ObjectiveTo construct a large language model （LLM）-based intelligent pre-consultation system for traditional Chinese medicine （TCM） to improve efficacy of clinical practice. MethodsA TCM large language model was fine-tuned using DeepSpeed ZeRO-3 distributed training strategy based on YAYI 2-30B. A weighted undirected graph network was designed and an agent-based syndrome differentiation model was established based on relationship data extracted from TCM literature and clinical records. An agent collaboration framework was developed to integrate the TCM LLM with the syndrome differentiation model. Model performance was comprehensively evaluated by Loss function， BLEU-4， and ROUGE-L metrics， through which training convergence， text generation quality， and language understanding capability were assessed. Professional knowledge test sets were developed to evaluate system proficiency in TCM physician licensure content， TCM pharmacist licensure content， TCM symptom terminology recognition， and meridian identification. Clinical tests were conducted to compare the system with attending physicians in terms of diagnostic accuracy， consultation rounds， and consultation duration. ResultsAfter 100 000 iterations， the training loss value was gradually stabilized at about 0.7±0.08， indicating that the TCM-LLM has been trained and has good generalization ability. The TCM-LLM scored 0.38 in BLEU-4 and 0.62 in ROUGE-L， suggesting that its natural language processing ability meets the standard. We obtained 2715 symptom terms， 505 relationships between diseases and syndromes， 1011 relationships between diseases and main symptoms， and 1 303 600 relationships among different symptoms， and constructed the Agent of syndrome differentiation model. The accuracy rates in the simulated tests for TCM practitioners， licensed pharmacists of Chinese materia medica， recognition of TCM symptom terminology， and meridian recognition were 94.09%， 78.00%， 87.50%， and 68.80%， respectively. In clinical tests， the syndrome differentiation accuracy of the system reached 88.33%， with fewer consultation rounds and shorter consultation time compared to the attending physicians （P＜0.01）， suggesting that the system has a certain pre- consultation ability. ConclusionThe LLM-based intelligent TCM pre-diagnosis system could simulate diagnostic thinking of TCM physicians to a certain extent. After understanding the patients' natural language， it collects all the patient's symptom through guided questioning， thereby enhancing the diagnostic and treatment efficiency of physicians as well as the consultation experience of the patients.

Hypoxia is the most common tumor microenvironment caused by rapid proliferation of tumor cells, and hypoxia-inducible factor (HIF) is the main transcription factor for tumor cells to adapt to hypoxia. Current research has found that HIF can interact with a variety of mesenchymal cells such as fibroblasts, endothelial cells and immune cells in the tumor microenvironment, leading to the transcription and expression of target genes in response to hypoxia, which ultimately promotes tumor angiogenesis, and induces physiological changes such as migration, invasion, and immune escape of tumor cells. However, the signaling pathways involved in the HIF regulatory mechanism are complex, and the mechanism of HIF in the tumor microenvironment need to be further investigated, also most HIF inhibitors are still in the preclinical research stage. This paper reviews the research progress on the effects of HIF on tumor mesenchymal stromal cells to provide a theoretical basis for the diagnosis, prevention and treatment of tumors targeting HIF.

Objective To explore the efficacy and safety of lung recruitment maneuver(LRM)on postoperative hypoxemia in patients with acute type A aortic dissection(ATAAD).Methods A total of 56 ATAAD patients with postoperative hypoxemia in the First Affiliated Hospital of Guangxi Medical University from November 2019 to May 2022 were selected and randomly divided into LRM group(n=36)and conventional treatment group(n=20).Patients in conventional treatment group received routine mechanical ventilation on the basis of lung protective ventilation.The patients in LRM group were treated with incremental positive end expiratory pressure(PEEP).Arterial blood gas analysis,respiratory parameters,hemodynamics parameters and serum interleukin(IL)-6 and IL-10 levels were compared between two groups before and after treatment.Results At 12h and 24h after treatment,arterial partial pressure of oxygen(PaO2),oxygenation index(OI),static compliance(Cstat)and dynamic compliance(Cdyn)in two groups were significantly higher than before treatment,the alveolar-arterial gradient of oxygen[PO2(A-a)],respiratory index(RI),peak inspiratory pressure(Ppeak)and plateau pressure(Pplat)were significantly lower than before treatment(P<0.05).PaO2,OI,Cstat and Cdyn in LRM group were significantly higher than those in conventional treatment group,PO2(A-a),RI,Ppeak and Pplat were significantly lower than those in conventional treatment group(P<0.05).Systolic blood pressure and mean arterial pressure decreased and central venous pressure increased during pulmonary reexpansion in LRM group(P<0.05),and all patients returned to baseline level after pulmonary reexpansion.At 12h after treatment,serum IL-6 and IL-10 levels in both groups were significantly lower than before treatment(P<0.05).Conclusion Incremental PEEP can improve oxygenation and lung compliance in patients with hypoxemia after ATAAD surgery,but it has transient effects on hemodynamics,and should be closely monitored during treatment.

BACKGROUND:At present,many drugs used in the treatment of polycystic ovary syndrome are super-designated drugs,and the treatment of patients with polycystic ovary syndrome still faces great challenges.Studies have shown that human umbilical cord mesenchymal stem cells can repair ovarian function,but few studies have reported their therapeutic effect on polycystic ovary syndrome. OBJECTIVE:To investigate the therapeutic effect of human umbilical cord mesenchymal stem cells on polycystic ovary syndrome,and to preliminarily explore the correlation between mitochondrial autophagy and the improvement of polycystic ovary syndrome by human umbilical cord mesenchymal stem cells. METHODS:Polycystic ovary syndrome mouse model was established by subcutaneous injection of dehydroepiandrosterone for 20 days into C57BL/6J mice.Human umbilical cord mesenchymal stem cells(2×106)were injected through the caudal vein.After treatment,vaginal secretions were collected for 10 consecutive days to detect the estrus cycle of mice.At 2 weeks after treatment,the levels of sex hormones in the peripheral blood of mice,including luteinizing hormone and follicle-stimulating hormone,were detected by ELISA.Hematoxylin-eosin staining was used to evaluate ovarian histopathology.Finally,mitochondrial autophagy in ovaries was observed by transmission electron microscopy. RESULTS AND CONCLUSION:(1)After human umbilical cord mesenchymal stem cell therapy,follicles at different stages(primitive follicles,primary follicles,and secondary follicles)appeared in the ovary of polycystic ovary syndrome mice,and luteal tissue could be seen,indicating that ovulation function of mice was effectively improved.(2)Polycystic ovary syndrome mice treated with human umbilical cord mesenchymal stem cells had sex hormone levels.(3)Untreated polycystic ovary syndrome mice were found to be in the estrous stage for a long time,lacking estrous interphase and estrous phase,but after human umbilical cord mesenchymal stem cell therapy,the estrous cycle returned to a normal level.(4)After treatment with human umbilical cord mesenchymal stem cells,the mitochondrial autophagy of polycystic ovary syndrome mice was significantly reduced.(5)The results show that human umbilical cord mesenchymal stem cells can effectively improve the symptoms of endocrine disorders and promote ovulation in polycystic ovary syndrome mice,which may be related to the inhibition of mitochondrial autophagy.

Objective To explore the bidirectional causal relationships between periodontitis and asthma using the two-sample Mendelian randomization(MR)method to provide a basis for exploring the etiology and formulating preven-tive and therapeutic measures of periodontitis and asthma.Methods We performed two-sample bidirectional Mende-lian randomization analysis using publicly released European genome-wide association studies(GWAS)statistics for periodontitis(n=34 615)and asthma(n=408 422).The inverse variance weighted(IVW)method was employed as the main approach to estimate the bidirectional causal relationships between periodontitis and asthma.In addition,weighted median(WM),MR-Egger regression,maximum likelihood,and Mendelian randomization robust adjusted profile score(MR-RAPS)were used as supplementary analyses.Sensitivity analyses were conducted using Cochran's Q test,Mende-lian randomization pleiotropy residual sum and outlier(MR-PRESSO),and leave-one-out analysis.Results A total of 12 and 43 single-nucleotide polymorphisms(SNPs)were included as instrumental variables for periodontitis and asthma,respectively.The results of IVW,WM,MR-Egger regression,maximum likelihood,and MR-RAPS showed that periodon-titis was not causally related to the risk of asthma(IVW:OR:1.003,95%CI:0.973-1.035,P=0.828,WM:OR:0.990,95%CI:0.951-1.031,P=0.641;MR-Egger regression:OR:0.988,95%CI:0.960-1.028,P=0.573;maximum likeli-hood:OR:1.003,95%CI:0.972-1.035,P=0.834;MR-RAPS:OR:1.002,95%CI:0.970-1.036,P=0.890)among the European population,and no causal effect of asthma on periodontitis was found(IVW:OR:1.021,95%CI:0.938-1.111,P=0.633,WM:OR:1.011,95%CI:0.894-1.142,P=0.866;MR-Egger regression:OR:1.042,95%CI:0.824-1.319,P=0.731;maximum likelihood:OR:1.021,95%CI:0.938-1.112,P=0.631;MR-RAPS:OR:1.017,95%CI:0.931-1.110,P=0.713)among the European population.Cochran's Q test showed no heterogeneity among the included in-strumental variables,MR-PRESSO test found no horizontal pleiotropy,and the leave-one-out method did not identify out-lier SNPs.Conclusion The results of this study,based on European genetic data,do not support a bidirectional causal association between periodontitis and asthma in the European population.

Objective:To investigate the role of an inflammatory microenvironment induced by Porphyromonasgingivalis ( P. gingivalis) in the occurrence of esophageal squamous cell carcinoma (ESCC) in mice. Methods:A total of 180 C57BL/6 mice were randomly divided into 6 groups, i.e. control group, P. gingivalis group, 4NQO group, 4NQO + P. gingivalis group, 4NQO + P. gingivalis + celecoxib group, and 4NQO + P. gingivalis + antibiotic cocktail (ABC, including metronidazole, neomycin, ampicillin, and vancomycin) group, with 30 mice in each group, using the random number table. All mice were normalized by treatment with ABC in drinking water for 2 weeks. In the following 2 weeks, the mice in the control group and the P. gingivalis group were given drinking water, while the other 4 groups were treated with 30 μg/ml 4NQO in the drinking water. In weeks 11-12, the mice in the P. gingivalis group, the 4NQO + P. gingivalis group, the 4NQO + P. gingivalis + celecoxib group, and the 4NQO + P. gingivalis + ABC group were subjected to ligation of the second molar in oral cavity followed by oral P. gingivalis infection thrice weekly for 24 weeks in weeks 11-34. In weeks 13-34, the mice in 4NQO + P. gingivalis+celecoxib group and 4NQO + P. gingivalis + ABC group were administered with celecoxib and ABC for 22 weeks, respectively. At the end of 34 weeks, gross and microscopic alterations were examined followed by RT-qPCR and immunohistochemistry to examine the expression profiles of inflammatory- and tumor-molecules in esophagi of mice. Results:At 34 weeks, 4NQO treatment alone did not affect the foci of papillary hyperproliferation, diseased area, and the thickness of the esophageal wall, but significantly enhanced the foci of hyperproliferation (median 1.00, P＜0.05) and mild/moderate dysplasia (median 2.00, P＜0.01). In addition, the expression levels of IL-6 [8.35(3.45,8.99)], IL-1β [6.90(2.01,9.72)], TNF-α [12.04(3.31,14.08)], c-myc [2.21(1.80,3.04)], pSTAT3, Ki-67, and pH2AX were higher than those in the control group. The pathological changes of the esophageal mucosa were significantly more overt in the 4NQO + P. gingivalis group in terms of the foci of papillary hyperproliferation (median 2.00), diseased area (median 2.51 mm 2), the thickness of the esophageal wall (median 172.52 μm), the foci of hyperproliferation (median 1.00, P＜0.05), and mild/moderate dysplasia (median 1.00, P＜0.01). In mice of the 4NQO + P. gingivalis group, the expression levels of IL-6 [12.27(5.35,22.08)], IL-1β [13.89(10.04,15.96)], TNF-α [19.56(6.07,20.36)], IFN-γ [11.37(8.23,20.07)], c-myc [2.62(1.51,4.25)], cyclin D1 [4.52(2.68,7.83)], nuclear pSTAT3, COX-2, Ki-67, and pH2AX were significantly increased compared with the mice in the control group. In mice of the 4NQO + P. gingivalis group, the diseased area, invasive malignant foci as well as pSTAT3 and pH2AX expression were significantly blunted by celecoxib. Treatment with ABC markedly reduced the papillary hyperproliferative foci, invasive malignant foci, and pSTAT3 expression but not pH2AX. Conclusions:P. gingivalis promotes the occurrence of esophageal squamous cell carcinoma in mice by inducing an inflammatory microenvironment primed with 4NQO induced DNA damage. Clearance of P. gingivalis with ABC or anti-inflammatory intervention holds promise for prevention of esophageal squamous cell malignant pathogenesis via blockage of IL-6/STAT3 signaling and amelioration of inflammation.

Objective:To investigate the role of an inflammatory microenvironment induced by Porphyromonasgingivalis ( P. gingivalis) in the occurrence of esophageal squamous cell carcinoma (ESCC) in mice. Methods:A total of 180 C57BL/6 mice were randomly divided into 6 groups, i.e. control group, P. gingivalis group, 4NQO group, 4NQO + P. gingivalis group, 4NQO + P. gingivalis + celecoxib group, and 4NQO + P. gingivalis + antibiotic cocktail (ABC, including metronidazole, neomycin, ampicillin, and vancomycin) group, with 30 mice in each group, using the random number table. All mice were normalized by treatment with ABC in drinking water for 2 weeks. In the following 2 weeks, the mice in the control group and the P. gingivalis group were given drinking water, while the other 4 groups were treated with 30 μg/ml 4NQO in the drinking water. In weeks 11-12, the mice in the P. gingivalis group, the 4NQO + P. gingivalis group, the 4NQO + P. gingivalis + celecoxib group, and the 4NQO + P. gingivalis + ABC group were subjected to ligation of the second molar in oral cavity followed by oral P. gingivalis infection thrice weekly for 24 weeks in weeks 11-34. In weeks 13-34, the mice in 4NQO + P. gingivalis+celecoxib group and 4NQO + P. gingivalis + ABC group were administered with celecoxib and ABC for 22 weeks, respectively. At the end of 34 weeks, gross and microscopic alterations were examined followed by RT-qPCR and immunohistochemistry to examine the expression profiles of inflammatory- and tumor-molecules in esophagi of mice. Results:At 34 weeks, 4NQO treatment alone did not affect the foci of papillary hyperproliferation, diseased area, and the thickness of the esophageal wall, but significantly enhanced the foci of hyperproliferation (median 1.00, P＜0.05) and mild/moderate dysplasia (median 2.00, P＜0.01). In addition, the expression levels of IL-6 [8.35(3.45,8.99)], IL-1β [6.90(2.01,9.72)], TNF-α [12.04(3.31,14.08)], c-myc [2.21(1.80,3.04)], pSTAT3, Ki-67, and pH2AX were higher than those in the control group. The pathological changes of the esophageal mucosa were significantly more overt in the 4NQO + P. gingivalis group in terms of the foci of papillary hyperproliferation (median 2.00), diseased area (median 2.51 mm 2), the thickness of the esophageal wall (median 172.52 μm), the foci of hyperproliferation (median 1.00, P＜0.05), and mild/moderate dysplasia (median 1.00, P＜0.01). In mice of the 4NQO + P. gingivalis group, the expression levels of IL-6 [12.27(5.35,22.08)], IL-1β [13.89(10.04,15.96)], TNF-α [19.56(6.07,20.36)], IFN-γ [11.37(8.23,20.07)], c-myc [2.62(1.51,4.25)], cyclin D1 [4.52(2.68,7.83)], nuclear pSTAT3, COX-2, Ki-67, and pH2AX were significantly increased compared with the mice in the control group. In mice of the 4NQO + P. gingivalis group, the diseased area, invasive malignant foci as well as pSTAT3 and pH2AX expression were significantly blunted by celecoxib. Treatment with ABC markedly reduced the papillary hyperproliferative foci, invasive malignant foci, and pSTAT3 expression but not pH2AX. Conclusions:P. gingivalis promotes the occurrence of esophageal squamous cell carcinoma in mice by inducing an inflammatory microenvironment primed with 4NQO induced DNA damage. Clearance of P. gingivalis with ABC or anti-inflammatory intervention holds promise for prevention of esophageal squamous cell malignant pathogenesis via blockage of IL-6/STAT3 signaling and amelioration of inflammation.

Objective: To compare the clinical outcomes and biomechanical characteristics of 1-, 2-, and 3-level pedicle subtraction osteotomy (PSO), and establish selection criteria based on preoperative radiographic parameters. Methods: Patients undergone PSO to treat ankylosing spondylitis from February 2009 to May 2019 in Sun Yat-sen Memorial Hospital of Sun Yat-sen University were enrolled. According to the quantity of osteotomy performed, the participants were divided into group A (1-level PSO, n = 24), group B (2-level PSO, n = 19), and group C (3-level PSO, n = 11). Clinical outcomes were assessed before surgery and at the final follow-up. Comparisons of the radiographic parameters and quality-of-life indicators were performed among and within these groups, and the selection criteria were established by regression. Finite element analysis was conducted to compare the biomechanical characteristics of the spine treated with different quantity of osteotomies under different working conditions. Results: Three-level PSO improved the sagittal parameters more significantly, but resulted in longer operative time and greater blood loss (p < 0.05). Greater stress was found in the proximal screws and proximal junction area of the vertebra in the model simulating 1-level PSO. Larger stress of screws and vertebra was observed at the distal end in the model simulating 3-level PSO. Conclusion Multilevel PSO works better for larger deformity correction than single-level PSO by allowing greater sagittal parameter correction and obtaining a better distribution of stress in the hardware construct, although with longer operation time and greater blood loss. Three-level osteotomy is recommended for the patients with preoperative of global kyphosis > 85.95°, T1 pelvic angle > 62.3°, sagittal vertical alignment > 299.55 mm, and pelvic tilt+ chin-brow vertical angle > 109.6°.

Objective: To compare the clinical outcomes and biomechanical characteristics of 1-, 2-, and 3-level pedicle subtraction osteotomy (PSO), and establish selection criteria based on preoperative radiographic parameters. Methods: Patients undergone PSO to treat ankylosing spondylitis from February 2009 to May 2019 in Sun Yat-sen Memorial Hospital of Sun Yat-sen University were enrolled. According to the quantity of osteotomy performed, the participants were divided into group A (1-level PSO, n = 24), group B (2-level PSO, n = 19), and group C (3-level PSO, n = 11). Clinical outcomes were assessed before surgery and at the final follow-up. Comparisons of the radiographic parameters and quality-of-life indicators were performed among and within these groups, and the selection criteria were established by regression. Finite element analysis was conducted to compare the biomechanical characteristics of the spine treated with different quantity of osteotomies under different working conditions. Results: Three-level PSO improved the sagittal parameters more significantly, but resulted in longer operative time and greater blood loss (p < 0.05). Greater stress was found in the proximal screws and proximal junction area of the vertebra in the model simulating 1-level PSO. Larger stress of screws and vertebra was observed at the distal end in the model simulating 3-level PSO. Conclusion Multilevel PSO works better for larger deformity correction than single-level PSO by allowing greater sagittal parameter correction and obtaining a better distribution of stress in the hardware construct, although with longer operation time and greater blood loss. Three-level osteotomy is recommended for the patients with preoperative of global kyphosis > 85.95°, T1 pelvic angle > 62.3°, sagittal vertical alignment > 299.55 mm, and pelvic tilt+ chin-brow vertical angle > 109.6°.

Objective: To compare the clinical outcomes and biomechanical characteristics of 1-, 2-, and 3-level pedicle subtraction osteotomy (PSO), and establish selection criteria based on preoperative radiographic parameters. Methods: Patients undergone PSO to treat ankylosing spondylitis from February 2009 to May 2019 in Sun Yat-sen Memorial Hospital of Sun Yat-sen University were enrolled. According to the quantity of osteotomy performed, the participants were divided into group A (1-level PSO, n = 24), group B (2-level PSO, n = 19), and group C (3-level PSO, n = 11). Clinical outcomes were assessed before surgery and at the final follow-up. Comparisons of the radiographic parameters and quality-of-life indicators were performed among and within these groups, and the selection criteria were established by regression. Finite element analysis was conducted to compare the biomechanical characteristics of the spine treated with different quantity of osteotomies under different working conditions. Results: Three-level PSO improved the sagittal parameters more significantly, but resulted in longer operative time and greater blood loss (p < 0.05). Greater stress was found in the proximal screws and proximal junction area of the vertebra in the model simulating 1-level PSO. Larger stress of screws and vertebra was observed at the distal end in the model simulating 3-level PSO. Conclusion Multilevel PSO works better for larger deformity correction than single-level PSO by allowing greater sagittal parameter correction and obtaining a better distribution of stress in the hardware construct, although with longer operation time and greater blood loss. Three-level osteotomy is recommended for the patients with preoperative of global kyphosis > 85.95°, T1 pelvic angle > 62.3°, sagittal vertical alignment > 299.55 mm, and pelvic tilt+ chin-brow vertical angle > 109.6°.