A type 2 diabetes drug taken orally and in widespread
use for more than a decade has been found to have distinct
advantages over nine other, mostly newer medications used
to control the chronic disease, according to a study by
researchers at Johns Hopkins.
In their report, published online July 16 in the
journal Annals of Internal Medicine, the Johns
Hopkins team found that metformin, first approved by the
U.S. Food and Drug Administration in 1995 (and sold as
Glucophage, Riomet and Fortamet), not only controlled blood
sugar levels but also was less likely to cause weight gain
and more likely than other medications to lower bad
cholesterol levels in the blood.
Researchers say these health benefits are important
because they can potentially ward off heart disease and
other life-threatening consequences of diabetes. More than
15 million Americans have type 2 diabetes.
"Sometimes newer is not necessarily better," said lead
study author Shari Bolen, an internist and instructor at
School of Medicine. "Issues like blood sugar levels,
weight gain and cost could be significant factors to many
patients struggling to stay in good health."
In what is believed to be the largest drug comparison
of its kind, the scientists showed that all the commonly
used oral medications worked much the same at lowering and
controlling blood sugar levels, and were equally safe. But
metformin stood out because it offered the same level of
effectiveness without lowering glucose measurements too
much, and it did so for a lower price.
Metformin was found to lower LDL, or bad cholesterol,
by about 10 milligrams per deciliter of blood, while newer
medications studied, such as pioglitazone (Actos) and
rosiglitazone (Avandia), or so- called thiazolidinediones,
were found to have the opposite effect, increasing levels
of the artery-clogging fat by the same amount.
Researchers say the main drawbacks to metformin are
digestive problems and diarrhea. Previous reports have
found evidence that the medication leads to the buildup of
lactic acid in the blood of people with moderate kidney or
heart disease, and the researchers note that it should not
be prescribed to anyone with either of these conditions.
The main advantages to both newer thiazolidinediones were a
small increase in HDL, or good cholesterol, and less
too-low blood sugar levels than three other older, cheaper
drugs studied — glimepiride (Amaryl), glipizide
(Glucotrol) and glyburide (Micronase, DiabBeta, Glynase
PresTab), known as second-generation sulfonylureas.
Annual treatment with metformin or the sulfonylureas,
they note, costs on average $100, roughly one-fourth the
cost of oral diabetes medications approved by the FDA more
recently, including the two newer thiazolidinediones
approved in 1999. (Their price is expected to drop once
generic versions become available.)
"When you are dealing with an epidemic like diabetes,
it is important for people to weigh their treatment options
with their physician and to make informed decisions about
which medication best suits their needs," Bolen said.
In the study, Bolen and her colleagues reviewed the
scientific evidence from 216 previous studies and compared
each drug for its clinical effectiveness, risks and costs.
In addition to metformin, the thiazolidinediones and
sulfonylureas, drugs included in their analysis were
repaglinide (Prandin), miglitol (Glyset), acarbose
(Precose) and nateglinide (Starlix).
Among the team's other findings were that glimepiride,
glipizide and glyburide led more frequently to too-low
blood sugar levels than the other drugs. The sulfonylureas
and acarbose appeared to have no effect on bad cholesterol.
And except for metformin and acarbose, drug treatment led
to an increase in weight from two to 11 pounds.
Researchers also noted the increased risk of heart
failure, albeit small (less than three people in 100), in
people taking thiazolidinediones who did not have a history
of heart disease. They also caution that despite recent
reports about the potential for increased risk of heart
attack from rosiglitazone, there is not yet sufficient
information to verify the finding.
Researchers say further studies are needed to compare
the long-term effectiveness of one treatment to another and
to compare drug effects on quality of life and life
expectancy. Additional research also will be needed to
compare these findings with results for injectible
medications for diabetes, most notably insulin, which was
not included in the latest report.
The study, conducted solely at Johns Hopkins, was
supported with funding from the Agency for Health Care
Research and Quality; the federal agency has posted the
analysis, along with a question- and-answer document, on
its Web site at
watchdog publication Consumer Reports has posted a related
report at www.CBestBuyDrugs.org.
In addition to Bolen, researchers involved in the
study were Leonard Feldman, Jason Vassy, Lisa Wilson,
Hsin-Chieh Yeh, Spyridon Marinopoulos, Crystal Wiley,
Elizabeth Selvin, Renee Wilson, Eric Bass and Frederick