Long-term prognostic impact of fasting plasma glucose and myocardial flow reserve beyond other risk factors and heart disease phenotypes

Abstract

Abstract Cardiometabolic risk factors, including high fasting plasma glucose (hFPG), are emerging prognostic determinants in patients with coronary artery disease (CAD) or heart failure (HF). Coronary microvascular dysfunction might be a comprehensive risk predictor in these patients. The purpose of this study was to assess whether hFPG and global myocardial blood flow (MBF) reserve measured by positron emission tomography (PET), expressing global coronary function, predict long-term prognosis beyond other risk factors and presence of obstructive CAD or left ventricular (LV) dysfunction associated with HF. We retrospectively collected long-term follow-up data in 103 patients (mean age 61 ± 10 years, 74 males) with stable chest pain or dyspnoea who underwent cardiac PET/computerized tomography and coronary angiography. Disease phenotypes included obstructive CAD (35%), LV dysfunction without obstructive CAD (43%), or none (22%). At multivariable logistic regression analysis, MBF reserve lower than the median value (OR 1.8, 95% CI 1.5–2.2) was significantly associated with male gender (OR 3.45, 95% CI 1.21–9.83) and hFPG (OR 3.87, 95% CI 1.17–12.84) among all risk factors. In a median follow-up of 10.9 years (interquartile range 7.8–13.9), 39 patients (37.8%) died (13.6% cardiac death). At multivariable Cox analyses including all risk factors and disease phenotypes, age (HR 1.07, 95% CI 1.02–1.12), hFPG (HR 2.18, 95% CI 1.02–4.63), and depressed MBF reserve (HR 4.47, 95% CI 1.96–10.18) were independent predictors of death (global χ2 37.41, P = 0.0004). These results suggest a strong long-term prognostic role of hFPG and depressed MBF reserve in a high-risk population of patients with a high prevalence of obstructive CAD or HF.