Accepted Manuscript
Title: 210Po poisoning as possible cause of death: forensic investigations and toxicological analysis of the remains of Yasser Arafat
Author: Pascal Froidevaux Francois Bochud Sebastien Baechler Vincent Castella Marc Augsburger Claude Bailat Katarzyna Michaud Marietta Straub Marco Pecchia Theo M. Jenk Tanya Uldin Patrice Mangin
PII: S0379-0738(15)00411-9
DOI: http://dx.doi.Org/doi:10.1016/j.forsciint.2015.09.019
Reference: FSI8162
To appear in: FSI
Received date: 28-5-2015
Revised date: 25-9-2015
Accepted date: 30-9-2015
Please cite this article as: P. Froidevaux, F. Bochud, S. Baechler, V. Castella, M. Augsburger, C. Bailat, K. Michaud, M. Straub, M. Pecchia, T.M. Jenk, T. Uldin, P. Mangin, 210Po poisoning as possible cause of death: forensic investigations and toxicological analysis of the remains of Yasser Arafat, Forensic Science International (2015), http://dx.doi.org/10.1016/j.forsciint.2015.09.019
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Highlights
Yasser Arafat died of an unknown illness
We found unexpected high levels of Po in some of Yasser Arafat belongings
210 210
We found higher than expected level of Pb and Po in Yasser Arafat bone remains Bayesian analysis moderately supports the proposition that Mr. Arafat has been poisoned by
Po poisoning as possible cause of death: forensic investigations and toxicological analysis of the remains of Yasser Arafat
Pascal Froidevauxa, François Bochuda*francois.bochud@chuv.ch, Sébastien Baechlera, Vincent Castellab, Marc Augsburgerb, Claude Bailata, Katarzyna Michaudb, Marietta Strauba, Marco Pecchiac, Theo M. Jenkc, Tanya Uldinb, Patrice Manginb
aInstitute of Radiation Physics, Lausanne University Hospital, Lausanne, Switzerland bUniversity Centre of Legal Medicine, Lausanne and Geneva, Switzerland cPaul Scherrer Institute, Villigen, Switzerland
Prof. François Bochud, Institute of Radiation Physics, Lausanne University Hospital, Lausanne, Switzerland
Abstract
The late president of the Palestinian Authority, Yasser Arafat, died in November 2004 in Percy Hospital, one month after having experienced a sudden onset of symptoms that included severe nausea, vomiting, diarrhoea and abdominal pain and which were followed by multiple organ failure. In spite of numerous investigations performed in France, the pathophysiological
mechanisms at the origin of the symptoms could not be identified. In 2011, we found abnormal
levels of polonium-210 ( Po) in some of Arafat's belongings that were worn during his final hospital stay and which were stained with biological fluids. This finding led to the exhumation of
Arafat's remains in 2012. Significantly higher (up to 20 times) activities of 210Po and lead-210
(210Pb) were found in the ribs, iliac crest and sternum specimens compared to reference samples
from the literature (p-value < 1%). In all specimens from the tomb, Po activity was supported
210 210 by a similar activity of Pb. Biokinetic calculations demonstrated that a Pb impurity, as
identified in a commercial source of 3 MBq of 210Po, may be responsible for the activities measured in Arafat's belongings and remains 8 years after his death. The absence of
myelosuppression and hair loss in Mr Arafat's case compared to Mr Litvinenko's, the only
known case of malicious poisoning with Po, could be explained by differences in the time delivery-scheme of intake. In conclusion, statistical Bayesian analysis combining all the
evidence gathered in our forensic expert report moderately supports the proposition that Mr
Arafat was poisoned by Po.
Keywords: polonium, Yasser Arafat, toxicology, poisoning, human remains Introduction
The following presentation of the background events leading to President Yasser Arafat's death is based on the reading and analysis of the medical records provided to us by Mrs Suha Arafat, widow of Yasser Arafat and next of kin (1).
President Arafat was born on August 4th, 1929. For over three years prior to his death, he remained secluded in the quarters of the Palestinian Authority in Ramallah. He was said to be in general good health, and did not drink alcohol or smoke tobacco. On the evening of October 12th, 2004, approximately four hours after a meal, Yasser Arafat experienced severe nausea, vomiting, and abdominal pain followed by watery diarrhoea. On October 25th, 2004, a bone marrow
aspiration analyzed in Tunis showed a well populated bone marrow with mild myelodysplastic changes, the presence of megakaryocytes and an increased number of macrophages with hemophagocytic changes. An infectious work-up, including blood, fecal and urine cultures and an immunological evaluation were considered as normal. On October 29th, 2004, President Yasser Arafat was transferred to the French Military Hospital of Percy in Clamart, France. At the time he arrived in France, the diagnostic list included inflammatory enterocolitis, severe disseminated intravascular coagulation (DIC) with significant thrombocytopenia and bone marrow hemophagocytosis. Shortly after admission, the patient developed liver problems with cholestatic jaundice. The patient did not develop fever and the infectious work-ups, including AIDS tests, remained negative. All the microbiological investigations, as well as a urine test for the presence of gamma ray emitting radionuclides, were negative. Infectious, autoimmune or vascular processes, as well as a pathological tumour were decided to be very unlikely. Toxicological tests carried out by the Institute of Criminal Investigation of the French Gendarmerie (Institut de Recherche Criminelle de la Gendarmerie, IRCG ) on blood, urine, cerebrospinal fluid and stool taken on the 1st and 4th of November, 2004, revealed only the presence of the medication prescribed by the hospital. Further analyses were performed in urine and faeces on November 8th by the Radiological Protection Service of the French Armed Forces to search for radioactive substances based only on gamma emission, without positive results. The patient developed acute renal failure on November 6th, 2004. On November 11th, 2004 at 3:30 am, President Arafat died as a result of cerebral herniation secondary to cerebral haemorrhage. Despite the number of medical experts involved and the numerous investigations performed, the pathophysiological mechanisms at the origin of the group of symptoms experienced by Yasser Arafat could not be identified and no autopsy was performed. According to two letters dated
from April 2nd, 2012 and June 11th, 2012 from the Directorate of Radiation Protection of the Army (Service de Protection Radiologique des Armées, France) to our attention and one letter dated April 2nd, 2012 from the Directorate of the IRCG to the attention of Mrs Suha Arafat, the biological samples for toxicological and radio-toxicological analyses that were taken during his hospitalisation were destroyed in 2008.
Following upon the assassination of Alexander Litvinenko in London in November 2006 by
ingestion of polonium-210 ( Po), a suggestion was made that Arafat could have been poisoned
in a similar manner. Po is a short lived (T1/2 =138 d) alpha emitter (2) , which typically creates damage on a submillimetre scale, notably to the gastro-intestinal tract if ingested and then to inner organs and bone marrow via blood distribution.
On February 2012, Mrs Suha Arafat presented a bag containing Yasser Arafat's belongings from his final hospital stay, to the University Centre of Legal Medicine (CURML). She gave us her consent that we carry out all the necessary toxicological investigations that could potentially lead to new information about her late husband's death. She also gave us a signed letter giving explicit permission to publication in the scientific literature once our Final Expert Forensic Report had been handed to her and subsequently made public. The Institute of Radiation Physics
(IRA) works in close collaboration with CURML in the use of radiometric measurements in
forensic science (3; 4). The search for 210Po in the belongings was proposed within the context of
current knowledge regarding Litvinenko's poisoning. From a medico-legal perspective, the
hypothesis of poisoning with Po was justified given the absence of a characterised cause of illness and the absence of toxicological evidence, including gamma radiations. Additionally, to medico-legal experts, the sudden onset of violent gastrointestinal symptoms following a meal in a patient otherwise in general good health may indicate poisoning.
After studying the medical files in detail, we performed a workup on genetic identification as
well as toxicology and radioactivity measurements on Yasser Arafat's belongings that we
received. Our preliminary results of Po identification in the biological stains at unexpected levels of activity (5) led to the exhumation of the body, which took place in Ramallah on
November 27th, 2012. At this time, several specimens of bones, scalp, shroud, tooth and soil
were taken from the corpse and the grave for toxicological analysis, focusing mainly on Po
measurement. We report here the results of our full investigation concerning Yasser Arafat's
death and the possible involvement of Po as a cause of his death. Materials and Methods Exhumation procedure
We were requested to act as an expert team for the Palestinian Authority and Mrs Suha Arafat (hereinafter called the Swiss team). The Swiss team was composed of a medico legal expert, a radiation physicist, a radiochemist, a forensic anthropologist and a mortuary technician. A French team, acting on behalf of the French Justice Department after Suha Arafat launched a court case in Paris, and a Russian team, acting as an expert team on behalf of the Palestinian
Authority, were also present during the exhumation. The evening before the opening of the tomb,
we perforated a 12 mm diameter hole through the grave slab to measure the 222Rn gas
concentration (AlphaGuard, Genitron Instuments, Germany). It was essential to determine the
222 222 210 210
Rn air volume activity because Rn is the precursor of diagenetic Pb and Po which will
deposit on the grave surfaces and possibly contaminate the remains. In fact, during diagenesis, a
bone may accumulate contaminants from the burial environment, yielding a radionuclide activity
that is in excess of the bone's biogenically derived activity (4). The exhumation took place the
following morning, on November 27th, 2012. During the exhumation, the almost completely skeletonised corpse was not moved, but specimens from the remains (mainly fragments of bones and scalp) and surrounding materials (soil and shroud) were collected by a Palestinian medico legal expert in the presence of representatives from all teams. The state of Arafat's remains meant that it was not possible to identify organs specifically targeted by polonium such as soft tissues. This was a major drawback since polonium is not a bone-seeking radionuclide. All three teams received 20 similar specimens, whose description can be found in the supplementary information (SI) provided on the website of the journal. As far as we can judge, tampering with the corpse was impossible before the opening of the grave and during the manipulation of the remains.
Genetic identification
Genetic analyses were performed on a bone fragment and a tooth collected during exhumation. The material from the exhumation was ground and prepared by drilling at low speed. DNA was then extracted using a phenol/chloroform protocol. Identical nuclear DNA (nDNA) profiles comprising 15 and 13 loci (NGM Select kit from Life Technologies) were retrieved from the bone fragment and the tooth, respectively. Those nDNA profiles corresponded to those previously obtained from the belongings of Yasser Arafat (5).
Toxicological screening
We selected four human remains sampled during the exhumation for systematic toxicological analysis, according to international recommendations (6). Gas chromatography coupled to mass spectrometry (GC-MS) was used for the detection and identification of compounds. The
specimens were diluted into methanol before GC-MS analyses (Agilent 5973N). Specimens were also analyzed after acetylation. Details of the procedure are given in the SI.
210Po determination
To obtain a high activity, which is necessary in case of poisoning, Po is produced artificially
by irradiating a 209Bi target with neutrons in a nuclear reactor. However, 210Po is also the decay
product of 210Pb (Ti /2- 22 y), a naturally-occurring radionuclide which is the result of the decay
222 238 210
of the natural radioactive gas Rn in the U decay chain. When Pb contaminates a sample,
Po will accompany it with a similar activity after about 2 years as a radioactive secular
equilibrium establishes between both radionuclides. In this case, Po is said to be supported by
210 210
Pb. In this sense supported Po is apparently of natural origin. In Yasser Arafat's case, the
210 210 210 presence of natural Po and Pb will possibly mask the presence of artificial Po, which - if
present as a poisoning material at the time of death - would have decreased by more than two
million by the time of exhumation, due to radioactive decay (number of Po elapsed half-lives
of 21). The main problem for determining the level of artificial Po in Yasser Arafat's remains,
210 210
buried eight years earlier, is the potential presence of diagenetic Pb and Po.The diagenetic
210 210 contribution of Pb to the total activity of 210Pb is the contribution that is not the consequence
of biological uptake during life time. The diagenetic activity is solely derived from accumulation
of contaminants from the burial environment. If diagenesis is present, it will yield a radionuclide
activity that is in excess of the bone's biogenically derived activity Therefore, solubility
profiling was used to remove diagenetic contaminants, after residual tissue extraction using a
basic solution of 1 M NaOH and mineralization in 10% oxygenated water in a pressurized
microwave digester (Ultraclave IV, Milestone, Germany). Solubility profiling is based on the
differential solubility of bone hydroxyl-apatite from biogenic or diagenetic origin (7). A soft
acidic buffer (10 ml, ammonium acetate 0.2 M at pH 4.5) was used to remove layers of potentially contaminated bone hydroxyl-apatite (1.0 g) in six successive washings. The washing time was 5 minutes each in an ultrasonic bath, afterwards the bone matrix was totally dissolved in 1 M HCl. After six washings, about 10% of the mass of the bone was removed.
Po was spontaneously deposited on a silver disk, using its specific electrochemical properties. Then the disc was measured using alpha spectrometry on a PIPS detector (Canberra Alpha Analyst, France), using a 209Po tracer as an internal standard (4). To quantify 210Pb activity, gamma spectrometric measurements were performed on several grave specimens (soil, bone and
shroud) using a Canberra p-type HPGe well detector (GCW4523, Canberra, France) in a 4 ml
well-geometry. Typically, 210Pb can be quantified using gamma-spectrometry if the activity of
the sample is above 0.1 Bq. For a lower activity level, determining the Pb was carried out by
separating lead from polonium on a strongly basic resin (Dowex AG1x8, Bio-Rad) because
2 2+ 210 [PoCl6] - anion is extracted on the column, while Pb is not. Then, polonium-free Pb was left
to rest for at least three months to allow the secular equilibrium to partially re-establish. After a
210 210 three month period, Po activity reaches 36% of the initial Pb activity and was measured
using alpha spectrometry as described previously.
Several studies dating back to 1963 were used to obtain a dataset of Po reference
concentrations in human bones (3; 4; 8-11). For statistics, these reference values were separated
into two sub-sets, the first one containing the Po data from autopsy (n=95) and the second one
containing Po data from bones sampled after exhumation or found in the environment, e.g.
buried in soil, coming from forensic cases (n=36). We compared these datasets to the dataset of
Po values found in Yasser Arafat bones (n=16).
Unfortunately, an international inter-laboratories comparison for Po determination in bone
samples does not exist. Thus, the quality control was done through the participation to several
international inter-laboratories exercises on Po determination in urine samples organized by the PROCORAD association (12), using a somewhat similar procedure to the one described above (2008-2011, n= 5; average bias: 4.2%).
Pb impurity: calculation and analysis
To check for 210Pb impurity in Po made by irradiating a Bi target, a 3 MBq 210Po source
(certificate n° 9031-OL-501/13) was bought from the Czech Metrology Institute. Polonium was
210 210 separated from lead to authenticate the presence of 210Pb as an impurity. Pb was measured
using gamma-spectrometry both before and after chemical separation and its presence confirmed
by proportional counting of its short half-life (5.01 d) daughter product Bi (Tennelec LB4100w). We also separated the lead from four bone specimens, two soils and a scalp
207 206
subsample to measure the ratio of stable lead Pb/ Pb by mass spectrometry (Element 2,
Thermo Scientific), with an accuracy better than 0.1 % (13). The rationale for this measurement
206 210 210 was that Pb is also the daughter product of Po, thus the past presence of a high level of 210Po
207 206
will yield a lower Pb/ Pb ratio after decay. Biokinetic calculation
The systemic biokinetic model of polonium proposed by Legett and Eckerman (14) was
implemented in the simulation modeling tool Ecolego (15) and in SAAM II (16) in order to
210 210
calculate the typical retention of 210Po and 210Pb in organs and tissues in case of intake. Since ingestion is the most probable route of intake for poisoning, the systemic biokinetic model was
coupled to the human alimentary tract model as described in ICRP Publication 100 (17).
Absorption of Po is assumed to occur exclusively from the small intestine and is characterized by f1=0.1 for inorganic form.
Further details of the radiometric, toxicological, genetic and statistical tests, as well as the exhumation procedure and biokinetic calculation procedure can be found in the SI.
Bayesian analysis
Statistical Bayesian analysis is well suited to coherently combine evidence from various origins, in our case medical records, toxicological and radiotoxicology measurements, as well as personal judgements of probability (18). It fit the goals of the present forensic expert report, which was to
estimate the probability of occurrence of a unique event and to evaluate all the evidence in the
light of two alternative hypotheses: H0 Mr Arafat has not been poisoned by Po; Hi Mr Arafat
has been poisoned by Po. It was not our task to give an estimate of the probability of Hi, which contains too many elements that are out of our scope. Our task was to evaluate the likelihood ratio (LR):
P (E|H0)
where P(E|H1) is the probability of observing a given evidence E if Hi is true and P(E|H0) is the probability of observing a given evidence E if H0 is true. We cannot overemphasize that P(E|H1) is not equal to P(H1|E). The former is the probability of observing a given evidence E if Mr Arafat has been poisoned by polonium, whereas the latter is the probability that Mr Arafat has been poisoned by polonium if we observe evidence E.
The likelihood ratio LR measures the relative strength of support which evidence E gives to H1. One of the main advantages of the LR is that it can easily be updated with new evidence: the updated LR is simply the previous LR multiplied by LR associated with the new evidence.
Results
Before presenting and discussing the results of our analyses carried out on Yasser Arafat's remains, it is important to mention that in a previous study (5), toxicological tests on Yasser
Arafat's belongings remained negative with only the patient's medication and metabolites
indentified, but that Po tests were positive on some specimens, particularly on a possible urine stain on his underwear (Figure 1).
4.0 -q
U z^ -
J2 2.0-
O 1.5-
underwear A # 1 underwear A # 5 chapka #1 toothbrush bristles other samples
I I I | I I I—I—I—rr
4.8 5.0 5.2
Energy [MeV]
reference samples
1 I 1 I I | I" 4.8 5.0 5.2 Energy [MeV]
Figure 1. Polonium alpha spectra of Yasser Arafat's belongings visually stained by biological fluids (left) and reference samples (right). Reference samples included belongings from Yasser
Arafat that had not been worn and fabric samples from some IRA collaborators kept in an attic in plastic bags for 10 years. Spectra are normalized to the internal tracer 209Po (50 mBq).
Statistics revealed strong differences between the activities of Yasser Arafat's worn belongings and Yasser Arafat's unworn belongings (p-value < 1 %) and between Yasser Arafat's worn belongings and IRA reference fabrics (p-value < 1 %). Conversely, no statistical difference was
found between Yasser Arafat's unworn belongings (internal references) and our own reference
fabrics. Interestingly, 42 % of the Po from the most contaminated sample was supported by
210 210 210
Pb. Full results of the Po and Pb measurements in the belongings can be found in the SI
of Froidevaux et al. (5).
These initial findings prompted the exhumation of Yasser Arafat's remains for further toxicological analyses. Genetic identification of the human remains from the grave guaranteed that all the specimens were from Yasser Arafat. Only Arafat's medication and its metabolites
were found in the human remains. The activity of the Rn gas in the grave was 12000 Bq/m , which is a value not especially elevated (for instance, our own measurements in Switzerland performed with the same instrumentation led to a mean value of 35,000 Bq/m on the Swiss Plateau and 90,000 Bq/m in the Jura Mountains (19)). This level validates the comparison of the activities of Yasser Arafat's remains with the activities of bones from other exhumations. Surface activity (mBq/cm2) was determined in three soil specimens, in six shroud fragments, in one scalp fragment and two fragments of a flat bone (iliac). One of the soil specimens was sampled below the abdomen cavity; it contained 42 % relative humidity (Hr) and had a strong black colour compared to 26 % Hr for a red soil sampled far from the corpse (reference soil from the front right corner of the tomb). Thus the stained soil is believed to contain decayed body fluids. Its surface activity was 30±4 mBq/cm whereas the surface activity of the reference soil
was only 1.7±0.3 mBq/cm . The surface activities of the shroud and the reference soil (located away from the corpse) can be considered as representative of samples not contaminated by the corpse and possibly only exposed to radon decay product. The mean activity was
3.9±0.7 mBq/cm. Wax (decomposed tissues, from 0.5 to 1 mm thickness) was carefully
removed from a surface of the iliac flat bone and its Po activity determined. Wax is known to
be a good interceptor of airborne radioactive particles and paraffin wax is used near nuclear
installations to check for radioactive aerosols. Therefore any contamination of the iliac by Rn
210 210
progeny ( Pb and Po) should be concentrated in the wax coating the bone.
The soil specimen located left from the corpse was not statistically different, whereas the surface activities of the iliac crest waxed tissue, the scalp and the black stained soil located under the abdominal cavity were all significantly higher (p-value < 10-5).
Results are presented in Figure 2.
shroud specimens
iliac crest
"D O .Q
Figure 2. Surface activity of Po (mBq/cm ) of some specimens sampled from the grave during the exhumation. Deposition on the iliac crest was calculated using activity measured on the wax coating the iliac, divided by the surface of the iliac flat bone fragment.
210 210 210 210Po and 210Pb were measured in 16 fragments of bone and results showed that the Po was
210 210 210 supported by Pb, e.g. the activity of Pb was at least equivalent to the activity of 210Po.
Nevertheless, some of Arafat's bone fragments contained much higher (up to 20 times) Po and
Pb activities than the reference bones (Figure 3). Furthermore, the isotopic ratio of stable
207 206 207Pb/206Pb
was lower in the rib and scalp specimens, which were precisely the specimens with
the highest Po activity (see SI for details).
The Yasser Arafat results were compared to previously published autopsy data (4; 8-11). The
210 210
Po activities differed significantly with 8 out of 16 Po values with a p-value lower than 1 %.
To take into account the possibility of diagenesis, even after solubility profiling, the Yasser
Arafat dataset was further compared to the exhumation dataset (3; 20; 21). The Po activities
differed even more significantly, with 15 out of 16 210Po activities having a p-value lower than 1%.
"с? CL
800 ■
600 ■
400 ■
200 ■
□ vertebra X rib A femur X sternum + iliac
smoker: YES NO unknown
-лр-
unknown 0
---1-1-1— —Г
10 20 30 40 50 Age at death [years]
x n "^aifb aft
YA (75)
ro О га 1т m JE
References O cortical / trabecular bones
This case □ vertebra X rib A femur 1X1 sternum + iliac
1 Г unknown 0
-<£-^Sj> 0<>Y-O-jOD-
10 20 30 40 50 60 Time after death [years]
1170 1270 YA (8)
Figure 3
3a. 210Po activities (mBq/g Ca) of all bone specimens from Yasser Arafat (right) compared to references activities from the autopsy dataset, reported versus the age at death.
3b. 210Po activities (mBq/g Ca) of all bone specimens from Yasser Arafat (right) compared to references activities from the forensic dataset, reported versus the known or estimated time since death.
In addition, several sub-specimens of the iliac crest were measured. The Po activities were
found to increase from the inner part to the outer part of the iliac bone (Figure 4). In all of the
210 210 210 iliac sub-specimens, Po activities were also supported by Pb. The results of all the Po and
Pb measurements carried out in specimens taken from the grave can be found in Table SI-2.
iv3 s*
<£> %
Figure 4. Locations and Po activities in the iliac crest bone.
210 210 Since 210Po was supported by Pb in all of the measured specimens, it is possible that the
presence of 210Po was natural, in spite of all the precautions taken to decontaminate the bones.
Nevertheless, diagenesis in Yasser Arafat's remains should be lower compared to other exhumed remains because his body was not buried. It simply rested on a thin layer of 10 cm of soil, wrapped in a shroud at the time of inhumation. Thus bone specimens were not spoiled by soil
particles and the corpse was protected by the shroud, at least during the time necessary for the
shroud to decompose. Another explanation could be that Pb was present as a by-product of
Po production by neutron irradiation, and not separated from polonium due to the formation of a volatile lead polonide, PbPo (22; 23). Calculations carried out by the Paul Scherrer Institute,
Switzerland, showed that the production of some 210Pb during 209Bi activation by neutrons is
210 210 -12 208 possible, but with a very low Pb/ Po activity ratio of around 10" , depending on the 208Pb
impurity into the 209Bi target taken into account. Nevertheless, an impurity of 210Pb at a much
higher level was found in a commercial source of 210Po of 3 MBq (see SI for details). The results
210 210 of the measurements show that Pb was present as an impurity with an activity ratio of Pb/
210 7 5
210Po of 1.5x10 . Only an elapsed time of 2.5 years is needed to reach an impurity factor of 10- , starting from the commercial source that was investigated. Taking into account the difference in
half-life between 210Po (138 d) and 210Pb (22 y), 210Pb rapidly becomes an impurity that might
210 210 210 explain the Pb and Po activity levels found in Yasser Arafat's remains. Additionally, 210Pb
is a bone-seeking radioelement that will accumulate in the skeleton ten times more compared to
210 210
Po (24). A calculation using a scenario where a lethal activity of 1 GBq of 210Po was ingested
with a 210Pb impurity ratio of 10-7, 10-6 and 10-5 is presented in Figure 5.
_CD CD
10 105 104 103
'I...........I...........I...........I...........I1
2 3 4 5 6 Time after death [years]
Figure 5: Total activity in the skeleton following a Po intake of 1 GBq and death 30 days
afterward. Three scenarios were computed with impurities of 210Pb in the source of 0.1 kBq
(impurity factor of 10"7), 1 kBq (10-6) and 10 kBq (10-5). The activities of 210Po and 210Pb reach
equilibrium after about 8.5 years, 7.1 years and 6 years, respectively.
Discussion
Radioanalytical results
The time elapsed since Yasser Arafat's death and this study is eight years. During these eight
years, artificial 210Po in Yasser Arafat, if present at death, would have decayed by about two
million (number of elapsed Po half-lives of 21). This is the major limitation of our overall
forensic report because by the time we performed our measurements any residual artificial 210Po
has possibly reached the natural level of activity found in human bones. Nevertheless, a
significant activity of Po was found in Yasser Arafat's belongings and remains. Certain bone
specimens, such as rib, iliac and sternum, had much higher Po activity compared to the
published literature on this subject. Thus the Po activities found in Yasser Arafat's remains do
not fall within a normal range of activity for unexposed persons. These findings are in
accordance with the unexpected level of Po previously found in some of Yasser Arafat's
belongings (5). However, Po could not be attributed unequivocally to the ingestion of an
artificial source of polonium given that it was supported by the same activity of Pb. In this
210 222 respect, Po may appear natural and its presence explained as the consequence of Rn decay
within the tomb.
We ruled out this hypothesis as an explanation for the apparent secular equilibrium between
210 210
Po and Pb because a very careful decontamination process based on solubility profiling of the bone matrix was carried out. During this process, 10% of the mass of the bone is removed, to expose only inner, and therefore contamination-free, surfaces. Additionally, most of the bones were still covered by a thin layer of wax that was mechanically and chemically removed before solubility profiling. This means that the wax should have been the receptor of diagenetic
contamination rather than the bone structure. In our laboratory, 15 cases of forensic interest were
210 210
treated similarly in a previous study and revealed no such 210Po and 210Pb activities (3) as those
found in Yasser Arafat's rib, sternum and iliac. In addition, the dataset from Swift (20) and Ziad
et al. (21) contain Po measurements of bones from exhumation that were only decontaminated
mechanically, mainly by abrasion with sand paper; the maximum Po activity measured was 15 mBq/g Ca, a value far below those measured in Yasser Arafat's rib, sternum and iliac. In a prior study, we used a dataset of 20 exhumed vertebrae which had been buried for 20 years, to test the level of diagenesis in the trabecular bones in contact with soil particles (4). Without decontamination, the maximum activity found in bone with strong diagenetic contamination was
300 mBq/g Ca, an activity which lowered further to 36 mBq/g Ca after 6 washings with acetate
buffer. Once again, no Po activity such as those measured in Yasser Arafat's remains was
210 210
found. Another finding points to a biogenic origin of Po and Pb; strong heterogeneity was found in the activities measured along an inner-outer transect in the iliac bone, with a steady
increase in activity from the inner to the outer part (Figure 4). Additionally, only highly
210 210
vascularised bones such as the iliac, rib and sternum contained very significant Pb- Po
activity, while cortical bone such as the femur had somewhat close to background activity. This
210 210
finding is not compatible with a diagenetic origin of Pb- Po but matches a distribution pattern coherent with an incorporation of a bone-seeking element shortly before death (24).
207 206
While we do not put emphasis on the results of the stable 207Pb/206Pb ratio because of large
uncertainties in the measurement, the trend of low ratio values for the two specimens with the
210 210 highest measured Po activities might indicate the presence of a high 210Po level in the past.
210 210
Additionally, if the Po and Pb measured in the body were of a diagenetic origin, similar surface activities on the reference soil and the shroud would be expected on the one hand, and on
the black stained soil and the scalp on the other hand. This was not the case and the latter values
are significantly higher. Therefore, and knowing that the activity concentration of Rn
measured before opening the tomb was relatively low for a soil, we are confident that the 210Po 210
and Pb measured in this study are of biogenic origin.
Nevertheless, the question remains regarding the provenance of Pb in Yasser Arafat's bones.
We measured a commercial Po source of 3 MBq and found it to contain a significant activity
210 210 210 of Pb impurity able to explain the high Po and Pb activities found in Yasser Arafat's
210 210
remains, as well as the presence of a secular equilibrium between Po and Pb eight years
after death. Furthermore, Pb is a bone-seeker and has a long half-life of 22 years. Its presence
210 210 210 210 as an impurity in a Po ingested source will leave a Pb- Po trace for much longer than Po
alone, as was demonstrated using a biokinetic calculation (Figure 5). Additionally, there is strong
evidence that 210Pb may be a significant impurity in a 210Po source made by irradiation of 209Bi.
Medical file
As already mentioned, the assassination of Alexander Litvinenko represents the only openly
documented Po poisoning case. Although it has not been described in the conventional scientific literature, the evidence from the Litvinenko case have been made public on the internet (25). The inquiry showed that Mr Litvinenko's contamination probably resulted from one substantial intake, although the results of autoradiography measurements of alpha emitter performed on his hair suggest that a much smaller contamination may have occurred a couple of days earlier (26; 27) . The first reported symptoms were vomiting and diarrhoea, together with abdominal pain followed by alopecia and myelosuppression. Then multiple organ failure, including kidney and liver deficiencies, led to his death.
Comparatively, Yasser Arafat's medical records essentially showed gastrointestinal symptoms at the onset, followed by liver and renal damage and ended with multiple organ failure. Multiple organ failure is, however, a non-specific final issue that can frequently not be attributed to a specific triggering event. Although mild myelodysplastic changes of the bone marrow were found about two weeks after the first symptoms, myelosuppression and alopecia were not observed in Yasser Arafat. At first glance, this striking difference between the two medical records suggests polonium poisoning in the case of Yasser Arafat must be ruled out. However, before drawing a definite conclusion, it is important to take into account the difference between
external irradiation and radionuclide intake, and how the time delivery-scheme of intake could affect the observed symptoms.
Ionizing radiations, including gamma rays and alpha particles are known to damage biological
molecules, including DNA. Po is an alpha emitter, which is particularly destructive at the
cellular level as alpha particles deposit all their energy within tens of micrometers. Therefore,
radiological damages induced by Po are not uniform within the whole body, because they are directly related to the location of the individual polonium atoms, which is defined by their specific biokinetic. This is very different than external irradiation with penetrating radiation, like photons or neutrons that tend to produce relatively homogenous dose deposition within all exposed organs. While persons exposed to high doses of external irradiation tend to develop haematopoietic, gastrointestinal, cardiovascular, central nervous system and skin reactions, the
reaction following the intake of alpha-emitting nuclides may be very different and is much less
documented. In the case of 210Po intake, the toxicity is influenced by the specific tissue sensitivities and the accumulation of the radionuclide in the different organs.
After ingestion of Po, it is considered that the gut is the primary target for the initiation of prodromal responses, including vomiting and diarrhoea (28). The toxic substance is then absorbed into the blood (f1 = 10%) and redistributed to the organs where it can be incorporated and distributed, mainly to the liver and the kidneys (14; 29).
Tissue effects depend on the organ and have a dose threshold that varies from person to person, and below which nothing is detectable without conducting a clinical test. This means that
depending on the amount of intake and its time delivery-scheme the organs that show an effect
could vary. For Po, the available animal and human data indicate that a sufficiently high intake
can lead to death after multiple organ failure. For lower intakes, animal studies have shown protracted effects resulting only in kidney damage and then death (27). Other animal studies have shown that low intakes did not provoke haematological abnormalities and that only mild kidney lesions could be noticed (29; 30). The same studies also showed that the initially observed damage to the bone marrow could be followed by an active regeneration.
All these considerations allow us now to examine the plausibility of observing the reported
symptoms of Mr Arafat in light of 210Po poisoning. In this forensic context, we considered different possible time delivery-schemes. We will therefore discuss in the next paragraphs two situations: one in which a single high activity was ingested and another in which smaller activities were ingested repeatedly.
The fact that myelosuppression and hair loss are missing in Mr Arafat's case does not support a
single high activity intake of Po. However, side-effects observed with high activities injected or ingested in nuclear medicine therapy include nausea, vomiting, fatigue and abdominal pain, but less frequently hair loss. Furthermore, myelosuppression varies greatly among patients treated either with beta-emitters (31) or alpha-emitters (32). As a matter of fact, the extent of myelosuppression to be expected following intakes of radionuclides will differ according to their chemical and physical characteristics. For example, the effect of alpha particles emitted from radium isotopes deposited in bone mineral will be limited mainly to peripheral marrow and other parts of the marrow will be unaffected. Recent treatments with bone-seeking alpha-emitters such
as Ra confirmed that myelosuppression was then minimal (32; 33). According to the ICRP
intake model, Po tends to be deposited on the bone surface during the first month after intake. Because of the short range of alpha particles, the dose delivered from the bone surface to the active marrow is highly heterogeneous and large fractions of the active marrow are spared (33).
The uptake of 210Po on the bone surface is thus not expected to systematically cause severe
myelosuppression. It is then only the amount of Po absorbed into the blood that contributes to
the bone marrow dose. All this considered, the absence of myelosuppression and hair loss does
not support a single high intake of Po but does not exclude the possibility.
In the scenario of repeated intakes of small Po activity, we can picture that the "last straw" trigged the clinical symptoms, which were subclinical after previous intakes. In Yasser Arafat's case, low marrow cellularity (i.e. increased adipocyte concentration) associated with aging might also have attenuated the myelosuppression. Indeed, as marrow cellularity decreases, alpha particles increasingly encounter adipocytes and the energy deposited into active marrow is thus decreased (34). This could potentially reduce myelosuppression in the elderly. It is also possible that the intake could be too low to damage the red bone marrow or that the transiting damage could recover during the interval between two successive intakes as it was observed in animal studies (30). The same argument could possibly explain the absence of hair loss through recovery of the skin basal cells.
As already mentioned, the alimentary tract is considered to be the primary radiation target of
Po involving the release of a neurotransmitter from intestinal cells that activates the brain-vomiting centre (26) . This is further supported by the fact that the present biokinetic model probably underestimates the dose as it does not take into account the possibility of retention in the gastrointestinal tracts (28). This explains the presence of vomiting and diarrhea in the Litvinenko case. For Yasser Arafat, no explanations for these observed symptoms were found despite very extensive clinical investigations. Harrison et al. (28) also reported results from a rat study suggesting that damage to the gut mucosa was the possible cause of death in a case of
210 210 ingestion of 210Po. After ingestion of Po, the gastrointestinal syndrome, associated with
multiple organ failure, could therefore be a predominant cause of death (35; 36). Russian forensic expert report
Two other teams also took part in the exhumation of Yasser Arafat and analyzed similar samples: a French team commissioned by the French justice department and a Russian team commissioned by the Palestinian Authority. Contrary to the French, the Russians published their
results (37) . Their analytical results were very similar to our measurements. They also measured
210 210
bone activities of Po and Pb at equilibrium that were much higher than normal. They tested
two hypotheses: contamination by radon in the tomb and ingestion of a Pb source shortly
before death. They rejected both hypotheses and concluded that "unfortunately, it is not possible
to state clearly the cause of the Pb radionuclide presence in the samples only on the basis of
the physical research results". Concerning the medical file, they considered that Po poisoning should have led to a full spectrum of acute radiation syndrome. Specifically, they were expecting myelosuppression and hair loss. They did not take into account our measurements on the
personal belongings (5) and concluded that "a direct causal link between the presence of high
210 210
content of [210Po and 210Pb] in the remains of the deceased and his death should be excluded". Bayesian analysis
The evidence gathered during this expert report are not clear-cut: we cannot exclude Po as a
cause of death, but we cannot be sure that Po was the cause of death. We have three main contributions to take into account for calculating LR: the measurements performed on the
belongings, the measurements performed on the exhumed body remains, and the analysis of the medical files. We estimate each contribution separately.
The measurements performed on the personal effects showed large quantities of unsupported
Po associated with the presence of biological fluids. Although this has been published for two years (5), no alternative explanation to polonium poisoning has been proposed.
The measurements performed on the remains showed a high quantity of supported Po but were about 20 times higher than all other exhumed cadavers documented in the scientific literature.
These high activities cannot be explained by the presence of radon in the tomb but are coherent
210 210 with the ingestion of a Po source containing an impurity of Pb akin to what we measured in
a commercial source.
Concerning the medical file, the situation is similar. Yasser Arafat did not show all possible symptoms observed in the case of acute radiation syndrome. In particular, no myelosuppression and no hair loss were documented. In a forensic expert report, we have to factor in the variability of reactions that are probably present. For instance, Mr Arafat was 32 years older than Mr Litvinenko at the time of their deaths. Mr Litvinenko lost his hair and showed a clear myelosuppression, but, as discussed before, this is not always documented in radiation therapy patients of nuclear medicine. If we open the range of possible poisoning scenarios, we should also consider possible repeated smaller intakes. As mentioned above, this type of scenario could lead to gastro-intestinal symptoms that could appear once a certain threshold has been exceeded and then develop into multiple organ failure.
We can now process all this information using a Bayesian analysis, and estimate the LRs associated with each piece of evidence. We estimated that the LR of high activity of unsupported
Po in the personal belongings is larger than one: in other words, it is more likely to observe
unsupported Po in the biological stains if Mr Arafat had been poisoned than if he had not. For
the high activities of supported Po found in the remains, the LR is also larger than one because, to our knowledge, no documented cadaver has been measured with such an activity.
Furthermore, no plausible alternative explanation has been proposed and the presence of a small
210 210
impurity of 210Pb in a Po source could lead to what has been observed in Mr Arafat's bones. The medical file analysis points toward the opposite, because of the differences between Mr
Arafat and Mr Litvinenko. Consequently, we estimated that the corresponding LR is smaller than
one. However, LR is certainly not zero because a repeated series of small intakes of Po or a specific weaker cellular reaction of Mr Arafat could also explain the observed symptoms and, mostly, because the Percy Hospital physicians did not ultimately identify the cause of Yasser Arafat's death.
Putting all of this together, we estimate that the LR associated with the medical file is not small enough to compensate the relatively large LRs associated with the measurements on Mr Arafat's personal belongings and remains. Therefore, we have to deduce that our radiotoxicology
analyses and the observed medical symptoms support the proposition that Mr Arafat was
poisoned by Po. This support is maybe not strong, but certainly not slight. Conclusions
Despite extensive clinical investigations, the cause of the gastrointestinal symptoms followed by
multiple organ failure developed by Yasser Arafat was not determined at time of his death. Our
study presents analytical results that support the hypothesis of Po poisoning essentially on the
basis of unexplained Po activities found in Yasser Arafat's belongings worn shortly before his
210 210
death and on unexplained Pb and Pb activities found in his remains.
Alexander Litvinenko's case is at this time the only documented case related to acute radiation
syndrome following upon Po malicious administration. The comparison of these two cases showed clinical similarities except for alopecia and myelosuppression experienced by Alexander
Litvinenko. Nevertheless, this difference is not an argument to exclude with sufficient certitude
the hypothesis of Po poisoning if we consider that the administration scheme could have been different in the two cases: an acute poisoning by administration of a substantial dose versus poisoning by the repetition of smaller doses. Moreover, it should also be taken into account all possible interfering factors such as age, previous health status and genetic predisposition—all of which could lead to different symptoms.
In conclusion, even if our findings reasonably support the hypothesis of Po poisoning, Yasser Arafat's cause of death will probably remain a cold-case.
Acknowledgement
Role of the founding sources
There are no financial or other relations that could lead to a conflict of interest. However, the costs of the analyses were covered in equal part by Mrs. Suha Arafat and the Palestinian Authority (PA). In addition, due to security concerns, the PA accommodated the Swiss team during its stay in Ramallah. Neither Mrs. Suha Arafat, nor the PA had a role in study design, data gathering, data analysis, data interpretation, or writing of the report. All the other costs of the
expertise were covered internally by the Lausanne University Hospital. The Expert Forensic Report concerning the Late President Arafat that we wrote in our quality of Swiss Experts Team was handed jointly to both Mrs. Suha Arafat (represented by her legal counsel) and to the PA officials during a meeting in Geneva, November 5th, 2013. The corresponding author had full access to all the data in the study and had final responsibility for the decision to submit for publication.
References
1. Arafat S. (Nov 18, 2004). Medical records from Percy Hospital provided by Mrs Suha Arafat. Available on http://www.yaf.ps/yaf/news_details.php?pid=81.
2. LNHB. Recommended Data. http://www.nucleide.org/DDEP_WG/DDEPdata.htm
3. Schrag B, Uldin T, Mangin P, Bochud F, Froidevaux P. 2014. Dating human skeletal remains using Sr-90 and Pb-210: Case studies. Forensic Science International 234
4. Schrag B, Uldin T, Mangin P, Froidevaux P. 2012. Dating human skeletal remains using a radiometric method: Biogenic versus diagenetic Sr-90 and Pb-210 in vertebrae. Forensic Science International 220:271-8
5. Froidevaux P, Baechler S, Bailat CJ, Castella V, Augsburger M, et al. 2013. Improving forensic investigation for polonium poisoning. Lancet 382:1308-
6. TIAF. International Association of Forensic Toxicologists. http://www.tiaft.org/data/uploads/documents/tiaft-sta-laboratory-guidelines.pdf
7. Sillen A. 1986. Biogenic and diagnentic Sr/Ca in pliopleistocene fossils of the Omo Shungura formation. Paleobiology 12:311-23
8. Ham GH, SA; Youngman, MJ; Etherington, G; Stradling, GN. 2003. Review of autopsy, in vivo and bioassay measurements of the member of the public in the UK. National Radioprotection board.
9. Henshaw DL, Hatzialekou U, Randle PH. 1988. Analysis of alpha particle autoradiographs of bone samples from adults and children in the UK at natural levels of exposure Radiation Protection Dosimetry 22:231-42
10. Holtzman RB, Ilcewicz FH. 1966. Lead-210 and polonium-210 in tissues of cigarette smokers. Science 153:1259-60
11. Osborne RV. 1963. Lead-210 and polonium-210 in human tissues. Nature 199:295-&
12. PROCORAD. Association de promotion des controles qualite en radiotoxicologie et des bonnes pratiques de laboratoire. www.procorad.org
13. Yoshinaga J. 1996. Isotope ratio analysis of lead in biological materials by inductively coupled plasma mass spectrometry. Tohoku Journal of Experimental Medicine 178:37-47
14. Leggett RW, Eckerman KF. 2001. A systemic biokinetic model for polonium. Science of the Total Environment 275:109-25
15. Ecolego. 2011. ECOLEGO-a toolbox for radio ecological risk assessement. Proceedings of the International Conference on the protection from the Effects of Ionizing Radiation. http://ecolego.facilia. se/ecolego/show/HomePage :229-32
16. Barrett PHR, Bell BM, Cobelli C, Golde H, Schumitzky A, et al. 1998. SAAM II: Simulation, Analysis, and Modeling Software for tracer and pharmacokinetic studies. Metabolism-Clinical and Experimental 47:484-92
17. ICRP. 2006. Human alimentary tract model for radiological protection: ICRP Publication 100. ICRP 36
18. Franco Taroni SB, Alex Biedermann, Paolo Garbolino, Colin Aitken,. 2010. Data Analysis in Forensic Science: A Bayesian Decision Perspective. John Wiley and Sons, Ltd.
19. Kropat G, Bochud F, Jaboyedoff M, Laedermann JP, Murith C, et al. 2015. Predictive analysis and mapping of indoor radon concentrations in a complex environment using kernel estimation: An application to Switzerland. Science of the Total Environment 505:137-48
20. Swift B, Lauder I, Black S, Norris J. 2001. An estimation of the post-mortem interval in human skeletal remains: a radionuclide and trace element approach. Forensic Science International 117:73-87
21. Ziad N, Zarki R, Benmansour M, Sayerh T, Laissaoui A. 2012. Determination of Pb-210 in human skeletal remains from Morocco: implications for time since death assessment. Journal of Radioanalytical and Nuclear Chemistry 292:315-9
22. Feuerstein H, Oschinski J, Horn S. 1992. BEHAVIOR OF P0-210 IN MOLTEN PB-17LI. Journal of Nuclear Materials 191:288-91
23. Loewen E. 2005. Investigation of polonium removal systems for lead-bismuth cooled FBRs. Progress in Nuclear Energy 47:586-95
24. Leggett RW. 1993. An age-specific kinetic-model of lead metabolism in humans.
Environ Health Perspect 101:598-616
25. Litvinenko. The Litvinenko Inquiry. https://www.litvinenkoinquiry.org/evidence
26. INQ016745. The Litvinenko Inquiry. https://www.litvinenkoinquiry.org/evidence
27. INQ007633. The Lirtvinenko Inquiry. https://www.litvinenkoinquiry.org/evidence
28. Harrison J, Leggett R, Lloyd D, Phipps A, Scott B. 2007. Polonium-210 as a poison. Journal of Radiological Protection 27:17-40
29. INQ014961. The Litvinenko Inquiry. https://www.litvinenkoinquiry.org/evidence
30. Casarett GW, Metcalf RG, Boyd GA. 1950. Pathology studies on rats injected with polonium, plutonium and radium. NatlNucl. Energy Ser. Div. VI-3 343-88
31. Baechler S, Hobbs RF, Jacene HA, Bochud FO, Wahl RL, Sgouros G. 2010. Predicting Hematologic Toxicity in Patients Undergoing Radioimmunotherapy with Y-90-Ibritumomab Tiuxetan or I-131-Tositumomab. Journal of Nuclear Medicine 51:1878-84
32. Bruland OS, Nilsson S, Fisher DR, Larsen RH. 2006. High-linear energy transfer irradiation targeted to skeletal metastases by the alpha-emitter Ra-223: Adjuvant or alternative to conventional modalities? Clin. Cancer Res. 12:6250S-7S
33. Hobbs RF, Song H, Watchman CJ, Bolch WE, Aksnes AK, et al. 2012. A bone marrow toxicity model for Ra-223 alpha-emitter radiopharmaceutical therapy. Physics in Medicine and Biology 57:3207-22
34. Watchman CJ, Bourke VA, Lyon JR, Knowlton AE, Butler SL, et al. 2007. Spatial distribution of blood vessels and CD34(+) hematopoietic stem and progenitor cells within the marrow cavities of human cancellous bone. Journal of Nuclear Medicine 48:645-54
35. Casarett GW. 1964. HEMATOLOGY OF SINGLE INTRAVENOUS DOSES OF POLONIUM. Radiation Research S:322-&
36. Casarett GW. 1964. PATHOLOGY OF ORALLY ADMINISTERED POLONIUM. Radiation Research S:361-&
37. V.V. Uyba KVK, L.A. Ilyin, Yu.E. Kvacheva, Yu.V. Abramov,, I.A. Galstyan AKG, B.A. Kukhta, N.M. Nadezhina,, V.A. Stebelkov AGT, S.M. Shinkarev, V.N. Iatsenko. 2015. Polonium-210 Version of Arafat's Death: the Results of Russian Investigation. Journal of Medical Radiology and Radiation Safety 60:50-7