Hypersensitivity to Gadolinium-Based Contrast Media

Gadolinium-based contrast agents (GBCAs) are frequently used in magnetic resonance imaging (MRI) examinations to increase sensitivity in diagnoses. Recently, an increase in the description of hypersensitivity reactions to GBCAs has been detected. We performed research in PubMed, PubMed, SCOPUS, and EMBASE until September 2021, searching for studies regarding immediate and delayed hypersensitivity reactions to gadolinium-based contrast agents in which an allergy study was performed. The initial research identified 149 articles written in English. After excluding articles duplicated and articles that had irrelevant designs, 26 articles were included. Finally, 17 studies concerning immediate reactions, six studies concerning non-immediate reactions, and three concerning both that performed allergy evaluations were selected. In the review, we analyzed the characteristics of immediate and delayed reactions and the results of the allergy study and cross-reactivity. Skin tests seem to have acceptable accuracy, but drug provocation tests are still needed when skin tests are negative o to find alternative agents. Although cross-reactivity patterns are not well established, cross-reactivity seems to exist among macrocyclic agents. Notwithstanding, the number of patients analyzed is low and further studies are required. A management algorithm is suggested.


INTRODUCTION
Gadolinium-based contrast agents (GBCAs) are used in radiology to increase the sensibility and specificity of magnetic resonance imaging (MRI) examinations. After approval of gadopentetate dimeglumine by the US Food and Drug Administration (FDA) in 1988, the use of GBCAs has multiplied, aiding lesion depiction and therapeutic guidance in over 500 million patients worldwide (1).
As gadolinium administered directly is toxic, all GBCAs add to their composition a chelating substance of linear or cyclic morphology that binds to gadolinium and improves the stability, solubility, and safety of the central gadolinium heavy metal ion. Thus, GBCAs are categorized as linear or macrocyclic based on the molecular structure of the organic ligand and as non-ionic or ionic based on their net charge in solution (2) (Figure 1; Table 1). The most used GBCAs are extracellularly distributed and mainly eliminated via the kidneys, with some liver excretion demonstrated for a few agents.
The adverse event rate for GBCAs administered at clinical doses (0.1-0.2 mmol/kg) ranges from 0.07 to 2.4% (3). Most adverse events are type A, generally mild reactions, including coldness, warmth, or pain at the injection site; nausea, vomiting; headache; paresthesias; dizziness; myalgia and arthralgia; and thrombophlebitis. If the patients are also in a situation of great stress, it is not infrequent that they can suffer a vasovagal syncope (bradycardia, hypotension, paleness, weakness, nausea, vomiting, and sweating), making it difficult to distinguish it from an allergic reaction. Type A reactions are probably caused by the direct toxicity of the contrast on the cell due to increased osmolality (concentration of solutes in a fluid evaluating the number of particles per weight -kilo-of fluid).
Hypersensitivity reactions (HRs) are uncommon and vary in frequency from 0.004 to −0.7% (4). They can be classified as immediate and non-immediate. Immediate hypersensitivity reactions (IHRs) are urticaria, angioedema, rhinitis, conjunctivitis, bronchospasm, gastrointestinal symptoms, anaphylaxis, or anaphylactic shock occurring within 1-6 h after drug administration. Non-immediate hypersensitivity reactions (NIHRs) include delayed urticaria, maculopapular eruptions, fixed drug eruptions, vasculitis, and Severe Cutaneous Adverse Reactions (SCAR); they may occur at any time as from 1 h after the initial drug administration (5).
The most frequent reactions are immediate reactions where skin manifestations are presented in 75-100% of cases, including urticaria, rashes, pruritus, and limited facial edema (6). Anaphylaxis occurs in 0.01% of cases (7). The death rate due to gadolinium contrast-induced anaphylaxis is 0.0019% and of the total deaths reported by the FDA is 0.00008%. In 2015, a case was published of a 66-year-old patient who underwent a gadoteridol scan and, upon completion of the infusion, immediately began with dry cough and facial erythema followed by seizures and dyspnea, with hypotension and ventricular fibrillation and subsequent death. Postmortem serum tryptase levels of 1,220 mcg/liter were found, suggesting a massive tryptase release during anaphylaxis (8).
In immediate hypersensitivity reactions, whether IgE mediated or not, the release by mast cells and basophils of histamine and other mediators can give rise to cutaneous manifestations, nasal congestion, dyspnea, bronchospasm, nausea, vomiting, diarrhea, hypotension, and tachycardia. In non-IgE-mediated cell activation, the mechanisms causing this degranulation include the direct effect of the contrast on the cell membrane (related to the chemical structure of the contrast and osmolality), the activation of the complement (anaphylatoxins, C3a and C5a, which are capable of activating mast cells by binding to membrane receptors other than those of IgE) and the generation of bradykinin. Considering the risk factors for developing immediate adverse reactions to GBCAs, the pattern is similar to the reactions that appear after exposure to iodinated contrast media (ICM), female gender being a predisposing factor for developing immediate hypersensitivity reactions with GBCAs (9). Although these differences are unexplained, animal studies suggest that specific sex hormones may be related to the increased incidence in females (10). Antecedents of allergy and history of a previous reaction to GBCAs are relevant determinant factors. In the analyzed studies, age and concomitant treatments were not found to be risk factors, even if other studies reported them as such (6).
Concerning non-immediate or delayed hypersensitivity reactions to GBCAs, Power et al. (11) reported an incidence of 0.05%. The reported symptoms were urticaria (66%), rash (33%), and pruritus (6.6%). Delayed reactions appeared on the same day in 46% of cases, on the following day in 20% of cases, and in 33% of cases, the moment of manifestation was uncertain (11).
Most of the publications that address an allergy evaluation did not appear until the 2000s, and most describe cases of immediate hypersensitivity. The allergic assessment of patients who have suffered a suspected hypersensitivity reaction has shown the involvement of specific immunological mechanisms, IgE-or T cells-mediated, in a significant percentage of patients. This allergy evaluation could also be helpful for the study of cross-reactivity patterns.
The study aimed to examine through a systematic review of the studies published regarding the allergic evaluation of immediate and delayed hypersensitivity reactions after the administration of GBCAs used for MRI. Another objective was to analyze the cross-reactivity patterns between GBCAs in both immediate and non-immediate reactions.

Literature Search
A bibliographic search on studies was performed, including the available scientific evidence up to September 2021. The primary sources for the search were PubMed, SCOPUS, and EMBASE.

Inclusion and Exclusion Criteria
• Original articles, cases reports, case series, and systematic reviews were selected. • Only articles in English were considered.
• Only articles explicitly dealing with hypersensitivity reactions were included. • No age restriction was considered.
• At least two blinded researchers independently reviewed titles and abstracts from the initial search, and eligibility criteria determined their inclusion or exclusion.
Studies about the allergic evaluation in suspected immediate and delayed hypersensitivity reactions were prioritized. After analyzing the reviewed articles, reactions were classified as IHRs or NIHRs according to the criteria mentioned above (5).

RESULTS
Initially, database searches showed 149 results. Articles not peerreviewed, conference proceedings, editorials, or commentaries to review articles were excluded after abstracts were evaluated. After applying inclusion and exclusion criteria, 26 articles were included in the review (Figure 2). In most reviewed studies, the diagnosis of hypersensitivity to GBCAs was made based on the clinical history and skin test results (ST). Skin tests were usually performed following the EAACI-ENDA methodological recommendations (12). Most manuscripts refer to previously published guidelines concerning the positivity or negativity criteria, mainly that of ENDA's guideline (12).
Nineteen articles that evaluated immediate reactions and performed an allergy evaluation were found in the bibliographic search (3,7,(13)(14)(15)(16)(17)(18)(19)(20)(21)(22)(23)(24)(25)(26)(27)(28)(29). Most of them were isolated cases or small series of cases ( Table 2). In all of them, skin tests were carried out with the eliciting contrast media; in some of them, evaluation with contrast media different than those that had originated the reaction was carried out. The quality of the study was considered acceptable if STs and controlled drug provocation tests (DPTs) had been performed and suboptimal if DPTs had not been performed. Whether DPTs were performed is specified in Tables 2, 3.
In IHRs, in most articles, skin prick tests (SPTs) with undiluted commercial solutions and intradermal tests (IDTs) diluted at 1:10 were generally used (3,17,19,25,29), actual concentrations were not mentioned in the articles. In the case of severe reactions, IDTs began with higher dilutions (35). No study specifies the concentration used for skin testing; they only set the dilutions. That is an omission that has to be remarked in this review. STs were performed with the GBCAs involved in the reaction if known, although some authors recommend using a panel of GBCAs (28,29).

Unmet needs and future research
To improve knowledge of mechanisms involved in IHRs and NIHRs To identify the epitopes involved in immune reactions to GBCAs To determine sensitivity and specificity of skin tests in IHRs and NIHRs

Development of in vitro studies for diagnosis
To further clarify cross-reactivity among the different GBCAs To assess the usefulness of premedication in the prevention of reactions IHRs, Immediate hypersensitivity reactions; NIHs, Non-immediate hypersensitivity reactions; GBCAs, Gadolinium-based contrast agents.
Hasdenteufel et al. (15) reported two anaphylactic shocks with gadoterate meglumine. Both patients had positive SPTs and IDTs with gadoterate meglumine but negative with three linear GBCAs. Galera et al. (17) described two anaphylaxis, one with gadoteridol and another with gadobenate dimeglumine. In both cases, STs were positive only with the culprit GBCAs. In a recent study (29), our group presented five patients with immediate reactions to gadobutrol; 4 out of 5 had positive STs with the culprit GBCA, and two positive STs with another macrocyclic agent. When evaluating immediate reactions to GBCAs, SPTs were safer but had lower sensitivity than IDTs. Nonetheless, some reported cases had positive SPTs with GBCAs (14,15,17,18,(22)(23)(24). Notwithstanding, most cases diagnosed by positive STs were due to positive IDTs. Nevertheless, as skin tests' sensitivity is suboptimal, DPTs are necessary to diagnose and demonstrate tolerance to other GBCAs. The DPT is considered the gold standard for the diagnosis of drug hypersensitivity reactions (5).
In many patients, particularly in the case of severe reactions, an alternative GBCA can be tried to verify tolerance or assess reactivity. Such tests can confirm or exclude the diagnosis when there is no other available evidence and can be used to look for an alternative GBCA. Thus, Tomás et al. (7) described two cases of hypersensitivity to gadopentetate dimeglumine and gadoteridol, with negative STs and DPTs with alternative GBCAs. The first patient had a negative DPT with gadoteridol. The second case, due to gadoteridol, tolerated gadobenate dimeglumine (7). Another study (19) showed an excellent negative predictive value for GBCA skin tests in 11 patients with IHRs for 10 of them and NIHRs for one. The 11 patients re-exposed to GBCAs tolerated without reaction the GBCA that yielded a negative ST (3 of them with pretreatment). Notwithstanding, Moreno-Escobosa et al. (25) published one anaphylaxis with gadobutrol with positive STs to all studied agents (gadobenate dimeglumine, gadodiamide, and gadolinium), except gadoteridol (a macrocyclic agent). The DPT with gadoteridol triggered an immediate reaction, suggesting the need for DPTs despite negative STs.
A study showed the results of provocations with low-doses of GBCAs (3). The authors performed single-blind placebo DPTs with 1 mL of the GBCA solution (corresponding to one-tenth of the usually injected GBCA for MRI) during an 8-h hospital stay. They included 14 patients that underwent controlled exposure and skin tests with gadolinium contrasts. Twelve patients had presented immediate hypersensitivity reactions, one a delayed reaction, and another a reaction that could not be classified. A positive allergy workup was found in 5/14 patients, with positive GBCA IDTs in three cases, and positive DPTs in two cases (one immediate and one delayed). After the allergic study, 10 out of 14 patients were contacted, and two underwent a new injected GBCA using the contrast tested negatively during DPTs, without premedication. Analyzing the results, the authors calculate a negative predictive value (NPV) of skin tests of 86%. Other studies have also found a good NPV of skin tests (27,29).
Hojreh et al. (26) did a retrospective study on 17 children who had presented a reaction with GBCAs. Twenty-one DPTs were performed in ten patients for at least two substances-either gadoterate meglumine, gadobutrol, or gadobenate dimeglumine. None of the patients who underwent DPT showed any sign of a hypersensitivity/allergic-like reaction. Two patients exhibited dizziness, nausea and flush, and one patient experienced vomiting, all categorized as chemotoxic responses but not as allergic reactions.

Cross-Reactivity
Cross-reactivity between gadolinium chelates is still unclear. Kolenda et al. (22) described 33 patients with immediate reactions to GBCAs, finding cross-reactivities, more frequently between gadoterate meglumine and gadobutrol, both macrocyclic agents, although they described three patients monosensitized to gadobutrol. Moulin et al. (18) reported an anaphylactic reaction to gadoterate meglumine and a strongly positive SPT with gadoterate but negative STs to four GBCAs, linear and macrocyclic. They performed a DPT with gadobenate dimeglumine with good tolerance. We have described similar sensitization patterns: patients with selective responses to gadobutrol and patients sensitized to macrocyclic agents (29). Notwithstanding, the cross-reactivity between macrocyclic and linear GBCAs has not been elucidated to date, and crossreactivity between linear agents has not been addressed. Recently, Mankouri et al. (27) found cross-reactivity in 7 of 18 allergic patients (38%). Overall, amongst the 18 patients in whom both linear and macrocyclic GBCAs were tested (either as culprit agents or alternatives), the cross-reactivity rate was 27.7% among macrocyclic agents, 5.5% among linear agents, and 5.5% between both.
A lower rate of immediate allergic adverse events has recently been reported using non-ionic, non-protein-bound agents and linear GBCAs (gadodiamide) (27,38). However, the potential GBCA deposition in the brain observed preferentially with linear GBCAs has limited their use (39). In some countries, the linear GBCA agents gadodiamide and gadopentetate dimeglumine have been withdrawn from authorization of use by medical authorities, with difficulty to find substitutive GBCAs in case of HRs to macrocyclic GBCA.

Premedication
It does not appear that premedication with corticosteroids prevents a future reaction. A 2019 systematic review and meta-analysis sought patients with immediate reactions to GBCAs who were undergoing repeated administration of GBCAs (40). The study concluded that in patients with a history of immediate reaction to GBCA, the repeated HRs to the same GBCA may occur in ∼39%, despite the use of adequate premedication with corticosteroids. They found no difference in the rate of reactions when comparing macrocyclic with linear-ionic GBCAs (40). Other studies have also found no utility of premedication to prevent future reactions with GBCAs (26,(41)(42)(43).
One case (31) involved an acute generalized exanthematous pustulosis (AGEP) due to gadobutrol, with PTs yielding positive results to gadobutrol on days 2 and 4 and negative results to gadoterate meglumine. DPTs were not performed. Another case involved an erythematous maculopapular rash following the administration of gadobutrol (30). PTs were performed with gadobutrol, gadoteridol, and gadoterate meglumine (all macrocyclic agents), as well as gadodiamide and gadopentetate meglumine (both linear agents), with positive results only to gadobutrol. As in the previous case, DPTs were not performed. The third case was a severe delayed reaction with cutaneous and cardiac symptoms. Neither PTs nor DPTs were performed (32).
In 2021, Gauthier et al. (33) reported a delayed reaction to GBCA, showing a positive delayed IDT with undiluted gadoteridol. DPTs were not performed. The same year our group published two case reports. A case of delayed exanthema with Gadobutrol (macrocyclic) with tolerance to gadoxetate disodium (linear) (36). A woman consulted our Allergology Service because 24 h after administration of gadobutrol developed mild itching and a generalized erythematous and generalized erythematous rash. PT and IDTs tests were negative. Given the likely need for future MRI scans, that the skin reaction had not been severe, and the diagnosis was uncertain, the patient gave written informed consent for a DPT with the usually injected gadobutrol dose for MRI, suffering a delayed rash. Subsequently, a new exposure test with a full dose of gadoxetate disodium was tolerated (36). That is the first reported case of a NIHR with gadobutrol in which a complete allergy study was performed, including DPTs, to confirm the diagnosis and offer a safe alternative for subsequent administration of GBCAs.
The second case report was a 13-year-old boy of a drug rash with eosinophilia and systemic symptoms (DRESS) syndrome in a child sensitized to chemically and antigenically unrelated substances: antibiotics, NSAIDs, and gadolinium-based contrast media (34). The boy had a reaction with gadobutrol -a macrocyclic non-ionic GBCA-and a doubtfully positive result with disodium gadoxetate, a linear ionic GBCA. Although guidelines for DRESS usually recommends not performing DPTs with the suspicious drug and structurally related drugs due to the risk of eliciting a new reaction (44), providing an alternative GBCA was considered essential for disease management by the attending physicians, so a DPT was performed with gadoteric acid. The patient developed symptoms again, which confirmed DRESS.
The main limitation of this review is that the quality of the evidence was not formally assessed, as the studies included individual patients or few patients; hence, the quality of the evidence is very low. Consequently, rates of hypersensitivity reactions to individual GBCAs are not available. Table 4 enumerates the unmet needs that should be the object of future research.

CONCLUSION
Hypersensitivity reactions to GBCAS are rising due to the worldwide increase in nuclear magnetic resonance diagnostic techniques using GBCAs. Although many immediate reactions after the administration of GBCAs are not genuine allergic reactions, reactions due to specific immunological mechanisms, both immediate and delayed, are increasingly being described and confirmed by positive results in skin or challenge tests. The usefulness of skin tests with GBCAs is not yet well established because most studies have included low figures of patients. Although some studies of immediate reactions after GCBA administration have suggested a high negative predictive value of STs, DPTs are still necessary to confirm or exclude the diagnosis or find alternative GBCAs.
Cross-reactivity patterns between different GBCAs are not well established. Notwithstanding, data from recent studies point to a partial cross-reactivity among macrocyclic agents and no or very low cross-reactivity between macrocyclic and linear agents. Nevertheless, further studies are required to confirm these data. In Figure 3, we suggest a possible algorithm for the study and selection of GBCAs in case of purported reactions to these agents.

DATA AVAILABILITY STATEMENT
The original contributions presented in the study are included in the article/supplementary material, further inquiries can be directed to the corresponding author/s.