Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
5 Toxic Effects of Uranium on Other Organ Systems Chapters 3 and 4 evaluated the effects of uranium on the kidneys and the lungs, respectively. This chapter examines effects on other organ systems. CENTRAL NERVOUS SYSTEM EFFECTS Human Studies Animal studies have demonstrated uranium uptake by and possible func- tional effects on the brain, but human studies have not confirmed or substanti- ated a relationship between uranium exposure and neurologic disease (IOM 2000; Craft et al. 2004). McDiarmid et al. (2000) found a statistically significant relationship between increased concentration of depleted uranium (DU) and decreased performance on automated tests that assess performance accuracy in Gulf War veterans who have embedded fragments. The authors cautioned that the number of people with increased urinary uranium was small, and a few vet- erans who had complex histories may have contributed appreciably to the ob- served variance. Neurocognitive function was assessed with a battery of tests consisting of traditional (paper and pencil) and automated measures. The traditional measures involved generating a neuropsychologic index from several tests: the California Verbal Learning Test; the Trail Making Test, Parts A and B; the Shipley Insti- tute of Living Scale; and the Digit Span, Arithmetic, and Digit Symbol subtests of the Wechsler Adult Intelligence Test-Revised (Wechsler 1981). Three scores were developed from automated measures of neurocognitive function from the Automated Neuropsychological Assessment Metrics test library (Kane and Reeves 1997): A-IIac (accuracy), A-IIrt (speed), and A-IItp (computed score combining accuracy and speed). Psychiatric assessment included the Wide Range Achievement Test 3 (Jastak and Wilkinson 1993), the Symptom Check- 48
Toxic Effects of Uranium on Other Organ Systems 49 list-90 Revised (Derogatis 1983), the Beck Depression Inventory (Beck et al. 1996), the Beck Anxiety Inventory (Beck et al. 1988), and the Mississippi Scale for Posttraumatic Stress Disorder (Keane et al. 1988). There were no statistically significant (p <0.05) differences between the high and low uranium-exposure groups in any neurocognitive measure at any of the five surveillance visits, although a higher impairment score on the A-IIac was consistently observed in the high-exposure group during the 1997 visits. For that time, results on the automated tests demonstrated a statistically significant relationship between urinary uranium concentration and lowered performance efficiency. More robust regression analyses that controlled for confounding fac- tors, such as emotional status and general intellectual level, were conducted on A-IIac impairment scores from the 2001 and 2003 visits and urinary uranium concentrations. The results revealed a marginal association between measured urinary uranium and the accuracy index; however, the authors commented that the relationship in both years was driven by two cases with persistent complica- tions due to combat injuries and their high urinary uranium concentrations (Hooper et al. 1999; McDiarmid et al. 2000, 2001a, 2004b, 2006). Animal Studies Description of morphologic changes in the central nervous system caused by exposure to uranium is limited to early experiments at high exposures. How- ever, more recent studies with lower exposures have identified functional elec- trophysiologic effects and perhaps neuropsychologic (behavioral) effects. Stud- ies performed many years ago showed that exposure of dogs to near-lethal doses of uranyl nitrate produced changes in the epithelium of the choroid plexus (Pur- jesz et al. 1930). In a more recent study that used an in situ brain-perfusion tech- nique, Lemercier et al. (2003) showed that uranium reached the brain paren- chyma by blood circulation (microcirculation) and the vascular space. In a 30-d inhalation study, dogs exhibited muscular weakness and gait in- stability on day 13 of exposure to uranyl fluoride gas at a uranium concentration of 1.8 mg/m3 but showed no effects at lower concentrations (Dygert et al. 1949). In a study to determine the LD50, groups of 10 male Sprague-Dawley rats and 10 male Swiss mice were given a single subcutaneous dose of uranyl acetate at 1.25, 2.5, 5, 10, 20, and 50 mg/kg. The LD50 in rats was 8.3 mg/kg, and that in mice was 20.4 mg/kg. Rats and mice that survived 6 d or longer showed central cholinergic neurologic signs (piloerection, tremors, hypothermia, papillary size decrease, and exophthalmos) that persisted until termination of the study at 14 d (Domingo et al. 1987). The relevance of those studies to potential human expo- sure has been questioned (IOM 2000). More recently, Pellmar et al. (1999a) surgically implanted DU pellets in the gastrocnemius muscle of rats at three doses (low dose, four DU and 16 tanta- lum pellets; medium dose, 10 DU and 10 tantalum pellets; and high dose, 20 DU pellets). After 1 mo, the uranium concentrations in the brain were statistically
50 Risks to Military Personnel from Exposure to Depleted Uranium significantly higher in the high-dose rats than in controls; after 18 mo, there were dose-related increases in uranium concentrations in several areas of the brain in all three groups of DU-implant rats. The highest uranium concentrations were in the motor and frontal cortex, midbrain, and cerebellar vermis. The study demonstrated that uranium absorbed from embedded DU pellets may accumu- late in the central nervous system. In a followup study, Pellmar et al. (1999b) assessed the electrophysiologic changes in the hippocampus of rats that had implanted DU fragments. At 12 mo, the amplitude of synaptic potentials was significantly greater in tissues derived from the high-dose (20 fragments) DU-implant rats than in controls. In the same animals, uranium did not affect locomotor activity, discrimination learning, or a battery of general neurologic functional measures. The abnormal electrophysi- ologic measurements were not apparent 18 mo after exposure to 20 DU pellets. The authors suggest that by 18 mo the effects of aging and DU exposure con- verge, and the convergence obscures the effects of uranium. Thus, uranium from embedded DU pellets clearly can accumulate in the brain, but the significance of the effect on the function of the hippocampus is difficult to interpret. Because of the inherent stress of combat and the potential for stress to al- ter blood-brain barrier permeability, Barber et al. (2005) investigated the impact of forced-swim stress on the temporal and regional distributions of brain ura- nium after a single peritoneal injection of uranyl acetate at a uranium concentra- tion of 1 mg/kg in Sprague-Dawley rats. Uranium concentrations in serum, hip- pocampus, striatum, cerebellum, and frontal cortex were measured with inductively coupled plasma-mass spectrometry 6 h, 24 h, 7 d, and 30 d after ex- posure. Uranium entered the brain rapidly and was initially concentrated in the hippocampus and striatum but was distributed among various regions. Prior ex- posure to stress significantly reduced hippocampal and cerebellar uranium 24 h after exposure and tended to reduce uranium in all regions of the brain 7 d after exposure. That effect of stress is not peculiar to uranium: brain zinc concentra- tions also are reduced by stress (Izgut-Uysal et al. 2000). Although stress ap- pears to reduce the brain burden of uranium, the combined effects of stress and uranium on brain function remain unclear. The stressed hippocampus has been shown to be hypersensitive to anticholinesterases (Meshorer et al. 2002). Spontaneous locomotion activity increased and spatial working memory decreased after repeated inhalation exposure of rats to DU at 197 mg/m3 30 min/d, 4 d/wk for 3 wk (Monleau et al. 2005). When rats ingested uranyl nitrate in drinking water (at a uranium concentration of 40 mg/L) for 1.5, 6, and 9 mo, acetylcholinesterase activity in the cerebellum was transiently impaired after 6 mo, and monoamine concentrations were low after 9 mo. Those effects might be caused by uranium accumulation in the brain or changes in oxidation balance, which could account for changes in neurobehavior (Bussy et al. 2006). In an- other study, decrease in food intake and shorter paradoxical sleep were observed 3 d after intraperitoneal injection of DU at 144 Â± 16 Âµg/kg but not after a lower dose, 70 Â± 8 Âµg/kg (Lestaevel et al. 2005). The studies indicate that exposure to uranyl nitrate and DU results in accumulation of uranium in the brain and possi-
Toxic Effects of Uranium on Other Organ Systems 51 ble neurochemical and neurobehavioral effects in rats. A study in the same labo- ratory found a correlation between uranium accumulation in the brain and an increase in paradoxical sleep and a reduction in spatial working memory capaci- ties after exposure to 4% enriched uranium for 1.5 mo. The authors suggest that radiologic activity induces the effects of uranium (Houpert et al. 2005). Arfsten et al. (2007) surgically implanted DU pellets in young adult Sprague-Dawley rats and after 150 d did not find any evidence of effects on behavior and toxic end points. REPRODUCTIVE AND DEVELOPMENTAL EFFECTS Human Studies A few studies (as reviewed by Craft et al. 2004) have indicated that ura- nium exerts effects on the reproductive system and consequently induces changes in the developing organism. Muller et al. (1967) reported that uranium exposure affects the sex ratio; the unusually high frequency of female births in uranium-mining workers indicates altered sperm function or reproductive- capacity dysfunction. Zaire et al. (1997) noted a significant reduction in testos- terone concentrations and altered gonadal function in uranium miners; however, study limitations included inadequate characterization of exposure, a small sam- ple, and the presence of other confounding factors. McDiarmid et al. (2004b) conducted the most comprehensive followup study of a group of Gulf War veterans exposed to DU. Of a cohort of 74 veter- ans who were exposed to DU during friendly-fire incidents in February 1991, 39 were examined in Baltimore in April-July 2001. The items measured included serum follicle-stimulating hormone (FSH), luteinizing hormone (LH), prolactin, thyroid-stimulating hormone (TSH), free thyroxine, and testosterone concentra- tions. In addition, semen characteristicsâincluding volume, sperm concentra- tion, total sperm count, and functional measures of sperm motilityâwere deter- mined. The study included veterans drawn from the cohort of DU-exposed veterans evaluated by McDiarmid et al. (2000, 2001b) in two previous investiga- tions. The three studies demonstrate that some observations are contradictory even though they used subsets of the same population; this raises issues regard- ing the biologic relevance of the findings. For example, prolactin concentrations were significantly higher in the low-uranium group (n = 26; defined as less than 0.1 Âµg/g creatinine) in the 2004 study. However, in the 2000 study, the high- uranium group (n = 13; defined as over 0.1 Âµg/g creatinine) had higher prolactin concentrations; and in the 2001 study, there was no marked difference between the two groups. The relevance of the findings with respect to the use of prolactin as a biomarker of DU actions on the reproductive system is questionable in light of the variation in observations in the three studies. In all three studies, McDiarmid et al. found no marked changes in the concentrations of serum FSH, LH, testosterone, TSH, or free thyroxine. Another followup study (McDiarmid
52 Risks to Military Personnel from Exposure to Depleted Uranium et al. 2007) of 34 veterans found similar nonsignificant changes. Thus, the data suggest that DU exposure did not markedly affect neuroendocrine functions in humans. The semen characteristics of 35 DU-exposed Gulf War veterans were also examined by McDiarmid et al. (2004b, 2006). Eight veterans were excluded for such reasons as vasectomy and other medical conditions. No significant differ- ences were observed between the low-exposure group (n = 16) and the high- exposure group (n = 11) in semen volume, sperm concentration, total sperm count, or percentage motile sperm. No significant effects were observed in ura- nium-exposed veterans in the following categories defined by the World Health Organization (WHO 1987): percentage progressive sperm, total progressive sperm, percentage rapid progressive sperm, and total rapid progressive sperm. Although the sample was small, the values obtained were all within the WHO normal guideline range. Similarly, McDiarmid et al. (2007) found no significant alterations in semen quality in a followup study of 24 veterans. Thus, DU con- centrations did not appear to affect reproductive capacity in the men studied. Information gathered through the continuing surveillance program of the male DU-exposed Gulf War veterans at the Baltimore Veterans Administration Medical Center contains no evidence of congenital anomalies or abnormal re- productive capacity. During the 8-y span from exposure in 1991 to the 1999 surveillance visit, 50 of the DU-exposed veterans fathered 35 children, all with- out birth defects (McDiarmid et al. 2001b). Animal Studies The reproductive system and fetal development are targets of uranium in animals (Domingo 2001). Paternain et al. (1989) administered uranyl acetate orally to mice at 5, 10, or 25 mg/kg per day. Male and female mice received uranyl acetate for 60 and 14 d, respectively, before mating. Females were ex- posed throughout mating, gestation, parturition, and lactation. At 25 mg/kg per day, uranyl acetate was lethal to some embryos and reduced growth rate in sur- viving offspring. The authors concluded that uranyl acetate did not markedly alter fertility, reproductive measures, or offspring survival at the lower doses. Similarly, Domingo et al. (1989) found that oral administration of uranyl acetate at 0.05, 0.5, 5, or 50 mg/kg per day from day 13 of pregnancy until weaning on postnatal day (PND) 21 significantly decreased mean litter size in Swiss mice only at 50 mg/kg per day. In addition, viability was reduced, and the lactation index was diminished at the highest dose. However, no marked changes were observed in sex ratio, mean litter size, pup body weight, or pup length in the other groups. Domingo et al. (1989) estimated a safety factor of 1,000 between the no-observed-effect level (NOEL) in the rodent study and typical human exposure concentrations. The influence of uranyl acetate on the male reproductive capacity of Swiss mice was examined by Llobet et al. (1991). Groups of mice were given uranyl
Toxic Effects of Uranium on Other Organ Systems 53 acetate in drinking water for 64 d at daily doses of 0, 10, 20, 40, or 80 mg/kg. To evaluate fertility, a male was mated with untreated females for 4 d. Uranium treatment was associated with a significant but non-dose-related decrease in pregnancy rate: groups of 16 mice had 13, four, five, four, and six pregnancies after treatment at 0, 10, 20, 40, and 80 mg/kg per day, respectively. However, there were no marked effects on the number of implantations, the number of resorptions, the number of dead fetuses, or the number of live fetuses. At 80 mg/kg, there was a significant decrease in body weight but no change in testicu- lar weight. Testicular function and spermatogenesis were not affected by ura- nium. At 80 mg/kg, there was vacuolation of Leydig cells. Given uranium in drinking water at 100 Âµg/L, a 70-kg human consuming 2 L of water a day would receive a daily dose of 0.003 mg/kg, which is less than one-thousandth of the lowest dose at which apparent reproductive effects were observed in this study. Subcutaneous administration of uranyl acetate at 1 and 2 mg/kg per day significantly decreased fetal body weight, increased the number of dead fetuses at birth, and increased the number of resorptions. Concerning teratogenic in- dexes, uranium was found to induce cleft palate, dorsal and facial hematomas, and skeletal malformations (Bosque et al. 1992, 1993). There were no apparent effects at 0.5 mg/kg per day. It should be noted that the route of exposure is critical. In the studies de- scribed above, uranium was administered directly orally or subcutaneously, and some effects were noted at high doses. The studies described below differed with respect to exposure in that the route was muscle implantation of uranium and later transport to the target site. Several studies were conducted to determine the effects of DU embedded in soft tissue and whether it might affect reproduction and development in ro- dents. The study was undertaken to simulate the human condition in which DU fragments from friendly fire would become embedded in a person. Arfsten et al. (2005) estimated human equivalent exposures from the pellets implanted in the rat. The surface area of four pellets, each 2 Ã 1 mm in diameter, is about 31 mm2 and corresponds to 0.1% of the estimated body surface area of an adult rat (0.025 mm2). Twelve 2 Ã 1-mm DU pellets are the equivalent of one 30-mm APFSDSâT DU solid projectile (about 425 mg of DU, 28 cm long). Twenty 2 Ã 1-mm, DU pellets in a 250-g rat is the equivalent of about 0.22 kg of DU in a 70-kg man. As discussed above, Pellmar et al. (1999a) implanted four, 10, or 20 DU pellets into gastrocnemius muscle of rats for 18 mo. No uranium was present in the testes of the control or four-pellet group after 18 mo, but testicular uranium was found in the 10- and 20-pellet groups, indicating transport and translocation of uranium from the embedded site to testes. In female Sprague-Dawley rats, implantation of 32 DU pellets in the gastrocnemius muscle for 82 d resulted in uranium in the ovaries (Benson 1998). Furthermore, Benson and McBride (1997) implanted four, eight, or 12 DU pellets in female Sprague-Dawley rats and then bred them within days and euthanized them on gestational day 20. The
54 Risks to Military Personnel from Exposure to Depleted Uranium presence of uranium in placenta, whole fetus, fetal liver, and maternal kidney indicated placental transfer and fetal accumulation. Arfsten et al. (2005) implanted 12 DU pellets (six pellets per gastrocne- mius muscle) in male Sprague-Dawley rats and four, eight, or 12 DU pellets in females. Thirty days after surgery, the males and females were cross-mated, and parental females were monitored for weight gain on gestational days 5, 10, 15, and 20. On PND 4, the litters were culled to eight pups, and females nursed until PND 20. Pups were studied until PND 90. DU did not affect reproductive capac- ity, survival, or weight gain in parental males and females throughout the study period. There was no significant effect on birth weight, litter size, number of litters, percentage surviving pups on PND 4 and 20, or pup weight gain. In pups, there was no adverse effect on sperm motility or concentration and no signifi- cant difference in percentage motile sperm, curvilinear velocity, straight-line velocity, mean sperm path velocity, amplitude of lateral sperm-head displace- ment, or sperm-head beat-cross frequency up to PND 90. Surprisingly, uranium content was not detected in whole-body measurements of pups on PND 4 and 20. In a later study, Arfsten et al. (2006) implanted 20 uranium pellets (each 1 Ã 2 mm; equivalent to 0.22 kg of DU in a 70-kg man) into Sprague-Dawley rats, which were then assessed for male reproductive performance on postimplanta- tion day 150. The concentration, motion, and velocity of sperm were not mark- edly affected. Furthermore, there was no evidence of adverse effects on mating or reproductive success 3-45 and 120-145 d after implantation. The findings of Arfsten et al. (2005, 2006) clearly are at odds with those Benson and McBride (1997), who noted placental uranium transfer. Arfsten et al. (2005) attributed the differences to variations in methods of measuring ura- nium. That implanted DU pellets had no effects on breeding success, fetal de- velopmental defects, and sperm quality in the Arfsten et al. studies (in contrast with the finding of uranium administered orally or subcutaneously) suggests that the presence of DU fragments at tissue sites other than the endocrine- reproductive system does not have a marked effect on reproductive capacity or neonatal development in rodents. The route of exposure appears to be a critical factor in the consequences on reproduction and development in rodents. It should be noted that the amount of uranium that produced an adverse reproduc- tive effect in the Domingo et al. (1989) study was markedly higher than that used in the Arfsten et al. studies. Mitchell et al. (2005) exposed embryos of the African clawed frog (Xenopus laevis), a sentinel species for environmental exposure, to DU at 5-78 mg/L in water for 96 h. That there were no marked effects on embryonic mortal- ity, malformation, or growth indicates that DU did not induce teratogenic altera- tions. At high concentrations of DU (well above those in municipal drinking water), there was a delay in metamorphosis. With respect to humans, exposure from DU fragments is a more plausible scenario than exposure by ingestion; data indicate that uranium would not exert a significant effect on the reproduc- tive system or fetal development.
Toxic Effects of Uranium on Other Organ Systems 55 HEMATOLOGIC EFFECTS Human Studies Uranium miners who worked less than 5-20 y showed small but signifi- cant decreases in hemoglobin and mean corpuscular hemoglobin and significant increases in red blood cell (RBC) and mean corpuscular volume (Vich and Krik- lava 1970). All values were within the normal ranges according to the authors. The exposure concentrations were not reported. Studies of workers exposed to uranium have shown no excess mortality from leukemia, but there is uncertainty about lymphoma induction because a few studies have shown excess lymphoma mortality (see Chapter 6 and Appendix B). Animal Studies Zhao and Zhao (1990) reported no hematologic effects in a single animal exposed by inhalation to powdered uranium tetrafluoride for 5 min (concentra- tion not reported). Hematologic effects have been observed in the rat, dog, and rabbit after subacute inhalation exposure (ATSDR 1999); however, ATSDR characterized the effects as minor because of statistical insignificance or incon- sistency with other hematologic measures. Specifically, Dygert et al. (1949) exposed rats and rabbits to diuranate and ammonium diuranate, respectively, 6 h/dy for 30 d. The lowest observed-adverse-effect level (LOAEL) of uranium was 6.8 mg/m3 for diuranate for decreased RBCs and hemoglobin in the rat. The LOAEL of uranium was 6.8 mg/m3 for ammonium diuranate for increased neu- trophils and decreased lymphocytes in the rabbit. Similarly, Roberts (1949) ex- posed rats, dogs, and rabbits to uranyl nitrate hexahydrate daily (time of day not specified) for 30 d. The LOAEL of uranium was 9.5 mg/m3 for decreased RBCs and hemoglobin in the rat, 2.1 mg/m3 for slightly decreased fibrinogen in the dog, and 0.13 mg/m3 for increased plasma prothrombin and fibrinogen in the rabbit. Rothstein (1949) exposed rats to uranium trioxide 6 h/d for 28 d; this resulted in a LOAEL of uranium of 16 mg/m3 for increased myeloblasts and lymphoid cells of the bone marrow. Maynard and Hodge (1949) and Maynard et al. (1953) exposed rats orally to uranyl nitrate hexadydrate daily for 2 y. The LOAEL was 16.6 mg/kg per day for mild (defined as low-grade) anemia and increased number of leucocytes. HEPATIC EFFECTS Human Studies Humans exposed once to uranyl acetate dihydrate demonstrated increased serum alanine amino transferase, aspartate amino transferase, and gamma glu- tamyl transpeptidase at 131 mg/kg per day (Pavlakis et al. 1996). Those en-
56 Risks to Military Personnel from Exposure to Depleted Uranium zymes can be associated with cellular necrosis of the liver. Studies of workers exposed to uranium have not shown any excess mortality from hepatic cirrhosis (see Table 5-1) or liver cancer (see Chapter 6). Animal Studies There are several studies of the hepatic effects of DU after chronic inhala- tion exposure of monkeys, rats, and dogs. Daily exposure of monkeys to ura- nium dioxide for 5 y resulted in a no-observed-adverse-effect level (NOAEL) of uranium of 5.1 mg/m3 (Leach et al. 1970). Exposure of rats to uranium tetrachlo- ride or uranium hexafluoride for 1 y resulted in a uranium NOAEL of 0.2 mg/m3 for each compound (Stokinger et al. 1953). Exposure of dogs to uranium tetra- chloride or uranyl nitrate hexahydrate for 1 y resulted in a uranium NOAEL of 0.2 or 2.0 mg/m3, respectively, for increased bromosulfalein retention (Stokinger et al. 1953). The retention of bromosulfalein in plasma is an indication of im- paired hepatic function. Under healthy conditions, it is rapidly removed from plasma by the liver. Focal necrosis of the liver was reported in rats exposed to uranium tetrafluoride at a uranium concentration of 0.4 mg/m3 for 6 h/d for 30 d (Dygert et al. 1949). IMMUNOLOGIC EFFECTS The immune system differs from other organ systems in that it is not con- fined to a single site in the body. Rather, it comprises numerous lymphoid or- gans and diverse cell populations. Toxicity in the immune system can be evi- denced by decreased immune function (immunosuppression as evidenced by tumor production or increased infection) or enhanced immune function (allergy or autoimmunity). Inhaled uranium oxide particles deposited in the lung are transported by alveolar macrophages to the draining lymph nodes, where they may remain for several years. Retained alpha particles may cause some lymphocyte death as lymphocytes pass through the lymphoid tissues, but the decrease is unlikely to affect lymphoid function (Royal Society 2002). No report has been identified that indicates immune system dysfunctionâeither immunosuppression or im- munoenhancementâin humans resulting from DU exposure. Furthermore, an Institute of Medicine report (IOM 2000) concluded that there is inadequate or insufficient evidence to determine an association between exposure to uranium and lymphocytic cancer or bone cancer. As discussed below, immunotoxic ef- fects of DU have been studied in laboratory animals by using inhalation and oral exposure and implantation DU pellets. In vitro studies also have been per- formed.
Toxic Effects of Uranium on Other Organ Systems 57 TABLE 5-1 Standardized Mortality Ratios (95% Confidence Intervals) and [Observed Number of Deaths] for Hepatic Cirrhosis in Uranium Workers Study Hepatic Cirrhosis Reference Colorado Plateau uranium-mill 0.52 (0.21-1.07)  Waxweiler et al. 1983 workers (with no history of uranium mining) Fernald fabrication of uranium 0.92 (0.64-1.29)  Ritz 1999 products Savannah River nuclear-fuels 0.51 (0.34-0.72)  Cragle et al. 1988 production Linde uranium-processing facility 0.93 (0.66-1.28)  Dupree et al. 1987; (1943-1949) Teta and Ott 1988 Springfields, UK: mortalitya 1.05 (0.69-1.53)  McGeoghegan and Binks 2000a Capenhurst, UK: 235U-enrichment 0.63 (0.16-1.72)  McGeoghegan and plant, mortalitya Binks 2000b Total Observed/Expected Cases 134/160 a Data only on those classified as radiation workers. Animal Studies No lymph node tumors were found in rats exposed for 65 d to uranium ore at 19 mg/m3 or 50 mg/m3 (Mitchell et al. 1999). Studies in dogs suggest that cancers do not develop in the lymph nodes after inhalation of uranium dioxide (uranium at 5 mg/m3; mass median diameter [MMD], 1 Âµm) for 5 y (Leach et al. 1973). The short-term effect of DU on mucosal immunity of the digestive tract was studied in rats. The rationale for the study was that the digestive tract lumen is the first biologic system exposed to the chemical in ingestion. Rats received a single dose of uranyl nitrate dissolved in water (204 mg/kg in 1.5 mL; pH 3; Dublineau et al. 2006). That dose is known to be toxic to kidneys (equivalent to 1-3 Âµg/g of wet kidney). The intestine was assessed for cell proliferation, differ- entiation, and apoptosis 1 d later and then again 3 d later. Results indicated that DU was not toxic to the intestine although changes were noted in the production of chemokines and in the expression of cytokines. Production of monocyte chemoattractant protein-1 was decreased, and expression of interferon-gamma was increased. No changes were noted in the localization or density of neutro- phils, Th1 lymphocytes, or cytotoxic T lymphocytes after DU administration. Long-term effects of DU were not studied. As discussed above, Arfsten et al. (2005) studied the effects of DU pellets surgically implanted in adult rats on reproductive success and development in two consecutive generations. Immune function in the offspring was assessed.
58 Risks to Military Personnel from Exposure to Depleted Uranium Results indicated no effect on T-lymphocyte or B-lymphocyte function, NK-cell function, delayed hypersensitivity, or organ weights. In Vitro Studies In vitro effects of uranium compounds on macrophage viability and func- tion have been reported. In one study, particulate uranium dioxide at increasing concentrations was added to rat alveolar macrophages that had been isolated after bronchoalveolar lavage (Tasat and deRey 1987). Cell viability and incorpo- ration of uranium particles were assessed; the alveolar macrophages were able to phagocytose uranium particles despite toxicity in cell membranes and eventual cell death. The in vitro effect of soluble DU-uranyl chloride on the J774 mouse macrophage cell line was studied with flow cytometric analysis (Kalinich et al. 2002); treatment with 1-100 ÂµM uranyl chloride resulted in apoptosis and de- creased viability within 24 h. The effect of DU on mouse macrophages and T lymphocytes was studied. Peritoneal macrophages and splenic CD4+ T cells were elicited in response to thioglycollate administration, and cells were assessed for viability, function, and cytokine gene expression after in vitro exposure of cells to increasing concentra- tions of uranyl nitrate (Wan et al. 2006). In both cell lines, cytotoxicity was dose-dependent. Apoptosis and necrosis of macrophages occurred within 24 h of exposure to 100 ÂµM DU, but 50 ÂµM was noncytotoxic. CD4+ T cells died when exposed to 500 ÂµM DU, whereas 100 uM was noncytotoxic. DU altered gene- expression patterns in both cell types; genes related to signal transduction (for example, c-jun and NF-Îº Bp65) were the most differentially expressed. Up- regulation of IL-10 and IL-5 was noted after in vitro exposure of cells to DU. MUSCULOSKELETAL EFFECTS Human Studies There is a clinical report of deliberate ingestion of 15 g of uranyl acetate (Pavlakis et al. 1996). The dose was estimated to be the equivalent of 131 mg/kg for a 70-kg man (ATSDR 1999). The patient suffered from increasing rhabdo- myolysis (characterized by increased serum creatine kinase). The condition was resolved at 6 mo after the ingestion. The etiology of the effect is unknown, but the presence of confounding factors in the suicide attempt makes interpretation difficult. Uranium accumulates in bone, affects bone metabolism, and, when in- gested in drinking water, increases urinary excretion of calcium and phosphate, important components of bone structure. To demonstrate those effects in hu- mans, Kurttio et al. (2005) studied people who drank well water with high con- centrations of natural uranium. On the basis of slightly increased concentrations of osteocalcin and serum type I collagen carboxy-terminal telopeptide in men,
Toxic Effects of Uranium on Other Organ Systems 59 but not women, the authors suggested that bone may be a target of uranium tox- icity. Medical followup of a cohort of Gulf War veterans who were exposed to DU during combat has been carried out since the early 1990s (McDiarmid et al. 2004b). Findings reveal a persistent increase in urinary uranium more than 10 y after exposure in veterans who had retained fragments; however, no evidence of musculoskeletal effects has been observed. Animal Studies Data from the available animal studies suggest that oral exposure to ura- nium does not cause detectable damage to the musculoskeletal system. Exami- nation of muscle after exposure to uranyl nitrate in drinking water showed no effects in Sprague-Dawley rats after 28 d (uranium at up to 40 mg/kg per day; Gilman et al. 1998a) or 91 d (uranium at up to 53 mg/kg per day; Gilman et al. 1998a) or in New Zealand rabbits after 91 d (uranium at up to 43 mg/kg per day; Gilman et al. 1998c). Two animal studies funded by the Department of Defense in response to concerns about the effects of DU fragments have been completed. The study of the distribution of uranium in rats with implanted DU pellets (Pellmar et al. 1999a) concluded that bone is one of the primary reservoirs of uranium redistri- bution from intramuscularly embedded fragments. In the other study, Hahn et al. (2002) evaluated the carcinogenic response to implanted tantalum metal, an in- jected colloidal suspension of radioactive thorium dioxide (Thorotrast), and im- planted DU in the muscle tissue of rats. Squares (2.5 Ã 2.5 Ã 1.5 mm or 5.0 Ã 5.0 Ã 1.5 mm) or pellets (2.0 Ã 1.0 mm in diameter) of DU were surgically im- planted in the thigh muscles of male Wistar rats. Tantalum was implanted as four squares (5.0 Ã 5.0 Ã 1.1 mm) per rat. Thorotrast was injected at two sites in the thigh muscles of each rat as a positive control. Control rats had only a surgi- cal implantation procedure. Each treatment group included 50 rats. After life- time observation, the incidence of soft-tissue sarcomas (malignant fibrous histiocytomas and fibrosarcomas) was increased significantly around the 5.0 Ã 5.0-mm squares of DU (in nine of 49). A slightly increased incidence occurred in rats with the 2.5 Ã 2.5-mm DU squares (in three of 50) and with 5.0 Ã 5.0-mm squares of tantalum (in two of 50). No tumors were seen in rats with DU pellets or in the controls. Uranium dioxide powder (0.125 g/kg of body weight) was implanted sub- cutaneously in rats to evaluate the effects of an internal source of an insoluble form of uranium on bone. After 30 d, rats exposed to uranium weighed less than controls. Bone-formation activity in endochondral ossification and bone growth were lower in the experimental animals (Diaz Sylvester et al. 2002). The toxic effect of uranium on bone modeling and remodeling in the periodontal cortical bone was studied in rats. Uranyl nitrate (2 or 0.8 mg/kg of body weight) was injected intraperitoneally, and the rats were killed 14, 30, and
60 Risks to Military Personnel from Exposure to Depleted Uranium 60 d after injection (Ubios et al. 1991). There was a decrease in bone formation and an increase in bone resorption at 14 d. The distribution and retention of intravenously injected 233U (uranium VI) in the skeleton of the female rat have been investigated (Priest et al. 1982). About one-third of the injected uranium was deposited in the skeleton, where it was retained with an initial biologic half-life of about 40 d. The study also showed that uranium is initially deposited onto all types of bone surface but preferentially onto types that are accreting; uranium is deposited in the calcify- ing zones of skeletal cartilage; bone accretion results in the burial of surface deposits of uranium; bone resorption causes the removal of uranium from sur- faces; uranium removed from bone surfaces enters the bloodstream, where most is either redeposited in bone or excreted via the kidneys; and the recycling of resorbed uranium in the skeleton tends to produce a uniform concentration of uranium contamination throughout mineralized bone. Ubios et al. (1994) reports that 30 daily cutaneous applications of 2% or 4% insoluble triuranium octaoxide (vehicle and amount not specified) to Wistar rats resulted in impairment of bone formation. Tibiae and mandibles of the rats were affected. Miller et al. (2003) examined the effects of DU on human osteoblasts in vitro. The exposure caused genomic instability manifested as delayed reproduc- tive death and micronuclei formation. CARDIOVASCULAR EFFECTS Few studies are available to evaluate the effects of uranium exposure on the cardiovascular system. Studies of workers exposed to uranium have not shown any excess mortality due to cardiovascular or cerebrovascular disease (see Table 5-2). No cardiac effects were observed in inhalation studies in several test spe- cies, including rats exposed to uranium (as uranium hexafluoride) at 0.2 mg/m3 for 1 y (Stokinger et al. 1953) and rats, mice, guinea pigs, and rabbits exposed to uranium (as triuranium octaoxide) at 4.8 mg/m3 for 26 d (Dygert et al. 1949). Filippova et al. (1978) instilled 235U-enriched soluble tetravalent and ura- nium hexahydrate salts into the tracheae of rats and found abnormalities of blood vessels and cardiac enlargement. Those findings are probably due to a radiation effect of uranium rather than a chemical effect. Gilman et al. (1998c) did not find any cardiovascular effect in rabbits exposed to uranyl nitrate in drinking water (at 0.96, 4.8, 24, 120, and 600 mg/L) for 91 d. OCULAR EFFECTS Chemical burns of the eyes have been reported in humans after accidental exposure to uranium hexafluoride (Kathren and Moore 1986). Conjunctivitis and
TABLE 5-2 Standardized Mortality Ratios with (95% Confidence Intervals) and [Observed Numbers of Deaths] for Circulatory, Heart, and Cerebrovascular Disease in Uranium Workers Cerebrovascular Study Circulatory Disease Heart Disease Disease Reference Colorado Plateau uranium-mill workers 0.81 (0.71-0.92) 0.84 (0.75-0.94) 0.79 (0.56-1.08) Waxweiler et al. 1983; Pinkerton (with no history of uranium mining) a  a et al. 2004 TEC/Y12 (1943-1947): Oak Ridge 0.85 (0.81-0.88) â â Polednak and Frome 1981 uranium conversion/enrichment, all [2,571] workers TEC/Y12 (1943-1947): Oak Ridge 0.83 (0.78-0.89) â â Polednak and Froome 1981 uranium conversion and enrichment, alpha  and beta chemistry departments Y12 (1947-1974): Oak Ridge uranium- 0.87 (0.81-0.93) â â Checkoway et al. 1988; Loomis metal production and recycling  b and Wolf 1996 Mallinckrodt uranium-processing workers 0.89 (0.81-0.97) â â Dupree-Ellis et al. 2000  Fernald fabrication of uranium products 0.78 (0.71-0.86) 0.80 (0.71-0.89) 0.81 (0.59-1.07) Ritz 1999    Portsmouth gaseous diffusionc 0.70 (0.60-0.81) â â Brown and Bloom 1987  Savannah River nuclear-fuels production 0.81 (0.74-0.88) 0.81 (0.73-0.88) 0.71 (0.52-0.94) Cragle et al. 1988    Linde uranium-processing facility 0.94 (0.86-1.02) â 0.71 (0.54-0.92) Dupree et al. 1987; Teta and (1943-1949)   Ott 1988 United Nuclear Corp. nuclear-fuels 0.82 (0.64-1.04) 0.82 (0.65-1.02) 0.68 (0.27-1.41) Hadjimichael et al. 1983 fabricationd    (Continued) 61
62 TABLE 5-2 Continued Cerebrovascular Study Circulatory Disease Heart Disease Disease Reference Florida phosphate workersd â 0.88 (0.83-0.93) 0.96 (0.85-1.08) Checkoway et al. 1996 [1,232]  Atomic Weapons Establishment, UK 0.91 (0.72-1.14) â â Beral et al. 1988  Springfields, UK: mortalitye 0.90 (0.86-0.94) 0.91 (0.86-0.96) 0.95 (0.85-1.06) McGeoghegan and Binks 2000a    Capenhurst, UK: 235U enrichment 0.86 (0.76-0.96) 0.92 (0.80-1.06) 0.66 (0.47-0.90) McGeoghegan and Binks 2000b plant, mortalitye    Total Observed/Expected Casesf 7,857/8,853 3,806/4,376 847/912 a From Waxweiler et al. 1983. b Ischemic heart disease. c Includes only âSubcohort I,â which consists of those who at some time worked in one of the departments considered to have uranium exposure. d SMRs were similar for white and nonwhite men, so results for the combined groups are presented. e Data only on those classified as radiation workers. f Sums do not include row labeled âTEC/Y12 (1943-47): Oak Ridge uranium conversion and enrichment, alpha and beta chemistry departments,â be- cause those workers were already included in the TEC/Y12 row for all workers.
Toxic Effects of Uranium on Other Organ Systems 63 eye irritation have also been reported in animals after exposure to uranium hexafluoride (Spiegel 1949) and uranium tetrachloride (Dygert et al. 1949). Uranium hexafluoride rapidly dissociates into hydrofluoric acid and uranyl fluo- ride on contact with moisture in the air (ATSDR 1999). Similarly, uranium tet- rachloride forms hydrochloric acid. Therefore, ocular effects were probably due to direct contact of caustic vapors or aerosols, rather than uranium, with the eye. No studies of ocular effects in humans or animals after exposure to other ura- nium compounds by other routes (oral, dermal, and implantation) were located. GASTROINTESTINAL EFFECTS No gastrointestinal effects were observed in various animals exposed orally to uranyl nitrate at 664 mg/kg per day for up to 2 y (Maynard and Hodge 1949) or in rats exposed to uranyl nitrate hexahydrate in drinking water at 40 mg/kg per day for 28 d, or in rabbits exposed to uranyl nitrate hexahydrate in drinking water at up to 600 mg/L for 91 d (Gilman et al. 1998a,b,c). Gastrointes- tinal tracts did not show abnormalities in dogs that ingested uranium dioxide or triuranium octaoxide. Mild hemorrhage was observed in animals that received the highest dosage of uranium tetrafluoride (20 mg/kg per day) (Maynard and Hodge 1949). DERMAL EFFECTS Uranium compounds have been found to be slightly irritating to the skin of laboratory animals, but there is no evidence of skin sensitization or skin cancer. Three of the more water-soluble uranium compounds (uranyl nitrate, uranium tetrachloride, and uranium pentachloride) caused mild to moderate transient irri- tation when applied to the skin of rabbits (Orcutt 1949). In rats, 30 daily topical applications of triuranium octaoxide caused epidermal atrophy and increased the permeability of the skin (Ubios et al. 1997). Local skin toxicity seems to be mainly limited to irritation, which may be caused by direct cutaneous contact with some of the compounds. ATSDR (1999) concluded that there are no studies relating skin cancers to uranium compounds. As discussed in Chapter 2, ura- nium can be absorbed through the skin, and resulting toxicity depends on the solubility of the uranium compounds and the vehicle used. SUMMARY â¢ Neurobehavioral studies in Gulf War veterans are inconclusive. â¢ Studies in experimental animals have shown that uranium crosses the blood-brain barrier and accumulates in the brain. Accumulation after large expo- sure is associated with abnormal electrophysiologic effects, changes in monoam- ine metabolism, and neurobehavioral changes.
64 Risks to Military Personnel from Exposure to Depleted Uranium â¢ Forced-swim stress reduces uranium and zinc in the hippocampus and cerebellum. â¢ Continuing surveillance (14 y) of DU-exposed 1991 Gulf War veterans has yielded no evidence of reproductive-system dysfunction in males, abnor- malities in sperm, or alterations in neuroendocrine function after DU exposures that occurred in friendly-fire incidents. Nor is there evidence of excess sponta- neous abortions, fetal mortality, congenital anomalies, developmental delays, or abnormal infant neurobehavioral difficulties to date in the offspring of those veterans. â¢ The findings from animal studies are inconsistent with respect to ura- nium placental transfer and the presence of metal in the fetus. However, it should be noted that uranium concentrations to which dams were exposed were high (equivalent to about 200 g in a 70-kg person) in studies that reported pla- cental uranium transfer. â¢ Although studies cited here indicated hematologic and possibly heptatic effects of DU, the exposure durations or concentrations do not appear to be ap- propriate for extrapolation to human exposure conditions. â¢ No effects of DU on the human immune system have been identified. Rats given high doses of DU by gavage did not demonstrate intestinal musosal immunotoxicity. â¢ The effects of uranium compounds on the musculoskeletal system in laboratory animals are changes in bone formation and remodeling after oral, intraperitoneal, intravenous, and implantation exposure. It is not known whether DU has those effects, but there is not much potential for the large doses used in the studies to enter the body by inhalation or dermal exposure. Large fragments of DU embedded in muscle tissue caused soft-tissue sarcomas in experimental studies with rats and may have deleterious effects (sarcomas) in soldiers if large DU fragments are left in place over a lifetime. â¢ There are no reports of chemical toxicity in the cardiovascular system from uranium exposure. â¢ Absorption of uranium across the gastrointestinal mucosal lining is solubility-dependent, and most uranium compounds are not readily absorbed. In animal studies, no toxicity was observed after ingestion of uranyl nitrate, uranyl nitrate hexahydrate, uranium dioxide, or triuranium octaoxide. Mild hemorrhage occurred in animals exposed to a high dose. â¢ Studies in several animal species have shown that lethality and toxicity can result from cutaneous exposure to uranium compounds solublizied in a vari- ety of vehicles. Soluble uranium compounds cause transient irritation of the skin, and repeated applications may cause changes in epidermal integrity. RECOMMENDATIONS â¢ Monitoring of Gulf War veterans for neurologic and neurobehavioral effects should be continued.
Toxic Effects of Uranium on Other Organ Systems 65 â¢ Experimental studies should be conducted to determine the nature of and the LOAEL for neurologic and neurobehavioral effects of exposure to DU. â¢ On the basis of available reproductive-toxicity and developmental- toxicity data, samples of blood, urine, or semen of DU-exposed military person- nel should be collected for the measurement of uranium content and signs of abnormal reproductive function in men and women. In addition, the reporting of spontaneous abortions and congenital anomalies should be continued. â¢ The committee does not recommend additional studies of the hema- tologic or hepatotoxic effects of DU.