A retrospective cohort evaluation of the effect of multiple administrations of gadopentetate dimeglumine on brain magnetic resonance imaging T1-weighted signal.

Abstract

Gadolinium deposition occurs following repeated administration of gadolinium-based contrast media. However, few studies have evaluated factors that lead to increased detection of deposition or the individual differences among patients. To measure the effect of repeated dosages of gadopentetate dimeglumine on pediatric brains and to determine the factors that influence signal intensity changes. A retrospective study evaluated magnetic resonance imaging (MRI) in patients <18years of age who received >5 doses of gadopentetate dimeglumine. Regions of interest were placed in 30 locations in the brain on axial precontrast T1 images. Signal intensity ratios were evaluated throughout the brain. The effect of increasing gadopentetate dimeglumine exposure on signal intensity ratios was assessed using linear mixed models adjusted for gender, age, imaging sequence type (fast spin echo or gradient echo), MRI manufacturer (General Electric, Philips or Siemens), and field strength (1.5 tesla [T] or 3T). Finally, the variance of the random slope in the linear mixed models was tested to determine if there were differences in the rate of signal intensity ratio change among individuals. Fifty patients (M:F=25:25; mean age at first imaging: 6.4years) with a mean of 21.5 gadopentetate dimeglumine administrations (range: 6-86) were included. There were significant increases in T1 signal in the globus pallidus, dentate nucleus and pulvinar with an increasing number of contrast administrations. Patient gender, age, and MRI field strength were not associated with changing signal intensity ratios. However, MRI sequence type and vendor significantly impacted some measured signal intensity ratios. Finally, significant differences in the slopes of the ratios were present among patients for multiple locations. Repeated administration of gadopentetate dimeglumine is associated with T1 hyperintense signal in the dentate nucleus, globus pallidus and pulvinar. Detection is significantly affected by MRI sequence type and scanner vendor. Even when accounting for these differences, there are individual differences in the slope of signal intensity change suggesting a patient-level effect influences gadolinium deposition.