Retrospective evaluation of predictors and clinical impact of postoperative delirium in liver transplant recipients

This study aimed to analyze the epidemiology, predictors, and clinical outcomes of postoperative delirium (POD) in liver transplant (LT) recipients. Clinical data from 427 LT recipients from the Third Xiangya Hospital were retrospectively reviewed. The epidemiology (incidence, onset, and duration), predictors, and associated clinical outcomes of POD were analyzed using standardized delirium assessment tools and multivariate regression. POD occurred in 19.0% of recipients, with a median onset at 26.0 hours posttransplant. Multivariate analysis identified 4 independent pre and postoperative predictors, including a higher model for end-stage liver disease score and hepatic encephalopathy within 24 hours before LT, infection within 2 months prior to LT, and abnormal sodium within 1 hour after LT (all P < .05). In a sensitivity analysis treating early death or persistent coma as a composite outcome with POD (n = 436), the same 4 independent predictors were identified, with consistent effect estimates. Regarding clinical outcomes, POD was significantly associated with prolonged intensive care unit stays, extended mechanical ventilation requirements, and increased use of renal replacement therapy (all P < .01). Although recipients with POD showed higher 1- and 6-month all-cause mortality in unadjusted analyses, POD was not an independent predictor of mortality after adjustment for confounding factors. POD occurs frequently and early after LT, with identifiable pre and postoperative risk factors. While POD markedly increases perioperative morbidity and resource utilization, it does not independently predict short-term all-cause mortality in this cohort.