Gadolinium Found in Brain Tissue

Pauline Anderson

March 26, 2015

Researchers have found deposits of gadolinium in the brains of deceased patients who had undergone multiple contrast-enhanced MRI examinations during their lifetime.

A new study shows a dose-dependent relationship between use of intravenous gadolinium-based contrast agents (GBCAs) and deposition of gadolinium in neural tissues.

GBCAs contain the rare earth metal gadolinium and an organic ligand. GBCAs, approved by the US Food and Drug Administration (FDA) in 1988, are now the most commonly used MRI contrast agents.

The findings of the study, published online March 5 in Radiology, were a bit of a surprise, said lead study author Robert McDonald, MD, PhD, a resident in the Department of Radiology, Mayo Clinic, Rochester, Minnesota.

"The fact that a tiny fraction of the administered dose of these contrast agents is depositing in neural tissues was unexpected," he told Medscape Medical News. "These agents were designed to remain in the bloodstream and not deposit in tissues."

It's too early to determine the significance of the study results, the researchers say. "We know of no health risk at this time," commented another author, Laurence Eckel, MD, staff radiologist, Neuro-radiology Division, Department of Radiology, Mayo Clinic. "These FDA-approved agents have been safely provided to tens of millions of patients per year."

Surprising Finding

The study was prompted by a previous study from Japan that noted a progressive increase in signal in certain parts of the brain after administration of successive doses of gadolinium. "This initial paper found increases in the precontrast T1-weighted signal over time, which they postulated was indirect evidence of deposition," said Dr McDonald.

The new study from the Mayo Clinic included 23 patients, mostly from Olmsted County in Minnesota, who had consented to an autopsy. All patients had undergone MRI of the brain, which included an unenhanced T1-weighted axial sequence of the entire brain. A control group of 10 patients had undergone one to six unenhanced MRI examinations without use of a GBCA.

The contrast group of 13 patients had received 4 to 29 GBCA-enhanced MRI examinations of the brain between 2000 and 2014. The brain imaging was performed with the intravenous agent gadodiamide.

Study participants had had MRI of the brain because of a concern about or a known intracranial pathologic abnormality. Most in the contrast group had undergone serial brain MRI for surveillance of known or resected central nervous system tumors. Ten of the 11 patients in this exposed group with intracranial malignancies had tumors that were limited to the supratentorial region without posterior fossa involvement.

Patients in the contrast group were significantly younger than the control group (mean age at time of death, 51 years vs 83.5 years) and were significantly younger at the time of their first MRI. The median interval between the last MRI examination and death was significantly shorter in the contrast group.

Despite slight differences in baseline renal and hepatobiliary function, which represent the two biological excretion pathways for GBCAs, most patients in the contrast-exposed group had normal kidney and liver function at the time of MRI. Renal function is commonly assessed before each MRI examination to screen for chronic renal failure.

As per protocol, gadolinium contrast material is withheld from any patient with severe chronic kidney disease. This, said Dr Eckel, is because these patients are at a slightly increased risk for an "exceptionally rare" skin disorder called nephrogenic systemic fibrosis (NSF). None of the patients in this study had severe chronic renal disease or evidence of NSF.

Harvesting Tissue

From all patients, the researchers harvested neuronal tissues from the dentate nuclei in the cerebellum and the pons and from two deep brain regions, the globus pallidus and thalamus. Using inductively coupled plasma mass spectrometry, transmission electron microscopy, and light microscopy, they quantified, localized, and assessed the effects of gadolinium deposits in the tissue.

The researchers found that all the control patients had undetectable levels of gadolinium.

In contrast, all patients exposed to multiple doses of a GBCA had elevated levels of elemental gadolinium in the four neuroanatomic regions of interest, ranging from 0.1 to 58.8 μg of gadolinium per g of dried neural tissue.

The dentate nucleus contained the highest median concentration of elemental gadolinium, but deposits were also found in the thalamus and globus pallidus. The pons had the lowest overall absolute change in T1-weighted signal intensity with gadolinium.

"For reasons that are not yet clear to us, the gadolinium appears to be preferentially depositing in the dentate nucleus of the cerebellum," said Dr McDonald.

Gadolinium deposition, he said, appears to be dose dependent, varying with the number of contrast-enhanced MRI examinations a patient underwent during his or her lifetime. However, Dr Eckel stressed that these deposits were detected in "very trace amounts."

The dose-dependent relationship between use of GBCAs and subsequent neuronal tissue deposition was independent of the patient's age, sex, baseline renal and hepatobiliary function, and interval between gadolinium exposure and death.

Crossing the Barrier

Transmission electron microscopy showed that most of the gadolinium deposits were in the endothelial walls. However, 18% to 42% of the gadolinium appeared to have crossed the blood-brain barrier and was deposited in the neural interstitium.

"Certainly, that wasn't expected; we didn't think these agents could do that," said Dr McDonald.

It's important to note that pathologists investigating the tissue with the gadolinium deposits found little to worry about. "They couldn't detect evidence for neural tissue damage," said Dr McDonald.

The fact that these experts couldn't see any differences between gadolinium-naive and exposed tissues is reassuring, said Dr Eckel. "In the unlikely event that there is a cellular response to this deposition, it is likely very minimal and without any structural damage to the brain."

While some people might feel that any metal in the brain can't be a good thing, Dr McDonald pointed out that certain metals, such as iron and calcium, normally deposit in the same areas of the brain.

"Is gadolinium different? It's a rare earth metal and not something that's ordinarily found in the environment in significant quantities, so it's worth investigating" said Dr. McDonald, "But we can't automatically assume that it's necessarily bad."

Since it was first used as a contrast agent, gadolinium has "revolutionized" MRI, said Dr McDonald. "If we took gadolinium out of the equation, the clinical utility of MRI as a test to identify and monitor disease would, in many cases, be reduced."

There are many examples where gadolinium plays a crucial role in our ability to diagnose disease, added Dr McDonald.

In addition to brain abnormalities, GBCAs also help diagnose various diseases in other parts of the body, including the heart, liver, bowel, breast, and prostate.

One of the next steps for the research group is to determine which of the approximately half-dozen types of GBCAs preferentially deposit in tissue and in what amounts, since different agents have different chemical structures, said Dr. Eckel.

"People are investigating which agents might be involved in deposition and which might not be; no one knows the answer to that at this point."

Going forward, investigators are developing an animal model to better study this phenomenon and to determine whether there's a cellular response to this deposition by using gene expression profiling. "Using these techniques, we will be able to determine if and how cells in the brain are reacting to these deposits," said Dr McDonald.

Asked to comment, imaging expert Marc Nuwer, MD, PhD, professor, neurology, University of California at Los Angeles, said that the findings are "both interesting and unexpected," but it's unclear whether they have any clinical significance.

The data, said Dr Nuwer "reasonably lead to the conclusion" that gadolinium is leaking through the "normal" blood-brain barrier.  

"These authors have tried to exclude known causes of blood-brain barrier breakdown, leading to the clear suggestion that this retention of gadolinium in the endothelium and brain parenchyma is not due to local effects of tumor or inflammation." 

The authors call for further study that would include determining any long-term effects of the brain deposits of gadolinium. "We don't know of any adverse effects," said Dr Nuwer. "I agree it deserves some long-term follow-up studies or animal research."

Dr McDonald, Dr Eckel, and Dr Nuwer have disclosed no relevant financial relationships.

Radiology. Published online March 5, 2015. Abstract


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.