Transarterial chemoembolization (TACE) is one of the standard locoregional treatments for intermediate stage hepatocellular carcinoma (HCC). Transarterial radioembolization (TARE) using β-emitting yttrium-90 (90Y) integral to the glass matrix of the microspheres has been developed as an alternative to TACE in recent years. Thus, we conducted a meta analysis to evaluate the safety and efficacy of TARE versus TACE for unresectable HCC. We searched PubMed, EMBASE, Web of science and the Cochrane Library for clinical trials comparing TARE with TACE for unresectable HCC. Response rate, overall survival (OS), time to progression (TTP), hospitalization time days and clinical complications were analyzed and compared. Eight studies published from 2009 to 2014, with a total of 1,499 patients, were included in this meta-analysis. The pooled results showed that TARE (90Y) is significantly better in OS (HR = 0.74; 95% CI: 0.61-0.90), 3-year OS rates (RR = 1.75; 95% CI = 1.01-3.03, p = 0.05), TTP (HR = 0.61; 95% CI: 0.41-0.89), hospitalization time days (mean difference = -2.66; 95% CI: -4.08 - -1.24) and some complications (abdominal pain [RR = 0.30, 95% CI: 0.11-0.83, p = 0.02]) for patients with HCC, but did not affect tumor response (CR [RR = 1.06; 95% CI = 0.51-2.22], PR [RR = 1.24; 95% CI = 0.79-1.94], SD [RR = 1.13; 95% CI = 0.92-1.39], PD[RR = 0.75; 95% CI = 0.37-1.51], over-all tumor control [RR = 1.16; 95 % CI = 0.94-1.44]). The current meta-analysis suggests that TARE (Y90) is significantly better in OS, 3-year OS rates, TTP, hospitalization time days and some complications for patients with HCC.