Abstract 16452: Myocardial Fibrosis Quantified by Cardiovascular Magnetic Resonance is Associated with Diabetes and Predicts Mortality

Abstract

Objectives: We hypothesized that: 1) diabetes would be associated with quantitative, validated, and reproducible cardiovascular magnetic resonance measures of myocardial fibrosis, i.e., extracellular volume fraction (ECV); 2) ECV in diabetic individuals would predict subsequent mortality; and 3) ACE inhibitor (ACEI) use would be associated with lower ECV.

Methods: We prospectively enrolled 877 consecutive patients referred for clinical CMR without amyloidosis or hypertrophic cardiomyopathy. We computed ECV from hematocrit and T1 of blood and myocardium pre and post contrast. Unlike isolated post contrast T1, ECV is not confounded by comorbidities prevalent in diabetes (kidney disease, obesity). ECV measures excluded areas of myocardial infarction (MI) evident by late gadolinium enhancement (LGE), but did include nonischemic scar. We did not want spatial variation of myocardial fibrosis (rendering it detectable by LGE) to confound its quantification. Unlike ECV, LGE has not been validated for collagen volume fraction quantification. Linear regression models predicted ECV; Cox regression models predicted mortality.

Results: Diabetic individuals had higher median ECV than those without diabetes (30.2% (IQR 26.9-32.7) versus 28.0% (IQR 25.8-30.9), respectively, p<0.001). Diabetes remained significantly associated with ECV in multivariable linear regression models adjusting for demographics, comorbidity, medications, ejection fraction (EF) and left ventricular mass index (p<0.001). Over a median of 1.0 years (IQR 0.5-1.5), there were 18 deaths among diabetic individuals. In the subset of 183 individuals with diabetes, myocardial fibrosis was associated with mortality (HR 1.42, 95% CI1.08-1.85 per 3% increase) in a model that adjusted for EF and stratified by MI size; ECV was also linked with mortality in stepwise selection models. In multivariable linear regression models adjusting for baseline differences, ACEI was associated with lower ECV.

Conclusions: These data suggest that myocardial fibrosis is associated with diabetes, predicts mortality in diabetic individuals, and may be ameliorated with ACEI use. Myocardial fibrosis may be an important intermediate phenotype in diabetic individuals that is treatable.