Current Epidemiology and Management of Radiocontrast-Associated Acute- and Delayed-Onset Hypersensitivity: a Review of the Liter
Total Page：16
File Type：pdf, Size：1020Kb
REVIEW ARTICLE Current Epidemiology and Management of Radiocontrast-Associated Acute- and Delayed-Onset Hypersensitivity: A Review of the Literature Eric M Macy, MD, MS Perm J 2018;22:17-072 E-pub: 12/29/2017 https://doi.org/10.7812/TPP/17-072 ABSTRACT China. They prospectively collected data Radiocontrast-associated acute-onset hypersensitivity reactions now occur less frequently on all cases. They identified 506 (0.4%) than before 1990, when high-osmolar, ionic, radiocontrast agents were widely used. Premedi- hypersensitivity reactions, of which 90.0% cation with corticosteroids and antihistamines does not reliably prevent recurrent low-osmolar were mild, 7.7% were moderate, and 1.4% radiocontrast-associated acute hypersensitivity reactions. Corticosteroid prophylaxis for acute were considered severe. Hypersensitivity hypersensitivity currently causes more morbidity than benefit. The specific radiocontrast agent reactions were more common with iso- that is associated with a patient’s adverse reaction must be displayed in the drug intolerance osmolar agents than with hypo-osmolar or drug “allergy” field of their electronic health record to enable effective management and radiocontrast media. Risk factors for acute prevention of future reactions. The term iodine allergy should never be used in the context of reactions included previous acute reac- radiocontrast-associated adverse reactions because it leads to poorer clinical outcomes. The tions, asthma, doses higher than 100 mL, time to onset of the reaction and the nature of the reaction must be noted in enough detail in and injection rates higher than 5 mL/s. the drug intolerance comment fields in the electronic health record to determine the potential The authors were unable to reliably col- mechanism for the reaction and to enable selection of the appropriate radiocontrast material lect delayed-onset reaction data because for future exposures. Most individuals with a history of radiocontrast agent hypersensitivity many patients were in the facility only for can be effectively managed by selecting an alternative radiocontrast agent, without any the procedure. premedication. Radiology Departments, catheterization laboratories, and all physicians who We reported in 2012 that of 2,375,424 use parenteral radiocontrast media must have management plans in place to treat severe Kaiser Permanente Southern California acute reactions when they occur. Patients should be informed that delayed-onset reactions, (KPSC) Health Plan members who had a mostly benign rashes within one week of exposure, are as common or more common than health care visit and at least 11 months of acute reactions. Future radiocontrast-associated acute and delayed-onset reactions can be health care coverage during 2009, a total minimized, but never completely avoided, by using an appropriate alternative agent. of 0.5% of females and 0.3% of males had a radiocontrast agent “allergy.”7 During INTRODUCTION benign delayed-onset, and severe delayed- 2009 a new radiocontrast “allergy” was The epidemiology and optimal manage- onset. Acute onset, which occurs less than reported in 0.1% of females who had at ment of radiocontrast-associated adverse one hour after exposure, is typically caused least 1 other drug “allergy,” in 0.1% of reactions have changed dramatically since by mast cell activation, either directly or, males who had at least 1 other drug “al- 1985, with the almost exclusive current use rarely, is immunoglobulin E (IgE) medi- lergy,” 0.04% of females who had no other of low-osmolality nonionic radiocontrast ated.3 Delayed onset is defined as starting drug “allergy,” and in 0.02% of males who agents and very low rates of use of high- more than one hour but typically starting had no other drug “allergy.” Individuals osmolality ionic radiocontrast agents.1 more than three hours and up to two to with any drug “allergies” were more likely This review will concentrate on reports five days after exposure; these reactions to use more health care services and thus published since 2010. The goal of this re- are thought to be T-cell-mediated, de- might be more likely to be exposed to view is to add clarity and specificity to the layed-type hypersensitivity.4 These reac- radiocontrast media. Between January general suggestions given in the American tions rarely rise to the level of a serious 1, 2014, and April 30, 2017, there were College of Radiology’s ACR Manual on cutaneous adverse reaction, such as toxic 372 serious, acute-onset, radiocontrast- Contrast Media, Version 10.3, last updated epidermal necrolysis or Stevens-Johnson associated reactions reported by KPSC on May 31, 2017.2 syndrome.5 Radiology Departments, 335 (90.1%) as- sociated with iohexol, 19 (5.1%) associated REVIEW OF THE LITERATURE Epidemiology with iodixanol, 1 (0.3%) associated with Radiocontrast Agent Hypersensitivity Li and coworkers6 in 2016 reported on diatrizoate, and 17 (4.6%) with the associ- Mechanisms 120,822 individuals receiving iopromide, ated radiocontrast agent not reported. It is There are four general categories of iodixanol, iopamidol, ioversol, iobitridol, or currently not possible to accurately iden- radiocontrast-associated hypersensitivity iohexol between January 2014 and March tify all exposures to radiocontrast agents reactions: benign acute-onset, anaphylaxis, 2016 at a single institution in Chongqing, throughout the entire KPSC health care Eric Macy, MD, MS, is an Allergy Specialist and Researcher in the Department of Allergy at the San Diego Medical Center in CA. He is a Partner Physician with the Southern California Permanente Medical Group, and an Assistant Clinical Professor of Medicine at the University of California, San Diego. E-mail: [email protected]. The Permanente Journal/Perm J 2018;22:17-072 1 REVIEW ARTICLE Current Epidemiology and Management of Radiocontrast-Associated Acute- and Delayed-Onset Hypersensitivity: A Review of the Literature system or to capture all new acute and Table 1. Radiocontrast agent use in Kaiser Permanente Southern California, delayed-onset reactions because of the 2014 through 2016 poor quality of the adverse drug reaction Agent 2014 2015 2016 Total reporting in the electronic health record Iohexola (mL) 19,088,565 20,385,128 22,064,825 61,538,518 (EHR). Even when reported, the specific Iodixanolb (mL) 1,650,500 1,361,000 1,352,500 4,364,000 radiocontrast agent implicated is virtu- Total (mL) 20,739,065 21,746,128 23,417,325 65,902,518 ally never noted in the EHR, nor are the Approximate exposures 414,782 434,922 468,346 1,318,050 symptoms described in enough detail to at 50 mL per exposure confidently determine a mechanism. It a Typically used at 10 mL to 100 mL per exposure. was, however, possible to identify how b Typically used at 10 mL to 250 mL per exposure. much radiocontrast medium was pur- chased each year for KPSC and then estimate annual exposures. The amount cases had any previous exposure to radio- subsequent radiocontrast exposures, with of iohexol and iodixanol used annually in contrast agents. The authors concluded a median of 2 exposures and a range of 1 KPSC in 2014 through 2016 is displayed that “risk” factors for fatal anaphylaxis to 11. There were 19 individuals (38.0%) in Table 1. In KPSC we annually used included any history of asthma, allergic who had at least 1 additional episode of about 11.5 to 16 times as much iohexol rhinitis, atopic dermatitis, multiple al- radiocontrast-associated hives, for a total as iodixanol. We had approximately 1 lergies, drug allergy, food allergy, previ- of 26 events (19.5%) in the 133 imaging reported severe acute reaction for ev- ous radiocontrast exposure, β-blocker or studies. Paradoxically, individuals who ery 183,697 mL (about 3674 exposures nonsteroidal anti-inflammatory drug use, were premedicated were more likely to [range, 1837-18,370]) of iohexol and 1 and any preexisting condition including have hives with subsequent exposures. reported severe acute reaction for ev- any cardiovascular, renal, hematologic, The was no premedication given before 89 ery 229,684 mL (about 4594 exposures autoimmune, or metabolic disease. This (66.9%) of the scans. Those premedicated [range 1199-22,968]) of iodixanol. list is of questionable utility, with almost with diphenhydramine had an adjusted Scheinfeld and colleagues8 at Albert as many “risk” factors as reported cases. odds ratio of 1.2 (95% confidence interval Einstein College of Medicine in New There have been only rare reported = 0.2-7.3, p = 0.85). Those premedicated York, NY, reported in 2014 that of cases of radiocontrast-associated serious with corticosteroids had an adjusted odds 927,000 total “allergies” documented dur- cutaneous adverse reactions, specifically ratio of 14.3 (95% confidence interval = ing a 10-year period in their EHR, virtu- Stevens-Johnson syndrome or toxic epi- 4.1-50.4, p < 0.0001). Those premedi- ally none of the more than 7000 patients dermal necrolysis.5 There has been one cated with corticosteroids and diphen- with “allergies” reported to “contrasts” case of recurrent iopromide-associated hydramine had an adjusted odds ratio of or “iodine” had a specific radiocontrast Stevens-Johnson syndrome reported, 8.3 (95% confidence interval = 1.8-37.9, agent listed. with three distinct episodes.11 p = 0.006). The authors concluded that Dean et al9 reported in 2015 that premedication may not be necessary, but adverse reactions after radiocontrast- Prevention of Recurrent radiology personnel need to be aware of enhanced computed tomography (CT) Radiocontrast-Associated Reactions prior reaction history and be knowledge- scans were reported at a lower overall rate Kolbe and coworkers12 at the Mayo able in recognition and treatment of these in inpatients compared with outpatients, Clinic in Rochester, MN, reported in reactions. but the reactions reported were more se- 2014 data from 245 individuals with Mervak and coworkers13 in 2015 re- vere. Less than 10% of the reported reac- reactions (0.08%, or 1 in 1222), of ported on 626 inpatients with a history tions were delayed onset.