Perioperative Hyperglycemia Raises Risks Inflammation/Hormones Increase Adverse Outcomes
Ioanna Apostilidou, MD, and Richard C. Prielipp, MD
Hyperglycemia and glucose intolerance are common manifestations of perioperative stress in many hospitalized patients. Diabetic patients have more frequent, more prolonged, and more expensive hospital admissions that result in increased morbidity and mortality than nondiabetics. Diabetic patients also require more frequent surgical interventions and are more often admitted to the intensive care unit (ICU). Moreover, it is common for even nondiabetic surgical and ICU patients to develop acute hyperglycemia during stress. This hyperglycemia is mediated by the release of proinflammatory cytokines (e.g., TNF-alpha and IL-6) and elevated concentrations of catecholamines, growth hormone, glucagon, and glucocorticoids. These mediators induce metabolic alterations in carbohydrate balance that alter peripheral glucose uptake and utilization, increase gluconeogenesis, depress glycogenesis, and induce glucose intolerance and insulin resistance.
Hyperglycemia produces deleterious effects on the immune system, neutrophil function, and on the response to endotoxin. As a consequence, acute hyperglycemia adversely affects patient outcomes. Diabetic patients undergoing cardiac surgery managed with tight perioperative glycemic control have a lower rate of sternal wound infection and hospital mortality.24 In a large nonrandomized study, 2,467 diabetic cardiac surgical patients were classified in 2 sequential groups, the control group with usual sliding scale insulin glucose control and the study group with continuous intravenous insulin infusion to maintain blood glucose <200 mg/dL.2 Continuous insulin infusion resulted in lower glucose levels and was associated with significantly lower incidence of sternal wound infection (0.8 vs. 2%) and lower postoperative mortality (2.5 vs. 5.3%). In a subsequent analysis of 4,864 diabetic patients who underwent open-heart procedures, the investigators reported that a 3-day continuous insulin infusion that kept glucose levels <150 mg/dL was a key factor in improved outcomes.4 Modulation of the metabolic state during cardiac ischemia and inhibition of lipolysis by insulin stimulates nitric oxide production and may confer cardiac protection. For instance, in a prospective randomized study of 141 coronary artery bypass graft (CABG) patients, Lazar and colleagues found that tight glycemic control (serum glucose, 125200 mg/dL) decreased the incidence of recurrent wound infections, episodes of recurrent ischemia, atrial fibrillation, and postoperative length of stay.5 Outcome in patients without diabetes undergoing cardiac surgery also improved with tight glycemic control.69 An increase of only
20 mg/dL in the mean intraoperative glucose was linked to an increase of more than 30% in adverse outcomes.8
ICU and Similar Patient Groups
Numerous prospective, randomized trials confirm that maintenance of normoglycemia in critically ill patients (plasma glucose between 80110 mg/dL) improves ICU outcomes.614 Euglycemia can be achieved in ICU patients with insulin infusion protocols and reduces
- <LI class=style1>ICU mortality (32%) <LI class=style1>in-hospital mortality (34%) <LI class=style1>serious infections rate <LI class=style1>onset of acute renal failure <LI class=style1>neuropathy
- duration of ventilatory dependence.10,11
While these benefits are more difficult to document in medical ICU patients,12 it is clear that appropriate use of insulin decreases complications from hyperglycemia associated with the response to acute disease, with or without a direct impact on the primary disease process itself.1214
Other patients with acute illness and hyperglycemia are also at risk. The Diabetes and Insulin-Glucose Infusion in Acute Myocardial Infarction (DIGAMI 1) study revealed that intensive glycemic control during the peri-infarction period reduced long-term mortality rate (1 year, 28%; 3.4 years, 25%).15 That benefit was evident regardless of the antidiabetic regimen used (DIGAMI 2) emphasizing the importance of maintaining euglycemia.16 Acute stroke patients have higher mortality rates and poorer recovery when blood glucose exceeds 110 mg/dL.17 Thus, evidence supports the use of aggressive insulin protocols to manage hyperglycemia in patients admitted to acute care hospitals for myocardial infarction, stroke, those with a previous diagnosis of diabetes, and those patients undergoing surgery.18,19
Management Caveats
Tight glucose control demands frequent measurement (at least hourly initially) of glucose concentration and a consistent approach to management. Ideally, a glucose control protocol must fulfill these criteria:
- <LI class=style1>Ability to make rapid, precise, consistent modifications in blood sugar <LI class=style1>Ability to maintain, increase, or decrease blood sugar depending on clinical situation
- Ability to monitor glucose levels quickly, close to real time with trend detection to allow preemptive glucose management. (See the appendix below for a protocol example from the University of Minnesota.)
The risk of hypoglycemia and difficulty of attaining normoglycemia with a tight glycemic control protocol is an important safety concern in both cardiac and other ICU patients.20 In 2 recent studies, a novel approach, the
hyperinsulinemic normoglycemic clamp technique, achieved normoglycemia even during especially high stress such as cardiac surgery. This technique involves a fixed, relatively high-dose infusion of insulin and then uses a variable rate of glucose infusion to clamp the blood glucose concentration at an appropriate level.2123 However, this methodology is incredibly labor and time intensive, too.
Although the methodology for administering insulin and glucose may be debated, the clinical end-point is not. The American College of Endocrinology position statement recommends maintaining blood glucose ≤110 mg/dL (<6.1 mM) in intensive care patients to decrease perioperative complications and in-hospital morbidity and mortality.24 Most insulin protocols for ICU patients target glucose levels in the physiologic range of 80110 mg/dL.1012 However, we still need to elucidate the exact biochemical mechanisms by which the benefit of normoglycemia is actually conferred.6 Indeed, although insulin is the primary agent available to lower blood sugar, recently available pharmacologic agents, such as the incretin mimetics, amylin and exenatide, which can actually lower glucagon release, may confer metabolic advantages distinct from insulin treatment alone. Other strategies to ameliorate the perioperative stress response in surgical patients include interventions like epidural or spinal blockade to reduce catecholamine secretion and improve insulin responsiveness.
In summary, we believe that whenever hyperglycemia and/or insulin resistance occur, early detection and effective insulin therapy is indicated. Clearly, the potential of hypoglycemia remains the most serious safety issue. Recent clinical reports suggest hypoglycemia may be associated with multiple factors, including misunderstanding of the insulin administration protocol, rebound response from concomitant intravenous bolus of corticosteroids, and other complex insulin and drug-patient interactions. Therefore, there is intense interest in continuous glucose level monitoring technology, which promises a means of avoiding, undiagnosed and untreated hypoglycemia. We also await the findings of additional important clinical studies regarding these issues.25
Dr. Apostolidou is Associate Professor of Anesthesiology at the University of Minnesota in Minneapolis, MN. Dr. Prielipp is Professor and Chair of the Department of Anesthesiology at the University of Minnesota in Minneapolis, MN. Dr. Prielipp is also Chair of the APSF Committee on Education and Training and a member of the APSF Executive Committee.
, then why not just put that same patient to sleep and try intubating asleep without paralytic before going the whole nine yards?
