Are gadolinium contrast media really less nephrotoxic than iodine agents in radiographic examinations? A comparison in relation to their ability to attenuate x-rays in a pig model

Abstract: Purpose: To confront the statement that gadolinium contrast media (Gd-CM) are less nephrotoxic than iodine contrast media (I-CM) when used in x-ray angiographic and computed tomographic (CT) examinations.

Methods: I) For CT measurements (mean density in Hounsfield numbers) 20-mL syringes filled with I- and Gd-CM at 0.01, 0.02, 0.05 and 0.1 mmol attenuating atoms/mL were scanned in air and in a 30 cm polystyrene phantom. For measurements on radiofluoroscopy (RF), X-ray angiography (XA) and direct digital systems (DX) systems (relative contrast values) 20-mL syringes were filled with 0.5M Gd-CM and I-CM at 35, 50, 70, 90, 110 and 140 mg I/mL. The syringes were placed in phantoms equivalent to 13 (thin) and 20 cm (thick) water. Syringes filled with distilled water served as a constancy reference. II) In a non-crossover design in three separate studies 3 ml of each test solution were injected in 8 pigs/study at a rate of 20 mL/min into the balloon-occluded (10-minutes) right renal artery of left-sided nephrectomized pigs. Test solutions: 1) 0.5M gadopentetate (1.96 Osm/kg H2O), 0.5M gadodiamide (0.78 Osm/kg), 0.5M iohexol (190 mg I/mL; 0.42 Osm/kg), 0.18M iohexol (70 mg I/mL; with an x-ray attenuation equal to that of 0.5M Gd-CM at 80 kVp) and saline; 2) 0.5M gadopentetate, 0.5M gadodiamide, 0.5M iohexol and mannitol solutions iso-osmotic to these CM; 3) 1.0M gadobutrol (1.6 Osm/kg), 0.5M gadodiamide, iodixanol 150 and 320 mg I/mL (290 mOsm/kg) and iopromide 150 mg I/mL (340 mOsm/kg). The plasma half-life elimination time of a GFR-marker were used to compare their effects on glomerular filtration rate 1-3 hours post-injection. III) After the experiments the kidneys were evaluated histomorphologically.

Results: I) In vitro measurements indicate that 0.5M Gd-CM are equal attenuating with 60-80 mg I/mL at commonly used 70-90 kVp for XRA and with 110 mg I/mL at 120 kVp CT using a body phantom. II) Gadopentetate and iso-osmotic mannitol as well as gadobutrol caused severe impairment of renal function. Gadodiamide caused a 90% prolongation of plasma half-life relative to saline, significantly longer than 0.5M iohexol with a 35% prolongation. GFR following injections of iohexol 70, iopromide 150, and iodixanol 150 and 320 were in the same range as that following saline. III) Gd-CM with the highest osmolality caused marked necroses and haemorrhage/congestion correlating with their marked impairment of renal function, while the plasma iso-osmotic I-CM caused no or only minimal changes.

Conclusions: Gd-CM are more nephrotoxic than equal volumes of I-CM resulting in the same or even better attenuation of x-rays. Thus, Gd-CM should not be used as a substitute for I-CM in patients with renal impairment when performing radiographic examinations.

Key words: Angiography; Computed tomography; Contrast media, toxicity; Gadolinium; Glomerular filtration; Iodine; Kidney failure; Nephrotoxicity.

Key words: Angiography; Computed tomography; Contrast media, toxicity; Osmolality; Renal impairment; Gadolinium, Kidney failure; Iodine, nephrotoxicity, equal attenuation, equi-molar, osmotic load, attenuation, half-life, porcine model