Objective To explore the clinical efficacy of anatomic liver resection in treatment of hepatocellular carcinoma (HCC) with microvascular invasion (MVI).Methods The retrospective cohort and case-control study was conducted.The clinical data of 150 HCC patients with MVI who were admitted to the Anhui Medical University Affiliated Provincial Hospital from June 2007 to June 2012 were collected.Sixty patients undergoing anatomic liver resection were allocated into the AR group and 90 undergoing non-anatomic liver resection in the NR group.Patients in the AR group underwent anatomic liver resection according to results of preoperative ICG R15 test,and patients in the NR group underwent non-anatomic liver resection.Observation indicators:(1) operation situations:operation time,volume of intraoperative blood loss,number of patients with blood transfusion;(2) postoperative recovery situations:time of drainage-tube removal,duration of hospital stay,Clavein grade of complication within 30 days postoperatively,number of patients with hepatic failure within 30 days postoperatively and number of death within 30 days postoperatively;(3) follow-up:postoperative median survival time,5-year overall survival rate and 5-year tumor-free survival rate;(4) prognostic factors analysis of 150 HCC patients with MVI.Measurement data with normal distribution were represented as-x±s and comparison between groups was analyzed using the independent-sample t test.Count data were represented as the chi-square test or Fisher exact probability.The survival rate was calculated using the Kaplan-Meier method and survival analysis was done using Log-rank test.The univariate analysis and multivariate analysis were done using the COX regression model.Results (1) Operation situations:all the 150 patients received successful radical resection of HCC.Operation time,numbers of patients with volume of intraoperative blood loss ≥ 500 mL and with volume of intraoperative blood loss ＜ 500 mL and number of patients with blood transfusion were (165±39) minutes,12,48,15 in the AR group and (136±30)minutes,34,56,38 in the NR group,respectively,with statistically significant differences between the 2 groups (t =29.172,x2=5.351,4.673,P＜0.05).(2) Postoperative recovery situations:time of drainage-tube removal and duration of hospital stay were (2.7± 1.1)days and (5.2± 1.3) days in the AR group,(3.8±1.6)days and (7.1±2.3)days in the NR group,respectively,with statistically significant differences between the 2 groups (t =4.641,5.812,P＜0.05).Numbers of patients with grade Ⅰ-Ⅱ of Clavein grade and with grade Ⅲ-Ⅳ and number of death within 30 days postoperatively were 45,15,1 in the AR group and 61,29,2 in the NR group,respectively,with no statistically significant difference between the 2 groups (x2 =0.906,P＞0.05).Number of patients with hepatic failure within 30 days postoperatively in the AR and NR group were respectively 4 and 17,with a statistically significant difference (x2=4.467,P＜0.05).(3) Follow-up:all the 150 patients were followed up for 1-106 months,with a median time of 26 months.The postoperative median survival time,5-year overall survival rate and 5-year tumor-free survival rate were 46 months,33.3％,21.7％ in the AR group and 18 months,15.6％,2.2％ in the NR group,respectively,with statistically significant differences in overall survival and tumor-free survival between the 2 groups (x2=23.718,63.932,P＜ 0.05).(4) Prognostic factors analysis of 150 HCC patients with MVI:result of univariate analysis showed that maximum diameter of tumor,tumor capsule,TNM stage,Edmondson grade and surgical procedures were relative factors affecting overall survival and tumor-free survival of HCC patients with MVI,with statistically significant differences (x2=5.519,2.790,13.639,8.321,42.470,31.057,15.963,19.594,23.718,63.932,P＜ 0.05).Result of multivariate analysis showed that missing tumor capsule,stage Ⅲ-Ⅳ of TNM stage,grade Ⅲ-Ⅳ of Edmondson grade and non-anatomic liver resection were independent factors affecting poor overall survival and tumor-free survival of HCC patients with MVI,and maximum diameter of tumor ＞5 cm was an independent factor affecting poor overall survival of HCC patients with MVI,with a statistically significant difference [HR =0.527,0.683,0.333,0.522,0.576,0.514,0.523,0.268,95％ confidence interval (CI):0.355-0.782,0.475-0.983,0.219-0.504,0.361-0.755,0.389-0.852,0.358-0.737,0.342-0.800,0.174-0.413;HR=0.559,95％CI:0.370-0.845,P＜0.05].Conclusions Anatomic liver resection in the treatment of HCC patients with MVI is safe and effective,with good short-term and long-term outcomes,and it can also improve prognosis of patients.Missing tumor capsule,stage Ⅲ-Ⅳ of TNM stage,grade Ⅲ-Ⅳ of Edmondson grade and non-anatomic liver resection are independent factors affecting poor overall survival and tumor-free survival of HCC patients with MVI,and maximum diameter of tumor ＞5 cm is an independent factor affecting poor overall survival of HCC patients with MVI.