Assessment of left ventricular ejection fraction by four different methods using 99mTc tetrofosmin gated SPECT in patients with small hearts: correlation with gated blood pool

AIM: To compare the currently available gated SPECT software programs, quantitative gated SPECT (QGS), Emory Cardiac Toolbox (ECTb), Left Ventricular Global Thickening Fraction (LVGTF), and the recently developed Layer of Maximum Count (LMC) method with equilibrium Gated Blood Pool (GBP) scintigraphy in calculating the ejection fraction in patients with small hearts. METHODS: Twenty patients with small hearts (end diastolic volume <85 ml) were collected for the study. Gated myocardial perfusion SPECT and planar GBP were performed for all patients. The four methods QGS, ECTb, and LVGTF and LMC were used for volumes estimation and ejection fraction calculation. RESULTS: ANOVA analysis revealed significant differences among the methods in ejection fraction estimation (P<0.0001). The mean ejection fraction by GBP was significantly overestimated by QGS and ECTb and LVGTF (P<0.0001, P<0.0001 and P=0.006, respectively). The mean ejection fraction by GBP was not significantly different from that by the LMC method (P=0.213). Ejection fraction measurements by QGS and ECTb yielded moderate correlation with GBP values (r=0.588, P=0.006; and r=0.564, P=0.010, respectively). The ejection fraction by the LMC method was marginally correlated but LVGTF showed a non-significant correlation with GBP (r=0.438, P=0.053; and r=0.155, P=0.515, respectively). Agreement analysis for ejection fraction estimation by QGS and ECTb demonstrated a non-significant correlation between the difference and the mean. The LMC method showed a non-significant trend to decrease the difference with GBP as the mean increased. However, the LVGTF method significantly increased the difference as the mean increased. CONCLUSION: The currently available gated SPECT methods have moderate to poor correlations in addition to wide agreement limits with gated blood pool studies in patients with small hearts. Improvement of these methods to achieve better results in such patients is recommended. The newly developed LMC method yielded better results in the group with small hearts but with low interchangeability with GBP studies.