance and supplies to cover the floor contamination 1
before the detective examined the conference room. Health 4
Physicist C responded to Health Physicist A's request for
assistance and papered the floor.
I
18 l
l
l
l
l
i
Health Physicist A examined the 50cc centrifuge tube found
by the authorized User the previous evening -and noted that
it contained a small amount of clear liquid. The NIH Police
Detective allowed Health Physicist A to remove the liquid
from the centrifuge tube. Health Physicist A noted that the
centrifuge tube contained approximately 1 ml of liquid. The
liquid was analyzed by liquid scintillation counting and
identified as containing 3000 Bq (80 nanocuries) of P-32 per
ml.
. Health Physicist C reported that she surveyed and
inventoried all the items in the refrigerators located in
the SC conference room and found no contamination. Health
Physicist C also took smears of the refrigerators and the
contents of the refrigerators and reported finding no
contamination. Health Physicist C surveyed the interior of
the microwave and found no contamination. Health Physicist
C found a brown paper bag, with significant contamination
(200.000 CPM). on to
the conference room.pTheof an oldwas
bag computer box
lying on its in theand
side back of
open
and ap) eared to have the imprint of a round bowl creased
into tie paper. Health Physicist B stated that the paper
bag was not present when he left the conference room at
9:00 p.m. on the previous evening, as he had removed
cardboard from the computer box and used it to cover the
c.ontamination in front of the refrigerator.
A total of 12 spots of contamination were found in the
conference room (200 to 10.000 CPM). not including the spot
found in front of the refrigerator. Health Physicist A
measured the spot in front of the refrigerator and found
greater than 500.000 CPM through the cardboard placed over
the spot by Health Physicist B the previous evening.
Researchers A and B arrived at the RSB office and Researcher
A provided additional urine samples. At 11:15 a.m. Health
Physicists D and E surveyed Researcher A with a pancake GM
(26% efficiency for P-32) and noted the following: 3 alm
right hand 12.000 CPM: back of right hand 10.000 cpi; left
palm 10.000 CPM: back of left hand 6000 CPM: right foot 350
CPM: left foot 250 CPM: front of body at chest 6000 CPM:
back of body 500 CPM: hair at forehead 10.000 CPM: hair at
sides 8000 CPM: and hair on back of head 5000 CPM. Health
Physicists D and E also took wet and dry smears of
Researcher A's hands and completed a whole body count using
the licensee's Canberra whole body counter. The smears were
negative for contamination. Researcher A was taken to NIH's
OMS where blood was drawn for a complete hematological
profile. Researcher A and B were provided with several
plastic containers and instructed to collect 24-hour samples
of Researcher A's urine.
19
_ . _ _ _ . _ . _ _ _ _ . . _ . . _ _ _ _ . _ . . _ . _ . _ _ _ _ _ _ _ _ _ _ _
.i
l .
I
Health Physicist D surveyed the car driven by Researchers A
and B. Researcher A reportedly vomited in the passenger i
. seat of the car when she returned from the hos) ital at i
approximately 4:30 a.m. that morning. Health ?hysicist D l
noted contamination inside of the passenger's side of the i
car of between 200 and 1200 CPM. The car was ;
decontaminated. The contaminated floor mat and a box of
~
crackers had to be taken for storage in the RSB Office.
l The RSO called an emergency meeting of the NIH Radiation i
l Safety Committee.
l NIH Police directed that the SC conference room door lock be l
l
changed. The contaminated paper bag found by Health ;
Physicist C in the SC conference room was turned over to the ;
l NIH police. !
Health Physicists F and G surveyed the a)artment belonging
to Researchers A and B using a pancake Gi with an efficiency
of 26% for P-32. Health Physicists F and G found fixed
contamination in one 3-inch by 5-inch s)ot on the concrete
<
- patio (approximately 40.000 to 50.000 C)M) in front of a
green outdoor carpet, and contamination (100.000 to 120.000
CPM) in one area of the green outdoor carpet. This
contamination reportedly occurred on Thursday morning.
June 29, when Researcher A vomited on the patio (this
indicates that Researcher A's ingestion of P-32 occurred
before the morning of June 29). An attempt to remove the
contamination from the car)et was made, but the carpet was
eventually removed to the Radiation Safety Office for
storage.
A plastic bucket containing peanut shells and other trash
items was found to be contaminated. Researcher A said that
she had vomited into the bucket. The bucket and its
contents were stored for decay. j
A basket containing the couple's laundry was surveyed. Four I
pair of underwear (500 to 5000 CPM), a pair of pants (500 to l
1000 CPM), and a shirt (minimal contamination), all )
belonging to. Researcher A, were found to be contaminated ;
Researcher A reportedly wore the pants and the shirt on both !
June 27 and June 28. Researcher A reported that the four i
pair of underwear were worn on June 28 and June 29.
Contamination was not noted on other clothes in the basket
belonging to Researchers A and B. or on clothes hanging in
both the bedroom and office closets. Two hand towels found
in the bathroom were found to be slightly contaminated (150
to 200 CPM). Contamination was detected inside of the
,
toilet bowl (150 to 1000 CPM). Minor contamination (150 to
! 200 CPM) was found on the li) of the bathroom sink. No
contamination was found in t1e kitchen, on any of the
20
___ _ _
_ _. _ ___ __ __ - _
_ . _ _ _ _ . _ ~ _
appliances, or in food located in the couple's
refrigerator / freezer. Utensils and kitchenware, including
the bowls that contained food consumed by Researchers A and
B on. June 28 and 29, were clean and exhibited no sign of
contamination.
At 5:00 a.m. Health Physicist F took Researcher A to NIH's
Nuclear iedicine Department for a scan utilizing one of the
Department's gamma cameras. A urine sample produced by
Researcher A immediately after the scan was retained by the
Nuclear Medicine physician in order to attempt to quantify
the contamination seen on the scan. The physician commented
that the scan appeared to show a higher concentration of
radioactive material in the liver and-spleen, meaning the
blood, and not in the bone masses.
The licensee communicated the results of urine and blood
analyses performed thus far to RIDIC and requested that
RIDIC estimate a dose to Researcher A and her fetus. The
licensee estimated an ingestion of 9.6 MBq (260 microcuries)
of P-32 using NUREG/CR-4884 (" Interpretation of Bioassay
Measurements").
The RSO decided to perform urine bioassays on anyone who had
access to the contaminated conference room.
NIH police expressed concern that RSB and AIT inspection
activities would compromise the criminal investigation, and
therefore would not permit interviews by RSB or AIT of
Researchers A and B and the Authorized User.
July 3 A urine sample taken from Researcher B was found to contain
approximately 100 DPM of P-32 per ml of urine.
An additional researcher was found to have approximately
16 DPM of P-32 per ml in his urine. This researcher was
bioassayed as part of the RS0's program, initiated on
June 30. to screen anyone who had access to the contaminated
conference room.
The RSB reported that Researcher A provided a container of
urine measuring more than 7200 milliliters. Because of the
large volume of urine produced, the RSB believed that the
sample was collected over a two day period.
An inspector with the State of Maryland Department of the
Environment was onsite to review NIH efforts to
decontaminate the apartment and car belonging to Researchers
A and B.
NIH stated that the contamination event occurred on June 28.
based on the contamination found on Researcher A's clothes,
21
. -- . -. ~ . - - . - . - . . - . - . - - - - . - . . - , - -.
I
!
I
and arbitrarily placed the time of the event at 11:00 a.m. ,
because Researchers A and B normally ate their lunch at this !
time.
July 5 RIDIC reported that, based on urine data through June 30th.
the P-32 intake was a) proximately 9.8 MBq. (265 microcuries).
RIDIC requested that ilH continue with urine collection to
refine the intake estimate.
July 6 The RSB reported that Researcher A )rovided a second
container of urine measuring more tlan 7200 milliliters.
'
The RSB said that they determined from interviews of !
Researchers A and B that this sam)le was collected over a ;
,
24-hour aeriod on July 3 and 4. tlat the urine was produced
'
only by Researcher A. and that no water or other liquid was
added to fill the container. The RSB also learned that the
greater than 7200 milliliter urine sample provided to the
RSB on July 3 was a 24-hour sample collected on June 30 and ;
July 1.
'
A second whole body count was performed by RSB. The
licensee reported that RSB counted for ten minutes with a ,
window setting from 10 kev to 2000 kev. and measured the
bremsstrahlung peak height at approximately 850 counts. The ;
previous peak height measured on the morning of June 30th !
was approximately 1700 counts. No further quantification ;
was made.
A second Nuclear Medicine scan was performed. The physician
noted that the activity was no longer concentrated in
Researcher A's body in the same configuration as was seen on
the previous scan. The Nuclear Medicine physician also :
performed a whole body count using a NaI detector.
An inspector with the State of Maryland Department of the :
Environment conducted surveys of Researcher A's and B's
apartment and confirmed the survey results previously
reported by NIH.
The RSO reported that they had identified two additional
researchers from the fifth floor of Building 37 who had
Jositive urine bioassay results. One bioassay result was 30 '
l
)PM per ml of urine, and the second result was 180 DPM per
ml of urine. The RSO scheduled follow up bioassays for both
individuals. One of the individuals reportedly only used
tritium, the second individual used 0.4 to 0.7 MBc (10 to 20
microcuries) of P-32 every few weeks. but reportec that he
had not used P-32 in over 2 weeks.
While checking P-32 inventories on the fifth floor of
Building 37, the AIT found an unsecured laboratory (5D12)
with an unlocked refrigerator containing approximately
22
- .- -. .-
740 MBq (20 millicuries) of hydrogen-3 and 92 MBq (2.5
millicuries) of carbon-14. The RSB confiscated all of the
radioactive material.
July 7 The licensee ex3ected Researcher A to provide additional
urine samples, lowever, no samples were delivered.
July 10 The licensee began urine bioassays of everyone on the fifth
floor of Building 37.
The RSB reported that a reanalysis of the liquid 3 resent in
the centrifuge tube showed P-32 and some P-33. T1e RSB
reanalyzed urine samples received from Researcher A and
determined that only P-32 was present in her urine.
Researcher A provided no urine samples to the RSB.
The Director of the Developmental Therapeutics Program l
instituted new requirements: that all requests to use
radioactive material in the Program be made directly to him,
and that he would control issuance of the material. The
Director also assigned responsibility for surveying all
restricted and unrestricted areas of the fifth floor to
specific researchers.
l
,
1
l
!
l
l
I
!
j
.
l
l
23
_. . _ _ _ . _ - - _- . . _ _ _ _ _ __ _ _ _ . _
l
l
l July 11 RSB reported no contact with Researchers A and B. The
licensee was uncertain whether more urine samples would be
received or whether the opportunity to perform additional
whole body scans would be provided.
No additional positive urine bioassays of other researchers
were identified. The RSB completed bioassays on 26 persons
i thus far.
1
The RSO stated that they would complete surveys of all fifth
floor labs by close of business. ;
July 12 The licensee reported that Researcher A provided two more
24-hour urine samples for analysis.
RIDIC stated to the licensee that it would not require i
collection of all of Researcher A's urine to get the !
excretion data needed to do the dose assessment--just a few
more 24-hour samples taken next week, a few days a) art.
RIDIC was also interested in additional Canberra w1 ole l
bodyscans and "quantification" of the scan information to ;
get some retention data. RIDIC advised that, even though
only a rough quantification of the whole body scan
information was possible, this quantification would provide
" guideposts" which would allow RIDIC to have more confidence
in its assessment.
July 13 The third whole body count was performed. The i
bremsstrahlung peak height was now 425 counts.
The physician who supervised the Nuclear Medicine scans
estimated that Researcher A had a P-32 intake of
approximately 29.6 MBq (800 microcuries).
The RSB reported that they would analyze two more 24-hour
urines and perform an additional whole body count.
July 14 Surveys of all rooms on the fifth floor were almost
complete. NIH said that only a few rooms remained to be
checked and no contamination had been found
At midday, the RSB reported that it continued screenings of
the approximately 120 researchers who work on the fifth I
floor of Building 37. Samples from 82 individuals had been
processed, samples from 17 individuals that had been
received the prior evening currently were being processed,
and RSB was picking up the samples from the remaining 20 or
so researchers. NIH reported identifying P-32 contamination
in three additional researchers. This brought the total to
eight individuals, including Researchers A and B. One
researcher had 140 DPM per ml of urine: the other two
researchers had less than 100 DPM per ml in urine. NIH
24
~ _ __ _ __ .__ _
tried to identify commonalities among the individuals with
positive bioassays.
Because of the identification of several additional positive
, urine bioassays in the urine sam)les processed by the
!
licensee the previous evening, t1e RSB surveyed the water
l coolers and coffee stations on the fifth floor of Building
37 and identified contamination in a water cooler located in
the west hallway, between the C and D corridors. Earlier
surveys by the RSB reportedly did not look at this area
because the east hallway supported the foot traffic between
l the conference room in the C corridor and the lab where
i
Researchers A and B worked on the D corridor and surveys
]erformed by researchers did not include the hallways. The
158 reported to the AIT at approximately 3:15 p.m. that the
water cooler spigot was contaminated, and that water inside
of the drip pan contained approximately 300 DPM per ml of
- P-32 contamination (see Appendix E). The RSO said that
'
because the water from the dri) pan was not metabolized NIH
intended to try and identify tie chemical species by thin-
layer chromatography and mass spectrographic analysis (see .
Appendix E). The RSB reported that the water bottle was l
removed and the water inside of the bottle had no detectable :
activity. The RSB also reported that the water present in
the reservoir exhibited no contamination. The RSB stated !
that when the water was drained from the reservoir, 60,000
CPM of contamination located on the interior surface of the ,
reservoir was measured with a pancake GM.
The licensee determined that a delivery of water bottles was
made to Building 37 on July 13th. The water company
reported that some em)ty bottles were stored at its Jessup,
Maryland, facility, w1ereas others had gone to its Lorton,
Virginia, bottling plant. At the Lorton facility, where
52.000 bottles are processed daily, the bottles are washed
in hot soapy water, rinsed, and refilled with potable water.
The water company agreed to hold the bottles for analysis.
Urine screening was requested of anyone who drank from the
I water cooler. Rescreening of urine samples for individuals
! previously identified as being free of P-32 was begun.
25
. . - -. --. . - - - - - - . - - - ..- - - - - - - --
,
The RSB also reported finding contaminated coffee cups, an
individual-size water bottle, a microwave oven, and coffee
.
!
pots, all of which were confiscated by the RSB. The
contaminant was identified as P-32.
The FBI assumed the lead in the investigation. ,
July 15 Samples of wash water from the Lorton facility as well as
samples of water from the refilled water bottles were
analyzed by NIH. No contamination was identified in any
l sample.
i
July 17 The RSO identified 24 persons other than Researchers A and B
who had P-32 contamination in their urine (an additional
individual was identified on July 19). The RSO stated that i
those individuals having 100 DPM per ml of urine or more i
would be followed with 24-hour urine samples, and that those ;
! individuals having less than 100 DPM per ml of urine would
be spot-tested. The RSO reported that the maximum
contamination was 229 DPM per ml of urine, with six persons !
total having contamination greater than 100 DPM per ml of i
urine. For dose assessment purposes, the RSB estimated that
l
I
the intakes occurred on June 28. 1995 (at the same time that
it is believed Researcher A received her contamination). !
resulting in a maximum intake estimate of 1.4 MBq (37
~
, microcuries) or 5 percent of the ALI. The RSO reported to
! the AIT that the average intake of P-32 was 0.3 MBq (7.7
microcuries). The RSB reported that most of the affected
individuals were researchers, one was a housekeeping
contractor, and one was an adult summer student. Two
children who visited the area with a parent were tested and
no contamination was detected in their urine. Other
visitors were also tested and found to have no P-32
contamination.
The RSO and NIH management met with the research staff from
the fifth floor of Building 37 to discuss the contamination
of Researcher A and the subsequent contaminations of
additional individuals from the contaminated water cooler.
The research staff were asked to bring their own drinks to
work rather then rely on the facility's refreshments. Staff
were also told to maintain a high level of surveillance.
Those who received an intake of P-32 were advised of the
dose consequences in writing. All staff were asked to
complete questionnaires (Appendix F) to assist the RSB in
identifying when the contamination of the water cooler
occurred and to ensure that all of the contamination came
from the affected cooler. A list of telephone contacts was
provided to each researcher as well as a copy of NRC
Regulatory Guide 8.29 (" Instruction Concerning Risks from
Occupational Radiation Exposure"). Urine screenings were
made available to anyone who requested a screening. The RSB
l 26
-. .
l
__ . . . _ . _ _ _ _ ._ ._ ._ _._ _ _ _ _ _ . . _ . _ . . _ _ . _ . _ _ _ .
reported processing a total of more than 200 urine samples.
RSB staff performed surveys at the water cooler company's
Jessup facility. No contamination was identified. State of l
Maryland inspectors conducted a confirmatory survey at the
Jessup facility.
Two individuals who previously had no contamination in their i
urine when they were tested on July 12 were found to have
minor P-32 contamination in their urine (30 DPM per ml) when I
they were tested on July 14th. ,
l
RSB staff surveyed all other water coolers in Building 37
and reported finding no contamination. Questionnaires '
i indicated that all contaminated individuals drank water from ;
l
'
the contaminated water cooler. However, not all individuals
who drank water from the contaminated cooler had measurable
l levels of P-32 in their urine. Although the information
! provided on the questionnaires did not point to a specific
time that the water cooler was contaminated, the AIT noted
, that urine sam)les 3rovided as early as July 3, by
l individuals otler tlan Researcher A, exhibited internal P-32
t contamination.
The NIH survey contractor performed a complete resurvey of
all areas or the fifth floor of Building 37. No additional ,
unexpected contamination was identified. l
4 DOSE ASSESSMENT l
NIH performed an assessment of the intake of P-32 by Researcher A, the
resultant radiation exposure received by Researcher A, and the radiation '
exposure received by Researcher A's fetus. These assessments were I
initially performed in collaboration with the Radiation Internal Dose
Information Center (RIDIC) at the Oak Ridge Institute for Science and
Education (0 RISE). The licensee completed the assessments without the
assistance of RIDIC. The licensee also performed assessments of the
P-32 intakes and resultant radiation exposures received by the 26
additional individuals. These assessments were performed without the
assistance of RIDIC.
RIDIC, serving as a scientific consultant to the AIT and using bioassay
data provided by NIH, performed an assessment of the intake of P-32 by
Researchar A, the resultant radiation exposure received by Researcher A.
,
27
l
!
,
, ,4\ ,- -r -- - ,
. __ _ _ - ___ _ . - _ _ _ _ _ _ _ _ _ _ . . _ _ . _ . . ___ _
,
i
l '
1 !
'
and the radiation exposure received by Researcher A's fetus. RIDIC also l
performed assessments of the P-32 intakes and resultant radiation !
exposures received by the 26 additional individuals. l
Because of the differences in the results of the assessments performed
by the licensee and the AIT's scientific consultant. NRC contracted with
a third party. Lawrence Livermore National Laboratory (LLNL)
(independent consultant), to independently review the assessments
performed by the licensee and the AIT's scientific consultant.
l 4.1 Researcher A Intake Estimates
The AIT determined that 39.6 MBq (1070 microcuries) is the best
,
estimate of Researcher A's P-32 ingestion. This value is based on
i the best available bioassay information and, as recommended by
! NRC's independent consultant, is the average of the best
individual estimates obtained by the scientific consultant and the
independent consultant. These values-range from 30.3 to 48.1 MBq
(820 to 1300 microcuries).
The AIT did not use the NIH assigned estimate as summarized below
and discussed in detail in Appendix G. Appendix G also describes
the raw data, data processing, data analysis methodologies,
results and conclusions. (Note: Althougn NIH staff and the AIT's
scientific consultant were in communication with each other. NIH
elected to perform its own intake estimates.)
,
NIH assigned a 18.5 MBq (500 microcurie) intake to Researcher A.
The AIT scientific consultant's best estimate of intake was
30.3 MBq (820 microcuries). Both of these estimates were based on
NIH's urine bioassay data. The independent third party arrived at
intake estimates of 40.7 MBq (1100 microcuries) from the urine
data set. 48.1 MBq (1300 microcuries) from the Department of
Nuclear Medicine quantified bremsstrahlung scan performed June 30.
1995, and 38.8 MBq (1050 microcuries) from the combined Department
of Nuclear Medicine bremsstrahlung scan and whole body count data ;
set. The ICRP-30 metabolic model for inorganic P-32 was used by
all three groups in making these estimates.
All estimates were based upon specific components of the following
available bioassay data for Researcher A: 25 urine specimens ;
collected between June 29 and July 27: 2 blood specimens collected i
June 29 and June 30: 2 quantified nuclear medicine bremsstrahlung
images performed June 30 and July 6: 1 sodium iodide detector
whole body count performed by the Nuclear Medicine staff July 6: i
and 4 lithium-germanium detector body counts performed by the
Radiation Safety Branch staff June 30. July 6.13. and 20.
!
l
28
.- -.
_ ... _ __ ____ _ _ _ _ _. _ _ _ _ _ . _ _ _ _ .
i
l
'
f In summary, the difference between the NIH estimate and the other
'
two for the urine data is due to fundamental differences in '
interpretation and use of the raw data. The AIT scientific
consultant (RIDIC) and the third-party independent consultant . ;
(LLNL) both elected to exclude the first few days.of urine data.
These data were excluded because they did not cover true 24 hour2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br />
urine collections. The uncertainty associated with Researcher A's
i total urine output and P-32 output in the specimens collected for
l this period was too high to calculate meaningful 24 hour2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> values.
,
The assumation of a constant concentration for the first two data
! points, w1en the two rapid excretion components would have been
important, would result in an underestimated intake. . Further, the
errors associated with'these calculated values would significantly. ,
affect later data analysis since the evaluation technique NIH used ;
to compare the urine data to the ICRP-30 model puts higher weight i
on the early data points. Each group used essentially the same t
numerical values for the bioassay data points after day-4. +
The AIT concurs with the conclusions of both the AIT scientific
consultant and the third-) arty independent consultant that the NIH
approach underestimated t1e expected excretion at longer times and ,
does not provide as good a fit to the bioassay data.
~
The differences between the estimates of the third-party
independent consultant and the others are based on additional i
bioassay data not used by either NIH or the scientific consultant i
and the methods used to evaluate the bioassay data. The i
differences due to evaluation methods can be seen in the urine
bioassay results obtained
(820 microcuries) by the) arty
and the third scientific consultant
40.7 MBq 30.3 MBq ).
(1100 microcuries l
Although these values appear to )e numerically different, they are i
within the range of acceptable values once the uncertainties
associated with urine collection, measurements data processing, j
and data evaluation are considered.
The AIT concurs with the third-party independent consultant's '
,
conclusion that the best estimate. 39.6 MBq (1070 microcuries), is
obtained by averaging the four best estimates from their three t
analyses and that of RIDIC.
4.2 Researcher A's Committed Effective Dose Equivalent Estimates
The AIT determined that 104 mSv (10.4 rem) is the best estimate of !
Researcher A's committed effective dose equivalent. This value
was based on reference woman and, as recommended by the third-
party independent consultant. is the average of the best
individual estimates obtained by the scientific consultant and the '
third-party independent consultant. These values range from 80 to
127 mSv (8.0 to 12.7 rem). ;
The AIT did not use NIH's assigned effective dose equivalent
because, as discussed below and in Appendix G. it was based upon
29 !
-. . -
- - - .. --
l
reference man and NIH's estimated ingestion for Researcher A.
NIH staff assigned an individual effective dose equivalent of 41.7
mSv (4.17 rem). The scientific consultant's best estimate of the
dose was 80 mSv (8.0 rem). The third-party independent
consultant's best dose estimates were 108 mSv (10.8 rem) from the
- urine data set: 103 mSv (10.3 rem) from the NIH Department of
Nuclear Medicine quantified bremsstrahlung scan performed June 30.
1995, and 127 mSv (12.7 rem) from the combined Department of
Nuclear Medicine bremsstrahlung scan and whole body count data
l
set.
Both NIH and the scientific consultant determined the effective
dose equivalents by dividing the estimated intake by the allowable
limit of intake for P-32 by ingestion, and multiplying the results
by 50 mSv (5 rem). This method uses the mathematical relationship
between an ALI and 50 mSv (5 rem). NIH used the value given for
an ALI in 10 CFR Part 20. Appendix B. Table 1. This value is for
reference man. Provided the estimate of intake is correct, this
computational technique may be used by licensees to demonstrate
compliance with Part 20. The scientific consultant concluded that
the approximately 53-kilogram (116 pound-lb) researcher was more
a)propriately represented by the 57-kg (125 lb) reference woman
tlan the 70-kg (154 lb) reference man and elected to use the ALI
for P-32 and reference woman. The independent third party did not
use the ALI to determine the dose but used the " Code for Internal
Dosimetry (CINDY)" to calculate weighted organ doses directly from
the three estimated P-32 ingestion values and correcting the
appropriate organ dose results for the differences between
reference man and reference woman.
4.3 Dose Estimates for Researcher A's Fetus
The AIT determined that 66 mSv (6.6 rem) is the best estimate for
the fetal dose equivalent. This value, as recommended by the
third-party independent consultant, is the average of the best
individual estimates obtained by the scientific consultant and the
third-party independent consultant. These values range from 51 to
81 mSv (5.1 to 8.1 rem).
The AIT did not use NIH's assigned fetal dose equivalent because,
as discussed below, it was based upon NIH's estimated ingestion
for Researcher A.
NIH staff's assessment resulted in an estimated fetal dose
equivalent of 32 mSv (3.2 rem). The scientific consultant's best
estimate was 51 mSv (5.1 rem). The third-party independent
consultant's estimates were 69 mSv (6.9 rem) from the urine data
set. 65 mSv (6.5 rem) from the NIH Department of Nuclear Medicine
quantified bremsstrahlung scan performed June 30, 1995, and 81 mSv
(8.1 rem) from the combined De)artment of Nuclear Medicine
bremsstrahlung scan and whole aody count data sets.
All three groups arrived at the fetal dose equivalent estimate by
assuming a dose of 0.078 mSv (0.0078 rem) per 0.037 MBq
30
. _ _ _ _ .
(microcurie) introduced into the maternal transfer compartment.
The number of megabecquerels (microcuries) introduced into the
maternal transfer compartment is 80 percent of the estimated
megabecquerel (microcurie) intake value.
, 4.4 Other Contaminated individuals
!
There were 26 contaminated individuals, other than Researcher A,
with estimated ingestions ranging from 0.07 to 1.4 MBq (2 to 37
microcuries) .
NIH staff calculated committed effective dose equivalent values
from the available bioassay data for each individual, took the
highest estimated value, and assigned a range for the estimated
dose. The ranges were in 0.5 mSv (50 millirem) increments from 0
, to 3.5 mSv (0 to 350 millirem). There were 18 individuals with
'
estimated doses less than 1.0 mSv (100 millirem), and eight with
estimated doses above 1.0 mSv (100 millirem) but less than 3.5 mSv
(350 millirem).
Although all 26 individuals were employed at NIH, for pur)oses of
limitation of exposure to radiation, the AIT determined tlat 21
individuals were occupational workers and 5 individuals were
members of the public. Of the,21 occupational workers, no one
received a radiation ex)osure in excess of the annual limit of
50 mSv (5 rem) establis1ed for occupational workers. Of the 5
individuals identified as members of the public, one individual
received a radiation exposure in excess of the annual limit of
1.0 mSv (100 millirem) established for members of the public.
This individual had an estimated committed dose equivalent of
2 mSv (200 millirem) based on a 24-hour urine sample. !
l
The AIT determined that the use of dose ranges is acceptable given
that there is considerable uncertainty associated with the dose
% equivalent estimates for these individuals. The P-32 model used
9 for the estimates is designed for single uptakes. The AIT noted
that some individuals, especially those in coffee groups,
probablyingested P-32 on more than one occasion. Also the
majority of estimates were made from spot samples with reference
man or woman urinary output values assigned for the 24-hour
volumes. Only 10 of these individuals were asked to give 24-hour
urine specimens and in all cases the actual urinary output was
considerably different from reference man or woman. The
differences varied from 200 to 1500 milliliters (with an average
of 670 milliliters) from the appropriate reference man or woman
value. If the exact time and number of P-32 ingestions were
known, the estimated doses probably would have been lower.
The AIT determined the NIH ranges repcesent the doses to these
l
individuals, and our scientific consultant confirmed these doses.
'
l 5 SAFETY SIGNIFICANCE
!
The following evaluations of the safety significance of the exposures to
the pregnant researcher and to her fetus have been reviewed by the NRC's
31
.
l
l
medical consultant and are based on his draft report dated November 21,
1995. Although during July. August, and November, the medical
consultant received informatiori verbally from NIH's Occupational Medical
l Services and the researcher's obstetrician, issuance of the medical
consultant's final report can not occur until the recei]t and review of
l the 3 regnant researcher's actual medical records. If t1e opinion stated
by t1e medical consultant in his final report differs from that stated
herein, an addendum to this AIT report will be issued.
5.1 Effect on the Pregnant Researcher
Serious medical consequences are not probable to the exposed
researcher as a result of the ingestion of P-32.
- The researcher reported episodes of vomiting on June 29 and 30
l that may have persisted for several days thereafter, and
l right-sided back pain that continued into July. Based on the
limited medical information available to the medical consultant,
he was unable to determine whether any of the symptoms reportedly
experienced by the researcher were related to ingestion of P-32.
Based on AIT's best estimate dose of 104 mSv (10.4 rem) and range
of doses of 80 to 127 mSv (8.0 to 12.7 rem) to the researcher. it
is highly unlikely that these symptoms are radiation related but,
because the chemical form of the ingested P-32 has not been
determined, chemical toxicity cannot be entirely excluded.
For purposes of assessing the potential medical effects of this
P-32 ingestion, the medical consultant used both the average
radiation dose to the researcher and to her fetus, as recommended
by the inde)endent third-party consultant, and the range of doses
determined )y the AIT scientific consultant and the third-party
independent consultant.
5.1.1 Potential Deterministic Consequences of P-32 Ingestion
Assuming an effective dose to the contaminated researcher of
104 mSv (10.4 rem) and a dose range of 80 to 127 mSv (8.0 to
12.7 rem), no deterministic effects are expected. Based on
Appendix 4 of the dose estimate report from Lawrence
Livermore National Laboratory, the red marrow dose would
approximate 390 mSv (39 rem) in a female subject whose
effective dose was 104 mSv (10.4 rem). The red marrow dose
would range from 300 to 470 mSv (30 to 47 rem) for a dose of
80 to 127 mSv (8.0 to 12.7 rem). Discernible hematopoietic
system effects can be detected after single exposures to the
bone marrow of doses as small as 500 mSv (50 rem), but would
be quite unlikely with a smaller dose. Moreover, the
modulating influence of low dose rate must be considered in
an instance of internal exposure with a radionuclide having
a half-life of 14.3 days. The limited hematological data
available for the contaminated researcher do not show
evidence of a deterministic effect on the hematopoietic
system.
5.1.2 Potential Stochastic Consequences of P-32 Ingestion
32
. . _. _ - . _ - - _ _ . ~ - _ _ _ _ _ _ _ _ _ ._. . _ _ _
.
l
l
{
The stochastic effect of concern in the contaminated
researcher is radiation-induced cancer. Based on the risk
estimates published in " Health Effects of Exposure to Low
Levels of Ionizing Radiation: BEIR V." (Washington D.C.:
National Academy Press: 1990:175). the lifetime additicnal
! risk for fatal cancer with a 104 mSv (10.4 rem) exposure is
i 1.24% for a woman exposed at age 25 and 0.59% for a woman :
'
exposed at age 35. These lifetime additional risk estimates
for exposures between 80 and 127 mSv (8.0 and 12.7 rem)
range from 0.95 to 1.51% for a woman exposed at age 25 and l
from 0.45 to 0.72% for a woman exposed at age 35 The
actual additional risk in this case can be reduced by a
factor of 2.0 to 2.5 because the exposure was delivered at a
low dose rate. For comparison, the lifetime risk of fatal
cancer without radiation exposure is approximately 20%.
5.2 Effect on the Fetus of the Pregnant Researcher
Serious medical consequences are not probable to the exposed
researcher's fetus as a result of the ingestion of P-32.
5.2.1 Potential Deterministic Consequences of P-32. Ingestion
Assuming a fetal effective dose of 66 mSv (6.6 rem) and a
range of 51 to 81 mSv (5.1 to 8.1 rem), no deterministic
effects are expected with a fetus of approximately 16-weeks
age. The deterministic effect of most concern at this stage
of fetal development would be impairment of brain
development (manifested by retardation or reduced
intelligence test scores). The studies evaluating the risk ,
of injury to the developing brain have stratified fetuses !
into those 8 to 15 weeks of age (post-conce) tion) and those ;
16 to 25 weeks of age (post-conception). T1e age of the l
fetus at the time of exposure in this case (16.5 weeks post-
menstrual age, or 14.5 weeks post-conception age) is at the
border between these two groups. Additionally, the
radiation dose from P-32 would be delivered over a
protracted period of time, thus clearly extending part of
the exposure into the older-age stratum. Moreover, it is
unknown whether modulation for a low dose rate must also be
considered. Irrespective of these points of uncertainty,
radiation-induced severe mental retardation is unlikely in
this fetus.
Although the available data are generally taken to indicate
a linear, non-threshold response, they are also consistent
with a threshold in the range of 200 to 400 mSv (20 to 40 1
rem) for the 8 to 15 week fetus, with no definite increase ,
in mental retardation evident at doses of less than 200 mSv :
(20 rem). In fetuses between 16 and 25 weeks of age, no ,
definite increase in mental retardation is evident at doses !
below 500 mSv (50 rem). With regard to less severe
impairment of brain development, the results of intelligence
test scores suggest a significant radiation-related decrease
in 8 to 15 week fetuses and a less marked effect at 16 to 25
33
.
- . - .-. .. -- ._ . - - - . - - - - . - _ . - - .
weeks. With utilization of a linear model, a 21 to 33 point
diminution of 10 score is expected with a 1000 mSv (100 rem)
acute exposure to a 8 to 15 week fetus. Hence ignoring
dose-rate effects, a theoretical reduction in IQ of the
exposed fetus in this case of 1.4 to 2.2 points (assuming a
-
fetal dose of 66 mSv (6.6 rem)) might be expected. The
range of theoretical reductions in 10 scores extends from a
low value of 1.1 points to a high value of 2.7 points, with
fetal doses ranging from 51 to 81 mSv (5.1'to 8.1 rem).
Such an effect would be undetectable and any putative
relationship to radiation exposure unprovable.
5.2.2 Potential Stochastic Consequences of P-32 Ingestion
l Although there is moderate uncertainty in the data used for
cancer risk estimation as a result of in utero radiation
l exposure, a reasonable estimate of the risk during the first
,
10 to 14 years of life for leukemia and other childhood
'
cancers after in utero radiation exposure is approximately
0.05% per rem. Accordingly, in this case with an average
fetal effective dose of 66 mSv (6.6 rem). an excess risk of !
0.33% is estimated. An excess risk of 0.26 to 0.40% is
estimated for effective doses ranging from 51 to 81 mSv (5.1
to 8.1 rem). It is unknown whether this risk estimate :
should be reduced because of the low dose rate associated
with this internal ex)osure. For com]arative purposes, the
natural risk of childlood cancer is a)out 0.1%.
5.3 Effect on Other Contaminated Individuals
l
No deterministic or stochastic consequences are expected for the
'
additional 26 individuals who were internally contaminated with
P-32.
All of the 26 individuals received' radiation exposures of less
than 10 percent of the annual limit of 50 mSv (5 rem) established
for occupationally exposed workers. This 50 mSv (5 rem) annual
limit was established to ensure that there would be no
deterministic consequences for an individual who received an
exposure within the annual limit and to minimize the probability
of stochastic consequences for an individual who received the
annual limit in each year of her or his working life.
6 ASSESSMENT OF THE LICENSEE'S RESPONSE TO THE INCIDENT
Researcher B initially attempted to notify the Authorized User after the
identification of Researcher A's contamination. Unable to accomplish
this Researcher B notified the NIH Fire Department through use of the
emergency call number (116). The Fire Department appropriately
identified the problem as involving radioactive material and contacted
the Deputy RSO at his home. The Deputy RSO is at the to) of the call
list for emergencies involving radioactive materials. W1en the
Authorized User was notified of the incident. he contacted the RSB
directly. Both notifications initiated a RSB res)onse. These
notifications were performed in accordance with t1e licensee's emergency
34
_
_. .._ _ _ _ ._ _
,
procedure.
Initial response to the incident by RSB technical staff and management
, was immediate and appropriate. Proper surveys of Researcher A were
)erformed to evaluate and classify the contamination before transporting
lesearcher A offsite. Bypassing this evaluation in favor of immediate
trans)ortation offsite and hospitalization would have risked the
possi)le spread of contamination and been warranted only if Researcher
A's condition was life threatening (e.g. significant bleeding or
cardiac arrest).
The final survey of Researcher A. performed before her transport
offsite, was the collection of the initial urine sample. A meter survey
of this sample proved that Researcher A was internally contaminated.
The decision to collect and analyze the initial urine sample and to use
NUREG CR-4884 to obtain a rough estimate of Individual A's intake was
- appropriate and timely.
i The licensee's decision to contact radiation emergency medical
professionals at REAC/TS and arranging for the REAC/TS physician to
speak directly with the emergency room physician, as well as the
'
licensee's decision to utilize the expert resources at RIDIC to assist
with the evaluation of the P-32 intake, the radiation dose to Researcher
A and to her fetus, was appropriate given the estimated level of P-32
i internal contamination. Researcher A's pregnancy, and the lack of
experience dealing with significant P-32 internal contamination both at
- NIH and at other institutions.
1 The RSO advised the attending physician at Holy Cross Hospital of the
i need to collect 24-hour samples of Researcher A's urine for
i determination of her P-32 intake and radiation dose. The Authorized
'
User reportedly later instructed the attending physician to collect only
4
a portion of each void but to measure and document the total volume of
i each void. The attending physician did not believe that the two
instructions were substantially different but he modified the
instructions such that a portion of each urine sample was collected and
the remainder of each sample was placed in a single " pooled urine"
container. These instructions were provided to Researchers A and B when
Researcher A was discharged from Holy Cross Hospital early on the
morning of June 30. Researchers A and B followed these instructions.
Later on the morning of June 30. the RSO provided Researchers A and B
with an additional container for urine collection and instructed them to
begin using this container. Although NIH used these early samples in
their analysis of the P-32 intake. RIDIC and LLNL did not use them
because they were collected only over a 12 hour1.388889e-4 days <br />0.00333 hours <br />1.984127e-5 weeks <br />4.566e-6 months <br /> period and both RIDIC
and LLNL believed that the use of this data to represent the average
P-32 concentration in urine over a 24 hour2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> period, at a time when the
rate of P-32 excretion was rapidly changing, was inappropriate.
The licensee's analytical capabilities enabled them to accurately
identify and quantify P-32 contamination in urine and the licensee's
decision to perform urine bioassays and whole body counts for assessment
35
_ . . . _ . _ _ _ _ . _ _ .__ __~ .. _ ._ _ . . . _ _ _ _ _ _ _
,
. ,
l
of the P-32 intake was appropriate. However, the licensee did not
quantify the whole body counting data due to concerns that the
quantification could not be performed accurately and that the data would
be misinterpreted. The AIT recognized the difficulties inherent in
cuantifying activity based on bremsstrahlung radiation however, these
cata could have been used as guideposts to support other determinations ;
of intake.
l The RSB should have contacted the NIH Police Department-immediately
!
after receipt of allegations that Researcher A was intentionally
contaminated rather than waiting until the following morning. Early'
notification would have allowed the NIH Police the opportunity to secure
the affected areas on the evening of June 29. possibly preserving
evidence. .
Although surveys of laboratory 5018, the SC conference room, and the
corridors / hallway in between the laboratory and the conference room were
extensive. the licensee did not survey additional areas on the fifth :
floor. These additional areas, including water coolers and coffee
stations, were not surveyed because the licensee believed that the ;
method of ingestion was known. This focus on one ingestion pathway, to
'
the exclusion of others, was reenforced by statements made by
Researchers A and B on June 29 that they brought all of their food and
drink to the lab from their home. Early urine bioassays suggest that
the contamination of the water cooler occurred before July 3. Had the !
surveys been more comprehensive and other pathways been considered, the >
contaminated water cooler may have been identified earlier and fewer ,
people contaminated. *
The RSB's review of P-32 inventories on the fifth floor of Building 37
on June 30 was appropriate. However, because researchers do not always- !
account for radioactive decay on inventory records and researchers
dispose of unfinished vials of radioactive material to waste without
quantification of the activity remaining in the vial, the accuracy of ,
these records is questionable. Problems with the accuracy of I
radioactive material inventory records are exacerbated by the fact that
radioactive material suppliers frequently provide more activity than is i
requested by the purchaser. While these problems precluded the i
identification of the source of the material which caused the
contamination. the AIT noted that a knowledgeable person (i.e. , an
authorized user or a supervised user of radioactive materials) could
falsify inventory records to disguise unauthorized use.
The licensee's involvement of NIH management beginning on June 29 and
the Radiation Safety Committee on June 30 was appropriate.
Urine samples from the pregnant researcher were analyzed promptly. The
AIT also determined that the licensee analyzed these samples accurately
as confirmed by the analyses performed for the AIT at ORISE and by the
NRC's Region I laboratory (Appendix H). The periodic reanalysis of
samples by the licensee to ensure that the samples contained no
additional radioactive contaminates was appropriate.
36
,-
.. . __ . _ _ _ - _ _ __ _ _ - _ __ _ _.. .
!
! Radiation surveys of Researcher A's car and apartment were comprehensive
and timely, and appropriate action was taken to secure contaminated
materials.
,
The RSB did not review and evaluate the adequacy of radiation surveys :
conducted by others. As previously stated, bioassay evidence suggests '
that the contamination of the water cooler occurred before July 3,1995.
On July 10 the Director of the Developmental Therapeutics Program
assigned responsibility for surveying all restricted and unrestricted
areas of the fifth floor to specific researchers. It was later
i determined that the Director did not consider the fifth floor hallways.
l including the west hallway where the contaminated water cooler was
located to be within the scope of the survey area. These surveys also
failed to identify the contaminated coffee pots etc. , found by the RSB
staff and contractors during surveys of fifth floor unrestricted areas
conducted on July 14th. The complete resurvey of the 5th floor by
contractors and staff after discovery of the contaminated water cooler
on July 14 was appropriate.
The assessment of the P-32 intake from scans performed in the licensee's
Nuclear Medicine Department was consistent with AIT estimates of intake
based on urine bioassay.
,
37
. . . _ _ _ - - _ _ .- - . -.- . - __. .. . _ . . - _ _ __.
i
The licensee's initial decision to bioassay additional persons who had
access to the contaminated conference room was appropriate and most
samples were analyzed promptly. The later decision to perform bioassays
of everyone who worked on the fifth floor after identification of
internally contaminated individuals who did not use the SC conference
room was appropriate. However, one sample collected on July 5. 1995 was
analyzed on July 18. 1995, because the licensee's system to ensure
tracking of samples did not identify that this sample had been mailed to
the RSB. Furthermore analysis of some urine samples may have been
delayed by up to three days, possibly delaying identification of the
l contaminated water cooler (see Appendix I).
! NIH records show that 55 bioassay samples were counted beginning at
3:32 p.m. on July 13th, and that 10 of these samples were eventually ,
determined to be positive for P-32 contamination. At midday on July 14. '
the RS0 reported to the AIT Team Leader that, thus far, bioassay samples i
from 82 individuals who worked on the fifth floor of Building 37 had i
been processed, sam)les from 17 individuals that they received on the
evening of July 13t1 were being processed, and RSB staff had been sent
to pick up the samples from the remaining 20 or so researchers. The RSO
reported that, at that time. P-32 contamination had been identified in
three additional researchers, bringing the total to eight individuals,
including Researchers A and B.
Expansion of the urine bioassay program on July 14 to everyone who drank
from the contaminated water cooler was appropriate and the AIT
determined that these urine samples were processed promptly.
On July 17. the RSO and NIH management met with NIH staff from the fifth
floor of Building 37 to discuss the contamination of Researcher A and
the subsequent contaminations of additional individuals from the
contaminated water cooler. The research staff were asked to bring their !
own drinks to work rather then rely on the facility's refreshments.
Staff were also told to maintain a high level of surveillance. Those
who received an intake of P-32 were advised of the dose consequences in
writing. All staff were asked to complete questionnaires (Appendix F)
to assist RSB in identifying when the contamination of the water cooler
occurred and to ensure that all of the contamination resulted from the ,
water cooler. A list of telephone contacts was provided to each
researcher as well as a copy of Reg Guide 8.29 (" Instruction Concerning
Risks from Occupational Radiation Exposure"). Urine screenings were
made available to anyone who requested a screening. Although some
licensee staff believed that the results of urine sample analyses were
not adequately communicated to all of the persons on the fifth floor who
were bioassayed. the RSB staff stated that their program to maintain
radiation exposures as low as reasonably achievable recuired them to
communicate and deal with the maximally exposed indivicuals first. The
AIT determined that the licensee's efforts to communicate with Jersonnel
from the fifth floor were appropriate. However, interviews witi NIH
staff indicated that wider and earlier distribution of information
regarding the contamination of the pregnant researcher, specifically the
l
fact that the pregnant researcher did not use P-32 in her research.
l
, 38
__ . _ _ . _ _ . - _ _ _ . . _ _ _ _ _ . . _ _ . _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
would have raised their level of awareness and resulted in their taking
additional safety precautions.
The licensee's use of questionnaires to try to identify when the
contamination of the water cooler occurred and to ensure that all of the '
contamination came from the affected cooler was appropriate. However,
when five of the persons contaminated with P-32 did not respond to the
questionnaire, the licensee was not prompt in contacting those
individuals directly to ascertain whether they had consumed water from
the affected water cooler.
The RSB stated that, after the identification of the contaminated water !
cooler on July 14. they were reluctant to contact the water cooler i
company. However, the RSB did contact the water cooler company, '
identified that the bottles removed from NIH were processed at the
company's facilities in Jessup, Maryland and Lorton. Virginia, and ;
i conducted surveys at these facilities adequate to confirm that no l
l residual contamination existed at these facilities. '
After the identification of the contaminated water cooler on July 14
the licensee recognized the need to ensure that the full scope of the l
l contamination event was known and to reduce the possibility of further
'
contamination events. To this end, after discussions with the NRC. the
licensee committed to: (1) survey all of Building 37: (2) perform urine
bioassays on all willing occupants of Building 37: (3) develo) a
statistically valid plan to perform urine bioassays on other ilH
employees: (4) develop an augmented' radiological survey program: (5)
assist in the development of security plans for food storage and
preparation areas; and (6) make permanent the revised interim security
policy adopted as a permanent policy by the RSC on July 20, 1995,
including the enforcement policy that specified that breaches of
security by researchers would result in mandatory suspension of
privileges to use radioactive materials. These commitments were
documented in Confirmatory Action Letters dated July 21, 1995
(Appendices B and C). NRC determined that these actions were adequate 4
to obtain the required assurances. 1
7 ASSESSMENT OF THE BYPRODUCT MATERIAL PROGRAM AT NIH IN AREAS RELATED TO
THE ACCESS AND CONTROL OF LICENSED RADI0 ACTIVE MATERIAL
7.1 Purchase and Receipt of Radioactive Material
Only authorized users can purchase radioactive materials. All
radioactive material shipments are required to be delivered to the '
RSB Office. The RSB established this requirement independently I
with all radioactive material manufacturers. The inspector noted
that centralized receipt is an effective method for controlling
- the use of radioactive materials.
Requests for the purchase of radioactive materials are submitted
to the RSB on form NIH 88-1. Aside from ordering information the
39
l
l
l
_ _ _ . _ _ _ _ . _ _ _. _ _ _ _ _ _ _ ._ . . _ _ _ _ . . _
'
,
NIH 88-1 form requires the name of the authorized user, the names
of the individuals who will use the radioactive material, and the
signature of the authorized user. The RSB places all orders for
radioactive material. The RSB reviews each request for purchase
of radioactive material against the authorized user's ,
authorization before placing the order. Shipments of radioactive '
material to Building 21 are received by RSB contractors in
accordance with the requirements of 10 CFR 20.1906. and the order
is checked against the authorized user's current inventory. When
surveys and order verification are com)leted the shi) ment is
transported by the RSB contractor to t1e individual w1o ordered
the material or her/his designee.
NRC recently determined that the licensee replaced the NIH 88-1
form with an electronic version of the document which is- '
transmitted via the NIH computer system to the RSB. AIT
determined that this procedure is not in accordance with the
requirements of the NIH license because the electronic document
does not include the signature of the authorized user.
l l
All radioactive material orders and recei)ts are maintained in a i
'
database by the RSB. The RSB also uses t11s information to keep a
running inventory of radioactive materials possessed under the
license. The licensee adjusts the inventory for decay and reduces I
the inventory after waste disposal. Authorized Users are recuired l
to maintain inventories of radioactive material possessed uncer '
their permit. The RSB provides each authorized user with a
monthly report of purchases made under their authorization.
The inspector noted that the NIH 88-1 forms used to purchase the
radioactive material used by Researchers A and B in September and
October 1994. before their receipt of formal training in November
1994, indicated that the material would be used by Researcher C
and the Authorized User. These NIH 88-1 forms did not list
Researchers A and B as users. The AIT determined that the
Authorized User improperly completed these forms as he did not
identify Researchers A and B as the users of these materials.
7.2 Radioactive Material Use and Inventory Control
NIH places ultimate responsibility for the proper use of
radioactive material on the authorized user who purchased the
material, Authorized users are permitted to purchase and share
radioactive material with other users and a supervised user may
work under more than one authorized user. If an authorized user
wishes to transfer responsibility for material purchased under
!
her/his authorization, an NIH 88-1 form must be completed ,
! transferring responsibility to another authorized user. The RSO l
i stated that routine laboratory audits include checks to see who 1s
4 using radioactive material and that unauthorized use is dealt with
{
severely.
40
-
- - -
.
_ _ . _ _ . _ . ._ _-_ - _ _ . - _ _ _ _ _ _ _ _ _ . _ _ _ _ _
'
,
l
!
,
As previously stated, the AIT determined that Researchers A and B ,
were permitted to use licensed material before their receipt of-
required radiation safety training and without required dosimetry.
In addition, the licensee's audit program failed to identify these
inadequacies.
Authorized users are required to maintain radioactive material
inventory records. The AIT noted that the accuracy of inventory
records is questionable because_ researchers only estimate the
amount of material removed from each vial, radioactive decay is
rarely accounted for and, if the vial is not emptied because the -
expiration date has passed, the users do not check the balance
before disposal. The AIT noted that inventories of higher-
activity vials (greater than 37 MBq (1 millicurie)) are more
accurate because the higher-activity vials normally contain
radioactive raw material that is used'in its entirety in a single
chemical syntheses. This contrasts with low-activity vials, which
.
'
normally contain prelabeled compounds that are used a few
kilobecquerels (microcuries) at a time over a period of days or
weeks. The AIT identified one researcher who only documented the
recei
vial.pt This
of a radioactive
researcher stated material thatvial and detailed
a more the disposal of the
inventory form
was not useful in her case because of the numerous, small
withdrawals she and her coworkers routinely made from the vial. ,
This researcher stated that, before leaving Building 37 at the end '
of May 1995, she routinely ordered P-32 labeled nucleotides for
use in her lab and for use in the neighboring lab. Although the
radioactive material was ordered under the authorization of her
authorized use, the material was also used by individuals working
for another authorized user. It was also noted that when this
researcher left Building 37 at the end of May 1995, she left the
remaining P-32 for use by her coworkers and did not record the
disposal of the vials as was her practice. The inspector noted
that the original quantity of P-32 in the two vials, allowing for
decay, was not nearly enough to have been responsible for the
contamination event discovered on June 29, 1995. The AIT was able
to identify from an inventory that was conducted by the licensee
and AIT on June 30th that the vials were present in the laboratory
on that date and were awaiting disposal. An interview of the ,
researchers who routinely used the material showed that the !
material was, in fact, used. The AIT determined that the
researcher failed to maintain the required inventory of
radioactive material in that she did not document the disposal of
these materials to waste.
7,3 Waste Disposal :
NIH, by procedure, forbids the disposal of radioactive liquid i
waste to the sanitary sewer via individual hot sinks. Researchers ;
are required to collect liquid waste in plastic carboys and
document the amount and type of radioactive material entered into
each carboy. Liquid waste is picked up, by appointment, by RSB
41
. _. , _- , - _
. _ _ _ _ . _ _ _ _ . . _. .- _ _ --__ _ _ - _ . _ _ _ . _ _ _ _ . _ _ .. _
l
i
contractors: each carboy is analyzed before disposal to make sure
that the licensee stays within the regulatory limits for disposal
to the sewer. Analyses of irdividual carboys indicates that the
researcher is almost always accurate in his/her statement as to
l the type of radioactive material that was placed into the carboy.
-
but frequently is wrong in his/her estimate of the amount of
radioactive material placed in the carboy. The inspector noted
, that the analyses are high as often as they are low. The
l inspector noted that one cause of inaccurate estimates is that ;
l some researchers account for decay in their documentation and l
l others do not.
NIH requires that researchers document the type and quantity of I
radioactive material placed in solid waste. Jaste is picked up by
appointment by RSB contractors and is shipped offsite for
incineration or compaction at a commercial facility. Analysis of
solid waste is extremely difficult (if not impossible) from a
quantitative standpoint, and NIH does not evaluate the accuracy of
statements made by researchers regarding the amount or type of 1
radioactive material placed in solid waste.
1
The AIT determined that waste inventory records are unreliable for l
the reasons stated above. The AIT concluded that the P-32 used in '
the contamination of Researcher A and 26 other individuals could
have been retrieved or diverted from waste by a knowledgeable
individual without detection.
,
7.4 Security of Radioactive Materials ]
1
The requirements of 10 CFR 20.1801 and 20.1802 are absolute in i
that there are no s)ecified activity thresholds. The licensee l
established a thresloid amount for the security of radioactive
materials located in laboratories at 10 CFR Part 20 Appendix C.
quantities. The licensee's position is based on its understanding
of the answer to question #129 (NUREG CR-6204), posed to the NRC
after issuance of the revised 10 CFR Part 20 in January 1994. The
answer to question #129 indicates, in part. that the security
requirements described in 10 CFR 20.1801 and 20.1802 will not be
enforced for quantities of radioactive material that are exempt
from labeling by 10 CFR 20.1905(a). These are the quantities l
described in Appendix C of Part 20. )
,
The inspectors who conducted the routine inspection of NIH on
June 26 through 30, 1995. concluded that the security of
radioactive materials at NIH was adequate and improved. This
conclusion was reached by inspectors who toured ap3roximately 50
research laboratories in the Bethesda facility in Buildings 21. l
10, 41, 37. 18, 49. 4. and 26T. and in the Rockville facility at o
the Key West Life Science Center and 5 Research Court. The
ins)ectors concluded that most of the improvement was gained by
loccing stock vials and other concentrated samples in
refrigerators and in shields. The inspectors did identify one
42 I
_. .
__
. .-
.
__ ..
l
security violation related to an incident identified by the
licensee on July 13. 1994. This incident occurred on July 12.
, 1994, and involved the loss of a package containing 70 microcuries
l of an iodine-125-labelled protein.
On July 6,1995, the AIT identified one laboratory on the fifth
Floor of Building 37 where vials containing megabecquerel l
(millicurie) quantities of H-3 and C-14 were found unsecured in a
refrigerator. Once identified, the RSB confiscated these items. ;
On July 20, 1995. after the identification of P-32 internal
cont'mination, the licensee adopted a new NIH enforcement policy ,
that specified that breaches of security by researchers would l
result in a mandatory suspension of privileges to use radioactive
material. 1
The security of radioactive material at NIH is similar to the
security of radioactive material at other academic biomedical 1
research facilities. There currently is no evidence to suggest l
that inadequate security of radioactive materials was the
proximate cause of this incident. The possibility exists that the l
contamination resulted from the acts of an individual authorized
to use the material, either as an authorized user or a supervised
user. The possibility also exists that the material could have
been brought onto the NIH campus from another licensed facility.
,
l
,
43
_ _. _ -.. _ _ _ _ . _ . _ - _ . _ _ _ _ _ _ . _ . _ . - _ _ _ _ _
!
1
8 CONCLUS:0NS
The following conclusions are listed in order of safety significance:
8.1 Cause of the Incident
The AIT concluded that these internal contaminations were not the
result of accidental acts. As a result. 27 individuals were
,
unnecessarily exposed to radiation.
!
! 8.2 Licensee's Initial Response
' The AIT concluded that the licensee's initial response after !
identification of the incident on June 29 was appropriate.
Notification of the RSB occurred promptly as designed and the RSB
response was timely. The radiation surveys of the pregnant
researcher. the decision to perform urine bioassays to assess the 1
researcher's intake of P-32. and the licensee's use of expertise l
at REAC/TS and RIDIC were appropriate. Most urine bioassay '
samples from the pregnant researcher and individuals who had
access to the contaminated conference room were promptly analyzed.
8.3 Licensee's Follow-up l
i
The response to the pregnant researcher focused on one ingestion !
pathway, the possible contamination of food consumed by the j
pregnant researcher. and did not consider the possibility of other 1
ingestion pathways. Radiation surveys of the fifth floor of
Building 37 performed by the RSB after identification of the
incident on June 29 were limited in scope. and subsequent surveys
performed by research staff in these areas were not adequately
reviewed. Had complete surveys of the fifth floor been performed,
the contaminated water cooler may have been identified before
July 14 and fewer individuals would have been affected by this
contamination. When the bioassay program was expanded to include
everyone who worked on the 5th floor of Building 37, processing of
all urine bioassay samples was not prompt. The 3rocessing of some
.
urine bioassay samples were delayed as much as tvee days. Had
these samples been processed earlier. the RSB may have been
prompted to survey for additional sources of contamination before ,
July 14 and fewer individuals would have been affected by the !
contaminated water cooler. Processing of urine bioassay samples !
after the identification of the contaminated water cooler was
prompt.
.
44
I
i
.mer , , , , , +--7, --4 _ 4_, r.- + , > < w
, _ _ _ _ _ _ _.._ _ _ _ _ _ _ _ _ _ _ _ . . _ _
,
t
8. 4. Security of Radioactive Material
There is currently no evidence to suggest that inadequate security
of radioactive materials was the proximate cause of the internal
contaminations. The AIT concluded that the security of
radioactive materials at NIH was generally. adequate and comparable
to the security of these materials at otner similar institutions. !
The AIT did identify one laboratory on the fifth floor of Building
37 where megabecquerel (millicurie) quantities of- carbon-14 and
hydrogen-3 were left unattended in an unlocked laboratory. >
Furthermore, special inspections of security conducted in October !
and November,1995, identified weaknesses in the security of
radioactive material. Weaknesses may have resulted in ease of
access to radioactive material which could be a contributing
factor.
However, the possibility also exists that the contamination
resulted from the acts of an individual authorized to use the
material. either as an authorized user or a supervised user.
Further, the possibility exists that the material could have been
retrieved or diverted from radioactive waste or brought onto the
NIH campus from another licensed facility.
8.5 Dose to Contaminated Individuals
8.5.1 Researcher A
AIT determined 39.6 MBq (1070 microcuries) is the best
estimate of Researcher A's P-32 ingestion. This value is
based on the best available bioassay information and, as i
recommended by NRC's independent consultant, is the average j
of the best individual estimates obtained by the scientific
consultant and the independent consultant. .These values
range from 30.3 to 48.1 MBq (820 to 1300 microcuries). AIT
did not use the NIH assigned estimate as discussed in detail,
in Appendix G.
AIT determined 104 mSv (10.4 rem) is the best estimate of
Researcher A's committed effective dose equivalent. This
value was based on reference woman and, as recommended by
NRC's independent consultant, is the average of the best
individual estimates obtained by the scientific consultant
and the independent consultant. These values range from 80
to 127 mSv (8.0 to 12.7 rem). AIT did not use NIH's
assigned effective dose equivalent because, as discussed in
Appendix G. it was based upon reference man and NIH's ;
estimated ingestion for Researcher A. l
l The AIT noted that this exposure exceeds the annual limit
for occupationally exposed workers specified in 10 CFR l
20.1201(a)(1). As stated in Section 5.1 of this report.
! 45
. _
. - - - , . - . ., .-,
. . _ _ _ _ _. _ . _ _ _ ._.___ _ _ _ _ _ .__ . . _ _ _ _ . _ _
serious medical consequences are not probable to the exposed
researcher as a result of the ingestion of P-32,
8.5.2 Researcher A's Fetus ,
i
o AIT determined 66 mSv (6.6 rem) is the best estimate for the !
I fetal dose equivalent. This value, as recommended by NRC's
l independent consultant, is the average of the best
.
individual estimates obtained by the scientific consultant
l
and the independent consultant. These values range from 51
to 81 mSv (5.1 to 8.1). AIT did not use NIH's assigned
fetal dose equivalent because it was based upon NIH's
estimated ingestion for Researcher A.
The AIT determined that a radiation exposure of 66 mSv (6.6
rem) would constitute an overexposure of a declared pregnant
worker's fetus pursuant to 10 CFR 20.1208. As stated in
'
Section 5.2 of this report, serious medical consequences are
not 3robable to the exposed researcher's fetus as a result
of t1e ingestion of P-32.
l 8.5.3 Other Contaminated Individuals !
The licensee assessed the intake of P-32 and estimated ,the
resultant radiation exposure to 26 other individuals who
were internally contaminated, including the pregnant
researcher's husband. The radiation exposures ranged
between 1 and 3.5 mSv (100 to 350 millirem) for eight
individuals, and less than 1 mSv (100 millirem) for 18
individuals. The AIT determined that these assessments were l
conservative in that they assumed every individual's intake !
occurred at the earliest Jossible time and that every ;
individual's intake was tie result of a single exposure to '
the contaminate.
The AIT also determined that 5 of the 26 contaminated
individuals (other than the pregnant researcher) did not
receive radiation safety training because they did not work ;
in a restricted area and they were not expected to come into !
contact with radioactive materials during the course of
'
their employment at NIH. The AIT determined that these 5
individuals are members of.the public under 10 CFR 20 and
are therefore limited by 10 CFR 20.1301 to the receipt of
'
less than 1 mSv (100 millirem) annually. The AIT noted that
one of these individuals is estimated by the licensee to
have received.approximately 2 mSv (200 millirem), an
exposure in excess of the limits for a member of the public
specified in 10 CFR 20.1301. As previously stated in
Section 5.3 of this report, no deterministic or stochastic
consequences are expected for the additional 26 individuals
who were internally contaminated with P-32.
l
46
- _ , _ _ _
.. - -
- . . _ _ . - . . - . - _ - . - _ - - - . - . -- - ------.
! !
I
'
8.6 Training of Researchers A and B
During the course of this inspection. the AIT determined that i
Researchers A and B used licensed radioactive material before i
their receipt of formalized radiation safety training on ,
November 29, 1994. The NIH license permits the use of radioactive i
materials by individuals under the supervision of an authorized
user before receipt of formalized training, as long as the
authorized user certifies that she or he has provided the training
,
described in the " Radiation Safety Orientation for New Personnel
l Planning to Use Radioactive Material" training packet. The ;
licensee reported that the Authorized User never certified that
'
i
this training was provided to Researchers A and B. The use of l
radioactive material by Researchers A and B before November 29.
1994, was not in accordance with the requirements of the NIH l-
license.
8.7 Analytical Capabilities l
The AIT concluded that the licensee's analytical capabilities
allowed it to accurately identify and quantify P-32 contamination
in urine samples. This conclusion was reached after the '
comparison of analyses performed by NIH with the analyses i
performed for the AIT by ORISE and by NRC Region I's laboratory. '
8.8 Purchase and Use of Radioactive Material
The AIT determined that NIH 88-1 forms used to order radioactive
materials received in laboratory 5D18 between August 31 and ,
December 14, 1994, were not completed properly by the Authorized )
User as he failed to list Researchers A and B as users of these '
materials. In addition, the Authorized User allowed Researchers A J
and B to use radioactive material under his su)ervision before
certifying their receipt of training required )y the NIH license.
The AIT also identified one researcher who failed to maintain the
required inventory of radioactive material in that she did not
document the disposal of these materials to waste.
1
47 l
l
l
., .-- - . _. - , -. --- .
_ _
l
I
i
'
!
l
8.9 Collection and Evaluation of Bioassay Samples
l
The licensee's decision to perform urine bioassays to assess the I
intake of P-32 was approariate. The AIT is aware that the ;
emergency room staff at ioly Cross Hospital received two different i
instructions for urine sample collection and that the instructions ,
provided to the pregnant researcher and her husband for urine ;
i sample collection may not have been completely understood.
However, the AIT is not aware of any urine sample data which were '
.
lost or compromised as' a result of these difficulties. Although !
NIH used these early samples in their analysis of the P-32 intake,
RIDIC and LLNL did not use them because they were collected only ,
over a 12 hour1.388889e-4 days <br />0.00333 hours <br />1.984127e-5 weeks <br />4.566e-6 months <br /> period and both RIDIC and LLNL believed that the i
use of this data to represent the average P-32 concentration in l
urine over a 24 hour2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> period, at a time when the rate of P-32
l excretion was rapidly changing, was inappropriate. The licensee's !
decision to perform whole body scans in order to assess the
'
!'
clearance of P-32 from the pregnant researcher's body was
appropriate: however, the AIT is concerned that the licensee
rejected quantification of this information because of concern
that quantification could not be performed accurately, and that '
i the data would be misinter)reted. The AIT believes these data ;
would provide guideposts w1ich would increase confidence in the ;
assessment of the intake. Similarly, although the licensee .
arranged for the performance of scans and an assessment of the !
pregnant researcher's intake by its Nuclear Medicine Department, !
the licensee did not consider this assessment in its overall l
evaluation of the intake because of their concerns for the ;
accuracy of this information. NRC's third party independent. !
consultant used these data to refine the dose estimate.
8.10 Use of External Dosimetry in Laboratory SD18 !
>
'
RSB records indicated that Researchers A and B each were first
issued a film badge to monitor whole body radiation exposure and a j
thermoluminescent dosimeter (TLD) (ring badge) to monitor ,
radiation exposures to their extremities in October 1994. The AIT
noted that, althouah Researchers A and B had not been issued ,
dosimetry at the time of their first use of radioactive material
'
during the third week of September, the NIH license does not
require that either whole body or extremity dosimetry be worn by
researchers who use low energy beta-emitting isotopes such as
S-35. Whole body film badaes were required to be worn beginning
on November 14, 1994, when P-33 use began, and extremity dosimetry '
was required to be worn beginning on December 14, 1994 when more
than 18.5 MBq (0.5 millicuries) of P-32 was used. Film badges and
TLDs for Researchers A and B were sent monthly from the RSB
directly to the Authorized User for distribution to Researchers A
and B. The AIT determined that whole body dosimetry was never
,. distributed to Researchers A and B by the Authorized User and ;
l therefore, Researchers A and B did not wear required dosimetry
during their use of radioactive material. The AIT also determined
48
,, -
. - . _ .. - .-. . . .
!
l
l
l
that the licensee's external dosimetry program failed in that it
did not identify that Researchers A and B were not wearing the
dosimetry that was issued to them, even though monthly dosimetry
reports indicated that the dosimetry was not worn.
l 8.11 Program Audits
l
l The AIT determined that the licensee's audit program failed to
identify, during an audit of laboratory 5D18 conducted on
November 17. 1994, that Researchers A and B were using radioactive
material before toeir receipt of required training. This audit
also failed to identify that P-33 was being used in the
laboratory. The licensee's audit program failed again on May 19.
1995, when it did not identify that Researchers A and B were not
wearing required dosimetry.
9 EXIT MEETING
A Technical Debrief was held with senior licensee managers on August 8.
1995.
1
1
i
49
l
10 ABBREVIATIONS AND DEFINITIONS
Adenine - CsHs N,, -- a purine constituent of nucleic acids
Adenosine - An organic compound composed of adenine and ribose that is a
structural component of nucleic acids.
ATP - Adenosine tri)hos3 hate
ADP - Adenosine diplosplate
AMP - Adenosine monophosphate
AIT -
Augmented Inspection Team
ALARA - As Low As Reasonably Achievable
Autorad. - Abbreviation for autoradiograph
Autoradiography - A technique used to visualize the location of
radioactive materials by developing x-ray film after it
has been placed over the radioactive sample for a
specific time.
BTP - Base triphosphate
Borate - A salt of boric acid
CAL -
Confirmatory Action Letter
Cytosine - C Hs3N 0 -- a pyrimidine base that is a constituent of both
ribonucleic acids and deoxyribonucleic acids.
CTP - Cytosine triahosahate
CDP - Cytosine diplosplate
CMP - Cytosine monophosphate
CPM - Counts per minute
ID - One-dimensional
2D - Two-dimensional
dist. - Distilled
DPM - Disintegrations per minute
FBI -
Federal Bureau of Investigation
GM -
Geiger Mueller
Guanine -CH ss Ns0 -- a purine constituent of both ribonucleic and
deoxyribonucleic acids
GTP - Guanosine tri Shos) hate
GDP - Guanosine diplosplate
GMP - Guanosine monophosphate
HO
2
- Chemical symbol for water
Inosine -
ITP -
Inosine triphosphate
IDP - Diphosphate
IMP - Monophosphate
IRF -
intake retention function
kev -
Kilo-electron volt
KeVp - Kilo-electron volt peak
50
- . - . - -- - - -
'
1
l
1
l
j KH,PO, - Potassium phosphate l
l
l LiCl -
Lithium Chloride
l LSC -
Liquid Scintillation Counter l
l LLNL - Lawrence Livermore National Laboratory '
M - Abbreviation for molar which is used to designate a solution
that contains one mole of solute per liter of solution.
Me0H - Abbreviation for methanol ,
,
'
! mSv -
Millisievert
mole - The amount of a substance w1th a weight in grams numerically
equal to the molecular weight of the substance,
ml - Milliliter
NaFormate - Sodium formate
Nal -
Sodium Iodide
NCI -
National Cancer Institute
NIH -
National Institutes of Health
NIST - National Institute for Science and Technology
NH, - Chemical symbol for ion ammonium
nucleoside- A compound made up of a sugar and a purine or ayrimidine base.
nucleotide- Organic compounds composed of a nucleoside comained with
phosphoric acid.
OMS -
NIH's Occupational Medical Service .
ORISE - Oak Ridge Institute for Science and Education !
PEI cellulose - Polyethylene imide
pH - Abbreviation for hydrogen potential which is a measure of the
acidity or alkalinity of a solution. numerically equal to 7 for
neutral solutions, increasing for increasing alkalinity and ;
decreasing for increasing acidity.
pH front - The location usually trailing the solvent front in the
migrating liquid where the pH is equal to the pH of the
migrating solution.
P, - Phosphoric acid
PP, - Pyrophos ahoric acid ;
pyrimidine- An organic 3ase with the formula of C.H.N and a group of compounds l
having a chemical structure similar to a pyrimidine in a
'
nucleotide component, for example uracil, thymidine.
purine - A chemical compound with a formula of C sH.N and a group of
naturally occurring organic compounds derived from or having
the molecular structure related to purine, for example adenine,
guanine.
REAC/TS - Radiation Emergency Assistance Center / Training Site
RIDIC - Radiation Internal Dose Information Center
R$B -
NIH's Radiation Safety Branch
RSC -
Radiation Safety Committee
RSO -
Radiation Safety Officer
RSOS - NIH's Radiation Safety Operations Section
51
. _ _ _ _ _ _ _ _ _ - . _- . . . _ . - .- _. ._. ___.
l std. - Standard
l
l TAR -
Technical Assistance Request
! TLC - Thin-layer chromatography which is a technique using a thin
l layer of chromatography material on a flat inert support to
l separate chemical compounds.
!
TLD -
Thermoluminescent Dosimetry
Thymine - A pyrimidine base C H N 0, that is an essential constituent of
3,
deoxyribonucleic acid.2
Thymidine - C ,H 3N,0s -- a nucleoside composed of thymine and deoxyribose.
TTP - Thymidinetriahos) hate
l
TDP - Thymidine diplosplate
TMP - Thymidine monophosphate ,
1
Uracil - A pyrimidine C H,N,0, which is constituent of ribonucleic acids. l
UTP - Uracil triahos) hate i
UDP - Uracil diplosplate !
UMP - Uracil monophosphate I
i
l
l
l
52
..
,
[ @ MCo
q#
APPENDIX A
m
0,,
8 g
UNITED STATES
- " E NUCLEAR REGULATORY COMMISSION
- REGloN I
k. ..... 475 ALLENDALE RoAo
KING oF PRUSSIA, PENNSYLVANIA 19406-1415
June 30, 1995
MEMORANDUM T0: Susan Frant Shankman, Deputy Director
Division of Radiation Safety and Safeguards
FROM: #
Thomas T. Martin
e nal Administrator [
SUBJECT: ,
AUGMENTED TEAM INSPECTION CHARTER FOR THE REVIEW 0F THE
DEPARTMENT OF llEALTH AND HUMAN SERVICES NATIONAL' IN
0F HEALTH (NIH)
l
On June 30, 1995, the National Institutes of Ilealth (NIH) informed an on-site
NRC inspector that they had responded at 6:00 p.m. on June 29, 1995, to a
reported contamination of a 32-year old pregnant female researcher. An '
initial urine sample indicated internal contamination levels of 16,000 dpm per
milliliter.
.
1
The Division of Radiation Safety and Safeguards (DRSS) is assigned the
responsibility for the overall conduct of the augmented inspection. l
Attachment I represents the Charter for the AIT and details the scope of the
inspection. The inspection shall be conducted in accordance with NRC
Management Directive 8.3, NRC Inspection Manual 2800, Inspection Procedures
83822, 87100 and 87103, Regional Office Instruction 1010.1 and this
memorandum.
Mr. James P. Dwyer is appointed as the AIT leader. Other AIT members are
identified in Attachment 2. DRSS is responsible for the timely issuance of
the inspection report, the identification and processing of potentially
generic issues and the identification and completion of any other action
warranted as a result of the team's review.
Attachments: As stated (2)
!
,
A-1
4
]
.
h0GMENTEDINSPECTIONTEAMCHARTER
The general objectives of this AIT are to:
9
1. Conduct a thorough and systematic review of the circumstances
surrounding the Department of Health and Human Services National
Institutes of Health internal contamination reported to the NRC on
June 30, 1995, including an incident chronology detailing the sequence
of events associated with the contamination event.
2. Assess the safety significance of the event and communicate to Regional
and Headquarters management the facts and safety concerns related to the
event so that appropriate follow-up actions can be taken. Include an
analysis of the actual and potential dose consequences.
' '
3. Collect, analyze, 'and document factual information and evidence
sufficient to determine the cause(s), conditions, and circumstances
pertaining to the event.
4. Examine any procedural or management failures and identify associated
root causes.
5. Prepare a report documenting the results of this review for the Regional
' Administrator within thirty days of the completion of the inspection.
In addition to the above, in coordination with NMSS, examine and assess the
adequacy.of the NRC procedures and processes for responding to on-going events
including a medical emergency. Document any lessons learned and recommended
changes in a separate document within sixty days of completion of the
inspection.
Attachment (1)
A-2
v
,
APPENDIX B
pm Rf Cg
UNITED STATES
E ?, NUCLEAR REGULATORY COMMISSION
$~ ! nEcion i
i 475 ALLENDALE ROAD
'
.
,
,o'f KING OF PRUSSIA, PENNSYLVANIA 1940 41415
.
-
July 21, 1995
Docket No. 030-01786 License No. 19-00296
Michael M. Gottesman, M.D.
Deputy Director for Intramural Research
Department of Health and ' Human Services j
National Institutes of Health
NIH Building 1, Room 114
9000 Roci;ville Pike ,
Bethesda, MD 20892 l
SUBJECT: CONFIRMATORY ACTION LETTER 1-95-011
Dear Dr. Gottesman:
! The National Institutes of Health continues to investigate the circumstances
"
surrounding contamination of a water cooler and NIH staff. Followup actions
were addressed in a letter from you to Thomas T. Martin, Regional
'
Administrator, Region I, on July 19, 1995. These actions were clarified and l
. amplified in a telephone conversation on July 20, 1995, between yourself and '
'
others of your staff, and myself and others of my staff, and in a subsequent
'
conversation on July 21, 1995 between Richard G'. Wyatt,' Assistant Director for
i Intramural Affairs and Mr. Martin. The purpose of this letter is to document
your commitments.
Pursuant to the July 19, 1995 letter, and the telephone conversations of July
20 and 21, 1995, it is our understanding that you have taken or will take the
following actions by the dates specified: -
1. Complete the radiological survey of all areas in Building 37 by July 21,
1995.
]
l 2. Obtain urine samples from~ all individuals in Building 37, and complete j
the analysis of each sample by July 28, 1995. '
3. Develop by July 21,.1995, and implement a statistically valid plan to l
collect and analyze urine samples from other NIH staff.
4. Develop and begin implementation of an augmented. continuing radiological
survey program for the NIH campus by July 28, 1995.- The program will
include a survey of all water coolers and food storage areas on.the NIH
campus, or individual declinations from the owners that a survey is not
desired.
t
B-1
_ . _ _ _ . _ _ ._ _ _ _ . _ . ._ _ _ . _ _ . . _
7
_
_ __
'
-
1
)
,
M. Gottesman, M.D. 2
5. Develop individualized food security plans for work groups in Building
37. These security plans are to be developed under the direction of the
designated safety officer for Building 37, using the expertise of the -
authorized users in the work, groups. These plans should emphasize
security of the food storage and preparation areas' and will be complete
and in place by July 28, 1995.
6. At a meeting of the Radiation Safety Committee on July 20, 1995, the
Enhanced Interin Security Policy submitted on May 18, 1994, by Richard
G. Wyatt, Assistant Director for Intramural Affairs, and revised on
j September 13, 1994' as submitted by Ted W. Fowler, Acting Radiation -
Safety Officer, was made a permanent policy effective August 1,1995.
Training on this policy will be provided by August 1,1995, and will
include an enforcement policy that specifies that breaches .of security
by researchers will result in mandatory suspension of privileges to use
licensed materials.
Pursuant to Section 182 of the Atomic Energy Act, 42 U.S.C. 2232, you are
required to:
1. Notify me immediately if your understanding differs from that set forth
above;
2. Notify me if for any reason you cannot complete the actions within the
specified schedule and advise me in writing of your modified schedule in
advance. of the change; and
.
3. Notify me in writing when you have completed the actions addressed in ;
this Confirmatory Action Letter. j
i
Issuance of this Confirmatory Action Letter does not preclude ~ issuance of an ,
order formalizing the above commitments or requiring other actions on the part )
of the licensee; nor does it preclude the NRC from taking enforcement action i
for violations of NRC requirements that may have prompted the issuance of this
letter. In addition, failure to take the actions addressed in this
Confirmatory Action Letter may result in enforcement action.
The responses directed by this letter are not subject to the clearance
procedures of the Office of Management and Budget as required by the Paperwork
~
Reduction Act of 1980, Pub. L. No.96-511.
In accordance with 10 CFR 2.790 of the NRC's " Rules of Practice," a copy of
this letter and your response will be placed in the NRC Public Document Room
(PDR). To the extent possible, your response should not include any personal '
privacy, proprietary, or safeguards information so that it can be placed in
4
B-2
i
j
_. _ . - . _ . . _ , _ _ . . . . _ _ _ _ _ _ _ . _ _ ._. . _ . _ _ _ . _ _ _ . .
-
,
t
M. Gottesman, M.D. 3
the PDR without redaction. However, if you find it necessary to include such
information, you should clearly indicate the specific information that you
desire not to be placed. in the PDR, and provide the legal basis to support
y'our request for withholding the information from the public. -
Sincerely, .
+/
Charles W. Hehl', Director
Division of Radiation Safety
and Safeguards !
~
Docket No. 03'0-01786
License No. 19-00296
cc:
State of Maryland
Robert Zoon, RS0
.
.
t
.
P
B-3
.
'
.
APPENDIX C
sA*ECu
e 9 UNITED STATES
f h . NUCLEAR REGUI.ATORY COMMISSION
$ j REGION I
e
,' g
g- 475 ALLENOALE ROAD
KING oF PRUSSIA, PENNSYLVANIA 1940G-1415
July 21, 1995
-
-
. -
Docket No. 030-01786 License No. 19-00296
Michael M. Gottesman, M.D.
Deputy Director for Intramural _ Research
~
Department of Health and Human Services
National Institutes of Health
NIH Building 1, Room 114
9000 Rockville Pike .
~
Bethesda, MD 20892
'
.
SUBJECT: CONFIRMATORY ACTION LETTER ~ 1-95-011, REV. 1
Dear Dr. Gottesman:
.
The National Institutes of Health continues to investigate the circumstances .
surrounding contamination of a water cooler.and NIH staff. Followup actions
were addressed in a letter from you to Thomas T. Martin, Regional
Administrator, Region I, on July 19, 1995. These actions were clarified and
amplified in a.telepho.ne conversation on July 20, 1995, be, tween yourself and
others of your, staff, and myself and others of my staff,. and in a subsequent
conversation on July 21, 1995 between Richard G. Wyatt; Assistant Di~ rector for
Intramural Affa' irs and Mr. Martin. The purpose of this letter is to document
your commitments. Revision of our original letter is based on refinements
discussed during telephone conversations between yourself and Mr. Martin, also
on ~ July 21, 1995.
1
Pursuant to the July 19, 1995 letter, and the telephone conversations of July l
20 and 21, 1995, it is our understanding that you have taken or will take the
following actions by the dates specified:
1. Complete the radiological survey of all areas 'in Building 37 by July 24,
1995.
2.' Obtain urine samples from all available and willing individuals in
.
Building 37, and complete the analysis of each sample by July 28, 1995.. 1
~
3. Develop by July 21, 1995, and implement a statistically valid plan to
- l collect and analyze urinc. samples from other NIH staff.
I
.
i
s
C-1
s
.
. , . - . - - - _ - - . - - - - _ . - . - . . - . .
. _
i. .
,
.
-
!
l *
. . .
-M. Gottesman, M.D. 2
i
4. Develop and begin implementation of an augmented continuing radiological
survey pregram for the NIH campus by July 28, 1995.- The program will
include a survey of al.1. publicly accessible water: coolers and food ,
storage areas' in laboratory buildings. The program will also include
. notification of all NIH staff that any other water coolers or food
storage areas will be surveyed upon request.
5. Assist the safety committee for Building 37 to develop and implement
plans for security of the food storage and preparation areas. These .
plans will be complete and in place by July 28, 1995;
6. At a meeting of the Radiation Safety Committee on July 20, 1995, the
l
Enhanced Interim Security Policy submitted on May 18, 1994, by Richard
G. Wyatt,- Assistant Director for. Intramural Affairs, and revised .on . '
September 13, 1994 as submitted by Ted W. Fowler, Acting Radiation -
l Safety Officer, was made a permanent policy effective August 1,1995.
!
Dissemination of this policy will be achieved by August 1,1995, and
l will include an enforcement policy that specifies that breaches of
' security by researchers will result in mandatory suspension of
privileges to use licensed materials.
'
l Pursuant to Section 182 of the Atomic. Energy Act, 42 U.S.C. 2232, you are
required'to:
1. Notify me.immediately if your understanding differs from that set forth
! above;- ,
.
.
.
l 2. Notify me if for any reason you cannot complete the actions within the
I speci.fied schedule' and advise me in writing of your modified schedule in
l advance of the change; and-
i
3. Notify me in writing when you have completed the actions addressed in !
this Confirmatory Action Letter. '
'
Issuance of this Confirmatory Action Letter does not preclude issuance of an
order formalizing the above commitments or requiring other actions on the part
of the licensee; nor does it preclude the NRC from taking enforcement action 1
for violations of NRC requirements that may have prompted the issuance of this
letter. In addition, failure to take the actions addressed in this
Confirmatory Action Letter may result in enforcement action.
.
.
. r
The responses directed by this letter are not subject to the clearance
procedu.res of the Office of Hanagement and Budget as required by the Paperwork
Reduction Act of 1980, Pub. L. No.96-511. .
f
'
!
.
'
,
.
l C-2 . . .
1
- -
,
M. Gottesman, M.D. 3
In accordance with 10 CFR 2.790 of the NRC's " Rules of Practice," a co'py of
this letter and your response will be placed in the NRC Public Document Room
(PDR).. To the extent pc.ssible,.your response s.hould not include any personal ..
privacy, proprietary, or safeguards information so that it can be placed'in
the PDR without redaction. However, if you find it necessary to include such
information, you should clearly indicate the specific information that you
desire not to be placed in the PDR, and provide the legal basis to support
your request for withholding the information from the public.
Sincerely,
~
W .
' Charles W.11ehl, Director -
Division of Radiation Safety
and Safeguards
Docket No. 030-01786
License No. 19-00296
CC"
State of'Haryland )
-Robert Zoon, RSO ,
i
.
4
e
9
C-3
-
-
._ _ _ __ ._ _ _ . _ . _ . _ . _ _ . _ _ . . . _ _ _ _ _ _ _ _ _ _ _ . . _ _ _ _ _ _ _ _ . _ _ . . .
. APPENDIX D . -
- - - . -
z
'
,,,.,, 1 - - - 1 - >
_.
1
g1 ...
dhrkkFM ,
,
7
- ENI i :
-
. ;1b i w %. 6 ,gj b& *
Xo
-
e
-
s a.
.
'
.
.
ggy,-i de i
.
'
. .m
s.
.
. .... .
i.
.. .
X.
i t,
.
.
'
(S ~ g
-- -
/jiL '
s
+
s
'N
s
+ i I
s
+
- ,
'
c ,
g
7 g 4i 9 ,
-
12 e,- p, t
g-
.
o. e -
'
a 4 4pa
a In qs -
- 1:n %y >n, ifyff T 7
5 8 1
~ gT'YT. ;! :
8
9 , -a b r v -
N !5 ,' 5 ! l' I ir : -
.
- . . . .m., . . . . . . . . . . ;-
J
- ! i n r a t t = =
~!
-
+dj
.
l +iL g +
.gf-p i I
g
,
a +
s
,
- ig ;
e at. . . a .un n .
u <e -
d 4 M 4! 5 8 #- 8 p p.4 ' Mr*,k O I:
.
! .
-
-
-
,:
p
- n
i
M y
.n
!
+
=
i
=
V
g
.
-
3 .
= P -
i
.
tx D
. . . o1,ur, . . . - . . . . . . :
C
-
,
Cl g n. . a ~ g ; ~ ~
'
g 3
g fN f- h + ! I 7 ph C
f
"
'
-
r *
.
- =
g - + T- i) 3 g -
d dM.L /h ' 7 ha =
hdd8 -
. .
a x, a
a - a-s vs o. ~ -
'
NF f 0"hh k
. ! I .
-
Y "~
9 9 (I -
if *
_ _ _
- -
F i I, E i i ii E is ! 7 I F.S ? 1* -
"
% . . uriury a conc as tie . m XJ %,3I 3 -
.
a a
' A+
<
-
.
E,
g
-4
5
h
i ; i ! i 14 #
g
- =
-
~=!
! ! i { i
.
liic*
?
d
-
$I*
C
M."JL./MLd.adw /02 4 hLdkI I.JE:/M -
.
4I
i i $7 s E
rn . *3 $ 4 8 $ $ 9 E
._
r [ I I I I 1 I I I I I I I I I I I I I I I iI I I I I I 1 >[~
FLOOP, PLAN
NIH BUILDING 37 " 5TH FLOOR
_ . _ _ _ .__ _ _ _ _ _ _ . _ _ _ _ _ _ _ _ _ _ _ . _ _ . . _
APPENDIX E
CONTAMINATED WATER COOLER RADIONUCLIDE ANALYSIS
.
1. BACKGROUND
On July 14, 1995, within the National Institutes of Health (NIH) the
water cooler on the fifth floor west hallway of Building 37 was'found to
be contaminated with radioactive material. This water cooler may have ,
been the source of the radioactive material ingested between late June i
and mid-July by the individuals working on the fifth floor.
l In the initial survey, NIH Radiation Safety Branch staff reported
l detecting radioactivity in the stainless steel reservoir (part 11 in
'
Figure E-1) and the dri > pan (part 15. Figure E-1) but not in, or on, ,
the water jug .itself. Jsing a pancake probe attached to a Geiger-
Mueller survey instrument NIH staff detected 60,000 counts per minute
in the reservoir. Conventional liquid scintillation techniques were . ,
used to measure 300 disintegrations per minute in a 1-milliliter aliquot ;
drawn from the approximately 15-milliliters of water removed from the !
!
drip pan. J
>
A number of efforts were made to identify the radionuclide and its form. !
These efforts will be summarized below.
'
.
2. IDENTIFICATION OF THE RADIONUCLIDE ,
The Radiation Safety Branch identified the radionuclide as I
phosphorous-32 (P-32). The identification was made using the energy l
l spectrum in the tritium, carbon-14, and P-32 windows from conventional
liquid scintillation counting.
3. IDENTIFICATION OF THE-CHEMICAL FORM
The radionuclide was identified as the same radionuclide ingested by the
individuals working on the fifth floor. Since there was only a limited
amount of the water sample available for testing, the types of com)ounds
on the fifth floor labeled with P-32 were looked at to determine t1e
best approach to identifying the chemical form. P-32 was available in a
number of forms in the laboratories on the floor. Inorganic phosphate
was used to label other molecules, prelabeled P-32 nucleotides were
used, and the radioactive waste contained any number of different P-32
labeled biochemical byproducts from experiments.
Since the water cooler was probably contaminated before July 3 the
environmental conditions surrounding it could affect the stability of
P-32 labeled compounds in the water. The cooler was located on the west
hallway, which ran the length of the building. The up)er half of the
exterior wall contained sealed windows and the bottom 1alf was a solid
wall. The interior wall contained several built-in (but out-of-service)
water fountains, and a number of double wooden doors leading into either
'
the laboratory corridors or the electrical / mechanical utility access
E-1
. _ . . - . .-_
.
_ _ ._ ___ _ _ _ _ _ _ _ _ . . _.._._ _ _ . _ _ _ _ _
i
,
4
I ways between the-backs of the laboratories.
-
This hallway received the
j full afternoon sun without adequate air-conditioning to reduce this
- additional heat load. Signs on the corridor doors reminding the workers
3
to kee) the corridor doors closed to conserve the air-conditioning in
- the la] oratories. The weather from the end of June to the middle of
- July was in the high 90's and hit 100 degrees fahrenheit some days. The r
j west corridor was hot in the afternoons and the water cooler drip pan
j would have been hot also.
Both these environmental conditions and the length of time the P-32
i material was in the drip pan increase the probability of chemical and
4
biochemical degradation of the P-32 labeled material.
3
!' Tests were performed by NIH and E. I. DuPont de Nemours & Co., Inc.
- (Dupont) (under contract with NRC) on the water sample to determine the
i chemical form of the P-32 labeled material. The general procedures and
l results are described below.
j 3.1 NIH Tests
l Dr. Michael C. Cashel of the National Institute of Child Health and
j Human Development was contacted to identify the P-32 labeled material.
l His report is available at NRC's Public Document Room. Table E-1,
prepared by NRC. provides a summary of his experiments. He confirmed l
l the radioactivity of the water sample using Cerenkov radiation i
- measurements made with a liquid scintillation counter and determined the
i sample had a pH of approximately. 5. using pH paper (i.e.. was slightly !
i acidic). Dr. Cashel expected the water to be slightly acidic because
the dissolved carbon dioxide from the air forms carbonic acid to make ,
i normal water slightly acidic. Nucleotides are unstable in alkaline j
! conditions and are stable in acidic conditions. '
I
i Although his first experiment to determine the nature (i.e., organic or
'
inorganic) of the labeled material indicated it was an organic, later
polyethylene imine (P.E.I.) cellulose thin-layer chromatography tests
a
'
showed the major radioactive com]onent migrated with the same properties
as inorganic phosphoric acid. Tiere were two additional " tiny"
- identifiably radioactive components with different migration pronerties.
! but there was not enough of this material to allow further
identification.
$' His final conclusions were that the major radioactive component is in
the chemical form of inorganic phosphate, as judged by co-chromatography
on P.E.I. cellulose thin layers. However the radioactivity binds to
charcoal, which is typical of nucleotdyl phosphate and aty)ical of
- inorganic phosphate. Furthermore, there is no certainty t1e original
l chemical form was inorganic phosphate. He indicated the material could
be a dearadation product of a nucleotide labeled with P-32.
3.2 DuPont Test
AIT initially contacted DuPont for an independent evaluation of Dr.
E-2
_ _ . _ _ . _ _ . _ _ _. _ - _ _ _ .. _ __._. _ _ _ _ _ _ ._...__ _ _
l
l
Cashel's report and to determine if there were additional tests that
could be performed on the sample to identify the chemical form of the '
radioactive component. The response (Enclosure 1) confirmed that Dr. i
Cashel performed the same type of experiments they would have done. '
AIT also contracted with DuPont Medical-Products Department to do a mass ;
spectral analysis on the P-32 contaminated water from the drip tray.
Two separate mass spectral analyses were run; the first was on AIT's ;
submitted sample and the second was on Dupont's in-house water purified
by reverse osmosis to establish a background in the mass spectrum. The ,
experimental details, observations, and mass spectrums are provided in
Enclosure 2.
The AIT sample had three intense peak and background peaks not found in
the Dupont water sample. The masses of the intense peaks (135, 196. and
230 Daltons) were too high to be consistent with the molecular weights >
l 4 P 4
L of either inorganic
(inosine, phosphates (H P0. Na3 )0. K P0 ) or nucleic acidsTbeam
cytidine, uridine,ortbymidine.
in the sample was also determined to be so small that it would be very
difficult to detect by the Desorption Chemical Ionization Mass Spectral
technique used. Therefore. the peaks seen could be unrelated t
degradation products or impurities. ;
.
4. CONCLUSIONS
l The information currently available indicates the only radionuclide
detected in the water. sample was P-32.
I Neither the initial nor the final form of the P-32 material (i.e.. the
material put in the water cooler and the material retrieved from the
'
,
drip aan) was ever confirmed by the tests performed on the water sample ,
by NIi and Dupont. The tests performed by Dr. Cashel were the )
l appropriate tests to determine if the radioactive component were i
organic, inorganic, or specific radiolabeled nucleotide. Dr. Cashel
determined the final form both adhered to activated charcoal the way a
P-32 labeled organic molecule would, and migrated on P.E.I. cellulose
thin-layer chromatography like an inorganic phosphate.
Dupont's mass spectroscopy tests of the sample revealed a number of
unidentified peaks that are not normally found in water samples but
these peaks were neither common nucleotide, phosphoric acid, nor
pyrophosphate peaks.
I
l
! E-3 l
l
l
- ,
.
'
l
- i
l _ _ , _ _ _ _ '
_ _ _ _ _ _ _ _ _ _ _ _ __ __ __ - _ - _ _ _ _ _ _ - _ _
TABLE E-1
SUMMARY OF DR. CASHEL"S
ACTIVATED CHARC0AL AND THIN-LAYER CHROMATOGRAPHY EXPERIMENTS
Experiment Objective: To Procedure Results Conclusion
Number determine if the
P-32 labeled
material is:
0 Measure Cerenkov 166 cmp of Cerenkov Cerenkov measurement of
radiation by liquid radiation (approximately activity agreed with NIH
scintillation 332 dpm) conventional liquid
counter and measure pH approximately 5 scintillation counting.
pH. The water solution was
slightly acidic.
1 In an inorganic Pass water sample Majority of P-32 activity P-32 is most probably
or organic form through activated remained in activated tagged to an organic
charcoal filter. charcoal, as expected for molecule.
an organic molecule.
2 Tagged to a 0.85 molar KHJ)0, Majority of P-32 activity Eliminated possibility P-
triphosphate solvent one- migrated with the pH 32 material was ATP. CTP.
nucleoside dimensional front in region expected GTP or UTP.*
Polyethylene imide for pyrimidine
cellulose thin- monophosphates.
layer
chromatography.
E-4
___________-_- ___- - ___-________ - __________ - _ - _ _ _ _ _ _ - _ _ - - _ - - - _ _ _ _- . - - - - -_. __ __ _ - _ - _ _ _ _ _ _ _ _ _ _ - _ _ - _ _ - _ _ - _ - -
._
TABl.E E-1 (CONTINUED)
Experiment Objective: To Procedure Results Conclusion ,
Number determine if the
P-32 labeled
material is:
3 Tagged to Two-dimensional Majority of P-32 ran Eliminated the
triphosphate polyethylene with first solvent front triphos) hate ribo- and
ribo- or cellulose thin- ahead of the pH front and deoxyri )onucleotides.
deoxyribo layer faster than either
nucleotides chromatography with triphos) hate ribo- or
ammonium formate deoxyri )onucleotides. ;
borate as first
solvent and 0.85
molar KH,PO, as
second solvent
4 Tagged to a 0.4 molar lithium Majority of P-32 activity P-32 material migrated in
monophosphate chloride solvent ran just below CMP *. with position expected for '
'
one- dimensional two very small amounts inosine monophosphate.
polyethylene imine below 2'3'GMP* and at the The small amount below
cellulose thin- origin. 2'3'GMP may be a di- or
layer tri-phosphate.
chromatography
i
E-5
_ _ _ _ _ _ _ _ _ _ . ________ - _ __ _ __- _____
_ _ _ _ _ _ - _ _ _ _ _ _ _ _ .__ _ _ _ _ _ _ _ _ _ _ _ _ _ - _ _ _ _ _ _ _ - _ _ _
TABLE E-1 (CONTINUED)
Experiment Objective: To Procedure Results Conclusion
Number determine if the
P-32 labeled
material is:
5a In IMP.* Tested 3.3 molar AMP. IMP. IMP and 2'3' AMP Procedure was not used.
ammonium formate and 2'3'GMP could be
and 4.2 percent resolved, but migration
borate (pH 7) in 1 differences were not
to 5 ratio solvent large.
one-dimensional
polyethylene imine
cellulose thin-
layer
chromatography for
separation of AMP.*
Sb In IMP. 0.4 molar NaFormate Better system for P-32 migration confirmed
solvent one- resolving IMP from CMP material was not AMP. CMP.
dimensional and AMP: GMP. or IMP.
polyethylene imine majority of P-32 material Material is clearly not a
cellulose thin- ran behind IMP. but ahead common nucleotide.
layer of GMP, with two small
chromatography amounts at origin and
just above origin.
t
E-6
n_ _ _ = _ _ _ _ _ . _ . _ _ _ _ _ _ _ _ _ _ _ - _ _ _ _ _ _ _ _ _ - _ - _ _ . . _ _ _ _ _ - - - . __ __
_ _ - _ _ . ._____ __ ________-__ ___ _ _ _.
TABLE E-1 (CONTINUED)
Experiment Objective: To Procedure Results Conclusion
Number determine if the
P-32 labeled
material is:
6a ortho- Run authentic P-32 Authentic P-32 labeled Authentic P-32 labeled
phosphoric acid labeled H3 PO, in 0.4 H2 PO, migrated to the same H PO, migrated with same
3
molar formate location as the major properties as the major
solvent activity of the P-32 activity of the water
polyethylene imine water sample in sample.
cellulose thin- experiment 5b.
layer
chromatography.
6b ortho- Run authentic P-32 Authentic P-32 labeled Authentic P-32 labeled
phosphoric acid labeled H3 P0, in 0.4 H PQ, migrated to the same
3
H3 PO, migrated with same
~-
molar LiCL* solvent location as the major properties as the major
one- dimensional activity of the P-32 activity of the water
polyethylene imine water sample in sample.
cellulose thin- experiment 4.
layer
chromatography.
- ATP - Adenosine tri)hos) hate
- CTP - Cytosine triplosplate >
- - Gaanosine triphosphate
'
GTP
- UTP - Uracil monophosphate
- CMP - Cytosine monophosphate
- GMP - Guanosine monophosphate
- IMP -
Inosine Monophosphate ,
- AMP - Adenosine monophosphate ,
- LiCL - Lithium Chloride
E-7 l
s
.
'
. _ _ _ __ . - _ . .
r
..
APPENDIX E
D
. . . .
F
l *?f 'j) 5
'
. i
l ~ N'. ' ,
,) ." 34 e ?/ 0
j .' / '
- ,
s %y7 l
'
7 26
,, 4
,
,,-
,
30 \ 4
33
I
~
s '
,
) 2
Ig
,
2~
.1 p ,
s
.
'
I
(k ,
'
!
l ,
{ 's s
s
'
I, :. L 14
\ \ p \
ft
.'
I( - - s
cq.
..- .(: ;
4- .
l -
, s
I f
I s i e
'
I
'
> 8
l
,/ ,, 16
' i
, s %
' N
's. -3
's. \
s $
'
's 33 '
19 - 22
f
!3
l l
l
'
11
!l ' ' 10
I LE rA
7 2 / 2
m ,,
~% 2,
x ,,
c
FIGURE E-1 j
- {
- . . _ __ _ _. . _ _ _ _ __. _ _ . _ . _ . _ _ _ _. . _ . _ _ _ _ _
APPENDIX F
i
,
BUILDING 37, STH FLOOR SURVEY
flame Position Title
Work address l
Lab / Branch,Section Telephonc# 1
Rooms in Bldg. 37 used Radiation Safety Branch User ID#
Supervisor
Authorized users who order radioactive materials that you
use
Radioactive Materials (Isotope, Chemical compound) Used in the Past Two Months
Have you ingested water (drinking, tea, coffee, soup, etc.)in the past 4 weeks from the Bldg. 37,5th floor
water cooler, f4W corner, between C and D corridor (Old Georgetown Rd side of 37)?
Have you shared tea, coffee, soup, etc. made with this water with anyone else?
Please provide names
Please indicate on the calendar the following: (1) Mark with an X days when you definitely could not have
ingested water (out of town, etc.)
(2) Estimate total water consumed for each of the days on the calendar in standard conical paper cups, in
ounces, or in ml l
Monday Tuesday Wednesday Thursday Friday Saturday Sunday
June 19 June 20 June 21 June 21 June 23 June 24 June 25
Monday Tuesday Wednesday Thursday Friday Sunday
Saturday
June 26 June 27 June 28 June 29 June 30 July 2
July 1
Monday Tuesday Wednesday Thursday Friday Saturday Sunday
July 3 July 4 July 5 July 6 July 7 July 8 July 9
Monday Tuesday Wednesday Thursday Friday Saturday Sunday
July 10 July 11 July 12 July 13 July 14 July 15 July 16
Have you changed any water bottles (removed or installed)in the 5th floor, NW corner water cooler in the
past 4 weeks? Have you noticed any bottles tampered with in any way? -
Have you had any visitors in the past 4 weeks who may have used the water cooler?
Date(s) of visit (s)
Please attach any additional comments you would like to make.
Please place the survey in a scaled envelope and give to your administrative of t'ce or to your radiation
safety officer.
. - - - - - - - . - - - . - - . . _ _ - - . -.. - - - . - .
i
l APPENDIX G l
1
INGESTION AND COMMITTED EFFECTIVE DOSE EQUIVALENT ESTIMATES
1. BACKGROUND
i
l A total of 27 people working on the fifth floor of Building 37 at the
i
'
National Institutes of Health (NIH) were identified by either direct
measurement or urine bioassay specimens as having detectable uptakes of
3hosphorus-32 (P-32). Ingestion was assumed to be the intake pathway,
]ecause of the measurable P-32 uptake in individuals who did not work or
frequent areas with P-32 use and the lack of evidence of widespread
laboratory contamination. The actual time of ingestion or chemical form
of the P-32 is unknown. Available bioassay data were used to estimate
the initial uptakes and the committed effective dose equivalents.
The first person identified with P-32 u]take was a pregnant researcher.
She also had the largest body burden. 9IH's Radiation Safety Branch
staff focused its initial attention on her activities June 29, 1995.
(the date she was discovered to be contaminated), and on radiation
surveys of her work area, home, and car. Physical evidence from
contaminated clothing indicated the most probable date of ingestion to
be June 28, 1995. The most probable time of ingestion was from mid-
morning through early evening that day, because of her presence at NIH
at those times. For this reason, NIH used 11:00 a.m. June 28. 1995, as
the initial ingestion time for the 27 individuals involved to calculate
their committed effective dose equivalents. This date may be very
conservative for some of the individuals.
NIH identified a radioactively contaminated water cooler in the hallway
of the fifth floor, as well as contaminated mugs and coffee pots on i
July 14, 1995. (The cooler was taken out of service at this time.)
Initial results of water, smear, and swab samples indicated only the
presence of P-32 contamination in the water in the drip pan and on the
surface of the water cooler, the metal reservoir and the drip pan under
the faucets. The water in the water bottle did not have measurable 4
contamination. l
Groups of individuals having P-32 u)take were linked to the contaminated 1
coffee pots and mugs. Water from tie contaminated water cooler was used ,
to make the coffee and tea in those pots and mugs. NIH distributed a
questionnaire to personnel in Building 37 to obtain data on water cooler
use (Appendix F). All individuals with measurable P-32 uptakes reported
drinking water from the contaminated water cooler between June 28 and
July 14. However, a number of individuals who reported drinking large !
amounts of water every day from the cooler did not have measurable P-32
uptakes.
l
1
!
G-1
- - -.
--.-.- ---- - . - ~ - - - . - - - - _ - - - - - .
i
'
l
'
1
4
Questionnaires were reviewed for 42 individuals without measurable
,
uptakes who indicated they used water from the cooler. Twenty-three
i individuals with measurable uptakes were interviewed about their water
,
use habits. No discernable )attern was found that could explain why
,
some individuals had measuraale uptakes and others did not. Those with
j uptakes drank water from the cooler early in the morning, midmorning, at !
j lunch, in the afternoon, and early evening. They also ranged in amounts j
,
consumed from one or two small glasses to many liters.
-
l 2. COMMITTED EFFECTIVE DOSE EQUIVALENT ESTIMATE FOR THE PREGNANT RESEARCHER 1
"
The most extensive set of bioassay data was collected from the pregnant
>
researcher. To estimate the amount of P-32 she ingested, her committed
l effective dose equivalent (CEDE), and the total effective dose
- equivalent (TEDE) to her fetus, it was important not only to obtain as
- many bioassay samples as possible but also to obtain both retention and-
1 excretion bioassay data. Not knowing the form of P-32 ingested, the
'
amount ingested, or the time of ingestion complicated these calculations
, and increased the uncertainty of the estimated values. 1
i Preliminary committed effective dose equivalent estimates for the
l pregnant researcher made from excretion bioassay data collected in the
1 first week were too low to account for the amounts of P-32 that
! continued to be excreted in her urine a month later. All estimates are
4
also affected by the lack of bioassay data for approximately the first
i
day and a half and early 24-hour urine samples. Once the researcher was-
identified as having an uptake, urine samples were collected for
, analysis but the first true 24-hour urine sample was not obtained until
4
day three. By this time the researcher was actively pushing fluids and
- eating a diet high in phosphorus in an attempt to remove as much
4
radioactive phosphorus from her system as possible. Whether this action
4 was effective or affected the initial bioassay samples is unknown. A
- total of 25 urine samples (including eight true 24-hour samples and one
- 12-hour sample) were collected during the next month. There were also
- three different retention bioassay studies.
2.1 Retention Study 1
,
Approximately two and a quarter days after the ingestion, whole body
. bremsstrahlung images were made at the NIH Nuclear Medicine Department.
l This procedure was repeated six days later. The physician performing
4 the study noted that, due to the difficulties in detecting the P-32 beta
i radiation and its corresponding bremsstrahlung, the estimates obtained
from this study were crude estimates. The pregnant researcher was
2,
estimated to have a total of 31.9 MBq (862 microcuries) of P-32 retained
i at the time of the initial scans and 12.7 MBq (343 microcuries) when the
second set of scans were made. Most of the activity in the first scans
(estimated to be 15.1 MBq (409 microcuries)) was in the chest and upper
. abdomen, in the region of the liver.
!
'
A Biad dual-headed gamma camera with a medium energy collimator and
'
energy window of 100 (+/- 50%) kiloelectron volts was used to take the
i
j G-2
2
-, - - - - .-.
_ . _ _ _ _ _ _ _ _ _ _ . . _ _ - _ _ _ _ _ _ . _ _ _ _ _ _ _ > _ . _ . _ _ _ _ . _
.
c
o
{
- images. Three images were obtained, i .e. , images of the skull, chest'
! and up)er abdomen, and r.id-abdomen down to the lower thighs. A 5-gallon
, '
water ]ottle with 128 MBq (3.45 millicuries) of P-32 in aqueous solution
was used as a phantom to quantify the bremsstrahlung images. An earlier-
- image with the same phantom containing 4.9 MBq (132 microcuries) of P-32
was taken-but it did not have sufficient activity. The anterior and ;
posterior images were summed and smoothed for the researcher, phantom. '
i
and background counts. Background was subtracted from the patient ,
! images and the activity in the field of view was compared with the l
j phantom to quantify the results. l
.
I 2.2 Retention Study 2 i
'
i
i The NIH Department of Nuclear Medicine performed the second retention
i bioassay study immediately after the second set of scans. The
i researcher was counted at a distance of 5.1 meters from a probe with a !
!
sodium iodine crystal that was normally used for iodine bioassay
i measurements. The measurement was made with an open window. A
l background count was taken with the same configuration. A retention
i
estimate of 452 microcuries was obtained with this procedure. The same
phantom with 3.45 millicuries of P-32 in aqueous solution was used to
4
quantify the results. NIH cautioned that the results of this study were
only expected to be a crude estimate of the P-32 activity retained by
i the researcher.
.
l 2.3 Retention Study 3
4
"Whole body" gross counts were obtained from the pregnant researcher by
L having her stand 15 inches from the Radiation Safety Branch's Canberra
lithium-germanium. detector. The detector was centered at the
. researcher's naval which placed the researcher's trunk in the detector's
- field of view. Ten minute counts were made on June 30. July 6. July 13.
and July 20. The resulting gross counts were 280110, 170903. 107949.
- and 73479, respectively, for the single peak with a height of 1179. 714.
! 443, and 338 counts. The center of this peak was at approximately 140
kiloelectron volts. A characteristic x-ray was observed at 69
i kiloelectron volts. NIH did not quantify these results,
i 2.4 Blood Specimen Study
f Two whole blood specimens were drawn: the first at the hospital the
j night of June 29 and the second during the afternoon of July 30. The
-
June 29 specimen contained 4.400 disintegrations per minute per
. milliliter of blood. The June 30th specimen contained 5.577
1
disintegrations per milliliter. An aliquot drawn from the clear portion
of the settled June 30 specimen had 7.800 disintegrations per milliliter
when counted again July 3 and decay corrected to June 30.
l 2.5 NIH Estimates from Excretion Data
NIH staff estimated the pregnant researcher had an intake of 500
!
microcuries and committed dose equivalent of 4.17 rem. The pregnant
i G-3
.
,r - -
. . _ _. _. ._ _ _ _ _ _ _ _ . _ _ . _ . _ _ _ _ _ _ _
researcher's fetus was estimated to have received a' fetal dose
equivalent of 3.2 rem (Table G-1). These estimates were based on 12
data points representing 24 hour2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> urine excretions of P-32 Table G-2.
Seven of these data points were obtained from true 24 hour2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> urine
specimens. The remaining five data points (i .e. . days 2. 3, 4, 9, and
21) were estimates derived from measured urine specimens made by
adjusting the values of the volumes to construct 24 hour2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> intervals.
l The day-two data point was obtained by increasing the spot sample volume
! from the 14 milliliters collected to a value estimated for an 8-hour
I collection time. NIH staff assumed the researcher had a 24-hour urine
I output of 3200 milliliters. The day three and day four data points were
derived by splitting the activity determined from a 24-hour urine sample
between two days and scaling each value to complete the two 24 hour2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br />
! periods. The volumes were adjusted for two urine samples to estimate
the day nine data point. Finally the volume from the 12-hour urine
sample was scaled up to an expected 24-hour volume to obtain the day 21
data point.
A scientist at NIH stated that organic materials labeled with P-32 would
be expected to be degraded to inorganic phosphate after ingestion. NIH
staff used the ICRP 30 model for inorganic P-32 to calculate the amount
of unknown P-32 initially ingested.
The estimated ingestion amount was derived using the intake retention
fractions, "f" values from NUREG/CR-4884, " Interpretation of Bioassay
Results." The unweighted least squares fit recommended by NUREG/CR-4884
!
for multiple samples gave an estimate of 11.1 MBq (300 microcuries).
The weighted least squares fit method of calculations, recommended by
Skrable, et al . *(Skrable, et al. Chapter 14. "Use of Multiple
- Compartment Models for Retention for Internally Deposited
Radionuclides"), gave an estimated intake of 18.5 MBq-(500 microcuries).
NIH staff elected to use the weighted intake estimate as the basis for
calculating the pregnant researcher's effective dose equivalent and the
fetal dose equivalent (Table G-3). NIH staff stated the additional
calculations using INDOSE **(Skrable Enterprises Inc.) verified the
results. The dose estimate is calculated from the 22.2 MBq (600
microcurie) annual limit of intake that would give reference man a dose
of 50 mSv (5 rem).
2.6 Radiation Internal Dose Information Center (RIDIC) Estimates from
Excretion Data
Oak Ridge Institute for Science and Education's RIDIC staff provided AIT
with an analysis of the NIH bioassay data and dose determinations, as i
well as the RIDIC dose estimates (Table G-1). The 30.3 MBc (820 ;
l
microcuries) RIDIC estimate of initial ingestion committec effective 1
'
dose equivalent of 80 mSv (8 rem), and fetal effective dose equivalent
of 52 mSv (5.2 rem) was greater than the NIH estimates. ,
1
Although derived from the same raw rta, the RIDIC and NIH estimates i
differed in the way the data were used. RIDIC staff evaluated the '
G-4
l
l i
1
'
\
1
- _ __ -. _ . _ . __ _. ._ _._._ _ _ _ _ _ . . .
results _ from the first-14 bioassay samples and determined that the
uncertaintles associated with assigning either activity or volume for
the 24-hour Seriod did not warrant using these data in determining the
initial intate. Further, only two of the ten data points RIDIC used
were estimated. The remaining eight data points were actual 24-hour
urine specimen results (Table G-2).
' RIDIC staff adjusted the volume of the two urine specimens to estimate
the day nine data point. The estimated 24-hour urine volume of 3000
milliliters was lower than the 3200-milliliter value estimated by NIH,
_ This difference is not significant and only tended to lower the RIDIC
l
value for this data point. The day 21 data point was estimated by
,
'
doubling the activity of the 12-hour urine specimen (NIH scaled this by
a factor of 2.4). Again this difference only tended to lower the RIDIC
l value for the data point.
RIDIC used the 10 data points to estimate the biological half-life of
P-32 and compare this with the long-term clearance half-time of the ICRP
model for inorganic phosphate. The estimated value was 18.3 days and
the ex]ected value from the model was 19 days. RIDIC staff concluded
'from tais agreement that application of the ICRP 30 model for inorganic
- phosphate to the data was appropriate for determining the initial oral
l ingestion.
The 24-hour urinary excretion data in NUREG/CR 4884 for an oral intake
of P-32 were used with the observed urinary excretion activities.
!
'
Weighted and unweighted least squares estimates were calculated
(Table G-3). RIDIC staff concluded the weighted estimate provided the
best overall fit to the data. NIH's review of the RIDIC data showed
that the results varied depending on whether the NUREG/CR 4884 data were
corrected to remove consideration of radioactive decay. Decay
- correction gave a ingestion value of 30.3 MBq (820 microcuries). An
ingestion value of 27.4 MBq (740 microcuries) was obtained without decay
correction.
2.7 Lawrence Livermore National Laboratory (LLNL) Third Party Estimates from
Excretion Data
NRC contracted with the Internal Dosimetry Assistance Team of the
Fission Energy and System Safety Program at LLNL to provide an
independent third-party estimate and review of the ingestion and dose
assessment estimates from the bioassay data collected by NIH.
Using the urine bioassay data, the LLNL estimated the ingestion to be
40.7 MBq (1100 microcuries): Researcher A's committed effective dose
equivalent to be 108 mSv (10.8 rem): and the fetal effective dose
equivalent to be 69 mSv (6.9 rem). LLNL averaged available multiple
measurements, decay- corrected measurements from the time of analysis to
. the midpoint of the sample collection period, and pooled results from
l some specimens, to form 24-hour samples. l
1
The above urine bioassay estimate was obtained from nine data points
l G-5
. . -
. -- . . .. . . - .
(Table G-2). Like the RIDIC approach. LLNL elected to exclude the ;
bicassay data of the first 14 urine specimens. Unlike the RIDIC l
approach LLNL assigned the 24-hour specimen to the midpoint of the
'
collection period and the nearest tenth of a day (RIDIC assigned it to
the end of the last day of the collection period): corrected each data
point to 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br />; and elected to merge urine s)ecimens 18 and 19 with
the previous 24-hour urine sample and correct t1e value back to 24
hours. The net result however was to have a data set almost identical ,
to the RIDIC data set and the NIH data set for data points higher than ;
four.
The ICRP-30 metabolic model for inorganic P-32 was used to estimate the 1
initial ingestion. The computer program " Code for Internal Dosimetry" :
(CINDY) was used to statistically evaluate the data. The four !
statistical methods used by the program are the ratio of the means. '
average of the slopes, unweighted least squares, and a user-defined
weighted least squares analysis using inverse weights. All four l
techniques give the "best fit" but the plots of the data and best-fit !
curves showed that the " average of the slopes" provided a better
statistical and visual fit to the data.
The CEDE estimate was obtained by integrating the area under the best-
fit curve for the average of the slopes. The fetus is estimated to
receive 0.0078 rem for each microcurie of P-32 in the maternal transfer
compartment. The microcuries of P-32 in the maternal transfer
compartment is 80 )ercent of the ingestion estimate. The RIDIC
estimates for intace. CEDE, and fetal dose fell within the ranges
estimated by the four statistical best-fit methods.
LLNL also calculated 24-hour urine values for urine samples 1 through 2
and 3 through 14 and assigned these values to days 1.4 and 1.9 (Table
G-3). The time ratio used to calculate the 24-hour urine values assumes
the rate of radioactive excretion for the measured data is the same as
that for the unmeasured period. This calculation was done in an attemat
to use all the urine bioassay data. LLNL was able to show that the nil :
results were within the range of estimates it obtained for the four '
statistical methods (2.1 to 33.3 Mbq (58 to 900 microcuries) ingested).
None of the methods gave a good "best fit" for all the data points.
l
l
G-6
l
_ _m._.__ .. _ _ _ _ . _ ... _ _ _. _ .._ _ ._ _ ___ _ . _ . _ _ _ _
'
LLNL also evaluated NIH's 12 data points by using three of the four '
statistical evaluations methods included in CINDY. (A weighted least
squares evaluation was not performed because NIH did not provide the
measurement error values which are used for the inverse weights by
CINDY.) The estimated ingestions were 16.3 MBq (440 microcuries) for
the unweighted least squares (630 microcuries) for the ratio of means. 1
and 37 MBq (1000 microcuries) for the average of the slopes evaluations. *
The average of the slopes method gave the best visual and statistical
I
fit. The NIH weighted least squares estimate of 18.5 MBq (500
microcuries) was within the range of values estimated by LLNL but the *
NIH unweighted least squares estimate fell below this range. i
.
'
2.8 Lawrence Livermore National Laboratory LLNL Estimates from Other
Bioassay Data l
LLNL evaluated the Nuclear Medicine whole body scan and count data
in two ways (Table G-3). The first method of evaluation was to
take the three points (i.e.. one data point from each whole body
scan and one data point from the whole body count) and use CINDY
w make a )rojection cf the bioassay results after intake. LLNL
averaged tie projected intakes and rounded the number up to 48.1
MBq (1300 microcuries). The estimated dose equivalent to
Researcher A from this ingestion was 127 mSv (12.7 rem) and the
!
estimated dose to the fetus was 81 mSv (8.1 rem).
The second method was to use first principle calculations to
j calculate the ingestion from the measured 31.9 MBq (862
t
microcuries) retained in Researcher A's body on June 30. The
first principle calculation uses the effects of radioactive decay
and biological elimination to determine a retention fraction.
This resulted in an estimated ingestion of 38.8 MBq (1050
microcuries), effective dose equivalent for Researcher A of
103 mSv (10.3 rem) and fetal dose of 65 mSv (6.5 rem).
LLNL also tried to estimate the total P-32 ingested by calculating
the total systemic P-32 in the whole blood volume for reference
woman using the radioactivity measurements made from the two blood ,
samples collected June 29 and June 30. The estimated P-32
ingestion was 3.4 MBq (92 microcuries). LLNL did not use the
blood data in determining Researcher A's or her fetus's dose when
they concluded the radioactivity measurements had little
credibility because of the inherent difficulties in counting blood
samales. This was demonstrated by the large chemiluminescence
peat in the first blood sample. LLNL believed this indicated
chemicals in the blood samples may have interfered with the liquid
scintillation counter measurements and resulted in lower estimated
L blood activities.
l
G-7
.
- - , , _- - .- -_ . - - -
_ _ . _ - _ _ _ _ _ _ _ _ _ . _ _ _ _ _ _ _ _ _ . _ _ _ . _ .
i I
- 1
I
3. ESTIMATED DOSES FOR OTHER INDIVIDUALS WITH MEASURABLE P-32 INTAKES
i
Twenty-six individuals had estimated ingestions that ranged from 0.074
to 1.4 MBq (2 to 37 microcuries). One individual, originally estimated ;
to have ingested 1.4 kBq (37 microcuries), was out of town from June 28 :
to July 7. and may have ingested only a maximum of 14 microcuries. if '
the ingestion occurred after his return. NIH staff calculated
committed effective dose equivalent values from the available bioassay '
data for each individual took the highest estimated value, and assigned
a range for the estimated dose. The ranges were in 0.5 mSv.(50- l
millirem) increments from 0 to 3.5 mSv (0 to 350 millirem). There were .
18 individuals with estimated doses less than 1 mSv (100 millirem), and ;
8 with estimated doses above 1 mSv (100 millirem) but less than 3.5 mSv 4
l (350 millirem). (These number remain the same if the individual with j
'
the 1.4 kBq (37-microcurie) ingestion had a maximum ingestion of 0.53 ;
MBq (14 microcuries)).-
AIT's scientific consultant evaluated the bioassay data from these
individuals and estimated their ingestion and committed dose
equivalents. The consultant used the ICRP-30 model to estimate
ingestions from the bioassay data available for each individual. When ;
multiple urine samples were available, multiple ingestions and dose
estimates were made. The resultant dose estimate was reported both as :
the arithmetic mean of the estimates and the weighted least squares
estimate. A logarithmic interpolation was made to determine the
a)propriate intake retention functions (IRF's) from NUREG/CR-4884. for
t1ose days post ingestion not explicitly included in the tables.
<
Although AIT's scientific consultant developed a table of single ,
estimated ingestions and doses. AIT staff censiders the NIH ranges to be l
the more a]propriate representation of the doses to these individuals. 4
based on tie considerations discussed in 4.4 below. j
There were five non-radiation workers among the 26 individuals with i
measurable P-32 uptakes. One of these had an estimated committed dose
equivalent of 2 mSv (200 millirem) based on a 24-hour urine sample. The
estimated committed effective dose equivalents for these individuals
ranged from 0.25 to 2 mSv (25 to 200 millirem). -
Initial liquid scintillation results for urine samples provided by 1
30 individuals (in addition to Researcher A) were above the lower '
limits of detection NIH's Radiation Safety Branch set for tritium. '
carbon-14. and P-32 energy windows. On subsequent counting of the
same or new urine samples. 2 were determined to be below the lower !
limits of detection for the tritium and carbon-14 energy windows. l
One of the individuals was removed from the list of P-32 Jositives i
when the urine sample had the same activity days later. ilH staff j
concluded that if the individual were contaminated with P-32, the ;
counts would have been lower due to radioactive decay. NIH '
l assumed this individual had elevated levels of potassium-40 or ,
other long-lived, naturally occurring radionuclide. NIH concluded :
G-8
!
,
,,e-- , , , - , -
__ _ _ _ _ _ _ __ _ __ _ _ _ ._ _ _ _ _. _ _ . - _ _ _
i
,
the urine sample with elevated counts in the tritium window !
demonstrated very high chemiluminescense and was not considered
positive.
'
4. DISCUSSION
l
4.1 Relationship between NIH, RIDIC, AIT, and LLNL Evaluations ,
NRC contracted with LLNL to provide independent dose calculations based i
on the NIH bioassay data. Previous ingestion and dose estimates
l performed by NIH and AIT's scientific consultant involved initial joint i
! collaboration and resulted in different final estimates.
l
l NIH and AIT separately requested assistance from the RIDIC scientific
consultant in evaluating the scope of the ingestion and the dose :
evaluations for the pregnant researcher and her fetus. Although the '
scientific consultant was in communication with NIH, NIH elected to
perform its own evaluation of researcher's A's ingestion and resulting i
doses. Further, communications between NIH and RIDIC resulted in
refinements to the NIH estimates, but NIH did not agree with the ,
scientific consultant's exclusion of the June 29 and June 30 bioassay :
data points and the RIDIC final ingestion and dose estimates. Since the '
two groups;ap) eared to have different estimates. NRC contracted with a
third party (_LNL) to provide an inde)endent estimate of the ingestion
and doses and independent review of t1e NIH bioassay data, the NIH
estimates, the RIDIC estimates.
4.2 Errors Associated with the Time of Ingestion
NIH assumed the time of ingestion was 11:00 a.m. on June 28, 1995.
(i .e. , lunch time). If the ingestion occurred at mealtime it could also
have occurred at dinnertime (5:00 to 6:00 p.m.). NIH, the scientific
consultant, and the independent third party all used the 11:00 a.m. time
as the basis for their calculations.
The scientific consultant assigned 24-hour urine data to days post
ingestion and did not distinguish between fractions of a day.
Therefore.. a shift in the time of ingestion by 6 hours6.944444e-5 days <br />0.00167 hours <br />9.920635e-6 weeks <br />2.283e-6 months <br /> in the same day
would not have affected his assignment of post ingestion days or his
evaluations. LLNL assigned 24-hour urine data to the nearest tenth of a
day with the days calculated from the time of ingestion. A 6-hour
change in the ingestion time would not have made a difference in the
ingestion estimated from the urine data: resulted in a 2 percent
decrease in the estimate from the nuclear medicine quantified
bremsstrahlung scan performed June 30. 1995: and a 7 percent decrease in
the ingestion estimate from the combined scan and whole body count data
set. These differences are not significant, and the 11:00 a.m. time
provides conservative estimates for the assessments.
4.3 Errors and Assumptions Associated with Assigning 24-Hour Values to Data
Collected for Shorter Periods
G-9
_ _ __ _ _ - . - _ __ _ ,
_ _ _ _ . _ _ _ _ . _ _ . _ _ - . - _ _ _ _ _ _ _ _ _ _ _ . _ _ _ . _ _ _
.
i
! ,
1
As seen in the radioactive concentrations measured in Researcher A's
l first 14 urine specimens, there is considerable variation between the .
! radioactive concentrations excreted in consecutive urine collections. !
[ (Considerable variability continues to be seen in the later 24-hour
- urine. specimens.) Further, the two short half-life components of the !
4
ICRP-30 inorganic phosphate model have their greatest effect on ;
- radioactive excretion in the first 5 days. This part of the model has t
, the greatest rate of change in excreted radioactivity. Both of these '
+ effects, in addition to the large uncertainty associated with assigning
an appropriate 24-hour urinary output for Researcher A in the first
'
,
l week s post ingestion, raise questions about the appropriateness of f
,
calculating 24-hour urine excretion values for the first 5 days after '
.
ingestion.
!
- The inherent problems associated with assigning P-32 excretion values to
i the first few days can also be seen in LLNL's assignment of 1.5 and 2.1
1 microcuries to the same days NIH staff assigned 9.68 and 9.1 ,
i microcuries. '
!
I In defending its estimated intake value. NIH staff places considerable -
! weight on the argument that its estimate is correct because the high
4
P-32 excretion rate expected by the RIDIC and LLNL estimates are not
-
seen in the early data points. If, in fact, the first three data points
l were true 24-hour urine specimens, this argument could be justified, but j
i they are not. As the RIDIC reports point out, there is too much {
- uncertainty associated with Researcher A's urinary output and rate of ;
! radioactivity excretion to assign values for these dates. The lack of l
j sensitivity in the NIH approach also is demonstrated in its assignment '
j of 11.7 microcuries to day 4, because some (7/24) of the true 24-hour ,
specimen was collected on day 4. This resulted in assigning all the
'
.
)
activity to day 4, because the rate of radioactive excretion was assumed
j to be constant.
- 4.4 Use of Ranges vs. Numerical Values for Committed Effective Dose
- Equivalent Values for the 26 Individuals with Detectable P-32 Uptakes
!
'
Both NIH and the scientific consultant calculated values for the
! estimated committed effective does equivalents. These estimates were
, based on a number of standard assum3tions, such as time of ingestion, a
single ingestion, and accuracy of tie data.. These assumptions may have
resulted in considerable uncertainty associated with calculating their
committed effective dose equivalent estimates because the assumptions
1
probably did not accurately reflect the actuai P-32 ingestions for these
individuals. Some individuals. especially those in coffee groups,
probably ingested P-32 on more than one occasion. The P-32 model used
-
for the estimates is designed for single uptakes. The majority of
i estimates were made from spot samples with reference man or woman
urinary output values assigned for the 24-hour volumes. Only 10 of
- these individuals were asked to give 24-hour urine specimens, and in all !
! cases, the actual urinary output was considerably different from !
!
reference man or woman. The differences varied from 200 to 1500 l
j milliliters (with an average of 670 milliliters) from the appropriate ]
1
'
i'
G-10
l
. )
)
-_ _ . . _ _ _ _ _ _- ___ ,
reference man or woman value. When this degree of variability is added
to the inherent variability in radioactive excretion in a single urine
sample, the difficulty in assigning a more precise number is obvious.
Additionally, if the exact time and number of P-32 ingestions were
known, the estimated doses would probably have been lower.
Reporting.the committed effective dose equivalents as a range more
accurately emphasizes the uncertainties associated with the estimates.
The maximum value estimated by NIH was use<J.to determine which group the
individual was placed in. Although the maximum value may not accurately
reflect the true dose in all cases it counterbalances underestimates
derived from lower radioactive excretion rates in specific grab samples
and the problems associated with multiple ingestions that may not be
seen in the lack of data between collection periods.
5. CONCLUSIONS
5.1 Best Estimates for P-32 Ingestion and Dose Assignments for Researcher A
and Her Fetus
Based on the best available data and the results of two
evaluations of the data by AIT's consultants. AIT has determined
the best estimates for Researcher's P-32 ingestion to be 39.6 MBq
(1070 microcuries), her committed effective dose equivalent to be
104 mSv (10.4 rem) and the effective dose equivalent to her fetus
to be 66 mSv (6.6 rem).
AIT concludes that until a more definitive time of ingestion is
determined. 11:00 a.m. June 28. 1995, is an appropriate time to use for
these assessment.
AIT agrees with its consultants that the dose to Researcher A
should be based on reference woman models. Researcher A's weight
of 53 kg indicates she is closer to reference woman's 57 kg weight
than reference man's 70 kg weight.
AIT agrees with it's consultants that the most appro)riate urine
bioassay data base should exclude the calculated 24-lour urine
values derived from the urine specimens collected June 29 and 30.
This data should be excluded because the uncertainties associated
with Researcher A's excretion of P-32 and urinary output during
this period are too great to permit confident and realistic
estimates for the resultant data points. Further, these
unreliable initial data points would carry disproportionate weight
in all subsequent calculations and evaluations. On the other
hand, the majority of the remaining data points provide actual
indications of Researcher A's P-32 excretion for 24-hour time
periods.
AIT concludes that the retention studies performed by the Department of
Nuclear Medicine support the ingestion estimates obtained by the
consultants from the urine bioassay data when the June 29 and 30
G-11
-- . -- - _.
.
. _ . . _ . _ _ _ _,_____. _ _.-_._._._ _. _ . _ _ _ . _ . _ _ . _
[
specimens are excluded. i
AIT recognizes that NRC regulations and guidance permit NIH to use any '
one of several methods to estimate Researcher A's ingestion of P-32 and
the corresponding doses to Researcher A and her fetus. In general. y
depending on the method selected, a licensee's results may show either -
compliance with the regulations or a realistic estimate of the ingestion
or ideally both. While recognizing the licensee's option of selecting
the methodology with respect to regulatory concerns. AIT has determined
that it is appropriate in this case to provide Researcher A with the
, best available estimate of her P-32 ingestion, her dose and the dose to
i
her fetus, c
5.2 Best Estimates for Committed Dose Equivalent Assignments to Other
Individuals
AIT concluded the committed dose equivalents assigned to the other I
individuals with measurable P-32 uptakes should be reported as a range '
for each individual. The assignment of the range for the ) articular
individual is based on the maximum dose estimate provided ay NIH. The .
ranges are to be reported in increments of 0.5 mSv (50 millirem) from 0 l
to 3.5 mSv (0 to 350 millirem). Although there is a high probability
that the individual with the highest dose may have received a lower dose
because he was not at NIH from June 28 through July 7 there is still
enough uncertainty in the exact time of the P-32 ingestion for each
individual prevent the dose from being lowered at this time. ;
l
1
l
, ,
G-12
1
- - . , - - . - - . , , . - - . . - - . , -. , - , ,
__
TABLE G-1
SUMMARY OF BEST ESTIMATES FOR RESEARCHER A AND HER FETUS
ORGANIZATION INGESTION CEDE CEDE FETAL DOSE
MBq REFMANmSv{ rem} REFWOMANmSv{ rem} EQUIVALENTmSv{ rem)
{microcuries)
NIH* 18.5 (500) 41.7 (4.17) 32 (3.2)
4
'
ORISE* (Scientific 30.3 (820} (68) ({6.8}) 80 {8.0} 51 (5.1}
Consultant)
LLNLo (Third Party)
Urine (9 Points) 40.7 {1100} (100) ({10}) 108 {10.8} 69 {6.9}
Whole Body
(1st principles) 38.8 {1050} (96) ({9.6}) 103 {10.3} 65 {6.5}
Whole Body
(Computer Model) 48.1 (1300) (120) ((12)) 127 (12.7) 81 (8.1) I
l
AIY 39.6 (1070) 104 (10.4) 66 (6.6) '
- NIH -
National Institutes of Health
- ORISE - Oak Ridge Institute for Science and Education
- LLNL - Lawrence Livermore National Laboratory
- Ref MAN -
'
reference man
- Ref WOMAN -
reference woman
mSv -
Millisieverts
MBq -
Megabecquerrel
- (N) indicates the value was calculated but not used for the best dose estimate
I
,
J
1
l
G-13
TABLE G 2
RESEARCHER A URINE EXCRETION DATA BASES
P-32 Activity (microcuries)
NIH* Sample No. LLNL* Time NIH ORISE* Time NIH ORISE LLNL
Description Assigned Assigned
(days) (days)
1-2 1.4 1.5
3-14 1.9 2.1 i
1-10 2 9.68
11-15 3 9.1
15 4 11.73
15 2.8 3 11.6 11.87
16 5.6 6 6.72 6.57 5.81
17 7.8 8 4.76 4.68 4.64
18 & 19 9 7.31 7.06
20 9.8 10 9.33 9.24 9.33
21 13.8 14 3.57 3.54 3.57
22 18.3 19 1.41 1.35 1.40
23 21.1 21 3.24 3.24 3.25
24 27.8 28 0.95 0.94 0.95
25 28.8 29 1.39 1.37 1.39
- NIH -
National Institutes of Health
- LLNL - Lawrence Livermore National Laboratory
- ORISE - Oak Ridge Institute for Science and Education
!
G-14
TABLE G-3
RESEARCHER A INGESTION AND DOSE ESTIMATES
DESCRIPTION
_
URINE DATA INGESTION CEDE CEDE FETAL EDE
MBq REF MAN mSv REF WOMAN mSv mSv(rem}
{microcuries) (rem) { rem}
NIH* (WEIGHTED) 18.5 {500} 41.7 {4.17} --
32 {3.2}
NIH (UNWEIGHTED) 11.1 (300)
ORISE* (WEIGHTED)(DECAY CORR) 30.3 (820) (64) ((6.4)) 80 (8.0) 51 (5.1) l
ORISE (WEIGHTED)(NOT DECAY CORR) (27.4) ((740)) (58) ({5.8)) (72) ((7.2)) (46)({4.6)) )
ORISE (UNWEIGHTED)(DECAY CORR) (22.9) ({620)) l
ORISE (UNWEIGHTED)(NOT DECAY CORR) (21.8) (!590))
l
l
LLNL* Best fit (WITH 11 P0lNTS) (33.3) ({9CO)) (83) ({8.3}) (88) ({8.8)) (56)((5.6)) l
LLNL Best Fit (WITH 9 CCINTS) 40.7 (1100) (100) ((101) 108 (10.8) 69 (6.9) l
NUCLEAR MEDICINE WHOLE BODY DATA
LLNL (First Principles) 38.8 {1050} (96)({9.6}) 103 {10.3} 65 {6.5}
LLNL (Computer Model) 48.1 {1300} (120) ({12}) 127 {12.7} 81 {8.1}
BLOOD SAMPLES
LLNL (8) ({0.8})
.
- CEDE - Committed effective dose equivalent
- NIH - National Institutes of Health
- ORISE - Oak Ridge institute for Science and Education
- LLNL - Lawrence Livermore National Laboratories
(N)-Indicates the value was determined but not used in the best estimate
G-15
- . . ._. _ _ _ _ _ . - _ _ . _ _ _ - _ - _ _ . _ _ _ _ _ _ _ _ _ _ . __
'
,
.
!
t
!
,
1
APPENDIX H !
SPLIT BI0 ASSAY SAMPLE PROCESSING AND DATA ANALYSIS
1. BACKGROUND -
The National Institutes of Health (NIH) provided the Nuclear Regulatory *
Commission (NRC) with three different sets of bioassay samples taken from .
urine or blood specimens. These specimens were provided by the NIH researcher !
with the highest ingestion of phosphorus-32 (P-32). from the June 1995 NIH '
P-32 incident. AIT obtained aliquots of these samples to perform an
independent analysis of a representative number of bioassay samples collected -;
from this researcher and to compare the results with NIH's radioactivity
determinations. !
In the first set of samples, two equal aliquots were taken on July 3.1995. !
from each of one blood and three urine samples. These split samples were
divided into two groups. The first was processed and counted at NIH by NIH i
Radiation Safety Branch staff. The second was shipped to Oak Ridge Institute ;
for Science and Education (0 RISE) for processing and counting. Each set of
samples was placed in 10-milliliter liquid-scintillation vials )rovided by
NIH. An additional empty vial was added to each set for a blanc. Table 1A
describes the samples provided July 3.1995.
In the second set. AIT received aliquots from four urine specimens on
September 1. 1995. This set was not prepared as split samples and there was
no new NIH analysis available for these samples. (NIH recounted, for other
purposes. some of the urine specimens included in the second set on August 29,
1995.) The split samples were processed and counted at NRC Region I. The
four samples contained three new specimens and one sample drawn from a i
specimen that was tested in the first set. This repeat specimen provided a
limited means of comparing all three (i.e. NIH, ORISE, and AIT) counting
results. Table 1B describes the samples received September 1.1995.
On September 7.1995. AIT received the third set of urine samples. These were
prepared as split samples with NIH analysis available for comparison. AIT's
split samples were processed and counted at NRC Region I. This set contained
aliquots from the remaining urine specimens obtained from the researcher from
July 1 to July 27, 1995. Table 1C describes the samples received September 7,
1995.
H-1
._ _ __
Table 1A
July 3, 1995. Split Samples
Samale Type (NIH NIH Volume ORISE Volume Time and Date of
Num)er sam)le Dispensed. Transferred, ml Initial Collection
num)er) ml
1 Urine (1) 0.250 0.13 7 p.m. June 29
2 Urine (14) 1.00 0.88 6 p.m. June 30
3 Urine (15) 2.00 1.85 6 p.m. June 30 to
6 p.m. July 1
4 Blood Plasma 0.100 0.05 3 p.m. Junc 30
Table IB
September 1. 1995. Samples
Type (NIH 1
Sam)1e NIH Volume NRC Volume Time and Date of l
Num)er samale Dispensed. Transferred, ml Initial Collection ;
num)er) ml
5 Urine (15) 6-10 1.00 6 p.m. June 30 to ,
6 p.m. July 1 l
4
6 Urine (18) 6-10 1.00 5 p.m. 6 July
7 Urine (21) 6-10 1.00 7 p.m. 11 July to
7 p.m. 12 July
8 Urine (24) 6-10 1.00 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br />
25 to 26 July
,
H-2
.
- _ _ .. _. _ . _ _ _ . _ . . _ _ _ _ _ _ _ _ _ _ _ _ . _ . _ . _ . _ - . . _
!
1
i
4
i Table 1C
'
September 7. 1995. Samples
Sam)le Type (NIH NIH Volume NRC Volume Time and Date of
Num]er sam)le Dispensed. Transferred, ml Initial Collection
j num)er) ml
!
'
9 Urine (16) 6-10 1.00 11 a.m. 3 July .
2 p.m. 4 July
$
10 Urine (17) 6-10 1.00 4 p.m. 5 July to
j 3 p.m. 6 July e
i
- 11 Urine (19) 6-10 1.00 6
- 30 p.m. 6 July
I
! 12 Urine (20) 6-10 1.00 7 p.m. 7 July to i
j 7 p.m. 8 July ;
j 13 Urine (22) 6-10 1.00 7 a.m. 16 July to
j 7 a.m. 17 July
a
! 14 Urine (23) 6-10- 1.00 7 a.m. to 7 pm
j 19 July
l 15 Urine (25) 6-10 1.00 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br />
4
26 to 27 July
i
l 2. ADMINISTRATIVE PROCESSING l
-
l
l 2.1 NIH Procedures i
i
NIH processed the split samples the same vtay it routinely processes
i urine bioassay samples.
j! Administratively, each urine bioassay s)ecimen is assigned a sequential
j " urine sample number" (or "U-number") tlat is unique for that year. The
! numbers are assigned consecutively, starting with U-001 in January and
i ending with U-XXX in December. A " urine bioassay sample data" sheet
l (Figure 1) is generated for each individual providing a specimen. The
i U-number, as well as the dates the sample was submitted, received, and
i
'
prepared, and the dates the results were verified, received, and
recorded are entered on the urine bioassay sample data sheet. Once the
sample-is prepared for counting, a " sample counting request" form
(Figure 2) is filled out and the next available number on the " Record of
- Analysis Numbers" form (Figure 37 is assigned as the " analysis number."
l The " sample counting recuest" form has entries describing the sample
- (liquid, urine), the racionuclide counted, counter used, and counting
4 time. NIH attaches the countirg data computer printout (Figure 4) from
, the liquid scintillation counter for each analysis run to the " sample
- counting request" form.
li
.-
I
I
_ ._ - _ _ _ _ . _ _ _ _ . . _ . . _ _ _ _ _ .
. . . - -. . .. - .-
'
Both the original urine specimen and the counted urine sample are j
identified by U-number. This number is written on the counting data
printout to correlate the specific sample with its counting position
number in that liquid scintillation counter run. This system provides ;
traceability directly from the urine specimen to the liquid
scintillation counting data.
The only difference between the administrative processing of the split
sam)les and routine urine bioassay samples was that sample numbers
1 t1 rough 4 were assigned instead of U-numbers.
2.2 ORISE Processing
ORISE staff established and maintained a chain of custody for the
materials, on receipt. The samples were logged in under one number and
ORISE staff transcribed the NIH identification information written on
the 10-milliliter vial tops to the new 20-milliliter vial tops. The
samples remained in these vials throughout the sample preparation and
counting procedures. Data records relate the sample identifications to
the liquid scintillation counting positions, to provide a traceability
of the analysis data to each sample.
2.3 NRC Processing
Once NRC received the samples, each set was logged in under the next
available analysis number. Both the analyses number and the NIH sample
identification numbers were transcribed on the top of each 20-milliliter
liquid scintillation counting vial used for subsequent counting. Data
records relate the sample identifications to the liquid scintillation
counting positions. to provide a traceability of the analysis data to
each sample.
3. SAMPLE PREPARATION
3.1 NIH Procedures
The routine urine bioassay preparation procedure is to draw a
1-milliliter aliquot from each urine specimen and place it in either a
10-milliliter or a 20-milliliter liquid scintillation vial. (Although
draft written procedures specify using a 20-milliliter vial, the actual
size used appeared to be determined by personal ) reference.) The split
samples were placed in 10-milliliter vials. Bec(man " Ready Safe" liquid
scintillation counting cocktail is added to the vial before adding the
urine sample. (The blank vial receives 1-milliliter of distilled
water.) NIH uses this liquid scintillation counting cocktail because it
is designed to be used with aqueous solutions, it mixes well with urine,
and in NIH staff's experience has a low chemical luminescence. The
10-milliliter vials receive 5-milliliters of liquid scintillation
counting cocktail, and the 20-milliliter vials receive 15-milliliters.
The solutions were mixed in the vials before counting. Each vial is l
identified by its unique U-number, which is written on the vial cap. l
The split sample preparation procedures differed from the typical urine
'
H-4
i
i
. _ _. . _.- _ _ _ _.--_ _ . _ - _ _ _ _ ._ _ _ _ _ . _ _ . -
,
l
bioassay procedures in two respects for the first set. First, because
AIT selected the set to send to ORISE a'fter the samples were ) laced in
the vials, the liquid scintillation fluid was added to the NIi set after
the samples were put in the vials. Secondly, there was only one
1-milliliter aliquot sample. The original volumes of urine sample 1 and
blood sample 4 were limited and smaller aliquots were taken in case the. t
material. was needed for later tests. Since there were several liters of
i urine sam)1e 3 available, a larger aliquot (2-milliliters) was taken.
l For the t1ird set of samples. NIH 3rovided AIT with larger volumes, from
l .which 1-milliliter aliquots could 3e drawn.
1
1 3.2 ORISE Procedures
The 10-milliliter scintillation vials used for shipping the split
samples were incompatible with the ORISE liquid scintillation counter.
This necessitated the samples being transferred to 20-milliliter
scintillation vials. Medical syringes were used to make the transfer.
Although this method maximized the recovery and transfer of material it
introduced significant uncertainties in determining the volumes of the '
smaller samples. The volume of liquid transferred for each sample is
provided in Table 1A.
,
Reagent water (1-milliliter) was added to the blank vial. (Reagent
water consists of ta) water that has been passed through the following:
anion and cation exc1ange beds to remove most of the ions, a carbon bed
to remove most of the chlorine and organics and finally reverse osmosis
and additional ion exchange beds to remove the residual ions and a
carbon bed to remove residual organics.) This process removes salts but
leaves the naturally occurring hydrogen-3 (tritium) in the reagent
water. ORISE estimates the natural tritium concentration in the reagent
water to be 35 millibecquerells per milliliter 0.7 picocuries per ,
milliliter (pCi/ml). j
a
Two milliliters of BioSol were added to each vial, the vial caps screwed
on tightly, and the vials placed on a hot plate set on low heat
(estimated to be 70 to 80 degrees centigrade) for 2 hours2.314815e-5 days <br />5.555556e-4 hours <br />3.306878e-6 weeks <br />7.61e-7 months <br />, with
occasional stirring. -The heat activates the BioSol which is used to
break down organics. The vials were allowed to cool to room temperature
before they were opened and 4 drops of 30 percent hydrogen peroxide were
added to each vial to remove color. " Ultima Gold XR" liquid
scintillation counting cocktail was added to all five vials and an
additional background vial was prepared by adding BioSol and Ultima Gold
XR to a sixth (and empty) vial.
H-5
l
i
j
_ _ _ , . , .
- - - - - -. . .- . ..
i
3.3 NRC Procedures
On September 1. 1995. NRC staff drew 1-milliliter aliquots from the four
samples received that day, placed each aliquot in a separate 20-
i
milliliter liquid scintillation vial and added 10-milliliters of liquid
scintillation counting cocktail. Ultima Gold XR. to each vial.
A fifth background vial was pre)ared by placing 1-milliliter of
demineralized water in the fift1 vial and adding 10-milliliters of
" Ultima Gold XR" to the vial.
On September 8. 1995. NRC re)eated the above procedures for the seven
new urine samples receivod t1e day before and the background vial.
4. INSTRUMENT CALIBRATION AND CONSTANCY CHECKS
4.1 NIH Procedures
'
NIH samples were counted on a Beckman LS 6000TA liquid scintillation
counter that was calibrated by Radiation Safety Branch staff for
tritium, carbon-14. sulfur-35. P-32, and chromium-51. The tritium,
carbon-14. and P-32 energy windows were set by the manufacturer. The
instrument was calibrated with a set of quench standards traceable to
the National Institute for Science and Technology (NIST). This
calibration is retained in the liquid scintillation counter's
computerized library. Counts of scintillation events per minute in the
P-32 energy window are automatically converted to disintegrations )er
minute. The distribution of total counts per energy interval can )e
displayed to compare with the expected spectrum for P-32 to confirm
that the counts are from P-32.
NIH uses liquid scintillation standards for carbon-14 and chlorine-36 on
a daily basis, to ensure the constancy of the liquid scintillation
counter. Every 6 months. the value of the chlorine-36 standard is
adjusted to account for loss of chlorine in the standard. If the daily
values differ by more than 2 standard deviations. the constancy check is
run again. If it still differs by 2 standard deviations, the counter is
taken out of service and repaired.
4.2 ORISE Procedures
ORISE samples were counted on a Packard TRI-CARB 1900CA licuid
scintillation counter. A strontium-89 (Sr-89) liquid stancard traceable
to NIST was prepared and counted with the split samples, to determine
the counting efficiency of each run that provides an internal run
calibration. The Sr-89 standard was prepared using the same quantities
, of BioSol, hydrogen peroxide, and Ultima Gold XR as the split samples. ,
ORISE in)uts the counts per minute for each vial in the batch run into a !
spread sleet and calculates the counting efficiency for each batch. The
'
Sr-89 standard is used because its beta energy spectrum is similar to
that of P-32.
. j
H-6 I
4 __m 2
l
l
The run efficiency for P-32 is determined using the following equation:
efficiency - (CPM Sr-89 - CPM BKG) / (pCi Sr-89)
l where CPM Sr-89 is the counts per minute from the Sr-89 standard
solution vial, CPM BKG is the counts per minute from the background
vial, and PC/ b80 is the known activity of the Sr-89 standard solution
in picocuries. The background counts per minute for the split samples
l was determined by averaging the reagent water and empty vial counts per
!
minute.
l 4.3 NRC Procedures
NRC samples were counted on a Packard TRI-CARB 2250CA liquid
scintillation counter. The efficiency of the instrument is determined
using a carbon-14 standard traceable to NIST. The efficiency was
determined before each sample counting run. The liquid scintillation
counter has a constant counting efficiency in the carbon-14 and higher
energy ranges. The instrument automatically uses an efficiency tracing
technique to convert the counts per minute into disintegrations per
minute.
5. RESULTS
5.1 NIH
NIH counted the split samples July 3, 1995, for 5 minutes, using a
preset program that counted scintillation event in the hydrogen-3,
carbon-14, and p-32 energy windows. The counts per minute were
converted to disintegrations per minute and all positive disintegrations
per minute were in the P-32 energy window. The results were converted i
to disintegrations per minute per milliliter. ;
I
NIH also obtained a liquid P-32 standard from Isotope Products
Laboratory that was traceable to NIST. Two 1-milliliter aliquot samples
were 3repared and counted August 29, 1995, on the Radiation Safety
Branc1's Beckman LS 6000TA liquid scintillation counter. The results
are seen in Figure 5.
5.2 ORISE
ORISE counted the split sam]les July 7, 1995. The vial of the split
samples was counted until t1e counting statistics gave a value of 2 ,
'
percent for a 2 sigma standard deviation or 60 minutes, whichever was
shorter. Three energy windows were used. The first window (1- to 700- i
kilo electron volts (kev)) is set to capture all the low-energy beta l
emitters; the second (50 to 1700 kev) to capture all but tritium and
carbon-14: and the third (5 to 1700 kev) to capture all but tritium.
The results were reported in picocuries per milliliter (pCi/ml). The 95
percent confidence levels are based on the counting statistics only.
ORISE staff indicated that there were higher measurement uncertainties
(about 20 percent) associated with measurement volumes, especially with
l
H-7
. _ _ . _ _ . .-- _ . _ _ _ _._-.___. _ -.._...___ ___ ___._
!
>
the blood plasma sample.
ORISE staff used its only available Sr-89 standard solution to prepare .
the Sr-89 standard July 7, 1995. The last recommended date of use for ,
this solution was May 31. 1995. On August 31. 1995. ORISE compared a
liquid scintillation data from the July 7. 1995, standard and its new l
Sr-89 standard solution prepared July 27. 1995. The results for the new !
and the old standards agreed within 2.3 percent and verified the initial l
calibration and all associated data. The new standard was ordered when >
the urine samples were received and had a last recommended date of use
of January 10. 1996.
NRC converted the ORISE results to disintegrations per minute and decay-
corrected the data to July 3, 1995, for Table 2A, so that the ORISE ;
results could be compared with the NIH data.
5.3 NRC
l
'
NRC counted the second set of samples September 1. 1995. Each sample !
was counted for 20 minutes. The original NIH data for each sample were j
decay-corrected to September 1. 1995, for Table 2B. so that the NRC -
results could be compared with the NIH data.
NRC counted the third set of sam)les Se)tember 8, 1995. Each sample was i
'
counted twice for 20 minutes eac1. Bota NIH's original data and the
l average of NRC's two counts were decay-corrected for each sample to
September 7,1995. for Table 2C. so that the original NIH and NRC data i
can be compared. The third set of samples was prepared as split
samples, and NIH counted its split samples on September 7.1995. These
I data were compared with NRC's decay-corrected data in Table 2D. ,
l
'
I
l
l
.
i H-8
_ - --
, , _
Table 2A
NIH/0 RISE July 3.1995. Split Sample Results
Sample Number ORISE pCi/ml ORISE dpm/ml NIH dpm/ml NIH/0 RISE
7 July 95 3 July 95 3 July 95
1 6430+130 17400+352 13200 0.76
2 506+11 1370+30 1360 0.99
3 1262+25 3420+68 3460 1.01
4 1982+67 5370+181 6850 1.28
.
'
H-9
i
Table 2B
NIH/NRC September 1. 1995. Sample Results
(Compared to Decay-Corrected Original Measurements)
2
Sample Number NIH dpm/ml NIH dpm/ml NRC. dpm/ml NIH/NRC
Original data 1 Sep 95 1 Sep 95 ,
1
5 3200 174 162 1.07 1
6 5158 345 336 1.03
7 2448 216 202 1.07 i
8 1109 203 101 2.00
Table 2C l
NIH/NRC September 7, 1995. Sample Results ]
(Compared to Decay-Corrected Original Measurements)
Sample Number NIH, dpm/ml. NIH, dpm/ml. NRC dpm/ml. NIH/NRC
Original data 7 Sep 95 7 Sep 95
9 1897 86 64 1.34
10 3718 177 172 1.03
11 5123 256 244 1.05 i
12 3461 217 204 1.06
'
13 777 73 70 1.04
14 1450 123 130 0.95
15 1471 201 121 1.66
l
l
I
.
I
H-10
,
!
l
Table 2D
NIH/NRC September 7. 1995. Split Sample Results
Sample Number NIH, dpm/ml, NRC, dpm/ml, NIH/NRC
7 Sep 95 7 Sep 95
9 57 64 0.89
10 161 172 0.94
11 255 244 1.05
12 211 204 1.05
13 78 70 1.11
14 137 130 1.05
15 132 121 1.09
6. DISCUSSION
6.1 NIH/0 RISE Analysis
The discre)ancies between the data for aliquots less than 1-milliliter
were attri)uted, by ORISE staff, to higher uncertainties introduced by
volumetric changes resulting from shipment and transfer. Radioactivity
inhomogeneities in the bioassay sample may also have contributed to data
variations.
Since both sets of split samples were prepared by one person at
essentially the same time using the same pipette, initial volume
measurement uncertainties would be the same for both sets. The original
urine specimens were mixed before the aliquots were removed, to minimize
inhomogeneities between split sam)les. The blood sample was not mixed
so that the plasma portion could Je tested. This may have resulted in
sample inhomogeneities for this split sample. Shipping and transfer of
the samples from the 10-milliliter to 20-milliliter vials by ORISE staff
introduced additional sample- transfer and volumetric-measurement
uncertainties into the ORISE measurements.
The results show larger discrepancies between the ORISE and NIH data for
the split samples that were less than 1-milliliter (samples 1 and 4).
In both cases, only 50 percent of the initial volume could be
transferred. The discrepancies for these samples were similar to the
ORISE estimated 20 percent volume measurement uncertainty estimates.
Radioactivity distribution inhomogeneity between the materials
transferred and that left behind in these split samples may also
contribute to the discrepancies, since significant volumes were not
transferred. The loss of material for the one- and two-milliliter
samples was only 12 and 7 percent, respectively, and there was close
agreement between the counting data.
H-11
. - -. . - - - - - - . - - .
L !
,
6.2 NIH/NRC Analysis
i
NIH did not prepare split samples when it prepared the second set of'
urine samples. Therefore. NRC had to compare its results with the
original data. NRC analyses of the second set of samples resulted in
close agreement.between NIH's original. measurements and NRC's
measurements for .three of the four samples. Since the discrepancy for
the fourth was so large, the sample was both recounted and a new aliquot
taken for a second analysis, to ensure that it was prepared and counted
correctly. Both actions confirmed NRC's first measurements for that
urine sample.
l NIH was informed of the discrepancy between the September 1 results and
I
its original results. asked if it could explain the discrepancy, and
, requested to provide additional samples of the urine bioassay samples,
from July 1 to July 27 that had not been analyzed by either ORISE or
l NRC.
NIH staff responded by submitting the third set of samales (which were ;
prepared as split samples), and informing NRC that it lad noticed the
same discrepancies when it had prepared a separate set of split samples
l on August 29, 1995, for the researcher. It observed, at that time, that
l some of the urine specimens had increased precipitation and what
i appeared to be bacterial growth on the bottom of the containers. This
was especially true for NIH sample numbers 24 and 25 (NRC sample numbers
8 and 15); it attributed the discrepancy to urine sample degradation.
NRC analyzed the third set of urine samples and compared the results to
- both the original data (Table 2C) and NIH's split samples prepared
l September 7, 1995 (Table 20). NRC's results were also compared to NIH's
samples drawn August 29, 1995 in Table 3-
'
i
l
l I
1
l
.
l
j H-12
,
. , _ _ , .- . .-. -
-_- - . . - - - _ - - - --.- - .. . -. - . - . - .
i
.,
,
i Table 3
NIH New Aliquots Counted August 29, 1995
j NRC Aliquots Counted September 1. 1995. and September 8. 1995. Results
j Samp1e Number NIH, dpm/ml, NIH, dpm/ml. NRC. dpm/ml. NIH/NRC
- Counted. 1 Sep 95 or 1 Sep 95 or
29 Aug 95 7 Sep 95 7 Sep 95
j 5 190 164 162 1.01
l 6 336
7 252 218 202 1.08
8 136 118 101 1.17
9 133 86 64 1.34
! 10 267 172 172 1.00
11 244
j
i
12 330 213 204 1.04
13 110 71 70 1.01
14 207 134 130 1.03
i 15 212 137 121 1.1 L ,
j 7. CONCLUSIONS
'
7.1 NIH/0 RISE Analysis
- NRC concluded that the ORISE independent analysis of 1-milliliter (or
higher) aliquots verified the validity of the NIH liquid scintillation
counting results within the uncertainties of normal counting statistics.
l The differences between sample preparation for the two sets was not
observable for these aliquots. The discrepancies between the data for
the less than 1-milliliter aliquots was attributed to higher
uncertainties introduced by volumetric changes resulting from shipment ,
! and transfer.
7.2 NIH/NRC Analysis
> Based on comparisons between NIH*s original measurement data and the
recently resam) led data. NRC concluded that there was significant loss
of measurable L32 in some of the urine specimens (samples 8.'9, and 15)
that could not be attributed to variability in aliquot preparation or
counting techniques. Table 3 shows that by August 29. 1995, samples 8
and 15 were already showing signs of degradation but that sample 9 had
not started to degrade. NIH acknowledged the loss of P-32 radioactivity
in these samples and attributed it to biological degradation of the
4 samples with time. NRC further concluded that the closer agreements
-
(ranging from 3 to 11 percent) between the split sample results affirm
~
H-13
. -
. . ...-._=_ -. ~. - -. . .. .- . - . . .. .- - . . . .
i
l
l
the degree of confidence NRC has in NIH's ability to measure the P-32
radioactivity in the researcher's urine bioassay specimens as they were ;
collected.
I
l
l
l 1
'
l
i l
\ l
l
i
l
l
l
I
i
l
H-14
l
l
1
APPENDIX I
5 HOUR DELAY IN PROCESSING BI0 ASSAY DATA
I-1
- _ . . _ _ . . . . - -__ __- -
APPENDIX I :
SUMMARY OF NIH URINE BI0 ASSAY COLLECTION AND ;
COUNTING CHRONOLOGY
Sample Date Number of Number of New Repeat Time / Day
New Repeat Positives Positives Counted
Individuals Individuals
1
July
3 4 -
1 -
17:24 3 July
5 12 1 -
1 12:50 6 July
'
5 1 -
{1} -
10:10 18 July
6 8 -
1 -
12:50 6 July
6 2 - - -
10:10 18 July
5 or 6 6 -
1(1) -
12:50 6 July
10 3 - - -
15:32 13 July
11 42 -
9(1) -
15:32 13 July
11 2 -
2 -
14:22 14 July
12 9 -
1(1) -
15:32 13 July 1
11 or 12 1 - - -
15:32 13 July
12 6 -
2 -
14:22 14 July 1
13 7 -
3 -
14:22 14 July I
13 2 -
1 -
16:15 14 July
13 2 - - -
17:22 14 July
14 12 26 2{1} 12 17:22 14 July
No Date 2 6 - -
17:52 14 July
14 6 3 -
1 18:54 17 July
17 9 4 (1C) 1 18:54 17 July
17 4 - - -
20:09 18 July
18 12 3 -
1 18:02 18 July
18 2 - - -
20:09 18 July
Researcher A and Her Husband are Omitted from this Chronology
(N) Indicates recount of urine sample was below the lower limit of detection for P-32
energy window
(1C) Indicates recount of urine sample was below the lower limit of detection in carbon-
14/ sulfur-35 energy window
{1} Indicates first positive urine sample for this individual was put in NIH internal
mail and not analyzed until July 18, 1995
I-2
. __
,
-
DIVISION OF
,
- NUCl.EAll M EDICINE
-
- , --
[_ mat t lNCkRODI ]
l lNSillUI( Of R ADIOtOGd
m m M m Im1T4
uuurmmmrermumus
4 September 1996
James P. Dwyer
Division of Radiation Safety and Safeguards
U.S. Nuclear Regulatory Commission
Region 1
475 Allendale Road
King of Prussia, PA 19406-1415
Re: Report of Medical Consultant to Augmented Inspection Team
Internal P-32 Contamination of Researcher at National Institutes of Health (NIH)
Docket No. 030-01786; License No. 19-00296-10; Event No. 29008
Dear Mr. Dwyer:
The following Medical Consultant's repon is provided in response to the letter from Jenny M.
Johansen dated 5 July 1995. In preparing this report, I have had access to several different
sources of data and information. These included (1) numerous telephone conversations and
electronic mail communications with you, as well as conversations with Donna Beth Howe
(NMSS) and other members of NRC staff, on multiple occasions since I was first contacted by
Larry W. Camper on 30 June 1995; (2) a telephone conversation with Shen-sho Tseng, M.D.,
the contaminated researcher's private physician on 21 July 1995; (3) telephone conversations
with Shawn W. Googins, CHP, of the NIH Radiation Safety Branch on 23 July and 3
November 1995; (4) telephone conversations with James Schmitt, M.D., of Occupational
Medical Services at NIH on 26 July,15 August, and i November 1995; and (5) telephone
conversations with David Hickman, Ph.D of the Special Projects Division at Lawrence
Livermore National Laboratory (LLNL) on 3 and 16 November 1995. I have also reviewed
various documents including (1) an event chronology prepared by James Dwyer; (2) an i I
July 1995 summary of whole-body scanning date prepared by Jorge A. Carrasquillo, M.D.,
Acting Chief of the NIH Nuclear Medicine Department;(3) summaries of background infor-
mation on phosphate metabolism and of studies to determine the chemical identity of the
ingested P-32 (provided to me by Shawn Googins); (4) a letter dated 8 November 1995 to
Shawn Googins from Ronald F. Goans, M.D., Ph.D. of the Radiation Emergency Assistance
Centerffraining Site (REAC/TS) at the Oak Ridge Institute for Science and Education
(ORISE);(5) several August 1995 dosimetry estimate summaries and a supplement dated 17
October 1995 prepared by Michael Stabin, CHP, of the Radiation Internal Dose Information
Center (RIDIC) at ORISE; (6) the dosimetry estimate summaries dated 26 October and 10
November 1995 prepared by staff of the Special Projects Division at LLNL;(7) copics of the
contaminated researcher's medical records (consisting almost entirely of laboratory data)
provided to the NRC by her attorney on 22 April 1996; and (8) a copy of a letter from a
consultant, David A. Dooley, Ph.D., dated 15 April 1996 to the contaminated researcher's
attorney, sent to me by the NRC on 27 April 1996.
Despite the fact that I have had access to a substantial amount of information as indicated
above, I believe it is important for me to indicate that the useful medical information available
to me in this case is actually quite limited. Selected medical information has been obtained
Enclosure (2)
510 South Kmgshighway ikulevard
St. Lou 6, Wsouri 63110 1070
(3t O 302 2809 FAX (314) 362 2o06
. ..
.
1
'
,
Repoit of Medical Consultant to Augmented Inspection Team
. Docket No. 030-01786; License No. 19-00296-10; Event No. 29008
'
4 September 1996 l
Page 2 ;
l
l
through various telephone conversations and from the medical records supplied. Ilowever, the
information pertaining to any symptoms or signs of illness that the contaminated researcher
may have experienced before or after the presumed date of exposure is quite sketchy. I have
essentially no information (other than laboratory records and the delivery record) concerning
the researcher's medical condition since late July /carly August. As you are aware, I have not
had an opportunity to interview or examine the contaminated researcher directly.
Relevant Medical History and Medical Aspects of Event: The researcher is a 32- 1
year-old Chinese woman, who reportedly had been in good health prior to the event. She l
underwent pre-employment evaluation at the NIH Occupational Medical Services on 14 l
September 1994. Her evaluation at that time was reportedly normal. At the time of the internal l
contamination event, the individual was approximately 17 weeks pregnant (this is presumed to l
represent the post-menstrual age of the pregnancy based on the recorded date of her last
menstrual period of I March 1995 and her delivery of a full-term infant on 1 December 1995).
Her pregnancy was undeclared. She reportedly had complained of" morning sickness" during
the first three months of her pregnancy, but these symptoms had subsided during the first few i
weeks of June. Hemoglobin and hematocrit obtained on 12 June 1995 were 11.6 g/dL and
35.1%, respectively. This borderline " anemia" is a common finding in pregnancy, and is
largely dilutional in most cases, as a result of plasma volume expansion. Her white blood cell
count on that date was 7,660/pL and her platelet count was 249,000/ L, both values within
normal limits. Her white blood cell differential count was essentially normal (78.5%
neutrophils,14.1% lymphocytes,5.4% monocytes,0.4% cosinophils,0.4% basophils, and
1.2% atypical lymphocytes).
The investigation performed by the Radiation Safety Branch at NIH and by the NRC suggests
that the internal contamination with P-32 occurred midday on 28 June 1995. The researcher
i reportedly developed right-sided back pain on that day or the next (the infonnation available to
me is unclear regarding this point). The researcher apparently reported this symptom to her
private physician, but also indicated that the symptom was relieved by wrapping a towel
around her abdomen, thereby suggesting to her physician that the discomfort was muscu-
loskeletal in origin. Additionally, recurrent nausea and vomiting also apparently developed on
either 28 or 29 June, and may have persisted for several days thereafter.
When the internal contamination was discovered on the evening of 29 June 1995, the
researcher was seen in the Emergency Room at Holy Cross Hospital. Staff of the Radiation
Safety Branch at NIH had contacted the Radiation Emergency Assistance Center /Fraining Site
(REAC/TS). The REAC/TS physician spoke with the emergency room physician at Holy
Cross Hospital and recommended hydration and administration of stable phosphate as treat-
ment to maximize excretion of P-32. Hydration was performed, but the physician at Holy
Cross Hospital elected not to administer phosphate.
4
Occupational Medical Services at NIH obtained a blood sample for assessment of the
researcher's hematological profile on 30 June 1995. At that time, her hemoglobin and hemat-
ocrit were 10.7 g/dl and 31.3%, respectively. These values were mildly decreased by
comparison with the measurements made on 12 June. The results, however, were still within
the expected range for dilutional" anemia" of pregnancy. The fluids administered on the previ-
ous evening to hydrate the researcher may also have contributed to this result. Her white blood
cell count was 7,300/pL and her platelet count was 230,000/pL. These results are normal.
The white blood cell differential count remained essentially normal (79% neutrophils,13%
lymphocytes,7% monocytes, and 1% cosinophils).
.
.
Repon of Medical Consultant to Augmented Inspection Team
Docket No. 030-01786; License No. 19-00296-10; Event No. 29008
-
4 September 1996
Page 3
Only minimal subsequent medical history is available. The researcher apparently continued to
experience episodes of vomiting, but the severity, frequency, and duration of these symptoms
are unclear. The patient's private physician obtained additional laboratory data on 22 July
1995. At that time, her hematological profile was essentially unchanged by comparison with
that on 30 June 1995 (hemoglobin 10.5 g/dL; hematocrit 31.6%; white blood cell count
6,710/pL; platelet count 223,000/gL; 76.4% neutrophils,15.5% lymphocytes,6.1% mono-
cytes; 0.4% cosinophils, 0.4% basophils, and 1.3% atypical lymphocytes). Serum
electrolytes were normal (potassium 3.6 mcq/L; sodium 137 meq/L; and chloride 105 meq/L).
lier total scrum protein and scrum albumin were both mildly depressed at 5.5 g/L and 3.4 g/L,
respectively (normal ranges 6.3-8.4 g/L and 3.3-5.3 g/L, respectively). Additionally, her
blood urea nitrogen (5 mg/dL), serum creatinine (0.4 mg/dL), and uric acid (1.8 mg/dL) were
all below the normal range. I inteipret these findings to be most consistent with plasma
volume expansion during pregnancy. There was mild elevation of the scrum aspartate amino-
transferase. (SGOT) concentration at 40 U/L (normal range 0-39 U/L) and of the alanine
aminotransferase (SGPT) at 82 U/L (normal range 0-53 U/L). Elevations in the levels of these
enzymes most often indicate hepatocellular injury, but can be seen with many different disor-
ders involving many different organs and tissues. In the absence of other medical historical
infonnation and follow-up laboratory data, I am unable to determine the clinical significance of
these enzyme abnormalities. The remainder of her chemistry profile on this date was normal.
Additional blood laboratory studies were performed on 5 August 1995. Her mild anemia
remained unchanged and her SGOT and SGIYT levels had returned to normal. Her white blood
cell count was still normal, but the differential count showed a slight left shift (75% neutrophils
and 7% bands), a finding of uncertain significance.
In a telephone conversation on 15 August 1995, Dr. Schmitt at Nili Occupational Medical
Services reported to me that the researcher had declined to have further contact with the
physicians at Occupational Medical Services. Dr. Schmitt further reported that he had spoken
to the researcher's private physician who indicated that the researcher claimed to have persis-
tent severe nausea and vomiting. Dr. Schmitt asked the private physician to contact me directly
if he were willing to do so, but I never received a call from him.
Additional blood laboratory studies on 30 August 1995 were essentially unchanged, except for
mild hyperglycemia (glucose 118 mg/dL). Fetal ultrasonography on i September 1995 was
normal, and the measurements indicated a postmenstrual fetal age of about 27 weeks (which
indicates a postmenstmal fetal age of about 18 weeks on 28 June 1996). Repeat hematological
profile on 13 September 1995 again showed mild anemia (hemoglobin 10.6 g/dL) and 7% l
bands, and another on i1 November 1995 still showed mild anemia (hemoglobin i1.0 g/dL);
with a normal white blood cell differential count.
The researcher was admitted in labor to the Columbia liospital for Women Medical Center on
30 November 1995 and uneventfully delivered a grossly normal male infant on i December
1995. The infant's Apgar scores were 9 and 9 at I and 5 minutes, respectively. Newspaper
accounts stated that the baby was apparently healthy.
I have had access to no subsequent medical information concerning the researcher. In a
telephone conversation with Dr. Schmitt on 1 November 1995,I confirmed that he had had no
follow-up contact with the researcher since he and I had last spoken on 15 August 1995.
.
.
.
.
'
Repon of Medical Consultant to Augmented Inspection Team
, Docket No. 030-01786; License No. 19-00296-10; Event No. 29008
4 September 1996
! Page 4
,
Based on the medical information available to me, I am unable to determine whether any of the
reported symptoms experienced by the researcher are related to ingestion of P-32. liowever,
as noted below, it is highly unlikely that these symptoms are radiation related. Since the
chemical form of the ingested P-32 remains uncertain, chemical toxicity cannot be entirely
excluded.
- Radiation Dosimetry: As you are well aware, there is some degree of uncertainty regard-
'
ing the amount of P-32 ingested by the researcher. This reflects the inherent variability of
urine bioassay data and, in the case of a pure beta-emitter such as P-32, the difficulty in
i
precisely quantifying retained activity by whole-body counting. Moreover, the interpretation
of these data depends on the use of statistical fitting and mathematical modeling techniques that
may properly characterize the kinetics of the radionuclide in an " average" subject, but do not
necessarily reflect the real kinetics in any individual subject. Some additional uncertainty in
this case is introduced by the fact that the chemical identity of the P-32 compound ingested is
not known. It is not clear whether the ingested material was in the form of P-32 inorganic
phosphate or a P-32 labeled nucleoside or nucleotide. For practical purposes, these organic
forms would functionally behave on ingestion like inorganic phosphate (because of degrada- i
tion of the orgame form in the gastrointestinal tract). The possibility that the ingested P-32 was
incorporated m an entirely different chemical form, with much different absorption, excretion,
metabolism, and biodistribution compared with inorganic phosphate, has not been excluded.
Despite these inherent limitations in the data and in the assumptions used to evaluate the
ingested activity and radiation dosimetry, the estimates obtained by the licensee, by RIDIC,
and by LLNL are all reasonably close. With the several different calculation methods
,
'
employed, estimates of the ingested activity range from 300 Ci to 1300 pCi. The best esti-
mate of intake obtained by Michael Stabin at RIDIC is 820 Ci. The best estimates obtairH
by David Hickman at LLNL, based on several different computational methods, are 1100,
1300, and 1050 Ci, respectively. The average of these four intake estimates is 1068 pCi.
The effective dose equivalent to the researcher is estimated (using reference woman rather than
ieference man assumptions) to be 8.0 rem by RIDIC, and 10.8,12.7, and 10.3 rem, respec-
tively, for the three different computational methods employed by LLNL. The average effec-
tive dose, taken as the average of these four estimates, is 10.4 rem. The corresponding fetal
doses are estimated to be 5.1,6.9,8.1, and 6.5 rem, respectively. The average fetal dose,
taken as the average of these four estimates,is 6.6 rem
As I have indicated in cailier communications with NRC staff, I do not consider myself to be
an expert in the interpretation of bioassay data or in the estimation of internal dosimetry.
Accordingly, I have not presumed to second-guess the estimates made by Mr. Stabin and Dr.
Hickman, who are acknowledged experts in these matters. For the purposes of assessing of
potential medical effects of this P-32 ingestion, I have used the average doses and the dose
ranges given above.
Potential Deterministic Consequences of P-32 Ingestion: Assuming an effective
dose to the contaminated researcher of 10.4 rem (range 8.0 to 12.7 rem), no deterministic
effects are expected. With P-32, the red marrow dose is of potential greatest concern for
adverse deterministic effects. By extrapolation from the Appendix 4 of the dose estimate report
prepared by LLNL, the red marrow dose would approximate 39 rem in a female subject whose
effective dose was 10.4 rem (range 30 to 47 rem marrow dose for an effective dose range of
8.0 to 12.7 rem). Discernible hematopoietic system effects can be detected following single
exposures to the bone marrow of doses as small as 50 rem, but would be quite unlikely with a
- - - -- - . . -- _ .
.
.
Report of Medical Consultant to Augmented inspection Team
,
Docket No. 030-01786; 1.icense No. 19-00296-10; Event No. 29008
i 4 September 1996
Page 5
smaller dose. Moreover, the modulating influence of low dose rate must be considered in an
i
instance of internal exposure with a radionuclide having a half-life of 14.3 days. The available
hematological data available for the contaminated researcher do not show direct evidence for a
deterministic effect on her hematopoietic system.
With an assumed fetal effective dose of 6.6 rem (range 5.1 to 8.1 rem), no deterministic effects
are expected with a fetus of approximately 17-weeks age. The deterministic effect of most
i concern at this stage of fetal development would be impairment of brain development
(manifested by retardation or reduced intelligence test scores). The studies evaluating the risk
of injury to the developing brain have stratified fetuses into those 8-15 weeks of age (post-
conception) and those 16-25 weeks of age (post-conception). The age of the fetus at the time
- of exposure in this case (17 weeks post menstrual age or 15 weeks post-conception age) is at
the border between these two groups. Additionally, the radiation dose from P-32 would be i
'
'
.
delivered over a protracted period of time, thus clearly extending part of the exposure into the
older-age stratum. Moreover,it is unknown whether modulation for a low dose rate must also
be considered. Irrespective of these points of uncertainty, radiation induced severe mental
, retardation is unlikely in this fetus. Although the available data are generally taken to indicate a
j linear, non-threshold response, they are also consistent with a threshold in the range of 20-40
rem for the 8-15 week fetus, with no definite increase in mental retardation evident at doses
less than 20 rem. In fetuses between 16 and 25 weeks of age, no definite increase in mental
, retardation is evident at doses below 50 rem. With regard to less severe impairment of brain
development, the results of intelligence test scores suggest a significant radiation-related
decrease in 8- to 15-week fetuses and a less marked effect at 16-25 weeks. With utilization of
a linear model, a 21- to 33-point diminution of IQ score is expected with a 100-rem acute
'
.
exposure to a 8- to 15-week fetus. Hence, ignoring dose-rate effects, a theoretical reduction in
IQ of the exposed fetus in this case of 1.4 to 2.2 points (assuming an average dose of 6.6 rem)
might be expected. (The range of theoretical reductions in IQ scores extends from a low value
of 1.1 to a high value of 2.7 points, with doses ranging from 5.1 to 8.1 rem.) Needless to
say, such an effect would be undetectable and any putative relationship to radiation exposure
unprovable.
1
Potential Stochastic Consequences of P-32 Ingestion: The stochastic effect of
concem in the contaminated researcher is radiation-induced cancer. Based on the risk estimates
in Health Effects of Exposure to Low Levels oflanizing Radiation: BEIR V. (Washington,
D.C.: National Academy Press; 1990:175), the lifetime risk for fatal cancer with a 10.4-rem
exposure is 1.24% for a woman exposed at age 25 and 0.59% for a woman exposed at age 35.
(The ranges of these risk estimates are 0.95 to 1.51% for exposure at age 25 and 0.45 to
0.72% for exposure at age 35, for effective doses of 8.0 to 12.7 rem.) The actual risk in this
case can be reduced by a factor of 2.0 to 2.5 because the exposure was delivered at a low dose
rate. For comparative purposes, note that the lifetime risk of fatal cancer without radiation
exposure is approximately 20%.
Although there is mod.: rate uncertainty in the data used for cancer risk estimation as a result of 4
in utero radiation exposure, a reasonable estimate of the risk during the first 10-14 years of life l
for leukemia and other childhood cancers following in utero radiation exposure is approxi-
mately 0.05% per rem. Accordingly, in this case, with an average fetal effective dose of 6 6
rem, an excess risk of 0.33% is estimated (range 0.26 to 0.40% for effective doses ranging
from 5.1 to 8.1 rem). For comparative purposes, the natural risk of childhood cancers is about
0.1%. Thus, the risk is increased by about 330% (range 260 to 400%). However, stated
another way, the probability that the exposed fetus will NOT develop a radiation-induced
1
J
_ _ _
Report of Medical Consultant to Augmented Inspection Team
Docket No. 030-01786; License No. 19-00296-10; Event No. 29008
, 4 September 1996
,. Page 6
childhood cancer is 99.67% (range 99.60 to 99.74%). It is unknown whether this risk esti-
mate should be reduced because of the low dose rate associated with this internal exposure i
from P-32.
Dr. David Dooley, who provided consultant services for the researcher and her attorney stated .
I
in his letter of 15 April 1996 that the risk of chilhood cancer in this infant was 30 to 150 times
that in an unexposed child. He then " reasons" that the uncertainties in the dose estimates in
this child and in the published estimates of the risk of radiation-induced childhood cancer
justify inflation of the relative risk to 1,000 times that in an unexposed child (which is the
number he stated in a 7 December 1995 letter to the Washington Post.). I see no scientific
justification for the latter reasoning. Moreover,I believe that his calculation of the relative risk
is flawed. Specifically,it appears that he has divided the estimated total-childhood risk in
this infant by the annual risk in an unexposed child:
In summary, serious medical consequences are not probable to either the exposed researcher or
to her infant as a result of the ingestion of P-32.
Please let me know if you need additional information.
Sincerely yours,
%U &
Barry A. Siegel, M.D.
Professor of Radiology and Medicine
Director, Division of Nuclear Medicine
BAS:mte
l
l