Limitations of four-slice multirow detector computed tomography in the detection of coronary stenosis.

ABSTRACT

BACKGROUND: Our aim was to compare 4-slice spiral computed tomography with conventional coronary angiography in the detection of significant (> 50%) coronary stenosis. METHODS: Sixty-two patients (41 males, 21 females, mean age 60 +/- 8 years) with suspected coronary artery disease were submitted to coronary angiography and then to multislice spiral computed tomography (GE Light Speed 4 slice) performed 12 +/- 5 days later. RESULTS: We excluded 25% of the patients from analysis because of a heart rate > 70 b/min or because of frequent ectopic beats. We also excluded from analysis 23% of all the angiographic segments judged not evaluable at multislice spiral computed tomography. Within these limits, the sensitivity was 65%, the specificity 98%, the positive predictive value 88%, and the negative predictive value 92%. CONCLUSIONS: By considering the intrinsic limitations such as its low temporal and spatial resolution, 4-slice spiral computed tomography has a limited applicability and has to be used with caution in the evaluation of native coronary arteries.