红霉素提高成年危重患者肠内营养耐受性的系统评价和Meta分析

Erythromycin for improving enteral nutrition tolerance in adult critical patients： a systematic review and Meta-analysis

目的 系统评价红霉素用于提高成年危重患者肠内营养（EN）耐受性的有效性和安全性.方法 检索美国国立医学图书馆PubMed数据库、荷兰医学文摘EMbase数据库、Cochrane临床试验数据库、中国知网CNKI和万方数据库,查找从建库至2013年6月有关成年危重患者使用红霉素提高EN耐受性或提高EN置管成功率的随机对照临床试验（RCT）.由2位研究人员独立筛选文献、提取资料和评价质量后,采用RevMan 5.2软件进行荟萃分析（Meta分析）.结果 最终纳入16个RCT,其中改善EN耐受性的研究10个,共668例患者;提高胃肠置管成功率的研究6个,共353例患者.Meta分析结果显示：红霉素与安慰剂相比能显著提高鼻肠管幽门后置管成功率[相对危险度（RR）=1.82,95％可信区间（95％CI）为1.40～2.37,P＜0.000 01],而与甲氧氯普胺相比则差异无统计学意义（RR=1.04,95％CI为0.79～ 1.36,P=0.799）.对需要进行早期EN的患者,红霉素组试验5d时胃内喂养成功率较安慰剂或空白对照组明显升高（RR=1.89,95％CI为1.19 ～3.00,P=0.007）.对EN不耐受的患者,红霉素较甲氧氯普胺可显著增加24 h胃内喂养成功率（RR=1.30,95％CI为1.02～ 1.66,P=0.03）、72 h胃内喂养成功率（RR=1.57,95％CI为1.15 ～2.14,P=0.005）、144 h胃内喂养成功率（RR=2.04,95％ CI为1.23～3.37,P=0.006）;红霉素组喂养失败的中位时间晚于甲氧氯普胺组.5个研究进行了不良反应的报告,除了红霉素与甲氧氯普胺联合治疗组腹泻发生率明显高于红霉素单药治疗外,其他差异均无统计学意义.结论 红霉素能显著提高危重患者幽门后营养管置管成功率,可以成为床旁无引导条件下置管前的一种辅助手段.目前的证据支持对EN不耐受的危重患者静脉给予小剂量（3 mg/kg）红霉素治疗.

Objective To systematically review the efficacy and safety of erythromycin on enteral nutrition （EN） tolerance in adult critical care patients.Methods Databases including PubMed,EMbase,the Cochrane Library,CNKI and Wangfang data were retrieved up to June,2013 to collect the randomized controlled trial （RCT） concerning erythromycin in improving EN tolerance or increasing the successful rate of postpyloric EN tube as compared with other treatments.Two reviewers independently screened the literature,extracted the data,and assessed the quality of methodology.Then Meta-analysis was performed using RevMan 5.2 software.Results A total of 16 RCTs were included.Ten RCTs involving 668 patients were included for evaluating erythromycin in improving EN tolerance.Six RCTs involving 353 patients were included for evaluating erythromycin to increase the successful rate of postpyloric EN tube.The result of Meta-analysis showed that compared with placebo,erythromycin could significantly improve the successful rate of postpyloric EN tube placement [relative risk （RR） =1.82,95％ confidence interval （95％CI） 1.40-2.37,P＜0.000 01],while there was no significant difference between erythromycin and metoclopramide （RR=1.04,95％ CI 0.79-1.36,P=0.799）.In patients who needed early EN,compared with placebo or blank control,erythromycin had higher successful gastric feeding rate over 5 days （RR =1.89,95％ CI 1.19-3.00,P=0.007）.In patients who failed EN,compared with metoclopramide,erythromycin could significantly increase the successful gastric EN rate for 24 hours （RR=1.30,95％CI 1.02-1.66,P=0.03）,72 hours （RR=1.57,95％CI 1.15-2.14,P=0.005） and 144 hours （RR=2.04,95％CI 1.23-3.37,P=0.006）.The median time of EN intolerance was postponed in erythromycin group than that in metoclopramide group.Adverse reactions were reported in 5 studies.There was no statistic difference except for the higher diarrhea rate in the combination treatment group compared with erythromycin group.Conclnsions Postpyloric EN tube placement rate can be improved by erythromycin,which could be a choice of substitute for bedside intubation without fluoroscopy or endoscopic assistance.Based on the evidence,we recommended that intravenous erythromycin in a small dose of approximately 3 mg/kg weight as an option for EN intolerance in critical patients.