Diagnosis of delirium, confusion, and altered mental status in the emergency department using head computed tomography: A meta-analysis

Background and objective Patients in the emergency department (ED) often have delirium, unconsciousness, and mental state fluctuation, which often hide deep structural lesions in the brain. In the ED, head computed tomography (CT) was a commonly used method to help judge the patient's condition. The purpose of this meta-analysis is to analyze the specific abnormal rate of brain CT in ED patients with delirium, unconsciousness, or mental state fluctuation. Methods This study systematically searched and included 13 observational studies involving a total of 8316 ED patients with delirium, unconsciousness, or mental state fluctuation, all of whom underwent head CT. The revised Newcastle-Ottawa Scale (NOS) was used to assess the research level, and the study summarized the prevalence rate using a random-effects model (restricted maximum likelihood [REML] method. Results The overall quality selected was medium to high (NOS score is 6-8). Among delirious ED patients, the incidence of abnormal head CT was 28% (95% confidence interval [CI]: 18%–38%). Among specific findings, the prevalence of intracranial hemorrhage was 30% (95% CI: 20%–40%), that of cerebral infarction was 35% (95% CI: 19%–52%), and that of brain tumor was 11% (95% CI: 6%–16%). The odds ratio for abnormal CT findings in patients with focal neurological deficits (compared to those without) was 0.05 (95% CI 0.02-0.14, P < 0.00001), indicating a strong protective effect. Conclusion Among patients who come to the ED because of delirium, or abnormal mental condition, the rate of structural abnormalities on head CT is high, with intracranial hemorrhage and cerebral infarction being the most common findings. This highlights the clinical necessity of routine neuroimaging in this population to exclude critical intracranial pathologies. Furthermore, focal neurological deficits are significantly associated with abnormal CT findings and can serve as an important clinical indicator for identifying high-risk patients.