Registration Dossier

Administrative data

Link to relevant study record(s)

Description of key information

A toxicokinetics assessment was performed for gadolinium oxide and gadolinium oxalate, two poorly water soluble gadolinium compounds. The assessment was performed using the physicochemical characteristics of both compounds in combination with experimental information reported in literature on the absorption, distribution, and/or elimination/excretion of gadolinium. Preliminary conclusions based on these data were then checked for consistency with the available toxicological data on these compounds.


Oral, respiratory and dermal absorption are expected to be (very) limited, due to the poor water solubility of gadolinium oxide and gadolinium oxalate at physiologically relevant pH levels, as well as the complexation behaviour of gadolinium with phosphates, carbonates, etc. Gadolinium is not expected to be bioavailable to a significant extent at potential uptake sites in the body. The oral, respiratory and dermal absorption factors for gadolinium after exposure to gadolinium oxalate/oxide are set at 1, 1 and 0.1%, respectively, as a worst case for risk assessment purposes.


Based on all available information on distribution, it can be concluded that – although absorption of gadolinium after oral, dermal or inhalation exposure to gadolinium oxide/oxalate is expected to be (very) limited – any gadolinium that may be taken up in the blood would complex with proteins and form colloids (e.g., with carbonates, phosphates, …) that would be removed from the blood stream by phagocytic cells and temporarily stored in organs of the reticuloendothelial system such as liver and spleen.


Although there is not much experimental information informing on elimination/excretion mechanisms of gadolinium, the limited available information as well as information available for other lanthanides, allows drawing the conclusion that elimination via the faeces is by far the most important route of elimination/excretion after oral administration. Limited amounts of gadolinium that might be absorbed can be expected to end up as depositions in organs/tissues that are part of the reticuloendothelial system, from which slow clearance is expected. No experimental information is however available on the exact mechanisms for excretion of such accumulated material, apart from the potential role of biliary excretion for material stored in the liver. For clearance of depositions from other organs/tissues, slow urinary excretion might be involved, but this is not experimentally confirmed at this point in time.

Key value for chemical safety assessment

Bioaccumulation potential:
no bioaccumulation potential
Absorption rate - oral (%):
Absorption rate - dermal (%):
Absorption rate - inhalation (%):

Additional information


Some experimental data (published scientific literature) were identified informing on the toxicokinetic behaviour of gadolinium and its compounds. These data, together with information on the physicochemical characteristics of gadolinium oxide/oxalate as well as the toxicological information available on these compounds, are used to perform a qualitative assessment of the absorption, distribution/metabolism and elimination/excretion of these gadolinium compounds.

Note that in this document, a Klimisch score is only reported for the toxicity studies that are included in IUCLID (study summaries in the dossiers for gadolinium oxide and/or gadolinium oxalate). However, for this assessment, all available data providing information that could contribute to the assessment of the toxicokinetic profile of gadolinium oxide and gadolinium oxalate is considered, and the publications only informing on toxicokinetics are not scored for reliability as they are not entered in IUCLID. Therefore, for such studies, no Klimisch score is provided in this assessment.

Gadolinium oxide as well as gadolinium oxalate are solid gadolinium compounds with gadolinium in its most prevalent oxidation state, i.e. +III. Both compounds are rather insoluble in water at neutral pH and higher. Water solubility of gadolinium oxide at pH 4.0, 6.4, 7 and 9 was 2.5 g/L, 3.5, 0.53 and 0.12 mg/L, respectively, in a water solubility study at 20°C (Jean-Baptiste, 2017a). Water solubility of gadolinium oxalate was determined to be 5.1 mg/L at pH 4.0, 1.0 mg/L at pH 6.4, and 1.3 mg/L at pH 7 and 9 at 20°C (Jean-Baptiste, 2017b). When comparing dissolved gadolinium measurements, the results are even more similar (2.1 g Gd/L, 3.1, 0.46 and 0.11 mg Gd/L at pH 4.0, 6.4, 7 and 9, respectively, for gadolinium oxide, and 2.8, 0.6, 0.7 and 0.7 mg Gd/L at pH 4.0, 6.4, 7 and 9, respectively, for gadolinium oxalate). A clear decrease of water solubility with increasing pH was observed. At physiologically relevant pH levels, water solubility (gadolinium-based) of both compounds is rather similar.

Since it is generally assumed that for metals and metal compounds, the metal ion (regardless of the counterparts of the metal in the respective metal compounds) is responsible for the potential systemic effects after uptake, information on other gadolinium compounds can be used in this assessment as long as their similarities or differences in terms of inherent properties, such as water solubility, are discussed to predict the behaviour of the compound of interest. As indicated in ECHA’s guidance on QSARs and grouping of chemicals (ECHA, 2008), comparison of the water solubility can be used as a surrogate to assess the bioavailability of metals, metal compounds, and other inorganic compounds. In the case of gadolinium compounds, this simplistic approach assumes that a specific poorly water soluble gadolinium compound will show similar toxicokinetic behaviour and toxicological hazards as other equally poorly water soluble gadolinium compounds. Therefore, studies evaluating the toxicokinetic behaviour of gadolinium oxide are also considered relevant for gadolinium oxalate. Further, studies evaluating the toxicokinetic behaviour of water soluble gadolinium compounds (e.g., gadolinium trichloride) – which may give rise to slightly higher levels of bioavailable gadolinium – may also be considered relevant to a certain extent, as long as the potential relevant differences are described.


The toxicokinetic behaviour of the counter ion is thus not evaluated in this document (e.g., the oxalate in gadolinium oxalate). For considerations on the oxalate with regard to the toxicological properties of gadolinium oxalate, reference can be made to the read across document attached to IUCLID Section 13.



Oral/gastro-intestinal (GI) absorption

Gadolinium oxide and gadolinium oxalate have a molecular weight of 362.5 and 578.56 g/mol, respectively, with gadolinium itself having a molar mass of 157.25 g/mol. The latter value being well below 500 g/mol (molecular weights < 500 g/mol are favourable for absorption), it can be assumed that uptake of bioavailable species (free gadolinium cation and/or potentially other water soluble gadolinium species with relatively low molar mass) is possible. The extent of uptake, if any, would however be determined by the bioavailability of gadolinium in the gastro-intestinal tract. Gadolinium oxide and gadolinium oxalate are poorly soluble in water (Jean-Baptiste, 2017a,b) at pH levels relevant for the uptake sites in the gut. The pH at the entrance of the duodenum is about 6, gradually increasing to pH 7.4 in the terminal ileum (Fallingborg, 1999). Although a higher solubility of gadolinium oxide/oxalate can be expected in the stomach, any dissolved gadolinium is expected to be rapidly precipitated when it enters the gut. Precipitation will be further enhanced in the gastro-intestinal tract due to the presence of ligands that have strong affinity for gadolinium. Phosphate for instance forms strong, insoluble complexes with gadolinium, in a pH-independent way. Further, precipitation of carbonate complexes of gadolinium becomes more important with increasing pH. Therefore, it is expected that the bioavailability of gadolinium for uptake in the small intestine will be very limited. Complexation and precipitation as a function of the presence of inorganic ligands and pH has been confirmed by modelling in Visual Minteq (see also the document for justification of waiving further algal growth inhibition experiments with rare earth compounds, attached to IUCLID Section 13).

Studies directly evaluating the absorption of gadolinium following oral exposure to gadolinium oxalate/oxide in animals and humans are not available. For gadolinium trichloride however (a water soluble gadolinium compound), gadolinium accumulation was measured in different organs of rats repeatedly exposed via oral gavage to doses up to 1000 mg/kg bw/day gadolinium trichloride for 28 days, followed by 14 days recovery (Ogawa et al., 1992, Klimisch 4, only abstract available). Gadolinium was reported to be accumulated in the liver, kidney, spleen and femur in a dose dependent manner. Unfortunately, the level of accumulation was not reported in the abstract. The same authors have performed a similar study for yttrium trichloride (yttrium is a transition metal chemically similar to the lanthanides) (Ogawa et al., 1994, full publication available), from which it could be deduced that the accumulated yttrium represented only a very small % of the daily dose (and certainly of the total dose administered over 28 days). The highest concentrations were found in the kidney after 28 days of exposure to the highest test dose (1000 mg/kg bw/day, as YCl3.6H2O): ca. 7 µg Y/g organ dry weight (dw) in females and ca. 8 µg Y/g organ dw in males. Considering the body weight of the rats (ca. 170 g for males, ca. 130 g for females), the administered dose of yttrium was ca. 39.5 mg Y/day in males and ca. 30 mg Y/day in females. Although dry and wet kidney weights are not reported, it is clear from these figures that the accumulated yttrium represents a very small % of the daily dose and certainly of the total dose administered over 28 days. Further, because accumulated concentrations in organs of rats of the recovery group (14 days recovery after 28 days of exposure) were not much lower than after 28 days of exposure, and the relative distribution between organs was still rather similar, it can be safely assumed that the accumulated concentrations were mainly determined by absorption during the exposure period. It was therefore concluded that the absorption of yttrium via the gastro-intestinal tract after repeated oral exposure to yttrium trichloride is rather limited. Because of the common behaviour of rare earths and similar transition metals such as yttrium, it is expected that similar observations were made for gadolinium after repeated oral exposure to gadolinium trichloride.

The limited absorption after exposure to water soluble rare earth compounds is not surprising: because of the complexation behaviour and pH-dependent solubility of rare earths in general, water soluble rare earth compounds are not expected to give rise to much higher levels of bioavailable rare earth at the uptake sites in the gut compared to insoluble/sparingly soluble rare earth compounds. Therefore, information supportive of limited uptake after oral exposure to water soluble rare earth compounds is also considered relevant for less soluble rare earth compounds as a worst case. If absorption is observed to be limited after oral exposure to water soluble rare earth compounds, it will be at least equally or even more limited after oral exposure to insoluble/sparingly soluble rare earth compounds.

Then, it is investigated whether the available toxicological data for the gadolinium compounds under consideration in this toxicokinetics assessment (i.e. gadolinium oxide and gadolinium oxalate) support the indications of limited oral absorption.


Gadolinium oxide was shown not to be harmful or toxic to rodents after single oral administration, in four different studies (key studies: LD50 rats > 2000 mg/kg bw, Clouzeau, 1994, Klimisch 1; LD50 rats > 5000 mg/kg bw, Shapiro, 1990, Klimisch 1; supporting studies: LD 50 rats > 1000 mg/kg bw (i.e. the only dose tested in this study), Bruce et al., 1963, Klimisch 3; LD 50 mice (after injection into the stomach) > 10,000 mg/kg bw, Mogilevskaya and Roshchina, 1964, Klimisch 4). No clinical signs, deleterious effects, apparent abnormalities at macroscopic examinations or mortality were noted following exposure to the test item.


After single oral administration of 2000 mg gadolinium oxalate/kg bw to female rats, no mortality or clinical signs were recorded (Mátyás, 2016, Klimisch 1). The LD50 was concluded to be greater than 2000 mg/kg bw.


The absence of systemic effects after single oral administration of gadolinium oxide and gadolinium oxalate is supportive of limited bioavailability of gadolinium for uptake in the small intestine, as discussed above.


In a combined repeated dose toxicity study with the reproduction/developmental toxicity screening test in Wistar rats, gadolinium oxide was tested at 110, 330 and 1200/1008 mg/kg bw/day (1200 mg/kg bw/day until day 17 of the study, 1008 mg/kg bw/day from day 18 to the end of the study due to an erroneous dosing). The NOAEL (No Observed Adverse Effect Level) for systemic toxicity of the parent animals was ≥ 1200/1008 mg/kg/day based on the absence of adverse effects: at this test dose, there was no mortality among parent animals, no clinical findings (daily or weekly), no differences in the functional observational battery (including grip strength and locomotor activity), no differences in mean absolute or relative organ weights, and no overt macroscopical or microscopical findings of toxicological relevance (Papineau, 2017, Klimisch 1).


Regarding the reproductive data, there were no effects of the treatment on mating, fertility, and gestation indexes of the parents, no effects on viability, clinical signs, body weight, and sex ratio of the pups, and no test-item related findings in the pups at post-mortem examination. However, the NOAEL for reproductive performance and toxic effects on progeny was reported to be 330 mg/kg bw/day, based on a higher non-statistically significant mean post-implantation loss. These results were due to the contribution of 2 females of the high dose group which delivered 7 live pups. Although a high variability was observed at this dose level (standard deviation: 20.2), an effect of treatment with the test item could not be excluded and might be the result of limited uptake of Gd3+. However, it should be noted that the observed effects do not trigger classification.


Further, the results of a less detailed supporting study, in which guinea pigs received a dose of 2000 mg/kg bw of gadolinium oxide by oral gavage every other day for a month, are in line with the findings of Papineau (2017) on systemic effects in parent animals, as no significant adverse effects related to exposure to the test item were observed (Mogilevskaya and Roshchina, 1964, Klimisch 4).


The available toxicological data reported above (oral exposure) demonstrate no overt adverse effects for most endpoints after exposure to gadolinium oxide or gadolinium oxalate and do not trigger classification. Based on the equally poor water solubility of gadolinium oxide and gadolinium oxalate, it is expected that the bioavailability/bioaccessibility of the common element Gd3+will be equally low and too low to elicit effects in living organisms. The results of these studies do not provide reasons to deviate from the assumption that absorption of gadolinium following oral exposure to gadolinium oxide and gadolinium oxalate is limited.


Absorption from the gastro-intestinal lumen can occur by passive diffusion but also by specialised transport systems. Regarding absorption by passive diffusion, the lipid solubility and the ionisation are important. However, metal compounds such as gadolinium oxalate/oxide are usually not lipid soluble and are thus expected to be poorly absorbed by passive diffusion (Beckett et al., 2007). It has been demonstrated that several metals can cross cell membranes via ion channels or specific carriers intended for endogenous substrates (Beckett et al., 2007). For gadolinium compounds, there is no direct experimental evidence for the existence of a mechanism of transport responsible for uptake of Gd3+. It has however been described that gadolinium is very close to the divalent calcium ion as reflected by size, bonding, coordination, and donor atom preference (Evans, 1990). The ionic radius of Gd3+ (107.8 pm) is close to that of Ca2+ (114 pm) and therefore this rare earth element could be a calcium channel blocker. In certain in vitro/ex vivo studies, Gd3+ (from soluble forms) was demonstrated to be a blocker for several types of voltage-gated calcium channels at nano- to micromolar concentrations (Konoha et al., 2004; Palasz and Czekaj, 2000; Zhang and ter Keurs, 1996), which suggests that bioavailable Gd3+ might be taken up into the systemic circulation via such channels. This would of course require the presence of free Gd3+ in the system, which is unlikely to occur to a significant extent after oral exposure to insoluble/poorly soluble compounds such as gadolinium oxide/oxalate (see further).


Based on the abovementioned information on the physicochemical characteristics of gadolinium oxalate/oxide, the indications from literature on the limited absorptionafter oral exposure to gadolinium trichloride (a water soluble gadolinium compound), and the available toxicological data for gadolinium oxalate/oxide, the absorption factor for gadolinium after oral exposure to gadolinium oxalate/oxide is set at 1% as a worst case for risk assessment purposes.


Respiratory absorption

Low exposure to gadolinium oxalate/oxide is expected based on the inherent properties of these compounds. No vapour pressure value has been determined as these compounds do not melt below 300°C (Demangel, 2016a,b). Therefore, inhalation of gadolinium oxalate/oxide as a vapour is not likely to occur.

Gadolinium oxalate is typically manufactured as wet powder. Thus, the formation of respirable suspended particulate matter is unlikely. Occasionally manufactured dry powders typically have a D50 > 50 µm, which implies that no or only a very limited fraction of the powders may be respirable and capable of reaching the alveolar region of the lungs. Consequently, human exposure by inhalation is considered not significant for this compound.


The particle size distribution of gadolinium oxide samples which can be considered representative for what ismanufacturedis as follows: D10, D50 and D90 were determined to be in the ranges 0.587-1.2 µm, 1.3-3.10 µm, and 3.4-8.76 µm, respectively. Based on this particle size information, the inhalation route of exposure should be taken into consideration for this compound. Therefore, the absorption of gadolinium from potentially inhaled/respired particles of gadolinium oxide is assessed here below.

Gadolinium oxide is a poorly soluble inorganic compound. Moreover, its solubility decreases sharply with increasing pH (Jean-Baptiste, 2017a). Due to its poor solubility, once deposited on the walls of the airways, the concentration of bioavailable forms – such as (solely or predominantly) the free Gd3+ cation – in the bronchoalveolar fluid is expected to be very low, assuming a pH of about 6.6 (for healthy individuals). Therefore, absorption from the lungs to the circulatory system is expected to be minimal. Since respired gadolinium oxide particles are expected to accumulate in the lungs mainly as particulate material, they are expected to be engulfed from the alveolar region by alveolar macrophages. The macrophages will then either translocate particles to the ciliated airways or carry particles into the pulmonary interstitium and lymphoid tissues. Deposited material may also be transported out of the respiratory tract and swallowed through the action of mucociliary clearance mechanisms, especially material settled in the tracheobronchial region. In that case it will contribute to the gastro-intestinal absorption rather than to the absorption via inhalation.

The absorption of gadolinium following inhalation exposure to gadolinium oxide as well as the fate of undissolved respired gadolinium oxide particles was investigated in a number of studies that are described below. Although of low reliability, these studies confirm the theoretical assumptions set out above.


Ball and Van Gelder (1966, Klimisch 3) conducted a repeated dose toxicity study via inhalation in mice using gadolinium oxide. 256 mice were exposed for 20 to 120 days via inhalation chambers to gadolinium oxide powder (Mass Median Aerodynamic Diameter (MMAD) < 1 µm, mean particle size 0.312 µm), 6 hours/day, 5 days/week, at an average concentration of 30.0 mg/m³. Localised accumulations of metal-containing macrophages were observed in lungs of mice exposed for 20 days or more. The metal-containing macrophages were usually located free within the alveoli. Less commonly, they were found in the interstitium of the alveolar septa. Metal was always present in the tracheobronchial lymph nodes, where it was concentrated in the reticuloendothelial cells of the medulla. Based on these observations, the pulmonary clearing mechanism could be concluded to be insufficient to remove the accumulated rare earth oxide from the lungs, resulting in clean-up by macrophages and consequent translocation to the tracheobronchial lymph nodes. The adverse effects that have been observed in the exposed animals were loco-regional and may be related to lung overload considering the high level of exposure (30.0 mg/m3) used in this study. No clear systemic effects have been observed in this study. An increased susceptibility to / mortality from pneumonia was reported among the exposed animals, however, pneumonia and related mortalities also occurred in the control group, indicating that the results may have been confused by bacterial contamination.


Similar observations were made by Abel and Talbot (1965, 1967, both Klimisch 3) and Talbot et al. (1965, Klimisch 3). For instance, Abel and Talbot (1965) investigated the toxicity of gadolinium oxide dust (MMAD < 0.563 µm, mean particle size 0.22 µm) after repeated exposure by inhalation in guinea pigs (6 hours/day, 5 days/week, 20.0 mg/m³, for 40-120 days). After 40 days of exposure, the most apparent change was mild swelling and proliferation of septal cells in the lung. After 80 days of exposure, the predominant finding was the presence of numerous septal and alveolar macrophages. After 120 days of exposure, thickening of the alveolar walls was seen. Lungs exhibited heavy infiltration of lymphocytic nodules. A similar experiment by Talbot et al. (1965) with mice for a period up to 120 days revealed a similar incidence of effects in lungs: after a 100-day exposure period to gadolinium oxide dust (MMAD < 1 µm, mean particle size 0.312 µm), macrophages containing dust particles were observed in the tracheobronchial lymph nodes. Accumulations of macrophages were present in multiple foci throughout the lymph node sections examined. High power microscopic examination revealed typical foamy dust laden macrophages similar in appearance to those seen in the lung parenchyma. Multinuclear giant cells containing dust particles were also seen, probably evidencing coalescence of two or more macrophages. No lesions were seen in organs other than the lungs and tracheobronchial lymph nodes. It is assumed that gadolinium was not soluble enough to be absorbed into the blood stream at amounts high enough to affect other organs histologically.


Although no reliable studies are available on the toxicity of gadolinium oxide after repeated exposure by inhalation, the available studies show a trend of loco-regional effects that may be related to lung overload conditions but absence of systemic effects that could be related to absorption of bioavailable gadolinium. Next to these studies, a reliable (Klimisch 1) acute inhalation toxicity study has recently become available. The acute effects of a 4-hour nose-only exposure of gadolinium oxide as aerosol at concentration of 5 mg/L (nominal concentration) were investigated in 6 rats (3 males and 3 females) with a 14-day post-exposure observation period (Tóth, 2016). The acute LC50 was determined to be greater than 5.04 mg/L (analytically determined concentration), as no mortality or any test item-related gross changes were observed. This result supports the assumption of very low respiratory absorption.


Based on the poor solubility of gadolinium oxide, as well as available information on the fate and toxicity of gadolinium after inhalation exposure to gadolinium oxide, the respiratory absorption factor is set at 1 % for risk assessment purposes, as a worst case. Although the inhalation route of exposure is considered less relevant for gadolinium oxalate (see above), the same absorption factor is put forward for this compound, based on similar poor water solubility as gadolinium oxide at pH levels relevant for the lung.


Dermal absorption

Studies evaluating absorption following dermal exposure in humans or animals are not available. Therefore, a qualitative assessment of the toxicokinetic behaviour based on the physicochemical properties of gadolinium oxide/oxalate is performed, taking toxicological data on these compounds, obtained after dermal exposure, into consideration.

Gadolinium oxalate/oxide would have to dissolve in the moisture of the skin prior to penetrating the skin by diffusive mechanisms. However, as the solubility of both compounds was shown to be limited at physiologically relevant pH levels (relevant to skin), no significant dermal uptake is expected.

The acute dermal toxicity of gadolinium oxalate/oxide was not investigated. This was not considered necessary based on the results of the acute oral toxicity studies (see paragraph on oral absorption, LD50 > 2000 mg/kg bw) and the absence of systemic effects observed in in vivo studies with dermal exposure (see the in vivo skin irritation and skin sensitisation studies described below), in agreement with the recently adjusted section 8.5.3, Column 2 adaptations of Annex VIII of the REACH Regulation.


Gadolinium oxide and gadolinium oxalate are not classified as skin irritant. Gadolinium oxide was tested in an in vivo acute dermal irritation study (Lambert et al., 1993, Klimisch 2). A single 24-hour, occluded application of the test item to intact or abraded skin of six rabbits produced no signs of dermal irritation over the 72-h observation period. Gadolinium oxalate was tested in a reconstructed human epidermis test to assess its potential for skin irritation (Kanizsai, 2016, Klimisch 1). Following exposure to gadolinium oxalate, the mean relative viability of three individual tissues after 15 minutes exposure to the test item and 42 hours post-incubation was 79.7% compared to the negative control value. This is above the threshold of 50%, therefore the test item was considered as being non-irritant to skin.


Further, results of a skin sensitisation study in guinea pigs (Tarcai, 2016, GPMT, OECD 406, Klimisch 1) indicated that gadolinium oxalate induced no skin sensitisation response in the guinea pig after intradermal, dermal and challenge exposures. No signs of systemic or local toxicity and no mortality were observed in this study.


As gadolinium oxalate/oxide do not meet the CLP criteria for classification as skin irritant or skin sensitiser, the expected low dermal absorption is not expected to be enhanced by any irritating/sensitising effects.


In the absence of measured data on dermal absorption, ECHA guidance (2017) suggests the assignment of either 10% or 100% default dermal absorption rates. However, the currently available scientific evidence on dermal absorption of some metals (e.g. Zn sulphate, Ni acetate; based on the experience from previous EU risk assessments) indicates that lower figures than the lowest proposed default value of 10 % could be expected (HERAG, 2007).


Based on the inherent properties of gadolinium oxalate/oxide, the toxicological data available, demonstrating the absence of systemic toxicity, and the experience from HERAG, no significant dermal absorption is expected. Therefore, a dermal absorption factor of 0.1% is suggested for risk assessment purposes as a worst case.


Distribution and accumulation

From the above discussion, absorption of gadolinium following exposure to gadolinium oxalate/oxide via the oral, inhalation, or dermal pathway is expected to be (very) limited. In the absence of accurately determined absorption factors, absorption factors for risk assessment have been set sufficiently high, as a worst-case approach. Although absorption is expected to be (very) limited, the available information on distribution and accumulation of gadolinium is discussed below, in order to describe its most likely behaviour once ending up in the circulatory system. The focus is on information obtained in studies in which gadolinium compounds such as gadolinium oxide or gadolinium trichloride (i.e. a water soluble gadolinium compound) were administered via a relevant pathway for assessment in view of REACH registration. Nevertheless, since lots of information is also available from intravenous studies with gadolinium trichloride, some relevant information from such studies is briefly summarised too. Note that only data obtained with gadolinium compounds containing stable gadolinium have been included below. Lots of data are available obtained with gadolinium radioisotopes or gadolinium compounds containing radioisotopes of gadolinium but these have not been included because absorption and distribution of these radioisotopes may be different. Finally, data available on chelated organic gadolinium complexes such as those used in medical imaging are not described either because they are designed especially for behaving differently inside the body.

Oral administration

The most relevant study yielding information on distribution and accumulation after oral exposure is the repeated dose toxicity study for gadolinium trichloride in rats (Ogawa et al., 1992), which was already mentioned above (partim oral absorption). After 28 days of exposure at the highest dose tested (1000 mg/kg bw/day), gadolinium was observed to be accumulated in liver, kidney, spleen and femur. However, no details on concentration or comparison to the administered dose are given. Note that a similar study by the same authors with yttrium trichloride, for which more details are available, indicated that the level of accumulation in the organs that were studied was very low in relation to the daily and total administered dose (Ogawa et al., 1994, see above, partim oral absorption).


Administration via inhalation/intratracheal instillation

No studies have been identified in which distribution and accumulation of gadolinium in different organs after exposure via inhalation have been extensively studied. The most relevant study in this section was also discussed already above (partim respiratory absorption) (Ball and Van Gelder, 1966). In this study, gadolinium oxide was given as an aerosol to mice for an extended period. Exposed mice were characterised by localised accumulations of metal-containing macrophages within the lungs. Metal was also found in the tracheobronchial lymph nodes, where it was concentrated in the reticuloendothelial cells of the medulla. The clean-up mechanism by macrophages and consequent transport to/accumulation in the lymph nodes associated with the lungs is indicative of probable lung overload and insufficient lung clearance. The authors did not study any other organs with respect to gadolinium levels but given the observation of clean-up of undissolved material by macrophages and the fact that no overt systemic effects were observed during the study, absorption in the circulatory system and distribution to/accumulation in other organs after respiratory exposure is expected to be limited.


Intravenous administration

Spencer et al. (1997) administered a gadolinium trichloride solution of 0.07-0.35 mmol/kg (i.e. 18.45-92.26 mg/kg) to rats via a single intravenous injection and investigated histopathological changes in all organs from high-dose and control animals necropsied 48 h post-dose and in tissues that showed treatment-related changes from all other rats either necropsied 48 h or 14 d post-dose. Relevant observations were mineral deposition in capillary beds (particularly lung and kidney) and phagocytosis of mineral by the mononuclear phagocytic system. Accumulations of phagocytic cells were observed in tissues of various organs such as kidney, liver, spleen, bone marrow. X-ray microanalysis performed on some of these accumulations showed that these intracellular deposits contained gadolinium, calcium, and phosphorous.


The observation of mineral deposition in capillary beds indicates limited stability of free gadolinium in the blood. This was already suggested by Evans (1990) in an extensive review on the metabolism of lanthanides. When entering the blood stream, gadolinium ions (as well as other lanthanide ions) form soluble complexes with proteins like albumin in the plasma. When the complexing ability of the plasma is exceeded, colloidal precipitates with anions like hydroxides, carbonates and phosphates are formed. These insoluble precipitates are then taken up by the reticuloendothelial system in the liver where they most likely localise in lysosomes. This was demonstrated by Dean et al. (1988), after intravenous injection of gadolinium trichloride in rats.


Another reference providing evidence for the mechanism described by Evans (1990) is Barnhart et al. (1987), who intravenously injected gadolinium trichloride (at a dose of 100 µmol/kg, i.e. 26.36 mg/kg) in rats. After injection, gadolinium was found primarily in liver and spleen. A rapid accumulation was described in these organs, reaching 72% of the injected dose after 2 h. The results of the study suggest that after gadolinium trichloride was injected it was subject to complexation with proteins and colloid formation. The biodistribution of gadolinium trichloride was hence mostly depending on uptake of the colloids by phagocytic cells of the reticuloendothelial system and consequent storage of the material in organs that are part of this system, such as liver and spleen. This is in line with the suggestions and findings of the other studies mentioned above.


In conclusion, based on all available information on distribution, it can be concluded that – although absorption of gadolinium after oral, dermal or inhalation exposure to gadolinium oxide/oxalate is expected to be (very) limited – any gadolinium that is taken up in the blood will complex with proteins and form colloids (e.g., with carbonates, phosphates, …) that are removed from the blood stream by phagocytic cells and temporarily stored in organs of the reticuloendothelial system such as liver and spleen.


As an element, gadolinium is neither created nor destroyed within the body. There are no indications of transformation to more hazardous forms in the liver or kidney, which is also supported by the fact that both gadolinium oxalate and gadolinium oxide were demonstrated not to be mutagenic in vitro (Varga-Kanizsai, 2016a,b, Klimisch 1; Hargitai, 2017a, Klimisch 1) and gadolinium oxalate was demonstrated not to be clastogenic in Chinese hamster cells V79 (Hargitai, 2017b, Klimisch 1), both in the presence and absence of metabolic activation. As discussed above, several studies have observed that gadolinium concentrated in granular inclusions in lysosomes of phagocytic cells (e.g., in the lungs after intratracheal instillation of gadolinium oxide, in liver and spleen after intravenous injection of gadolinium trichloride), where it may appear under different forms (e.g., complexed with proteins, inorganic ligands such as phosphates or carbonates, etc.). This can be considered as a detoxification mechanism. No other information relevant for this section has been identified.



No studies described the elimination/excretion of gadolinium following oral, dermal, or inhalation exposure.

Since there are sufficient indications for oral absorption to be (very) limited after exposure to poorly soluble compounds such as gadolinium oxide and gadolinium oxalate (see above, partim oral absorption), most of the administered/ingested dose can be expected to be eliminated via faeces, without reaching the general circulation and undergoing distribution to organs. Because no experimental evidence is available on the elimination/excretion of gadolinium after oral administration of poorly soluble/insoluble gadolinium compounds, reference can be made to experiments performed with poorly soluble/insoluble compounds of other lanthanides. For instance, following oral administration of lanthanum carbonate to rats and dogs, the great majority of the dose was found to be eliminated unabsorbed via the faeces: 99% and 93% of the dose was recovered in the faeces of rats (Damment and Pennick, 2007) and dogs (FDA (2002, 2004), respectively. The extremely small absorbed fraction was found to be excreted predominantly via the liver into bile (Pennick et al., 2006; Damment and Pennick, 2007). A similar elimination/excretion pathway can be expected in the case of gadolinium oxide and gadolinium oxalate.

Also after dermal and inhalation exposure to gadolinium oxide and gadolinium oxalate, (very) limited absorption is expected (see above, partim repiratory absorption and dermal absorption). As described above, in case exposure conditions are such that the capacity of the mucociliary clearance system is exceeded, macrophages can transport the undissolved material to the tracheobronchial lymph nodes, from where elimination can be expected to occur very slowly. Unfortunately, no experimental information could be identified on the excretion mechanism of such stored material.

Although no information could be found on gadolinium oxalate/oxide, there are quite some studies available evaluating the excretion of gadolinium after intravenous administration of gadolinium compounds in rats or mice. However, most of these studies concern the excretion of chelated organic gadolinium complexes which are specifically designed for desirable behaviour (distribution and excretion) allowing their use in e.g. medical imaging and thus, these studies are not described in this assessment. In comparison to these chelated forms, ‘free gadolinium’ such as from gadolinium acetate is mentioned to be slowly cleared (without specifying the mechanisms or pathways for further excretion) from the organs where it accumulated (liver, lungs, spleen) after intravenous administration in mice (Wedeking and Tweedle, 1988). This could be expected based on the findings described above on the clean-up of colloidal precipitates of gadolinium from blood by the reticuloendothelial phagocytic clearance system and consequent storage in organs such as the liver and spleen.

Because so little information is available on gadolinium, reference could be made to studies performed with other lanthanides, informing on excretion after intravenous administration. For instance, after a single intravenous 0.3 mg/kg dose of lanthanum trichloride in rats, recovery of lanthanum over 42 days was 76.4 ± 5.7% of the administered dose, predominantly in the faeces (Damment and Pennick, 2007). In contrast, only 1.94 ± 0.24% of the dose was excreted in urine over the 42-day collection period. The data imply that the kidneys are not significantly involved in the clearance of absorbed lanthanum. Because of the similar behaviour of lanthanides, it can be assumed that urinary excretion is of low importance for excretion of gadolinium too.


Although there is not much experimental information informing on elimination/excretion mechanisms of gadolinium, the limited available information as well as information available for other lanthanides, allows drawing the conclusion that elimination via the faeces is by far the most important route of elimination/excretion after oral administration. Limited amounts of gadolinium that might be absorbed can be expected to end up as depositions in lysosomes of organs/tissues that are part of the reticuloendothelial system, from which slow clearance is expected. No experimental information is however available on the exact mechanisms for excretion of such accumulated material, apart from the potential role of biliary excretion for material stored in the liver. For clearance of depositions from other organs/tissues, slow urinary excretion might be involved, but this is not experimentally confirmed at this point in time.



Abel M, Talbot RB, 1965. In Vitro Pulmonary Compliance and Physiopathology of Guinea Pig Lungs Exposed In Vivo to Gadolinium Oxide Dust. Biology and Medicine (UC-48) TID-4500, COO-1170.


Abel M, Talbot RB, 1967. Gadolinium oxide inhalation by guinea pigs: A correlative

functional and histopathologic study. J Pharmacol Exper Thera 157: 207-213.


Ball BA, Van Gelder G, 1966. Chronic toxicity of gadolinium oxide for mice following exposure by inhalation. Archives of Environmental Health 13, 601-608.


Barnhart JL, Kuhnert N, Bakan DA, Berk RN, 1987. Biodistribution of GdCl3 and Gd-DTPA and their influence on proton magnetic relaxation in rat tissue. Magn Reson Imaging 5:221-231.


Beckett WS, Clarkson TW, Nordberg G, 2007. Routes of exposure, dose and metabolism of metals. Chapter 3 in Handbook on the toxicology of metals (3rd Edition), p. 39-64.


Bruce DW, Hietbrink BE, DuBois KP, 1963. The Acute Mammalian Toxicity of Rare Earth Nitrates and oxides. Toxicology and Applied Pharmacology 5: 750-759.


Clouzeau, 1994. Acute Oral Toxicity in Rats - Oxyde de Gadolinium 99.99.


Damment SJ, Pennick M, 2007. Systemic lanthanum is excreted in the bile of rats. Toxicology Letters 171 : 69-77.


Dean PB, Niemi P, Kivisaari L, Kormano M, 1988. Comparative pharmacokinetics of gadolinium DTPA and gadolinium chloride. Invest. Radiol. 23 (Suppl.1): S258–S260.


Demangel B, 2016a. Melting point test by DSC method on digadolinium trioxide [12064-62-9]


Demangel B, 2016b. Melting point test by DSC method on Tris[oxalato(2-)]digadolinium [867-64-1]


ECHA Guidance on Information Requirements and Chemical Safety Assessment Chapter R.7c: Endpoint specific guidance, Version 3.0, November 2017.


ECHA Guidance on Information Requirements and Chemical Safety Assessment Chapter R.6: Guidance on QSAR and grouping of chemicals (May 2008).


Evans C. Biochemistry of the Lanthanides. Plenum Press: New York, 1990. 303–315.


Fallingborg J, 1999. Intraluminal pH of the human gastrointestinal tract. Dan Med Bull 46:183-196.


FDA (2002, 2004). Center of Drug Evaluation and Research Package for Application No. 21-468 Fosrenol® (Shire Pharmaceuticals).


Hargitai J, 2017a. Digadolinium oxalate: In Vitro Mammalian Cell Gene Mutation Test (HPRT Assay).


Hargitai J, 2017b. Digadolinium oxalate: In vitro Mammalian Chromosome Aberration Test.


Health risk assessment guidance for metals (HERAG) fact sheet (2007). Assessment of occupational dermal exposure and dermal absorption for metals and inorganic metal compounds. EBRC Consulting GmbH (2007).


Jean-Baptiste C, 2017a. Solubility in water of one batch of digadolinium trioxide (CAS No. 12064-62-9).


Jean-Baptiste C, 2017b. Solubility in water of one batch of gadolinium oxalate.


Kanizsai B, 2016. Digadolinium oxalate: In Vitro Skin Irritation Test in the EPISKIN(TM) (SM) Model.


Konoha K, Sadakane Y, Kawahara M, 2004. Effects of gadolinium and other metal on the neurotoxicity of immortalized hypothalamic neurons induced by zinc. Biomed Res Trace Elements 15: 275-277.


Lambert CE, Barnum EC, Shapiro R, 1993. Acute toxicological evaluation of gadolinium oxide.


Mátyás A, 2016. Digadolinium oxalate – Acute Oral Toxicity Study in Rats.


Mogilevskaya OY, Roshchina PA, 1964. Toxicity of the oxides of Rare-Earth elements gadolinium and dysprosium. Labor Protection and Safety Engineering in Power Metallurgy 3: 256-258.


Ogawa Y, Suzuki S, Naito K, Saito M, Hirose A, Kaneko T, Kurokawa Y, 1992. A 28-day repeated dose toxicity study test of Europium and Gadolinium in Wistar Rat in comparison with Lanthanum and Yttrium. Division of Toxicology, Biological Safety Research Center, National Institute of Hygienic Sciences.


Ogawa Y, Suzuki S, Naito K, Saito M, Kamata E, Hirose A, Kaneko T, Chiba M, Inaba Y, Kurokawa Y, 1994. Studies on the biological effects of yttrium. Effects of repeated oral administration in rats. Jpn. J. Toxicol. Environ. Health 40: 374-382.


Palasz A, Czekaj P, 2000. Toxicological and cytophysiological aspects of lanthanides action. Acta Biochimica Polonica 47: 1107-1114.


Papineau, 2017. Digadolinium Trioxide: Combined Repeated Dose Toxicity Study with the Reproduction/Developmental Toxicity Screening Test by Oral Route (Gavage) in Rats.


Pennick M, Dennis K, Damment SJ, 2006. Absolute bioavailability and disposition of lanthanum in healthy human subjects administered lanthanum carbonate. Journal of Clinical Pharmacology 46: 738-746.


Shapiro, 1990. Acute oral limit test – Gadolinium oxide.


Spencer AJ, Wilson SA, Batchelor J, Reid A, Rees J, Harpur E, 1997. Gadolinium chloride toxicity in the rat. Toxicol Pathol 25: 245-255.


Talbot RB, Davison FC, Reece WO, 1965. Inhalation Exposure of Mice to an Aerosol of Gadolinium Oxide. Biology and Medicine (UC-48) TID-4500, COO-1170-6, 20 pp.


Tarcai Z, 2016. Digadolinium oxalate – A Skin Sensitisation Study in the Guinea Pig using the Magnusson and Kligman Method.


Tóth B, 2016. Digadolinium Trioxide: Acute Inhalation Toxicity Study (Nose-Only) in the Rat.


Varga-Kanizsai B, 2016a. Digadolinium trioxalate: Bacterial Reverse Mutation Assay.


Varga-Kanizsai B, 2016b. Digadolinium trioxide: Bacterial Reverse Mutation Assay.


Wedeking P, Tweedle M, 1988. Comparison of the biodistribution of 153Gd-labeled Gd(DTPA)2-, Gd(DOTA)-, and Gd(acetate) in mice.Int J Rad Appl Instrum B 15: 395-402.


Zhang Y, ter Keurs HE, 1996. Effects of gadolinium on twitch force and triggered propagated contractions in rat cardiac trabeculae. Cardiovasc Res 32: 180-188.