We aimed to clarify the correlates of left ventricular mass and secondarily, left ventricular volume, in a cohort of prevalent hemodialysis recipients. Left ventricular hypertrophy is common and left ventricular mass is a widely-accepted surrogate for clinical outcomes in dialysis recipients, who are often subjected to chronic pressure and volume overload. However, the precise pathophysiologic mechanisms of left ventricular hypertrophy in this unique population have not been well understood. This was a cross-sectional study of patients receiving conventional thrice-weekly dialysis in Toronto, Canada. Left ventricular mass and volume were assessed with cardiac magnetic resonance and indexed to the patient's height to the power of 2.7. Fibroblast growth factor-23 concentration was measured using a C-terminal enzyme-linked immunosorbent assay. Patient demographics, comorbidities, dialysis-associated blood pressures and ultrafiltration volumes, biochemical and hematologic parameters, vascular access and medications were extracted from clinical records. Multivariable linear regression was used to identify independent correlates of left ventricular mass index (LVMI) and the left ventricular end diastolic volume index (LVEDVI). We enrolled 56 patients, of whom 23 (41.1 %) were women with mean age 54 ± 12 years. Mean LVMI was 31.1 ± 6.8 g/m(2.7). In multivariable analyses, systolic blood pressure and LVEDVI were the only factors significantly associated with LVMI. Post-dialysis weight, percent reduction in urea and the presence of a permanent form of vascular access were associated with LVEDVI. Fibroblast growth factor-23 was not associated with either LVMI or LVEDVI. Blood pressure and left ventricular dilatation are independent determinants of elevated left ventricular mass. Aggressive blood pressure reduction and avoidance of volume overload may confer LVM regression and improve clinical outcomes.