Johns Hopkins Diabetes Guide

Key Studies in Diabetes Care: Efficacy of Therapies

Thomas Donner, M.D.; Rita Rastogi Kalyani, M.D., M.H.S.; Christopher Saudek, M.D.

Medical therapies for type 1 and 2 diabetes

  • Metformin: UKPDS 34[1] studied the effect of intensive glycemic control with metformin in newly diagnosed, overweight people with type 2 diabetes. 1,704 patients who were hyperglycemic after 3 months of diet alone were randomly continued on "conventional" management mainly with diet alone, or metformin or sulfonylureas or insulin. Metformin reduced any diabetes-related endpoint (p=0.0034), all-cause mortality (p=0.021) and stroke (p=0.032). In an unexpected and unexplained result, early addition of metformin in the sulfonylurea treated group increased diabetes-related deaths (p=.039), although there was no overall association of metformin plus sulfonylurea with increased death risk. Conclusion was that metformin may be the first-line therapy of choice[1].
  • Metformin: The Multicenter Metformin Study Group[2] Randomized trial comparing effect of metformin versus placebo in 289 patients with type 2 diabetes who were moderately obese. After 29 weeks, the metformin group had lower fasting plasma glucose (189 mg/dl versus 244 mg/dl) and hemoglobin A1c (7.1% versus 8.6%) compared to placebo. Combination therapy with glyburide compared to glyburide alone resulted in lower fasting plasma glucose (187 mg/dl versus 261 mg/dl) and HbA1c (7.1% vs 8.6%). However, 18% had hypoglycemia in the combination group, compared to only 2% with metformin alone and 3% with glyburide alone[2].
  • Sulfonylureas: The University Group Diabetes Program[3]An early and much-debated study of new-onset type 2 diabetes, the UGDP randomized to treatment with a first-generation sulfonylurea (SU) (tolbutamide), phenformin or insulin. Found deaths from lactic acidosis in the phenformin group, and slightly but significantly more deaths from cardiovascular disease in the SU group. Serious critiques were leveled regarding methods and conclusions, for instance, by A.R. Feinstein[4]. The UGDP finding of adverse effects of SU have not been confirmed or widely accepted, but continue to prompt a black box warning in the package insert. SUs have been used safely in myriad studies since Meinert[3].
  • Glyburide[5]: Randomized 31 patients with type 2 diabetes previously inadequately controlled on diet alone to once daily NPH insulin or glyburide. Baseline HbA1c ~10% in both groups. After 9 months, HbA1c dropped similarly in both groups by ~3%.
  • Acarbose[6]: Randomized 96 patients with type 2 diabetes and dietary failure to acarbose versus glibenclamide versus placebo. After 24 weeks, mean HbA1c dropped by 1.1% with acarbose and 0.9% with glibenclamide compared to placebo. Acarbose also lower postprandial insulin increase[6].
  • Troglitazone[7]: An early description of how thiazolidinediones (TZDs) (troglitazone in this case) works by reducing insulin resistance in 18 obese patients with either normal or impaired glucose tolerance. While troglitazone is no longer available, this mechanism is a class effect for TZDs.
  • Pioglitazone[8]: Randomized 408 patients to placebo or four different doses of pioglitazone monotherapy. HbA1c decreased on average between 1 - 1.6% on the three highest doses of pioglitazone (15 - 45 mg daily) after 26 weeks compared to placebo. Improvements in fasting glucose were observed after two weeks of therapy, and were maximal at 10-14 weeks but maintained until study completion (-39 to -65 mg/dl versus placebo). The improvement in glycemic control was greatest in those who were treatment naive (HbA1c difference from placebo of -2.55%)[8].
  • Pioglitazone[9]: In the PROActive trial, 5,238 patients with type 2 diabetes and evidence of macrovascular disease were randomized to pioglitazone versus placebo. After an average follow-up of 34.5 months, the primary composite endpoint of all-cause mortality, non-fatal myocardial infarction, acute coronary syndrome, endovascular or surgical intervention in the coronary or leg arteries, and amputation above the ankles was not significantly different between groups. However, the composite secondary endpoint of all-cause mortality, non-fatal myocardial infarction and stroke was 16% lower in the pioglitazone group (p=0.027)[9].
  • Pioglitazone[10]: The PERISCOPE Trial performed coronary intravascular ultrasonography (IVUS) on 543 patients with coronary disease and type 2 diabetes. Patients were then randomized to receive glimepiride (a sulfonylurea) or pioglitazone (a thiazolidinedione) followed by repeat IVUS in 360 patients after 18 months. Percent atheroma volume increased 0.73% (95% CI, 0.33% to 1.12%) with glimepiride and decreased 0.16% (95% CI, -0.57% to 0.25%) with pioglitazone (P = .002). Using this extremely fine measurement, the conclusion was that pioglitazone was more favorable than SU in slowing progression of coronary atherosclerosis[10].
  • Pioglitazone versus Rosiglitazone Effects on Lipids[11]: Many of these studies are company-sponsored. In this, subjects with type 2 diabetes and dyslipidemia, not previously on insulin or lipid lowering-agents, were treated with pioglitazone (n=400) or rosiglitazone (n=402) for 12 weeks. Triglyceride levels fell by 51 ± 7.8 mg/dl with pioglitazone, but increased by 13.1 ± 7.8 mg/dl with rosiglitazone (P < 0.001 between treatments). Pioglitazone also increased HDL cholesterol (5.2 ± 0.5 vs. 2.4 ± 0.5 mg/dl; P < 0.001) and increased LDL cholesterol less (12 ± 1.6 vs. 21 ± 1.6 mg/dl; P < 0.001). LDL particle concentration was reduced and LDL particle size was increased more with pioglitazone (p=0.005), both considered favorable changes, suggesting that pioglitazone and rosiglitazone have significantly different effects on plasma lipids (pioglitazone being more favorable).
  • Rosiglitazone[12]: A meta-analysis of 42 published and unpublished, small and large studies, which found that use of rosiglitazone was associated with a marginally significant increase in deaths from myocardial infarction (odds ratio=1.43, 95% CI 1.03-1.98; p=0.03) and an increase in death from cardiovascular disease that had borderline significance (odds ratio=1.64, 95% CI 0.98 - 2.74; p=0.06). Immediately controversial.
  • Rosiglitazone[13]: An updated meta-analysis of 56 trials of rosiglitazone at least 24 weeks in duration found that rosiglitazone significantly increased risk of myocardial infarction (odds ratio = 1.28, 95%Y. CI 1.02-1.63, p=0.04) but not cardiovascular mortality (odds ratio=1.03, 95%Y. CI 0.78-1.36, p=0.86b).
  • Rosiglitazone[14]: RECORD is a large (n=4,447 patients) randomized trial of the cardiovascular events following addition of rosiglitazone to metformin or sulfonylurea therapy in type 2 diabetes. After 5.5 years, the study found that rosiglitazone treatment was associated with a significant increase in body weight (~4kg), an increased risk of heart failure (hazard ratio 2.10, 95% CI 1.35 - 3.27) and overall bone fractures (relative risk 1.57, 95% CI 1.26-1.97), mainly among women and primarily in upper and lower limb fractures. However, no overall increase in cardiovascular morbidity or mortality was found in this study[14].
  • Comparative Efficacy[15]: The ADOPT study (n=4,360) evaluated longevity of glycemic control with several oral agents in type 2 diabetes over 5 years. Monotherapy failure was defined as fasting plasma glucose of >180 mg/dl. Glyburide had the most significant early hypoglycemic effect but the highest failure rate (34%) at 5 years. 15% failed rosiglitazone, and 21% failed metformin. Rosiglitazone was associated with more weight gain and edema, and an increased rate of new fractures[15].
  • Exenetide versus Insulin Glargine[16]:An early open-label comparison of exenatide (an incretin mimetic) versus insulin glargine in 138 persons with type 2 diabetes who had failed oral agents. Found that exenatide had similar glycemic efficacy after 16 weeks (~38% had HbA1c< 7%), but exenatide had more weight reduction (difference of -2.2 kg between groups, p< 0.001), as well as a higher incidence of gastrointestinal side effects (42.6% versus 3.1% with nausea).
  • Inhaled Insulin[17]: A safety and efficacy trial of inhaled insulin in 580 adults with type 1 diabetes, showing efficacy and safety over 2 years, although a small decrease in FEV1 during the first 3 months that was not progressive thereafter and cough was noted. Inhaled insulin was marketed briefly and then removed from the market, because it was not well accepted, and did not sell. Subsequently, the FDA announced a worrisome finding: while there were few deaths among study participants, there was an imbalance of deaths due to lung cancer in those taking inhaled insulin.
  • Sitagliptin (Study 021 Group)[18]: A safety and efficacy trial of sitagliptin as monotherapy in type 2 diabetes. Randomized 741 patients to sitagliptin (100 or 200 mg) or placebo. After 24 weeks, overall significant reductions in A1c in sitagliptin group (100 or 200 mg) of -0.79 and -0.94%, respectively compared to placebo. A1c lowering effects greatest when baseline A1c >9% (~-1.50%). Sitagliptin was weight neutral, with similar incidence of hypoglycemia and slightly higher GI side effects compared to placebo[18].
  • Canagliflozin[19]: A trial of 755 subjects with type 2 diabetes uncontrolled on metformin and a sulfonylurea. Subjects were randomized to canagliflozin 300 mg daily or sitagliptin 100 mg daily. After 52 weeks, the change in HbA1c was -1.03% with canagliflozin vs -0.66% with sitagliptin. Body weight change was -2.3 kg with canagliflozin vs +0.1 kg with sitagliptin. An increased number of mycotic genital infections and osmotic diuresis-related side effects were seen with canagliflozin use.

Medical therapies for gestational diabetes[Top]

  • Glyburide[20]: Randomized 404 women with singleton pregnancy and gestational diabetes requiring treatment to receive glyburide or insulin between 11 - 33 weeks gestation. Mean concentrations of glucose before treatment ~ 115 mg/dL and after treatment ~105 mg/dL in both groups. No significant differences in macrosomia, hypoglycemia, lung complications, fetal anomalies, or requirement for neonatal intensive care between infants in both groups. Suggested that glyburide could be a clinically effective alternative to insulin in management of gestational diabetes. However, later studies suggested that although glyburide may be as effective as insulin, it may be associated with increased incidence of preeclampsia and phototherapy, and these risks require further study[21].
  • Metformin[22]: Randomized 149 patients with gestational diabetes inadequately controlled on diet to metformin versus glyburide. Failure to achieve adequate glycemic control was significantly higher in metformin group (35%) versus glyburide group (16%), requiring insulin.

Surgical therapies for diabetes and related complications[Top]

  • Islet transplantation[23]: The first demonstration that islet transplantation is feasible. Seven patients with type 1 diabetes who had history of severe metabolic instability and hypoglycemia were transplanted with 2-3 donor pancreases, and managed with a steroid-free regimen of sirolimus, tacrolimus and daclizumab. After 1 year, insulin independence was achieved in all 7 patients, though longer term follow-up showed some deterioration. The positive results on insulin independence were thought to be due to advances in the islet isolation techniques and the steroid-free immunosuppressive regimen.
  • Islet transplantation[24]: An expanded, international trial of the Edmonton Protocol for islet transplantation in 36 patients with type 1 diabetes. After 1 year, 16 patients (44%) had insulin independence and adequate glycemic control, 10 patients (28%) had partial function, and 10 patients (28%) had complete graft loss. At 2 years, 5 of the 16 subjects (31%) who were free of insulin at 1 year remained insulin independent. The authors concluded that insulin independence after islet transplantation is possible but not usually sustained; however, protection from severe hypoglycemia and improved HbA1c are benefits of persistent islet function[24].
  • Implantable insulin pump[25]: Randomized 171 male patients with type 2 diabetes on insulin therapy to implantable insulin pump (IIP) versus multiple daily insulin injections (MDI). After 1 year, similar decline in blood glucose levels (~8 mmol/L) and hemoglobin A1c was observed. IIP had significantly reduced glucose fluctuations, mild clinical hypoglycemia and weight gain, while improving quality of life. However, 25% of patients had insulin under delivery due insulin microprecipitates within the implantable insulin pump[25].
  • Gastric banding[26]: One of the few randomized trials of laparoscopic adjustable gastric banding bariatric surgery for type 2 diabetes. Of 60 obese patients enrolled, 73% in the surgery group had remission of type 2 diabetes versus only 13% in the lifestyle modification group (relative risk 5.5, 95% CI 2.2-14.0). The benefit was directly related to weight loss; in the surgical group, mean weight was reduced by 20.7% versus 1.7% in lifestyle modification group[26].
  • Bariatric Surgery[27]: The LABS consortium report evaluated safety of bariatric surgery in 4,776 patients, about three-quarters with Roux-en-Y gastric bypass, the rest with adjustable gastric banding. Overall, 4.3% had at least one major adverse outcome, and 30 day death rate was 0.3%. Extreme obesity was associated with an increased risk of adverse outcomes[27].
  • Bariatric surgery[28]: Long-term mortality after the most common bariatric surgery procedure, gastric bypass (n=9,949), compared to a retrospectively evaluated control group of obese patients (n=9,628), was reduced by 40% (p< 0.001) particularly deaths from diabetes, heart disease, and cancer after 7 years. However, the rate of deaths from accidents and suicides were 58% higher (p=0.04) in the surgery versus control group.
  • Bariatric Surgery[29]:The SOS study followed 4,047 obese subjects for up to 11 years of whom 2,010 underwent bariatric surgery and 2,037 received conventional treatment. Weight loss was +/-2% in the conventional group for up to 15 years compared to the surgical group after 1-2 years and 10 years, respectively, as follows gastric bypass (32%, 25%), vertical-banded gastroplasty (25%, 16%), banding (20%, 14%). The sex, age, and risk factor adjusted mortality rate was 29% lower in the surgery- versus conventionally-treated group (p=0.01)[29].
  • Cardiac Bypass Surgery[30]: The BARI trial compared the efficacy of coronary-artery bypass grafting (CABG) with percutaneous transluminal coronary angioplasty (PTCA) in 1,829 patients with multivessel disease. There was no significant difference in mortality between the two groups after an average follow-up of 5.4 years. However, when the analysis was restricted to persons with diabetes (n=353), a subgroup that had not been prespecified, survival was significantly better in the CABG compared to the PTCA group (80.6% versus 65.5%, p=0.003). The study put into question the use of PTCA in people with diabetes, with the caveat that PTCA methods are constantly improving[30].
  • Cardiac Bypass Surgery[31]:The BARI 2D study enrolled 2,368 people with type 2 diabetes and stable ischemic heart disease. They were randomly assigned to either prompt intervention with revascularization (CABG) or percutaneous coronary intervention (PCI), versus optimization of medical therapy. The type of prompt intervention was selected according to severity of the disease. With an average 5 year follow-up, optimal medical intervention resulted in no difference in deaths or major cardiovascular events compared to prompt revascularization, unless the patient was selected for CABG (had more severe disease), in which case CABG had significantly lower rate of major cardiovascular events than medical management (22.4% versus 30.5%, p=0.01)[31].


  • Only the relative efficacy of oral and non-insulin injectable therapies were included in this list.
  • Trials of drug efficacy may be subject to publication bias; however, several prominent medical journals now require that clinical trials be registered in a public database from the start, reducing the possibility that only positive results will get published.
  • Relatively greater glucose-lowering effects may be observed when HbA1c is higher (i.e. >9%) at baseline or in treatment naive patients.
  • The rosiglitazone versus pioglitazone controversy has been particularly active. For perspective, there are clear, known, side effects of both these thiazolidinediones: they each cause fluid retention, an increased incidence of congestive heart failure; weight gain; and increased fracture rate. One head-to-head study[11] (above) found a significantly better lipid profile for pioglitazone. The controversy is about whether rosiglitazone also causes more CVD than does pioglitazone. The evidence is mainly from the Nissen meta-analysis (above). Most of the large studies such as RECORD (above) have not found a difference in CVD. A panel was convened by the FDA and announced in September 2010 that rosiglitazone will stay on the market, but restricted its use.
  • Surgical therapies are primarily available for complications of diabetes (e.g. obesity, cardiovascular disease) but novel treatments such as islet transplantation for type 1 diabetes are being investigated as possibilities for curative treatment.


  1. Effect of intensive blood-glucose control with metformin on complications in overweight patients with type 2 diabetes (UKPDS 34). UK Prospective Diabetes Study (UKPDS) Group. Lancet 352:854, 1998  [PMID:9742977]

    Comment: UKPDS Group Study

  2. DeFronzo RA, Goodman AM: Efficacy of metformin in patients with non-insulin-dependent diabetes mellitus. The Multicenter Metformin Study Group. N Engl J Med 333:541, 1995  [PMID:7623902]

    Comment: The Multicenter Metformin Study Group

  3. Meinert CL et al: A study of the effects of hypoglycemic agents on vascular complications in patients with adult-onset diabetes. II. Mortality results. Diabetes 19:Suppl:789, 1970  [PMID:4926376]

    Comment: The University Group Diabetes Program

  4. Feinstein AR: Clinical biostatistics. 8. An analytic appraisal of the University Group Diabetes Program (UGDP) study. Clin Pharmacol Ther 12:167, 1971 Mar-Apr  [PMID:4324447]
  5. Nathan DM, Roussell A, Godine JE: Glyburide or insulin for metabolic control in non-insulin-dependent diabetes mellitus. A randomized, double-blind study. Ann Intern Med 108:334, 1988  [PMID:3124685]

    Comment: Efficacy of glyburide on glycemic control

  6. Hoffmann J, Spengler M: Efficacy of 24-week monotherapy with acarbose, glibenclamide, or placebo in NIDDM patients. The Essen Study. Diabetes Care 17:561, 1994  [PMID:8082525]

    Comment: Essen Study

  7. Nolan JJ et al: Improvement in glucose tolerance and insulin resistance in obese subjects treated with troglitazone. N Engl J Med 331:1188, 1994  [PMID:7935656]

    Comment: Troglitazone Benefits on Insulin Resistance

  8. Aronoff S et al: Pioglitazone hydrochloride monotherapy improves glycemic control in the treatment of patients with type 2 diabetes: a 6-month randomized placebo-controlled dose-response study. The Pioglitazone 001 Study Group. Diabetes Care 23:1605, 2000  [PMID:11092281]

    Comment: Pioglitazone 001 Study Group

  9. Dormandy JA et al: Secondary prevention of macrovascular events in patients with type 2 diabetes in the PROactive Study (PROspective pioglitAzone Clinical Trial In macroVascular Events): a randomised controlled trial. Lancet 366:1279, 2005  [PMID:16214598]

    Comment: PROactive Study

  10. Nissen SE et al: Comparison of pioglitazone vs glimepiride on progression of coronary atherosclerosis in patients with type 2 diabetes: the PERISCOPE randomized controlled trial. JAMA 299:1561, 2008  [PMID:18378631]

    Comment: PERISCOPE Study

  11. Goldberg RB et al: A comparison of lipid and glycemic effects of pioglitazone and rosiglitazone in patients with type 2 diabetes and dyslipidemia. Diabetes Care 28:1547, 2005  [PMID:15983299]

    Comment: Comparative effects of pioglitazone versus rosiglitazone on lipids

  12. Nissen SE, Wolski K: Effect of rosiglitazone on the risk of myocardial infarction and death from cardiovascular causes. N Engl J Med 356:2457, 2007  [PMID:17517853]

    Comment: Meta-analysis of rosiglitazone on cardiovascular outcomes

  13. Nissen SE, Wolski K: Rosiglitazone Revisited: An Updated Meta-analysis of Risk for Myocardial Infarction and Cardiovascular Mortality. Arch Intern Med Jun 28  [PMID:20656674]

    Comment: Updated meta-analysis of 52 trials.

  14. Home PD et al: Rosiglitazone evaluated for cardiovascular outcomes in oral agent combination therapy for type 2 diabetes (RECORD): a multicentre, randomised, open-label trial. Lancet 373:2125, 2009  [PMID:19501900]

    Comment: RECORD Study

  15. Kahn SE et al: Glycemic durability of rosiglitazone, metformin, or glyburide monotherapy. N Engl J Med 355:2427, 2006  [PMID:17145742]

    Comment: ADOPT

  16. Barnett AH et al: Tolerability and efficacy of exenatide and titrated insulin glargine in adult patients with type 2 diabetes previously uncontrolled with metformin or a sulfonylurea: a multinational, randomized, open-label, two-period, crossover noninferiority trial. Clin Ther 29:2333, 2007  [PMID:18158075]

    Comment: Exenetide versus Insulin Glargine

  17. Skyler JS et al: Two-year safety and efficacy of inhaled human insulin (Exubera) in adult patients with type 1 diabetes. Diabetes Care 30:579, 2007  [PMID:17327324]

    Comment: Inhaled Insulin

  18. Aschner P et al: Effect of the dipeptidyl peptidase-4 inhibitor sitagliptin as monotherapy on glycemic control in patients with type 2 diabetes. Diabetes Care 29:2632, 2006  [PMID:17130196]

    Comment: Sitagliptin

  19. Schernthaner G et al: Canagliflozin Compared With Sitagliptin for Patients With Type 2 Diabetes Who Do Not Have Adequate Glycemic Control With Metformin Plus Sulfonylurea: A 52-week randomized trial. Diabetes Care Apr 5  [PMID:23564919]
  20. Langer O et al: A comparison of glyburide and insulin in women with gestational diabetes mellitus. N Engl J Med 343:1134, 2000  [PMID:11036118]

    Comment: Glyburide for treatment of gestational diabetes

  21. Jacobson GF et al: Comparison of glyburide and insulin for the management of gestational diabetes in a large managed care organization. Am J Obstet Gynecol 193:118, 2005  [PMID:16021069]

    Comment: Glyburide for treatment of gestational diabetes

  22. Moore LE et al: Metformin compared with glyburide in gestational diabetes: a randomized controlled trial. Obstet Gynecol 115:55, 2010  [PMID:20027034]

    Comment: Metformin versus glyburide for treatment of gestational diabetes

  23. Shapiro AM et al: Islet transplantation in seven patients with type 1 diabetes mellitus using a glucocorticoid-free immunosuppressive regimen. N Engl J Med 343:230, 2000  [PMID:10911004]

    Comment: Islet transplantation

  24. Shapiro AM et al: International trial of the Edmonton protocol for islet transplantation. N Engl J Med 355:1318, 2006  [PMID:17005949]

    Comment: The Edmanton protocol for islet transplantation.

  25. Saudek CD et al: Implantable insulin pump vs multiple-dose insulin for non-insulin-dependent diabetes mellitus: a randomized clinical trial. Department of Veterans Affairs Implantable Insulin Pump Study Group. JAMA 276:1322, 1996 Oct 23-30  [PMID:8861991]

    Comment: Department of Veteran Affairs Implantable Insulin Pump Study Group

  26. Dixon JB et al: Adjustable gastric banding and conventional therapy for type 2 diabetes: a randomized controlled trial. JAMA 299:316, 2008  [PMID:18212316]

    Comment: University Obesity Research Center

  27. Longitudinal Assessment of Bariatric Surgery (LABS) Consortium et al: Perioperative safety in the longitudinal assessment of bariatric surgery. N Engl J Med 361:445, 2009  [PMID:19641201]

    Comment: LABS Consortium

  28. Adams TD et al: Long-term mortality after gastric bypass surgery. N Engl J Med 357:753, 2007  [PMID:17715409]

    Comment: Gastric bypass surgery long-term mortality

  29. Sjöström L et al: Effects of bariatric surgery on mortality in Swedish obese subjects. N Engl J Med 357:741, 2007  [PMID:17715408]

    Comment: Swedish Obesity Subjects study

  30. Comparison of coronary bypass surgery with angioplasty in patients with multivessel disease. The Bypass Angioplasty Revascularization Investigation (BARI) Investigators. N Engl J Med 335:217, 1996  [PMID:8657237]

    Comment: The Bypass Angioplasty Revascularization Investigation

  31. BARI 2D Study Group et al: A randomized trial of therapies for type 2 diabetes and coronary artery disease. N Engl J Med 360:2503, 2009  [PMID:19502645]

    Comment: BARI 2D Study Group

Key Studies in Diabetes Care: Efficacy of Therapies is a sample topic found in Johns Hopkins Guides.

To find other Johns Hopkins Guides topics please log in or purchase a subscription.

Content Manager
Related Content
Key Studies in Diabetes Care: Management of Complications
Key Studies in Diabetes Care: Prevention and Glycemic Control

more ...