To assess the application value of three-dimensional reconstruction with the multislice spiral computed tomography (MSCT) in preoperative evaluation and predicting the postoperative liver compensatory situation in patients with hepatocellular carcinoma.

Thirty-eight patients with hepatocellular carcinoma (31 males and 83 females; 33 to 83 years of age; 54 years of mean age) who underwent radical hepatectomy in Hepato-biliary and Pancreatic Surgery Department of China-Japan Union Hospital of Jilin University, from February 2010 to February 2011 were enrolled in this prospective study. Local ethical committee approval had been received and that the informed consent of all participating subjects was obtained. Three-dimensional reconstruction was made by MPR, MIP and VR before the operation. The vascular visualization rate of three reconstruction pattern was compared by Chi-square and Fisher exact probability. Rank sum test was used to compare the difference between the predictive volume and actual volume of the liver. The relation between the percentage of remnant liver and the postoperative liver function of Child-Pugh grade was analyzed by one-way ANOVA and Student-Newman-Keuls q test.

The abdominal aorta, celiac trunk, common hepatic artery, proper hepatic artery, portal system and branches of the 38 patients were clearly visualized by MSCT images. Different characteristics of tumors and vascular appearance were observed in the three patterns. The smallest visualized branches of hepatic artery and portal vein were both grade Ⅵ. The visualization rate of grade Ⅲ hepatic artery were 8%(3/38) by MPR, 100%(38/38) by MIP, 100%(38/38) by VR respectively. The visualization rate of MIP and VR were significantly higher compared with MPR(P<0.001). The visualization rate of grade Ⅵ hepatic artery by MIP and VR were 16%(6/38), 3%(1/38), The MIP images was significantly better than the VR (P=0.047). The visualization rate of grade Ⅲ portal vein were 5%(2/38) by MRP, 100%(38/38) by MIP, 100%(38/38) by VR. The visualization rate of MIP and VR were significantly higher than that of the MPR(P<0.001). There was no significant difference between the predictive liver volume [116(12-1 002 cm³] and the actual volume [125(8-987) cm³] by MSCT before operation (Z=-0.099, P=0.921). There was no significant difference between expected resection volume [215(36-1 294)cm³] and actual resection volume [212(40-1 234)cm³] was not statistically significant (Z=-0.31, P=0.975). The percentage of remnant liver for Child-Pugh grade A patients was (83.1±6.3)%, for Child-Pugh grade B was (61.7±5.4)% and for Child-Pugh grade C was (38.9±2.3)%. There was significant difference among the three grades (F=97.425, P<0.001). The differences of each other among the three grades were statistically significant (all in P<0.001).

The liver lesion and vascular can be clearly visualized by three-dimensional reconstruction with MSCT preoperatively and the tumor volume and expected resection liver volume can also be accurately measured. The technique may be helpful for preoperative evaluation and predicting the postoperative liver compensatory function. MPR three-dimensional reconstruction is useful in making operation plans, while the MIP and VR techniques are helpful for specific operation design.