目的

研究重症肺炎（SP）患者肠道微生物群落的宏基因组学及其对抗菌药物疗效的预测价值。

方法

回顾性选择2022年3月至2024年3月于石家庄市人民医院就诊的126例SP患者为研究对象，根据其使用抗菌药物后的疗效分为有效组（84例）和无效组（42例）。收集两组患者的粪便样本并进行DNA提取，随后进行16S rRNA基因V4区域的测序分析，将测序分析结果纳入宏基因组学研究。采用独立样本t检验、主坐标分析（PCoA）、非参数协方差分析（PERMANOVA）和相似性分析（ANOSIM）分析两组患者肠道菌群丰度、微生物网络和基因功能等差异。

结果

两组患者年龄、性别、BMI、饮酒史、吸烟史、并发症、是否合并基础疾病和基础疾病用药差异均无统计学意义（P均＞0.05）。稀释性曲线和物种累积曲线显示，该体系测序数量充足，测序深度合理。α多样性分析结果显示，两组间Shannon指数和Simpson指数差异均有统计学意义（t = 3.958、P = 0.025，t = 6.583、P ＜0.001）；β多样性分析结果显示两组样本菌群显著分离（F = 6.665、P ＜ 0.001）。两组样本在门层级水平上共检测出18个菌门，主要为厚壁菌门、变形菌门、拟杆菌门、放线菌门和软壁菌门等；在属层级水平上，共检测出952个菌属，其中有8个菌属的相对丰度在3%以上，两组患者的优势菌种不一致。丰度比较结果显示：在门水平上，有效组厚壁菌门和变形菌门相对丰度低于无效组（t = 14.889、P ＜ 0.001，t = 2.508、P = 0.013），有效组拟杆菌门、放线菌门和软壁菌门相对丰度高于无效组（t =17.776、P ＜ 0.001，t = 16.802、P ＜ 0.001，t = 1.997，P = 0.048）；在属水平上，有效组肠球菌属、链球菌属、克雷伯菌属和瘤胃球菌属相对丰度高于无效组（t = 16.663、P ＜ 0.001；t = 6.313、P ＜0.001；t = 9.826、P ＜ 0.001；t = 13.158、P ＜ 0.001），有效组双歧杆菌属、大肠埃希菌属和梭菌属相对丰度低于无效组（t = 16.642、P ＜ 0.001，t = 18.814、P ＜ 0.001，t = 12.762，P ＜ 0.001）。网络分析显示，有效组微生物网络节点和边的数量显著高于无效组，微生物网络结构比无效组更加复杂。基因功能富集分析显示，有效组相对丰度较高的基因功能是ATP结合、IMP生物合成过程和氯酸盐生物合成过程；无效组相对丰度较高的基因功能是胞质、核糖体的结构成分等。通路富集分析显示，有效组相对丰度较高的信号通路为淀粉降解、蔗糖降解和异亮氨酸生物合成；无效组相对丰度较高的信号通路为异戊二烯生物合成和甲基赤四醇磷酸途径。

结论

抗菌药物治疗有效的SP患者样本丰度较高的菌群为肠球菌属、链球菌属和克雷伯菌属等；抗菌药物治疗无效的SP患者样本丰度较高的菌群为肠球菌属、大肠埃希菌属和双歧杆菌属等，肠道微生物群落的宏基因组学对抗菌药物的疗效具有一定预测价值。

Objective

To investigate the metagenomics of gut microbiota in patients with severe pneumonia （SP） and the predictive value for the efficacy of antimicrobial agents.

Methods

Total of 126 patients with SP treated in Shijiazhuang People's Hospital from March 2022 to March 2024 were selected as the research objects， retrospectively.According to the efficacy of antibiotics， 126 cases were divided into effective group （84 cases） and ineffective group （42 cases）.Stool samples were collected from both groups for DNA extraction， followed by sequencing analysis of the V4 region of the 16S rRNA gene.The results of sequencing analysis were included in the metagenomics study.The abundance， microbial network and gene function of the patients were analyzed by independent sample t test， principal coordinate analysis（PCoA）， non-parametric analysis of covariance （PERMANOVA） and analysis of similarity （ANOSIM）.

Results

There were no significant differences in age， gender， BMI， drinking history， smoking history，complications， comorbidities and medication for underlying diseases between the two groups （all P ＞ 0.05）.The dilution curve and species accumulation curve showed that the sequencing amount of this system was sufficient and the sequencing depth was reasonable.The results of α diversity analysis showed that there were significant differences in Shannon index and Simpson index between the two groups （t = 3.958， P = 0.025；t = 6.583， P ＜ 0.001）.β diversity analysis showed that the two groups of samples were significantly separated（F = 6.665， P ＜ 0.001）.Total of 18 phyla were detected in the two groups of samples， mainly including Firmicutes， Proteobacteria， Bacteroidetes， Actinobacteria and Tenericutes.At the genus level， a total of 952 bacterial genera were detected， of which 8 genera had a relative abundance above 3%.The dominant species of the two groups were not consistent.The abundance comparison showed that： at the phylum level，the relative abundance of Firmicutes and Proteobacteria in the effective group was lower than that of the ineffective group （t = 14.889， P ＜ 0.001； t = 2.508， P = 0.013）， and the relative abundance of Bacteroidetes，Actinobacteria and Tenericutes in the effective group was higher than that in the ineffective group （t =17.776， P ＜ 0.001； t = 16.802， P ＜ 0.001； t = 1.997， P = 0.048）.At the genus level， the relative abundance of Enterococcus， Streptococcus， Klebsiella and Ruminococcus in the effective group was higher than that of the ineffective group （t = 16.663， P ＜ 0.001； t = 6.313， P ＜ 0.001； t = 9.826， P ＜ 0.001； t = 13.158， P ＜ 0.001）.The relative abundance of Bifidobacterium， Escherichia coli and Clostridium in effective group was lower than that of the ineffective group （t = 16.642， P ＜ 0.001； t = 18.814， P ＜ 0.001； t = 12.762， P ＜ 0.001）.Network analysis showed that the number of nodes and edges of the effective group was significantly higher than that of the ineffective group， and the structure of the microbial network was more complex than that of the ineffective group.Gene function enrichment analysis showed that the gene functions with higher relative abundance in the effective group were ATP binding， IMP biosynthesis process and chlorate biosynthesis process.The functions of genes with higher relative abundance in the ineffective group were cytoplasm and structural components of ribosomes.Pathway enrichment analysis showed that the signaling pathways with high relative abundance in the effective group were starch degradation， sucrose degradation and isoleucine biosynthesis.The signaling pathways with high relative abundance in the ineffective group were isoprene biosynthesis and methylerythritol phosphate pathway.

Conclusions

Enterococcus， Streptococcus and Klebsiella were the most abundant bacteria in samples of patients with SP effectively treated with antibiotics.Enterococcus， Escherichia coli and Bifidobacterium were the most abundant bacteria in samples of patients with SP ineffectively treated with antibiotics， and the metagenomics of intestinal microbial community has a certain predictive value for the efficacy of antibiotics.