Gated SPECT offers improved interobserver agreement compared with echocardiography.

Area of Science:

• Cardiology
• Nuclear Medicine
• Medical Imaging

Background:

• Left ventricular ejection fraction (EF) is crucial for prognosis in coronary artery disease.
• Accurate measurement of EF and ventricular volumes is essential for patient management.
• Interobserver variability in imaging can impact clinical decision-making.

Purpose of the Study:

• To compare interobserver differences between gated SPECT (GSPECT) and echocardiography for left ventricular volumes and EF.
• To evaluate the reliability of different GSPECT software packages (QGS, ECT, 4D-MSPECT).
• To assess modality performance, particularly in patients with small ventricles.

Main Methods:

• Eighty-four patients underwent 1-day GSPECT with Tc-99m-tetrofosmin.
• Two raters analyzed GSPECT images using QGS, ECT, and 4D-MSPECT software.
• Two raters measured echocardiographic volumes.
• Interobserver reliability was assessed using intraclass correlation coefficients (ICC).

Main Results:

• GSPECT demonstrated significantly higher interobserver reliability (ICC 0.88-0.94) compared to echocardiography (ICC 0.61).
• Quantitative Gated SPECT (QGS) showed the highest ICC (0.94), particularly for small ventricles (ICC 0.90 vs. 0.58 for echo).
• GSPECT measurements for end-diastolic and end-systolic volumes were significantly larger than echocardiography, leading to different EF values.

Conclusions:

• GSPECT offers superior interobserver reliability for assessing left ventricular volumes and EF compared to echocardiography.
• Quantitative Gated SPECT (QGS) software appears particularly robust, especially in cases with small ventricles.
• These findings suggest GSPECT may provide more consistent measurements for cardiac function assessment.