Anesthesiology and Perioperative Science最新文献

Purpose

To compare ChatGPT, a general-purpose large language model (LLM), with OpenBioLLM, a domain-specific biomedical model, in generating clinically appropriate anesthesia plans, and to assess the impact of advanced prompt engineering (PE).

Methods

A comparative observational study analyzing anonymized clinical records of 100 cardiac surgery patients using three LLMs under simple querying and PE. Plans were evaluated by anesthesiologists using a double-blind approach. The main outcome measures included clinical alignment, reasoning quality, medication selection, dosage accuracy, omissions, and risk of harm. Scores were rated on a 5-point Likert scale by both experienced and trainee anesthesiologists, and differences analyzed via paired t-tests and analysis of variance.

Results

In clinical matching accuracy, logical reasoning, medication selection, and safety evaluation metrics, ChatGPT consistently outperformed OpenBioLLM. GPT-4o demonstrated superior performance compared to GPT-3.5, with significant performance enhancements achieved through prompt optimization. Both physician groups reported notable improvements in their average scores for the ChatGPT model when utilizing advanced PE, whereas OpenBioLLM showed comparatively smaller score improvements.

Conclusion

ChatGPT’s adaptability and clinical accuracy, enhanced by PE, make it a valuable tool for anesthesia planning, especially in resource-limited settings. However, over-reliance on AI by less experienced clinicians poses risks, underscoring the need for physician oversight and tailored training. Further research is necessary to validate these findings across diverse clinical scenarios.

Since its introduction in 2002, transcatheter aortic valve replacement (TAVR) has gained rapid worldwide adoption, significantly increasing the survival rates of high-risk surgical patients. As key members of the multidisciplinary TAVR team, anesthesiologists play a crucial role in perioperative care. Initially, anesthesiologists administered general endotracheal anesthesia, maintained hemodynamic stability, monitored cardiac function with transesophageal echocardiography, and managed adverse events. Improvements in techniques and valve design have expanded TAVR indications, enabling minimally invasive procedures. As a result, anesthesia practices have shifted from general anesthesia to monitored anesthesia care or local anesthesia. Today, anesthesia efforts focus on refining perioperative strategies to improve patient comfort while reducing complications, length of hospital stay, costs, and mortality. These developments highlight the essential role of anesthesiologists in optimizing TAVR outcomes.

Ancient China has a rich history of acupuncture anesthesia (AA), with both theoretical and clinical foundations that were notably effective in the 1970s. This paper provides an in-depth analysis of AA's historical context and clinical applications, juxtaposing it with conventional anesthesia methods. It also aims to forecast the future trajectory of AA and examine the factors influencing its evolution. Despite experiencing a decline after its peak—primarily due to the precision, effectiveness, visibility, and cost-effectiveness of conventional anesthesia—AA's potential remains significant. Its application in perioperative period has been demonstrated to significantly reduce side effects associated with traditional methods, leading to improved clinical outcomes. However, in recent years, the concept of AA has been associated with perioperative acupuncture. There are numerous successful AA cases reported worldwide, and the field is in urgent need of continuously being evaluated and refined to keep pace with modern medical advancements and societal needs. This nonsystematic narrative review manifests the potential for further development in AA is considerable, and ongoing research and integration efforts suggest a promising future for this ancient practice in contemporary medicine.

Patients with chronic liver disease (CLD) are at increased risk of adverse perioperative outcomes. Effective preoperative risk stratification and optimization are critical to minimizing perioperative complications. This article summarizes key aspects of perioperative care in patients with CLD, including pathophysiology, preoperative risk assessment, and evidence-based perioperative care strategies. For patients with hepatocellular carcinoma require liver resection, a multidisciplinary approach is recommended to optimize surgical candidacy and outcomes. In addition, altered bioavailability of opioids and other analgesics necessitates individualized postoperative pain management, emphasizing multimodal approaches tailored to hepatic function.

Graphical Abstract

Purpose

Patients undergoing transurethral prostate surgery often experience catheter-related bladder discomfort (CRBD), which can significantly impairs postoperative recovery. The anatomical convergence between the ilioinguinal and iliohypogastric (II-IH) nerve pathways (T12–L2) and the bladder innervation provides a rationale for evaluating the II-IH nerve block as a potential treatment. This study aimed to investigate the efficacy of ultrasound-guided II-IH nerve block versus intravenous tramadol in reducing postoperative CRBD.

Methods

Ninety-four patients undergoing transurethral prostate surgery were equally randomized to receive either 1.5 mg/kg intravenous tramadol or bilateral ultrasound-guided II-IH nerve block with 15 mL of 0.375% ropivacaine per side before extubation. Intravenous tramadol was used as rescue medication for moderate-to-severe CRBD. The primary outcome was the overall incidence of CRBD within postoperative 24 h. Secondary outcomes included CRBD severity, pain scores, adverse events, patient satisfaction, hospitalization duration, and International Prostate Symptom Scores (IPSSs).

Results

The overall incidence of CRBD within postoperative 24 h was 56.5% in the II-IH block group versus 65.2% in the tramadol group, with an absolute risk difference of −8.70% (95% confidential interval: −24.71% to 10.92%; p = 0.393). Secondary outcomes, including the incidence of moderate-to-severe CRBD, pain scores at all assessed timepoints, rescue tramadol use, length of hospitalization, and IPSSs were comparable between the two groups (all p > 0.05). However, the II-IH block group demonstrated significantly lower incidences of adverse events (including nausea and vomiting, headache, dizziness, and dry mouth) and a higher proportion of extremely satisfied patients (all p < 0.05). Importantly, no nerve block-related complications were observed in the II-IH block group.

Conclusions

Ultrasound-guided II-IH nerve block appeared to be a comparable alternative to intravenous tramadol for reducing CRBD after transurethral prostate surgery, while offering a more favorable safety profile and higher patient satisfaction.

Trial registration

Chinese Clinical Trial Registry, ChiCTR2300078552. Registered 12 December 2023, https://www.chictr.org.cn/showproj.html?proj=209712

Purpose

Postoperative delirium (POD) is common in older adults following major surgery. Ketamine has opioid-sparing properties and potential neuroprotective effects, yet whether intraoperative ketamine reduces the incidence of postoperative delirium remains uncertain. This systematic review aims to determine whether ketamine use is associated with a lower incidence of POD in older patients undergoing spine or orthopedic surgery.

Methods

We conducted comprehensive searches in PubMed, Web of Science, EMBASE, Cochrane CENTRAL, and CINAHL Complete from inception to June 26, 2025. Both keywords and index terms (Mesh/EMTREE) were used to develop broad and sensitive searches tailored to each database to retrieve all relevant articles. EMBASE results were limited to articles and articles in press. Eligible studies included patients aged ≥ 60 years undergoing surgery with intraoperative ketamine administration and reporting POD as an outcome. Two reviewers independently screened studies, extracted data, assessed risk of bias, and performed quality appraisal using RoB 2 for randomized controlled trials (RCTs) and ROBINS-I for observational studies. Random-effects meta-analyses were performed to calculate odds ratios (ORs) with 95% confidence intervals (CIs).

Results

We included 24 studies in the final analysis and five specifically studied patients undergoing spine or orthopedic surgery. The effect of ketamine on POD showed an OR of 1.50 (95% CI 0.38–5.91, P = 0.56 random effects model) for the two RCTs. Including the additional three cohort studies was marginally associated with an increase in the occurrence of postoperative delirium, OR 1.30 (95% CI 1.01–1.67), P = 0.0425. However, the imprecision of the studies was deemed to be serious due to small sample sizes, and only two were RCTs.

Conclusions

This systematic review and meta-analysis did not find a statistically significant association between intraoperative ketamine use and POD in older patients undergoing spine or orthopedic surgery in the RCTs. However, because of the small number of available studies together with imprecision, this conclusion should only be considered as preliminary. Larger, high-quality randomized trials are needed to determine whether ketamine influences POD risk and to assess different dosing strategies for their effects on postoperative cognitive outcomes in older adults.

This review provides an evidence-based overview of procedural sedation and analgesia (PSA) practices in emergency department settings, addressing pharmacologic agent selection, monitoring strategies, pediatric- and adult-specific considerations, and current guideline recommendations to support safe and effective PSA across diverse clinical scenarios. Literature published between 2014 and 2024 was reviewed using PubMed, Cochrane Library, Embase, Scopus, and Web of Science, with inclusion limited to English-language, full-text human studies of methodological quality and clinical relevance. Ketamine, propofol, midazolam, and fentanyl remain the most commonly used agents, often administered in combination to balance efficacy and adverse effects, while newer agents such as remimazolam, esketamine, and ciprofol are emerging but lack robust evidence in emergency settings. In pediatric populations, noninvasive administration routes and age-appropriate dosing are emphasized, and adherence to guidelines is consistently associated with improved procedural standardization and patient safety. Point-of-care ultrasound has gained increasing prominence in monitoring strategies, complementing established practices. Overall, when guided by appropriate patient selection, careful pharmacologic planning, and adherence to evidence-based protocols, PSA can be safely and effectively performed in emergency care, and the integration of standardized training, clinical guidelines, and novel monitoring tools will be critical for optimizing outcomes across diverse patient populations. 

Anaesthesiology is evolving from a clinical-based discipline focused primarily on maintaining intraoperative safety into an evidence-based perioperative medicine that emphasizes comprehensive patient care throughout before, during and after surgery journey. This shift has given rise to several important interdisciplinary research directions, which warrant systematic summarisation to better define future perspectives. In this review, we discuss four key intersecting domains: anaesthesia and neuroscience, perioperative organ protection, anaesthesia and cancer, and anaesthesia therapy. Furthermore, based on emerging establishment of perioperative medicine databases that integrate diverse-source data and leverage artificial intelligence for large-scale data analysis, we propose four new perspectives of anaesthesiology including personalized and perioperative management strategy, anaesthetic and organ-protective drugs, anaesthesia techniques and intelligent devices.

Graphical Abstract

Purpose

Sepsis is a complex and life-threatening condition that arises from an abnormal and dysregulated host response to an infection. Developing a practical and interpretable machine learning model to predict mortality in patients with sepsis is essential. To address the limitations of existing sepsis mortality prediction models (insufficient validation and limited interpretability), we developed a streamlined extreme gradient boosting (XGBoost) model that predicts mortality risk using only 7 routine clinical features.

Methods

Medical Information Mart for Intensive Care-IV (MIMIC-IV) data were used to train and internally validate our model. External validation was performed using the eICU-Collaborative Research Database (eICU-CRD). The area under the receiver operating characteristic curve (AUC) was used as the key evaluation index. Differences in model performances were assessed using DeLong's test. The optimal model was simplified using a recursive feature elimination algorithm, and model interpretability was evaluated using SHAP.

Results

Our 7-feature model demonstrated strong performance across training (AUC 0.920), internal validation (AUC 0.900) and external validation (AUC 0.839) cohorts. The selected features—platelet count (at discharge), leukocyte levels (post-antibiotic therapy and at discharge), BUN (post-antibiotic therapy), anion gap (post-antibiotic therapy and at discharge), and ICU length of stay—are objective (eliminating subjective/recall bias) and readily available in clinical practice. SHAP analysis showed that the prognostic contributions of these variables differed before and after antibiotic treatment.

Conclusion

The XGBoost model, built on 7 routine and objective clinical features, quantifies inter-feature relationships to provide mortality risk and treatment thresholds, thereby enabling clinicians to formulate evidence-based therapeutic strategies.

Purpose

Emerging evidence suggests that retinal changes may indicate brain disorders. However, the association between retinal nerve fiber layer (RNFL) thickness and delirium risk has not been investigated. This study examined whether reduced RNFL thickness could predict an increased risk of delirium.

Methods

A total of 42,449 participants without a history of delirium were recruited from the UK Biobank. Optical coherence tomography (OCT) was used for RNFL thickness measurement. Cox proportional hazard models were applied to calculate hazard ratios and 95% confidence intervals for incident delirium risk.

Results

Over a median follow-up of 13.7 years, 1.0% of participants (425) developed delirium. Each 5-μm decrease in RNFL thickness was associated with a 13.6% higher hazard of delirium. The lowest RNFL tertile was linked to a 47.7% higher hazard compared to the highest tertile. A similar correlation was observed for postoperative delirium.

Conclusion

Reduced RNFL thickness is associated with increased delirium risk, highlighting the potential of OCT-based RNFL measurements as a non-invasive tool to identify potential delirium patients.