Clinical Diabetes
25:98
2007
DOI: 10.2337/diaclin.25.3.98
© 2007 by the American Diabetes Association
Intraoperative Intensive Insulin Therapy Did Not Improve Outcomes Among Patients Undergoing Cardiac Surgery
Michael Pignone, MD, MPH
 |
STUDY
|
|---|
Gandhi GY, Nuttall GA, Abel MD, Mullany CJ, Schaff HV, O'Brien PC, Johnson
MG, Williams AR, Cutshall SM, Mundy LM, Rizza RA, McMahon MM: Intensive
intraoperative insulin therapy versus conventional glucose management during
cardiac surgery: a randomized trial.
Ann Intern
Med 146: 233-243,2007[Abstract/Free Full Text]
|
SUMMARY
|
|---|
Design. Randomized, controlled trial at a single academic teaching
hospital.
Subjects. Four hundred patients undergoing on-pump cardiac surgery.
Mean age was 63 years, 69% were men, 96% were white, and 20% had a history of
diabetes.
Methods. Patients were assigned randomly to receive either
conventional intraoperative glucose control (treatment for blood glucose >
200 mg/dl) or intensive control (insulin infusion to maintain blood glucose
between 80 and 100 mg/dl). After surgery, both groups were treated with
intensive glucose control. The main outcome was the combined incidence of
death, sternal infection, prolonged mechanical ventilation, arrhythmia,
stroke, or renal failure within 30 days.
Results. Intensive therapy lowered mean blood glucose: 114 mg/dl in
the intensive group versus 157 mg/dl in the conventional group at the end of
surgery. However, no difference was observed in the incidence of the main
composite outcome; 44% of patients in the intensive group had an event versus
46% in the conventional group (risk ratio 1.0; 95% confidence interval 0.8,
1.2). Deaths (4 vs. 0) and strokes (8 vs. 1) were more common in the intensive
treatment group. Outcomes appeared similar for patients with and without
previously diagnosed diabetes.
Conclusions. Intensive intraoperative insulin therapy did not confer
additional benefit compared with conventional intraoperative management in
patients undergoing cardiac surgery.
|
COMMENTARY
|
|---|
Gandhi et al. found no additional benefit of intensive intraoperative
glucose control through insulin infusion compared with conventional
intraoperative management in patients receiving intensive postoperative
glucose control. The risk ratio for the combined outcome of interest (death,
sternal infection, prolonged mechanical ventilation, arrhythmia, stroke, or
renal failure within 30 days) was 1.0, with a confidence interval between 0.8
and 1.2. Although the trial was not large enough to exclude a small
possibility of a beneficial effect, the findings of this trial suggest that a
large benefit is unlikely. In addition, two of the most important constituents
of the main outcome, deaths and strokes, were increased in the intensive
therapy group. This finding could have occurred by chance, but it makes the
possibility even smaller that the net benefits of intensive intraoperative
therapy exceed the downsides. It is important to recognize that the negative
results of this study should not suggest that postoperative intensive glucose
management in the intensive care unit is ineffective, because both groups
received such intervention.
|
Footnotes
|
|---|
Michael Pignone, MD, MPH, is an associate professor of medicine at the
University of North Carolina Department of Medicine in Chapel Hill. He is also
an associate editor of Clinical Diabetes.
 CiteULike  Del.icio.us  Digg  Reddit  Technorati What's this?
|
|
|
Top
STUDY
SUMMARY