探讨三维手术规划系统在原发性肝癌（肝癌）切除术前评估中的应用价值。

本前瞻性研究对象为2012年6月至2013年6月在兰州大学第二医院普通外科收治的44例肝癌患者。所有患者均签署知情同意书，符合医学伦理学规定。其中男32例，女12例，平均年龄（60±12）岁。按照随机数字表法将患者随机分为计算机体层摄影血管造影术（CTA）组和三维手术规划组，并根据肿瘤大小、累及范围及既往手术史将两组进一步分为复杂肝癌亚组和非复杂肝癌亚组。CTA组的复杂肝癌亚组8例，非复杂肝癌亚组14例；三维手术规划组的复杂肝癌亚组6例，非复杂肝癌亚组16例。CTA组采用CTA对肝癌切除术进行术前评估。三维手术规划组采用肝脏三维手术规划系统对肝癌切除术进行术前评估。以手术所见作为金标准，观察CTA与三维手术规划系统对肝癌、肝癌与周围组织毗邻关系的显示情况，以及对12条腹腔血管、肝动脉变异、肿瘤侵犯血管、胆管扩张的显示率。分析三维手术规划系统预测的肿瘤体积与实际切除肿瘤质量的关系。非正态分布资料采用M（Q₂₅，Q₇₅）表示，两种检查方法的显示率比较采用χ²检验或Fisher确切概率法。三维手术规划系统预测的肿瘤体积与实际切除肿瘤质量的关系分析采用Spearman秩相关分析。

两种方法均能显示肝癌病灶，三维手术规划系统还能清晰显示肝癌与周围组织的毗邻关系。在复杂肝癌患者中，CTA组对腹腔血管的显示率为81%（78/96），三维手术规划组为100%（72/72），差异有统计学意义（χ²=15.12，P<0.05）。CTA组对胆管扩张均未能显示，三维手术规划组对胆管扩张的显示率为3/6。在非复杂肝癌患者中，CTA组对腹腔血管的显示率为90%（151/168），三维手术规划组为100%（192/192），差异有统计学意义（χ²=20.39，P<0.05）。CTA组对胆管扩张均未能显示，三维手术规划组对胆管扩张的显示率为19%（3/16）。在复杂肝癌患者中，三维手术规划系统预测肿瘤体积的中位数为218（129，429）ml，实际切除肿瘤质量为194（112，429）g，两者呈正相关（r =0.943，P<0.05），平均误差率为6.5%。在非复杂肝癌患者中，三维手术规划系统预测肿瘤体积为368（89，560）ml，实际切除肿瘤质量为395（126，578）g，两者呈正相关（r =0.958，P<0.05），平均误差率为6.3%。

与CTA相比，三维手术规划系统能更好地显示肿瘤与周围组织的毗邻关系、腹腔血管以及胆管扩张情况，并能较准确地预测肿瘤切除体积，尤其适用于复杂肝癌患者。

To investigate the application value of three-dimensional surgery planning system in the preoperative evaluation of primary liver cancer (PLC) resection.

A total of 44 patients with PLC [32 males and 12 females, mean age of (60±12) years old] in Department of General Surgery, Lanzhou University Second Hospital from June 2012 to June 2013 were enrolled in this prospective study. The informed consents of all patients were obtained and the ethics committee approval was received. According to random number table method, the patients were randomly divided into 2 groups: computed tomography angiography (CTA) group and three-dimensional surgery planning (3D) group. Subgroups of complex PLC and non-complex PLC were further defined in each group according to tumor size, tumor invasive extent and history of surgery. In CTA group, 8 cases was assigned in complex PLC subgroup and 14 cases in non-complex PLC subgroup, and 6, 16 cases respectively in 3D group. CTA was used in the preoperative evaluation of PLC resection in CTA group. Liver three-dimensional surgery planning system was used in the preoperative evaluation of PLC resection in 3D group. The intra-operative finding was taken as a gold standard. The visualization of PLC, the adjacent relationship between PLC and peripheral tissues, the display rates of 12 abdominal vessels, variation of hepatic artery, vascular invasion of tumor, cholangiectasis by CTA and three-dimensional surgery planning system were observed. The relationship between estimated tumor volume by three-dimensional surgery planning system and actual weights of resected tumor was analyzed. Non-normal distribution data were expressed in M(Q₂₅,Q₇₅). The display rates by 2 methods were compared using Chi-square or Fisher's exact probability test. The relationship between estimated tumor volume by three-dimensional surgery planning system and actual weights of resected tumor was analyzed using Spearman rank correlation analysis.

PLC lesions could be both visualized by 2 methods. The adjacent relationship between PLC and peripheral tissues could also be clearly visualized by three-dimensional surgery planning system. For patients with complex PLC, the display rate of abdominal vessels was 81% (78/96) in CTA group, and was 100% (72/72) in 3D group, where significant difference was observed (χ²=15.1, P<0.05). Cholangiectasis could not be visualized in CTA group and the display rate of cholangiectasis was 3/6 in 3D group. For patients with non-complex PLC, the display rate of abdominal vessels was 90% (151/168) in CTA group, and was 100% (192/192) in 3D group, where significant difference was observed (χ²=20.39, P<0.05). Cholangiectasis could not be visualized in CTA group and the display rate of cholangiectasis was 19% (3/16) in 3D group. For patients with complex PLC, the median estimated tumor volume by three-dimensional surgery planning system was 218(129,429)ml and the actual weights of resected tumor was 194(112,429)g, where positive correlation was observed (r =0.943, P<0.05) with an average error rate of 6.5%. For patients with non-complex PLC, the estimated tumor volume by three-dimensional surgery planning system was 368(89,560)ml and the actual weights of resected tumor was 395(126,578)g, where positive correlation was observed (r =0.958, P<0.05) with an average error rate of 6.3%.

Compared with CTA, three-dimensional surgery planning system can better display the adjacent relationship between tumor and peripheral tissues, abdominal vessels, cholangiectasis and estimate the volume of resected tumor more accurately. It is especially suitable for patients with complex liver cancer.