| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Opinion |
1 From the Department of Radiology, University of Pittsburgh Medical Center–Presbyterian, 200 Lothrop St, Room D-132, Pittsburgh, PA 15213-2582; Department of Radiology, Loma Linda University Medical Center, Loma Linda, Calif (D.R.B.); Department of Radiology, Emory University Hospital, Atlanta, Ga (D.R.M.); Department of Diagnostic Sciences, Faculty of Health Sciences, University of Copenhagen, Copenhagen, Denmark (H.S.T.); and Department of Diagnostic Radiology, Copenhagen University Hospital, Herlev, Denmark (H.S.T.). Received July 18, 2007; revision requested July 27; revision received August 2; accepted August 10; final version accepted August 16. Address correspondence to E.K. (e-mail: ekanal{at}pitt.edu).
On May 23, 2007, the Food and Drug Administration (FDA) requested that a "black box" warning (one of the more extreme mechanisms that the FDA may invoke for calling attention to observed serious adverse reactions) regarding the potential risk of nephrogenic systemic fibrosis (NSF) in patients with renal failure be added to the product descriptions of all five FDA-approved gadolinium-based magnetic resonance (MR) contrast agents marketed in the United States: Magnevist (gadopentetate dimeglumine; Bayer Healthcare, Wayne, NJ), MultiHance (gadobenate dimeglumine; Bracco Diagnostics, Princeton, NJ), Omniscan (gadodiamide; GE Healthcare, Oslo, Norway), OptiMARK (gadoversetamide; Tyco-Mallinckrodt, St Louis, Mo), and ProHance (gadoteridol; Bracco Diagnostics). All five agents were treated equally, in this regard, despite data suggesting that certain agents possessed a seemingly greater risk for precipitating NSF than did others. The authors of this commentary believe this implied equality among the five gadolinium-based agents available in the United States to be imprudent in light of present information.
We are concerned as to how the implication of equivalent risk among gadolinium-based contrast agents may impact agent selection for renally impaired patients. As such, we provide the following points of information so that all health care providers may make a more fully informed decision regarding the selection of contrast agents for use in patients with renal impairment.
1. Dr Shawn Cowper, who first described NSF in 2000, maintains the Yale NSF Registry, which contains nearly 250 patients. A retrospective review of this database has shown that among patients in whom an exposure could be verified within 2–3 months of NSF onset, approximately 85% of these patients had been administered Omniscan prior to receiving a diagnosis of NSF; the rest had been administered Magnevist (unpublished data, July 16, 2007). Similarly, Thomsen reported, "More than 150 patients have developed NSF after exposure to a Gd-based contrast medium. The overwhelming majority (~90%) had had gadodiamide with certainty" (1).
2. As of March 12, 2007, a review of the peer-reviewed literature (2) revealed a total of 74 cases of NSF, 72 of which were associated with prior Omniscan administration; one, with prior Magnevist administration; and one, with no gadolinium-based agent identified. There is also one case reported in the peer-reviewed literature (3) of NSF developing in a patient who had received MultiHance and who subsequently received Omniscan before developing NSF.
3. Although the FDA MedWatch database is an unmonitored and largely unverified reporting system, as of April 17, 2007, there were 160 Omniscan-associated cases of NSF, 73 Magnevist-associated cases, and three OptiMARK-associated cases. There were also several known confounded cases reported in which more than one gadolinium-based agent had been administered. There were no reports of NSF after isolated MultiHance or ProHance administration.
4. The UK Commission on Human Medicines in conjunction with the European Pharmacovigilance Party (4) published that, as of June 26, 2007, 180 cases of NSF worldwide had been reported with Omniscan and 78, with Magnevist (with one case in a patient who had received MultiHance and Omniscan.) Several cases have also been reported with OptiMARK in the United States.
5. The June 26, 2007, public assessment report issued by the Commission on Human Medicines together with the European Pharmacovigilance Working Party (4) states that since the 1980s more than 30 million patients had received Omniscan and 80 million had received Magnevist. Of importance, this market share was not proportionate to the number of cases reported for each of these contrast agents, as recorded either in the literature or in the FDA MedWatch system.
6. Studies performed by Bayer Healthcare–Schering (5) indicated a substantial discrepancy among various gadolinium-based contrast agents with regard to the incidence of NSF-like skin lesions in animals following prolonged high-dose administration of the various contrast agents tested. While these initial experiments have not yet been published, we believe these results to be sufficiently concerning and compelling to warrant our mention of them (5).
7. The dominant theory regarding NSF development and gadolinium deposition in tissues of patients with renal disease is related to the possibility of transmetallation and release of free gadolinium from the chelating agent. The thermodynamic conditional stability constant defines the affinity of the various gadolinium-based agents, at physiologic pH, to bind the gadolinium ion and prevent its release as toxic free gadolinium. Subject to considerations of in vitro versus in vivo comparison, it is notable that this constant varies substantially among the FDA-approved contrast agents and is lowest for Omniscan and OptiMARK by 100- to 1000-fold, compared with the values for Magnevist, MultiHance, and ProHance (6). Omniscan and OptiMARK are also the only contrast agents with a substantial amount of excess chelating agent added to the commercially distributed preparation (with Omniscan containing less excess chelating agent than OptiMARK [6]).
8. When dosing information was provided, the majority of Omniscan-associated NSF cases reported in the peer reviewed literature involved high-dose (>0.1 mmol · kg–1) administrations. The FDA update recommends that "[w]hen administering a GBCA [gadolinium-based contrast agent], do not exceed the dose recommended in product labeling" (7). It is important to note, however, that the product labeling for Omniscan (8) permits a total of triple the standard dose (ie, 0.3 mmol · kg–1) to be administered within 20 minutes. Despite strong evidence that suggests a relationship between administered dose and the likelihood of developing NSF, no comment is made in the FDA update regarding the advisability of ensuring that the administered dose is kept as low as is absolutely necessary.
9. On February 7, 2007, the UK Medicine and Healthcare Products Regulatory Agency chose to recommend that Omniscan be contraindicated for patients with severe or end-stage renal disease (9). On June 26, 2007, in an updated statement from that group, Magnevist was added to the list of contraindicated contrast agents (4) in patients with severe or end-stage renal disease, with Omniscan also contraindicated in patients who have undergone or who are awaiting liver transplantation and who have any level of renal dysfunction. In this document they explicitly state the following:
Risk of NSF is considered to be highest with Omniscan and OptiMARK, which carry no molecular charge, are arranged in a linear structure with excess chelate, and seem more likely to release free Gd3+ into the body. Those that are cyclical in structure (eg, ProHance, Gadovist, and Dotarem) are least likely to release free Gd3+ into the body. Between these two groups are those that carry a molecular charge and have a linear structure (eg, Magnevist, MultiHance, Primovist, and Vasovist) (4).
10. Thomsen noted that from 2002–2005, while Omniscan was used at his site, a total of 27 cases of NSF were diagnosed at his institution. Yet, in the more than 1 year (since March of 2006) that his institution has discontinued the use of this agent in favor of a macrocyclic gadolinium-based agent, their site has not seen any new cases of NSF develop despite continued administration of the macrocyclic agents to patients with renal insufficiency in whom contrast-enhanced MR imaging was medically indicated (2) (written communication, H. Thomsen, July 10, 2007).
The FDA did note the following:
Where a specific agent was identified, the most commonly reported agent was Omniscan, followed by Magnevist and OptiMARK. NSF has also developed following the sequential administration of Omniscan and MultiHance and Omniscan and ProHance. The distribution of the number of reports for the individual GBCAs may relate to multiple factors, including more limited use of some GBCAs, under-reporting of NSF, characteristics of the agent and a lack of patients' complete GBCA exposure history (7).
However, the FDA update does not list the possibility that the unequal distribution of NSF reported among the various gadolinium-based agents may relate to unequal relative risks among those agents, and it does not acknowledge that the actual risk of developing NSF following Omniscan administration may be higher than that after administration of some or all of the other agents.
The FDA did specifically state, "Because reports incompletely describe exposure to gadolinium-based contrast agents, it is not possible to know if the extent of risks for developing NSF is the same for all agents" (10). They also stated, "The lack of complete information of GBCA [gadolinium-based contrast agent] exposure, exposure to multiple GBCAs, along with similarities among all these contrast agents, make it impossible at present to definitively determine whether the extent of risks for developing NSF are the same for all the GBCAs or vary for some of them. Therefore, FDA considers that NSF may occur following the administration of any of the approved GBCAs to certain patients" (11). However, the FDA has now mandated that black box warnings and other warnings be added to the product labeling of all gadolinium-based contrast agents despite, the FDA's acknowledgment that the data only "suggest that GBCAs play a role in the development of NSF" but have not been "definitively determined" to do so (12).
In light of the evidence cited above, it is our collective opinion that the risk of a patient with renal disease developing NSF appears to not be the same among the various gadolinium-based MR contrast agents available in the United States and abroad today. We share the opinion and concern that at least theoretically, NSF might develop after the administration of any gadolinium-based agent in a renally impaired patient, especially if the patient has severe or end-stage renal disease. However, we believe it to be premature to conclude that the relative risk among agents is equivalent (13). We believe that to do so ignores the available data and also foregoes an opportunity to minimize risk among patients with renal disease for whom the use of gadolinium-based contrast agents is unavoidable or possibly life saving.
Is it possible that other explanations (other than a truly greater incidence of NSF) account for this discrepancy between Omniscan-associated NSF incidence relative to its market share? As scientists we would be remiss if we did not entertain that possibility. This must be considered as we design research studies to further investigate this issue. As physicians, however, in view of presently available incidence data, we cannot gamble with the safety and health of our patients.
It is important to note that the present focus on Omniscan results merely because the incidence of reported NSF cases, to date, is so clearly out of proportion to the market share data available for these agents. However, the substantial number of cases reported to have been preceded by administration of Magnevist strongly suggests that NSF very likely can result from exposure to this agent as well, albeit possibly at a lower relative frequency. We believe, therefore, that caution is also clearly warranted at this point regarding administration of Magnevist in patients with impaired renal function. Moreover, despite the low number of cases reported to date, the fact that OptiMARK has such a small market share suggests that caution is strongly urged with this agent as well. Theory suggests that MultiHance, a linear agent with thermodynamic stability similar to that of Magnevist, may be expected to have a similar associated risk for NSF. However, the reported data have not yet shown this to be true. An additional factor to be considered with MultiHance may be that its increased relaxivity (14,15) allows it to be administered at a lower dose than the other FDA-approved gadolinium-based agents while still providing similar contrast enhancement. Primovist (gadoxetic acid disodium; Bayer-Schering Pharma, Berlin, Germany), although not FDA approved, is used in Europe and is another linear agent that may also benefit from lower-dose administrations. Primovist also has 50% clearance by means of liver excretion into bile, thus providing an alternative elimination pathway in patients with renal insufficiency. Macrocyclic agents (of which ProHance is the only such agent available at this time in the United States) may be in a class of their own, with release of free gadolinium and therefore a possibility of NSF development far smaller than that of the linear agents. This may indeed be theoretically logical, but we reiterate that continued analysis is required.
We urge all MR practitioners to become and remain well informed about issues relating to the development of NSF and the association with gadolinium-based contrast agents and to keep all aspects of the available data in mind as decisions are made regarding patients presenting for contrast-enhanced MR procedures. As with all medical procedures, the practitioner must weigh the risk of contrast-enhanced MR imaging in patients with impaired renal function (particularly in patients with severe renal insufficiency, stage 4 or higher, or with a glomerular filtration rate <30 mL/min/1.73 m2) against the risk of performing nonenhanced MR only, of not performing the test at all, or of performing an alternative test such as contrast-enhanced CT, where risks of radiation or iodinated contrast material–induced nephropathy may exceed the risk of NSF after use of a carefully reasoned selection of a gadolinium-based agent.
 |
FOOTNOTES |
|---|
|
References |
|---|
 TOP References |
|---|
This article has been cited by other articles:
|
|
 |
|
  M. R. Prince, H. Zhang, M. Morris, J. L. MacGregor, M. E. Grossman, J. Silberzweig, R. L. DeLapaz, H. J. Lee, C. M. Magro, and A. M. Valeri Incidence of Nephrogenic Systemic Fibrosis at Two Large Medical Centers Radiology, September 1, 2008; 248(3): 807 - 816. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. Wertman, , BA, E. Altun, D. R. Martin, D. G. Mitchell, J. R. Leyendecker, R. B. O'Malley, D. J. Parsons, E. R. Fuller III, and R. C. Semelka Risk of Nephrogenic Systemic Fibrosis: Evaluation of Gadolinium Chelate Contrast Agents at Four American Universities Radiology, September 1, 2008; 248(3): 799 - 806. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
