Patients:
This was a prospective study done on 80 lesions in 59 patients over a period of 6 months during September 2013 to March 2014.
These patients were referred for MRI of abdomen in the Department of Radiology of our hospital from different oncology units.
Only those lesions, which underwent FNAC or biopsy under US or CT guidance or had excision done during the abdominal surgery or had typical imaging features on MR or other imaging modalities,
were included in the study.
Atypical lesions were correlated with pathological diagnosis.
Informed consent was taken from all the patients after explaining the entire procedure,
its benifits and risks.
All patients were subjected to full clinical evaluation,
laboratory investigations and radiological assessment using conventional MRI and Diffusion-weighted MRI with or without post contrast imaging.
MRI protocol for liver:
MRI studies were performed on 1.5 T unit (GE Signa HDx1.5T MRI,
GE Healthcare system) by using body phased-array coil with patient in supine position.
1.
Axial T1-weighted ( TR/TE of 600/30ms; non fat sat) and T2-weighted (TR/TE of 200/100ms; Fat sat).
For both T1 &T2 the slice thickness was 6mm,
FoV of 34 cm and an acquisition matrix of 256 x 256.
These images were used for assistance in lesion detection and characterization.
2.
Diffusion-weighted Imaging was performed using respiratory triggered protocol at b value 0 and 800,
with the single shot echo-planar imaging (EPI) technique in axial plane,
matrix=128 x 128; slice thickness 6mm and interslice gap of 1mm.
Qualitative assessment at different b values and quantitative assessment by measuring ADC values were done.
3.Dynamic post contrast (Gd-DTPA) T1-weighted axial 3D fat suppressed GRE were performed in some of the cases.
Images were acquired in all 3 phases after bolus injection at the rate of 0.2mmol/kg body weight of Gd-DTPAflushed with 20ml of sterile 0.9% of saline.
Arterial phase at 20 sec,
portal venous phase at 60sec and delayed at 180sec after contrast injection.
Diagnosis of malignant liver lesions (primary or metastatic) were confirmed by histopathological diagnosis.
The diagnosis of benign liver lesions were confirmed either by histopathological diagnosis or with typical imaging findings on US,
CT or MRI or if they remained stable on follow up scans.
Statistical analysis:
Data were analyzed using statistical software R version 3.1.2 with p<0.001.
Quantitative data of different variables were expressed as mean ± standard deviation (SD).
Reciever operating characteristics (ROC) was drawn to identify cut off values for the ADC that best classified the liver lesions as benign and malignant.