Randomized Clinical Trial Comparing Procedural Amnesia and Respiratory Depression Between Moderate and Deep Sedation With Propofol in the Emergency Department

Objectives: The objective was to determine if there is a difference in procedural amnesia and adverse respiratory events (AREs) between the target sedation levels of moderate (MS) and deep (DS) procedural sedation. Methods: This was a prospective, randomized clinical trial of consenting adult patients planning to undergo DS with propofol between March 5, 2015, and May 24, 2017. Patients were randomized to a target sedation level of MS or DS using the American Society of Anesthesiologist's definitions. Drug doses, vital signs, observer's assessment of alertness/sedation (OAAS) score, end-tidal CO ₂ (ETCO ₂ ), and the need for supportive airway maneuvers (SAMs; bag-valve mask use, repositioning, and stimulation to induce respirations) were monitored continuously. A standardized image was shown every 30 seconds starting 3 minutes before the procedure continuing until the patient had returned to baseline after the procedure. Recall and recognition of images were assessed 10 minutes after the sedation. Subclinical respiratory depression (RD) was defined as SaO ₂ ≤ 91%, change in ETCO ₂ ≥ 10 mm Hg, or absent ETCO ₂ at any time. The occurrence of RD with a SAM was defined as an ARE. Patient satisfaction, pain, and perceived recollection and physician assessment of procedure difficulty were collected using visual analog scales (VASs). Data were analyzed with descriptive statistics and Wilcoxon rank-sum test. Results: A total of 107 patients were enrolled: 54 randomized to target MS and 53 to DS. Of the patients randomized to target MS, 50% achieved MS and 50% achieved DS. In the target DS group, 77% achieved DS and 23% achieved MS. The median total propofol dose (mg/kg) was lower in the MS group: MS 1.4 (95% confidence interval [CI] = 1.3–1.6, IQR = 1) versus DS 1.8 (95% CI = 1.6–2.0, IQR = 0.9). There were no differences in median OAAS during the procedure (MS 2.4 and DS 2.8), lowest OAAS (MS 2 and DS 2), percentage of images recalled (MS 4.7% vs. DS 3.8%, p = 0.73), or percentage of images recognized (MS 61.1% vs. DS 55%, p = 0.52). In the MS group, 41% patients had any AREs compared to 42% in the DS group (p = 0.77, 95% CI difference = –0.12 to 0.24). The total number of AREs was 23% lower in the MS group (p = 0.01, 95% CI = –0.41 to –0.04). There was no difference in patient-reported pain, satisfaction, or recollection VAS scores. Provider's rating of procedural difficulty and procedural success were similar in both groups. Conclusions: Targeting MS or DS did not reliably result in the intended sedation level. Targeting MS, however, resulted in a lower rate of total AREs and fewer patients had multiple AREs with no difference in procedural recall. As seen in previous reports, patients who achieved MS had less AREs than those who achieved DS. Our study suggests that a target of MS provides adequate amnesia with less need for supportive airway interventions than a target level of DS, despite the fact that it often does not result in intended sedation level.