Minerva anestesiologica最新文献

Background: Ultrasound-guided regional anesthesia is commonly used to enhance perioperative analgesia. However, the technique is limited by inter-operator variability in identifying anatomical structures. Recent advances in artificial intelligence, particularly large language models, have raised the possibility of providing real-time interpretation support.

Methods: This prospective observational study was conducted at a tertiary-level hospital between 20/10/2024, and 28/02/2025. A total of 111 adult patients (aged 18-70) undergoing ultrasound-guided regional anesthesia were included. Ultrasound images were obtained intraoperatively during routine care. Exclusion criteria included poor image quality, incomplete clinical metadata, and non-standardized block protocols. Each image was submitted to ChatGPT-4 with standardized prompts requesting: 1) identification of the block type; 2) labeling of key anatomical landmarks; and 3) assessment of block success based on local anesthetic spread. Model outputs were compared to expert anesthesiologist assessments.

Results: A total of 147 ultrasound images were analyzed. ChatGPT-4 correctly identified anatomical landmarks in 141 cases (95.9%), classified block types in 107 cases (72.8%), and predicted block success in 138 cases (93.9%). Model performance was highest in commonly performed blocks such as transversus abdominis plane (TAP), erector spinae plane (ESP), and femoral nerve blocks. Accuracy was relatively lower in more complex or less frequently performed blocks.

Conclusions: ChatGPT-4 demonstrated high accuracy in identifying anatomical structures and predicting block success on ultrasound images. While performance in classifying certain block types was lower, these findings support the potential of large language models as decision-support tools in ultrasound-guided regional anesthesia.

Background: The Cormack-Lehane (CL) classification and the percentage of glottic opening (POGO) score are used for glottic views during direct laryngoscopy; however, no standard scoring exists for video-laryngoscopy. The Video Laryngoscopic Intubation and Difficult Airway Classification (VIDIAC) model was recently proposed to grade glottic views during video-laryngoscopy. This study aimed to evaluate the diagnostic accuracy of the VIDIAC score in predicting first-pass nasal intubation success using the Touren video laryngoscope in patients undergoing head and neck surgery.

Methods: After ethical approval and clinical trial registration, patients aged 18-65 years, planned for surgery under general anesthesia with nasal intubation, were recruited. Patient with a mouth opening ≤2.5 cm, planned for awake intubation, emergency procedure or rapid sequence induction was excluded. Nasal intubation was done using an appropriate size flexible endotracheal tube using Touren VL. VIDIAC scoring was done on the best view of the glottis in a single attempt, without using an external laryngeal maneuver (ELM). Use of ELM, Magill forceps and any airway-related adverse events were noted.

Results: A total of 131 participants completed the study. 121 (92.4%) patients had successful first-pass nasal intubation. For first-pass nasal intubation, the sensitivity, specificity, positive predictive values and negative predictive values of the VIDIAC score were 92.6%, 70%, 97.5%, and 43.8%, respectively. The VIDIAC score demonstrated an area under the Receiver Operating Characteristic curve of 0.876 (95% CI: 0.728-1.000).

Conclusions: With high sensitivity and positive predictive value, the VIDIAC score shows potential in predicting successful first-pass nasal intubation; however, low specificity and negative predictive value limit its overall clinical utility.

Background: Invasive pulmonary aspergillosis (IPA) is increasingly recognized in non-neutropenic patients, where coexisting bacterial infections, particularly with Gram-negative pathogens, may impair susceptibility. However, validated tools for early risk stratification in this population remain unavailable.

Methods: We retrospectively analyzed 437 non-neutropenic adults with bacterial co-infection (derivation N.=331; validation N.=106) admitted between 2019 and 2024. Independent predictors of IPA were identified through multivariable logistic regression and incorporated into both a weighted clinical risk score and an ensemble machine learning (ML) model. Model performance was assessed using discrimination, calibration, and decision curve analysis, with subgroup validation in Gram-negative infection, intensive care unit (ICU) admission, and diabetes.

Results: Seven independent predictors of IPA were identified: nodular shadow, chronic respiratory disease, Gram-negative infection, corticosteroid exposure, ICU admission, smoking history, and diabetes. Gram-negative pathogens accounted for nearly half of infections, with Pseudomonas aeruginosa predominating. The ensemble score achieved a strong performance (area under the curve [AUC] 0.922 derivation; 0.862 validation) with superior calibration compared to traditional approaches. Risk stratification at a threshold score of ≥7.5 significantly enriched 28-day IPA incidence (log-rank P<0.001). Subgroup analyses confirmed score robustness in Gram-negative infection (AUC=0.898), ICU admission (AUC=0.888), and diabetes (AUC=0.914). However, the predictive contributions of respiratory disease and corticosteroid exposure were attenuated in diabetic patients.

Conclusions: Gram-negative bacterial co-infection synergistically amplifies IPA risk in non-neutropenic patients. The ensemble ML model integrating seven pragmatic predictors provides accurate, interpretable, and clinically actionable stratification, enabling precision prophylaxis and early antifungal intervention. Prospective multicenter validation is warranted before clinical implementation.

Background: Postoperative nausea and vomiting (PONV) remain a significant clinical challenge in pediatric squint surgeries, with potential serious complications. This randomized controlled trial aimed to evaluate the efficacy of preoperative ondansetron lozenges in preventing PONV among pediatric patients undergoing squint surgeries.

Methods: Eighty pediatric patients (aged 4-15 years) undergoing squint surgeries were randomly assigned to two equal groups: Group S received 4 mg ondansetron lozenge for two hours preoperatively, while Group C served as the control.

Results: The ondansetron lozenge group demonstrated significantly lower PONV incidence (22.5% vs. 52.5%, P=0.006, with relative risk (RR) of 0.43 (95% CI 0.22:0.82), with reduced severity across 0-2-, 2-12-, and 12-24-hour intervals. Time to onset of emesis was prolonged (269.11±214.87 vs. 115.24±65.68, P=0.005), and the oculocardiac reflex incidence was reduced (12.5% vs. 37.5%, P=0.018). Parental satisfaction was significantly higher in the Ondansetron group (8.95±1.38 vs. 7.83±1.99, P=0.004).

Conclusions: Preoperative ondansetron lozenges effectively reduce PONV incidence, severity, and complications in pediatric squint surgery, suggesting a promising prophylactic intervention.