The diagnostic value of attenuation correction in gated myocardial perfusion SPECT

Purpose: To evaluate the value of gated with attenuation correction in improving the diagnostic sensitivity and accuracy of myocardial peffusion SPECT. Methods: Sixty-four subjects (Male 40, female 24, age from 43 to 78 years old, average 58 years old) underwent stress and rest gated single photon emission computed tomography myocardial perfusion imaging with attenuation correction using 99mTc-MIBI. After acquisition, the images were reconstructed into vertical long axis slices, horizontal long axis slices and short axis slices with and without attenuation correction respectively. The myocardium was divided into 8 segments. The images were viewed in 2 ways: gated with and without attenuation correction. Images were assessed by 2 experienced observers individually with the knowledge of the left ventricular wall movement and wall thickness, but without the results of coronary artery angiography. The results of the coronary angiography was as a standard. The sensitivity and diagnostic accuracy were calculated. Results: The coronary artery angiography showed that 76 segments supplied artery stenosis not less than 50% and 436 segments supplied artery stenosis less than 50% or normal in 64 patients. Eight slightly infarcts were ignored and 5 false positive were found when gated SPECT was used only, When gated with attenuation correction were used, only 3 ischemia area was not found and 8 false positive area were found in septal, anterior or lateral wall. But the wall movement could tell 5 false positive areas from artifacts. Overall the sensitivity/accuracy were 89.5%/91.6% and 96.1%/98.8 % respectively, for the gated without and with attenuation correction. Conclusions: The combination of gated with attenuation correction provides the highest diagnostic accuracy of myocardial perfusion SPECT than gated without attenuation correction. The attenuation correction was not always the reliable method for detecting the coronary artery disease without combing with the left ventricular wall movement. (authors)