From Reuters Health Information
News Author: Karla Gale, MS
CME Author: Désirée Lie, MD, MSEd
July 2, 2010 –— In apparently healthy middle-aged adults with a screening hemoglobin (Hb)A1c below 6.0%, rescreening at 3-year intervals won't discover many cases of diabetes, according to a longitudinal study in Tokyo with more than 16,000 subjects.
But for those with an initial HbA1c of 6% or higher, annual rescreening "would be a reasonable strategy," the authors advise. In the first year alone, more than 16% of such individuals were diagnosed with diabetes.
According to the American Diabetes Association, diabetes screening should be
considered in all adults who are overweight and have additional risk factors, such as a family history, a high-risk ethnic background, or history of cardiovascular disease. In normal weight individuals without risk factors, they recommend that testing start at age 45. If results are normal, they say that testing should be repeated at least every 3 years.
In Diabetes Care published online June 21st, lead author Dr. Osamu Takahashi, from St. Luke's International Hospital, Tokyo, and colleagues analyzed data for 16,313 people (mean age at baseline, 49.7 years) who underwent four annual health exams between 2005 and 2008.
"In June last year, the American Diabetes Association made a major change to the diagnosis of diabetes by switching to the use of HbA1c at a threshold of 6.5%," Dr. Takahashi told Reuters Health by e-mail. "However, this is also more costly and hence the timing of reassessment will be important. We have used a large cohort study to assess the appropriate interval for re-checking."
At baseline, 1828 subjects had an HbA1c less than 5.0%, 7609 were between 5.0% and 5.4%, 4592 were between 5.5% and 5.9%, and 771 were between 6.0% and 6.4%.
The cumulative incidence of diabetes varied greatly depending on the baseline HbA1c level. At three years, the cumulative incidence was 0.05% in the groups with initial HbA1c of less than 5.0% or 5.0-5.4%. The cumulative incidence increased to 1.2% for those with baseline values of 5.5-5.9%, and to 20% for those with values of 6.0-6.4%.
When the baseline value was 6.0-6.4%, the incidence of diabetes rose most steeply in the first year, to 16.2%. It rose to 17.1% at year 2 and 20.2% at year 3.
Patterns were similar when diabetes diagnosis was based on a fasting plasma glucose above 126 mg/dL.
Logistic regression analysis suggested that other significant risk factors besides baseline HbA1c were body mass index (odds ratio 1.14 per kg/m2) and fasting plasma glucose (odds ratio, 1.06 per mg/dL). Age, gender, systolic blood pressure and LDL-cholesterol did not have significant impact.
Dr. Takahashi recommends retesting patients with an HbA1c over 6.0% to provide a more timely diagnosis in some patients. However, he believes a cost-effectiveness analysis is warranted before the study's findings are used in clinical practice.
The authors note that their data may not be generalizable to populations outside of Japan, especially since adult body mass index averages 22-23 kg/m2 in Asia, versus 25-27 kg/m2 across North America and Europe.
Diabetes Care. Published online June 21, 2010. Abstract
Reuters Health Information 2010. © 2010 Reuters Ltd.
Clinical Context
A shift to using HbA1c level for the diagnosis of diabetes has been recommended because it integrates long-term glucose levels and has better preanalytic stability. However, the optimal interval for rechecking HbA1c levels in nondiabetic patients with HbA1c levels below the threshold of 6.5% has not been determined.
This is an examination of retrospective data from a screening program that used annual HbA1c to test for diagnosis in healthy adults, to assess optimal intervals for testing using HbA1c levels
No comments:
Post a Comment