Risks of treating diabetes 2
Posted by medconsumers on August 8, 2011
In keeping with the theme of this website—reporting the lack of evidence for common medical treatments—attention this week goes to drugs for type 2 diabetes. A new meta-analysis of the best studies on this topic shows strong hints that these drugs may cause more harm than good when intensive glucose (blood sugar) lowering is the goal. At best, there is a small reduction in non-fatal heart attacks among people treated intensely, at worst, an increase in cardiovascular deaths and deaths from all causes. All because blood sugar was intentionallly reduced to extremely low levels—ironically, in an effort to save type 2 diabetics from an early demise.
The meta-analysis included 13 randomized controlled trials with a combined total of over 33,000 type 2 diabetics given drugs to get their glucose levels down to what is defined as normal. The analysis of these trials was conducted by an international team of researchers and published recently in the British Medical Journal. To make sense of the findings, I turned to one of its co-authors, James M. Wright, MD, PhD, University of British Columbia, Vancouver, who acknowledged they were surprising. “One would expect that lowering glucose levels to normal [with drugs] would make people live longer; but we could find no reduction in mortality in these trials,” he explained. “If anything, mortality was increased by 4% in the people treated intensively [below normal].”
What exactly is “normal” and why was intensive lowering of glucose levels considered a treatment goal? To put the new findings in perspective, here’s the back-story for how type 2 diabetes became a common diagnosis:
-In 1998, the definition of diabetes was changed from a fasting blood sugar of 140 mg/dL to a blood sugar of 126 mg/dL, thus making millions of formerly healthy Americans into type 2 diabetics overnight and, not incidentally, candidates for long-term drug therapy. In time the A1C blood test, which provides a three- to four-month picture of past glucose levels, would be considered the better measure for assessing type 2 diabetes and the need for treatment.
-Most of the physician experts who served on the panel that expanded the definition of type 2 diabetes had financial ties to companies that make drugs for the condition. click here By 2009, an equally conflicted international panel of experts announced “the goal of therapy is to achieve a hemoglobin A1C of 6.5% or less.” click here Unfortunately, there is no evidence to prove this current recommendation is safe or effective.
-Lowered thresholds for normal initially seemed to make sense. Observational studies showed type 2 diabetics die of heart disease at rates two to four times higher than people without it. Diabetics also have more heart attacks and strokes. What’s more, risks start to increase at glucose levels lower than what had previously been defined as normal.
-Will glucose-lowering drugs also lower the chances of dying a premature death or having a heart attack or stroke? The first clinical trial designed to answer that question, the ACCORD trial, had to be stopped 18 months ahead of schedule in 2008. More deaths had occurred in the participants treated intensively, compared to those treated to normal blood glucose levels. There was also higher rate of hypoglycemia in the intensively treated people, 10% of whom required medical care.
This disturbing recent history brings us full circle to the new meta-analysis of the 13 clinical trials (one of which is the ACCORD). The 18,315 study participants of these trials who had been treated intensively (to A1C of 6.6%) were compared with the 16,218 treated less intensively (to A1C of 7.5%). All participants were taking the usual drugs prescribed for type 2 diabetes; those treated intensively were more likely to use combinations of drugs and higher doses. The meta-analysis produced this alarming conclusion: “The overall results do not show a benefit of intensive glucose lowering treatment on all cause mortality or cardiovascular death. A 19% increase in all cause mortality and a 43% increase in cardiovascular mortality cannot be excluded.”
Dr. Wright, co-author of the meta-analysis and practicing physician, was asked how he treats his patients. “My target is a HbA1C of 7% to 8%, never lower,” he responded. “If a patient comes to me with a 7.9% A1C test result, I would say that’s fine. “When someone on drug treatment has a 6.5% A1C, I would stop one of the drugs. When you get into the lower range—that is, below 7% there’s an increased risk of causing hypoglycemia, which can be severe and life threatening. The only exception would be the person on diet alone where hypoglycemia is not a risk and where you would encourage the patient to continue their diet and exercise to maintain the level of 6.5%.”
Added September 22, 2012 : Soon after this post went up on our website, this review was published in the Cochrane Library. It did not find any significant reduction in either death from any cause or death from heart disease when intensive glycaemic control was compared with conventional control.
Maryann Napoli, Center for Medical Consumers©