Please cite this article in press as: Boussageon R, et al. Effects of pharmacological treatments on micro- and macrovascular complications of type 2 diabetes: What is the level of evidence? Diabetes Metab (2014), http://dx.doi.org/10.1016/j.diabet.2013.12.010 ARTICLE IN PRESS +Model DIABET-589; No. of Pages 7 Available online at ScienceDirect www.sciencedirect.com Diabetes & Metabolism xxx (2014) xxx–xxx Review Effects of pharmacological treatments on micro- and macrovascular complications of type 2 diabetes: What is the level of evidence? R. Boussageon a,∗ , F. Gueyffier b,c , C. Cornu b,c,d a Département de médecine générale, faculté de Poitiers, 11, route du Clos-Bardien, 86000 Poitiers, France b UMR 5558, CNRS, 69622 Villeurbanne cedex, France c Clinical Pharmacology, Louis-Pradel Hospital, CHU de Lyon, 69677 Bron cedex, France d Inserm, Clinical Investigation Center (CIC201), 69677 Bron cedex, France Received 27 August 2013; received in revised form 19 December 2013; accepted 22 December 2013 Abstract Antidiabetic drugs for type 2 diabetes receive marketing authorization if they show efficacy in reducing levels of HbA 1c . However, efficacy on this biological criterion does not necessarily reflect clinical benefit to patients. Several randomized clinical trials have shown that antidiabetic drugs reduce HbA 1c without a corresponding reduction in clinical events. This suggests a need to focus on the clinical effectiveness (morbimortality criteria) of our available antidiabetic drugs. In this non-extensive review of the literature, it was found that none of the current antidiabetic drugs have clearly proven their superiority over placebo in the gold standard double-blind randomized clinical trials. Thus, in 2013, the level of evidence for the clinical efficacy of antidiabetic drugs is disappointing and does not support the millions of prescriptions being written for them. © 2014 Elsevier Masson SAS. All rights reserved. Keywords: Antidiabetic drugs; Type 2 diabetes; Clinical efficacy; Level of evidence 1. Introduction The treatment of type 2 diabetes (T2D) is based on a seem- ingly simple principle. Observational studies have shown that hyperglycemia is a risk factor for excess mortality, cardiovas- cular events and microvascular complications [1]. It therefore appears logical that T2D patients would benefit from any treat- ment reducing hyperglycemia, and any drug with proven efficacy on the intermediate outcome of lowering HbA 1c may be con- sidered efficacious at preventing the clinical complications of T2D. Indeed, the US Food and Drug Administration (FDA) now approves marketing authorizations for new antidiabetic drugs if they reduce HbA 1c and show an excess relative risk of cardiovas- cular events that is clearly < 80% (upper limit of the confidence interval, or CI) [2]. On this basis, several new antidiabetic drugs have received marketing authorization, such as the dipeptidyl peptidase-4 (DPP-4) inhibitors, glucagon-like peptide (GLP)-1 analogues, and even new insulins and insulin analogues. ∗ Corresponding author. Tel.: +33 054 966 3531. E-mail address: remy.boussageon2@wanadoo.fr (R. Boussageon). HbA 1c has, until now, been considered a reliable surro- gate outcome despite the lack of any formal demonstration in randomized controlled trials (RCTs) using relevant clinical out- comes (such as morbimortality criteria). However, this should probably now be questioned. Several randomized trials with a high level of evidence have disproved the idea that reducing HbA 1c is beneficial for patients with T2D [3–5]. There was an increased all-cause and cardiovascular mortality (which led to premature termination of the study) in patients receiving inten- sified treatment in the Action to Control Cardiovascular Risk in Diabetes (ACCORD) [3] trial, even though their HbA 1c lev- els were lowered by 1.1% on average. In the Veterans Affairs Diabetes Trial (VADT) [4], there was a difference of 1.5% in HbA 1c values between the two groups throughout the follow- up (6.9% vs 8.4%), yet no differences were observed in total mortality [risk ratio (RR) = 1.08; 95% CI: 0.83–1.41], cardio- vascular mortality (RR = 1.22; 95% CI: 0.78–1.92) and non-fatal myocardial infarctions (RR = 0.78; 95% CI: 0.55–1.11). Ben- fluorex, rosiglitazone and pioglitazone were recently removed from the French marketplace, even though they reduce HbA 1c . The reason was that no convincing reduction in morbimortal- ity factors was seen with these drugs. Also, whenever serious 1262-3636/$ – see front matter © 2014 Elsevier Masson SAS. All rights reserved. http://dx.doi.org/10.1016/j.diabet.2013.12.010