Myocardial blood flow quantification by widely available conventional nai SPECT improves detection of hemodynamically significant coronary stenosis

Background Relative myocardial perfusion imaging (MPI) with single-photon emission computed tomography (SPECT) is widely used for evaluating coronary artery disease but may underestimate ischemia in multi-vessel disease or microvascular dysfunction. Absolute myocardial blood flow (MBF) quantification enhances diagnostic assessment, yet its clinical use is limited by restricted availability of positron emission tomography and cadmium-zinc-telluride systems. Whether widely available conventional sodium iodide (NaI) SPECT can provide clinically meaningful flow quantification against invasive physiological reference standards remains uncertain. Purpose To determine whether MBF and myocardial flow reserve (MFR) quantified by dynamic conventional NaI SPECT improve detection of hemodynamically significant coronary stenosis compared with relative MPI, using invasive coronary angiography (ICA) and fractional flow reserve (FFR) as reference standards. Methods In this prospective single-center study, 202 patients with suspected or known stable coronary artery disease underwent rest–adenosine stress dynamic SPECT using a conventional NaI camera. Stress MBF and MFR were quantified with a one-tissue compartment model. Relative MPI was assessed using a 17-segment visual scoring system. All patients subsequently underwent ICA with FFR for intermediate lesions. Hemodynamically significant stenosis was defined as diameter stenosis >90% or FFR ≤ 0.80. Diagnostic performance was evaluated per vessel using receiver-operating characteristic analysis and generalized estimating equations. Results Among 580 vessels, 175 (30.1%) showed hemodynamically significant stenosis. Diseased vessels had lower stress MBF and MFR than non-significant lesions (1.47±0.63 vs. 2.05±0.64 mL/min/g and 1.95±0.79 vs. 2.58±0.75; both P<0.001). Stress MBF (AUC 0.778; 95% CI 0.734–0.822) and MFR (AUC 0.774; 95% CI 0.729–0.820) outperformed relative MPI (AUC 0.637; 95% CI 0.598–0.676; both P< 0.001), with no difference between quantitative indices (P=0.718). Sensitivity was higher for stress MBF and MFR (70.9% and 69.7%) than MPI (38.3%; both P<0.001), with greater negative predictive values (86.5% and 86.2% vs. 77.0%), whereas MPI showed higher specificity (89.1% vs. 80.5% and 81.7%). In intermediate stenoses, quantitative parameters maintained superior discrimination (AUC 0.750 and 0.758 vs. 0.535; both P<0.001). Adding stress MBF or MFR to MPI significantly improved discrimination (AUC 0.786 and 0.783) and yielded significant reclassification gains. Conclusions MBF quantification by widely available conventional NaI SPECT significantly enhance detection of hemodynamically significant coronary stenosis compared with relative MPI, maintaining advantage in intermediate lesions. These findings support broader clinical implementation of dynamic flow quantification using conventional SPECT and warrant multi-center validation.ROC: Quantitative Flow vs. MPI