摘要目的:探讨合并血流动力学不稳定的重症患者早期开始肠内营养耐受性及其对临床预后的影响。方法:2014年5月至2016年5月在浙江省人民医院连续招募预期在重症医学科（ICU）住院时间超过48 h，且血流动力学不稳定的重症患者。前瞻性采集患者临床、实验室和生存数据，并评估急性胃肠损伤（acute gastrointestinal injury，AGI）分级。患者接受血管活性药物支持，维持血流动力学稳定（平均动脉压≥65 mmHg）6 h后开始尝试肠内营养（enteral nutrition,EN），并按照EN开始时机分为3组：早期EN组（EN开始<48 h）、晚期EN组（EN开始≥48 h）和7 d未喂养组。结果:201例患者入选，平均年龄和APACHE Ⅱ评分分别为（65.3±16.4）岁和（22.4±6.85）岁；其中，191例（95.0%）患者接受机械通气治疗。早期EN组与晚期EN组患者胃储留、腹泻和消化道出血症状差异无统计学意义（ P>0.05）,而7 d未喂养组患者胃储留发生率较少(16.7% vs.33.3%， P=0.05)，但消化道出血风险增加（47.2% vs.26.1%， P=0.02）。与晚期EN组和7 d未喂养组相比，早期EN组患者28 d（30.4% vs. 47.9% vs..55.6%， P=0.01）和60 d病死率（38.0% vs. 53.4% vs 63.9%， P=0.017）显著降低。进一步多因素Cox回归分析显示：EN喂养时机（早期 vs晚期： χ 2≥5.83， P≤0.03；早期 vs 7 d未喂养： χ 2≥7.90， P<0.01）、血肌酐（ χ 2≥5.06， P≤0.02）、白蛋白水平（ χ 2≥6.41， P<0.05）、AGI分级（ χ 2≥8.15， P<0.01）和APACHE Ⅱ评分（ χ 2≥9.62， P≤0.001）分别是预测血流动力学不稳定的重症患者28 d和60 d死亡的独立危险因素。 结论:接受稳定剂量血管活性药物维持血流动力学稳定的重症患者可耐受早期EN，且早期启动EN与其28 d和60 d病死率减少显著相关。

abstractsobjective:To investigate the tolerability of early enteral nutrition (EN), and to further explore the association of early EN with clinical outcome in critically ill patients with hemodynamic instability.Methods:The adult patients from Zhejiang Provincial People’s Hospital with an expected admission to ICU for at least 24 h were consecutively recruited from May 2014 to May 2016, and all clinical, laboratory, and survival data were prospectively collected. The AGI grade was daily assessed on the first week of ICU admission. Enteral nutrition (EN) started after 6 h of hemodynamic stability (MAP ≥ 65 mmHg) when the patients took vasoactive medication. The patients were divided into three groups based on the timing of EN initiation: early EN group (EN initiation within 48 h of ICU admission), late EN group (EN initiation at more than 48 h of ICU admission), and no initiation of enteral feeding within 7 days of ICU admission.Results:Of 201 patients enrolled, the mean age was 65.3 ± 16.4 years, APACHE II score was 22.4 ± 6.85, and 191 patients (95.0%) took mechanical ventilation. There were no differences in high gastric residual volume, diarrhea, and gastrointestinal (GI) bleeding between the early EN group and late EN group ( P>0.05). Whereas, patients in the no initiation of EN within 7 days of ICU admission had a lower prevalence of gastric residual volume (16.7% vs. 33.3%, P=0.05), but higher prevalence of GI bleeding (47.2% vs. 26.1%, P=0.02). Compared with those in the late EN group and in no initiation of EN within 7 days of ICU admission, patients in the early EN group had lower 28- (30.4% vs. 47.9% vs. 55.6%, P=0.01) and 60-day mortality rates (38.0% vs. 53.4% vs. 63.9%, P=0.017). Multivariate Cox regression analysis showed that the timing of EN initiation on the admission to ICU (early EN vs. late EN, χ 2≥5.83, P<0.05; early EN vs. no initiation of EN, χ 2≥7.90, P<0.01), serum creatinine ( χ 2=5.06, P<0.05), plasma albumin ( χ 2≥6.41, P<0.01), AGI grade ( χ 2≥8.15, P<0.01), and APACHE II score ( χ 2≥9.62, P<0.01) were independent predictors for 28- and 60-day mortality. Conclusions:Early EN on admission to ICU could be tolerated, and is significantly associated with lower risk of 28- and 60-day mortality in critically ill patients with vasoactive medication to maintain hemodynamic stability.