Use of multi-slice computed tomography in patients with chest-pain submitted to the emergency department

This study evaluates calcium scoring (CS) and computed tomography angiography (MSCTA) in patients >50 years with chest-pain submitted to the emergency department utilising CS as a "diagnostic filter" upfront. Results of CS and MSCTA performed by a 64-slice CT scanner were compared to invasive coronary angiography (ICA). 289 consecutive symptomatic patients (185 men, mean age 71.3 ± 6.4 years) were included. In patients with CS = 0 (Group I; n = 60) or CS > 400 (Group III; n = 95) we refrained from MSCTA, whereas patients with CS 1-400 (Group II; n = 134) underwent subsequent MSCTA. ICA detected significant coronary artery disease (CAD) in 162 patients (56.1%; male 98). None of Group I-patients showed CAD, but in Group III CAD prevalence increased to 82.1%. In Group II, MSCTA correctly identified 177/190 significantly diseased vessel segments. Compared to CS alone, our approach increased sensitivity to 98.1% (+1.8%), specificity to 82.6% (+27.5%) and negative predictive value (NPV) to 97.2% (+5.1%) as well as positive predictive value to 87.8% (+14.6%), respectively. Overall DA was 91.3%. Stratification of symptomatic patients into three different risk groups according to CS results with concomitantly increasing disease prevalence is possible. Zero calcium was found to exclude significant CAD, but needs further evaluation. Still server calcifications impair image quality in MSCTA. Thus direct referral to ICA might be a reasonable approach in case of high CS. In patients with intermediate CS, MSCTA is able to rule out significant CAD with a high NPV.