Postoperative delirium (POD) and postoperative cognitive dysfunction (POCD) are frequent complications in elderly patients undergoing cardiac and noncardiac surgery.1 They have been associated with delayed recovery, decreased quality of life, persistent neurocognitive disorders and increased morbidity and mortality within the first year of surgery. The concern that anaesthetic and sedative agents may be implicated in the pathogenesis of POD has prompted multiple studies. Regional anaesthesia, avoiding the use of sedative–hypnotic anaesthetics, rather than general anaesthesia, has thus long been held up as one strategy to prevent postoperative cognitive dysfunction. However, two recent large, randomised studies (the REGAIN and RAGA trials) comparing regional and general anaesthesia in elderly patients undergoing hip fracture surgery reported no significant difference between groups in the percentage of patients with POCD.2,3 In their accompanying editorial, Avidan et al.4 concluded that surgery with general anaesthesia was not implicated in persistent neurocognitive disorders at a population level. Even so, there might still be individuals or patient subgroups with specific vulnerabilities to POD5 including some older surgical patients with mild cognitive impairment (MCI), characterised by subclinical or unnoticed cognitive decline often missed by most of the tests, and where POD and early POCD can reflect a cognitive trajectory towards a clinically apparent cognitive decline.5 Important risk factors for postoperative neurocognitive disorders include pre-operative general and cognitive health, and medical comorbidities.5 Prevention of POD and POCD should, therefore, focus on improvement of pre-operative fitness, optimisation of pre-operative comorbidities and mitigating physiological and biochemical disturbances secondary to surgical stress and inflammatory reactions to avert organ dysfunction and peri-operative stroke.5 Measures to reach these goals should be part of good clinical practice in all patients but they have only been scantily tested specifically to prevent POCD. Neuroinflammation secondary to surgery is also considered a major contributor to the development of POCD,6 and ageing, atherosclerosis and neurodegenerative diseases may lead to an increased systemic and cerebral inflammatory response to surgery. Also, postoperative complications, such as infection, pulmonary complications and hypoxaemia may aggravate the postoperative inflammatory reaction. Several drugs with anti-inflammatory properties (steroids, NSAIDs, intravenous lidocaine, ketamine and dexmedetomidine) have been used to mitigate postoperative systemic inflammation and subsequent neuro-inflammation but without definitive conclusions.1,6 Enhanced recovery after surgery (ERAS) programmes, as those recommended by the ERAS Society (www.erassociety.org) and GRACE (www.grace-asso.fr), form a multidisciplinary, multimodal approach designed to control surgical stress response including inflammatory reaction in order to hasten postoperative recovery and decrease postoperative complications.7,8 Their efficiency has been widely demonstrated in all surgical specialties. Interestingly, several studies using ERAS programmes for colorectal surgery 9,10 and orthopaedic surgery 11,12 reported a significant reduction in the incidence of POD and POCD. These observations are unsurprising as these programmes combine several measures recommended by the European Society of Anaesthesiology and Intensive Care to prevent POD in elderly patients13 by shortening pre-operative fasting, avoidance of premedication, multimodal opioid-sparing analgesia, avoidance of unnecessary indwelling catheters, early mobilisation, early nutrition and discharge to home environment. All these measures are likely to facilitate return to normal circadian rhythm, hasten the recovery of patients’ usual behaviour in hospital and after discharge and improve patient orientation.13 Recommending ERAS programmes to prevent POD, therefore, appears justified especially as elderly patients, at a higher risk of POD and POCD, benefit much from these programmes.14 Likewise, the POSE-study highlights the importance of a comprehensive peri-interventional and patient-centred management for older patients.15 Disappointingly, the literature abounds with examples of large RCTs with short-term interventions for POD/POCD but without information or incorporation of enhanced recovery programmes.1,2,16,17 In summary, future studies on the prevention of POD should, therefore, focus less on hypothetical short-term interventions unless fully implemented ERAS programmes have been instituted, based on the pathophysiological background 7,8 and positive preliminary observations.9–12