Between April 2015 and January 2020, 172 consecutive patients with suspected or known stable CAD who underwent NH3-PET and CCTA within 3 months at a single center were enrolled. Patients with a history of coronary artery bypass graft surgery, congenital heart disease, or adenosine ineffectiveness were excluded. Finally, 133 patients were enrolled in this study and prospectively analyzed. The study was performed in compliance with the Declaration of Helsinki, and the protocol was approved by ethics committee of our hospital (authorization number: 5260). Informed consent was obtained from all patients.
NH3-PET image acquisition
Patients refrained from caffeine for ≥12 hours before NH3-PET study and did not take any antianginal medications on the morning of the test. Imaging was performed using a 3-dimensional PET system (Biograph, mCT, Siemens Healthcare). A sequential CT scan (120 kV, 20 mAs, and 3 mm slice collimation) was acquired for attenuation correction. Immediately after administering 185 MBq 13N-ammonia intravenously, electrocardiographic-gated acquisition was performed for 10 min with 16 frames per cardiac cycle using parallel list-mode acquisition. We visualized MBF images in the initial 2 min and MPI in the next 8 min. After the PET MPI at rest, an adenosine stress test was performed (0.12 mg/kg/min x 6 min). Approximately 555 MBq of 13N-ammonia was infused 3 min after the administration of the vasodilator, and the stress and rest MPIs were evaluated.
Calculation of MBF and MFR
Images were reconstructed using Fourier rebinning and filtered back-projection with a 12 mm 3-dimensional Hann window of the ramp filter. Automatic reorientation of the images, automatic extraction of the mean myocardial and cavity time-activity curves (TACs), and generation of polar maps of the absolute MBF and MFR were performed using a dedicated software (Syngo MI cardiology, Siemens Healthcare). MBF was estimated using the TAC of the left ventricle input and myocardial uptake using a 3-compartment model and dataset of list-mode images obtained in the first 2 min. MFR was determined as the ratio of hyperemic MBF to resting MBF. Global MFR <2.0 was considered abnormal [2,4].
Myocardial perfusion assessment using static imaging
The observers had access to polar maps and 16 slices each on the short axis, vertical long axis, and horizontal long axis. The images were interpreted using a 17-segment model and semi-quantitative scoring system (0=normal, 1=mildly abnormal, 2=moderately abnormal, 3=severely abnormal, and 4=complete defect) to detect the severity . Summed stress score (SSS), summed rest score (SRS), and summed difference score (SDS = SSS – SRS) were calculated. SSS ≥4 in each region of the three major territories was considered abnormal.
CCTA image acquisition
A 320-row CT scanner (Aquilion One Vison Genesis, Canon) with a section collimation of 320 × 0.5 mm and a gantry rotation time of 275 ms was used for CCTA imaging. Metoprolol (20 mg or 40 mg, oral) was administered for a target heart rate of <65 bpm 1 hour before CCTA was performed. If necessary, landiolol (0.125 mg/kg, intravenous) was additionally administered. Sublingual nitroglycerine was administered to all patients immediately before the scan. To determine the optimal scan timing for CCTA, a test-bolus injection of small contrast media was scanned in the ascending aorta at the slice level of the pulmonary trunk. CCTA with a tube current of 700 mA and 100 kV, and prospective electrocardiogram gating were performed at 75% of the R-R interval. A bolus of contrast media (Oypalomin 370, Fuji Pharma) was injected at a flow rate of 4-5 mL/s, followed by a saline flush.
Assessment of coronary artery stenosis and plaque
Coronary arteries with a diameter ≥2 mm were evaluated according to a 17-segment coronary artery model using axial, multi-planar reformation, maximum intensity projection, and cross-sectional images . CAD-RADS 0, 1, 2, 3, 4A, 4B, and 5 were defined as 0% (absence of CAD) stenosis; 1-24% stenosis; 25-49% stenosis; 50-69% stenosis; 70-99% stenosis in one or two vessels; 70-99% stenosis in three vessels or left main coronary artery stenosis ≥50%; and 100% stenosis (total coronary occlusion), respectively (Figure 1) [13-15]. The most severe stenosis was considered if a single vessel had multiple lesions. The V (vulnerable plaque), S (stent), or G (graft) category modifiers were not used for grouping.
Coronary lesions were also analyzed in terms of the four HRP characteristics as previously defined [18-20]. The stenosis severity and plaque characteristics on CCTA were analyzed using a dedicated workstation (Ziostation 2, Ziosoft Inc.). The CCTA assessment was determined by the consensus of two radiologists and two cardiologists with more than 10 years of experience in nuclear cardiology and cardiac CT.
Assessment of clinical outcomes
The endpoint was defined as the occurrence of MACE consisting of all-cause death, acute coronary syndrome, hospitalization due to heart failure, and cerebrovascular disease. The event data were retrospectively gathered from the patients’ records, including in- and out-of-hospital reports. Coronary events within 90 days were not included. Based on the results of CCTA and NH3-PET, it was left to the attending physician to decide whether coronary revascularization was required.
Patients were divided into two groups based on the thresholds of CAD-RADS 3 and MFR 2.0, and the prognosis was compared. Continuous variables are expressed as mean ± SD, while categorical variables are presented as absolute counts and percentages. Comparisons were made using the Mann–Whitney U test for continuous variables, the chi-square test or Fischer exact test for categorical variables, and the Wilcoxon rank-sum test for ordinal variables. The Cox proportional hazards regression model was used to calculate hazard ratios with 95% confidence intervals (CI). The ability of CAD-RADS and MFR to predict MACE was assessed using the C-statistic. Survival curves of the patient subgroups were created using the Kaplan-Meier method to clarify the time-dependent cumulative MACE-free rates and were compared using the log-rank test. A p-value <0.05 was considered to indicate a statistically significant difference. All analyses were performed using JMP Pro version 14 (SAS Institute Inc., Cary, NC).