Major Finding: Patients with type 2 diabetes and cardiovascular risk factors who took metformin had a 24% decrease in the risk of death over a 2-year period.
Data Source: A subanalysis of the international REACH Registry, focusing on 19,699 patients with type 2 diabetes.
Disclosures: The registry is sponsored by Sanofi-Aventis, Bristol-Myers Squibb, and the Waksman Foundation, Tokyo. Dr. Roussel disclosed that he has received research support or consulting fees from Sanofi-Aventis, Servier Laboratories, Roche, Eli Lilly & Co., Novo Nordisk Inc., Medtronic Inc., and LifeScan Inc.
ORLANDO — Metformin use was associated with a significant decrease in the risk of all-cause death among diabetic patients at risk for cardiovascular events.
The subanalysis of the Reduction of Atherothrombosis for Continued Health (REACH) Registry found that subjects with type 2 diabetes who took metformin were 24% less likely to die from all-cause mortality over 2 years than were those who did not take the drug. The association remained significant even after researchers controlled for age and gender, and after factoring in a number of baseline characteristics that varied significantly between the groups.
“Given the diversity of the 44 countries and widely different practice settings involved in the registry, we think these data are highly relevant,” Dr. Ronan Roussel said at the annual meeting of the American Diabetes Association. While perhaps not sufficient to make practice recommendations, he did say the results are strong enough to prompt clinical trials, especially when viewed in the context of the growing body of evidence about metformin's cardioprotective effects.
The REACH Registry was established to track outcomes in patients with atherothrombosis or atherothrombotic risk factors. Almost 70,000 patients were enrolled. They were either symptomatic, with documented cardiovascular, coronary artery or peripheral artery disease; or asymptomatic with at least three risk factors for atherothrombosis. Of this group, 19,699 had type 2 diabetes and 2-year outcomes data. Dr. Roussel of the Groupe Hospitalier Bichat-Claude Bernard, Paris, and his colleagues compared those who were taking metformin at baseline with those who were not. Metformin was taken by 40% of the patients.
There were some significant baseline differences between the groups, Dr. Roussel noted. Patients taking metformin were significantly younger (67 vs. 69 years), had a higher average fasting blood glucose (138 vs. 131 mg/dL), and higher systolic blood pressure (138 vs. 136 mm Hg). Prior arterial disease was present in 80% of those taking metformin and 75% of those not. Metformin users were also taking significantly more cardiovascular drugs, including aspirin (74% vs. 69%), statins (75% vs. 67%), and angiotensin-converting enzyme inhibitors (54% vs. 49%).
Over the 2-year follow-up period, there were 1,270 deaths. After researchers adjusted for gender and age only, metformin was associated with a 33% reduction in the risk of all-cause death. A Kaplan-Meier analysis showed that the mortality trajectories began to separate early, with a significant difference appearing around 6 months.
After adjustment for the other factors, the mortality difference still remained significant in favor of metformin use, with a 24% risk reduction in all-cause death.
In an age analysis, with subjects split into groups 40-65 years, 65-80 years, and older than 80 years, the risk reductions were significant for the youngest group (37%), and the middle group (23%). The oldest subjects did not have a survival advantage with the drug.
Metformin also improved the odds of survival in patients with existing congestive heart failure, conferring a significant 31% reduction in the risk of death.
While renal insufficiency is considered a contraindication to metformin use, Dr. Roussel noted that REACH subjects with moderately impaired renal function appeared to benefit from the drug. Those with a glomerular filtration rate of 30-60 mL/min had a significant 36% reduction in the risk of death; those with a GFR of less than 30 mL/min or greater than 60 mL/min did not gain a significant survival advantage.
Subjects who were taking insulin as well as metformin benefited more than did those who were taking metformin alone (hazard ratio 0.64 vs. 0.80).