Introduction Surveillance of valvular heart disease accounts for a significant proportion of outpatient attendances in cardiology. A biomarker that correlates with disease severity and helps guide the optimum timing for intervention would be of great interest. NT-proBNP has been extensively examined in cardiac disease but focus has turned to enzymes that control collagen turnover, specifically matrix metalloproteinases (MMPs) and their tissue inhibitors (TIMPs). There is little data, other than from tissue samples, in aortic stenosis. We sought to examine the relationship between LV geometry and remodelling and serum MMPs and TIMPs in isolated severe aortic stenosis.
Methods 46 patients with isolated severe AS (peak velocity >4 m/s or mean pressure gradient >40 mm Hg or aortic valve area <1 cm2) without obstructive coronary artery disease (on angiography) were recruited. Investigations were performed on the same day once listed for AVR and 6 months following surgery; venous blood samples were drawn, centrifuged within 30 min and frozen at −80°C until assayed. MMP 2, 3 and 9 and TIMPs 1, 2 and 4 and additionally NT-proBNP were measured using similarly designed non-competitive immunoluminometric assays. All samples both pre- and post-operative were thawed and analysed at the same time to ensure consistency of assays. LV geometry, function and remodelling were assessed with transthoracic echocardiography for diastolic function and cardiac magnetic resonance for cardiac mass/volumes and myocardial fibrosis on late gadolinium enhancement.
Results Three patients did not attend follow-up. The remaining 43-paired data-sets were used for analysis. Data that were not normally distributed (including all biomarker data) were log transformed before analysis. Change in biomarker values between pre- and post-op were examined using paired t-tests and results are shown in Abstract 075 table 1. The relationship between baseline biomarkers and functional parameters (both at baseline and the ratio of change; pre/post-operative) were examined using bivariate correlations and are shown in Abstract 075 table 2. There are significant falls in MMP 2 and 9 and TIMPs 1 and 2 despite no significant change in NT-proBNP by six months post AVR.
Discussion NT-proBNP is perhaps unsurprisingly associated with mainly left ventricular geometry. In contrast the MMPs and TIMPs seem to be more closely related to measures of diastolic function, namely septal E/e' (a measure of LVEDP) and left atrial volume along with right ventricular mass. The role of MMPs and TIMPs as biomarkers in aortic stenosis is of interest and warrants study in a larger patient cohort including asymptomatic patients under surveillance.
- Aortic stenosis
- matrix metalloproteinases