Modifiable risk factors for post-operative delirium in older adults undergoing major non-cardiac elective surgery: a multi-centre, trainee delivered observational cohort feasibility study and trainee survey

Background: Post-operative delirium (POD) is an acute brain failure which may occur following major surgery, with serious implications for participants and caregivers. Evidence regarding optimal anaesthetic management for older participants at higher risk of POD is conflicting. We conducted a feasibility study of our protocol in 5 centres to guide sample size estimation and inform future recruitment strategies for a larger cohort study.

Methods: Participants aged over 65 and scheduled for major surgery were recruited. They were assessed pre-operatively for delirium, cognitive impairment, depression, comorbidity, activity levels and alcohol use. Details of management during surgery, all medications and complications were recorded by a trainee-led research team. Participants were assessed for delirium in the immediate recovery period and then on post-operative days 1-4 using the 4 question attention test (4AT) with complications assessed at day 4 using the post-operative morbidity survey (POMS). Primary outcomes were the incident rates of POD. Secondary outcomes were number of eligible patients, recruitment rates and retention rates throughout the study, time required for data collection, preoperative risk factors assessment and daily postoperative delirium assessments. Also to assess the added value of employing the regional trainee research network (INCARNNET) to deliver the study. Specifically, what proportion of patient consent, data collection and post-operative testing is performed by anaesthesia trainees from this group, especially the success of weekend delirium assessment by trainees? A survey was completed at the end of the study by the trainees involved regarding their involvement in the study.

Results: Ninety-five participants were recruited, of whom 93 completed the study. Overall, POD occurred in 9 patients. Of these, three were detected in recovery and six on post-op days 1-4. Median length of stay was 6 days. Recruitment rates were high in all but one site. 59 (62%) participants were consented by trainees and 189 (63%) of post op delirium assessments were performed by trainees. A total of six patients declined the study (in a follow up survey of trainees). Pre-existing cognitive impairment, depression and problem drinking were detected in 4(4.3%), 3(3.2%) and 5(5.37%) participants, respectively. Co-morbidity was common with 55(59%) in class three or four of the geriatric index of morbidity. Overall, from a total of 641 data points, levels of missing data were as follows, site A = 9.3%, B = 13.5%, C = 15.4%, D = 10.9%, E = 11.1% (data could not be completed retrospectively).

Conclusions: A multi-centre observational cohort study of delirium carried out by UK trainee anaesthetists is feasible. Patients are content to undergo day of surgery consent and multiple short questionnaires pre-operatively. Proposed data, especially pharmacological, should be carefully considered for their relevance to modifiable mechanisms that can lead to POD. Future research to enable prognostic modelling of POD should involve large scale cohort studies of enriched populations to capture a higher POD incidence. POD remains a common complication in older persons undergoing major surgery in the UK and studies of interventions are urgently needed.

Trial registration: All methods were carried out in accordance with relevant guidelines and regulations. The study was retrospectively registered with ISRCTN94663125 on 07/02/2018.