Discussion
The 0.07% frequency of acute allergic-like reactions to gadolinium-containing contrast media obtained in this study is similar to that described in the few other large studies in the literature on the subject. Murphy et al. described a 0.1% frequency of allergic-like reaction to gadolinium-containing contrast media involving 21,000 patients over an almost 5-year period. Gadopentetate dimeglumine accounted for 98% of contrast material injections in that study. A retrospective survey involving 53 institutions determined that 241 allergic-like reactions occurred after 825,535 injections of gadolinium-containing contrast media (687,255 gadopentetate dimeglumine doses, 64,005 gadoteridol doses, and 74,275 gadodiamide doses), for a reaction frequency of 0.03%. It is likely that the rate of allergic-like reaction established in that study is an underestimate of the actual rate based on the inherent design of the investigation. A more recent study showed 19 allergic-like events involving 9,528 patients, for a slightly higher reaction frequency of 0.2%. Patients in that study experienced allergic-like reactions after the administration of gadoterate meglumine, gadodiamide, or gadopentetate dimeglumine.
The frequency of acute allergic-like reaction to gadolinium-containing contrast media observed in our study is less than that typically observed for iodinated contrast media in adults. In an investigation of IV-administered nonionic iopromide (Ultravist, Bayer HealthCare) in adult patients, Mortelé et al. described an adverse event rate of 0.7% (of which approximately 90% were allergic-like) in 29,508 patients. A lower frequency of reaction for nonionic iodinated contrast media was shown by Cochran et al. during a 9-year period in which this group observed an adverse event rate of 0.23% (of which approximately 90% were allergic-like). Although the frequency of acute allergic-like reactions to iodinated contrast media is lower in the study by Cochran et al. than that described by Mortelé et al., both rates are substantially higher than those observed in our study of gadolinium-containing contrast agents.
In the pediatric population, the risk of allergic-like reaction to IV gadolinium-containing contrast media also appears to be less than that previously observed for iodinated contrast agents. In 1975, a large study by Gooding et al. involving 12,419 IV administrations of ionic contrast material described a 0.4% frequency of allergic-like reactions in children undergoing excretory urography. In a recent series, Dillman et al. observed a pediatric incidence of acute allergic-like reaction to low-osmolality nonionic iodinated contrast media of 0.18%. Both rates are higher than the 0.04% frequency of reaction noted for pediatric patients in our study.
The pediatric frequency of acute allergic-like reaction to gadolinium-containing contrast media was 0.04% during our study period, and the adult rate was 0.07% (similar to the combined pediatric and adult rates). Thus, the adult frequency of contrast reactions is nearly two times that of the pediatric population, with a relative risk of reaction in adults compared with children of 1.53 (95% CI, 0.66-3.56). Although this difference in frequencies of reaction is not statistically significant, these results mirror an age-related trend observed in previous studies that evaluated the risk of allergic-like reaction to iodine-containing contrast media. The studies by Dillman et al. and Katayama et al., when stratified by age, both showed that pediatric patients have a decreased risk of contrast reaction when compared with adult patients after the IV administration of iodine-containing contrast media. The factors responsible for this difference in rates of reaction between pediatric and adult patients are uncertain.
Most acute allergic-like reactions to gadolinium-containing contrast media in our study were mild. This is not dissimilar to results presented in other studies that evaluated both gadolinium-containing and iodinated contrast material allergic-like reactions. In our investigation, 40 (74%) of the 54 documented acute allergic-like reactions were mild. Most of these mild events were self-limited isolated cutaneous reactions (65%, or 26/40 mild allergic-like reactions) such as urticaria or rash. Five of six pediatric acute allergic-like reactions were mild, but a sixth reaction was severe.
Overall, 10 acute allergic-like reactions were moderate in severity and four were severe. Consequently, 26% of reactions required either medical management, transfer to the emergency department, or hospital admission. It is essential that institutions and individual radiologists who perform contrast-enhanced MRI examinations be prepared to appropriately manage such acute reactions. Acute allergic-like reactions to gadolinium-containing contrast agents are typically treated in a manner identical to that for treating similar reactions after the administration of iodinated contrast material. Although no gadolinium-containing contrast material-related death occurred during the study period, such occurrences have been described.
Twenty-four (52%) of 46 adult patients who experienced acute allergic-like reactions had one or more apparent risk factors for allergic-like contrast reaction, such as a history of previous allergic-like reaction to either gadolinium- or iodine-containing contrast media, a prior allergic reaction to a substance other than contrast media, or documented asthma. Twenty-two percent of adults who experienced an acute allergic-like reaction had more than one risk factor. In comparison, two (33%) of six pediatric patients had an identifiable risk factor, including a patient with a history of prior allergic-like reaction to gadolinium-containing contrast media and a patient with a history of prior allergic reaction to a substance other than contrast medium. No pediatric patient had more than a single risk factor. Four (67%) of six pediatric and 22 (48%) of 46 adult patients had none of the described presumed risk factors.
Interestingly, in our study we found that female patients experienced 63% of adult and 83% of pediatric acute allergic-like reactions. When combining both the pediatric and adult rates, 65% of reactions involved female patients, whereas only 35% of reactions involved males. Although this sex difference is not statistically significant (the relative risk of reaction in women compared with men is 1.57 [95% CI, 0.90-2.74]), it may be a real finding. It is likely that this difference would be significant in a slightly larger patient population. The cause of such a difference between sexes is uncertain.
Both gadopentetate dimeglumine and gadobenate dimeglumine are linear (open-chain) ionic gadolinium chelates. No cyclic or nonionic gadolinium-containing contrast agents were included in this study (only a minimal number of doses of gadodiamide were administered at our institution during the study period). Some evidence suggests that cyclic chelates are generally more stable compounds than linear chelates. This difference in stability has been implicated by some as playing an important role in the pathogenesis of NSF. However, no definitive evidence in the literature suggests that this difference in structure between various gadolinium-containing contrast agents affects the observed rates of allergic-like reactions.
Our study has a few limitations. First, it is possible that certain allergic-like reactions included in this study were not directly related to the IV administration of gadolinium-containing contrast material. Although this is thought to be unlikely, if such an event occurred, it would artificially increase the frequency of reaction established in the study. Second, our results depend on the completion of a contrast material reaction form. If an acute allergic-like reaction was not documented (or if a contrast material reaction form was completed but lost), such a reaction would not be included in our investigation. Such occurrences would falsely lower the frequency of reaction calculated in this study. Third, when attempting to assess apparent risk factors for acute allergic-like reaction to gadolinium-containing contrast media, the prevalence of certain conditions, including history of asthma, history of allergic-like reaction to contrast media, and history of allergic reaction to substances other than contrast media, is not known over our entire pediatric or adult patient populations. Finally, the offending gadolinium-containing contrast agent was not documented for 44% of the reactions included in this study.
In conclusion, acute allergic-like reactions related to IV-administered gadolinium-containing contrast media in both children and adults are rare. They occur less frequently than do reactions to IV-administered iodinated contrast media. Most of these acute allergic-like reactions to gadolinium-containing contrast media are mild; however, moderate and severe reactions do occur. Although reactions were more frequent in adult than in pediatric and in female than in male patients, these differences were not statistically significant. Potential factors that may increase an individual's risk of an acute allergic-like reaction include a history of previous allergic-like reaction to IV-administered contrast medium (either gadolinium- or iodine-containing) and prior allergic reaction to a substance other than contrast media.