Cardiac magnetic resonance imaging assessment of ventricular mass: a prospective randomized study of the effect of icodextrin-based solution

Background: Left ventricular hypertrophy (LVH) is a risk factor for cardiovascular mortality in patients on peritoneal dialysis (PD). Icodextrin has a greater ultrafiltration power and could improve left ventricular mass by treating fluid overload. Findings from cardiac magnetic resonance imaging (MRI), however, are scarce. Methods: This is a randomized cohort that included prevalent patients on PD recruited from 2 tertiary hospitals. Patients were allocated to the glucose (GLU) or icodextrin (ICO) group. Clinical, demographic, biochemical, bioimpedance (extracellular water/total body water ratio – AEC/ACT) data, and cardiac MRI were evaluated at baseline and after 6 months. The outcome was a change in left ventricular mass adjusted by surface area (ΔLVMI). Results: 22 completed the study (GLU, N=12 and ICO, N=10). Patients from GLU and ICO groups had similar age, sex, underlying disease, and time on dialysis. At baseline, LVH was found in 17 patients (77.3%), with no difference between groups (p=0.748). According to ECW/TBW, 36.8% and 80% of patients from GLU and ICO groups, respectively, were considered hypervolemic (p=0.044). During follow-up, ΔLVMI was 3.9 (-10.7, 2.2)g/m in GLU and 5.2 (-26.8, 16.8) in ICO group (p=0.651). ΔLVMI correlated with change in ΔBNP (r=0.566, p=0.044), and remained significant in a multiple regression analysis. Conclusion: We found no superiority of icodextrin to glucose-based solution in improving LMVI in prevalent patients on PD. Whether icodextrin would improve LVMI in long-term follow-up deserver further evaluation.