Abstract 6130: Prospective Multicenter Trial Comparing Novel High-speed Myocardial Perfusion SPECT to Conventional SPECT: Quantitative Results of Myocardial Perfusion and Left Ventricular Function
Background: High-speed SPECT (D-SPECT) applied to myocardial perfusion imaging (MPI) has been shown in a pilot study to detect similar perfusion abnormality compared with conventional Anger camera SPECT (A-SPECT) and normal limits for quantitative analysis of D-SPECT MPI have been validated. This study reports quantitative perfusion/function results of a large prospective multicenter trial comparing D-SPECT to A-SPECT.
Methods: All patients (pts) [224, 127 male] had Tc-99m sestamibi rest/stress SPECT with D-SPECT and A-SPECT at four centers: Cedars-Sinai Medical Center (CSMC) (n=57), Vanderbilt University Medical Center (VUMC) (n=54), Brigham and Women’s Hospital (BWH) (n=58) and Miami Baptist Hospital (MBH) (n=55). D-SPECT MPI was performed within 30 min of A-SPECT. Rest/stress acquisition times were 20 and 15 min, respectively for A-SPECT, and 4 and 2 min, respectively for D-SPECT. Normal limits were constructed from studies of 60 pts with low likelyhood of CAD. The total perfusion deficit (TPD) was derived for rest and stress images. All perfusion and function measurments were automatically generated for the stress and rest images of D-SPECT and A-SPECT.
Results: D-SPECT TPD in the entire cohort correlated linearly to A-SPECT TPD over a wide range of perfusion abnormality at stress and rest (r=0.96 and 0.97, respectively, p<0.0001). High linear correlation between D-SPECT and A-SPECT stress TPD was also seen at each participating center (r=0.97, 0.96, 0.97, 0.97 for CSMC, VUMC, BWH, and MBH, respectively, p<0.0001). Normalcy rates (normal defined as TPD<5%) in 61 pts with low prescan likelihood of coronary artery disease, separate from pts used for normal limit development were 95% and 90% for D-SPECT and A-SPECT, respectively, p=ns. Post-stress EF, end-diastolic and end-systolic volumes by D-SPECT correlated linearly with A-SPECT over a wide range of EF and volumes (r=0.88, 0.96, 0.98, respectively, p<0.0001), with slopes close to 1, and small offsets.
Conclusion: The results of the prospective multicenter clinical trial demonstrats that high-speed SPECT provides objective quantitative measures of myocar-dial perfusion and function comparable to conventional SPECT in a faction (1/5th-1/7th) of the conventional SPECT acquisition time.