Background Liver transplantation has become one of the best treatment options for early hepatocellular carcinoma in cirrhosis. The purpose of this study was to correlate pathologic findings with outcomes after liver transplantation for hepatocellular carcinoma and to evaluate associations between various tumor characteristics and patient outcomes. Study Design One hundred patients with hepatocellular carcinoma underwent liver transplantation at our center during an 8-year period. Fifty-five patients received deceased donor livers and the remaining 45 received live donor grafts. A systematic literature search was performed. Using the truncated product method and Tippet’s method p values from the literature were combined with our own results, for a total of 3,107 patients. Results For survival, significant associations were found for tumor grading in our series. Tumor recurrence was significantly associated with α-fetoprotein levels, tumor grading, vascular invasion, Milan and University of California San Francisco criteria in univariate analysis, and with tumor grading and University of California San Francisco criteria by multivariable analysis. Meta-analysis showed significant associations for all variables evaluated (α-fetoprotein, tumor number, tumor size, lobar distribution, vascular invasion, tumor differentiation, Milan criteria, and University of California San Francisco criteria) both for survival and recurrence in the performed univariate truncated product method analyses. Interestingly, metaanalysis of the available multivariable studies showed no significant associations between Milan/University of California San Francisco criteria and survival or tumor recurrence. Conclusions Although results of this study add some corroborative evidence in this special field of transplantation for malignancy, they also open the discussion about the prognostic power of the current listing criteria.