Daily Anesthesiology Research Analysis
Three perioperative studies stand out today: a randomized trial shows remimazolam reduces postoperative delirium versus propofol in frail elderly hip surgery patients; a multicenter external validation demonstrates an ML model (S-PATH) can safely cut type-and-screen orders while maintaining high sensitivity for transfusion risk; and an updated meta-analysis confirms dexmedetomidine reduces delirium after cardiac surgery, albeit with increased bradycardia.
Summary
Three perioperative studies stand out today: a randomized trial shows remimazolam reduces postoperative delirium versus propofol in frail elderly hip surgery patients; a multicenter external validation demonstrates an ML model (S-PATH) can safely cut type-and-screen orders while maintaining high sensitivity for transfusion risk; and an updated meta-analysis confirms dexmedetomidine reduces delirium after cardiac surgery, albeit with increased bradycardia.
Research Themes
- Perioperative delirium prevention and anesthetic choice
- Machine learning decision support for transfusion planning
- Sedation strategies and hemodynamic safety trade-offs
Selected Articles
1. Remimazolam tosylate or propofol and delirium in frail elderly patients after hip surgery: A randomised controlled clinical trial.
In frail elderly hip surgery patients, remimazolam reduced postoperative delirium versus propofol (4.4% vs 17.6%; RR 0.25) and decreased induction hypotension and vasopressor use. EEG burst suppression was significantly less with remimazolam, suggesting gentler neurophysiologic effects.
Impact: This RCT provides direct evidence that anesthetic choice can reduce postoperative delirium in a high-risk population, with concurrent hemodynamic and EEG benefits.
Clinical Implications: Consider remimazolam for frail elderly undergoing hip surgery to lower delirium risk and hypotension; monitor protocols and availability. Incorporate EEG burst suppression minimization into anesthesia quality metrics.
Key Findings
- Postoperative delirium: 4.4% (remimazolam) vs 17.6% (propofol); RR 0.25, P=0.0143
- Induction hypotension: 23.5% vs 47.1%; RR 0.50, P=0.004; fewer vasopressor requirements
- Significantly less intraoperative burst suppression with remimazolam (shorter duration and proportion)
Methodological Strengths
- Randomized controlled design with trial registration
- Twice-daily 3D-CAM assessments and EEG burst suppression quantification
Limitations
- Single-centre, single-blind design may limit generalizability
- Short delirium assessment window (first 3 postoperative days)
Future Directions: Multicentre pragmatic RCTs across surgeries and comorbidities; mechanistic studies on EEG dynamics and delirium; cost-effectiveness analyses of remimazolam adoption.
BACKGROUND: Remimazolam tosylate, a novel short-acting benzodiazepine, is increasingly being used in general anaesthesia, but its role in the incidence of postoperative delirium is uncertain, particularly in frail elderly patients. OBJECTIVE: To compare the effects of remimazolam tosylate with propofol on the incidence of postoperative delirium in frail elderly patients undergoing hip surgery. DESIGN: Randomised, single-centre, single-blind controlled trial. SETTING: A tertiary teaching hospital in China, conducted from March to December 2023. PATIENTS: Frail elderly patients (Reported Edmonton Frail Scale Score ≥ 6) undergoing hip surgery under general anaesthesia. INTERVENTIONS: Patients were randomly assigned to either the propofol or remimazolam group. Both groups received total intravenous anaesthesia following a standardised protocol with either propofol or remimazolam tosylate for induction and maintenance. MAIN OUTCOME MEASURES: The primary outcome was the incidence of postoperative delirium within three postoperative days, assessed twice daily using the 3D Confusion Assessment Method (3D-CAM). The secondary outcomes included the quality of postoperative recovery and adverse events. RESULT: A total of 136 patients were enrolled. The incidence of postoperative delirium was significantly lower in the remimazolam group than in the propofol group [3 of 68 (4.4%) vs. 12 of 68 (17.6%), risk differece (RD) -13.2%, 95% CI -23.5% to -2.9%, relative risk (RR) 0.25, 95% CI 0.074 to 0.847, NNT 7.6, P = 0.0143]. The incidence of hypotension after induction was also lower in the remimazolam group [16 of 68 (23.5%) vs. 32 of 68 (47.1%), RD -23.5%, 95%CI -39.1% to -8.0%, RR 0.5, 95% CI 0.304 to 0.822, NNT 4.3, P = 0.004], with fewer patients requiring vasopressors [55 of 68 (80.9%) vs. 66 of 68 (97.1%), RD -16.2%, 95% CI -26.3 to -6.0, RR 0.8, 95% CI 0.737 to 0.942, NNT 6.2, P = 0.003]. Notably, the remimazolam group exhibited significantly less burst suppression compared with the propofol group, both in terms of burst suppression time (2.2 s [0 to 17.6] vs. 21.9 s [2.3 to 115.3] median difference = 11.98 s, 95% CI 2.44 to 27.90, P < 0.001) and its proportion relative to the total surgery time (0.3‰ [0 to 2.1] vs. 2.8‰ [0.2 to 14.7], median difference 1.30‰, 95% CI 0.27 to 3.34, P < 0.001). CONCLUSION: In frail elderly patients, remimazolam tosylate was associated with a lower incidence of postoperative delirium compared with propofol. TRIAL REGISTRATION: Chinese Clinical Trial Registry, Chictr.org.cn, identifier: ChiCTR2300068632.
2. Multicenter Validation of a Machine Learning Model for Surgical Transfusion Risk at 45 US Hospitals.
Across 3.28 million surgeries at 45 hospitals, S-PATH reduced type-and-screen recommendations by a median 17.9 percentage points at fixed 96% sensitivity and achieved higher AUROC (0.929) than MSBOS (0.857). This supports implementation of personalized transfusion risk prediction for perioperative decision support.
Impact: Massive multicenter external validation demonstrates robust generalizability and practical benefit (fewer unnecessary tests) at maintained sensitivity, a key step for clinical adoption of ML in perioperative medicine.
Clinical Implications: Hospitals can adopt S-PATH to personalize type-and-screen orders, reducing laboratory workload and costs without sacrificing safety; integrate into EHR with local monitoring and governance.
Key Findings
- At 96% sensitivity, S-PATH recommended type and screen for median 32.5% vs 51.6% with MSBOS (median difference 17.9 percentage points).
- S-PATH achieved higher AUROC (median 0.929) than MSBOS (0.857) across hospitals.
- External validation across 45 hospitals and 3,275,956 cases without local retraining.
Methodological Strengths
- Large-scale multicenter external validation with per-hospital performance reporting
- Head-to-head comparison to current standard (MSBOS) at fixed sensitivity threshold
Limitations
- Retrospective design with potential data quality and coding variability
- Did not assess downstream outcomes (e.g., costs, crossmatch delays, clinical adverse events)
Future Directions: Prospective implementation studies measuring safety, cost, workflow, and equity; calibration drift monitoring; extension to plasma/platelet planning.
IMPORTANCE: Accurate estimation of surgical transfusion risk is important for perioperative planning and effective resource allocation. Most machine learning models in health care are not validated or perform poorly in external settings. OBJECTIVE: To externally validate a publicly available machine learning algorithm (Surgical Personalized Anticipation of Transfusion Hazard [S-PATH]) to estimate red cell transfusion during surgery within a national sample of hospitals. DESIGN, SETTING, AND PARTICIPANTS: This retrospective cohort study evaluated all surgical cases performed in 2020 or 2021 at 45 US hospitals participating in the Multicenter Perioperative Outcomes Group. Obstetric and nonoperative cases were excluded. Data analysis was performed from February 2023 to March 2025. EXPOSURES: At each hospital, S-PATH was used to estimate surgical transfusion risk using patient- and procedure-specific characteristics without local retraining. A baseline model representing the standard-of-care maximum surgical blood ordering schedule (MSBOS) approach, which omits patient factors, was used for comparison. Risk thresholds above which a type and screen would be recommended were set for 96% sensitivity. Performance was evaluated at each hospital separately. MAIN OUTCOMES AND MEASURES: The primary outcome was the difference in the percentage of patients with type and screen order recommendations between S-PATH and MSBOS at each hospital. The secondary outcome was area under the receiver operating characteristic curve (AUROC). RESULTS: In this cohort study of 3 275 956 surgical cases (median [IQR] age, 57 [40-69] years; 53.1% female) performed at 45 hospitals (28 of 45 academic [62.2%]), S-PATH recommended type and screen orders for a median (IQR) of 32.5% (25.8%-42.0%) of cases, whereas the MSBOS approach recommended type and screens for a median (IQR) of 51.6% (46.9%-61.1%) of cases for the same sensitivity (median [IQR] difference, 17.9 [14.8-24.9] absolute percentage points). The median (IQR) S-PATH AUROC was 0.929 (0.915-0.946), whereas the median (IQR) MSBOS AUROC was 0.857 (0.822-0.884). CONCLUSIONS AND RELEVANCE: In this cohort study of 45 hospitals, a personalized surgical transfusion risk prediction algorithm demonstrated external validity and discrimination. S-PATH was consistently more effective than standard care, suggesting its potential for use as a perioperative clinical decision support tool.
3. Dexmedetomidine for delirium prevention after cardiac surgery: An updated systematic review and meta-analysis with trial sequential analysis.
Pooling 31 RCTs (n=5,628), dexmedetomidine reduced postoperative delirium after cardiac surgery (RR 0.61) with TSA-confirmed robustness and slightly shorter ICU stay, but increased bradycardia. No differences were observed in mortality or hypotension.
Impact: This comprehensive synthesis with TSA resolves prior inconsistencies and supports guideline-level consideration of dexmedetomidine for delirium prevention in cardiac surgery, while emphasizing hemodynamic vigilance.
Clinical Implications: Dexmedetomidine can be incorporated into cardiac surgery sedation/analgesia pathways to lower delirium risk, with protocols to monitor and manage bradycardia.
Key Findings
- Delirium risk reduction with dexmedetomidine: RR 0.61 (95% CI 0.49–0.75), P<0.001; consistent across subgroups and sensitivity analyses
- Slightly shorter ICU length of stay (MD -0.14 days), TSA confirmed robustness
- Increased bradycardia (RR 1.53); no differences in mortality, hypotension, atrial fibrillation, or intubation duration
Methodological Strengths
- Systematic review and meta-analysis of 31 RCTs with trial sequential analysis
- Extensive subgroup and sensitivity analyses to test robustness
Limitations
- Heterogeneity in dosing regimens, timing, and delirium assessment methods across trials
- Potential publication bias and varying risk of bias among included RCTs
Future Directions: Standardize dosing/timing protocols; identify patient phenotypes with greatest net benefit; pragmatic trials integrating hemodynamic monitoring bundles.
BACKGROUND: Postoperative delirium remains a common complication after cardiac surgery. The impact of dexmedetomidine (DEX) on preventing postoperative delirium is still controversial as recent randomized controlled trials (RCTs) have presented conflicting results. METHODS: We conducted an updated systematic review and meta-analysis of RCTs evaluating DEX efficacy in preventing delirium after cardiac surgery. A systematic search of MEDLINE, Embase, and Cochrane databases identified RCTs comparing DEX with placebo or other treatments in patients aged ≥18 years. Sensitivity, subgroup analyses, and trial sequential analysis (TSA) assessed the robustness of the findings. RESULTS: A total of 31 RCTs including 5628 patients were analyzed, 50.1% of them receiving DEX. Delirium incidence was significantly lower in the DEX group (RR 0.61; 95% CI, 0.49-0.75; P < 0.001). This protective effect remained across subgroup analyses based on age, control type, delirium assessment method, and after excluding trials at high risk of bias. DEX use was associated with a shorter intensive care unit length of stay (MD -0.14 days; 95% CI, -1.28 to -0.04; P < 0.01). TSA confirmed the result's robustness. However, DEX increased bradycardia risk (RR 1.53; 95% CI, 1.05-2.21; P = 0.02). No significant differences were found in mortality, intubation duration, hospital length of stay, atrial fibrillation, or hypotension. CONCLUSIONS: Dexmedetomidine significantly reduces postoperative delirium following cardiac surgery, with moderate evidence confirmed by TSA. While it demonstrates clinical benefits, careful bradycardia monitoring is warranted. SYSTEMATIC REVIEW PROTOCOL: PROSPERO (CRD42024593472).