Diagnostic performances of adding transition and hepatobiliary phase to washout in gadoxetic acid-enhanced MRI for subcentimeter hepatocellular carcinoma.

Objective: To compare the diagnostic performances of washout (WO) criteria confined to portal venous phase (PVP), extended to transition phase (TP) or hepatobiliary phase (HBP) for subcentimeter hepatocellular carcinoma (HCC) using gadoxetic acid-enhanced MRI.

Materials and methods: Two hundred patients (mean age, 53.4 ± 10.4 years) who were scanned with 1.5-T gadoxetic acid-enhanced MRI were included. Arterial phase hyperenhancement and signal intensity on PVP, TP and HBP images were evaluated. Three WO criteria were established: WO confined to the PVP, WO extended to the TP and WO extended to the HBP.

Assessment: Sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and accuracy with 95% confidence interval were calculated for WO confined to the PVP and WO extended to the TP or HBP, and generalized estimating equation was used to compare sensitivity, specificity, PPV, NPV and accuracy.

Results: A total of 200 surgically confirmed subcentimeter hepatic nodules (171 HCCs and 29 non-HCCs) were included. For solitary subcentimeter HCCs, WO extended to the TP showed significantly higher sensitivity and accuracy than WO confined to the PVP (70.80% vs. 49.64%, P < 0.05; 71.52% vs. 55.70%, P < 0.001) and AFP (70.80% vs. 51.09%, P < 0.05; 71.52% vs. 55.70%, P < 0.05), but no significant difference in specificity (76.19% vs. 95.24%, P > 0.05; 76.19% vs. 85.71%, P > 0.05). Extending WO to the HBP achieved significantly higher sensitivity (78.10% vs. 49.64%, P < 0.001; 78.10% vs. 70.80%, P < 0.05) and accuracy (75.32% vs. 55.70%, P < 0.001; 75.32% vs. 71.52%, P < 0.05) compared to WO confined to the PVP and WO extended to the TP, but lower specificity (57.14%% vs. 95.24%, P < 0.05) compared to WO confined to the PVP. The combination of AFP with WO confined to the PVP, WO extended to the TP, or WO extended to the HBP yielded significantly higher sensitivities and accuracies compared to each MRI washout criterion alone (all P < 0.001).

Conclusion: WO extending to TP rather than HBP may be a preferred alternative for diagnosing subcentimeter HCC, and the supplement of AFP may improve sensitivity and accuracy.