Abstract 3386: Matrix Metalloproteinase-9 Compared with Brain Natriuretic Peptide as a Biomarker of Ventricular Remodeling in a Large Heart Failure Cohort
Background: Plasma matrix metalloproteinase-9 (MMP-9) has been proposed as a biomarker for matrix and ventricular remodeling, but has not been studied in a large heart failure cohort. We hypothesized that peripheral MMP-9 levels would be independently associated with indices of ventricular remodeling, and we compared the strength of this association with that of BNP, an established marker of heart failure (HF).
Methods and Results: Patients (n=500) were recruited for the Penn Heart Failure Study, a prospective cohort study of patients with a broad spectrum of systolic and diastolic HF. Clinical data and peripheral bloods were collected at study entry. Plasma MMP-9 was quantified using ELISAs (R & D Systems), and plasma BNP was quantified using the ArchitectTM BNP assay (Abbott). Left ventricular end-diastolic dimension (LVEDD) was measured using 2D echocardiography by readers blinded to all other data. In univariate analyses, MMP-9 (p=0.05) and BNP (p<0.001) were each associated with LVEDD. Multivariate analyses showed the highest tertile of MMP-9 was independently associated with LVEDD after adjusting for age, gender, race, BSA, BMI, HF type, HF etiology, hypertension, diabetes, and medical therapies (p=0.03). Addition of BNP tertiles to the model attenuated this association, indicating a stronger relationship between BNP and LVEDD (Figure⇓).
Conclusion: These findings demonstrate that peripheral MMP-9 levels are independently associated with LVEDD in patients with a broad spectrum of HF. This mirrors the relationship between BNP and LVEDD, albeit with a smaller effect size, and suggests a role for MMP-9 as a marker or mediator of ventricular remodeling in human HF.