Accuracy of PET versus SPECT for myocardial perfusion imaging (MPI)

Objectives Although conventional SPECT has become the accepted technique for MPI, cardiac PET has begun to gain widespread utilization in the evaluation of coronary artery disease (CAD). The aim of this study is to evaluate the accuracy of PET vs SPECT for MPI when compared to cardiac catheterization (cath) results.

Methods The study consisted of 798 patients referred for MPI; 398 PET and 400 SPECT. PET MPI utilized a dipyridamole rest-stress imaging protocol with attenuation correction and Rb-82 using a dedicated PET system. SPECT MPI was conducted using a dual isotope rest-pharmacological stress imaging protocol. Studies were evaluated as normal, abnormal, or equivocal and location of defect was noted. Abnormal cath results were defined as ≥50% lesion of the left main and/or ≥70% lesion in the left anterior descending, left circumflex, or right coronary artery. Cath and previous image results were compared for matching lesions and defects, assessing accuracy of the respective imaging procedure. A total of 174 patients (79 PET, 95 SPECT) were referred for cath. Of them, 106 had CAD (51 PET, 55 SPECT).

Results Image and cath results were correlated. Table 1 displays image, cath correlation results, sensitivity and specificity of PET and SPECT MPI.

Conclusions Both PET and SPECT MPI are sensitive tests for diagnosing and evaluating the extent of CAD. PET is superior to SPECT because it offers increased image quality and speed, with less radiation exposure. The low specificity noted is likely due to referral bias (normal studies not referred for cath). Results may be limited due to other factors not considered. Further examination is warranted

• Download figure
• Open in new tab
• Download powerpoint

Table 1. Correlation Data