Preload dependence of new Doppler techniques limits their utility for left ventricular diastolic function assessment in hemodialysis patients

Left ventricular (LV) hypertrophy leads to diastolic dysfunction. Standard Doppler transmitral and pulmonary vein (PV) flow velocity measurements are preload dependent. New techniques such as mitral annulus velocity by Doppler tissue imaging (DTI) and LV inflow propagation velocity measured from color M-mode have been proposed as relatively preload-independent measurements of diastolic function. These parameters were studied before and after hemodialysis (HD) with ultrafiltration to test their potential advantage for LV diastolic function assessment in HD patients. Ten patients (seven with LV hypertrophy) underwent Doppler echocardiography 1 h before, 1 h after, and 1 d after HD. Early (E) and atrial (A) peak transmitral flow velocities, peak PV systolic (s) and diastolic (d) flow velocities, peak e and a mitral annulus velocities in DTI, and early diastolic LV flow propagation velocity (V(p)) were measured. In all patients, the E/A ratio after HD (0.54; 0.37 to 1.02) was lower (P < 0.01) than before HD (0.77; 0.60 to 1.34). E decreased (P < 0.01), whereas A did not. PV s/d after HD (2.15; 1.08 to 3.90) was higher (P < 0.01) than before HD (1.80; 1.25 to 2.68). Tissue e/a after HD (0.40; 0.26 to 0.96) was lower (P < 0.01) than before HD (0.56; 0.40 to 1.05). Tissue e decreased (P < 0.02), whereas a did not. V(p) after HD (30 cm/s; 16 to 47 cm/s) was lower (P < 0.01) than before HD (45 cm/s; 32 to 60 cm/s). Twenty-four hours after the initial measurements values for E/A (0.59; 0.37 to 1.23), PV s/d (1.85; 1.07 to 3.38), e/a (0.41; 0.27 to 1.06), and V(p) (28 cm/s; 23 to 33 cm/s) were similar as those taken 1 h after HD. It is concluded that, even when using the newer Doppler techniques DTI and color M-mode, pseudonormalization, which was due to volume overload before HD, resulted in underestimation of the degree of diastolic dysfunction. Therefore, the advantage of these techniques over conventional parameters for the assessment of LV diastolic function in HD patients is limited. Assessment of LV diastolic function should not be performed shortly before HD, and its time relation to HD is essential.