Effect of seasons on delirium in postoperative critically ill patients: a retrospective analysis

Background and objectives: Postoperative delirium is common in critically ill patients and is known to have several predisposing and precipitating factors. Seasonality affects cognitive function which has a more dysfunctional pattern during winter. We, therefore, aimed to test whether seasonal variation is associated with the occurrence of delirium and hospital Length Of Stay (LOS) in critically ill non-cardiac surgical populations. Methods: We conducted a retrospective analysis of adult patients recovering from non-cardiac surgery at the Cleveland Clinic between March 2013 and March 2018 who stayed in Surgical Intensive Care Unit (SICU) for at least 48 hours and had daily Confusion Assessment Method Intensive Care Unit (CAM-ICU) assessments for delirium. The incidence of delirium and LOS were summarized by season and compared using chi-square test and non-parametric tests, respectively. A logistic regression model was used to assess the association between delirium and LOS with seasons, adjusted for potential confounding variables. Results: Among 2300 patients admitted to SICU after non-cardiac surgeries, 1267 (55%) had postoperative delirium. The incidence of delirium was 55% in spring, 54% in summer, 55% in fall and 57% in winter, which was not significantly different over the four seasons (p = 0.69). The median LOS was 12 days (IQR = [8, 19]) overall. There was a significant difference in LOS across the four seasons (p = 0.018). LOS during summer was 12% longer (95% CI: 1.04, 1.21; p = 0.002) than in winter. Conclusions: In adult non-cardiac critically ill surgical patients, the incidence of postoperative delirium is not associated with season. Noticeably, LOS was longer in summer than in winter.