Allergy evaluation of immediate hypersensitivity reactions to radiocontrast media

INTRODUCTION — Hypersensitivity reactions to both iodinated contrast used in computed tomography (CT), angiography, and urography, and gadolinium-based contrast agents used in magnetic resonance imaging (MRI) are well described. These can be divided into immediate (occurring within minutes to an hour of administration) or nonimmediate (delayed; occurring after hours to days). The allergy evaluation of immediate reactions is discussed in this topic review.

●The clinical manifestations, epidemiology, diagnosis, and treatment of immediate contrast reactions are discussed separately. (See "Diagnosis and treatment of an acute reaction to a radiologic contrast agent".)

●The use of premedications in patients with past immediate hypersensitivity reactions (IHRs) to iodinated contrast for the purpose of preventing subsequent reactions is also discussed separately. (See "Patient evaluation prior to oral or iodinated intravenous contrast for computed tomography", section on 'Patients with past reactions to contrast'.)

IODINATED CONTRAST AGENTS — All modern intravenous contrast agents injected for computed tomography (CT) are iodine based. A list of contrast agents used in CT can be found in Appendix A of the Manual of the American College of Radiology [1]. In this topic, we refer to iodinated radiocontrast agents as "contrast" or "contrast agents."

Iodinated contrast agents are classified as iso-osmolal (IOCM), low-osmolality contrast media (LOCM), or high-osmolality contrast media (HOCM). Only IOCM and LOCM are used intravenously; HOCM is no longer used. (See "Patient evaluation prior to oral or iodinated intravenous contrast for computed tomography", section on 'Types of contrast material'.)

Immediate reactions — Immediate contrast reactions occur within minutes to an hour of exposure and can be divided into two types, called "hypersensitivity" and "physiologic," although various guidelines use additional terms, as discussed below [1]. Both occur rapidly after contrast administration, but they have different signs and symptoms and presumed pathogenesis (table 1). Physiologic reactions are more common than hypersensitivity reactions.

Signs and symptoms

●Physiologic: These reactions are also called toxic or chemotoxic reactions in different guidelines. Physiologic reactions typically present with transient warmth/flushing (or chills), metallic taste, nausea, vomiting, isolated pruritus, chest pain, arrhythmia, seizure, hypertension, and vasovagal signs (ie, hypotension and bradycardia).

●Hypersensitivity reactions: Hypersensitivity reactions, also called allergy-like or allergic-like reactions, are relatively rare and have features of immediate hypersensitivity, including urticaria or angioedema, widespread erythema, hoarseness or stridor (due to upper airway or laryngeal edema), bronchospasm, and hypotensive shock. In this topic, the term immediate hypersensitivity reaction (IHR) is used to refer to these reactions.

Among hypersensitivity reactions, there is growing evidence that some of these reactions, particularly those that are severe, may be immunoglobulin E (IgE) mediated. Evidence for an IgE- or mast cell-mediated mechanism includes positive skin tests [2], tryptase and histamine release during the reaction [3], and basophil activation tests [4,5]. Among patients with severe anaphylaxis, positive skin test results are common. (See 'Sensitivity and specificity' below.)

The recognition and acute treatment of IHR and toxic reactions to iodinated contrast are described separately. (See "Diagnosis and treatment of an acute reaction to a radiologic contrast agent".)

Premedication and changing contrast agent — For patients with past IHRs to contrast who require future radiologic studies, it is common practice to administer premedications and use a different contrast agent from the one that caused the previous reaction, if a repeat exposure to contrast is judged to be reasonable. Evidence suggests that both interventions are effective in reducing but not eliminating retreatment reactions, and that changing the contrast agent may be the more important component. Even with these precautions, the risk of recurrence is higher in patients with severe initial reactions compared to those with milder symptoms. Studies of the efficacy of these interventions are reviewed separately. (See "Patient evaluation prior to oral or iodinated intravenous contrast for computed tomography", section on 'Prevention'.)

Role of skin testing — The role of skin testing in the evaluation of IHRs to contrast is evolving. In the past, skin testing was usually negative and therefore not recommended, and patients with severe past reactions were managed primarily with avoidance. However, older high-osmolar contrast agents had a several-fold greater number of physiologic and hypersensitivity reactions including severe reactions [6,7], and use of these agents has been largely discontinued. It is hypothesized that the severe reactions that now occur with low- and iso-osmolar agents are more likely to be immunologic (and possibly IgE-mediated) in nature, and therefore skin testing may be more relevant.

The evolution of practice is reflected in variations in the guidelines currently in use: The 2010 American College of Radiology and an American Drug Allergy Practice Parameter from the same year do not endorse skin testing with contrast [8], while a subsequent European position paper on drug allergy and an international expert panel on controversies in drug hypersensitivity initiated by the American Academy of Allergy and Clinical Immunology do recommend its use [9-11]. The latter consensus statements are more recent and reflect growing evidence for a higher prevalence of IgE sensitization with newer radiocontrast media.

At present, there is only limited evidence that a skin testing-based approach is superior to the combination of premedication and the use of an alternative contrast agent (ie, different from the agent that caused the patient's past reaction) with low osmolality, because studies comparing the two approaches have not been performed. However, an international panel of drug allergy experts was convened in 2018 to review controversies in contrast hypersensitivity and concluded that there was evidence to suggest that skin testing for immediate hypersensitivity to contrast can potentially identify safe alternative(s) for re-exposure [11]. It was the opinion of most members of the panel, including the author of this topic, that skin testing to contrast is most useful in patients with recent, severe IHRs, and that in this context, it can be useful in choosing an alternative contrast agent that the patient can safely receive.

It is postulated that skin testing performed promptly (two to six months) after a reaction is able to differentiate between allergic (ie, IgE-mediated) and nonallergic IHRs, and that patients with severe reactions are often skin test-positive to the culprit agent. Skin testing can further be used to identify skin test-negative alternative agents, which are subsequently tolerated [2]. In addition, patients with milder reactions (eg, isolated urticaria) tend to be skin test-negative, and the pathophysiology of these reactions remains unknown. Recurrence of these milder reactions may be unlikely and/or adequately prevented by changing the contrast medium and premedication.

Indications for skin testing — The goal of skin testing in a patient with a past IHR to contrast is to identify one or more alternative agents to which the patient's skin test is negative, so that these alternative agents can be used in the future. We limit skin testing with contrast to specific clinical situations:

●The most straightforward indication is the patient with recent anaphylaxis to contrast who requires another procedure with contrast in the near future. Patients with severe past IHRs are at greater risk of developing recurrent reactions on re-exposure, despite glucocorticoid premedication [12,13]. Patients who have experienced life-threatening reactions with cardiovascular symptoms should be managed with the utmost care, as they are most likely to be skin test-positive and be allergic to one or more contrast agents [14].

●A more controversial indication is the patient with a less severe IHR to contrast, such as urticaria/angioedema or isolated bronchospasm [10,11,14]. At least one study suggests that these reactions are non-IgE mediated [2].

●Skin testing can also be helpful in patients with recurrent (ie, "breakthrough") IHRs despite premedication.

Of note, there is universal consensus that skin testing has not been shown to be useful in identifying patients at risk for IgE-mediated IHRs, if the patient has not already experienced this type of reaction [15-17]. Stated differently, skin testing cannot be used to "screen" patients without past reactions for possible future reactions.

Various described risk factors for allergic-like reactions to contrast (eg, sex, atopy, allergy to other drugs) are not indications for performing an evaluation for contrast allergy. Similarly, asthma, cardiac disease, mastocytosis, and treatment with beta-blockers, all of which may increase the severity of an IHR to radiocontrast media, are not indications for allergy testing [10].

Referral — If skin testing is pursued, it should be performed by allergy experts experienced in drug allergy and trained in treating acute allergic reactions, since there is a very small risk of systemic allergic reactions to the testing itself [18]. However, testing is safe when properly performed. In a meta-analysis including 726 patients with past IHRs who underwent skin testing, only one systemic reaction was reported in response to intradermal testing [19].

Timing of evaluation — Ideally, skin testing should be performed within two to six months of the original reaction, as the incidence of positive skin tests to the culprit agent appears to be substantially lower both before and after this time period [18,19]. A multicenter study evaluated 122 patients with past IHRs and found a much higher rate of skin test positivity among patients who were tested within a few months of their reactions. Skin testing was performed with the culprit contrast if known, as well as several other commonly used agents. Positive intradermal skin tests were seen in 26 percent of patients with IHRs overall, although subjects were statistically more likely to have positive skin test results if tested within two to six months of their reactions (50 percent positive), compared with longer or shorter periods of time (18 percent positive) [18].

The observation that there appears to be a brief window of time in which positive skin tests are demonstrable (ie, a few months) following the initial severe hypersensitivity reaction is not typical of other IgE-mediated allergies (ie, foods, venoms) and may indicate that contrast hypersensitivity is shorter-lived compared to other allergies. There are well-studied forms of IgE-mediated drug allergy (ie, penicillins, cephalosporins) in which skin testing tends to revert to negative over time in most patients, although loss of reactivity occurs over years in the case of penicillins. In the case of contrast hypersensitivity, this seems to happen much more rapidly.

However, if patients with severe past IHRs present for evaluation years after the initial reaction, then skin testing can still be performed with a panel of possible alternative agents and an agent to which the patient is negative on skin testing can be chosen for future use. In this situation, it would be prudent to also give premedication and to give the contrast in a setting equipped to manage anaphylaxis.

Skin testing techniques

What to include in testing — Skin testing should be performed with the contrast agent involved in the past reaction, if known, as well as several possible alternative agents. Discussion with the radiologist can help identify possible alternative agents. An institution-specific panel can be developed that reflects the contrast used in that radiology department [10,11]. Histamine and saline should be included as positive and negative controls, respectively, to demonstrate normal skin reactivity.

Concentrations — Skin prick testing with undiluted contrast is performed first to detect the rare, extremely sensitive patient who is at risk for a systemic reaction with intradermal testing, although skin prick testing is not sufficiently sensitive to identify many sensitized patients and it is important to proceed with intradermal testing when skin prick testing is negative [18,19].

Intradermal testing is the more sensitive method [18,20]. Intradermal testing is conducted on either the volar forearm or the back. A small amount of the test solution (0.02 to 0.05 mL) is injected into the skin to produce a bleb of 4 to 5 mm in diameter. The optimal concentration for intradermal testing is not uniformly agreed upon, although most studies have used a 1:10 dilution, which corresponds to a concentration of approximately 30 to 32 mg iodine/mL. This concentration appears to be specific, but at the price of some sensitivity. (See "Evaluating diagnostic tests", section on 'Balancing sensitivity and specificity'.)

The use of a 1:10 dilution (approximately 30 mg iodine/mL) for intradermal testing was recommended by the 2013 European Network on Drug Allergy and European Academy of Allergy and Clinical Immunology (ENDA/EAACI) position paper [9]. In this paper, the specificity of this dilution was >95 percent in controls, whereas similar data are lacking for undiluted contrast. The 1:10 dilution has been used in the majority of studies published subsequently.

The need for relatively high concentrations of contrast to obtain positive results on intradermal testing is characteristic of skin testing for several many other small molecular weight drugs (eg, proton pump inhibitors, platins, antibiotics) in contrast to many other types of proven IgE-mediated allergy to food or insect venom proteins, which are usually demonstrable with even very dilute testing solutions.

Interpretation of results

Sensitivity and specificity — The sensitivity of skin testing for the detection of patients with past IHRs is variable, depending on the severity of the reaction and the time elapsed between the reaction and testing. In the largest study to date, positive skin tests were observed in 26 percent of 220 patients reporting a past immediate reaction of any severity: 3 percent had positive skin prick tests, and 25 percent had positive intradermal tests [18].

The specificity of skin testing with contrast media is high and estimated at 95 percent for skin prick testing, and 91 to 96 percent for intradermal testing in largely nonexposed controls [18,21].

Multiple studies and a meta-analysis have demonstrated that patients with recent severe IHRs are more likely to have positive intradermal skin test results (when adequate concentrations are used for intradermal testing, such as 1:10 dilution [corresponding to approximately 30 mg iodine/mL]) [5,14,19-25].

●In a meta-analysis of patients from 21 studies, intradermal skin test results were positive in 17 percent of 726 patients with IHRs of any severity, and up to 52 percent of those with severe IHRs [19].

●In a series of 51 patients with contrast-induced anaphylaxis, 65 percent had positive intradermal skin testing to the culprit agent; among patients with anaphylaxis with hypotension, 82 percent had positive skin tests [22].

Meaning of a positive test — A skin prick test for immediate hypersensitivity is considered positive if a wheal of ≥3 mm in diameter, with surrounding erythema, develops within 15 to 20 minutes. An intradermal test for immediate hypersensitivity is regarded as positive if the initial wheal size has increased by at least 3 mm in diameter and is surrounded by erythema after 20 minutes.

It is assumed, but not proven, that a positive skin test indicates an allergic (possibly IgE-mediated) mechanism. A positive skin test demonstrates that the agent has the ability to cause immediate release of inflammatory mediators from the cutaneous mast cells of that patient, through mechanisms that are either IgE-mediated or involve contrast media-specific direct mast cell activation (by a yet-unknown mechanism). Thus, if the culprit agent yields a positive result, it indicates that the reaction was indeed mast cell mediated and that the negative results obtained with alternative agents have a high likelihood of predicting that a similar reaction will not occur.

Note that different control populations have been used in the studies. Whereas most often patients who tolerated contrast years earlier or who are contrast-naïve were used as controls, individuals who recently tolerated contrast (within the past few months) might be a better control population.

Positive predictive value — The positive predictive value (PPV) of skin testing with contrast is considered to be high but not precisely known, as intentionally administering a skin test-positive contrast agent would be unethical. A case report described a patient with past anaphylaxis and positive skin testing results who was accidentally given the culprit drug again and had a repeat reaction of moderate severity, despite premedication [20]. However, the PPV does not appear to be 100 percent, because a single patient out of 1053 (without histories of previous IHRs) who were skin tested just before contrast administration had a positive result, and this individual tolerated the contrast in question [17].

Meaning of a negative test — Provided that skin testing is performed within a few months of the reaction, negative results are more often seen in patients with less severe reactions, such as isolated urticaria or angioedema. Recurrent reactions of this type may not be IgE mediated and are often preventable with premedication and changing to a different contrast agent. It is also important to note that for the contrast agent that caused the past reaction, the absence of skin test reactivity does not necessarily guarantee the absence of clinical reactivity, because the mechanism could still be non-IgE mediated, and for this reason, use of a different agent, to which the patient is skin test-negative, is preferred.

Negative predictive value — The negative predictive value (NPV) of skin testing to contrast agents is an important parameter, because the primary purpose of skin testing is to identify alternative contrast agents that the patient can safely receive in future radiologic procedures. In studies in which patients with IHRs to contrast underwent skin testing to identify alternative agents, the NPV of skin testing ranged from 91 to 100 percent [5,20,23,26-28]. A 2015 meta-analysis pooled results from available studies and calculated a NPV for IHRs in 116 patients of 95 percent (95% CI 89-98 percent) [19].

The largest study investigating the NPV of skin tests in patients with previous skin test-proven contrast allergy included 423 patients with past IHRs [23]. Re-exposure to contrast occurred in 233 patients (39 percent) and was tolerated in 16 of 17 (94 percent) with at least one positive skin test, and 201 of 216 (93 percent) with all negative skin tests. Among the 16 patients who experienced repeat reactions, the intensity was similar in 4, milder in 10, worse in 1 (although this was determined to be nonallergic in retrospect), and unknown in 1 patient. It was concluded that skin testing for potential contrast IHR is an effective means of identifying safe alternative(s) for re-exposure.

Sensitization to multiple agents — Patients may demonstrate skin testing positivity to more than one contrast agent, which may represent either immunologic cross-reactivity or exposure to other agents. In a meta-analysis that included 726 subjects with past IHRs, 39 percent of patients (95% CI 29-50 percent) were reactive to two or more contrast agents on skin testing or graded challenge [19]. Predictable patterns of multiple sensitizations have not been recognized.

Cross-reactivity related to the structure of the contrast agent might exist, although studies are conflicting. In the largest study on cross-reactivity among contrast agents, the authors proposed a classification of different contrast subgroups: A (ioxitalamate, iopamidol, iodixanol, iomeprol, ioversol, iohexol), B (iobitridol), and C (ioxaglate) and hypothesized that the risks of reacting were higher within these groups [29]. However, these groupings are not consistent with the results of other studies and further investigation will be necessary to determine if there are predictable patterns [2,18].

In contrast, cross-reactivity does not appear to be related to common excipients contained in contrast media, nor to ionicity. In addition, larger studies did not demonstrate cross-reactivity correlating to the contrast classes ionic tri-iodized monomers, ionic hexa-iodized dimers, nonionic tri-iodized monomers, and nonionic hexa-iodized dimers [18,23,29].

Therefore, the choice of alternative agents for skin testing should be based on which agents are available at the specific facility in question, as well as the patient's past exposures.

Graded challenge (rarely performed) — The administration of a small "test dose" of radiocontrast, without preceding skin testing to identify agents to which the patient is not sensitized, should not be performed because it was shown not to predict the occurrence of IHR with a full dose of contrast and can precipitate severe reactions [30]. However, the safety is significantly enhanced if challenge is preceded by skin testing to identify contrast agents to which the patient is not sensitized. The expert panel (mentioned previously) that was convened in 2018 to discuss controversies in radiocontrast hypersensitivity concluded that graded challenge procedures, when preceded by skin testing to identify contrast agents to which the patient was not sensitized and performed in an environment prepared to manage anaphylaxis, could be useful in certain selected situations [11]. Studies of graded challenge (also called drug provocation testing or DPT) from European centers indicate that this approach can be safely performed [2,5,21,23,26,27,31].

Considering the high NPV of skin testing to contrast in general, there may be relatively few situations in which the additional step of a DPT would benefit the patient [23,31]. One example might be an individual with a severe past IHR who is likely to require radiocontrast in the future (eg, repeat cardiac angiography), in which case it would be better to have the patient's initial re-exposure to contrast take place under the guidance of allergists experienced in the management of anaphylaxis, rather than in the middle of a procedure when the patient is potentially unstable and allergists are not present. Thus, the decision about performing a DPT should be made on a patient-by-patient basis after a risk-benefit analysis, after appropriate negative skin testing with the candidate radiocontrast media. If performed, DPT with contrast should be done in well-equipped centers and by personnel who are trained to recognize and manage repeat reactions [10].

Intravenous administration of iodinated contrast can cause acute kidney damage. Therefore, if a DPT is performed, low-osmolality or isosmolar agents are preferred, and clinicians should check renal function before the procedure. Patients with an estimated glomerular filtration rate <30 mL/min/1.73 m² and those with an ongoing episode of acute kidney injury may require specific preventative measures. Management of such patients is presented separately. (See "Prevention of contrast-induced acute kidney injury associated with computed tomography", section on 'Prevention among high-risk patients'.)  

Protocols — Available protocols for DPT are not yet standardized or validated, and there is no consensus regarding the dose of contrast needed for a provocation test. Studies have reported provocation doses ranging from 10 to 120 mL. Most have not included premedication:

●One center administered 5, 15, 30, and 50 mL (cumulative dose of 100 mL) of contrast intravenously in saline at 45-minute intervals, without premedication [5].

●Another group used a two-day protocol of 5, 30, and 60 mL of contrast at 30-minute intervals on day 1, and 120 mL on day 2, without premedication [27].

●A third group used 10 mL of saline (placebo) followed two hours later by 10 mL of contrast agent without premedication [32].

●Another protocol used 0.05, 0.5, 1, 5, 7.5, 10, and 25 mL in 30-minute intervals for a total of 49.05 mL [2,20].

It is the author's practice to perform a one-day protocol with 1, 5, 15, and 50 mL doses (cumulative dose of 71 mL), given at 60-minute intervals, although this has not been published.

Future use of a skin test-negative contrast agent — Due to the relatively limited nature of the evidence concerning skin testing to contrast, it is common practice in the United States and elsewhere to premedicate all patients with previous IHRs, even if the contrast agent being administered has yielded negative skin test results. However, accumulating data from studies of DPTs in European centers show good tolerance of skin test-negative alternatives in patients with moderate to severe anaphylaxis, positive skin test to the culprit, and negative skin test to the alternative, indicating that premedication in this patient group might be unnecessary [2]. Premedication protocols are reviewed separately. (See "Patient evaluation prior to oral or iodinated intravenous contrast for computed tomography", section on 'Premedication regimens'.)

GADOLINIUM-BASED CONTRAST AGENTS — Gadolinium-based contrast agents (GBCAs) are paramagnetic macrocyclic or linear gadolinium-containing chelates used for contrast in magnetic resonance imaging (MRI) [33]. Generic and brand names and the molecular properties of the commercially available GBCAs can be found in Appendix A of the Manual of the American College of Radiology [1].

Immediate reactions — Immediate hypersensitivity reactions (IHRs) to GBCA have been reported [33-38]. They are considered less frequent than IHRs caused by iodinated contrast, although this was questioned in one study [14].

The clinical manifestations of IHRs to GBCAs are similar to IHRs to iodinated contrast [33,34]. The most common symptoms are urticaria (50 to 90 percent) and nausea, whereas anaphylaxis is reported only rarely [34-37]. However, anaphylaxis has been reported even with initial exposure [38]. In one study, abdominal MRI examinations more frequently elicited IHRs (1 in 10,000), compared with examinations of the brain (5 in 100,000) or spine (3 in 100,000) [39].

A meta-analysis reviewing data through 2017 identified 662 IHRs to GBCAs from a total of 716,978 administrations [34]. The overall and severe rates of GBCA IHRs were 9.2 and 0.52 per 10,000 administrations, respectively:

●The majority (81 percent) were mild: self-limited, without evidence of progression and including limited urticaria, pruritus, or cutaneous edema, limited itching or "scratchy" throat, nasal congestion, sneezing, conjunctivitis, and rhinorrhea

●Thirteen percent were moderate

●Six percent were severe

It is unclear if reaction rates are related to the structure of the GBCAs [35,36,39,40]. There are four molecular structure classes: linear ionic, linear nonionic, macrocyclic ionic, and macrocyclic nonionic. In the meta-analysis mentioned previously, the linear nonionic chelate gadodiamide had the lowest rate of reactions, at 1.5 per 10,000 administrations, which was significantly less than that of linear ionic GBCAs at 8.3 per 10,000 administrations and less than that for macrocyclic nonionic GBCAs at 16 per 10,000 administrations. There were no differences between ionic macrocyclic and nonionic macrocyclic GBCAs. Ionic linear GBCAs known to be associated with protein binding (gadoxetate, gadofosveset, and gadobenate) had a higher rate of reactions, at 17 per 10,000 administrations compared with the same chelate classification without protein binding (gadopentetate dimeglumine), at 5.2 per 10,000 administrations [34].

In a retrospective study of 102 patients who experienced 112 IHRs to GBCAs, the rate was higher in women and in patients with allergies and asthma [35]. The highest risk was in patients with a previous history of IHRs to GBCAs: They had a higher rate of recurrence after re-exposure to GBCAs (30 percent) compared with the incidence rate in all exposed patients. The incidence of IHRs increased the more times patients were exposed to GBCA [35].

Allergy evaluation — As with iodinated contrast, the presence of positive skin tests in a small minority of patients with IHRs to GBCAs points to a possible IgE-mediated mechanism in some individuals, particularly those with severe reactions [14,38,41-45]. However, in most patients, IHRs are attributed to a nonallergic mechanism that has not been elucidated. Allergy evaluation includes a historical review of past reactions, any tryptase levels measured at the time of the reaction, skin testing, and possibly other investigative forms of testing, such as the basophil activation test (BAT).

Skin testing — By analogy with radiocontrast media, skin testing should be performed with the GBCA that caused the initial reaction, if known, as well as some possible alternative agents of different molecular structure classes. If positive skin tests are obtained, those GBCAs should be avoided, and agents that yielded negative skin test results should be used preferentially [42].

Skin prick testing with undiluted GBCA and intradermal testing with a 1:10 dilution is recommended [14,38,42-46]. Intradermal testing with undiluted GBCA preparations can be irritative and nonspecific [38]. The percentage of patients with positive skin tests is not yet known, nor is the relationship to reactive status. One report included 36 cases of IHRs to gadolinium-based contrast media (GBCM) [14]. Skin testing was positive in 28 percent (on skin prick testing in two patients and on intradermal testing with a 1:10 dilution in the others). An additional 8 of 36 patients had positive intradermal testing only to undiluted GBCM. Skin test positivity increased with the severity of the reaction. Tryptase and histamine concentrations were more likely to have been increased in patients with positive skin testing to 1:10 versus undiluted GBCM.

The negative predictive value (NPV) of GBCA skin testing appears to be high, based on a report of 27 patients with a history of IHR to GBCA over a five-year period [44]. All 11 patients (of the total of 27) who were re-exposed to a skin test-negative GBCA tolerated a new agent.

Other testing — Another study described 33 patients with IHRs to GBCA who were evaluated with skin testing and BATs [45]. Two patients had positive intradermal testing only with undiluted GBCA. Based on positive skin testing, 19 of 33 patients were considered allergic, while 14 patients were considered nonallergic. BAT was negative in 13 of 14 skin test-negative patients, corresponding to a specificity of 93 percent. When offered re-exposure in a challenge procedure, five declined, five BAT-negative, skin test-negative patients tolerated the implicated drug, and four tolerated another GBCA. However, in the patients with positive skin tests, BAT was positive in just 13 of 19 (sensitivity of 68 percent). Eleven of these 19 patients tolerated another GBCA to which they were skin test-negative. Thus, in this study, BAT appeared to be a useful adjunct test. Of note, BAT is considered investigational in the United States. Technical issues with the BAT are discussed separately. (See "Overview of in vitro allergy tests", section on 'Basophil tests'.)

Future radiologic procedures — It is recommended that patients with previous reactions to GBCA either avoid gadolinium preparations completely and use an iodinated contrast agent instead, or undergo skin testing with the culprit agent, as well as with one or more GBCAs with a different molecular structure, to identify a skin test-negative product that can be given instead [33,47].

In a retrospective study of 185 patients with mild IHRs to GBCA, recurrence rates were compared for patients who were given various GBCAs again [48]. Patients given the identical (culprit) GBCA again had a recurrence rate of 26 percent. Rates were lower for patients given a different GBCA with the same molecular structure (12 percent), and lowest for those given a GBCA with a different molecular structure class (6 percent).

The benefit of premedication has not been demonstrated for the prevention of recurrent IHRs to GBCAs. In the study discussed above, a single dose of premedication (either chlorpheniramine [4 mg intravenously (IV)] 30 minutes before or chlorpheniramine [4 mg IV] plus methylprednisolone [40 mg IV] one hour before) did not provide any additional protection from recurrence beyond that arising from the change of agent [48]. However, studies of premedication for radiocontrast have found that regimens containing at least two doses of glucocorticoid (with one dose ≥12 hours in advance) are more effective than single-dose regimens, so this study did not use what is considered to be a maximally effective approach to premedication. Options for future radiologic studies are presented separately. (See "Patient evaluation before gadolinium contrast administration for magnetic resonance imaging", section on 'Measures to prevent or anticipate a recurrent reaction'.)

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Hypersensitivity to iodinated and gadolinium-based contrast agents".)

SUMMARY AND RECOMMENDATIONS

●Hypersensitivity reactions to both iodinated contrast used in computed tomography (CT), angiography, and urography, and gadolinium-based contrast agents (GBCAs) used in magnetic resonance imaging (MRI) are well-described and are categorized as either immediate (occurring within minutes to an hour of administration) or nonimmediate (also called delayed; occurring after hours to days). (See 'Introduction' above.)

●Immediate reactions to iodinated contrast can either be immediate hypersensitivity reactions (IHRs), also called "allergic-like" or "allergy-like" in the radiology literature, or physiologic reactions, also called toxic or chemotoxic reactions. These have different signs and symptoms and presumed pathogenesis (table 1). IHRs involve different combinations of pruritus, erythema, urticaria, hoarseness or stridor (due to upper airway or laryngeal edema), bronchospasm, and hypotensive shock. (See 'Immediate reactions' above.)

●For patients with mild past IHRs to iodinated contrast (eg, isolated urticaria) who require future radiologic studies, it is common practice to administer premedications (table 2) and use a different contrast agent from the one that caused the previous reaction, if a repeat exposure to contrast is necessary. This approach has been used for decades and is usually successful. A more detailed discussion of the efficacy of these measures is found separately. (See "Patient evaluation prior to oral or iodinated intravenous contrast for computed tomography", section on 'Prevention'.)

●For patients with moderate to severe past IHRs, the strategy of using different agents and giving premedications can reduce the rate of reactions, but recurrent reactions can still occur and can be severe. There is growing appreciation that skin testing is helpful in this population, as these more severe reactions are more likely to be allergic (possibly IgE-mediated) in nature. Thus, the role of skin testing is evolving and a moderate to severe reaction to iodinated contrast is now the primary indication for its performance. (See 'Role of skin testing' above.)

•Skin testing is optimally performed by an allergy specialist. The contrast agent involved in the past reaction, if known, as well as several possible alternative agents should be included. Skin prick testing with undiluted contrast is performed first. If negative, this is followed by intradermal testing, most commonly with a 1:10 dilution (corresponding to 30 to 32 mg iodine/mL). Testing should ideally take place within two to six months of the reaction, as the likelihood of obtaining a positive skin test to the culprit agent decreases with time. If positive skin tests are obtained, those agents should be avoided, and agents that yielded negative skin test results should be used preferentially. (See 'Referral' above and 'Timing of evaluation' above and 'Skin testing techniques' above.)

•In patients with a past moderate to severe IHR, a negative intradermal skin test result to a nonculprit iodinated contrast agent has a high negative predictive value, which has been estimated at approximately 94 percent. (See 'Interpretation of results' above.)

●Graded challenge procedures to confirm tolerability to a skin test-negative agent are generally not necessary. However, there may be high-risk individuals who are likely to need urgent contrast procedures in the future, such as patients with severe previous IHRs who may require repeat cardiac angiography. For such patients, challenges can be safely performed by allergists at a time when the patient is stable. (See 'Graded challenge (rarely performed)' above.)

●IHRs to GBCAs most commonly present with urticaria (50 to 90 percent) and nausea, whereas anaphylaxis is uncommon.

•The utility of premedication to a prevent future reaction is less well-studied, and it is recommended that patients with previous IHRs to GBCA either avoid gadolinium preparations in the future and use an iodinated contrast agent instead, or undergo allergy evaluation if GBCAs are likely to be needed again and use an alternative skin test-negative agent. (See 'Immediate reactions' above.)

•The approach to evaluation of patients with past severe IHRs to GBCA is identical to that used for radiocontrast reactions, although less is known about the negative predictive value. (See 'Allergy evaluation' above.)