Assessment of cardiac structure and function in patients without and with peripheral oedema during rosiglitazone treatment

Background: Thiazolidinediones cause peripheral oedema, the aetiology of which remains poorly understood. Methods: In a sub-study of a 6-month trial comparing rosiglitazone (Rsg) versus placebo, we compared those with versus without oedema among the 74 subjects treated with Rsg with respect to peak oxygen consumption indexed to fat-free mass (VO_2peak-FFM), cardiac MRI and markers of plasma volume expansion. Results: Almost half (49%) of the Rsg-treated patients developed oedema. Baseline VO_2peak-FFM was not different between those with versus without oedema (25.8 versus 28.2 ml/kg/min; p = 0.22) and declined 5% in the oedema group (Δ -1.3 ml/min/kg; p = 0.005) with no change in those without oedema. Stroke volume increased in both groups (Δ 8.7 and 8.8 ml; p < 0.001 for each); end-diastolic volume increased only in those with oedema (+13.1 ml; p = 0.001). No other cardiac function changes were observed. In both groups, weight increased (3.6 and 2.2 kg) and haematocrit decreased (-3.2% and -2.1%; p < 0.001 for each). In those with oedema, albumin decreased (-0.2 g/dl) and brain natriuretic peptide increased (11.9 pg/ml; p < 0.03 for each). Conclusions: Oedema was associated with a small decline in VO_2peak-FFM, no adverse effects on cardiac function, and changes in selected measures suggesting that volume expansion underpins Rsg oedema.