The Committee considered the heterogeneity and uncertainty in the studies and the resulting ICERs. It concluded that the large degree of heterogeneity and uncertainty arose mainly from the individual response to anaesthesia, the case mix and the variation in administering anaesthesia in clinical practice.
The Committee was advised that population groups considered to be at high risk of adverse events from anaesthesia varied with changes in anaesthesia practice, but that the type of surgery, patient's age, BMI and comorbidities were known risk factors.
The Committee was advised that post-traumatic stress disorder following awareness during surgery can be severe and have far-reaching consequences for the patient's quality of life beyond those considered within the health context (for example, marital breakdown and loss of employment). The Committee also noted that people who experience awareness during surgery can become averse to any contact with the healthcare system and may not seek treatment for conditions in the future. This might mean that the impact of awareness and the costs of treating its consequences have been underestimated.
The Committee was advised that unintended awareness during surgery in patients who receive muscle relaxants is associated with more severe psychological harm than in patients who do not receive muscle relaxants.
The Committee noted that the risk of awareness during surgery in patients receiving inhaled anaesthesia can be reduced using structured anaesthesia protocols such as measuring end-tidal anaesthetic concentration with audible alarms and using MAC values, but the use of such protocols is not possible in patients receiving total intravenous anaesthesia. In patients who are more sensitive to anaesthetic and who are therefore at higher risk of receiving an excess of anaesthetic, such as older patients, the standard levels within such protocols may not be appropriate.
The Committee noted that unintended awareness during surgery could still occur with the use of depth of anaesthesia monitors or structured protocols for measuring end-tidal anaesthetic concentration, but that the use of these interventions lowered the risk. The Committee considered that it is uncertain if the depth of anaesthesia monitors reduce the risk of consciousness without recall.
The Committee acknowledged that distinguishing between late psychological symptoms and post-traumatic stress disorder was difficult but concluded that the adverse impact on quality of life was the same. The Committee noted that the 2 groups had been separated in the cost-effectiveness analyses, and the costs associated with post-traumatic stress disorder were not applied to the group with late psychological symptoms. Therefore, the Committee concluded that the clinical benefits of monitoring could have been underestimated in the cost-effectiveness analyses.
The Committee considered there was uncertainty about the effects of excessively deep levels of anaesthesia. The Committee was advised that there was evidence suggesting an increase in morbidity and mortality associated with excessively deep anaesthesia (for example, an increase in the incidence of stroke or myocardial infarction). They also noted that there was weak evidence showing that excessively deep anaesthesia resulted in post-operative cognitive dysfunction. The Committee noted that these outcomes had not been included in the cost model, and that their absence meant that the clinical benefits of avoiding excessively deep levels of anaesthesia were likely to have been underestimated in the cost-effectiveness analyses.
Although the Committee considered that the clinical benefits associated with reducing adverse outcomes from anaesthesia were underestimated in the model, the Committee also discussed the uncertainty about the extent to which depth of anaesthesia monitoring could reduce these adverse effects and the consequent uncertainty about the cost savings. The Committee noted the possibility that the clinical benefits of monitoring may have been overestimated in the cost-effectiveness analyses.
The Committee noted that potential cost savings associated with reductions in operating theatre time and recovery time were not included in the model. The Committee considered that incorporating the cost savings associated with these outcomes might improve the cost effectiveness of the monitors, but the time savings were too small to significantly benefit clinical practice.
The Committee considered the wide variation in price for the sensors for the 3 monitors (from under £1 to over £14). The Committee noted that there may be technical differences in the sensors, which could affect the accuracy of the monitors, but that there was no evidence of a substantial clinical difference when the sensors are used by anaesthetists well trained in depth of anaesthesia monitoring.
The Committee noted anecdotal evidence that the BIS monitor and sensors could be procured locally at a lower cost than that used in the model.
The Committee noted that despite many large studies, particularly of the BIS monitor, uncertainties remained about the probability of unintended awareness during surgery and the benefits of avoiding excessively deep levels of anaesthesia and, therefore, the extent to which depth of anaesthesia monitors could reduce adverse outcomes. The Committee considered the value of additional research studies before making its recommendations, but concluded that the size, complexity, cost and time requirements of such studies could unduly delay the uptake by the NHS of what is likely to be a beneficial technology.
The Committee concluded that additional research is desirable and should be undertaken by both the manufacturers and clinical researchers to provide additional information about the benefits and costs associated with the use of these technologies. In particular, information is needed about the clinical effectiveness of E-Entropy and Narcotrend-Compact M in reducing unintended awareness during surgery, as is further information about the effectiveness of all 3 monitors in reducing all adverse outcomes of general anaesthesia (including post-operative cognitive dysfunction). The Committee also wished to encourage further research into the clinical implications of accidental awareness during surgery, and the impact of the length and depth of anaesthesia on short- and long-term morbidity and mortality.
The Committee noted that the Royal College of Anaesthetists and the Association of Anaesthetists of Great Britain and Ireland have commissioned a National Audit Project (NAP5 – Accidental Awareness during General Anaesthesia [AAGA]) that will collect data on all reported cases of accidental awareness during general anaesthesia during a 1-year time period. The results of this audit are expected to be published in 2014. The Committee felt that the data from this audit may be of some benefit when this guidance is reviewed. The Committee discussed the potential impact of this guidance on the validity of the audit and concluded that an adverse impact was unlikely.
The Committee noted that only literature written in the English language was included in the assessment, and therefore some studies, particularly on the Narcotrend-Compact M monitor, may not have been included in the evidence base. It was also noted that observational studies comparing the different technologies were not included in the evidence base.
The Committee noted that the modelling gave base-case ICERs for BIS that were above the usual levels accepted by NICE for the adoption of a technology. The Committee noted the considerable uncertainty in many of the parameters of the model and noted that the ICERs were very sensitive to small changes in the parameters. In addition, the Committee noted that the depth of anaesthesia monitors were relatively low-cost interventions, and it was likely that the clinical benefits of using the monitors were underestimated in the base case, particularly those benefits associated with avoiding excessively deep levels of anaesthesia. The Committee considered that the avoidance of uncommon but catastrophic events for patients was an important factor in accepting a technology with an ICER that appeared to be higher than usually acceptable in the base-case results.
The Committee noted that E-Entropy and Narcotrend-Compact M both had ICERs in the acceptable range, but that there was greater uncertainty about their clinical benefit than for the BIS monitor.
Notwithstanding the uncertainty in the evidence base, the Committee considered that depth of anaesthesia monitoring is most likely to be cost effective and of clinical benefit in patients receiving total intravenous anaesthesia and in patients considered at higher risk of unintended awareness or of excessively deep levels of general anaesthesia.
The Committee considered that anaesthetists using depth of anaesthesia monitors should ensure that they have appropriate experience with these monitors and appreciate the potential pitfalls in their use in clinical practice. The Committee considered it important to note that the use of the monitors might require significant changes to clinical practice to achieve clinical benefit, and the skill and experience of the anaesthetist in using the depth of anaesthesia monitor are highly likely to influence the clinical effectiveness of the technique.
The Committee considered possible equality impacts and concluded that the recommendations would be unlikely to disadvantage any groups protected under equalities legislation.