The Current Drug Treatment Landscape for Diabetes and Perspectives for the Future.

The escalating global epidemic of type 2 diabetes mellitus has focused attention on the devastating consequences of protracted hyperglycemia. Early and effective intervention to control blood glucose is a fundamental principle of treatment guidelines, requiring assiduous use of current therapies. However, many patients do not achieve or maintain glycemic targets, emphasizing the need for further therapies. This narrative review assesses the available medicinal options to address hyperglycemia and the opportunities to develop novel agents.

Safety of antidiabetes medications: An update.

Adverse iatrogenic effects of especial relevance for antidiabetes medications include hypoglycemic episodes, major adverse cardiovascular (CV) events, cancer, bone fractures, pancreatic effects, genital/urinary tract infections, and weight gain. Here, recent clinical studies addressing safety profiles of antidiabetes medications are reviewed. On balance, new prospective and population-based studies continue to indicate that the benefits of improved glucose control outweigh the risks associated with antidiabetes medications in most patients with type 2 diabetes.

Efficacy and safety of dapagliflozin monotherapy in people with Type 2 diabetes: a randomized double-blind placebo-controlled 102-week trial.

AIMS: To assess initial pharmacotherapy of Type 2 diabetes with the sodium-glucose cotransporter-2 inhibitor dapagliflozin. METHODS: This double-blind, placebo-controlled trial, randomly allocated people with Type 2 diabetes aged 18-77 years and inadequate glycaemic control on diet and exercise [HbA1c 53-86 mmol/mol (7.0-10.0%)] to receive placebo (n = 75) or dapagliflozin monotherapy 2.5 mg (n = 65), 5 mg (n = 64) or 10 mg (n = 70) once daily in the morning. After 24 weeks, low-dose double-blind metformin 500 mg/day was added to the placebo group regimen (placebo+low-dose metformin group). Changes in HbA1c level, fasting plasma glucose and body weight, as well as adverse events, were assessed over 102 weeks. RESULTS: Of the 274 participants randomized, 167 completed the study (60.9%). At 102 weeks, significant differences vs placebo+low-dose metformin with dapagliflozin 5 and 10 mg were observed for HbA1c (-5.8 mmol/mol [-0.53%], P = 0.018; and -4.8 mmol/mol [-0.44%], P = 0.048), respectively); and for FPG (-0.69 mmol/L, P = 0.044; and -1.12 mmol/l, P = 0.001, respectively). For body weight, the difference between the dapagliflozin 10-mg group and the placebo+low-dose metformin group was significant (-2.60 kg; P = 0.016). Hypoglycaemic events were uncommon, with rates of 5.3% for placebo+low-dose metformin group and 0-4.6% for the dapagliflozin groups. Genital infections and urinary tract infections were more common in the dapagliflozin groups than in the placebo+low-dose metformin group. CONCLUSIONS: Dapagliflozin as monotherapy in treatment-naïve people with early Type 2 diabetes improved glycaemic control and reduced weight without increasing hypoglycaemia over 102 weeks. Dapagliflozin may provide an alternative initial pharmacotherapy in such people.

Pronounced reduction of postprandial glucagon by lixisenatide: a meta-analysis of randomized clinical trials.

AIM: Glucagon-like peptide-1 (GLP-1) receptor agonists improve islet function and delay gastric emptying in patients with type 2 diabetes mellitus (T2DM). This meta-analysis aimed to investigate the effects of the once-daily prandial GLP-1 receptor agonist lixisenatide on postprandial plasma glucose (PPG), glucagon and insulin levels. METHODS: Six randomized, placebo-controlled studies of lixisenatide 20 µg once daily were included in this analysis: lixisenatide as monotherapy (GetGoal-Mono), as add-on to oral antidiabetic drugs (OADs; GetGoal-M, GetGoal-S) or in combination with basal insulin (GetGoal-L, GetGoal-Duo-1 and GetGoal-L-Asia). Change in 2-h PPG and glucose excursion were evaluated across six studies. Change in 2-h glucagon and postprandial insulin were evaluated across two studies. A meta-analysis was performed on least square (LS) mean estimates obtained from analysis of covariance (ANCOVA)-based linear regression. RESULTS: Lixisenatide significantly reduced 2-h PPG from baseline (LS mean difference vs. placebo: -4.9 mmol/l, p < 0.001) and glucose excursion (LS mean difference vs. placebo: -4.5 mmol/l, p < 0.001). As measured in two studies, lixisenatide also reduced postprandial glucagon (LS mean difference vs. placebo: -19.0 ng/l, p < 0.001) and insulin (LS mean difference vs. placebo: -64.8 pmol/l, p < 0.001). There was a stronger correlation between 2-h postprandial glucagon and 2-h PPG with lixisenatide than with placebo. CONCLUSIONS: Lixisenatide significantly reduced 2-h PPG and glucose excursion together with a marked reduction in postprandial glucagon and insulin; thus, lixisenatide appears to have biological effects on blood glucose that are independent of increased insulin secretion. These effects may be, in part, attributed to reduced glucagon secretion.

Dapagliflozin monotherapy in drug-naïve patients with diabetes: a randomized-controlled trial of low-dose range.

AIMS: Many patients with type 2 diabetes are suboptimally managed with currently available therapies. Dapagliflozin, a sodium-glucose co-transporter-2 inhibitor, has shown efficacy in reducing diabetic hyperglycaemia. This study assessed efficacy of three lower doses in recently diagnosed patients. METHODS: This phase 3, randomized, double-blind, placebo-controlled study assigned treatment-naïve patients to placebo or dapagliflozin monotherapy (1, 2.5 or 5 mg) daily for 24 weeks. Patients were antidiabetic drug-naïve with inadequate glycaemic control [haemoglobin A1c (HbA1c) ≥7.0 and ≤10.0%]. The primary efficacy endpoint was change in HbA1c from baseline. Secondary endpoints included changes in body weight and fasting plasma glucose (FPG), and proportions achieving HbA1c <7%. RESULTS: A total of 282 patients with type 2 diabetes were randomly assigned to one of four treatment groups. Baseline characteristics were similar across groups. At week 24, mean HbA1c reduction was significantly greater with dapagliflozin: -0.68% for 1 mg, -0.72% for 2.5 mg, -0.82% for 5 mg, versus 0.02% for placebo (p < 0.0001); compared to mean baseline values of 7.8-8.1%. Mean FPG reduction was significantly greater for all dapagliflozin groups versus placebo (p < 0.02), as was mean weight reduction (p < 0.003). During the treatment period, 19.1% of placebo-treated patients received rescue medication or discontinued because of poor glycaemic control versus 6.9, 4.1 and 5.9% for dapagliflozin 1, 2.5 and 5 mg, respectively. Percentages of patients experiencing ≥1 adverse event were similar across groups. CONCLUSION: Dapagliflozin at doses of 1, 2.5 and 5 mg/day is effective in reducing glycaemic levels and body weight in treatment-naïve patients with type 2 diabetes. Dapagliflozin was generally well tolerated. This insulin-independent mechanism suggests a new treatment for type 2 diabetes.

Rotational co-culture of clonal ß-cells with endothelial cells: effect of PPAR-γ agonism in vitro on insulin and VEGF secretion.

AIM: Delayed graft revascularization impedes the success of human islet transplantation. This study utilized rotational co-culture of insulin secreting ß-cells with human umbilical vein endothelial cells (HUVECs) and a peroxisome proliferator-activated receptor gamma (PPAR-γ) agonist to promote insulin and vascular endothelial growth factor (VEGF) secretory function. METHODS: Clonal BRIN-BD11 (D11) cells were maintained in static culture (SC) and rotational culture (RC) ± HUVEC and ± the TZD (thiazolidinedione) rosiglitazone (10 mmol/l) as a specific PPAR-γ agonist. HUVECs were cultured in SC and RC ± D11 and ± TZD. D11 insulin secretion was induced by static incubation with low glucose (1.67 mmol/l), high glucose (16.7 mmol/l) and high glucose with 10 mmol/l theophylline (G+T) and assessed by enzyme-linked immunosorbent assay (ELISA). HUVEC proliferation was determined by ATP luminescence, whereas VEGF secretion was quantified by ELISA. Co-cultured cells were characterized by immunostaining for insulin and CD31. RESULTS: D11 SC and RC showed enhanced insulin secretion in response to 16.7 mmol/l and G+T (p < 0.01); without significant alteration by the TZD. Co-culture with HUVEC in SC and RC also increased D11 insulin secretion when challenged with 16.7 mmol/l and G+T (p < 0.01), and this was slightly enhanced by the TZD. The presence of HUVEC increased D11 SC and RC insulin secretion in response to high glucose and G+T, respectively (p < 0.01). Addition of the TZD increased SC and RC HUVEC ATP content (p < 0.01) and VEGF production (p < 0.01) in the presence and absence of D11 cells. CONCLUSIONS: Rotational co-culture of insulin secreting cells with endothelial cells, and exposure to a PPAR-γ agonist may improve the prospects for graft revascularization and function after implantation.

Patient adherence to medication requirements for therapy of type 2 diabetes.

Type 2 diabetes is a complex, progressive endocrine and metabolical disease that typically requires substantial lifestyle changes and multiple medications to lower blood glucose, reduce cardiovascular risk and address comorbidities. Despite an extensive range of available and effective treatments, < 50% of patients achieve a glycaemical target of HbA(1c) < 7.0% and about two-thirds die of premature cardiovascular disease. Adherence to prescribed therapies is an important factor in the management of type 2 diabetes that is often overlooked. Inadequate adherence to oral antidiabetes agents, defined as collecting < 80% of prescribed medication, is variously estimated to apply to between 36% and 93% of patients. All studies affirm that a significant proportion of type 2 diabetes patients exhibit poor adherence that will contribute to less than desired control. Identified factors that impede adherence include complex dosing regimens, clinical inertia, safety concerns, socioeconomic issues, ethnicity, patient education and beliefs, social support and polypharmacy. This review explores these factors and potential strategies to improve adherence in patients with type 2 diabetes.

Acute and long-term effects of peroxisome proliferator-activated receptor-γ activation on the function and insulin secretory responsiveness of clonal beta-cells.

Thiazolidinediones (TZDs) are used as antidiabetic therapy. The purpose of the present study was to examine whether the TZD rosiglitazone has direct actions on pancreatic beta-cells that contribute to its overall effects. Effects of acute and prolonged (48 h) exposure to rosiglitazone, as a model glitazone compound, were assessed in clonal pancreatic BRIN-BD11 beta-cells maintained in standard, glucotoxic and lipotoxic cultures. In acute 20-min incubations, rosiglitazone (0.2-100 µM) did not alter basal or glucose-stimulated insulin secretion. However, rosiglitazone (6.25 µM) enhanced (p<0.001) the acute insulinotropic action of GLP-1. Prolonged exposure to 6.25 µM rosiglitazone in standard media had no effect on cell viability or cellular insulin content, but slightly reduced the insulin secretory response to glucose and alanine (p<0.05). Prolonged (48 h) exposure to glucotoxic or lipotoxic conditions reduced beta-cell viability (p<0.05), cellular insulin content (p<0.001 and p<0.05, respectively), and insulin release in response to glucose and a range of secretagogues. The adverse effect of lipotoxicity on beta-cell viability was prevented by concomitant exposure to 6.25 µM rosiglitazone. Culture with 6.25 µM rosiglitazone further decreased acute insulin release under glucotoxic conditions. However, when insulin secretion was expressed as percentage cellular insulin content, rosiglitazone (6.25 µM) significantly improved many of the adverse effects of gluco- and lipotoxic conditions on insulin secretory responsiveness. The results suggest that despite decrease in cellular insulin content TZDs exert direct beneficial effects on beta-cell viability and function during gluco- or lipotoxicity.

Bromocriptine: old drug, new formulation and new indication.

Bromocriptine is an ergot alkaloid dopamine D(2) receptor agonist that has been used extensively in the past to treat hyperprolactinaemia, galactorrhoea and Parkinsonism. It is known that hypothalamic hypodopaminergic states and disturbed circadian rhythm are associated with the development of insulin resistance, obesity and diabetes in animals and humans. When administered in the early morning at the start of the light phase, a new quick release (QR) formulation of bromocriptine appears to act centrally to reset circadian rhythms of hypothalamic dopamine and serotonin and improve insulin resistance and other metabolic abnormalities. Phase II and III clinical studies show that QR-bromocriptine lowers glycated haemoglobin by 0.6-1.2% (7-13 mmol/mol) either as monotherapy or in combination with other antidiabetes medications. Apart from nausea, the drug is well tolerated. The doses used to treat diabetes (up to 4.8 mg daily) are much lower than those used to treat Parkinson's disease and have not been associated with retroperitoneal fibrosis or heart valve abnormalities. QR-bromocriptine (Cycloset™) has recently been approved in the USA for the treatment of type 2 diabetes mellitus (T2DM). Thus, a QR formulation of bromocriptine timed for peak delivery in the early morning may provide a novel neurally mediated approach to the control of hyperglycaemia in T2DM.

Effects of metformin on BRIN-BD11 beta-cell insulin secretory desensitization induced by prolonged exposure to sulphonylureas.

AIMS: Prolonged exposure of pancreatic beta-cells in vitro to the sulphonylureas tolbutamide and glibenclamide induces subsequent desensitization of insulinotropic pathways. Clinically, the insulin-sensitizing biguanide drug metformin is often administered alongside sulphonylurea as antidiabetic therapy. The present study examines the functional effects of metformin (200 µM) on tolbutamide- and glibenclamide-induced desensitisation. METHODS: Acute and prolonged (18 h) effects of exposure to tolbutamide and glibenclamide alone, or in the presence of metformin, were examined in insulin-secreting BRIN-BD11 cells. RESULTS: In acute 20 min incubations at 1.1 mM glucose, metformin increased (1.2-1.7-fold; p < 0.001) the insulin-releasing actions of tolbutamide and glibenclamide. At 16.7 mM glucose, metformin significantly enhanced glibenclamide-induced insulin release at all concentrations (50-400 µM) examined, but tolbutamide-stimulated insulin secretion was only augmented at higher concentrations (300-400 µM). Exposure for 18 h to 100 µM tolbutamide or glibenclamide significantly impaired insulin release in response to glucose and a broad range of insulin secretagogues. Concomitant culture with metformin (200 µM) prevented or partially reversed many of the adverse effects on K(ATP) channel dependent and independent insulinotropic pathways. Beneficial effects of metformin were also observed in cells exposed to glibenclamide for 18 h with significant improvements in the insulin secretory responsiveness to alanine, GLP-1 and sulphonylureas. The decrease of viable cell numbers observed with glibenclamide was reversed by co-culture with metformin, but cellular insulin content was depressed. CONCLUSIONS: The results suggest that metformin can prevent the aspects of sulphonylurea-induced beta-cell desensitization.

Tailoring treatment to the individual in type 2 diabetes practical guidance from the Global Partnership for Effective Diabetes Management.

Good glycaemic control continues to be the most effective therapeutic manoeuvre to reduce the risk of development and/or progression of microvascular disease, and therefore remains the cornerstone of diabetes management despite recent scepticism about tight glucose control strategies. The impact on macrovascular complications is still a matter of debate, and so glycaemic control strategies should be placed in the context of multifactorial intervention to address all cardiovascular risk factors. Approaches to achieve glycaemic targets should always ensure patient safety, and results from recent landmark outcome studies support the need for appropriate individualisation of glycaemic targets and of the means to achieve these targets, with the ultimate aim to optimise outcomes and minimise adverse events, such as hypoglycaemia and marked weight gain. The primary goal of the Global Partnership for Effective Diabetes Management is the provision of practical guidance to improve patient outcomes and, in this article, we aim to support healthcare professionals in appropriately tailoring type 2 diabetes treatment to the individual. Patient groups requiring special consideration are identified, including newly diagnosed individuals with type 2 diabetes but no complications, individuals with a history of inadequate glycaemic control, those with a history of cardiovascular disease, children and individuals at risk of hypoglycaemia. Practical guidance specific to each group is provided.

Orotic aciduria and uridine monophosphate synthase: a reappraisal.

Three subtypes of hereditary orotic aciduria are described in the literature, all related to deficiencies in uridine monophosphate synthase, the multifunctional enzyme that contains both orotate: pyrophosphoryl transferase and orotidine monophosphate decarboxylase activities. The type of enzyme defect present in the subtypes has been re-examined by steady-state modelling of the relative outputs of the three enzymic products, uridine monophosphate, urinary orotic acid and urinary orotidine. It is shown that the ratio of urinary outputs of orotidine to orotate provides a means of testing for particular forms of enzyme defect. It is confirmed that the type I defect is caused by loss of uridine monophosphate synthase activity. Cells and tissue of type I cases have a residual amount of activity that is qualitatively unchanged: the relative rates of the transferase and decarboxylase do not differ from those of wild-type enzyme. The single claimed case of type II, thought to be due to specific inactivation of orotidine monophosphate decarboxylase, is shown to have a product spectrum inconsistent with that claim. It is proposed that this type II form does not differ sufficiently to be accepted as separate from type I. The third subtype, hereditary orotic aciduria without megaloblastic anaemia, occurs in two cases. It has the product spectrum expected of a defect in orotidine monophosphate decarboxylase. This form is the only one that appears to have a qualitatively different uridine monophosphate synthase. The possibility that orotidine monophosphate may control flux through the pyrimidine biosynthesis pathway in hereditary orotic aciduria is discussed.

Recent advances in antidiabetic drug therapies targeting the enteroinsular axis.

The enteroinsular axis (EIA) constitutes a physiological signalling system whereby intestinal endocrine cells secrete incretin hormones following feeding that potentiate insulin secretion and contribute to the regulation of blood glucose homeostasis. The two key hormones responsible are named glucagon-like peptide-1 (GLP-1) and glucose-dependent insulinotropic polypeptide (GIP). Recent years have witnessed sustained development of antidiabetic therapies that exploit the EIA. Current clinical compounds divide neatly into two classes. One concerns analogues or mimetics of GLP-1, such as exenatide (Byetta) or liraglutide (NN2211). The other group comprises the gliptins (e.g. sitagliptin and vildagliptin) which boost endogenous incretin activity by inhibiting the enzyme dipeptidyl peptidase 4 (DPP 4) that degrades both GLP-1 and GIP. Ongoing research indicates that further incretin and gliptin compounds will become available for clinical use in the near future, offering comparable or improved efficacy. For incretin analogues there is the prospect of prolonged duration of action and alternative routes of administration. This review focuses on recent advances in pre-clinical research and their translation into clinical studies to provide future therapies for type 2 diabetes targeting the EIA.

Sustained insulin secretory response in human islets co-cultured with pancreatic duct-derived epithelial cells within a rotational cell culture system.

AIMS/HYPOTHESIS: Loss of the trophic support provided by surrounding non-endocrine pancreatic cell populations underlies the decline in beta cell mass and insulin secretory function observed in human islets following isolation and culture. This study sought to determine whether restoration of regulatory influences mediated by ductal epithelial cells promotes sustained beta cell function in vitro. METHODS: Human islets were isolated according to existing protocols. Ductal epithelial cells were harvested from the exocrine tissue remaining after islet isolation, expanded in monolayer culture and characterised using fluorescence immunocytochemistry. The two cell types were co-cultured under conventional static culture conditions or within a rotational cell culture system. The effect of co-culture on islet structural integrity, beta cell mass and insulin secretory capacity was observed for 10 days following isolation. RESULTS: Human islets maintained under conventional culture conditions exhibited a characteristic loss in structural integrity and functional viability as indicated by a diminution of glucose responsiveness. By contrast, co-culture of islets with ductal epithelial cells led to preserved islet morphology and sustained beta cell function, most evident in co-cultures held within the rotational cell culture system, which showed a significantly (p < 0.05) greater insulin secretory response to elevated glucose compared with control islets. Similarly, insulin/protein ratio data suggested that the presence of ductal epithelial cells is beneficial for the maintenance of beta cell mass. CONCLUSIONS/INTERPRETATION: The data indicate a supportive role for ductal epithelial cells in islet viability. Further characterisation of the regulatory influences may lead to novel strategies to improve long-term beta cell function both in vitro and following islet transplantation.

Fixed-dose single tablet antidiabetic combinations.

Combinations of two or more oral agents with different mechanisms of action are often used for the management of hyperglycaemia in type 2 diabetes. While these combinations have customarily been taken as separate tablets, several fixed-dose single tablet combinations are now available. These are based on bioequivalence with the separate tablets, giving similar efficacy to the separate tablets and necessitating the same cautions and contraindications that apply to each active component. Fixed-dose combinations can offer convenience, reduce the pill burden and simplify administration regimens for the patient. They increase patient adherence compared with equivalent combinations of separate tablets, and this is associated with some improvements in glycaemic control. Presently available antidiabetic fixed-dose combinations include metformin combined with a sulphonylurea, thiazolidinedione, dipeptidylpeptidase-4 inhibitor or meglitinide as well as thiazolidinedione-sulphonylurea combinations, each at a range of dosage strengths to facilitate titration. Anticipated future expansion of multiple drug regimens for diabetes management is likely to increase the use of fixed-dose single tablet combinations.

Partnering with governments and other institutions: driving change in diabetes care.

The human and material cost of type 2 diabetes is a cause of increasing concern for health professionals, representative organisations and governments worldwide. The scale of morbidity and mortality has led the United Nations to issue a resolution on diabetes, calling for national policies for prevention, treatment and care. There is clearly an urgent need for a concerted response from all interested parties at the community, national and international level to work towards the goals of the resolution and create effective, sustainable treatment models, care systems and prevention strategies. Action requires both a 'bottom-up' approach of public awareness campaigns and pressure from healthcare professionals, coupled with a 'top-down' drive for change, via partnerships with governments, third sector (non-governmental) organisations and other institutions. In this review, we examine how existing collaborative initiatives serve as examples for those seeking to implement change in health policy and practice in the quest to alleviate the health and economic burden of diabetes. Efforts are underway to provide continuous and comprehensive care models for those who already have type 2 diabetes; in some cases, national plans extend to prevention strategies in attempts to improve overall public health. In the spirit of partnership, collaborations with governments that incorporate sustainability, long-term goals and a holistic approach continue to be a driving force for change. It is now critical to maintain this momentum and use the growing body of compelling evidence to educate, inform and deliver a long-term, lasting impact on patient and public health worldwide.

Geometric implications of the tibial wedge osteotomy for the treatment of cranial cruciate ligament disease in dogs.

The tibial wedge osteotomy (TWO), a procedure that reduces the tibial plateau slope, has become an established surgical technique for the treatment of cranial cruciate ligament injuries in dogs, yet variation from the desired postoperative tibial plateau slope of 6 degrees has been noted. The objectives of this study were to investigate the geometric implications of this procedure and to identify factors that affect the postoperative angles. The records of 35 consecutive cases that had a TWO performed were reviewed and comparisons were made between the alignment of the cortices and level of the osteotomy using duplicated tracings of a preoperative radiograph. The existing method for calculating the size of wedge to be removed was found to result in a postoperative slope greater than the expected 6 degrees. Our results indicate that the desired postoperative angle is more likely to be achieved if the cranial cortices are aligned and the osteotomy is performed proximally.

A double-blind randomized study comparing the effects of continuing or not continuing rosiglitazone + metformin therapy when starting insulin therapy in people with Type 2 diabetes.

AIMS: To compare the efficacy and safety of either continuing or discontinuing rosiglitazone + metformin fixed-dose combination when starting insulin therapy in people with Type 2 diabetes inadequately controlled on oral therapy. METHODS: In this 24-week double-blind study, 324 individuals with Type 2 diabetes inadequately controlled on maximum dose rosiglitazone + metformin therapy were randomly assigned to twice-daily premix insulin therapy (target pre-breakfast and pre-evening meal glucose < or = 6.5 mmol/l) in addition to either rosiglitazone + metformin (8/2000 mg) or placebo. RESULTS: Insulin dose at week 24 was significantly lower with rosiglitazone + metformin (33.5 +/- 1.5 U/day, mean +/- se) compared with placebo [59.0 +/- 3.0 U/day; model-adjusted difference -26.6 (95% CI -37.7, -15,5) U/day, P < 0.001]. Despite this, there was greater improvement in glycaemic control [HbA(1c) rosiglitazone + metformin vs. placebo 6.8 +/- 0.1 vs. 7.5 +/- 0.1%; difference -0.7 (-0.8, -0.5)%, P < 0.001] and more individuals achieved glycaemic targets (HbA(1c) < 7.0% 70 vs. 34%, P < 0.001). The proportion of individuals reporting at least one hypoglycaemic event during the last 12 weeks of treatment was similar in the two groups (rosiglitazone + metformin vs. placebo 25 vs. 27%). People receiving rosiglitazone + metformin in addition to insulin reported greater treatment satisfaction than those receiving insulin alone. Both treatment regimens were well tolerated but more participants had oedema [12 (7%) vs. 4 (3%)] and there was more weight gain [3.7 vs. 2.6 kg; difference 1.1 (0.2, 2.1) kg, P = 0.02] with rosiglitazone + metformin. CONCLUSIONS: Addition of insulin to rosiglitazone + metformin enabled more people to reach glycaemic targets with less insulin, and was generally well tolerated.