Multimodality approach to low-flow, low-gradient aortic stenosis: sex-specific ct calcium thresholds validated by dobutamine stress echocardiography

Abstract Background Diagnosing true severe aortic stenosis (AS) in patients with low-flow, low-gradient AS (LFLG-AS) remains challenging. While dobutamine stress echocardiography (DSE) is recommended to diagnose true AS severity, computed tomography (CT)-based aortic valve calcium (AVC) scoring provides quantitative anatomical evaluation. Current guideline thresholds for AVC (2,000 AU in men and 1,200 AU in women) are based on concordant severe AS, and may not be applicable to LFLG-AS. Purpose To determine sex-specific AVC thresholds on CT for diagnosing true severe AS in patients with LFLG-AS, using DSE as the reference standard. Methods We retrospectively analyzed patients with LFLG-AS who underwent both ECG-gated CT and DSE between April 2020 and March 2025. True severe AS was defined by AVA < 1.0 cm², mean gradient < 40 mmHg, stroke volume index < 35 mL/m², and presence of contractile reserve on DSE. AVC scores were calculated non-contrast-enhanced Agatston scores. Results Seventy patients were initially identified. After excluding 7 patients due to inconclusive DSE, 63 patients (35 male and 28 female；mean age 83 ± 7 years) were included in the final analysis. The AUC for AVC predicting true severe AS was 0.792 in male (95% CI: 0.642–0.942, p<0.001) and 0.834 in female (95% CI: 0.685–0.983, p<0.001). Optimal calcium thresholds were 1430 AU for male and 948 AU for female—both lower than guideline-recommended cutoff values. These thresholds yielded sensitivity and specificity of 0.79 and 0.81 in male, and 0.76 and 0.91 in female, respectively. Conclusion In LFLG-AS, sex-specific CT calcium thresholds for true severe AS are lower than current guideline values. AVC alone may underestimate AS severity, especially in female. Combing CT and DSE improves diagnostic accuracy and supports the use of multimodality approach.