Gender dependence of normal myocardial perfusion distributions using the XACT imaging system

Objectives A new solid-state SPECT imaging system employing a respiratory motion matched, ultra-low dose CT-based attenuation correction (AC) has shown improved radiotracer uniformity in the normal myocardium compared to uncorrected (NC) reconstructions. The purpose of this study was to investigate the differences in the radiotracer distribution in normal myocardium as a function of gender with and without AC.

Methods We evaluated 70 randomly selected patients (35M, 35F) with a pre-test likelihood of <=5% for coronary artery disease who were referred for myocardial perfusion SPECT. Patients were imaged on the Digirad XACT system using a standard one-day rest-stress ^99mTc protocol. Images were motion corrected as needed, and AC images were reconstructed with the nSPEED algorithm using measured attenuation maps from the XACT system; NC images were reconstructed without the inclusion of the attenuation map. All images were reformatted into short axis volumes and input to the Corridor4DM software to provide normal databases for each gender and correction. The normal distributions from each database were compared using a 9 segment model and t-test statistic.

Results NC gender databases were significantly different in 4 of 9 segments (basal and distal anterior, p<0.05, basal and distal inferior, p<0.001). AC gender database were significantly different only in the inferior-basal segment (p=0.03). Comparisons of NC and AC databases showed the most significant post-AC improvements in the basal anterior and lateral segments for females (correction of breast attenuation) and in the basal and distal inferior segments for males (correction of diaphragmatic attenuation).

Conclusions XACT AC distributions in males and females approach equivalence in this pilot population. Further evaluation with a larger population will more clearly define the significance of these findings with the expectation that gender independence with XACT AC is achievable