Cerebral oximetry monitoring using near-infrared spectroscopy during adult procedural sedation: a preliminary study

Background and objectives: We sought to evaluate the effect of adult procedural sedation on cerebral oxygenation measured by near-infrared spectroscopy (rSo₂ levels), and to assess whether respiratory depression occurring during procedural sedation was associated with decreases in cerebral oxygenation. Methods: We performed a prospective, observational preliminary study on a convenience sample of adult patients (>18 years) undergoing unscheduled procedural sedation in the ED from August 2017 to September 2018 at Hennepin County Medical Center in Minneapolis, Minnesota. The primary outcome measures were rSo₂ values by level of sedation achieved and the incidence of cerebral hypoxaemia during procedural sedation (absolute rSo₂ ≤60 or decrease ≥20% from baseline). The secondary outcome is the decrease in rSo₂ during episodes of respiratory adverse events (AEs), defined by respiratory depression requiring supportive airway measures. Results: We enrolled 100 patients (53% female). The median (IQR) rSo₂ values (%) by each level of sedation achieved on the Observer Assessment of Alertness and Sedation (OAAS) scale 1.5, respectively, were 74 (69.79), 74 (70.79), 74 (69.79), 75 (69.80), 72 (68.76). The incidence of cerebral hypoxaemia at any point within the sedation (absolute rSo₂ <60%) was 10/100 (10%); 2 out of 10 had rSo₂ reduction more than 20% from baseline value; the median (IQR) observed minimum rSo₂ in these patients was 58 (56.59). We observed respiratory depression in 65 patients via standard monitoring; of these, 39 (60%) required at least one supportive airway measure, meeting the definition of a respiratory AE. During these AEs, 15% (6/39) demonstrated cerebral hypoxaemia with a median (IQR) minimum rSo₂ of 58 (57.59). Four patients (4%) had cerebral hypoxaemia without a respiratory AE. Conclusion: Cerebral oximetry may represent a useful tool for procedural sedation safety research to detect potential subclinical changes that may be associated with risk, but appears neither sensitive nor specific for routine use in clinical practice.