Subjects
- 85 consecutive patients,
who underwent CCTA from September from December 2011 to February 2012,
were prospectively enrolled.; All patients had suspected or diagnosed coronary artery disease and were examined based on the clinical requirement.
- Excluded 11 patients; who had inappropriate heart beats (i.e.,
very fast [>75 bpm] or irregular heart rates) in spite of beta-blocker administration.
- 74 patients were enrolled as subjects
– 24 women,
50 men
– mean age,
58 ± 10.9 years (range,
23–76 years)
Data Acquisition and Processing
- All 74 patients underwent CCTA with a 320-row multidetector CT (Aquilion ONE,
Toshiba Medical Systems,
Otawara,
Japan); 100-120kV,
400-550mA based on body mass index (BMI)
– Standard breath hold method; from carina to diaphragm
– Free-breathing method; from 1–2 cm above the level of carina
to diaphragm (as a compensation for possible range extension
by respiration)
– Nonionic contrast material (Xenetix®; iobitridol 350 mgI/ml;
Guerbet,
Cedex,
France) was injected using a biphasic injection
protocol [injection rate of 4 ml/s,
median volume,
70 ml (range,
65–100 ml)]
- The breath-holding CCTAs were acquired during an inspiratory
breath-hold,
whereas the free-breathing CCTAs were
performed without interruption of breathing by muting the
automatic play of breath-holding announcement.
– Prospectively-triggered data acquisition (70–80% of the R-R
interval);heart rate was sufficiently slow (≤ 65 bpm)
– Retrospective data acquisition with ECG-based tube current
modulation;
fast heart rate (> 65 bpm) even after administering beta-
blocking agents,
patients who required left ventricular function
measurements
- An initial interpretation was performed with reconstructed images from the 75% cardiac phase,
additional phases were reconstructed as needed.
– Systolic phase images (at 40% cardiac phase) or automatically
suggested ‘best phase (msec)’ were reconstructed for fast
heart patients.
– Image matrix; 512X512,
slice thickness; 0.5mm,
interval;
0.25mm)
- All data sets were transferred to Vitrea Workstation (Vital images,
Minnesota,
USA) and image analysis.
Image Analyses
– coronary artery was subdivided into 16 segments [American
Heart Association (8)]
– Whole coronary artery with a luminal diameter larger than
1.5 mm was analyzed.
– The image quality score was defined as:
1 (excellent); no motion artifacts,
excellent visualization of
vessel lumen
2 (good); some artifacts,
but still sufficient for interpretation
3 (poor); severe motion artifacts,
inadequate for
interpretation
- Signal-to-noise ratio (SNR)
– We measured attenuation value in Hounsfield unit (HU) and its
standard deviation (SD) of the aortic root on each axial image
SNR= vessel density (HU aorta)/ image noise (SD aorta)
- Contrast-to-noise ratio (CNR):
– Vessel contrast: difference of mean HU between contrast-
enhanced arterial lumen and adjacent
perivascular tissue.
– Image noise: SD of the attenuation value in a region of
interest (ROI) at aortic root.
CNR= vessel contrast (HU RCA (or LM) – HU perivascular tissue)
/ image noise (SD aorta).
– Scan length,
volume CT dose index (CTDIvol),
and dose length
product (DLP)
– Effective dose; calculated using a conversion coefficient for
chest and coronary arteries (k = 0.014mSv/ [mGy·cm]) (9)
– The effective doses were compared between the two
scanning modes in this study.