目的:基于16S核糖体RNA(rRNA)高通量测序分析特重度烧伤患者肠道菌群的动态变化规律。方法:本前瞻性观察性研究纳入2017年2—6月中国科学院大学宁波华美医院收治的符合入选标准的5例男性特重度烧伤患者(年龄32～48岁)，采集其休克期(伤后3 d内)、急性感染早期(伤后4～14 d)、急性感染中期(伤后15～28 d)、急性感染后期(伤后29 d至出院前1周)及出院前1周内的粪便样本，各时期样本数均为5。使用pH计测定粪便样本的pH值，高通量测序技术对粪便样本的16S rRNA V3、V4区进行测序。QIIME分析软件分析肠道菌群操作分类单元(OTU)数目、α多样性(Chao1指数、Shannon指数)和门、科的相对丰度，非加权组平均法聚类分析肠道菌群β多样性，Tax4Fun预测肠道菌群功能变化。对数据行单因素重复测量方差分析、Bonferroni法、配对样本Wilcoxon秩和检验及Bonferroni校正。结果:(1)本组特重度烧伤患者急性感染早、中期粪便pH值分别为7.40±0.45、7.56±0.45，明显高于休克期的6.68±0.36(n P<0.05或n P<0.01)。(2)共获得2 333 584条有效高质量序列，序列长度为415 bp左右，共获得1 209个OTU，所有标本测序覆盖度均达99.0%以上。本组特重度烧伤患者急性感染早、中、后期OTU数目、Chao1指数明显低于休克期(n Z值均为2.023，n P<0.05)；出院前1周内OTU数目、Chao1指数明显高于急性感染早、中、后期，Shannon指数明显高于急性感染早、中期(n Z值均为2.023，n P<0.05)。(3)本组特重度烧伤患者休克期样本菌群结构与出院前1周内相似度高，与急性感染早、中、后期相似度低。各个单独样本的分析显示，大部分样本聚类规律与分期样本一致。休克期肠道菌群加权Unifrac距离明显短于急性感染早、中、后期(n Z＝3.326、2.570、2.690，n P<0.05或n P<0.01)，其他时期肠道菌群加权Unifrac距离相近。(4)门水平上，与休克期比较，急性感染早、中、后期厚壁菌门细菌相对丰度降低，变形菌门和拟杆菌门细菌相对丰度升高；上述3个菌门细菌相对丰度在出院前1周内与休克期相近。伤后不同时期科水平上相对丰度前5的优势菌有较大差异，休克期的前5优势科细菌相对丰度在急性感染早、中、后期降低。休克期的非优势科如肠杆菌科、链球菌科、拟杆菌科细菌相对丰度在急性感染早、中、后期大幅升高，成为这些时期新的优势科；出院前1周内部分产酸菌科细菌相对丰度恢复至接近休克期水平。(5)急性感染早、中期肠道菌群的某些氨基酸代谢、糖酵解和糖异生、丙酮酸代谢等功能较休克期减弱，急性感染后期肠道菌群某些氨基酸和糖类代谢较休克期增强，休克期与出院前1周内肠道菌群功能基因分布相似。n 结论:特重度烧伤患者肠道内环境及菌群结构在急性感染早、中期发生了明显变化，pH值升高，菌群种类及多样性减少，尤其是产酸菌科细菌的相对丰度明显降低，但随着患者治疗结束而恢复。可考虑将粪便pH值和肠道变形菌门及产酸菌科细菌变化作为反映特重度烧伤患者肠道菌群紊乱水平的指标。“,”Objective:To analyze the dynamic change rule of gut microbiota in patients with extremely severe burns using 16S ribosomal RNA (rRNA) high-throughput sequencing technology.Methods:Five patients with extremely severe burns who were admitted to Hwa Mei Hospital of University of Chinese Academy of Sciences from February to June 2017 and conformed to the inclusion criteria were included in the prospective observational study. All patients were males with age of 32-48 years. Fecal samples were collected in the shock stage (within 3 days after injury), early stage of acute infection (4-14 d after injury), middle stage of acute infection (15-28 d after injury), late stage of acute infection (from 29 d after injury to 1 week before discharge) and within 1 week before discharge. The number of samples was 5 in each stage. The fecal pH value was measured using a pH meter. High-throughput sequencing technology was applied for sequencing of 16S rRNA V3 and V4 regions of fecal samples. QIIME software was used to analyze the number of operational taxonomic units (OTUs), α diversity (Chao1 index and Shannon index), and the relative abundance of gut microbiota at the phylum and family levels. Unweighted pair group method with arithmetic mean clustering method was used to analyze the β diversity of gut microbiota, and Tax4Fun was used to predict functional changes of gut microbiota. Data were statistically analyzed with one-way analysis of variance for repeated measurement, Bonferroni method, Wilcoxon rank sum test for paired samples, and Bonferroni correction.Results:(1) The pH value of feces in the early and middle stages of acute infection in patients with extremely severe burns in this group was 7.40±0.45 and 7.56±0.45 respectively, which were significantly higher than 6.68±0.36 in the shock stage (n P<0.05 orn P<0.01). (2) A total of 2 333 584 efficient and high-quality sequences were obtained, and the length of the sequences was about 415 bp. A total of 1 209 OTUs were obtained. The sequencing coverage of all samples was over 99.0%. The number of OTUs and Chao1 index in the early, middle, and late stages of acute infection in patients with extremely severe burns in this group were significantly lower than those in the shock stage (n Z=2.023, n P<0.05). The number of OTUs and Chao1 index within 1 week before discharge were significantly higher than those in the early, middle, and late stages of acute infection, and Shannon index within 1 week before discharge was significantly higher than that in the early and middle stages of acute infection (n Z=2.023, n P<0.05). (3) The structure of gut microbiota in the shock stage in patients with extremely severe burns in this group was highly similar to that within 1 week before discharge, and lowly similar to that in the early, middle, and late stages of acute infection. The analysis of individual sample showed that the clustering rule of most of the samples was in accordance with that of the staged samples. The weighted Unifrac distance of gut microbiota in the shock stage was significantly shorter than that in the early, middle, and late stages of acute infection (n Z=3.326, 2.570, 2.690, n P<0.05 orn P<0.01), while the weighted Unifrac distance of gut microbiota in the other stages was similar. (4) At the phylum level, compared with that in the shock stage, the relative abundance ofn Firmicutes was decreased in the early, middle, and late stages of acute infection, while the relative abundance of n Bacteroidetes and n Proteobacteria increased. However, the relative abundance of the above three phyla within 1 week before discharge was similar to that in the shock stage. At the family level, the top five dominant bacteria in relative abundance in different stages after injury were quite different. The relative abundance of dominant five family bacteria in the shock stage was decreased in the early, middle, and late stages of acute infection. The relative abundance of non-dominant bacteria such as n Enterobacteriaceae, Streptococcaceae, and n Bacteroidaceae in the shock stage increased significantly in the early, middle, and late stages of acute infection, which became new dominant families in these stages. The relative abundance of some acid-producing bacteria within 1 week before discharge resumed to the similar level in the shock stage. (5) Functions such as some amino acid metabolism, glycolysis and gluconeogenesis, and pyruvate metabolism of gut microbiota were obviously weaker in the early and middle stages of acute infection than those in the shock stage. Functions such as some amino acid metabolism and carbohydrate metabolism of gut microbiota were significantly enhanced in the late stage of acute infection compared with that in the shock stage. The distributions of functional genes in gut microbiota were similar between the shock stage and within 1 week before discharge.n Conclusions:The internal environment and gut microbial compositions in extremely severe burned patients change significantly in the early and middle stages of acute infection. The pH value increases, the bacterial species and diversity decrease, especially the relative abundance of acid-produced bacteria is significantly reduced, which gradually recover with the improvement of the patient′s condition. The pH value and the changes of n Proteobacteria and acid-producing bacteria could be considered as suitable parameters for reflecting the disorder level of gut microbiota in patients with extremely severe burns.n