Purpose: To compare vascular conspicuity and ability to connect pulmonary arterial branches on pulmonary angiograms obtained with helical multi-detector row computed tomography (CT) with those on pulmonary angiograms obtained with helical single-detector row CT.
Materials and methods: Of 93 consecutive patients suspected of having pulmonary embolism, 48 underwent scanning with multi-detector row CT and 45 with single-detector row CT; scans were obtained in 9 seconds and 28 seconds with 2.5-mm and 3.0-mm collimation, respectively. The lungs were divided into three zones: central, middle, and peripheral. Two independent observers used five-point grading scales.
Results: Conspicuity of pulmonary arteries in the central zone was ranked equal (median of 5), but in the middle and peripheral zones it was significantly higher at multi-detector row CT than at single-detector row CT (median 5 vs 4 and 4 vs 3, P < .001, respectively). In addition, multi-detector row CT improved the ability to connect peripheral arteries with their more centrally located pulmonary artery of origin in the peripheral but not the middle zone on transverse images and in both zones on multiplanar images. Viewing with a modified window setting (width, 1,000 HU; level, -100 HU) significantly increased pulmonary arterial conspicuity. Contrast material column in the pulmonary arteries was significantly more homogeneous at multi-detector row CT.
Conclusion: Use of multi-detector row CT significantly improves pulmonary arterial visualization in the middle and peripheral lung zones.