Echocardiographic Evaluation of Hemodynamics in Patients With Systolic Heart Failure Supported by a Continuous-Flow...

Hemodynamics assessment is important for detecting and treating post-implant residual heart failure, but its accuracy is unverified in patients with continuous-flow left ventricular assist devices (CF-LVADs).

We determined whether Doppler and 2-dimensional transthoracic echocardiography reliably assess hemodynamics in patients supported with CF-LVADs.

Simultaneous echocardiography and right heart catheterization were prospectively performed in 50 consecutive patients supported by using the HeartMate II CF-LVAD at baseline pump speeds. The first 40 patients were assessed to determine the accuracy of Doppler and 2-dimensional echocardiography parameters to estimate hemodynamics and to derive a diagnostic algorithm for discrimination between mean pulmonary capillary wedge pressure ≤15 versus >15 mm Hg. Ten patients served as a validation cohort.

Doppler echocardiographic and invasive measures of mean right atrial pressure (RAP) (r = 0.863; p < 0.0001), systolic pulmonary artery pressure (sPAP) (r = 0.880; p < 0.0001), right ventricular outflow tract stroke volume (r = 0.660; p < 0.0001), and pulmonary vascular resistance (r = 0.643; p = 0.001) correlated significantly. Several parameters, including mitral ratio of the early to late ventricular filling velocities >2, RAP >10 mm Hg, sPAP >40 mm Hg, left atrial volume index >33 ml/m(2), ratio of mitral inflow early diastolic filling peak velocity to early diastolic mitral annular velocity >14, and pulmonary vascular resistance >2.5 Wood units, accurately identified patients with pulmonary capillary wedge pressure >15 mm Hg (area under the curve: 0.73 to 0.98). An algorithm integrating mitral inflow velocities, RAP, sPAP, and left atrial volume index was 90% accurate in distinguishing normal from elevated left ventricular filling pressures.

Doppler echocardiography accurately estimated intracardiac hemodynamics in these patients supported with CF-LVAD. Our algorithm reliably distinguished normal from elevated left ventricular filling pressures.