Review of the Toxicologic and Radiologic Risks to Military Personnel from Exposures to Depleted Uranium During and After Combat (2008)

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.

2 Toxicokinetics of Depleted Uranium Natural uranium is a mixture of three isotopes—234U, 235U, and 238U—that behave the same chemically. As discussed in Chapter 1, depleted uranium (DU) is a byproduct of uranium enrichment and has a decreased fraction of 235U and 234 U. Uranium is found in the body in soluble form only as tetravalent and hexavalent complexes with carbonate ions and proteins (Berlin and Rudell 1986). The absorption, distribution, metabolism, and elimination of uranium compounds depend on their physical and chemical characteristics. Their toxic potency depends primarily on their solubility because solubility affects tissue dose, which is directly related to toxicity. Table 2-1 lists uranium compounds according to rough solubility (ATSDR 1999). Aerosol byproducts of DU muni- tions would primarily be the insoluble oxides uranium trioxide, triuranium oc- taoxide, and uranium dioxide (IOM 2000). ABSORPTION Table 2-2 provides information on absorption of uranium compounds on the basis of exposure route and solubility. Overall, absorption of uranium com- pounds is low except for soluble particles in the lung. Absorption of inhaled uranium particles is determined by the aerodynamic diameter and solubility of the particles. Large particles (over 10 µm in activity median aerodynamic diameter [AMAD]) are exhaled or may deposit in the ante- rior nasal passages. Deposited particles that are not cleared by mechanical means, such as nose blowing, may be transported to the posterior nasal passages where they are swallowed. The small fraction of inhaled particles that deposits in the tracheobronchial compartments is cleared by the mucociliary escalator that transports them from there to the pharynx, from which they are swallowed and enter the gastrointestinal tract (IOM 2000), or by phagacytosis, which trans- fers the particles to the tracheobronchial lymph nodes. Smaller particles (less 18 

Toxicokinetics of Depleted Uranium 19 TABLE 2-1 Solubility of Uranium Compounds Uranium Compound Chemical Formula More water-soluble compounds Uranyl nitrate UO2(NO3)2 Uranyl nitrate hexahydrate UO2(NO3)2●6H2O Uranium hexafluoride UF6 Uranyl fluoride UO2F2 Uranium tetrachloride UCl4 Uranium pentachloride UCl5 Ammonium uranyl tricarbonate (NH4)4UO2(CO3)3 Less water-soluble compounds Sodium diuranate (yellow oxide of Na2U2O7●6H2O uranium) Ammonium diuranate U2(NH4)2O7 Uranyl acetate UO2(CH3CO2)2 Insoluble compounds Uranium tetrafluoride UF4 Uranium trioxide UO3 Uranium dioxide UO2 Uranium peroxide UO4 Triuranium octaoxide U3O8 Source: ATSDR 1999. TABLE 2-2 Absorption by Exposure Route Absorption of Absorption of Exposure Route Soluble Compounds Insoluble Compounds Inhalation 5% or more <1% Ingestion 0.1-2% 0.01-0.2% Skin contact <1% <1% Source: Data from ATSDR 1999. than 10 µm) are deposited predominantly deeper in the terminal bronchioles and alveoli. Soluble particles in the lungs and tracheobronchial lymph nodes are taken up into the systemic circulation within days (IOM 2000). Less-soluble particles are likely to remain in pulmonary tissue and associated lymph nodes for weeks. Relatively insoluble compounds are least likely to enter the systemic circulation and may remain in the lung and tracheobronchial lymph nodes for several years or decades. Most inhaled uranium aerosol is cleared from the respiratory tract via the gastrointestinal tract, but a fraction is absorbed into the body fluids and distrib- 

20 Risks to Military Personnel from Exposure to Depleted Uranium uted throughout the body. Solubility in body fluids is the most important factor in determining absorption. The International Commission on Radiological Pro- tection (ICRP 1994a, 1995b) classifies inhaled aerosols broadly in terms of the rate of their absorption into the body as fast (F), moderate (M), or slow (S). Type F uranium compounds include uranium hexafluoride, uranyl fluoride, and uranyl nitrate; of the fraction not excreted via the gastrointestinal tract, 100% is absorbed with a half-life of 10 min. Type M uranium compounds include ura- nium trioxide, uranium tetrafluoride, uranium tetrachloride, and triuranium oc- taoxide, (the latter may behave like a type S compound under some circum- stances, particularly if produced at high temperature); of the fraction not excreted via the gastrointestinal tract, 10% is absorbed with a half-life of 10 min, and the remaining 90% is absorbed with a half-life of 140 d. Type S ura- nium compounds include uranium dioxide (and, as indicated above, triuranium octaoxide under some circumstances) and have very low and slow absorption; most of the insoluble aerosol is excreted relatively quickly via the gastrointesti- nal tract, and of the remainder, 99.9% is absorbed from the respiratory tract with a half-life of 7,000 d (about 19 y). Of the particulate material cleared from the respiratory tract via the gastrointestinal tract, only 0.2% of type S compounds or 2% of type F and M compounds is absorbed (ICRP 1995b). Uranium may enter the body through ingestion of food and water. As indi- cated above, gut absorption of uranium is poor, and only a small fraction of the uranium ingested is absorbed in the gut. Numerous observations in humans have shown that the uptake fraction of soluble uranium is about 1-2% and that of the insoluble oxide forms about one-tenth that (Wrenn et al. 1989a; Harduin et al. 1994; Medley et al. 1994; Leggett and Harrison 1995). Dermal absorption of uranium compounds has not been characterized in humans, but animal studies indicate that they can penetrate the skin. The study results are difficult to compare because vehicles with differing physical charac- teristics were used. Orcutt (1949) found that large doses of uranyl nitrate, uranyl fluoride, uranium pentachloride, uranium trioxide, sodium diuranate, and am- monium diuranate were absorbed through the skin and caused poisoning and death in experimental animals. However, the insoluble oxides uranium dioxide, uranium tetroxide, and triuranium octaoxide did not cause toxicity. A later study confirmed the effect of water solubility on dermal toxicity; de Rey et al. (1983) found that uranyl nitrate and ammonium uranyl tricarbonate were more toxic (in terms of lethality and body-weight changes) than uranyl acetate and ammonium dianurate. In that study, de Rey et al. visualized the movement of uranyl nitrate through the skin with x-ray microanalysis and scanning electron microscopy. Electron-dense areas could be seen in the epidermis 15 min after application, and the electron-dense material could be seen in the capillary endothelial cells of the upper epidermis 24 h after one application; no traces were left in the skin 48 h after application. Topical application of uranyl nitrate to rats caused kidney and bone effects that were related to exposure area and time (Lopez et al. 2000); uranyl acetate and ammonium diuranate caused less toxicity, and uranium diox- ide caused no toxicity. In vitro studies with uranyl nitrate have shown that 

Toxicokinetics of Depleted Uranium 21 0.04% of the applied dose was absorbed by human skin after 2 h (Tymen et al. 2000), and the steady-state flux across hairless rat skin was 0.13 ng/cm2 per hour (Petitot et al. 2004). Open wounds, burns, or other conditions in which the skin barrier is disrupted would enhance absorption through the skin. The release of uranium from embedded fragments of DU has been studied in soldiers and laboratory animals. A 10-y followup of soldiers involved in “friendly-fire” incidents with DU munitions showed urinary uranium concentra- tions that suggested that DU was still being absorbed into the body from frag- ments (McDiarmid et al. 2004b). Several studies have investigated implantation of DU pellets in rats (Pellmar et al. 1999a,b; Arfsten et al. 2005, 2006). They all used DU and tantalum (control) pellets that were 1 mm in diameter and 2 mm long and weighed 38 mg. They changed the dose by surgically implanting dif- ferent numbers of pellets (from four to 20) in the rats’ gastrocnemius muscle. Twenty pellets (760 mg of DU) in a 250-g rat is the equivalent of about 0.22 kg (0.5 lb) of DU in a 70-kg (154-lb) person (Arfsten et al. 2006). Absorption stud- ies in rats showed that DU release began within a day and that release continued throughout 18 months of observation (Pellmar et al. 1999a). DISTRIBUTION Once in the blood, hexavalent and tetravalent uranium compounds form complexes with carbonate ions and proteins (Berlin and Rudell 1986). About 47% of natural uranium forms a complex with bicarbonate in plasma, 32% binds to plasma proteins, and 20% binds to erythrocytes (Chevari and Likhner 1968, as cited in IOM 2000). After pulmonary intubation of rats, 50% of the 233U was bound to transferrin, 25% to citrate, and 25% to bicarbonate (Cooper et al. 1982). Absorbed uranium is found in all tissues but deposits preferentially in bone and kidney (ATSDR 1999). Initial distribution is affected by the route of entry, particle size, and solubility of the compound. After inhalation of uranium dioxide dust, the lungs and lymph nodes account for more than 90% of the body burden (Leach et al. 1970). Morris et al. (1990) exposed rats to a uranium diox- ide aerosol with a median particle diameter of 2.7-3.2 µm. At 720 d, 82% of the body burden was in the lung and 10% in the lymph nodes. A normal adult’s body burden of uranium is about 90 µg. It has been estimated that about 66% of that is in bone, 16% in the liver, 8% in the kidneys, and 10% in other tissues (ICRP 1975; ATSDR 1999). In bone, uranium replaces calcium in hydroxyapa- tite (IOM 2000) and has a half-life of 300 d (Harley et al. 1999). In the kidneys, uranium accumulates primarily in the proximal tubules, where it is reabsorbed (IOM 2000). After in vitro and in vivo dermal exposures to uranium as a nitrate solution, the majority of the uranium was localized to the epidermis after 2 h, and nearly all the applied uranium was associated with the skin (Tymen et al. 2000). 

22 Risks to Military Personnel from Exposure to Depleted Uranium An 18-mo study of DU implantation in rats showed interesting tissue- distribution profiles (Pellmar et al. 1999a). Kidney and bone concentrations were highest 1 d after implantation and constituted the only significant changes. At 1 mo, bone and the kidneys still had the highest concentrations, but differ- ences from controls were also seen in urine, muscles, the liver, the spleen, se- rum, and the brain. Kidney concentrations appeared to peak 6 mo into continu- ous exposure but remained increased. There were low concentrations of DU in the urine throughout the study, and they were relatively constant between 6 and 18 mo. Tibia and skull concentrations continued to increase over the course of the study. Brain concentrations were increased in the motor and frontal cortex, midbrain, cerebellum, and vermis at 18 mo. Other tissues that showed increases at 18 mo were lymph nodes, the testicles, the teeth, the heart, and the lungs. Uranium can cross the placenta after parenteral administration (0.01- 0.03% of an intravenous dose) in rats (Sikov and Mahlum 1968) and can cross the blood-brain barrier of rats (Pellmar et al. 1999a; Barber et al. 2005). Whole- body retention of hexavalent uranium after an intravenous injection in beagles was 17% at 7 d, 10% at 94 d, 7.6% at 1 y, and less than 5% at 2 y (Stevens et al. 1980). METABOLISM There is little information on the metabolism of uranium compounds (Ber- lin and Rudell 1986), but oxidation of tetravalent uranium to hexavalent uranium is likely to occur in the blood (ATSDR 1999). ELIMINATION In mammals, inhaled soluble uranium compounds are eliminated primarily by the kidneys and to a small extent in the feces. Inhaled insoluble uranium compounds are eliminated primarily in the feces. More than 90% of intrave- nously injected hexavalent uranium is excreted by the kidneys and less than 1% in the feces (Berlin and Rudell 1986). Renal clearance is rapid; human studies of comatose, terminally ill patients with brain tumors showed that two-thirds of an intravenous dose (0.07-0.28 mg/kg of body weight) left the bloodstream in 6 min and that 49-84% of the dose was excreted in the urine within 24 h (Lues- senhop et al. 1958). In contrast, McDiarmid et al. (2004b) showed that uranium from DU fragments was still being cleared by the kidneys 10 y after friendly-fire incidents involving U.S. troops. In humans, there appears to be a two-phase pat- tern of excretion after occupational exposure to soluble and insoluble com- pounds (Berlin and Rudell 1986). In the case of soluble compounds, there is a fast phase in which about 70% of the dose is cleared in the first 24 h, followed by a slow phase with a half-life of months. In the case of insoluble compounds, the fast-phase half-life is 11-100 d, and the slow-phase half-life is 120-1,500 d. 

Toxicokinetics of Depleted Uranium 23 Ingested uranium that is not absorbed is eliminated primarily (94-99%) in the feces (Hursh et al. 1969; Leggett and Harrison 1995; ATSDR 1999). BIOKINETIC MODELS Biokinetic models provide the means by which the dose of uranium, and the associated risk, can be assessed. Such models mathematically characterize the processes by which DU is taken up by various tissues, distributed in the body as a function of time, and cleared or excreted from various tissues and the body itself. Although human experience with uranium spans 2 centuries and uranium was once used therapeutically as a treatment for diabetes mellitus, the biokinetics of uranium are not well known. Over the years, a number of sys- temic models and refinements have been proposed (Bernard and Struxness 1957; ICRP 1977,1979; Lipsztein 1981; Durbin 1984; Wrenn et al. 1985a; ICRP 1988; Wrenn et al. 1989b; Fisher et al. 1991; Leggett 1992; Harduin et al. 1994; ICRP 1995a; Leggett and Harrison 1995; Leggett and Pellmar 2003). Generally, the models indicate that most DU absorbed into the systemic circulation by what- ever route of entry is quickly excreted in urine. A small fraction of absorbed DU may be taken up by and incorporated into some tissues and irradiate them at the cellular level. Skeleton, liver, and kidneys are considered the primary sites of deposition of uranium. The dose delivered from the small fraction of uranium that is not quickly excreted determines the potential for radiologic effects. Perhaps the most widely used and accepted biokinetic model for ura- nium—certainly for radiation-protection purposes—is the current comprehen- sive model put forth by ICRP (1995a,b), as shown in Figure 2-1. It is a recycling model; that is, uranium in the blood flows into organs and then back to the blood. Major compartments of the model are the skeleton, liver, kidneys, and a generic soft-tissue compartment. The generic soft-tissue compartment is mod- eled as all other organs and tissues and receives all the distribution from the blood not accounted for by other organs or excretion. Most of the uranium in soft tissue is retained only briefly, but a small amount (less than 1%) is retained essentially indefinitely. The skeleton is modeled as two compartments: one represents trabecular bone and one cortical bone. Additional subcompartments represent movement of uranium from the bone surfaces to bone volume and back. The skeleton receives about 15% of the uranium leaving the blood, but most of that is returned fairly rapidly (retention half-lives, 30 d or less); a very small amount of the uranium is retained in the skeleton for periods (years) con- sistent with normal turnover rates. The bone marrow is also included in the skeleton compartment to allow calculation of dose to red marrow. The liver is modeled as two compartments: one that receives uranium from the blood but returns most of it with a 7-d half-life and one that receives uranium from the other liver compartment and retains it with a half-life of about 10 y. About 1.5% of the uranium leaving the blood goes to the liver. The kidney is modeled as two 

24 Risks to Military Personnel from Exposure to Depleted Uranium FIGURE 2-1 Biokinetic model for uranium. Source: ICRP 1995a. Reprinted with per- mission; copyright 1995, International Commission on Radiological Protection. compartments: kidney tissue and a urinary path that flows only one way, to the bladder and excretion. The kidney tissue itself receives only about 0.05% of the uranium leaving the blood but retains it with a moderately long half-life of about 5 y.1 Excretion is the largest pathway of removal of uranium from the blood. Of the uranium in the blood, 63% is modeled as being transported directly to the bladder; another 12%, which passes through the kidneys (renal tubules), is re- tained briefly (for days) and then passes on to the bladder. Systemic uranium is eliminated virtually entirely through urine. Using this or other systemic models and the ICRP respiratory tract model for inhalation exposure, one can calculate the radiation dose to individual tissues and organs from a known intake of DU and thus gain an indication of the risk of stochastic effects (see Chapter 6). 1 The cellular basis of renal retention of uranium is not well understood. Unlike reten- tion of cadmium and lead, uranium retention has not been associated with an increase in a specific protein. Leggett (1989) suggested that concentration and insolubilization of ura- nium in lysosomes might be a protective mechanism. Alternatively, uranium may be retained by macrophages of interstitial cells, as has been shown to occur in alveoli (Tasat and de Rey 1987). 

Toxicokinetics of Depleted Uranium 25 Recent studies have evaluated and applied the basic models. Chen et al. (2004) used the ICRP model to calculate a table of kidney burdens from inhala- tion and oral intakes of soluble and moderately soluble uranium. Leggett and Pellmar (2003) evaluated models for application to uranium migration from em- bedded DU fragments. The time-dependent rate of uptake into plasma from pel- lets was adjusted to match the urinary data in the DU-implanted rats from the Pellmar et al. (1999a) study. Leggett and Pellmar used data from Morris et al. (1990) on inhalation of insoluble uranium dioxide to mimic slow release from pellets. The data after the first 6 mo fit well, but the data for the first 6 mo did not, perhaps because of variable urinary excretion or some peculiarity of the biokinetics of uranium release from embedded DU metal. Leggett and Pellmar concluded that it is reasonable to apply ICRP’s updated biokinetic model for uranium to assess chemical risk to soldiers who have embedded DU fragments. SUMMARY • Absorption of uranium compounds is low (less than 1%) by all expo- sure routes, except for soluble compounds, of which 5% or more is absorbed after inhalation. • Initial distribution of uranium compounds depends on their solubility and the route of absorption. Uranium compounds complex with ions and pro- teins in the blood, are distributed to all tissues, and preferentially deposit in bone and kidneys. Uranium can cross the placenta and the blood-brain barrier. • The metabolism of uranium is not well understood, but oxidation of tet- ravalent uranium to hexavalent uranium is likely to occur. • After inhalation, elimination of soluble uranium is primarily by the kidneys, and insoluble uranium is eliminated primarily in the feces. In general, soluble compounds are cleared more rapidly than insoluble compounds. • Release of DU from embedded particles is slow and occurs over many years. • Well-established models are available to predict the toxicokinetics of uranium from inhalation exposures.