We included 97 patients with suspected coronary spastic angina who underwent ECG-gated CT and an ergonovine provocation test. The EFV was measured with CT data sets using dedicated software. Coronary spasm was defined as total or subtotal occlusion (compared with the relaxed state after nitroglycerin) that was associated with ischemic ECG changes and concurrent chest pain.
Coronary spasms were observed in 27 patients (28%). The EFV was significantly higher in patients with spasms compared with those without spasms (175.0 ± 57.8 cm3 vs. 129.7 ± 57.8 cm3, p < 0.01). Univariate logistic regression analyses indicated significant relationships between coronary spasms and EFV (per 10 cm3, odds ratio (OR): 1.13; p < 0.01), male gender (OR: 3.34; p < 0.01), and smoking (OR: 3.42; p < 0.01). In the multivariate model, EFV (per 10 cm3, OR: 1.10; p = 0.03) and male gender (OR: 5.94; p = 0.02) remained significant predictors of coronary spasm. The optimal EFV threshold for predicting coronary spasm was 149.4 cm3, identified with a receiver operating characteristic curve.
Increased EFV was associated with ergonovine-induced epicardial coronary artery spasms.