1

Describe the pathophysiology of type 1 and type 2 diabetes mellitus.

DM is a metabolic disease arising from defects in insulin secretion, insulin responsiveness, or both. Insulin is normally released from the endocrine pancreas, specifically from pancreatic β cells, in response to increases in blood glucose. Insulin receptors are present on cell membranes of several tissue types. It is responsible for storing excess nutrients as glycogen in the liver, as fat in adipose tissue, and as protein in muscle. These stored nutrients are available during periods of fasting to maintain glucose delivery to the brain, muscle, and other organs.

The diagnostic criteria for DM are listed in Box 26-1 . Any of the listed criteria can establish the diagnosis of DM; it is not a requirement to meet all criteria. However, in the absence of hyperglycemia symptoms, the test is repeated before solidifying the diagnosis. In the acute setting, hyperglycemia symptoms include polyuria, polydipsia, weight loss, and blurred vision. The most extreme cases of acute hyperglycemia, diabetic ketoacidosis (DKA) and hyperglycemic hyperosmolar state (HHS), can be life-threatening (see Questions 6 and 7). Uncontrolled chronic hyperglycemia is detrimental to multiple organ systems.

Type 1 DM is caused by an absolute deficiency of insulin secretion; this accounts for only 5%–10% of cases. Classically, type 1 DM is caused by cell-mediated autoimmune destruction of pancreatic β cells. Immune-mediated DM commonly occurs in childhood and adolescence but can occur at any age. Individuals with type 1 DM require exogenous insulin administration for survival. In the absence of insulin administration, these individuals rapidly become symptomatic from hyperglycemia. Carbohydrate, fat, and protein metabolism are dramatically disturbed. Frequently, the first clinical presentation of a type 1 diabetic is DKA. As a result of significantly impaired glucose utilization, blood ketoacid concentration is elevated from increased lipolysis of fat and subsequent conversion of fatty acids to ketoacids.

Type 2 DM, which accounts for 90%–95% of the disease burden, is caused by a combination of tissue insulin resistance and inadequate compensatory insulin secretory response. In contrast to type 1 diabetics, these patients typically do not require exogenous insulin administration to survive, at least initially. The risk of developing type 2 DM increases with age, obesity, and inactivity. At earlier stages, it is asymptomatic and frequently goes undiagnosed for years as hyperglycemia develops gradually. However, these patients are already at risk for developing chronic microvascular and macrovascular complications associated with DM. Table 26-1 outlines the major differences between type 1 and type 2 DM.

Other rare causes of DM include genetic defects of the pancreatic β cells or genetic abnormalities in insulin action. Diffuse injury to the pancreas from pancreatitis, trauma, infection, cancer, or pancreatectomy can cause diabetes, if extensive. Acromegaly, Cushing syndrome, and pheochromocytoma can cause DM. Gestational DM, which affects 7% of all pregnancies, is glucose intolerance that is first recognized during pregnancy. Insulin resistance usually resolves postpartum, but these patients are at increased risk for developing type 2 DM later in life.

2

What are the end-organ effects of diabetes mellitus, and how do they affect the perioperative course?

DM affects nearly every organ system. Preoperative evaluation should focus on systems that are most relevant in the perioperative period.

3

Discuss the oral medications and insulin preparations available to treat diabetes mellitus and how they should be managed perioperatively.

Oral agents used in the treatment of DM are outlined in Table 26-2 . Insulin preparations for subcutaneous use are presented in Table 26-3 .

Metformin, an oral hypoglycemic agent that is typically the first-line treatment for newly diagnosed type 2 DM, warrants special mention because of the rare but potentially life-threatening side effect of lactic acidosis associated with this drug. The perioperative period may manifest with conditions leading to increased risk for lactic acidosis, including impaired renal or liver function, hemodynamic instability, and decreased tissue perfusion. Metformin should be held the morning of surgery, and use of metformin should be resumed only when renal function and circulatory status are stabilized postoperatively. Metformin is also discontinued before and for 24–48 hours after the administration of iodinated contrast material for radiologic examination.

When administered as a solo agent, metformin does not cause hypoglycemia. Other oral diabetic agents are associated with hypoglycemia, particularly the sulfonylureas. These medications should be held the morning of surgery and throughout the perioperative period, as long as the patient is fasting.

Type 1 and type 2 diabetics who are not adequately managed with oral agents typically receive basal-bolus insulin maintenance regimens. Basal insulin is meant to replace the patient’s baseline insulin that would be produced during fasting. The total daily dose of insulin is usually divided into a 50% basal component and a 50% prandial component, either continuously via an insulin pump or by subcutaneous injection of a long-acting insulin once (insulin glargine) or twice (insulin detemir) daily plus short-acting boluses with food. Basal insulin can be continued in the perioperative period because it covers the fasting state. However, when type 2 diabetics use long-acting insulins as their sole insulin, this is not meant to be basal dosing, and the dose should be reduced perioperatively to avoid hypoglycemia. Intermediate-acting agents, such as NPH (neutral protamine Hagedorn) and insulin lispro protamine, display peaks in activity (6 hours and 4–12 hours, respectively) that can cause hypoglycemia during fasting. These agents require dosing adjustments in the perioperative period. Patients are typically instructed to take half of their usual dose on the morning of surgery. Insulin pumps can be continued intraoperatively, usually for procedures lasting <2 hours, if secured away from the surgical field.

In-hospital glycemic management is best achieved with the use of insulin rather than oral diabetic agents. In the perioperative period, frequent changes in fasting status, unpredictable intestinal absorption of medications, and rapidly changing clinical conditions make the use of oral agents potentially less effective and unpredictable, and there is a risk for hypoglycemia. Correction of hyperglycemia in the perioperative period can be accomplished with subcutaneous doses of ultra-rapid-acting insulin or regular insulin for ambulatory patients or patients in the postanesthesia care unit who will be admitted to a regular nursing floor. Rough estimates for dosing are 1–4 units of insulin per 50 mg/dL decrease in glucose desired. Outside of the immediate perioperative period, it is important to include a basal insulin component meant to suppress gluconeogenesis between meals. This is particularly essential for type 1 diabetics, who without basal insulin can rapidly develop DKA.

For critically ill patients, glycemic control is best achieved with intravenous insulin infusions. These patients can have unpredictable subcutaneous absorption of insulin because of edema, hypotension, and peripheral vasoconstriction. The same is true for patients undergoing procedures where large fluid shifts or hemodynamic lability are expected because skin perfusion can vary greatly under these conditions.

4

What impact does hyperglycemia have on perioperative morbidity and mortality?

Hyperglycemia is common in the perioperative period both in diabetics and in nondiabetics. The neuroendocrine stress response to surgery causes release of counterregulatory hormones glucagon, epinephrine, and cortisol, which inhibit insulin secretion, increase insulin resistance, mobilize glycogen, and increase gluconeogenesis. The severity of insulin resistance and resulting hyperglycemia are directly related to the degree of surgical trauma. Hyperglycemia is especially common after cardiac surgery and major abdominal surgery. It is also more common for procedures lasting a long duration and in open procedures more so than laparoscopic procedures. Inhaled anesthetics contribute to perioperative hyperglycemia by depressing insulin secretion in response to increasing blood glucose levels. Perioperative steroid administration further exacerbates the propensity toward hyperglycemia.

Hyperglycemia in the perioperative period is associated with increased risk of infection secondary to impaired leukocyte function—specifically, impaired chemotaxis, phagocytosis, and intracellular bacterial killing. It is also associated with impaired collagen synthesis and decreased nitric oxide production, reducing local perfusion and delaying wound healing. Vascular reactivity can be altered with increased levels of angiotensin II and enhanced systemic vascular resistance. Elevated glucose levels are also associated with renal injury, pulmonary complications, myocardial infarction, cerebrovascular insult, longer hospital and intensive care unit (ICU) stays, and increased mortality. These adverse outcomes are found to be more prevalent in patients without a previous diagnosis of DM who develop hyperglycemia perioperatively than in patients with a known history of DM.