Study On The Relationship Between Nasal And Gut Microbiome And Clinical Events In Patients With Sepsis

Background:Sepsis should be defined as life-threatening organ dysfunction caused by a dysregulated host response to infection,with high morbidity and mortality,and its total mortality is 10%to 52%.In sepsis,It isn’t understanded sufficiently about the relationship between intestinal and nasal microbiome character and the development of the sepsis.This study will further explore relevant issues using the second-generation sequencing technology.Purpose:1.Analyze the characteristics of nasal and gut microbiome in sepsis using microbiology.2.To investigate the relationship between nasal and intestinal microbiome and clinical events.3.Screening of sepsis-related differential microorganisms to establish a microbiome prognosis model for sepsis.Methods:Nasal and fecal specimens were collected from sepsis and healthy volunteers from the ICU department of a hospital.In cross-section,we collected nasal swabs from 89 patients with sepsis,65 patients with non-sepsis and 78 healthy volunteers,and anal swabs from 80 patients with sepsis,14 patients with non-sepsis and 80 healthy volunteers.In longitudinal,we dynamically collected 281 nasal swabs and 235 anal swabs from 26 patients with sepsis.Total DNA was extracted,amplified,and sequenced.we analyze the micro-ecological community structure characteristics of sepsis and healthy volunteers combined with clinical information and QIIME.The random forest model was used to construct the sepsis microbiological index,and to mine the relationship between clinical indicators and microbiome.Results:Chapter One1.There are significant difference of nasal microbiome between patients with sepsis and non-sepsis,healthy people.2.The nasal flora of patients with sepsis were mainly composed of Corynebacterium,Staphylococcus,Acinetobacter and Pseudomonas.There were three nasal bacterial type:Corynebacterium,Staphylococcus,Acinetobacter.,Acinetobacter type was the longest compared with hospitalization time in ICU in this three nasal type,and Staphylococcus type was the shortest.The mortality rate of Acinetobacter type with composition ratio>0.5 or more than 19 hospitalization was even higher compared with other nasal type.3.The disorder of nasal microbiome was closely related to the hospitalization time in ICU,antibiotic use,nasal feeding,and tracheostomy in sepsis.4.The random forest model was a good predictor of death in sepsis,AUC:0.8012.Chapter two1.The difference of intestinal flora between patients with sepsis and healthy people was significant.2.The intestinal flora of patients with sepsis were mainly composed of Enterococcus,Corynebacterium,Bacteroides,Acinetobacter,etc.There were three enterotype:bacteroides,Peptoniphilus,Enterococcus.3.The alpha diversity of sepsis was effected by hormones and platelets(PLT).Machine learning showed that antibiotic time,the hospitalization time in ICU,and oxygenation index affected intestinal microbiome significantly.Chapter three1.There was a close relationship between the nasal flora and intestinal flora of patients with sepsis.The common bacteria of patients with sepsis were greatly increased compared with healthy people.The main pathogens were:A cinetobacter\Pseudomonas\Dyella\Enterococcus\Peptoniphilus et al.2.The nasal and intestinal microbiota of patients with sepsis showed higher levels of Acinetobacter,and different underlying diseases and invasive operations were associated with pathogenic bacteria.3.The bacterial type of the deterioration group changed fast compared with the improvement group in the nasal and intestinal flora of sepsis.And the stability level of nasal microbiota was higher than intestinal microbiota.In the nasal cavity and intestine,Staphylococcus,Bradyrhizobium,Veillonella,Prevotella,Bacteroides were negatively correlated with hospitalization time,and Acinetobacter was correlated with hospitalization time positively.Conclusion:1.The features of nasal and intestinal microbiome of patients with sepsis was unique,and multiple clinical events could influence the microbiota.And the rapid changes and higher instability in bacterial microbiota predict adverse outcomes of sepsis.2.The nasal flora is expected to be used in the microecological diagnosis of sepsis in the future.3.Random forest model is a good predictor of death in patients with sepsis,combining with clinical information and nasal flora.