| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Original Research |
1 Department of Radiology, University of California, San Francisco, 505
Parnassus Ave., Rm. M-391, San Francisco, CA 94143.
2 Research Department, American College of Radiology, Reston, VA.
4 Yale University School of Medicine, New Haven, CT.
Received March 7, 2005;
accepted after revision November 10, 2005.
Address correspondence to B. M. Elicker
(brett.elicker{at}radiology.ucsf.edu).
Abstract
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
MATERIALS AND METHODS. Surveys were e-mailed to 2,000 radiologists. The survey included questions on the issues just described and respondent demographics.
RESULTS. The response rate was 21%. Serum creatinine is the most commonly used screening method: 92% of respondents for inpatient examinations and 66% for outpatient examinations. Only 2% use estimated creatinine clearance. The average threshold serum creatinine value used to determine that patients should not receive IV contrast material is 1.5 mg/dL in 35%, 1.7 mg/dL in 27%, and 2.0 mg/dL in 31% (mean, 1.78 mg/dL). Diabetes slightly lowers the threshold creatinine (mean, 1.68 mg/dL). Fewer than 30% of respondents frequently administer IV contrast material to patients with a renal transplant or multiple myeloma. The most commonly used preventive measures include hydration (93%), reduction of contrast dose (77%), and administration of acetylcysteine (39%). Overall, no important differences were found among practice settings or level of specialization.
CONCLUSION. A wide variation of practice patterns is apparent in the screening and prevention of contrast nephropathy. In some cases, these patterns conflict with recommendations from the literature. The results of this study identify opportunities for further research and areas in need of improved practice guidelines.
Keywords: contrast media • contrast nephropathy • CT • radiology practice • renal disease • screening
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The survey was composed of 12 questions about general practices regarding the identification, screening, and prevention of contrast nephropathy in patients receiving iodinated IV contrast material for CT examinations. Because this is one of the first surveys to address these issues, we chose to focus on broad themes to discern overall practice patterns within the community, keeping in mind that these practices are often individualized to the particular patient. As is the case with practice guidelines in general, this survey identified a starting point on which to base decisions in dealing with the patient receiving IV contrast. We also attempted to focus on those practices that seem to be widely variable. Some scenarios (e.g., renal transplant, multiple myeloma, and gadolinium usage) are not commonly encountered in all practice environments, but we felt it was important to address them, given the absence of a widely accepted standard of care. The survey (Appendix 1), which also included four background questions regarding the respondent's specialty, experience, and practice environment, was developed in conjunction with the ACR research division.
|
|
The next four questions dealt with the respondents' use of laboratory values in the decision to administer iodinated IV contrast for CT. These questions focused on the use of serum creatinine or estimated creatinine clearance. In particular, we investigated the threshold levels above which IV contrast administration was deemed contraindicated. The influence of the risk factors of advanced age, diabetes, multiple myeloma, and renal transplant on this threshold level was also examined.
The frequency with which radiologists administer IV contrast to patients with a renal transplant and multiple myeloma was investigated. Respondents could choose a frequency of "Never," "Some-times," "Frequently," or "Almost Always."
Four questions about the use of preventive measures included IV hydration, acetylcysteine administration, fenoldopam administration, and a reduced IV iodinated contrast dose. One question asked which of these measures the respondent uses. The second question inquired about the frequency of using preventive measures in patient populations with specific risk factors. Two questions followed on the use of gadolinium chelate agents as a substitute for iodinated IV contrast and the types of examinations for which these agents are used.
Demographic questions focused on the respondents' practice environment, their level of specialization, and their experience. The Analytical Group collected and compiled the data.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Because potential respondents were chosen randomly from the ACR database, there were multiple responses from different individuals within single institutions. Of these, there were four separate responses from one institution, three separate responses from four institutions, and two separate responses from 21 institutions. Thus, of 421 respondents, 58 were from institutions from which more than one individual responded to the survey.
Screening for Patients at Risk for Contrast Nephropathy
When evaluating an inpatient before administering iodinated IV contrast for
CT, 369 (92.3%) of 400 respondents stated they use serum creatinine level as
one of the primary screening tests, whereas only eight (2.0%) use estimated
creatinine clearance. Overall, very few radiologists (21/400 = 5.3%; 95%
confidence interval [CI], 3.1-7.5%) use clinical history alone without
obtaining laboratory tests. However, when evaluating only outpatients, the
number of radiologists who use clinical history alone increased to 120 of 419
(28.6%; 95% CI, 24-32%), and 275 (65.6%; 95% CI, 61-70%) still use serum
creatinine as a primary screening test.
Regarding the use of laboratory tests, 82 (19.5%) of 420 respondents stated that all patients are screened with laboratory values. The percentage of the 420 respondents who obtained laboratory values on patients with specific risk factors is as follows: advanced age, 337 (80.2%); diabetes, 393 (93.6%); and multiple myeloma, 338 (80.5%).
Discriminatory Levels of Renal Insufficiency
Serum creatinine was the most common laboratory test obtained to assess for
underlying renal insufficiency. Most radiologists use a specific creatinine
threshold above which they withhold IV contrast in the average patient. This
threshold level was assessed in different patient subgroups. The average
threshold serum creatinine value for all patients was 1.78 mg/dL. There was a
wide SD (0.29), with 143 (34.9%) of 410 using 1.5 mg/dL, 111 (27.2%) using 1.7
mg/dL, and 127 (31.0%) using 2.0 mg/dL as a serum creatinine threshold. Fewer
than 6% of radiologists used a threshold of less than 1.5 or greater than 2.0
mg/dL.
|
|
Figures 3 and 4 show the frequency with which radiologists administer iodinated IV contrast to patients with a renal transplant or multiple myeloma.
Measures to Prevent Contrast Nephropathy
We then turned our attention to the use of measures to prevent contrast
nephropathy. The percentage of the 421 radiologists who used these measures is
as follows: IV hydration, 391 (92.9%); acetylcysteine, 165 (39.2%);
fenoldopam, 20 (4.8%); reduced volume of IV contrast, 325 (77.2%); and
replacement by iodixanol (Visipaque, Nycomed), 29 (6.9%).
Table 1 depicts the frequency of use of preventive measures in particular subgroups at risk for contrast nephropathy.
|
Replacing iodinated IV contrast with IV gadolinium chelate is a technique used by 68 (16.1%) of 421 radiologists. It was used for the following indications: evaluation of acute abdominal processes, 45/68 (66.2%); cancer-staging CT scans, 17/68 (25%); aortic dissection, CT 39/68 (57.4%); and pulmonary embolism protocol CT scans, 17/68 (25%).
Overall, there were no statistical differences in any of the responses between academic and private practice radiologists, different practice environments, or level of specialization (body imaging specialists vs non-body imaging specialists). Table 2 shows some important statistical comparisons.
|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Radiologists are continually faced with the decision of whether to administer IV iodinated contrast material for CT examinations, and one of the primary concerns (in addition to the risk of anaphylactoid reactions) is contrast nephropathy. The ACR's Manual on Contrast Media [8] is a valuable source for information on contrast nephropathy. It discusses risk factors for its development and recommendations for its prevention. However, the practical value of this manual is limited in that it provides broad recommendations but not specific parameters to guide clinical practice. For instance, the manual provides a recommendation to measure serum creatinine before iodinated IV contrast administration but does not address what specific creatinine thresholds are appropriate. There is a perceived lack of consensus in the radiology community regarding many of the issues of screening and prevention of contrast nephropathy. Even within single institutions, practice patterns seem to vary widely.
Screening for Patients at Risk for Contrast Nephropathy
The well-established risk factors for contrast nephropathy include
preexisting renal insufficiency, the combination of preexisting renal
insufficiency and diabetes, and the dose of IV contrast administered
[9-12].
Dehydration, advanced age, diabetes without preexisting renal insufficiency,
renal transplantation, and multiple myeloma are also possible contributors,
but they are not as well established
[13-16].
Patients with a combination of renal insufficiency and diabetes are at the
highest risk for developing contrast nephropathy
[7,
10,
16,
17]. Most of these risk
factors are easily discernible by clinical history alone, but renal
insufficiency may be clinically silent. Thus, much of the evaluation of a
patient before receiving IV contrast focuses on determining his or her
probability of having renal insufficiency.
In our study, most responding radiologists (92%) use serum creatinine as their primary screening test before administering iodinated IV contrast to inpatients. Very few use estimated creatinine clearance (2%), despite the fact that it is easily calculated by the Cockcroft-Gault equation and is more predictive of renal insufficiency than serum creatinine [18-20]. In one study, 15.2% of out-patients with a normal serum creatinine had an estimated creatinine clearance of 50 mL/min/1.73 m2 or less [21] (normal is 90 mL/min/1.73 m2 or more).
In outpatients, respondents were more apt to use clinical history alone (29%) in the evaluation of occult renal insufficiency, but serum creatinine was still frequently used (65.6%). The literature suggests that in outpatient and emergency department settings, most patients with renal insufficiency can be identified without using laboratory values. A simple questionnaire can identify more than 97-99% of outpatients or emergency department patients with renal insufficiency [22, 23]. This would certainly provide an easier and more cost-effective way to evaluate patients before giving iodinated IV contrast for CT examinations. In inpatients with multiple comorbidities, it is less clear who should be screened.
Discriminatory Levels of Renal Insufficiency
Serum creatinine clearly is the most commonly used laboratory value as a
screening examination before CT. The incidence of contrast nephropathy
increases dramatically after a threshold serum creatinine of 1.2-1.5 mg/dL is
reached [7,
24]. Yet, in practice, the
threshold creatinine used by radiologists varies widely. The average threshold
creatinine used in patients with no additional risk factors was 1.78 mg/dL.
However, the values used by radiologists vary between 1.5 and 2.0 mg/dL. An
individual's preferred threshold is a fine balance between the risk of giving
a patient contrast nephropathy and the risk of a nondiagnostic CT scan. The
appropriate balance is difficult to quantify scientifically and depends on
each individual's comfort level. Although the threshold should be
individualized to each patient, most radiologists (94%) use thresholds of
1.5-2.0 mg/dL. In terms of renal function, this is an extremely wide range.
Further investigation of an appropriate threshold for the average patient thus
is warranted.
The degree by which this threshold serum creatinine should be altered in the presence of additional risk factors is also uncertain. In our survey, the presence of advanced age, diabetes, renal transplantation, and multiple myeloma lowered the average serum creatinine threshold but only by a small amount. For instance, in the presence of diabetes, only 16% of respondents lowered their threshold creatinine value. The average threshold serum creatinine value in this patient subgroup was 0.1 mg/dL lower than in patients without other risk factors. It is clear from the literature that the combination of diabetes and renal insufficiency portends a significantly higher risk for contrast nephropathy than renal insufficiency alone. Thus, the appropriate average threshold in patients with additional risk factors is unclear.
Little consensus exists on how patients with renal transplant and those with multiple myeloma should be handled with regard to receiving IV contrast. The incidence of contrast nephropathy in renal transplantation patients is highly variable in the literature. In transplantation patients with a healthy renal function, Moreau et al. [25] showed no increased incidence; a more recent study, however, showed an incidence of contrast nephropathy of 21.2% in transplantation patients with a unenhanced average serum creatinine value of 2.3 mg/dL [26].
Multiple myeloma was originally thought to be a significant risk factor for developing contrast nephropathy. Subsequently, the risk has been shown not to be related to myeloma per se but rather to the presence of associated dehydration and hypercalcemia [27]. The incidence of contrast nephropathy in myeloma patients is approximately 0.6-1.25% compared to a risk of 0.15% in the general public [13].
In our study, radiologists were very conservative in using iodinated IV contrast in patients with a renal transplant or multiple myeloma. Interestingly, the participants were most conservative with myeloma patients despite evidence in the literature that these patients have a relatively lower risk of contrast nephropathy compared with patients with other risk factors such as renal insufficiency.
Measures to Prevent Contrast Nephropathy
The use of preventive measures to decrease the incidence of contrast
nephropathy is one of the areas in which there appears to be general agreement
within the radiology community and support in the literature. Using these
measures in at-risk patients enables radiologists to raise their threshold for
giving iodinated IV contrast. The most commonly used preventive measures are
IV hydration, reduction of IV contrast dose, and administration of
acetylcysteine. Seven percent of respondents used Visipaque to replace
standard iodinated IV contrast. Given that this choice had to be selected in
the "Other" category, the use of Visipaque is likely
underreported.
More than half of the respondents stated that they use some form of preventive measure in 75-100% of patients with renal insufficiency or with the combination of diabetes and renal insufficiency. Good hydration before receiving IV contrast and administering a reduced contrast dose are the preventive measures shown in the literature to be most consistently effective [28]. Thus, it seems appropriate that these are the most used preventive measures. The benefit of acetylcysteine, as the third most used measure, is promising but still controversial. A recent meta-analysis showed a reduction of 56% in the relative risk of contrast nephropathy in patients treated with acetylcysteine [29]. Although used in the past to prevent contrast nephropathy, a multicenter trial with fenoldopam showed no benefit [30]. How the use of preventive measures should affect a radiologist's serum creatinine or estimated creatinine clearance threshold is unclear.
Some radiologists use gadolinium to replace iodinated IV contrast in high-risk patients. Gadolinium has a significantly lower incidence of contrast nephropathy in the doses commonly used for MRI. The doses commonly used for CT examinations are higher than those used for MRI, but they still do not provide equal X-ray attenuation compared with iodine; examinations thus are usually limited by suboptimal vascular opacification. To achieve equal vascular opacification to iodine, a gadolinium dose significantly higher than that used for MRI would be required [31]. In animal models, equal attenuating X-ray doses of gadolinium show at least the same level of renal toxicity as iodinated contrast material [32]. In early clinical studies, gadolinium has shown varied results, with some studies showing a decreased incidence of contrast nephropathy [33] and others showing no difference when compared with iodinated contrast [34]. Most of these studies, however, have focused on angiography, not CT.
Only 16% of radiologists use gadolinium as a replacement for iodinated IV contrast, with the most common indications being abdominal-pelvic CT scans for acute processes and aortic dissection. The role of gadolinium in replacing standard iodinated contrast for CT examinations is very unclear. The relationship of gadolinium dose and its associated renal toxicity compared with the ability to interpret CT images adequately must be examined.
In conclusion, this survey illustrates the many issues in the screening and prevention of contrast nephropathy for which there is no general agreement in the radiology community. In several areas, however, there is a consensus, primarily the use of creatinine as a screening test before giving IV contrast and the selection of preventive measures. In some cases, these practices have a scientific basis; in other cases, however, they conflict with recommendations derived from the literature. Although research is ongoing with regard to many of the issues addressed in this survey, we propose the following guidelines based on the literature (Table 3).
|
The discrepancies between current practice patterns and recommendations from the literature may be explained by several factors. Misinformation, conflicting information, or a lack of information likely plays a role. Of course, in a busy clinical practice, it is more expedient to simply obtain a serum creatinine rather than calculate an estimated creatinine clearance. Medical-legal factors are also likely important. Radiologists tend to be overly conservative given that they are provided with little background information about patients. Although education about these issues is important, just as crucial is building an infrastructure to facilitate obtaining all the necessary background information about patients and to communicate with clinicians about these issues. This will enable radiologists to make more educated decisions and take a more active role in patient management.
The issues listed in Appendix 2 are ones with no consensus in the community and little scientific basis in the literature. These are areas that need further investigation before the development of guidelines.
As is evident in Appendix 2, the areas that lack data in the literature primarily involve the selection of patients who should receive iodinated IV contrast. These are also the areas for which there is the least consensus within the radiology community. One important step in resolving these issues would be to clearly define the risk of contrast nephropathy in patients with different risk factors, different levels of renal insufficiency, and different preventive measures taken. This risk has not been extensively studied with the particular contrast agents and doses used for CT. If these data were available, the radiologist could calculate the precise risk of giving a patient contrast nephropathy and then weigh that risk against the need for a contrast-enhanced examination. For instance, a particular patient with diabetes with an estimated creatinine clearance of x and having received a typical prehydration treatment would have a calculated risk of contrast nephropathy of 5%. This would provide a more rational means for evaluation before administering the contrast.
The primary limitation of this study is the general nature of the questions in the survey. As the first survey to address many of these issues, we decided to focus on overall themes in practice patterns. Obviously, these patterns are individualized to each patient and a survey of this type is unable to capture the nuances of a radiologist's decision-making process. The radiologist must take into account many factors, including the indication for the examination, prior administrations of contrast agents, the patient's comorbidities, and so on. It is important to keep in mind, however, that any guideline is simply a starting point and is adapted in each individual patient. In addition, radiologists often do not have this vital clinical information, making guidelines even more important.
Several important issues were not addressed in this survey primarily because of space limitations. The use of isoosmolar contrast material and bicarbonate in the prevention of contrast nephropathy is an issue that deserves attention and will have an essential role in future guidelines.
The response rate of 21%, although low when compared with regular mail surveys, is actually higher than most other e-mail surveys, whose response rates are in the range of 10-15% [35].
In summary, practice patterns among radiologists vary widely in the screening and prevention of contrast nephropathy. In many cases, these patterns conflict with recommendations derived from the literature or are not based on any scientific data. The results of this study identify opportunities for further research and areas in need of improved practice guidelines.
APPENDIX 1:Survey of Practices Regarding Administration of Iodinated IV
Contrast Material for CT
APPENDIX 2: Ongoing
Issuesa
|
a No consensus in the community and little information in the literature to propose guidelines.
Acknowledgments
We thank Jonathan Sunshine and the Research Division at the American
College of Radiology for their help in the development and administration of
the survey for this article.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  J. C. Weinreb Which Study When? Is Gadolinium-enhanced MR Imaging Safer than Iodine-enhanced CT? Radiology, October 1, 2008; 249(1): 3 - 8. [Full Text] [PDF]
|
|
|
|
|
|
B. R. Herts, E. Schneider, E. D. Poggio, N. A. Obuchowski, and M. E. Baker Identifying Outpatients with Renal Insufficiency before Contrast-enhanced CT by Using Estimated Glomerular Filtration Rates versus Serum Creatinine Levels Radiology, July 1, 2008; 248(1): 106 - 113. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
