Prevalence of post-stroke delirium in acute settings: A systematic review and meta-analysis

Background: Although post-stroke delirium is a frequent complication in acute settings and has significant long-lasting implications for patients, the prevalence of post-stroke delirium diagnosed according to gold-standard criteria remains uncertain. This highlights the importance of updating the current evidence. Objective: This review aims to provide a precise estimation of the prevalence of delirium in acute post-stroke settings and to explore potential differences in the reported prevalence according to the characteristics of the study and stroke survivors. Design: Systematic review and meta-analysis. Setting and participants: Post-stroke survivors with delirium in acute settings. Data sources: Five English and two Chinese language databases from inception to June 2023, and the reference lists of published reviews. Methods: Studies reporting the prevalence of post-stroke delirium according to the Diagnostic and Statistical Manual of Mental Disorders criteria with the description of assessment method in the absence of interventions for delirium were eligible. Two reviewers independently screened searched records and extracted data from eligible studies. A meta-analysis was conducted using a random-effects model to evaluate the prevalence of post-stroke delirium, and predefined subgroup analyses, sensitivity analyses, and mixed-effects meta-regression were conducted to explore the contribution of heterogeneity. Results: Twenty-nine eligible studies involving 8839 survivors were included. In acute settings, the overall weighted prevalence of post-stroke delirium was 25 % (95 % CI, 20-30; I-2, 95 %). Subgroup analyses revealed age-dependent variations in delirium prevalence, with rates of 20 %, 25 %, and 34 % for ages 60-64, 65-74, and 75-79, respectively. Delirium prevalence varied by assessment duration (7 days: 31 %, > 7 days: 24 %, < 7 days: 20 %). Initial screening significantly reduced reported prevalence to 15 %, compared with 30 % when no initial screening was conducted. Delirium prevalence was lower when nurses were involved in assessments (13 % vs. 29 %). No significant effects were detected for economic status, publication year, female percentage, onset-to-admission time, stroke type, and assessment frequency. Meta-regression analysis revealed that higher delirium prevalence was associated with increased survivors' age (beta = 0.051; 95 % CI, - 0.0002 to 0.103; P = 0.05; R-2 = 13.36 %). Conclusion: A high prevalence of post-stroke delirium was observed in acute settings. Nurses are required to integrate and coordinate the prevention, early detection, and management of delirium following stroke into their nursing practice, particularly for older stroke survivors. The findings necessitate the need for incorporating specialized training within nursing education to identify and manage the delirium effectively and accurately. This reinforces the pivotal role of nurses in the early detection and prevention of delirium within the acute stroke care continuum. Tweetable abstract: Evidence suggests 25 % prevalence of post-stroke delirium in acute settings, with higher rates in older patients, emphasizing the need for prevention, early detection, and management in nursing practice. (c) 2024 Elsevier Ltd. All rights reserved.