Echocardiography-based qualification and response assessment to cardiac resynchronisation therapy in patients with chronic heart failure. The matrix metalloproteinase-9 substudy.

Background: The concept of cardiac resynchronisation therapy (CRT) is based on biventricular pacing in symptomatic, chronic heart failure (HF) patients with systolic left ventricular (LV) dysfunction and QRS >= 120 ms. The response to CRT is determined by clinical and echocardiographic parameters. The change of biochemical status (e.g. natriuretic peptides or metalloproteinase levels) caused by CRT is not well explored. Aim: To analyse the clinical and haemodynamic changes caused by CRT in relation to patients' biochemical status and to assess factors determining a favourable response to CRT. Methods: Fifty patients with chronic systolic HF (NYHA IV: two patients), wide QRS complex (160 +/- 31 ms) and reduced LV ejection fraction (26 +/- 5.8%) under optimal pharmacotherapy, who underwent CRT, were enrolled. Data on NT-proBNP and C-reactive protein serum levels, as well as standard echocardiography with tissue Doppler measurements, were collected before CRT and after six months of pacing. The levels of matrix metalloproteinase-9 (MMP-9) were assessed in a subgroup of 18 patients. Patients were regarded as responders if LVend-systolic volume decreased by 10% compared to baseline. Results: Thirty five (70%) patients responded favourably to CRT. Cardiac resynchronisation therapy resulted in an improvement of max. ventilatory oxygen uptake (12.9 +/- 3.8 vs 16.6 +/- 4.7 mL/kg/min; p < 0.05), a of NT-proBNP decrease (2,579 +/- 2,598 vs 1,339 +/- 1,088 pg/mL, p < 0.05), and decrease of atrio-, inter- and intra-LV clyssynchrony. A greater baseline dyssynchrony was observed in responders. A decrease of MMP-9 level following CRT was observed in 12 (67%) patients. Significant MMP-9 decrease was observed only in the subgroup of ischaemic HF patients (26,100 +/- 7,624 pg/mL vs 23,360 +/- 6,258 pg/mL; p = = 0.03). In patients with MMP-9 decrease during CRT, a lower C-reactive protein concentration at baseline was observed (2.12 +/- 1.6 vs 4.7 +/- 4.1 mg/L). The reduction in LV end-diastolic diameter correlated with the changes in MMP-9 level (r = = 51; p = 0.03). Baseline left atrial end-diastolic diameter measured in parasternal long-axis view <= 46 mm had a sensitivity of 83% and a specificity of 67% in predicting MMP-9 decrease (AUC 0.83; 95% CI 0.59-0.96). Conclusions: The CRT induces favourable myocardial remodelling, resulting in NT-proBNP level decrease, improvement of regional and global biventricular function, and MMP-9 level reduction, in ischaemic HF patients. The changes of MMP-9 level may be predicted by baseline left atrial end-diastolic diameter and correlate with LV end-diastolic diameter change during CRT.