Reproducibility of LV ejection fraction, volume and mass by gated SPECT perfusion imaging

Abstract

Background: Gated SPECT perfusion imaging provides quantitative data on left ventricular [LV] ejection fraction [EF], end-distolic volume [EDV], end-systolic volume [ESV] and mass [LVM]. A number of factors may affect the reproducibility of these measurements. This study tested the reproducibility in 238 patients. Material and methods: Each patient had separate days stress and rest, high-dose Tc-99m tetrofosmin SPECT, with the same camera system. All patients had normal stress and rest image. Results: There were 108 men and 130 women, aged 53±13 years. The results are shown in the table. Study EF% EDV[ml] ESV[ml] LVM[gm] Post-stress 62 ± 10 88 ± 30 35 ± 20 140 ± 25 Rest 61 ± 10 91 ± 31 37 ± 22 141 ± 28 There were excellent agreements between repeated measurements for EF, r=0.72, p=0.001, Bland-Altman: mean difference = 1.0±7.7 and for LVM, r= 0.78, p<0.001 and Bland-Altman: mean difference= 0.9±25. The correlations were similar in men and women except for the fact that the EF was higher in women than man [ 65±10 VS 56±8, p<0.0001] and the EDV smaller in women [76±24 VS 103±30, p< 0.0001]. Least variability was observed in ESV. Conclusion: In this large sample, we have shown that gated SPECT imaging provided reproducible information on LV EF, EDV, ESV and mass. These data are crucial for future studies to ascertain sample size and intervention effects.