Prognostic value of quantitative anatomical and functional measures of coronary artery disease in diabetic and non-diabetic patients

Abstract Background Diabetes is associated with increased risk of coronary artery disease (CAD). However, the relative prognostic importance of anatomical and functional findings of CAD in the presence vs. absence of diabetes is incompletely understood. Technological advances have made quantitative analysis of coronary artery plaques and stenosis feasible by coronary computed tomography angiography (CTA). In turn, positron emission tomography (PET) enables absolute quantification of myocardial blood flow (MBF). Purpose To study long-term prognostic value of quantified severity of coronary atherosclerosis and myocardial perfusion in non-diabetic vs. diabetic patients with suspected CAD. Methods From two academic medical centres, we identified symptomatic patients with suspected CAD who had undergone coronary CTA and [15O]H2O PET myocardial perfusion imaging during adenosine stress. Based on artificial intelligence -guided quantitative analysis of coronary CTA scans, the anatomical severity of CAD was measured as the presence of obstructive CAD (≥50% diameter stenosis) and percent atheroma volume (PAV; higher vs. lower than median). The functional severity of CAD was measured as the presence of regional myocardial ischemia (≥2 adjacent segments with stress MBF <2.3 ml/g/min) and global stress MBF (<2.2 ml/g/min vs. ≥ 2.2 ml/g/min) by PET. Annual rates of adverse events (AER) including all-cause mortality, myocardial infarction (MI), and unstable angina pectoris (UAP) were evaluated. Results Among 1311 patients, 251 (19%) had diabetes. Patients with diabetes had more frequently obstructive CAD (55% vs. 43%, p<0.001) and regional myocardial ischemia (51% vs. 43%, p=0.015), higher plaque burden by PAV (12% vs. 7%, p<0.001), and lower global stress MBF (2.78 ml/g/min vs. 3.05 ml/g/min, p<0.001). During median follow-up of 7.1 years, 171 (13.0%) patients experienced AE (85 deaths, 56 MIs, and 30 UAPs). Patients with diabetes had higher AER than non-diabetic patients (3.1% vs. 1.6%, p<0.001). The presence of obstructive stenosis was associated with increased AER in non-diabetic (2.6% vs. 1.0%, p<0.001) and diabetic patients (4.2% vs. 1.9%, p<0.001). High plaque burden by PAV was associated with increased AER in non-diabetic (2.8% vs. 0.7%, p<0.001) and diabetic patients (3.9% vs. 1.7%, p<0.001). The presence of regional myocardial ischemia was associated with increased AER in non-diabetic (2.5% vs. 1.1%, p<0.001), but not in diabetic patients (3.4% vs. 2.7%, p=0.238). Low global stress MBF was associated with increased AER in non-diabetic (2.9% vs. 1.3%, p<0.001), but not among diabetic patients (3.2% vs. 3.0%, p=0.286). Conclusions Diabetes was associated with more advanced CAD. Quantified severity of anatomical findings predicted long-term adverse outcome in non-diabetic and diabetic patients. In contrast, functional imaging risk stratified non-diabetic patients, whereas diabetic patients had impaired long-term outcome irrespective of perfusion findings.