Background: Machine-learning techniques are useful for creating prediction models in clinical practice. This study aimed to construct a prediction model of postoperative 30-day mortality based on an automatically extracted electronic preoperative evaluation sheet. Methods: We used data from 276,341 consecutive adult patients who underwent non-cardiac surgery between January 2011 and December 2020 at a tertiary center for model development and internal validation, and another dataset from 63,384 patients between January 2011 and October 2021 at another center for external validation. Postoperative 30-day mortality was 0.16%. We developed an extreme gradient boosting (XGB) prediction model using only variables from preoperative evaluation sheets. Results: The model yielded an area under the curve of 0.960 and an area under the precision and recall curve of 0.216, which were 0.932 and 0.122, respectively, in the external validation set. The optimal threshold calculated by Youden’s J statistic had a sensitivity of 0.885 and specificity of 0.914. In an additional analysis with balanced distribution, the model showed a similar predictive value. Conclusion: We presented a machine-learning prediction model for 30-day mortality after non-cardiac surgery using preoperative variables automatically extracted from electronic medical records and validated the model in a multi-center setting. Our model may help clinicians predict postoperative outcomes.

Interleukin 2 (IL-2) influences the development and severity of pain due to its antinociceptive and immunomodulatory effects. Its production is influenced by the increased expression of c-Cbl (Casitas B-lineage lymphoma proto-oncogene) and Cbl-b E3 ubiquitin ligases. We evaluated the effects on IL-2-mediated changes in c-Cbl and Cbl-b expression in a rat model of chronic neuropathic pain. Peripheral neuropathy was induced in adult male Sprague-Dawley rats weighing 250–300 g by chronic spinal nerve ligation. Half of the spinal cord ipsilateral to the nerve injury was harvested at 1, 3, and 6 weeks, and the expression levels of IL-2, c-Cbl, Cbl-b, phospholipase C-γ1 (PLC-γ1), ZAP70, and protein kinase Cθ (PKCθ), as well as ubiquitin conjugation, were evaluated. Total IL-2 mRNA levels were significantly decreased at 3 and 6 weeks after nerve injury compared to those in sham-operated rats. The mRNA levels of c-Cbl and Cbl-b, as well as the level of ubiquitin conjugation, were significantly increased at 3 and 6 weeks. In contrast, the levels of phosphorylated ZAP70 and PLC-γ1 were decreased at 3 and 6 weeks after spinal nerve ligation. Ubiquitination of PLC-γ1 and PKCθ was increased at 3 and 6 weeks. Our results suggest that ubiquitin and the E3 ubiquitin ligases c-Cbl and Cbl-b function as neuroimmune modulators in the subacute phase of neuropathic pain after nerve injury.

BACKGROUND: Remimazolam is not only associated with a lower incidence of respiratory depression than propofol but also in itself has the risk of respiratory depression. OBJECTIVE: We investigated respiratory depression following remimazolam infusion, targeting different effect-site concentrations using target-controlled infusion. DESIGN: A prospective, double-blind, randomised controlled study. SETTING: Tertiary hospital, Suwon, South Korea, from April 2022 to November 2022. PARTICIPANTS: One hundred and seven patients scheduled for general anaesthesia were randomised into three groups targeting remimazolam effect-site concentrations of 500 (RMZ-500) (n = 36), 1000 (RMZ-1000) (n = 35) and 1500 ng ml⁻¹ (RMZ-1500) (n = 36). INTERVENTIONS: Remimazolam was solely infused for 10 min according to target effect-site concentrations. According to the degree of SpO₂ decrease, oxygen desaturations were managed with the following respiratory supports: jaw-thrust for SpO₂ less than 97%, 100% oxygen delivery for SpO₂ less than 93% and assisted ventilation for SpO₂ less than 90%. MAIN OUTCOME MEASURES: The incidence of each respiratory support, along with respiratory variables (at baseline, 5 min and 10 min after remimazolam infusion) and loss of consciousness were observed for 10 min after remimazolam target-controlled infusion. RESULTS: Both RMZ-1000 and RMZ-1500 required more frequent respiratory support than RMZ-500 (both P < 0.001), with nearly identical frequencies between RMZ-1000 and RMZ-1500. In terms of respiratory support, the incidence of assisted ventilation was significantly lower in RMZ-500 (2.8%) than RMZ-1000 (48.6%) and RMZ-1500 (50%) (P < 0.001). RMZ-1000 and RMZ-1500 achieved loss of consciousness in all patients; RMZ-500 only achieved loss of consciousness in 86.1% of patients (P = 0.010). In patients who maintained spontaneous respiration, tidal volume decreased by 41 to 48% and respiratory rate increased by 118 to 158% at 5 and 10 min, significantly compared to baseline in all groups (P < 0.001). CONCLUSIONS: Remimazolam infusion, like that of other benzodiazepines, led to respiratory depression, which was more prominent at higher target effect-site concentrations. Therefore, appropriate countermeasures should be developed to prevent oxygen desaturation. TRIAL REGISTRATION: CRIS (https://cris.nih.go.kr), identifier: KCT0006952

Lee HJ, et al. J Gastric Cancer. 2025 May;25:e27. https://doi.org/10.5230/jgc.2025.25.e27

Remifentanil has been used to suppress peri-extubation cough. Palonosetron, a 5-HT₃ receptor antagonist, is an effective antiemetic, and 5-HT receptors mediate the cough reflex. We assessed the impact of palonosetron on effect-site concentration (Ce) of remifentanil for preventing emergence cough in females. Forty-five female patients undergoing laparoscopic cholecystectomy randomly received 0.075 mg of palonosetron (n = 21) or normal saline (n = 24) intravenously at the end of surgery. The remifentanil Ce for 50% (EC₅₀) and for 95% (EC₉₅) of patients were estimated via Dixon’s up-and-down method or isotonic regression. Using Dixon’s method, EC₅₀ in the control group (1.33 ± 0.38 ng/mL) was comparable to that of the palonosetron group (1.42 ± 0.75 ng/mL) (p = 0.813). Using isotonic regression, EC₅₀ (83% CIs) and EC₉₅ (95% CIs) did not reveal significant differences between the control and the palonosetron groups (1.17 (0.86–1.43) and 1.90 (1.45–1.96) ng/mL and 0.88 (0.78–1.23) and 2.43 (1.94–2.47) ng/mL, respectively). No difference was found in the remifentanil Ce to suppress emergence cough in the palonosetron group compared with the control group. It may indicate no effect of palonosetron on antitussive activity of remifentanil.