Purpose: Computations of left and right ventricular (LV and RV) gated blood pool SPECT (GBPS) ejection fraction (EF) have been well validated against other imaging modalities. As GBPS images depict the entire extent of both blood pools, it is possible to compute not only global but also regional biventricular function parameters, which have the prospect of being clinically useful for planning cardiac resynchronization therapy. This investigation sought to establish LV and RV count-based GBPS regional functional normal limits and to quantify their reproducibility.
Methods and materials: Count-versus-time curves were fit to third-order Fourier series for each of 17 LV and RV sub-volumes to compute global and regional EF, timing, phase and dyssynchrony parameters. Algorithms were applied to data for 40 normal controls (NLs) and 15 patients with CHF. To assess reproducibility, data were reprocessed a second time, blinded to initial calculations.
Results: There were no statistically significant differences between any initial and reprocessed LV or RV parameters for NLs or patients with CHF. Percent of subjects categorized as abnormal were the same for initial and reprocessed parameters (McNemar's differences = 0-7%, P > 0.05 for each parameter). Most parameters were significantly different for patients with CHF versus NLs. Normal limits for the new technique agreed well with the literature for other imaging methods, and RV normal limits closely paralleled LV limits.
Conclusion: GBPS global and regional LV and RV normal limits are reproducible, and application of these normal limits to patients with CHF results in reproducible detection of functional abnormalities.