Iodixanol (Visipaque) and iohexol (Omnipaque) are dimeric and monomeric, respectively, non-ionic X-ray contrast media (CM), with well-characterized pharmacokinetics in healthy volunteers. This study was undertaken to study the pharmacokinetics of the contrast media in patients with severely impaired renal function. A total of 16 patients referred for preoperative abdominal angiography were randomized to form two groups of eight patients, receiving either iodixanol 320 mgI ml-1 or iohexol 350 mgI ml-1. Urine and faeces were sampled before the examination and collected quantitatively for five days afterwards, and blood samples were drawn frequently. The concentrations of iodine and contrast medium in urine and in serum, and the amount of iodine in faeces were determined. Mean baseline creatinine clearance was 13.6 and 9.9 ml min-1 1.73 m-2 in the iodixanol and iohexol groups, respectively. Patients in the iodixanol group received on average 0.34 gI per kg bodyweight (bw) and those in the iohexol group 0.39 gI per kg bw. The semilogarithmic plots of serum concentration of CM vs. time indicated elimination according to a two-compartment model. The mean elimination half-life was 23.0 h for iodixanol and 27.2 h for iohexol, and the mean apparent volume of distribution was similar for the two CM, ranging from 0.20 to 0.30 1 per kg bw. Mean plasma clearance of iodixanol was 10.4 ml min-1 1.73 m-2 and 6.9 ml min-1 1.73 m-2 for iohexol, whereas the mean renal clearances were 8.7 and 6.1 ml min-1 1.73 m-2, respectively. Mean faecal recovery was 8.2% for iodixanol and 6.1% for iohexol, and the respective figures for that in urine were 76.1 and 74.8%. Renal clearance of radiolabelled iothalamate, a marker of glomerular filtration rate (GFR), measured simultaneously, indicated that both CM were eliminated by the kidneys by glomerular filtration only. Thus, both media are suitable as GFR markers.