ARTICLE IN PRESS
Experimental and Toxicologic Pathology 58 (2007) 323–330
EXPERIMENTAL ANDTOXICOLOGIC PATHOLOGY www.elsevier.de/etp
Toxicological assessment of gadolinium release from contrast media Simona Bussia,, Xavier Fouilletb, Alberto Morisettia a
Centro Ricerche Milano, Bracco Imaging SpA, Milan, Italy Bracco Research S.A., Geneva, Switzerland
b
Received 22 February 2006; accepted 20 September 2006
Abstract In vivo gadolinium release was evaluated for MultiHances, Omniscans and Gadovists estimating gadolinium content in liver, kidneys, spleen, femur and brain after single or repeated intravenous administrations to rats at 1 mmol/kg. Gadolinium acetate (GdAc) at a daily dose of 0.03 mmol/kg and physiological saline were used as positive and negative controls, respectively. No changes in blood chemistry, haematology nor histopathology were seen with any of the tested contrast media, whereas an increase in white blood cell count and in serum cholesterol were found after GdAc at 0.18 mmol/kg cumulative dose. Analogously, gadolinium content in target organs (as % of injected dose) after any of contrast media was 100–200 times lower than after GdAc, either after single or repeated administrations. Under these experimental conditions, the rank of residual gadolinium found in these organs was GdAcbOmniscans4Gadovists4MultiHances. Depopulation of lymphocytes in periarteriolar lymphatic sheaths (PALS) areas of the spleen was noted in rats treated with a single dose of GdAc sacrificed 24 h post-dosing, but not in repeated dose rats sacrificed 48 h after last dosing. It was, therefore, concluded that this was a transient phenomenon and that PALS are rapidly repopulated with lymphocytes. With all contrast media, gadolinium content after a 2-day washout following a 3-week repeated administration period was lower than the amount found 24 h after a single administration. Accordingly, the observed gadolinium content in organs should actually be in a complexed form (possibly the injected complex) which is subjected to elimination. r 2006 Elsevier GmbH. All rights reserved. Keywords: Gadolinium release; ICP; Gadolinium acetate; Intravenous; Rat
Introduction Clinical diagnostics has dramatically evolved in the last few decades, taking advantages from magnetic resonance imaging (MRI). Paramagnetic contrast media Corresponding author. Tel.: +39 02 21772454; fax: +39 02 21772794. E-mail address:
[email protected] (S. Bussi).
0940-2993/$ - see front matter r 2006 Elsevier GmbH. All rights reserved. doi:10.1016/j.etp.2006.09.003
(mainly gadolinium complexes) are injected intravenously to enhance the signal from water protons, thus allowing recognition of tissues with different water content. Molecules constituting contrast media were shown to be remarkably safe in clinical use, but they still retain a degree of toxicity, which is thought to be responsible of adverse reactions that are observed in a small percentage of patients (Runge et al., 1991; Niendorf et al., 1994; Gemery et al., 1998; Terzi and
ARTICLE IN PRESS 324
S. Bussi et al. / Experimental and Toxicologic Pathology 58 (2007) 323–330
Sokmen, 1999; Murphy et al., 1999). One of the mechanisms which has been proposed to explain the toxicity of paramagnetic gadolinium complexes used as contrast agents for MRI is the in vivo dissociation and/ or metabolism of the complex itself to yield metal and free ligands (Cacheris et al., 1990). Both free ligands and metal were shown to possess a certain degree of acute toxicity (in terms of LD50) at least 20 times higher than that of their respective complexes, the free ligand acting as a chelator of physiological positive ions (Ca2+, Zn2+), the metal by taking place of those ions in body tissues through a transmetallation process (Evans, 1990). However, while the newly formed complexes are excreted by the urinary system, the metal typically accumulates in tissues, thus being a suitable indicator of complex dissociation. Data from literature show that intravenous administration of GdCl3 to rats results in gadolinium deposition in liver, spleen, kidneys and bone marrow (Kasokat and Urich, 1992). The use of GdCl3 given intravenously as a positive control, may, however, lead to intravascular flocculation by interactions with the reticuloendothelial cells (RES) (Spencer et al., 1997), thus leading to artifacts that makes GdCl3 unsuitable for comparison with water-soluble paramagnetic contrast agents as far as the toxicological effects and deposition kinetics is regarded. A better model of transmetallation would be therefore represented by a gadolinium salt such as gadolinium acetate (GdAc), which would simulate more closely the gadolinium release from paramagnetic complexes (Wedeking and Tweedle, 1988). The aim of this study was to compare the in vivo stability of three gadolinium complexes used for MRI, MultiHances, Omniscans and Gadovists by estimating the gadolinium content in organs thought to be major sites of deposition of free gadolinium (liver, kidneys, spleen, femur and brain).
Materials and methods Compounds: Gadobenate dimeglumine (MultiHances), a sterile and apyrogenic solution without any excipients (de Hae¨n et al., 1999), produced by Bracco S.p.A., Milan, Italy; gadodiamide (Gd-DTPABMA, Omniscans, Nycomed Amersham plc, Norway) and gadobutrol (Gadovists, Schering AG, Germany), all formulated as 0.5 M aqueous solution, were obtained as commercial products. GdAc, 0.015 M solution was prepared in house. Unless otherwise indicated, the generic names of contrast agents refer to their European pharmaceutical formulations for injection. Omniscans and Gadovists formulations contain excess chelator, while in the approved MultiHances
formulation there is no excess chelator (Kirchin and Runge, 2003). Animals: All the procedures involving animals were conducted according to the national and international laws on animal experimentation (L.D. 116/92; C.D. EEC 86/609). No validated non-animal alternatives were known to meet the objectives of this study. Eight, 6–7-week old, Sprague–Dawley Crl:CD(SD)IGS BR male rats per product and per sacrifice time were purchased from Charles River Italia, Calco (Lc, Italy). Animals were randomised on the basis of body weight 3 days after arrival. An acclimation period of at least 3 days was allowed between allocation of animals to groups and beginning of treatment. During this period, a veterinary review of animal health was undertaken by a veterinary officer. Test and control articles were injected using a Harvard infusion pump at an injection rate of 2 mL/min via a lateral tail vein. There were five groups of rats (Saline, GdAc, Multihances, Omniscans and Gadovists) receiving a single injection and sacrificed 24 h after this injection, and five groups of rats receiving administrations twice a week for three consecutive weeks and sacrificed 48 h after the last administration. Animals were weighed on the day of allocation to treatment groups, once during each week of treatment and at sacrifice. The day before scheduled sacrifice the animals were fasted overnight (water ad libitum) being housed in metabolic cages for 16–20 h in order to obtain urine samples. On the day of sacrifice, the animals were anaesthetised by administering intravenously a mixture of Ketavet and Rompuns (ketamine chlorhydrate 100 mg/mL, Farmaceutici Gellini, Aprilia, Latina, Italy; xylazine 2% w/v, Bayer AG, Leverkusen, Germany). Haematology and plasma chemistry: Red cell count (RBC), white cell count (WBC), haemoglobin (Hb), packed cell volume (PCV), mean corpuscular volume (MCV), platelet count (PLT), mean platelet volume (MPV), mean corpuscular haemoglobin (MCH) and mean corpuscular haemoglobin concentration (MCHC) were measured in blood using EDTA as anticoagulant. Plasma samples prepared with sodium citrate as anticoagulant were analysed for prothrombin time (PT) and activated partial thromboplastin time (aPTT). Plasma samples prepared with sodium heparin as anticoagulant were analysed for glutamic–oxaloacetic transaminase (GOT or AST), glutamic–pyruvic transaminase (GPT or ALT), alkaline phosphatase (AP), glucose (GLUC), total cholesterol (CHOL), triglycerides (TRI), total protein (TPRT), (UREA), creatinine (CREAT), bilirubin (BIL), albumin (ALB), sodium (Na), potassium (K), calcium (Ca) and phosphate (Phos). Urinalysis: The volume of all urine samples was measured and specific gravity, pH, leucocytes, protein, glucose, ketones, urobilinogen, bilirubin and blood were determined using Combur10Test strips. The deposit
ARTICLE IN PRESS S. Bussi et al. / Experimental and Toxicologic Pathology 58 (2007) 323–330
obtained after centrifugation (172g for 15 min, at room temperature) was examined microscopically for the presence of epithelial cells (E), leucocytes (L), erythrocytes (R) and casts (C). Pathology: Complete macroscopic post-mortem examinations were performed on all rats. Liver, kidneys, spleen, right femur and brain were carefully dissected and properly trimmed to remove fat and other contiguous tissues in a uniform manner. For the first six animals of each group, these tissues were weighed and processed for gadolinium content analysis. The tissues from the remaining two animals of each group were fixed in 10% neutral-buffered formalin and then embedded in paraffin wax, sectioned at a nominal thickness of 4–5 mm, and stained with the haematoxylin–phloxin–saffron (HPS) trichromic stain for all organs plus the periodic acid Schiff (PAS) stain for the kidney. Assay of gadolinium in biological samples: All samples (liver, brain, kidneys, spleen and femur) were analysed by inductively coupled plasma-atomic emission spectrometry (ICP-AES) using a Jobin-Yvon Mod 24 spectrometer. A different sample preparation for each organ was adopted. Liver and brain were dried firstly by means of a freeze drying process (indicatively with a shelf temperature ranging from 40 to +30 1C and a chamber pressure up to 0.2 mbar). Liver was then homogenised by grinding in a mortar. Liver solutions were prepared by suspending 200 mg of tissue powder, which were accurately weighed, in 1.5 mL of nitric acid (65% v/v). Dried brains, accurately weighed, were suspended in 1.5 mL of nitric acid (65% v/v). Spleen solutions were prepared by suspending the fresh spleen, accurately weighed, in 1.5 mL of nitric acid (65% v/v). Kidney solutions were prepared by suspending each kidney, accurately weighed, in 1.5 mL of nitric acid (65% v/v). Right femurs were prepared by suspending the femur, accurately weighed, in 1.5 mL of nitric acid (65% v/v). The destruction of the organic matrix was
Table 1.
325
performed by subjecting these samples to a wet ashing process with a microwave oven system. Statistical analysis: Statistics was performed to test differences among negative control (saline), contrast media and positive control (GdAc) groups. One-way ANOVA was performed on all experimental groups. If significant value of F was found (Po0:05), Dunnett’s test was employed to test significance of each group vs. negative control. The statistical tests were performed by using inerSTAT-a v1.3, an MSExcel-based statistical software (Vargas, 1999).
Results Clinical observation No clinical signs nor changes in body weight were noted in any treated animal.
Haematology (Tables 1 and 2) No relevant changes were observed after both single or repeated administrations for all the tested compounds. A significant increase in Hb concentration was noted after a single administration in animals treated with Omniscans and GdAc, while an increase in PLT was noted at the end of the repeat administration treatment with Omniscans, (all Po0:05) but these changes are considered to have no biological relevance in absence of other concomitant signs of haemotoxicity.
Blood chemistry (Tables 3 and 4) After the single administration, significant increases (Po0:01) in GOT, GPT and triglyceride levels were noted with GdAc, in agreement with results from
Haematology: single treatment
Compound/dose
RBC (106/mL)
Hb (g/dL)
PCV (%)
MCV (mm3)
MCH (pg)
MCHC (g/dL)
PLT (103/mL)
MPV (mm3)
WBC (103/mL)
PT (s)
aPTT (s)
Physiological saline 2 mL/kg MultiHances 1 mmol/kg Omniscans 1 mmol/kg Gadovists 1 mmol/kg GdAc 0.03 mmol/kg
6.74 (0.34) 6.92 (0.21) 7.12 (0.39) 6.91 (0.31) 7.08 (0.36)
14.3 (0.63) 14.90 (0.55) 15.04* (0.54) 14.84 (0.53) 15.03* (0.61)
37.2 (2.2) 38.7 (2.6) 38.7 (2.1) 38.1 (1.8) 38.1 (1.6)
55.13 (1.81) 56.0 (3.2) 54.4 (1.4) 55.25 (0.71) 53.9 (1.1)
21.15 (0.33) 21.48 (0.63) 21.11 (0.62) 21.44 (0.45) 21.20 (0.52)
38.5 (1.2) 38.5 (1.5) 38.90 (0.94) 38.93 (0.70) 39.43 (0.52)
851 (61) 932 (85) 933 (76) 876.0 (54) 770 (105)
7.10 (0.35) 7.0 (0.0) 6.90 (0.35) 7.0 (0.0) 7.00 (0.53)
10.5 (3.2) 8.9 (2.0) 11.4 (2.2) 10.5 (2.7) 10.2 (2.2)
9.66 (0.42) 9.37 (0.23) 9.80 (0.44) 10.01 (0.41) 10.13 (0.86)
18.2 (1.6) 17.3 (1.1) 17.7 (1.3) 17.1 (1.1) 17.0 (1.1)
*Po0:05. Mean values (standard deviation), n ¼ 8.
ARTICLE IN PRESS 326
Table 2.
S. Bussi et al. / Experimental and Toxicologic Pathology 58 (2007) 323–330
Haematology: repeated treatment
Compound/ dosage
Hb (g/ RBC (106/mL) dL)
PCV (%)
MCV (mm3)
MCH (pg)
MCHC PLT MPV (g/dL) (103/mL) (mm3)
WBC (103/mL)
PT (s)
aPTT (s)
Physiological saline 2 mL/kg/ day MultiHances 1 mmol/kg/day Omniscans 1 mmol/kg/day Gadovists 1 mmol/kg/day GdAc 0.03 mmol/kg/ day
7.31 (0.34)
15.41 (0.68)
38.1 (1.9)
52.0 (1.2)
21.04 (0.41)
40.38 (0.54)
795 (58)
7.0 (0.0)
11.0 (1.8)
9.85 (0.78)
21.2 (1.7)
7.35 (0.15) 7.32 (0.32) 7.36 (0.23) 7.52 (0.54)
15.40 (0.34) 15.16 (0.60) 15.58 (0.26) 15.31 (0.83)
38.0 (1.2) 37.1 (1.4) 38.33 (0.58) 37.6 (2.5)
51.6 (1.1) 50.6 (1.1) 52.1 (2.2) 50.1 (1.6)
20.91 (0.39) 20.66 (0.33) 21.14 (0.83) 20.35 (0.81)
40.45 (0.44) 40.79 (0.56) 40.59 (0.40) 40.68 (0.73)
811 (31) 876* (55) 825 (89) 852 (56)
7.13 (0.35) 7.13 (0.35) 7.13 (0.35) 7.13 (0.35)
11.0 (1.5) 13.8 (1.8) 11.3 (1.2) 13.2 (5.8)
9.88 (0.71) 10.43 (0.65) 10.40 (0.39) 10.21 (0.76)
19.1 (1.8) 20.1 (1.6) 21.1 (2.0) 19.8 (1.2)
*Po0:05. Mean values (standard deviation), n ¼ 8:
literature (Spencer et al., 1997). Analogously, a significant increase in GOT (Po0:01) and cholesterol levels was noted after repeated treatments with GdAc. Serum calcium levels showed erratic, though statistically significant, variations after single (reduction) or repeated treatments (reduction for MultiHances and positive controls, increase for Omniscans and Gadovists). These changes, however, were not considered to be related to the treatment based on the absence of parallel reduction in albumin plasma levels (Hall, 2001).
Urinalysis No significant changes in urinary parameters were found either after single or repeated treatment with any administered compounds.
Pathology (Figs. 1–5) No changes in gross pathology and organ weights were observed for animals treated with any of the tested compounds. The only microscopic abnormal finding was a depopulation of lymphocytes in the periarteriolar lymphatic sheaths (PALS) of the white pulp of the spleen in two out of eight rats treated with a single dose of GdAc and sacrificed 24 h post-dosing (Fig. 3). Such finding was not observed in rats after a single or repeated dosing with the contrast agents (Figs. 1 and 2) or after the repeated dose treatment with GdAc (animals sacrificed 48 h after the last dosing), demonstrating that it is a transient phenomenon and that the PALS are rapidly repopulated with lymphocytes. No significant differences were observed histopathologically in the liver, spleen, kidneys, brain and femur among the rats treated with MultiHances, Omniscans
and Gadovists and the corresponding controls treated with saline under the same experimental conditions. In particular, there were no abnormal findings in the kidney after the repeated administrations with the MRI contrast agents (Fig. 5). Assay of gadolinium in biological samples (Tables 5 and 6) Gadolinium content in liver, kidneys, brain, spleen and femur was calculated as per cent of injected dose, both after single administration (contrast media: 1 mmol/kg, positive control GdAc: 0.03 mmol/kg) and repeated administrations (contrast media: 1 mmol/kg/ day, cumulative dose 6 mmol/kg, positive control GdAc: 0.03 mmol/kg/day, cumulative dose 0.18 mmol/kg). Since in all biological samples of control animals the gadolinium content was found equal to zero, the statistical evaluation was made matching MultiHances values versus comparators and positive control (GdAc). Single administration: Target organs for gadolinium deposition were confirmed to be liver, spleen and bone, as demonstrated by their respective gadolinium content after the injection of the positive control (58%, 6%, 0.3% of the injected dose, for these organs, respectively). In comparison, the values for all tested contrast media were smaller by one or two orders of magnitude, without significant differences among contrast media, in liver, brain, spleen and bone. However, the kidneys showed a significantly higher (Po0:01) gadolinium content after Omniscans and Gadovists than after MultiHances and GdAc administration. This can be explained, at least in part, by the dual (urinary and biliary) route of elimination for MultiHances and by the selective localisation of the free metal in target organs for GdAc. Repeated administrations: The results obtained after single administration were confirmed in this part of the
1.57 (0.23) 1.74 (0.29) 1.75 (0.20) 1.59 (0.19) 1.39 (0.29)
50.5 (1.6)
51.5 (2.3) 48.1 (6.4) 50.8 (1.9) 49.9 (5.1)
8.0 (1.3)
7.9 (1.1) 8.5 (1.2) 8.1 (1.2) 7.77 (0.90)
Physiological saline 2 mL/kg/ day MultiHances 1 mmol/kg/day Omniscans 1 mmol/kg/day Gadovists 1 mmol/kg/day GdAc 0.03 mmol/kg/ day
*Po0:05. **Po0:01. Mean values (standard deviation), n ¼ 8:
26.1 (1.4) 25.0 (1.6) 26.91* (0.75) 24.6 (1.7)
3.67 (0.91) 3.75 (0.93) 4.11 (0.91) 3.65 (0.62) 3.72 (0.59)
31.9 (3.5) 29.8 (4.3) 28.7 (8.6) 30.5 (4.0) 34.2 (4.6)
20.6 (1.5) 26.0 (6.2) 22.9 (3.1) 24.3 (5.4) 31* (11)
1.27 (0.24) 1.45 (0.16) 1.70 (0.39) 2.70** (0.55)
0.48 (0.19) 0.51 (0.26) 0.34 (0.18) 0.39 (0.14)
0.43 (0.12) 3.60 (0.66) 4.14 (0.43) 4.16 (0.51) 4.11 (0.46)
3.84 (0.54) 41.0 (4.6) 43.7 (6.2) 39.7 (5.3) 42.7 (4.1)
37.8 (3.9) 19.9 (2.3) 21.6 (8.1) 19.9 (5.2) 25.4 (3.4)
20.9 (3.8)
CHOL TRI UREA CREAT GPT (mmol/L) (mmol/L) (mmol/L) (mmol/L) (U.I./L)
25.16 1.41 (0.82) (0.15)
GLUC TPRT ALB (mmol/L) (g/L) (g/L)
0.410 (0.079) 0.46 (0.11) 0.50 (0.11) 0.475 (0.084) 0.62* (0.17)
CHOL TRI UREA CREAT GPT (mmol/L) (mmol/L) (mmol/L) (mmol/L) (U.I./L)
Blood chemistry: repeated treatment
Compound/ dosage
Table 4.
*Po0:01. Mean values (standard deviation), n ¼ 8:
25.00 (0.86) 26.5 (2.6) 26.7 (1.7) 26.1 (1.0) 26.1 (1.1)
7.9 (1.8) 7.5 (1.8) 8.37 (0.98) 7.5 (1.2) 7.3 (1.0)
Physiological saline 2 mL/kg MultiHances 1 mmol/kg Omniscans 1 mmol/kg Gadovists 1 mmol/kg GdAc 0.03 mmol/kg
50.6 (1.4) 52.2 (3.8) 53.4 (3.4) 50.4 (1.8) 51.4 (2.2)
GLUC TPRT ALB (mmol/L) (g/L) (g/L)
Blood chemistry: single treatment
Compound/ dose
Table 3.
69.6 (8.1) 76 (18) 66.3 (8.6) 94** (12)
70.6 (8.1)
GOT (U.I./L)
66.9 (6.5) 79 (14) 74 (13) 75.3 (6.9) 123* (28)
GOT (U.I./L) 370 (88) 440 (93) 405 (102) 449 (70) 415 (78)
142.5 (8.4) 159 (39) 143.0 (1.8) 144.4 (1.2) 142.9 (8.5)
3.19 (0.18) 3.24 (0.85) 3.19 (0.22) 3.59 (0.63) 3.29 (0.23)
2.536 (0.041) 2.480* (0.082) 2.499* (0.057) 2.490* (0.039) 2.413* (0.087)
8.07 (0.42) 8.19 (0.97) 7.91 (0.29) 8.48 (0.85) 7.82 (0.39)
3.10 (0.61) 2.76 (0.85) 2.43 (0.77) 2.65 (0.58)
3.01 (0.40)
362 (103) 292 (61) 294 (48) 279 (43)
315 (48)
150.00 (0.61) 149.30 (1.0) 149.95 (0.81) 149.90 (2.3)
149.5 (1.5)
3.22 (0.17) 3.46 (0.28) 3.41 (0.11) 3.44 (0.20)
3.38 (0.18)
2.584** (0.071) 2.75** (0.19) 2.69** (0.20) 2.441** (0.064)
2.608 (0.063)
7.98 (0.36) 8.41 (0.50) 7.92 (0.23) 7.67 (0.63)
7.85 (0.39)
BIL AP (U.I./ Na K (mmol/ Ca Phos (mmol/L) L) (mmol/L) L) (mmol/L) (mmol/L)
1.6 (1.4) 1.80 (0.19) 2.09 (0.26) 1.86 (0.63) 1.83 (0.66)
BIL AP (U.I./ Na K Ca Phos (mmol/L) L) (mmol/L) (mmol/L) (mmol/L) (mmol/L)
ARTICLE IN PRESS
S. Bussi et al. / Experimental and Toxicologic Pathology 58 (2007) 323–330 327
ARTICLE IN PRESS 328
S. Bussi et al. / Experimental and Toxicologic Pathology 58 (2007) 323–330
Fig. 1. Spleen: normal aspect in a rat sacrificed 24 h postdosing with saline. The bar represents 50 mm. HPS stain.
Fig. 2. Spleen: normal aspect in a rat sacrificed 24 h postdosing with 1 mmol/kg Multihances. The bar represents 50 mm. HPS stain.
Fig. 3. Spleen: note cellular losses in periarteriolar lymphocyte sheats of spleen in a rat sacrificed 24 h post-dosing with 0.03 mmol/kg GdAc. The bar represents 50 mm. HPS stain.
Fig. 4. Kidney: detail from a rat sacrificed 48 h after 3-week repeat treatment with saline. The bar represents 50 mm. HPS stain.
Fig. 5. Kidney: no abnormal finding in a rat sacrificed 48 h after 3-week repeat treatment with Multihances. The bar represents 50 mm. HPS stain.
study. Moreover, since organs were analysed 48 h after the last administration, a more robust assessment of the gadolinium deposition following a suitable washout period was allowed. After GdAc, the gadolinium content in liver, kidneys and femur, expressed as per cent of the cumulative dose, was in the same order of magnitude as after single injection, thus indicating a proportional, additive deposition. After repeat administration, spleen showed a per cent of gadolinium almost double than the value noted after single administration, thus indicating that this organ is more prone to accumulate free gadolinium than the other organs. With all contrast media, gadolinium values were one or two orders of magnitude lower than that observed after GdAc administrations. In kidneys, gadolinium content still demonstrated the same pattern of significance as
ARTICLE IN PRESS S. Bussi et al. / Experimental and Toxicologic Pathology 58 (2007) 323–330
Table 5.
329
Gadolinium levels: single treatment
Compound/dose
Liver
Brain
Spleen
Kidneys
Femur
MultiHances 1 mmol/kg
0.0582 (0.0049) 0.165 (0.051) 0.076 (0.015) 58.4* (6.4)
0.000232 (0.000035) 0.000344 (0.000041) 0.000147 (0.000048) 0.0067* (0.0020)
0.00277 (0.00054) 0.00541 (0.00098) 0.00378 (0.00060) 5.6* (1.8)
0.254 (0.016) 0.448* (0.058) 0.402* (0.092) 0.288 (0.059)
0.00151 (0.00018) 0.0048 (0.0011) 0.00139 (0.00024) 0.262* (0.017)
Omniscans 1 mmol/kg Gadovists 1 mmol/kg GdAc 0.03 mmol/kg
*Po0:01. Mean values % injected dose/organ (standard deviation), n ¼ 6:
Table 6.
Gadolinium levels: repeated treatment
Compound/dosage
Liver
Brain
Spleen
Kidneys
Femur
MultiHances 1 mmol/kg/day
0.0159 (0.0017) 0.0495 (0.034) 0.0249 (0.0028) 58.3* (6.2)
0.0000979 (0.0000070) 0.000191 (0.000021) 0.0000611 (0.0000085) 0.00162* (0.00061)
0.000793 (0.000075) 0.00202 (0.00026) 0.00161 (0.00046) 11.7* (2.4)
0.141 (0.029) 0.283* (0.047) 0.235* (0.062) 0.203 (0.019)
0.00099 (0.00016) 0.00381 (0.00063) 0.00055 (0.00012) 0.224* (0.044)
Omniscans 1 mmol/kg/day Gadovists 1 mmol/kg/day GdAc 0.03 mmol/kg/day
*Po0:01. Mean values % injected dose/organ (standard deviation), n ¼ 6:
that observed after single injection, i.e. a greater residual amount of gadolinium after Omniscans and Gadovists than with MultiHances or GdAc.
Conclusion This study was aimed to assess gadolinium release in rats after single or repeated administrations of paramagnetic contrast media for MRI, and to evaluate possible pathological changes caused by such a release. Contrast media used in this study are well known by their pre-clinical and clinical profiles (Harpur et al., 1993; Morisetti et al., 1999; Tombach and Heindel, 2002). They are thermodynamically stable under physiological conditions, they do not undergo metabolism in animals or man, they are excreted by more than 90% via the renal route (Lorusso et al., 2005). An exception to the latter pattern is represented by MultiHances, which shows a unique hepatic tropism leading to about 50% biliary excretion in rat, whereas it accounts to about 5% in man (de Hae¨n et al., 1995; de Hae¨n et al., 1996). This intracellular penetration of a molecule bearing a potentially dangerous element calls for even more in-depth safety investigations. This study confirms that gadolinium release is unlikely to occur after single injection of paramagnetic contrast media, even at doses 10-fold higher than those
recommended for diagnostic purposes (Kirchin and Runge, 2003). Although the analytical method here employed (ICP) could not discriminate between bound and unbound gadolinium (i.e. the complex or the released gadolinium), the kinetics of deposition after administration of GdAc clearly indicates that liver, bone and spleen are the target organs for free gadolinium to take place of cations (typically Ca2+, Mg2+, Zn2+) in these tissues (Cacheris et al., 1990). None of the contrast media used in this study induced a level of gadolinium higher than 0.5% of the injected dose in any organs after a single administration or 3-week repeated administrations twice a week. In the context of these small quantities, the highest gadolinium content was noted in the kidneys, in relation to the predominant renal excretion of the contrast media. Although for MultiHances a dual route of elimination has been demonstrated, the short half lives (approximately 10 min) of these contrast media render them indistinguishable at the 24 h examination time. According to the model provided by GdAc, uncomplexed gadolinium should not be found in kidneys, but rather in spleen, bone or liver. It can, therefore, be assumed that the amount of gadolinium detected in the kidneys is related to the unmetabolised complex. The overall gadolinium content in target organs after single or repeat dose administration was found as
ARTICLE IN PRESS 330
S. Bussi et al. / Experimental and Toxicologic Pathology 58 (2007) 323–330
GdAc b Omniscans4Gadovists4MultiHances for the different contrast media. No evidence of biologically significant changes in laboratory investigations, either after single or repeated administrations was observed. No histopathological changes was observed in any animals, except a transient lymphocytic depopulation of the PALS in the spleen white pulp after GdAc administration. Based on the animal data obtained at doses 10-fold as high as the diagnostic dose, the risk of toxic effects by gadolinium release from paramagnetic complexes should be considered as low.
References Cacheris WP, Quay SC, Rocklage SM. The relationship between thermodynamics and the toxicity of gadolinium complexes. Magn Reson Imag 1990;8:467–81. de Hae¨n C, Lorusso V, Tirone P. Hepatic transport of the magnetic resonance imaging contrast agent gadobenate dimeglumine in the rat. Acta Radiol 1995;2:232–8. de Hae¨n C, Lorusso V, Tirone P. Hepatic transport in TR rats of the MRI contrast agent gadobenate dimeglumine. Acta Radiol 1996;3:S452–4. de Hae¨n C, Cabrini M, Aknana L, Ratti D, Calabi L, Gozzini L. Gadobenate dimeglumine 0.5 M solution for injection (MultiHances): pharmaceutical formulation and physicochemical properties of a new magnetic resonance imaging contrast medium. J Comput Assist Tomogr 1999;23(Suppl 1): S161–8. Evans CH. Biochemistry of the lanthanides. New York: Plenum Press; 1990. p. 339–417. Gemery J, Idelson B, Reid S, Yucel EK, Pagan Marin H, Ali S, et al. Acute renal failure after arteriography with a gadolinium-based contrast agent. Am J Roentgenol 1998; 171:1277–8. Hall RL. Principles of clinical pathology for toxicology studies. In: Hayes AW, editor. Principles and methods of toxicology. Philadelphia: Taylor & Francis Publishers; 2001. p. 1001–38. Harpur ES, Worah D, Hals PA, Holtz E, Furuhama K, Nomura H. Preclinical safety assessment and pharmacokinetics of gadodiamide injection, a new magnetic resonance imaging contrast agent. Invest Radiol 1993;28:S28–43.
Kasokat T, Urich K. Quantification of dechelation of gadopentetate dimeglumine in rats. Arzneimittelforschung 1992;42(6):869–76. Kirchin MA, Runge VM. Contrast gents for magnetic resonance imaging: safety update. Top Magn Reson Imag 2003;14:426–35. Lorusso V, Pascolo L, Ferretti C, Anelli PL, Uggeri F, Tiribelli C. Magnetic resonance contrast agents: from the bench to the patient. Curr Pharmaceut Des 2005;11:4079–98. Morisetti A, Bussi S, Tirone P, de Hae¨n C. Toxicological safety evaluation of gadobenate dimeglumine 0.5 M solution for injection (MultiHances), a new magnetic resonance imaging contrast medium. J Comput Assist Tomogr 1999;23(Suppl 1):S207–17. Murphy KPJ, Szopinski KT, Cohan RH, Mermillod B, Ellis JH. Occurrence of adverse reactions to gadoliniumbased contrast material and management of patients at increased risk: a survey of the American Society of Neuroradiology Fellowship Directors. Acad Radiol 1999;6:656–64. Niendorf HP, Alhassan A, Geens VR, Clauss W. Safety review of gadopentetate dimeglumine: extended clinical experience after more than five million applications. Invest Radiol 1994;29:S179–82. Runge VM, Bradley WG, Brant-Zawadzki MN, Carvlin MJ, DeSimone DN, Dean BL, et al. Clinical safety and efficacy of gadoteridol: a study in 411 patients with suspected intracranial and spinal disease. Radiology 1991;181(3): 701–9. Spencer AJ, Wilson SA, Batchelor J, Reid A, Rees J, Harpur E. Gadolinium chloride toxicity in the rat. Toxicol Pathol 1997;25:245–55. Terzi C, Sokmen S. Acute pancreatitis induced by magneticresonance-imaging contrast agent. Lancet 1999;354: 1789–90. Tombach B, Heindel W. Value of 1.0-M gadolinium chelates: review of preclinical and clinical data on gadobutrol. Eur Radiol 2002;12(6):1550–6. Vargas MH. InerSTAT-a v1.3 – simple statistics tool for Excel. Instituto Nacional de Enfermedades Respiratorias, Hospital General ‘‘Dr. Manuel Gea Gonza´lez’’, Mexico, 1999. Wedeking P, Tweedle M. Comparison of the biodistribution of 153 Gd-labeled Gd(DTPA)2, Gd(DOTA), and Gd(acetate)n in mice. Nucl Med Biol 1988;15:395–402.
