Introduction: Microvascular invasion {MVl) has been demonstrated to be a strong predictor of tumor recurrence  and poor survival after liver transplantation (LT) and liver resection for HCC.l,2 As MVI cannot be determined preoperatively, it is, therefore, of great importance  to try to identify predictors of MVI prior to LT. Methods: A retrospective analysis of preoperative and pathological data of 79 consecutive patients who underwent living donor liver transplantation (LDLT) between 2002 and 2009 for HCC was conducted. MVI was defined as pathological evidence of microscsopic  involvement of the vessels (portal vein or hepatic vein) within the peritumoralliver tissue. The chi-square test and Student t test were used for univariate  analysis. Overall survival  and disease-free survival rates were analyzed using Kaplan-Meier estimates. Results: Patients were divided into two groups. Group I had no MVI and included 55 (70.6%) patients and Group II had MVI and included 24 (30.4%) patients. Recurrence in group II (MVI group) was significantly higher than in group I (25% Vs 4%, P = 0.008).Among the preoperative factors, the tumors beyond Milan criteria, number, size and tumor grade were significant predictors of MVI. MVI  was 6.7% in well differentiated HCC in comparison to 46.8%  in moderately and poorly differentiated HCC, respectively (P=O.002) and was 26% versus 83.3% when tumor number was less than 3 and more than 3, respectively (P=0.009).MVIwas 25.6% in tumors less than 5 em and 71.4% in tumors more than 5 em in size (P=0.02). HCC within Milan criteria  had statistically  significant  lower incidence  of MVI than those beyond Milan criteria (P=0.004). Conclusion: Microvascular invasion associated with higher HCC recurrence rate. Tumor grade, number and size were useful in predicting MVI before LT for HCC. LT for patients within Milan criteria is  associated with   lower   incidence of  pathologically evident MVI.