Deposition and Retention of Plutonium in the United States General Population (Plutonium Injection Experiment)

Page 1

DEPOSITION AND RETENTION OF PLUTONIUM IN THE UNITED STATES GENERAL POPULATION Since 1959, a Los Alamos Scientific Laboratory (LASL) study has analyzed over 5000 tissues from 1100 individuals of the nonoccupationally exposed general population for fallout plutonium. These data have been useful in determining the tissue distribution and the annual baseline levels of environmental plutonium in the U.S. population. The effects of age, sex, date of death, cause of death, and geographic location of residence on the observed plutonium deposition have been evaluated. Because of the different biological turnover times of plutonium in the various organs of the body and the changing concentrations of plutonium in the atmosphere, the plutonium concentration ratios between tissues have changed as a function of time. However, our data indicate that over the past ten years, the highest concentrations in the general population are found in the tracheobronchial lymph nodes and the liver, and the lowest concentrations are in the spleen, gonads, and kidney. The median body burdens of plutonium in the U.S. population are estimated to have reached 12 pci during the 1 9 6 0 ' s and have declined to about 2 pCi in 1977. Large errors in estimated skeletal burdens of plutonium may exist because of small specimen sample sizes and a lack of knowledge concerning the relative distribution of plutonium among the various bones of the human body. The Los Alamos Scientific Laboratory (LASL) has collected human autopsy specimens over the past twenty years (1959-1979) to determine their plutonium content. Tissues and organs from the U.S. general population, exposed primarily to fallout plutonium resulting from atmospheric testing of nuclear weapons, have been radiochemically analyzed. The tissues collected at the beginning of the program were the lung, tracheobronchial lymph nodes, liver, kidney, and bone (usually a rib or vertebral wedge). Whole organs were requested but, typically, one lung and half of the liver were all that were received. Beginning in 1973, we added the gonads, spleen, and thyroid to the list of requested organs. The samples are obtained through the cooperative efforts of pathologists located throughout the united States. At present, we have collected specimens from 26 states, with

Page 2

most of the samples coming from Colorado, Georgia, Illinois, New Mexico, New York, Pennsylvania and Washington. The major thrust of this study is to measure the actual human exposure (depositions) to plutonium resulting from environmental sources before a major expansion of the nuclear industry occurs. This information has been useful in establishing the current annual baseline levels of plutonium in the tissues of the general population because of worldwide or regional environmental fallout from radioactive contamination and will aid in the identification of the uptake and distribution patterns of plutonium in the human body. The following is a report to this workshop of work still in progress and is not intended to represent our final analysis and interpretation of the data we are continuing to collect. The tissue specimens requested are identified and weighed by the pathologist at the time of autopsy. Each sample is individually packaged in prelabeled plastic bags and frozen. The sets of tissues from each autopsy case are packed in dry ice and sent to our laboratory by air freight. Upon receipt, the specimens are thawed, reweighed, grossly dissected to remove extraneous tissue if present, and analyzed We request, for each autopsy case, information concerning the cause of death, date of death, age, sex, race, weight, and the residential and occupational histories. The residential and occupational data are not always readily available to the pathologist; therefore the data we receive are not as complete as we would like it to be. The detailed analytical methods used for the analyses of plutonium in human autopsy specimens are described elsewhere in this publication (Bo81). In summary, a known amount of tracer solution containing 242Pu is added to each tissue. The specimens are dry ashed in a furnace for 4 days with a gradual increase of the temperature from 1 5 0 " to 500' C. The ash is alternately wet and dry ashed until all visible carbonaceous material is destroyed. A l l samples ace subsequently treated by adding a mixture of LiN03-NaN03 to the residue and continuing the wet and dry ashing until the sample will dissolve completely in 7.8M H N 0 3 . A n aliquot of the HN03 tissue solution is adjusted to the appropriate pH; the plutonium is isolated by anion exchange and electrodeposited onto The activity is assayed by alphastainless steel planchets. spectrometry, counting the alpha activity on the planchet with a 300-mm silicon surface-barrier detector for 50,000 sec and correcting the measured 239,240Pu-activity by the average background o € the detector and the fraction of the tracer solution recovered. In our spectrographic anal ses, we are unable to distinguish between the 5 . 1 5 MeV alpha of the f39Pu and the 5 . 1 6 MeV alpha of 240Pu; therefore, all

Page 3

Pu in the U.S. General Population 163 results we report as 239Pu are actually the sum of any 239Pu and 2 4 0 ~ u The count data from each detector are stored in the memory of a multichannel analyzer, transEerred to a magnetic tape, read into the computer for evaluation by a computer code that propagates the errors in each measurement. The final result and the associated error of each measurement, along with the appropriate information on the case number, tissue type, fraction analyzed, and the fraction of the tracer recovered are entered into a general purpose data base management system which operates on a C!X 6600 computer. This enables the rapid retrieval of all data in a form determined by the investigator for various RESULTS AND DISCUSSION The results to date of our analyses for 239Pu in the tissues of the U.S. general population have recently been published in toto (Mc79). Tissue concentrations from 1100 persons in seven geographic regions throughout the United States were reported. Preliminary attempts to summarize the data in the above report involved the estimation of median tissue concentrations from log-normal plots of individual tissue data collected since the beginning of the study. We now realize that the plutonium tissue concentrations change with time (see discussion under Time Trends) and that annual median tissue concentrations are necessary for making regional comparisons. Tables 1-4 show the regional median tissue concentrations of 239Pu in lung, tracheobronchial lymph nodes, liver and vertebrae (vertebral bodies) €or the years 1970-77. The annual median concentrations of all available data for each tissue type are indicated as "All Data". For comparison, the annual concentrations in each tissue predicted by Bennett (Be76), using the measured and inferred concentrations of fallout plutonium in the New York City area and the ICRP Task Group Lung Model to estimate plutonium deposition in the lung and transfer to other body organs, are also shown. The agreement in magnitude between the predicted tissue concentrations and the observed values is reasonable for lung and liver data, with 75% to 95% of the values within a factor of two of the predicted value. The lymph node and vertebrae, however, vary significantly from the model. The observed lymph node concentrations are all an order of magnitude or more lower in concentration than the predicted values. This suggests that the model parameters for transfer of plutonium into and from the pulmonary lymphatic system may be grossly in error or that the analytical process is not measuring the plutonium in this tissue accurately because of the small mass of the sample. Measurements of tracheobronchial lymph node concentrations of fallout plutonium in New York City autopsy cases have been made recently by other investigators (Fi79). Lymph node specimens were pooled to increase the mass being analyzed. Their results confirm a much lower concentration than that predicted by the ICRP model.

Page 8

The bone concentrations appear to be in close agreement with Bennett's skeletal values. However, our concentrations are based on wet weights of the bone and any marrow present, while the reported concentrations are calculated on the basis of a 5-kg mineral bone skeleton. In extrapolating to whole skeleton burdens, our data are higher by a factor of two and also suggest a shorter residence time than that used by ICRP. We are currently fitting the model to our data with the expectation that better estimates of the model parameters will result in a compartmental model able to predict more accurately the distribution and retention of inhaled fallout plutonium in the human Fox et al. have examined the data we have collected to identify possible regional differences and time and age trends in plutonium tissue concentrations (Fo79). They have also tested each of our autopsy cases to determine possible bias in the data and to identify outliers in our analyses. They have reported the following conclusions. Autopsies do not, in general, represent a random sampling of -all deaths. Tramatic deaths, unattended deaths, and deaths from unknown or medically interesting causes are more likely to result in an autopsy. To verify that the cause of death in our autopsy population was not biasing the amount of plutonium found in the tissues, the medical information provided by the pathologist and the results of our tissue analyses for plutonium were tested using a chi-square test of independence to measure the association between the cause of death and plutonium concentration in tissues. The lung and liver were selected as the tissues containing the largest amounts of plutonium and, because of their mass, the tissues in which we have the most confidence in the measured values. Chi-square values of 32.5 and 36.4 with 28 degrees of freedom indicate no detectable association and, thus, no cause for concern about bias in our data because of autopsy selection. In every large set of data, one finds outliers (observations that do not appear to be consistent with the bulk of the data). These may result from possible errors in analysis and transcription of information, from contamination of the sample, small sample sizes, etc. The concentrations of plutonium for the general population are near background and even a slight contamination may have a significant effect on the values measured. Certain a-emitting nuclides may interfere with the accurate determination of 239Pu. These may be present in the chemical reagents used, including the tracer added to each sample, on the stainless steel planchets, or on the electrodeposition apparatus from an earlier processing of a sample with high activity. The amount of contaminant analyzed with the autopsy specimen may have been equal to the activity in the sample, thus causing the current analysis to give erroneously high results. Other contributors to outliers are analysis of small aliquots, small tissue

Page 9

Pu in the U.S. General Population 169 samples, and improperly prepared samples with undissolved solids that interfere with the measurement process. Finally, there is the possibility that despite all efforts to identify the history of all samples, some occupationally exposed cases may have been included in the general population data. Using the Grubbs' statistic as a test for single outliers and the Tietjen-Moore test (Ti72) as a test for multiple outliers, 139 high results out of 4373 observations, or 3.2% of the data, were declared outliers. These specimens are being reanalyzed where samples are still available; if no sample remains, the outliers are removed from the data base or are identified as being a questionable result. The age at death of the persons included in our study is an uncontrolled variable and may not be typical of the population at large. The age distributions for all cases collected since the beginning of the program are shown in Figure 1 . Although the distributions resemble each other generally, the New York City data are a clear exception: the individuals from that population are much younger than those from other areas. This may be because the New York City samples were largely from traumatic deaths, which are more frequent in younger males. If age has an effect on the deposition and retention of plutonium, it is important to identify such trends and to adjust for them before making geographical comparisons. Excluding the New York City data, separate regression analyses indicated no dependence between the age at Figure 1 Age distributions of geographical groups. The dots represent the loth, 50th, and 90th percentiles. The end points of the rectangles are at the 25th and 75th percentiles so that they include the middle 50% of the data.

Page 10

death and geographical location of residence. In order to test for agerelated trends on the deposition of plutonium in tissues with the autopsy tissue data presently available, four very short segments of time (1968-69, 1970-71, 1972-73, 1974-75) were selected as periods when the trends would be neafly constant. For the liver and lung tissue data (over all ages and locales), the plutonium concentration versus age at death were fitted to a linear relationship by least squares for each of the four short time periods. Another line was fitted to the data (for each tissue separately) over the whole time period (1968-75). Tests of whether the slope of the line was significantly different from zero were made. For the liver, the slopes for each time period are consistently different from zero, but a single line for the whole time period fits as well as separate lines for each time period. This indicates that the slopes for different time periods are very similar. It was concluded is the liver concentration of 239Pu in dpm per kg tissue, best represents the effect of age on liver concentration. Over an 80-year lifetime, an increase of about 1 pCi could be expected in the liver. From age 4 0 to age 80, the increase would be about 0.6 pCi in the liver due to age alone. ' For lung tissue, the evidence of age-related trends on the deposition of plutonium was not convincing. For the kidney, tracheobronchial lymph nodes, rib, and male gonadal tissue, there was no detectable effect of age for any of the time periods. For vertebrae, the slope of the regression line relating concentration versus age at death was significantly different from zero for the 1974-75 data and the 1968-75 data. Significantly, perhaps, the slopes for this tissue are negative (or near zero), supporting the hypothesis that the skeleton is being remodeled with age and the plutonium is being transferred from the skeleton to the liver. There are roughly twice as many males in the study as females. To test the hypothesis of sex differences, the Colorado 1970-77 data, adjusted for age trends, was examined using the Mann-Whitney test (see Table 5 ) . The results show that there are no significant differences in the concentration of plutonium in the tissues shown due to sex. Gonadal tissues were not compared because of the obvious anatomical differences between the testes and ovaries. To compare the levels of plutonium concentrations in the populations of various geographic areas within the United States, we attempted to eliminate the dependence of age at death and year of death by considering very short …