February 26, 2008

Diabetics’ Deaths Tied to Lower Blood Sugar

Federal researchers who tried to reduce heart attacks among diabetics by driving their blood sugar to low levels found that more patients were dying instead. The finding, announced yesterday, prompted officials to halt part of a major study of diabetes and heart disease.

Researchers said they don't know what caused the spike in deaths but said it can't be blamed on Avandia -- a diabetes medication linked last year to an increased risk of heart attacks -- even though some volunteers were taking the drug.

"Clearly we did not anticipate the findings," said Dr. Elizabeth G. Nabel, director of the National Heart, Lung and Blood Institute of the National Institutes of Health, which is running the clinical trial.

The troubling results call into question the popular theory that radically lowering blood sugar levels in patients with Type 2 diabetes could prevent heart attacks and stroke.

"We have to be careful, because more is not always better," said Dr. W. Douglas Weaver, president-elect of the American College of Cardiology and head of cardiology at the Henry Ford Health System in Detroit. "The importance of this study is: Sometimes the unintended happens. In this case, lowering blood sugar to normal levels had poor outcomes."

The ACCORD study (Action to Control Cardiovascular Risk in Diabetes) enrolled 10,251 participants at 77 sites in the U.S. and Canada, including the Baltimore Veterans Affairs Medical Center.

Of these, 257 in the intensive therapy group died over the course of four years, compared with 203 in the standard treatment group.

Using modified diet, exercise and glucose-lowering medication, the goal for the standard treatment group was to reduce blood sugar levels to a point known to be safer for diabetics -- but still higher than normal.

In the intensive therapy group, doctors were more aggressive, aiming to drive blood sugar levels down to those found in nondiabetics. They saw those patients more often, had them test their blood sugar more frequently and put them on more medications, researchers said.

Overall, the death rate in both groups was well below normal for Type 2 diabetics -- probably the result of the extra care and monitoring that ACCORD patients received.

The study found some beneficial effects from aggressive blood sugar lowering. For example, there were 10 percent fewer nonfatal cardiovascular events -- such as heart attacks -- in the intensive group, as opposed to the standard treatment group.

Large-scale study "However, it appeared that, if a heart attack did occur, it was more likely to be fatal," said Dr. William Friedewald, a professor of medicine at Columbia University and chairman of the study's steering committee. "In addition, the intensive treatment group had more unexpected sudden deaths, even without a clear heart attack."

Dr. Bruce Hamilton, chief of endocrinology at the Baltimore V.A., which enrolled 180 patients in the study, said the results caught him somewhat by surprise because death rates at his hospital were about equal between the two patient groups.

"We never picked up anything," he said, noting that it took a study with thousands of patients to reveal the danger.

Hamilton speculated that some patients might have had moments where their blood sugar dipped dangerously low, a condition called hypoglycemia. When this happens, the nervous system responds by quickening the pulse and constricting blood vessels. That, in turn, can trigger a heart attack or arrhythmia.

From the outset, ACCORD study volunteers were a high-risk population -- they had Type 2 diabetes for an average of 10 years, above-average blood sugar levels, heart disease and other risk factors such as high blood pressure or high cholesterol. Some were obese or smokers.

Those in the intensive treatment group will remain part of the study -- which also looks at lowering blood pressure and cholesterol. But their blood sugar goals will be relaxed for the remaining 18 months, officials said.

Heart disease, stroke More than 20 million Americans have Type 2 diabetes, and an estimated 284,000 die from it each year. Heart disease and stroke account for about 65 percent of their deaths. Adults with Type 2 diabetes are two to four times more likely to have a heart attack or stroke, or to die from heart disease, than nondiabetics.

A large body of research has shown that lowering blood sugar levels can significantly lower the risk of some complications from diabetes, including eye, nerve and kidney disease. One study has shown that patients with a more recent onset of diabetes than those in the ACCORD study showed a trend toward fewer heart attacks when their blood sugar levels were reduced.

But this was the first major clinical trial to study whether lowering diabetics' blood sugar levels to those of healthy people reduces the risk of heart disease.

The answer in this case was no.

"Obviously no one would have expected this result," said Dr. Steven Nissen, a cardiologist at the Cleveland Clinic who led the Avandia study. "It's a wake-up call, and it really does tell you drugs have complicated effects."

Nissen said the findings will have an immediate impact on his practice. Earlier this week, if a diabetic with heart disease had arrived in his office with a blood sugar level of 7.5 percent (compared with less than 6 percent in a normal person), "I would add another drug." he said.

Raising doubts "Today I have to wonder if that's the right thing to do," he said. Conversely, he would have patted himself on the back if a patient under of his showed up with blood sugar of 6.2 percent. Now, he said, he would wonder, "Am I hurting or helping?"

ACCORD researchers said the goal of those in the standard treatment group (including the former intensive therapy patients) is a blood sugar level between 7 and 8 percent. Their advice to others: Diabetics with increased risk for heart disease should not be encouraged to reduce levels further than that.

"I'm grateful to know what the glucose target should be," said Dr. Sheria Golden, an endocrinologist at the Johns Hopkins Hospital.

Too far down Golden said she's always mindful that driving blood sugar too low can trigger a hypoglycemic episode. "I'll get the glucose down as far as I can without complicating the patient with a hypoglycemic risk," she said.

ACCORD researchers found it difficult to get volunteers' blood sugar levels down to those seen in healthy people -- they reached only 6.4 percent instead of the under-6 target.

Doctors said counterintuitive results are one reason why clinical trials are conducted.

"Essentially, we do need evidence when something seems like a good idea ... to see whether it pans out in real life," said Dr. Mary M. Newman, an internist in Lutherville. "It sounds biologically plausible but that doesn't mean it works out that way."