United Kingdom Prospective Diabetes Study (UKPDS 33)

The UKPDS 33 investigated whether tight control of blood glucose in type 2 diabetic patients reduced the risk of microvascular or macrovascular disease. Patients over age 65 were excluded from this multicenter RCT; mean age was 54. Of the 3867 patients, 2729 were randomized to open-label intensive drug therapy with either chlorpropamide, glibenclamide, or insulin, with the goal of lowering FPG to less than 108 mg/dL. The 1138 other patients were randomized to conventional therapy that aimed for the lowest FPG possible with diet alone (although drugs were added if FPG reached as high as 270 mg/dL or in the presence of hyperglycemic symptoms).

Compared to conventional therapy, intensive therapy significantly reduced the risk of microvascular complications, but not macrovascular complications or macrovascular subclinical surrogate endpoints over the 10-year study period. Median hemoglobin A_1c levels were significantly lower in the intensive group (7%) than in the conventional group (7.9%).

Three grouped endpoints were assessed: all-cause mortality, diabetes-related deaths [myocardial infarction (MI), stroke, peripheral vascular disease, renal disease, hyperglycemia or hypoglycemia, and sudden death], and any diabetes-related endpoint (death from any of the preceding, as well as nonfatal MI, angina, heart failure, stroke, renal failure, amputation, vitreous hemorrhage, retinopathy, blindness, or cataract).

Only diabetes-related endpoints were significantly reduced with intensive therapy, with a risk reduction of 12% (95% CI, 1%-21%). Most of this reduction was due to reduction in microvascular outcomes, especially the need for retinal photocoagulation, which was reduced 25% (95% CI, 7%-40%). All three intensive drug regimens reduced microvascular endopoints equally.

Neither all-cause mortality nor diabetes-related deaths differed by intensity of treatment, although a nonsignificant trend (p = 0.052) toward reduction in risk of myocardial infarction was observed in the intensive therapy group, with an RRR of 16% (95% CI, 0-29%). Compared to conventional treatment, intensive therapy was associated with a mean 6.4-pound weight gain (most prominently in the insulin group) and an increased risk of hypoglycemia. Treatment for hypertension was more common with chlorpropamide than with any other regimen, including conventional therapy.

A substudy examined 342 newly diagnosed obese diabetic patients randomized to metformin during UKDPS 33 and compared them to the overweight patients who were included in the UKDPS 33 study already discussed [conventional treatment (411) or intensive treatment (951)]. Based on absolute risk (events per 1000 patient-years) and relative to conventional therapy, metformin treatment reduced risk of any diabetes-related endpoint by 32% (NNT, 10; CI, 6-32), diabetes-related deaths by 42% (NNT, 19; CI, 10-97), and all-cause mortality by 36% (NNT, 15; CI, 8-83).

{embed="db/static-menu-geriatric"}Relative risk of myocardial infarction was also significantly reduced. Metformin was also significantly better than all other drugs used in the intensively treated group in reducing risk of reaching any diabetes-related endpoint, stroke, and all-cause mortality. In these already overweight patients, weight gain was less common with metformin than with other intensive therapies, as were hypoglycemic episodes.