Electroencephalography-guided anaesthetic administration does not impact postoperative delirium among older adults undergoing major surgery: an independent discussion of the ENGAGES trial

Delirium is a reversible state of impaired cognition, inattention, and altered level of consciousness occurring in ∼20% of adults after major surgery.1, 2 Internationally, healthcare stakeholders have highlighted delirium as a major public health priority, given its strong association with delayed recovery and excess healthcare resource utilisation. The pragmatic Electroencephalography Guidance of Anesthesia to Alleviate Geriatric Syndromes (ENGAGES) trial was designed to investigate whether limiting anaesthetic dose, guided by minimising electroencephalogram suppression during major cardiac and non-cardiac surgery, decreases the incidence of postoperative delirium.³ The premise for this landmark, single-centre study was based on meta-analyses of randomised trials that concluded targeted EEG-guided administration of anaesthesia may decrease postoperative delirium incidence by >33%.⁴ Low bispectral index values (BIS <45) are associated with a higher risk of mortality.¹ However, the causality between EEG-guided depth of anaesthesia and adverse outcomes cannot be inferred from observational studies, as unknown confounders or poorly delineated mechanisms may account for this apparent association. For example, postoperative delirium is more likely in individuals who report frequent preoperative falls or who undergo cardiac surgery.5, 6 This link may merely reflect that patients who are susceptible to delirium also exhibit frequent electroencephalogram suppression during general anaesthesia,⁷ as a consequence of pre-existing neurological and cardiovascular co-morbidity that predisposes to postoperative complications. For example, burst suppression is observed in patients with brain injury, but not during sleep.⁸

The basic premise of the ENGAGES trial was that avoiding burst suppression, an electroencephalographic pattern suggesting excessively deep anaesthesia, may reduce delirium. Using EEG to guide the administration of anaesthesia ‘dosing’ may conceivably minimise burst suppression,⁹ and therefore, directly address the hypothesis that delirium occurs as a result of burst suppression. Electroencephalogram suppression during surgery has been associated with postoperative delirium. In ENGAGES, patients were randomly assigned to either usual anaesthesia care or to EEG guidance of anaesthesia, but also examined whether the type of surgery (cardiac vs non-cardiac) and a history of falls in the 6 months preceding surgery may reveal individuals at particular risk.¹⁰ Here, we consider the entire findings and inferential reproducibility of the ENGAGES trial using a structured independent discussion.¹¹ This process requires the author of the independent discussion (G.L.A.) to only have access to the article's Introduction, Methods, and Results, but not to the Abstract or the Discussion.

Main finding

Original discussion.

‘The primary finding of this trial was that electroencephalography guidance of anesthesia in older adults undergoing major surgery did not decrease the incidence of postoperative delirium, despite successfully reducing anesthetic exposure and duration of electroencephalogram suppression’.³

Independent discussion

The ENGAGES trial failed to find any effect of EEG-guided anaesthesia administration on the primary outcome—postoperative delirium—in older adults undergoing major surgery. The incidence of delirium up to 5 days after surgery was 157/604 (26.0%) in the EEG-guided group, whereas delirium occurred in 140/609 (23.0%) patients in the usual-care group (difference: 3.0%; 95% confidence interval [CI]: –2.0 to 8.0; P=0.22). Logistic regression analysis that adjusted for preselected prognostic co-variates considered likely to be associated with postoperative delirium also failed to detect a beneficial effect on postoperative delirium attributable to the minimisation of electroencephalographic suppression. Taken together, the analysis does not provide evidence to suggest that an intervention to guide burst suppression influences delirium after major surgery.

Commentary

The original and independent discussions of the ENGAGES trial are concordant in their interpretation of the main study finding. The original discussion emphasised that the intervention (a practice instruction) had resulted in separation between the groups in anaesthetic exposure and electroencephalographic suppression.

Relationship of main finding to previous studies

Original discussion.

‘This finding contrasts with recent meta-analyses that reported greater than a one-third reduction in delirium incidence with electroencephalography guidance of anesthesia.4, 11, 12 However, the evidence from these meta-analyses has been appraised as moderate in quality.⁴ The following methodological differences between the current trial and trials comprising the meta-analyses might partially explain the discrepant findings: anaesthetic techniques, compliance with trial protocols, population risk profiles, effect of electroencephalogram guidance on anesthetic management, rigor in delirium ascertainment, and reporting of missing data. In the study by Chan et al.,¹² 921 of 1000 patients undergoing noncardiac surgery were randomized, and 902 were assessed for delirium. In contrast to the current trial, patients were healthier and, on average, underwent shorter surgical procedures. Bispectral index guidance of anesthesia was associated with a significant reduction in anesthetic administration, bispectral index values, and delirium incidence. However, the primary focus was postoperative cognitive dysfunction, and delirium was a secondary outcome only assessed daily with the CAM. Methodological details on the number of missing delirium assessments and training of raters were not reported. In the study by Radtke et al.,¹³ 1277 patients of 1600 patients undergoing non-cardiac surgery were randomized, and 1155 patients were assessed for delirium. Patients were healthier than in the current trial. Bispectral index guidance of anaesthesia was not associated with a significant decrease in average bispectral index values but was associated with a significant decrease in delirium incidence. However, clinicians unblinded themselves for a quarter of the patients in the control group. With per-protocol analysis, by inclusion of these patients in the bispectral index guided group, the difference in delirium incidence between groups was not significant. Delirium was assessed twice daily in this study, but the number of missing assessments was not reported. In the study by Whitlock et al.,¹⁴ 310 patients undergoing mainly cardiac surgery were examined and delirium as a secondary outcome. In contrast to the study by Chan et al.,¹² higher volatile anaesthetic concentration was associated with lower delirium incidence.¹⁴ Bispectral index guidance was not associated with meaningful differences in median anaesthetic concentrations or bispectral index values between study groups. Trained intensive care unit nurses assessed patients for delirium twice daily with the CAM-ICU instrument; there were few missing assessments, and there was not a statistically significant difference in delirium incidence between groups.’

Independent discussion

Preceding studies have been observational, limiting comparisons. Retrospective observational studies have reported that patients who experienced electroencephalogram suppression at lower volatile anaesthetic concentrations had a higher incidence of postoperative delirium. This association remained significant after adjusting for patient and surgery, and the duration of electroencephalogram suppression. Previous observational data from the same investigators reported that the duration of electroencephalogram suppression >4.5 min was associated with markedly higher incidence of delirium (∼45% vs 25% incidence of delirium in individuals with ≤4.4 min of burst suppression). Moreover, no further increase in incidence of delirium was evident with more than three-fold longer episodes. The ENGAGES intervention reduced the median cumulative time spent with electroencephalogram suppression (7 min in EEG guided vs 13 min in control group; difference: –6.0; 95% CI: –9.9 to –2.1). However, the duration for both groups in ENGAGES breached the threshold values found previously to be associated with a marked increased risk of delirium. These data suggest that, unless pilot data using the intervention limited burst suppression to <4 min, the likelihood of ENGAGES showing a reduction in delirium was quite small.

The lack of randomised controlled trials has hampered our understanding as to whether burst suppression or BIS guidance may impact on the development of delirium after surgery. However, such association studies may merely reflect patient susceptibility to haemodynamic compromise, which promotes delirium, or encompass an endotype where patients at risk of delirium who are susceptible to delirium could coincidentally be prone to electroencephalogram suppression during general anaesthesia.⁷ ENGAGES contributes new data that suggest the mechanism underlying delirium is not related to burst suppression.

Commentary

There are important distinctions in how the original and independent discussions relate the ENGAGES trial to prior literature. The original discussion contrasts the primary outcome of the ENGAGES trial with the conclusions of prior meta-analyses, which have favoured EEG guidance. They then briefly discuss the three largest trials included in these prior meta-analyses, and emphasise that differences in the study cohort, exposure, and protocol rigour might contribute to the contrasting results. The independent discussion is thematically similar to the extent that it acknowledges the inadequate quality of previous evidence, but it additionally emphasises a methodological concern rooted in prior literature—that the EEG-guided group in the ENGAGES trial, whilst having a significantly shorter exposure to EEG suppression, still had an exposure that exceeded the threshold previously observed by the ENGAGES authors to be associated with reduced incidence of delirium.⁹ The independent discussion thus questions whether the pragmatic intervention of the ENGAGES trial was likely to succeed. Elsewhere, the original discussion emphasises the separation in exposure between the groups as a study strength, but does not address how the absolute duration of exposure compares to prior evidence.

Additional (secondary) findings

Original discussion.

‘Moreover, attempting to minimize anesthetic exposure is labor intensive and may distract from other priorities. Unintended negative consequences might occur, such as undesirable patient movement during surgery. On the other hand, the lower 30-day mortality in the guided group warrants further investigation’.³

Independent discussion

A large number of inter-related, pre-specified exploratory delirium outcomes were stated in the original protocol article. ENGAGES failed to find any impact of EEG-guided therapy on the incidence of severe delirium, the duration of delirium, time to onset of delirium, or the incidence of delirium in the four randomisation strata (cardiac surgery with no history of falls, cardiac surgery with history of falls, non-cardiac surgery with no history of falls, and non-cardiac surgery with history of falls). Similarly, exploratory 30 day outcomes, including falls and additional cognitive tests, did not differ between patients randomised to either usual anaesthesia care or EEG-guided care. Some other pre-specified secondary endpoints included health-related quality of life, and falls at 1 yr were not reported in this article. The supplementary data also show that EEG-guided therapy was associated with higher 30 day survival; four/614 (0.7%) deaths occurred in the EEG-guided group, compared with 19/618 (3.1%) in the usual-care group (difference: –2.42%; 95% CI: –4.25 to –0.81).³ This post hoc finding is difficult to explain, particularly given the lack of correlation between significant morbidity and EEG-guided therapy.

Commentary

The original and independent discussions are substantively different in the depth with which they address secondary and exploratory findings. The original discussion is lean, focusing almost entirely on the primary outcome, with only one direct reference to any of the secondary findings (30 day mortality), and this is not discussed in any detail. The independent discussion does include a succinct summary of the negative secondary outcomes, and is complete in mentioning secondary outcomes that were defined a priori in the published protocol,¹⁰ but not reported in the article. Both discussions place emphasis on the positive 30 day mortality; the original discussion comments only that it warrants further investigation, whilst the independent discussion suggests that the finding is difficult to explain.

Relationship of additional (secondary) findings to previous studies

Original discussion.

The original discussion does not include any direct comment on how the secondary results relate to previous studies.

Independent discussion

There are limited data on the additional delirium measures reported in ENGAGES. Non-standardised reporting of outcomes in many previous studies makes comparability with ENGAGES difficult. Moreover, given that such outcomes were reported in observational studies, they appear to be of limited comparability to ENGAGES. Many previous studies focused on low BIS values, which appear to be epiphenomenal. Previous ‘depth of anaesthesia’ studies have found an inconsistent relationship with mortality, bearing in mind that these analyses have considered mortality as a secondary outcome. The BALANCED Anaesthesia Study,¹⁵ which enrolled 6500 patients, will make an additional important contribution to resolving this controversy, which has lingered since the first observational study reported an association between hypnotic depth during surgery and 1 yr postoperative mortality.¹⁶

Commentary

Here, the discussions are similar in their omission of any emphasis. The original discussion does not elaborate on prior studies related to the secondary findings at all. The independent discussion makes only a brief summary comment that the existing literature on anaesthetic depth and mortality is inconsistent and with substantial limitations.

Limitations

Original discussion.

‘This trial had several limitations. First, particular practice patterns at the study's single center might have negated the benefit of the intervention. Ongoing multi-center trials, like ENGAGES-Canada (NCT02692300) and the BALANCED Anesthesia Study,¹⁵ might refine the interpretation of this trial. Second, delirium can be difficult to diagnose,¹⁷ with no corroborative biomarkers. Attempts to minimize this limitation included following established methods for delirium assessment10, 17 and blinding assessors to treatment assignment. Third, delirium is a fluctuating disorder and could be missed with interval assessments. To address this issue, CAM assessments were complemented with independent structured medical record review for evidence of delirium. Fourth, enrolment in a clinical trial focused on the prevention of delirium could have decreased the likelihood of delirium occurring. However, the overall delirium incidence (24.5%) was in accordance with the a priori estimate for this patient population.¹⁰ Fifth, a multicomponent intervention, including postoperative medication simplification, educational materials, and postoperative fall safety planning, was implemented for all patients in the trial.¹⁰ This implementation might have affected study outcomes, especially falls and quality of life, but the effect should have been the same in both groups. Sixth, the findings might not apply to general anaesthesia based on intravenous anesthetic agents. Seventh, the bispectral index monitor's suppression ratio parameter might underestimate electroencephalogram suppression.²⁰ To mitigate this issue, clinicians were educated to recognize suppression from the electroencephalogram waveforms and not to rely on this derived parameter.’

Independent discussion

This was a single-centre trial, which limits generalisability. Masking the intervention to attending anaesthesiology staff was clearly impossible. Given previous observational data from the same investigators, pilot data showing that the implementation could reduce the duration of electroencephalogram suppression to <4.5 min would have been particularly instructive. In this regard, the likelihood of ENGAGES showing a reduction in delirium was low. During surgeries, researchers provided real-time feedback to anaesthesia clinicians on the cardinal features of the electroencephalogram waveforms. It is unclear how this may have influenced anaesthesia practice in this, or other, regards. For example, although the median duration of hypotension at various thresholds was not different between groups (Supplement 3 eTable 7 in original article), and more phenylephrine was administered in the usual-care group. Undesirable intraoperative movement was also reported in 22.3% of patients in the guided group and 15.4% in the usual-care group (difference: 6.9%; 95% CI: 2.5–11.4); this occurred despite >90% of patients receiving neuromuscular block and modestly lower anaesthetic doses. The explanation for these differences remains unclear, but suggests that real-time engagement with investigators could have influenced care in an unanticipated manner. No biological samples were collected to address additional potential contributory factors underlying postoperative delirium.

Commentary

The original and independent discussions have some concordance in the limitations they emphasise. Both discussions mention the primary limitation as ENGAGES being a single-centre trial, and both comment on the challenges associated with the definitive diagnosis of delirium. However, there are a number of notable substantive differences. The independent discussion again emphasises that the shorter exposure to EEG suppression in the EEG-guided group still exceeded the beneficial threshold observed in the authors' previous study.⁹ The independent discussion also places particular emphasis on the possibility that knowledge of the EEG state by clinical providers may have influenced care in undetermined but relevant ways, and cites two examples that may reflect as much. With regard to such unintended observer or other study effects, the original discussion does not address knowledge of the EEG on other aspects of care, but does acknowledge that participation in a trial evaluating the prevention of delirium might itself lead to a lower probability of delirium. The original discussion also acknowledges that the fluctuating course of delirium means that the diagnosis may be missed, which was a point raised in their discussion of the limitations of previous studies.

Strengths

Original discussion.

‘The current trial had methodological strengths. First, barriers to successful conduct of the trial were proactively identified and addressed in a pilot phase, where predetermined milestones were achieved.¹⁸ Second, the intervention was successful in modifying anesthetic exposure. This success is important, because without demonstrable effect on anesthetic practice parameters, the biological plausibility of a positive finding (ie, a significant decrease in delirium incidence) would be questionable, and the relevance of a negative finding (ie, no significant decrease in delirium incidence) would be diminished. Third, delirium incidence was sufficiently high to allow detection of clinically meaningful intervention effects. Fourth, the CAM instrument has been validated against reference standard Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition and Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition, Text Revision criteria in multiple studies, and has been appraised as 1 of the 2 most reliable instruments for detecting delirium in a research context.²¹ The CAM has been demonstrated to have excellent psychometric properties for both hypoactive and hyperactive delirium.²² Fifth, with programmatic training coupled with highly structured use of the CAM, excellent inter-rater reliability of the researchers was demonstrated.18, 19 Sixth, by complementing the CAM with validated chart review,²³ delirium detection was bolstered and missed primary outcome assessments were minimized’.³

Independent discussion

ENGAGES was a preregistered randomised clinical trial with clear, predefined, clinically relevant patient-centred outcomes. The statistical plan for these secondary analyses was also finalised and approved before data review. Preoperative preparation was exemplary, with patients in both groups receiving a multicomponent safety intervention, including review of their baseline medications by a geriatric psychiatrist, information on improving safety in the hospital after surgery, and educational material about making the home environment safer to decrease the risk of falls and related injuries. Once enrolled, patients were randomised one to one in blocks of 20 in four strata based on plausible modifiers of risk for postoperative delirium, namely, type of surgery (cardiac vs non-cardiac surgery) and presence/absence of falls in the 6 months preceding surgery.¹⁰

In efforts to optimise the quality of the trial, anaesthesiologists were instructed on typical EEG morphology during volatile-anaesthetic-based general anaesthesia;10, 24, 25 this was reinforced during the study and through online educational modules.²⁵ During surgeries, researchers also provided real-time feedback to anaesthesia clinicians on EEG waveforms. Fidelity checklists were provided to all anaesthesia clinicians during the surgery to complete and sign. Impressively, a checklist was completed for >99.5% of patients in the ENGAGES trial. Indeed, median end-tidal volatile anaesthetic concentration was lower in the guided group, suggesting that the protocol was likely adhered to.

The primary outcome was based on the established criteria for incident delirium, namely, any positive CAM²⁶ or CAM-ICU²⁷ assessment, or chart review determination. Reassuringly, the incidence of the ENGAGES primary outcome of delirium during postoperative Days 1–5 was similar to the incidence predicted by the authors', and by others', preceding work. Pre-specified post hoc sensitivity analyses were conducted, which included modelling improved clinician fidelity to the guided protocol; none of these analyses showed any difference in delirium incidence between groups. Lastly, the investigators made trial data available to interested researchers to facilitate transparency and reproducibility.

Commentary

There is reasonable concordance between the original and independent discussions in the factors recognised as study strengths. Both place particular emphasis on the use of the CAM as a superior instrument for the assessment of delirium, and although with slightly differing focus, both mention the benefits of a high degree of study planning and rigour. However, there are also important differences between the two discussions. Most notably, the original discussion mentions the separation between groups in anaesthetic exposure as a fundamental study strength; the independent discussion does not disagree, but elsewhere cited inadequate shortening of exposure to EEG suppression in the treatment group (which could be viewed as inadequate separation) as a limitation of the study. Reciprocally, the independent discussion sees the multicomponent delirium prevention intervention applied to both study arms as a strength and evidence of rigour, whereas the original discussion cites this as a potentially confounding limitation.

Future directions

Original discussion.

‘The American Geriatric Society, the European Society of Anesthesiologists, and the UK's National Institute for Health and Care Excellence all recommend that intraoperative electroencephalogram monitoring should be considered to prevent excessive anesthetic administration to patients at high risk of postoperative delirium.28, 29, 30 The majority of UK anesthesiologists have not adopted this recommendation.³¹ The results of the current trial challenge the evidence underpinning the recommendation. Ongoing multi-center trials, like ENGAGES-Canada (NCT02692300) and the BALANCED Anesthesia Study,¹⁵ might refine the interpretation of this trial’.³

Independent discussion

Reducing delirium in this population may prove challenging given the lack of mechanistic understanding and confounding contribution of various co-morbidities. A systematic exploration of the interactions between delirium and other common concomitant morbidity may provide additional insights. Further translational work is required to understand the pathophysiology and risk factors associated with such syndromic states. Surgical factors (e.g. bone cementation³²) could be further explored. Strategies involving physical and occupational therapy may help patients to recover function, as championed by enhanced recovery protocols.

Commentary

The original and independent discussions adopt different themes in their consideration of possible future directions. The original discussion makes a strong point that the ENGAGES trial calls into question some of the evidence used in recommendations issued by a number of societies, and effectively suggests that a near-term future undertaking should be critical re-evaluation of those guidelines. They also note the future publication of two large multi-institutional trials as important developments that will address some of the limitations and add to the interpretation of ENGAGES. In contrast, the independent discussion focuses on other approaches to the prevention of postoperative delirium.

Conclusion

Original discussion

‘Among older adults undergoing major surgery, electroencephalography-guided anesthetic administration, compared with usual care, did not decrease the incidence of postoperative delirium. This finding does not support the use of electroencephalography-guided anesthetic administration for this indication’.³

Independent discussion

The ENGAGES trial failed to find any effect of EEG-guided anaesthesia administration on postoperative delirium up to 5 days after surgery in older adults undergoing major surgery.

Commentary

The original and independent discussions are concordant in their conclusion.

Inferential reproducibility

The major inferential difference between the discussions relates to whether the pragmatic intervention in the ENGAGES trial achieved a sufficient reduction in exposure to EEG suppression for there to be a reasonable likelihood of success. The original discussion emphasises as the strength of the study that the intervention did achieve separation between the groups. The independent discussion acknowledges the separation, but stresses that, despite this, the intervention group still had exposure times that exceeded what some evidence would suggest being the threshold at which benefit is likely. The inference is that the negative result of the ENGAGES trial has not eliminated the possibility that a clinical protocol leading to a greater reduction in EEG suppression might protect against the development of postoperative delirium. The other notable inferential difference was the weight given to the secondary finding of reduced 30 day mortality in the treatment group. Although neither discussion elaborated on this finding in detail, the original discussion viewed it as a basis for further investigation, whilst the independent discussion was more restrained, emphasising that it was difficult to explain post hoc. The discussions contained differences of emphasis in other areas, but to a large extent, these reflected differences in approach or less significant points, rather than substantive disagreements on the main inferences to be drawn from this landmark perioperative study.

Authors' contributions

Writing independent discussion: GLA.

Writing commentary on discussions: KOP.

Both authors agreed with the article's results and conclusions, approved the final version of this article, and have read and confirm that they met the International Committee of Medical Journal Editors criteria for authorship.

Declaration of interest

GLA is an editor and KOP is an associate editor of the British Journal of Anaesthesia. GLA is British Oxygen Company/Royal College of Anaesthetists research chair in Anaesthesia [2016–2021].