Abstract
OBJECTIVE: Several studies have confirmed high diagnostic performance of positron emission tomography (PET) myocardial perfusion imaging (MPI) in patients with known or suspected coronary artery disease. However, whether the superior diagnostic accuracy could translate into improved mortality outcomes remains unknown. The aim of this study was to define the prognostic value of normal PET MPI. METHODS: PubMed and EMBASE were searched to identify related studies up to June 2016. All studies using PET MPI to evaluate subjects with known or suspected coronary artery disease and providing absolute number of patients with a negative test and primary data on clinical outcomes with a follow-up time ≥3 months were included for analysis. RESULTS: The search yielded 11 studies comprising 20,471 patients for final analysis. The negative-predictive value (NPV) for cardiac death, all-cause death and major adverse cardiovascular events (MACE) were 98.80% [95% confidence interval (CI), 97.64%-99.39%], 94.89% (95% CI: 92.99-96.30%) and 90.26% (95% CI: 78.01-96.03%), over 36.9 months of follow-up for cardiac death, over 26.8 months for all-cause death and 35.7 months for MACE. The corresponding annualized event rates were 0.39%, 2.29% and 3.27%, respectively. In subgroup analyses of different imaging analysis methods for PET MPI, studies using perfusion abnormity had a similar NPV as compared with those using coronary flow reserve (98.46% vs 98.86%, p-value = NS), with a corresponding annualized event rate after negative tests (equal to 1 - NPV) as 0.45% and 0.42%, respectively. CONCLUSION: Normal PET has a high NPV for cardiac death, MACE and all-cause mortality. Different indexes used for PET imaging analysis have a comparable prognostic value. Advances in knowledge: A normal PET MPI conferred a very low risk of cardiac death of 0.39% per year, which is close to that of a normal aged-matched population.