Diffusion weighted MR imaging versus F18 FDG PETC CT in identifying small hepatic metastases.

Abstract

e14675 Background: This exploratory study assessed the ability of diffusion weighted MRI imaging in identifying small liver metastases as compared to F18- FDG PETCT. Methods: Patients with diagnosis of carcinoma of breast, lung, colorectal, pancreas, neuroendocrine and HCC who were staged for their disease using PET CT and MRI were included in the study. We randomly analyzed and compared PETCT and MRI images of 40 patients with hepatic metastases for sensitivity in detection of metastatic lesions in the liver on PETCT and MRI. Where possible, biopsy was performed if MRI/PET CT showed one or two small lesions in the liver. We performed ANCOVA to detect differences between PETCT and MRI controlling for primary sites of disease as a covariate. Results: The mean number of small lesions detected on PETCT was 0.92 where as it was 3.96 in MRI. There was a significant increase in small (p=0.02) and total number of lesions (p=0.01) seen on MRI compared to PET CT but not in large lesions. MRI was able to detect subcentimeter lesions more accurately than PET CT. Random biopsies done on small lesions found on MRI were positive for disease. MRI was able to detect small subcentimeter lesions three times more accurately than PET CT. Conclusions: MR diffusion is more sensitive in indentifying subcentimeter metastatic deposits in the liver that could help in accurate staging when PET CT is negative for liver metastases or when locoregional treatment for oligometastases is contemplated. However, pixels of DWI MRI (like reduced Signal to noise ratio at high b value, artefacts at liver edges and left lobe due to exaggerated susceptibility weighting and spin dephasing) need to be addressed by using other morphological images.