A comparison of low- with high-osmolality contrast agents in cardiac angiography. Identification of criteria for selective use.

BACKGROUND Controversy exists as to whether low-osmolality radiographic contrast agents, which have less detrimental pharmacological effects but are considerably more expensive than high-osmolality agents, should be used universally or only for selected high-risk patients. METHODS AND RESULTS A randomized, double-blind study was used to compare the frequency and severity of adverse events in 2245 consecutive patients undergoing diagnostic cardiac angiography. Two thousand one hundred sixty-six patients were successfully randomized to either iohexol, a low-osmolality contrast agent, or diatrizoate (as Hypaque 76), a high-osmolality agent. The end point event included clinically important adverse events (which jeopardized the patient or required aggressive treatment), contrast agent-related procedure abbreviations, and conversion to open-label contrast agent. Clinically important end point events were associated with increased age, New York Heart Association functional class, left ventricular end-diastolic pressure, arteriovenous oxygen difference, severity of coronary artery disease, and history of a previous reaction to contrast agent. End point events were less frequent in patients receiving iohexol (2.6% versus 4.6%; adjusted odds ratio, 1.59; 95% confidence interval, 0.97-2.60; P = .07). The difference in event frequency between iohexol and diatrizoate was confined to the highest-risk quartile of the patient population. An algorithm was developed to classify patients as being at high or low risk for an event based on patient age, New York Heart Association class, history of a prior contrast reaction, and left ventricular end-diastolic pressure. Application of this algorithm for selective use of low-osmolality agents only for high-risk patients to a theoretical population of 1000 patients reduced contrast agent costs 66% without increasing the frequency of contrast agent-related adverse events. CONCLUSIONS The advantages of low-osmolality contrast agents are clinically important in patients with severe heart disease but are not in less ill patients. Universal use of low-osmolality agents for cardiac angiography in an unselected population is not necessary. Appropriately guided selective use of low-osmolality contrast agents is feasible and has the potential to reduce cost substantially without compromising safety or effectiveness.