OBJECTIVE To determine the efficacy and adverse effects of different neoadjuvant and adjuvant therapies compared to surgery alone or surgery and placebo/supportive therapy when given to improve relapse and survival rates for operable hepatocellular carcinoma. METHODS Electronic databases, conference proceedings, bibliographies of identified publications. METHODS All truly randomised and quasi-randomised clinical trials that compared hepatocellular carcinoma patients who were given and not given neoadjuvant/adjuvant therapy as a supplement to curative liver resection. METHODS Study data was extracted independently by two reviewers and discrepancies were resolved by consensus. A total of eight randomised controlled clinical trials were identified, totaling 548 randomised patients. Seven of the eight trials reported survival and disease-free survival curves and the results of hypothesis testing (log-rank test). The remaining trial reported only the mean survival times. None reported the hazard ratio and only one did a sample size calculation. The survival and disease-free survival curves were compared using their one, two and three-year survival rates, median survival times and the result of the hypothesis tests. RESULTS The size of the randomised clinical trials ranged from 40 to 115 subjects. Both preoperative (neoadjuvant) and postoperative (adjuvant), systemic and locoregional (+/- embolization), chemo- and immunotherapy interventions were tested. None were comparable in terms of both treatment regimen and participants selected, so no pooling was done. Only one regimen using preoperative transcatheter arterial chemoembolization with doxorubicin was approximately duplicated. Seven of the eight trials reported no survival benefit from adjuvant therapy. Only one trial reported a statistically significant difference for survival and disease-free survival for the treatment arm, but the results of both its arms were very poor when compared to other studies. Two of the trials that did not report any absolute survival advantage reported statistically significant differences in disease-free survival. Five of the eight trials did not perform intention-to-treat analysis. The highest toxicity rate was in a trial using oral 1-hexylcarbamoyl 5-fluorouracil which resulted in 12 out of 38 subjects stopping because of adverse events. CONCLUSIONS There is no evidence for efficacy of any of the adjuvant protocols reviewed. In order to detect a realistic treatment advantage, larger trials will have to be conducted.