Joo Cho MD, Presenter: Nothing to Disclose

Salar Hakham DO, Abstract Co-Author: Nothing to Disclose

John David Towle MD, Abstract Co-Author: Nothing to Disclose

Paul Kim, Abstract Co-Author: Nothing to Disclose

Younghun Han PhD, Abstract Co-Author: Nothing to Disclose

Hoon Ji MD, PhD, Abstract Co-Author: Nothing to Disclose

The purpose was to conduct meta analysis on the diagnostic performance of dynamic multidetector CT (MDCT), dynamic MRI with gadolinium, diffusion-weighted imaging (DWI), Gadoxetic Acid-EOB-DTPA (Eovist) Hepatobiliary phase MRI, and combined DWI and Eovist MRI in detection of hepatocellular carcinoma. 

PubMed was searched with key phrases of “DWI and HCC” and “DTPA-EOB and HCC”, which resulted in relevant original articles published from 1993 to 2013. After reviewing each article, relevant articles were pooled for subgroup data analysis with commercially available statistical software.

Thirty articles with 1948 patients, 2489 HCCs, and 972 benign lesions were analyzed. The overall DWI sensitivity was 84.2% (95% CI: 82.7%-85.7%), specificity 93.1% (95% CI: 91.7%-94.4%), and the sensitivity for HCC ≤1 cm was 68.8% (95% CI: 64.5%-73.1%). The overall Gadoxetic Acid-EOB-DTPA Hepatobiliary MRI sensitivity was 92.2% (95% CI: 91.0%-93.4%), specificity 97.4% (95% CI: 96.4%-98.3%), and the sensitivity for HCC ≤1 cm was 80.3% (95% CI: 76.5%-84.1%). The diagnostic sensitivity of dynamic MDCT and dynamic MRI were lower than the DWI or the hepatobiliary phase studies with overall sensitivity of 78.7% (95% CI: 76.3%-81.0%) and 79.5% (95% CI: 77.4%-81.7%) respectively for all HCC lesions and 23.0% (95% CI: 14.1%-32.0%) and 60.1% (95% CI: 54.8%-65.3%) respectively for HCC measuring ≤1 cm. The diagnostic specificity was high in all image modalities. Meta analysis of studies that combined the diagnostic performance of DWI and Eovist MRI resulted in sensitivity of 92.9% (95% CI: 91.1%-94.7%) and specificity of 97.1% (95% CI: 96.3%-98.0%). The overall diagnostic odds ratio for DWI was 71.9 (95% CI: 53.4-96.8), Eovist MRI 442.8 (95% CI: 281.5-696.49), dynamic MDCT 74.9 (95% CI: 53.9-104.2), dynamic MRI 52.3 (95% CI: 39.2-69.8), and DWI+Eovist MRI 438.1 (95% CI: 281.8-681.1).

DWI or Gadoxetic Acid-EOB-DTPA Hepatobiliary phase MRI offer added value to the diagnosis of HCC and demonstrates superior diagnostic sensitivity, especially in HCC ≤1 cm, when compared to the conventional dynamic MDCT or MRI. Combined DWI and Eovist MRI resulted in high sensitivity and specificity, but not statistically different from Eovist MRI. 

Further refinement of the AASLD and LI-RAD guidelines with incorporation of DWI and hepatobiliary phase MRI, leading to improved HCC detection and clinical outcome.

Cho, J, Hakham, S, Towle, J, Kim, P, Han, Y, Ji, H, Utility of Diffusion-weighted Imaging and Hepatobiliary Phase MRI in Detection of Hepatocellular Carcinoma: A Meta Analysis.  Radiological Society of North America 2013 Scientific Assembly and Annual Meeting, December 1 - December 6, 2013 ,Chicago IL. http://archive.rsna.org/2013/13044201.html