SATURDAY, 21 NOVEMBER 2026
Quiet MRI room in a clinic with the edge of the scanner bore, and in front of it on a stainless steel tray a small glass vial of clear liquid with a blank white label next to a capped syringe, with a folded light blue blanket on the examination table. AI-generated image.

Gadolinium in MRI Contrast: Retention, Side Effects, and Risk After Your Scan

Gadolinium from the MRI contrast agent demonstrably stays in the body in small amounts, including in the brain. Harm from the deposits in the brain has not been proven in people with normal kidney function, but long-term studies are lacking [1][2][3]. Gadolinium is established as dangerous in severe kidney failure: nephrogenic systemic fibrosis (NSF) has become rare since precautions were introduced; among 529 cases in the literature with documented administration, the administration occurred after 2008 in only 7 [4]. In people without NSF, the EMA also links rare skin plaques to linear agents [5].

Linear agents release more gadolinium and deposit more in the brain than macrocyclic agents; for this reason, three linear agents for intravenous use have been suspended in the EU following the decision of November 23, 2017, and in Germany since February 28, 2018 [1][6]. Whether lasting symptoms after an MRI come from gadolinium is open: there is no recognized diagnosis, and there is no proof that chelation (detox) works [5][7]. So you should not postpone a medically justified MRI with contrast. Ask beforehand whether the contrast agent is necessary and which agent will be used, and tell the staff your kidney values.

This article is part 10 of our series “How Do Humans Poison Themselves?” and does not replace medical advice. Published so far: Mercury Poisoning, Lead Poisoning, Cadmium in Food, Arsenic in Rice, Aluminum in Deodorant and Foil, Glyphosate and Pesticides, PFAS and Forever Chemicals, Bisphenol A and Plasticizers and Fluoride in Toothpaste and Drinking Water. Last updated: November 2026.

Key Points at a Glance

  • Volume in Germany: Raw official figures in a manufacturer’s G-BA dossier (G-BA: the German Federal Joint Committee) yield a calculated roughly 3.0 million contrast-enhanced MRI scans per year (inpatient, all patients 2022: 806,709; outpatient, statutory health insurance patients only 2021: 2,210,255). These are examinations, not “doses”; by calculation, about three quarters of all MRI scans are done without contrast agent [8].
  • EU decision of November 23, 2017: Intravenous use is suspended for gadodiamide, gadopentetic acid and gadoversetamide; the BfArM (German Federal Institute for Drugs and Medical Devices) most recently extended the suspension to February 28, 2028. Macrocyclic agents remain authorized, at the lowest effective dose [1][9].
  • Stability in human serum after 15 days (lab experiment, 37 °C): Optimark 21 percent and Omniscan 20 percent gadolinium release; Gadovist, ProHance and Dotarem below the detection limit of 0.1 percent [10].
  • NSF: 639 biopsy-confirmed cases in the literature; in only 7 of the 529 cases with documented administration did the administration occur after 2008 [4]; 0 cases among 4,931 patients with severe kidney impairment after group II agents as classified by the American College of Radiology (gadobenic acid and three macrocyclic agents), upper limit of the risk 0.07 percent [11].
  • Retention yes, harm in the brain not proven: Gadolinium is found in brain, bone and skin [12][13][14]; the EMA sees no evidence that the deposits in the brain have harmed patients [1]; the ESUR writes in 2025: “No relevant neurological symptoms have yet been confirmed” [3]. The EMA also links rare skin plaques to linear agents [5].

What Is Gadolinium in MRI Contrast Agents?

A gadolinium-based contrast agent is a compound of the metal gadolinium and a molecule that encloses it; it is usually injected into a vein so that certain tissues appear brighter on an MRI. When people talk about a gadolinium MRI, this is exactly the examination they mean. Gadolinium is one of the rare earth metals. After injection, it acts on the water molecules in the tissue, which then give off a stronger signal [15].

The surrounding molecule, the chelator, has a purpose. Free gadolinium is considered toxic: in cell experiments, it can bind to calcium binding sites and inhibit calcium channels [16]. In the contrast agent, it is therefore meant to stay firmly enclosed. There are two designs: linear chelators (open-chain) and macrocyclic chelators (cage-like) [5].

The kidneys excrete the contrast agent: according to the EMA assessment, the route of elimination is 100 percent through the kidneys for most agents and 50 percent for gadoxetic acid [5]. A small remainder nevertheless stays in the body longer; more on that below.

How often are MRI scans with contrast agent performed in Germany? There is no official figure for “doses per year.” The widely circulated figure of 2.4 to 3 million doses cannot be traced back to any primary source.

An order of magnitude can be calculated from raw figures that a manufacturer cites in a dossier submitted to the G-BA and that trace back to Destatis and WIdO: in 2022 there were 1,842,799 inpatient MRI scans, of which 806,709 with contrast agent (about 44 percent); in the outpatient setting in 2021, 2,210,255 of 11,152,078 MRI scans in people with statutory health insurance were done with contrast agent (just under 20 percent). Together, that is about 3.0 million contrast-enhanced examinations. The inpatient figure covers all patients, the outpatient figure only people with statutory health insurance; privately insured outpatients are missing [8]. The dossier itself calculates 2.9 to 3.4 million for people with statutory health insurance and concedes possible overestimation [8]. It counts examinations, not administered doses. The dossier comes from the manufacturer of a gadolinium agent and is therefore suitable only for an order of magnitude.

MRI Contrast Agent: Linear or Macrocyclic?

Since 2018, macrocyclic agents have remained authorized in the EU for intravenous use, at the lowest effective dose and only if an image without contrast agent is not sufficient; of the linear agents, two remain only for liver MRI and three are suspended. How it came to this: in 2014, Kanda and co-authors first described a brighter signal in deep brain nuclei after earlier contrast agent administrations [17]. The EMA opened a review on March 17, 2016, and the EU Commission decision followed on November 23, 2017 [18]. The Rote-Hand-Brief (German “Dear Healthcare Professional” letter) followed on January 8, 2018 [6]. The BfArM ordered the suspension as of February 28, 2018 and extended it several times, most recently to February 28, 2028 [9].

The EMA’s reasoning: gadolinium deposits in the brain, and according to stability, laboratory and animal data, linear agents release more of it from the chelator than macrocyclic agents. At the same time, the EMA writes that there is currently no evidence that the deposits in the brain have harmed patients [1]. So it was a matter of precaution. Suspension also does not mean revocation: the restrictions can be lifted if manufacturers demonstrate a benefit for a specific patient group or modify the product so that hardly any gadolinium is released [1].

The US agency FDA took a different path: on December 19, 2017, it required a warning and a Medication Guide for all gadolinium-based contrast agents but left the agents on the market. Its key statements: retention is not directly linked to adverse health effects in people with normal kidney function, the benefit of all approved agents outweighs possible risks, and only NSF is known as a consequence. The FDA also sees linear agents as having more retention than macrocyclic ones [2].

Gadolinium contrast agents and their status in the EU (as of the 2017 EMA decision, gadopiclenol authorization 2023)
Active substanceBrand name (examples)TypeStatus in the EUSource
GadobutrolGadovistmacrocyclicremains (intravenous use)[1]
Gadoteric acidDotarem, Artirem, genericsmacrocyclicremains (vein and joint)[1][6]
GadoteridolProHancemacrocyclicremains (intravenous use)[1]
GadopiclenolEluciremmacrocyclicauthorized since December 7, 2023[15]
Gadobenic acidMultiHancelinearliver MRI only[1]
Gadoxetic acidPrimovistlinearremains (liver MRI)[1]
Gadopentetic acidMagnevistlinearvein: suspended; joint: remains[1]
GadodiamideOmniscanlinearsuspended[1]
GadoversetamideOptimarklinearsuspended, additionally withdrawn from the market by the manufacturer[1][3]

Gadopiclenol is the newest agent; according to the EMA, it can be given with half the gadolinium dose and delivers the same contrast as other non-specific gadolinium-based agents [15]. Manufacturer-independent long-term data in humans are not yet available for it.

And are there contrast agents without gadolinium? A manganese agent (Teslascan) was authorized in the EU until 2012 and was revoked at the request of the holder for commercial reasons, not because of a safety signal [19]. According to a review article, new manganese agents are in the preclinical or early clinical phase [20]; one example is Mn-PyC3A, which was tested in animals in the cited paper [21]. We are not aware of an authorized gadolinium-free manganese-based agent in the EU.

How Stable Are Gadolinium Compounds?

In the test tube, linear agents release markedly more gadolinium than macrocyclic ones. In a laboratory experiment with human serum (15 days, 37 °C, pH 7.4), the nonionic linear agents Optimark and Omniscan released 21 percent and 20 percent of their gadolinium, and the ionic linear agents Magnevist, MultiHance and Primovist 1.1 to 1.9 percent. Gadovist, ProHance and Dotarem were below the quantification limit of 0.1 percent [10]. With added phosphate (10 millimoles per liter, simulating conditions in kidney failure), the initial release of Omniscan and Optimark rose to roughly one hundred times; for the three macrocyclic agents, no measurable release occurred [10]. The first author worked at a manufacturer (Bayer Schering Pharma), and a lab experiment does not show what happens in the body over years.

Horizontal bar chart, gadolinium release in human serum after 15 days at 37 °C: Optimark 21 percent, Omniscan 20 percent, Magnevist 1.9, MultiHance 1.9, Primovist 1.1 percent, Gadovist, ProHance and Dotarem each below 0.1 percent.
Gadolinium release in human serum after 15 days: linear agents up to 21 percent, macrocyclic agents below 0.1 percent. Laboratory experiment, not directly transferable to the body. Source: Frenzel et al., Investigative Radiology 2008.
Use the graphic on your own site

You may use the graphic freely with attribution and a link. Copy the code:

<a href="https://gesundefakten.com/en/health/gadolinium-mri-contrast/"><img src="https://gesundefakten.com/wp-content/uploads/2026/11/gadolinium-grafik-en.webp" alt="Infographic: gadolinium release in human serum after 15 days, linear contrast agents up to 21 percent, macrocyclic agents below 0.1 percent" width="1600" height="1000"></a><br>Source: <a href="https://gesundefakten.com/en/health/gadolinium-mri-contrast/">gesundefakten.com</a>

The claim that “all” gadolinium compounds break apart in the body is therefore not accurate. For the three classic macrocyclic agents, serum data show release below the detection limit, and in the rat brain after macrocyclic agents the gadolinium was found practically entirely as intact agent (for gadobutrol and gadoteric acid “most likely,” as the authors write): 100 percent after gadoteridol, 97 to 102 percent of the amount of gadolinium found in the soluble fraction after gadoteric acid; after the linear gadodiamide, it was only 18.2 percent in the cerebellum [22][23][24]. In two of these three animal studies, authors worked at manufacturers (Bayer, Guerbet). In humans the question is open: “It is not known whether the deposited gadolinium remains chelated,” writes the ESUR in 2025 [3]. A study of two patients suggests that gadolinium in the cerebellum might detach from the chelator; this is a hypothesis with n = 2 [25].

Gadolinium Retention in the Brain, Bone and Skin

Gadolinium demonstrably deposits in brain, bone and skin after contrast agent administration, more after linear agents than after macrocyclic ones, but in very small amounts. The EMA also names liver, kidney and muscle [5]. The deposits occur independently of kidney function, that is, also with healthy kidneys [3].

Brain: first in the image, then at autopsy

In 2014, Kanda found in 381 patients a brighter signal in the dentate nucleus and globus pallidus the more contrast agent examinations had preceded [17]. A follow-up study of 127 patients found the abnormality only in connection with linear, not macrocyclic agents [26]; Radbruch confirmed this in 50 patients each for gadopentetic acid versus gadoteric acid [27]. The signal in the image is, however, nonspecific and can also arise from manganese, iron or calcium; tissue analyses are more sensitive [3].

Autopsies provided the direct evidence: in 13 deceased people with at least four contrast MRI scans, gadolinium was found in brain tissue (0.1 to 58.8 micrograms per gram), the more administrations, the more; in 10 controls without contrast agent it was not detectable [12].

Bone and skin

More stays behind in bone than in the brain: in 9 deceased people who had received only a single agent from a group less burdened with NSF, the median value in bone was 23 times higher than in the brain [13]. The ESUR names very small amounts in bone, liver and skin: larger than in the brain, but without symptoms in bone and liver [3]. In the skin, gadolinium decreases over time in people with healthy kidneys [28][14]. The EMA, however, is aware of rare skin plaques (gadolinium-associated plaques) in people without NSF and links them to linear agents [5]; a review of 40 studies found no conclusive evidence of a causal link with skin changes in people with healthy kidneys [28].

After macrocyclic agents: less, but not zero

Macrocyclic agents also leave traces. In an autopsy study with 28 gadoteridol and 9 gadobenic acid cases, brain retention was 3.0 to 6.5 times higher after the linear gadobenic acid; in the white matter and in the skin, it declined with time after the last administration for both agents [14]. In 52 children with a median of seven gadobutrol administrations, the MRI evaluation pointed to deposition in the globus pallidus and putamen [29].

Overview of studies on gadolinium deposition in humans
StudyWho / how manyFindingSource
Kanda 2014 (MRI)381 patientsSignal in deep brain nuclei increases with the number of earlier administrations[17]
Kanda 2015 (MRI)127 patientsAbnormality in 7.1 percent, association only with linear agents[26]
McDonald 2015 (autopsy)13 deceased, 10 controls0.1 to 58.8 µg gadolinium per g of brain tissue, controls below detection limit[12]
Murata 2016 (autopsy)9 deceased, one agent eachGadolinium in all brain regions, bone median 23 times higher than brain[13]
Kobayashi 2021 (autopsy)28 gadoteridol, 9 gadobenic acidGadobenic acid: brain 3.0 to 6.5 times, bone 4.4 times, skin 2.9 times more than gadoteridol[14]
Gräfe 2023 (MRI, children)52 children, gadobutrolIndication of deposition in globus pallidus and putamen[29]

What is not known

Three things are open: in what chemical form gadolinium is present in humans after macrocyclic agents, at what amount something might happen, and whether small effects go unnoticed. According to the ESUR, all studies on brain retention in humans are retrospective; their clinical significance is “not yet known” [3].

Does Gadolinium Retention Make You Sick?

Harm from the deposits in the brain has not been proven in people with normal kidney function, but in the absence of long-term studies it has not been ruled out either. In 2017 the international MRI society ISMRM summarized that there are no data in humans or animals showing clinical disadvantages from the brain deposits [30]. The EMA’s precaution was based on a different consideration: possible effects on fine motor skills or thinking would be “delayed and subtle,” the committee could not name a safe threshold or a group without retention, and it considered clinical studies within a reasonable time unfeasible [5]. It also named as plausible but not shown that the deposits could worsen existing inflammatory diseases [5].

Registry data on the risk of Parkinson’s disease contradict each other. In Ontario, among 246,557 patients (99,739 with, 146,818 without contrast agent), there was no increase (adjusted hazard ratio 1.04, where a value of 1.00 means no increased risk; 95 percent confidence interval 0.98 to 1.09) [31][5]. The figures come from the EMA’s account; it considered the informative value limited because only 2.5 percent had more than three contrast MRI scans and the ten-year observation period might be too short for late effects [5]. In Korea, a study of 175,125 adults (3,835 with contrast agent) found, after matching the groups, more Parkinson’s cases with contrast agent (0.9 percent after linear and 0.5 percent after macrocyclic agents versus 0.0 and 0.04 percent, respectively, in the corresponding comparison groups without contrast agent), but no difference between the two types of agent [32]. A second study of 222,977 people found for parkinsonism overall a hazard ratio of 1.13 after linear agents (97.5 percent confidence interval 1.08 to 1.19) and 1.00 after macrocyclic agents [33]. Such data are retrospective and cannot clarify whether the contrast agent is the cause or the condition for which the MRI was done.

In multiple sclerosis, people with and without signal abnormality did not differ in motor function and thinking [34]; an 18-year study found an association with lower word fluency that, according to the authors, does not prove a cause [35]. In four glioblastoma patients with 53 to 71 administrations, nothing attributable to the contrast agent was found after 13 to 15 years; the authors attributed mental impairments in two of those affected to age, illness and prior education [36]. As groups requiring particular attention, the FDA names multiple administrations over a lifetime, pregnant women, children and inflammatory diseases [2].

Nephrogenic Systemic Fibrosis (NSF): The Established Danger

Gadolinium is established as harmful in severe kidney failure, as nephrogenic systemic fibrosis, and this disease has become very rare today. In 2006, Marckmann and co-authors described 13 confirmed cases, all after gadodiamide; 7 patients were severely disabled, one died [37]. According to the ESUR, NSF usually begins days to three months after administration, rarely years afterward, with pain, itching, swelling and redness of the skin, mostly on the legs; later come thickening of the skin and joint stiffening, and when internal organs are affected it can be fatal [3].

Risk factors are reduced kidney function (especially an eGFR below 15 ml/min per 1.73 m², a measure of the kidneys’ filtering performance), dialysis and the dose. As estimated frequency in severe kidney failure, the ESUR gives 3 to 18 percent after gadodiamide and 0.1 to 1 percent after gadopentetic acid [3].

The measures appear to have worked: of 639 biopsy-confirmed cases in the literature, 529 had a documented contrast agent administration, and in only 7 of these did the administration occur after 2008; 70.8 percent (296 of 418 cases with information on this) had movement restrictions [4]. For the group II agents as classified by the American College of Radiology (gadobenic acid, gadobutrol, gadoteric acid, gadoteridol), a meta-analysis of 4,931 patients with severe kidney impairment found not a single NSF case; 3,167 of them had received gadobenic acid, a linear agent. As the upper limit of the risk (95 percent confidence interval), the paper gives 0.07 percent [11]. The ESUR lists gadopiclenol, gadobutrol, gadoteric acid and gadoteridol in the class with the lowest risk and does not consider a prior kidney function measurement mandatory for these; with a known or suspected eGFR below 30, the usual benefit-risk assessment applies [3].

Symptoms After an MRI: “Gadolinium Deposition Disease”?

Take symptoms after a contrast MRI seriously. But there is no recognized diagnosis of “gadolinium deposition disease,” and it is not proven that gadolinium is the cause.

What affected people report

The term goes back to online surveys. In 2016, Semelka and co-authors described 42 participants from two online self-help groups: headache (28), bone pain (26), pain in arms and legs (26) and skin thickening (22); 29 reported clouded thinking or headache lasting more than three months, 21 after a single administration [38]. In a survey with 50 participants, all attributed their symptoms to gadolinium; in 66 percent they began immediately after administration [39]. Both papers have no control group, and the participants selected themselves. The symptoms are thus not invented, but the data cannot show the connection.

What authorities and guidelines say

The EMA wrote in 2017: apart from the skin plaques, clinical signs or symptoms due to gadolinium deposition in people without NSF are not confirmed; the reports of individual patients left the cause open [5]. The FDA: reports of events in several organ systems with normal kidney function, a causal relationship could not be established [2]. The ESUR 2025: no relevant neurological symptoms have yet been confirmed [3]. A review of 40 studies on the skin found no conclusive evidence of a causal link with normal kidney function [28]. Controlled studies on these symptoms are not available.

If you develop symptoms after an MRI, note the date, agent and dose (the ESUR requires that the agent and dose be recorded in the patient file [3]) and have the symptoms evaluated by a doctor. Whether the contrast agent is a possible factor is then one question among several.

Urine test and blood test: what they do not tell you

An elevated gadolinium level in urine does not prove a disease, because it is expected after every administration. In 13 people with healthy kidneys after the first administration, the mean values in 24-hour urine were 1,944 micrograms (day 3), 301 (day 10) and 34 (day 30); all three mean values were above the reference value of 0.7 micrograms that applies to people never exposed [40]. After gadobutrol, according to the authors’ model, in 24 fully examined patients 95 percent of serum values were below the reference value after about 56 days and of urine values after about 132 days; none reported gadolinium symptoms, and the authors consider the reference values not applicable for five months after administration [41]. Also in a group of 45 people with symptoms (authors from the Semelka group), the value was above the reference for up to three months, similar for all agents [42]. According to these data, people without symptoms and people with symptoms cannot be told apart by the urine value.

Chelation with DTPA: uncontrolled and disputed

There is no controlled study on chelation. The best-known work is a series of 25 patients with “presumed” disease who received three treatment rounds with the chelator Ca- and Zn-DTPA as an infusion: gadolinium in the urine rose 30.3-fold (monthly) and 12.9-fold (weekly), respectively; symptoms improved in 13, stayed the same in 10 and were worse in 2, and 11 experienced a temporary worsening of symptoms (“flare-up”). There was no control or placebo group, and the improvement was rated by the patients themselves [7]. The rise in urine shows that gadolinium is mobilized, not that this helps. A follow-up paper by the same authors from 2024 titled “Near-cure” was retracted on November 25, 2024, according to the publisher because of undisclosed conflicts of interest that compromised the peer review process; the authors dispute this [43]. In a rat experiment (3 animals each, gadodiamide, zinc DTPA given simultaneously up to 8 hours after administration), gadolinium did not fall significantly in any organ [44]; this says nothing about late chelation. For sauna, zeolite or dietary supplements as “gadolinium detox,” there is no proof of efficacy whatsoever.

You know the pattern from part 1 and part 2: in 1999, the Human Biomonitoring Commission of the German Environment Agency (Umweltbundesamt) stated on DMPS and DMSA that the DMPS mobilization test for mercury is not valid, there are no reference values for stimulated excretion, the safety of long-term therapy is largely unknown, and chelators could redistribute metals in the body, including into the brain [45]. This concerns other chelators and other metals; we found no explicit German professional society position on gadolinium. As a warning against uncontrolled self-experimentation, it fits nonetheless.

Gadolinium in Water

Gadolinium from contrast agents ends up in waterways via urine and wastewater because sewage treatment plants barely retain it [46]; the concentrations are in the range of nanograms to micrograms per liter and thus, according to a review, at least one order of magnitude below the level at which free gadolinium would be toxic to humans [47]. In six large German cities, human-made gadolinium was detectable in tap water; the authors see it as an indicator of other wastewater substances, not as a toxicological assessment of gadolinium itself [46]. According to a model, Germany, France and Italy together account for 40 percent of the European gadolinium flux into the sea [48]. According to the review, it is open whether intact contrast agents, breakdown products or free gadolinium play a role for health [47]. The ESUR writes: “The risks of this gadolinium in the environment are not yet known” [3]. An official risk assessment is not yet available.

Fact Check: What Is True About Common Claims

None of the nine claims is true without qualification; the table shows what is right and what is not in each.

Gadolinium claims: finding and source
ClaimFindingSource
“In Germany, 2.4 to 3 million doses are administered every year.”Partly. The primary source of the figure is missing; calculated, it is about 3.0 million contrast-enhanced MRI examinations (inpatient all patients, outpatient only people with statutory health insurance), not doses.[8]
“Almost all MRI contrast agents are based on gadolinium.”Largely. The common MRI contrast agents contain gadolinium; the manganese agent Teslascan was revoked in 2012, new manganese agents are in early development. Most MRI scans are done without any contrast agent.[5][19][20][8]
“Stable compounds that are excreted intact within a short time.”Partly. Most of it is excreted via the kidneys; a small remainder stays in tissue for weeks to years.[5][2][41][13]
“All gadolinium compounds break apart in the body, this is clearly proven.”False as a generalization. Proven for linear agents (up to 21 percent in 15 days in serum), for macrocyclic agents below 0.1 percent in serum; in humans the form of the deposited gadolinium is unknown.[10][3]
“Because of its similarity to calcium, gadolinium deposits in brain, nerves, muscles, skin, organs, vessels and bone.”Partly. Deposition in brain, bone, skin, liver, kidney and muscle is proven, in very small amounts; the similarity to calcium is described for free gadolinium in cell experiments, not shown as the cause of the deposition.[5][16][25]
“Gadolinium causes poisoning (gadolinium toxicity) with a long list of symptoms.”Not proven. The symptom lists come from online surveys without a control group; EMA, FDA and ESUR do not confirm a cause. NSF in severe kidney failure is confirmed; the EMA also links rare skin plaques to linear agents; whether they are caused by gadolinium in normal kidney function is open.[38][39][5][2][3]
“With a urine or provocation test and chelation, gadolinium can be detected and removed.”Not proven. Urine values remain elevated for months after every administration; DTPA chelation only in one uncontrolled series, a follow-up paper was retracted.[40][41][7][43]
“Macrocyclic agents are completely safe.”Overstated. They deposit less and have a very low NSF risk, but gadolinium is also found in tissue after them.[13][11][3]
“Linear agents are banned in the EU.”Imprecise. Three linear agents are suspended for intravenous use, two remain authorized for liver MRI; gadopentetic acid remains for joints.[1][9]

Before the MRI: Questions for the Radiology Team

You should not postpone medically justified MRI scans with contrast agent; six questions help beforehand. The FDA writes that necessary examinations with contrast agent should not be avoided or deferred [2]; the ESUR: never refuse a clinically well-justified examination with contrast agent [3]. According to the EMA decision and the Rote-Hand-Brief (German “Dear Healthcare Professional” letter), gadolinium should be used only if the necessary information cannot be obtained without contrast agent, and then at the lowest effective dose [1][6].

Questions for the radiology team before an MRI with contrast agent

  1. Do I really need the contrast agent for the clinical question, or is the scan without it enough?
  2. Which agent will be used, is it macrocyclic, and can I get the name and the dose for my records? [3]
  3. Is my kidney value known? Tell the practice if you have reduced kidney function or are on dialysis [3]. If your creatinine value appears in your report in a different unit, the creatinine converter helps; it only converts units and gives no eGFR.
  4. How many examinations with contrast agent have I already had, and was there ever a reaction to it? The FDA advises avoiding repeated, closely spaced examinations [2]; after an earlier reaction, the triggering agent should be recorded [3].
  5. Am I pregnant or breastfeeding? The ESUR 2025: with a very strong indication, the smallest possible dose of a macrocyclic agent can be given during pregnancy, and special examinations of the newborn are not necessary afterward. After macrocyclic agents, breastfeeding may continue as normal; pumping and discarding milk for 24 hours is considered unnecessary, and those who want to do so should discuss it with the doctor. With additional kidney impairment, no gadolinium should be given [3]. For the data on pregnancy, see the paragraph below the list.
  6. Is it about a child? The FDA counts children among the groups requiring particular attention [2]; European neuroradiologists in 2026 recommend weighing administration individually in pediatric neuroradiology and considering gadolinium-free protocols [49].

On pregnancy, there is a registry study from Ontario with 397 women who had a contrast MRI during pregnancy. It found a higher risk of stillbirth or neonatal death (adjusted relative risk 3.70, confidence interval 1.55 to 8.85; 7 cases) and of rheumatological, inflammatory or infiltrative skin diseases in the children (adjusted hazard ratio 1.36, confidence interval 1.09 to 1.69); NSF-like diseases were not significantly more common. The case numbers were small, the agent is unknown, and the reason for the examination can distort the numbers [50][5]. This is consistent with the ESUR recommendation to give a contrast agent in pregnancy only with a very strong indication.

Emergency: shortness of breath, hoarseness, swelling in the face or circulation problems after a contrast agent? These can be signs of a severe allergic reaction. Tell the staff at the imaging center immediately or call 911. Immediate reactions usually occur within one hour, rarely up to six hours; delayed skin reactions can occur from one hour up to a week, rarely up to eight weeks. Clarify mild skin rash or itching with the practice or your doctor [3].

Frequently Asked Questions About Gadolinium After an MRI

Does gadolinium stay in the body after an MRI?

A small part, yes. Most of it is excreted via the kidneys; a remainder is demonstrably found in brain, bone and skin [5][12][13][14].

Is gadolinium in the brain safe, or dangerous?

Harm in the brain has not been proven in people with normal kidney function; the EMA sees no evidence, and the ESUR confirms no relevant neurological symptoms. Long-term studies are lacking, so precaution applies [1][3].

Which MRI contrast agent side effects are proven?

Proven as gadolinium side effects are NSF in severe kidney failure, very rare today [4], and rare skin plaques that the EMA links to linear agents [5]. In addition, there are allergic reactions to contrast agents, which usually occur within one hour [3]. For the newest agent, gadopiclenol, the EMA names headache and injection-site reactions as the most common side effects [15]. Other lasting symptoms due to the deposits are not proven [5].

Which agent is the safer one?

Macrocyclic contrast agents release less gadolinium, deposit less and have the lowest NSF risk; for this reason they remain authorized [1][3]. None is safe in the sense of “zero residue” [13].

Who should be especially careful?

People with reduced kidney function, especially an eGFR below 30 or on dialysis, as well as pregnant women, children and people with many earlier examinations with contrast agent should raise it before the appointment [3][2].

Can gadolinium be removed from the body?

Proof that it works is lacking: chelation with DTPA has been studied only in one uncontrolled series of 25 patients, and a follow-up paper was retracted [7][43].

Does a urine test show whether I am poisoned with gadolinium?

No. Urine values are elevated for weeks to months after every administration, including in people without symptoms [40][41].

May I refuse the contrast agent?

After being informed, the decision is yours. Ask whether the scan without contrast agent is enough. If the examination is medically well justified, according to the FDA and ESUR it should not be avoided or postponed [2][3].

How many MRI scans with contrast agent are too many?

The EMA, FDA and ESUR name no fixed limit. The FDA advises avoiding repeated, closely spaced examinations, and the ESUR requires that the agent and dose be documented [2][3].

Conclusion: Gadolinium After an MRI Is a Question of Weighing, Not a Poisoning Case

Gadolinium from MRI contrast agents stays in the body in small amounts. Whether you receive one is therefore always a weighing of the benefit of the examination against open long-term questions. Proven are deposits in brain, bone and skin, more after linear than after macrocyclic agents, NSF in severe kidney failure, and rare skin plaques that the EMA links to linear agents. Not proven are harm in the brain from the deposits in normal kidney function, “gadolinium toxicity” as a diagnosis, and any benefit of chelation. Symptoms after an MRI should be evaluated; urine and provocation tests do not help with that.

When people talk generally about MRI risk or MRI danger, it is often about the magnetic field or implants, and that is a different topic. Here it was only about the contrast agent, and most MRI examinations are done entirely without one. If you need a contrast agent, ask about the agent and dose, tell the staff your kidney values and earlier reactions, and do not postpone a justified examination.

Suggested citation: gesundefakten.com (2026): Gadolinium After MRI: Retention, Side Effects, and Risk of the Contrast Agent. https://gesundefakten.com/en/health/gadolinium-mri-contrast/

Sources

  1. EMA: EMA’s final opinion confirms restrictions on use of linear gadolinium agents in body scans. EMA/625317/2017, November 23, 2017. ema.europa.eu.
  2. U.S. FDA: Drug Safety Communication “FDA warns that gadolinium-based contrast agents (GBCAs) are retained in the body; requires new class warnings”, December 19, 2017, updated May 16, 2018 (archived version of January 3, 2024; the fda.gov page could not be accessed at the time of checking). web.archive.org.
  3. ESUR Contrast Media Safety Committee: Guidelines 2025 (chapters on NSF, gadolinium retention, pregnancy and breastfeeding, environment). esur.org (PDF).
  4. Attari H, Cao Y, Elmholdt TR, Zhao Y, Prince MR: A systematic review of 639 patients with biopsy-confirmed nephrogenic systemic fibrosis. Radiology 2019; 292(2): 376–386. PubMed.
  5. EMA/PRAC: Assessment report, Referral under Article 31, Gadolinium containing contrast agents. EMA/411650/2017, July 6, 2017. ema.europa.eu (PDF).
  6. AkdÄ, marketing authorization holders in coordination with EMA and BfArM: Rote-Hand-Brief (German “Dear Healthcare Professional” letter) “Gadolinium-haltige Kontrastmittel: Aktualisierte Empfehlungen nach Bewertung von Gadoliniumablagerungen im Gehirn und anderen Geweben”, January 8, 2018. akdae.de (PDF).
  7. Semelka RC, Ramalho M, Jay M, Hickey L, Hickey J: Intravenous calcium-/zinc-diethylene triamine penta-acetic acid in patients with presumed gadolinium deposition disease: a preliminary report on 25 patients. Investigative Radiology 2018; 53(6): 373–379. PubMed.
  8. Guerbet GmbH: Dossier zur Nutzenbewertung nach § 35a SGB V, Gadopiclenol (Elucirem), Modul 3 A, as of March 28, 2024, published by the G-BA (manufacturer document; contains figures from the Destatis OPS statistics and from WIdO). g-ba.de (PDF).
  9. BfArM: Risk information “Contrast agents containing gadolinium: Deposits in the brain and other tissues” (Stufenplanverfahren, i.e., German drug safety procedure; extension of the suspension most recently to February 28, 2028). bfarm.de.
  10. Frenzel T, Lengsfeld P, Schirmer H, Hütter J, Weinmann HJ: Stability of gadolinium-based magnetic resonance imaging contrast agents in human serum at 37 °C. Investigative Radiology 2008; 43(12): 817–828 (first author employed at Bayer Schering Pharma). PubMed.
  11. Woolen SA, Shankar PR, Gagnier JJ et al.: Risk of nephrogenic systemic fibrosis in patients with stage 4 or 5 chronic kidney disease receiving a group II gadolinium-based contrast agent: a systematic review and meta-analysis. JAMA Internal Medicine 2020; 180(2): 223–230. PubMed.
  12. McDonald RJ, McDonald JS, Kallmes DF et al.: Intracranial gadolinium deposition after contrast-enhanced MR imaging. Radiology 2015; 275(3): 772–782. PubMed.
  13. Murata N, Gonzalez-Cuyar LF, Murata K et al.: Macrocyclic and other non-group 1 gadolinium contrast agents deposit low levels of gadolinium in brain and bone tissue: preliminary results from 9 patients with normal renal function. Investigative Radiology 2016; 51(7): 447–453. PubMed.
  14. Kobayashi M, Levendovszky SR, Hippe DS et al.: Comparison of human tissue gadolinium retention and elimination between gadoteridol and gadobenate. Radiology 2021; 300(3): 559–569. PubMed.
  15. EMA: Elucirem (Gadopiclenol), EPAR overview, EU authorization December 7, 2023. ema.europa.eu.
  16. Rogosnitzky M, Branch S: Gadolinium-based contrast agent toxicity: a review of known and proposed mechanisms. Biometals 2016; 29(3): 365–376. PubMed.
  17. 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(3): 834–841. PubMed; DOI.
  18. EMA: Gadolinium-containing contrast agents. Referral (Article 31 procedure EMEA/H/A-31/1437; initiation March 17, 2016, CHMP opinion July 20, 2017, EU Commission decision November 23, 2017). ema.europa.eu.
  19. EMA: Teslascan (Mangafodipir), EPAR; authorization 1997, revocation 2012 at the request of the holder for commercial reasons. ema.europa.eu.
  20. Remiszewski P, Drabik MA, Filipowicz P et al.: Manganese-enhanced magnetic resonance imaging: from translational research to first-in-human trials in oncology. Molecular Diagnosis & Therapy 2026. PubMed.
  21. Erstad DJ, Ramsay IA, Jordan VC et al.: Tumor contrast enhancement and whole-body elimination of the manganese-based magnetic resonance imaging contrast agent Mn-PyC3A. Investigative Radiology 2019; 54(11): 697–703. PubMed.
  22. Frenzel T, Apte C, Jost G, Schöckel L, Lohrke J, Pietsch H: Quantification and assessment of the chemical form of residual gadolinium in the brain after repeated administration of gadolinium-based contrast agents: comparative study in rats. Investigative Radiology 2017; 52(7): 396–404 (first author employed at Bayer). PubMed.
  23. Gianolio E, Bardini P, Arena F et al.: Gadolinium retention in the rat brain: assessment of the amounts of insoluble gadolinium-containing species and intact gadolinium complexes after repeated administration of gadolinium-based contrast agents. Radiology 2017; 285(3): 839–849. PubMed.
  24. Strzeminska I, Factor C, Robert P, Szpunar J, Corot C, Lobinski R: Speciation analysis of gadolinium in the water-insoluble rat brain fraction after administration of gadolinium-based contrast agents. Investigative Radiology 2021; 56(9): 535–544 (co-authors employed at Guerbet). PubMed.
  25. Bücker P, Richter H, Radbruch A et al.: Deposition patterns of iatrogenic lanthanum and gadolinium in the human body depend on delivered chemical binding forms. Journal of Trace Elements in Medicine and Biology 2021; 63: 126665. PubMed.
  26. 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(3): 803–809. PubMed.
  27. Radbruch A, Weberling LD, Kieslich PJ et al.: Gadolinium retention in the dentate nucleus and globus pallidus is dependent on the class of contrast agent. Radiology 2015; 275(3): 783–791. PubMed.
  28. Parillo M, Mallio CA, Van der Molen AJ et al.: Skin toxicity after exposure to gadolinium-based contrast agents in normal renal function, using clinical approved doses: current status of preclinical and clinical studies. Investigative Radiology 2023; 58(8): 530–538. PubMed.
  29. Gräfe D, Simion SH, Rosolowski M et al.: Brain deposition of gadobutrol in children: a cross-sectional and longitudinal MRI T1 mapping study. European Radiology 2023; 33(7): 4580–4588. PubMed.
  30. Gulani V, Calamante F, Shellock FG, Kanal E, Reeder SB (ISMRM): Gadolinium deposition in the brain: summary of evidence and recommendations. Lancet Neurology 2017; 16(7): 564–570. PubMed.
  31. Welk B, McArthur E, Morrow SA et al.: Association between gadolinium contrast exposure and the risk of parkinsonism. JAMA 2016; 316(1): 96–98. PubMed.
  32. Kim C, Kim C, Tae BS, Kwon DY, Lee YH: Assessing the association between gadolinium-based contrast agents and Parkinson disease: insights from the Korean National Health Insurance Service database. Investigative Radiology 2025; 60(8): 487–492. PubMed.
  33. Shin NY, Park SK, Kim B et al.: Risk of parkinsonism after exposure to different types of gadolinium-based contrast agents: a nationwide population-based cohort study of 222,977 individuals. Korean Journal of Radiology 2026; 27(3): 276–288. PubMed.
  34. Scaravilli A, Tranfa M, Pontillo G et al.: MR imaging signs of gadolinium retention are not associated with long-term motor and cognitive outcomes in multiple sclerosis. American Journal of Neuroradiology 2023; 44(4): 396–402. PubMed.
  35. Forslin Y, Shams S, Hashim F et al.: Retention of gadolinium-based contrast agents in multiple sclerosis: retrospective analysis of an 18-year longitudinal study. American Journal of Neuroradiology 2017; 38(7): 1311–1316. PubMed.
  36. Vymazal J, Krámská L, Brožová H, Růžička E, Rulseh AM: Does serial administration of gadolinium-based contrast agents affect patient neurological and neuropsychological status? Fourteen-year follow-up of patients receiving more than fifty contrast administrations. Journal of Magnetic Resonance Imaging 2020; 51(6): 1912–1913. PubMed.
  37. Marckmann P, Skov L, Rossen K et al.: Nephrogenic systemic fibrosis: suspected causative role of gadodiamide used for contrast-enhanced magnetic resonance imaging. Journal of the American Society of Nephrology 2006; 17(9): 2359–2362. PubMed.
  38. Semelka RC, Ramalho J, Vakharia A et al.: Gadolinium deposition disease: initial description of a disease that has been around for a while. Magnetic Resonance Imaging 2016; 34(10): 1383–1390. PubMed.
  39. Burke LM, Ramalho M, AlObaidy M, Chang E, Jay M, Semelka RC: Self-reported gadolinium toxicity: a survey of patients with chronic symptoms. Magnetic Resonance Imaging 2016; 34(8): 1078–1080. PubMed.
  40. Alwasiyah D, Murphy C, Jannetto P, Hogg M, Beuhler MC: Urinary gadolinium levels after contrast-enhanced MRI in individuals with normal renal function: a pilot study. Journal of Medical Toxicology 2019; 15(2): 121–127. PubMed.
  41. McDonald JS, Day PL, Spears GM et al.: Serum and urine gadolinium reference intervals in patients with normal renal function following gadobutrol administration. Investigative Radiology 2025; 60(9): 586–591. PubMed.
  42. Ramalho J, Ramalho M, Semelka RC: Gadolinium elimination in a gadolinium deposition disease population after a single exposure to gadolinium-based contrast agents. Investigative Radiology 2025; 60(7): 429–433 (two of three authors report an unpaid board position at GadTTRAC). PubMed.
  43. Semelka RC, Ramalho M: Near-cure in patients with gadolinium deposition disease undergoing intravenous DTPA chelation. Frontiers in Toxicology 2024; 6: 1371131; retracted on November 25, 2024 (Retraction Notice: doi 10.3389/ftox.2024.1528757). PubMed.
  44. Prybylski JP, Coste Sanchez C, Jay M: Impact of chelation timing on gadolinium deposition in rats after contrast administration. Magnetic Resonance Imaging 2019; 55: 140–144. PubMed.
  45. Umweltbundesamt, Kommission „Human-Biomonitoring“: Einsatz von Chelatbildnern in der Umweltmedizin? Bundesgesundheitsblatt 1999; 42: 823–824 (concerns DMPS and DMSA, not gadolinium). umweltbundesamt.de (PDF).
  46. Schmidt K, Bau M, Merschel G, Tepe N: Anthropogenic gadolinium in tap water and in tap water-based beverages from fast-food franchises in six major cities in Germany. Science of the Total Environment 2019; 687: 1401–1408. PubMed.
  47. Brünjes R, Hofmann T: Anthropogenic gadolinium in freshwater and drinking water systems. Water Research 2020; 182: 115966. PubMed.
  48. Pereto C, Lerat-Hardy A, Baudrimont M, Coynel A: European fluxes of medical gadolinium to the ocean: a model based on healthcare databases. Environment International 2023; 173: 107868. PubMed.
  49. Righini A, Löbel U, Galijasevic M et al.: ESR Essentials: “Gadolinium-free” pediatric neuroimaging, practice recommendations by the European Society of Neuroradiology. European Radiology 2026. PubMed.
  50. Ray JG, Vermeulen MJ, Bharatha A, Montanera WJ, Park AL: Association between MRI exposure during pregnancy and fetal and childhood outcomes. JAMA 2016; 316(9): 952–961. PubMed.