Note: This copy is for your personal, non-commercial use only. To order presentation-ready copies for distribution to your colleagues or clients, contact us at www.rsna.org/rsnarights. Delayed Adverse Reaction to Contrast-enhanced CT: CONSTRAST MEDIA n A Prospective Single-Center Study Comparison to Control Group without Enhancement 1 ORIGINAL RESEARCH Shaun Loh , MD , MBA Purpose: To prospectively assess the incidence of delayed adverse Sepideh Bagheri , MD reactions (DARs) in patients undergoing contrast material– Richard W. Katzberg , MD enhanced computed tomography (CT) with the low osmo- Maxwell A. Fung , MD lar nonionic contrast agent iohexol and compare with the Chin-Shang Li , PhD incidence of DARs in patients undergoing unenhanced CT as control subjects. Materials and Institutional review board approval and informed written Methods: consent for this prospective study were obtained. The study was HIPAA compliant. Patients undergoing CT for routine indications were enrolled from a random next-available scheduling template by an on-site clinical trials monitor. All subjects received a questionnaire asking them to indicate any DAR occurring later than 1 hour after their examina- tion. Sixteen manifestations were listed and included rash, skin redness, skin swelling, nausea, vomiting, and dizziness, among others. To ensure maximal surveillance, a clinical tri- als coordinator initiated direct telephone contact for further assessment. Patients suspected of having moderately severe cutaneous reactions were invited to return for a complete dermatologic clinical assessment including skin biopsy, if in- dicated. Statistical analysis was performed by using a two- sided Wilcoxon-Mann-Whitney test, a logistic regression utilizing a x 2 test to adjust for sex and age, and a two-sided Fisher exact test. Results: A total of 539 patients (258 receiving iohexol and 281 not receiving contrast material) were enrolled. DARs were observed in 37 (14.3%) of 258 subjects receiving iohexol and in seven (2.5%) of 281 subjects in the control group ( P , .0001, x 2 test) after adjusting for sex and age. Specifi c manifestations of DARs that were signifi cantly more frequent at contrast-enhanced CT were skin rash ( P = .0311), skin redness ( P = .0055), skin swelling ( P = .0117), and headache ( P = .0246). DARs involving the skin included generalized rashes of the face, neck, chest, back, and extremities and were often associated with swelling, erythema, and pruritus. 1 From the Departments of Diagnostic Radiology (S.L., Conclusion: This study substantiates a frequent occurrence of DARs at R.W.K.), Dermatology (S.B., M.A.F.), Pathology (M.A.F.), and Biostatistics (C.S.L.), University of California, Davis Medical contrast-enhanced CT compared with that in control sub- Center, 4860 Y St, Suite 3100, Sacramento, CA 95817. jects. Continued growth in the use of contrast-enhanced From the 2009 RSNA Annual Meeting. Received October 4, CT suggests a need for greater awareness and attention to 2009; revision requested November 13; revision received prevention and management. November 23; accepted December 10; fi nal version ac- q cepted December 17. Address correspondence to S.L. RSNA, 2010 (e-mail: [email protected] ). Supplemental material: http://radiology.rsna.org/lookup q RSNA, 2010 /suppl/doi:10.1148/radiol.10091848/-/DC1 764 radiology.rsna.org n Radiology: Volume 255: Number 3—June 2010 CONSTRAST MEDIA: Delayed Adverse Reaction at CT Loh et al elayed adverse reactions (DARs) Trying to defi ne the actual occurrence effects with DARs in patients undergo- resulting from the parenteral of late reactions has been diffi cult, and ing unenhanced CT as control subjects. administration of iodinated con- the reported incidence for late reactions D trast agents are more common than to monomeric nonionic low osmolar Materials and Methods previously appreciated. The defi nition contrast media varies from 1% to 23% of DARs has varied somewhat but usu- ( 3 ). In addition, there is the so-called Bracco Diagnostics (Princeton, NJ) has ally means adverse events that begin 1 background noise, whereby anywhere provided fi nancial support for this study. hour or longer after the administration from 3% to 12% of surveyed subjects The authors had control of the data and of the contrast agent. The majority oc- undergoing unenhanced computed to- information submitted for publication. cur 6–12 hours after initial contrast mography (CT) reported adverse re- Our study was approved by the institu- material injection. Other than contrast actions. Indeed, in our survey of the tional review board, and informed writ- material–induced nephropathy, DARs literature, there were only two prospec- ten consent was obtained from each that are of most frequent concern are tive studies ( 4,5 ) that examined DARs patient participating in the study. The cutaneous. The incidence of delayed and compared patients receiving intra- study was Health Insurance Portability adverse cutaneous reactions has been venous contrast material with a similar and Accountability Act compliant. This reported to range from 0.5% to 9%, population undergoing CT examination prospective surveillance was conducted of which some are moderate to severe without contrast material. Interestingly, over the course of 2 years (2006–2008) in distribution and associated symp- these studies showed no overall differ- at a tertiary academic medical center toms ( 1 ). These reactions are often not ence between the subjects receiving and included patients who were re- brought to the attention of the radiolo- intravenous contrast material and the ferred for chest, abdominal, pelvic, gist and are ascribed to other causes control group relative to the occurrence head, neck, or extremity CT with or because contrast agents have a biologic of DARs. Schild et al ( 5 ), however, did without contrast material administra- half-time of only about 1½ hours, are report that delayed adverse cutaneous tion. Patients were excluded if they (a) too small to function unbound as anti- events were noted signifi cantly ( P , were younger than 18 years old, (b) gens, and are minimally protein bound. .05) more often with a dimeric nonionic were pregnant or lactating, (c) had a Because many radiologists are unaware agent than with a monomeric nonionic prior moderate-to-severe adverse reac- that such reactions occur, DARs are of- contrast agent. tion to contrast media, (d) had renal ten mistakenly thought to be caused by The purpose of this prospective clin- dysfunction, (e) had previously partici- another inciting agent. ical surveillance study was to assess the pated in a similar study, (f) were sched- There have been a relatively small incidence of DARs in patients undergo- uled for CT with injection of a contrast number of prospective studies that have ing contrast-enhanced CT with the mo- medium other than iohexol (350 mg of examined the incidence and character- nomeric nonionic low osmolar contrast iodine per milliliter), and (g) anticipated istics of DARs; the fi rst was by Panto material iohexol and to compare these not being available for telephone con- and Davies in 1986 ( 2 ), who reported tact or were unwilling to complete the 5% of subjects experiencing a rash and 14% showing a variety of other reac- Implications for Patient Care tions to high osmolar contrast media. n DARs to contrast-enhanced CT Published online before print occur more often than is gener- 10.1148/radiol.10091848 Advances in Knowledge ally recognized, may be moderate Radiology 2010; 255:764–771 n In this prospective surveillance of in severity, and are often subjects undergoing contrast- ascribed to causes other than Abbreviation: DAR = delayed adverse reaction enhanced CT, we observed a sig- contrast media. nifi cantly ( P , .0001) higher rate n Given the continual and dramatic Author contributions: of delayed adverse reactions increase in the utilization of Guarantors of integrity of entire study, S.L., S.B., R.W.K.; (DARs) in subjects receiving non- contrast-enhanced CT, the impli- study concepts/study design or data acquisition or data analysis/interpretation, all authors; manuscript drafting or ionic monomeric low osmolar cations are for a substantial manuscript revision for important intellectual content, all contrast material (14.3% [37 of occurrence of patient morbidity authors; manuscript fi nal version approval, all authors; 258]) than in a CT control group in need of clinical strategies for literature research, S.L., S.B., R.W.K., M.A.F.; clinical that did not receive contrast treatment. studies, S.B., R.W.K., M.A.F.; statistical analysis, S.L., S.B., C.S.L.; and manuscript editing, S.L., S.B., R.W.K., M.A.F. material (2.5% [seven of 281]). n Our fi ndings should alert radiolo- n The most commonly observed gists about the common occur- Funding: DARs were cutaneous in nature, rence of DARs secondary to This research was supported by the National Center for may be moderate in severity and contrast-enhanced CT and raise Research Resources (grant no. UL1 RR024146) and the Na- duration, and may require the level of awareness of the tional Institutes of Health Roadmap for Medical Research. treatment. referring clinicians. See Materials and Methods for pertinent disclosures. Radiology: Volume 255: Number 3—June 2010 n radiology.rsna.org 765 CONSTRAST MEDIA: Delayed Adverse Reaction at CT Loh et al mail-in questionnaire. After a patient Figure 1 was deemed eligible for participation in the study, the study was described in full, and informed written consent was obtained by a clinical trials moni- tor during regular working hours and during evening hours and weekends by another clinical trials monitor who was employed for 3 months at the midpoint of the study. The patients were ran- domly included in the study. Each pa- tient was approached sequentially and was invited to participate, with each of the daily imaging sessions having been determined by schedulers unaware of the study and routinely employing a next-available scheduling template.