ACADEMIC EMERGENCY MEDICINE 2011; 18:1005–1009 © 2011 by the Society for Academic Emergency Medicine
Objectives: Despite increasing attention to the long-term risks of radiation exposure and contrast-induced nephropathy (CIN), institutional guidelines and patient consent procedures for contrast-enhanced computed tomography (CECT) imaging in the emergency department (ED) setting have focused primarily on more immediate complications, directly attributable to the administration of intravenous (IV) iodinated contrast administration. Thus, this study sought to define the risk of these immediate complications with the overall aim of improving institutional guidelines and patient consent procedures.
Methods: This was a prospective, consecutive cohort study of patients undergoing CECT of any body region in the ED, for complications occurring within 1 week of contrast administration, using predefined implicit definitions. Severe complications were defined as any of the following requiring medical or surgical intervention: bronchospasm with acute respiratory failure, airway obstruction, anaphylactoid shock, or acute pulmonary edema. The development of compartment syndrome, lactic acidosis, or pulmonary edema within 1 week of contrast administration was also considered a severe complication.
Results: Of 633 patients, only five (0.8%, 95% confidence interval [CI] = 0.3% to 1.8%) reported any immediate complications, all of which were classified as minor. No patient developed a reaction meeting the study definition of a severe complication.
Conclusions: The frequency of severe, immediate complications from CECT imaging that includes IV contrast is less than 1%, and the frequency of mild complications is less than 2%. The authors conclude that CECT is associated with a very low rate of severe immediate complications.
Recent reports have focused on the toxic effects of radiation and contrast from computed tomography (CT) imaging. Several reports have found a 7% to 11% frequency of contrast-induced nephropathy (CIN) during a more intermediate period (2 to 7 days) after CT angiography.1–3 However, many patients who develop CIN followed by severe renal failure have multiple other potential causes for their kidney injury. So, the precise contribution of the contrast load as the cause of the renal failure remains a matter of debate. Additionally, several reports have estimated a 0.1% to 0.7% attributable risk of developing cancer as a long-term consequence of exposure to radiation from CT imaging to the torso.4 However, the primary basis for these risk estimates relies upon highly extrapolated information sources, such as data obtained in Hiroshima, Japan, after detonation of the atomic bomb in 1945. Thus, the cause–effect relationship between contrast-enhanced CT (CECT) and acute renal failure and the magnitude of the cause–effect relationship between CT and cancer remain uncertain.
From the viewpoint of patients, and to a lesser extent, clinicians, CECT can cause several immediate complications that would undoubtedly be ascribed as being directly caused by the imaging procedure. These include mishaps related to injection of the contrast, primarily from contrast extravasation, hypersensitivity reactions, complications from acute volume overload, and lactic acidosis resulting from concurrent metformin use. In theory, the most severe manifestations of these potential complications include compartment syndrome, bronchospasm with acute respiratory failure, airway obstruction, anaphylactoid shock, acute pulmonary edema, or severe lactic acidosis. From experience, most physicians would describe the occurrence of these events in clinical practice, particularly severe complications, as very low, but these potential complications remain the major focus of the consent process when discussing the use of iodinated contrast agents with patients. Furthermore, concern for these complications has been the major focus of institutional policies guiding the use of these agents and aftercare instructions to patients.5 With the use of iodinated contrast agents rapidly increasing in emergency medicine (EM) practice, the primary aim of this study was to document the incidence of these immediate complications in patients also being followed for CIN. Our primary hypothesis was that severe immediate complicates of CECT are rare and occur in <1% of patients undergoing CECT in the emergency department (ED) setting. Patients were followed for both outcomes, and this secondary analysis was preplanned prior to the initiation of the study.2 With the aim of placing the occurrence of immediate complications of CECT in the context of CIN, we will summarize those data in this report. However, a full report of the incidence of CIN and related outcomes has been previously published.2
This was a planned, secondary observational study of a consecutive cohort of ED patients undergoing imaging for another study. This analysis was conducted with institutional review board approval.
Study Setting and Population
The study was performed at Carolinas Medical Center in Charlotte, North Carolina, which is an urban, academic center with an annual ED volume >110,000 patients per year. The patients were enrolled from June 2007 through November 2008 and had received intravenous (IV) iodinated contrast, followed by CECT imaging of various anatomical locations.2
The ED uses two multidetector Siemens Somatom Sensation 64-slice scanners (Siemens Medical Solutions USA, Inc., Malvern, PA) 24 hours a day, 7 days a week. The institutional protocols for imaging studies represented in this study use 120 mL of Iopamidol-370 (Isovue-370, Bracco Diagnostics, Princeton, NJ), delivered IV via automated injector (2.5 to 5 mL/sec) through an IV catheter (minimum of 20 gauge) placed in either upper extremity, proximal to the wrist and distal to the axilla. We reviewed the imaging and gantry reports to verify the volume, route, timing, and mechanism of delivery of contrast administered for the CECT study.
With regard to immediate complications, our institution has the following existing protocols: 1) pretreatment of patients with any reported severe allergy to iodinated contrast IV with diphenhydramine (25 to 50 mg), plus either methylprednisolone (125 mg) or hydrocortisone (100 mg) prior to emergent imaging. Treatment is recommended 1 hour prior to contrast administration, but exceptions for clinical urgency are allowed. Severe allergies qualifying for this protocol include a history of severe bronchospasm, severe laryngeal edema, severe angioedema, unresponsiveness, seizure activity, cardiac arrhythmias, or cardiopulmonary arrest following the administration of iodinated contrast; 2) discontinuation of metformin for 48 hours following contrast administration; and 3) postprocedural observation (minimum of 2 hours) with recommended orthopedic consultation for the cutaneous extravasation of 120 mL of iodinated contrast, followed by written instructions for follow-up with a radiologist or radiology nurse within 24 hours and instructions to return to the ED for any increased pain or the development of skin or sensory changes.
We followed this cohort of patients for known complications of IV, iodinated contrast including any of the following: hypotension, respiratory symptoms, oral swelling, rash (or subjective pruritus), abdominal pain, nausea or vomiting, altered mental status, IV extravasation, and the development of lactic acidosis within 1 week of contrast administration. Severe immediate complications (primary outcome) were defined as any of the following requiring medical or surgical intervention: bronchospasm with acute respiratory failure, airway obstruction, anaphylactoid shock, or acute pulmonary edema. The development of compartment syndrome, lactic acidosis, or pulmonary edema within 1 week of contrast administration was also considered a severe complication.
Patients were also followed for the outcomes of CIN, severe renal failure, and death from renal failure.2 Briefly, CIN was defined as an increase in serum creatinine of ≥0.5 mg/dL or ≥25% of baseline within 2 to 7 days of contrast administration. The combined outcome of severe renal failure and death from renal failure was defined as either of the following within 45 days of contrast administration: 1) an absolute increase in serum creatinine from <3.0 mg/dL to ≥3.0 mg/dL and/or 2) death from renal failure as determined by two of three blinded reviewers.
The methods of follow-up have also been previously published and include beside data collection, a follow-up visit with study personnel within 1 week of contrast administration for collection of a blood sample,2,6 a review of the ED chart from the index visit and electronic medical records, and patient interviews that were conducted by blinded study personnel.2,6 Patients were enrolled in real time at the time that the CECT order was placed and were followed throughout their ED stay by study personnel. Patients were also interviewed in person during the follow-up blood draw. For patients who missed their initial follow-up blood draw, up to five attempts were made to contact the patient by telephone for a brief interview. We also attempted to reschedule the blood draw during these telephone calls, if time permitted. Starting at 7 days, electronic medical records were reviewed for both immediate complications of CECT and evidence of CIN. Telephone interview, electronic medical record review, and a search of the social security death index were also used to identify cases of severe renal failure and death from renal failure at 45 days.2
We reported overall outcome incidence and population characteristics, as proportions with associated 95% confidence intervals (CIs), calculated using the Clopper-Pearson method (STATSDirect V3.3 software, Chesire, UK). Age data are reported as a mean with the associated standard deviation (SD).
We enrolled and followed 633 patients for the outcomes of severe and minor complications of CECT. The enrollment of patients, has been previously published.2 Briefly, we approached 1,110 ED patients and enrolled 664 patients. We excluded 23 patients whose CECT was cancelled or converted to a noncontrasted study after enrollment and 8 patients who requested to be withdrawn, leaving 633 patients included in this analysis. Population characteristics of this cohort are reported in Table 1.7 With the exception of one patient whose contrast injection was halted for extravasation, all patients received 120 mL of iopamidol per the institutional imaging protocols.
|Characteristic||n = 633|
|Age (yr), mean (±SD)||47 (±15)|
|Female sex||57 (53–61)|
|African American||40 (36–44)|
|Prior reaction to iodinated contrast*||5 (4–7)|
|Asthma or COPD||6 (4–8)|
|Allergy to one or more medication†||3 (2–4)|
|Congestive heart failure||7 (5–9)|
|Diabetes mellitus||17 (15–21)|
|Vascular disease||10 (8–13)|
|Baseline renal insufficiency‡||8 (6–10)|
Within this cohort, only 5 patients (0.8%, 95% CI = 0.3% to 1.8%) reported any immediate complications. Subjective pruritus was reported by three patients, none with documented urticaria or other skin findings, and only one required treatment. Nausea was reported by one patient who did not require treatment. None of these patients reported a preexisting contrast allergy or history of reactions of any kind to IV contrast.
Of the 33 patients who did report a prior history of allergic or other reactions to contrast, treating physicians identified only two patients with a history of symptoms prompting any pretreatment with any form of diphenhydramine and/or steroid. None of the remaining 31 patients developed immediate complications. One patient reported a history of hypotension and respiratory distress with iodinated contrast, prompting the administration of IV diphenhydramine and methylprednisolone, given 30 minutes prior to imaging. This patient remained asymptomatic following contrast administration. The second patient presented with a history of asthma and clinical findings of bronchospasm and reported a history of worsening bronchospasm following iodinated contrast administration. The patient was given oral prednisone and nebulized albuterol prior to imaging, but was not given diphenhydramine. The patient’s respiratory symptoms were improving prior to contrast administration and continued to improve following contrast administration. The same patient was the only patient within the cohort with an episode of contrast extravasation. The infusion was stopped at 100 mL and was felt by the treating physician to represent the full volume extravasated into the soft tissue of the forearm. This was treated with the application of an ice pack. No additional treatments or specialty consultation were required for extremity symptoms, and the patient was discharged with a 3-day course of oral steroids following resolution of the respiratory symptoms. The patient remained asymptomatic (extremity and respiratory) at follow-up performed at 48 hours following contrast administration and at 45 days. A total of 34 patients reported metformin use at the time of contrast administration. None of these patients developed lactic acidosis (0/34, 95% CI = 0 to 10%). No patient (0/633, 95% CI = 0 to 0.6%) developed evidence of a primary outcome, a severe immediate complication as defined by this study.
Comparatively, the incidence of CIN in this population was 11% (70/633, 95% CI = 9% to 14%). Following the development of CIN, six patients went on to develop severe renal failure, and four of the six patients with severe renal failure died. Reviewers determined that renal failure significantly contributed to all four of these deaths. One patient, who did not develop CIN initially, did have an episode of severe renal failure within 45 days of contrast administration and survived beyond the 45-day follow-up period. Thus, the relative risk of the combined outcome of severe renal failure and/or death from renal failure following CIN was 48 (95% CI = 8 to 302).2
The most notable finding of this study is the low incidence of even minor immediate complications of iodinated contrast and the absence of severe immediate complications. To place this study in perspective, it is commonplace to have multiple institutional protocols aimed at mitigating these potential adverse events. These protocols can include mandatory screening, pretreatment regimens that potentially delay imaging to protocols that result in major changes in medical therapies, mandatory specialty consultation, and added follow-up. More so, these are the major focus of education and clinical surveillance and dominate the current patient consent process for these studies. While these protocols do likely assist in the management of rare but potentially serious complications, the risk of more delayed complications, such as CIN and radiation exposure are less well recognized, despite recent evidence of greater prevalence. In fact, CIN occurred in 11% of patients in this study and was associated with a significant risk of severe renal failure and death from renal failure.2
Historically, the incidence of immediate adverse complications following the administration of low-osmolar agents is around 3%, which is considerably decreased from 15% seen with older, high-osmolar agents.8 The lower rate seen in this study has also been observed in recent literature, which suggests that the rate of complications is even lower among patients undergoing CT imaging. The results of this study are similar to the rate of immediate adverse events recently published for iodixanol use in CT imaging.9 Iodixanol, like iopamidol, is a commercially available, low-osmolar, nonionic contrast agent currently used at several large institutions in the United States and Europe. In this multicenter study, the incidence of any adverse immediate reaction occurred in less than 1% of patients, and serious events occurred in only 0.05%. In the only blinded, prospective study comparing low-osmolar, nonionic agents, Gomi et al.10 reported that the incidence of adverse events was as high as 4% (iomeprol); among the 9,000 patients undergoing CT imaging using five agents, fewer than 1% required any treatment for adverse events, and none had events that were classified as serious. However, this study only followed patients for 1 hour after contrast administration. As a result, complications such as lactic acidosis, soft tissue complications, and delayed pulmonary edema could not be detected. Furthermore, this study also used contrast volumes at least 20% lower (<100 mL) than those used in this study.10 These two studies represent the sum total of previously published prospective data documenting the incidence of immediate reactions from IV, nonionic, low- or nonosmolar iodinated contrast.
Notably, there are several retrospective studies of immediate reactions to both IV and intraarterial iodinated contrast media. The largest of these is a retrospective study of 298,491 doses of low-osmolar agents. Immediate complications occurred in 0.2% of patients and were primarily limited to urticaria and nausea. Only 0.03% required any treatment and only 15 patients were classified as having a severe reaction. In contrast, this study defined immediate complications as only those occurring within 30 minutes of contrast administration and did follow patients for soft tissue complications.11
Patients who are considered for CECT are routinely asked about prior episodes of severe hypersensitivity reactions to iodinated contrast agents, usually in the same manner as patients are questioned about allergic reactions to medications. Patients with a clear history of severe reactions to iodinated contrast agents are usually not offered CECT and instead undergo noncontrasted CT and/or other imaging modalities. This study is not designed to identify these cases; rather, this study was designed to detect reactions that are unanticipated or occurring in patients with a history of a minor reaction or in whom history is not as clear or convincing or is unavailable. Furthermore, the constraints of patient care in the ED do not allow for pretreatment initiated 12 to 24 hours prior to imaging, as is currently supported by the literature.8 In other words, this design evaluates the incidence of immediate complicates following standard clinical evaluation and physician decision-making. This study was performed at a single academic, urban ED, and the results may vary in other settings. Only 60% of patients approached for enrollment agreed to participate. This may reflect the inclusion requirement for the patient to be willing and able to provide two blood samples, to return to the ED, or to accommodate a home health visit and agree to a telephone interview and medical record review. In fact, 32% of enrolled patients did not complete the follow-up blood draw. However, this would only affect the detection of lactic acidosis, which is a severe enough complication that these patients would have also likely required medical attention or died. Based on prior studies documenting the reliability of telephone and medical record review in this population, we are reasonably confident of the accuracy of these methods in identifying this outcome.6 Finally, the protocol excluded critically ill or severely injured patients, who account for approximately 20% of all CECT in our ED.12
We interpret these data to indicate that the frequency of severe, immediate, complications from CT imaging that includes IV iodinated contrast is less than 1%, and the frequency of mild complications is less than 2%. We conclude that CECT is associated with a very low rate of severe immediate complications.