Objective:By analyzing the clinical data of patients with primary duodenal adenocarcinoma (PDA), the risk factors affecting the postoperative prognosis of PDA patients were discussed.Methods:The clinical data of 191 patients diagnosed with PDA in Peking University First Hospital from Jan 2009 to Dec 2022 were collected. The survival rate was calculated and the survival curve was plotted by Kaplan-Meier method. Univariate analysis was performed by Log-Rank test, and multivariate analysis was performed by COX proportional hazards regression model to obtain independent risk factors.Results:The median age of onset in patients with PDA is 65 years old, and the most common symptoms are abdominal pain and abdominal distension. Prognostic analysis showed that the survival rates at 1, 3 and 5 years were 73.8%, 44.6%, and 23.0%. The analysis of Cox risk proportional regression model showed that preoperative CA19-9 level, depth of tumor invasion, degree of differentiation, TNM stage, and surgical mode were independent risk factors for the prognosis of PDA (all P<0.01). Conclusion:The overall incidence of PDA is low, but the prognosis is rather poor. Multvariable factors are associated with its prognosis and surgery is still the mainstay for hope of cure.

【Objective】 To investigate the effects of preoperative ureteroscopy (URS) on the intravesical recurrence (IVR) in patients with upper tract urothelial carcinoma (UTUC) after radical nephroureterectomy (RNU). 【Methods】 The clinical data of 241 UTUC patients treated during May 2012 and Jan.2020 in the Second Hospital of Laozhou University were retrospectively analyzed. The patients were divided into URS before RNU group (URS group) and non-URS before RNU group (non-URS group). The cumulative IVR rate, progression-free survival (PFS) and overall survival (OS) after RNU were compared, and the survival curve was drawn. Cox proportional hazards models were used to assess risk factors affecting IVR. 【Results】 Of the 241 patients, 64 (26.6%) were included in the URS group and 177 (73.4%) in the non-URS group. In the URS group, 49 underwent biopsy and 15 did not. All patients were followed up for a median of 44 (3 to 122) months, with a median time to recurrence of 12 (3 to 56) months. IVR occurred in 18 patients (28.1%) in the URS group and 25 (14.1%) in the non-URS group. Kaplan-Meier survival analysis showed that the cumulative IVR rate was higher in the URS group than in the non-URS group (all P<0.05), regardless of whether patients had a history of bladder cancer (BC) or not, while PFS was lower in the URS group than in the non-URS group (P=0.007). Cox multivariate regression analysis showed that URS (P=0.031) and complicated renal pelvis tumor and ureteral tumor (P=0.004) were independent risk factors for IVR. 【Conclusion】 Preoperative URS increases the incidence of IVR in patients with UTUC, and routine preoperative use of URS is not recommended.

Objective: More than 75% of ovarian cancer patients are diagnosed at advanced stages and die of tumor cell metastasis. This study aimed to identify new epigenetic and transcriptomic alterations associated with ovarian cancer metastasis. Methods: Two cell sublines with low- and high-metastasis potentials were derived from the ovarian cancer cell line A2780. Genome-wide DNA methylome and transcriptome profiling were carried out in these two sublines by Reduced Representation Bisulfite Sequencing and RNA-seq technologies. Cell-based assays were conducted to support the clinical findings. Results: There are distinct DNA methylation and gene expression patterns between the two cell sublines with low- and high-metastasis potentials. Integrated analysis identified 33 methylation-induced genes potentially involved in ovarian cancer metastasis. The DNA methylation patterns of two of them (i.e., SFRP1 and LIPG) were further validated in human specimens, indicating that they were hypermethylated and downregulated in peritoneal metastatic ovarian carcinoma compared to primary ovarian carcinoma. Patients with lower SFRP1 and LIPG expression tend to have a worse prognosis. Functionally, knockdown of SFRP1 and LIPG promoted cell growth and migration, whereas their overexpression resulted in the opposite effects. In particular, knockdown of SFRP1 could phosphorylate GSK3β and increase β-catenin expression, leading to deregulated activation of the Wnt/β-catenin signaling. Conclusion Many systemic and important epigenetic and transcriptomic alterations occur in the progression of ovarian cancer. In particular, epigenetic silencing of SFRP1 and LIPG is a potential driver event in ovarian cancer metastasis. They can be used as prognostic biomarkers and therapeutic targets for ovarian cancer patients.

Objective:To evaluate the safety of double and a half layered esophagojejunal anastomosis in radical gastrectomy.Methods:This prospective, multi-center, single-arm study was initiated by the Affiliated Cancer Hospital of Zhengzhou University in June 2021 (CRAFT Study, NCT05282563). Participating institutions included Nanyang Central Hospital, Zhumadian Central Hospital, Luoyang Central Hospital, First Affiliated Hospital of Henan Polytechnic University, First Affiliated Hospital of Henan University, Luohe Central Hospital, the People's Hospital of Hebi, First People's Hospital of Shangqiu, Anyang Tumor Hospital, First People's Hospital of Pingdingshan, and Zhengzhou Central Hospital Affiliated to Zhengzhou University. Inclusion criteria were as follows: (1) gastric adenocarcinoma confirmed by preoperative gastroscopy;(2) preoperative imaging assessment indicated that R0 resection was feasible; (3) preoperative assessment showed no contraindications to surgery;(4) esophagojejunostomy planned during the procedure; (5) patients volunteered to participate in this study and gave their written informed consent; (6) ECOG score 0–1; and (7) ASA score I–III. Exclusion criteria were as follows: (1) history of upper abdominal surgery (except laparoscopic cholecystectomy);(2) history of gastric surgery (except endoscopic submucosal dissection and endoscopic mucosal resection); (3) pregnancy or lactation;(4) emergency surgery for gastric cancer-related complications (perforation, hemorrhage, obstruction); (5) other malignant tumors within 5 years or coexisting malignant tumors;(6) arterial embolism within 6 months, such as angina pectoris, myocardial infarction, and cerebrovascular accident; and (7) comorbidities or mental health abnormalities that could affect patients' participation in the study. Patients were eliminated from the study if: (1) radical gastrectomy could not be completed; (2) end-to-side esophagojejunal anastomosis was not performed during the procedure; or (3) esophagojejunal anastomosis reinforcement was not possible. Double and a half layered esophagojejunal anastomosis was performed as follows: (1) Open surgery: the full thickness of the anastomosis is continuously sutured, followed by embedding the seromuscular layer with barbed or 3-0 absorbable sutures. The anastomosis is sutured with an average of six to eight stitches. (2) Laparoscopic surgery: the anastomosis is strengthened by counterclockwise full-layer sutures. Once the anastomosis has been sutured to the right posterior aspect of the anastomosis, the jejunum stump is pulled to the right and the anastomosis turned over to continue to complete reinforcement of the posterior wall. The suture interval is approximately 5 mm. After completing the full-thickness suture, the anastomosis is embedded in the seromuscular layer. Relevant data of patients who had undergone radical gastrectomy in the above 12 centers from June 2021 were collected and analyzed. The primary outcome was safety (e.g., postoperative complications, and treatment). Other studied variables included details of surgery (e.g., surgery time, intraoperative bleeding), postoperative recovery (postoperative time to passing flatus and oral intake, length of hospital stay), and follow-up conditions (quality of life as assessed by Visick scores).Result:[1] From June 2021 to September 2022,457 patients were enrolled, including 355 men and 102 women of median age 60.8±10.1 years and BMI 23.7±3.2 kg/m2. The tumors were located in the upper stomach in 294 patients, mid stomach in 139; and lower stomach in 24. The surgical procedures comprised 48 proximal gastrectomies and 409 total gastrectomies. Neoadjuvant chemotherapy was administered to 85 patients. Other organs were resected in 85 patients. The maximum tumor diameter was 4.3±2.2 cm, number of excised lymph nodes 28.3±15.2, and number of positive lymph nodes five (range one to four. As to pathological stage,83 patients had Stage I disease, 128 Stage II, 237 Stage III, and nine Stage IV. [2] The studied surgery-related variables were as follows: The operation was successfully completed in all patients, 352 via a transabdominal approach, 25 via a transhiatus approach, and 80 via a transthoracoabdominal approach. The whole procedure was performed laparoscopically in 53 patients (11.6%), 189 (41.4%) underwent laparoscopic-assisted surgery, and 215 (47.0%) underwent open surgery. The median intraoperative blood loss was 200 (range, 10–1 350) mL, and the operating time 215.6±66.7 minutes. The anastomotic reinforcement time was 2 (7.3±3.9) minutes for laparoscopic-assisted surgery, 17.6±1.7 minutes for total laparoscopy, and 6.0±1.2 minutes for open surgery. [3] The studied postoperative variables were as follows: The median time to postoperative passage of flatus was 3.1±1.1 days and the postoperative gastrointestinal angiography time 6 (range, 4–13) days. The median time to postoperative oral intake was 7 (range, 2–14) days, and the postoperative hospitalization time 15.8±6.7 days. [4] The safety-related variables were as follows: In total, there were 184 (40.3%) postoperative complications. These comprised esophagojejunal anastomosis complications in 10 patients (2.2%), four (0.9%) being anastomotic leakage (including two cases of subclinical leakage and two of clinical leakage; all resolved with conservative treatment); and six patients (1.3%) with anastomotic stenosis (two who underwent endoscopic balloon dilation 21 and 46 days after surgery, the others improved after a change in diet). There was no anastomotic bleeding. Non-anastomotic complications occurred in 174 patients (38.1%). All patients attended for follow-up at least once, the median follow-up time being 10 (3–18) months. Visick grades were as follows: Class I, 89.1% (407/457); Class II, 7.9% (36/457); Class III, 2.6% (12/457); and Class IV 0.4% (2/457).Conclusion:Double and a half layered esophagojejunal anastomosis in radical gastrectomy is safe and feasible.

Objective:To evaluate the safety of double and a half layered esophagojejunal anastomosis in radical gastrectomy.Methods:This prospective, multi-center, single-arm study was initiated by the Affiliated Cancer Hospital of Zhengzhou University in June 2021 (CRAFT Study, NCT05282563). Participating institutions included Nanyang Central Hospital, Zhumadian Central Hospital, Luoyang Central Hospital, First Affiliated Hospital of Henan Polytechnic University, First Affiliated Hospital of Henan University, Luohe Central Hospital, the People's Hospital of Hebi, First People's Hospital of Shangqiu, Anyang Tumor Hospital, First People's Hospital of Pingdingshan, and Zhengzhou Central Hospital Affiliated to Zhengzhou University. Inclusion criteria were as follows: (1) gastric adenocarcinoma confirmed by preoperative gastroscopy;(2) preoperative imaging assessment indicated that R0 resection was feasible; (3) preoperative assessment showed no contraindications to surgery;(4) esophagojejunostomy planned during the procedure; (5) patients volunteered to participate in this study and gave their written informed consent; (6) ECOG score 0–1; and (7) ASA score I–III. Exclusion criteria were as follows: (1) history of upper abdominal surgery (except laparoscopic cholecystectomy);(2) history of gastric surgery (except endoscopic submucosal dissection and endoscopic mucosal resection); (3) pregnancy or lactation;(4) emergency surgery for gastric cancer-related complications (perforation, hemorrhage, obstruction); (5) other malignant tumors within 5 years or coexisting malignant tumors;(6) arterial embolism within 6 months, such as angina pectoris, myocardial infarction, and cerebrovascular accident; and (7) comorbidities or mental health abnormalities that could affect patients' participation in the study. Patients were eliminated from the study if: (1) radical gastrectomy could not be completed; (2) end-to-side esophagojejunal anastomosis was not performed during the procedure; or (3) esophagojejunal anastomosis reinforcement was not possible. Double and a half layered esophagojejunal anastomosis was performed as follows: (1) Open surgery: the full thickness of the anastomosis is continuously sutured, followed by embedding the seromuscular layer with barbed or 3-0 absorbable sutures. The anastomosis is sutured with an average of six to eight stitches. (2) Laparoscopic surgery: the anastomosis is strengthened by counterclockwise full-layer sutures. Once the anastomosis has been sutured to the right posterior aspect of the anastomosis, the jejunum stump is pulled to the right and the anastomosis turned over to continue to complete reinforcement of the posterior wall. The suture interval is approximately 5 mm. After completing the full-thickness suture, the anastomosis is embedded in the seromuscular layer. Relevant data of patients who had undergone radical gastrectomy in the above 12 centers from June 2021 were collected and analyzed. The primary outcome was safety (e.g., postoperative complications, and treatment). Other studied variables included details of surgery (e.g., surgery time, intraoperative bleeding), postoperative recovery (postoperative time to passing flatus and oral intake, length of hospital stay), and follow-up conditions (quality of life as assessed by Visick scores).Result:[1] From June 2021 to September 2022,457 patients were enrolled, including 355 men and 102 women of median age 60.8±10.1 years and BMI 23.7±3.2 kg/m2. The tumors were located in the upper stomach in 294 patients, mid stomach in 139; and lower stomach in 24. The surgical procedures comprised 48 proximal gastrectomies and 409 total gastrectomies. Neoadjuvant chemotherapy was administered to 85 patients. Other organs were resected in 85 patients. The maximum tumor diameter was 4.3±2.2 cm, number of excised lymph nodes 28.3±15.2, and number of positive lymph nodes five (range one to four. As to pathological stage,83 patients had Stage I disease, 128 Stage II, 237 Stage III, and nine Stage IV. [2] The studied surgery-related variables were as follows: The operation was successfully completed in all patients, 352 via a transabdominal approach, 25 via a transhiatus approach, and 80 via a transthoracoabdominal approach. The whole procedure was performed laparoscopically in 53 patients (11.6%), 189 (41.4%) underwent laparoscopic-assisted surgery, and 215 (47.0%) underwent open surgery. The median intraoperative blood loss was 200 (range, 10–1 350) mL, and the operating time 215.6±66.7 minutes. The anastomotic reinforcement time was 2 (7.3±3.9) minutes for laparoscopic-assisted surgery, 17.6±1.7 minutes for total laparoscopy, and 6.0±1.2 minutes for open surgery. [3] The studied postoperative variables were as follows: The median time to postoperative passage of flatus was 3.1±1.1 days and the postoperative gastrointestinal angiography time 6 (range, 4–13) days. The median time to postoperative oral intake was 7 (range, 2–14) days, and the postoperative hospitalization time 15.8±6.7 days. [4] The safety-related variables were as follows: In total, there were 184 (40.3%) postoperative complications. These comprised esophagojejunal anastomosis complications in 10 patients (2.2%), four (0.9%) being anastomotic leakage (including two cases of subclinical leakage and two of clinical leakage; all resolved with conservative treatment); and six patients (1.3%) with anastomotic stenosis (two who underwent endoscopic balloon dilation 21 and 46 days after surgery, the others improved after a change in diet). There was no anastomotic bleeding. Non-anastomotic complications occurred in 174 patients (38.1%). All patients attended for follow-up at least once, the median follow-up time being 10 (3–18) months. Visick grades were as follows: Class I, 89.1% (407/457); Class II, 7.9% (36/457); Class III, 2.6% (12/457); and Class IV 0.4% (2/457).Conclusion:Double and a half layered esophagojejunal anastomosis in radical gastrectomy is safe and feasible.

Objective:To investigate the effects of compound fermented milk on intestinal microbiota, short chain fatty acid (SCFA), intestinal motility and mucosal barrier in mice with constipation.Methods:Twenty-seven C57BL/6JNifdc mice were randomly divided into three groups: control group, model group and intervention group. The model group and the intervention group were given loperamide intragastrically for two weeks. Starting from the second week, the intervention group was treated with compound fermented milk for 7 d. The control group was given normal saline. Food intake, water intake, weight change, fecal moisture content, time of first-time black stool and small intestine propulsion rate were detected. Expression of serotonin C receptor (5-HTR2C), zona occludins-1 (ZO-1) and mucin-2 (MUC-2) at mRNA level in colon was analyzed. Western blot was used to measure the expression of Raf/ERK/MAPK-related proteins. SCFA level in intestinal tract was detected by gas chromatography. Intestinal microbiota diversity was analyzed by high-throughput sequencing.Results:Compared with the control group, the first black stool excretion time was significantly prolonged in the model group ( P<0.01). Moreover, fecal moisture content, small intestine propulsion rate and the expression of 5-HTR2C and ZO-1 at mRNA level in colon were significantly decreased ( P<0.01). Compared with the model group, the first black stool excretion time was significantly shortened ( P<0.05); fecal moisture content, small intestine propelling rate ( P<0.05), the expression of colon 5-HTR2C and ZO-1 at mRNA level ( P<0.05), phosphorylation of Raf/ERK/MAPK pathway in the colon, intestinal SCFA-producing bacteria and intestinal SCFA content were increased in the intervention group. Conclusions:Compound fermented milk had a therapeutic effect on constipation in a mouse model by increasing the abundance of SCFA-producing bacteria and SCFA content, enhancing the phosphorylation of the Raf/ERK/MAPK pathway to up-regulate the expression of 5-HTR2C at mRNA level in the colon, and increasing the expression of ZO-1 at mRNA level in the colon. Intestinal peristalsis and intestinal mucosal barrier function were enhanced, thus improving the symptom of constipation.

Objective:To observe the effect of different clinical factors on the level of parathyroid hormone (PTH) and to predict the possibility of permanent hypothyroidism (PHP) after total thyroidectomy (TT) by monitoring the levels of PTH in serum and drainage fluid after TT.Methods:Retrospective analysis was made on 150 patients who underwent TT for papillary thyroid carcinoma (PTC) in the Department of Thyroid Head and Neck Surgery in Jilin Cancer Hospital from Jan. 2020 to Aug. 2021. The changes of serum PTH were recorded at 1, 3, 7, 30 days and 6 months after surgery. The risk factors of postoperative hypoparathyroidism (HP) were investigated by single factor and multi factor methods. The impairment of parathyroid function was predicted combined with the level of PTH in the drainage fluid 1 day after operation.Results:After TT, serum PTH returned to normal value in most patients 1 month after operation. The proportion of PHP was 3.33% (5/150). Univariate analysis showed that bilateral central lymph node dissection, Hashimoto’s thyroiditis, tumor diameter ≥2 cm and intraoperative selective parathyroid autologous transplantation were risk factors for temporary hypoparathyroidism (THP). Multivariate analysis showed that BCND ( OR=0.322, P=0.001) , intraoperative selective parathyroid autograft ( OR=5.442, P=0.001) and tumor diameter ≥2 cm ( OR=2.247, P=0.003) were independent risk factors for THP. ROC curve was used to compare the predictive effect of postoperative serum and drainage PTH levels on postoperative PHP. The statistical results showed that the highest predictive effect of postoperative PHP was found on the first day of drainage PTH level within 1 week after operation (AUC 0.81) . 54 cases whose serum PTH was lower than normal value on the first day after operation were divided into 4 groups according to the level of PTH in drainage fluid from high to low. The results showed that the lower the level of PTH in drainage fluid, the greater the possibility of PHP ( P<0.05). Conclusions:Most of the patients with low PTH one month after operation develop PHP, while bilateral central lymph node dissection, intraoperative selective parathyroid transplantation, tumor diameter ≥2 cm are risk factors for THP after TT. If the serum PTH is lower than the normal value on the first day after operation, there is a possibility of PTH. The lower the PTH in the drainage fluid, the greater the possibility of PHP, which should be paid attention to in clinical practice.

Objective: To investigate the relationship between gut microbiota structure and biochemical changes in patients with different types of nonalcoholic fatty liver disease (NAFLD), in order to provide evidence for clinical diagnosis and prevention of NAFLD. Methods: Forty-eight NAFLD cases (NAFLD group), 40 NAFLD cases with type 2 diabetes mellitus (NAFLD combined with type 2 diabetes mellitus group) and 30 healthy cases (healthy group) were randomly enrolled, and their body mass index, serum alanine aminotransferase, aspartate aminotransferase, total bilirubin, total cholesterol, triglyceride, high density lipoprotein, low density lipoprotein and uric acid were measured. Serum levels of TNF-alpha and fasting insulin were measured using ELISA, and then insulin resistance index was calculated. The gut microbiota of three groups of subjects was detected using 16S rDNA-based high-throughput sequencing. Lastly, the correlations between the various factors were analyzed. The comparison among groups was conducted by 2 test, and one-way ANOVA was used for comparison among groups with normal distribution and homogeneity of variance. Furthermore, the LSD method was used to compare the two groups. K-W rank sum test was used for comparison among groups without normal distribution or homogeneity of variance. Results: Body mass index, aspartate aminotransferase, triglyceride, total cholesterol, low density lipoprotein, uric acid, tumor necrosis factor-alpha, fasting insulin and insulin resistance index of NAFLD group were higher than healthy group, while the high-density lipoprotein was lower in the healthy group, and the difference was statistically significant (P< 0.05). Compared with NAFLD group, the life expectancy, fasting blood glucose and insulin resistance index of NAFLD combined with type 2 diabetes mellitus group were higher, while the body mass index, aspartic acid aminotransferase, total cholesterol and HDL levels were decreased, and the difference was statistically significant (P< 0.05). NAFLD group (P= 0.016) had decreased abundance of firmicutes than healthy group, and the abundancy of the firmicutes in the NAFLD combined with type 2 diabetes group was significantly lower (P< 0.001). The abundance of bacteroidetes in NAFLD combined with type 2 diabetes group was higher than healthy group, and the difference was statistically significant (P= 0.006). At the "genus level," the abundance of Roseburia and Subdoligranulum in the NAFLD group was decreased, while the Roseburia in the NAFLD group with type 2 diabetes group was significantly lower (P< 0.05). In addition, the abundance of Faecalibacterium, Blautia, Anaerostipes and Fusicatenibacter in NAFLD combined with type 2 diabetes group was lower than healthy group, and the difference was statistically significant (P< 0.001). Fusicatenibacter, Blautia, Anaerostipes, Faecalibacterium, and Roseburia were negatively correlated with fasting blood glucose and insulin resistance index levels (r< 0,P< 0.05), and positively correlated with high-density lipoprotein levels (r> 0,P< 0.05). Fusicatenibacter was negatively correlated with tumor necrosis factor-alpha (r= -0.211,P= 0.044), and Lachnoclostridium was positively correlated with body mass index, alanine aminotransferase, aspartate aminotransferase levels (r> 0,P< 0.05). Fusobacterium was positively correlated with aspartate aminotransferase level (r= 0.245,P= 0.019). Escherichia-shigella was positively correlated with fasting blood glucose, low-density lipoprotein, alanine aminotransferase, aspartate aminotransferase levels (r > 0,P< 0.05). Megamonas was negatively correlated with high-density lipoprotein levels (r= -0.231,P= 0.027). Conclusion A structural change of gut microbiota had occurred in patients with NAFLD, suggesting changes in some of these bacterial genuses had relation to insulin resistance and inflammatory response, which may become a new target for the treatment of NAFLD.

Objective To analyze the impact of Helicobacter pylori standard strain (Hp P12) and its virulence factor vacuolating cytotoxin A ( VacA) on DNA damage and homologous recombination ( HR) repair in a human gastric epithelial cell line (GES-1). Methods Strains of Hp P12 and vacA gene knock-out Hp P12 ( Hp P12 ΔvacA) were respectively used to infect GES-1 cells at a multiplicity of infection of 100. GES-1 cells treated with etoposide (50μmol/L) or mitomycin (0. 5μg/ml) for 2 h were used as posi-tive control. Western blot and immunofluorescence were performed to detect the expression of DNA damage marker protein γH2AX and key HR repair proteins (Rad51, pMRE11, CtIP and pCtIP) and the recruitment of them at DNA damage sites. Human embryonic kidney HEK-293 ( DR-GFP) cells were infected with Hp P12 and Hp P12 ΔvacA strains to verify the impact of VacA on HR repair efficiency. Results The expres-sion and recruitment of γH2AX and key HR repair proteins ( Rad51, pMRE11, CtIP and pCtIP) were in-creased in Hp P12-infected cells as compared with that in uninfected and Hp P12 ΔvacA-infected cells ( all P<0. 05). To evaluate the HR repair efficiency, I-SceⅠ plasmid-transfected HEK-293 (DR-GFP) cells were infected with Hp P12 and Hp P12 ΔvacA and the results showed that green fluorescent protein ( GFP)-positive cells were decreased after infection, especially in Hp P12 ΔvacA-infected cells (both P<0. 05). Conclusions Hp P12 infection could cause DNA damage and promote HR repair in GES-1 cells, in which the virulence factor VacA played an important role.

Objective To investigate the effects of cholesterol-lowering probiotics, DM9054 com-bined with 86066, on the intestinal mucosal barrier and gut microbiota in mice with nonalcoholic fatty liver disease ( NAFLD) induced by high-fat diet and the possible mechanisms. Methods Twenty-four male mice deficient in the low-density lipoprotein receptor gene ( Ldlr- / - mice ) were randomly divided into three groups including control, NAFLD model and probiotic intervention groups. Mice in the three groups were given normal chow diet+normal saline, high-fat diet ( HFD)+normal saline, and HFD+cholesterol-lowering probiotics, respectively. The mouse model of NAFLD was established by feeding mice with high-fat diet (45％ of calories derived from fat diet) for 12 weeks. qPCR was performed to measure the expression of liv-er and intestinal inflammatory genes and liver cholesterol synthesis genes. Western blot assay was used to de-tect the expression of intestinal tight junction proteins and HMG-CoA reductase ( HMGCR ) . Pathological changes in tissues were evaluated by HE staining. Features of gut microbiota were analyzed by 16S rRNA gene sequencing. Results Cholesterol-lowering probiotics intervention attenuated HFD-induced hepatic steatosis, inflammatory responses and obesity and decreased the synthesis of liver cholesterol (P<0. 05). Moreover, inhibited gut inflammatory responses and improved intestinal barrier function were detected in the probiotic intervention group (P<0. 05). The composition of gut microbiota in mice of the probiotic intervention group was different from that of the model group, but similar to that of the control group. Con-clusions Cholesterol-lowering probiotics might attenuate NAFLD in mice through reducing liver cholesterol synthesis, alleviating liver and intestinal inflammation, improving intestinal mucosal barrier function and reg-ulating intestinal microbiota.