<b><i>Background and Aims:</i></b> Conventional transcatheter arterial chemoembolization (C-TACE) and drug-eluting bead (DEB)-based TACE are current treatments for hepatocellular carcinoma (HCC). We compared the therapeutic efficacies and adverse events of these methods in a single-center retrospective cohort study. <b><i>Methods:</i></b> We enrolled 174 patients treated between January 2010 and October 2016; 98 and 76 underwent C-TACE and DEB-TACE, respectively, with 76 and 22 of the former group and 49 and 27 of the latter group classified as Child-Pugh class A and B, respectively. Therapeutic outcomes, progression-free survival (PFS), and adverse events were evaluated. <b><i>Results:</i></b> The PFS rates in the C-TACE and DEB-TACE groups were 8.1 and 6.1 months, respectively (<i>p</i> = 0.79). The response and disease control rates were 64 and 71% in C-TACE patients and 69 and 78% in DEB-TACE patients, respectively (<i>p</i> = 0.25). Postprocedural pain, vomiting, and fever were more frequent following C-TACE than DEB-TACE (<i>p</i> &#x3c; 0.001). In contrast, the incidences of bilomas and arterio-portal shunts were significantly higher following DEB-TACE (<i>p</i> &#x3c; 0.001); the incident rates of arterio-portal shunt formation were 8.1 and 48.7% in patients undergoing C-TACE and DEB-TACE, respectively. Child-Pugh class A was significantly associated with arterio-portal shunt formation after DEB-TACE on multivariate analysis. <b><i>Conclusions:</i></b> There were no significant differences in the therapeutic efficacies of C-TACE and DEB-TACE. However, the frequency of arterio-portal shunt formation was significantly higher in HCC patients with Child-Pugh class A undergoing DEB-TACE. Our findings imply that C-TACE should be selected for HCC patients with Child-Pugh class A and DEB-TACE should be chosen for those with Child-Pugh class B.