Lung Ultrasound-Guided Dry Weight Assessment and Echocardiographic Measures in Hypertensive Hemodialysis Patients: A Randomized Controlled Study

Rationale & Objective: Left ventricular (LV) hypertrophy and dysfunction are associated with adverse outcomes in hemodialysis patients. Hypertension and hypervolemia play important roles in these cardiac abnormalities. We report on the prespecified secondary outcome, echocardiographic indexes of LV function, from a previously reported study of the effect of lung ultrasound (US)-guided dry weight reduction on systolic blood pressure. Study Design: Single-blind randomized trial. Settings & Participants: 71 clinically euvolemic hypertensive hemodialysis patients in Greece and Slovenia. Intervention: The active intervention group's (n = 35) volume removal was guided by the total number of lung US B-lines observed every week before a midweek dialysis session. The usual-care group (n = 36) was treated using standard-ofcare processes that did not include acquisition of US data. Outcomes: 2-dimensional and tissue Doppler echocardiographic indexes at baseline and study end (8 weeks) that evaluated left and right heart chamber sizes, as well as systolic and diastolic function. Results: Overall, 19 (54%) patients in the active intervention and 5 (14%) in the usual-care group had ultrafiltration intensification (P < 0.001) during follow-up; changes in US B-lines (-5.3 +/- 12.5 vs +2.2 +/- 7.6; P < 0.001) and dry weight (-0.71 +/- 1.39 vs +0.51 +/- 0.98 kg; P < 0.001) significantly differed between the active and usual-care groups. Inferior vena Gaya diameter decreased in the active compared with the usual-care group (-0.43 +/- 4.00 vs 0.71 +/- 4.82 cm; P = 0.03) at study end. Left (LA) and right (RA) atrial dimensions decreased more in the active group (LA surface, -1.09 +/- 4.61 vs 0.93 +/- 3.06 cm(2); P = 0.03; RA surface -1.56 +/- 6.17 vs 0.47 +/- 2.31; P = 0.02). LA volume index nominally decreased more in the active group (-2.43 +/- 13.14 vs 2.95 +/- 9.42 mL/m(2)), though this was of borderline statistical significance (P = 0.05). Reductions in LV enddiastolic diameter and volume were marginally greater in the active group. The change in LV filling pressures was significantly different in the active compared with the usual-care group (early transmitral diastolic velocities ratio [E/e'], -0.38 +/- 3.14 vs 1.36 +/- 3.54; P = 0.03; E wave deceleration time, 35.43 +/- 85.25 vs -18.44 +/- 50.69; P= 0.002]. Systolic function indexes were unchanged in both groups. In multivariable analysis, US B-line reduction was associated with a reduction in the E/e' LV ratio (OR, 4.542; 95% CI, 1.266-16.292; P= 0.02). Limitations: Exploratory study; small sample size. Conclusions: A US-guided strategy for dry weight reduction is associated with decreased cardiac chamber dimensions and LV filling pressure, but no difference in systolic performance compared with usual care in hypertensive hemodialysis patients.