Effect of gadolinium chelate contrast agents on diffusion weighted MR imaging of the liver, spleen, pancreas and kidney at 3 T.

PURPOSE

To retrospectively test the null hypotheses that the qualitative appearance of DWI and the signal intensity values in DWI and corresponding ADC values of the liver, spleen, pancreas and kidneys are identical before and after the administration of gadolinium.

MATERIALS AND METHODS

Following IRB approval, DWI was acquired in 50 patients (25 male; mean age 54.9 years) prior to and after contrast administration, using single-shot echo planar imaging with b-values of 50 s/mm2 and 800 s/mm2 at 3 T. Binomial analysis was used to determine which image set was more significantly preferred in conveying the diffusion information. Pre- and post-gadolinium DWI and ADC values of corresponding regions of each organ were analyzed using standardized signal intensity measurements.

RESULTS

Pre-contrast DWI images of the liver, spleen, and pancreas were preferred 52%, 49%, and 58%, respectively, with none of the differences being statistically significant. DWI of the kidneys was preferred on pre-contrast images in 83% (p<0.001). In the liver and spleen, contrast caused a significant increase in the post-contrast DWI signal intensity values at b=50 (p<0.02) and b=800 (p<0.05) but had no statistically significant effect on the ADC value (p>0.40). Pancreatic DWI signal intensity and ADC values pre- and post-contrast were also not significantly different (p=0.489). In the renal parenchyma, significant decrease in the values of DWI at b=50 (p<0.01) and b=800 (p<0.01) as well as ADC (p<0.02) was demonstrated following gadolinium administration.

CONCLUSION

Intravenous gadolinium administration does not make a statistically significant difference in the qualitative appearance or ADC measurements of the liver, spleen, or pancreas when comparing pre-contrast to post-contrast DWI. In the kidneys, however, ADC values are significantly lower post-contrast with the pre-contrast diffusion weighted images also being qualitatively preferred.