2026年 · 第29卷第03期

Immune dysregulation is one of the leading causes of mortality in patients with sepsis. T cell immunoglobulin and mucin domain-containing protein 3 (TIM-3), a negative costimulatory molecule, is pivotal for immune regulation during sepsis. The effects of TIM-3 appear to be bidirectional: in the early stages of sepsis, upregulation of TIM-3 may help attenuate inflammation, whereas its sustained overexpression in later stages and ligand binding promotes immune apoptosis or exhaustion, which suppresses immune responses. Furthermore, TIM-3 synergizes with other immune checkpoint molecules (e.g., programmed cell death receptor-1), exacerbates immunosuppression, and increases the risk of secondary infections. Blocking the TIM-3 signaling pathway can restore immune cell function and represent a novel therapeutic strategy for sepsis. Although TIM-3 holds promise as both a biomarker and a therapeutic target, its mechanisms are complex and may vary across disease stages, which necessitates further research to optimize targeted interventions. Future studies should focus on elucidating the dynamic signaling pathways of TIM-3, developing combination immunotherapies, and conducting clinical trials to validate its safety and efficacy in sepsis treatment.

Mitophagy has emerged as a key regulator in lung injury, presenting new avenues for therapeutic intervention. Lung injury, often caused by infections, trauma, or inhalation of toxic gases, disrupts lung tissue integrity and function, frequently leading to pulmonary fibrosis. Mitophagy serves a dual purpose: it removes damaged mitochondria, reducing oxidative stress and preventing cell death, thereby offering protection in acute lung injury. However, excessive mitophagy can deplete mitochondria, impair energy metabolism, and aggravate tissue damage. Throughout the progression of lung injury, mitophagy finely tunes inflammation, immune responses, and cell survival, helping to modulate cytokine storms and delay fibrosis. Additionally, it influences metabolic reprogramming and intercellular communication, affecting critical cell types such as alveolar epithelial cells, macrophages, and fibroblasts, which are essential for tissue repair and regeneration. Although the precise molecular mechanisms remain under investigation, mitophagy is increasingly recognized as a promising therapeutic target for lung injury and fibrosis. The future challenge lies in achieving a precise balance in the regulation of mitophagy to maximize its protective effects while minimizing potential harm, thereby opening new pathways for innovative therapeutic strategies.

Purpose: Alveolar type II (AT2) epithelial cells play a vital role in lung injury repair, where their directed migration toward damaged regions is essential for epithelial regeneration. However, the underlying regulatory mechanisms remain poorly understood. This study aimed to determine whether direct-current electric fields (EFs) act as directional cues for AT2 cell migration and to identify the key molecular mediators and intracellular signaling pathways involved. Methods: RNA-seq data from acute respiratory distress syndrome patient–derived AT2 cells were integrated with charged membrane protein signatures to identify candidate genes associated with electrotaxis. Cell migration and calcium dynamics were assessed using time-lapse imaging and live-cell calcium imaging, while pharmacological inhibition, quantitative polymerase chain reaction, and Western blot analyses were employed to investigate molecular mechanisms. Results: Integrative transcriptomic analysis identified Piezo1 as a candidate regulator associated with AT2 cell electrotaxis. Functional assays in A549 cells demonstrated that exposure to a direct-current EF (100 mV/mm) significantly promoted cathode-directed migration and increased Piezo1 expression at both mRNA and protein levels. Inhibition of Piezo1 using the specific antagonist GSMTx4, or chelation of intracellular calcium with BAPTA-AM, markedly disrupted EF-induced directional migration, as evidenced by reduced migration velocity and loss of directionality. Live-cell calcium imaging confirmed that Piezo1 is essential for EF-induced calcium influx. Moreover, Western blot analysis revealed that EF stimulation elevated the phospho-phosphatidylinositol 3-kinase and protein kinase B, which was significantly attenuated upon Piezo1 inhibition. Conclusion: Piezo1-mediated calcium influx and subsequent phospho-phosphatidylinositol 3-kinase/protein kinase B activation drive the electrotaxis of AT2 cells. These findings identify Piezo1 as a key bioelectrical sensor linking EFs to intracellular calcium signaling and directional migration, and suggest its potential as a therapeutic target for promoting epithelial regeneration in acute lung injury.

Purpose: Penetrating neck trauma can frequently be life-threatening, including those that appear minor. Consequently, thorough examination and quick operative treatment have been the standard measure. In addition, modern precision medicine has shifted the procedure towards more selective non-operative management, whereas advanced imaging like computed tomography angiography plays a key role. The diagnostic and therapeutic protocol remains to be adapted and optimized. The purpose of this study was to present our experience, outcomes, and management strategy, and contribute to establishing an evidence-based algorithm for stab wounds to the neck area. Methods: This was a retrospective, observational cohort study conducted at 3 hospitals involving 5 patients with stab wounds to the neck between Jan, 2013 and Dec, 2023. Preoperative interventions included airway and circulation restoration. Ancillary investigations, including imaging and electronic laryngoscopy, were used to assess potential vascular, nerve, and cervical vertebra injury. Surgeries included exploration, debridement, and closure under general anesthesia, with blood transfusion when necessary. Patients were followed up for recovery status and provided postoperative management. Results: Among the 5 patients aged 21–64 years, 4 were males and 1 was female. All wounds were located on the lateral aspect of the neck. Four patients had pharyngocutaneous fistulas with hematemesis/hemoptysis, and 3 of them experienced hemorrhagic shock. One patient had nerve damage, and 1 patient had an airway compressed. All patients underwent successful surgery. Patients were followed up for more than 6 months, and examinations revealed good wound healing, partial recovery of facial nerve function, and no other residual functional impairments in any of the 5 patients. Conclusion: Adequate preoperative preparation and assessment, timely and effective exploratory surgery, efficient multidisciplinary cooperation, and comprehensive postoperative management can maximize the chances of life-saving and functional recovery for penetrating stab wounds in the neck area.

Purpose: To develop and validate an artificial intelligence model based on focused assessment with sonography for trauma (FAST) for the semi-quantitative grading of intra-abdominal hemorrhage resulting from blunt abdominal trauma, particularly for use in prehospital or resource-limited settings. Methods: Nine Bama miniature pigs, mean weight (31.46 ± 3.73) kg were enrolled. Graded hemorrhage from 0 to 1000 mL was simulated by infusing 100 mL of autologous arterial blood into the peritoneal cavity at each step. The hemorrhage volume was mapped to 3 grades based on total blood volume (estimated at 65 mL/kg): Grade I (< 15%), Grade II (15% – 30%), and Grade III (> 30%). FAST ultrasound videos were acquired from 6 standard sites: right upper quadrant-1, right upper quadrant-2, left upper quadrant-1, left upper quadrant-2, right pelvic cavity, and left pelvic cavity. The pixel area of hemorrhage was obtained by manually segmenting the frame with the largest fluid collection using ITK-SNAP, and the corresponding scanning depth was recorded. A linear mixed-effects model was used to assess the impact of scanning depth on pixel area. A deep neural network, incorporating class weighting and dynamic probability threshold optimization, was constructed using a multimodal feature set including animal weight, pixel areas and scanning depths from each site, and the total pixel area. A 3-grade classification was performed. The model's performance was evaluated using leave-one-out cross-validation on an animal basis and compared with logistic regression, random forest, gradient boosting decision tree, and support vector machine. Results: A total of 797 raw videos were acquired, with 522 videos comprising 87 data groups (each covering 6 sites) included after screening. As hemorrhage volume increased, heart rate and shock index rose, while systolic blood pressure decreased; at 800 mL of hemorrhage, the shock index was 2.31 ± 0.38. The mixed-effects model revealed a significant negative correlation between scanning depth and pixel area (β= - 2099.00, SE = 1041.13, z = - 2.02, p = 0.044). The proposed model achieved an overall accuracy of 81.19%, outperforming support vector machine (73.77%), gradient boosting decision tree (70.63%), random forest (69.52%), and logistic regression (65.99%). Conclusion: In a porcine model of blunt abdominal trauma, a multimodal artificial intelligence approach based on FAST multi-site pixel area features, combined with a deep neural network optimized by class weighting and dynamic probability thresholds, can achieve semi-quantitative grading of intra-abdominal hemorrhage.

Purpose: The study was to compare the predictive performance of multiple models for clinical outcomes in trauma patients. To provide decision-making support for the clinical management of trauma patients by analyzing and ranking the importance of predictive indicators. Methods: The data were collected from the sub-system of the national trauma union database (2022 – 2024) in Shanghai East Hospital. The terminal outcomes (death or survive) were the final predictive variables. Lasso regression was employed to screen factors affecting clinical trauma treatment pathways, which were then used as indicators for the final model construction. Logistic regression, random forest, support vector machine, XGBoost, and neural network were used for model evaluation. The predictive performance of each model was assessed through 5 evaluation index including area under curve, accuracy, recall, precision, and F1 score. SHapley Additive exPlanations (SHAP) analysis was employed to assess the importance of variables and visualize the contribution and impact of variables on trauma patient outcomes. Results: Lasso regression selected a total of 9 variables for the final model construction. The variables were: trauma index score, diastolic blood pressure, body temperature, blood oxygen saturation, venous access, injury severity score (ISS), abbreviated injury scale (AIS) score, residence time in emergency department, and number of surgeries. XGBoost had the highest sensitivity (0.800) and Youden's index (0.775), demonstrating strong recognition ability for positive samples. Logistic regression achieved the best accuracy (0.985) and Kappa coefficient (0.598), with stronger overall classification consistency. Neural network had the lowest sensitivity (0.450) and Youden's index (0.441), and its recognition ability for positive samples was relatively weak. All models had relatively high specificity (0.976 – 0.991), indicating generally good recognition ability for negative samples, but the differences in Precision and F1 score reflected the performance divergence of different models in "reliability of predicting positives" and "precision-recall balance". SHAP analysis showed that the feature "venous access" occupied the most important position among all features, with a mean absolute SHAP value of about 0.012, which had the most significant impact on the model's prediction and may be the key factor in the model's decision-making. In addition, the AIS score had a mean absolute SHAP value of about 0.011, and the ISS was about 0.008, both of which significantly affected the model's prediction. Moreover, other features such as "truama index Score" and "residence time in emergency room" also played important roles in the model's prediction, although their impacts were relatively smaller. Conclusion: We analyzed the optimal model from 5 aspects: (1) receiver operating characteristic curve, (2) 5 indicators, (3) the trade-off between precision and recall, (4) scenario application: emergency settings, and (5) economic benefits. XGBoost was recommended as the first choice based on the performance, but external validation was required to ensure clinical applicability. SHAP analysis enhanced model transparency, helping rescue staff to understand the degree of influence of variables and provide a reference for clinical decision-making. Furthermore, key variables (such as ISS and AIS scores) aligned with medical consensus, verifying the credibility of the model. In the future, the researches should validate the model in real-world scenarios and continuously optimize features and algorithms to enhance practicality.

Purpose: It is often difficult for surgeons to accurately implant the cannulated screws for fixing femoral neck fractures under C-arm fluoroscopy. Our simulation study attempts to help surgeons identify the optimal positions of screws on anteroposterior and lateral radiographs. Methods: A retrospective study including patients from 2014.01.01 to 2017.12.31 was conducted. Softwares were used for 3-dimensional reconstruction and cross-sections construction of 60 femoral necks. The contours of all cross-sections along the axis of the femoral neck were stacked to identify a safe zone for screws placement. A simulation experiment was carried out with 3 circles of diameter 6.5 mm in the safe zone. The absolute and relative positions of every screw in the anteroposterior radiographs and lateral radiographs were measured Normal distribution, continuous data are presented in mean ± standard deviation, and analyzed by t-test. The category data are presented by n (%) and analyzed by the Chi-square test. Results: The morphologies of the safe zones for screws placement were anteriorly flat ellipses. The best dispersion was obtained with the inverted triangle shape in this zone: the inferior was positioned near the lower 1/10 on the anteroposterior view and near the anterior 3/5 on the lateral view. The anterosuperior and posteriosuperior screws were positioned near the upper 1/5 and 1/4, respectively, on the anteroposterior view, and near the anterior 1/10 and posterior 1/5, respectively, on the lateral view. Conclusion: Simulated screws implantation demonstrates the distribution characteristics of screws in anteroposterior and lateral radiographs. This can help decrease the risk of perforation and iatrogenic injury.

Purpose: To evaluate the prognostic performance of 4 CT scoring systems in pediatric patients with moderate-to-severe traumatic brain injury (TBI) and develop a simplified, interpretable predictive model based on machine learning. Methods: This retrospective study included 103 pediatric patients with moderate-to-severe TBI who were admitted to a tertiary children's hospital in Southwest China from September 2020 to December 2023. CT images were assessed using the Marshall score, Rotterdam score, Helsinki score, and Stockholm score. Clinical outcomes were defined by the Glasgow outcome scale (GOS) and categorized as favorable results (GOS 4–5) or unfavorable results (GOS 1–3). The dataset was divided into a training set (n=83) and a test set (n=20). Class imbalance was corrected using the random over-sampling examples method. Eight classification models were compared through 5-fold cross-validation. The Naive Bayes model showed the best performance and was simplified to include 5 key predictors. An interactive online application was developed to provide individualized prognostic estimation and visualization. Results: The Helsinki score demonstrated the highest predictive accuracy among the 4 CT scores (area under the curve (AUC)=0.906), followed by the Stockholm score (AUC=0.897), Rotterdam score (AUC=0.837), and Marshall score (AUC=0.764). The simplified Naive Bayes model achieved an AUC of 0.930, with 100% sensitivity and 65.9% specificity in the test set. The model enables real-time outcome prediction and visual interpretation of contributing factors. Conclusions: The simplified Naive Bayes model outperforms traditional CT scoring systems in predicting outcomes of pediatric moderate-to-severe TBI. Its easy use, interpretability, and web-based implementation support its potential for clinical application. Further prospective and multicenter studies are needed to validate these findings.

Purpose: Emergency thoracotomy (ET) is a life-saving procedure performed during emergency resuscitation of patients in extremis. The controversy around such an invasive and resource-consuming procedure still lingers after so many years of its introduction. ET can serve as one of several indicators of trauma center efficiency, providing information about the trauma system's overall preparedness, resources, and training. Our goal was to examine the clinical determinants of improved outcomes and review our ET experience over the past 8 years. Methods: All patients who underwent ET during the study period from January 2014 to December 2021 at our hospital were included in this retrospective study. Patients in whom the time of arrest was not known or who had incomplete data were excluded. Also, patients with fixed, dilated pupil on arrival with a significant blunt trauma were excluded. Patients undergoing urgent thoracotomy (not as a part of primary survey/resuscitation) were also not included in the study. All data were collected from the prospectively maintained hospital registry and patient case files. The primary outcome measure was in-hospital mortality and various parameters affecting it including but not limited to patient demographics, prehospital information, injury characteristics and scores, parameters of primary survey and perioperative data. The secondary outcome measures were the clinical spectrum of injury in patients with torso trauma who underwent ET including complications. The data was analysed using Fisher's exact test, Chi-square test, or Wilcoxon rank-sum test based on the category of variables. A Kaplan-Meier survival curve was plotted on relevant factors which determined clinical outcomes. Results: Sixty-seven patients underwent ET, with the majority (94.0%) being young males with a median age of 27 years. Penetrating injuries (61.2%) were more common than blunt trauma. Overall survival was 46.3%, with the best outcomes seen with penetrating stab (61.1%) injuries and patients with isolated intrathoracic injuries (67.5%). Patients who underwent resuscitative thoracotomy had dismal outcomes (mortality: 100%). On arrival, hemodynamic parameters like airway status, heart rate, systolic blood pressure, signs of life, and shock at presentation are statistically significant predictors of mortality (p < 0.05). The log-rank test for equality of survivor functions revealed the mechanism of injury (p = 0.010), the status of the airway (p = 0.002), shock on presentation (p = 0.001), and initial GCS (p = 0.040) to be significantly associated with mortality. Conclusion: ET can be a life-saving procedure with good outcomes, provided a careful selection of patients based on the mechanism of injury, location of major injury, and signs of life. It is prudent to have a system with proper protocol and a swiftly acting trauma team performing it to optimize the outcomes.