Association of 3‐Day Postoperative Blood Glucose Variability With All‐Cause Mortality in ICU Patients After Surgical Repair of Acute Type A Aortic Dissection: A Retrospective Cohort Study Using the MIMIC‐IV Database

ABSTRACT Background Stanford type A aortic dissection (TAAD) remains associated with substantial mortality despite prompt surgical repair. Glucose variability has been associated with mortality in intensive care unit (ICU) patients. However, its clinical significance after open TAAD repair remains unclear. Methods We conducted a retrospective cohort study using the MIMIC‐IV v3.1 database (2008–2022). Glucose variability was assessed using the coefficient of variation (%CV) of blood glucose measurements within the first 3 days of postoperative ICU admission. Associations between 3‐day postoperative %CV and outcomes were evaluated using Kaplan–Meier (K–M) analysis, Cox proportional hazards (PH) models, and restricted cubic spline (RCS) analyses. Results A total of 201 adult survivors of TAAD open repair surgery who were subsequently admitted to the ICU were included in the study. In the fully adjusted model, each 1‐standard deviation (SD) increase in the 3‐day postoperative %CV was independently associated with higher ICU mortality (adjusted hazard ratio aHR = 1.820, 95% confidence interval CI: 1.239–2.675, p = 0.0023) and all‐cause in‐hospital mortality (aHR = 1.627, 95% CI: 1.134–2.334, p = 0.0082). RCS analyses showed a positive association between the 3‐day postoperative %CV and each outcome, consistent with an approximately linear pattern. In addition, the 3‐day postoperative %CV showed stronger discriminatory ability for both outcomes than the 3‐day mean and median glucose levels. Conclusions Higher 3‐day postoperative glucose variability was independently associated with increased all‐cause ICU mortality and in‐hospital mortality after open TAAD repair. Postoperative glucose variability may reflect postoperative physiological instability in this high‐risk population.