ObjectiveObesity, a global chronic metabolic disease, is closely associated with disruptions in lipid metabolism and gut microbiota. Current intervention strategies still have limitations in terms of safety and microecological regulation, necessitating the exploration of novel natural regulatory approaches. Based on the early pathological characteristics of obesity, this study innovatively employs a rectal delivery method alongside a high-fat diet (HFD)-induced obesity model to systematically evaluate the inhibitory effects, safety, and gut microbiota regulation mechanisms of leek-derived and konjac-derived extracellular vesicles on obesity development. By simulating early clinical intervention scenarios, this study aims to explore the preventive potential of plant-derived extracellular vesicles during the initial stages of obesity onset. MethodsExtracellular vesicles from leek and konjac were isolated using ultracentrifugation combined with density gradient centrifugation. Their nanoscale properties were characterized by dynamic light scattering (DLS), transmission electron microscopy (TEM), and nanoparticle tracking analysis (NTA). Male C57BL/6J mice were randomly divided into four groups: normal control (NC), high-fat diet (HFD), leek-derived extracellular vesicles (LEVs), and konjac-derived extracellular vesicles (KEVs). Beginning simultaneously with HFD feeding, mice in the intervention groups received 20 g/L vesicles rectally every 3 d for 4 weeks. Body mass and body composition were monitored throughout. At endpoint, mouse serum, adipose tissue, and colonic contents were collected. Serum biochemical indices (lipid profile, liver and kidney function, cardiac markers) were assessed to evaluate safety and metabolic efficacy, while 16S rRNA sequencing was employed to analyze gut microbial structure and diversity. ResultsDLS, NTA, and TEM confirmed that both LEVs and KEVs exhibited typical cup-shaped nanostructures with average particle sizes of approximately 284 nm and 223 nm, respectively. LEVs and KEVs treatment significantly suppressed HFD-induced weight gain and elevation of body-fat percentage (P<0.05), and reduced accumulation of abdominal white and epididymal adipose tissue. Serological analyses showed that both vesicles lowered total cholesterol, triglycerides and LDL-cholesterol, and ameliorated liver enzyme profiles (ALT, AST), demonstrating lipid-metabolic regulation and hepatoprotective effects. No hepatic, renal or cardiac dysfunction was observed, indicating favorable safety. Gut microbiota analyses revealed that vesicle intervention partially restored HFD-depleted microbial diversity and reshaped community structure. Notably, LEVs markedly increased the relative abundance of the beneficial taxon Lachnospiraceae at the family level, which is known for producing short-chain fatty acids and enhancing intestinal barrier function. Furthermore, Phylogenetic Investigation of Communities by Reconstruction of Unobserved States (PICRUSt) functional prediction suggested that LEVs and KEVs modulated gut microbial functions through distinct mechanisms: LEVs downregulated pathways related to ribosomes and DNA replication while enhancing xenobiotic degradation, whereas KEVs tended to upregulate energy metabolism and protein synthesis toward healthy levels. ConclusionRectally administered LEVs and KEVs exhibit excellent safety and pronounced metabolic benefits during the early phase of obesity, suppressing weight gain, correcting lipid dysregulation, and exerting effects via modulation of gut microbial composition and function. This study provides systematic experimental evidence supporting plant-derived exosome-like vesicles as an early intervention strategy against obesity.

ObjectiveObesity, a global chronic metabolic disease, is closely associated with disruptions in lipid metabolism and gut microbiota. Current intervention strategies still have limitations in terms of safety and microecological regulation, necessitating the exploration of novel natural regulatory approaches. Based on the early pathological characteristics of obesity, this study innovatively employs a rectal delivery method alongside a high-fat diet (HFD)-induced obesity model to systematically evaluate the inhibitory effects, safety, and gut microbiota regulation mechanisms of leek-derived and konjac-derived extracellular vesicles on obesity development. By simulating early clinical intervention scenarios, this study aims to explore the preventive potential of plant-derived extracellular vesicles during the initial stages of obesity onset. MethodsExtracellular vesicles from leek and konjac were isolated using ultracentrifugation combined with density gradient centrifugation. Their nanoscale properties were characterized by dynamic light scattering (DLS), transmission electron microscopy (TEM), and nanoparticle tracking analysis (NTA). Male C57BL/6J mice were randomly divided into four groups: normal control (NC), high-fat diet (HFD), leek-derived extracellular vesicles (LEVs), and konjac-derived extracellular vesicles (KEVs). Beginning simultaneously with HFD feeding, mice in the intervention groups received 20 g/L vesicles rectally every 3 d for 4 weeks. Body mass and body composition were monitored throughout. At endpoint, mouse serum, adipose tissue, and colonic contents were collected. Serum biochemical indices (lipid profile, liver and kidney function, cardiac markers) were assessed to evaluate safety and metabolic efficacy, while 16S rRNA sequencing was employed to analyze gut microbial structure and diversity. ResultsDLS, NTA, and TEM confirmed that both LEVs and KEVs exhibited typical cup-shaped nanostructures with average particle sizes of approximately 284 nm and 223 nm, respectively. LEVs and KEVs treatment significantly suppressed HFD-induced weight gain and elevation of body-fat percentage (P<0.05), and reduced accumulation of abdominal white and epididymal adipose tissue. Serological analyses showed that both vesicles lowered total cholesterol, triglycerides and LDL-cholesterol, and ameliorated liver enzyme profiles (ALT, AST), demonstrating lipid-metabolic regulation and hepatoprotective effects. No hepatic, renal or cardiac dysfunction was observed, indicating favorable safety. Gut microbiota analyses revealed that vesicle intervention partially restored HFD-depleted microbial diversity and reshaped community structure. Notably, LEVs markedly increased the relative abundance of the beneficial taxon Lachnospiraceae at the family level, which is known for producing short-chain fatty acids and enhancing intestinal barrier function. Furthermore, Phylogenetic Investigation of Communities by Reconstruction of Unobserved States (PICRUSt) functional prediction suggested that LEVs and KEVs modulated gut microbial functions through distinct mechanisms: LEVs downregulated pathways related to ribosomes and DNA replication while enhancing xenobiotic degradation, whereas KEVs tended to upregulate energy metabolism and protein synthesis toward healthy levels. ConclusionRectally administered LEVs and KEVs exhibit excellent safety and pronounced metabolic benefits during the early phase of obesity, suppressing weight gain, correcting lipid dysregulation, and exerting effects via modulation of gut microbial composition and function. This study provides systematic experimental evidence supporting plant-derived exosome-like vesicles as an early intervention strategy against obesity.

Sleep disorder is a common complication in elderly patients with cerebral infarction, which seriously affects the rehabilitation process and quality of life of patients. Currently, the treatment of sleep disorders mainly consists of pharmacological and non-pharmacological interventions. Although pharmacological treatment has a certain effect, there are many adverse reactions, especially for elderly patients with declining body function, whose use carries a higher risk. It is of great significance to develop a reasonable individualized non-pharmacological intervention to prevent and treat the accompanying sleep disorders after cerebral infarction. This paper provides a brief review of present status, influencing factors and non-pharmacological interventions of sleep disorders in patients with acute cerebral infarction.

OBJECTIVE To explore and evaluate the machine learning model for prediction of bacterial bloodstream infections established based on routine test data.METHODS By means of retrospective survey,a total of 5 421 pa-tients who were hospitalized in 3 medical institutions from Jan.2015 to Dec.2022 were recruited as the research subjects,1 914 of whom were assigned as the bloodstream infection group,and 3 507 were assigned as the non-bloodstream infection group.The baseline data including gender and age and the results of routine laboratory tests were collected from the enrolled patients.The 3 types of machine learning algorithms,logistic regression,support vector machine and random forest,were respectively used for the screening of the optimal prediction model;the contribution of feature variables to the predictive capability of the model was interpreted through SHAP.The fea-ture variables of the model were optimized by using recursive feature elimination method,and the predictive effi-ciency of the model was evaluated by the area under the curve(AUC)of receiver operating characteristic(ROC)curves.RESULTS Totally 26 variables involving age,gender and blood routine test indexes were included.The random forest was chosen as the optimal machine learning algorithm for the establishment of prediction model for bloodstream infections,and the accuracy of the model was 0.709,with the AUC 0.706.The result of SHAP ex-planation indicated that the age,hematokrit and erythrocyte volume distribution width-CV had remarkable effect on the model's making right decisions.17 variables of the prediction model showed more remarkable effect than 26 variable on distinguishing from the gram-positive bacteria bloodstream infections from the gram-negative bacteria bloodstream infections,with the AUC 0.715,the sensitivity 0.701,the specificity 0.632.CONCLUSIONS The prediction model that is established based on the blood routine test indexes by machine learning algorithm can pre-dict the bacterial bloodstream infection.Meanwhile,the feature selection strategy can further improve the predic-tive efficiency of the model on basis of lowering the dimensionality.

Objective:To describe the epidemiological characteristics and changes of febrile seizure(FS)among children under 6 years old in Ningbo City,Zhejiang Province from 2015 to 2021.Methods:Based on the Ningbo Regional Health Information Platform,a dynamic cohort was established using vacci-nation registration information,and the cases of FS were identified by the diagnostic results of Chinese terms or International Classification of Diseases 10th revision(ICD-10)R56.0 code in the electronic medical records.The first visit of FS during the observation period was defined as a new case,and a re-currence case was defined as the case with a visit interval of more than 7 days.The 95％confidence in-terval(CI)of FS incidence density was calculated by the Poisson distribution.Results:From January 2015 to June 2020,there were 1.3 million children under 6 years old in Ningbo,with male accounting for 52.87％.The median follow-up time was 2.83(1.55-4.00)years.During the follow-up period,12 776 new onset cases had FS,with more males than females,with an overall incidence density of 4.34(95％CI:4.27-4.40)/1 000 person-years and a recurrence rate of 21.63％.There was a higher inci-dence density in children who were male,born in Ningbo and of non-mobility.The incidence density of FS was higher in urban areas than in rural and rural-urban fringe areas,and the incidence density was different among districts and counties.The peak density was found in children aged 18-23 months[8.42(95％CI:8.11-8.74)/1 000 person-years].From 2015 to 2019,the incidence density in-creased with calendar year(Ptrend＜0.001),and the highest was 5.62(95％CI:5.43-5.81)/1 000 person-years.The incidence density of FS decreased significantly during the period between 2020 and 2021.The incidence density was higher in winter.Conclusion:From 2015 to 2019,the overall inci-dence density of FS in children under 6 years old in Ningbo City presented an increasing trend.More at-tention should be paid to the health education,the improvement of the health maintenance model,the en-hancement of the cognition of FS,the identification and treatment of FS among high-risk population and regions so as to prevent its recurrence and reduce the disease burden during the corona virus disease 2019(COVID-19)epide-mic.

The incidence of osteoporotic fractures has demonstrated a significant upward trend, with osteoporotic thoracolumbar fracture accounting for nearly 50% of all the osteoporotic fractures. Following osteoporotic thoracolumbar fracture, the patients face significantly increased risks of disability and even mortality, severely compromising their quality of life and overall health. In recent years, the proposal and application of the concept of very high risk of fracture in osteoporosis have provided new perspectives for stratified treatment approaches in osteoporosis management. Study results demonstrate that fracture risk stratification-based treatment strategies can maximize clinical benefits for osteoporosis patients. However, osteoporosis management in China faces significant challenges characterized by high disease prevalence yet low awareness, diagnosis, and treatment rates, creating substantial barriers to implementing stratified care. To improve the screening and clinical management of very high fracture-risk patients, the authors reviewed the research progress in the diagnosis and treatment of patients with very high risk of osteoporotic thoracolumbar fracture, aiming to provide evidence-based references for formulating treatment strategies for such patients.

Objective:To analyze the characteristics of highly cited articles on spine and spinal cord injuries published in Chinese Journal of Trauma (hereinafter referred to as the "Journal") from 1985 to 2024. Methods:All the articles related to spine and spinal cord injuries published in the Journal from 1985 to 2024 were retrieved from the Wanfang Database and the articles with citation frequency ≥15 times were defined as highly cited, with the retrieval cut-off date of April 15, 2025. For articles related to spine and spinal cord injuries, the total publication count, number of cited articles, total citation frequency, average citation frequency per article, number of fund-supported articles were recorded and analyzed. For the highly cited articles among them, the total number of articles, types of literature, distribution periods, number of fund-supported articles, core authors, institution distribution, region distribution, frequency of and key words were also statistically analyzed.Results:From 1985 to 2024, the Journal published 1 182 articles on spinal and spinal cord injuries, of which 993 (84.01%) were cited, with a total citation frequency of 10 739 and an average of 10.81 citations per article. A total of 388 articles (32.83%) were supported by research funds. There were 184 highly cited articles on spinal and spinal cord injuries, with the following characteristics: (1) The most common type of highly cited literature was original articles (150 articles, 81.52%), while consensus and guideline articles achieved the highest average citation rate (44.75 citations per article). (2) The period 2015-2019 had the highest number of highly cited articles, accounting for 33.70%. (3) Fifty-four highly cited articles (29.35%) were supported by research funds, including 19 by national-level funds and 35 by funds of other levels. (4) Among the highly cited articles, 8 (4.35%) were authored by one person, and 176 (95.65%) were co-authored by two or more persons. Sixty-two authors who published ≥3 highly cited articles were identified as core authors. (5) A total of 123 highly cited articles (66.85%) were authored by researchers affiliated with a single institution, whereas 61 articles (33.15%) were the result of inter-institutional collaboration. Fifteen first-author institutions published≥3 highly cited articles. (6) Regions with ≥15 highly cited articles were Zhejiang Province (28 articles), Chongqing Municipality (23 articles), Guangdong Province (18 articles), Shaanxi Province (18 articles), Jiangsu Province (16 articles), and Shanghai Municipality (15 articles). (7) The top five key words in the highly cited articles were "spinal fractures" (71 times), "thoracic vertebrae" (45 times), "lumbar vertebrae" (44 times), "spinal cord injuries" (27 times), and "vertebroplasty" (27 times).Conclusions:Over the 40-year period, original articles constitute the majority of highly cited articles on spine and spinal cord injuries in the Journal, while consensus and guideline articles demonstrate the highest average citation frequency. Nearly 30% of the highly cited articles receive research funding support. Core authors and institutions show a relatively concentrated distribution, with spinal fractures, spinal cord injuries, and vertebroplasty emerging as research hotspots in the field.

AIM To compare total sugar,total protein,total phenol,total flavonoid,calycosin-7-O-β-D-glucoside,liquiritin,lobetyolin,quercetin,isoferulic acid,hesperidin,glycyrrhizic acid contents and their antioxidant activities,hypoglycemic activities of big honey pill,small honey pill,water pill,concentrated pill,granule,mixture and decoction of Buzhong Yiqi Recipe.METHODS Anthraquinone-sulfuric acid method,Coomassie brilliant blue method,Folin-phenol colorimetry method,sodium nitrite-aluminum nitrate method and HPLC were adopted in the content determination of total sugar,total protein,total phenol,total flavonoid and seven constituents,respectively,after which the scavenging capacities,reducing powers on DPPH·free radical,ABTS+free radical,hydroxyl free radical,and inhibition capacity on α-glucosidase activity were detected.Subsequently,correlation analysis was performed.RESULTS Total sugar,total protein,total phenol and total flavonoid contents demonstrated significant differences among different dosage forms(P＜0.05,P＜0.01).Calycosin-7-O-β-D-glucoside,glycyrrhizin,codonoside and quercetin displayed the highest contents in the decoction,while those of isoferulic acid,hesperidin and glycyrrhizin were observable in the mixture.The water pill exhibited the strongest antioxidant activity,while those of the concentrated pill and mixture were weak;the big honey pill exhibited the strongest hypoglycemic activity,while that of the decoction was the weakest.Total protein,total phenol,total flavonoid and liquiritin contents displayed significant positive correlations between antioxidant activity(P＜0.05,P＜0.01),while hesperidin content displayed significant negative correlation between the latter(P＜0.05);total protein,calycosin-7-O-β-D-glucoside,codonoside and quercetin contents displayed significant negative correlations between hypoglycemic activity(P＜0.05,P＜0.01).CONCLUSION Active constituent contents and their biological activities of Buzhong Yiqi Recipe with different dosage forms exist differences,total sugar,total protein,total flavonoids,calycosin-7-O-β-D-glucoside,licorice glycoside,hesperidin,codonoside and quercetin can be taken as quality control indices for this prescription.

OBJECTIVE To explore and evaluate the machine learning model for prediction of bacterial bloodstream infections established based on routine test data.METHODS By means of retrospective survey,a total of 5 421 pa-tients who were hospitalized in 3 medical institutions from Jan.2015 to Dec.2022 were recruited as the research subjects,1 914 of whom were assigned as the bloodstream infection group,and 3 507 were assigned as the non-bloodstream infection group.The baseline data including gender and age and the results of routine laboratory tests were collected from the enrolled patients.The 3 types of machine learning algorithms,logistic regression,support vector machine and random forest,were respectively used for the screening of the optimal prediction model;the contribution of feature variables to the predictive capability of the model was interpreted through SHAP.The fea-ture variables of the model were optimized by using recursive feature elimination method,and the predictive effi-ciency of the model was evaluated by the area under the curve(AUC)of receiver operating characteristic(ROC)curves.RESULTS Totally 26 variables involving age,gender and blood routine test indexes were included.The random forest was chosen as the optimal machine learning algorithm for the establishment of prediction model for bloodstream infections,and the accuracy of the model was 0.709,with the AUC 0.706.The result of SHAP ex-planation indicated that the age,hematokrit and erythrocyte volume distribution width-CV had remarkable effect on the model's making right decisions.17 variables of the prediction model showed more remarkable effect than 26 variable on distinguishing from the gram-positive bacteria bloodstream infections from the gram-negative bacteria bloodstream infections,with the AUC 0.715,the sensitivity 0.701,the specificity 0.632.CONCLUSIONS The prediction model that is established based on the blood routine test indexes by machine learning algorithm can pre-dict the bacterial bloodstream infection.Meanwhile,the feature selection strategy can further improve the predic-tive efficiency of the model on basis of lowering the dimensionality.

The incidence of osteoporotic fractures has demonstrated a significant upward trend, with osteoporotic thoracolumbar fracture accounting for nearly 50% of all the osteoporotic fractures. Following osteoporotic thoracolumbar fracture, the patients face significantly increased risks of disability and even mortality, severely compromising their quality of life and overall health. In recent years, the proposal and application of the concept of very high risk of fracture in osteoporosis have provided new perspectives for stratified treatment approaches in osteoporosis management. Study results demonstrate that fracture risk stratification-based treatment strategies can maximize clinical benefits for osteoporosis patients. However, osteoporosis management in China faces significant challenges characterized by high disease prevalence yet low awareness, diagnosis, and treatment rates, creating substantial barriers to implementing stratified care. To improve the screening and clinical management of very high fracture-risk patients, the authors reviewed the research progress in the diagnosis and treatment of patients with very high risk of osteoporotic thoracolumbar fracture, aiming to provide evidence-based references for formulating treatment strategies for such patients.