Effects of frequent hemodialysis on ventricular volumes and left ventricular remodeling

Abstract

Background and objectives: Higher left ventricular volume is associated with death in patients with ESRD. This work investigated the effects of frequent hemodialysis on ventricular volumes and left ventricular remodeling.

Design, setting, participants, & measurements: The Frequent Hemodialysis Network daily trial randomized 245 patients to 12 months of six times per week versus three times per week in-center hemodialysis; the Frequent Hemodialysis Network nocturnal trial randomized 87 patients to 12 months of six times per week nocturnal hemodialysis versus three times per week predominantly home-based hemodialysis. Left and right ventricular end systolic and diastolic volumes, left ventricular mass, and ejection fraction at baseline and end of the study were ascertained by cardiac magnetic resonance imaging. The ratio of left ventricular mass/left ventricular end diastolic volume was used as a surrogate marker of left ventricular remodeling. In each trial, the effect of frequent dialysis on left or right ventricular end diastolic volume was tested between predefined subgroups.

Results: In the daily trial, frequent hemodialysis resulted in significant reductions in left ventricular end diastolic volume (-11.0% [95% confidence interval, -16.1% to -5.5%]), left ventricular end systolic volume (-14.8% [-22.7% to -6.2%]), right ventricular end diastolic volume (-11.6% [-19.0% to -3.6%]), and a trend for right ventricular end systolic volume (-11.3% [-21.4% to 0.1%]) compared with conventional therapy. The magnitude of reduction in left and right ventricular end diastolic volumes with frequent hemodialysis was accentuated among patients with residual urine output<100 ml/d (P value [interaction]=0.02). In the nocturnal trial, there were no significant changes in left or right ventricular volumes. The frequent dialysis interventions had no substantial effect on the ratio of left ventricular mass/left ventricular end diastolic volume in either trial.

Conclusions: Frequent in-center hemodialysis reduces left and right ventricular end systolic and diastolic ventricular volumes as well as left ventricular mass, but it does not affect left ventricular remodeling.

Trial registration: ClinicalTrials.gov NCT00264758 NCT00271999.

Figures

Figure 1.

Relative changes in left ventricular end diastolic volume (LVEDV), right ventricular end diastolic volume (RVEDV), left ventricular end systolic volume (LVESV), and right ventricular end systolic volume (RVESV) in patients randomized to either conventional (three times per week) or intensive (six times per week) dialysis. Left panel shows results of the daily trial; right panel shows results of the nocturnal trial. Error bars show ± SD.

Figure 2.

Effect of baseline residual urine volume on LVEDV, RVEDV, LVESV, and RVESV in the daily and nocturnal trials. Left panel shows results of the daily trial; right panel shows results of the nocturnal trial. Different scales were used for the vertical axis to accommodate wider confidence intervals for the nocturnal trial.

Figure 3.

Scatter plots relate change in LVEDV to change in predialysis systolic BP from baseline to 12 months with nonparametric loess regression lines. Left panel shows results of the daily trial; right panel shows results of the nocturnal trial.

Figure 4.

Changes from baseline to 12 months in LVEDV and LVMI. In upper panels, open circles designate the baseline LVEDV (x coordinate) and left ventricular mass index (LVMI) (y coordinate) values for each patient. For those patients with at least a 10% change in either LVEDV or LVMI, arrows are drawn from the baseline LVEDV and LVMI values to month 12 LVEDV and LVMI values. Open circles and arrows are colored red if both LVMI and LVEDV increased; they are colored blue if both decreased, green if LVMI increased but LVEDV decreased, and aquamarine if LVMI decreased but LVEDV increased. Lower panels contain corresponding bar charts indicating the percentage of patients falling into the four categories with at least a 10% change in one or both of LVMI and LVEDV within each treatment group.