Aim: The aim of this study was to analyse the relationship between left ventricular ejection fractions (EF) obtained using four different instruments.
Material and methods: Eighty-five EF values were analysed. Gated acquisition was performed with the same gammacamera (SOPHA) using 99mTc-labelled red cells. Each EF was obtained using four processing systems: NXT (SOPHA), Entegra (Gems), Odyssey (Philips) and Esoft (Siemens), always working in automated mode. The paired student's t-test, Spearman correlation and Bland-Altman analysis were used to compare methods, and Deming regression was applied.
Results: Mean values and standard deviations for each program were: NXT: 61 +/- 9; Entegra: 60 +/- 10; Odyssey: 60 +/- 9; Esoft: 60 +/- 10. Although no significant differences were found as a whole and the values were linearly related, the methods are not interchangeable.
Conclusions: The same program should be used in the follow-up of each patient, which is now easily achievable by means of the DICOM standard.