Duke Medicine HealthLine
Published: May 15, 2008
Updated: May 13, 2010
Clearing the confusion about controlling blood sugar
Duke endocrinologist Mark Feinglos, MD, is worried. He’s been hearing lots of questions and concerns from patients and providers about the national ACCORD study, which made headlines in 2008 when a portion of the study was shut down.
ACCORD is looking at how lowering glucose levels in people with diabetes affects their risk for heart disease and stroke; researchers quickly halted the study group that used multiple medicines to reduce their glucose to below the recommended level, after it was shown that participants in that group had a higher incidence of heart-related death.
This development was surprising as much as it was disturbing -- why would the group with the lowest glucose levels have the highest mortality? Feinglos, who is principal investigator on the Duke section of the study, says that more research is needed to answer that question -- for now, he’s more concerned about the false conclusions that may be drawn.
“What scared me about the ACCORD study is that many people read the news stories to mean that tight control of blood sugar is bad,” says Feinglos. “I’m afraid that there are patients out there who have newly diagnosed diabetes, who take small doses of one medication, see their blood sugar levels go down, then hear about the ACCORD study and think, ‘Oh no, this is dangerous, I now have to stop taking my meds and go eat doughnuts.’”
Instead, he says, it’s important to note that the study group in question doesn’t reflect the general population. “It was a very specific group of older people who either have cardiovascular disease or are at high risk for it,” he says. “These people have had diabetes for a long time and were taking a lot of medications to get down to this blood sugar level.”
What’s more, the absolute number of deaths in the study was not only fairly small, but also much smaller than the rate of death in the general population with diabetes. “The mortality rate in both arms of the study was less than it is for people outside of the study. It’s just that one arm had more deaths than the other, so we took out that arm.”
Participants in the canceled study group were trying to reach an A1C level of 6 percent. A1C is a measure of blood glucose control over time, and an A1C of 7 percent or less is what’s recommended by the American Diabetes Association. But Feinglos points out that there’s really nothing magical about the number seven.
“The reason that 7 percent is the guideline is actually because of type 1 diabetes,” he says. “In early studies on controlling diabetes complications, they tried to lower blood glucose in folks with type 1 -- and they couldn’t ever really get patients’ A1C lower than seven. That’s why they set the recommendation there.”
If this ACCORD finding means anything, says Feinglos, it’s for only a very specific group of people. “If you’ve had diabetes for many years, if you’re older and you already have heart disease or its risk factors, then it may not be good to use a whole lot of different medications to try to push your A1C way down. But for someone with newly diagnosed diabetes -- for anyone whose blood sugar can be controlled using diet and one or two medications -- the goal should still be to have an A1C that is as low as possible.
“Diabetes management is not a one-size-fits-all deal,” Feinglos says. “Were we surprised by what happened in the ACCORD study? Yeah. But does it have bearing for everybody? No, it really doesn’t.”