Postoperative delirium screening tools for post‐anaesthetic adult patients in non‐intensive care units: A systematic review and meta‐analysis

Abstract Aims and Objectives To identify the most accurate postoperative delirium screening tools for detecting postoperative delirium among patients who underwent general anaesthesia surgery in general wards. Background The lack of detection of postoperative delirium can negatively affect the patient's condition, along with their postoperative treatment and rehabilitation, and it can prolong their hospitalisation, persists cognitive dysfunction and increases mortality. Screening for postoperative delirium in hospitalised patients as nursing assessment is routine clinical practice for early detection. Design A systematic review and meta‐analysis. Methods MEDLINE, Embase, CINAHL, KoreaMed and Cochrane electronic databases were searched using the key words delirium, postoperative, assessment or screening, and adult for articles published up to April 2020, with no limit on the year of publishing. Only prospective cohort studies reporting sensitivity and specificity values were included. We followed the recommendations of the Cochrane Handbook of Diagnostic Test Accuracy Reviews and the PRISMA checklist. The Quality Assessment of the Diagnostic Accuracy Studies‐2 tool was used for data extraction and quality assessment, while a bivariate random‐effects meta‐analysis model was used for pooling and comparing diagnostic accuracy and providing a summary of evidence. Results Six delirium assessment tools were evaluated from nine papers including 3088 patients. Due to the limited number of papers, the meta‐analysis included the Confusion Assessment Method (CAM) and its variants, Delirium Detection Score (DDS) and Nurses' Delirium Screening Checklist (NuDESC). Overall, NuDESC demonstrated higher sensitivity than CAM or DDS, while all showed high specificity (0.90 or greater). Conclusion This review suggested that NuDESC can be employed as an accurate screening tool with high specificity for assessing postoperative delirium during routine checkups. However, it is necessary to consider suitable cut‐off values, which is the reference point, in accordance with the clinical setting and the patients’ condition. Relevance to clinical practice NuDESC reported the best evidence of diagnostic accuracy, and we recommend clinical nurses to employ this easy‐to‐use and validated tool for daily screening of postoperative delirium in general wards to facilitate its early detection and the accurate estimation of its prevalence.