Effect of Extrinsic Warming of Low-Osmolality CT Contrast Media (Iohexol 350) on Extravasations and Patient Reaction Rates: A Retrospective Study

Extrinsic warming of iodinated CT contrast media to body temperature reduces viscosity and injection pressures. However, studies examining the effect of extrinsic warming on clinical adverse events are limited in number and provide conflicting results. Therefore, consensus practice recommendations have been sparse. The purpose of this study is to compare rates of extravasation, allergic and allergiclike reactions, and physiologic reactions between iohexol 350 mg I/mL warmed to body temperature (37°C) versus this agent maintained at room temperature. This retrospective study compared adult patients who received CT examinations using IV iohexol 350 that had either been warmed to body temperature or maintained at room temperature. At our institution, contrast media had historically been warmed to body temperature before a protocol change unrelated to this investigation. Information on the patient and CT examination was extracted from the electronic medical record. Adverse events, including extravasations, allergic and allergiclike reactions, and physiologic reactions, were compared between groups. A total of 3939 patients received contrast media warmed to body temperature before the protocol change; 3933 patients received contrast media at room temperature after the protocol change. The body temperature group experienced 11 (0.28%; 95% CI, 0.14-0.50%) adverse events, all extravasations; the allergic and allergic-like reaction rate was 0.00% (97.5% CI, 0.00-0.09%). The room temperature group experienced 17 (0.43%; 95% CI, 0.25-0.69%) adverse events: 13 (0.33%; 95% CI, 0.17-0.56%) extravasations and four (0.10%; 95% CI, 0.03-0.26%) allergic and allergiclike reactions. No physiologic reaction occurred in either group. The two groups were not different in terms of overall reaction rate ( = .19), extravasation rate ( = .69), allergic and allergiclike reaction rate ( = .06), or physiologic reaction rate ( > .99). Logistic regression adjusting for patient and CT characteristics (age, sex, conventional CT vs CTA, contrast media volume, injection location) showed no significant association of patient group and adverse reaction rate (odds ratio, 2.19; 95% CI, 0.68-7.00). Multivariable regression modeling showed an excess of 0.27 adverse events per 100 patients within the room temperature group, which is below a 0.6% noninferiority margin. The data suggest that maintaining iohexol 350 at room temperature is noninferior to warming the agent to body temperature before inj