目的 基于ESPEN标准，探讨营养风险评估表（NRS⁃2002）、营养不良通用筛查表（MUST）、患者主观整体评估表（PG⁃SGA）、控制营养状态评分（CONUT）评估肝细胞肝癌患者术前营养状态并进一步探讨与预后相关性。方法 选取我院在2014年1月至2017年1月期间就诊的245例（hepatocellular carcinoma，HCC）患者。所有患者营养筛查于入院24 h 内进行。收集NRS⁃2002、MUST、PG⁃SGA、CONUT 分数。结果 NRS⁃2002、MUST、PG⁃SGA、CONUT与ESPEN标准的Kappa值分别为：0.339、0.426、0.483、0.483。多因素Cox比例风险回归显示：BMI、肝硬化（是）、肿瘤分期（Ⅱ⁃Ⅲ）、PG⁃SGA（营养不良）、CONUT（营养不良）是HCC患者复发的独立危险因素（P＜0.05）；肝硬化（是）、PG⁃SGA（营养不良）、CONUT（营养不良）是HCC患者死亡的独立危险因素（P＜0.05）。结论 PG⁃SGA、CONUT均可以作为HCC患者术前营养筛查工具并与术后复发及死亡存在联系。但是PG⁃SGA临床运用繁琐及CONUT的一致性较高，推荐采用CONUT进行营养筛查工具。

Objective To evaluate the nutritional risk assessment form（NRS⁃2002），the malnutrition universal screening form（MUST），the patient subjective overall assessment form（PG⁃SGA），and the controlled nutritional status score（CONUT）to evaluate liver based on ESPEN standards. The preoperative nutritional status of patients with hepatocellular carcinoma and further explore its correlation with prognosis. Methods A total of 245 patients with HCC who were treated in our hospital from January 2014 to January 2017 were selected. All patients were screened for nutrition within 24 hours of admission. Collect NRS⁃2002，MUST，PG⁃SGA，and CONUT scores. Results The Kappa values of NRS⁃2002，MUST，PG⁃SGA，CONUT and ESPEN standards are：0.339，0.426，0.483，0.483. Multivariate Cox proportional hazard regression showed that BMI，liver cirrhosis（yes），tumor
staging（Ⅱ⁃Ⅲ），ESPEN standard（malnutrition），PG⁃SGA（malnutrition），and CONUT（malnutrition）were in⁃dependent risk factors for recurrence in HCC patients（P < 0.05）Cirrhosis（yes），ESPEN standard（malnutrition），PG⁃SGA（malnutrition），and CONUT（malnutrition）were independent risk factors for death in patients with HCC（P < 0.05）. Conclusion Both PG⁃SGA and CONUT can be used as preoperative nutrition screening tools for HCC patients，and they are related to postoperative recurrence and death. However，the clinical application of pg⁃sga is complicated and the consistence of conut is high. It is recommended to use conut as a nutritional screening tool.