Initial characterization of a dedicated cardiac camera using normal databases: Comparison with conventional SPECT

Objectives Myocardial perfusion SPECT is a widely used diagnostic method. Recently, faster and radiation dose-reduced examinations are possible with dedicated cameras. We compared normal databases from non-attenuation corrected conventional SPECT system with data from such a device.

Methods From two clinically similar patient cohorts acquired with a fast system (n=375, 1VD:30%,2VD:28%,3VD:21%), D-SPECT,DataSpectrum, stress:3MBq/kg, ACQ 10min, rest: 9MBq/kg 5min) and a conventional system (n=348,1VD:25%,2VD:26%,3VD:36%, Siemens ECAM, (4MBq/kg,20min,rest:12MBq/kg,20min), 20 ♀ and 20 ♂ with normal angiograms were selected. Polar maps were calculated and from those, mean (M) and standard deviation maps (SD) were generated. These M+SD maps were analyzed in 17 segments, the mean of the M+SD maps, average and absolute differences, and Pearson's r² were calculated in: a) between rest and stress and between the genders, and b) the differences between the gender-specific databases with data from the conventional camera.

Results There was no significant difference between rest and stress for both genders (r²=0.9 ,ΔM=0±1%). Thus, in the following rest and stress will be combined. Between the genders only moderate differences were found in the mean uptake (r²=0.8, ΔM=3±2% ) - except for the area of the basal inferolateral wall (ΔM=3±4%). In these areas, the SD maps also differed more (♂ 6% vs. ♀ 9%). In comparison to the normal maps from the conventional camera a comparable variability was found in the SD maps (♂ 6±1% vs. 6±1%, ♀: 7±1% vs. 8±2%), however, the difference between the sexes is more pronounced using conventional SPECT (M r²: 0.8 vs. 0.5, SD 0.4 vs. 0.1).

Conclusions The normal databases of dedicated system are more homogeneous and show smaller differences between males and females than conventional SPECT. Further investigations will show whether these results can be used to produce a higher tolerance against attenuation artifacts to improve the diagnostic accuracy.