Subsequent MRI of pediatric patients after an adverse reaction to Gadolinium-based contrast agents

Background Gadolinium-based contrast agent (GBCA)-enhanced magnetic resonance imaging (MRI) scans often must be used repeatedly in pediatric oncologic patients. Although GBCAs are usually well tolerated, severe and life-threatening allergic reactions might occur, which can result in overly cautions adherence to special precautions in patients. Purpose To evaluate the management of the reported GBCA-associated adverse reactions in subsequent contrast-enhanced MRIs in pediatric patients, distinguishing non-allergic and allergic reactions. Materials and methods In this retrospective, cross-sectional study, consecutive pediatric neurooncological patients who underwent GBCA-enhanced MRI at our university hospital, between 2007 and 2016, were eligible. The patients’ history was evaluated with regard to any adverse events after GBCA administration. In a subset of patients with reported adverse reactions, the institutional premedication regime or an allergy work-up in clinical practice were performed, using either skin-prick tests or intravenous provocation tests in a double-blind procedure. Results Included were 8156 contrast-enhanced MRI scans in 2109 patients. Nineteen acute adverse events (19/8156; 0.23%) in 17 patients (17/2109; 0.81%) were reported. Despite a premedication regime in 14 patients, three patients (3/14; 21.4%) reported a breakthrough reaction. None of the 12 patients who underwent skin-prick tests or intravenous provocation tests showed allergic reactions. At least one well-tolerated GBCA was identified in almost every tested patient. Conclusion A fast-track allergy work-up can help to distinguish non-allergic and allergic reactions and to identify a well-tolerated GBCA, thus avoiding unnecessary premedication for subsequent GBCA administrations.

Introduction (NSAIDS), local anesthetics, non-beta lactam antibiotics, and other drugs for which safer tests do not exist or are not standardized [17].
To our knowledge, there are two reported experiences about intravenous provocation testing to rule out GBCA-associated allergic reactions [18] and iodine contrast agent-associated allergic reactions in adult patients [19].
The purpose of this retrospective study was to evaluate the management of the reported GBCA-associated adverse reactions in subsequent contrast-enhanced MRIs in our pediatric center, distinguishing non-allergic and allergic reactions.

Materials and methods
The institutional review board of Medical University of Vienna approved the study and waived the requirement for informed patients' and parental consent, because the study was a retrospective data analysis. (IRB 1321/2018). All procedures performed in the study that involved human participants were in accordance with the ethical standards of the institutional review board and with the 1964 Helsinki declaration and its later amendments.
All patients of the Department of Pediatrics and Adolescent Medicine with intracranial lesions, who underwent an MRI scan in the Department of Biomedical Imaging and Imageguided Therapy between January 2007 and December 2016, were reviewed retrospectively. Patients with GBCA-enhanced MRI scans were evaluated with regard to the presence or absence of adverse reactions and their severity according to the ACR and the ESUR guideline classification system, and reactions were categorized as mild, moderate, or severe [3,4].
In the case of a reported GBCA-associated adverse reaction, there are three methods in our pediatric center with which to manage the subsequent MRI: either the institutional pediatric premedication regime, modified according to Greenberger et al. [3,10], or one of the two allergy work-ups, which are used in clinical practice, either by skin prick testing or by intravenous provocation tests.

Institutional pediatric premedication regime
Our institutional premedication guidelines for the prevention of allergic-like reactions in children include a combination of intravenous prednisone 0.5-0.7 mg/kg (up to 50 mg) and intravenous diphenhydramine 1.25 mg/kg (up to 50 mg) one hour prior to contrast agent injection [3,10].

Skin-prick test
Skin testing was typically performed with a set of four GBCAs (Gadoterate meglumine, Gadobutrol, Gadoteridol, and Gadobenate dimeglumine) for optimal evaluation of potential crossreactivity and identification of alternatives. Briefly, skin-prick tests were performed with the undiluted, commercially available solution [14]. Evaluation for reactions to GBCAs was performed 15 minutes after skin-prick tests, and, for non-immediate hypersensitivity reaction, there was a delayed reading of the skin-prick tests. The immediate-reading skin-prick test was considered positive if the size of the wheal was at least 3 mm in diameter, with surrounding erythema after 15 minutes. Immediate and optional delayed reading of skin-prick tests was performed according to the international guidelines of the European Society of Contact Dermatitis [20].

Intravenous provocation test
Patients were provoked in a double-blind procedure with two different intravenous GBCAs on two different days, each preceded or followed by a placebo. Patients were informed that they would receive placebos, but were not told when they would be administered. The intravenous provocation was performed with the routinely administered doses of each GBCA. All provocation tests were performed under strict hospital surveillance, with emergency room facilities equipped to handle any anaphylactic reactions.
Patients were observed for at least 120 minutes after the last intravenous administration. Vital signs and physical examination were performed at baseline and every 30 minutes. If a reaction occurred, vital signs and physical examination were repeated.
The adverse reactions were classified according to the recommendation of the ESUR guidelines on contrast agents into hypersensitivity/allergy-like reactions, including cutaneous, respiratory and cardiac reactions, or chemotoxic/non-allergic reactions, including gastrointestinal and vasovagal reactions, which resolved spontaneously with no changes in vital signs, cardiac arrhythmia or convulsions [3,4].

Statistics
Descriptive statistics only were used for this study.
In total, 8156 GBCA-enhanced MRI examinations and 1669 non-contrast MRIs were performed during this observation time (Table 1).
In our collective, acute adverse reactions were reported in 17 patients (17/2109; 0.81%; 11 males and six females, mean age, 10 ± 4.47 years; range, 3-17 years at the first adverse reaction), either by the patients, parents, or radiologists. Two of these patients (patients 2 and 5) reported two acute adverse reactions after two different GBCAs (Table 2). In total, there were 19 acute adverse reactions reported after 8156 GBCA i.v. administrations (0.23%).
Twelve patients underwent an allergy work-up. There were three patients with a breakthrough reaction, eight patients after a premedication regime without a breakthrough reaction,  and one patient with no premedication ( Table 3). The allergy work-up included skin-prick tests in three patients and intravenous provocation tests in ten patients (patient 6 had both tests). Two patients (patients 7 and 15) moved away, and the parents of three patients (patients 1, 4, and 11) refused an allergy work-up (Table 3). Skin-prick tests in three patients for Gadoterate meglumine, Gadobutrol, Gadoteridol, and Gadobenate dimeglumine revealed skin-test negativity in all three patients, and in one of the patients a hypersalivation to Gadoteridol was reported (patient 6), which is a chemotoxic response, according to the ESUR guidelines [4]. Although the patients showed skin-test negativity, in clinical practice, only patient 9 had uneventful subsequent GBCA-enhanced MRIs without any premedication. Patient 2 still received premedication before GBCA administration and patient 6 underwent intravenous provocation tests (Table 3).
Twenty-one intravenous provocation tests were performed in ten patients for at least two substances-either Gadoterate meglumine, Gadobutrol, or Gadobenate dimeglumine ( Table 3).
None of the patients who underwent intravenous provocation tests showed any sign of a hypersensitivity/allergic-like reaction. Two patients (patients 3 and 5) exhibited dizziness, nausea and flush, and one patient (patient 10) experienced vomiting (Table 3), which are all categorized as chemotoxic responses, but not as allergic-like reactions, according to European Society of Urogenital Radiology (ESUR) guidelines on contrast agents [4]. None of the patients needed any medication thereafter or showed signs of renal failure in follow-up examinations. Only in patient 3 were both tested GBCAs not well tolerated. In all other provoked patients, at least one well-tolerated GBCA without any reaction could be identified. All but one of the provoked patients had uneventful subsequent GBCA-enhanced MRIs without any premedication.

Discussion
In our patient collective with reported GBCA-associated adverse events, there was no confirmed allergic reaction to GBCA, either with skin-prick tests or with intravenous provocation tests, although, the 0.23% incidence of GBCA-associated adverse events in our study was within the range of the published incidences in earlier pediatric studies, where an adverse reaction frequency of 0.04% -19.3% was reported [21][22][23]. The administration of the well-tolerated GBCA identified by intravenous provocation tests or skin-prick tests in subsequent MRI examinations was uneventful.
Repeated GBCA administration is necessary in several clinical indications, including in the diagnosis and follow-up of pediatric oncological patients. Acute, life-threatening adverse reactions to GBCAs are rare, but have occurred, even though GBCAs are well tolerated by most patients [22][23][24][25].
A substantial number of patients with a history of prior reaction to GBCAs require additional contrast-enhanced examinations, which would expose them to the same or a similar contrast agent [8]. Self-reported allergy is always subjective, and, in many cases, inaccurate [26,27]. Nevertheless, a suspected allergy to contrast media might escalate into a life-threatening anaphylactic event [28], leading to uncertainty in radiological departments, which usually establish special precautions to deal with these patients.
In patients with a previous adverse reaction to GBCA, the risk of a repeated adverse event is increased [3,4,6]. In many institutions, prophylactic premedication with corticosteroids and antihistamines is administered, to reduce the risk of allergic-like reactions in patients with a history of previous adverse events [3,9]. However, adverse events to GBCA have been reported to occur despite premedication with corticosteroids and antihistamines, the so-called "breakthrough reactions" [7,8]. In our series, we observed breakthrough reactions in 21.4% of the patients after premedication, which required further evaluation.
When an allergic-like reaction is suspected, skin testing may be useful to confirm an allergy and to identify alternative agents for subsequent studies [29]. Skin-testing revealed an excellent negative predictive value in patients with suspected GBCA hypersensitivity [30]. Although skin-testing can identify safe alternatives for GBCA re-exposure, and potentially discriminate between allergic and non-allergic reactions, the interpretations of the allergist, the radiologist, and the patient are often not well aligned [15]. Similar work in patients with penicillin allergy indicates that the allergy label often persists despite a negative work-up [16]. Patients with negative skin test results continued to avoid contrast media, mainly because of personal concerns or the radiologist's concerns [15]. One reason might be that skin tests do not reflect the reallife situation, as we also observed in two patients who refused a GBCA-enhanced subsequent MRI, even though no allergic reaction was confirmed by skin testing.
The European Academy of Allergy and Clinical Immunology Executive Committee considers drug provocation tests the gold standard for the diagnosis of an allergy to drugs [17]. In our study, the intravenous provocation testing was performed in ten patients under strict hospital surveillance with emergency room facilities. The intravenous provocation tests were well tolerated by all ten patients. None of the provoked patients showed symptoms attributable to an allergic reaction. Three patients reported dizziness, nausea, or vomiting, which were classified as clinically mild chemotoxic, non-allergic adverse reactions. We were able to identify at least one GBCA in nine of ten provoked patients that caused no reaction at all using intravenous provocation tests. All but one of the provoked patients had an uneventful, contrastenhanced subsequent MRI with the well-tolerated GBCA identified by the intravenous provocation test. Based on our experience, the intravenous provocation test led to a remarkable reduction of concerns in patients, parents and radiologists, probably due to the better simulation of the real-life situation and uneventful subsequent MRI examinations with a tested GBCA. In a recent study, the experience with an intravenous provocation test with which to rule out a GBCA associated allergic reaction in adult patients was reported [18].
Because of the retrospective character of this study, our single-center study contains several limitations. The flow rate of the GBCA injection during the MR imaging (manually or via power injection) could not be verified, which has been discussed with regard to the appearance of acute dyspnea after i.v. administration of Gadoxetate disodium [31]. There was no blood sample collection within one hour after the acute adverse reaction [14]. Since reactions to Gadolinium are very rare, skin-prick tests and intravenous provocation tests could be performed in only a small sample size. Both tests, the skin-prick tests and the intravenous provocation tests were applied in one patient only; otherwise, the tests were performed in different patient groups, so these could not be compared in the same patients.
A prospective, multi-center study could verify our study results.

Conclusion
In conclusion, a fast-track allergy work-up can help to distinguish non-allergic adverse reactions from allergic reactions in pediatric patients with reported GBCA-associated adverse events, and to identify a well-tolerated GBCA, thus avoiding unnecessary premedication for subsequent GBCA administrations.