Is contrast use in MRI safe?


Radiology. 2014 Mar;270(3):834-41. doi: 10.1148/radiol.13131669. Epub 2013 Dec 7.

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.

Kanda T, Ishii K, Kawaguchi H, Kitajima K, Takenaka D.


To explore any correlation between the number of previous gadolinium-based contrast material administrations and high signal intensity (SI) in the dentate nucleus and globus pallidus on unenhanced T1-weighted magnetic resonance (MR) images.


The institutional review board approved this study, waiving the requirement to obtain written informed consent. A group of 381 consecutive patients who had undergone brain MR imaging was identified for cross-sectional analysis. For longitudinal analysis, 19 patients who had undergone at least six contrast-enhanced examinations were compared with 16 patients who had undergone at least six unenhanced examinations. The mean SIs of the dentate nucleus, pons, globus pallidus, and thalamus were measured on unenhanced T1-weighted images. The dentate nucleus-to-pons SI ratio was calculated by dividing the SI in the dentate nucleus by that in the pons, and the globus pallidus-to-thalamus SI ratio was calculated by dividing the SI in the globus pallidus by that in the thalamus. Stepwise regression analysis was undertaken in the consecutive patient group to detect any relationship between the dentate nucleus-to-pons or globus pallidus-to-thalamus SI ratio and previous gadolinium-based contrast material administration or other factors. A random coefficient model was used to evaluate for longitudinal analysis.


The dentate nucleus-to-pons SI ratio showed a significant correlation with the number of previous gadolinium-based contrast material administrations (P < .001; regression coefficient, 0.010; 95% confidence interval [CI]: 0.009, 0.011; standardized regression coefficient, 0.695). The globus pallidus-to-thalamus SI ratio showed a significant correlation with the number of previous gadolinium-based contrast material administrations (P < .001; regression coefficient, 0.004; 95% CI: 0.002, 0.006; standardized regression coefficient, 0.288), radiation therapy (P = .009; regression coefficient, -0.014; 95% CI: -0.025, -0.004; standardized regression coefficient, -0.151), and liver function (P = .031; regression coefficient, 0.023; 95% CI: 0.002, 0.044; standardized regression coefficient, 0.107). The dentate nucleus-to-pons and globus pallidus-to-thalamus SI ratios in patients who had undergone contrast-enhanced examinations were significantly greater than those of patients who had undergone unenhanced examinations (P < .001 for both).


High SI in the dentate nucleus and globus pallidus on unenhanced T1-weighted images may be a consequence of the number of previous gadolinium-based contrast material administrations.

J Neurol Neurosurg Psychiatry. 2017 Jul 14. pii: jnnp-2017-316102. doi: 10.1136/jnnp-2017-316102. [Epub ahead of print]

Lack of T1 hyperintensity in the dentate nucleus after 15 administrations of a macrocyclic contrast agent in multiple sclerosis. 

Eisele P, Szabo K, Alonso A, Ong M, Platten M, Schoenberg SO, Gass A

Figure: Increased signal intensity particularly in the dentate nucleus (picture to the left) is observed with linear gadolinium-based contrast agents.

For all the advantage an MRI scan provides in terms of diagnosis and disease monitoring, it is also probably the safest and most non-invasive investigation we have available to us in the 21st century. But, like all developments, what is considered superior at the time will at some point be considered inferior. Since its discovery 40 years ago, the first major problem with MRIs becomes apparent, striking at its main functionality, which is its ease of repetition. It is yet not known what the long-term effects of the high signal intensity seen in the brain (specifically the dentate nucleus and the globus pallidus) of those repeatedly receiving gadolinium, but it cannot be ignored.

It has emerged that linear gadolinium-based contrast agents (GBCAs) may be to blame (see figure above), even in the absence of kidney impairment (most things are got rid of from the body by the kidneys). Based on these findings, on the 10th March 2017m the EMAs Parmacovigilance and Risk Assessment Committee (PRAC) recommended suspension of the marketing authorisations for four gadolinium contrast agents. There is insufficient evidence to suggest whether exposure to macrocyclic GBCAs leads to deposition of gadolinium in the brain. The letter to JNNP (see above for the reference) suggests that this may be a safe alternative. They sampled 22 PwMS who had had 15-26 (average 17) contrast scans with a follow-up period of 3-7 years (average 4.5 years), and found no increase in signal intensity in the MRI examinations using the macrocylic contrast agents gadoterate meglumine and gadobutrol. 

I’m not certain, that this sufficiently placates my concern over gadolinium use. Overall, this new development is likely to impact significantly on clinical decision making. One recommendation, maybe, is to use surveillance MRIs less frequently (this will impact on clinicians practicing ‘no evidence of disease activity’), or to rely more on non-contrast scans. Another, maybe, to use alternative modes of monitoring, such as clinical or blood/CSF biomarkers.

About the author

Neuro Doc Gnanapavan


  • Abstract
    Background: Previous studies have postulated an association between dentate nucleus T1 hyperintensity
    and multiple sclerosis (MS)-related progressive neurodegeneration. Therefore, MS patients have been
    excluded from most studies investigating brain deposition of gadolinium-based contrast agents (GBCAs).

    Conclusion: Dentate nucleus T1 hyperintensity in MS patients is associated with Gd-DTPA (but not Gd-
    BT-DO3A) administration, suggesting an alternative explanation for the association of T1 hyperintensity
    with disease duration and severity.

    • It means that the "ends" of the molecule which holds gadolinium inside a large cycle (macrocycle) do not flip and flop independently. As a result, the cyclic molecule holds gadolinium more tightly, and any release of gadolinium would take much longer. There are many ways (both known and hypothetical) to further improve gadolinium contrast agents. The interest in pursuing this area decreased dramatically with several approved contrast agents being commercially available and widely used. If the need is there, it will not take long to make dramatic improvements rather rapidly.

    • The publications are indicating that this is the case, but we need to keep an open mind.

  • I refused such an MRI as it just didn't seem like a good idea – making bits of my brain "soak up" gadolinium. Or taxing my kidneys. Glad I followed my instincts.



Recent Posts

Recent Comments