Defining the prognostic value of [15O]H2O positron emission tomography derived myocardial ischemic burden

Background A myocardial ischemic burden (IB) of 10% is used to denote high-risk patients with coronary artery disease (CAD). This threshold has primarily been assessed by single-photon emission computed tomography. Differences in the pharmacokinetics of the utilized traces, higher resolution of positron emission tomography (PET), and lastly differences in assessment of IB might lead to a higher prognostic threshold for quantitative PET. Purpose To determine a [15O]H2O PET derived IB to identify low (annualized event rate (AER) <1.0%) and high (AER ≥3%) risk CAD patients. Methods 623 patients who underwent [15O]H2O PET because of suspected CAD and in whom follow-up was obtained were included. The IB was defined as the percentage of myocardium with a hyperemic blood flow (hMBF) ≤2.3 ml/min/g and by a coronary flow reserve (CFR) of ≤2.5. The endpoint was a composite of death and non-fatal myocardial infarction (MI). Time-dependent ROC curves were constructed for the prediction of the endpoint within the first 5-years, based on these curves thresholds were selected for which specificity was maximized and sensitivity was at least 80%. Patients were classified as having a high IB if their respective IBs were above the prognostic thresholds and as having a low IB if not. Results During a median follow-up of 6.7 years, 34 (6%) patients died and 28 (4%) experienced a MI resulting in 62 (10%) endpoints. An IB of 24% and 28% for hMBF and CFR, respectively, was found to be the optimal threshold to define prognosis. Patients with a high hMBF or high CFR IB had worse outcome compared to patients with a low IB (log-rank p<0.001 for both), with AERs of 0.6% vs. 2.8%, and 0.6% vs. 2.4% (p<0.001 for both), respectively. Patients with a concordantly high hMBF and CFR derived IB had the worst outcome (AER: 3.1%), whereas patients with a concordantly low or discordant IB result had a similarly low event rate of 0.5% and 0.9% (p=0.953), respectively (log-rank p=0.445). A concordantly high hMBF and CFR IB was an independent predictor of adverse outcome beyond clinical characteristics (adjusted hazard ratio: 3.52, p<0.001). Conclusion An IB of 24% and 28% for hMBF and CFR was found to be the optimal prognostic threshold. Both measurements can be used to determine patients outcome. However combining hMBF and CFR IB results leads to a further refinement of risk-stratification allowing for the identification of low (concordant low or discordant IB result) and high (concordant high IB result) risk CAD patients. Funding Acknowledgement Type of funding source: None