Some ideas, like "why don't
we try a round wheel," or "how about sliced bread," are
so beautifully simple in their conception, it seems incredible
no one had thought of them before. Incredulity and the
stakes increase when it comes to medical discoveries �
because these can save lives.
Greet Van den Berghe, PhD of the
University of Leuven in Belgium, figured that most hospital
ICUs let glucose levels climb too high before administering
insulin to bring them down. In a landmark 2001 paper
in the New England Journal of Medicine, he demonstrated
that hospitals can cut their overall mortality by about
a third by adopting the simple practice of giving intravenous
insulin to patients in intensive care. These results
have now been confirmed in a real hospital setting and
published in the August issue of Mayo Clinic Proceedings.
Typically, hospitals only take
action when the glucose level climbs above 200mg/dL.
Professor Van den Berghe endeavoured to maintain blood
glucose levels lower than 110mg/dL by means of continuous
insulin infusion, while the new study headed by Dr James
Krinsley of Stamford Hospital in Connecticut aimed for
140mg/dL. Another difference was that the Belgian study
population was a relatively homogeneous group of surgical
patients, while Dr Krinsley's was more heterogeneous.
His study groups consisted of the last 800 patients
to pass through the ICU before it set the new
standard of glucose control, compared to the first 800
test subjects who came in afterwards. The two groups
were well matched for age, sex, race, critical condition
and diabetes status.
Under the new protocol, the number
of new cases of renal insufficiency decreased by 75%,
and the number of patients undergoing transfusion of
packed red blood cells fell by 18.7%. Above all, length
of stay fell by 10.8%, and hospital mortality by 29.3%.
Professor Van den Berghe said further
research is needed to fine-tune the approach: "At this
stage, several important questions remain unanswered.
First, what level of blood glucose is considered ideal
to achieve the most benefit combined with the lowest
risk of adverse events? Second, how do the benefits
occur � is glycemic control the most important factor
or are other metabolic or even nonmetabolic effects
of insulin playing a more important role?"
But there are no calls for further
research to test the basic proposition. On the contrary,
Professor Van den Berghe suggested it would be unethical
to treat ICU patients by the old standard for purposes
of comparison, since some are bound to die if they aren't
subjected to tighter glycemic control.
And that, effectively, is that,
without caveats or qualifications. This dirt-cheap,
almost effortless intervention will cut the death rate
in the average hospital by a quarter to a third, depending
on the glucose level aimed at. If only everything in
medicine were so simple.
|
