Opioid-free procedural sedation for vertebral augmentation in older adults: a before-and-after cohort study of etomidate versus midazolam with propensity-score matching

Abstract Background Vertebral augmentation for osteoporotic fracture is done under procedural sedation because the patients are old, often carry several chronic conditions, and are unfit for general anesthesia. Guidelines warn against benzodiazepines in older adults and against pairing them with an opioid. Etomidate and midazolam have not been compared for this procedure. Methods This retrospective before-and-after cohort covers 476 consecutive vertebral augmentation procedures in 433 patients (aged 51 to 96 years; 40 procedures, 8.4%, in patients younger than 65) treated by one surgeon, using midazolam 0.04 mg/kg until December 2020 and etomidate 0.2 mg/kg thereafter. The surgeon gave the sedative, no opioid was used, and all patients were prone. Propensity-score matching on 19 covariates yielded 136 matched pairs of procedures. Co-primary outcomes, selected before analysis, were recalled pain rated after emergence on a numeric rating scale and the number of movements that interrupted the procedure. Because sedative choice followed calendar period, segmented regression was applied to the full series with autocorrelation-robust standard errors. Results Recalled pain differed by − 1.23 (95% CI − 1.55 to − 0.90) and pain of 7 or above was recalled by 5.9% versus 29.4%; movements interrupting the procedure differed by − 0.60 (− 0.80 to − 0.39). Peak Ramsay depth differed by + 0.24 (+ 0.00 to + 0.47; p = 0.056), and Ramsay 3 or above was reached by 87.5% versus 75.0%. No statistically detectable difference was found in cardiorespiratory events. Patient satisfaction differed by + 1.55 (+ 1.25 to + 1.85) and recovery-area discharge came 6.3 min earlier. Segmented regression showed a step change at the switch, but placebo cutoffs within the midazolam era produced changes of similar or larger size in recalled pain. Conclusions Procedures performed during the etomidate era were associated with lower recalled pain scores, fewer recorded interruptions from patient movement, and shorter recovery-room stays than matched procedures from the midazolam era. Because sedative choice was determined entirely by calendar period, no outcome was assessed blind, and respiratory and drug-specific adverse events were incompletely captured, these findings are hypothesis generating and do not establish comparative efficacy or safety.