Evaluation of absolute myocardial blood flow and flow reserve using [¹³N]ammonia-PET/CT in comparison with 99mTc-tetrofosmin-CZT SPECT

Introduction: Evaluation of absolute myocardial blood flow and flow reserve using [¹³N]ammonia-PET/CT in comparison with 99mTc-tetrofosmin-CZT SPECT Background and Aim: The gold standard for quantitative assessments of myocardial blood flow (MBF) is the PET/CT, but the issues due to the availability of short-lived radio tracers, and the expensive costs are still a problem. The introduction of SPECT based on Cadmium Zinc Telluride (CZT) detectors allows for quantitative assessments of MBF by dynamic acquisitions. The aims of this study were to compare MBF measurements of 99mTc-tetrofosmin-CZT with [¹³N]ammonia-PET/CT and to evaluate the impact of attenuation correction (AC) in SPECT imaging. Materials and Methods: 54 patients were retrospectively evaluated. The patients underwent both CZT-based-SPECT and [¹³N]ammonia-PET/CT; the PET scan was performed only for the patients with CZT-based-SPECT examination inconclusive. As myocardial stress agent we used regadenoson for both SPECT and PET/CT scans. The single-day CZT-based-SPECT fixed activities of 99mTc-tetrofosmin were 185 MBq for the stress and 370 MBq for the rest acquisition. The delay between stress and rest scan was 2 hours. For stress imaging, an initial dose of 37 MBq was used to localize the patient’s heart. Residual resting activity was subtracted from stress data and images reconstructed with CT-based attenuation correction (AC-CZT), obtained from the [¹³N]ammonia-PET/CT, and without AC (NC-CZT). MBF and CFR were assessed by Corridor4DM Reserve software for both AC-CZT and NC-CZT. The single day [¹³N]ammonia-PET/CT activities were 370 MBq for both stress and rest. MBF and CFR were assessed by Corridor4DM. Stress MBF ≥ 1.7 and MFR ≥ 2 were considered as normal.

Results: The [¹³N]ammonia-PET/CT MBFs (mean ± standard deviation) were 1.87±0.45ml/mg/min for stress and 0.76±019ml/mg/min for rest; the CFR was 2.52±0.56ml/mg/min. The AC-CZT MBFs were 1.62±0.68 (P <0.01) and 0.77±0.24ml/mg/min (P ns) for stress and rest respectively; the CFR was 2.75±0.86ml/mg/min (P <0.01) . The NC-CZT MBFs were 2.36±1.09ml/mg/min (P <0.01) for stress and 1.14±0.40ml/mg/min (P <0.01) for rest; the CFR was 2.18±0.96ml/mg/min (P <0.01). The sensitivity and the specificity of the AC-CZT to predict MFR < 2 were 84% and 66% and stress MBF < 1.7 were 86.4% and 63%, respectvely. The sensitivity and the specificity of the NAC-CZT to predict MFR < 2 were 72.9% and 67% and stress MBF < 1.7 were 81% and 66%.

Conclusions: Reliable measures of MBF and MFR can be obtained by CZT with dynamic acquisition. A correct methodology is mandatory, applying crosstalk subtraction of residual activity, motion correction during dynamic acquisition and attenuation correction are required to obtain physiological MBF values.