Echocardiographic and biomarker characteristics in diabetes, coronary artery disease or both: insights from HOMAGE trial

Background: Coronary artery disease (CAD) and diabetes mellitus (DM) can induce changes in myocardial structure and function, thereby increasing the risk of heart failure (HF). We aimed to identify the alterations in echocardiographic variables and circulating biomarkers associated with DM, CAD, or both and to assess the effect of spironolactone on them. Methods: The “Heart OMics in AGEing” (HOMAGE) trial evaluated the effect of spironolactone on circulating markers of fibrosis over 9 months of follow-up in people at risk for HF. From the initial population (N = 527) of the HOMAGE trial, a total of 495 participants (mean age 74 years, 25% women) were categorized according to clinical phenotype (DM-/CAD + vs. DM+/CAD- vs. DM+/CAD+), while the DM-/CAD- group was excluded due to the low sample size (N = 32). Multivariable linear regression analysis was used to assess the relations between variables and DM/CAD status. Results: At baseline, participants with DM, whether or not they had CAD, showed lower markers of type I collagen synthesis (procollagen type I C-terminal propeptide; β [95% CI]: DM+/CAD-: -6.973 [-13.778; -0.167]; DM+/CAD+: -9.039 [-15.174; -2.903]), reduced left ventricular volumes (β [95% CI]: end-diastolic, DM+/CAD-: -6.323 [-9.696; -2.951]; DM+/CAD+: -2.503 [-5.531; 0.526]; end-systolic, DM+/CAD-: -2.905 [-4.817; -0.992]; DM+/CAD+: -1.400 [-3.120; 0.320]) and higher levels of galectin-3 (Exponential β [95% CI]: DM+/CAD-: 1.127 [1.050; 1.209]; DM+/CAD+: 1.118 [1.048; 1.192]), and growth differentiation factor-15 (Exponential β [95% CI]: DM+/CAD-: 1.542 [1.360; 1.747]; DM+/CAD+: 1.535 [1.370; 1.720]), along with an elevated E/e’ ratio (β [95% CI]: DM+/CAD-: 1.355 [0.462; 2.248]; DM+/CAD+: 0.879 [0.067; 1.690]), compared with DM-/CAD + individuals (all p < 0.05). At follow-up, the effect of spironolactone on echocardiographic variables and circulating biomarkers was not significantly different across DM/CAD phenotypes (all p-interaction > 0.05), except for a more pronounced reduction in GDF-15 in the DM+/CAD + group at the 1-month visit (p-interaction = 0.03). Conclusions: Among HOMAGE trial participants, diabetes was a powerful driver of biomarker and echocardiographic alterations irrespectively of CAD. These alterations were mainly related to the domains of inflammation and diastolic function. Graphic abstract: Summary of the study design and key findings. Abbreviations: CAD, coronary artery disease; DM, diabetes mellitus; LV, left ventricular; M0, baseline; M1, 1-month follow-up; M9, 9-month follow-up; PICP, procollagen type I C-terminal propeptide. (Figure presented.)