CHICAGO – A decade of diet and exercise had no effect on cardiovascular morbidity and mortality in overweight and obese adults with type 2 diabetes, according to a randomized, federally funded study that was stopped 2 years early due to futility.
"However, there are many reasons why, I would argue, that you should be encouraging patients with diabetes to lose weight," said Rena R. Wing, Ph.D., the trial’s chair. The results showed that the participants who were in the diet and exercise group, "are less likely to have kidney disease, have less incidence of depression, lower hospital costs, and less medication costs, so there are still many advantages to intense lifestyle interventions," Dr. Wing said during the presentation of the findings at the annual scientific sessions of the American Diabetes Association.
The results also showed that the participants were able to maintain a modest weight loss during a 10-year period.
The Look AHEAD (Action for Health in Diabetes) trial aimed to fill a gap in existing data about whether intensive lifestyle intervention would decrease cardiovascular morbidity and mortality of patients with type 2 diabetes in the long term.
Researchers recruited more than 5,100 patients and randomized them to the intensive lifestyle intervention program, which decreased calorie intake and increased exercise, or to the control group, which provided diabetes support and education. The goal of the intervention was achieving and maintaining weight loss of at least 7% through group and individual counseling.
The trial’s primary outcome was a composite of death from cardiovascular causes, nonfatal myocardial infarction, nonfatal stroke, or hospitalization for angina. The trial was planned to have a greater than 80% probability of detecting an 18% difference in cardiovascular events between the two arms, assuming that two-sided alpha was 0.05, a primary outcome rate of 2% per year in the control group, and the planned maximum follow-up of 13.5 years.
But in September 2012, the trial was stopped at the request of its primary sponsors, based on a futility analysis. The median follow-up at the time was 9.6 years.
By then, 403 patients in the intervention group and 418 patients in the control group had met the trial’s primary outcome, with no statistically significant difference between the two groups (1.83 and 1.92 events per 100 person-years, respectively).
Why didn’t it work?
Researchers listed several possible explanations for the results. The study may have had insufficient power, although that wouldn’t explain the negative results, the authors wrote. It is also possible that a higher sustained weight loss was needed in the intervention group to reduce the risk of cardiovascular disease. Meanwhile, the educational sessions and increased use of statins in the control group may have lessened the difference between the two groups, the authors noted. And intensified medical management of cardiovascular risk factors for both groups may have made the relative benefit of the intensive lifestyle intervention more difficult to demonstrate, they added.
Earlier intervention during the course of diabetes may be needed, they added. The results were published online simultaneously with the presentation (N. Engl. J. Med. 2013 June 24 [doi:10.1056/NEJMoa1212914]).
In an editorial accompanying the published results, Dr. Hertzel C. Gerstein wrote that the inclusion of "the somewhat unreliable outcome of hospitalization for angina in the primary composite outcome may have added noise and further obscured any emerging signal" (N. Engl. J. Med. 2013 June 24 [doi:10.1056/NEJMe1306987]).
Dr. Hertzel Gerstein
During the presentation, an audience member suggested that since the 1-year and 4-year findings of the trial were published and well publicized, the participants in the control group might have taken it upon themselves to take action and lose weight. The panel of investigators said that is also a possibility.
Dr. Peter H. Bennett, who is one of the trial’s investigators and was cochair of eligibility, said that stopping the trial "was a big mistake," and that it should have been carried on at least until the planned termination point. "They stopped this trial because they were projecting that if they carried on for 2 more years, the differences in the primary endpoint were unlikely to be statistically significant. But there are several problems. First, you prejudice the power to analyze any other events, and just as we’ve learned from natural history of diabetes-related mortality, it’s clear that you don’t see excess diabetes-related mortality until they’ve had at least 15 years of diabetes, and that’s just at the point that the trial was stopped," Dr. Bennett, scientist emeritus at the National Institutes of Health – one of the main funders of the trial – said in an interview.