Postoperative delirium guidelines

This Invited Commentary accompanies the following article: Aldecoa C, Bettelli G, Bilotta F, et al. European Society of naesthesiology evidence-based and consensus-based guideline on postoperative delirium. Eur J Anaesthesiol 2017; 34:192–214. Postoperative delirium can be considered the bane of modern surgery particularly as surgical patients are getting older. Although at first neglected, surgeons, anaesthesiologists and intensive care physicians have now recognised the importance of this topic for our everyday practice. A recent survey of surgeons listed delirium among the most important 'Proposed competencies in geriatric patient care for use in assessment for initial and continued board certification of surgical specialists'.1 We also seem to have finally reached the point where it is generally accepted that postoperative delirium is a complex syndrome that can be triggered by a multitude of predisposing and precipitating factors and is not just a consequence of the anaesthesiologist having a bad day. This journal has recently published the European Society of Anaesthesiology policy on guideline development.2 'Asking the right questions' and 'evaluation of the evidence' are crucial steps when guidelines are written, particularly if the amount of published material on the topic is large. The task force screened almost 6000 hits for relevance when developing the European Society of Anaesthesiology Evidence and Consensus-based Guidelines on postoperative delirium.3 In view of the complexity of the subject, it is encouraging to see that newer guidelines addressing delirium are being developed by multidisciplinary task forces. The current European guideline is no exception. It was drawn up by a group of experts, including anaesthesiologists, surgeons, geriatricians and psychiatrists. However, asking the right questions and appropriate interpretation of data, particularly when much of it is of less than optimal quality, remain at least to some extent subjective, with the possible consequence of introducing bias. It may, therefore, be interesting to compare the current European recommendations with those of another multidisciplinary expert group: the American Geriatrics Society expert panel on postoperative delirium in older adults, published in January 2015.4 One of the strengths of the European guideline is the fact that it concentrates not only on geriatric patients as does the American guideline but includes all age groups and specifically also includes a section on paediatric postoperative delirium. Not surprisingly, the two expert groups agree on many points such as the necessity to implement nonpharmacological measures to prevent delirium, the importance of identifying and managing triggers of delirium as early as possible and the significance of adequate (opioid sparing) postoperative pain control, even if the levels of evidence are not always graded identically. Nevertheless, there are two interesting issues that I would like to highlight. First, although American experts state that there is insufficient evidence to recommend using processed electroencapholagram monitoring intraoperatively, European experts recommend monitoring of depth of anaesthesia. What are the reasons for this difference of opinion? The first is that the guidelines do not share an identical literature base. Although both depend for their conclusion on work by Sieber et al.,5 Chan et al.6 and Radtke et al.,7 the Europeans also included data from Whitlock et al.8 that were not available at the time the American guidelines were developed. In contrast, the Americans, but not the Europeans, included data by Santarpino et al.,9 whose methodology is not comparable with the other studies and cannot be used to explain the discrepancy. The study by Whitlock et al.8 was in itself negative regarding the use of depth of anaesthesia monitoring. However, they provided a meta-analysis, including not only their data but also those of Sieber et al.5, Chan et al.6 and Radtke et al.7 and found that anaesthesia guided by bispectral index (BIS) monitoring was associated with less risk of postoperative delirium, with a summary odds ratio of 0.56 (95% confidence interval, 0.42 to 0.73). However, these three publications supporting depth of anaesthesia monitoring5–7 were available to both task forces, so this is, at least in my opinion, an example where groups of experts interpret the available data differently and reach a different consensus. This highlights the difficulty of developing guidelines. The recommendation to use depth of anaesthesia monitoring will continue to be controversial not least because its implementation would be expensive. Perhaps, our hope as anaesthesiologists is that there may be at least one tool to help us to adapt intraoperative management to prevent delirium that could be justified after all. Future, well-designed prospective trials will hopefully help us to better understand the complex relationships between effects of anaesthetics on the (ageing) brain, depth of anaesthesia and outcome. The second point I would like to highlight is the role of antipsychotics in the treatment of delirium. Although the European guideline suggests (conditional recommendation, Grade of Recommendation B) using low-dose haloperidol or low-dose atypical neuroleptics to treat postoperative delirium, the American guideline differentiates between hypoactive and other forms of delirium. It recommends (weak recommendation statement) the use of typical or atypical antipsychotics '…at the lowest effective dose for the shortest possible duration…to treat delirious patients who are severely agitated or distressed or who are threatening substantial harm to self and/or others', but does not recommend prescribing antipsychotics for the treatment of hypoactive delirium. The authors do, however, acknowledge that this might lead to increased suffering in some patients as hallucinations and delusions, which may be present in hypoactive delirium, might be resolved by such medication. Overall, both panels are clearly reluctant to endorse an uncritical use of our main weapon in the daily battle against delirium. The role of antipsychotics here is a difficult topic because effects are not consistently documented, as a recent meta-analysis indicates.10 The authors were not able to show an effect of antipsychotics on prevention, duration or severity of delirium, on hospital or ICU length of stay, or on mortality. However, neither guideline is able to offer a valid evidence-based alternative to antipsychotic drugs. Guidelines are developed with '…the noble aim of assisting healthcare providers in the choice of appropriate interventions in specific clinical situations'.2 However, implementing guidelines is altogether another topic, and delirium is in my opinion a problem where successful implementation of a guideline is a very demanding undertaking. A systematic review11 concluded that strategies incorporating multiple components are more successful than the use of one single strategy. This is clearly the case for this guideline, as a multitude of interventions is recommended. The time when a magic bullet was thought to be the answer to postoperative delirium is undeniably over. Other important factors that were identified in this review are characteristics of the guidelines themselves. Guidelines that are easy to understand and do not require specific resources have a greater chance of implementation. This is where things become difficult with delirium. Although the guideline is well written and easy enough to understand, the amount of specific resources required even for only a partial implementation is considerable. These range from additional intraoperative monitoring to the most expensive resource: time of our co-workers, time to evaluate preoperative cognitive impairment (using which instrument?), time to evaluate the nutritional status or time to perform delirium screening in each shift for up to 5 postoperative days. Finally, patient characteristics such as co-morbidity, which is present in the majority of delirious patients, also reduce the chance that guidelines are successfully implemented. This suggests that despite the tremendous amount of work that has been invested in these guidelines and for which the task forces should be congratulated, they represent only one more, albeit important, step to reduce the burden of postoperative delirium. Nevertheless, the potential difficulties with implementation should not lead us to accept delirium as a regrettable part of modern surgery and persuade us to concentrate on the implementation of other guidelines that are easier to put into practice. We should not forget that the long-term associations with delirium – dementia, institutionalisation and possibly mortality12–14 – are of utmost importance to our patients. One finding that illustrates this, in a study investigating patients on average 30 months after elective or urgent hip replacement, is that the risk of dementia or mild cognitive impairment at follow-up was almost doubled in those with postoperative delirium compared with at-risk patients without delirium. Half of the patients with delirium were institutionalised at follow-up compared with 29% of the controls.15 How should we go from here? Perhaps an approach similar to that of the Enhanced Recovery After Surgery (ERAS),16 where a variety of interventions is available and the choice of which to implement is left to the individual unit, would be an important step forward in the management of postoperative delirium. ERAS also requires additional resources, some of which involve collecting data and providing feedback to the clinicians. These data not only allow monitoring of how consistently the chosen interventions are implemented in the unit concerned but also the extent to which the goals of the ERAS concept were achieved and how this compares with other units implementing ERAS. Such a feedback system for delirium management would perhaps permit an efficient use of our limited resources, allowing for a targeted fine tuning of our clinical efforts to the needs of a particular patient group or unit. In my opinion, initially determining the incidence, duration and perhaps severity of delirium would be a good start. Despite the fact that we have a well-written up-to-date guideline available, we are far from achieving substantial improvements regarding the burden of delirium in postoperative patients. Perhaps Molière can provide some encouragement: 'The greater the obstacle, the more glory in overcoming it'. Acknowledgements relating to this article Assistance with invited commentary: none. Financial support: none. Conflict of interest: LAS has received speaker honoraria from Covidien (BIS) within the last 5 years. Comment from the editor: this invited commentary was checked and accepted by the editors but was not sent for external peer review.