To the Editor Thiazolidinediones are widely used in patients with type 2 diabetes. In addition to lowering plasma glucose, this class of drugs may reduce the risk of cardiovascular diseases in patients with diabetes. Despite the beneficial effect of thiazolidinediones, their use is sometimes associated with peripheral oedema. This is the most frequent adverse effect leading to discontinued treatment with thiazolidinediones in Japan. However, the mechanism by which thiazolidinediones induce oedema remains to be established. Recently, Guan et al. reported that, in a mouse model, thiazolidinediones caused fluid retention through peroxisome proliferator-activated receptor-γ stimulation of renal salt absorption mediated by epithelial Na channel (ENaC) [1]. The authors speculated that an increase in plasma volume causes thiazolidinedione-induced peripheral oedema. We, however, doubt that this is the sole cause of oedema in humans. ENaC is a sodium channel located in collecting tubules. This channel is also the site of action of the hormone aldosterone, a mineralocorticoid that increases sodium reabsorption. Primary aldosteronism is a disease caused by hypersecretion of this hormone. Due to excess sodium reabsorption, the circulating blood volume increases. The resulting headache and hypertension are well documented. However, peripheral oedema does not usually develop [2]. This is a well-known outcome of this disease. Oedema, moreover, also does not develop as a result of the increased plasma volume in the syndrome of inappropriate secretion of antidiuretic hormone [3]. Antidiuretic horomone is a pituitary hormone that activates the water channel aquaporin 2 in collecting tubules, thereby increasing water reabsorption. These observations in relation to endocrine diseases suggest that blood volume expansion is not sufficient for peripheral oedema to develop. We would like to highlight another important factor that might also be involved in the process. Previously, we reported that thiazolidinediones increase the plasma levels of vascular endothelial growth factor (VEGF) [4]. Since VEGF is a potent vascular permeability factor, an increase in VEGF levels may contribute to thiazolidinedioneinduced oedema. However, because of the difficulty of quantifying the severity of oedema, no one has demonstrated a clear correlation between peripheral oedema and plasma VEGF levels. Assessing the presence of peripheral oedema by body weight can be misleading, because thiazolidinediones increase body fat mass [5] and circulating blood volume [1]. In our previous report, for example, a weak correlation between body weight and plasma VEGF concentration was observed only in female patients, but not in male patients or groups comprising both sexes [4]. Here, we demonstrate a strong correlation between clinically apparent peripheral oedema and plasma VEGF concentration. Oedema is defined as a clinically apparent increase in the interstitial fluid volume. Thus, we judged the existence of peripheral oedema by the presence of pitting after pressure was applied to the bilateral lower extremities. We re-evaluated the clinical data collected for our previous study [4]. Briefly, a total of 30 patients with type 2 diabetes Diabetologia (2006) 49:2217–2218 DOI 10.1007/s00125-006-0313-5
Read full abstract