Tuesday, 18 August 2015

Response of bispectral index to neuromuscular block in awake volunteers. BJA July 2015.

This is an Australian study out of Cairns that harks back to earlier days of anaesthetic research where advances were made by anaesthetists experimenting on each other.

In 2003 a study was published in Anaesthesia and Analgesia by Messner et al demonstrating that the BIS was dropped significantly following administration of succinylcholine in awake volunteers. The explanation for this was unclear but it was postulated that EMG was affecting the BIS in a way that the monitor was not compensating for. The study was conducted using an early form of BIS monitor and newer models have attempted to address this problem. Subsequent versions have included an additional electrode designed to identify signal contents related to muscle activity and the software now generates an indicator of EMG. The contribution of the EMG to the BIS score in these newer monitors is known only to the manufacturers. It is unclear if it’s role relates simply to assessing signal quality or if it contributes directly to the score its self.

This study published in July’s BJA replicates Messner’s assessment of the level of awareness monitor using these newer BIS models. Inclusion criteria were that the subject was an Anaesthetist, was ASA 1 or 2, 25-60 year old, no GORD, no signs of difficult airway, BMI <25, no anxiety disorder. Each subject was paralysed initially with 1.5mg/kg sux and BIS was monitored until return of muscle activity. At a later date the experiment was repeated with 0.7mg/kg rocuronium followed by reversal with sugammadex at 20 minutes. Consciousness was confirmed with an isolated forearm technique. Subjects were asked to conduct math problems answering with their hands, and remember facts from stories in order to demonstrate full awareness. A rescue dose of propofol would be administered if significant distress was demonstrated.

11 anaesthetists were recruited for the study. BIS was monitored for 12 minutes following sux and 20 minutes following rocuronium. Following administration of muscle relaxant all patients demonstrated a two-stage drop in their BIS during the first few minutes of paralysis that was maintained until the return of muscle activity. Most subjects spent a significant period of the experiment with a BIS less than 70 (rough average of 15-40% of the observational period) whilst some subjects recorded a BIS < 60 for 10-20% of the time.

This result questions the reliability of BIS as a measure of awareness in the paralysed patient.

Important points:
  • The original BIS algorithms were generated from a database of EEGs recorded in anaesthetised patients who had not received muscle relaxants
  • The dataset used to develop BIS would not have included an awake and paralysed patient and it may be insufficiently designed to identify this clinical scenario
  • Both BIS and Entropy monitors include EMG activity in the synthesis of a level of awareness value, but it is not known how the EMG contributes to these scores by anyone except the manufacturers.
  • In this study, patients were most commonly found to be in the range of 60-80 when they were paralysed but fully conscious. The manufacturers have suggested this level of anaesthesia is appropriate for the end of surgery. With the introduction of sugamadex many anaesthetists are maintaining deeper paralysis upto the end of the procedure. In this setting over reliance on the BIS may lead to increased incidence of awareness.
  • This paper does not dismiss all the potential benefits of using BIS or Entropy but does highlight a major weakness that it maybe be unreliable when you depend on it the most
  • Despite the findings of this paper a recent Cochrane review demonstrated no increased incidence of awareness in patients using Level of Awareness monitors, even though this group received significantly less anaesthetic agents.
  • The findings of this study support an approach that avoiding paralysis where possible reduces the risk of awareness. In NAP 5 the risk of awareness of the unparalysed patient was 1:136,000 but this rose to 1:8,000 when muscle relaxant was used
For those interested, several other editorials and original scientific papers relating to risk of awareness and awareness monitoring were published in the same July edition of the BJA.


No comments:

Post a Comment