ORIGINAL ARTICLE Asian Pacific Journal of Allergy and Immunology Comparison of immediate hypersensitivity reaction rate according to the type of radiocontrast media Hye Jung Park,1* Hye Sun Lee,2 Sinae Kim,2 Chul Hwan Park3 Abstract Background: Various types of radiocontrast media (RCM), which can be substituted for each other, are used in clinical settings. However, data on comparisons of hypersensitivity reactions (HSR) based on types of RCM are lacking. Objectives: We aimed to compare HSR rates to facilitate the choice of appropriate RCM types. Methods: From January 2011 to December 2015, we retrospectively reviewed 209,027 clinical events involving seven different RCM in Gangnam Severance Hospital. We analyzed immediate HSR events using an internally developed adverse drug reaction reporting system. Furthermore, we defined risk factors for HSR, including RCM type. Results: During the 5-year study period, 209,027 RCM usage occurred in 97,374 patients. During that period, 108 HSR happened in 88 patients (0.05% events). The HSR rate for iodixanol (0.16%) was the highest, followed by that for iopamidol (0.08%), iopromide (0.04%), iobitridol (0.04%), ioversol (0.02%), ioxitalamic acid (0.01%), and iohexol (0.01%). Multivariate generalized estimating equation analysis showed age (odds ratio [OR], 0.980; 95% confidence interval [CI], 0.969-0.990; P < 0.001), urticaria (OR, 1.936; 95% CI, 1.003-3.737; P = 0.049), drug allergy (OR, 9.212; 95% CI, 5.353- 15.853; P < 0.001), iopamidol use (OR, 8.327; 95% CI, 1.129-61.401; P = 0.038, compared to iohexol), iodixanol use (OR, 18.780; 95% CI, 2.564-137.560; P = 0.004, compared to iohexol), and use in 2014-2015 (OR, 2.704; 95% CI, 1.780-4.108; P < 0.001) were significant HSR-associated risk factors. Conclusion: Among the RCMs, iodixanol (iso-osmolar RCM) and iopamidol (low-osmolar RCM) showed a significantly higher risk for immediate HSR than did other agents. Key words: contrast media, hypersensitivity, iodixanol, iopamidol, iopramide From: * Corresponding author: 1 Department of Internal Medicine, Gangnam Severance Hospital, Hye Jung Park Yonsei University College of Medicine, Seoul, Korea Department of Internal Medicine, Gangnam Severance Hospital, 2 Biostatistics Collaboration Unit, Yonsei University College of Medicine Yonsei University College of Medicine, 211 Eonju-ro Gangnam-gu, 3 Department of Radiology, Gangnam Severance Hospital, Seoul 06273 Yonsei University College of Medicine, Seoul, Korea E-mail: [email protected] Introduction The use of radiocontrast media (RCM) has increased along iso- and low-osmolar RCM, which can be easily substituted with the increase in the use of imaging modalities, especially each other, have been used. computed tomography (CT).1 The hypersensitivity reaction However, recent large-scale studies have revealed that ad- (HSR) rate associated with RCM is known to be fairly high;2,3 verse drug reaction (ADR) rates differ significantly according therefore efforts to reduce the HSR rate has been the focus of to the type of RCM. Gomi et al.14 compared the incidence of numerous studies. Non-ionic RCM were introduced to replace acute ADRs among five low-osmolar RCM including iomeprol, ionic RCM, which reduced the HSR rate.4,5 Moreover, the HSR iopamidol, iohexol, iopromide, and ioversol. They discovered rate reduced consderably after high-osmolar RCM were re- that the use of iopamidol, iohexol, and ioversol significantly placed by iso- and low-osmolar types.6-8 Further, there have reduced the incidence of ADRs. Kim et al.15 reviewed 286,087 been reports that HSR rates among iso- and low-osmolar RCM cases of low-osmolar RCM use and found that compared to were not significantly different.9-13 To date, various types of iohexol, iopamidol, and iopromide, iobitridol significantly Asian Pac J Allergy Immunol DOI 10.12932/AP-191017-0182 reduced immediate HSR. Morales-Cabeza et al.16 showed im- skin-mucosal tissue, respiratory compromise, reduced BP or mediate reaction was significantly lower in iopamidol and io- associated symptoms, or persistent gastrointestinal symptoms) versol rather than iomeprol. To date, data on the comparison that occur rapidly after exposure to a likely allergen; reduced BP of HSR rates based on the type of RCM are lacking. Therefore, after exposure to known allergen.20 we aimed to compare the HSR rate to facilitate the selection of appropriate and safe RCM in clinical practice. Statistical analysis We used the Student’s t-test with an analysis of variance Methods (ANOVA) and chi-square tests to identify any differences in Data collection continuous and categorized variables between groups, respec- We retrospectively reviewed 215,135 incidences of tively. Univariate and multivariate analyses were used to deter- RCM use that occurred at Gangnam Severance Hospital from mine significant factors associated with HSR using a logistic January 2011 to December 2015. Gangnam Severance Hospi- regression analysis and generalized estimating equation (GEE) tal, which is a branch of Severance Hospital, implemented a analysis. The Statistical Package for the Social Sciences (SPSS) previously described spontaneous reporting program at Sever- v18.0 (IBM Corp, Armonk, NY, USA) was used, and a P < 0.05 ance Hospital to report ADRs from 2005.15,17 Doctors, nurses, was considered statistically significant. and paramedics have spontaneously reported ADRs using this program. We analyzed HSR events among ADRs in patients in Ethics whom RCM were used, by using this internally developed ADR This study was exempted from requiring an approval by the reporting system. Institutional Review Board of the Gangnam Severance Hospi- tal, Yonsei University Health System (approval number: 3-2016 RCM analyzed in this study -0207) The RCM investigated in this study were iohexol (Bonorex® 300, GE Healthcare, Amersham, UK), ioxitalamic acid (Tele- Results brix® 30, Guerbet, Sulzbach, Germany), ioversol (Optiray® 320, Clinical characteristics Guerbet, Sulzbach, Germany), iobitridol (Xenetix® 300, Guerbet, During the 5-year study period, 209,027 incidences of RCM Sulzbach, Germany), iopromide (Ultravist® 300, Bayer, Leverku- use occurred in 97,374 patients. During that period, 108 HSR sen, Germany), iopamidol (Pamiray® 300, Bracco, Milan, Italy), events occurred in 88 patients (0.05% events). We divided the and iodixanol (Visipaque®, GE Healthcare, Amersham, UK). patients into subjects without and with HSR (n = 97,286 and 88, Among them, ioxitalamic acid is an ionic RCM and iodixanol is respectively). Age, sex, and the prevalence of asthma, allergic a non-ionic iso-osmolar RCM. Except for these two, the others rhinitis, atopic dermatitis, hypertension, diabetes mellitus, liver are non-ionic low-osmolar RCM. disease, and chronic renal disease were not significantly differ- ent between the two groups. However, prevalence of urticaria Definition of HSR and severity of HSR (10.2%) was significantly higher in subjects with HSR than that HSR was defined as a particular condition causes the im- in subjects without HSR (3.3%, P = 0.003). The prevalence of mune system to overreact and included all allergic reactions food allergy (3.4%) and drug allergy (19.3%) was also signifi- (which is defined as one type of HSR when immune system cantly higher in subjects with HSR than that in subjects without overreacts to harmless substances) and various non-allergic re- HSR (0.6 and 1.2%), respectively (Table 1). actions according to American Academy of Allergy Asthma & Immunology definition.18 Immediate HSR was defined as a HSR Clinical manifestations of HSR occurring within 1 hour after administration of a RCM. Among the 108 HSR events, cutaneous symptoms including The HSR severity was classified as follows: mild (itching, rash, pruritus, and urticaria were the most frequent manifes- urticaria, flushing, nausea, mild vomiting, and/or chills, i.e. tations (88.9%). Nausea/vomiting (13.0%), angioedema (5.6%), symptoms that are self-limited, without evidence of progres- and dyspnea (3.7%) were also frequently observed. Chest pain, sion), moderate (severe vomiting, systemic urticaria, mild unconsciousness, and shock were rarely observed (0.9%, Table bronchospasm, face and/or larynx angioedema, throat tight- 2). ness, tachycardia, bradycardia, mild dyspnea, hypoxia, and/or temporary hypotension, i.e. symptoms that are more pro- HSR rates according to time nounced and commonly require medical management), and The HSR rates in 2014 and 2015 (0.09% and 0.07%) were severe (shock, pulmonary edema, prolonged hypoxia, cardio- significantly higher than those in 2011-2013 (0.03%, P < 0.001). pulmonary arrest, and/or convulsions, i.e. symptoms that are Further, upon categorizing them into 2011-2013 and 2014- often life-threatening and can result in permanent morbidity or 2015, the differences were still significant (Table 3). death if not managed appropriately).19 In addition, we defined anaphylaxis as a serious allergic reaction that is rapid in onset HSR rates according to the type of RCM and may cause death , when any one of the following 3 criteria The HSR rate for iodixanol (0.16%) was the highest, followed are fulfilled: acute onset of an illness with involvement of the by that for iopamidol (0.08%), iopromide (0.04%), and iobitri- skin, mucosal tissues, or both and other sympotms (respisratory dol (0.04%). Ioversol (0.02%), ioxitalamic acid (0.01%), and compromise or reduced BP or associated symptoms of end-or- iohexol (0.01%) were associated with relatively low incidences gan dysfunction); two or more