Predictors of longer recognition time in acute aortic dissection: a retrospective observational study

Acute aortic dissection (AAD) carries a high mortality risk closely tied to diagnostic timeliness. Delays remain prevalent, driven by atypical presentations-particularly in patients without a chief complaint of pain. This study identified predictors of a longer recognition time, as indicated by prolonged door-to-CT times, and developed a cumulative prediction model for patients presenting without pain complaints.

We conducted a retrospective study of 197 adult patients diagnosed with AAD at a tertiary emergency department (ED) from 2016 to 2023. Patients were classified as having a shorter (< 74 min) or longer (≥ 74 min; cohort median) door-to-CT completion time. Predictors were identified by least absolute shrinkage and selection operator (LASSO)-penalized logistic regression and decision tree analysis. A subgroup analysis characterized pain-complaint-free patients using a stepwise identification algorithm.

A longer door-to-CT time was associated with female sex, older age, and physiological instability. The strongest multivariable predictors were absence of abrupt pain (odds ratio [OR] 2.81, 95% confidence interval [CI] 1.47-5.46) and non-severe triage classification (OR 4.70, 95% CI 1.71-15.29). Dyspnea (OR 2.14) and elevated diastolic blood pressure (OR 1.02/mmHg) also independently predicted a longer door-to-CT time. Among pain-complaint-free patients (n = 46), a stepwise combination of mediastinal widening on chest radiography, Charlson Comorbidity Index ≥ 3, and Shock Index > 1.0 was present in all identified cases (cumulative case coverage 46/46; 95% CI 92.3%-100%).

This study identifies a high-risk profile for a longer recognition time of AAD, particularly in older female patients and those presenting without sudden pain. For patients unable to report pain, chest radiography, shock index, and comorbidity burden may be considered to help identify atypical presentations warranting urgent evaluation.