Abstract
BACKGROUND/OBJECTIVE: Delirium is a common complication in critically ill patients with a negative impact on hospital length of stay, morbidity, and mortality. Little is known on how neurological deficits affect the outcome of commonly used delirium screening tools such as the Confusion Assessment Method for the Intensive Care Unit (CAM-ICU) and the Intensive Care Delirium Screening Checklist (ICDSC) in neurocritical care patients. METHODS: Over a period of 1 month, all patients admitted to a neurocritical care and stroke unit at a single academic center were prospectively screened for delirium using both CAM-ICU and ICDSC. Tool-based delirium screening was compared with delirium evaluation by the treating clinical team. Additionally, ICD-10 delirium criteria were assessed. RESULTS: One hundred twenty-three patients with a total of 644 daily screenings were included. Twenty-three patients (18.7%) were diagnosed with delirium according to the clinical evaluation. Delirium incidence amounted to 23.6% (CAM-ICU) and 26.8% (ICDSC). Sensitivity and specificity of both screening tools were 66.9% and 93.3% for CAM-ICU and 69.9% and 93.9% for ICDSC, respectively. Patients identified with delirium by either CAM-ICU or ICDSC presented a higher proportion of neurological deficits such as impaired consciousness, expressive aphasia, impaired language comprehension, and hemineglect. Subsequently, generalized estimating equations identified a significant association between impaired consciousness (as indexed by Richmond Agitation and Sedation Scale) and a positive delirium assessment with both CAM-ICU and ICDSC, while impaired language comprehension and hemineglect were only associated with a positive CAM-ICU result. CONCLUSIONS: A positive delirium screening with both CAM-ICU and ICDSC in neurocritical care and stroke unit patients was found to be significantly associated with the presence of neurological deficits. These findings underline the need for a more specific delirium screening tool in neurocritical care patients.