A New Path for the ACCORD (The Action to Control Cardiovascular Risk in Diabetes) trial: Does Being Sweeter Save Lives?

February 8, 2008

90px-glukometr_ot.jpgCommentary by Melissa Freeman MD, Endocrinology Section Editor

The ACCORD trial is an ongoing 5-year, North American, randomized study that began in 2001 to evaluate potential interventions to decrease cardiovascular (CV) events in adults living with DM2.  The trial enrolled 10,251 adults, aged 40- 82, with DM2 for 10 or more years, and a history of CV disease or two CV risk factors in addition to DM2. All participants were randomized at enrollment into intensive versus standard glucose control. In addition, participants were also enrolled into one of two additional arms of the study looking at either tight blood pressure control versus standard blood pressure control or treatment of lipids by a fibrate plus statin versus a statin alone.  The intensive blood sugar treatment group aims for an A1C of less than 6% and blood sugars comparable to non-diabetics, achieved by using multiple glucose lowering agents used including metformin, TZD’s, insulin, sulfonylureas, acarbose, and exenatide.  The primary outcome being measured is the occurrence of major CV events after randomization.

The sponsor of the ACCORD trial, The National Heart, Lung, and Blood Institute, has put the brakes on the intensive blood sugar lowering treatment arm due to serious safety concerns expressed by the study’s Data and Safety Monitoring Board. What happened?  After roughly four years, 257 participants in the intensive-treatment group died compared to 203 in the standard treatment group resulting in a 20% higher rate of all-cause death in the intensive glucose lowering group.  Over half of the deaths were from CV causes.  Interestingly, the rate of primary cardiac outcome events (i.e. nonfatal heart attacks) was 10% lower than in the standard group, but the study cites that when heart attacks in this group occurred they were more likely to be fatal. Furthermore, the death rate in the intensive treatment group was lower than the death rate in similar general populations of such adults with DM2.  Still, after risk/benefit comparisons, it was decided that the intensive glucose lowering group posed more harm than good.  The additional arms of the ACCORD study will go on, but all participants will now be treated according to the protocol for standard blood sugar control (A1C 7- 7.9%) with a comparison of outcomes in the participants originally in the intensive blood sugar group to evaluate whether further differences will persist. 

What is the cause of the cited increased mortality?  Neither the medications or the increased number of hypoglycemic episodes have been found to be legitimate culprits.  Was the medical strategy too brave and too intense by going below the current recommendations? Are the participants too high-risk to be generalizable? Did the blood pressure or lipid interventions in this treatment arm confound the results?  Was the study properly designed?  Was more time needed before making conclusions?  In light of this months NEJM article by Gaede et al. showing that tight glucose regulation results not only in CV benefits but in decreased mortality, the current findings in the ACCORD trial must be viewed in the larger context of what is known.  

Still, after a resultant news frenzy that even landed the front page of the NY Times, patients will be asking whether aggressive treatment is right for them.  It may only be a matter of time before A1C’s of 14 will be the new craze.  Nevertheless, at present, it would be premature to say that sweetness saves lives.