Diabetes Control:
Thanks for the Memories
Can Good Glycemic Control for a Short Time,
Reduce Diabetes Complications Long Term?
Evan
D. Rosen, M.D., Ph.D.
Assistant Professor of Medicine, Harvard Medical
School
A
little more than a decade ago, a major debate
in diabetes was put to rest. The question
centered on whether the complications of diabetes,
such as kidney, eye, nerve, and cardiovascular
disease, were the direct result of high glucose
levels. Opponents of this so-called "glucose
hypothesis" posited that these complications
resulted from other diabetes-related abnormalities
(like elevated lipid levels) or from a common
inherited predisposition to both diabetes
and to complications.
The publication of the Diabetes Control and
Complications Trial (DCCT) in 1993 answered
this question definitively, at least with
respect to type 1 diabetes. In that landmark
study, more than 1,400 people with type 1
diabetes were split into two groups. The first
group was given "conventional" insulin
treatment, which led to them having hemoglobin
A1c values of around 9%. The second group
received "intensive" insulin therapy,
which brought their A1c down to 7.4%. After
an average of 6.5 years, a significant reduction
in complications was noted in the intensive
treatment group. The benefit was clear in
terms of eye, kidney, and nerve disease. There
were fewer heart attacks and strokes in the
intensive group as well, but the overall number
of these events was so low that it was difficult
to draw meaningful conclusions about cardiovascular
disease and diabetes control.
The results of the DCCT study ushered in a
sea change in the way we approach diabetes
control. We now recommend that patients with
diabetes attempt to control their glucose
levels with a target hemoglobin A1c of <7%
(lower even than the levels achieved with
intensive therapy in the DCCT). These recommendations
are the same for patients with type 2 diabetes,
as later studies in Japan and the United Kingdom
indicated the validity of the glucose hypothesis
for this population as well.
When a clinical trial is over, the participants
are told the results and encouraged to manage
their condition in the best possible way given
the results of the study. This is exactly
what happened to the people in the DCCT, with
one exception. After the DCCT ended, the participants
continued to be monitored for progression
of complications in what was termed the Epidemiology
of Diabetes Interventions and Complications
(EDIC) Study. The goal of the EDIC study was
to see what would happen to these patients
after they were taken out of the two original
study groups.
First of all, there were big changes in hemoglobin
A1c in both groups. The patients in the conventional
group saw a reduction in their mean hemoglobin
A1c levels down to 8.2% (averaged over the
8 years of the EDIC study); not terrific to
be sure, but much better than the 9.1% they
had carried through the DCCT years. The intensive
group had the opposite trend—liberated
from the rigorous control imposed by the DCCT
and its enforcement by study nurses and doctors,
this group let their glucose control slip
a bit, and they settled out at a mean A1c
of 8.0%. The net result was that both groups
ended up at about the same A1c level within
the first year after the DCCT, and stayed
there for the remaining 7 years of EDIC.
This "leveling" of glucose control
allowed the EDIC researchers to ascertain
whether the rate of complications would also
even out over time. In fact, this is NOT what
happened. What happened was that the complication
rate remained significantly lower in the patients
that used to be in the intensive group. In
a paper published last year, it was noted
that the intensively treated folks had fewer
cases of retinopathy. Now, a new paper by
the same authors points out that the people
who had been treated intensively have a 60%
reduced likelihood of developing small amounts
of protein in the urine (an early sign of
diabetic kidney disease) and an 80% reduced
likelihood of developing more serious amounts
of urine protein. Although not enough patients
had such severe progression of their kidney
disease that they required dialysis or transplantation,
it was interesting to note that there were
fewer cases like this in the people who had
been treated intensively in the DCCT. Finally,
and perhaps most intriguingly, there was a
significantly reduced risk of high blood pressure
in the intensive group. And all of this benefit
accrued to the intensive group despite the
fact that their control was no better than
the conventional group over the most recent
eight years.
The results of EDIC indicate that there may
be a "metabolic memory," where the
benefits of a period of good glucose control
continue even after A1c levels have slipped.
This memory may work in the opposite direction
as well. One of the findings of the original
DCCT was that it took several years for the
intensively treated group to experience reduced
complications relative to the conventional
group, as if their bodies needed time to get
over the preceding years of poor control.
One would predict then that over the next
several years, the complication rates of the
two groups in EDIC will begin to look more
and more alike, as the body’s "memory"
resets to a hemoglobin A1c around 8%.
For patients, it almost certainly doesn't
mean that they can strive for a period of
good control and then let go, secure in the
knowledge that they will always enjoy the
benefits of their metabolic memory. But it
would be fascinating to learn the cellular
and molecular basis of the memory phenomenon.
It would then be theoretically possible to
identify medications that would trick the
body into believing that sugar levels were
near normal levels all the time, thus reducing
complications without the risks associated
with tight glucose control, such as hypoglycemia
and weight gain. Until that time, however,
there is no substitute for good control, and
patients should strive to have the lowest
hemoglobin A1c levels possible with an acceptable
rate of adverse events.
This information was last reviewed November
6, 2003.
References:
Journal of the American Medical Association
2002 287:2563
Journal of the American Medical Association
2003 290:2159
New England Journal of Medicine 1993 329:977
For other Features from Dr. Rosen, go
to Dr. Rosen’s Archives
