An exploratory study on the heterogeneity of postoperative delirium: Preoperative cognitive screening does not detect hallucinatory delirium risk.

ABSTRACT

INTRODUCTION: This study compared the clinical characteristics of patients who had hallucinations and those who did not during delirium after elective cancer surgery. MATERIALS AND METHODS: This study retrospectively observed 1137 consecutive patients who had preoperative screening by a trained nurse before elective major urologic cancer surgery in our department. We compared the patient characteristics, including mini-cognitive assessment instrument (Mini-Cog) and Geriatric-8 (G8) scores, between those who developed postoperative delirium and those who did not, and also between those who had hallucinations and those who did not during delirium. RESULTS: Out of 1137 patients, 68 developed postoperative delirium, and 12 of them had hallucinations. A hierarchical cluster analysis based on the G8 and Mini-Cog scores divided the patients into two groups one with high G8 and cognitive function (36 patients) and one with low G8 and cognitive function (32 patients). Hallucinations during delirium were more frequent in the high G8 and cognitive function group (11 out of 36 patients) than in the low G8 and cognitive function group (one out of 32 patients). Patients who had hallucinations during delirium also had higher preoperative Mini-Cog scores (P = 0.002) and G8 scores (P = 0.03) than those who did not, indicating better cognitive function and less frailty. DISCUSSIONS We identified a patient population that is prone to hallucinations that preoperative screening tools cannot detect. This suggests the heterogeneity of postoperative delirium and the need for further research.