Cardiopulmonary bypass with hypothermia may induce alterations in insulin secretion and resistance and administration of cardioplegia can constitute a glucose load, both of which can lead to severe hyperglycemia during cardiac surgery. Further, hyperglycemia can occur in both diabetic and non-diabetic patients. Patients undergoing cardiac surgery are at risk for neurological dysfunction (both overt and subtle) and hyperglycemia has been shown to worsen neurological outcome after focal and global cerebral ischemia. Controversy exist as to how aggressively hyperglycemia should be treated, especially in a previously normoglycemic patient. Common sense might suggest that treatment of hyperglycemia could be accomplished with insulin administration during surgery as needed when glucose levels rise. However, this is difficult to accomplish once patients become cool. Further, there is much inter-patient variation in how high and when glucose levels rise.

The authors describe a study in which 20 non-diabetic patients were randomized to two treatment groups. One group (N=10) was treated with an IV insulin infusion protocol for "tight" control of glucose levels during surgery and one group (N=10) had no treatment of glucose levels during surgery. Surprisingly, despite therapy in one group, both groups had similar significant increases in glucose level during surgery. Disturbingly, the treatment group suffered more hypoglycemia after surgery with 40% of the treatment group needing glucose administration in the ICU. The authors concluded that normoglycemia during cardiac surgery was unattainable with their insulin infusion protocol.

In addition to investigating glucose control, the authors also investigated the rise in serum markers thought to correlate with subtle neurological injury. These were CPK-BB and S-100 protein. Interestingly, both groups had similar increases in these markers suggesting that treatment of glucose level intraoperatively did not affect this surrogate marker of neurological function.

This is an interesting study because it addresses a common problem for anesthesiologists taking care of patients undergoing cardiac surgery with hypothermic cardiopulmonary bypass. This study suggests that aggressive intraoperative treatment of glucose level in a previously normoglycemic patient may be both futile (no difference in intraoperative glucose levels between groups) and potentially dangerous (postoperative hypoglycemia in treatment group vs. control). The mechanisms causing abnormal glucose regulation during cardiac surgery are not well understood and a method to effectively prevent a rise in glucose level may not be attainable until the mechanism is elucidated completely. Of note, the trend to "tepid" or "warm" cardiopulmonary bypass may result in less abnormal glucose levels during surgery and may lead to a patient in whom administration of insulin may be more effective. This remains to be fully investigated.

A major limitation of this study is the small sample size. Further, no formal neurological testing was done to document differences in neurological function as the result of attempted glucose control. This study would also be interesting to repeat in diabetic patients (especially non-insulin dependent diabetic patients), who represent a significant percentage of patients presenting for cardiac surgery.