The randomized trial enrolled 80 adults scheduled for elective arthroscopic meniscectomy and analyzed 71 patients who completed the postoperative QoR-40 at approximately 24 hours. Investigators reported mean global QoR-40 scores on POD 1 of 166.1 ± 16.8 for the Propofol group and 170.6 ± 13.6 for the Remimazolam group. The between-group mean difference was −4.5 (95% CI −11.8 to 2.7), with a P value of 0.218. The authors treated the global QoR-40 as the primary endpoint and concluded there was no statistically significant difference between the two intravenous anesthetic strategies for this patient population at the 24-hour assessment.
The QoR-40 instrument spans five domains and yields a global score from 40 (worst) to 200 (best). The study team performed both parametric and nonparametric supporting analyses because the primary outcome distribution deviated from normality; results were reported as consistent with the primary complete-case analysis. The trial achieved its planned sample size requirement after accounting for expected dropout, and 71 patients exceeded the calculated minimum of 70 required for the predefined effect size and power assumptions.
When investigators examined domain-level scores, they observed higher emotional-state scores in the Remimazolam group (39.2 ± 2.5) compared with the Propofol group (37.0 ± 4.3). The reported mean difference was −2.2 with a 95% CI of −3.9 to −0.6 and a P value of 0.009. The authors described this as an exploratory, domain-specific finding. They also noted that secondary and domain analyses were performed without multiplicity adjustment and should therefore be interpreted cautiously.
Investigators recorded mean arterial pressure and heart rate at predefined intraoperative and PACU time points. The trial reported that intraoperative hypotension requiring ephedrine occurred more frequently in the Propofol group: 44.4% versus 17.1% in the Remimazolam group (P = 0.013). The study team therefore documented greater intraoperative hemodynamic stability with remimazolam relative to propofol. Other perioperative hemodynamic comparisons were analyzed using linear mixed-effects models and presented as exploratory between-group contrasts at individual time points.
Recovery timing differed between groups. Time to full alertness (assessed every 5 minutes in PACU by a blinded investigator) was longer in the Remimazolam group: median 37 minutes (IQR 31–45) versus 18 minutes (IQR 15–24) for Propofol (P < 0.001). PACU length of stay was likewise longer after remimazolam: median 38 minutes (IQR 30–50) compared with 30 minutes (IQR 30–39) in the Propofol group (P = 0.008). The authors emphasized this trade-off: remimazolam provided fewer vasopressor-requiring hypotensive episodes but at the cost of slower return to full alertness and modestly prolonged PACU monitoring.
The protocol specified standardized adjunctive analgesia (IV acetaminophen 1 g at induction, intraoperative remifentanil TCI, fentanyl 50 µg and ramosetron at end of surgery) and rescue analgesia thresholds in PACU and ward. Pain scores and postoperative nausea and vomiting (PONV) were reported as secondary, exploratory outcomes; the article presents these data without adjustment for multiple comparisons.
This was a single-center, randomized controlled trial with allocation concealment using opaque envelopes. Attending anesthesiologists were not blinded because the infusion methods for propofol (TCI pump using Marsh model) and remimazolam (weight-based syringe pump) differed; however, patients and the investigator conducting postoperative assessments remained blinded. Both groups received remifentanil TCI for analgesia. Propofol target effect-site concentrations and remifentanil Ce ranges, and remimazolam infusion rates for induction and maintenance, were specified by the authors and applied to maintain a patient state index (PSI) between 25 and 50. Mechanical ventilation and intraoperative hemodynamic management thresholds (systolic BP <90 mmHg or MAP <60 mmHg treated with ephedrine) were predefined.
The authors conducted a complete-case primary analysis (n = 71) and also described multiple imputation for missing POD 1 QoR-40 data across all randomized patients as a sensitivity approach; the nine missing questionnaires resulted from administrative failure to distribute the instrument rather than from withdrawals or refusals.
The trial had several limitations that the authors reported. Nine randomized patients lacked POD 1 QoR-40 data due to administrative omission; the primary analysis therefore used a complete-case set of 71 patients, although multiple imputation was described as a sensitivity analysis. Attending anesthesiologists were not blinded to group allocation, which could introduce performance bias during intraoperative management despite blinded outcome assessment. The study was single-center and focused on a specific short-stay orthopedic procedure, which may limit generalizability to other surgical populations or more complex operations. Secondary and domain analyses were exploratory and conducted without correction for multiplicity. These limitations were explicitly reported by the investigators and remain open questions for replication or broader multicenter evaluation.