Previous studies showed that both high-osmolality and low-osmolality iodinated contrast media cause temporary albuminuria and enzymuria (presence of enzymes in urine) in normal rats. Whether the same is true with ionic high-osmolality and nonionic low-osmolality magnetic resonance (MR) contrast media is unknown. We studied urine profiles and histology after intravenous injection of four types of contrast media in rats with normal kidneys. Urine profiles were monitored 4, 24, 48, and 72 hr after intravenous injection of saline, diatrizoate, iohexol, gadopentetate dimeglumine, and gadodiamide (4.59 mmol/kg of body weight) in normal rats. Each group included 20 male rats. After sacrifice, both kidneys were removed for examination by light microscopy (LM) and electron microscopy (EM). All four contrast agents caused a temporary (< 22 hr) increase in the excretion of albumin (2-5 times) and of cytoplasmic (30-100 times) and brush border (10-100 times) renal enzymes when compared with saline. The degree of albuminuria correlated well (r = 0.90) with the osmolality of the injected media, whereas the increased level of enzymuria was unrelated to the osmolality. No major differences in the enzymuric effects of the four agents were noted. LM revealed vacuoles in all kidneys exposed to radiologic contrast media but not in kidneys exposed to MR contrast media or saline. Slight vacuolation was revealed by EM after the use of MR contrast media, and significant vacuolation was evident via EM after the use of radiologic contrast media. No difference between ionic and nonionic media within each drug group was detected by either LM or EM. Transient renal effects are induced by both ionic and nonionic high-osmolality and low-osmolality radiologic and MR contrast media in normal rats. Both osmotic (e.g., albuminuria) and chemotoxic (e.g., enzymuria) mechanisms seem to be involved. From a morphologic point of view, the chemotoxic mechanisms seem to be of major importance.