中文摘要：

目的 评价肠外营养（PN）中添加ω-3多不饱和脂肪酸对脓毒症患者预后的影响.方法 通过计算机检索MEDLINE、EMBASE、Cochrane Library、中国知网CNKI和万方生物医学期刊全文数据库（1996年1月至2013年6月）,全面收集有关PN中添加ω-3多不饱和脂肪酸治疗脓毒症患者的随机对照临床试验（RCT）文章,并进行质量评估.治疗组为PN中添加ω-3多不饱和脂肪酸,对照组为常规PN治疗.应用RevMan 5.0软件进行统计学分析.结果 共纳入12篇RCT、721例患者.结果显示：PN中添加ω-3多不饱和脂肪酸可以降低脓毒症患者28 d病死率[相对危险度（RR）=0.77,95％可信区间（95％CI）为0.59～ 0.99,P=0.04],缩短重症监护病房（ICU）住院时间[加权均数差（WMD） =-3.10,95％CI为-5.98 ～-0.21,P=0.04]及总住院时间（WMD=-3.12,95％CI为-4.65 ～-1.60,P＜0.000 1）,但不能缩短机械通气时间（WMD=1.33,95％CI为-5.09～7.75,P=0.69）.结论 本荟萃分析（Meta分析）显示,PN中添加ω-3多不饱和脂肪酸能改善脓毒症患者临床预后,但由于纳入文献质量大多较低,须谨慎看待此Meta分析结果.

英文摘要：

Objective To determine whether the supplementation of ω-3 polyunsaturated fatty acids in parenteral nutrition （PN） confers treatment benefits to outcomes of septic patients. Methods The databases of MEDLINE, EMBASE, Cochrane central register of controlled trials, the CNKI and the Wanfang biomedicine were searched to identify prospective randomized controlled trials （RCTs） which compared the ω-3 polyunsaturated fatty acids supplemented with the standard PN on outcomes of adult patients with sepsis from January 1996 to June 2013. The methodological quality of the included studies was evaluated, and the Cochrane Collaboration RevMan 5.0 was used for data analysis. Results A total of 12 studies enrolling 721 patients were included. Significant reduction in 28-day mortality [relative risk （RR） 0.77, 95% confidence interval （95%CI） 0.59 to 0.99, P=0.043, short intensive care unit [ICU, weighted mean difference （WMD）=-3.10, 95%CI-5.98 to -0.21, P=0.043 and hospital length of stay （WMD =-3.12, 95% CI -4.65 to - 1.60, P 〈 0.000 1 ） were observed in patients receiving ω-3 polyunsaturated fatty acids supplemented PN. There was no differences in days of mechanical ventilation between patients with or without adding ω-3 polyunsaturated fatty acids in PN （WMD =1.33, 95% CI-5.09 to 7.75, P=0.69）. Conclusions Meta-analysis results demonstrated that PN supplemented with ω-3 polyunsaturated fatty acids was beneficial in improving the outcomes of patients with sepsis. However, this conclusion must be interpreted with caution due to the low quality of the enrolled trials.