摘要目的:目前对于ICU老年患者的营养实施缺乏指南推荐意见，主要原因在于此领域内研究证据稀少。本研究旨在探究早期蛋白供给量对ICU老年患者临床预后的影响。方法:本研究是一项多中心、整群随机对照试验（NEED研究）的事后分析，NEED研究旨在评估喂养流程对ICU患者营养实施和预后的影响。本研究计划纳入NEED研究中年龄≥70岁的高龄患者，入组后第3天仍未开始营养治疗、住ICU时长不足7 d以及主要结局指标缺失的患者被排除。本研究的主要结局指标是入组后28 d内病死率。根据早期蛋白供给量的三分位数，将患者分为Q1[<0.6 g/(kg·d）]，Q2[0.6~0.83 g/(kg·d）]，Q3[≥0.83g/(kg·d）]3组。使用对数秩检验（log-rank test）来比较28 d病死率的Kaplan-Meier生存曲线。通过Cox比例风险模型检测不同蛋白分组对患者28 d病死率的影响。同时依据是否存在高营养风险（mNUTRIC ≥5分）以及是否合并急性肾功能损伤，对研究人群进行亚组分析。结果:本研究共纳入了789名高龄（≥70岁）患者，患者入ICU后第3~7 d内的平均蛋白量为0.69 (0.53, 0.91) g/(kg·d），Q1低蛋白组、Q2中等蛋白组、Q3高蛋白组的平均蛋白量分别为0.46（0.36, 0.53），0.69（0.63, 0.76）以及1.03（0.91, 1.23）g/(kg·d）。结果显示，相比于高蛋白组，中等蛋白组与更低的28 d病死率相关，通过Cox多因素回归分析对可能的混杂因素进行控制后，中等蛋白组与更低的28 d病死率之间的相关性仍然成立。在高营养风险亚组中（mNUTRIC ≥5分），中等蛋白组与更低的28 d病死率之间显著相关。结论:通过本项大样本量的事后分析发现早期高蛋白供给对高龄ICU患者并无益处，相比于高蛋白组，中等量蛋白供给与更低的28 d病死率相关。本研究可为ICU老年患者早期蛋白供给最佳剂量提供理论依据，以及为临床实施提供参考。

abstractsObjective:To investigate the effect of early protein supplementation on the clinical outcomes of elderly ICU patients with critically ill.Methods:The study was a post-hoc analysis of a multicenter, cluster randomized controlled trial (NEED trial), which aimed to evaluate the impact of feeding protocol on nutritional implementation and outcomes in ICU patients. It was planned to include elderly patients aged ≥70 years from the NEED trial, and patients who had not started nutritional therapy by the Day 3 after enrolment, stayed in the ICU less than 7 days, missing the primary outcome were excluded. The primary outcome of this study was 28-day mortality of enrolment. Patients were categorized into Q1 (<0.6 g/kg/d), Q2 (0.6-0.83 g/kg/d), and Q3 (≥0.83 g/kg/d) groups according to the tertiles of protein supply. The log-rank test was used to compare the Kaplan-Meier survival curves for 28-day mortality. The associations between different protein groups and 28-day mortality were tested by Cox proportional hazards regression models. Subgroup analysis was conducted in patients with high (mNUTRIC score≥5) nutritional risk or patients with baseline acute kidney injury.Results:A total of 789 elderly (≥70 years) patients was included in the study, with a mean protein amount of 0.69 (0.53, 0.91) g/(kg·d) during days 3-7 after ICU admission, and mean protein amounts in the Q1 low-protein group, the Q2 medium-protein group, and the Q3 high-protein group were 0.46 (0.36, 0.53), 0.69 (0.63, 0.76), and 1.03 (0.91, 1.23) g/(kg·d), respectively. The results showed that the medium protein group associated with lower 28-day mortality compared to the high protein group, and the association between the medium protein group and lower 28-day mortality still held after controlling for possible confounders by Cox multivariate regression analysis. In the high-nutritional risk subgroup (mNUTRIC≥5), a significant association was also found between the medium protein group and lower 28-day mortality.Conclusions:Early high protein supply are not beneficial for elderly ICU patients by this large sample size post-hoc analysis, and medium protein supply associate with lower 28-day mortality compared with the high protein group. This study may provide a theoretical basis for the optimal dose of early protein supply in elderly ICU patients, as well as a reference for clinical implementation.