Comparison of the diagnostic performance of 64-slice computed tomography coronary angiography in diabetic and non-diabetic patients with suspected coronary artery disease

BACKGROUND: Diabetics have high prevalence of subclinical coronary artery disease (CAD) with typical characteristics (diffuse disease, large calcifications). Although 64-slice multidetector computed tomography (MDCT) coronary angiography has high diagnostic accuracy to detect CAD, its diagnostic performance in diabetics with suspected CAD is unknown. To compare the diagnostic performance of 64-slice MDCT between diabetics and non-diabetics with suspected CAD scheduled for invasive coronary angiography (ICA). METHODS: We enrolled one hundred and five diabetic patients (92 men, age 65 +/- 9 years, Group 1) and 105 non-diabetic patients (63 men, age 63+/-5 years, Group 2) with indication to ICA for suspected CAD undergoing coronary 64-slice MDCT before ICA. RESULTS: In Group 1, the overall feasibility of coronary artery visualization was 93.8%. The most frequent artifact was blooming due to large coronary calcifications (54 artifacts, 67%). In Group 2, the overall feasibility was significantly higher vs. Group 1 (97%, p < 0.0001). In Group 1, the segment-based analysis showed a MDCT sensibility, specificity, positive predictive value, negative predictive value and accuracy for the detection of ≥50% luminal narrowing of 77%, 90%, 70%, 93% and 87%, respectively. In Group 2, all these parameters were significantly higher vs. Group 1. In the patient-based analysis, specificity, negative predictive value and accuracy were significantly lower in Group 1 vs. Group 2. CONCLUSIONS: Although MDCT has high sensitivity for early identification of significant CAD in diabetics, its diagnostic performance is significantly reduced in these patients as compared to non-diabetics with similar clinical characteristics.