Radiation accidents with multi-organ failure in the United States

DOI: 10.1259/bjr/27824773
Radiation accidents with multi-organ failure in the United
States
1R E GOANS, PhD, MD, MPH and 2N WALD, MD
1Senior Medical Consultant, MJW Corporation and Clinical Associate Professor, Tulane University School of Public
Health and Tropical Medicine and 2Professor, Environmental and Occupational Health, University of Pittsburgh Graduate
School of Public Health, USA
Abstract. Only a small number of radiation accidents in the United States have been severe enough to result in
multi-organ failure (MOF). Medical details of selected medical misadministration and criticality cases are
reviewed, with an emphasis on pathophysiology. The four criticality cases are particularly relevant for analysis
of MOF, since medical treatment was supportive and did not appreciably alter the clinical evolution of
radiation injury.
Introduction
begun, but one source was not retracted afterwards and
remained in place for 4 days until it dislodged. Hospital
Only a small number of radiation accidents in the
staff ignored warning signals, believing that safety
United States post World War II have been severe enough
equipment was giving a false alarm, and the source was
to result in multi-organ failure (MOF). The cases with
not discovered until it was transferred to a medical
radiation-induced MOF are summarised in Table 1. Since
incinerator. The patient died 5 days after the exposure [1,
1945, 4 deaths have resulted from criticality accidents, 21
5, 6].
from medical misadministration in the course of radiation
therapy or diagnostic procedures, and 1 from a lost source
[1]. In addition, three people died of blunt trauma at the
Industrial accelerator malfunction case
SL/1 reactor accident (Idaho National Engineering and
Environmental Laboratory, 1961). Details of this accident
Van de Graaff accident involving amputation of
have been described elsewhere [2]. Medical details of
forearms and legs, with life-saving bone marrow
selected medical misadministration and criticality cases are
transplant (1969)
reviewed below, with an emphasis on pathophysiology.
The four US criticality cases are particularly relevant for
An experienced Van de Graaff accelerator operator at
analysis of MOF, since medical treatment was supportive
the
Gulf
Oil
Corporation
Research
Laboratory
in
and did not appreciably alter the clinical evolution of
Pittsburgh was exposed to intense bremmstrahlung radia-
radiation injury.
tion while unscrewing target cooling tubes, unaware that
multiple safety interlocks had failed. From film badge
readings, the operator was estimated to have received a
Medical misadministration cases
mean marrow dose of 6 Gy, with approximately 60 Gy to
the forearms and 20–30 Gy to the lower legs. The operator
Therac-25 cases
underwent a successful bone marrow transplant 9 days
post exposure from his identical twin, reversing a severe
A series of misadministrations occurred when software
granulocytopenia and lymphopenia. Nevertheless, the
errors caused an automated radiation therapy machine to
deliver massive overdoses in six cases in the United States
and Canada in the period 1985–87. Estimated doses were
Table 1. The US experience with multi-organ failure
reported in the range 80–250 Gy. One patient experienced
Accident
Type
Dates
Fatalities
multi-organ injury and died 5 months after exposure. One
patient had severe injury to the brain and brain stem,
Therac-25
Radiation therapy
1985–87 3
quickly became disoriented, comatose and died 3 weeks
accidents
misadministration
after exposure. Three other patients received severe local
(estimated dose
80–250 Gy)
injury, and one died from radiation-related complications.
Indiana, PA,
High dose brachytherapy
1992
1
All patients were aware of the misadministration and most
192Ir accident
accident
clinical symptoms appeared over a period of weeks [3, 4].
Van de Graaff
Industrial accelerator
1969
0
accident
malfunction
LANL-1
Solid criticality accident
1945
1
Indiana, PA, iridium-192 incident
LANL-2
Solid criticality accident
1946
1
LANL-3
Liquid criticality accident 1958
1
An 82-year-old woman was diagnosed with anal cancer
Wood River
Liquid criticality accident 1964
1
and treated with high dose rate brachytherapy at Indiana
Junction
Regional Cancer Center, Indiana, Pennsylvania, on 16
November 1992. High-intensity 192Ir brachytherapy was
LANL, Los Alamos National Laboratory.
E 2005 The British Institute of Radiology
41

---

R E Goans and N Wald
highly exposed forearms and lower legs required sequential
Wood River Junction case
amputations. Although confined to a wheelchair there-
This liquid process accident occurred on 24 July 1964, at
after, the operator led an active life until passing away
the United Nuclear Fuels Recovery Plant, Wood River
from a heart attack many years later at the same age as his
Junction, RI. A chemical processing plant [12] was
twin brother [7–11].
designed to recover highly enriched uranium from scrap
material left over from the production of fuel rods. Uranyl
nitrate solution U(93) was poured into a carbonate reagent
Criticality cases
vessel. The critical excursion occurred when nearly all of
the uranium had been transferred, resulting in approxi-
Los Alamos plutonium sphere cases
mately 1.16E+17 fissions. It is probable that the system
oscillated, resulting in a series of excursions with total
Two criticality events occurred with the same 6.2 kg
energy release equivalent to 1.36E+17 fissions. The acute
delta-phase plutonium sphere at Los Alamos National
dose to the operator was estimated to be 100 Gy. Two
Laboratory (LANL). The first incident occurred on 21
supervisory personnel received approximately 1 Gy and
August 1945, when a worker was preparing a critical
0.6 Gy, respectively. The operator died 49 h post accident.
assembly by stacking tungsten carbide bricks around the
plutonium core as a reflector. He moved the final block
over the assembly but, noting that this block would make
Clinical course of the criticality cases
the assembly supercritical, he withdrew it. The brick fell
onto the centre of the assembly, resulting in a super-
Case 1 – Los Alamos plutonium sphere
prompt critical state of approximately 66E+15 fissions
(Haematopoietic Syndrome; Cutaneous Radiation
[12]. The worker sustained an average whole body dose of
Injury Syndrome; whole body dose approximately
approximately 5.1 Gy and a dose to the right hand of
5.1 Gy, dose to right hand 100–400 Gy)
approximately 100–400 Gy. The patient died of sepsis 28
The patient was a 26-year-old male whose past medical
days post accident.
history was significant only for Wolff-Parkinson-White
The second criticality accident occurred in 1946 during
Syndrome diagnosed 3 years prior to the incident [13]. On
an approach to criticality demonstration at which several
admission to hospital, his vital signs were within normal
observers were present. The operator used a screwdriver as
limits and his only initial complaint was numbness and
a lever to lower a hemispherical beryllium shell reflector
tingling of both hands. The initial physical examination
into place. While holding the top shell with his left thumb
was also within normal limits.
in an opening at the spherical pole, the screwdriver slipped
Within 30 min post accident, the patient’s right hand
and caused a critical configuration. The fission yield in this
had become diffusely swollen. Emesis began approximately
accident was estimated at 36E+15 fissions. The operator
1K h post event, and nausea continued intermittently for
received an estimated acute whole body dose of approxi-
the next 24 h. The patient experienced subjective improve-
mately 21 Gy, with a dose to the left hand of 150 Gy and
ment but had a mild temperature, gastric distress and
somewhat less to the right hand. Seven observers were
weakness during days 3–6. By day 5 the patient
exposed in the dose range 0.27–3.6 Gy [12]. The operator
experienced a distinct rise in temperature with tachycardia
died 9 days later.
and began to appear increasingly toxic. On day 10 he
developed severe stomatitis, a paralytic ileus and diar-
rhoea. Clinical signs of pericarditis were noted on day 17,
and the patient’s mental status became irrational. The
Los Alamos liquid criticality event
clinical course is notable for progressive pancytopenia.
Figure 1 demonstrates an exponential decrease of lym-
On 30 December 1958, during purification and con-
phocytes during the first 4 days post accident.
centration of plutonium, unexpected plutonium-rich solids
Within 36 h post accident, blisters were noted on the
were washed from two vessels into a single large vessel that
volar aspect of the right third finger, and within 24 h
contained layered, dilute aqueous and organic solutions.
thereafter, extensive blistering was noted both on the
The tank contained approximately 295 l of a caustic
palmar and volar surfaces of the hand. A decision was
stabilised organic emulsion [12]. The added nitric acid
made on day 3 to surgically drain the blisters, but by the
wash is believed to have separated the liquid phases.
third week the right hand had progressed to a dry
Accident analysis shows that the aqueous layer was
gangrene. Desquamation of the epidermis involved almost
initially slightly below delayed critical (approximately
all of the skin of the dorsum of the forearm and hand. In
203 mm thick, critical thickness 210 mm). When the stirrer
addition, epilation was almost complete at the time of
was started, the central portion of the liquid system was
death.
thickened, changing system reactivity to super-prompt
On day 28, the patient’s temperature had risen to
critical. The excursion yield was approximately 1.56E+17
41.1˚C. He had lost a great deal of weight, developed
fissions. Bubble generation was the negative feedback
thoracic–abdominal erythema and had signs of sepsis. On
mechanism for terminating the first neutron spike. The
day 28 the patient because comatose and died. During the
system was driven permanently subcritical by mixing of
patient’s clinical admission, treatment consisted of fluid
the two layers. This accident resulted in the death of the
support, penicillin antibiotic therapy, thiamine and two
operator 36 h post accident. The dose to the upper
blood transfusions.
extremity was estimated to be 120 Gy¡50%. Two other
On autopsy, severe skin necrosis was observed as well
persons received acute doses of 1.34 Gy and 0.53 Gy,
as overt dry gangrene. The cardiorespiratory system
respectively [12].
was significant for pericarditis, cardiac hypertrophy,
42
Radiation-induced multi-organ involvement and failure: a challenge for pathogenetic,
diagnostic and therapeutic approaches and research

---

Radiation accidents with MOF in the United States
Figure 1. Lymphocyte kinetics in the 1945
Los Alamos National Laboratory LANL-1
criticality accident.
pulmonary oedema and alveolar haemorrhage. The spleen
aspiration pneumonia. The spleen exhibited no germinal
was noted to have no germinal centres, and the mucosa of
centres, and the mucosa of most of the gastrointestinal
the large bowel was ulcerated as was the buccal mucosa.
tract showed sloughing, most pronounced in the jejunum
The bone marrow was noted to be hypoplastic, and lymph
and ileum. Widespread degenerative changes were noted in
nodes also showed significant lymphocyte depletion. The
the adrenal cortex, and hyaline degeneration in the renal
testes demonstrated significant atrophy with aspermia. A
tubular epithelium. Examination of the red bone marrow
solitary ulcer was noted in the large colon in addition to a
showed it to be of liquid consistency.
right renal infarct.
Case 3 – Los Alamos Liquid Criticality Event
Case 2 – Los Alamos plutonium sphere
(Central Nervous System (CNS) Syndrome; dose to
(Gastrointestinal Syndrome; Cutaneous Radiation
the upper extremity 120 Gy ¡ 50%)
Injury Syndrome; acute dose approximately 21 Gy,
The patient was a 50-year-old male with no significant
dose to the left hand 150 Gy)
past medical history [14]. The clinical course has been
The patient was a 32 year-old male, admitted to hospital
divided into four separate phases. Phase 1 (20–30 min post
within 1 h post accident. His medical history was generally
event): immediate physical collapse and mental incapacita-
unremarkable. His occupational history was significant
tion, progressing eventually into semi-consciousness; phase
only for several prior, generally chronic occupational
2 (90 min): signs and symptoms of cardiovascular shock
exposures, none exceeding 0.005 Gy in a week. The patient
accompanied by severe abdominal pain; phase 3 (28 h):
complained of nausea in the hour prior to admission and
subjective minimal clinical improvement; phase 4 (2 h):
vomited once in the first hour post accident.
rapidly appearing irritability and mania, progressing to
The general condition of the patient was quite good in
coma and death. The clinical course was remarkable for
the first 5 days post accident. On the fifth day there was a
continuing,
profound
hypotension,
tachycardia,
and
precipitous drop in his leukocyte count and his condition
intense dermal and conjunctival hyperaemia. The patient
began to decline rapidly. The patient rapidly lost weight,
died 35 h post exposure.
became mentally confused on day 7 post event, became
On autopsy, examination of the bone marrow was most
comatose, and died quietly on the ninth day.
significant for absence of mitotic activity. The lungs
Medical therapy during the 9-day course was largely
showed
pycnotic,
degenerating
cells
in
the
pleura,
symptomatic.
Penicillin
was
given
intramuscularly
degenerating lymphocytes and neutrophils in the sub-
50 000 U every 3 h beginning on day 5 because of
pleural connective tissue, and many areas of focal
granulocytopenia. Blood transfusions were also given
atelectasis interspersed with foci of emphysema. All
daily after the fifth day. On day 6, fever and tachycardia
lymph nodes were markedly atrophic and lymphoid
developed, and on the seventh day the patient developed a
follicles in the spleen were greatly depleted.
severe paralytic ileus. The patient died in cardiovascular
Examination of the heart showed acute myocarditis,
shock on the ninth day. At the time of death both hands
myocardial oedema, cardiac hypertrophy and a fibrinous
showed extensive radiation damage.
pericarditis. Examination of the brain demonstrated
On autopsy, examination of the skin was remarkable for
cerebral oedema, diffuse vasculitis and cerebral haemor-
early vesicle formation in the abdominal skin and marked
rhage. The gastrointestinal system showed necrosis of the
epidermal damage. The cardiorespiratory system was
anterior gastric wall parietal cells, acute upper jejunal
remarkable for cardiac haemorrhage and myocardial
distention, mitotic suppression throughout the entire
oedema, and the terminal bronchi showed features of
gastrointestinal tract and acute jejunal and ileal enteritis.
E 2005 The British Institute of Radiology
43

---

R E Goans and N Wald
Case 4 – Wood River Junction (CNS syndrome;
approximately 100 Gy)
The patient was a 38-year-old male with a negative
medical history. Following the initial criticality excursion
[15], the patient appeared stunned, ran from the building
and immediately vomited. He also experienced immediate
diarrhoea and complained of severe abdominal cramping,
headache, thirst and profuse perspiration. His initial vital
signs showed borderline blood pressure elevation and
tachycardia.
Approximately
4 h
post
accident,
the
patient experienced transient difficulty in speaking, hypo-
tension and tachycardia. A portable chest radiograph 16 h
post
admission
showed
hilar
congestion
(Figure 2).
Physical examination also showed the left hand and
forearm to be oedematous as well as left-sided conjuncti-
vitis and periorbital oedema (Figure 3). On day 2 the
patient became very disoriented, hypotensive and anuric.
The patient died 49 h post accident in cardiovascular
shock.
At autopsy, interstitial oedema of the left hand, arm and
abdominal wall was noted. Examination of the heart,
lungs and abdominal cavity revealed acute pulmonary
oedema, bilateral hydrothorax, hydropericardium, abdom-
inal ascites, acute pericarditis, interstitial myocarditis and
Figure 3. Wood River Junction patient 24 h post accident.
Note oedema in the left arm. Figure reprinted with permission
inflammation of the ascending aorta. Examination of the
of the New England Journal of Medicine.
gastrointestinal tract showed severe subserosal oedema of
the stomach and the transverse and descending colon. The
bone marrow was noted to be aplastic, and lymph nodes,
Early prediction of MOF in severe radiation
spleen and thymus were depleted of lymphocytes. The
accidents
brain showed minimal change, with rare foci of microglial
It is possible in many cases to predict when MOF will
change. The testes showed interstitial oedema and overt
be clinically important. Figure 4 presents data (N5114
necrosis of the spermatogonia.
cases) for the time to emesis post accident [16, 17] as
a function of dose. For time to emesis of less than 1 h
post accident, the median dose in this cohort is found
to be 6.5 Gy (interquartile range 5–11 Gy). Therefore,
patients who experience radiation-induced emesis within
1 h after an incident will likely experience MOF. The
sensitivity and specificity of emesis as a medical test
has been found to approach unity for absorbed dose
D.3 Gy.
Following either a c accident or criticality accident with
dose greater than a threshold of approximately 1 Gy c
equivalent, lymphocytes decrease from the peripheral
blood according to an equation of the form:
L(t)~L(0)e{Kt
ð1Þ
where L(0) is the lymphocyte count pre accident, L(t) is
the lymphocyte count at time t post accident, and the rate
constant K has been shown to be a linear function of dose
D [18]. Using serial complete blood count (CBC), if K.1
(lymphocytes decrease by more than a factor of 1/e in
24 h), then the predicted dose is greater than 8–10 Gy and
MOF is likely.
One of the most interesting aspects of the US criticality
cases is the high incidence of cardiac pathology (pericar-
ditis, pericardial effusions, myocarditis, etc.) noted at
autopsy. Whilst it is often noted that the heart is relatively
radioresistant, many of these historic accidents occurred
with the source near the thoracic level, likely giving
both a non-uniform dose distribution and a relatively
Figure 2. Portable chest radiograph 16 h post event in the
Wood River Junction accident. Figure reprinted with permis-
high dose to the heart and mediastinum. Radiation-
sion of the New England Journal of Medicine.
induced heart disease is generally described as a delayed
44
Radiation-induced multi-organ involvement and failure: a challenge for pathogenetic,
diagnostic and therapeutic approaches and research

---

Radiation accidents with MOF in the United States
Figure 4. Time to emesis as a function of dose with exponential fit y5ae2x/b. From the single exponential fit, the parameters (¡SE)
are found to be: a512.9¡1.2 Gy and b51.7¡0.2 h; r250.41.
Figure 5. Time to death as a function of dose for the four US criticality accidents. Power function fit: ln(y)5a+b ln(x);
b520.84¡0.04; r250.99.
manifestation of therapy, but acute effects both in humans
References
and in animal models have been well described in high
1. Ricks RC, Berger ME, Holloway EC, Goans RE. REAC/TS
dose cases [19–22].
Radiation Accident Registry: update of accidents in the
The US criticality cases are historically important since
United States, IRPA Proceedings, T-21. 1999. www.irpa.net/
they were fatal radiation accidents and medical treatment
irpa10/cdrom/00325.pdf
did not appreciably alter the clinical course of disease.
2. Voelz G. The SL-1 reactor. In: Stacy, Susan M, editors.
These cases can therefore serve as useful models of the
Proving the principle – a history of the Idaho National
natural course of acute radiation-induced tissue damage.
Engineering and Environmental Laboratory, 1949–1999. http://
Figure 5 presents the time to death in these four cases as a
www.inel.gov/proving-the-principle/chapter_15.pdf
[accessed
function of dose. A power function fit is presented to guide
25 January 2005].
the eye.
3. Rose BW. Fatal dose: radiation deaths linked to AECL
computer errors. June 1994. www.ccnr.org/fatal_dose.html
[accessed 25 January 2005].
Acknowledgment
4. Leveson N, Turner CS. An investigation of the Therac-25
accidents — Part III, Reprinted with permission. IEEE
Figures 2 and 3 are reprinted with permission of the
Computer 1993;26(7):18–41. http://courses.cs.vt.edu/ ,cs3604/
New England Journal of Medicine.
lib/Therac_25/Therac_2.html [accessed 25 January 2005].
E 2005 The British Institute of Radiology
45

---

R E Goans and N Wald
5. Gottfried K-LD, Penn G, editors. Radiation in medicine: a
14. Shipman TL, Lushbaugh LL, Peterson DF, Langham WH,
need for regulatory reform. Washington DC: Institute of
Harris PS, Lawrence JNP. Acute radiation death resulting
Medicine, National Academy Press, 1996:24–6.
from an accidental nuclear critical excursion. J Occup Med
6. USNRC. Loss of an iridium-192 source and therapy
1961;Suppl. March:145–92.
misadministration
at
Indiana
Regional
Cancer
Center
15. Karas JS, Stanbury JB. Fatal radiation syndrome from an
Indiana, Pennsylvania, on November 16, 1992. NUREG-
accidental nuclear excursion. N Engl J Med 1965;272:755–61.
1480 (February 1993).
16. Baranov AE, Guskova AK. Acute radiation disease in
7. Gilberti MV, Wald N. The Pittsburgh accelerator accident:
Chernobyl accident victims. In: Ricks RC, Fry SA. The
twenty-three-year follow-up of clinical and psychological
medical basis for radiation accident preparedness II. clinical
aspects. In: Ricks RC, Berger ME, O’Hara FM, editors.
experience and follow-up since 1979. New York, NY:
The
medical
basis for accident
preparedness
III, the
Elsevier, 1990:79–87.
psychological perspective. New York, NY: Elsevier Science
17. Goans RE. Clinical care of the radiation accident patient:
Publishing Co, 1991:199–208.
patient presentation, assessment, and initial diagnosis. In:
8. Gilberti MV. The 1967 radiation accident near Pittsburgh,
Ricks RC, Berger ME, O’Hara FM Jr, editors. The medical
Pennsylvania and a follow-up report. In: Hubner KF, Fry
basis for radiation accident preparedness, the clinical care of
SA, editors. The medical basis for radiation protection
victims. Proceedings of the Fourth International REAC/TS
preparedness. New York, NY: Elsevier Science Publishing
Conference on the Medical Basis for Radiation Accident
Co., 1980:131–40.
Preparedness; 2001 March; Orlando, FL. Boca Raton, FL:
9. Thomas ED, Rudolph RH, Fefer A, Storb R, Slichter S,
The Parthenon Publishing Group, 2002.
Buckner CD. Isogeneic marrow grafting in man. Exp
18. Goans RE, Holloway EC, Berger ME, Ricks RC. Early dose
Hematol 1971;21:16–7.
assessment in criticality accidents. Health Phys 2001;81:
10. Schenk R, Gilberti MV. Four extremity radiation necroses.
446–9.
Arch Surg 1970;100:729–34.
19. Fajardo
LF,
Berthrong
M,
Anderson
RE.
Radiation
11. Wald N, Pan S, Thomas ED. Cytogenetic observations in
pathology. New York, NY: Oxford University Press, 2001.
accidental human radiation exposure treated by marrow
20. Fajardo LF, Stewart JR, Cohn KE. Morphology of
transplantation. In: International Society of Hematology
radiation-induced
heart
disease.
Arch
Pathol
1968;86:
Congress; 1968; Boston, MA. International Society of
512–9.
Hematology. Abstract A-7, P. 4.
21. Gillette EL, McChesney SL, Hoopes PJ. Isodose curves for
12. McLaughlin TP, Monahas SP, Pruvost NL, Frolov VV,
radiation-induced cardiomyopathy in the dog. Int J Radiat
Ryazanov BG, Sviridov VI. A review of criticality accidents
Oncol Biol Phys 1985;11:2091–7.
2000 revision, LA-13638. Los Alamos, NM: Los Alamos
22. Stewart JR, Fajardo LF. Radiation-induced heart disease. An
National Laboratory, May 2000.
update. Prog Cardiovasc Dis 1984;27:173–94.
13. Hempelmann LH, Lisko L, Hoffman JG. The acute radiation
syndrome: a study of nine cases and a review of the problem.
Ann Intern Med 1952;36:279–510.
46
Radiation-induced multi-organ involvement and failure: a challenge for pathogenetic,
diagnostic and therapeutic approaches and research

---