Review article

Thiazolidinediones and heart failure Parminder S. Chaggar1, Steven M. Shaw1 and Simon G. Williams1

Diabetes & Vascular Disease Research 6(3) 146­–152 © The Author(s) 2009 Reprints and permission: http://www. sagepub.co.uk/journalsPermission.nav DOI: 10.1177/1479164109338772 http://dvr.sagepub.com

Abstract DM is an independent risk factor for the development of HF and its presence confers an adverse prognosis for those already diagnosed with HF.TZDs are potent insulin-sensitisers associated with a number of beneficial cardiovascular effects. However,TZDs increase renal sodium and water reabsorption, leading to fluid retention and overt signs of HF in patients with diabetes. Rosiglitazone has also been associated with an increased risk of myocardial infarction and cardiovascular mortality. However, pioglitazone may have macrovascular benefits. The majority of data on the cardiovascular safety of TZDs are based on non-cardovascular outcome trials and meta-analyses. Concerns regarding the risk of HF and cardiovascular safety of TZDs have led to restrictions on their use in patients with HF. This review addresses the latest evidence for HF with each of the TZD drugs currently available and reflects on the current guidelines regarding their prescription in at-risk patients. Key words heart failure, thiazolidinediones, diabetes

Introduction HF and type 1 and 2 DM are common medical problems associated with significant morbidity and mortality.1, 2 DM is an independent risk factor for HF and patients with both conditions have a worse outcome than those with HF alone.3, 4 Tight glycaemic control in DM is associated with significant morbidity and mortality benefits.5, 6 In addition, insulin resistance has been shown to be proportional to HF severity.7 This is independent of the aetiology of HF, suggesting a mechanism beyond the presence of atheroma. Certainly, DM is associated with a number of cellular and metabolic changes that impact negatively on cardiomyocyte function and culminate in the syndrome of diabetic cardiomyopathy.8 However, no trials have directly assessed the prognostic outcome of improving insulin sensitivity in HF. TZDs, including rosiglitazone and pioglitazone, are potent insulin sensitisers which act via the nuclear transcription factor, peroxisome proliferator-activated receptor  (PPAR).9 In addition to their insulin-sensitising effects, TZDs have a number of effects that are of potential benefit to patients with HF, including blood pressure lowering, angiotensin II reduction, endothelial function and lipid profile improvements and slowing of the progression of atherosclerosis.9-11 However, TZDs also increase sodium reabsorption in the distal nephron,12,13 leading to fluid retention and peripheral oedema, which may be of particular concern in patients with HF. Recent meta-analyses have indicated that pioglitazone use in DM increases the risk of HF14, 15 and that rosiglitazone increases the risk of both HF and MI.16-18 These reports have resulted in a consensus statement from the AHA and the ADA recommending caution in prescribing TZDs to

patients with NYHA class I–II HF and completely avoiding TZDs in patients with NYHA class II I–IV HF.19 Even more recently, randomised controlled trials directly assessing cardiovascular outcomes with TZD therapy have been published.20, 21 Here we review the latest evidence exploring the association between HF and currently available TZDs (table 1) and guidance regarding their prescription in at-risk patients.

HF Risk with TZDs Rosiglitazone Initial trials assessing the efficacy and tolerability of rosiglitazone did not evaluate cardiovascular side-effects as pre-specified outcomes. However, these trials noted an increased incidence of clinical HF in patients treated with rosiglitazone. 22, 23 The ADOPT trial22 assessed the efficacy of rosiglitazone compared with the established anti-diabetic treatments of metformin and glyburide in 4,360 patients in a double-blind, randomised controlled manner. Over the median follow-up period of 4 years, rosiglitazone was significantly associated with more cases of oedema and diuretic

1

The Northwest Heart and Transplant Centre, Wythenshawe Hospital, Manchester, M23 9LT, United Kingdom Corresponding author: Dr S.G. Williams, The Northwest Heart and Transplant Centre, Wythenshawe Hospital, Southmoor Road, Manchester, M23 9LT, United Kingdom. Email: [email protected]

RCT of rosiglitazone vs. metformin or glyburide

RCT of rosiglitazone vs. placebo

Updated MA of Nissen and Wolski16 with ADOPT, DREAM and RECORD data

MA of 7 RCT of rosiglitazone and pioglitazone

MA of 4 RCT of >12 months’ duration with CVS endpoints

RCT of rosiglitazone vs. no rosiglitazone as add-on therapy

MA of 19 RCT of pioglitazone

RCT of pioglitazone vs. placebo

RCT of pioglitazone vs. glyburide in patients with HF (LVEF