RESUMO
BACKGROUND: While echocardiography (ECHO)-measured left ventricular mass (LVM) predicts adverse cardiovascular events that are common in hemodialysis (HD) recipients, cardiac magnetic resonance imaging (CMR) is now considered the reference standard for determination of LVM. This study aimed to evaluate concordance between LVM measurements across ECHO and CMR among chronic HD recipients and matched controls. METHODS: A single-centre, cross-sectional study of 41 chronic HD patients and 41 matched controls with normal kidney function was performed to compare LVM measurements and left ventricular hypertrophy (LVH) designation by ECHO and CMR. RESULTS: In both groups, ECHO, compared with CMR, overestimated LVM. Bland-Altman analysis demonstrated wider agreement limits in LVM measurements by ECHO and CMR in the chronic HD group (mean difference, 60.8 g; limits -23 g to 144.6 g) than in the group with normal renal function (mean difference, 51.4 g; limits -10.5 g to 113.3 g). LVH prevalence by ECHO and CMR in the chronic HD group was 37.5% and 22.5%, respectively, while 17.5% and 12.5% had LVH by ECHO and CMR, respectively, in the normal kidney function group. Intermodality agreement in the designation of LVH was modest in the chronic HD patients (κ = 0.42, P = 0.005) but strong (κ = 0.81, P < 0.001) in the patients with preserved kidney function. Agreement was strong in assessing LVH by ECHO and CMR only in those with normal kidney function. CONCLUSIONS: Our results suggest that the limitations of LVM measurement by ECHO may be more pronounced in patients receiving HD, and provide additional support for the use of CMR in research and clinical practice when rigourous assessment of LVM is essential.