Delirium in acute stroke: screening tools, incidence rates and predictors: a systematic review

Delirium is a common complication in acute stroke yet there is uncertainty regarding how best to screen for and diagnose delirium after stroke. We sought to establish how delirium after stroke is identified, its incidence rates and factors predicting its development. We conducted a systematic review of studies investigating delirium in acute stroke. We searched The Cochrane Collaboration, MEDLINE, EMBASE, CINHAL, PsychINFO, Web of Science, British Nursing Index, PEDro and OT Seeker in October 2010. A total of 3,127 citations were screened, full text of 60 titles and abstracts were read, of which 20 studies published between 1984 and 2010 were included in this review. The methods most commonly used to identify delirium were generic assessment tools such as the Delirium Rating Scale (n = 5) or the Confusion Assessment Method (n = 2) or both (n = 2). The incidence of delirium in acute stroke ranged from 2.3-66%, with our meta-analysis random effects approach placing the rate at 26% (95% CI 19-33%). Of the 11 studies reporting risk factors for delirium, increased age, aphasia, neglect or dysphagia, visual disturbance and elevated cortisol levels were associated with the development of delirium in at least one study. The outcomes associated with the condition are increased morbidity and mortality. Delirium is found in around 26% of stroke patients. Difference in diagnostic and screening procedures could explain the wide variation in frequency of delirium. There are a number of factors that may predict the development of the condition.