| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Contrast Media |
1 From the Department of Radiology, University of Bonn, Hospital and Medical School, Sigmund Freud Str 25, 53105 Bonn, Germany (H.H.S., C.K.K., I.B.); and Department of Radiology, Charité, Humboldt University, Medical School, Berlin, Germany (U.H., U.S.). Received April 4, 2005; revision requested June 2; revision received July 4; accepted July 21; final version accepted September 8. Address correspondence to H.H.S. (e-mail: schild{at}uni-bonn.de).
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONADVANCES IN KNOWLEDGEReferences |
|---|
Materials and Methods: Local ethics committee approval and written informed consent were obtained. Patients were randomly assigned to the dimeric group, monomeric group, or concurrent control group. Acute (occurring during or within 1 hour after contrast agent injection) and delayed (occurring between 1 hour and 7 days after contrast agent injection) AEs were evaluated and categorized (mild, not requiring treatment; moderate, self-limiting AE requiring simple treatment; severe, AE requiring extensive treatment or endangering life). The two-sided Fisher exact test and the Mann-Whitney U and Wilcoxon signed rank tests were used for statistical analysis.
Results: A total of 895 patients were recruited. Acute AEs were noted significantly (P < .05) more often in the monomeric group (44.8% [133 of 297 patients]) than in the dimeric (23.7% [71 of 300 patients]) or control (9.4% [28 of 298 patients]) groups. Two moderate acute AEs were noted in one patient in each contrast agent group; all other acute AEs were mild. There was no significant difference in the overall incidence of delayed AEs between the dimeric and monomeric groups (53.1% [139 of 262 patients] vs 50.8% [131 of 258 patients]). Delayed cutaneous AEs were noted significantly (P < .05) more often in the dimeric group. A total 16 AEs (2.0%) were moderate; no AE was severe.
Conclusion: The dimeric contrast agent caused fewer acute AEs than the monomeric contrast agent; however, the dimeric and monomeric agents caused a similar overall number of delayed AEs. Delayed cutaneous symptoms were noted more often with the dimeric contrast agent. Both contrast agents were safe in that no severe AEs and only a few moderate AEs were observed.
© RSNA, 2006
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONADVANCES IN KNOWLEDGEReferences |
|---|
In spite of these advantages, dimeric contrast agents have only reluctantly been used on a broader scale. Apart from cost considerations, the limited use of dimeric contrast agents is probably due to reports that suggested a higher incidence of delayed adverse events (AEs) in association with dimeric contrast agents (1). However, published results are heterogeneous in this regard (6,13,14), and the majority of these AEs were categorized as mild (not requiring specific treatment).
AEs observed after a contrast media–enhanced examination may be caused not only by the injection of contrast media but also by (a) the patient's underlying disease (which can cause nausea), (b) concomitant medication (which can cause diarrhea), or (c) food intake (which can cause skin reactions). AEs (eg, nasal congestion) may even occur spontaneously, without any detectable cause. If the true incidence of contrast agent–related side effects is to be investigated, one must therefore include a reference group of patients in whom the contrast agent will not be administered. This is especially important when analyzing nonspecific (ie, mild AE) side effects or AEs that occur some time after contrast agent injection.
Thus, the purpose of our study was to prospectively evaluate AEs after intravenous injection of nonionic monomeric or dimeric contrast agents for computed tomography (CT) and to compare these effects with AEs in patients who underwent unenhanced CT.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONADVANCES IN KNOWLEDGEReferences |
|---|
Schering (Berlin, Germany) provided the dimeric contrast medium used in this study; however, the authors controlled the data and information submitted for publication.
This prospective single-blinded randomized controlled trial was conducted over a 6-year period (1996–2002) at an academic tertiary care center and included patients who were referred for body or craniofacial CT. A detailed history was taken before the CT examination. Patients were excluded if they (a) were younger than 18 years, (b) were pregnant or lactating, (c) were known to have or suspected of having hyperthyroidism, (d) had undergone injection of dimeric contrast media within the past 4 weeks, (e) had cardiac insufficiency and an unstable clinical condition, or (f) would require medical or surgical therapy within the 7-day follow-up period. Patients were also excluded if further administrations of contrast agent(s) were planned or expected during the 7-day observation period or if medical treatment would probably be modified within the observation period. When a patient was deemed eligible for inclusion, details of the study were explained and informed consent was obtained. However, whether a patient was eventually included in the study depended on the randomization, as will be explained.
Sampling Method
The decision to administer an intravenous contrast agent was based on the individual patient's clinical indication for CT. The type of examination (with or without contrast media enhancement) that was indicated on clinical grounds was compared with the method indicated by the next free random number (ie, contrast-enhanced or unenhanced CT). If the next random number was for contrast-enhanced CT and the patient was referred for unenhanced CT, he or she was not included in the study, even if all other inclusion criteria were met. Likewise, if the next random number was for unenhanced CT and an intravenous contrast agent was required on clinical grounds, the patient was not included. A rationale for this was to exclude any seasonal variables that might have confounded the results; without this precaution, recruitment of control patients who underwent unenhanced CT for paranasal sinus problems would have likely been skewed toward winter. Since seasonal differences in the incidence of AEs are known to exist, this type of sampling was used even if it slowed recruitment.
If injection of an intravenous contrast agent was indicated on clinical grounds, the patients were randomized to either the nonionic monomeric group (iopromide [Ultravist 300; Schering]) or the nonionic dimeric group (iotrolan [Isovist 280; Schering]). Osmolality of iopromide and iotrolan was 607 and 290 mosm/kg H2O, respectively; iodine content of iopromide and iotrolan was 300 and 280 mg iodine per milliliter, respectively. Viscosity at 37°C was 4.7 mPa · sec for iopromide and 6.8 mPa · sec for iotrolan. Patients in whom contrast material was not administered served as a concurrent comparison group and were referred to as the control group.
CT Examinations
CT was performed in accordance with current standards of care and with different acquisition techniques, depending on the clinical indication; two single-section spiral CT systems were used (Somatom Plus, Siemens Medical Systems, Erlangen, Germany; Tomoscan, Philips Medical Systems, Best, the Netherlands). If injection of an intravenous contrast agent was indicated on clinical grounds, a standardized volume (100 mL) was injected at a standardized injection rate of 2 mL/sec with a power injector. Patients in the control group underwent CT, but intravenous contrast material was not administered.
Data Collection
To record acute AEs (those that occur during or 1 hour after contrast agent injection), all patients were closely observed during contrast agent administration and for 1 hour after contrast agent administration (in the control group, patients were observed for 1 hour after the start of CT); thereafter, a detailed physical examination (I.B., H.H.S., and seven CT attending physicians) was conducted. The attending physician recorded all AEs that were observed by the attending physician or described by the patient. No attempt was made to categorize any symptoms in terms of the likelihood with which they were actually due to or associated with the use of the contrast agent.
To record delayed AEs (ie, those occurring more than 1 hour and up to 7 days after the injection), patients received a questionnaire (Figure) on which they were instructed to record all changes in their health status during the 7 days after CT. They were instructed to send the completed questionnaire to the study site after 7 days. Patients were instructed to call the study site if they observed unusual symptoms or if minor symptoms did not resolve or required treatment. A telephone number was provided in the questionnaire; the decision as to whether a patient should undergo a physical examination was based on the symptoms the patient described on the telephone.
|
Data Analysis
For data analysis, the severity of AEs was further characterized in accordance with current clinical practice, as follows (1): Mild AEs were defined as those that did not require treatment. Moderate AEs were those that were usually self-limiting and required simple treatment. Severe AEs were those that required extensive or invasive treatment, required treatment for an extended period of time, or endangered the patient's life or well-being.
As a subgroup of acute AEs, allergy-like acute AEs (rash, breathing difficulty, or angioedema [Quincke edema]) were analyzed separately.
Statistical Analysis
Demographic details and anamnestic variables at study entry, as well as AEs during the 7-day follow-up period, were analyzed according to their frequencies and with statistical testing at a descriptive level.
AEs recorded more than once by the same patient were treated as a single AE, with onset and resolution coinciding with the first and last times, respectively, these AEs were recorded. For comparison of the incidence of acute or delayed reactions and for comparison of the incidences of acute or delayed allergic reactions, the two-sided Fisher exact test was used. Age and body mass index were compared with the Mann-Whitney U and Wilcoxon signed rank tests between (a) the monomeric group and the dimeric group and (b) each contrast agent group and the control group. For selected questions, the incidences of events in treatment groups were compared with the Fisher exact test. A P value of .05 was used to indicate statistical significance. Since the study protocol did not include any confirmatory statistical analyses, these results should be interpreted as only descriptive. Before analysis, thorough plausibility checks were conducted. Statistical software (SAS, version 8.02; SAS Institute, Cary, NC) was used for analysis.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONADVANCES IN KNOWLEDGEReferences |
|---|
Participants in each study group were comparable in most aspects (Table 1). Participants had a personal history of various diseases or symptoms, which most frequently affected the circulatory system (201 [22.4%] of 895 patients), joints (148 [16.5%] of 895 patients), or gastrointestinal tract (120 [13.4%] of 895 patients); there was no significant difference between the groups. Concomitant medication was taken in all groups, with no differences regarding frequency of intake or type of medication.
|
While the treatment groups (ie, the monomeric group and the dimeric group) did not differ to a significant degree regarding any of the demographic or medical data that were collected, there were important differences between the treatment groups and the control group: Clinical indications for unenhanced versus contrast-enhanced CT are different, and since the underlying disease or clinical question determines the indication, there were also differences regarding the respective reasons for referral; patient age; prevalence of ear, nose, and throat (ENT) symptoms (nasal congestion); and allergic predisposition (Table 1) between the treatment and control groups.
In the treatment groups, oncologic indications were most prevalent, occurring in 505 (85.4%) of 591 patients. In the control group, CT was performed mostly for nononcologic (especially ENT-related and orthopedic indications) reasons in 266 (92%) of 289 participants. The large number of patients with ENT indications and fluctuating sinus-related symptoms that were due to allergies explains the differences regarding age, prevalence of ENT symptoms (nasal congestion), and prevalence of allergic predispositions between the control group and the treatment groups. Preexisting symptoms (those recorded before the start of CT) are listed in Table 2.
|
|
The majority of acute AEs in all groups (230 of 232 AEs [99.1%]) were categorized as mild (70 [99%] of 71 acute AEs in the dimeric group, 132 [99.2%] of 133 AEs in the monomeric group, and all 30 AEs in 28 patients in the control group).
Of the patients who developed an acute AE, 36 (27%) of 133 patients in the monomeric group, 16 (23%) of 71 patients in the dimeric group, and 13 (46%) of 28 patients in the control group had a history of general allergic reactions. These findings reflect almost exactly the incidence of allergic disposition in the three patient groups.
Allergy-like Acute AEs
The number of patients who developed allergy-like AEs was low: Allergy-like acute AEs were observed in three (1.0%) of the 300 patients in the dimeric group, in five (1.7%) of the 297 patients in the monomeric group, and—surprisingly—in three (1.0%) of the 298 patients in the control group. The overall incidence was too low for detection of possible differences in the incidence between groups.
Delayed AEs
Of the 895 enrolled patients, 772 returned the questionnaires, with no significant difference between the three groups. Of the 772 patients, 16 (2.0%) developed delayed AEs that were categorized as moderate. Moderate delayed AEs were reported by 11 (4.2%) of the 262 patients in the dimeric group, by four (1.6%) of the 258 patients in the monomeric group, and—surprisingly—by one (0.4%) of the 252 patients in the control group (Table 4). The difference between the dimeric group and the control group was significant (P < .006); there was no significant difference between the other groups. In all 16 patients, skin rash or itching was treated with oral antihistamines, cortisone ointment, or both. Interestingly, premedication of antihistamines, a corticoid, or both was administered in three of the 15 patients in the treatment group (one patient in the dimeric group and two patients in the monomeric group).
|
In the control group, the high incidence of delayed AEs was due to a high reported incidence of nasal congestion; typical ENT-related symptoms were seen more frequently in the control group than in the treatment groups. When patients with ENT symptoms were excluded from the control group, a more meaningful analysis of the respective incidence rates of delayed AEs was possible. A total of 93 control patients without ENT symptoms returned the questionnaire. The size of this control group seemed large enough to serve as a reference for calculating the baseline value (ie, the spontaneous incidence of symptoms that were counted as AEs).
The overall rate of delayed AEs (moderate and mild) did not differ between the dimeric and monomeric groups (123 [46.9%] of 262 patients versus 117 [45.5%] of 258 patients, respectively). If the control group was "cleared" of ENT patients, a delayed AE rate of 40% (37 of 93 patients) was observed; this rate was significantly (P < .001) lower than the delayed AE incidence rates observed in patients in the dimeric group.
Delayed cutaneous symptoms (itching or skin rash) were reported significantly more often (P = .027) in the dimeric group (43 of 262 patients [16.4%]) than in the monomeric group (25 of 258 patients [9.7%]) (P < .03) or the control group with (14 of 252 patients [5.5%]) (P < .001) or without (six of 93 patients [6.4%]) (P < .015) patients with ENT symptoms. In turn, some symptoms were observed more frequently in patients in the monomeric group (Table 4).
Regarding sex differences, delayed AEs were seen significantly more often (P < .01) in female patients in the dimeric group, while there was no sex-related difference regarding the incidence of delayed AEs in the monomeric or control group.
Delayed AEs were reported more often in patients who had a history of allergies in the control group (P = .022); this was not the case in the monomeric and dimeric contrast media groups (P = .136 and P = .09, respectively).
The majority of delayed AEs were reported during the first 4 days of the observation period (Table 5). A peak for the incidence of cutaneous symptoms was observed on day 2 of the observation period in the dimeric group.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONADVANCES IN KNOWLEDGEReferences |
|---|
In our cohort of 895 patients, acute AEs were observed in 232 (25.9%) patients: 204 (34.1%) acute AEs were observed in 597 patients in the treatment groups, and 28 (9.3%) acute AEs were observed in 298 patients in the control group. This latter value indicates that almost one-third of the 34.1% incidence of acute AEs in the treatment groups may not be attributed to the use of a contrast agent. Still, this rate is higher than that reported for acute AEs in other trials (4,15,16).
In our study, the high rate of acute AEs was probably due to several causes. One cause was the fact that patients were closely observed throughout the acute phase (ie, during and up to 1 hour after contrast agent injection). A second cause was that we did not attempt to categorize AEs according to the likelihood that they were due to contrast agent administration. Instead, all symptoms that were observed or found during physical examination or reported by the patient were recorded. The rationale was to exclude any observer bias; however, this necessarily increased the number of AEs (particularly mild AEs) reported in this trial compared with that reported in other trials. This possible overestimation of the incidence of acute AEs is accounted for when the rate of AEs in the control group is taken into account. The emerging risk for contrast agent–induced acute AEs is in fact comparable to reported values for both the monomeric and the dimeric contrast agents.
Overall, acute AEs were observed significantly more often in the monomeric group than in either the dimeric group or the control group. Heat sensation was the one symptom that was most frequently reported as an acute AE in both treatment groups; however, it was reported significantly more often in the monomeric group. This was expected; in fact, heat sensation is only questionably classified as an AE, as it could also be considered a physical or physiologic effect (16) of the intravenous injection of any hyperosmolar solution. Even when the patients who reported heat sensation as the only AE were excluded from analysis, the incidence of acute AEs was significantly higher in the monomeric group than in the control group; no such difference was observed in the dimeric group. The incidence of acute allergy-like reactions was low in all groups—too low, in fact, to investigate differences between groups.
Virtually all acute AEs were categorized as mild and did not require treatment. Moderate acute AEs were reported in only two patients (one from each treatment group). In both cases, the AE was a rash that was successfully treated with histamine-1 or histamine-2 receptor antagonists and, in the case of one patient, cortisol injection. No acute AE was categorized as severe.
While the detection, documentation, and analysis of acute AEs are usually straightforward processes, the same is not true for the study of delayed AEs (17). To study delayed AEs, we asked patients to complete a questionnaire. This approach is not ideal, as differences in patients' perception or ability to observe themselves and differences in the understanding of what is asked for in the questionnaire may confound the results. However, the control group data should compensate for these effects. The alternative to self-reporting, direct patient observation, was pursued (18) in a prospective study of 7000 inpatients in whom iohexol was administered for CT or urography and who were followed up for 7 days by their respective physicians. However, contrast material concentration, volume, or both were not standardized between patients. More important, observations were made by an unknown number of physicians at 131 hospitals, and it was at the discretion of each physician to interpret the symptoms as an AE.
About half of the patients in the dimeric, monomeric, and control groups (53.1%, 50.8%, and 48.4%, respectively) reported symptoms that were categorized as delayed AEs during the 7-day follow-up period; the overall rates of delayed AEs were comparable in all three study groups. This equivalent incidence across treatment and control groups was probably caused by the many patients with ENT symptoms in the control group who accounted for the high incidence of nasal congestion or paranasal sinus–related symptoms during follow-up. When patients with ENT symptoms were excluded from the control group, the incidence of delayed AEs decreased to 40% (37 of 93 patients), which is significantly different when compared with that in the dimeric group (P = .03).
The incidence of delayed AEs in both contrast agent groups was high compared with that in previous reports (13,18,19). The reasons for this high incidence are probably the same as the ones mentioned previously for our higher incidence of acute AEs. In addition, a prospective study will yield more AEs than will a retrospective study. In one study (13), the low rates of delayed AEs could probably be explained by the retrospective design of the study, particularly the long interval (from several months to 1 year) between the time of contrast agent injection and the time of patient contact.
A factor that will increase the number of AEs, irrespective of other confounding factors, is the number of symptoms mentioned in the self-reporting questionnaire: The more symptoms from which patients may choose, the more likely AEs are to be reported. Our questionnaire offered approximately 30 choices of symptoms—as such, it probably revealed higher rates of AEs than a questionnaire that offered only three symptoms, such as the one used by Hosoya et al (16), who reported an incidence of only 9.4% for delayed AEs. On the other hand, Sakai et al (20) used a questionnaire with 12 choices, and they reported delayed AEs in 45.8% of the patients after intravenous injection of a nonionic monomeric contrast agent (iomeprol), which is in good agreement with our findings.
If the incidence of delayed AEs was analyzed by group, the following statements could be made: As opposed to previous reports (5,15,16), in our prospective randomized controlled trial we did not find a significant difference in the overall incidence of delayed AEs in the dimeric group compared with that in the monomeric group. Only when we analyzed AEs by specific type did we find a significantly higher incidence of cutaneous symptoms (itching or skin rash) in the dimeric group; this is in agreement with previously published data (14,17). Delayed skin reactions were reported by 16.5% of patients in the dimeric group and by 9.7% of those in the monomeric group. When the incidence of delayed cutaneous AEs in the control group was taken into account, delayed cutaneous AEs attributable to the injected contrast agent seemed to occur in less than 11% of the patients in whom dimeric contrast agents were administered and in more than 4% of the patients in whom monomeric contrast agents were administered. These incidence rates are again in line with published data (14).
Sex was a risk factor for a delayed AE after injection of the dimeric contrast agent, as female patients developed significantly more delayed AEs. This finding was in line with the observation that female sex was identified as an independent risk factor for adverse drug reactions in general (21).
In our cohort, more than 80% of delayed AEs occurred more than 1 day after the contrast agent was administered (ie, when the patient was probably no longer under medical observation). While this is a concern, one must understand that the majority of AEs were categorized as mild and were, to a substantial extent, also observed in the control group (ie, they were not related to the contrast agent). None of the delayed AEs were severe, and few reports of severe delayed AEs can be found in the literature on monomeric or dimeric (ie, iodixanol) contrast agents (19,22).
Our study had limitations. As stated previously, one limitation was the use of a questionnaire. Another potential limitation was the fact that the composition of the control group did not exactly match that of the treatment groups. However, controlled comparisons between patients in the monomeric or dimeric group and patients in the control group are inevitably subject to fundamental difficulties—which is probably why such a study had not been undertaken. As the ideal randomized double-blind comparison of identical examinations with and without contrast agent administration is impossible for both practical and ethical reasons, we designed a study that approached this design as closely as was practicable. Our control group was not ideal; however, it was better to have an imperfect control group than to not have any control group. Our data provide insight into the prevalence and incidence of AE-like symptoms in patients who underwent CT but in whom no contrast agent was administered. Notably, and we feel more important, the treatment groups were randomized, so they matched in every aspect and were comparable.
In addition, another preselection bias resulted from the fact that critically ill patients and patients who were scheduled to undergo surgery or major interventions during the week after CT were excluded. These patients are known to exhibit higher incidences of AEs (18); thus, it is conceivable that the actual incidence of AEs in unselected groups of patients—specifically, critically ill patients—may be higher.
In conclusion, we performed a prospective randomized controlled trial to analyze acute and delayed AEs after CT with and without injection of either a monomeric or a dimeric contrast agent. Compared with the monomeric contrast agent, the dimeric contrast agent led to a significantly lower incidence of acute AEs and a significantly higher incidence of delayed skin reactions. The vast majority of acute or delayed AEs after injection of monomeric or dimeric contrast agents were categorized as mild (ie, not requiring any treatment); only a small fraction (1.9%) were categorized as moderate, with no significant difference between groups.
Accordingly, if there is a choice between nonionic monomeric and dimeric contrast agents, the increased risk of delayed (usually mild) skin reactions must be weighed against the benefit of reduced pain during contrast agent injection (especially arterial injection) and—in high-risk patients—the potential benefit of reduced incidence of contrast material–induced nephropathy.
|
ADVANCES IN KNOWLEDGE |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONADVANCES IN KNOWLEDGEReferences |
|---|
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
Abbreviations: AE = adverse event • ENT = ear, nose, and throat
Authors stated no financial relationship to disclose.
Author contributions: Guarantors of integrity of entire study, all authors; study concepts/study design or data acquisition or data analysis/interpretation, all authors; manuscript drafting or manuscript revision for important intellectual content, all authors; manuscript final version approval, all authors; literature research, all authors; clinical studies, H.H.S., C.K.K., U.H., I.B.; statistical analysis, H.H.S., C.K.K., U.S.; and manuscript editing, H.H.S., C.K.K., U.H.
|
References |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONADVANCES IN KNOWLEDGEReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  M J Fernandez Cabezudo, G Petroianu, B Al-Ramadi, and R D Langer Iosimenol, a new non-ionic dimeric contrast medium, does not induce immunoreactivity in the popliteal lymph node assay Br. J. Radiol., September 1, 2007; 80(957): 713 - 718. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
