| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Letters to the Editor |
Department of Clinical Radiology, Kent and Canterbury Hospital, Ethelbert Road, Canterbury, Kent CT1 3NG, United Kingdom
e-mail: mark.downes{at}ekht.nhs.uk
 |
Editor: |
|---|
 TOP Editor: ReferencesReferences |
|---|
The small number of patients who developed acute renal failure.—The NEPHRIC study was designed to evaluate rises in serum creatinine level, and the diagnosis of acute renal failure was a local clinical decision. The study was powered specifically to examine the hypothesis and as such met this requirement (and satisfied the publishing journal's reviewers). It is noteworthy, however, that six patients did (according to local definition) develop acute renal failure. All were in the low-osmolar arm of the study, and although six of 65 may be a low number, it is not a figure that a clinician can take lightly.
The significant differences in proteinuria.—The NEPHRIC study showed no correlation of high urinary albumin to serum creatinine level and therefore was not considered relevant with regard to renal tolerance. Morcos et al (4) do not give albuminuria a high rating as a risk factor. A literature search does not reveal any controlled prospective studies on the correlation between urinary protein excretion and the incidence of contrast agent–induced nephropathy.
The duration of diabetes.—There are no published data supporting the concept that the duration of diabetes is a predictor of contrast agent–induced nephropathy. Morcos et al (4) suggest that diabetes alone, without renal impairment, may not be a significant risk factor.
The use of angiotensin-converting enzyme (ACE) inhibitors.—I can find no reference to this in the article published by Aspelin et al (2), so it is unclear what was meant by this phrase.
The lack of standardized vigorous hydration.—All centers in the NEPHRIC trial had local regimens of hydration. Block randomization used in the study suggests that any local variations in hydration would affect both groups equally.
Dr Heinrich and colleagues, having questioned the validity of the NEPHRIC trial, proceed to comment on other trials.
The Rapid Protocol for the Prevention of Contrast-induced Renal Dysfunction, or RAPPID, study by Baker et al (5) was not designed to compare the incidence of contrast agent–induced nephropathy in patients receiving different contrast agents but rather the potential benefit of intravenously administered acetylcysteine. A higher rate of contrast agent–induced nephropathy in this study is irrelevant without knowing what the rate would have been in a comparative arm of similar patients receiving a low-osmolar contrast agent.
The abstract (to my knowledge this has not been published as a full article) presented at the 2003 RSNA Annual Meeting by Hardiek et al (6) was from a study with a total of 102 patients. Although two patients in the low-osmolar arm of the study developed contrast agent–induced nephropathy compared with no patients in the iso-osmolar arm, this was not a significant difference. It has to be considered that this is possibly due to underpowering of the study, and clinicians need to be careful when taking evidence from smaller studies that result in no difference between comparators.
The study by Katayama et al (7) that was cited was actually a review of different data and referred, in turn, to data on file at Bracco Diagnostics (Princeton, NJ). These data on file concerned a study comparing a low-osmolar with an iso-osmolar agent, although the available information does not state whether the patients included in the trial had any risk factors for contrast material–induced nephropathy.
In addition to the unconvincing nature of the arguments put forward by Dr Heinrich and coauthors as to the evidence that contradicts the NEPHRIC findings, they have also failed to include mention of evidence supporting the study.
Chalmers and Jackson published their article "Comparison of Iodixanol and Iohexol in Renal Impairment" (8) in the British Journal of Radiology in 1999. As with the article by Aspelin et al (2), results of this double-blind randomized study showed a significant advantage in relation to changes in serum creatinine level for the iso-osmolar contrast agent, iodixanol, over a low-osmolar contrast agent.
In addition, in 1995 European Radiology published "Renal Tolerance of Iotrolan 280: A Meta-analysis of 14 Double-Blind Studies" by Clauss et al (9). As the title suggests, authors of this metaanalysis reviewed data from 14 randomized double-blind studies to compare a number of low-osmolar contrast agents with an iso-osmolar contrast agent, iotrolan. The authors concluded that the iso-osmolar agent had less effect on renal function than did the low-osmolar comparators.
In summary, results of a number of prospective randomized clinical trials have shown that the reduction in osmolality seen when changing from high-osmolar to low-osmolar or from low-osmolar to iso-osmolar agents has a positive effect on the incidence of nephrotoxicity.
The continuing debate over the mechanism of contrast agent effects on the kidney is healthy, and all available data, clinical and nonclinical, should be considered. I do, however, believe that it is important that clinical decisions should be based on clinical outcome studies where these exist.
|
References |
|---|
|
TOPEditor:ReferencesReferences |
|---|
Department of Diagnostic and Interventional Radiology, University Hospital of Saarland, Kirrbergerstr, Homburg, Saarland 66421, Germany
e-mail: ramhei{at}uniklinik-saarland.de
Our comments about the NEPHRIC trial should demonstrate that our in vitro data do not necessarily conflict with the clinical data, because the NEPHRIC trial is discussed very controversially. It was not our aim to provide a complete discussion of all studies comparing monomeric and dimeric contrast media but to point out that there are also clinical studies in which investigators found no reduced nephrotoxicity of dimeric contrast media in comparison with monomeric contrast media. The comments questioning the validity of the NEPHRIC trial were not made by us but were mainly cited from a series of letters to the editor published in the same journal as the NEPHRIC trial (1).
In studies regarding the prevention of contrast agent–induced nephropathy, it is believed that not only rises in serum creatinine levels but also more clinically relevant end points such as persistent renal damage, need for dialysis, or mortality should be addressed (2). We believe that the number of patients who had developed acute renal failure (six patients) is too small to draw definitive conclusions about the comparative nephrotoxicity of two contrast media. A large number of clinical studies including more than 1000 patients and a metaanalysis with pooled data from 31 trials has been required to conclude that there is a statistically significant benefit for low-osmolar contrast agents compared with high-osmolar ones in patients with preexisting reduced renal function (3–5). It was suggested that although the results of the NEPHRIC trial are encouraging, one should not conclude that iodixanol is the answer to contrast agent–induced neuropathy and that, since previous clinical studies have yielded conflicting results, there is reason to believe that future studies of iodixanol may provide conflicting data as well (5).
In the NEPHRIC study, the duration of diabetes was significantly greater in the iohexol group, and more than twice as many patients in the iohexol group had proteinuria. It was suggested that these differences might have confounded the results (1,6). It is well known that reduced renal function in combination with diabetes is the most important risk factor for contrast agent–induced nephropathy. It is reasonable to assume that patients with a longer history of diabetes have a higher probability for diabetic nephropathy. This might be reflected by the proteinuria. In contrast with the statements of Dr Downes, there are a number of recent studies, including thousands of patients, that demonstrate diabetes to be an independent risk factor for contrast agent–induced nephropathy (7–10). Also, the duration of diabetes was considered to be a risk factor for contrast agent–induced nephropathy (11). Moreover, in a study including 260 patients, proteinuria was demonstrated to be one of the independent risk factors for contrast agent–induced nephropathy (12).
The higher rate of use of ACE inhibitors in the iodixanol group (78% in the iodixanol group vs 55% in the iohexol group [1]) might represent a further disadvantage for the iohexol group because, in a study including 7230 consecutive patients, it was demonstrated that pretreatment with ACE inhibitors was associated with a lower risk of contrast agent–induced nephropathy in patients with impaired renal function (10). However, the literature concerning the influence of ACE inhibitors on the incidence of contrast agent–induced nephropathy is not conclusive (10).
It was suggested that differences in hydration status caused by a nonstandardized hydration not adjusted for body weight might have influenced the results (1). In this context it is interesting to note that the patients in the iohexol group had a significantly greater body mass index than did those in the iodixanol group. Furthermore, it was criticized that there was no vigorous hydration (mean volume, 977 mL in the iodixanol group vs 934 mL in the iohexol group) because the volumes of hydration were markedly greater in other studies (1). Hydration is the most important measure for preventing contrast agent–induced nephropathy. It is believed that appropriate hydration may override any smaller differences in injected contrast media osmolality or viscosity (13) and that the importance of well-controlled and sufficient hydration status control cannot be overestimated (6).
Dr Downes suggests that it is irrelevant to compare the incidence of contrast agent–induced nephropathy in the NEPHRIC trial with that in other trials. However, in the NEPHRIC study itself such comparisons were made. It was stated that this study, in which a 3% incidence of contrast agent–induced nephropathy was found with iodixanol, appeared to have had a better outcome than did previous studies that used low-osmolar contrast medium alone and similar or better results than those in studies that additionally included acetylcysteine or fenoldopam (14). Therefore, it is important to note that the incidence of contrast agent–induced nephropathy in the iohexol group is in accordance with findings in the literature, whereas the incidence of contrast agent–induced nephropathy in the iodixanol group is much lower than that in the other studies that used iodixanol in similar groups of patients (33% in the study by Stone et al [15], 21% in the study by Baker et al [16], and 12% in the study by Boccalandro et al [17]). In a pooled analysis of results of the recent randomized trials in patients with impaired renal function undergoing angiographic procedures, it was demonstrated that the risk of contrast agent–induced nephropathy with iodixanol appears to be as low as that with iopamidol, iopromide, or iomeprol, whereas the risk with iohexol appears higher compared with iodixanol and iopamidol (18). This demonstrates also that one should bear in mind that the monomeric contrast media are chemically very different substances and that it is not possible to generalize from one monomeric contrast medium to the whole class of monomeric contrast media as was done in the NEPHRIC study.
We agree with Dr Downes that clinical decisions should be based on results of well-designed large-scale clinical studies. However, we agree with the other authors that the available data do not provide clear evidence that iodixanol is less nephrotoxic than the low-osmolar class of contrast media (19,20). In the guidelines from the European Society of Urogenital Radiology, it is stated that we must await results of further studies before a definitive conclusion can be reached, and it is recommended to use low- or iso-osmolar contrast media in patients with risk factors for contrast agent–induced nephropathy (21).
|
References |
|---|
|
TOPEditor:ReferencesReferences |
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
