Fibrosis in heart failure and atrial fibrillation assessed by TIMP-1 serum levels in comparison with echocardiographic measurement oft left atrial strain

Anett Große (Jena)1, K. Kirsch (Jena)1, L. Herzer (Jena)1, N. Duckwitz (Jena)1, F. Mettke (Jena)1, K. Grün (Jena)1, M. Franz (Jena)1, C. Schulze (Jena)1, R. Surber (Jena)1

1Universitätsklinikum Jena Klinik für Innere Medizin I - Kardiologie Jena, Deutschland


Background: Tissue inhibitor of metalloproteinase- 1 (TIMP-1) levels are strongly associated with cardiac extracellular matrix accumulation and atrial fibrosis. Left atrial (LA) strain is an echocardiographic method to detect LA- function. Heart failure (HF) and atrial fibrillation (AF) are associated with structural and functional changes in the LA. The aim of the study was 1) to measure TIMP-1 plasma levels and the LA strain in HF- and AF pts compared to pts without structural heart disease and 2) to search for a correlation between TIMP-1 plasma levels and echocardiographic measurement of LA strain.

Methods: This study included 11 pts with HF who underwent CRT- implantation (HF group), 10 pts with persistent AF who underwent LA catheter ablation (AF group) and 8 pts without structural heart disease or AF, who underwent electrophysiological study (control group). Blood was collected at the beginning of the procedures from the coronary sinus (CS), superior vena cava (SVC) and aorta (Ao). Level of TIMP-1 (ng/ml) in the CS, SVC and aorta was measured. The measurement of reservoir (LASr), conduit (LAScd) and contraction (LASct) LA- strain- using two-dimensional speckle tracking echocardiography (2D-STE) was performed in all patients before the procedure.

Results: Patients characteristics: HF group: mean age 74y, 82% male, LV- EF 27%, AF group: mean age 66y, 80% male, LV- EF 54%, control-group: mean age 52y, 63% male, LV- EF 60%.

At the various collection sites (CS, SVC and Ao), no significant differences between the 3 groups were noted in TIMP-1 levels. The level of TIMP-1 in the Ao was in the HF group significantly higher 223 ± 88 ng/ml vs 146 ± 32 ng/ml in controls (p= 0.04). In the AF- group (174 ± 48 ng/ml) it was not significantly different from the control group (p= 1.0) (fig 1).

The LASr was normal in controls (mean 41%), but reduced in HF (mean 18%) and AF pts (mean 23%), the differences to controls were significant (p<0.05). Otherwise, the LA strain between the AF and HF group was not different.

The TIMP-1 serum levels are moderate negative correlated to LA strain (r= -0.406) (fig 2).

Discussion: TIMP-1 serum levels in HF patients are significantly elevated compared to a control group without structural heart disease. No significant differences in levels of TIMP-1 were seen between AF patients and the control group. This is an expression of the advanced disease in the HF patients. In HF and AF patients the LA strain is significantly reduced in the same way compared to healthy ones. The serum level of TIMP 1 is negative correlated to the LA strain. Whether an effective treatment of HF and AF will lower TNC and TIMP-1 levels and improve LA- strain due to slow down the disease progression needs to be evaluated in further studies.


Diese Seite teilen