Standards Might Rise on Monitors for Diabetics By GARDINER HARRIS

July 19, 2009
Standards Might Rise on Monitors for Diabetics By GARDINER HARRIS

Federal officials may soon require improvements for the glucose monitors used by more than 11 million diabetics in the United States.

The rise in the use of home glucose monitors, even by hospitals, is pushing the action by the Food and Drug Administration, which for decades has followed international standards that allow the devices to be wrong by as much as 20 percent. Such a wide error rate can leave patients vulnerable to severe problems, including seizures, unconsciousness and coma.

In June, the agency pressed the international group that sets the standards to tighten them. If the group refuses to act, the agency “may instead recognize other (higher) performance standards” on its own, according to a June letter from Dr. Margaret A. Hamburg, the agency commissioner.

A change in the international standards is the easiest and best option, officials said. The International Organization for Standardization, which sets the standards, can act quickly and broadly. But the F.D.A. can change the rules itself through a more time-consuming and cumbersome process.

Officials said they would keep pushing until monitor accuracy improves, a promise that diabetes doctors cheered. In a May letter, the American Association of Clinical Endocrinologists formally asked that the agency act on the issue.

“Because of the highly variable quality of the meters and the glucose testing strips in widespread use, the safety of our patients who depend upon those meters is threatened,” the letter said.

Khatereh Calleja, a spokeswoman for the Advanced Medical Technology Association, which represents monitor manufacturers, responded, “We think the present standard is working.”

Diabetes has been diagnosed in 18 million people in the United States, and another 6 million are estimated to have the disease without knowing it. It is the seventh leading cause of death and costs the United States an estimated $174 billion a year, with the federal Medicare program spending $1 billion on diabetes test strips alone.

Of particular concern to federal officials is the increasing use of home glucose monitors in hospitals. A landmark 2001 study published in The New England Journal of Medicine found that using insulin to maintain low blood sugar levels in critically ill patients, even those without diabetes, reduced hospital deaths by 34 percent — a result so astonishing that hospitals around the world soon adopted the practice.

But instead of buying the highly accurate and expensive glucose monitors used in the study, many hospitals bought cheaper home models never approved for hospital use. More recent studies have shown that critically ill hospital patients whose glucose levels were kept low suffered more problems — the opposite result from 2001.

The difference, F.D.A. officials said, may have resulted because many patients in the second study were checked with home monitors.

“We think this technology is not up to par for some of the protocols we see out there” like hospital treatment of critically ill patients, said Dr. Alberto Gutierrez, deputy director of the agency’s in vitro diagnostics office. “We feel passionately that this is an important issue.”

Besides having a wide error rate, many home monitors give the wrong result if patients are taking certain drugs like Tylenol or even vitamin C. The Accu-Chek monitors made by Roche can be confounded by drugs commonly used in dialysis. Julie A. Vincent, a Roche spokeswoman, said, “Every blood glucose monitor on the market has some limitation or interferences.”

The F.D.A. issued warnings about the drug-related problems, but doctors complain that they have a hard time keeping straight which drugs conflict with which monitors.

“In the hospital setting, you really don’t know how many deaths are due to things that may be related to meter accuracy,” said Dr. Richard Hellman, a former president of the endocrinology group. “I don’t know how common it is, but I don’t think it’s rare.”

A study by government researchers found that when comparing tests from five different popular monitors, results varied by as much as 32 percent. For a class science project recently, Morgan DiSanto-Ranney, 16, of Bishop O’Connell High School in Arlington, Va., bought seven different glucose monitors and had her father, a diabetic, use all of them.

“What I found was that almost all of the meters were off from one another by 60 to 75 points,” Morgan said in an interview. Two of the meters — Ascensia Breeze and Ascensia Breeze II, both made by Bayer — differed by an average of 62 points, she said.

Staci Gouveia, a Bayer spokeswoman, said her company’s monitors meet federal requirements. “If the F.D.A. standards change, Bayer will work with the F.D.A. to meet their requirements and assure the accuracy and effectiveness of our meter,” Ms. Gouveia said.

Morgan’s mother is Emilia DiSanto, a staff investigator for Senator Charles E. Grassley, Republican of Iowa. Briefed on Morgan’s test and other studies, Mr. Grassley sent a letter to the F.D.A. in June asking officials to review the problem.

As a result of her project, Morgan’s father lost faith in glucose monitors. “He doesn’t use them as much anymore,” she said.

That reaction is exactly what federal officials are hoping to avoid by quietly pressing manufacturers to improve accuracy. Multiple studies make clear that diabetics who routinely use monitors are healthier and suffer fewer serious complications than those who do not.

Manufacturers have long complained that any requirement to improve accuracy would lead them to raise prices, which would discourage use.

“If we decrease the use of meters, you will have some fairly dire consequences to health,” Dr. Gutierrez said, but requiring stricter accuracy standards “seems a reasonable and safe practice to do.”

Every year, the F.D.A. receives reports of several deaths and thousands of injuries related to glucose monitor failures, but the reports represent only a fraction of the actual toll. Insulin-dependent diabetics slip into unconsciousness once a year on average, and 40 percent suffer seizures or coma in their lifetimes because of low blood sugar levels, according to the American Diabetes Association, which has long advocated stricter accuracy standards for monitors.

“Insulin is a dangerous drug, and if someone makes the wrong decision about its use because of a bad test, they could die,” said Dr. David Sacks, an associate professor of pathology at Harvard Medical School.