Peters AL, Davidson MB, Scriger DL, Hasselblad V for the Meta
-analysis Research Group
A clinical approach for the diagnosis of diabetes mellitus: An
analysis using glycosylated hemoglobin levels
J Am Med Assoc
(Oct) 276:1246-1252 1996

Guidelines
for
diabetes screening are presented by the ADA in the January 1996
supplement of Diabetes
Care (see "Practice Guidelines" at the bottom of the home page of HDCN).
The mainstay is the
fasting plasma glucose level (> 115 mg/dl indicates the need for further
testing) or a random
plasma glucose level > 160 mg/dl, followed by and oral glucose tolerance
test (OGTT).
The paper by Peters and colleagues suggests a different approach. They did a
meta-analysis of all
papers in which both an OGTT was done and glycosylated hemoglobin levels
(HbA1c) were measured, to
see if the latter could be used as a substitute test. HbA1c had the least
variance in normal
subjects. Using a database of about 9000 subjects, they found that when the
HBA1c level was greater
than 7.0%, 89% patients had diabetes, 7% had impaired glucose tolerance, and
4% were normal. They
suggest that patients with HbA1c levels > 7% need pharmacologic
intervention (based on no data),
whereas persons with levels %lt; 7% may need only tx with diet and exercise
(again, based on no
data). The approach is superficially attractive, however. It would be
interesting to assess the
variance of repeated measurement of the OGTT vs. HbA1c (John T. Daugirdas,
M.D., University of
Illinois at Chicago)
The full abstract is available at
the JAMA site. You will need to
register with the JAMA before being allowed to access this
material.
Once you have registered, read the
abstract.