Artificial intelligence-derived splenic response in cardiac positron emission tomography is associated with major adverse cardiovascular events: a multi-site study.
Ramirez Giselle G, Dharmavaram Naga L NL, Shanbhag Aakash A, Miller Robert J H RJH et al.
Inadequate pharmacologic stress may limit the diagnostic and prognostic accuracy of myocardial perfusion imaging (MPI). The splenic ratio (SR), a measure of stress adequacy, has emerged as a potential imaging biomarker. We developed an artificial intelligence (AI) method to derive SR and evaluated the prognostic value in a large multicentre 82Rb-PET cohort undergoing regadenoson stress testing. We retrospectively analysed 16 650 patients from six sites in the REFINE PET registry with clinically indicated MPI and linked clinical outcomes. SR was calculated using fully automated algorithms as the ratio of splenic uptake at stress vs. rest. Patients were stratified by SR into decile groups. The primary outcome was major adverse cardiovascular events (MACE). Survival analysis was conducted using Kaplan-Meier and Cox proportional hazards models adjusted for clinical and imaging covariates, including myocardial flow reserve (MFR ≥2 vs. <2) and presence of ischaemia. The cohort had a median age of 69 years, with 59% male patients. Common risk factors included hypertension (81%), dyslipidaemia (77%), diabetes (35%), and prior coronary artery disease (31%). Median follow-up was 3.4 years. Patients with high SR in the testing cohort (n = 950) had an increased risk of MACE [hazard ratio (HR) 1.19, 95% confidence interval (CI) 1.06-1.34, P = 0.005]. Among testing patients with preserved MFR (≥2; n = 8067), high SR remained independently associated with MACE (HR 1.43, 95% CI 1.20-1.70, P < 0.001). Even among patients with preserved MFR and no ischaemia in the testing cohort(MFR≥2, SDS<2; n = 6508), high SR remained independently associated with MACE (HR 1.54 95% CI 1.26-1.88, P < 0.001). Elevated AI-derived SR was independently associated with adverse cardiovascular outcomes, including among patients with preserved MFR. These findings support SR as a novel, automated imaging biomarker for risk stratification in 82Rb PET MPI.