Evaluation of contrast media used in coronary arteriography.

Coronary arteriography was first performed in dogs in 1933 by injection of a large quantity of con- trast medium into the root of the aorta (Rousth6i, 1933).The same "flush" technique was used in man in 1945 by Radner.The injection of contrast medium directly into each coronary artery consecu- tively, selective coronary arteriography, was de- veloped in 1962, and has been shown to be a rela- tively safe and effective means of obtaining precise information on the distribution of lesions in the coronary arterial tree (Sones and Shirey, 1962).In addition to the hazard ofmechanical damage to the arteries, the injection of contrast medium produces changes in the electrocardiogram (Benchimol and McNally, 1966) and may lead to asystole or ven- tricular fibrillation.The risk of these complications is not great even with selective injections provided effective treatment is readily available, but the mode of production of the electrocardio- graphic changes should be fully understood, and if differences exist between media the safest should be used.We report the results of an experiment designed to test contrast media in current use for their toxic