Awareness in Anaesthesia – B-Aware and B-Unaware trials.

Bispectral Index Monitoring to Prevent Awareness During Anaesthesia: the B-Aware Randomised Controlled Trial.

Myles PS, Leslie K, McNeil J et al.

Lancet. 2004 May 29; 363 (9423): 1757-63.

Aim of Study

“… assess whether BIS monitoring decreases the incidence of awareness during relaxant general anaesthesia in routine surgical patients at high risk of awareness.” (pg. 1757)

Design and Location

Prospective multicenter, double-blind, randomised controlled trial

• Mainly Australia and New Zealand, some input from UK and Thailand

Methodology

Inclusion Criteria:

• Surgical patients undergoing relaxant general anaesthesia
  • ≥ 18 y.o.
  • ≥ 1 risk factor for awareness

Exclusion Criteria:

• Inadequate comprehension of the English language, traumatic brain injury, memory impairment, psychosis, known or suspected EEG abnormality (e.g. epilepsy, previous brain resection, or scarring), or were not expected to be available for interview postoperatively

Total of 2463 patients included; randomised to ‘BIS’ vs. ‘routine care’

• BIS group n = 1225; target BIS 40 – 60
• Routine Care group n = 1238; BIS monitor turned off
• Anaesthetic technique at discretion of anaesthestist

Primary Outcome

• Incidence of confirmed awareness
  • Assessed by ‘structured questionnaire’ following recovery from GA (2 – 6 hours), 24 – 36 hours and 30 days post-op
  • Committee of 3 experienced anaesthetists to identify “Awareness”, “Possible awareness” or “No awareness”

Secondary Outcomes

• Possible awareness
• Recovery times
• Hypnotic drug administration
• Incidence of marked hypotension
• Anxiety and depression
• Patient’s satisfaction
• Major complications
• 30-day mortality

Statistics

• Primary Outcome – Fisher’s exact test; expressed as proportions, odds ratios with 95% confidence intervals (CIs) and p values
• Secondary Outcomes – Variety of other statistical analyses (mainly Fisher’s exact test or Χ² test)

Results

• 2 reports of “Awareness” in BIS group vs. 11 reports of “Awareness” in routine care group (p = 0.022)
  • Reduced risk of awareness by 82% (95% CI 17 to 98%) if BIS used
  • In both reports of “Awareness” in BIS group, BIS readings were >60 for 5 mins and 9 mins!
• Dose of Midazolam (2 mg vs. 2.5 mg, p = 0.017) and Target Plasma Conc. (if TIVA used; 2 mg/L vs. 2.4 mg/L, p = 0.016) were significantly less in BIS vs. routine care group
• Time to eye opening (mins, 9 vs. 10, p = 0.003) was quicker (i.e. faster recovery) in BIS vs. routine care group

Conclusions/Discussions

• Use of BIS reduces risk of awareness in adult population at higher risk of awareness undergoing relaxant GA
• Number needed to treat = 138
• Cost of preventing one case of awareness = $2,200

Stated Limitations from the Study

• Only focused on at-risk patients à ?Generalisability for entire adult population
• Lack of control in ‘routine care’ group à ?Prevention of awareness due to increased vigilance when using BIS
• No direct comparison of BIS to other standard forms of anaesthetic monitoring (e.g. MAC or E_TVolatile)

***************************************************

Anesthesia Awareness and the Bispectral index. (B-Unaware Trial)

Avidan MS, Zhang L, Burnside BA et al.

New England Journal of Medicine. 2008 Mar 13; 358 (11): 1097-108.

Aim of Study

“… determine whether, in patients at high risk, the incidence of anesthesia awareness is reduced when clinicians follow a BIS-guided protocol rather than an ETAG-guided protocol.” (pg. 1098)

Design and Location

Prospective single-center, randomised controlled trial

• Barnes-Jewish Hospital (Missouri, US)

Methodology

Inclusion Criteria:

• Surgical patients undergoing general anaesthesia with volatiles
  • ≥ 18 y.o.
  • High risk for awareness (≥ 1 major criteria or ≥ 2 minor criteria)

Exclusion Criteria:

• Surgical procedure/positioning prevented BIS monitoring
• Wake-up test required intra-operatively
• Dementia or history of stroke with residual neurological deficits
• Unable to provide informed consent

Total of 1941 patients included; randomised to ‘BIS-guided’ vs. ‘end-tidal anaesthetic gas (ETAG)’

• BIS-guided group n = 967; target BIS 40 – 60 with no MAC range
• ETAG group n = 974; target MAC 0.7 – 1.3, no BIS number visible

Primary Outcome

• Incidence of confirmed awareness
  • Assessed by Brice questionnaire within 24 hours, 24 – 72 hours and 30 days post-extubation
  • 3 – 4 experts, required ≥ 2 to be in agreement for “anesthesia awareness”, “might have had anesthesia awareness” or “did not have anesthesia awareness”

Statistics

• Primary Outcome
  • Anticipated incidence of awareness was 1% for ETAG group, 0.1% for BIS-guided group
  • Aimed for a 0.9% difference with a one-tailed alpha of 0.05 and a power of 80% using Fisher’s exact test
  • Confidence intervals for absolute risk reduction calculated using Newcombe’s method without continuity correction
• Other comparisons – Variety of other statistical analyses inc. Χ² test, Fisher’s exact test, unpaired t-test and unpaired Mann-Whitney test

Results

• 2 cases definite “anesthesia awareness” in each group (i.e. 2 in BIS-guided and 2 in ETAG group)
  • Absolute difference between groups was 0% (95% CI -0.56 to 0.57%)
• Additionally, 4 cases of “possible anesthesia awareness” in BIS-guided group and 1 case of possible anesthesia awareness in ETAG group
• BIS > 60 in 1 of the 4 cases of definite “anesthesia awareness” and 3 of the 9 cases of definite+possible “anesthesia awareness”
• MAC < 0.7 in 3 of the 4 cases of definite “anesthesia awareness” and 7 of the 9 cases of definite+possible “anesthesia awareness”

Conclusions/Discussions

• “We did not reproduce the results of previous studies that reported a lower incidence of anesthesia awareness with BIS monitoring… Anesthesia awareness occurred even when BIS values and ETAG concentrations were within the target ranges.” (pg. 1097)

Stated Limitations from the Study

• Only focused on at-risk patients – ?Generalisability for entire adult population
• Only focused on patients having volatile GA – Cautioned that use of TIVA is an additional risk factor for awareness and recommended using BIS when using TIVA!
• Diagnosis of awareness has a subjective element
  • Repeated questioning might induce false memories

Discussion from Journal Club Meeting (?Change of Practice)

• What about BIS and TIVA?
  • Yes, big multi-centre RCT done in China has shown BIS-guided TIVA reduces risk of awareness vs. routine-TIVA practice
  • Ref – Zhang et al. Bispectral index monitoring prevent awareness during total intravenous anesthesia: a prospective, randomized, double-blinded, multi-center controlled trial. Chin Med J (Engl). 2011 Nov; 124(22):3664-9.
• Re: B-Unaware Trial, any data for MAC < 0.7 but no awareness?
  • Data not explicitly available
• What about Entropy (Kingston standard for brain monitoring)?
  • Far fewer studies, none of which have been RCTs in prevention of awareness

Summary by Dr J Kua. Journal Club Meeting 28 March 2019.

Image from https://commons.wikimedia.org/wiki/File:BIS_JPN.jpg