Gadolinium Contrast Media for Radiographic Examinations 115 17 Gadolinium Contrast Media for Radiographic Examinations Henrik S. Thomsen

CONTENTS Gadolinium based contrast agents are in general known be safe and not nephrotoxic in the usual MRI 17.1 Introduction 115 doses up to 0.3 mmol/kg body weight (BW) However, 17.2 Gadolinium Preparations 115 17.3 Pharmacokinetics 116 the dose requirement for a satisfactory diagnostic study differs between MR and X-ray examination 17.4 Toxicity (LD50) 116 17.5 Incidence of General Reactions to Gadolinium Based because different properties of the gadolinium are Contrast Agents 116 being used in the two modalities. The use of gado- 17.6 Attenuation of X-Rays by Iodine linium-based contrast agents in radiographic exam- and Gadolinium 116 inations is contentious and the risks poorly under- 17.7 Clinical Studies 117 Thomsen Thomsen 17.8 Experimental Nephrotoxicity 118 stood ( et al. 2002; 2003). 17.9 MR Examinations and Nephrotoxicity 119 17.10 Conclusion 119 References 119 17.2 Gadolinium Preparations

17.1 The first four marketed gadolinium contrast media Introduction (gadopentate dimeglumine, gadoterate, gadodi- amide, gadoteridol) are available in a concentration It has been suggested that patients with significant of 0.5 mmol/ml. The same applies to gadobenate renal impairment and/or previous severe reactions dimeglumine which unlike the other four agents is to iodinated contrast media should receive gado- also excreted via the liver (1%) (Chap. 17). Recently, linium-based MRI contrast agents instead of the gadobutrol has been introduced in a concentration traditional iodinated radiographic contrast agents of 1 mmol/ml (see also Chaps. 9 and 16 for non- (Albrecht and Dawson 2000; Bittner et al. renal safety aspects). For all six agents there is one 1997; Engelbrecht et al. 1996). Another possible Gd-atom in each molecule, so the molar concentra- indication could be before thyroid treatment with tion of the agent and of gadolinium is the same. radioactive iodine to avoid interference with iodine Traditionally, iodine radiographic contrast media uptake. are marketed based on the mg of iodine per ml. The At the kV (~70) used for digital angiography, the concentration of 300 mg I/ml is equal to 2.38 mmol attenuation of X-rays by gadolinium is approxi- I/ml. Since there are three iodine atoms per mol- mately the same as for iodine. At the kV (~120) used ecule, the molar concentration of the agent is only for CT, the attenuation of X-rays by gadolinium is 0.8 mmol/ml. approximately double that of iodine. Therefore the- The commonly used dose for body CT is 150 ml of oretically gadolinium could therefore replace iodine a 300 mg I/ml (2.38 mmol I/ml) solution. The stand- as a radiographic contrast agent. ard dose for contrast-enhanced MR examination is 0.2 ml/kg BW of a 0.5 mmol/ml gadolinium-based contrast agent. For body CT, a patient weighing 70 kg would receive 120 mmol of the iodinated agent molecule (0.8 mmol/ml u 150 ml) and 360 mmol of H. S. Thomsen, MD Professor, Department of Diagnostic Radiology 542E, Copen- iodine (2.38 mmol/ml u 150 ml). For MR examina- hagen University Hospital at Herlev, Herlev Ringvej 75, 2730 tion, this same 70 kg patient would receive 7 mmol Herlev, Denmark of the gadolinium based agent molecule and 7 mmol 116 H. S. Thomsen

of gadolinium [0.5 mmol/ml u 14 ml (0.2 ml/kg BW about 50 mmol iodine/kg. The LD50 of low osmo- u 70 kg BW)]. Thus, the number of iodinated con- lality nonionic monomers, e.g. iopromide, is much trast agent molecules administered would be almost higher, about 150 mmol iodine per kg) (Weinmann einmann 17 times that of gadolinium containing molecules, et al. 1990; W 1999). These LD50 values and the number of iodine atoms administered suggest that comparing attenuating atoms the acute would be 51 times that of gadolinium. For a patient intravenous toxicity of the gadolinium-based con- weighing 50 kg, the difference is even larger [~24 trast media is 6 to 25 times that of the nonionic times (molecule) and ~72 times (atom)], whereas for iodinated monomers. a patient weighing 100 kg it is less [~12 times (mol- ecule) and ~36 times (atom)].

17.5 Incidence of General Reactions to 17.3 Gadolinium Based Contrast Agents Pharmacokinetics General adverse reactions similar to those observed The gadolinium chelates have pharmacokinetics with iodinated contrast media may be seen follow- similar to those of iodinated radiographic contrast ing injection of gadolinium based contrast agents, agents with the exception of gadobenate dimeglu- but the frequency is lower with the incidence of mod- mine which is also excreted by the liver in small erate and severe reactions well below 1% (Niendorf amounts (Chap. 17), but is mainly used for non liver et al. 1991; Thomsen 1997). However, the number specific indications with the five other “extracellu- of patients exposed to unapproved dosages (above lar” gadolinium chelates (Chap. 16). Both types of 0.3 mmol/kg BW) is still too small to draw any con- agent are distributed in the extracellular space and clusion about the safety of these higher doses. In excreted by glomerular filtration. Thus, the T½ is the few published studies, varying doses of gado- almost the same, and both types of agents can be linium-based agents (20–440 ml) have been used used to measure the glomerular filtration rate. In and the number of patients has been small. Equally, patients with normal kidney function about 98% the prevalence of generalized reactions to very low of these agents is excreted within 24 h of injection. doses of iodinated contrast agents (e.g. 10 ml of a However, in patients with severe renal impairment, 300 mgI/ml solution) is not documented in a large excretion of gadolinium and iodinated agents dif- number of patients. fers. Nearly no gadolinium is found in the feces in patients with renal insufficiency, whereas up to 6% of the injected iodine has been recovered in the feces of such patients (Joffe et al. 1998). No free gadolin- 17.6 ium is found in the blood several days after injection Attenuation of X-Rays by Iodine and of gadolinium chelates in patients with end-stage Gadolinium renal failure despite the slow excretion (Joffe et al. 1998; Normann et al. 2000). Iodine has the atomic number 53 and an atomic weight of 127, whereas gadolinium has the atomic number 64 and an atomic weight of 157. Attenuation increases with the atomic number of the atom but 17.4 decreases with the energy (keV) of the X-ray photons, Toxicity (LD50) except at the K-edges. At photon energies between the K-edge of iodine [33 kilo electron Volt (keV)] Acute intravenous LD50 of contrast media in mice and that of gadolinium (52 keV), iodine attenuates is expressed as mmol iodine or gadolinium atoms approximately twice as many X-ray photons as does per kg BW. For the five gadolinium-based contrast gadolinium. At all other photon energies, the oppo- agents, dimeglumine gadopentate, gadobenate site prevails (Nyman et al. 2002). For CT, the maxi- dimeglumine, gadoterate, gadoteridol, gadodiamide, mal X-ray photon energy is between 120140 keV the figures are 6, 8, 8, 18 and 20 mmol gadolinium/ and the most common photon energies in the spec- kg, respectively. The LD50 for the conventional high trum are between 6070 keV. This is above the K- osmolality iodinated contrast agent diatrizoate is edge of gadolinium, so the attenuation by gadolin- Gadolinium Contrast Media for Radiographic Examinations 117 ium in this situation is about twice that of iodine; contrast agent. However, the molar doses and con- but since there are three iodine atoms per contrast centrations of the iodine and gadolinium atoms medium molecule, the iodine molecule attenuates were not comparable. Although the exact dose of 1.5 times more radiation than does a gadolinium- the iodinated contrast used for each patient could based contrast molecule. For common radiographic not be verified, between 30 and 60 g I was adminis- examinations, the maximal X-ray photon energy is tered. For the MR examinations between 0.2 and 0.4 between 7090 keV and the most common photon mmol/kg BW were used. Assuming that all patients energies in the spectrum are above and below the were standard (~70 kg), the dose of iodine atoms was K-edge of gadolinium (50 keV). Because of the range approximately 17 times higher than that of gadolin- of photon energies, attenuation is approximately the ium atoms. Thus, the doses were not comparable and same for iodine and gadolinium atoms. Hence, the had equi-attenuating doses been used, the results attenuation by the iodinated contrast agent mole- might have been different. cule is three times that of the gadolinium molecule Over recent years, gadolinium-based contrast (Nyman et al. 2002). Below 33 keV only very few agents have been used for examinations such as CT, photons will pass through the body. intravenous urography and digital subtraction angi- It should theoretically be possible to obtain radi- ography of various parts of the body (e.g. liver, renal ographic images of diagnostic quality with gado- and peripheral arteries). Albrecht and Dawson linium-based contrast agents, but the image qual- (2000) studied 15 patients receiving 0.3 mmol/kg BW ity will generally be inferior to that achieved with gadopentate dimeglumine; five had abdominal CT, iodinated contrast agents. This can be explained five abdominal DSA and five intravenous urography. by the difference in molar concentrations between No side-effects were reported, but generally the image gadolinium- and iodine-based contrast agents. A quality was inferior to that obtained subsequently 0.5 mmol/ml concentration of iodine atoms contains with standard doses of iodinated contrast media (50- 63 mgI/ml. Assuming that a 0.5 mmol/ml concen- -150 ml of a 300 or 350 mgI/ml solution). The authors tration of gadolinium attenuates to the same extent suggested that higher doses including more concen- as a 0.5 mmol/ml concentration of iodine, a patient trated solutions of gadolinium-based contrast media receiving these equi-attenuating concentrations might be useful (Albrecht and Dawson 2000). will receive only 1/3 of the iodine contrast medium Gadolinium-based contrast media have also been molecules compared to the situation with gadolin- used for endoscopic retrograde cholangiography, ium contrast medium molecules. Considering the cystography, urethrocystography, and retrograde molar concentration of an iodinated contrast agent pyelography and during percutaneous nephrostomy at 300 mgI/ml, the attenuation of this preparation is and biliary tract drainage with resultant adequate almost five times that of gadolinium preparations at image quality and no side-effects (Velmas and equi-volume. Thus, the volume of gadolinium prepa- Markkola 1998). Coche et al. (2001) reported suc- ration required to obtain “comparable” attenuation cessful detection of pulmonary embolism using is five times that of the iodine preparation. gadolinium-enhanced helical CT (0.4 mmol/kg gadodiamide) in a 77-year-old woman with pre- vious allergy-like reaction to iodinated contrast medium and renal insufficiency (serum creatinine 17.7 of 200 Pmol/ml) without any problems. A total of 14 Clinical Studies patients with abnormal S-creatinine levels underwent digital subtraction vena cavography with a gadolin- Prince et al. (1996) studied 64 patients undergo- ium-based contrast agent (maximum 0.4 mmol/kg ing MR examination with a gadolinium-based agent BW) for filter placement, thrombolysis or diagnosis. and a radiographic examination with an iodinated Three of the 14 patients had a significant increase contrast medium. They concluded that high-dose in serum creatinine (> 44 Pmol/ml), but there were gadolinium chelates are significantly less neph- other concurrent causes, which might account for rotoxic than iodinated agents, since eleven of the the deterioration of renal function (Kaufmann et 64 patients had a significant increase in serum cre- al. 1999). It was concluded that gadolinium-based atinine after intravenous or intraarterial admin- contrast agents were suitable for digital subtraction istration of iodine-based contrast media whereas venography in patients with renal insufficiency. none had increased serum creatinine levels after In an azotemic patient with suspected renal intravenous administration of a gadolinium-based artery stenosis, a total of 40 ml (0.5 mmol/ml) undi- 118 H. S. Thomsen luted dimeglumine gadopentate was injected arteri- (Schenker et al. 2001). Acute pancreatitis has been ally (Matchett et al. 1996). The serum creatinine seen both after intraarterial (Gemery et al. 1998) increased from 290 Pmol/l to 390 Pmol/l, but this and intravenous (Terzi and Sokman 1999) injec- might have been attributable to a myocardial inf- tion of a gadolinium-based contrast agent. arction which the patient developed 3 days after the procedure. Acute renal failure was described fol- lowing lower extremity arteriography with 80 ml of 0.5 mmol/ml (0.44 mmol/kg BW) of gadoteridol in 17.8 an insulin-dependent diabetic patient with neph- Experimental Nephrotoxicity ropathy (Gemery et al. 1998). S-creatinine tran- siently increased from 350 to 820 Pmol/ml and the Intravenous injection (9 ml/kg) of gadopentate deterioration was considered most likely due to the (0.1 mol/ml), iohexol (300 mgI/ml), metrizoate contrast agent. (300 mgI/ml) and normal saline in rabbits showed A total of 31 patients with azotemia or previous nephrotoxicity of the same order for all three con- severe adverse reaction to iodinated contrast media trast agents (Leander et al. 1992). The molar con- underwent digital subtraction angiography with centration and dose of iodine atoms was 24 times between 20 and 60 ml of 0.5 mmol/ml gadopen- higher than the molar concentration and dose of Hammer tate ( et al. 1999). In nine cases, CO2 was gadolinium atoms. Thus, the iodinated agents might also used and in eight cases between 6 and 40 ml of have had a lower nephrotoxic effect than the gado- iohexol 350 mgI/ml (mean 17.8 ml) were used. In no linium media if the two agents had been compared patient did S-creatinine increase more than 44 Pmol/ in equi-attenuating doses and concentrations. Rat l within 48 h. Spinosa et al. (1998) studied 13 renal studies where high equimolar doses (4.59 mmol/kg transplant patients with suspected vascular causes BW) of gadolinium (gadopentate and gadodiamide) of renal insufficiency and/or accelerated hyperten- and iodinated (diatrizoate and iohexol) contrast sion with both CO2 and a gadolinium-based contrast agents were injected intravenously showed no sig- agent (16–60 ml gadodiamide). Digital subtraction nificant deterioration in the function of normal and angiography was considered adequate in all patients. diseased kidneys (Thomsen et al. 1994, 1995). There In two patients renal failure progressed (> 44 Pmol/l was a significant correlation between albuminuria within 48 h), but concurrent causes of the renal dys- and the osmolality of the contrast medium; gado- function were also present; one had received 20 and pentate caused the highest excretion and gadiodi- the other received 60 ml of gadodiamide. During amide and iohexol the least. However, the degree peripheral arteriography Spinosa et al. (1999) found of albuminuria does not correlate with nephrotoxic that gadodiamide with an osmolality of 789 mOsm potential of a contrast medium. In these studies the per kilogram of water was less painful than gado- dose of iodine atoms was three times the dose of pentate dimeglumine with an osmolality of greater gadolinium atoms. than 1800 mOsm per kilogram of water. No effects In an ischemic rat model, intra-aortic injections on renal function were found. Later Spinosa et al. of 1.5 ml (0.5 mmol/ml) gadopentate (0.75 mmol Gd (2000) reported one of 18 azotemic patients (6%) atoms) and 2.6 ml 370 mgI/ml diatrizoate (7.6 mmol whose renal function deteriorated after undergo- iodine atoms) caused a significant decrease in cre- ing CO2 angiography supplemented with 0.5 mmol/ atinine clearance of similar magnitude, 50% and ml gadodiamide (20100 ml; mean volume 55 ml; 67%, respectively (Deray et al. 1990; Brillet et al. 0.13.04 mmol/kg). The affected patient received 70 1994). Gadoterate [1.5 ml (0.5 mmol/l)] alone caused ml gadodiamide (0.3 mmol/kg BW) no decrease in renal function in this model. The Injections of 80–440 ml of gadodiamide during dose of iodine was ten times higher than the dose of arteriography have also been reported (Gemette et gadolinium and the two different doses produced a al. 2001). A S-creatinine increase of 53 Pmol/ml or similar significant decrease in creatinine clearance. more occurred in eight of 20 patients (40 %) with Whether the iodinated contrast medium would pro- a preprocedural S-Cr of 115–548 Pmol/ml. In three duce less decrease in creatinine clearance than the of the eight patients, the creatinine values did not gadolinium medium if equimolar doses had been return to baseline value. Following peripheral gado- given remains speculative. linium arteriography, angioplasty and stent place- In an experimental model of renal ischemia in pigs, ment, a patient with renal insufficiency (340 ìmol/l) 0.5 molar gadopentate dimeglumine (3 ml/kg BW) developed acute renal failure and acute pancreatitis caused severe impairment of renal function; the low- Gadolinium Contrast Media for Radiographic Examinations 119 osmolar gadodiamide caused less deterioration in Several reports have shown the usefulness of renal function, and the low-osmolar iohexol (3 ml of gadolinium-based agent in radiographic examina- 190 mgI/ml per kg BW) caused even less (Elmståhl tions including CT when iodinated contrast agents et al. 2004). Three ml per kg BW of iohexol (70 mg were contraindicated for a variety of reasons. The iodine/ml), which for angiography is equi-attenuat- major drawback when using gadolinium-based ing with 0.4 molar gadopentate dimeglumine, caused contrast agents for CT or radiography is that com- no change in renal function. An in vitro study using mercially available contrast media have only one the isolated perfused rat kidney showed that a large gadolinium atom per molecule and a low molar dose of gadopentate dimeglumine (0.3 mmol/kg BW) concentration. In comparison, iodinated monomers did not cause significant reduction in renal function for radiographic examinations contain three iodine (Brown et al. 1993). However, an equimolar dose per atoms per molecule and have molar concentration kg BW of iodine atoms in a 70 kg man would be 10 ml five times that of gadolinium in the four gadolin- at concentration of 265 mg iodine/ml. ium-based contrast agents (dimeglumine gado- pentate, gadobenate dimeglumine, gadoteridol, gadodiamide, gadoterate). Hence, image quality is generally inferior when gadolinium-based contrast 17.9 media are used for radiography. Gadolinium-based MR Examinations and Nephrotoxicity contrast media should not be used for radiographic examinations (see Appendix). Sam et al. (2003) reported that in 3.5% of 195 patients with abnormal pre-examination creatinine clearance levels, acute renal failure (anuria) developed after gadolinium-based contrast medium administration. References For MR angiography the incidence was 1.9% and for digital subtraction angiography 9.5%. Dialysis was Albrecht T, Dawson P (2000) Gadolinium-DTPA as X-ray con- required in three of the seven patients who developed trast medium in clinical studies. Br J Radiol 73:878882 Bittner CA, Goodwin SC, Lu D, McNamara TO, Joseph T (1997) acute renal failure. The average creatinine clearance 2 Gadolinium-based contrast agents for angiographic use in the whole group was 38.2±1.6 ml/min/1.73 m and as a safe radiocontrast in patients with impaired renal in the seven patients who developed contrast medium function. J Vasc Interv Radiol [Suppl] 8:178 induced nephropathy it was 32.5±7.8 ml/min/1.73 Brillet G, Dubois M, Beaufils H, Bourboze R, Deray G (1994) m2. The doses of gadolinium-DTPA ranged from Renal tolerance of gadolinium-DOTA and gadolinium- 0.31 to 0.41 mmol/kg for MR angiography and 0.27 DTPA in rats. Invest Radiol 29:352354 Brown PWG, Haylor JL, Morcos SK, El Nahas AM (1993) The to 0.42 mmol/kg for digital subtraction angiography. functional effects of gadolinium-DTPA on the isolated Contrast medium induced nephropathy occurred perfused rat kidney. Eur J Radiol 16:8589 after a moderate (0.14 mmol/kg) – approved – dose of Coche EE, Hammer FD, Gofette PP (2001) Demonstration a gadolinium-based contrast medium in a patient with of pulmonary embolism with dynamic gadolinium- moderate to severe diabetic nephropathy and chronic enhanced spiral CT. Eur Radiol 11:23062309 heart failure (Thomsen 2004). In diabetic patients with Deray G, Dubois M, Martinez F, Baumelou B, Beaufils H, Bourboze R, Baumelou A, Jacobs C (1990) Renal effects multiple risk factors it may be appropriate to take the of radiocontrast agents in rats: A new model of acute renal same precautions before enhanced MR examinations failure. Am J Nephrol 10:507513 as before enhanced radiographic examinations. Elmståhl B, Nyman U, Leander P et al (2004) Gadolinium con- trast media are more nephrotoxic than low dose iodine medium employing doses with equal X-ray attenuation in renal arteriography: an experimental study in pigs. Acad Radiol 11:12191228 17.10 Engelbrecht V, Koch JA, Rassek M, Mödder U (1996) Gado- Conclusion diamide and gadolinium-DTPA as intravenous media in computed tomography. Röfo Fortschr Geb Rontgenstr Nephrotoxicity of the gadolinium-based contrast Neuen Bildgeb Verfahr 165:2428 agents when used for radiographic studies, CT and Gemery J, Idelson B, Reid S et al (1998) Acute renal failure after arteriography with a gadolinium-based contrast agent. MRI has now been described in both man and ani- AJR Am J Roentgenol 171:12771278 mals. Use of high doses (> 0.3 mmol/kg BW) of the Gemmete JJ, Forauer AR, Kazanjian S, Dasika N, Williams DM, gadolinium agents in patients with impaired renal Cho K (2001) Safety of large volume gadolinium angiog- function is contraindicated. raphy. J Vasc Interv Radiol 12 (part 2):S28 120 H. S. Thomsen

Hammer FD, Gofette PP, Maliase J, Mathurin P (1999) Gadolin- of renal insufficiency and accelerated hypertension. J Vasc ium dimeglumine: an alternative contrast agent for digital Interv Radiol 9:909916 subtraction angiography. Eur Radiol 9:128136 Spinosa DJ, Matsumoto AH, Hagspiel KD, Angle JF, Hartwell Joffe P, Thomsen HS, Meusel M (1998) The pharmacokinet- GD (1999) Gadolinium-based contrast agents in angiog- ics of gadodiamide injection in patients with severe renal raphy and interventional radiology. AJR Am J Roentgenol insufficiency treated conservatively or undergoing hemo- 173:14031409 dialysis or continuous ambulatory peritoneal dialysis. Spinosa DJ, Angle JF, Hagspiel KD, Kern JA, Hartwell GD, Mat- Acad Radiol 5:491502 sumoto AH (2000) Lower extremity arteriography with Kaufmann JA, Geller SC, Bazari H, Waltman AC (1999) Gado- use of iodinated contrast material or gadodiamide to sup- linium-based contrast agents as an alternative at vena plement CO2 angiography in patients with renal insuf- cavography in patients with renal insufficiency – early ficiency. JVIR 11:3543 experiences. Radiology 212:280284 Terzi C, Sokmen S (1999) Acute pancreatitis induced by Leander P, Allard P, Caillé JM, Golman K (1992) Early effect magnetic-resonance-imaging contrast agent. Lancet of gadopentate and iodinated contrast media on rabbit 354:17891790 kidneys. Invest Radiol 27:922926 Thomsen HS (1997) Frequency of acute adverse events to a Matchett WJ, McFarland DR, Rusell DK, Sailors DM, Moursi non-ionic low-osmolar contrast medium: the effect of MM (1996) Azotemia: gadopentate dimeglumine as con- verbal interview. Pharmacol Toxicol 80:108110 trast agent at digital subtraction angiography. Radiology Thomsen HS (2003) ESUR Guidelines in contrast media. AJR 201:569571 Am J Roentgenol 181:14611471 Niendorf HP, Gifford LM, Haustein J, Cornelius I, Alhassan A, Thomsen HS (2004) Gadolinium-based contrast media may Clauss W (1991) Tolerance data of Gd-DTPA: a review. Eur be nephrotoxic even at approved doses. Eur Radiol J Radiol 13:1520 14:16541656 Normann PT, Joffe P, Martinsen I, Thomsen HS (2000) Identifi- Thomsen HS, Almén T, Morcos SK, Members of Contrast cation and quantification of gadodiamide in serum, peri- Media Safety Committee of European Society of Uro- toneal dialysate and faeces from end-stage renal patients genital Radiology (2002) Gadolinium-containing contrast dosed with gadodiamide injection by inductively coupled media for radiographic examinations: a position paper. plasma-atomic emission spectroscopy and comparative Eur Radiol 12:26002605 analysis by high-performance liquid chromatography. J Thomsen HS, Dorph S, Larsen S et al (1994) Urine profiles Pharmaceut Biomed Anal 22:939947 and kidney histology after intravenous injection of ionic Nyman U, Elmståhl B, Leander P, Nilsson M, Golman K, Almén and nonionic radiologic and magnetic resonance contrast T (2002) Are gadolinium-based contrast media really safer media in normal rats. Acad Radiol 1:128135 than iodinated media for digital subtraction angiography Thomsen HS, Dorph S, Larsen S et al (1995) Urine profiles in patients with azotemia? Radiology 223:311318 and kidney histology after intravenous injection of ionic Prince MR, Arnoldus C, Frisoli JK (1996) Nephrotoxicity of and nonionic radiologic and magnetic resonance contrast high-dose gadolinium compared with iodinated contrast. media in rats with cisplatin nephropathy. Acad Radiol JMRI 1:162166 2:675682 Sam AD II, Morasch MD, Collins J, Song G, Chen R, Pereles Velmas T, Markkola T (1998) Gd-DTPA as an alternative con- FS (2003) Safety of gadolinium contrast angiography trast agent in conventional and interventional radiology. in patients with chronic renal insufficiency. J Vasc Surg Acta Radiol 39:223226 38:313318 Weinmann HJ (1999) Gadolinium chelates: physico-chemical Schenker MP, Solomon JA, Roberts DA (2001) Gadolinium properties, formulation and toxicology. In: Dawson P, Cos- arteriography complicated by acute pancreatitis and grove DO, Grainger RG (eds) Textbook of contrast media. acute renal failure. J Vasc Interv Radiol 12:393 Isis Medical Media, Oxford, p 228 Spinosa DJ, Matsumo AH, Angle JF, Hagspiel KD, Iassacs R Weinmann JH, Press WR, Gries H (1990) Tolerance of extra- (1998) Gadolinium based contrast and carbon dioxide cellular contrast agents for magnetic resonance imaging. angiography to evaluate transplants for vascular causes Invest Radiol 25:S49S50