I have read with great interest the comments of Drs. Liu and Xia. Their study on a similar model of streptozotocine-induced diabetic rats1confirmed my findings that myocardial signal transducer and activator of transcription 3 (STAT3) concentrations are significantly decreased in the diabetic heart, whereas normal STAT3 concentrations are essential for initiation of the protective process of sevoflurane postconditioning (postC) in diabetes. They also described their findings that STAT3 concentrations might recover in diabetic hearts in the presence of N -acetylcysteine. They suggest that N -acetylcysteine and allupurinol promote tolerance of the myocardium to ischemic injury by restoring adinopectin concentration, and thus, they propose that N -acetylcysteine might be effective in restoring sevoflurane postC in the diabetic heart.
The animal model used in the study2was type I diabetes, i.e. , uncorrected hyperglycemia for 1 month. Insulin was given in the last 48 h to bring the animals into a state of normoglycemia at the time of the experiment. I did not give insulin therapy to alleviate the symptoms of diabetes, but to perform the experiments in a normoglycemic state, with a goal to reduce blood sugar to less than 135 mg/dl, but I actually achieved better control as reflected in the study. The diabetic irresponsiveness of the myocardium is more complex than that expected by the hyperglycemic state alone. It might be attributed to multiple cellular changes in mitochondrial membrane potentials, which lead to certain degrees of mitochondrial uncoupling.3This in turn leads to a glycation reaction, causing cellular injury and accumulation of glycation end products,4depression of more signal transduction components such as phosphatidylinositol 3-kinase, STAT3,2and nitric oxide synthase, as well as glycogen synthase kinase-3β activation,5all leading to lipid accumulation, inflammation, and remodeling.6Therefore, I do not agree with the comment that strict preservation of normoglycemia during the experiment might eliminate all changes accumulated during the month of diabetic state. The changes in the heart are too substantial to be prevented by a single maneuver of replenishing antioxidant state.
From a practical view, giving higher doses of insulin to an animal would be irresponsible, because hypoglycemia in animals who are not observed closely most of the time would be detrimental. I am not aware of any animal study evaluating diabetes for long-term outcome that used conventional glucose control levels. Extrapolating from human data that use conventional glucose control, as suggested in the current letter to the editor, on patient postoperative long-term outcome, is controversial at best. Strict control of diabetes at normal levels was found hazardous due to severe side effects.7,8Attributing changes in sensitive cellular pathways, such as STAT3 and phosphatidylinositol 3-kinase, only to the natural lack of antioxidants seems simplistic. As discussed in the previous paragraph, diabetes effects on the heart are profound and the mechanism of injury is at multiple levels.
The authors claim that a longer period of insulin therapy might have a beneficial effect on restoring postC in diabetes. Unfortunately, I suggested differently, that changes in the heart caused by diabetes were not restored by a short period of normoglycemia. Insulin administration before complete ischemia may carry the risk of enhanced cellular metabolism in the no-flow state, a situation that may bring about accumulation of high tissue levels of lactate, which inhibits glycolysis and prevents any glycolytic stimulation by insulin.9,10Therefore, until proved experimentally, I cannot agree with the authors' final statement that insulin administration before ischemia may produce beneficial effects.