Attenuation correction in myocardial perfusion imaging affects the assessment of infarct size in women with previous inferior infarct

Background Myocardial perfusion imaging is a well-established diagnostic tool in patients with known or suspected coronary artery disease. Numerous clinical trials have shown that attenuation correction (AC) in single photon emission computed tomography (SPECT) improves the diagnostic accuracy of myocardial perfusion imaging over non-AC SPECT, differentiating between scar and attenuation artifacts. We have previously shown that attenuation artifacts produce an overestimation of the size of inferior infarcts in the male population. It is assumed that women are less affected by inferior attenuation artifacts than men. Purpose The aim of this study is to evaluate the role of AC in the assessment of infarct size in female patients with a history of myocardial inferior infarct. Patients and methods We studied a population of 66 consecutive women, with a history of previous inferior myocardial infarct, by SPECT/computed tomography (CT) with 370+370 MBq of technetium-99m labeled compounds by a 2-day stress-rest protocol. Both AC and uncorrected gated-SPECT/CT studies were reconstructed after scatter and motion correction by ordered-subset expectation maximization iterative reconstruction and resolution recovery. The coregistration of the transmission and emission scans was verified for all patients; any misalignment was realigned manually. Uncorrected and corrected SPECT images were analyzed by software QPS/QGS package using a 17-segment model. For each segment, perfusion and wall motion were quantified using a five-point score according to the American Society of Nuclear Cardiology guidelines. Summed stress, summed rest score (SRS), and summed difference score of the inferior left ventricle wall (inferior, inferoseptal, inferolateral, and apical inferior segments) were calculated. A linear correlation was used to assess the relationship between perfusion and the regional wall motion score as determined by uncorrected gated-SPECT. Results The results of quantitative analysis of non-AC and CT-AC SPECT images, respectively, were as follows: summed stress score: 9.47±5.01 and 6.58±4.77% (P<0.001); SRS was 6.05±5.02 and 4.14±4.12% (P<0.001); the summed difference score was 2.92±2.74 and 2.52±2.63% (P=NS), respectively. The correlation between corrected and uncorrected SRS and the regional summed wall motion score of the same segment was R 2 =0.31 versus R 2 =0.34. Conclusion In the female population, like in men, attenuation artifacts affect the calculation of the infarct size of the inferior wall, with overestimation of the infarct size in uncorrected images. The AC regional perfusion score (SRS) better correlates with the regional wall motion score of the inferior wall in women with previous inferior infarct.