Article Text

Download PDFPDF

Patients with a history of hypersensitivity reaction to iodinated contrast medium and given iodinated contrast during an interventional pain procedure
  1. Honorio T Benzon1,
  2. Jeffrey Schechtman1,
  3. Sophy C Zheng1,
  4. Jeffery A Katz1,
  5. Arpan Patel1,
  6. Geeta Nagpal1 and
  7. Benjamin P Liu2
  1. 1 Department of Anesthesiology, Northwestern University Feinberg School of Medicine, Chicago, Illinois, USA
  2. 2 Department of Radiology, Northwestern University Feinberg School of Medicine, Chicago, Illinois, USA
  1. Correspondence to Honorio T Benzon, Departments of Anesthesiology, Northwestern University Feinberg School of Medicine, Chicago, IL 60611, USA; h-benzon{at}


In patients with a history of a hypersensitivity reaction to iodinated contrast medium, iodinated contrast medium is avoided, antihistamine and steroid premedication are given, or a gadolinium-based contrast agent is employed. Six patients with a history of a hypersensitivity reaction to iodinated contrast medium and who were not premedicated had an unintentional injection of iodinated contrast. None of the patients developed a moderate or severe reaction. All patients had gadopentetate dimeglumine in one of their injections; three had repeated injections of the gadopentetate. The lack of a significant reaction may be due to any or all of the following: questionable history of iodinated contrast reaction, low dose of iodinated contrast given, concomitant injection of (epidural) steroid, and slower absorption from epidural compared with intravenous injection. While it is reassuring to know that there is a low possibility of a moderate to severe reaction in these patients, every effort should be made to avoid this scenario, appropriate drugs and resuscitation equipment should be immediately available, and the patients should be observed adequately and followed for the possibility of late reactions. Recent publications have called for caution in the use of gadolinium-based contrast agents.

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.


Hypersensitivity reactions (HRs) to iodinated contrast medium (ICM) can be considered as anaphylactoid (also called idiosyncratic reactions) or non-anaphylactoid (chemotoxic) in nature.1 Most HRs are anaphylactoid, usually secondary to nonspecific mast cell release. HRs are classified as mild (localized urticaria, itching, flushing, nausea, vomiting), moderate (systemic urticaria, severe vomiting, angioedema, mild bronchospasm, throat tightness, tachycardia/bradycardia), or severe (laryngeal edema, severe bronchospasm, cardiovascular collapse, loss of consciousness, seizures).2 Mild reactions are self-limited and do not require treatment; moderate reactions necessitate medical management, while severe reactions are life threatening and mandate resuscitative measures.

We noticed that patients with a history of a HR to ICM did not develop a moderate or severe HR when unintentionally given contrast during an interventional pain procedure. We therefore looked at the records of these patients and noted the following: reaction occurrence, treatment given, and follow-up.

Case reports

Our patients were taken from a list that we prospectively made since 2007, after noting that some patients with a history of HR to ICM did not develop a reaction when unintentionally given contrast. Our retrospective review of the patients’ charts was approved and a waiver of consent granted by our Institutional Review Board.

Four investigators (HTB, JS, SZ, AP) reviewed the patients’ charts and noted the following: (1) how the history of HR was documented; (2) circumstances surrounding the administration of ICM; (3) whether the patient had a contrast injection in another setting (CT, MRI, or angiography); and (4) the reaction and treatment to the unintentionally injected ICM.

Description of cases

The patients’ interventional pain procedures are described in the table 1. All patients had ‘allergy to contrast’ noted in their charts. Patient 2 had CT angiography in 2016 and 125 mL iopamidol (51000 mg) was injected. Premedicated with prednisone and diphenhydramine prior to the procedure, she had no HR to the contrast. Patient 3 had skin testing for triamcinolone (not contrast since the patient had previous iohexol without HR). Skin test was negative and the impression was that the patient’s HR was probably due to the contrast. Patient 4 had no HR to gadopentetate dimeglumine that was given during three cervical epidural steroid injections (ESIs) in 2014 (not included in table). In 2016, she had diphenhydramine and prednisone premedication for a CT of her abdomen. She developed itching and hives 15–20 min after the contrast was administered; the symptoms resolved after 1 hour with no treatment. Patient 6 had MR angiograms in 2010, 2012, and 2015 to monitor her multiple cerebral aneurysms. She had prednisone and diphenhydramine premedication each time and had no HR to the ICM.

Table 1

Patients with history of hypersensitivity reaction to iodinated contrast medium and given contrast for their interventional pain procedures

Unless noted, we could not identify the ICM used during the angiogram or CT since the radiology or vascular surgery procedure notes did not specify which ICM was used. In all the patients’ charts, ‘time out’ was noted in the procedure note.

In summary, we report six patients with history of HR to ICM who were not premedicated and who were unintentionally given contrast. None developed a moderate or severe reaction. Three had urticaria and pruritus after large volumes of ICM but had no HR to 1–3 mL that was given epidurally or in the sacroiliac joint. One patient had different reactions after antihistamine and steroid premedication: no reaction versus mild hives in separate ICM injections and minimal hives with gadopentetate.


Most HRs to ICM are anaphylactoid or idiosyncratic reactions.1 These reactions are not associated with release of immunoglobulin (Ig)G or IgE antibodies that are specific to the contrast agent.3 Although the exact mechanism(s) of anaphylactoid reactions are unknown, it is thought to be secondary to the release of vasoactive compounds, conversion of arginine to nitric oxide, and release of pseudoantigens and histamine resulting in allergy-like symptoms.1 The activation of Hageman factor of the coagulation cascade results in the production of bradykinin with vasodilatation, increased vascular permeability, and bronchospasm.2 Non-anaphylactoid or chemotoxic reactions are dose dependent and are influenced by the physicochemical properties of the contrast.1 4 Increased osmolality, ionicity, and iodine concentration raise the risk of non-anaphylactoid reactions.2 4 Although very rare in occurrence, some investigators supported a specific allergic mechanism, documented by positive skin and basophil activation tests.3 5

The overall incidence of HRs to ICM is 3%–13%; moderate to severe reactions are very rare, occurring in 0.04%–0.28% of cases.6 7 In response to a self-administered questionnaire, patients reported a prevalence of 8% adverse reactions (16 of 200) to ICM. The symptoms included pruritus, rash, shortness of breath and generalized pain.8 Limitations of this study include the lack of confirmation of the symptoms, assumption that the contrast was low osmolality and non-ionic, and the patients’ considering all their symptoms as HRs. Risk factors for HRs include previous reaction to the contrast, history of multiple allergies, patients with atopic tendencies including those with asthma, and female gender.1 8 9

Most HRs occur within the first 15 min of ICM administration. Late HRs can occur 1 hour to 1 week after exposure but are usually mild in nature.10 Late reactions are T cell-mediated immune reactions and include skin symptoms, nausea, vomiting, headache, muscle pain, and fever. These symptoms are associated with previous late HRs, interleukin-2 treatment, and non-ionic dimers.10

The management of acute HRs is symptomatic. An intravenous fluid is started for hypotension and administration of drugs. Diphenhydramine 25–50 mg oral or intramuscular or 25 mg intravenous is given for urticaria and intravenous atropine 0.5–1.0 mg for bradycardia.2 Oxygen, epinephrine 0.3 mg, or hydrocortisone 100–200 mg or methylprednisolone 20–40 mg intravenous are administered for laryngeal edema and bronchospasm.1 9 Emergency equipment, including airway supplies, defibrillator, and cardiovascular drugs, should be immediately available.

In patients with a history of HR to ICM and scheduled to undergo an interventional pain procedure, the options include avoidance of the iodinated contrast unless it is absolutely necessary, premedication with antihistamine and steroid, or a gadolinium-based contrast agent (GBCA) is used instead of an ICM. There are two commonly used premedication protocols: (1) 50 mg oral prednisone at 13 hours, 7 hours, and 1 hour before ICM administration, combined with 50 mg diphenhydramine given 1 hour before ICM administration11 or (2) 32 mg methylprednisolone orally 12 and 2 hours before.12 When contrast medium has to be administered within 12 hours, 40 methylprednisolone sodium succinate or 200 mg hydrocortisone sodium succinate intravenously every 4 hours can be given until the ICM is administered; 50 mg intravenous diphenhydramine is injected 1 hour before the procedure.13 Methylprednisolone sodium succinate 40 mg intravenously may be substituted for the hydrocortisone.1 Premedication is not uniformly effective in preventing HRs14 15; this was also seen in one of our patients. Prophylaxis for late reactions, in the absence of an immediate reaction, is not recommended and treatment is symptomatic.

The risk factors for HR on re-exposure included severe initial HR, use of the same contrast, larger dose, chronic urticaria, drug allergy other than ICM, and diabetes.1 16 Re-exposure to ICM in a patient with a history of HR does not necessarily result in another reaction. A retrospective study of 150 patients on the incidence of repeat HR to low osmolality contrast medium in patients with a history of moderate-to-severe HRs showed an overall recurrence rate of 19.5%, with 75% of the cases being moderate in severity.16 The recurrence rate was 67% lower when a different ICM was used. Although the amount of contrast was not stated, reexposure involved contrast-enhanced CT, signifying volumes greater than what was given in our patients.

Some investigators challenged patients with a history of HR to ICM by exposing them to an ICM. A study looked at the effect of a low dose (10 mL) ‘intravenous provocation test’ in 36 patients with a definite history of HR from ICM.17 Only one of the 36 patients who were administered 10 mL of ICM had a reaction,note that 15 patients were challenged with the culprit ICM.17 Other studies used larger volumes of up to 120 mL of ICM. In a study of five patients with positive skin test to one ICM, increasing doses (5, 15, 30, 50 mL at 45 min intervals) of a different ICM were injected intravenously to a cumulative dose of 100 mL.18 Two had a positive reaction, that is, it reproduced their original symptoms. In another study of five patients with a history of HR and a positive skin test, a total of 120 mL of an alternative negative-tested contrast was injected without premedication over 2 days. All the patients tolerated the contrast.5 In a third study, two patients with a history of anaphylactic reaction and a positive skin test to an ICM tolerated a total intravenous volume of 49 mL of an alternative ICM.19

Skin testing for ICM has not been generally accepted. This is because HRs to ICM have been traditionally considered as non-allergic reactions; hence, skin tests were regarded as inappropriate tools.3 Also, the sensitivity, specificity, and accuracy of skin testing for ICM remain to be determined.3 5 Finally, the accuracy of skin test decreases 2–6 months after exposure probably due to loss of sensitization over time.3

The lack of significant HR in our patients may have been due to any or all of the following: questionable history of HR, injection of a small amount of ICM, concomitant injection of (epidural) steroid, or slower absorption from an epidural injection compared with an intravenous injection. We doubt if the steroid that was injected epidurally prevented the development of HR. A study on the pharmacokinetics of interlaminar epidural triamcinolone showed a median time to maximum concentration of 24.2 hours and a median peak plasma concentration of 4.1 ng/mL.20 Most HRs occur within 1 hour, with 70% of the immediate symptoms occurring within the first 5–15 min.3

An issue is the duration of observation of patients with history of HR and unintentionally injected ICM. Since most HRs occur within the first 15 min of ICM administration, no additional observation is probably required in premedicated patients with a history of mild HR and have no symptoms. For patients with a history of mild HR who were not premedicated and had no reaction, a 1-hour monitoring period may be adequate since most allergy clinics observe their patients for an hour after skin testing. Two hours is ideal in patients with a history of moderate HRs but had no reaction to the ICM, based on the occurrence of HR 2 hours after the intravenous provocation test of Sese et al.17 Patients with history of late HRs need to be cautioned about this possibility and should be contacted within a week after exposure.

A gadolinium-based contrast agent (GBCA) is often used as an alternative in patients with known HR to ICM. This is because of the lower incidence of HRs with GBCAs: 0.013%–0.48% immediate HR and 0.01%–0.03% severe reactions.6 21 22 While one of our patients had a HR to gadopentetate, it was very mild. The risk factors associated with GBCAs are the same as those associated with ICMs. There appears to be no significant difference in the incidence between the macrocyclic and linear gadolinium-based agents.21 For linear GBCAs, the adverse rates are higher for the ionic agents (gadopentetate, gadobenate, gadoxetic acid, gadofosveset) compared with the non-ionic agents (gadodiamide, gadoversetamide),21 similar to that seen between ionic and non-ionic contrast agents.

An emerging and serious concern is the deposition of gadolinium in the brain with repeated administrations of this agent.23 The deposition may be related to the class of GBCA, occurring with the linear gadopentetate but not with the macrocyclic gadoterate.23 Such deposits could remain for months to years after exposure, although there is no definitely proven neurological consequence of this phenomenon at this time.6 24 Although a review of the literature showed no brain deposition after interventional pain procedures with a GBCA,24 reports of encephalopathy have been described after unintentional subarachnoid injection of gadolinium.25 26

In this report, we describe six patients with history of HR to ICM who did not have moderate or severe reaction when given contrast. If patients with HR to ICM are missed and unintentionally injected with ICM, we recommend a root cause analysis to identify the events that led to the mistake. While it is reassuring to know that there is a low possibility of a moderate/severe HR after ICM is given to patients with a history of HR, appropriate ‘time out’ should be observed. It is important to clarify whether a patient’s claim of HR is due to an ICM or to a GBCA. This is because a patient who had HR to ICM and had a subsequent uneventful GBCA may consider himself as not having a HR to an ICM, the patient confusing GBCA with ICM. This confusion between classes of contrast agents may lead to an ICM being given unintentionally. Regardless of the contrast being used, allergy premedication protocol should be instituted. If the HR is from an ICM, a different contrast should be used. The patient with HR should be treated appropriately and adequate follow-up observed. Prudence should be observed in using a GBCA in interventional pain procedures.


The authors appreciate the contributions of Anne M. Ditto, MD, Professor of Allergy and Immunology at Northwestern University Feinberg School of Medicine, in the initial version of the paper. We also thank Paul Fitzgerald, RN, BSN, MS, who helped us get IRB approval.



  • Competing interests None declared.

  • Ethics approval Approved and a waiver of consent granted by the Institutional Review Board.

Linked Articles