Objective:To summarize the experience of surgical methods without repairing the fistula for 92 cases with gastrointestinal intrathoracic fistula.Methods:The surgical methods without repairing the fistula were performed through VATS, small incision assisted with VATS or thoracotomy. The focus of the surgery was to promote lung expansion, eliminate the residual cavity of chest cavity and keep effective drainage. After entering the chest cavity from the affected side, wash chest cavity with a large amount of warm normal saline and sterilize intermittently with iodophor to ensure the sterile environment in the pus cavity. Then completely remove the pleural cellulose or fiberboard on visceral pleura to promote lung expansion, eliminate the residual cavity of the chest cavity. The fistula was covered tightly and supported firmly by the visceral pleura on the lung. Multiple T-tubes were placed in thoracic cavity and fistula to keep effective postoperative drainage.Results:Among 92 cases, 85 cases were cured and the cure rate was 92.4% (85/92).7 cases died and the mortality rate was 7.61% (7/92). The 7 dead cases include 5 cases with esophagogastric anastomotic fistula (the death of 3 cases was cause by aortic esophagogastric fistula, the death of 1 case was cause by thoracic gastric tracheal fistula and 1 case was dead because of pulmonary infection and respiratory failure), 1 case with esophageal rupture (the cause of death was septic shock ), and 1 case with esophageal perforation(the cause of death was pulmonary infection and respiratory failure).Conclusion:Most of the surgeries without repairing gastrointestinal intrathoracic fistula are conducted simply through VATS or small incision assisted with VATS., which is safe and effective.

Objective:To investigate the changes of intestinal microecology in the early stage of sepsis rat model by 16S rDNA sequencing.Methods:Sixty male Sprague-Dawley (SD) rats were randomly divided into cecal ligation and puncture (CLP) group and sham operation group (Sham group), with 30 rats in each group. In the CLP group, sepsis rat model was reproduced by CLP method; the rats in the Sham group only underwent laparotomy without CLP. At 24 hours after the operation, the intestinal feces and serum samples of 8 rats in each group were collected. The survival rate of the rest rats was observed until the 7th day. The level of serum tumor necrosis factor-α (TNF-α) was detected by enzyme-linked immunosorbent assay (ELISA). Intestinal feces were sequenced by 16S rDNA sequencing technology. The operational taxonomic unit (OTU) data obtained after sequence comparison and clustering was used for α diversity and β diversity analysis, principal coordinate analysis and linear discriminant analysis effect size analysis (LEfSe) to observe the changes of intestinal microecology in early sepsis rats and excavate the marker flora.Results:At 24 hours after the reproduction of the model, the rats in the CLP group showed shortness of breath, scattered hair and other manifestations, and the level of serum TNF-α increased significantly as compared with that in the Sham group (ng/L: 43.95±9.05 vs. 11.08±3.27, P < 0.01). On the 7th day after modeling, the cumulative survival rate of the Sham group was 100%, while that of the CLP group was 31.82%. Diversity analysis showed that there was no significant difference in α diversity parameter between the Sham group and the CLP group (number of species: 520.00±52.15 vs. 492.25±86.61, Chao1 richness estimator: 707.25±65.69 vs. 668.93±96.50, Shannon index: 5.74±0.42 vs. 5.79±0.91, Simpson index: 0.93±0.03 vs. 0.94±0.05, all P > 0.05). However, the β diversity analysis showed that the difference between groups was greater than that within groups whether weighted according to OTU or not (abundance weighted matrix: R = 0.23, P = 0.04; abundance unweighted matrix: R = 0.32, P = 0.01). At the phylum level, the abundance of Proteobacteria and Candidatus_sacchari in the CLP group increased significantly as compared with the Sham group [18.100% (15.271%, 26.665%) vs. 6.974% (2.854%, 9.764%), 0.125% (0.027%, 0.159%)% vs. 0.018% (0.008%, 0.021%), both P < 0.05]. At the genus level, the abundance of opportunistic pathogen including Helicobacter, Ruthenium, Streptococcus, Clostridium ⅩⅧ in the CLP group was significantly higher than that in the Sham group [5.090% (1.812%, 6.598%) vs. 0.083% (0.034%, 0.198%), 0.244% (0.116%, 0.330%) vs. 0.016% (0.008%, 0.029%), 0.006% (0.003%, 0.010%) vs. 0.001% (0%, 0.003%), 0.094% (0.035%, 0.430%) vs. 0.007% (0.003%, 0.030%), all P < 0.05], and the abundance of probiotics such as Alloprevotella and Romboustia was significantly lower than that in the Sham group [7.345% (3.662%, 11.546%) vs. 22.504% (14.403%, 26.928%), 0.113% (0.047%, 0.196%) vs. 1.229% (0.809%, 2.29%), both P < 0.01]. LEfSe analysis showed that the probiotics belonging to Firmicutes were significantly enriched in the Sham group, and Romboustia was the most significantly enriched species. Opportunistic pathogens such as Helicobacter, Streptococcus and Clostridium ⅩⅧ were significantly enriched in the CLP group, Helicobacter_NGSU_ 2015 was the most significantly enriched species. Conclusion:In the early stage of sepsis, the intestinal microbiota structure of rats is significantly changed, which mainly shows that the abundance of Alloprevotella and other probiotics is significantly reduced, while that of Helicobacter and other opportunistic pathogens is significantly increased.