BACKGROUND: Transcatheter aortic valve replacement (TAVR) has been established for patients with severe and symptomatic aortic valve stenosis (AS) at high and moderate risk. However, even after TAVR some patients remain at a high risk as a consequence of a maladaptive remodelling of the left ventricle. Recently, data from our group demonstrated, that a higher amount of myocardial fibrosis is associated with increased cardiovascular mortality and a delayed reverse remodelling in AS patients. Thus, pharmacological anti-fibrotic treatment of these patients might be beneficial. Here, we aimed to determine biomarkers for myocardial fibrosis that sufficiently predict maladaptive remodelling in patients from different AS-subtypes.

MATERIAL AND METHODS: Between 2017 and 2018, we prospectively enrolled around 200 patients scheduled for transfemoral transcatheter aortic valve implantation (TAVR). AS-patients were subdivided into four different subtypes determined by ejection fraction (EF) and aortic valve gradient according to the current guidelines. Myocardial fibrosis was assessed blinded to clinical and imaging data using quantitative morphometry in biopsies harvested from the basal anteroseptum after TAVR. 

The serum biomarkers PICP (C-terminal propeptide of procollagen type I), CITP (carboxy-terminal telopeptide of collagen I) and MMP-1 (matrix metalloproteinase-1) were assessed using the EIA MicroVue for PICP (Quidel Corporation), an ELISA based assay for CITP (Orion Diagnostica), and an alphaLISA for quantification of total serum MMP-1 levels (PerkinElmer). Further, the ratio of CITP and MMP-1 (CITP/MMP-1) was analysed constituting a biomarker of collagen cross-linking.

RESULTS: We found a negative correlation between CITP/MMP-1 serum levels and histological fibrosis in the heart exclusively in the group with reduced systolic function and low gradients. In contrast, the left ventricular end-diastolic diameter (LVEDD) correlated in a positive manner with CITP/MMP-1 serum levels of in the group with preserved or reduced systolic function and high gradients. Neither levels of fibrosis, nor LVEDD were correlated with CITP or MMP-1 serum levels in any of the groups. Further, PICP as a marker of collagen deposition was not correlated with fibrosis or LVEDD in any of the groups.

CONCLUSION: The positive correlation of serum CITP/MMP-1 with the LVEDD in AS patients with preserved systolic function (EF >50%) suggests that LV dilatation in this group occurs predominantly in hearts with lower levels of collagen cross-linking (i.e. when the serum CITP/MMP-1 ratio is high). Most importantly, we found a significant correlation between CITP/MMP-1 as a marker of collagen cross-linking and fibrosis biopsies of patients with reduced EF and low aortic valve gradient, which are known to have a high risk even after TAVR. Therefore, CITP/MMP-1 might serve as biomarker for maladaptive fibrotic remodelling in TAVR patients with low EF that might benefit from an antifibrotic therapy.