External validation of the SPAU2 score for 1-year risk stratification in acute type A intramural hematoma
S Mizushima, S KageyamaAbstract
Background
Acute type A intramural hematoma (IMH) is a high-risk acute aortic syndrome in which early identification of patients who may require urgent surgery is crucial, particularly in regions where initial medical management is often selected. We previously developed the SPAU2 score, an admission-based risk score integrating systolic blood pressure (SBP), pericardial effusion (PE), ascending aortic diameter, and ulcer-like projection (ULP), to predict a 1-year composite endpoint (cardiovascular death or operation).(1) The derivation study has been cited by the 2022 ACC/AHA guideline for the diagnosis and management of aortic disease,(2) highlighting the need for external validation.
Purpose
To externally validate the SPAU2 score in a newly established, temporally distinct cohort from the same institution.
Methods
We conducted an external validation study using a new single-centre cohort (2018–2026) and compared it with the original derivation cohort (2011–2016). To ensure strict comparability, we applied identical endpoint definitions and event adjudication rules as in the original study. The primary endpoint was a composite of cardiovascular death or operation for type A IMH within 1 year after onset; for consistency across analyses, the endpoint was redefined based on event dates within 365 days. Patients with missing follow-up were excluded, and follow-up was truncated at 365 days. Discrimination was assessed using ROC analysis (prespecified cutoff ≥2 points), and time-to-event occurrence was evaluated using Kaplan–Meier cumulative incidence.
Results
The validation cohort included 85 patients. Overall, 142 patients with available follow-up were analysed (derivation n=57; validation n=85). Baseline characteristics in the validation cohort were broadly comparable to those in the original report (mean age 74.1 years, 40% male, mean admission SBP 129 mmHg, mean ascending aortic diameter 46 mm). The prevalence of key imaging findings was also similar, with only modest differences: ULP in the ascending aorta and pericardial effusion tended to be slightly more frequent in the validation cohort than in the original derivation report (ULP 39.8% vs 34.5%, PE 50.6% vs 45.6%). Within 365 days, the composite endpoint occurred in 28/57 (49.1%) in the derivation cohort and 31/85 (36.5%) in the validation cohort (Figure 1). In the validation cohort, the SPAU2 score showed excellent discrimination for the primary endpoint with an AUC of 0.858 (Figure 2). Using a cutoff of ≥2 points, sensitivity was 96.8% and specificity was 61.1%.
Conclusion
In a temporally independent external validation cohort, the SPAU2 score retained strong performance for predicting cardiovascular death or need for operation within 1 year in acute type A IMH, supporting its use for early risk stratification at admission.Cumulative incidence of primary endpointFor image description, please refer to the figure legend and surrounding text.ROC curve of the SPAU2 scoreFor image description, please refer to the figure legend and surrounding text.