The study population comprised 9 normal volunteers and 34 patients. GBPS data were acquired by use of 360° rotation and 60 stops per head. All 60 (360°) and 30 (45° right anterior oblique to 45° left posterior oblique) pieces of projection data that were selected for reconstructing the 180° data were reconstructed and both ventricular functional parameters were automatically obtained by QBS software. The contour of the LV septal wall was concave in 6 patients (14 % ) when processed at 180°, whereas a concave septum at 360° processing was observed in only 1 patient (2 % ). The coefficients of correlation between 180° and 360° were 0.467 for the end-diastolic volume (EDV) and 0.648 for the end-systolic volume (ESV). The mean 180° EDV value (152.9 ± 46.1 mL) was significantly smaller than that of the 360° EDV (191 ± 70.8 mL) (P < .001). However, there was no significant difference between the 360° EDV (0.623) and 180° EDV (0.407) as compared by echocardiography (P = .218). The agreement of the EF between both methods was close (r = 0.894, P < .0001). The agreement of the right ventricular volumes between the 180° and 360° orbits was close (r = 0.800 for EDV and 0.706 for ESV). The EF was relatively dispersed between the 180° and 360° methods (r = 0.642).
This study showed that SPECT image acquisition by use of both the 180° method and the 360° method considerably underestimated LV volume quantification. In addition, the LV volume with the 180° method was significantly smaller than that with the 360° method. Thus a 360° acquisition orbit may be suitable for more quantitatively accurate results when blood pool imaging is performed with gated SPECT.