三种术前营养评价方法对消化道恶性肿瘤病人术后并发症预测价值对比

Comparison of the Predictive Value of Three Preoperative Nutritional Evaluation Methods for Postoperative Complications in Patients with Gastrointestinal Malignancies

摘要: 目的：对比欧洲营养风险筛查2002 (NRS2002)、主观整体判断法(SGA)与微型营养判断法(MNA)三种手术之前营养判断手段对于消化道恶性癌症病人手术以后并发症之预测价值。方法：前瞻性纳入2019年2月至2020年5月我院普通外科及肿瘤外科接收之300例消化道恶性癌症病人，其中结直肠肿瘤158例，胃部肿瘤106例，食管肿瘤36例。分别地运用NRS2002、SGA3以及MNA三种营养判断手段作手术之前营养判断，分别地对比此三项手段筛选出之具有营养不良(即营养风险)病人和营养正常(没有营养风险)病人手术以后并发症出现比例。结果：300例消化道肿瘤病人手术之前按照NRS2002评分，存在营养风险者123病人(41.0%)，没有营养风险病人177例(59.0%)，差异无统计学意义(P > 0.05)。根据MNA评分，营养不良病人89例(29.7%)，潜在之营养不良病人92例(30.7%)，营养正常病人119例(39.7%)，差异有统计学意义(P < 0.05)。根据SGA评分，重度之营养不良病人53例(17.6%)，中度之营养不良病人112例(37.3%)，没有营养不良病人135例(45.0%)，差异有统计学意义(P < 0.01)。多因素逻辑回归研究表明，SGA评分与MNA评分都是手术以后并发症之独立预测原因(都P < 0.01)。SGA评分预测手术以后并发症之敏感性比MNA评分的高(91.2% vs 79.8%)，特异性接近(50.0% vs 51.0%)。结论：SGA评分与MNA评分都可以有效地预测消化道恶性癌症病人手术以后并发症之出现状况；然而对比之下，SGA之预测敏感性更加的高。手术之前拟定营养支持措施的时候应该重点地参考SGA判断结果。

Abstract: Objective: To compare the predictive value of nutritional risk screening 2002 (NRS2002), subjec-tive global assessment (SGA) and mini nutritional assessment (MNA) in predicting postoperative complications in patients with gastrointestinal malignant cancer. Methods: From February 2019 to May 2020, 300 patients with gastrointestinal malignant cancer admitted to our hospital were prospectively enrolled, including 158 cases of colorectal cancer, 106 cases of gastric cancer and 36 cases of esophageal cancer. NRS2002, SGA3 and MNA were used to determine the nutritional status before operation, and the incidence of postoperative complications in patients with malnutrition (i.e. nutritional risk) and patients with normal nutrition (without nutritional risk) were compared. Results: According to NRS2002 score, 123 patients (41.0%) had nutritional risk, 177 (59.0%) had no nutritional risk, and the difference was not statistically significant (P > 0.05). According to MNA score, there were 89 malnutrition patients (29.7%), 92 potential malnutrition patients (30.7%) and 119 normal nutrition patients (39.7%). The difference was statistically significant (P < 0.05). According to SGA score, there were 53 cases (17.6%) with severe malnutrition, 112 cases (37.3%) with moderate malnutrition and 135 cases (45.0%) without malnutrition. The difference was statistically significant (P < 0.01). Multivariate logistic regression analysis showed that SGA score and MNA score were independent predictors of postoperative complications (both P < 0.01). The sensitivity of SGA score in predicting postoperative complications was higher than that of MNA score (91.2% vs 79.8%), and the specificity was similar (50.0% vs 51.0%). Conclusion: Both SGA score and MNA score can effectively predict the incidence of postoperative complications in patients with gastrointestinal malignant cancer; however, the sensitivity of SGA is higher than that of MNA score. The results of SGA should be taken into consideration when formulating nutritional support measures before operation.