Patient evaluation prior to oral or iodinated intravenous contrast for computed tomography

INTRODUCTION — Health care providers who refer patients for computed tomography (CT) must consider whether to request intravenous and/or enteric (oral or rectal) contrast administration as part of the exam. While contrast administration with CT is often necessary to obtain the diagnosis, there are small risks for an acute adverse reaction or contrast-induced nephropathy (CIN).

Standards for patient preparation and indications for contrast vary among radiology practices. However, guiding principles are presented here. Patient evaluation for oral and intravenous contrast administration prior to CT is described in this topic. Guidelines on the appropriate indications for contrast use and procedures for patient preparation before the CT exam are presented.

Related UpToDate content discusses:

●Pathogenesis, diagnosis, and treatment of acute allergic-like reactions to iodinated contrast (see "Diagnosis and treatment of an acute reaction to a radiologic contrast agent")

●Allergy evaluation and indications for referral in patients with past moderate-to-severe allergic-like reactions to iodinated contrast (see "Allergy evaluation of immediate hypersensitivity reactions to radiocontrast media")

●Pathogenesis and diagnosis of CIN (see "Contrast-associated and contrast-induced acute kidney injury: Clinical features, diagnosis, and management")

●Management of patients at risk of induced nephropathy (see "Prevention of contrast-induced acute kidney injury associated with computed tomography")

TYPES OF CONTRAST MATERIAL

Types of intravenous contrast — All modern clinical intravenous contrast injected for CT is iodine-based. A list of intravenous contrast agents used in CT can be found here [1].

In patients with a history of an acute adverse event with iodinated contrast, providers should try to determine the type of agent that caused the reaction because this affects patient counseling and CT planning. Intravascular iodinated contrast is also given in other fluoroscopic exams such as angiography and intravenous urography, so adverse events related to these exams also may be relevant. A history of reaction to intravenous contrast given for magnetic resonance imaging (MRI) is not relevant because MRI contrast and CT contrast are not molecularly similar and are not cross-reactive. No cross-reactivity between iodinated and gadolinium-based contrast media has been observed.

Providers need not specify a particular type of intravenous contrast material when ordering a CT examination; this is usually determined by the hospital or the practice formulary. For CT, there are no important differences among the modern agents with regard to safety or efficacy.

Iodinated contrast agents are classified as low-osmolality contrast media (LOCM), iso-osmolality contrast media (IOCM), or high-osmolality contrast media (HOCM). For intravascular use, LOCM and IOCM are in use; HOCM is no longer used. Many iodinated contrast agents have been designed and marketed but all are soluble derivatives of a tri-iodinated benzene ring. Low-, iso-, and high-osmolality contrast are approximately 600 mOsm/kg H₂O, approximately 300 mOsm/kg H₂O, and 1200 mOsm/kg H₂O, respectively. A list of intravenous contrast agents used in CT can be found in Appendix A of the 2020 American College of Radiology Manual of Contrast Media [1].

HOCM are primarily of historical interest and are no longer used for intravascular administration due to a higher adverse event rate. LOCM and IOCM are considered equivalent with respect to the risk of contrast-induced nephropathy (CIN) at CT.

Types of oral or rectal contrast — Some enteric (oral or rectal) contrast is iodine-based. All enteric contrast has the small potential to elicit an acute reaction. A list of contrast agents used in gastrointestinal CT examinations can be found in Appendix A of the 2020 American College of Radiology Manual of Contrast Media [1].

The choice of enteric contrast is made by the radiologist primarily based on the exam indication and on any prior history of acute reaction.

Enteric contrast can be positive or negative. Positive enteric contrast (ie, high attenuation on CT), consisting of a dilute suspension of barium sulfate or a dilute solution of an iodinated agent, is used for most indications. Negative enteric contrast (ie, water attenuation on CT) is used to delineate the bowel mucosa and improve detection of active inflammation [2,3] or active gastrointestinal bleeding [4]. Examples of negative contrast agents include very dilute solutions of barium, methylcellulose, and polyethylene glycol. Plain water can be used as a negative contrast material to distend the stomach and duodenum, but it does not distend the remainder of the bowel due to absorption.

DECIDING TO USE CONTRAST — Appropriate contrast utilization is necessary to obtain acceptable diagnostic accuracy. Omitting contrast when it is indicated, or giving it when it is not, can lead to diagnostic and treatment errors, and unnecessary morbidity and cost. Added or repeated tests, delayed diagnoses, and accumulated patient radiation exposure are all potential adverse effects of an imaging exam performed suboptimally.

Both the ordering provider and the radiologist share the clinical and legal responsibility for each patient undergoing an imaging exam. However, most third-party payers require that contrast administration be specified when the exam is ordered for their precertification process, thereby placing the provider in the position of having to make the initial choice. This decision should be made after mutual consideration of the risks and benefits. In cases in which the choice is not obvious, a dialogue between the ordering provider and the radiologist can be helpful.

Indications by type of study — Whether enteric (oral or rectal) or intravenous contrast should be administered varies with the clinical indications for the CT exam. The American College of Radiology's (ACR) appropriateness criteria provide consensus guidelines on the choice of imaging modality and contrast material for many clinical situations [1]. In addition, UpToDate topics that recommend CT for patient management may specify whether oral and/or intravenous contrast is required, preferred, or not necessary for best diagnostic performance.

Intravenous contrast — For the most common CT exams, general guidelines for intravenous contrast are as follows:

●Head CT – Intravenous contrast is not necessary for head trauma, acute stroke, or intracranial hemorrhage. Contrast is indicated for neoplasm, meningitis, encephalitis, focal neurologic deficits, skull base disorders, orbital and vision disorders, pituitary imaging, complicated sinonasal disease, seizures, and CT angiography [5-8].

●Cervical CT – Intravenous contrast is not necessary for trauma unless arterial injury is a possibility or the mechanism of injury is penetrating. Contrast is indicated for a cervical mass or lymphadenopathy, suspected tumor or infection, abnormalities of cranial nerves X, XI, and XII, and brachial plexopathy [9,10].

●Cardiothoracic CT – Intravenous contrast material is not necessary for coronary calcium scoring, pulmonary parenchymal evaluation (eg, high-resolution chest CT for interstitial lung disease), or lymph node evaluation (eg, lymphoma). Contrast is indicated for CT of the heart and thoracic vessels, trauma, and for staging primary thoracic neoplasms [11-15].

●Abdominopelvic CT – Intravenous contrast material is not necessary for CT colonography, renal stone evaluation, and extraparenchymal lymphoma. Contrast is indicated for virtually all other gastrointestinal, hepatopancreaticobiliary, genitourinary, and gynecologic indications [16-18].

●Musculoskeletal CT – Intravenous contrast is not necessary for most CT exams of the extremities and spine. Contrast is indicated for evaluation of soft tissue masses and suspected septic arthritis or infected prostheses [19,20].

●CT angiography – Intravenous contrast is not necessary for monitoring a known aneurysm for growth or for detection of a hematoma. Contrast is indicated for evaluating the lumen of an artery, vein, or a pseudoaneurysm, and to assess for end-organ ischemia outside the brain or lung. It is also necessary to detect active bleeding [4,21-23].

Oral and rectal contrast — Oral contrast is necessary to evaluate for gastric or small bowel perforation following penetrating trauma or surgery and may help to separate bowel loops in thin patients with minimal visceral fat (ie, body mass index <25 kg/m²). Because it distends the stomach and intestinal tract, it improves delineation of the bowel from the mesentery and other peritoneal content (eg, hematoma, ascites, tumor nodules, etc) [24,25]. However, oral contrast has been shown to be unnecessary for accurate diagnosis of appendicitis and diverticulitis [25-27]. Oral contrast is not indicated for evaluation of the liver, kidneys, collecting systems, bladder, spleen, adrenal glands, retroperitoneum, bones, or vasculature. Negative enteric contrast can be helpful in the detection of occult gastrointestinal bleeding when paired with multiphasic CT angiography.

Rectal contrast is indicated for the detection of rectal or distal colonic leak. For most other indications, it is unnecessary.

There are relative but not absolute contraindications to oral contrast administration. In patients who are at risk of aspiration, it should be withheld or given judiciously, since aspirated oral contrast can cause pneumonitis [28]. Thus, oral contrast is potentially harmful and unnecessary to evaluate patients with suspected high-grade bowel obstruction [29]. The risk of aspiration may be highest with hypertonic oral contrast media compared with lower osmolality or barium-based solutions, but this has not been rigorously tested.

Oral contrast administration for CT in the emergency department has been associated with a 45- to 60-minute delay in disposition, which has led some practices to eliminate it for many indications in this practice setting [25].

In patients who have or are at risk for bowel perforation or a leak from a surgical anastomosis, barium-based oral contrast should be avoided. Extraluminal barium is thought to increase the risk for mediastinitis or peritonitis, although the evidence supporting this principle is incomplete and involves old agents [30-32]. Patients with chronic leaks or fistulae are unlikely to be at risk for complications from barium-based oral contrast since these are cavities with chronic exposure to bowel content.

PATIENTS WITH PAST REACTIONS TO CONTRAST — Acute adverse reactions to contrast may occur after oral, intracavitary, or intravascular administration. For any history of an acute reaction to contrast, the provider should verify that the reaction was to an iodinated contrast (used for CT, radiography, fluoroscopy, or angiography) rather than to a noniodinated contrast agent used for other imaging modalities (eg, gadolinium used for magnetic resonance imaging [MRI]). The clinical details of the index reaction such as symptoms, their duration, and required therapy should also be obtained.

Prophylactic measures are unnecessary in patients who have hypersensitivity to other allergens unrelated to iodinated contrast, including gadolinium contrast used in MRI. The risk of an immediate hypersensitivity reaction to contrast material in patients with atopic risk factors (eg, allergies to unrelated substances or asthma) is approximately twice that of the general population [33]. However, shellfish allergy is not a specific risk factor for contrast reactions [34]. Shellfish allergies are to tropomyosin (a muscle protein), not iodine [35].

Evaluation of the index reaction — Information should be gathered about the index reaction, including implicated contrast agent, clinical features, timing of onset, and severity.

Culprit agent — The incidence of an acute adverse reaction was more than 10-fold higher with intravenous administration of high-osmolality contrast media (HOCM) than with low-osmolality contrast media (LOCM) or iso-osmolality contrast media (IOCM) [36]. If the index reaction occurred before 1995, the agent was likely HOCM because LOCM was not yet available. If the prior acute reaction was to HOCM, the patient should be informed it is less likely to recur since LOCM or IOCM are now universally used for intravenous administration. However, despite the lower incidence of acute adverse reactions with LOCM and IOCM, possible cross-reactivity is presumed to exist between all iodinated contrast, regardless of osmolality.

Acuity of past reaction — Contrast reactions are categorized as immediate (also called acute) or nonimmediate (also called delayed).

●Immediate: Immediate reactions occur within minutes to one hour of exposure and are the primary concern when evaluating patients for future contrast administration. In the general population, with modern intravenous LOCM or IOCM, the observed rate of an immediate adverse reaction is approximately 0.6 percent of patients; the rate of severe adverse reactions is approximately 4 in 10,000, and death occurs in less than 1 in 170,000 patients [33,37].

Immediate adverse reactions with oral and intracavitary contrast occur but are less common than the immediate adverse reactions to intravascular contrast. The rates and manifestations of reactions to enteric contrast have not been well-studied.

●Delayed: Nonimmediate or delayed reactions occur more than one hour (usually >6 hours) after contrast administration and can occur several days later [38]. The most common forms are maculopapular exanthema or delayed-onset urticaria. Maculopapular exanthema are usually mild and transient and are believed to be T cell-mediated in most cases [39]. Rare desquamative reactions have been reported. A delayed reaction may or may not recur, and premedication is usually not used in these patients because it has not been shown to prevent recurrence. Delayed contrast reactions are discussed in more detail separately. (See "Radiocontrast hypersensitivity: Nonimmediate (delayed) reactions".)

Clinical features of immediate reactions — For patients reporting a past immediate reaction, the clinical features of the index reaction should be reviewed in detail. Immediate reactions are further classified based on the underlying physiology (allergic-like or physiologic) and severity (mild, moderate, or severe) [1]. This classification predicts the likelihood and probable severity of a recurrent reaction, and determines the need for premedication prophylaxis (algorithm 1).

Allergic-like (hypersensitivity) reactions — Allergic-like reactions are idiosyncratic, are not related to dose, may occur in patients without any history of prior exposure, and do not predictably recur after each contrast exposure.

Symptoms and signs of allergic-like reactions include [1]:

●Pruritus, urticaria or angioedema

●Nasal congestion, sneezing, conjunctivitis, rhinorrhea

●Widespread erythema

●Hoarseness or stridor, itchy/scratchy throat, with or without hypoxia (ie, upper airway compromise)

●Repetitive cough, wheezing, or chest tightness, with or without hypoxia (ie, lower airway compromise due to bronchoconstriction)

●Anaphylactic shock (ie, hypotension and tachycardia)

Clinical manifestations of allergic-like reactions are identical to those of immediate hypersensitivity reactions to other drugs. In the past when these reactions were mostly caused by HOCM, the underlying mechanism was usually not immunoglobulin E (IgE)-mediated [40,41]. However, some severe immediate hypersensitivity reactions to LOCM and IOCM may be IgE-mediated, and allergy evaluations are being performed more commonly for patients with past severe immediate reactions. This is discussed in more detail separately. (See "Allergy evaluation of immediate hypersensitivity reactions to radiocontrast media", section on 'Immediate reactions'.)

Prevention of recurrent reactions is discussed below. (See 'Prevention' below.)

Physiologic reactions — Physiologic (also called toxic) reactions are dose-dependent and not mitigated by corticosteroid prophylaxis. They are concentration-dependent and likely related to direct chemotoxicity, osmotoxicity, or binding of endogenous molecules [42-44]. Symptoms and signs of physiologic reactions include:

●Transient flushing, warmth or chills

●Headache, dizziness, anxiety

●Nausea/vomiting

●Metallic taste

●Arrhythmia

●Hypertension

●Isolated chest pain

●Convulsions, seizures

●Vasovagal reaction (ie, hypotension and bradycardia)

Reactions characterized by both allergic-like and physiologic manifestations should be classified as allergic-like. If classification is unclear, an allergic-like reaction should be assumed [1]. Prevention of recurrent physiologic reactions is discussed below. (See 'Preventing recurrent physiologic reactions' below.)

Cardiopulmonary arrest and pulmonary edema can be nonspecific endpoints of either a severe allergic-like or physiologic reaction [1].

Severity — Immediate reactions to iodinated contrast are classified as mild, moderate, or severe based on the nature of symptoms, their duration, and requirement for therapy.

Reactions are classified by severity as follows:

●Mild – Reaction is self-limited, does not progress, and rarely requires treatment.

●Moderate – Reaction usually requires treatment and may progress to a severe reaction if untreated.

●Severe – Reaction is life-threatening and can cause significant morbidity. Nearly all life-threatening contrast reactions occur within the first 20 minutes after contrast medium injection [1].

In patients who have had a reaction, any subsequent reaction is likely to be of the same severity [45,46]. If repeat contrast administration is planned, medical staff with expertise to treat the anticipated symptoms should be readily available.

Prevention — When referring patients with a prior moderate or severe immediate reaction for contrast-enhanced CT, consultation with the radiology service that will be performing the exam is advised. Whether or not the imaging could be modified to avoid contrast and whether or not premedication prophylaxis is appropriate should be discussed. These choices are highly individualized since the potential risks of a recurrent reaction should be weighed against the potential benefits of achieving an accurate diagnosis. The exam indication, available alternatives for diagnostic testing, and patient preferences should all be considered in the decision making (algorithm 1).

If contrast administration is planned, preparations should be made to perform the exam in a setting where medical expertise is available to treat a reaction.

Preventing recurrent allergic-like reactions — Patients who have had a prior allergic-like reaction or unknown-type of reaction to contrast have an approximately fivefold increased risk of developing a future allergic-like reaction if exposed to the same class of contrast medium again [42]. Patients with multiple severe allergies or asthma are at greater risk as well, although probably not to the same extent [1].

The approach to prevention of subsequent reactions depends on the severity of the previous reaction.

●Severe allergic-like – Future use of contrast should be avoided altogether when possible because recurrent reactions can occur despite changing contrast agents and premedication, and retrospective studies suggest that patients with past severe reactions are at risk for recurrent reactions that are similar in severity [45-47]. Other diagnostic tests, including noncontrast CT, ultrasound, or MRI should be pursued.

However, if the clinical scenario requires that contrast be given (this situation should be rare; eg, CT angiography for suspected aortic injury when MRI is not feasible), the exam should be performed with the following precautions:

•Setting – Performance in a setting with expertise (ie, code team) and equipment for advanced cardiopulmonary resuscitation on standby, because changing the contrast agent and administering premedication do not eliminate the risk of a repeat reaction.

•Premedication – Premedication that includes at least two doses of corticosteroids plus antihistamines (table 1). (See 'Premedication regimens' below.)

If the urgency of the examination prevents administration of the full course of premedication, an abbreviated regimen should be given, as time allows, taking into account the risks and benefits of making the diagnosis and risk of repeat reaction (final row of the table) (table 1).

Other interventions: There are two additional interventions to prevent recurrence of allergic-like reactions that may be appropriate in some settings.

•Change agent – Retrospective reviews have found that use of a contrast agent that is different from the one that caused the past reaction may also lower the incidence of recurrent allergic-like reactions, including changing from one LOCM to another [46-48]. The data are reviewed below. (See 'Change of contrast agent' below.)

•Referral to an allergist – If there is sufficient time and access to an allergist familiar with emerging literature, the patient can be referred for skin testing with a panel of contrast agents to identify specific agents that are safer alternatives for that individual. In the past, when HOCM was routinely used, severe contrast reactions were almost never IgE-mediated and skin testing was not useful. However, now that LOCM is standard, severe allergic-like reactions are even more rare, but there is mounting evidence that some are IgE-mediated. Note that studies in support of skin testing are relatively new, and familiarity with this approach is not widespread, so communication between the radiologist and allergist to see if testing is available is essential. The optimal time period in which to perform skin testing appears to be within two to six months of the index reaction, although testing after this time period can still be informative. The allergy evaluation is reviewed in detail separately. (See "Allergy evaluation of immediate hypersensitivity reactions to radiocontrast media".)

Note that test doses (or test injections) are not recommended for identifying patients at risk for repeat allergic-like reactions, because they do not predict tolerance of a full dose of contrast and can induce a repeat severe reaction in susceptible patients [41,49,50].

●Moderate or mild allergic-like – If contrast is deemed necessary, the CT exam should be preceded by corticosteroid/antihistamine prophylaxis and should be performed in a setting with the expertise to treat a recurrent reaction readily available (table 1). If the urgency of the examination prevents administration of the full course of premedication, an abbreviated regimen should be given, as time allows, taking into account the risks and benefits of making the diagnosis and risk of repeat reaction (final row of the table) (table 1). In patients who are premedicated for a prior mild allergic-like contrast reaction, the likelihood of a severe breakthrough reaction is low (<1 percent) [46]. Similar to the prevention of severe reactions, using a contrast agent that is different from the one that caused the past reaction may also lower the risk of a recurrence. (See 'Change of contrast agent' below.)

Premedication regimens — Standard premedication regimens are usually administered orally, beginning 12 to 13 hours before the examination, or intravenously administered beginning five hours before. If an urgent examination requires contrast and there is insufficient time for premedication, the exam indication and the risks and benefits of contrast administration should be discussed with a radiologist. In general, if the prior reaction was mild, an emergency and necessary CT exam requiring contrast should not be delayed for premedication.

●Oral – Use of premedication varies considerably with each practice, as it is unclear whether it is helpful in preventing moderate and severe reactions. Regimens containing two or more doses of corticosteroid beginning 12 or 13 hours before the procedure are better supported by evidence than regimens consisting of a single dose given closer to the procedure. The inclusion of diphenhydramine has not been specifically studied and can be considered optional, although antihistamines are commonly included [1].

For elective CT in outpatients, premedication is given orally using one of the following two regimens (table 1) [1]:

•Regimen 1: Three doses of 50 mg oral prednisone administered 13, 7, and 1 hour prior to contrast administration, plus 50 mg oral diphenhydramine administered 1 hour prior to contrast administration [51,52].

•Regimen 2: Two doses of 32 mg oral methylprednisolone administered 12 and 2 hours prior to contrast material administration, plus 50 mg oral diphenhydramine administered 1 hour prior to contrast administration [53,54].

The choice between these two regimens is determined by local radiology and institutional practice. Oral regimens shorter than 12 hours have not been shown to be effective; a two-hour oral regimen was shown to be no more effective than placebo [1,53].

●Intravenous – For patients in the emergency department or admitted to the hospital, shorter, 5-hour intravenous regimens are preferred and are not inferior to a 13-hour oral regimen (breakthrough reaction rates of 2.5 versus 2.1 percent) (table 1) [55]. In contrast, intravenous regimens shorter than four to five hours have not been shown to be effective [1].

•Option 1: Two doses of 200 mg intravenous hydrocortisone administered 5 and 1 hour prior to contrast material administration, plus 50 mg intravenous diphenhydramine administered 1 hour prior to contrast material administration.

•Option 2: Two doses of 40 mg intravenous methylprednisolone administered 5 and 1 hour prior to contrast material administration, plus 50 mg intravenous diphenhydramine administered 1 hour prior to contrast material administration.

Efficacy and adverse effects — Although premedication decreases the likelihood of recurrent allergic-like reactions, it does not prevent them altogether [47]. Breakthrough reactions occur in approximately 2 percent of patients [45,46]. In patients with prior breakthrough reactions, the likelihood of a subsequent breakthrough reaction is higher, at approximately 10 percent [46].

The best evidence supporting premedication was with HOCM, where the associated rates of acute reaction were high. With LOCM, now used universally in part because of its much lower rates of acute reactions, the benefit is less clear. Premedication decreases the likelihood of a mild reaction, but its efficacy in preventing moderate and severe reactions has not been proven [33-35,37,39-41,45,46,51,52,54,56-60].

A randomized trial in the general population showed that premedication prior to LOCM administration reduced the likelihood of all immediate adverse events from 4.9 percent to 1.7 percent [54]. This trial was not sufficiently powered to evaluate the efficacy of premedication in the prevention of moderate or severe reactions.

The incidence of a recurrent allergic-like reaction in high-risk nonpremedicated patients is estimated to range from 10 to 35 percent [33,41,46]. With premedication, the risk is estimated to be approximately 10 percent [45,46].

Due to the small effect of premedication and the rarity of contrast reactions in general, the number needed to treat to prevent one reaction is large. The estimated number needed to premedicate with a 13-hour oral regimen to prevent one allergic-like contrast reaction is 69 for any severity of reaction, 569 to prevent a severe reaction, and greater than 50,000 to prevent a death [57,59,60].

The direct adverse effects of premedication are minor [53,54]. Transient leukocytosis and asymptomatic hyperglycemia have been observed. Diphenhydramine may cause drowsiness and should not be taken shortly before operating a vehicle.

The delay imparted by a 12- or 13-hour premedication regimen can expose hospitalized patients to indirect risks. One study of an inpatient population showed that a 13-hour oral premedication protocol was associated with a median 25-hour delay in time-to-CT and prolongation in length of stay, as well as an increased risk of nosocomial infection. For these reasons, a shorter five-hour intravenous regimen is preferred for inpatients [59].

Change of contrast agent — A small number of retrospective studies have suggested that changing contrast agents can help to reduce the risk of recurrent allergic-like reactions [47,48]. In the largest, 771 cases in which the patients had a previous adverse reaction to contrast underwent repeat radiologic procedures using either the same agent again or a different agent, with or without premedications (although only a subset received a regimen involving two doses of corticosteroid) [48]. In patients not receiving premedication, the rate of recurrent reactions was 28 percent when the same contrast was given, compared with 5.2 percent when the contrast was changed. The rate was 8 percent when the patient was changed from one LOCM to another.

A second study specifically examined 150 patients with a history of past moderate-to-severe hypersensitivity reactions to LOCM who were re-exposed during 328 subsequent radiologic exams to the same or different LOCM with various degrees of premedication [47]. The rates of recurrence were 19 and 24 percent in patients with moderate and severe past reactions, respectively. Recurrences occurred in 28 percent when the same LOCM was used, compared with 13 percent when the LOCM was changed.

Although the quality of this evidence is low, the conclusions are consistent with the widely-held belief among allergists that premedication, regardless of the regimen, does not absolutely prevent anaphylaxis in a susceptible patient who is re-exposed to the same drug. Thus, in patients with past severe allergic-like reactions or reactions with any features of anaphylaxis, it is logical to change the contrast agent, in addition to administering premedication.

Preventing recurrent physiologic reactions

●Severe physiologic reaction – Contrast should be avoided. Other diagnostic tests, including noncontrast CT, ultrasound, or MRI should be pursued. If the clinical scenario requires that contrast be given (this situation should be rare; eg, CT angiography for suspected aortic injury and MRI is not feasible), the exam should be performed with expertise for advanced cardiopulmonary resuscitation (ie, code team) on standby. Corticosteroid prophylaxis is not helpful or necessary if there is certainty that the prior reaction was physiologic and not allergic-like. If in doubt, consider it to have been allergic-like and administer premedication.

●Mild or moderate physiologic reaction – If contrast is deemed necessary, the patient should be advised that the symptoms of the previous reaction may recur and reassured that they are likely to be transient. Premedication prophylaxis is not necessary.

ASSESSING RISK FOR CONTRAST-INDUCED NEPHROPATHY — Patients with severely impaired baseline kidney function (estimated glomerular filtration rate [eGFR] <30 mL/min/1.73 m²) and those with acute kidney injury may be at risk for contrast-induced nephropathy (CIN) following intravenous iodinated contrast administration. In patients with eGFR ≥30 mL/min/1.73 m², intravenous iodinated contrast can be safely administered. Oral and rectal contrast are not known to have any effect on kidney function (algorithm 2).

Whether intravenous contrast administered with CT is nephrotoxic is a topic of debate [61-70]. It has been associated with a transient increase in serum creatinine in patients with baseline eGFR <30 mL/min/1.73 m², but not in patients with baseline eGFR ≥30 mL/min/1.73 m² [71]. A meta-analysis controlling for comorbidities found similar rates of dialysis and mortality following contrast-enhanced CT versus unenhanced CT, even in the subgroup of patients with chronic kidney disease [72]. This issue is discussed in detail elsewhere. (See "Prevention of contrast-induced acute kidney injury associated with computed tomography".)

Indications for screening kidney function — Assessment of eGFR should be obtained prior to contrast-enhanced CT in patients whose kidney function may be impaired (calculator 1).

In addition, all hospitalized patients should have eGFR assessed prior to contrast unless they are receiving chronic hemodialysis, have evolving acute kidney injury, or require urgent imaging for a life-threatening condition.

The clinical features that would warrant assessment of eGFR in outpatients vary somewhat with each practice but may include the following:

●History of kidney disease – Solitary or transplanted kidney, history of kidney surgery, tumor, or chronic kidney disease.

●Hypertension requiring chronic medical therapy.

●Diabetes mellitus.

●Age >60 years (optional).

The eGFR is not useful prior to contrast-enhanced CT in these scenarios:

●Evaluation of a life-threatening condition (eg, severe trauma, suspected aortic dissection); contrast possibly would be given regardless of kidney function.

●Anuric patient undergoing hemodialysis.

●Evolving acute kidney injury; serum creatinine is a lagging indicator of kidney damage and is not reliable.

As a rule of thumb, calculating eGFR within 30 days for outpatients and two days for inpatients suffices. But temporal criteria for eGFR measurement vary with each practice and are modified by individual patient's clinical status.

PATIENTS WITH IMPAIRED KIDNEY FUNCTION

Patients with eGFR ≥30 — In patients with normal or mild to moderately impaired kidney function (ie, estimated glomerular filtration rate [eGFR] ≥30 mL/min/1.73 m²) and without acute kidney injury, no special precautions to prevent contrast-induced nephropathy (CIN) are required before contrast-enhanced CT.

Patients with eGFR <30 — Patients with acute kidney injury and patients with severely impaired kidney function (ie, eGFR <30 mL/min/1.73 m² but who are not receiving chronic hemodialysis) may be at risk for CIN. Other diagnostic tests, including noncontrast CT, magnetic resonance imaging (MRI), or ultrasound should be considered.

If contrast is to be administered, intravenous volume expansion with isotonic fluids may be an effective option. Other commonly studied prophylactic measures, such as intravenous sodium bicarbonate infusion and n-acetyl cysteine, are not well-supported by the literature and are not indicated. This issue is presented in detail separately. (See "Prevention of contrast-induced acute kidney injury associated with computed tomography", section on 'Prevention among high-risk patients'.)

There is no standardized regimen for volume expansion. Typical regimens (using intravenously administered normal saline or Ringer's lactate) are discussed separately. (See "Prevention of contrast-induced acute kidney injury associated with computed tomography", section on 'Volume expansion'.)

Evidence for using volume expansion prophylaxis is primarily derived from patients undergoing angiography [73-75]. Benefits of volume expansion in patients undergoing contrast-enhanced CT have not been shown [76].

Oral volume expansion has not been rigorously tested.

Patients with acute kidney injury — Patients with or suspected of having acute kidney injury (eg, due to sepsis, myocardial infarction, or a large-volume hemorrhage) should be assumed to have severely impaired kidney function and may be at risk for CIN. Assessment of eGFR is not useful in such patients. They should be managed similarly to those with severely impaired kidney function. (See 'Patients with eGFR <30' above.)

Patients receiving dialysis — No precautionary measures are necessary prior to contrast-enhanced CT in anuric patients receiving chronic hemodialysis. Since the kidneys are already nonfunctional, there is no additional risk to the kidneys from iodinated contrast administration. In patients who are oliguric (ie, peritoneal dialysis, chronic hemodialysis but still making urine), or in patients who are receiving temporary dialysis for acute kidney injury, there is a theoretical but as yet undemonstrated risk of accelerating progression to anuria or permanent hemodialysis with intravenous iodinated contrast. These patients should be managed similarly to those with severely impaired kidney function. (See 'Patients with eGFR <30' above.)

Patients who received contrast within preceding 24 hours — Recent iodinated contrast administration is not a contraindication for contrast-enhanced CT, nor are prophylactic measures for CIN indicated in this setting in patients who otherwise would not be considered at risk for CIN.

No absolute maximum volume of intravascular iodinated contrast that can be administered safely within a given time period has been established. In patients with normal kidney function, the half-life of most modern CT contrast agents is approximately two hours and, after 20 hours, essentially all of a single dose of iodinated contrast is eliminated.

In patients receiving intravenous iodinated contrast, a dose-toxicity relationship for acute kidney injury has not been reported. A dose-toxicity relationship has been shown in patients undergoing coronary angiography, but that relationship may be confounded (eg, patient morbidity, catheter dwell-time in the suprarenal aorta, case complexity) [77].

If multiple doses of intravascular iodinated contrast material must be administered within a 24-hour period, there is no reason to recheck eGFR prior to each dose. Even if a prior dose was nephrotoxic, it would not induce a change in eGFR for many hours.

SPECIAL POPULATIONS

Children — Intravenous contrast has the same benefits and uses in children as in adults. Contrast may be indicated for suspected nonaccidental trauma. Children have a lower rate of immediate hypersensitivity contrast reactions than adults [78]. The risk of contrast-induced nephropathy (CIN) in children is theoretically similar to that of adults but has not been formally tested. Contrast dose in children is weight-based and is determined by the radiologist.

Pregnant women — Iodinated contrast administered during pregnancy is not known to cause fetal harm. Neonatal hypothyroidism was a theoretical concern in the past but has been largely refuted by retrospective observational studies [79,80]. Intravenous contrast media crosses the placenta and enters the amniotic fluid and fetal circulation. Since the primary concern of potential harm to the fetus is exposure to ionizing radiation, the imaging protocol should err on the side of giving contrast if it may be helpful to avoid the need for repeat scanning.

Breastfeeding women — Infant ingestion of iodinated contrast excreted in breast milk is not known to be harmful. The amount that might be ingested and subsequently absorbed systemically is extremely small compared with clinically administered intravenous and oral doses [81]. If the mother wishes to not expose her infant even to any small amount of contrast, she may choose to discard her breast milk for 24 hours after she receives contrast material, but it is not medically necessary to do.

Patients with allergies or asthma — Patients with allergies to foods or medications appear to be at a modest (ie, two- to threefold) increased risk for allergic-like reactions to contrast, but avoidance of contrast or routine premedication for this reason is not recommended [1]. Note that seafood or shellfish allergies do not confer additional risk beyond that associated with having an allergy in general [82,83]. Similarly, patients with asthma are at some increased risk of an allergic-like contrast reaction, but neither avoidance of contrast nor routine premedication are indicated [33].

Pheochromocytoma — There is no increased risk of adrenergic crisis in patients with pheochromocytoma receiving modern intravenous iodinated contrast (ie, low-osmolality contrast media [LOCM], iso-osmolality contrast media [IOCM]) [84].

Sickle-cell disease — There is no increased risk of sickle crisis for patients with sickle cell disease receiving modern intravenous iodinated contrast (ie, LOCM, IOCM) [85].

Patients taking metformin — Metformin can increase the risk for lactic acidosis in patients with severe renal dysfunction (ie, estimated glomerular filtration rate [eGFR] <30 mL/min/1.73 m²) and is contraindicated in this setting. If the patient has normal or mildly impaired renal function (ie, eGFR ≥30 mL /min/1.73 m²), metformin need not be withheld before the CT examination, and the patient's renal function does not need to be followed after contrast administration.

Radioactive iodine imaging or therapy — Iodinated contrast interferes with imaging or therapy that requires thyroid uptake of radioactive iodine. In patients undergoing or about to undergo procedures involving radioactive iodine, iodinated contrast administration should be deferred, if possible. If a contrast-enhanced CT is necessary, a washout period of four to eight weeks is needed before imaging or therapy with radioactive iodine [86]. Consult with a nuclear medicine clinician or radiologist for specific advice.

SUMMARY AND RECOMMENDATIONS

●Appropriate contrast utilization is necessary to obtain acceptable diagnostic accuracy. Omitting contrast when it is indicated, or giving it when it is not, can lead to diagnostic and treatment errors, and unnecessary morbidity and cost. Added or repeat tests, delayed diagnosis, and accumulated patient radiation exposure are all potential adverse effects of a suboptimal imaging exam. (See 'Introduction' above.)

●Acute adverse reactions to contrast may occur after oral, intracavitary, or intravascular administration. For any history of an acute reaction to contrast, the provider should verify that the reaction was to an iodinated contrast (used for computed tomography [CT], radiography, fluoroscopy, or angiography) rather than to a noniodinated contrast agent used for other imaging modalities (eg, gadolinium used for magnetic resonance imaging [MRI]). The clinical details of the index reaction such as symptoms, their duration, and required therapy should also be obtained. (See 'Patients with past reactions to contrast' above.)

●The clinical features of the previous immediate contrast reaction should be classified based on the severity (mild, moderate, or severe) and underlying physiology (allergic-like or physiologic). This classification predicts the likelihood and probable severity of a recurrent reaction and determines the need for premedication prophylaxis (algorithm 1). (See 'Clinical features of immediate reactions' above.)

●Prevention of allergic-like reactions in patients with previous reactions depends on the severity of the index reaction because the majority of recurrent reactions are similar in severity to the index reaction:

•For patients with a prior severe allergic-like reaction, the safest approach is future avoidance of contrast and use of alternate imaging modalities. However, if repeat administration of contrast is necessary, the CT should be performed in a setting with expertise (ie, code team) and equipment for advanced cardiopulmonary resuscitation on standby. In addition, we recommend administering premedication prophylaxis (Grade 1C), and we suggest using a contrast agent that is different from the one that caused the previous reaction (Grade 2C). (See 'Prevention' above.)

•For patients with prior mild-to-moderate allergic-like reactions, we suggest that such patients receive premedication prophylaxis (Grade 2C). In addition, in patients with moderate past reactions, we suggest using a contrast agent that is different from the one that caused the previous reaction, if available (Grade 2C). (See 'Prevention' above.)

Commonly used premedication regimens include at least two doses of corticosteroid plus an antihistamine (table 1). Standard premedication regimens are usually administered orally over 12 to 13 hours, or intravenously over five hours. If an urgent examination requires contrast and there is insufficient time for premedication, the exam indication and the risks and benefits of contrast administration should be discussed with a radiologist. In general, if the prior reaction was mild, an emergency and necessary CT exam requiring contrast should not be delayed for premedication. (See 'Prevention' above.)

Finally, there is recent and growing evidence that, in patients with severe past allergic-like reactions to low-osmolality contrast media (LOCM), referral to an allergist for skin testing can identify safer agents for that individual, but familiarity with this is not widespread among either radiologists or allergists. (See "Allergy evaluation of immediate hypersensitivity reactions to radiocontrast media".)

●Patients with severely impaired renal function (ie, estimated glomerular filtration rate [eGFR] <30 mL/min/1.73 m²) who are not anuric receiving chronic hemodialysis may be at risk for contrast-induced nephropathy (CIN). In addition, patients with or suspected of having acute kidney injury (eg, due to sepsis, myocardial infarction, or a large-volume hemorrhage) should be assumed to have severely impaired kidney function and may be at risk for CIN. In these patients, other diagnostic tests, including noncontrast CT, MRI, or ultrasound should be considered. If contrast is to be administered, intravenous volume expansion with isotonic fluids may be an effective option (algorithm 2). (See 'Patients with eGFR <30' above.)  

●Intravenous iodinated contrast can be safely administered to patients who have stable eGFR ≥30 mL/min/1.73 m² and in those who are anuric on chronic hemodialysis. (See 'Patients with eGFR ≥30' above.)