Efeito da valvuloplastia mitral percutânea sobre a função ventricular esquerda avaliada pela ecocardiografia tridimensional na estenose mitral

Abstract

Mitral stenosis (MS) is the main sequel of rheumatic fever (RF). Rheumatic disease produces valve deformation, especially in the mitral valve, causing serious consequences. The objectives of this study were to evaluate the left ventricular (LV) systolic function in patients with MS by the left ventricular ejection fraction (LVEF) before and after percutaneous mitral valvuloplasty (PMV), using the threedimensional echocardiography in real time (E3D); Correlate the variation in LVEF changes in natriuretic peptide (BNP) and hemodynamic parameters with the procedure; and verify the impact of LVEF changes on long-term adverse events. This is a prospective study, with 142 patients, aged 42.3 ± 12.1 years old, 89% female MS patients with indication for PMV, coming from the Valve Disease Clinic at the Hospital das Clinicas, Federal University of Minas Gerais between April 1, 2008 to October 31, 2014. The patients were followed up after the procedure to detect adverse events, setting up primary and secondary outcome such as cardiovascular death and death from all causes, repeat PMV, mitral valve replacement, new onset of atrial fibrillation or stroke. A control group consisting of 20 healthy subjects with age, gender and body surface similar to the cases and normal echocardiogram was selected. The baseline LVEF by E3D was found to be lower than in the control group (50.7 ± 8.8% vs 55.5 ± 5.4%; p <0.001), although within the reference range, with a significant increase (50.6 ± 8.7 vs. 55.3 ± 8.0; p <0.001) 24 hours after PMV. This variation was due to the significant increase in end-diastolic volume (68.1 ± 20.8 vs 71.7 ± 21.4 mL; p 0.002), the end-systolic volume remained unchanged (33.7 ± 12.9 vs 32.2 ± 10.9 ml; p 0.191), resulting in significant increase in the stroke volume (34.3 ± 11.2 vs. 40.8 ± 13.1 ml, p <0.001). There was a significant reduction of serum BNP levels with median [interquartile range] 181 [105/303] to 96 [50/223] pg / ml; p <0.001). This reduction was correlated to change in LVEF (r -0.23; p <0.04) in univariate analysis. There was improvement of hemodynamic parameters with PMV. However, in multivariate analysis, the only variable that correlated to change in LVEF was the change in cardiac index (CI) (r 0.31; p <0.007). During a mean follow-up of 21 months (ranging from 12 hours to 86 months), 21 adverse events occurred (3 deaths, 8 mitral valve replacement , 4 repeat MVP, 5 new onset of AF and 1 stroke). LVEF or its variation were not predictive of long-term adverse events. In multivariate analysis using the Cox model, the valve area and mean gradient after PMV were independent predictors of combined events. In conclusion, baseline LVEF by E3D was lower compared to healthy controls. A significant increase in LVEF was observed with the PMV, being similar to the values of control group. The increase in LVEF correlated to increased CI, but was not associated with late adverse events.