Cost has been the major factor preventing the universal conversion to non-ionic contrast agents. We assessed the costs and potential benefits of making this change in our department. During a 10-month-period the use of all intravascular contrast agents and reaction rates were audited prospectively. One thousand three hundred and ninety-four examinations were included. A local protocol for the use of ionic and non-ionic contrast media was already in place, based on the Royal College of Radiologists Guidelines. For each patient the contrast agent used, risk factors, and presence or absence of a contrast reaction were recorded. Non-ionic contrast agent usage exceeded ionic by a factor of 10. Patients receiving ionic agents intravenously had a reaction rate of 16.8% compared with 2.7% for non-ionic contrast media. The use of ionic contrast media was subsequently suspended and the effect of this on overall costs assessed by retrospectively and prospectively analysing hospital expenditure on contrast agents. No increase in costs was found. The reasons for this are two-fold. Firstly in our institution the protocol in place and the nature of the workload resulted in relatively small volumes of ionic contrast media being used compared with non-ionic agents. Secondly, since our department is a bulk purchaser of non-ionic agents, substantial discounts could be negotiated. The benefits of non-ionic contrast media are well recognized and our experience suggests that cost may no longer be a barrier to conversion to these lower risk agents.