Objective To evaluate the predictive value of the Model for End-Stage Liver Disease Excluding International Normalized Ratio （MELD-XI） score for long-term all-cause mortality in elderly patients after cardiac valve surgery. Methods This retrospective study incorporated 263 elderly patients （aged ≥ 65 years） who underwent cardiac valve surgery at the hospital between January and December 2013. The MELD-XI score was computed based on the preoperative levels of total bilirubin and serum creatinine， and the long-term survival status was acquired. Patients were classified into a low-score group （≤ 11.82） and a high-score group （> 11.82） according to the optimal cut-off value determined by receiver operating characteristic curve analysis. Univariate and multivariate Cox proportional hazards regression models were employed to analyze the association between the MELD-XI score and long-term mortality， which was validated by Kaplan-Meier survival curves and subgroup analysis. Results Over a median follow-up period of 7.9 years， 60 patients passed away. Multivariate Cox regression analysis indicated that for every 1-point increment in the MELD-XI score， there was a 9% rise in the mortality risk （HR = 1.09， 95%CI： 1.03 - 1.16， P = 0.002 4）. The high-score group exhibited a significantly greater mortality risk when compared to the low-score group （HR = 2.36， 95%CI： 1.29 - 4.35， P = 0.005 6）. Subgroup analysis revealed a consistent predictive value among patients with different clinical characteristics （all interaction P-values > 0.05）. Kaplan-Meier survival curves verified a significantly higher cumulative mortality in the high-score group （Log-rank P = 0.000 52）. Conclusion The preoperative MELD-XI score serves as an independent predictor of the long-term mortality risk in elderly patients following cardiac valve surgery， where higher scores are indicative of a poorer prognosis.

Objective To analyze the independent risk factors for enteral nutrition feeding intolerance （ENFI） in neurocritical patients， construct a risk prediction model， and validate its performance so as to provide a reference for the early clinical identification of high-risk patients. Methods A prospective cohort study was carried out on 248 neurocritical patients who were receiving enteral nutrition therapy at a tertiary hospital in Guangzhou from July 2023 to July 2025. Univariate and multivariate logistic regression analyses were conducted to identify independent risk factors. A prediction model was developed based on these factors， and its performance was assessed using the receiver operating characteristic （ROC） curve and decision curve analysis （DCA）. Results The incidence of FI was 67.3%. Multivariate analysis identified the following independent risk factors： the length of ICU stay （OR = 1.136， 95%CI： 1.043 - 1.238）， mechanical ventilation （OR = 2.831， 95%CI： 1.168 - 6.858）， targeted temperature management （OR = 3.595， 95%CI： 1.709 - 7.565）， the concomitant use of multiple sedatives （≥ 2 types） （OR = 2.786， 95%CI： 1.102 - 7.045）， and the use of dehydrating agents （OR = 3.692， 95%CI： 1.535 - 8.876）. The area under the ROC curve was 0.851 （95%CI： 0.802 - 0.900， P < 0.001）， with a sensitivity of 90.4% and a specificity of 57.6%， respectively. DCA confirmed the clinical utility of the model. Conclusions By incorporating variables specific to neurocritical care， such as targeted temperature management and combined sedative use， the constructed risk prediction model exhibits excellent performance. This model can function as an effective instrument for predicting the risk of enteral nutrition FI in neurocritical patients.

Objective To explore the application effect of digital intelligent diagnosis and treatment technology combined with nurses' full participation in assisting decision-making in preoperative patient-physician communication for grade Ⅳ hepatobiliary surgery， and to analyze the influence of gender and education level on patients' preoperative anxiety and satisfaction with communication， and to clarify the correlation between communication methods and preoperative anxiety as well as independent influencing factors. Methods This study employed a retrospective cohort design. From January 2022 to December 2025， a total of 146 eligible patients were enrolled and divided into a digital intelligent group and a traditional group. In the digital intelligent group， nurses assisted patients in surgical decision-making throughout the process. The research team brought patients into a conversation room equipped with digital intelligent three-dimensional model posters for popular science introduction， patient education tools， and displays， and conducted preoperative conversations based on individualized three-dimensional models made from patients' imaging data， showing surgical plans and procedures on the models. The traditional group only received conventional preoperative conversations. Both groups of patients filled out the Self-Rating Anxiety Scale （SAS） to assess their preoperative anxiety status； they also filled out a self-designed communication effectiveness questionnaire to evaluate communication effectiveness and satisfaction with the conversation. Subgroup analyses were conducted based on gender and education level， and Spearman correlation analysis was used to analyze the correlation between communication indicators and anxiety， while binary logistic regression analysis was used to identify independent influencing factors of preoperative anxiety. Results The preoperative SAS score of patients in the digital intelligent group was 37.5 （32.5， 43.8）， significantly lower than that of the traditional group （51.3， 50.0， 53.8）， with a statistically significant difference （P < 0.001）. The communication effectiveness score of patients in the digital intelligent group was 13.0 （12.0， 15.0）， significantly higher than that of the traditional group （12.0， 12.0， 13.0）， with a statistically significant difference （P < 0.001）. Subgroup analyses based on gender and education level showed no statistically significant differences in intervention effects （P > 0.05）； communication indicators were negatively correlated with SAS scores， and communication methods were an independent influencing factor for preoperative anxiety （OR = 33.261， P < 0.001）. Subgroup analyses based on gender and education level also showed no statistically significant differences. Conclusion Integrating digital intelligent technology with nurses' full participation in assisting decision-making into the preoperative communication process for grade Ⅳ hepatobiliary surgery can reduce patients' preoperative anxiety levels， improve the quality of doctor-patient communication， and enhance patients' participation in decision-making.

Objective To explore and compare the clinical efficacy of thoracoscopic lobectomy and sublobectomy in treating patients with sub-centimeter non-small cell lung cancer （NSCLC） （with a maximum tumor diameter ≤ 1 cm）， as well as their influence on perioperative indicators and postoperative complications. Methods From May 2018 to May 2024， 138 patients with sub-centimeter non-small cell lung cancer （NSCLC） were recruited as study subjects for a retrospective cohort analysis. Based on the differences in surgical treatment methods， they were divided into a lobectomy group （video-assisted thoracoscopic lobectomy， 67 cases） and a sublobectomy group （video-assisted thoracoscopic sublobectomy， 71 cases）. The surgical efficacy， perioperative indicators， lung function indicators， and postoperative complications were compared between the two groups， and a statistical analysis was conducted on the occurrence of postoperative complications. Results There were no significant differences in surgical efficacy and postoperative tumor markers between the sublobectomy group and the lobectomy group （P > 0.05）. The sublobectomy group exhibited better perioperative indicators compared to the lobectomy group， with statistical differences （P < 0.05）. The overall incidence of postoperative complications was significantly higher in the lobectomy group than in the sublobectomy group （P < 0.05）. The forced expiratory volume in 1 second （FEV1）， forced vital capacity （FVC）， maximum voluntary ventilation （MVV）， and FEV1/FVC in the sublobectomy group were higher than those in the lobectomy group after surgery （P < 0.05）. No significant difference was found in the 1-year survival rate between the two groups of patients after surgery （P > 0.05）. Conclusions In comparison with video-assisted thoracoscopic lobectomy， video-assisted thoracoscopic sublobectomy for sub-centimeter NSCLC can more significantly shorten the surgical time， reduce the trauma， expedite postoperative recovery， enhance postoperative lung function， and lower the incidence rates of postoperative complications on the premise of ensuring the same short-term oncological efficacy and survival. Therefore， video-assisted thoracoscopic sublobectomy is a safe and effective surgical approach.

Objective To explore the relationship between high-resolution CT signs of pulmonary ground glass nodules （GGN） and the pathological typing of lung adenocarcinoma. Methods A total of 104 patients with GGN confirmed by pathology from October 2021 to October 2025 were included in the study. Based on the World Health Organization （WHO） pathological classification criteria for lung adenocarcinoma， the patients were classified into the invasive adenocarcinoma （IAC） group， the minimally invasive adenocarcinoma （MIA） group， the adenocarcinoma in situ （AIS） group， and the atypical adenomatous hyperplasia （AAH） group. The baseline data， clinical indicators， high-resolution computed tomography （HRCT） sign features， and HRCT quantitative parameters of patients with different pathological types were compared. The HRCT sign features and quantitative parameters of IAC subtypes （wall-attached type， acinar type， and papillary type） were compared. Binary logistic regression analysis was employed to explore the influencing factors of IAC. The receiver operating characteristic （ROC） curve was utilized to analyze the diagnostic value of HRCT features for IAC. Results The proportions of lobulation sign， spiculation sign， clear and rough interfaces， vacuole sign， bronchial truncation sign， pleural indentation sign， and vessel convergence sign were significantly higher in the IAC group than those in the MIA， AIS， and AAH groups （P < 0.05）. The maximum nodule diameter， CT value， volume， and mass in the IAC group were significantly higher than those in the MIA， AIS， and AAH groups （P < 0.05）， whereas the relative CT value was significantly lower than that in the other three groups （P < 0.05）. There were no statistically significant differences in HRC qualitative features among the three IAC subtypes （P > 0.05）. The CT value， volume， and mass in the wall-attached type group were significantly lower than those in the acinar type and papillary type groups （P < 0.05）， while the relative CT value was significantly higher （P < 0.05）. Moreover， the CT value in the acinar type group was significantly lower than that in the papillary type group （P < 0.05）. The maximum nodule diameter， CT value， and volume were all influencing factors of IAC occurrence （P < 0.05）. The AUC values for predicting confirmed IAC using the lobulation sign， maximum nodule diameter， CT value， and volume were 0.668， 0.929， 0.827， and 0.909 respectively. Combined prediction yielded an AUC of 0.992， with a sensitivity of 97.01% and a specificity of 75.68%， indicating higher predictive value. Conclusions The high-resolution CT signs of pulmonary GGN are associated with the pathological typing of lung adenocarcinoma. In patients with IAC， the maximum nodule diameter， CT value， and volume are relatively large， and the lobulation sign is prominent. There are significant differences in CT signs between patients with IAC and those without IAC. Moreover， CT quantitative parameters and signs can also offer references for the clinical identification of IAC pathological subtypes.

Objective To compare the mid-term outcomes of chimney endovascular aneurysm repair （ch-EVAR） and fenestrated endovascular aneurysm repair （f-EVAR） in treating pararenal abdominal aortic aneurysm （PAAA）， and to explore the anatomical criteria for the selection of the procedure. Methods A retrospective analysis was carried out on the clinical data of 95 patients with PAAA who were treated at Tianjin Medical University General Hospital from January 2013 to December 2021. Among them， there were 52 patients in the ch-EVAR group and 43 in the f-EVAR group. The technical success rate， perioperative complications， and 36-month follow-up outcomes were compared between the two groups. Receiver operating characteristic （ROC） curve analysis was conducted to characterize the distribution of anatomical parameters， such as the tortuosity index and the number of planned reconstructed branches. Results The technical success rates were 98.08% in the ch-EVAR group and 97.67% in the f-EVAR group （P > 0.05）. The number of planned reconstructed visceral arteries was notably higher in the f-EVAR group compared to the ch-EVAR group （2.19 ± 1.22 vs. 1.23 ± 0.58， P < 0.001）. ROC curve analysis indicated a tortuosity index threshold of 1.08 （AUC = 0.925） and a reconstructed branch number threshold of 1.5 （AUC = 0.726） that could distinguish the two groups. The incidence of immediate type Ⅲ endoleak was greater in the f-EVAR group than in the ch-EVAR group （16.28% vs. 0.00%， P = 0.003）， while the overall endoleak rate did not show a significant difference between the groups. During a mean follow-up period of （35.4 ± 24.2） months， no significant differences were found in the cumulative survival rate， freedom from non-PAAA-related mortality， or freedom from secondary intervention between the two groups （P > 0.05）. Conclusions Ch-EVAR and f-EVAR exhibit comparable mid-term efficacy in the treatment of PAAA. According to the retrospective data from our center， f-EVAR might be the preferred choice for patients with a tortuosity index < 1.08 who need to reconstruct ≥ 2 visceral arteries. In contrast， ch-EVAR serves as a safe and effective surgical alternative for elderly patients with a tortuosity index > 1.08 who only require single-branch reconstruction.

Objective To explore the relationships among Paired box 8 （PAX8）， α-enolase （ENO1）， human epididymal protein 4 （HE4）， and neuropilin-2 （NRP-2）， as well as their associations with the severity of endometrial cancer and their value in predicting the prognosis of patients. Methods From January 2021 to November 2022， 168 patients diagnosed with endometrial cancer were recruited， all of whom underwent surgical treatment. Following the surgical intervention， a follow-up protocol was initiated， requiring patients to attend follow-up appointments every three months. The follow-up period extended until November 2025. During this period， 3 patients were lost to follow-up. Ultimately， 165 patients with endometrial cancer were included in the study， forming the endometrial cancer group， with a median follow-up duration of 48 months. Among these patients， 35 cases with a poor prognosis （poor prognosis was defined as any occurrence of distant metastasis， patient death， or recurrence of endometrial cancer） were classified as the poor prognosis group， while the remaining 130 cases were designated as the good prognosis group. Concurrently， 165 patients with benign endometrial diseases were selected as the benign endometrial diseases group， and 165 healthy women were recruited as the healthy control group. Clinical data and serum levels of PAX8， ENO1， HE4， and NRP-2 were compared across the groups. The serum levels of PAX8， ENO1， HE4， and NRP-2 in relation to different pathological features were analyzed. The relationship between the serum levels of PAX8， ENO1， HE4， and NRP-2 and the severity of endometrial cancer was examined using Spearman correlation analysis. By designating patients with a poor prognosis as positive cases and those with a good prognosis as negative cases， the predictive value of the serum levels of PAX8， ENO1， HE4， and NRP-2 for the prognosis of patients with endometrial cancer was evaluated using the receiver's operating characteristics （ROC） curve. Results The serum levels of PAX8， ENO1， HE4， and NRP-2 in the endometrial cancer group were significantly higher than those in the benign endometrial disease group and the healthy control group. Additionally， the levels in the benign endometrial disease group were significantly higher than those in the healthy control group （P < 0.05）. In patients， the levels of serum PAX8， ENO1， HE4， and NRP-2 in those with FIGO stage Ⅲ-Ⅳ， poor differentiation， myometrial infiltration ≥ 1/2， and lymph node metastasis were significantly higher than those in patients with FIGO stage Ⅰ-Ⅱ， medium-high differentiation， myometrial infiltration < 1/2， and no lymph node metastasis （P < 0.05）. The serum levels of PAX8， ENO1， HE4， and NRP-2 were positively correlated with FIGO stage， myometrial infiltration， and lymph node metastasis， while negatively correlated with differentiation （P < 0.05）. The serum levels of PAX8， ENO1， HE4， and NRP-2 in patients with a poor prognosis were significantly higher than those in patients with a good prognosis （P < 0.05）. The combined detection of the serum levels of PAX8， ENO1， HE4， and NRP-2 could predict the area under the curve （AUC） of endometrial cancer patients， and this AUC value was higher than that of individual detection （P < 0.05）. Conclusions Serum levels of PAX8， ENO1， HE4， and NRP-2 are abnormally expressed in patients with endometrial cancer. This abnormal expression is closely related to the severity of the disease. Moreover， the combined examination of these markers has high clinical value in predicting the prognosis of patients with endometrial cancer.

Objective To investigate the effects of different doses of dehydrocavidine （DHC） used as a single agent on the in vitro proliferation， migration， and invasion capabilities of human hepatocellular carcinoma Huh7 cells， as well as the in vivo tumor-forming capacity of mouse hepatocellular carcinoma Hep1-6 cells. Simultaneously， to clarify the differences in anticancer efficacy between DHC and sorafenib and the synergistic effects of combination therapy， thus providing experimental evidence for the treatment of hepatocellular carcinoma （HCC）. Methods Human hepatocellular carcinoma Huh7 cells were treated with DHC at gradient concentrations of 0.1， 0.2， and 0.3 mg/mL. Cell proliferation was assessed using the CCK-8 assay. Meanwhile， cell colony formation， migration， and invasion were evaluated using plate cloning， cell scratch， and Transwell assays， respectively. A suspension of Hep1-6 cells in the logarithmic growth phase was subcutaneously injected into the right axilla of BALB/c-nu nude mice to establish a subcutaneous liver cancer xenograft model. Once the tumour volume had reached approximately 30 mm³， the nude mice bearing xenografts were randomly divided into a control group， low-dose， medium-dose， and high-dose DHC groups to conduct an in vivo dose-response study of DHC monotherapy. Once the effective dose of DHC had been determined， additional groups were established， including a control group， a DHC monotherapy group， a sorafenib monotherapy group， and a DHC-sorafenib combination group， to conduct experiments comparing the efficacy of individual drugs and combination therapy. All groups were administered the drugs regularly via intraperitoneal injection every other day. During the experiment， the major and minor axes of the tumours were periodically measured to calculate tumour volume， and changes in the body weight of the nude mice were continuously monitored. Results DHC demonstrated a concentration- and time-dependent inhibitory effect on the proliferation of Huh7 cells （P < 0.05）. In all concentration groups， the wound healing width， the number of colony-forming units， and the number of cells crossing the membrane were significantly lower compared to the control group （P < 0.05）. In the in vivo single-agent administration experiments， the tumour volumes in all DHC dose groups were significantly smaller than those in the control group （P < 0.05）， showing a dose-dependent relationship. In the combination therapy experiments， the tumour volumes in the DHC and sorafenib group， the DHC monotherapy group， and the sorafenib monotherapy group were all significantly smaller than those in the control group （P < 0.05）. There were no statistically significant differences in the body weight among the groups of nude mice （P > 0.05）， and no obvious toxicity was observed. Conclusion DHC can inhibit， in a dose-dependent manner， the in vitro malignant phenotype of human hepatocellular carcinoma Huh7 cells and the in vivo tumourigenic capacity of mouse hepatocellular carcinoma Hep1-6 cells.

Objective To investigate the expression level of circGAS5 in the peripheral blood of patients with chemotherapy-resistant rectal cancer and to explore its role in chemotherapy resistance as well as its clinical value in the diagnosis of chemotherapy resistance. Methods A circRNA microarray was initially carried out to screen for circGAS5 as a target associated with chemotherapy resistance in rectal cancer. Subsequently， CCK-8 and colony formation assays were performed in tumor cells to explore its influence on the malignant phenotype of chemotherapy resistance. Clinically， 44 patients with chemotherapy-resistant rectal cancer and 108 patients with chemotherapy-sensitive rectal cancer were recruited. The expression level of circGAS5 in plasma was measured using droplet digital polymerase chain reaction （ddPCR）. Statistical analyses， including the t-test， chi-square test， univariate and multivariate logistic regression analyses， and the receiver operating characteristic （ROC） curve， were utilized to assess the relationship between circGAS5 and chemotherapy resistance in rectal cancer. Results In vitro experiments demonstrated that the expression of circGAS5 was significantly up-regulated in chemotherapy-resistant cells and their culture media， and it was accompanied by enhanced structural stability. Clinical data analysis indicated that the plasma expression levels of circGAS5 were significantly higher in patients with chemotherapy-resistant rectal cancer than in those with chemotherapy-sensitive disease （P < 0.001）. Multivariate logistic regression analysis further confirmed that high plasma circGAS5 expression was an independent risk factor for chemotherapy resistance （OR = 3.41， 95%CI： 2.15 - 5.41， P < 0.001）. ROC curve analysis showed a favorable diagnostic performance with an area under the curve （AUC） of 0.827 9， and the AUC remained stable in subgroup analyses， suggesting the robust diagnostic value of circGAS5. Functional assays showed that circGAS5 promoted the progression of the malignant phenotype associated with chemotherapy resistance in rectal cancer cells. Conclusion circGAS5 may serve as a promising biomarker in peripheral blood for evaluating chemotherapy resistance in rectal cancer.

Objective This study was designed to compare the peri-operative indicators， efficacy， and influencing factors between carbon dioxide （CO₂） laser therapy and loop electrosurgical excision procedure （LEEP） in treating cervical intraepithelial neoplasia grade 2 （CIN2） patients with fertility requirements. Methods We retrospectively included 267 patients with pathologically-confirmed CIN2 from November 2021 to June 2024. Among them， 154 patients received CO₂ laser treatment， and 113 patients received LEEP treatment. We compared the peri-operative indicators， lesion remission rates at 6 and 12 months after surgery， high-risk human papillomavirus （HR-HPV） clearance rates， and cervical cytology （TCT） clearance rates between the two groups. Stratified analyses were carried out according to age and human papillomavirus （HPV） infection type， and multivariate logistic regression was employed to analyze the factors influencing the efficacy. Results The CO₂ laser group exhibited significant advantages over the LEEP group in aspects such as operative time， intraoperative blood loss， postoperative bleeding duration， recovery time， complication rates， and treatment costs （P < 0.05）. Nevertheless， at 6 and 12 months after the operation， the lesion remission rates were notably higher in the LEEP group when compared to the CO₂ laser group （79.65% vs. 68.83%， P = 0.045； 78.76% vs. 66.23%， P = 0.022）. Moreover， the overall HPV clearance rate at 6 months after the operation was higher in the LEEP group （80.53% vs. 70.13%， P = 0.047）， yet no significant difference was detected between the two groups at 12 months after the operation. Stratified analysis indicated that LEEP showed a more prominent early-efficacy advantage in patients aged 25 - 45 years. Multivariate analysis verified that multiple human papillomavirus （HPV） infections were an independent risk factor influencing postoperative lesion remission， whereas the treatment modality itself was not an independent influencing factor. Conclusions CO₂ laser therapy presents significant advantages in terms of minimal invasiveness， rapid recovery， and effective cost control. In contrast， LEEP can lead to quicker early clearance of lesions in patients of reproductive age. Clinical decision-making ought to comprehensively take into account the patient's age， HPV infection status， and individualized requirements. Patients with multiple HPV infections need intensified postoperative monitoring. Future prospective studies with long-term follow-up are necessary to validate these conclusions.

Objective To explore a multi-factor model that integrates enhanced CT parameters with the assessment of tissue rich in cysteinyl acidic secretory protein-like 1 （SPARCL1）， serum vascular endothelial growth factor （VEGF）， and thymidine kinase-1 （TK-1） for evaluating the clinical value and prognosis of primary liver cancer. Methods A total of 213 patients with primary liver cancer admitted to our hospital from January 2021 to August 2024 were selected as the study group. Meanwhile， 213 patients who underwent surgery or were hospitalized for benign liver diseases， including liver cirrhosis， liver hemangioma， and liver cysts， during the same period were selected as the control group. The enhanced CT parameters （arterial phase enhancement difference， venous phase enhancement difference）， SPARCL1 levels measured by fluorescence quantitative polymerase chain reaction， VEGF levels， and TK-1 levels measured by enzyme-linked immunosorbent assay were compared between the two groups. The diagnostic efficacy of each index alone and in combination for liver cancer was evaluated. All patients in the study group underwent transarterial chemoembolization （TACE） treatment. Post-operative follow-up was carried out for 1 year， and 4 cases had missing data. The patients were classified into groups according to their prognosis， with 156 cases in the good prognosis group and 53 cases in the poor prognosis group. A logistic regression model was used to analyze the prognostic factors， and the receiver operating characteristic （ROC） curve was plotted to evaluate the predictive value of each index for poor prognosis. Results The differences in venous phase enhancement between the study group and the poor-prognosis group， along with the SPARCL1 mRNA level， were lower than those in the control group and the good-prognosis group （P < 0.05）. Meanwhile， the difference in arterial phase enhancement， as well as the levels of VEGF and TK-1， were higher than those in the control group and the good-prognosis group （P < 0.05）. The AUC value of enhanced CT combined with the detection of SPARCL1， VEGF， and TK-1 for diagnosing primary liver cancer was higher than that of each indicator alone （P < 0.05）， with a sensitivity of 76.06% and a specificity of 81.22%. Logistic regression analysis confirmed that a low degree of differentiation （OR = 3.284， 95%CI： 1.886 - 5.718）， tumor stage Ⅲ-Ⅳ （OR = 2.793， 95%CI： 1.486 - 5.249）， lymph node metastasis （OR = 2.702， 95%CI： 1.224 - 5.965）， a large difference in arterial phase enhancement （OR = 2.702， 95%CI： 1.182 - 6.179）， a high VEGF level （OR = 2.423， 95%CI： 1.279 - 4.590）， and a high TK-1 level （OR = 2.344， 95%CI： 1.264 - 4.347） were all risk factors for the poor prognosis of primary liver cancer patients （all P < 0.05）. A large difference in venous phase enhancement （OR = 0.379， 95%CI： 0.189 - 0.762）， and a high SPARCL1 level （OR = 0.360， 95%CI： 0.151 - 0.858） were all protective factors for the poor prognosis of primary liver cancer patients （all P < 0.05）. Based on this， a risk factor model was constructed as follows： logit（P） = 0.867 + 1.189 × Degree of Differentiation + 1.027 × Tumor Stage + 0.994 × Lymph Node Metastasis + 0.994 × Difference in Arterial Enhancement - 0.969 × Difference in Venous Enhancement - 1.021 × SPARCL1 + 0.885 × VEGF + 0.852 × TK-1. The combined prediction probability for each patient was calculated according to this model. The ROC curve was plotted with this probability serving as the test variable. The results indicated that the AUC value of this combined prediction model was 0.905， with a sensitivity of 81.13% and a specificity of 82.69%. The predictive value of this model was relatively high. Conclusions The enhanced CT parameters， along with the serum levels of SPARCL1， VEGF， and TK-1， exhibit significant differences among patients with primary liver cancer and those with benign liver diseases， as well as among patients with different prognoses after TACE surgery. The combined detection of multiple indicators holds good diagnostic value for primary liver cancer. A low degree of differentiation， tumor stage Ⅱ-Ⅳ， lymph node metastasis， a large difference in arterial-phase enhancement， a high VEGF level， and a high TK-1 level are risk factors for the poor prognosis of primary liver cancer. In contrast， a large difference in venous-phase enhancement and a high SPARCL1 level are protective factors. The combined prediction model based on these indicators demonstrates good predictive efficacy for the short-term prognosis of patients and can offer a reference for clinical prognosis assessment.

Rheumatoid arthritis （RA） is a chronic autoimmune disease characterized by persistent synovitis， progressive joint destruction， and multisystem involvement. Although current pharmacotherapies have significantly improved the clinical outcomes of RA， challenges remain， including adverse effects， high economic burden， and inadequate treatment response in some patients. Plant-derived extracellular vesicles （PDEVs）， as a novel class of natural nanocarriers， possess the dual attributes of endogenous bioactivity and drug delivery vehicles， offering advantages such as broad availability， convenient preparation， low immunogenicity， and high biosafety. In recent years， research on PDEVs in the treatment of RA has been increasing， and their application value in immunomodulation， anti-inflammatory intervention， and targeted delivery has garnered considerable attention， providing innovative strategies for RA therapy. This review systematically summarizes the structural characteristics， isolation， purification， and characterization methods of PDEVs， with a focus on their research progress and underlying mechanisms in RA treatment， aiming to provide a reference for related basic research and clinical translation.

Objective To develop organ-specific risk prediction models for organ involvement in systemic lupus erythematosus （SLE） based on multidimensional autoantibody profiles， and to validate their performance using a time-split cohort， thereby providing a quantitative tool for early identification of high-risk patients. Methods A retrospective single-center cohort study was conducted with model development and validation performed in chronological order. The development cohort included 170 SLE patients， and 60 patients from the same institution were enrolled as a time-extrapolated validation cohort. Organ/system involvement was defined as a binary outcome according to predefined criteria， including serosal， renal， pulmonary， cutaneous， hematological， articular， and hepatic involvement. Candidate variables were screened by univariate analysis （χ2 test， t-test， or Mann?Whitney U test）， and independent associated factors were identified by multivariate logistic regression. Model discrimination was assessed using ROC analysis， AUC， sensitivity， specificity， and Youden index； robustness was evaluated by stratified 10-fold cross-validation. Calibration was assessed with the Hosmer-Lemeshow test and Brier score. Clinical utility was evaluated by decision curve analysis （DCA）. Results Among 170 SLE patients， the prevalence of organ involvement was： hematological 79.41%， renal 51.76%， pulmonary 51.17%， cutaneous 38.82%， articular 33.53%， hepatic 19.18%， and serosal 18.02%. Multivariate regression demonstrated that anti-P0 and anti-NUC were associated with serosal involvement； anti-Ro52 was associated with renal involvement； anti-dsDNA， ln（age）， and ln（disease duration） were associated with cutaneous involvement； anti-nRNP/Sm was associated with hematological involvement. Anti-NUC was an independent risk factor for articular involvement， whereas anti-AMA-M2 was an independent protective factor. The AUC of individual organ-specific models ranged from 0.60 to 0.80. The overall model achieved an AUC of 0.88 （95%CI： 0.81 - 0.94）， with a mean AUC of 0.86 （95%CI： 0.78 - 0.93） in stratified 10-fold cross-validation. Hosmer-Lemeshow tests in development and validation cohorts were > 0.10， overall Brier score ≈ 0.12. DCA showed a net clinical benefit of the models at threshold probabilities between approximately 10% and 50%. Conclusion Organ-specific risk prediction models based on multidimensional autoantibody profiles demonstrate favorable discrimination and acceptable calibration in a time-split validation cohort， indicating that autoantibody profiles can serve as a useful auxiliary tool for risk stratification of organ involvement in SLE patients.

Objective To observe the change characteristics of C4d， complements C3 and C4 in patients with systemic lupus erythematosus and lupus nephritis （SLE-LN）， and to analyze their clinical value in the early diagnosis of SLE-LN， the evaluation of disease severity， and the prognosis prediction. Methods A total of 98 patients with systemic lupus erythematosus （SLE） who received treatment at the Affiliated Hospital of Chengde Medical University were recruited as the research subjects from January 2024 to January 2025. Based on the occurrence of lupus nephritis （LN）， they were divided into the SLE-LN group （38 cases） and the simple SLE group （60 cases）. The levels of plasma C4d， complements C3， and C4 were measured using the enzyme-linked immunosorbent assay （ELISA）. The 24-hour urine protein， serum creatinine （Scr）， blood urea nitrogen （BUN）， and uric acid （UA） were determined， and the estimated glomerular filtration rate （eGFR） was calculated. The correlation between C4d， C3， C4 and 24-hour urine protein， as well as renal function， was analyzed by Pearson correlation analysis. All 98 SLE patients were followed up for 12 months. According to the changes in renal function， they were divided into the poor prognosis group （with deteriorated renal function， 29 cases） and the good prognosis group （with stable renal function， 66 cases）. The predictive value of C4d， C3， and C4 for the prognosis of SLE was analyzed by receiver operating characteristic （ROC） curves analysis. Results The plasma C4d level in the SLE-LN group was significantly higher than that in the simple SLE group， whereas the C3 and C4 levels were significantly lower than those in the simple SLE group （P < 0.05）. The 24 - hour urine protein， Scr， BUN， and uric acid （UA） levels in the SLE-LN group were significantly higher than those in the simple SLE group， while the eGFR was significantly lower than that in the simple SLE group （P < 0.05）. Pearson correlation analysis indicated that the C4d level was significantly positively correlated with the 24-hour urine protein， Scr， BUN， and UA levels （P < 0.05）， and significantly negatively correlated with the eGFR （P < 0.05）. The C3 and C4 levels were significantly negatively correlated with the 24 - hour urine protein， Scr， BUN， and UA levels （P < 0.05）， and significantly positively correlated with the eGFR （P < 0.05）. The plasma C4d level in the poor prognosis group was significantly higher than that in the good prognosis group， while the C3 and C4 levels were significantly lower than those in the good prognosis group （P < 0.05）. Receiver operating characteristic （ROC） curve analysis showed that the area under the curve （AUC）， sensitivity， and specificity of the combined detection were 0.868， 79.31%， and 86.36% respectively， which had the highest predictive efficiency for the prognosis of SLE patients. Conclusions Compared with patients with simple SLE， the level of plasma C4d is higher， while the levels of complements C3 and C4 are lower in patients with SLE-LN. All three of these indexes are significantly correlated with 24-hour urine protein and renal function. The combined detection of plasma C4d， C3， and C4 has a higher predictive efficiency for the prognosis of SLE patients.

Objective Based on clinical features and imaging data， the influencing factors of atherosclerotic stenosis in the intracranial segment of the vertebral artery were explored. Methods This study adopted a retrospective analysis approach. It selected 216 patients who either visited the hospital because of symptoms like dizziness and vertigo associated with posterior circulation ischemia or underwent routine cerebrovascular assessment. These patients simultaneously completed brain-cervical artery ultrasound and CT Angiography （CTA） examinations between January 2021 and June 2025. According to the diagnostic criteria of CTA， patients with a stenosis rate of ≥ 50% in the intracranial segment of the vertebral artery were included in the case group （62 cases）， whereas those without stenosis or with a stenosis rate < 50% were included in the control group （154 cases）. The clinical data， ultrasound hemodynamic parameters， and plaque characteristics of the two groups were compared. Variables with P < 0.05 in the univariate analysis were incorporated into the multivariate logistic regression model to screen out independent influencing factors. Results The peak systolic velocity and the end-diastolic velocity of the vertebral artery in the case group were significantly higher than those in the control group （P < 0.05）. Meanwhile， the diameter of the vertebral artery in the case group was significantly smaller than that in the control group （P < 0.05）. The proportion of patients with a smoking history， hypertension， diabetes， and eccentric plaques in the case group was significantly higher than that in the control group （all P < 0.05）. The levels of low-density lipoprotein and plaque burden in the case group were also significantly higher than those in the control group （P < 0.05）. Single-factor analysis indicated that there were statistically significant differences in the above indicators between the two groups （all P < 0.05）. The variables with significant meanings from the single-factor analysis were then included in the multivariate Logistic regression analysis. The results demonstrated that having a smoking history （OR = 3.340， 95%CI： 1.258 - 8.865）， elevated low-density lipoprotein levels （OR = 2.392， 95%CI： 1.233 - 4.639）， hypertension （OR = 2.630， 95%CI： 1.108 - 6.242）， diabetes （OR = 2.863， 95%CI： 1.302 - 6.296）， eccentric plaques （OR = 2.649， 95%CI： 1.398 - 5.018）， and large plaque burden （OR = 2.482， 95%CI： 1.267 - 4.861） were all independent influencing factors of atherosclerotic stenosis of the intracranial segment of the vertebral artery （all P < 0.05）. Based on this， a model was constructed as follows： logit（P） = 0.685 + 1.206 × smoking history + 0.872 × increase in low-density lipoprotein level + 0.967 × co-occurrence of hypertension + 1.052 × co-occurrence of diabetes + 0.974 × eccentric plaque + 0.909 × large plaque load. When logit（P） > 0.648， the AUC is 0.891 （95%CI： 0.842 - 0.930）， χ² = 15.320， and the sensitivity and specificity are 83.87% and 83.12% respectively. Conclusions This study demonstrates that a history of smoking， hypertension， diabetes， elevated LDL-C levels， eccentric and high-load plaques are independent influencing factors for stenosis of the intracranial segment of the vertebral artery. Based on these indicators， a logistic model was successfully constructed. This model fits well and has a high predictive value for atherosclerotic stenosis of the intracranial segment of the vertebral artery. Clinically， for patients with the aforementioned characteristics， more imaging screening and risk assessment of the intracranial segment of the vertebral artery should be enhanced.

Objective To investigate the mechanism by which crocin alleviates high glucose-induced glomerular endothelial cell damage based on the inhibitor of nuclear factor kappa B kinase β （IKKβ）/nuclear transcription factor-κB （NF-κB） signaling pathway. Methods High glucose-induced human glomerular endothelial cells were treated with crocin at concentrations of 0， 12.5， 25， 50， and 75 μmol/L， and the optimal concentration of crocin was screened via the CCK-8 experiment. Glomerular endothelial cells were randomly assigned to the normal group， osmotic pressure control group， high glucose group， crocin group， phorbol ester （PMA） group， and crocin + PMA group. Except for the normal group and osmotic pressure control group， all the other groups were induced with high glucose and treated with crocin and the IKKβ/NF-κB signaling pathway activator PMA according to different groups for 48 hours. Immunofluorescence staining was employed to observe the cytoskeleton of the cells. The CCK-8 experiment， Transwell experiment， and TdT-mediated dUTP nick end labeling （TUNEL） staining were carried out to detect cell viability， permeability， and apoptosis， respectively. The oxidative stress and inflammation-related factors in the cells were measured. Western blot （WB） was utilized to detect the expression of proteins related to apoptosis， tight junctions， and the IKKβ/NF-κB signaling pathway. Results Compared with the normal group and the osmotic pressure control group， the high-glucose group exhibited a decrease in the activity of glomerular endothelial cells， the activities of superoxide dismutase （SOD） and glutathione peroxidase （GSH-Px）， the relative expression of occludin protein， and the relative expressions of zonula occludens-1 （ZO-1） and Bcl-2 protein （P < 0.05）. Meanwhile， it showed an increase in permeability， apoptosis rate， malondialdehyde （MDA） content， interleukin （IL）-6 and IL-1β levels， the relative expression of Bax protein， and the ratios of p-IKKβ/IKKβ， p-IκB-α/IκB-α， and p-NF-κB p65/NF-κB p65 （P < 0.05）. Compared with the high-glucose group， the crocin group showed an increase in the activity of glomerular endothelial cells， the activities of SOD and GSH-Px， the relative expression of occludin protein， and the relative expressions of ZO-1 and Bcl-2 protein （P < 0.05）. In addition， it had a decrease in permeability， apoptosis rate， MDA content， IL-6 and IL-1β levels， the relative expression of Bax protein， and the ratios of p-IKKβ/IKKβ， p-IκB-α/IκB-α， and p-NF-κB p65/NF-κB p65 （P < 0.05）. The trend of changes in various indicators in the PMA group was opposite to that in the crocin group （P < 0.05）. Compared with the crocin group， the crocin + PMA group showed a decrease in the activity of glomerular endothelial cells， the activities of SOD and GSH-Px， the relative expression of occludin protein， and the relative expressions of ZO-1 and Bcl-2 protein （P < 0.05）. Moreover， it presented an increase in permeability， apoptosis rate， MDA content， IL-6 and IL-1β levels， the relative expression of Bax protein， and the ratios of p-IKKβ/IKKβ， p-IκB-α/IκB-α， and p-NF-κB p65/NF-κB p65 （P < 0.05）. Conclusion Crocin alleviates the damage to glomerular endothelial cells induced by high glucose by inhibiting the activation of the IKKβ/NF-κB signaling pathway.

Objective To investigate the effects of intravenous thrombolysis with alteplase prior to endovascular treatment on the levels of 5-hydroxytryptamine （5-HT）/5-hydroxyindoleacetic acid （5-HIAA）， 3，4-dihydroxyphenylacetic acid （DOPAC）/homovanillic acid （HVA）， brain-derived neurotrophic factor （BDNF）， and nerve growth factor （NGF） in patients with acute ischemic stroke （AIS）. Methods A total of 174 AIS patients who were admitted to the hospital from January 2023 to June 2025 were divided into a control group and a study group using the random number table method， with 87 cases in each group. The control group was treated with simple endovascular treatment. The study group received intravenous thrombolysis with alteplase for injection prior to endovascular treatment and then underwent endovascular treatment. Both groups were followed up for 3 months after treatment. The efficacy， degree of neurological deficits， cognitive function， activities of daily living， cerebral hemodynamics， inflammation-related indicators， serum levels of 5-HT/5-HIAA， DOPAC/HVA， BDNF， NGF， and safety were compared between the two groups. Results Compared with the control group， the total effective rate of the study group was significantly higher 3 months after treatment （P < 0.05）. When compared with the pre-treatment values， the scores of the National Institutes of Health Stroke Scale （NIHSS） and the levels of serum lipoprotein-associated phospholipase A2 （Lp-PLA2）， matrix metalloproteinase-9 （MMP-9）， high-sensitivity C-reactive protein （hs-CRP）， and tumor necrosis factor-α （TNF-α） in both groups decreased 3 months after treatment， and the values in the study group were lower. The scores of the Mini-Mental State Examination Scale （MMSE）， Modified Barthel Index （MBI）， peak systolic velocity （PSV）， end-diastolic velocity （EDV）， mean velocity （Vm） of the middle cerebral artery， and the serum levels of 5-HT/5-HIAA， DOPAC/HVA， BDNF， and NGF in both groups increased， and the values in the study group were higher （all P < 0.05）. There was no significant difference in safety between the two groups （P > 0.05）. Conclusion Intravenous thrombolysis with alteplase prior to endovascular treatment can effectively enhance the treatment efficacy for patients with AIS， alleviate the degree of neurological deficit， improve cognitive function， activities of daily living， and cerebral hemodynamics， suppress the inflammatory response， regulate metabolic balance， and does not significantly increase the risk of complications， demonstrating good safety.

Objective To analyze the application value of variable-angle zero-profile interbody fusion cage （Zero-PVA） and titanium plate fusion cage （CP） in anterior cervical discectomy and fusion （ACDF） for single-level cervical spondylosis based on imaging indicators. Methods Clinical data of 204 patients with single-level cervical spondylosis who underwent ACDF in our hospital from January 2021 to September 2025 were retrospectively collected. Based on the type of cage utilized during the surgery， the patients were divided into the Zero-PVA group （n = 101） and the CP group （n = 103）. Perioperative indicators， imaging indicators， Bridwell grades of interbody fusion， Japanese Orthopaedic Association （JOA） scores， Neck Disability Index （NDI）， and complications within 3 months after the operation were compared between the two groups. Results The Zero-PVA group exhibited less intraoperative blood loss and a shorter hospital stay compared to the CP group （P < 0.05）. At 1 week and 3 months post-operation， the cervical lordosis angle， the height of the fusion segment， the Cobb angle of the fusion segment， and the Cobb angle of C_2-7 in both groups increased when compared to the preoperative values （P < 0.05）. At both time points， the Cobb angle of the fusion segment in the Zero-PVA group was larger than that in the CP group （P < 0.05）. At 3 months post-operation， there was a statistically significant difference in the Bridwell grades of interbody fusion between the two groups （P < 0.05）. At 3 months post-operation， the JOA scores increased and the NDI decreased in both groups compared to the preoperative values （P < 0.05）， but no significant differences were found in the JOA scores or NDI between the two groups either before or after the surgery （P > 0.05）. The postoperative complication rate in the Zero-PVA group was lower than that in the CP group （P < 0.05）. Conclusions In ACDF for single-level cervical spondylosis， both Zero-PVA and CP can effectively enhance the cervical function of patients. Nevertheless， Zero-PVA is more effective in minimizing intraoperative trauma， improving the Cobb angle of the fusion segment， facilitating bone fusion， and reducing the risk of postoperative complications.

Objective To investigate the relationship between the triglyceride-glucose index （TyG index） and coronary heart disease （CHD）. Methods Clinical data were collected from patients who were admitted to the Department of Cardiology of our hospital because of chest tightness and pain with suspected CHD and underwent coronary angiography between January 2019 and February 2025. Based on the angiography results， the participants were classified into a CHD group and a non-CHD group. According to the quartiles of the TyG index， the participants were further divided into Q1， Q2， Q3， and Q4 groups. Multivariate logistic regression and restricted cubic spline models were used to evaluate the correlation and trend between the clinical TyG levels and the risk of CHD. After adjusting for multiple covariates， smooth curve analysis was carried out to assess the risk of CHD. Meanwhile， based on the summary data from large-scale genome-wide association studies， a two-sample Mendelian randomization （MR） analysis was conducted to prospectively explore the causal effect of the TyG index on CHD. The inverse variance weighting method was used as the primary analytical approach， and multiple sensitivity analyses were conducted to verify the robustness of the causal relationship. Results Clinical data analysis revealed that participants in the CHD group had age， fasting blood glucose levels， triglycerides， TyG index， creatinine AST， patients with diabetes， taking antihypertensive drugs and hypoglycemic drugs were significantly higher than those in the non CHD group， while high-density lipoprotein was lower than that in the non CHD group （P < 0.05）. After adjusting for traditional cardiovascular risk factors， patients in the Q1 and Q4 groups of the TyG index had a significantly increased risk of developing CHD （OR=3.41， 95%CI = 1.41 - 8.26， P = 0.006 6）. Trend analysis further revealed a positive dose-response relationship between the TyG index and the risk of CHD. MR analysis demonstrated a causal association between a higher TyG index and an increased risk of CHD （OR = 2.00， 95%CI： 1.64 - 2.45， P < 0.001）. Conclusions This study confirms， via both clinical data analysis and MR， that a higher TyG index is positively and causally associated with an elevated risk of CHD. These findings imply that the TyG index could act as a biomarker for identifying individuals at a greater risk of CHD.

Objective To explore the prognostic value of regadenoson stress echocardiography in patients with chronic coronary syndromes after percutaneous coronary intervention （PCI）. Methods A total of 113 CCS patients who underwent PCI and completed regadenoson stress echocardiography at the Department of Cardiology， Second Affiliated Hospital of Guizhou University of Traditional Chinese Medicine， from October 2023 to June 2025， were enrolled. According to whether cardiovascular events （including heart failure and myocardial infarction） leading to rehospitalization or all-cause death occurred within 6 months after PCI， patients were divided into the good prognosis group （71 cases） and the poor prognosis group （42 cases）. Stress echocardiography indicators， including coronary flow velocity reserve （CFVR）， global longitudinal strain （GLS）， and left ventricular ejection fraction （LVEF）， were compared between the two groups， along with demographic information and relevant clinical data. Key factors affecting the prognosis of CCS patients after PCI were screened using Least Absolute Shrinkage and Selection Operator （LASSO） regression， and a predictive model for poor prognosis in CCS patients after PCI was constructed using logistic regression. Internally validate the predictive model using the Bootstrap resampling method and assess the clinical utility of the model using decision curves analysis. Analysis of covariance （ANCOVA） was used to adjust for age in a sensitivity analysis to further validate the prognostic value of CFVR and GLS. Results CFVR and GLS in the poor prognosis group were lower than those in the good prognosis group （P < 0.05）， and the age was higher than in the good prognosis group （P < 0.05）. LASSO regression analysis showed that CFVR， GLS， and age were key factors affecting poor prognosis in CCS patients after PCI. Multivariate logistic regression analysis indicated that reduced CFVR， reduced GLS， and advanced age were risk factors for poor prognosis in CCS patients after PCI （P < 0.05）. Receiver operating characteristic （ROC） curves showed that the area under the curve （AUC） for CFVR， GLS， age， and the combined prediction model were 0.835， 0.802， 0.607， and 0.875. Internal validation results and decision curve analysis show that this model has good calibration and can significantly improve clinical net benefit. After adjusting for age by ANCOVA， CFVR and GLS remained significantly lower in the poor prognosis group than in the good prognosis group （P < 0.01）， further confirming that reduced CFVR and GLS are independent risk factors for poor prognosis in CCS patients after PCI. Conclusions Reduced CFVR， reduced GLS， and advanced age are risk factors for poor prognosis in CCS patients after PCI. The combined model is highly feasible for evaluating postoperative prognosis in CCS patients after PCI.

Objective To investigate the association between cerebral blood flow during hemodialysis and white matter hyperintensities （WMHs） in hemodialysis patients. Methods A total of 105 hemodialysis patients from Beijing Shijitan Hospital were enrolled and divided into the WMHs group （n = 57） and the non-WMHs group （n = 48） based on the presence or absence of white matter hyperintensities. The mean flow velocity （MFV） of the middle cerebral artery during dialysis was measured using transcranial Doppler （TCD）， and the MFV reduction rate was calculated. Clinical data were compared between the two groups. Risk factors for WMHs were analyzed. Moreover， the association between the MFV reduction rate and the severity of WMHs was explored， and the predictive value of the MFV reduction rate for WMHs was evaluated. Results Age， dialysis vintage， and the reduction rate of MFV were significantly higher in the WMHs group compared to the non-WMHs group （P < 0.05）. Multivariate logistic regression analysis indicated that the reduction rate of MFV was an independent risk factor for WMHs （P < 0.05）. Multivariate linear regression analysis demonstrated that the reduction rate of MFV was independently associated with the Fazekas score of WMHs. The area under the curve of the reduction rate of MFV for predicting WMHs was 0.806. Conclusion Cerebral blood flow reduction during hemodialysis is significantly associated with white matter hyperintensities WMHs.

During the early stage of the COVID-19 pandemic， the decision-making regarding the treatment of hospitalized patients was marked by an urgent clinical requirement but a lack of high-quality evidence. The Randomised Evaluation of COVID-19 Therapy （RECOVERY） trial， leveraging its adaptive multi-arm platform design， quickly assessed multiple candidate therapies within a unified framework. It confirmed the clinical efficacy of dexamethasone， tocilizumab， and baricitinib in specific hospitalized patients， while disproving the effectiveness of several interventions such as hydroxychloroquine， azithromycin， and lopinavir/ritonavir with strong randomized evidence. Using RECOVERY as the core framework， this review explores how the trial transformed the treatment of hospitalized COVID-19 patients from widespread empirical use to stratified decision-making based on disease stage， inflammatory phenotype， host immune status， and treatment timing. It also outlines RECOVERY′s methodological， clinical， and evidentiary contributions for future emerging infectious diseases. We contend that the significant contribution of RECOVERY lies not only in identifying effective therapies but also in demonstrating that embedded platform trials can rapidly generate evidence that can change clinical practice and support the development of dynamically updated treatment strategies during public health emergencies.

Premature mortality represents a significant public-health challenge among patients with severe mental disorders （SMD）， as their life expectancy is considerably shorter than that of the general population. However， the patterns of causes of death and the underlying mechanisms have not been systematically synthesized yet. This article reviews the latest research on premature mortality in SMD and classifies the causes into unnatural and natural deaths， while emphasizing the exacerbating effects of healthcare-and caregiving-related factors. It further examines the differences in direct causes of death and high-risk stages among different types of SMD， and explores the underlying mechanisms from biological， psychological-behavioral， and social viewpoints. Based on these findings， stratified intervention strategies guided by cause-mechanism pathways are put forward to reduce premature mortality. Future studies integrating cohort designs， multidisciplinary collaboration， and artificial intelligence may enhance the precision and effectiveness of interventions and contribute to reducing the related public-health burden.