Previous studies have shown that some microRNAs (miRs) are intensively involved in the development of hepatocellular carcinoma. We analyzed the prognostic role of serum microRNA (miR-122) levels in hepatocellular carcinoma patients using a retrospective design. MiR-122 levels in 122 hepatocellular carcinoma patients were measured, and Cox regression analysis was performed to analyze the prognostic role of miR-122 in hepatocellular carcinoma, and the hazard ratio (HR) with 95 % confidence interval (95 %CI) was used to evaluate its prognostic role. Patients with large tumor size had lower levels of serum miR-122 (P = 0.04). However, there was no significant association of serum miR-122 levels with other clinical characteristics. Kaplan-Meier method showed that there was higher overall survival rate in hepatocellular carcinoma patients with high serum miR-122 levels compared with those with low miR-122 level (P < 0.01). When using Cox regression analysis, high serum miR-122 level was independently associated with better overall survival in hepatocellular carcinoma patients (HR = 0.26; 95 %CI 0.14-0.47, P < 0.01). Subgroup analysis by gender showed that high serum miR-122 level was independently associated with better overall survival in male patients (HR = 0.08; 95 %CI 0.03-0.22, P < 0.01), but not in female patients (HR = 0.48; 95 %CI 0.18-1.32, P = 0.16). Thus, the outcomes in the analysis suggest that high serum miR-122 level is independently associated with higher overall survival rate in hepatocellular carcinoma patients, and it is a good biomarker of better prognosis in patients with hepatocellular carcinoma.