Accessibility links

Scientists Find More Accurate Way to Test Blood Sugar Levels

  • Jessica Berman

FILE - Judith Garcia, 19, fills a syringe as she prepares to give herself an injection of insulin at her home in Commerce, California, April 29, 2012

FILE - Judith Garcia, 19, fills a syringe as she prepares to give herself an injection of insulin at her home in Commerce, California, April 29, 2012

In one-third of people with diabetes, the "gold standard" method for monitoring blood sugar levels is inaccurate. But scientists have found a way to fix the discrepancy, helping diabetics more effectively manage their disease.

There are 422 million people with diabetes worldwide, according to World Health Organization estimates, and they face such devastating complications of the disease as blindness, kidney failure, heart disease and limb amputation. Keeping blood sugar levels in a range that's as close to normal as possible is considered the best way to reduce the likelihood that a diabetic will have to deal with those complications.

But researchers from Harvard Medical School and Massachusetts General Hospital found in a study of 200 diabetics that the A1C test, considered the most accurate method for measuring glucose levels, is inaccurate in one in three people.

Three-month period

The A1C offers a snapshot of blood sugar levels based on a three-month analysis of glycated hemoglobin, or the hemoglobin to which glucose is bound. The test is considered more accurate than daily blood sugar levels, which can vary minute by minute.

FILE - A woman who suffers from diabetes is seen walking on a treadmill as part of an exercise program to help control the disease.

FILE - A woman who suffers from diabetes is seen walking on a treadmill as part of an exercise program to help control the disease.

But when scientists compared daily glucose monitoring to the A1C, they found discrepancies based on the age of the red blood cells, according to John Higgins, a professor of systems medicine at Harvard and a clinical pathologist.

Red blood cells live an average of 45 days. But they live longer in some people, prompting this comparison by Higgins.

"I think a simplistic analogy is just a sponge sitting on a damp counter," he said. "The more water you have on your counter, the more soaked that sponge is going to get. But also if you leave that sponge on a counter even longer, it's going to soak up even more water."

So, the A1C level of a person whose blood sugar seems well-controlled day to day may actually be elevated if his red blood cells are older.

"And the opposite is true as well," said Higgins. "If someone has had a really high level of blood sugar but his blood cells are pretty young, it may look like he is fine when in fact his blood sugar's been high."

FILE - A Novo Nordisk employee controls a machine at an insulin production line in a plant in Kalundborg, Denmark, Nov. 4, 2013.

FILE - A Novo Nordisk employee controls a machine at an insulin production line in a plant in Kalundborg, Denmark, Nov. 4, 2013.

When the age of each person's red blood cells was factored in, the error rate in the A1C's three-month blood sugar reading was cut in half in the study.

The findings were reported in the journal Science Translational Medicine.

Continuous measurement

One way to correct for the discrepancy, according to Higgins, would be for someone to briefly wear a device that continuously measures blood sugar level.

The device, which uses a needle implanted under the skin, takes glucose readings every five minutes. Those hundreds of readings could then be compared with an A1C reading. With the data, the difference between daily blood sugar levels and the A1C level could be permanently corrected for in the lab. By factoring in the age of each patient's red blood cells, the method offers a more precise, individualized method of diabetes control.

Most people think of personalized medicine as involving sophisticated, expensive genetics tests that help doctors tailor treatments for complex diseases. Higgins said adjusting a test for A1C is a form of personalized medicine using everyday lab tests.

"I think this study shows that we can look at laboratory tests that are already being collected and measured," he said. "And if we think mathematically about the physiology leading to those measurements, we can personalize monitoring and treatment of some of these diseases already."

Show comments

XS
SM
MD
LG