From Wiki Journal Club
Jump to: navigation, search
UKPDS Study Group. "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.". The Lancet. 1998. 352(9131):854-865.
PubMedFull textPDF

Clinical Question

Among overweight patients with T2DM, does metformin reduce DM-related complications and all-cause mortality when compared to diet or other early antiglycemic agents?

Bottom Line

Among overweight patients with T2DM, metformin reduces the rate of DM-related complications and all-cause mortality when compared to diet alone or other early-generation antiglycemic agents.

Major Points

Biguanides have many theoretical benefits over other agents in the treatment of T2DM including reducing hepatic gluconeogenesis, decreasing plasma insulin levels, and facilitating weight loss. However, the biguanide phenformin was associated with increased CV and all-cause mortality in UGDP (1975)[1] and is no longer used in clinical practice. The related agent metformin subsequently became the subject of intense interest, but studies of metformin's effect on clinical endpoints were lacking.

The 1998 UK Prospective Diabetes Study 34 (UKPDS 34) randomized 1,704 overweight patients with newly diagnosed T2DM to one of three arms: conventional therapy with diet alone, intensive therapy with metformin, or intensive therapy with an early-generation antiglycemic agents (chlorpropamide, glibenclamide, or insulin). The primary analysis compared metformin to diet alone, with a secondary analysis comparing metformin to intensive therapy with the other agents. With a median follow-up of 10.7 years, metformin was associated with a reduction in DM-related complications and all-cause mortality when compared to the other two arms of therapy. These benefits persisted at an additional 10 years of follow-up.[2]

This study forms the basis for the early administration of metformin among overweight patients with T2DM. Since the publication of UKPDS 34, several generations of antiglycemic agents have been developed, and direct comparisons with metformin have been performed both prospectively and retrospectively. The large multicenter SPREAD-DIMCAD trial (2013)[3] randomized patients with T2DM and CAD to either metformin and glipizide for 3 years. At a median follow-up of 5 years, both groups achieved goal HbA1c levels (7.1% vs. 7.0%) but metformin was associated with a 12% absolute risk reduction in the composite primary outcome of nonfatal MI, nonfatal stroke, revascularization, CV mortality, or all-cause mortality. The benefits of metformin monotherapy in overweight patients with T2DM was further solidified by a 2005 Cochrane review of 29 randomized trials comparing metformin to conventional therapy with diet or other modern antiglycemic agents.[4] This meta-analysis demonstrated a benefit to metformin monotherapy across a diverse range of outcomes including glycemic control, weight loss, lipid and blood pressure control, diabetes-related mortality, all-cause mortality, and incidence of myocardial infarction.


ADA Diabetes Care (2013, adapted)[5]

  • Metformin as initial medication for hyperglycemia in T2DM (A)
  • Insulin therapy if markedly symptomatic hyperglycemia or significant elevation in HbA1c (E)
  • If non-insulin monotherapy at maximal dose does not achieve the HbA1c goal over 3-6 months, add a second non-insulin agent (the authors suggest GLP-1 agonist) or insulin (A)


  • Prospective, randomized, comparative trial
  • N=753
    • Diet alone (n=411)
    • Metformin plus diet (n=342)
  • Setting: 23 centers in the UK
  • Enrollment: 1977-1991
  • Median follow-up: 10.7 years
  • Analysis: Intention-to-treat
  • Primary outcomes: DM-specific endpoints, DM mortality, all-cause mortality


Inclusion Criteria

  • Age 25-65 years
  • Newly diagnosed T2DM
  • Fasting glucose >108 mg/dL on two mornings
  • Fasting glucose 108 mg/dL after a 3-month trial of dietary modifications
  • Overweight

Exclusion Criteria

  • Ketonuria >54 mg/dL
  • Creatinine >2 mg/dL
  • MI within previous year
  • Current angina
  • Current CHF
  • >1 major vascular event
  • Retinopathy requiring laser treatment
  • Malignant HTN
  • Uncorrected endocrine disorder
  • Occupation precluding insulin therapy
  • Severe concurrent life-limiting illness

Baseline Characteristics

From the metformin group.

  • Demographics: Age 53 years, male 46%
  • Health data: Weight 85 kg, BMI 31.6 kg/m2, BP 139/85 mmHg
  • Social history: Smoker 25% (former 32%), alcohol dependence 1.5%, sedentary 29%
  • Laboratory: Fasting blood glucose 146 mg/dL, HbgA1c 7.3%, plasma insulin 116 pmol/L
    • Lipids: TG 247 mg/dL, TC 217 mg/dL, LDL 141 mg/dL, HDL 41 mg/dL
  • Medications: Diuretic 17%, antihypertensives 15%, lipid lowering 0%, HRT or OCP 0.3%


  • All patients were advised to follow a diet low fat, high fiber, high carbohydrate diet, with calorie restriction.
  • Overweight patients were randomized to:
    • Metformin: started at 850mg daily, increased to 2550mg daily in divided doses, titrated to GI symptoms; sulfonylurea and then insulin added for persistent hyperglycemia
    • Intensive: treatment with chlorpropamide, glibenclamide, or insulin
    • Diet: dietary advice q3mo; hyperglycemic patients were randomized to metformin, another oral agent, or insulin; insulin added for persistent hyperglycemia

This study also compared add-on use of sulonylureas to metformin in non-overweight diabetic patients that is outside of the scope of this review.


Results are number per patient-years, unless otherwise specified.

Primary Outcomes

Diabetes-specific endpoints
Defined as sudden death, hyper- or hypoglycemia-related death, MI, angina, HF, stroke, renal failure, amputation, vitreous hemorrhage, retinopathy requiring photocoagulation, and blindness.
Metformin vs. intensive: 29.8 vs. 40.1 (P=0.0034)
Metformin vs. diet: 29.8 vs. 43.3 (P=0.0023)
Diabetes-specific mortality
Defined as death from MI, stroke, PVD, kidney disease, hypo- or hyperglycemia, or sudden death.
Metformin vs. intensive: 7.5 vs. 10.3 (P=0.11)
Metformin vs. diet: 7.5 vs. 12.7 (P=0.017)
All-cause mortality
Metformin vs. intensive: 13.5 vs. 18.9 (P=0.021)
Metformin vs. diet: 13.5 vs. 20.6 (P=0.011)

Secondary Outcomes

Metformin vs. intensive: 11.0 vs. 14.4 (P=0.12)
Metformin vs. diet: 11.0 vs. 18.0 (P=0.01)
Metformin vs. intensive: 3.3 vs. 6.2 (P=0.13)
Metformin vs. diet: 3.3 vs. 5.5 (P=0.13)
Metformin vs. intensive: 1.6 vs. 1.2 (P=0.62)
Metformin vs. diet: 1.6 vs. 2.1 (P=0.57)
Microvascular complications
Metformin vs. intensive: 6.7 vs. 7.7 (P=0.39)
Metformin vs. diet: 6.7 vs. 9.2 (P=0.19)

Additional Analyses

Major hypoglycemic events, per year
Metformin: 0%
Diet: 0.7%
Requiring DM medications, diet group only
HbA1c at 10 years
Metformin vs. diet: 7.4% vs. 8.0%


  • Small absolute risk reductions with intensive glucose therapies[6]


  • Public sources: UK Medical Research Council, British Diabetic Association, UK Department of Health, NIH, British Heart Foundation
  • Private sources: Novo-Nordisk, Bayer, Bristol Myers Squibb, Hoechst, Lilly, Lipha, Farmitalia Carlo Erba

Further Reading

  1. University Group Diabetes Program. "A study of the effects of hypoglycemic agents on vascular complications on patients with adult-onset diabetes: V- Evaluation of phenformin therapy." Diabetes. 1975;24(suppl 1):65-184.
  2. Holman RR, et al. "10-year follow-up of intensive glucose control in type 2 diabetes." The New England Journal of Medicine. 2008;15:1577-1589.
  3. Hong J, et al. "Effects of metformin versus glipizide on cardiovascular outcomes in patients with type 2 diabetes and coronary artery disease." Diabetes Care. 2013;36(5):1304-1311.
  4. Saenz A, et al. "Metformin monotherapy for type 2 diabetes mellitus." Cochrane Database of Systematic Reviews. 2005;(3):CD002966.
  5. Multiple authors. "Summary of the revisions for the 2013 clinical practice recommendations." Diabetes care. 2013;36 sup1:S3.
  6. Multiple authors. "Correspondence: The UK prospective diabetes study." The Lancet. 1998;352(9144):1932-1934.