Gadolinium —
will anybody learn from the debacle?

Rinckside 2015; 26,9: 23-26.

or some months I have been follow­ing with inter­est the uproar about the find­ing of gado­li­nium deposits in brain tissue of some pa­tients af­ter serial MR exam­ina­tions with non­speci­fic ga­do­li­nium agents.

The news broke when some radiologists saw high si­gnal intensity stem­ming from the pi­tui­tary gland on T1-weighted MR images — on non­en­han­ced images. However, the patients had un­der­gone several en­han­ced studies ear­li­er in their life [1-3].

It is not clear whether gadolinium is still bound to the chelate of a contrast agent, whether it is ele­men­tal, or in another, newly formed compound; the latter seems most like­ly. There is strong evidence that the deposit of ga­do­li­nium can be traced back to the li­ne­ar agents gadodiamide (Omniscan) and gado­pen­te­tate di­meg­lu­mine (Magnevist). Both of these compounds have al­ready been involved in the NSF scandal ten years ago — but are still on the mar­ket [4].

As early as 1988, at one of the first big and in­de­pen­dent meet­ings on con­trast agents development for MR imaging we arranged, Michael F. Tweedle pointed out that Mag­ne­vist could become unstable in vivo and release free gadolinium whereas mac­ro­cyclic com­pounds such as Gd-DO3A (ProHance) and Gd-DOTA (Do­ta­rem) remain stable [5]. At the same meeting re­sear­chers from the com­pany producing Gd-DOTA presented similar results and stressed the im­por­tance of macrocyclic compounds “to minimize the in vivo dis­so­ci­a­tion process and avoid[s] potential biological disturbances produced by the pre­sen­ce of free species.” [6]

What happens to free gadolinium in the human body?

During the early preclinical development of Mag­ne­vist, Wein­mann and his co-authors [7] compared the pharmacokinetics of Gd-DTPA and of ga­do­li­nium trichloride in rats. 80% of Gd-DTPA was excreted from the organism in urine within three hours, after seven days 90% of the dose was recovered in urine, another 7% in the feces. Less than 0.3% was found in the body, with 0.08% detected in the liver and 0.1% in the kidneys.

Only two percent of free gadolinium was excreted after seven days.
The rest remained all over the body …

In the case of free gadolinium (i.e., ga­do­li­nium tri­chlo­ride) only 2% was ex­cre­ted after seven days. The rest remained all over the body, mostly in the liver and in the spleen, one sixth elsewhere. The con­clu­sion was that chelates such as DTPA can be extremely effective to remove the highly toxic but dia­gnos­ti­cally very helpful gadolinium from the body.

However, if the chelate doesn't work properly, pa­tients might be at risk.

A major review on metal ion release from pa­ra­mag­ne­tic che­lates (entitled What is tolerable?) published in late 1991 ended with the sentence [8]:

"Although MRI contrast agents are unlikely to be ad­mi­nis­ter­ed re­peat­ed­ly in patients, which could re­sult in accumulation of metal ion, the long-term ef­fects of such potential deposition have yet to be de­mons­trat­ed."

Two years later, the results of animal experiments were published [9]:

“Although intended for single administration in pa­tients, gadodiamide injection has been studied ex­ten­­si­ve­ly in a range of subchronic studies in rats and monkeys. The compound was well tolerated in mon­keys even when administered at doses up to 1.25 mmol/kg daily for 28 consecutive days. In rats, sig­nificant toxicity occurred only at high doses … the pattern of toxicity (involving the stomach, testes, and skin) suggested a disturbance of zinc me­ta­bo­lism.”

However, only rats injected with 50-fold the recom­mended clinical dose three times a week for three weeks developed severe lesions.

The amount of gadolinium one needs to enhance contrast in pathologies on T1-weighted MR images is minimal — particularly compared to the amounts of iodine needed for x-ray contrast agents. Nearly everybody be­liev­ed what they were told: There were hardly any acute side effects; in general, “gado” is safer that iodinated x-ray contrast agents — which is true. As always with drugs, the dose and the galenics make the distinction between poison and remedy.

spaceholder red600   My first encounter with what would become Mag­ne­vist happened around 1982. At that time in­de­pen­dent researchers at universities were more in­ter­est­ed in the performance of such agents than in their sa­fe­ty, in par­ti­cu­lar because the safety was gua­ran­teed by the manufacturers. We trust­ed the cha­rac­te­ris­tics of, e.g., the gadolinium-DTPA complex that were pre­sen­ted to us. The challenge was how to ap­ply the com­pound with­in the given li­mits — which was worked out step by step.

Soon the new compounds were used for all con­ceiv­able kinds of exa­mina­tions. One outstanding example is the follow-up of treatment of multiple scle­ro­sis patients — in some instances MS patients underwent contrast en­hanc­ed studies once a month for two or three years. The general attitude for all me­di­cal or not-so-medical indications was: “Gado is a dye and can be used re­pe­ti­tively as often as possible.”

I still remember people in R&D, even companies' lead­ing scientists, com­plai­ning that their hints and pro­po­sals were pushed aside by the marketing de­part­ment and the ma­nage­ment: "Off-label use is the re­spon­si­bi­li­ty of the doctors."

The recommended dose, best enhancement, and MR angiography

On some of the first MR images in animals and in humans parts of the body were highlighted after the injection of Gd-DTPA, but others turned black. The pi­tui­tary gland became bright like a streetlight, the blad­der dark. The ques­tion was: why; the answer was the correct dose to be injected.

Here the manufacturer of Magnevist was cautious and finally agreed upon 0.1 mmol/kg body weight. All other manufacturers followed this re­com­men­da­tion. This dose provides excellent enhancement at low and medium mag­ne­tic fields.

The reason is easily visible on the animated si­mu­la­tion shown in this figure (link).

A dose increase beyond the recommended dose may lead to loss of con­trast. This is because a T2 shor­ten­ing remains and can take over primary influence upon image contrast.

spaceholder red600   Parallel to the nonspecific gadolinium agents, an­gio­gra­phic blood pool agents were being developed because non-enhanced imaging technologies did not fulfill the requirements for high-resolution an­gio­gra­phy. The ma­nu­fac­tu­rers anticipated a substantial market because contrast-enhanced MR an­gio­graphy was to cut a big slice out of the conventional and CT an­gio­gra­phy cake. Yet, the development of the blood pool agents was slow and pla­gued by set­backs.

Then, suddenly a group of doctors proposed the use of the existing non­spe­cific agents together with spe­cial pa­tent­ed techniques and hardware to per­form MR angiography. They all underlined that high-dose ga­do­li­ni­um che­la­tes (up to 0.3 mmol/ kg) were sig­ni­fi­cant­ly less ne­phro­to­xic than iodinated con­trast agents [10]. In a textbook of contrast enhanced MR an­gio­gra­phy, Mar­tin R. Prince, Thomas M. Grist and Jörg F. De­ba­tin stated in 2003:

“From an image quality point of view, generally the more contrast the bet­ter… Gadolinium compounds have no clinically detectable nephrotoxicity. They can be used safely at the maximum dose in patients with renal failure.” [11]

Already in a US patent applied for in 1993, one finds the following description:

“The dose of the gadolinium chelate may be within the range of 0.05 millimoles/kilogram body weight to 0.7 millimoles/kilogram body weight depending upon the time required to obtain the image. It should be noted that the dose of the contrast should not be too high such that there may be un­de­sir­able toxicity or T2 effects.” [12]

Since the companies involved neglected the pa­tents, multimillion dollar lawsuits by the patent hol­ders, then li­cense, patent and con­sul­tan­cy agree­ments with nu­merous pharmaceutical and hardware com­pa­nies followed [13].

Disaster strikes

Then, early in 2006, there was an outbreak of a new di­sease in Austria: Ne­phro­ge­nic fibrosing dermopathy, later called Nephrogenic Systemic Fibrosis (NSF). Single cases had been described earlier, but not at­tri­but­ed to ga­do­li­nium:

“NFD was unknown before March 1997 and some au­thors suggest that the sudden occurrence of the disease in the last 8 years makes it likely that a new agent or technique of examination causes NFD/NSF” [14].

All cases were re­lat­ed to Omniscan.

Among others, Martin R. Prince did an about-face and reacted with a paper describing fifteen pa­tients who developed NSF after high-dose ga­do­li­ni­um-en­han­ced MR imaging compared to no patient with the stan­dard dose. The conclusion of the paper included the fol­low­ing sentence [15]:

“We recommend using no more than a standard dose of GBCA (0.1 mmol/ kg).”

spaceholder red600   Shortly after the publication of this article, a let­ter reached the editor of Radiology and was pub­li­shed. The author stated [16]:

“The finding that all patients who developed NSF had received a high dose of a gadolinum-based contrast agent (GBCA), but none of the 74,124 who had received a standard dose (0.1 mmol per kilogram of body weight) de­ve­lop­ed NSF, ir­re­spec­ti­ve of renal function, is of particular interest …

“Perhaps it is linked to the development of tech­ni­ques that re­quire the use of higher doses of GBCA, such as contrast agent-enhanced magnetic re­so­nan­ce (MR) angiography. For better visualization, espe­ci­al­ly of distal and small vessels, double or triple doses of GBCA are often advocated …

“This temporal coincidence may only be incidental, but it is nonetheless suggestive and may help to ex­plain the some­what mys­te­ri­ous timing of the ap­pea­ran­ce of NSF.”

This seemed not to be a serious problem for the ma­nu­fac­tu­rers of Mag­ne­vist or Omniscan. These con­trast agents are still available on the market to­day.

More so, the reaction of the US-American ma­nu­fac­tu­rer of Omniscan against the publication of yet an­other NSF outbreak in a Danish trial con­sist­ed in an at­tempt to silence the radiologist in charge. He had pre­sent­ed the late side effects of the trial, i.e., death or mutilation of twenty patients. This was brought to the light by articles in Pro Publica and the Sun­day Times [17].

As the English newspaper The Guardian reported, GE Healthcare dropped the libel action in 2010:

"Lawyers for leading Danish radiologist Henrik Thomsen said today: 'He will be obviously relieved. Now he won't have to worry about his future fi­nan­cial position, and won't have to keep looking over his shoulder before he says anything.' In agreed statements released today, Thomsen said: 'I stand by my publicly expressed opinion, based on my experience and research on pub­lished papers, that there is an association between the chemical formulation of gadolinium-based contrast agents and NSF.' He added: 'It was not my in­ten­tion to suggest on the basis of the evidence then available to me that GE Health­care had marketed Omniscan knowing that it might cause NSF.'" [18]

spaceholder red600   Still, more than ten years after the identification of the culprits, those re­spon­sib­le evade or dismiss the moral challenge they are confronted with, however plunge headfirst into new ones: gadolinium deposits in the brain.

Gadolinium deposits don't belong in the brain.

Admittedly nowadays one finds the following warning in the package insert:

“Do not exceed the recommended dose of … and allow a sufficient period of time for elimination of the drug from the body prior to any re­ad­mi­ni­stra­tion.” Yet, the positions of the US-American Food and Drug Administration and their European coun­ter­parts leave conflicting impressions, to say it politely; why don't they stop the sale of Omniscan and Magnevist to protect possible future victims?

The vultures are already circling in the air and the ambulance chasers flood the Internet with their websites: the lawyers are coming in great number. The motto is: “Are you gadolinium toxic? If yes, contact us.” They are rather cle­ver and business minded. And they will soon find the arguments with which they will milk the manufacturers, the radiologists — and, most of all, their cli­ents. They will argue like this:

"Lack of background knowledge of the radiologists and their in­cre­dib­le trust in pharmaceutical companies; the almighty dollar sign; and use of an inappropriate and unsafe drug, too high a dose, too many serial studies, too close a study after another, examinations too often with­out proper indication; and, finally, off-label use."

spaceholder red600   Debacle is as good a word as any to describe what has happened here; and many of the parties involved try to sweep the problem under the rug: ma­nu­fac­tu­rers, radiologists, and supposedly supervising administrations.

What is the clinical significance of gadolinium deposits in the brain and elsewhere? Nobody really knows. However, it doesn't belong there although at present there is no proof that it is harmful. NSF was a new iatrogenic di­sease. The (unlimited?) storage of gadolinium in the human body could be but a continuation of this disease. Already the idea attracts all hypochondriacs in town.

The recent events coincide with descriptions of the sudden appearance of gadolinium as anthropogenic contamination in tap water [19]. The cause is the use of MR contrast agents; gadolinium cannot be removed by water treat­ment plants.

What I am afraid of are possible long-term con­se­quen­ces for all of us.

spaceholder dark yellow   Important Note: Contrast enhanced MR exami­nations have life-saving benefits. I fully support the intravenous application of gadolinium-based con­trast agents for diagnostic purposes. This is not an article against contrast en­han­ced MR imaging. However, please ap­ply macrocyclic contrast agents and/or agents excreted by both the liver and the kidneys. Even they should only be used if a clear diagnostic advan­tage for the patient can be expected.


1. Kanda T, Ishii K, Kawaguchi H, Kitajima K, Takenaka D. High signal intensity in the dentate nucleus and globus pallidus on unenhanced T1-weighted MR images: relationship with in­crea­sing cumulative dose of a gadolinium-based contrast material. Radiology 2014; 270: 834–841.
2. Kanda T, Osawa M, Oba H, et al. High signal intensity in dentate nucleus on unenhanced T1-weighted MR images: association with linear versus macrocyclic gadolinium chelate ad­mi­ni­stra­tion. Radiology 2015, 275: 803–809.
3. Radbruch A, Weberling LD, Pascal J, et al. Gadolinium retention in the dentate nucleus and globus pallidus is dependent on the class of contrast agent. Radiology 2015, 275: 783-791.
4. Rinck PA. Radiologists meet with heavy collateral damage. Rinckside 2008; 19,3: 7-10.
5. Tweedle MF. Work in progress toward nonionic macrocyclic gadolinium (III) complexes. in: Rinck PA (ed). Contrast and contrast agents in magnetic resonance imaging. Proceedings of Contrast and Contrast Agents in Magnetic Resonance Imaging – A Special Topic Seminar; Trondheim, Norway; 12-13 September 1988. Trondheim and Mons: The European Workshop on Magnetic Resonance in Medicine (EMRF). 1989. 65-73.
6. Meyer D, Schaefer M, Doucet D. Physico-chemical properties of the macrocyclic chelate Gadolinium-DOTA. in: Rinck PA (ed). Contrast and contrast agents in magnetic resonance imaging. Proceedings of Contrast and Contrast Agents in Magnetic Resonance Imaging – A Special Topic Seminar; Trondheim, Norway; 12-13 September 1988. Trondheim and Mons: The European Workshop on Magnetic Resonance in Medicine (EMRF). 1989. 33-43.
7. Weinmann HJ, Brasch RC, Press W-R, Wesbey GE. Characteristics of Gadolinium-DTPA complex: a potential NMR contrast agent. AJR 1984; 142: 619-624.
8. Rocklage SM, Worah D, Kim S-H. Metal ion release from paramagnetic chelates: What is tolerable? Magn Res Med 1991; 22: 216-221.
9. Harpur ES, Worah D, Hals PA, Holtz E, Furuhama K, Nomura H. Preclinical safety assessment and pharmacokinetics of gadodiamide injection, a new magnetic resonance imaging contrast agent. Invest Radiol. 1993; 28 Suppl 1: S28-43.
10. Prince MR, Arnoldus C, Frisoli JK. Nephrotoxicity of high-dose gadolinium compared with iodinated contrast. J Magn Reson Imaging 1996; 6: 162-166.
11. Prince MR, Grist TM, Debatin JF. 3D contrast MR angiography. Berlin, New York: Springer Publishers. 3rd ed., 2003. 22-23.
12. Magnetic resonance arteriography with dynamic intravenous contrast agents. Inventor: Martin R. Prince, 202 Delafield St., Ann Arbor, Mich. 48105 [U.S.A.]. United States [of North Amer­ica] Patent; Patent Number 5,417,213. Patent filed: 7 June 1993; patent granted: 23 May 1995.
13. Ersoy H, Zhang HL, Prince MR. Peripheral MR Angiography. Journal of Cardiovascular Magnetic Resonance. 2006; 8: 517–528.
14. Grobner T. Gadolinium – a specific trigger for the development of nephrogenic fibrosing dermopathy and nephrogenic systemic fibrosis? Nephrol Dial Transplant (2006) 21: 1104–1108. Important erratum: Nephrol Dial Transplant (2006) 21: 1745.
15. Prince MR, Zhang H, Morris M et al. Incidence of nephrogenic systemic fibrosis at two large medical centers. Radiology 2008; 248: 807–816.
16. Gossner J. Letter to the Editor (concerning Prince MR, et al. Radiology 2008; 248: 807-816) and response by Prince MR, et al. Radiology 2009; 251: 612-613.
17. Gerth J (ProPublica) and Ungoed-Thomas J (The Sunday Times). GE suit hushes scientist critical of Omniscan. www.propublica.org/article/ges-omniscan-lawsuit-ratchets-up-volume-in-british-libel-debate-1219; and: The Sunday Times, 19 December 2009.
18. Leigh D. US drug firm drops libel action against scientist. The Guardian. 18 February 2010. www.theguardian.com/science/2010/feb/18/ge-healthcare-henrik-thomsen-libel.
19. Tepe N, Romero M, Bau M. High-technology metals as emerging contaminants: Strong in­crease of anthropogenic gadolinium levels in tap water of Berlin, Germany, from 2009 to 2012. Applied Geochemistry 2014; 45: 191-197.

Citation: Rinck PA. Gadolinium – will anybody learn from the debacle? Rinckside 2015; 26,9: 23-26.

A digest version of this column was published as:
Rinck PA. Gadolinium: Will anybody learn from the debacle?
Aunt Minnie Europe. Maverinck. 9 December 2015.

TurnPrevPage TurnNextPage

Rinckside • ISSN 2364-3889
is pub­lish­ed both in an elec­tro­nic and in a prin­ted ver­sion. It is listed by the Ger­man Na­tio­nal Lib­rary.


→ Print version (pdf).


The Author

Rinck is my last name, and a rink is an area in which a com­bat or con­test takes place, rink­side means “by the rink”; in a double mean­ing “Rinck­side” means the page by Rinck.

Sometimes I could also imagine “Rinck­sighs”, “Rinck­sights”, or “Rinck­sites” ... More


Bulletin Board