Propofol sedation in children: sleep trumps amnesia

Abstract

Objective: Detailed assessments of the effects of propofol on memory in children are lacking. We assessed the feasibility of measuring memory during propofol infusion, as commonly performed in sedation for MRI scanning. In addition, we determined the onset of memory loss in relation to the onset of sedation measured by verbal responsiveness.

Materials and methods: Children scheduled for sedation for MRI received a 10-min infusion of propofol (3 mg/kg) as they viewed and named 100 simple line drawings, one shown every five seconds, until they were no longer responsive (encoding). A control group receiving no sedation for MRI underwent similar tasks. Sedation was measured as any verbal response, regardless of correctness. After recovery from sedation, recognition memory was tested, with correct yes/no recognitions matched to sedation responses during encoding (subsequent memory paradigm).

Results: Of the 48 children who received propofol, 30 could complete all study tasks (6.2 ± 1.6 years, 16 males). Individual responses could be modeled in all 30 children. On average, there was a 50% probability of no verbal response 3.1 min after the start of infusion, with 50% memory loss at 2.7 min. Children receiving propofol recognized 65 ± 16% of the pictures seen, whereas the control group recognized 93 ± 5%.

Conclusion: Measurement of memory and sedation is possible in verbal children receiving propofol by infusion in a clinical setting. Despite propofol being an amnestic agent, there was little or no amnestic effect of propofol while the child was verbally responsive. It is important for sedation providers to realize that propofol sedation does not always produce amnesia while the child is responsive. CLINICALTRIALS.

Gov number: NCT02278003.

Keywords: Memory; Pediatrics; Sedation; Sleep.

Conflict of interest statement

Conflict of interest

The authors have no potential conflicts of interest.

The ICMJE Uniform Disclosure Form for Potential Conflicts of Interest associated with this article can be viewed by clicking on the following link: http://dx.doi.org/10.1016/j.sleep.2016.10.002.

Copyright © 2016. Published by Elsevier B.V.

Figures

Fig. 1

Top graph: Individual probabilities are represented by the faint plots and group averages are shown in bold. Sedation was measured by probability of verbal response. The probabilities of both sedation and subsequent recognition memory are plotted against time (in seconds), which was used as the predictor variable over the period of propofol infusion before scanning started. As the bold lines demonstrate, the probability of recognition memory loss occurred at an earlier time than the same probability of sedation. The lower graph represents the difference in time between 50% probabilities of memory loss and sedation in each individual. The distribution of individual time intervals revealed that most differences were positive (memory loss occurred somewhat before sedation).