To the Editor: It is well known that diabetes increases the risk of developing heart failure (HF) (1). However, it was not clear whether this increased risk also occurred in patients with new-onset diabetes. A manuscript has recently been published that analysed the risk of HF in patients with recent-onset type 2 diabetes (2). Interestingly, authors found that type 2 diabetes was a major risk factor for the development of HF, and also that HF developed soon after diagnosis. Thus, the risk of new HF in individuals with diabetes was approximately threefold greater than for age- and sex-standardised controls. This is in accordance with the data reported in the VALIANT trial (3), in which diabetes mellitus, whether newly diagnosed or previously known, was associated with poorer long-term outcomes after myocardial infarction in high-risk patients. On the other hand, it has been suggested that there are associations between patient and physician characteristics with the prescription of the recommended drugs that should be taken into account to translate guideline recommendations for application in general practice (4). In this context, although thiazolidinediones precipitate fluid retention in clinical trials, current guidelines advocate their use for patients with diabetes who are felt to be at low risk for HF (5). However, thiazolidinediones (pioglitazone and rosiglitazone) seem to increase the risk of HF (6, 7). This has been related to the activation of the peroxisome proliferator-activated receptor-gamma (PPAR-γ) that modifies the metabolism of fat acids, which are crucial in the heart metabolism, and secondarily might facilitate the development of HF (8). This is very relevant, not only in daily clinical practice but also in the outcomes of clinical trials. For example, despite diabetes is associated with an increased risk of hospitalisations for HF and that available data suggest that inhibition of renin angiotensin system may prevent the development of HF (9), the results of the TRANSCEND trial are at least surprising (10). In this study, telmisartan was compared with placebo in patients with cardiovascular disease or diabetes with end-organ damage intolerant to angiotensin-converting enzyme inhibitors. Although the drug had no significant effect on the primary outcome, which included hospitalisations for HF, when this condition was excluded, it modestly reduced the risk of the composite outcome of cardiovascular death, myocardial infarction or stroke. Interestingly also in ONTARGET study (11), hospitalisation for HF was less frequently reported in the ramipril group than in the telmisartan arm. This is even more unexpected as telmisartan reduced the incidence of left ventricular hypertrophy, an independent predictor of new-onset HF, especially in diabetics (10). How is it possible that the addition of telmisartan to the standard treatment of patients with cardiovascular disease did not prevent admission due to HF? In these studies, although the proportion of patients with diabetes was given as well as the oral antidiabetics, however, unfortunately the percentage of patients taking thiazolidinediones was not provided (10, 11). Did the use of this drug influence the final results of these trials? Even more, as telmisartan modulates the PPAR-γ (12), unlike the other renin angiotensin system inhibitors, did the possible interaction between rosiglitazone and telmisartan has any importance in the results obtained in the diabetic population? However, it should be noted that thiazolidinediones differ with respect to activity on PPAR receptors α and γ. They are selective but not specific for α and γ binding. Therefore, drug interactions may be complex with agents that have partial PPAR activity such as telmisartan (13). Anyway, further investigations are needed to assess a potential deleterious interaction of thiazolidinediones, particularly, rosiglitazone and telmisartan in the development of HF. None.