Depleted uranium munitions—where are we now? (2025)

Kinetic energy weapons that contain a penetrator of depleted uranium (DU) were first used in the Gulf War of 1991 and were subsequently used in the Balkans. DU penetrators are considered to have significant operational advantages over those made of tungsten as they are capable of penetrating the heavy armour of the modern battle tank. The use of DU rounds in military conflicts has, however, provoked a wide debate about the health consequences for soldiers and the local population since DU is a toxic metal and is radioactive.

The Royal Society became involved in the debate about the health hazards of DU munitions as a result of public concern, to produce an independent view on the science and the uncertainties, uninfluenced by the conflicting interests of governments and the military, who consider that the risks are very slight, and of other individuals and organisations, some of whom have suggested that hundreds of thousands of deaths from cancer may result from the use of DU in the Gulf War.

Large quantities of DU rounds were deployed in the Gulf War (about 340 tonnes) and much smaller amounts in the Balkans (about 11 tonnes). In both conflicts the majority of the DU rounds were fired from aircraft in strafing attacks where most of the penetrators miss their target and penetrate several metres into the soil. Consequently, large numbers of DU penetrators are believed to remain buried in the ground. Corrosion of these penetrators will occur with the possibility of a gradual rise in the uranium levels in local water supplies. About 10,000 larger calibre DU rounds were fired from tanks during the Gulf War, although these were not used in the Balkans. DU rounds that penetrate a target vehicle may pass straight through or, particularly if they hit heavy armour, may release a variable proportion of the penetrator as DU particles which ignite to produce an aerosol of DU oxides. The DU particles released during such impacts will be inhaled by those surviving within a struck tank or by those in the path of the DU aerosol. Unless adequate respiratory protection is used, DU particles will also be inhaled by those charged with cleaning up DU-contaminated vehicles. The fraction of a DU penetrator that is aerosolised and the fraction of the particles of DU oxides that are within the respirable range, as well as the solubility properties of the DU oxides, are not well documented and depend on the type of impact, but a range of values is available from test firings of DU rounds.

A number of exposure scenarios were considered in the two Royal Society reports on the Health Hazards of DU Munitions [1, 2] and central and worst-case intakes of DU were estimated from the range of values obtained from test firings. These estimated intakes, and the range of reported values of the properties of the DU oxides released during an impact or fire, were used to produce central estimates and worst-case estimates of risks for soldiers on the battlefield. Although there are uncertainties about the intakes of DU, and of the properties of the DU oxides, there is a clear view among radiation biologists that, given the equivalent doses to tissues, the excess lifetime risks of various fatal cancers can be estimated, perhaps with an order of magnitude of uncertainty. Therefore if the intakes of DU, or the properties of DU oxides, are in future better defined, the estimates of risk given in the Royal Society reports can be adjusted appropriately.

There are very different views on the health hazards of DU munitions. Most of the concerns of veterans and their advisors focus on the radiological effects of DU and consequently these are the focus of this editorial. Effects on the kidney and environmental consequences are, however, considered in the second of the Royal Society reports [2] and the main conclusions of both of the reports are outlined in the summary document published in this issue of the journal (page 131). The main radiologicalconcerns focus on the irradiation of lung tissues from inhaled DU particles and irradiation resulting from the translocation of inhaled particles to the thoracic lymph nodes. DU shrapnel is also a concern and the health of a group of US soldiers with retained shrapnel is being carefully monitored, with little signs so far of adverse health effects. However, soft tissue sarcomas have recently been reported around DU pellets implanted in the muscles of rats [3] and long-term monitoring of these soldiers is required. The overwhelming scientific view, presented in the two Royal Society reports and in other independent reviews, is that the main risks of exposure to DU aerosols are an increase in lung cancer and (from chemical toxicity) damage to the kidney, although these are likely to be evident only following substantial intakes. The equivalent doses to the thoracic lymph nodes following inhalation of DU particles are about ten times greater than those to the lung, but the former tissue is considered to be relatively insensitive to radiation-induced cancers [4].

Excepting lung cancer in a few soldiers who could be very heavily exposed to DU, any increases in other cancers, including lymphomas and leukaemias, are predicted to be too small to be detected, unless the intakes for large numbers of soldiers on the battlefield are very much greater than estimated, or the ICRP models greatly under-estimate the risks of inhaled particles of DU oxides. Some advisors to the veterans groups take the latter position, and argue that the alpha-particles from highly insoluble DU particles translocated to the thoracic lymph nodes are much more dangerous than implied by ICRP models, and that the risks of leukaemias and lymphomas are greatly under-estimated. These views are familiar to those who have followed the debate about leukaemia clusters and `hot particles' in discharges from the nuclear industry or increased leukaemias from the accident at Chernobyl. DU particles can hardly be called `hot' but large intakes of DU by a few soldiers working for protracted periods cleaning up contaminated vehicles without any respiratory protection, or survivors of struck tanks, could, under worst-case assumptions, result in large numbers of DU particles in the thoracic lymph nodes, providing doses of up to about 5 Gy over a 50 year period [1]. However, very much smaller doses are predicted for the great majority of soldiers on the battlefield and for those returning to live in the region.

It is therefore unlikely from current scientific knowledge that there will be any detectable increase in any cancers, except perhaps lung cancer if small groups of soldiers were exposed to large amounts of DU. This conclusion has been difficult to accept for some veterans and their advisors, who are searching for a cause of the illnesses in the veterans (including Gulf War Syndrome), and who suggest that large numbers of cancers and birth defects are being seen, for example in Iraq, and who criticise the use of models to estimate cancer risks. Such critics of ICRP models believe that cancer incidence data rather than models should be used to produce the estimates of risks. Of course they are correct. Ideally, risks of cancers should be determined from the excess mortality in a population exposed to known levels of radiation, but this approach is very rarely possible as exposures are typically too low. Data from the atomic bombs in Hiroshima and Nagasaki provide a major source of the data that underpin ICRP models but critics argue that this type of radiation exposure is so different from the internal exposures to alpha-emitting radionuclides that result from inhalation intakes of DU, or from `hot particles' from nuclear power plants, that they invalidate the use of ICRP models for estimating risks for the latter type of exposures. Risks from internal exposures to alpha-particles have been obtained from underground miners exposed to radon, since the number of exposed miners, and the excess mortality from lung cancer, are large enough to make robust measures of risk. These direct measurements of risk from internal exposures agree rather well with those from IRCP models [5, 6]. In fact the modelling approach may over-estimate the risk to the lung.

This agreement between ICRP models and direct measures of risks to the lung from internal exposure to alpha-emitting radionuclides tends to be ignored by critics of ICRP models who focus largely on increased risks of lymphomas and leukaemias from radioactive particles in the thoracic lymph nodes. The risks of lymphomas and leukaemias from inhaled alpha-emitting particles have been proposed by some to be more than 100-fold under-estimated by ICRP models [7]. However, this extreme conclusion has been proposed on the basis of highly controversial data on leukaemia clusters around nuclear facilities, or leukaemia rates following the Chernobyl accident, or a small numbers of cases of leukaemia in Italian soldiers who served in the Balkans. Any major shift in the current consensus scientific view of the risks of cancers from inhaled radioactive particles must be derived from robust epidemiological data. Epidemiological studies and their interpretation are fraught with difficulties and, unless the data are generally considered to be robust, and the conclusions from the data are broadly accepted, it is very unwise to adjust the risks obtained from ICRP models. However, we also need to be open-minded and to listen to, engage, and vigorously challenge where necessary, those whose views on radiation risks are heterodox.

The substantial literature on the mortality of workers in the uranium industry shows little evidence for any substantial increase in overall mortality or deaths from all cancers, or from specific cancers, among those working with uranium, except for underground miners where excess lung cancers are attributed to exposure to radon [1, 5]. Can we use this large body of epidemiological data to put some limits on the radiological risks of exposure to DU aerosols, or are the intakes of DU on the battlefield significantly different from those that occur in the uranium industry? There are clearly some important differences; inhalation exposure to pure DU oxide particles is very different to inhalation exposure to ores which contain a low proportion of uranium, but exposures to pure uranium oxides that are similar to those that occur following the military use of DU do occur in some parts of the uranium industry. However, it is very difficult to relate the chronic exposures that occurred in the early days of the uranium industry with those that could occur on the battlefield and it is unclear how much comfort we should take from the lack of a clear excess mortality or excess cancers from studies of uranium workers. However, they do seem to indicate that exposure to uranium in industrial settings has not led to any major increased risks of cancer. Mainstream scientific opinion therefore argues that there is little evidence of excess cancers among workers in the uranium industry, and little evidence from other groups exposed to internal doses of alpha-emitting radionuclides, or from animal studies with inhaled alpha-emitting radionuclides, to indicate that the estimates of cancer risks from inhaled DU particles produced by ICRP models are likely to be seriously in error [8, 9]. The suggestion that some cancer risks from inhaled alpha-emitting particles are more than 100-foldunder-estimated by ICRP models is difficult to reconcile with the studies of uranium workers, where a greater than 100-fold under-estimation of risk ought to lead to an obvious excess of leukaemias or lymphomas in these large epidemiological studies.

Are there any features of DU that could lead to a risk of cancer that is greater than expected from ICRP models? One possibility that has emerged from recent studies with a human osteosarcoma cell line is that uranium may also be chemically carcinogenic [10]. Increased levels of soluble uranium in cells around DU particles in lymph nodes, or around retained DU shrapnel, could enhance the radiological effects from the alpha-particles. At present there is no evidence that this occurs, and presumably if there are synergistic interactions between chemical and radiological effects, these would also occur in the uranium workers exposed to inhaled uranium oxide particles where clearly increased mortality from cancers are not observed.

How do we move forward? Perhaps the greatest knowledge gap is the extent of exposure to DU that occurs on the battlefield. Almost no measurements of uranium isotopes in urine have been made in veterans of the Gulf War or Balkans conflicts. This is unfortunate as reliable measures of uranium isotopes in urine would identify any soldiers who have been exposed to substantial intakes of DU and could lead to more sensitive epidemiological studies to establish if there are any links between DU exposure and ill health. Those few measurements of uranium isotopes that have been made give contradictory results. Thus, DU has been detected in some residents of Kosovo (data from Nick Priest), and in some UK and Canadian Gulf War veterans (data from Pat Horan), but DU has not been detected in hair or bone from other Canadian veterans from the Gulf War or from Kosovo [11]. Few laboratories have sufficient expertise in measuring uranium isotopes in urine containing small amounts of uranium and the reliability of some of these measurements has been questioned. The Ministry of Defence DU Oversight Board, which includes representatives from veterans groups and their advisors, is currently assessing the ability of a number of independent laboratories to achieve reliable and accurate measurements of DU in urine, which can then be used to look at levels of DU in the urine of veterans. It is hoped that a validated test will be available to veterans by the end of the year. Reliable validated measures of DU in urine will allow concerned veterans from the Gulf War and Kosovo to know if they have been exposed to substantial intakes of DU. If the tests are sufficiently sensitive it should be possible to detect inhalation intakes during the Gulf War of about 25 mg of DU (and lower intakes from the more recent Balkans conflicts). If this sensitivity is achieved in a fully validated test, veterans in which no DU is detected should have no health concerns from possible exposures to DU. Those that have measurable levels of DU in urine need to get the best possible scientific advice about any health implications from their estimated level of exposure, from scientists who do not have political reasons to underplay or overplay the risks of their exposure.

The Royal Society Working Group were unconvinced by the data used to support a much higher cancer risk from inhaled DU particles than that suggested by ICRP models, but a rational scientific debate about the risks from exposure of the thoracic lymph nodes to radiation is appropriate and the Royal Society recommended a thorough review by an independent expert group of the available data and the uncertainties.

Independent studies to evaluate the reports of increased cancers and birth defects in Iraq and the Balkans are also required, although these are dependent on high quality longitudinal data, which may not be available. It is well known that increased awareness of possible health concerns can lead to increased reporting of the relevant conditions and careful independent studies need to be carried out. So far, large-scale epidemiological studies of UK and US Gulf War veterans have shown no increase in mortality from cancer, or kidney disease, but these studies need to be continued to see if any significant excess mortality from these causes appears. As mentioned previously, more sensitive epidemiological studies should be possible if groups of soldiers who have known exposures to DU can be identified.

Adverse reproductive effects have been observed in rodents exposed to uranium [12] although most of these effects are evident after relatively large daily intakes of uranium. The possibility of effects on reproductive health (from DU or other toxic exposures) is being studied in both UK and US Gulf War veterans. Results from the UK epidemiological study are not yet available but self-reported birth defects were significantly increased in US Gulf War Veterans compared to a control group [13]. Such studies are subject to reporting bias and confirmation of these results using medical records is required.

Finally, there are multiple toxic, or potentially toxic, exposures in modern warfare (particularly during the Gulf War), and the Iraqi population has also been exposed to chemical agents. Thus, even if clear excesses of cancers or birth defects are established, it will be difficult to link them to DU, to any of the other exposures, or to some combination of toxic exposures. However, a clear independent scientific view on whether there are any excesses is a necessary first stage.

Depleted uranium munitions—where are we now? (2025)
Top Articles
Latest Posts
Recommended Articles
Article information

Author: Greg O'Connell

Last Updated:

Views: 5748

Rating: 4.1 / 5 (62 voted)

Reviews: 93% of readers found this page helpful

Author information

Name: Greg O'Connell

Birthday: 1992-01-10

Address: Suite 517 2436 Jefferey Pass, Shanitaside, UT 27519

Phone: +2614651609714

Job: Education Developer

Hobby: Cooking, Gambling, Pottery, Shooting, Baseball, Singing, Snowboarding

Introduction: My name is Greg O'Connell, I am a delightful, colorful, talented, kind, lively, modern, tender person who loves writing and wants to share my knowledge and understanding with you.