RESUMO
INTRODUCTION: Ultrasound is a widely utilized method of screening patients with chronic liver disease for hepatocellular carcinoma (HCC). However, the sensitivity of ultrasound for small tumours is limited. We have prospectively compared ultrasound screening with diffusion-weighted (DWI) MRI for detecting HCC. METHODS: Patients with chronic liver disease referred for ultrasound screening underwent a liver ultrasound and a liver MRI comprising free breathing DWI. Each test was independently read to determine the accuracy of each modality for detecting HCC. RESULTS: One hundred and ninety-two patients were recruited and HCC was diagnosed in six patients (3%); all of whom were detected at ultrasound screening, and five detected at MRI screening. Ultrasound had false-positive studies 20 times (10%) while DWI MRI had three false-positive examinations (2%) p≥0.05. The sensitivity, specificity, positive predictive value and negative predictive values for ultrasound are 100%, 90%, 23% and 100%, respectively, while for MRI are 83%, 98%, 63% and 99%. CONCLUSION: In patients with chronic liver disease undergoing surveillance for hepatocellular carcinoma, DWI MRI screening shows similar sensitivity to screening ultrasound but with a significantly lower false-positive rate.