Current issues of ACP Journal Club are published in Annals of Internal Medicine


Metformin reduced diabetes-related end points and all-cause mortality in overweight patients with type 2 diabetes

ACP J Club. 1999 Jan-Feb;130:3. doi:10.7326/ACPJC-1999-130-1-003

Source Citation

UK Prospective Diabetes Study Group. Effect of intensive blood-glucose control with metformin on complications in overweight patients with type 2 diabetes (UKPDS 34). Lancet. 1998 Sep 12;352:854-65.



In overweight patients with type 2 diabetes mellitus, what is the effect of intensive plasma glucose control with metformin on the risk for macrovascular and microvascular complications?


Randomized controlled trial with median follow-up of 10.7 years.


15 U.K. hospital-based diabetes clinics.


1704 patients (mean age 53 y, 54% women) who had body weight ≥ 120% of ideal, newly diagnosed type 2 diabetes, and fasting plasma glucose (FPG) levels between 6.1 mmol/L and 15.0 mmol/L after 3 months of dietary therapy. Follow-up was 97%.


Patients were allocated to dietary advice and intensive blood glucose control with metformin (n = 342), 1700 to 2550 mg/d; chlorpropamide (n = 265); glibenclamide (n = 277); or insulin (n = 409) aiming for FPG levels of < 6 mmol/L or to conventional dietary treatment (n = 411) aiming for normal body weight and FPG levels < 15 mmol/L.

Main outcome measures

A first diabetes-related predefined end point, deaths from diabetes, all-cause mortality, myocardial infarction (MI), stroke, peripheral vascular disease, and microvascular disease.

Main results

Analysis was by intention to treat. Median hemoglobin A1c values for 10 years of follow-up were 7.4% for patients receiving metformin and 8.0% for patients receiving conventional treatment. Compared with conventional treatment, metformin reduced diabetes-related end points (P = 0.002), diabetes-related deaths (P = 0.017), all-cause mortality (P = 0.011), and MI (P = 0.01) (Table). Patients receiving metformin had fewer diabetes-related end points (P = 0.003), less all-cause mortality (P = 0.021) (Table) or stroke (3.5% vs 6.3%, Kaplan-Meier P = 0.032), and fewer hypoglycemic episodes than those receiving other intensive treatments.


In overweight patients with type 2 diabetes mellitus, early use of metformin reduced the risk for diabetes-related end points and all-cause mortality.

Sources of funding: U.K. Medical Research Council; British Diabetic Association; U.K. Department of Health; National Institutes of Health; British Heart Foundation; 7 drug companies.

For correspondence: Professor R. Turner, U.K. Prospective Diabetes Study Group, Diabetes Research Laboratories, Radcliffe Infirmary, Oxford OX2 6HE, England, UK.

Table. Metformin vs conventional or other intensive treatments in overweight patients with type 2 diabetes mellitus over 10 years of follow-up*

Outcomes Metformin Conventional RRR (95% CI) NNT (CI)
Diabetes end points 29% 39% 26% (10 to 40) 10 (6 to 29)
Deaths from diabetes 8% 13% 39% (6 to 60) 19 (10 to 138)
All-cause mortality 15% 22% 32% (8 to 51) 14 (8 to 67)
Myocardial Infarction 11% 18% 36% (8 to 55) 16 (9 to 78)
Diabetes end points 29% 37% 22% (7 to 36) 12 (7 to 43)
All-cause mortality 15% 20% 27% (3 to 45) 19 (10 to 172)

*Abbreviations defined in Glossary; RRR, NNT, and CI calculated from data in article.


Intensive blood glucose control reduced type 2 diabetes mellitus-related end points

The Diabetes Control and Complications Trial (DCCT) clearly showed that tight glycemic control with intensified insulin therapy dramatically reduces microvascular disease in patients with type 1 diabetes (1). The small number of cardiovascular events in the DCCT precluded detection of any cardioprotective effects, despite a tantalizing trend. Although surrogate primary end points were studied, the effect size and consistency of the effect across different end points in this and other studies led to widespread acceptance of tight glucose control for prevention of microvascular complications in type 1 diabetes. Until now, the DCCT results were generalized to patients with type 2 diabetes, without strong direct evidence in this group.

The United Kingdom Prospective Diabetes Study (UKPDS) now provides direct evidence of a policy of intensive blood glucose control for patients with type 2 diabetes. In contrast to the DCCT, the UKPDS used "hard" clinical primary end points and a variety of interventions.

The UKPDS did not achieve stable glucose levels. The hemoglobin A1c level rose steadily in both groups. Furthermore, the 0.9% separation in median hemoglobin A1c throughout the first 10 years of the UKPDS study was considerably less than the 2% separation shown in the DCCT (1).

Despite achieving only modestly better glucose control, the UKPDS clearly showed that an intensive treatment policy reduced diabetes-related end points and microvascular disease. A strong trend in favor of reduced risk for MI was also shown (relative risk reduction = 16%, P = 0.052).

In patients who were overweight, the UKPDS showed that an intensive therapy policy beginning with metformin reduced diabetes-related end points and microvascular disease plus diabetes-related deaths, MI, and all-cause mortality. This approach was superior to the other intensive therapies. Results for the treatment with metformin need to be interpreted cautiously in light of a subsequent analysis showing that addition of metformin to therapy of both obese and nonobese patients with suboptimal control on sulfonylureas increased diabetes-related death and all-cause mortality.

In light of these results, what can we conclude? First, clear and consistent evidence now exists that hyperglycemia in diabetes is a continuous, modifiable risk factor for clinically important outcomes and that reduction in glucose is the key to improving outcomes. Interventions that maintain optimal glucose levels for long periods are likely to result in even larger clinical benefits. Second, intensive therapy with sulfonylureas and insulin, but not metformin, modestly increases the risk for severe hypoglycemia and weight gain, but these adverse effects are offset by the clinical benefits. The higher rate of these adverse effects with insulin suggests that oral agents should be the initial therapeutic choice. Third, these data support the choice of metformin as a first-line agent for obese patients, unless contraindicated. Fourth, type 2 diabetes is clearly a progressive disease. Increased effort and several types of therapies need to be applied over time for adequate glycemic control. Finally, the UKPDS provides no support for the belief that exogenous insulin causes cardiovascular disease; indeed, there was a strong trend in the opposite direction.

We now have convincing evidence that glucose control is an important goal for type 2 diabetes. Unless patients are seriously ill or have a short life expectancy, the long-term benefits of intensive therapy clearly outweigh the few risks. Merely eliminating symptoms of hyperglycemia is unsatisfactory. Our task is to convince patients of the value of this approach and to work with the health care team to teach patients how to improve glycemia and reduce their risk for the many complications of diabetes.

Hertzel C. Gerstein, MD
McMaster University Health Sciences CentreHamilton, Ontario, Canada


1. The effect of intensive treatment of diabetes on the development and progression of long-term complications in insulin-dependent diabetes mellitus. The Diabetes Control and Complications Trial Research Group. N Engl J Med. 1993;329:977-86.