| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Letters to the Editor |
Research and Development, GE Healthcare Bio-Sciences, Nycoveien 2, N-0401 Oslo, Norway
e-mail: tore.skotland{at}ge.com
Editor:
We would like to comment on the article entitled "Cytotoxic Effects of Ionic High-osmolar, Nonionic Monomeric, and Nonionic Iso-osmolar Dimeric Iodinated Contrast Media on Renal Tubular Cells in Vitro" by Dr Heinrich and colleagues (1), which was published in the June 2005 issue of Radiology. The authors describe the effects of exposure of five different contrast media for up to 24 hours on LLC-PK1 cells, from an epithelial cell line derived from porcine proximal tubules, by using conversion of tetrazolium salt (MTT) as a measure of cytotoxicity. Overall, the experimental part of the study appears to have been well conducted, but we have reservations regarding the interpretation and relevance of the data to safety assessment.
The data presented by Dr Heinrich and colleagues in figure 1 of the article (1) show that iomeprol at a concentration of 150 mg of iodine per milliliter has a rather small effect on the MTT conversion when half of the growth medium has been exchanged with the contrast medium for up to 5 hours (approximately 8%, 15%, and 45% less conversion than the control cells after incubations for 1, 5, and 24 hours, respectively). The authors performed all the other experiments with a 24-hour incubation time, which, of course, is not relevant to the in vivo situation with the rapid excretion of these agents. In view of the data published by Hardiek et al (2), who reported a reversible effect for another nonionic monomeric agent by using a very similar test system (incubation with LCC-PK1 cells for 24 hours and measuring MTT conversion), one wonders just how relevant this measure is to nephrotoxicity in vivo, which involves frank tubule necrosis and which may lead to acute renal failure. There are many examples in the literature which show that there is often no correlation between such in vitro cell data and those obtained in animal experiments. This is amply illustrated by comparing the in vitro data presented by Dr Heinrich and colleagues with data we have reported in an in vivo model (3). Furthermore, the problem of using in vitro cell systems for differentiating between the toxicity of different contrast agents has been reported by several previous investigators and has been illustrated by Peer et al (4), who showed that there was a different ranking of the four contrast agents they tested, dependent on which of the four cell lines was used.
Despite the admission of Dr Heinrich and colleagues of the shortcomings of in vitro systems, they attempt to use their data to claim "a greater potential for nephrotoxic effects of the dimeric contrast media molecules when they are compared with the monomeric contrast media molecules." This claim is based on comparison of monomers and dimers on an equimolar basis. When they compared the data at the same iodine concentration, slightly less cytotoxicity was reported for the dimeric agents than for the two nonionic monomeric agents, although this difference was stated not to be statistically significant. Their conclusion of a greater nephrotoxic potential for the dimeric contrast media fails to recognize that iodinated contrast media are dosed in terms of iodine content, with dimers administered at half the molar dose compared with monomers, because they contain twice the number of iodine atoms and therefore have twice the efficacy per molecule.
In conclusion, in vitro cell culture systems, although of value to examine certain aspects of the intrinsic chemotoxicity of iodinated contrast media, cannot be used to predict or reflect in vivo nephrotoxicity in either experimental animal or human studies, and attempts by authors of such in vitro studies to make recommendations to radiologists on human safety should be discouraged. From an evidence-based medicine perspective, there is currently only one type of study that should be used as the reference standard for the potential of different iodinated contrast media to induce human nephrotoxicity, and that is by careful observations in well-designed randomized human studies.
References
* Department of Diagnostic and Interventional Radiology, University Hospital of Saarland, Kirrberger Strasse, 66421 Homburg/Saar, Germany*
e-mail: ramhei@uniklinik-saarland.de
Department of Diagnostic Radiology, University Hospital of Erlangen, Erlangen, Germany
Drs Skotland and Grant advanced the opinion that a 24-hour incubation time is not relevant to the in vivo situation. But we know that the elimination half-life of the contrast media is increased to up to 70 hours in patients most at risk for developing contrast medium–induced nephropathy (CIN) due to preexisting renal impairment (1,2). Therefore it seems likely that the renal proximal tubular cells could be exposed to the contrast media for 24 hours in these patients.
Drs Skotland and Grant doubt the relevance of the MTT assay, indicating that the inhibition of MTT conversion by contrast media should not be reversible because CIN in vivo would involve frank tubule necrosis. However, frank necrotic cells are not a required finding in acute renal failure (3). Moreover, clear documentation of tubular necrosis associated with CIN from biopsies in humans is lacking, and the low fractional excretion of sodium frequently observed in CIN is an argument against tubular necrosis as the primary cause of CIN (4,5). Therefore, induction of tubular necrosis is not considered obligatory when evaluating the value of an experimental model of CIN (4). The MTT assay is a well-established method of evaluating the cytotoxic effects of various substances in cell cultures. There is evidence for contrast medium–induced mitochondrial dysfunction as the cause of the inhibition of MTT conversion, which is accompanied by an increase in extracellular adenosine and a depression in cellular adenosine triphosphate content, which explains the reversibility (6).
Drs Skotland and Grant question the use of in vitro cell systems and state that there is often no correlation with animal experiments. However, there is no experimental model of CIN or acute renal failure, whose validity in predicting effects in patients has been generally accepted. Thus, the relevance of animal studies using rats or rabbits to investigate the renal tolerance of contrast media is considered to be low to humans (7,8). As a validating factor for an in vivo or in vitro model of CIN, a better tolerability by means of exposure to low-osmolar nonionic contrast media in comparison with high-osmolar ionic contrast media was proposed due to the fact that this difference can be found also in humans (4,8). This is the case in our model, as well as in previously investigated cell cultures, which speaks for the validity of the model (4,8).
We agree that trials in humans are the reference standard for comparing the nephrotoxic potential of drugs in humans and that, of course, the results of any experimental model cannot simply be transferred to humans. However, clinical studies in the field of CIN are frequently hampered by methodologic limitations: multiple drugs, use of serum creatinine value as surrogate parameters for renal function, nonstandardized hydration, lack of long-term results, and small numbers of patients (9). The conflicting results about the prophylactic use of N-acetylcysteine despite a large number of human trials demonstrate this problem well (10). This applies even more to the comparison of the nephrotoxicity of iso-osmolar and low-osmolar contrast media, for which much less data from clinical studies are available. We acknowledge that experiments in cell cultures do not take into account the probably multifactorial pathogenesis of CIN. However, direct cytotoxic effects are thought to be an important part of the nephrotoxic potential of contrast media (5), and it is believed that this direct cytotoxicity is best studied in vitro, because of the absence of confounding variables (11).
Of course, contrast media are dosed in terms of iodine content in clinical practice, and, at equal iodine concentrations, dimerics contain half the number of molecules of monomerics. This is why we have also compared them at equal iodine concentrations. We did not find a smaller cytotoxic effect of the dimeric contrast media compared with the monomeric ones at equal iodine concentrations corresponding to results of previous cell culture studies (6,12,13). To compare the cytotoxic effects of the contrast media molecules, however, it is necessary to compare the contrast media at equal molarity under exclusion of differences in osmolality. Results of these experiments demonstrated a greater cytotoxic effect of the dimeric contrast media molecules compared with the monomeric contrast media molecules.
References
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
