Neurological Complications and ICU Workload After Emergency Repair of Acute Type A Aortic Dissection

Single-center retrospective cohort at China Medical University Hospital (Taichung, Taiwan) using fully de-identified electronic health records. Consecutive adults who underwent emergency repair of acute type A aortic dissection between 2021-01-01 and 2025-04-30 were pooled into one cohort. The study measures the incidence and patterns of early postoperative neurological complications and evaluates their association with intensive care unit (ICU) resource use, focusing on prolonged ICU length of stay (LOS ≥ 10 days), ICU and hospital LOS, and duration of mechanical ventilation (MV). No new data collection or patient contact occurs. Institutional Review Board (IRB) approval: CMUH114-REC1-139.

Adults (≥ 18 years) undergoing emergency repair for acute type A aortic dissection (ATAAD) at China Medical University Hospital (CMUH) during 2021-01-01 to 2025-04-30 were screened. Exclusions: missing key variables, preoperative stroke within 30 days or modified Rankin Scale (mRS) ≥ 4, pregnancy. The analytic cohort included 274 of 309 screened patients. Neurological complications comprise radiology-confirmed stroke, other focal neurological deficits, paraplegia, or sustained coma/encephalopathy recorded by neurology consultation. Prolonged ICU stay is prespecified as ICU LOS ≥ 10 days in the index admission. Outcomes: primary outcomes are prolonged ICU stay and any postoperative neurological complication; secondary outcomes are ICU LOS (days), duration of mechanical ventilation (hours), hospital LOS (days), in-hospital mortality, and in-hospital death within 30 days of the index surgery. Analyses are performed in R version 4.5.1. Continuous variables are summarized as medians with interquartile ranges (IQRs) and compared using the Mann-Whitney U test; categorical variables are presented as counts and percentages and compared using chi-square or Fisher exact tests. Multivariable logistic regression identifies factors associated with prolonged ICU stay (ICU LOS ≥ 10 days) among hospital survivors using a priori covariates (age, sex, chronic kidney disease stage 4-5, operation time, cardiopulmonary bypass \[CPB\] time, intraoperative red blood cell \[RBC\] units transfused, postoperative acute kidney injury \[AKI\], any postoperative neurological complication). A prespecified sensitivity analysis in the full cohort treats the endpoint as ICU LOS ≥ 10 days or in-hospital death using the same prespecified covariates. Data were de-identified before analysis and stored on secure hospital servers; the study is minimal risk and does not involve United States Food and Drug Administration (US FDA) regulated products.