齋藤 大之

PURPOSE: Right-sided double-lumen endobronchial tubes (DLTs) are often avoided because of technical difficulties and concerns about respiratory complications, although evidence that they increase hypoxemia during anesthesia remains limited and inconsistent. We evaluated the association between DLT side and hypoxemia during anesthesia in patients undergoing unilateral surgery for left-sided lung tumors. METHODS: This single-center retrospective cohort study included adult patients undergoing unilateral surgery for left-sided lung tumors between October 2008 and May 2023. The primary outcome was hypoxemia during anesthesia, defined as SpO2 ≤ 90% for at least 3 consecutive min during the anesthetic period. Propensity score overlap weighting was used to adjust for measured differences between patients receiving right-sided and left-sided DLTs. Odds ratios (ORs) and absolute risk differences (RDs) were estimated. RESULTS: Among 1,280 patients, 842 received right-sided DLTs and 438 received left-sided DLTs. After overlap weighting, measured covariates were well balanced. The incidence of hypoxemia during anesthesia was 20.0% with right-sided DLTs and 18.5% with left-sided DLTs (OR, 1.10; 95% confidence interval [CI], 0.68-1.79; P = 0.70). The RD was 1.51 percentage points (95% CI - 3.86-6.87). No clear between-group differences were observed in exploratory secondary respiratory outcomes. CONCLUSION: We found no clear evidence that right-sided DLT use was associated with a higher incidence of hypoxemia during anesthesia than left-sided DLT use. However, equivalence between the two devices was not established, and further studies are needed to determine whether clinically important differences in hypoxemia risk exist.

Abstract Background The utilization of neuromuscular blocking agents (NMBAs) in hypoxemic critically ill patients varies, and real-world patterns of NMBA discontinuation and subsequent restart remain insufficiently described. This study aimed to evaluate patterns of NMBAs utilization and explore associations with subsequent NMBA restart and ICU mortality. Methods A retrospective cohort study was conducted using an integrated intensive care database (MIMIC-III / IV, eICU, HiRID, AmsterdamUMCdb, and our institution). Mechanically ventilated patients with PaO ₂ /FiO ₂ (P/F) ratio < 200 who received NMBAs for > 24 h were included to analyze NMBA discontinuation ( n  = 829). For those who discontinued NMBA during ICU stay ( n  = 737), NMBA restart and ICU mortality were also evaluated. Results Prolonged NMBA use exceeding 48 h occurred in 332 of 829 patients (40.0%), with substantial inter-dataset variability. Younger age, higher body mass index, later NMBA initiation and lower respiratory system compliance were associated with longer NMBA duration. Among the 737 patients who discontinued NMBA during the ICU stay, 291 (39.4%) required NMBA restart and 260 (35.3%) died in the ICU. Although NMBA duration was significantly associated with subsequent NMBA restart (Odds Ratio=1.006, CI 1.003–1.009, p  < 0.001), it was not associated with ICU mortality (Odds Ratio=1.002, CI 0.999–1.005, p  = 0.250). Conclusion NMBA practices varied widely, and prolonged administration was common, with a high rate of restart following discontinuation. NMBA duration was associated with subsequent NMBA restart but not with ICU mortality. These findings highlight the need for further studies to identify clinical factors associated with NMBA discontinuation and restart and to better understand the clinical significance of these observed patterns in hypoxemic respiratory failure.

OBJECTIVES: To operationalize and temporally validate an electronic medical record (EMR)-integrated machine learning system (Big data-driven Evaluation of Survival and Treatment in Acute Illness [BEST-AI]) that generates hourly predictions for multiple ICU outcomes, with emphasis on discrimination, calibration, and workflow integration. DESIGN: Single-center hybrid study with stepwise clinical deployment and forward-in-time temporal validation. SETTING: Thirty-bed tertiary mixed medical-surgical ICU in Japan. PATIENTS: All ICU admissions from August 2017 to March 2025. Exclusions: age younger than 16 years or ICU stay less than 4 hours. Development cohort (n = 11,176; from August 2017 to July 2024) and temporal validation cohort (n = 1,127; from August 2024 to March 2025). INTERVENTIONS: EMR-integrated deployment of BEST-AI providing hourly probabilistic predictions to clinicians within the EMR; no protocolized clinical interventions were mandated. MEASUREMENTS AND MAIN RESULTS: Six prediction tasks (in-hospital mortality, ICU mortality, ICU discharge ≤ 72 hr, intubation ≤ 72 hr, extubation ≤ 72 hr, tracheostomy at ICU discharge) were evaluated. In temporal validation, the area under the receiver operating characteristic curves ranged from 0.856 to 0.960, and the area under the precision-recall curves from 0.302 to 0.786. Decile-based calibration showed overall good agreement; hospital mortality was slightly overestimated at higher predicted probabilities, whereas ICU mortality remained well aligned. The intubation task had comparatively lower discrimination and greater deviation from perfect calibration, consistent with low event counts and heterogeneous timing. A 24-hour landmark sensitivity analysis (one prediction per patient at 24 hr after ICU admission) preserved discrimination and calibration relative to the main analysis, supporting robustness beyond repeated-measures evaluation. The system was successfully maintained with automated hourly updates and EMR-embedded patient- and unit-level visualizations, without prescriptive alerts. CONCLUSIONS: A continuously deployed, EMR-integrated ICU prediction system achieved strong temporal discrimination and generally good calibration. Embedding real-time predictions into routine workflow was feasible, and the system was maintained with automated hourly updates. Prospective multicenter studies are warranted to assess transportability and clinical impact.