Effect of sedation level on the prevalence of delirium when assessed with CAM-ICU and ICDSC

Purpose We hypothesized that reduced arousability (Richmond Agitation Sedation Scale, RASS, scores −2 to −3) for any reason during delirium assessment increases the apparent prevalence of delirium in intensive care patients. To test this hypothesis, we assessed delirium using the Confusion Assessment Method for the Intensive Care Unit (CAM-ICU) and Intensive Care Delirium Screening Checklist (ICDSC) in intensive care patients during sedation stops, and related the findings to the level of sedation, as assessed with RASS score. Methods We assessed delirium in 80 patients with ICU stay longer than 48 h using CAM-ICU and ICDSC during daily sedation stops. Sedation was assessed using RASS. The effect of including patients with a RASS of −2 and −3 during sedation stop (“light to moderate sedation”, eye contact less than 10 s or not at all, respectively) on prevalence of delirium was analyzed. Results A total of 467 patient days were assessed. The proportion of CAM-ICU-positive evaluations decreased from 53 to 31 % (p < 0.001) if assessments from patients at RASS −2/−3 (22 % of all assessments) were excluded. Similarly, the number of positive ICDSC results decreased from 51 to 29 % (p < 0.001). Conclusions Sedation per se can result in positive items of both CAM-ICU and ICDSC, and therefore in a diagnosis of delirium. Consequently, apparent prevalence of delirium is dependent on how a depressed level of consciousness after sedation stop is interpreted (delirium vs persisting sedation). We suggest that any reports on delirium using these assessment tools should be stratified for a sedation score during the assessment