RESUMO
<p><b>OBJECTIVE</b>To assess the diagnostic value of magnetic resonance imaging (MRI) in the follow-up of patients with hepatocellular carcinomas treated with radiofrequency ablation (RFA) and to compare it with that of computed tomography (CT).</p><p><b>METHODS</b>From December 2009 to September 2011, 40 patients (47 hepatocellular carcinomas) were treated with RFA after transcatheter arterial chemoembolization and underwent MRI and CT for follow-up. RFA margins were assessed on a five-point scale with receiver operating characteristic curve analysis. Sensitivity, specificity, positive predictive value, negative predictive value, and accuracy were evaluated.</p><p><b>RESULTS</b>The interobserver agreement rate for MRI was significantly higher (Kappa=0.935) than for CT (Kappa=0.714; P < 0.05). The scores of 1 and 5 points for MRI, which confirms the presence or absence of residual tumor, accounted for 89.4% (84/94), while for CT accounting for only 31.9% (30/94). The area under the receiver operating characteristic curve of MRI was significantly higher than that of CT (P < 0.05), as were the sensitivity, specificity, positive predictive value, negative predictive value, and accuracy of detection rate (mean, 100%, 96.4%, 76.9%, 100%, and 96.8% for MRI, respectively, vs. 30.0%, 57.1%, 10.3%, 87.7%, and 63.8% for CT).</p><p><b>CONCLUSION</b>MRI is superior to CT in assessing the RFA margins in terms of the diagnostic accuracy and detection rate .</p>