Nephrogenic systemic fibrosis-like effects of magnetic resonance imaging contrast agents in rats with adenine-induced renal failure.

Nephrogenic systemic fibrosis (NSF) is a scleroderma-like disease associated with prior administration of certain gadolinium chelates (GCs). NSF occurs in patients with severe renal failure. The purpose of this study was to set up a rat model of GC-associated NSF to elucidate the mechanism of this devastating disease. Firstly, after characterization of the model, male Wistar rats received a 0.75% adenine-enriched diet for 8, 14, or 16 days to obtain various degrees of renal failure. Rats received five consecutive daily iv injections of saline or gadodiamide (2.5 mmol/kg/day). Secondly, the safety profile and in vivo propensity to dissociate of all categories of marketed GCs (gadoterate, gadobutrol, gadobenate, gadopentetate, and gadodiamide) were compared in rats receiving adenine-enriched diet for 16 days. Serial skin biopsies were performed for blinded histopathological study. Total Gd concentration in tissues was measured by Inductively Coupled Plasma Mass Spectrometry. Relaxometry was used to evaluate the presence of dissociated Gd in skin and bone. Gadodiamide-induced high mortality and skin lesions (dermal fibrosis, calcification, and inflammation) were related to adenine diet duration. No skin lesions were observed with other molecules. Unlike macrocyclic GCs, gadodiamide, gadopentetate, and gadobenate gradually increased the r(1) relaxivity value, consistent with in vivo dissociation and release of soluble Gd (or, in the case of gadobenate, the consequence of protein binding). Gadodiamide-induced cutaneous and systemic toxicity depended on baseline renal function. We demonstrate in vivo dissociation of linear GCs, gadodiamide, and gadopentetate, whereas macrocyclic agents remained stable over the study period.

Skin gadolinium following use of MR contrast agents in a rat model of nephrogenic systemic fibrosis.

PURPOSE: To detect the ultrastructural site of gadolinium retention in skin by using an animal model of nephrogenic systemic fibrosis and compare a linear, low-stability gadolinium chelate (formulated gadodiamide) with a macrocylic, high-stability gadolinium chelate (gadoterate meglumine). MATERIALS AND METHODS: Experimental procedures were performed according to rules and regulations laid down by the UK Home Office (Animal Procedures Act of 1986). Male Wistar rats were subjected to 5/6 subtotal nephrectomy (creatinine clearance, 25% normal). Gadolinium-based contrast agents, formulated gadodiamide (n = 9) and gadoterate meglumine (n = 11), were administered intravenously (2.5 mmol/kg for 5 days). After 28 days, skin was analyzed by means of morphometric and immunohistochemical techniques and electron microscopy. Data were compared with the Student t test. Skin gadolinium was located by means of energy-filtered transmission electron microscopy. RESULTS: Formulated gadodiamide produced a 40-fold greater increase in gadolinium in skin than did gadoterate meglumine. An electron-dense filamentous material, detected within extracellular matrix, displayed a "halo" appearance, associated with collagen fibrils and electron-dense intracellular fragments of collagen fibrils within activated fibroblasts. Both electron-dense features demonstrated the presence of gadolinium but were much less apparent following gadoterate meglumine administration, where the presence of gadolinium was not detected. Formulated gadodiamide increased dermal cell count, dermal thickness, and collagen bundle density with enhanced immunostain for CD34, fibroblast-specific protein 1,4-hydroxy-prolyl-hydroxylase, and factor XIIIa. Circular staining for α-smooth muscle actin indicated new blood vessel formation. Skin of rats receiving gadoterate meglumine remained unchanged. CONCLUSION: Gadolinium retention in skin following formulated gadodiamide administration was located to the collagen fibril, in both the extracellular matrix and within activated fibroblasts.

Gadolinium contrast agent associated stimulation of human fibroblast collagen production.

OBJECTIVES: Nephrogenic systemic fibrosis occurs in patients with poor renal function who receive gadolinium-based contrast agents (Gd-CAs). Several reports suggest that this is more likely to occur with the less stable forms of Gd chelates, suggesting a release of cytotoxic free Gd ions from these. There is evidence that Gd can stimulate human fibroblast proliferation but the evidence is less clear concerning the production of collagen by these cells. Our aim was to assess effects of Gd chelates on human skin cell activity and collagen production. MATERIALS AND METHODS: Keratinocytes and fibroblasts were cultured with 3 Gd chelates (Gd-EDTA, Omniscan [nonionic linear Gd-CA], and Dotarem [ionic macrocyclic Gd-CA]) for up to 7 days, and cell viability and collagen production were assessed using the colorimetric assays of MTT (3-(4, 5-dimethylthiazol-2-yl) 2, 5-diphenyltetrazolium bromide) and Sirius Red, respectively. The uptake of Gd by cultured fibroblasts was also undertaken using the techniques of inductively coupled plasma mass spectrometry and relaxometry. RESULTS: Our data show that Gd-EDTA and Omniscan significantly stimulated both fibroblast and keratinocyte viability and fibroblast (but not keratinocyte) collagen production. In contrast, Dotarem had little, if any, effect on these cultured cells. The Omniscan-induced increase in fibroblast collagen was around 40% over 7 days-a similar increase to that seen for cell viability, suggesting that collagen production was secondary to an initial stimulatory effect on fibroblast viability. Studies of the uptake of Gd by the cultured fibroblasts showed that these took up Gd when cultured with Omniscan for 7 days, and our study also suggests that some of this Gd was in a free dissociated form. CONCLUSIONS: We conclude that these results support a simple nephrogenic systemic fibrosis causative role of low-stability nonionic linear Gd-CA inducing dermal collagen via the release of dissociated Gd which enters fibroblasts and stimulates their activity and associated collagen production.

Comparative in vivo dissociation of gadolinium chelates in renally impaired rats: a relaxometry study.

PURPOSE: Investigation of dissociated versus chelated gadolinium (Gd) in plasma, skin, and bone of rats with impaired renal function after administration of ionic macrocyclic (gadoterate or Dotarem) or nonionic linear (gadodiamide or Omniscan) Gd chelates. MATERIALS AND METHODS: Subtotally nephrectomized Wistar rats were subjected to receive daily injections of 2.5 mmol/kg of Omniscan, gadodiamide without excess ligand caldiamide, Dotarem, or saline (n = 7-10 rats/group) for 5 consecutive days. The Gd concentration was measured by inductively coupled plasma mass spectrometer in skin, femur epiphysis, and plasma on completion of the study (day 11), and dissociated Gd(3+) was measured in the plasma at day 11 (liquid chromatography inductively coupled plasma mass spectrometry). The r(1) relaxivity constant was measured in skin (at day 4 and day 11) and bone (day 11) to investigate the dissociated or chelated form of Gd found in tissue samples. Clinical and skin histopathologic studies were performed. RESULTS: Subtotal nephrectomy decreased creatinine clearance by 60%. No macroscopic skin lesions were observed in the Dotarem and Omniscan groups in contrast with the gadodiamide group (2 rats survived the study period and 4 of 10 rats showed skin ulcerations and scabs). Skin histopathologic lesions were in the range gadodiamide > Omniscan > Dotarem (similar to control rats). At day 11, the skin Gd concentration was lower in the Dotarem group (161.0 ± 85.5 nmol/g) as compared with the Omniscan (490.5 ± 223.2 nmol/g) and gadodiamide groups (mean value, 776.1 nmol/g; n = 2 survivors). The total Gd concentration in the femur was significantly higher in the Omniscan group than in the Dotarem group. At day 11, the dissociated Gd(3+) concentration in plasma was below the limit of detection in the Dotarem group and was 1.5 ± 0.7 µmol/L in the Omniscan group corresponding to 62% ± 15% of the total Gd concentration. The dissociated Gd(3+) concentration was 1.1 µmol/L in gadodiamide rats (n = 2 survivors). In the skin, the in vivo r1 relaxivity value increased from 4.8 ± 0.7 mM(-1)s(-1) at day 4 to 10.5 ± 3.9 mM(-1)s(-1) at day 11 in the Omniscan group, P < 0.05 (in vitro r(1) in skin, 3.5 mM(-1)s(-1)) and gadodiamide group, whereas no significant change was observed in the Dotarem group (2.8 ± 0.2 and 4.9 ± 2.8 mM(-1)s(-1) at day 4 and 11, respectively, NS) (in vitro value in the skin, 3.2 mM(-1)s(-1)). In the femur, the in vivo r1 relaxivity was higher in the Omniscan group (8.9 ± 2.1 mM(-1)s(-1)) (in vitro relaxivity, 4.5 mM(-1)s(-1)) and gadodiamide group (8.8 mM(-1)s(-1), n = 2 survivors) than in the Dotarem group (3.8 mM(-1)s(-1), n = 1 rat with measurable r(1), since for 7 rats, 1/T(1) - 1/T(1(diamagnetic)) <10% of 1/T(1(diamagnetic)) because of low Gd concentration) (in vitro relaxivity value in the femur matrix, 3.1 mM(-1)s(-1)). CONCLUSIONS: Unlike Dotarem, Omniscan and gadodiamide induced histologic skin lesions. At day 11, a higher Gd concentration was found in both skin and femur of Omniscan- and gadodiamide-treated rats than in Dotarem-treated rats. Relaxometry results indicate gradual in vivo dechelation and release of dissociated Gd(3+) in a soluble form in renally impaired rats receiving Omniscan and gadodiamide, whereas Dotarem remained stable over the study period.

Nephrogenic gadolinium biodistribution and skin cellularity following a single injection of Omniscan in the rat.

OBJECTIVES: The development of nephrogenic systemic fibrosis (NSF) following MRI contrast examination has been associated with gadolinium (Gd) toxicity. Animal models should show the key features of NSF in man where, the only immutable epidemiological feature is renal impairment. A rat model of chronic renal insufficiency has been employed to establish whether tissue gadolinium retention and increased skin cellularity following a gadolinium based contrast agent (GBCA) can be correlated with a reduction in renal function. The GBCA chosen for investigation was Omniscan, the least stable of the commercially available agents. MATERIALS AND METHODS: Wistar rats were subjected to 5/6 subtotal nephrectomy (SNx) under isoflurane anesthesia. The glomerular filtration rate (GFR) was assessed from serum creatinine and creatinine clearance. Two SNx rats groups were established, following either 75% or 80% resection of the kidney, which reduced the GFR down to 40% and down to 20%, respectively, of sham-operated controls. Three months after surgery, rats received a single intravenous injection of either saline or Omniscan (gadodiamide 2.5 mmol/kg). Four weeks later, the Gd content of serum, skin, liver, and bone was measured by inductively coupled plasma mass spectrometry and skin cellularity determined. RESULTS: In sham-operated rats, Gd was detected in skin < liver < bone. SNx rats with the GFR reduced down to 20% normal, had an increased tissue Gd concentration in bone (2.5-fold), skin (3-fold), and liver (10-fold) compared with sham-operated controls. The Gd concentration in all 3 tissues showed a positive linear correlation with serum creatinine (P < 0.01). No external skin lesions were observed. The skin cellularity of rats with the GFR reduced down to 20% of normal was increased following Omniscan, together with positive immunostain for CD34 and prolyl-4-hydroxylase. CONCLUSIONS: The SNx rat is a sensitive model for investigating the pathophysiology of NSF. A positive linear correlation was obtained between tissue Gd and serum creatinine, the major clinical marker of renal function. An increase in skin cellularity, a feature of human NSF, was demonstrated in rats with a level of renal impairment equivalent of stage 4 chronic kidney disease following just a single intravenous dose of Omniscan. This response was obtained in the absence of ulcerogenic skin lesions, at skin Gd concentrations as low as 50 nmol/g.