Development and internal validation of a computed tomography angiography‐based prediction model for acute rupture in type A aortic dissection

Background: While emergency surgery for acute type A aortic dissection (TAAD) is lifesaving, real-world delays occur. Current risk stratification lacks objective imaging criteria. We aimed to develop a computed tomography angiography (CTA)-based model to predict imminent rupture risk in TAAD. Methods: This retrospective study included 121 consecutive patients with TAAD, diagnosed at Beijing Anzhen Hospital between January 2024 and March 2025. Patients were classified into acute rupture (n=46, 38%) and non-acute (n=75, 62%) groups. The primary endpoint was acute rupture-related death within 14 days of symptom onset, rigorously defined by CTA evidence and/or clinical findings. Clinical and imaging variables were screened using univariable logistic regression, followed by least absolute shrinkage and selection operator (LASSO) regression. Predictive performance was evaluated using receiver operating characteristic (ROC) curves, bootstrap validation, calibration, and decision curve analysis (DCA). Results: The overall acute rupture rate was 38% (46/121). The median age of the cohort was 56 years; 72% were men. Univariable analysis showed that rupture patients had higher inflammatory and ischemic marker levels and were more likely to present with circumferential dissection, aortic sinus entry tear, pericardial effusion, entry tear diameter >20 mm, and ascending aorta diameter >50 mm (all P