Risk factors and associated outcomes of postoperative delirium after open abdominal aortic aneurysm repair

OBJECTIVE: Open abdominal aortic aneurysm repair (OAR) is a major vascular procedure that incurs a large physiologic demand, increasing the risk for complications such as postoperative delirium (POD). We sought to characterize POD incidence, identify delirium risk factors, and evaluate the effect of delirium on postoperative outcomes. We hypothesized that POD following OAR would be associated with increased postoperative complications and resource utilization. METHODS: This was a retrospective study of all OAR cases from 2012 to 2020 at a single tertiary care center. POD was identified via a validated chart review method based on key words and Confusion Assessment Method assessments. The primary outcome was POD, and secondary outcomes included length of stay, non-home discharge, 90-day mortality, and 1-year survival. Bivariate analysis as appropriate to the data was used to assess the association of delirium with postoperative outcomes. Multivariable binary logistic regression was used to identify risk factors for POD and Cox regression for variables associated with worse 1-year survival. RESULTS: Overall, 198 OAR cases were included, and POD developed in 34% (n = 67). Factors associated with POD included older age (74 vs 69 years; P < .01), frailty (50% vs 28%; P < .01), preoperative dementia (100% vs 32%; P < .01), symptomatic presentation (47% vs 27%; P < .01), preoperative coronary artery disease (44% vs 28%; P = .02), end-stage renal disease (89% vs 32%; P < .01) and Charlson Comorbidity Index score >4 (42% vs 26%; P = .01). POD was associated with 90-day mortality (19% vs 5%; P < .01), non-home discharge (61% vs 30%; P < .01), longer median hospital length of stay (14 vs 8 days; P < .01), longer median intensive care unit length of stay (6 vs 3 days; P < .01), postoperative myocardial infarction (7% vs 2%; P = .045), and postoperative pneumonia (19% vs 8%; P = .01). On multivariable analysis, risk factors for POD included older age, history of end-stage renal disease, lack of epidural, frailty, and symptomatic presentation. A Cox proportional hazards model revealed that POD was associated with worse survival at 1 year (hazard ratio, 3.8; 95% confidence interval, 1.6-9.0; P = .003). CONCLUSIONS: POD is associated with worse postoperative outcomes and increased resource utilization. Future studies should examine the role of improved screening, implementation of delirium prevention bundles, and multidisciplinary care for the most vulnerable patients undergoing OAR.