Artificial intelligence (AI) is driving a profound transformation in the paradigm of surgery. Before the operation, AI promotes decision-making from intuitive experience to data-driven through deepening imaging diagnosis, empowering imaging-based predictive analytics, and providing dynamic interactive surgical planning. During the operation, AI serves as real-time assistance, enhancing surgeons’ perception, operation, and decision-making capabilities in areas such as computer vision navigation, surgical process analysis, robotic intelligence, and intraoperative rapid diagnosis. Post-operatively, AI prediction models based on continuous data streams can achieve early warning and proactive prevention of complications and empower home-based rehabilitation. Additionally, AI is transforming the surgical education model through objective skill assessment, knowledge empowerment by large language models, and high-fidelity simulation. However, the clinical translation of AI still faces core challenges such as insufficient clinical research evidence, data and algorithm bottlenecks, difficulties in system integration, and the need for improved ethical and legal regulatory frameworks. Future trends will involve building surgical foundation models, developing patient digital twins, and constructing human-machine integrated intelligent operating rooms, ultimately reshaping clinical surgery from “artisanal craftsmanship” to “intelligent science”.
Advances in the diagnosis and treatment of critical surgical patients stem from the integration of critical care medicine, general surgery, and multidisciplinary collaboration. Progress in Intensive Care Medicine (ICU) includes fluid resuscitation, mechanical ventilation, continuous renal replacement therapy, extracorporeal membrane oxygenation and point-of-care ultrasound. The De-escalation concept helps avoid complications such as ventilator-associated pneumonia and catheter-related infections from prolonged support care. The widespread application of damage control surgery has significantly improved the survival rate of critically ill surgical patients. Surgical interventions for critically ill patients encompass laparotomy, percutaneous drainage, removal or replacement of drainage tubes, bedside debridement and hemostasis, open abdomen therapy, and vascular interventional procedures. Planned laparotomy and staged surgeries are particularly suitable for critically ill abdominal surgery patients whose conditions remain unstable. Surgeons should select the appropriate timing and surgical approaches based on the patient’s condition. The key to treating surgical critically ill patients lies in the active cooperation of surgeons and ICU physicians. Acute care surgery has developed, and acute care surgeons are trained to specially treat surgical critical illness, trauma and emergency surgical patients. This integrated model has effectively enhanced surgeons’ ability to treat critically ill surgical patients and is worthy of promotion.
Pulmonary embolism during the perioperative period of general surgery elevates mortality risk. Individual and treatment-related factors, underlying diseases, and the awareness of surgeons to bleeding and thromboembolism risk evaluation may influence the diagnostic assessment, prevention, and treatment strategy of pulmonary embolism. The symptoms, signs, D-dimer levels, electrocardiogram, and ultrasound findings of pulmonary embolism lack specificity. Computed tomography pulmonary angiography (CTPA) may not be feasible in the patients with severe hemodynamic instability, thereby affecting diagnostic accuracy. Therefore, it is essential to optimize the diagnostic protocol, appropriately determine the timing of anticoagulation and reperfusion therapy, and standardize the perioperative management of anticoagulant medications, including discontinuation, bridging, and resumption. Furthermore, indications for interventional treatments such as catheter-directed thrombolysis and inferior vena cava filter placement should be strictly limited.
Intra-abdominal infection (IAI) is a common infectious disease in surgical practice, posing a substantial disease burden. In recent years, the isolation rates of carbapenem-resistant Gram-negative bacteria, vancomycin-resistant Enterococci, and azole-resistant Candida have shown an increasing trend. Concurrently, advances in pathogen detection methods such as next-generation sequencing and the development of novel antimicrobial agents have provided new strategies for the diagnosis and treatment of IAI caused by drug-resistant pathogens. The predominant pathogens involved in IAI are enteric Gram-negative bacteria, Gram-positive bacteria, anaerobic bacteria, and Candida species, often presenting as polymicrobial infections. Therapeutic approaches for IAI include empirical antibiotic regimens based on patient stratification, followed by a targeted treatment strategies against common IAI pathogens. The recently approved agents in China, including novel β-lactam/ β-lactamase inhibitor combinations, newer tetracycline derivatives, and contezolid, provide new treatment options for IAI.
The incidence of moderate-to-severe pain following abdominal surgery remains unacceptably high. The conventional passive analgesic paradigm dominated by opioids ignores the complexity of pain mechanisms and interindividual variability among patients, making it difficult to meet the evolving demands of enhanced recovery after surgery (ERAS), including attenuating surgical stress, facilitating recovery, and improving clinical outcomes. Precision analgesia, developed on the foundation of multimodal analgesia, integrates individualized considerations based on patient characteristics, disease profiles, and temporal factors. It emphasizes comprehensive perioperative management, accurate patient stratification, mechanism-based selection of analgesic agents and techniques, and the establishment of recovery-oriented therapeutic goals. Drawing on the practical experience of our pain-free ward, the clinical pathways for implementing precision analgesia should include preoperative assessment and risk stratification, patient education and psychological optimization, application of intraoperative regional anesthesia techniques, dynamic adjustment of postoperative multimodal analgesia, and structured post-discharge management. The successful implementation of precision analgesia relies on a multidisciplinary and patient-centered approach, facilitating the transition of perioperative pain management in abdominal surgery from empirically driven practices toward individualized and precision-based strategies.
Critically ill patients in general surgery often present with complex conditions and rapid clinical deterioration, frequently involving severe infection and multiple organ dysfunction. Traditional severity scoring systems can partially reflect disease severity but rely mainly on static indicators and lack dynamic predictive capability. With the development of medical big data and artificial intelligence (AI), machine learning and deep learning-based models have been increasingly applied in critical care medicine. These technologies show great potential in early disease detection, risk prediction, and clinical decision support. By integrating electronic health records, vital signs, laboratory results, and imaging data, AI can continuously analyze patient conditions and provide real-time predictions to assist clinicians in making timely decisions. AI has promising prospects in the application in early warning systems, complication prediction, and decision support for critically ill patients in general surgery. However, current challenges such as data quality, model interpretability, and ethical issues are discussed to provide references for the standardized application of AI in surgical critical care.
Critically ill patients after general surgery often experience hypermetabolic state and malnutrition due to surgical trauma, infection, and organ dysfunction. Nutritional therapy has become a crucial component of comprehensive treatment. Nutritional therapy for these patients includes phased metabolic regulation strategies, individualized protein targets, and precision decision-making driven by the assessment of muscle reserves based on body composition. Regarding nutritional pathways, early enteral nutrition remains the first choice. However, routine excessive monitoring of gastric residual volume should be abandoned in favor of a comprehensive evaluation guided by clinical tolerance. Parenteral nutrition, as a necessary supplement, should adhere to the principle of “enteral first, delayed combination, and adequate protein”. Furthermore, immunomodulatory nutrients, such as omega-3 fatty acids, hold promising application prospects in specific subgroups of patients.
Postoperative gastrointestinal dysfunction (POGD) is a common postoperative complication in patients undergoing general surgery, characterized by gastrointestinal mucosal injury, impaired barrier function, and gastrointestinal dysmotility. Clinical manifestations include nausea, vomiting, abdominal distension, and delayed passage of flatus or stool; in severe cases, it may lead to gut-derived infection and multiple organ dysfunction. The pathogenesis of POGD is complex, and there is currently no standardized diagnostic criterion. Although progress has been made in recent years in understanding and managing POGD, effective therapeutic strategies remain limited. Current approaches for the prevention and treatment of POGD mainly include reducing surgical trauma, implementing multimodal analgesia, optimizing perioperative fluid management, and adopting enhanced recovery after surgery strategies such as early enteral feeding. These measures are often combined with pharmacological interventions, including prokinetic agents, opioid receptor antagonists, as well as traditional Chinese medicine.
Objective To compare the effects of simulated surgical planning (based on preoperative CT three-dimensional reconstruction) combined with multimodal navigation (intraoperative ICG fluorescence and LUS) versus simulated surgical planning alone on liver function preservation and short-term outcomes during laparoscopic precise hepatectomy for primary liver cancer. Methods Clinical data of 82 consecutive patients with primary liver cancer admitted to the First Hospital of Jiaxing City between March 2021 and February 2024 were retrospectively analyzed. According to the period of technology application, patients were divided into a multimodal navigation group, 38 patients who received preoperative CT three-dimensional reconstruction combined with intraoperative indocyanine green fluorescence and laparoscopic ultrasound (LUS) guidance between September 2022 and February 2024, and a planning-only group, 44 patients who received only preoperative CT three-dimensional reconstruction guidance between March 2021 and August 2022. Liver function parameters (before surgery, on postoperative day 1 and day 7), operative time, blood loss, and complications were compared between the two groups. Results No statistically significant differences were found in baseline characteristics between the two groups (P>0.05). Before surgery, there were no significant differences in total bilirubin, alanine aminotransferase, aspartate aminotransferase, and albumin levels between the two groups (P>0.05); on postoperative day 7, albumin levels also showed no significant difference (P>0.05). Compared with patients in the planning-only group, patients in the multimodal navigation group had significantly lower levels of total bilirubin, alanine aminotransferase, and aspartate aminotransferase on postoperative day 1 and day 7, and significantly higher albumin level on postoperative day 1 (P<0.05). Patients in the multimodal navigation group also had significantly shorter operative time, duration of drain retention, and postoperative hospital stay, less intraoperative blood loss, and a lower rate of blood transfusion than those in the planning-only group (P<0.05). The overall postoperative complication rates were 13.2% in the multimodal navigation group and 31.8% in the planning-only group, with the multimodal navigation group showing a significantly lower rate (P<0.05). Conclusion Compared with laparoscopic hepatectomy guided by simulated surgical planning alone, simulated surgical planning combined with multimodal intraoperative navigation is a more effective treatment for primary liver cancer, which enables precise resection while better preserving hepatic function.
Objective To evaluate the impact of laparoscopic-endoscopic cooperative surgery combined with sentinel lymph node navigation surgery (LECS-SNNS) on gastric emptying function and quality of life (QoL) in patients with early gastric cancer (EGC). Methods A prospective randomized controlled trial (RCT) design was adopted. A total of 100 patients with EGC who were diagnosed and treated at Beijing Friendship Hospital Affiliated to Capital Medical University between August 2023 and August 2025 were consecutively enrolled, and randomly assigned at a 1∶1 ratio to the LECS-SNNS group (n=50) and the laparoscopic standard gastrectomy (LSG) group (n=50). We evaluated baseline demographics, perioperative parameters, nutritional status, QoL, and gastric emptying function. Specifically, gastric emptying half-time (GET1/2) was assessed via radionuclide scintigraphy. Results In the LECS-SNNS group, 4 cases were confirmed to have lymph node metastasis during the operation and were converted to LSG. Ultimately, 46 cases in the LECS-SNNS group were analyzed.There were no statistically significant differences in baseline characteristics between the two groups (P>0.05). LECS-SNNS showed shorter operative time (165.1 min vs. 206.2 min), reduced blood loss (21.5 mL vs. 52.3 mL), and faster recovery (flatus, diet, and discharge) compared to LSG, with statistically significant differences (all P<0.05). Regarding gastric emptying function, at 3 months postoperatively, there was no statistically significant difference in the GET1/2 between the LECS-SNNS group and its preoperative level (67.0 min vs. 57.0 min, P>0.05). In contrast, the GET1/2 of the LSG group was significantly prolonged compared with its preoperative level, with a statistically significant difference (129.0 min vs. 56.5 min, P<0.05). The nutritional assessment showed that at 3 months postoperatively, the levels of albumin and hemoglobin in both groups recovered well; however, the Prognostic Nutritional Index (PNI) of the LECS-SNNS group was significantly higher than that of the LSG group (P<0.05). At 3 months postoperatively, the LECS-SNNS group showed superior QoL scores compared to the LSG group, with significant improvements in global health status, functional scales, physical/mental component summaries, reflux symptoms, eating restrictions, meal size, and total symptom scores, with statistically significant differences (all P<0.05). No recurrence occurred during follow-up. Conclusion In strictly selected EGC patients, LECS-SNNS achieves short-term oncological safety with the benefits of functional preservation and minimal invasiveness, leading to improved gastric function recovery and short-term QoL. Thus, LECS-SNNS represents a viable alternative to LSG for individualized EGC treatment, optimizing the balance between oncological cure and functional retention.
Objective To evaluate the safety, oncological efficacy, and long term survival impact of concomitant vascular resection (VR) in the surgical treatment of retroperitoneal liposarcoma (RPLS). Methods Clinical and pathological data of 622 patients with retroperitoneal liposarcoma (RPLS) treated at the Department of Retroperitoneal and Soft Tissue Tumor Surgery, Zhongshan Hospital, Fudan University, between January 2010 and December 2023 were retrospectively analyzed. All the patients were enrolled in the study and divided into a non vascular resection group (non VR, n=533) and a vascular resection group (VR, n=89) based on whether concomitant VR was performed intraoperatively. Propensity score matching (PSM) at a 1∶1 ratio was applied, yielding 89 patients in each group. Clinical data was compared between the two groups. Results After PSM, the VR group exhibited significantly greater intraoperative blood loss, allogeneic blood transfusion volume, operative time, intensive care unit (ICU) length of stay, total hospital stay, and incidence of postoperative vascular related complications compared with the non VR group (P<0.05). No statistically significant differences were observed between the two groups in 30 day postoperative mortality or the proportion of patients with Clavien-Dindo grade ≥ Ⅲ complications (P>0.05). Multivariate Cox regression analysis identified R2 surgical margin as an independent risk factor for overall survival (OS) (R2 vs. R0/R1: HR=4.436, 95%CI 2.223-8.850, P<0.001). FNCLCC grade Ⅱ and Ⅲ were independent risk factors for both OS (grade Ⅱ vs. grade Ⅰ: HR=3.543, 95%CI 1.322-9.492, P=0.012; grade Ⅲ vs. grade Ⅰ: HR=5.015, 95%CI 1.794-14.021, P=0.002) and recurrence free survival (RFS) (grade Ⅱ vs. grade Ⅰ: HR=2.345, 95%CI 1.212-4.536, P=0.011; grade Ⅲ vs. grade Ⅰ: HR=3.785, 95%CI 1.817-7.885, P=0.002). The median OS in the non VR and VR groups was 60 months and 63.8 months, respectively; the median RFS was 20 months and 25.2 months, respectively. No statistically significant differences were found between the two groups in OS or RFS (OS: HR=0.96, 95%CI 0.61-1.52, P=0.872; RFS: HR=0.96, 95%CI 0.67-1.39, P=0.835). Conclusion Although concomitant VR is associated with increased surgical trauma and a higher risk of vascular related complications, it does not result in elevated rates of severe postoperative complications or 30 day mortality. Although VR itself is not an independent prognostic factor, it provides an opportunity for R0/R1 resection in locally advanced RPLS involving major vessels.
Objective To investigate the clinical characteristics and prognosis of patients with resectable pancreatic ductal adenocarcinoma (PDAC) exhibiting different enhancement patterns on contrast-enhanced computed tomography (CECT). Methods Clinical data of 148 patients with resectable PDAC who underwent surgical treatment at the Department of Pancreatic and Metabolic Surgery, Nanjing Drum Tower Hospital Clinical College of Nanjing Medical University, between March 2019 and March 2023 was retrospectively analyzed. Independent prognostic risk factors were analyzed. Patients were stratified into low-enhancement and high-enhancement groups based on the optimal cutoff value of ratio3. Survival outcomes and clinical characteristics were then compared between the two groups. Results The portal enhancement ratio, defined as the CT value of tumor-adjacent normal pancreatic parenchyma divided by the CT value of the tumor during the portal phase of CECT (ratio3), was identified as an independent risk factor for both overall survival (OS) and disease-free survival (DFS) in 148 patients. ROC curve analysis revealed that the cut-off value of ratio3 was 1.35, with an area under the curve (AUC) of 0.735 (P<0.001). The median OS and median DFS for all patients were 29.0 months and 21.0 months, respectively. Patients in the low enhancement group (ratio3>1.35, n=110) presented a median OS of 24.0 months and a median DFS of 15.0 months. In contrast, patients in the high enhancement group (ratio3≤1.35, n=38) exhibited a significantly more favorable prognosis, with both median OS and DFS not reached (P<0.001). Significant differences were also observed between the low enhancement group and the high enhancement group in the incidence of preoperative abdominal pain, tumor size and Ki-67 expression levels (all P<0.05). Conclusion Tumor enhancement characteristics on preoperative CECT can serve as a promising prognostic indicator for patients with resectable PDAC. High tumor enhancement is correlated with superior long-term survival, smaller tumor lesions, a lower incidence of preoperative abdominal pain, and reduced Ki-67 expression levels.
Liver-related coagulation dysfunction is commonly observed in liver cirrhosis, hepatocellular carcinoma, sepsis-associated liver injury, and post-hepatectomy states. The underlying pathology is a remodeling and dysregulation of the dynamic equilibrium among procoagulant, anticoagulant, and fibrinolytic systems, often manifesting as coexisting bleeding and thrombosis. The traditional coagulation cascade model and routine coagulation tests primarily focus on changes in coagulation factor levels, which fail to fully capture the multicellular synergistic regulation and dynamic hemostatic status within the hepatic microenvironment. Recent advances in the cell-based coagulation model, immunothrombosis theory, and hepatic sinusoidal biology multicellular interaction research have revealed that liver hemostasis regulation is a dynamic network involving hepatocytes, liver sinusoidal endothelial cells, Kupffer cells and inflammatory signals. Under physiological conditions, liver is responsible for the synthesis, supply, and clearance of coagulation factors. Under pathological conditions, this network undergoes remodeling, characterized by disrupted hepatic sinusoidal multicellular interactions and an imbalance in coagulation factor synthesis-clearance homeostasis. Further perspectives should reconceptualize liver hemostatic regulation from a multicellular network and dynamic systems viewpoint, integrating molecular biology techniques with dynamic functional monitoring to establish more precise assessment and intervention strategies.
Driven by a dual burden of infectious and non-infectious factors, severe infections exhibit persistently high mortality rates. Simply treating infections and providing organ support is often limited in its ability to prevent disease progression; systemic immune modulation may play a role as a potentially important strategy for improving outcomes. Although source control and antimicrobial treatments remain foundational, conventional “one-size-fits-all” anti-inflammatory approaches lack specificity and may exacerbate late-stage acquired immunosuppression. Consequently, personalized precision therapies guided by dynamic immune monitoring have gained substantial traction. Interventions such as extracorporeal blood purification to attenuate early inflammatory peaks, precise supplementation with lymphocyte proliferative agents and myeloid cell activators to reverse immune paralysis, immune checkpoint inhibitors for dual regulation, and metabolic reprogramming for exhausted immune cells are highly promising. Integrating these tailored strategies with damage control surgery principles provides a novel framework for reducing mortality and enhancing long-term recovery in critically ill surgical patients.
Against the backdrop of the persistent global shortage of liver graft resources, living-donor liver re-transplantation has become the core salvage treatment modality for patients with graft dysfunction after primary liver transplantation. Compared with deceased-donor liver re-transplantation, it possesses unique clinical advantages in terms of graft accessibility, surgical complications, and long-term survival prognosis. Biliary complications, vascular embolism, chronic rejection, and recurrence of the primary disease are the main inducements requiring re-transplantation; severe intra-abdominal adhesions, complex vascular and biliary reconstruction remain the major technical difficulties during surgery. With the application of ICG molecular fluorescence imaging technology and minimally invasive techniques, surgical complications have been reduced, and the long-term survival rate of patients has been significantly improved.
