The history and physical examination most accurately predict the risks of anesthesia and the likelihood of required changes in monitoring or therapy.
For diabetic patients, end-organ dysfunction and the degree of glucose control in the perioperative and periprocedural periods are the critical issues with regard to risk.
The keys to managing blood glucose levels in diabetic patients perioperatively are to set clear goals and then monitor blood glucose levels frequently enough to adjust therapy to achieve these goals.
Obesity is associated with multiple comorbid conditions, including diabetes, hyperlipidemia, and cholelithiasis, but the primary concern is derangements of the cardiopulmonary system.
Obstructive sleep apnea is important to recognize because of the increased sensitivity to and the consequence of the depressing effects of hypnotics and opioids on airway muscle tone and respiration, as well as the difficulty with laryngoscopy and mask ventilation.
Although no controlled, randomized prospective clinical studies have been performed to evaluate the use of adrenergic receptor blocking drugs in patients undergoing resection of pheochromocytoma, the preoperative use of such drugs is generally recommended.
For patients with hypertension, the routine administration of all drugs preoperatively is recommended, except angiotensin-converting enzyme inhibitors and angiotensin II antagonists.
Evaluation of a patient with cardiovascular disease depends on clinical risk factors, the extent of surgery, and exercise tolerance.
In patients with pulmonary disease, the following should be assessed: dyspnea, coughing and the production of sputum, recent respiratory infection, hemoptysis, wheezing, previous pulmonary complications, smoking history, and physical findings.
In patients with pulmonary disease, several strategies have been suggested, including cessation of smoking 8 weeks or more preoperatively.
Risk factors for perioperative renal dysfunction include advanced age, congestive heart failure, previous myocardial revascularization, diabetes, and increased baseline blood creatinine concentration.
One of the primary objectives for a patient with renal disease is ensuring that the renal dysfunction is not augmented and thereby increasing the chance for renal failure, coma, and death.
Mild perioperative anemia may be clinically significant only in patients with ischemic heart disease.
Careful management of long-term drug administration should include questions about the effects and side effects of alternative as well as prescription drugs.
This chapter reviews many conditions requiring special preoperative and preprocedure evaluation, intraoperative or intraprocedure management, or postprocedure care. Patients undergoing surgical procedures move through a continuum of medical care to which a primary care physician, an internist or pediatrician, an anesthesiologist, and a surgeon, gastroenterologist, radiologist, or obstetrician-gynecologist contribute to ensure the best outcome possible. It may also involve comanagement with a hospitalist. No aspect of medical care requires greater cooperation among physicians than does performance of a surgical operation or a complex procedure involving multiple specialists and the perioperative care of a patient. Moreover, nowhere else can counseling make so huge a difference in so many lives. The preoperative evaluation also represents a time when education on tobacco cessation, physical inactivity, brain health, and poor food choices can be discussed. The importance of integrating physicians’ expertise is even greater within the context of the increasing life-span of our population. As the number of older adults and very old adults (those >85 years old) grows, so does the need of surgical patients for preoperative consultation to help plan for comorbidity, frailty, and multiple drug regimens, the knowledge of which is crucial to successful patient management. At a time when medical information is encyclopedic, it is difficult, if not impossible, for even the most conscientious anesthesiologist to keep abreast of the medical issues relevant to every aspect of perioperative or periprocedure patient management. This chapter reviews such issues with primary emphasis on the anesthesiologist providing preoperative evaluation and care, rather than transferring these responsibilities to other providers.
As with “healthy” patients, the history and physical examination most accurately predict not only the associated risks but also the likelihood of whether a monitoring technique, change in therapy, or “prehabilitation” will be beneficial or necessary for survival. This chapter emphasizes instances in which specific information should be sought in history taking, physical examination, or laboratory evaluation. Although controlled studies designed to confirm that optimizing a patient’s preoperative or preprocedure physical condition would result in a less frequent rate of morbidity have not been performed for most diseases, it is logical to assume that such is the case. That such preventive measures would cost less than treating the morbidity that would otherwise occur is an important consideration in a cost-conscious environment.
Minimally invasive procedures such as cataract extraction, magnetic resonance imaging (MRI), or diagnostic arthroscopy, performed in conjunction with the best current anesthetic practices, may pose no greater risk than daily living and thus may not be considered an opportunity for special evaluation. Nevertheless, the preoperative evaluation may identify conditions that could change perioperative management and that may improve both throughput of surgery and the speed of recovery. Examples include the following: ensuring the administration of long-term medications such as a β-adrenergic blocking drug, aspirin for patients with coronary stents, or a statin (or any combination); administering a histamine type 2 (H 2 ) antagonist 1 to 2 hours before entry into the operating room; ensuring the availability of equipment to measure blood glucose levels; obtaining a history of the patient’s diabetic course and treatment from the primary care physician, as well as from the patient; and performing a fiberoptic laryngoscopic examination or procuring additional skilled attention.
The following conditions are discussed in this chapter:
Diseases involving the endocrine system and disorders of nutrition (discussed first because of its increasing importance to care)
Diseases involving the cardiovascular system
Disorders of the respiratory and immune system
Diseases of the central nervous system (CNS), neuromuscular diseases, and mental disorders
Diseases involving the kidney, infectious diseases, and disorders of electrolytes
Diseases involving the gastrointestinal (GI) tract or the liver
Diseases involving hematopoiesis and various forms of cancer
Diseases of aging or those that occur more commonly in older adults, as well as chronic and acute medical conditions requiring drug therapy
Role of the Primary Care Physician or Consultant
The roles of the primary care physician or consultant are not to select and suggest anesthetic or surgical methods but rather to optimize the patient’s preoperative and preprocedure status regarding conditions that increase the morbidity and mortality associated with surgery and to alert the anesthesia care team about these conditions. Within the context of shared decision making, the primary care physician may also be involved in the decision to proceed with surgery.
Quotations and a box in a Medical Knowledge Self-Assessment Program published by the leading organization representing internists, the American College of Physicians, highlight this role for the consultant :
Effective interaction with colleagues in other specialties requires a thorough grounding in the language and science of these other disciplines as well as an awareness of basic guidelines for consultation [ Box 32.1 ]. The consulting internists’ role in perioperative care is focused on the elucidation of medical factors that may increase the risks of anesthesia and surgery. Selecting the anesthetic technique for a given patient, procedure, surgeon, and anesthetist is highly individualized and remains the responsibility of the anesthesiologist rather than the internist.
Complete a prompt, thorough, generalist-oriented evaluation.
Respond specifically to the question or questions posed.
Indicate clearly the perioperative importance of any observations and recommendations outside the area of initial concern.
Provide focused, detailed, and precise diagnostic and therapeutic guidance.
Emphasize verbal communication with the anesthesiologist and surgeon, particularly to resolve complex issues.
Avoid chart notations that unnecessarily create or exacerbate regulatory or medicolegal risk.
Use frequent follow-up visits in difficult cases to monitor clinical status and compliance with recommendations.
Optimizing a patient’s preoperative and preprocedure condition and, in settings with a preoperative clinic, counseling a patient about needed future lifestyle changes such as exercise, food choices, and tobacco cessation are cooperative ventures between the anesthesiologist and the internist, pediatrician, surgeon, or family physician. If available, the primary care physician should affirm that the patient is in the very best physical state attainable (for that patient), or the anesthesiologist and primary care physician should do what is necessary to optimize that condition. Although not yet definitively proven, prehabilitation prior to surgery has been advocated by many groups.
Primary care physicians can prepare and treat a patient to provide optimal conditions for daily life. The preoperative clinic should collaborate with the primary care physician to start the process of preparing the patient for the needs of surgery or complex procedures. Although such education is more readily available and of better quality than in previous decades, and although cardiologic organizations have provided considerable data on the importance of this aspect of care, the primary care physician’s training, knowledge, and ability may not include an in-depth understanding of the perioperative evaluation. Without understanding the physiologic changes that occur perioperatively, appropriate therapy is difficult to prescribe. It is therefore part of the anesthesiologist’s job to guide the patient’s consultants about the type of information needed from the preoperative and preprocedure consultation.
Diseases Involving the Endocrine System and Disorders of Nutrition
Preoperative and Preprocedure Diabetes Mellitus
Diabetes mellitus is a heterogeneous group of disorders that have the common feature of a relative or absolute insulin deficiency. The disease is characterized by a multitude of hormone-induced metabolic abnormalities, diffuse microvascular lesions, and long-term end organ complications. The diagnosis of diabetes is made with a fasting blood glucose level greater than 110 mg/dL (6.1 mmol/L), and impaired glucose tolerance is diagnosed if the fasting glucose level is less than 110 mg/dL (6.1 mmol/L) but greater than 100 mg/dL (5.5 mmol/L). Diabetes can be divided into two very different diseases that share the same long-term end-organ complications. Type 1 diabetes is associated with autoimmune diseases and has a concordance rate of 40% to 50% (i.e., if one of a pair of monozygotic twins had diabetes, the likelihood that the other twin would have diabetes is 40%-50%). In type 1 diabetes, the patient is insulin deficient, principally from autoimmune destruction of the pancreatic β cells, and susceptible to ketoacidosis if insulin is withheld. Type 2 diabetes has a concordance rate approaching 80% (i.e., genetic material is both necessary and sufficient for the development of type 2 diabetes). [CR] How markedly the aging and end-organ effects of these genes are expressed is based on lifestyle choices of food and physical activity. These patients are not susceptible to the development of ketoacidosis in the absence of insulin, and they have peripheral insulin resistance through multiple defects with insulin action and secretion. Patients with non–insulin-dependent (type 2) diabetes account for the majority (>90%) of the diabetic patients in Europe and North America. These individuals tend to be overweight, relatively resistant to ketoacidosis, and susceptible to the development of a hyperglycemic-hyperosmolar nonketotic state. Plasma insulin levels are normal or increased in type 2 diabetes but are relatively low for the level of blood glucose. This hyperinsulinemia by itself is postulated to cause accelerated cardiovascular disease. Gestational diabetes develops in more than 3% of all pregnancies and increases the risk of developing type 2 diabetes by 17% to 63% within 15 years.
Type 1 and type 2 diabetes differ in other ways as well. Contrary to long-standing belief, a patient’s age does not allow a firm distinction between type 1 and type 2 diabetes; type 1 diabetes can develop in an older person, and clearly, type 2 diabetes can develop in overweight children. Type 1 diabetes is associated with a 15% prevalence of other autoimmune diseases, including Graves disease, Hashimoto thyroiditis, Addison disease, and myasthenia gravis.
Over the next decade, the prevalence of diabetes is expected to increase by 50%. This growth is primarily the result of the increase in type 2 diabetes caused by excessive weight gain in adults and now also in the pediatric population. Large clinical studies show that long-term, strict control of blood glucose levels and arterial blood pressure, along with regular physical activity, results in a major delay in microvascular complications and perhaps indefinite postponement of type 2 diabetes in patients.
The common administered drugs can be classified into eight major groups: acarbose, biguanides (e.g., metformin), dipeptidyl peptidase-4 inhibitors (e.g., sitagliptin, saxagliptin, vildagliptin), glucagon-like peptide-1 receptor agonists (e.g., albiglutide, dulagutide, or exenatide), meglitinide (e.g., repaglinide or nateglinide), sodium-glucose transport protein 2 inhibitors (e.g., canagliflozin or empagliflozin), sulfonylureas (e.g., glibenclamide, glipizide, glimepiride, gliquidone), and thiazolidinediones (e.g., pioglitazone or rosiglitazone). [CR]
Patients with insulin-dependent diabetes tend to be younger, nonobese, and susceptible to the development of ketoacidosis. Plasma insulin levels are low or un-measurable, and therapy requires insulin replacement. Patients with insulin-dependent diabetes experience an increase in their insulin requirements in the post-midnight hours, which may result in early morning hyperglycemia (dawn phenomenon). This accelerated glucose production and impaired glucose use reflect nocturnal surges in secretion of growth hormone (GH). Physiologically normal patients and diabetic patients taking insulin have steady-state levels of insulin in their blood. Absorption of insulin is highly variable and depends on the type and species of insulin, the site of administration, and subcutaneous blood flow. Nevertheless, attainment of a steady state depends on periodic administration of the preparations received by the patient. Thus it seems logical to continue the insulin combination perioperatively that the patient had been receiving after assessing previous blood glucose control. [CR]
The major risk factors for diabetic patients undergoing surgery are the end-organ diseases associated with diabetes: cardiovascular dysfunction, renal insufficiency, joint collagen tissue abnormalities (limitation in neck extension, poor wound healing), inadequate granulocyte production, neuropathies, and infectious complications. Thus a major focus of the anesthesiologist should be the preoperative and preprocedure evaluation and treatment of these diseases to ensure optimal preoperative and preprocedure conditions. Poor preoperative glucose control, as measured by the hemoglobin A 1C (glycosylated hemoglobin) level, is an independent predictor of worse perioperative outcome.
Long-term tight control of blood glucose has been motivated by concern for three potential glucotoxicities, in addition to the results from major randomized outcome studies involving diabetic patients.
Glucose itself may be toxic because high levels can promote nonenzymatic glycosylation reactions that lead to the formation of abnormal proteins. These proteins may weaken endothelial junctions and decrease elastance, which is responsible for the stiff joint syndrome (and difficult intubation secondary to fixation of the atlanto-occipital joint), as well as decrease wound-healing tensile strength.
Glycemia also disrupts autoregulation. Glucose-induced vasodilation prevents target organs from protecting against increases in systemic blood pressure. A glycosylated hemoglobin level of 8.1% is the threshold at which the risk for microalbuminuria increases logarithmically. A person with type 1 diabetes who has microalbuminuria of greater than 29 mg/day has an 80% chance of experiencing renal insufficiency. The threshold for glycemic toxicity differs for various vascular beds. For example, the threshold for retinopathy is a glycosylated hemoglobin value of 8.5% to 9.0% (12.5 mmol/L or 225 mg/dL), and that for cardiovascular disease is an average blood glucose value of 5.4 mmol/L (96 mg/dL). Thus different degrees of hyperglycemia may be required before different vascular beds are damaged. Another view is that perhaps severe hyperglycemia and microalbuminuria are simply concomitant effects of a common underlying cause. For instance, diabetic patients in whom microalbuminuria develops are more resistant to insulin, insulin resistance is associated with microalbuminuria in first-degree relatives of patients with type 2 diabetes, and persons who are normoglycemic but subsequently have clinical diabetes are at risk for atherogenesis before the onset of disease.
Diabetes itself may not be as important to perioperative outcome as are its end-organ effects. Epidemiologic studies segregated the effects of diabetes itself on the organ system from the effects of the complications of diabetes (e.g., cardiac, nervous system, renal, and vascular disease) and the effects of old age and the accelerated aging that diabetes causes. Even in patients requiring intensive care unit (ICU) management, long-standing diabetes does not appear to be as important an issue as the end-organ dysfunction that exists and the degree of glucose control in the perioperative or periprocedure and ICU periods. 6b,8-13
The World Health Organization’s surgical safety checklist bundle suggests control with a target perioperative blood glucose concentration of 6 to 10 mmol/L (acceptable range, 4-12 mmol/L) or 100 to 180 mg/dL. Poor perioperative glycemic control has a significant impact on the risk of postoperative infection across a variety of surgical specialities. Different regimens permit almost any degree of perioperative control of blood glucose levels, but the tighter the control desired, the greater the risk of hypoglycemia. Therefore, debate regarding optimal control during the perioperative period has been extensive. Tight control retards all these glucotoxicities and may have other benefits in retarding the severity of diabetes itself. Management of intraoperative glucose may be influenced by specific situations, such as the following: the type of operation, pregnancy, expected global CNS insult, the bias of the patient’s primary care physician, or the type of diabetes.
Much of the research on perioperative control is derived from studies in the ICU, as opposed to the operating room. The first major trial demonstrating the benefit of tight glucose control was in medical ICU patients in Leuven, Belgium. The most recent trial was from the NICE-SUGAR (Normoglycemia in Intensive Care Evaluation and Survival Using Glucose Algorithm Regulation) group. In this randomized controlled trial, the investigators examined the associations between moderate and severe hypoglycemia (blood glucose, 41-70 mg/dL [2.3-3.9 mmol/L] and ≤40 mg/dL [2.2 mmol/L], respectively) and death among 6026 critically ill patients in ICUs. Intensive glucose control leads to moderate and severe hypoglycemia, both of which are associated with an increased risk of death. The association exhibits a dose-response relationship and is strongest for death from distributive shock. The optimal perioperative management has been reviewed elsewhere. Guidelines have been developed on the use of insulin infusions in the critical care unit to achieve these goals ( Table 32.1 ).
|Society, Guideline||Patient Group||Trigger Blood Glucose Value to Start Insulin Infusion (mM [mg/dL])||Target range, (mM [mg/dL])||Rationale|
|Society of Critical Care Medicine’s clinical practice guideline||General recommendation||8.3 (150)||5.6-8.3 (100-150)|
|Cardiac surgical patients||<8.3 (150)||Decreased risk for deep sternal wound infection and death|
|Critically ill trauma patients||8.3 (150)||<10 (180)|
|Patients with traumatic brain injury||8.3 (150)||<10 (180)|
|Neurologic ICU patients |
Aneurysmal subarachnoid hemorrhage
|8.3 (150)||<10 (180)|
|American Diabetes Association guidelines||General recommendation||10 (180)||7.8-10 (140-180)|
|Adaptation||6.1-7.8 (110-140)||Adjust to lower target range in documented low rate of severe hypoglycemia|
|American Association of Clinical Endocrinologists||General recommendation||7.8-10 (140-180)|
|Surgical patients||Lower range||Only in units showing low rates of hypoglycemia|
|Surviving Sepsis Campaign||General recommendation||10 (180)||<10 (180)||Based on the NICE-SUGAR study|
|Clinical Practical Guideline from the American College of Physicians||General recommendation||7.8-11.1 (140-200)||If insulin infusion is applied; however, guideline does not recommend intensive insulin therapy|
|Spanish Society of Intensive Care Medicine and Coronary Units||General recommendation||<8.3 (150)|
|French Society of Anesthesia and Intensive Care||General recommendation||10 (180)|
|Surgical patients||<6.1 (110)|
|Cardiac patients||<6.1 (110)|
|Society of Thoracic Surgeons||Cardiac surgical patients||<10 (180) except <8.3 (150) for those with devices in place|
Diabetes and Accelerated Physiologic Aging
Adverse perioperative outcomes have repeatedly and substantially correlated with the age of the patient, and diabetes does cause physiologic aging. When one translates the results of the Diabetes Control and Complications Trials into age-induced physiologic changes, a patient with type 1 diabetes who has poor control of blood glucose ages approximately 1.75 years physiologically for every chronologic year of the disease and 1.25 years if blood glucose has been controlled tightly. A patient with type 2 diabetes ages approximately 1.5 years for every chronologic year of the disease and approximately 1.06 years with tight control of blood glucose and blood pressure. Thus when providing care for a diabetic patient, one must consider the associated risks to be those of a person who is much older physiologically; the physiologic age of a diabetic patient is considerably older than that person’s calendar age just by virtue of having the disease.
Obesity and lack of physical exercise seem to be major contributors to the increasing prevalence of type 2 diabetes. As with type 1 diabetes, tight control of blood glucose, increased physical activity, and reduction in weight appear to reduce the accelerated aging associated with type 2 diabetes, and possibly delay the appearance of the disease and aging from it substantially. Although such a reduction in aging should reduce the perioperative risk for diabetic patients, no controlled trials have confirmed this theory.
The key to managing blood glucose levels perioperatively in diabetic patients is to set clear goals and then monitor blood glucose levels frequently enough to adjust therapy to achieve these goals. [CR]
Other Conditions Associated With Diabetes
Diabetes is associated with microangiopathy (in retinal and renal vessels), peripheral neuropathy, autonomic dysfunction, and infection. Diabetic patients are often treated with angiotensin-converting enzyme (ACE) inhibitors, even in the absence of gross hypertension, in an effort to prevent the effects of disordered autoregulation, including renal failure.
Preoperatively, assessment and optimization of treatment of the potential and potent end-organ effects of diabetes are at least as important as assessment of the diabetic patient’s current overall metabolic status. The preoperative evaluation of diabetic patients is also discussed in Chapter 31 .
The presence of autonomic neuropathy likely makes the operative period more hazardous and the postoperative period crucial to survival. Evidence of autonomic neuropathy may be routinely sought before the surgical procedure. Patients with diabetic autonomic neuropathy are at increased risk for gastroparesis (and consequent aspiration of gastric contents) and for perioperative cardiorespiratory arrest. Diabetic patients who exhibit signs of autonomic neuropathy, such as early satiety, lack of sweating, lack of pulse rate change with inspiration or orthostatic maneuvers, and impotence, have a very frequent incidence of painless myocardial ischemia. Administration of metoclopramide, 10 mg preoperatively to facilitate gastric emptying of solids, may be helpful ( Fig. 32.1 ). Interference with respiration or sinus automaticity by pneumonia or by anesthetic agents, pain medications, or sedative drugs is likely the precipitating cause in most cases of sudden cardiorespiratory arrest. Measuring the degree of sinus arrhythmia or beat-to-beat variability provides a simple, accurate test for significant autonomic neuropathy. The difference between the maximum and minimum heart rate on deep inspiration is normally 15 beats/min, but it is 5 beats/min or less in all patients who subsequently sustain cardiorespiratory arrest.
Other characteristics of patients with autonomic neuropathy include postural hypotension with a decrease in arterial blood pressure of more than 30 mm Hg, resting tachycardia, nocturnal diarrhea, and dense peripheral neuropathy. Diabetic patients with significant autonomic neuropathy may have impaired respiratory responses to hypoxia and are particularly sensitive to the action of drugs that have depressant effects. These patients may warrant continuous cardiac and respiratory monitoring for 24 to 72 hours postoperatively, although this has not been tested in a rigorous, controlled trial. In the absence of autonomic neuropathy, outpatient surgery is preferred for a diabetic patient if possible (see Table 32.1 ). [CR]
Many diabetic patients requiring emergency surgery for trauma or infection have significant metabolic decompensation, including ketoacidosis. Frequently, little time is available to stabilize the patient, but even a few hours may be sufficient for correction of any fluid and electrolyte disturbances that are potentially life-threatening. It is futile to delay surgery in an attempt to eliminate ketoacidosis completely if the underlying surgical condition will lead to further metabolic deterioration. The likelihood of intraoperative cardiac arrhythmias and hypotension resulting from ketoacidosis will be reduced if intravascular volume depletion and hypokalemia are at least partially treated. During the initial resuscitation phase of ketoacidosis bicarbonate should initially be avoided with crystalloid fluids, potassium repletion, and intravenous insulin therapy favored. [CR]
Insulin therapy is initiated with a 10-unit intravenous bolus of regular insulin, followed by continuous insulin infusion. The rate of infusion is determined most easily by dividing the last serum glucose value by 150 (or 100 if the patient is receiving steroids, has an infection, or is considerably overweight [body mass index ≥35]). The actual amount of insulin administered is less important than is regular monitoring of glucose, potassium, and arterial pH. The maximum rate of glucose decline is fairly constant, averaging 75 to 100 mg/dL/h, regardless of the dose of insulin because the number of insulin binding sites is limited. During the first 1 to 2 hours of fluid resuscitation, the glucose level may decrease more precipitously. When serum glucose reaches 250 mg/dL, the intravenous fluid should include 5% dextrose.
The volume of intravenously administered fluid required varies with the overall deficit; it ranges from 3 to 5 L and may be as large as 10 L. Despite losses of water in excess of losses of solute, sodium levels are generally normal or reduced. Factitious hyponatremia caused by hyperglycemia or hypertriglyceridemia may result in this seeming contradiction. The plasma sodium concentration decreases by approximately 1.6 mEq/L for every 100 mg/dL increase in plasma glucose greater than normal. Initially, balanced crystalloid solution is infused at a rate of 250 to 1000 mL/h, depending on the degree of intravascular volume depletion and cardiac status. Some measure of left ventricular volume should be monitored in diabetic patients who have a history of myocardial dysfunction. Approximately one third of the estimated fluid deficit is corrected during the first 6 to 8 hours and the remaining two thirds over the next 24 hours. [CR]
The degree of acidosis is determined by analysis of arterial blood gases and detection of an increased anion gap (see also Chapter 48 ). Acidosis with an increased anion gap (≥16 mEq/L) in an acutely ill diabetic patient may be caused by ketones in ketoacidosis, lactic acid in lactic acidosis, increased organic acids from renal insufficiency, or all three disorders. In ketoacidosis, plasma levels of acetoacetate, β-hydroxybutyrate, and acetone are increased. Plasma and urinary ketones can be measured semiquantitatively with Ketostix and Acetest tablets. The role of bicarbonate therapy in diabetic ketoacidosis is controversial, but could be considered in severe acidemia and hemodynamic instability as myocardial function and respiration are known to be depressed at a blood pH lower than 7.00 to 7.10. This careful consideration is because rapid correction of acidosis with bicarbonate therapy may result in alterations in CNS function and structure. These alterations may be caused by (1) paradoxical development of cerebrospinal fluid and CNS acidosis from rapid conversion of bicarbonate to carbon dioxide and diffusion of the acid across the blood-brain barrier, (2) altered CNS oxygenation with decreased cerebral blood flow, and (3) the development of unfavorable osmotic gradients. After treatment with fluids and insulin, β-hydroxybutyrate levels decrease rapidly, whereas acetoacetate levels may remain stable or even increase before declining. Plasma acetone levels remain elevated for 24 to 42 hours, long after blood glucose, β-hydroxybutyrate, and acetoacetate levels have returned to normal; the result is continuing ketonuria. Persistent ketosis with a serum bicarbonate level less than 20 mEq/L in the presence of a normal glucose concentration is an indication of the continued need for intracellular glucose and insulin for reversal of lipolysis.
The most important electrolyte disturbance in diabetic ketoacidosis is depletion of total-body potassium. Deficits range from 3 to 10 mEq/kg body weight. Serum potassium levels decline rapidly and reach a nadir within 2 to 4 hours after the start of intravenous insulin administration. Aggressive replacement therapy is required. The potassium administered moves into the intracellular space with insulin as the acidosis is corrected. Potassium is also excreted in urine because of the increased delivery of sodium to the distal renal tubules that accompanies volume expansion. Phosphorus deficiency in ketoacidosis as a result of tissue catabolism, impaired cellular uptake, and increased urinary losses may give rise to significant muscular weakness and organ dysfunction. The average phosphorus deficit is approximately 1 mmol/kg body weight; no clear guidance for replacement exists, but replacement is appropriate in patients with cardiac dysfunction, anemia, respiratory depression, or if the plasma phosphate concentration is less than 1.0 mg/dL.
Anticipated Newer Treatments of Diabetes
At least three major changes in the care of diabetic patients have made it to the clinical trial stage:
Implanted (like a pacemaker) glucose analyzer with electronic transmission to a surface (watch) monitor
New islet transplantation medication that makes islet cell transplants much more successful and rejection medication less hazardous
Medications such as INGAP (islet neogenesis–associated protein) peptide, which may cause regrowth of normally functioning islet cells (without the need for transplantation)
Some of these treatments may radically change the therapies used in the perioperative period. If regrowth of islet cells becomes common, type 1 diabetes could all but disappear; if implanted minute-to-minute glucose reading is possible, tight control may be much easier and more expected.
Insulinoma and Other Causes of Hypoglycemia
Hypoglycemia in persons not treated for diabetes is rare. Hypoglycemia in nondiabetic patients can be caused by such diverse entities as pancreatic islet cell adenoma or carcinoma, large hepatoma, large sarcoma, alcohol ingestion, use of β-adrenergic receptor blocking drugs, haloperidol therapy, hypopituitarism, adrenal insufficiency, altered physiology after gastric or gastric bypass surgery, hereditary fructose intolerance, ingestion of antidiabetic drugs, galactosemia, or autoimmune hypoglycemia. The last four entities cause postprandial reactive hypoglycemia. Because restriction of oral intake prevents severe hypoglycemia, the practice of keeping the patient NPO (nothing by mouth) and infusing small amounts of a solution containing 5% dextrose greatly lessens the possibility of perioperative postprandial reactive hypoglycemia. The other causes of hypoglycemia can cause serious problems during the perioperative period.
Symptoms of hypoglycemia fall into one of two groups: adrenergic excess (tachycardia, palpitations, tremulousness, or diaphoresis) or neuroglycopenia (headache, confusion, mental sluggishness, seizures, or coma). All these symptoms may be masked by anesthesia, so blood glucose levels should be determined frequently in at-risk patients to ensure that hypoglycemia is not present. Because manipulation of an insulinoma can result in massive insulin release, this tumor should probably be operated on only at centers equipped with a mechanical pancreas. Perioperative use of the somatostatin analogue octreotide, which suppresses insulin release from such tumors, makes the perioperative period safer in anecdotal experience.
Disorders of Nutrition, Including Obesity
Hyperlipoproteinemia, Hyperlipidemia, and Hypolipidemia
Hyperlipidemia may result from obesity, estrogen or corticoid therapy, uremia, diabetes, hypothyroidism, acromegaly, alcohol ingestion, liver disease, inborn errors of metabolism, or pregnancy. Hyperlipidemia may cause premature coronary, peripheral vascular disease, or pancreatitis.
Coronary events can be decreased by treating individuals with even normal levels of low-density lipoprotein (LDL) cholesterol with statins (3-hydroxy-3-methylglutaryl–coenzyme A [HMG-CoA] reductase inhibitors )—through an increase in high-density lipoprotein (HDL) and a decrease in LDL cholesterol levels. This approach has markedly decreased the rate of myocardial reinfarction in high-risk patients. Secondary prevention efforts were successful when these high-risk patients stopped smoking, reduced their arterial blood pressure, controlled stress, increased physical activity, and used aspirin, folate, β-blocking drugs, angiotensin inhibitors, diet, and other drugs to reduce their levels of LDL and increase their levels of HDL.
Although controlling the diet remains a major treatment modality for all types of hyperlipidemia, the drugs fenofibrate and gemfibrozil, which are used to treat hypertriglyceridemia, can cause myopathy, especially in patients with hepatic or renal disease; clofibrate is also associated with an increased incidence of gallstones. Cholestyramine binds bile acids, as well as oral anticoagulants, digitalis drugs, and thyroid hormones. Nicotinic acid causes peripheral vasodilation and should probably not be continued through the morning of the surgical procedure. Probucol (Lorelco) decreases the synthesis of apoprotein A-1; its use is associated on rare occasion with fetid perspiration or prolongation of the QT interval, or both, and sudden death in animals.
The West of Scotland Coronary Prevention Study and its congeners produced convincing evidence that drugs in the statin class prevent the morbidity and mortality related to arterial aging and vascular disease, as well as their consequences, such as coronary artery disease (CAD), stroke, and peripheral vascular insufficiency. Thus, the statins—lovastatin (Mevacor), pravastatin (Pravachol), simvastatin, fluvastatin, atorvastatin (Lipitor), and rosuvastatin (Crestor)—are mainstays of therapy, limited by patient tolerance most commonly secondary to musculoskeletal complaints. [CR]
However, the report of Downs and coworkers from the Air Force/Texas Coronary Atherosclerosis Prevention Study went further. This report showed a 37% reduction in the risk for first acute major coronary events in patients who had no risk factors and normal (average) LDL cholesterol levels. In this study lovastatin did not alter mortality rates, but that had been true for many early short-term trials with the statins. Although much of the effect of the statins has been attributed to their lipid-lowering effects, statins also influence endothelial function, inflammatory responses, plaque stability, and thrombogenicity. In 2013, the American College of Cardiology (ACC) and the American Heart Association (AHA) released a new clinical practice guideline for the treatment of blood cholesterol in people at high risk for cardiovascular diseases. They now advocate statin therapy for the following:
Patients who have cardiovascular disease (coronary syndromes, previous myocardial infarction [MI], stable or unstable angina, previous stroke or transient ischemic attack, or peripheral artery disease)
Patients with an LDL cholesterol level of 190 mg/dL or higher
Patients with diabetes (type 1 or 2) who are between 40 and 75 years old
Patients with an estimated 10-year risk of cardiovascular disease greater than 7.5% (the report provided formulas for calculating 10-year risk)
The 2014 National Lipid Association Recommendations for Patient-Centered Management of Dyslipidemia further emphasize the use of statins as a first-line therapy for dyslipidemia, but emphasize the inclusion of non-high density lipoprotein in addition to LDL as markers for risk. They further advocated for management of other atherosclerotic cardiovascular disease risk factors including high blood pressure, tobacco use, and diabetes mellitus. [CR]
Statins are drugs that block HMG-CoA reductase, the rate-limiting enzyme of cholesterol synthesis. Their use is occasionally accompanied by liver dysfunction, CNS dysfunction, and severe depression not related to the high cost of each drug and its congeners. Based on the available evidence, statin therapy should be continued in patients already taking these drugs. Other drugs that reduce LDL and increase HDL cholesterol and decrease triglycerides are docosahexaenoic acid (an ω-3 fatty acid) and niacin. Statins also provide the substantial benefit of reversing inflammation in arteries, as evidenced by their ability to decrease highly specific C-reactive protein and pull cholesterol from plaque.
Hypolipidemic conditions are rare diseases often associated with neuropathy, anemia, and renal failure. Although anesthetic experience with hypolipidemic conditions has been limited, some specific recommendations can be made: continuation of caloric intake and intravenous administration of protein hydrolysates and glucose should be continued throughout the perioperative period.
Obesity is a risk factor for perioperative morbidity. In the study of Medicare claims in which obese patients were matched to non-obese patients undergoing surgery, the obese patients displayed increased odds of wound infection, renal dysfunction, urinary tract infection, hypotension, respiratory events, 30-day readmission, and a 12% longer length of stay. Although many conditions associated with obesity (diabetes, hyperlipidemia, cholelithiasis, gastroesophageal reflux disease, cirrhosis, degenerative joint and disk disease, venous stasis with thrombotic or embolic disease, sleep disorders, and emotional and altered body image disorders) contribute to chronic morbidity in these patients, the main concerns for the anesthesiologist have been the same since the 1970s—derangements of the cardiopulmonary system.
Morbid obesity with minimal or no coexisting pulmonary conditions (e.g., no obesity-hypoventilation syndrome or chronic obstructive pulmonary disease [COPD]) is referred to here as “simple” obesity. In simple obesity, the pathophysiology of mild alterations in daytime gas exchange and pulmonary function may also result from compression and restriction of the chest wall and diaphragm by excess adipose tissue. Typically, in obese patients, the expiratory reserve volume and functional residual capacity are most affected and are reduced to 60% and 80% of normal, respectively. Care must be taken with medication choice and dosing, as simple obese patients may be more sensitive to sedative and narcotic agents leading to hypoventilation. [CR]
Other Eating Disorders: Anorexia Nervosa, Bulimia, and Starvation
Many endocrine and metabolic abnormalities occur in patients with anorexia nervosa, a condition characterized by starvation to the point of 40% loss of normal weight, hyperactivity, and a psychiatrically distorted body image. Many anorectic patients exhibit impulsive behavior, including suicide attempts, and intravenous drug use is much more common than in the general population. Acidosis, hypokalemia, hypocalcemia, hypomagnesemia, hypothermia, diabetes insipidus, and severe endocrine abnormalities mimicking panhypopituitarism may need attention before patients undergo anesthesia. Similar problems occur in bulimia (bulimorexia), a condition that may affect as many as 50% of female college students and is even unintentionally present in many older adults. As in severe protein deficiency, anorexia nervosa and bulimia may be accompanied by the following: alterations on the electrocardiogram (ECG), including a prolonged QT interval, atrioventricular (AV) block, and other arrhythmias; sensitivity to epinephrine; and cardiomyopathy. Total depletion of body potassium makes the addition of potassium to glucose solutions useful; although, fluid administration can precipitate pulmonary edema in these patients and should be monitored judiciously. Esophagitis, pancreatitis, and aspiration pneumonia are more frequent in these patients, as is delayed gastric emptying. One review reported that in patients with severe anorexia, a body mass index less than 13 kg/m 2 , marked hypoglycemia or leukocytopenia lower than 3.0 × 10 9 /L, or both, potentially fatal complications frequently occur. Intraoperatively, glucose or catecholamine administration may lead to disturbance of electrolytes or fatal arrhythmia. Intensive care and early nutritional support as soon as possible postoperatively are important to prevent surgical site infection with close monitoring for refeeding syndrome.
Hyperalimentation (Total Parenteral or Enteral Nutrition)
Hyperalimentation (i.e., total parenteral nutrition [TPN]) consists of concentrating hypertonic glucose calories in the normal daily fluid requirements. The solutions contain protein hydrolysates, soybean emulsions (i.e., Intralipid), or synthetic amino acids (or any combination of these ingredients). The major benefits of TPN or enteral nutrition have been fewer complications postoperatively and shorter hospital stays for patients scheduled to have no oral feeding for 7 days or who were malnourished preoperatively. Starker and colleagues found that the response to TPN, as monitored by serum albumin levels, predicted the postoperative outcome. The group of patients demonstrating an increase in serum albumin concentrations from TPN had diuresis, weight loss, and fewer complications (1 of 15 patients) than did the group that gained weight and had a decrease in serum albumin (8 of 16 patients had 15 complications; Fig. 32.2 ). The Veterans Administration (former name for Veterans Affairs [VA used for both]) studies also found that the serum albumin level was one of the most powerful predictors of perioperative outcome.
The major complications of TPN are infection, metabolic abnormalities, and longer duration of ICU stay. [CR] The central lines used for TPN require an absolutely aseptic technique and should not be used as an intravenous access or as a route for drug administration during anesthesia and surgery. Major metabolic complications of TPN relate to electrolyte deficiencies, and the development of hyperosmolar states. Complications of hypertonic dextrose can develop if the patient has insufficient insulin (diabetes mellitus) to metabolize the sugar or if insulin resistance occurs (e.g., because of uremia, burns, or sepsis).
A gradual decrease in the infusion rate of TPN prevents the hypoglycemia that can occur on abrupt discontinuance. Thus the infusion rate of TPN should be decreased the night before anesthesia and surgery, or should be continued throughout the operation at its current rate. The main reason for slowing or discontinuing TPN before anesthesia is to avoid intraoperative hyperosmolarity secondary to accidental rapid infusion of the solution or hypoglycemia if the infusion is discontinued because of high levels of endogenous insulin and lower levels of glucose present in the usual crystalloid solutions. Hypophosphatemia is a particularly serious complication that results from the administration of phosphate-free or phosphate-depleted solutions for hyperalimentation. The low serum phosphate level causes a shift of the oxygen dissociation curve to the left. The resulting low 2,3-diphosphoglycerate and adenosine triphosphatase levels mean that cardiac output must increase for oxygen delivery to remain the same. Hypophosphatemia of less than 1.0 mg/dL of blood may cause hemolytic anemia, cardiac failure, tachypnea, neurologic symptoms, seizures, and death. In addition, long-term TPN is associated with deficiencies in trace metals such as copper (refractory anemia), zinc (impaired wound healing), and magnesium.
Three major classes of hormones—androgens, glucocorticoids, and mineralocorticoids—are secreted by the adrenal cortex. For each class, an excess or a deficiency of hormone produces a characteristic clinical syndrome. The widespread use of steroids can also make the adrenal cortex unable to respond normally to the demands placed on it by surgical trauma and subsequent healing. The increase in computed tomography (CT) abdominal imaging procedures has meant that many adrenal masses have unfortunately been discovered incidentally. These adrenal “incidentalomas,” as they are termed because they were initially thought a nuisance discovered by body scans, have proved more serious. As many as 30% are hormonally active; in one review of 2000 such masses, 82% were not hormonally active, 5.3% proved to be cortisone-secreting adenomas, 5.1% were pheochromocytomas, 4.7% were adrenal carcinomas, 2.5% were unsuspected metastatic disease, and 1% were aldosterone-secreting adenomas. “Incidentalomas” may therefore require serious pursuit; however, well accepted and utilized guidelines are absent, but caution should be taken during anesthesia.
Controlled comparisons of the perioperative management of patients who have disorders of adrenal function are lacking, although steroids are used more and more commonly, with the results of some controlled trials available for specific uses. However, a review of the possible pathophysiologic changes in the adrenal cortex and techniques for their management should enable physicians to improve the perioperative care of patients with adrenal abnormalities.
Physiologic Properties of Adrenocortical Hormones
Androstenedione and dehydroepiandrosterone, weak androgens arising from the adrenal cortex, constitute major sources of androgen in women (and have gained prominence for their abuse among athletes). Excess secretion of androgen causes masculinization, pseudopuberty, or female pseudohermaphroditism. With some tumors, androgen is converted to an estrogenic substance, in which case feminization results. No special anesthetic evaluation is needed for such patients. Some congenital enzyme defects that cause androgen abnormalities also result in glucocorticoid and mineralocorticoid abnormalities that should be evaluated preoperatively. Most of these patients are treated with exogenous glucocorticoids and mineralocorticoids and may require supplementation of these hormones perioperatively.
The principal glucocorticoid, cortisol, is an essential regulator of carbohydrate, protein, lipid, and nucleic acid metabolism. Cortisol exerts its biologic effects through a sequence of steps initiated by the binding of hormone to stereospecific intracellular cytoplasmic receptors. This bound complex stimulates nuclear transcription of specific mRNA molecules. These molecules are then translated to give rise to proteins that mediate the ultimate effects of hormones.
Most cortisol is bound to corticosterone-binding globulin (CBG, transcortin). The relatively small amounts of unbound cortisol enter cells to induce actions or to be metabolized. Conditions that induce changes in the amount of CBG include liver disease and nephrotic syndrome, both of which result in decreased circulating levels of CBG, and estrogen administration and pregnancy, which result in increased CBG production. Total serum cortisol levels may become elevated or depressed under conditions that alter the amount of bound cortisol, and yet the unbound, active form of cortisol is present in normal amounts. The most accurate measure of cortisol activity is the level of urinary cortisol (i.e., the amount of unbound, active cortisol filtered by the kidney).
The serum half-life of cortisol is 80 to 110 minutes. However, because cortisol acts through intracellular receptors, pharmacokinetic data based on serum levels are not good indicators of cortisol activity. After a single dose of glucocorticoid, serum glucose is elevated for 12 to 24 hours; improvement in pulmonary function in patients with bronchial asthma can still be measured 24 hours after glucocorticoid administration. Treatment schedules for glucocorticoid replacement are therefore based not on the measured serum half-life but on the well-documented prolonged end-organ effect of these steroids. Hospitalized patients requiring long-term glucocorticoid replacement therapy are usually treated twice daily, with a slightly higher dose given in the morning than in the evening to simulate the normal diurnal variation in cortisol levels. For patients who require parenteral “steroid coverage” during and after surgery (see later paragraphs), administration of glucocorticoid every 6 to 12 hours is appropriate pending the type of surgery and expected stress response. [CR] The relative potencies of glucocorticoids are listed in Table 32.2 . Cortisol is inactivated primarily in the liver and is excreted as 17-hydroxycorticosteroid. Cortisol is also filtered and excreted unchanged into urine.
|Steroids||Relative Glucocorticoid Potency||Equivalent Glucocorticoid Dose (mg)|
The synthetic glucocorticoids vary in their binding specificity in a dose-related manner. When given in supraphysiologic doses (>30 mg/day), cortisol and cortisone bind to mineralocorticoid receptor sites, and cause salt and water retention and loss of potassium and hydrogen ions. When these steroids are administered in maintenance doses of 30 mg/day or less, patients require a specific mineralocorticoid for electrolyte and volume homeostasis. Many other steroids do not bind to mineralocorticoid receptors, even at high doses, and have no mineralocorticoid effect (see Table 32.2). [CR]
Secretion of glucocorticoids is regulated by pituitary adrenocorticotropic hormone (ACTH). ACTH is synthesized from a precursor molecule (pro-opiomelanocortin) that is metabolized to form an endorphin (β-lipotropin) and ACTH. Episodic secretion of ACTH has a diurnal rhythm that is normally greatest during the early morning hours in men and later in women and is regulated at least in part by light-dark cycles. Its secretion is stimulated by release of corticotropin-releasing factor (CRF) from the hypothalamus. (An abnormality in the diurnal rhythm of corticoid secretion has been implicated as a cause of so-called jet lag.) Cortisol and other glucocorticoids exert negative feedback at both the pituitary and hypothalamic levels to inhibit the secretion of ACTH and CRF. If the CRF- or ACTH-producing cells are destroyed, the adrenal gland takes more than 30 days to atrophy to the point at which short-term administration of exogenous ACTH will cause almost no adrenal responsiveness.
Aldosterone, the major mineralocorticoid secreted in humans, comes from the zona glomerulosa of the adrenal cortex and causes reabsorption of sodium and secretion of potassium and hydrogen ions, thereby contributing to electrolyte and volume homeostasis. This action is most prominent in the distal renal tubule but also occurs in the salivary and sweat glands. The major regulator of aldosterone secretion is the renin-angiotensin system. Juxtaglomerular cells in the cuff of renal arterioles are sensitive to decreased renal perfusion pressure or volume and, consequently, secrete renin. Renin transforms the precursor angiotensinogen (from the liver) into angiotensin I, which is further converted by a converting enzyme, primarily in the lung, to angiotensin II. Angiotensin II binds to specific receptors to increase mineralocorticoid secretion, which is also stimulated by an increased potassium concentration and, to a lesser degree, by ACTH.
Adrenocortical Hormone Excess
Glucocorticoid excess (Cushing syndrome) resulting from either endogenous oversecretion or long-term treatment with glucocorticoids at higher than physiologic doses produces a moon-faced plethoric individual with a centripetal distribution of fat (truncal obesity and skinny extremities), thin skin, easy bruising, and striae. Skeletal muscle wasting is common, but the heart and diaphragm are usually spared. A test for this syndrome is to ask the patient to get up from a chair without using the hands with the inability to do so indicating proximal muscle weakness consistent with Cushing syndrome. These patients often have osteopenia as a result of decreased formation of bone matrix and impaired absorption of calcium. Fluid retention and hypertension (because of increases in renin substrate and vascular reactivity caused by glucocorticoid activity) are common. Such patients may also have hyperglycemia and even diabetes mellitus from inhibition of peripheral use of glucose, as well as anti-insulin action and concomitant stimulation of gluconeogenesis ( Table 32.3 ).
|Central obesity||Weight loss|
|Proximal muscle weakness||Weakness, fatigue, lethargy|
|Osteopenia at a young age||Muscle, joint, and back pain|
|Hypertension||Postural hypotension and dizziness|
|Psychiatric disorders||Anorexia, nausea, abdominal pain, constipation, diarrhea|
|Hypokalemic alkalosis||Hyperkalemia, hyponatremia|
|Glucose intolerance||Occasional hypoglycemia|
|Increased leukocyte count|
The most common cause of Cushing syndrome is the administration of glucocorticoids for such conditions as arthritis, asthma, and allergies. In these conditions, the adrenal glands atrophy and cannot respond to stressful situations (e.g., the perioperative period) by secreting more steroid; therefore, additional glucocorticoids may be required perioperatively (see the later section “Patients Taking Steroids for Other Reasons”). Spontaneous Cushing syndrome may be caused by pituitary production of ACTH (65% to 75% of all spontaneous cases), which is usually associated with pituitary microadenoma, or by nonendocrine ectopic ACTH production (principally by tumors of the lung, pancreas, or thymus). Ten percent to 20% of cases of spontaneous Cushing syndrome are caused by an ACTH-independent process, either an adrenal adenoma or carcinoma.
Special preoperative and preprocedure considerations for patients with Cushing syndrome include regulating blood glucose control, managing hypertension, and ensuring intravascular volume and electrolyte concentrations are normal. Ectopic ACTH production may cause marked hypokalemic alkalosis. Treatment with the aldosterone antagonist spironolactone stops the potassium loss and helps mobilize excess fluid. Because of the incidence of severe osteopenia and the risk of fractures, meticulous attention must be paid to positioning of the patient. In addition, glucocorticoids are lympholytic and immunosuppressive, thus increasing the patient’s susceptibility to infection. The tensile strength of healing wounds decreases in the presence of glucocorticoids, an effect that is at least partially reversed by the topical administration of vitamin A.
Ten percent to 15% of patients with Cushing syndrome exhibit adrenal overproduction of glucocorticoids from an adrenal adenoma or carcinoma. If either unilateral or bilateral adrenal resection is planned, the physician should begin administering glucocorticoids at the start of resection of the tumor. Despite the absence of definitive studies, 100 mg of hydrocortisone every 24 hours intravenously is reasonable. This amount can be reduced over a period of 3 to 6 days until a maintenance dose is reached. Beginning on day 3, the surgeons may also give a mineralocorticoid, 9α-fluorocortisol (0.05-0.1 mg/day). In certain patients, both steroids may require several adjustments. This therapy continues if the patient has undergone bilateral resection. For a patient who has undergone unilateral adrenal resection, therapy is individualized according to the status of the remaining adrenal gland. The incidence of pneumothorax in an open adrenal resection approach can be as high as 20%; the diagnosis of pneumothorax is sought and treatment is initiated before the wound is closed. The use of the laparoscopic technique has markedly decreased this complication.
Bilateral adrenalectomy (now performed laparoscopically) in patients with Cushing syndrome is associated with a perioperative morbidity rate up to 20% and a perioperative mortality rate up to 3%. This procedure often results in permanent mineralocorticoid and glucocorticoid deficiency. [CR] Ten percent of patients with Cushing syndrome who undergo adrenalectomy have an undiagnosed pituitary tumor. After cortisol concentrations are decreased by adrenalectomy, the pituitary tumor will likely enlarge. These pituitary tumors are potentially invasive and may produce large amounts of ACTH and melanocyte-stimulating hormone, thereby increasing pigmentation.
Approximately 85% of adrenal tumors are discovered incidentally during screening CT scans. Nonfunctioning adrenal adenomas are found in patients on autopsy, ranging from 1% to 32% in different series. Functioning adenomas are generally treated surgically; often, the contralateral gland resumes functioning after several months. Frequently, however, the effects of carcinomas are not cured surgically. In such cases, administration of inhibitors of steroid synthesis, such as metyrapone or mitotane ( o , p ′-DDD[2,2-bis(2-chlorophenyl4-chlorophenyl)-1,1-dichloroethane]), may ameliorate some symptoms, as these drugs and specific aldosterone antagonists may aid in reducing symptoms of ectopic ACTH secretion if the primary tumor is unresectable. Patients given these adrenal suppressants are also prescribed long-term glucocorticoid replacement therapy with the goal of therapy being complete adrenal suppression. Therefore, these patients should be considered to have suppressed adrenal function, and glucocorticoid replacement should be increased perioperatively.
Excess mineralocorticoid activity (common with glucocorticoid excess because most glucocorticoids have some mineralocorticoid properties) leads to potassium depletion, sodium retention, muscle weakness, hypertension, tetany, polyuria, inability to concentrate urine, and hypokalemic alkalosis. These symptoms constitute primary hyperaldosteronism, or Conn syndrome (a cause of low-renin hypertension because renin secretion is inhibited by the effects of the high levels of aldosterone).
Primary hyperaldosteronism is present in 0.5% to 1% of hypertensive patients who have no other known cause of hypertension. Primary hyperaldosteronism most often results from unilateral adenoma, although 25% to 40% of patients have been found to have bilateral adrenal hyperplasia. Intravascular fluid volume, electrolyte concentrations, and renal function should be restored to within normal limits preoperatively by administering the aldosterone antagonist spironolactone. The effects of spironolactone are slow in onset and increase for 1 to 2 weeks. Frequently, a period of at least 24 hours is required to restore potassium equilibrium as the deficit can be up to 400 mEq; however, normal serum potassium level does not necessarily imply correction of a total-body deficit of potassium. In addition, patients with Conn syndrome have a high incidence of hypertension and ischemic heart disease; hemodynamic monitoring should be tailored to the individual patient.
A retrospective anecdotal study indicated that intraoperative hemodynamic status was more stable when arterial blood pressure and electrolytes were controlled preoperatively with spironolactone than when other antihypertensive agents were used. However, the efficacy of optimizing the perioperative status of patients who have disorders of glucocorticoid or mineralocorticoid secretion has not been clearly defined. Therefore, we have assumed that gradual restoration of physiologic norms is good medicine and expect that it would decrease perioperative morbidity and mortality.
Adrenocortical Hormone Deficiency
Withdrawal of steroids or suppression of synthesis by steroid therapy is the leading cause of underproduction of corticosteroids (its management is discussed in the later section “Patients Taking Steroids for Other Reasons”). Other causes of adrenocortical insufficiency include the following: defects in ACTH secretion and destruction of the adrenal gland by autoimmune disease, tuberculosis, hemorrhage (e.g., Sheehan syndrome), or cancer; some forms of congenital adrenal hyperplasia (see previous discussion); and administration of cytotoxic drugs.
Primary adrenal insufficiency (Addison disease) is associated with local destruction of all zones of the adrenal cortex and results in both glucocorticoid and mineralocorticoid deficiency if the insufficiency is bilateral; common symptoms and signs are listed in Table 32.3 . Autoimmune disease is the most common cause of primary (nonexogenous) bilateral ACTH deficiency in the United States, whereas tuberculosis is the most common cause worldwide. Tuberculosis is associated not only with decreased adrenal function, but also large adrenal glands, which are a common finding in sarcoidosis, histoplasmosis, amyloidosis, metastatic malignant disease, heparin-induced thrombocytopenia, and adrenal hemorrhage. Further, destruction or injury by trauma, human immunodeficiency virus (HIV), and other infections (e.g., cytomegalovirus, mycobacteria, and fungi) is being recognized more frequently.
Autoimmune destruction of the adrenal glands may be associated with other autoimmune disorders, such as some forms of type 1 diabetes and Hashimoto thyroiditis. Enzymatic defects in cortisol synthesis cause glucocorticoid insufficiency, compensatory elevations in ACTH, and congenital adrenal hyperplasia. Because adrenal insufficiency usually develops slowly, such patients are subject to marked pigmentation (from excess ACTH trying to stimulate an unproductive adrenal gland) and cardiopenia (secondary to chronic hypotension).
Secondary adrenal insufficiency occurs when ACTH secretion is deficient, often because of a pituitary or hypothalamic tumor. Treatment of pituitary tumors by surgery or radiation therapy may result in hypopituitarism and subsequent adrenal failure.
If unstressed, glucocorticoid-deficient patients usually have no perioperative problems. However, acute adrenal crisis (addisonian crisis) can occur when even a minor stress is present (e.g., upper respiratory infection). Preparation of such a patient for anesthesia and surgery should include treatment of hypovolemia, hyperkalemia, and hyponatremia. Because these patients cannot respond to stressful situations, it was traditionally recommended that they be given a stress dose of glucocorticoids (≈200 mg hydrocortisone/day) perioperatively. However, Symreng and colleagues gave 25 mg of hydrocortisone phosphate intravenously to adults at the start of the operative procedure, followed by 100 mg intravenously over the next 24 hours. Because using the minimum drug dose that would produce an appropriate effect is desirable, this latter regimen seems attractive. Such a regimen has proved to be as successful as a regimen using maximum doses (≈300 mg hydrocortisone/day). Thus we now recommend giving the patient’s usual daily dose plus 50 to 100 mg of hydrocortisone before surgical incision and 25 to 50 mg of hydrocortisone every 8 hours for 24 to 48 hours, depending on the type and duration of surgery. [CR]
Hypoaldosteronism, a less common condition, can be congenital, can occur after unilateral adrenalectomy, or be a consequence of prolonged heparin administration, long-standing diabetes, or renal failure. Nonsteroidal inhibitors of prostaglandin synthesis may also inhibit renin release and exacerbate this condition in patients with renal insufficiency. Plasma renin activity is lower than normal and fails to increase appropriately in response to sodium restriction or diuretic drugs. Most symptoms are caused by hyperkalemic acidosis rather than hypovolemia; in fact, some patients are hypertensive. These patients can have severe hyperkalemia, hyponatremia, and myocardial conduction defects. These defects can be treated successfully by administering mineralocorticoids (9α-fluorocortisol, 0.05-0.1 mg/day) preoperatively. Doses must be carefully titrated and monitored to avoid an increase in hypertension.
Patients Taking Steroids for Other Reasons
Perioperative Stress and the Need for Corticoid Supplementation
The adrenal responses of normal patients to the perioperative period, as well as the responses of patients taking steroids for other diseases, indicate the following:
Perioperative stress is related to the degree of trauma and the depth of anesthesia. Deep general or regional anesthesia delays the usual intraoperative glucocorticoid surge to the postoperative period.
A few patients with suppressed adrenal function will have perioperative cardiovascular problems if they do not receive supplemental steroids perioperatively.
Although a patient who takes steroids on a long-term basis may become hypotensive perioperatively; glucocorticoid or mineralocorticoid deficiency is seldom the cause. Longer duration and higher home steroid dose increase the likelihood of deficiency. [CR]
Acute adrenal insufficiency rarely occurs, but can be life-threatening.
Giving these patients steroid coverage equivalent to 100 mg of hydrocortisone perioperatively has little risk. [CR]
In a well-controlled study of glucocorticoid replacement in nonhuman primates, the investigators clearly defined the life-threatening events that can be associated with inadequate perioperative corticosteroid replacement. In this study, adrenalectomized primates and sham-operated controls were given physiologic doses of steroids for 4 months. The animals were then randomly allocated to groups that received subphysiologic (one-tenth of the normal cortisol production), physiologic, or supraphysiologic (10 times the normal cortisol production) doses of cortisol for 4 days preceding abdominal surgery (cholecystectomy). The group given subphysiologic doses of steroid perioperatively had a significant increase in postoperative mortality. Death rates for the physiologic and supraphysiologic replacement groups were the same and did not differ from the rate for sham-operated controls. Death in the subphysiologic replacement group was related to severe hypotension associated with a significant decrease in systemic vascular resistance and a reduced left ventricular stroke work index. Filling pressures of the heart were unchanged when compared with those in control animals. No evidence of hypovolemia or severe congestive heart failure (CHF) was observed. Despite the low systemic vascular resistance, the animals did not become tachycardic. All these responses are compatible with the previously documented interaction of glucocorticoids and catecholamines, and thus suggest that glucocorticoids mediate catecholamine-induced increases in cardiac contractility and maintenance of vascular tone.
The investigators used a sensitive measure of wound healing involving accumulation of hydroxyproline. All treatment groups, including the group given supraphysiologic doses of glucocorticoids, had the same capacity for wound healing. Furthermore, perioperative administration of supraphysiologic doses of corticosteroids produced no adverse metabolic consequences.
This study confirmed long-standing intuitive impressions concerning patients who had inadequate adrenal function as a result of either underlying disease or administration of exogenous steroids—inadequate replacement of corticosteroids can lead to addisonian crisis and increased mortality, whereas the administration of supraphysiologic doses of steroids for a short time perioperatively can be safe. It is clear that inadequate corticosteroid coverage can cause death, but what is not so clear is what dose of steroid should be recommended for replacement therapy. Yong and colleagues reviewed the randomized controlled trials for a Cochrane Systemic Review and reported only two trials involving 37 patients that met the inclusion criteria. These studies reported that supplemental perioperative steroids were not required during surgery for patients with adrenal insufficiency, but neither study reported any adverse effects or complications in the intervention or control groups. The authors concluded that they were unable to support or refute the use of supplemental perioperative steroids for patients with adrenal insufficiency during surgery. Because the risk is low and the benefit is high, physicians should consider providing supplementation for any patient who has received steroids within a year.
How can one determine when adrenal responsiveness has returned to normal? The morning plasma cortisol level does not reveal whether the adrenal cortex has recovered sufficiently to ensure that cortisol secretion will increase adequately to meet the demands of stress. Inducing hypoglycemia with insulin has been advocated as a sensitive test of pituitary-adrenal competence, but it is impractical and is probably a more dangerous practice than simply administering glucocorticoids. If the plasma cortisol concentration is measured during acute stress, a value of greater than 25 μg/dL assuredly (and a value >15 μg/dL probably) indicates normal pituitary-adrenal responsiveness. In another test of pituitary-adrenal sufficiency, the baseline plasma cortisol level is determined. Then, 250 μg of synthetic ACTH (cosyntropin) is given, and plasma cortisol is measured 30 to 60 minutes later. An increase in plasma cortisol of 6 to 20 μg/dL or more is normal. A normal response indicates recovery of pituitary-adrenal axis function. A lesser response usually indicates pituitary-adrenal insufficiency, possibly requiring perioperative supplementation with steroids. [CR]
Under perioperative conditions, the adrenal glands secrete 116 to 185 mg of cortisol daily. Under maximum stress, they may secrete 200 to 500 mg/day. Good correlation exists between the severity and duration of the operation and the response of the adrenal gland. “Major surgery” would be represented by procedures such as laparoscopic colectomy and “minor surgery” by procedures such as herniorrhaphy. In a study of 20 patients during major surgery, the mean maximal concentration of cortisol in plasma was 47 μg/dL (range, 22-75 μg/dL). Values remained higher than 26 μg/dL for a maximum of 72 hours postoperatively. During minor surgery, the mean maximal concentration of cortisol in plasma was 28 μg/dL (range, 10-44 μg/dL).
Although the precise amount required has not been established, we usually intravenously administer the maximum amount of glucocorticoid that the body manufactures in response to maximal stress (i.e., approximately 200 mg/day of hydrocortisone). [CR] For minor surgical procedures, we usually give hydrocortisone intravenously, 50 to 100 mg/day. Unless infection or some other perioperative complication develops, we decrease this dose by approximately 50%/day until the standard home dose is resumed. For major surgical procedures, we usually give 50 mg every 6 hours to 100 mg every 8 hours. Again unless a complication develops, this is decreased 50%/day until the standard home dose is resumed. [CR]
Risks of Supplementation
Rare complications of perioperative steroid supplementation include aggravation of hypertension, fluid retention, inducement of stress ulcers, and psychiatric disturbances. Two possible complications of short-term perioperative supplementation with glucocorticoids are abnormal wound healing and an increased rate of infections. This evidence is inconclusive, however, because it relates to short-term glucocorticoid administration and not to long-term administration of glucocorticoids with increased doses at times of stress. In contrast to a deleterious effect of perioperative glucocorticoid administration on wound healing in rats, a study involving primates suggested that large doses of glucocorticoids, administered perioperatively, do not impair sensitive measures of wound healing. An overall assessment of these results suggests that short-term perioperative supplementation with steroids has a small but definite deleterious effect on wound healing that is perhaps partially reversed by topical administration of vitamin A.
Information on the risk of infection from perioperative glucocorticoid supplementation is also unclear as there are no controlled trials addressing these effects. In many studies of long-term use by patients and supplementation, no increased risk of serious infections was reported with long-term use of steroids alone. Data indicate that the risk of infection in a patient taking steroids on a long-term basis is real, but whether perioperative supplementation with steroids increases that risk is not clear.
Adrenal Cortex Function in Older Adults
Production of androgens by the adrenal gland progressively decreases with age; this change has no known implications for anesthesia. Plasma levels of cortisol are unaffected by increasing age. Levels of CBG are also unaffected by age, a finding suggesting that a normal fraction of free cortisol (1%-5%) is present in older patients. Older patients have a progressively impaired ability to metabolize and excrete glucocorticoids. In normal individuals, the quantity of 17-hydroxycorticosteroids excreted is reduced by half by the seventh decade. This decreased excretion undoubtedly reflects the reduced renal function that occurs with aging. When excretion of cortisol metabolites is expressed as a function of creatinine clearance, the age difference disappears. Further reductions in cortisol clearance may reflect impaired hepatic metabolism of circulating cortisol.
The rate of secretion of cortisol is 30% slower in older adults. This reduced secretion may be an appropriate compensatory mechanism for maintaining a normal cortisol level in the presence of decreased hepatic and renal clearance of cortisol. The reduced cortisol production can be overcome during periods of stress, and even extremely old patients (>100 years old) display an entirely normal adrenal response to the administration of ACTH and to stresses such as hypoglycemia.
Both underproduction and overproduction of glucocorticoids are generally considered diseases of younger individuals. The highest incidence of Cushing disease of either pituitary or adrenal origin occurs during the third decade of life. The most common cause of spontaneous Cushing disease is benign pituitary adenoma. However, in patients older than 60 years in whom Cushing disease develops, the most likely cause is adrenal carcinoma or ectopic ACTH production from tumors usually located in the lung, pancreas, or thymus.
Adrenal Medullary Sympathetic Hormone Excess: Pheochromocytoma
Less than 0.1% of all cases of hypertension are caused by pheochromocytomas, or catecholamine-producing tumors derived from chromaffin tissue. Nevertheless, these tumors are clearly important to the anesthesiologist as previously 25% to 50% of hospital deaths in patients with pheochromocytoma occurred during induction of anesthesia or during operative procedures for other causes. This high mortality has been reduced with the improvements in anesthesia management during our current era. [CR] Although usually found in the adrenal medulla, these vascular tumors can occur anywhere (referred to as paragangliomas), with a proportion of up to 20%. [CR] Malignant spread, which occurs in less than 15% of pheochromocytomas, usually proceeds to venous and lymphatic channels with a predisposition for the liver. This tumor is occasionally familial or part of the multiglandular-neoplastic syndrome known as multiple endocrine adenoma type IIa or type IIb, and is manifested as an autosomal dominant trait. Type IIa consists of medullary carcinoma of the thyroid, parathyroid adenoma or hyperplasia, and pheochromocytoma. What used to be called type IIb is now often called pheochromocytoma in association with phakomatoses such as von Recklinghausen neurofibromatosis and von Hippel–Lindau disease with cerebellar hemangioblastoma. Frequently, bilateral tumors are found in the familial form. Localization of tumors can be achieved by MRI or CT, metaiodobenzylguanidine nuclear scanning, ultrasonography, or intravenous pyelography (in decreasing order of combined sensitivity and specificity).
Symptoms and signs that may be solicited before surgery or procedures and are suggestive of pheochromocytoma are as follows: excessive sweating; headache; hypertension; orthostatic hypotension; previous hypertensive or arrhythmic response to induction of anesthesia or to abdominal examination; paroxysmal attacks of sweating, headache, tachycardia, and hypertension; glucose intolerance; polycythemia; weight loss; and psychological abnormalities. In fact, the occurrence of combined symptoms of paroxysmal headache, sweating, and hypertension is probably a more sensitive and specific indicator than any one biochemical test for pheochromocytoma ( Table 32.4 ).
|Test/Symptoms||Sensitivity (%)||Specificity (%)||Likelihood Ratio|
|Positive Result ∗||Negative Result †|
|Vanillylmandelic acid excretion||81||97||27.0||0.20|
|Abdominal computed tomography||92||80||4.6||0.10|
|Concurrent paroxysmal hypertension, headache, sweating, and tachycardia ‡||90||95||18.0||0.10|
The value of preoperative and preprocedure adrenergic receptor blocking drugs probably justifies their use as these drugs may reduce the perioperative complications of hypertensive crisis, the wide arterial blood pressure fluctuations during tumor manipulation (especially until venous drainage is obliterated), and the myocardial dysfunction. Mortality is decreased with resection of pheochromocytoma (from 40% to 60% to the current 0% to 6%) when adrenergic receptor blockade is introduced as preoperative and preprocedure preparatory therapy for such patients.
α-Adrenergic receptor blockade with prazosin or phenoxybenzamine restores intravascular plasma volume by counteracting the vasoconstrictive effects of high levels of catecholamines. This reexpansion of intravascular fluid volume is often followed by a decrease in hematocrit. Because some patients may be very sensitive to the effects of phenoxybenzamine, this drug should initially be given in doses of 20 to 30 mg/70 kg orally once or twice a day. Most patients usually require 60 to 250 mg/day. The Endocrine Society Task Force guidelines from 2014 recommend α-adrenergic receptor blockade for all patients with active tumors. [CR] The efficacy of therapy should be judged by the reduction in symptoms and stabilization of arterial blood pressure. For patients who have carbohydrate intolerance because of inhibition of insulin release mediated by α-adrenergic receptor stimulation, α-adrenergic receptor blockade may reduce fasting blood glucose levels. For patients who exhibit ST-T changes on the ECG, long-term preoperative and preprocedure α-adrenergic receptor blockade (1-6 months) has produced ECG and clinical resolution of catecholamine-induced myocarditis.
β-Adrenergic receptor blockade with propranolol is suggested for patients who have persistent arrhythmias or tachycardia, the reason being that these conditions can be precipitated or aggravated by α-adrenergic receptor blockade. It is important to remember that β-adrenergic receptor blockade should not be used without concurrent α-adrenergic receptor blockade lest the vasoconstrictive effects of the latter go unopposed and thereby increasing the risk of malignant hypertension.
The optimal duration of preoperative therapy with α-adrenergic receptor blockade has not been well studied. The Endocrine Society Task Force guidelines from 2014 recommend α-adrenergic receptor blockade at least 7 to 14 days prior to surgery; however, most centers report a preoperative treatment duration of 2 to 6 weeks. Most patients will require 10 to 14 days, as judged by the time needed to stabilize arterial blood pressure and ameliorate symptoms. The Endocrine Society Task Force guidelines further recommended a high sodium diet and fluid intake to reverse the catecholamine-induced volume contraction. [CR] Because the tumor spreads slowly, little is lost by waiting until medical therapy has optimized the patient’s preoperative condition. The following criteria are reasonable for assessing the adequacy of treatment:
No in-hospital arterial blood pressure reading higher than 165/90 mm Hg should be evident for 48 hours preoperatively.
Orthostatic hypotension is acceptable as long as arterial blood pressure when the patient is standing is not less than 80/45 mm Hg.
The ECG should be free of ST-T changes that are not permanent.
No more than one premature ventricular contraction (PVC) should occur every 5 minutes.
Other drugs, including prazosin, calcium channel blocking drugs, clonidine, dexmedetomidine, and magnesium, have also been used to achieve suitable degrees of α-adrenergic blockade preoperatively. Multiple case series have confirmed the clinical utility of this approach in adults before tumor excision, including in a hemodynamic catecholamine crisis. Magnesium therapy has shown efficacy for the resection of pheochromocytoma or paraganglioma during pregnancy. The dosing of magnesium for the management of pheochromocytoma has been reviewed elsewhere.
The key clinical components of ideal patient care include optimal preoperative preparation, slow and controlled induction of anesthesia, and good communication among members of the perioperative team. Virtually all anesthetic drugs and techniques (including isoflurane, sevoflurane, sufentanil, remifentanil, fentanyl, and regional anesthesia) have been used with success, although all drugs studied were associated with a high rate of transient intraoperative arrhythmias.
Because of ease of use, the preference is to give phenylephrine for hypotension and nitroprusside or nicardipine for hypertension. Phentolamine has too long an onset and duration of action. Painful or stressful events such as intubation often cause an exaggerated stress response in less than perfectly anesthetized patients who have pheochromocytoma. This response is caused by release of catecholamines from nerve endings that are “loaded” by the reuptake process. Such stresses may result in catecholamine levels of 200 to 2000 picograms (pg)/mL in normal patients. For a patient with pheochromocytoma, even simple stress can lead to blood catecholamine levels of ten times normal. However, infarction of a tumor, with release of products onto peritoneal surfaces, or surgical pressure causing release of products, can result in blood levels of 200,000 to 1,000,000 pg/mL—a situation that should be anticipated and avoided (if possible ask for a stay of surgery to increase vasodilator infusion). Once the venous supply is secured and if intravascular volume is normal, normal arterial blood pressure usually results. However, some patients may become hypotensive and occasionally require catecholamine infusions. Vasopressin has also been used for hemodynamic rescue in catecholamine-resistant vasoplegic shock after resection of a massive pheochromocytoma. On rare occasion, patients remain hypertensive intraoperatively. Postoperatively, approximately 50% of patients remain hypertensive for 1 to 3 days and initially have markedly increased but declining plasma catecholamine levels—at which time all but 25% will become normotensive. Other family members should be advised to inform their future anesthesiologist about the potential for such familial disease.
Hypofunction or Aberration in Function of the Sympathetic Nervous System (Dysautonomia)
Disorders of the sympathetic nervous system include Shy-Drager syndrome, Riley-Day syndrome, Lesch-Nyhan syndrome, Gill familial dysautonomia, diabetic dysautonomia, and the dysautonomia of spinal cord transection.
Although individuals can function well without an adrenal medulla, a deficient peripheral sympathetic nervous system occurring late in life poses major problems; nevertheless, perioperative sympathectomy or its equivalent is often recommended. A primary function of the sympathetic nervous system appears to be regulation of arterial blood pressure and intravascular fluid volume during changing of body position. Common features of all the syndromes with hypofunction of the sympathetic nervous system are orthostatic hypotension and decreased beat-to-beat variability in heart rate. These conditions can be caused by deficient intravascular volume, deficient baroreceptor function (as also occurs in carotid artery disease ), abnormalities in CNS function (as in Wernicke or Shy-Drager syndrome), deficient neuronal stores of norepinephrine (as in idiopathic orthostatic hypotension and diabetes), or deficient release of norepinephrine (as in traumatic spinal cord injury ). These patients may have a compensatory upregulation of available adrenergic receptors causing an exaggerated response to sympathomimetic drugs. In addition to other abnormalities, such as retention of urine or feces and deficient heat exchange, hypofunction of the sympathetic nervous system is often accompanied by renal amyloidosis. Thus electrolyte and intravascular fluid volume status should be assessed preoperatively. Because many of these patients have cardiac abnormalities, cardiac function and intravascular volume status may require invasive assessment with echocardiography, central venous catheter, or a pulmonary artery catheter per the treating physician’s discretion.
Because the functioning of the sympathetic nervous system is not predictable in these patients, slow and controlled induction of anesthesia and treatment of sympathetic excess or deficiency should be initiated through titratable direct-acting vasodilators (nicardipine/nitroprusside), vasoconstrictors (phenylephrine/norepinephrine), chronotropes (isoproterenol), or negative chronotropes (esmolol). A 20% perioperative mortality rate for 2600 patients after spinal cord transection has been reported, thus indicating that such patients are difficult to manage and deserve particularly close attention.
After reviewing 300 patients with spinal cord injuries, Kendrick and coworkers concluded that autonomic hyperreflexia syndrome does not develop if the lesion is below spinal dermatome T7. If the lesion is above that level (splanchnic outflow), 60% to 70% of patients experience extreme vascular instability. The trigger to this instability, a mass reflex involving noradrenergic release and motor hypertonus, can be a cutaneous, proprioceptive, or visceral stimulus (a full bladder is a common initiator). The sensation enters the spinal cord and causes a spinal reflex, which in normal persons is inhibited from above. Sudden increases in arterial blood pressure are sensed in the pressure receptors of the aorta and carotid sinus. The resulting vagal hyperactivity produces bradycardia, ventricular ectopia, or various degrees of heart block. Reflex vasodilation may occur above the level of the lesion and result in flushing of the head and neck. In the acute injury period, modest therapeutic hypothermia may provide benefit but many note that further large randomized trials are needed; the anesthesiologist must be vigilant to avoid hyperthermia and maintain normothermia—hypothermia during procedures. [CR]
Depending on the length of time since spinal cord transection, other abnormalities may occur. In the short term (i.e., <3 weeks from the time of spinal injury), retention of urine and feces is common and, through elevation of the diaphragm, may affect respiration. Hyperesthesia is present above the lesion; reflexes and flaccid paralysis are present below the lesion. The intermediate period (3 days to 6 months) is marked by a hyperkalemic response to depolarizing drugs. The chronic phase is characterized by return of muscle tone, Babinski sign, and, frequently, the occurrence of hyperreflexia syndromes (e.g., mass reflex [see earlier]).
Thus in addition to meticulous attention to perioperative intravascular volume and electrolyte status, the anesthesiologist should know—by history taking, physical examination, and laboratory data—the status of the patient’s myocardial conduction (as revealed by the ECG), the status of renal functioning (by noting the ratio of creatinine to blood urea nitrogen [BUN]), and the condition of the respiratory muscles (by determining the ratio of forced expiratory volume in 1 second to forced vital capacity). The anesthesiologist may also obtain a chest radiograph if atelectasis or pneumonia is suspected on the basis of history taking or the physical examination. Temperature control, the presence of bone fractures or decubitus ulcers, and normal functioning of the urination and defecation systems must be assessed.
The major thyroid hormones are thyroxine (T 4 ), a prohormone product of the thyroid gland, and the more potent 3,5,3-triiodothyronine (T 3 ), a product of both the thyroid and extrathyroidal enzymatic deiodination of T 4 . Under normal circumstances, approximately 85% of T 3 is produced outside the thyroid gland. Production of thyroid secretions is maintained by secretion of thyroid-stimulating hormone (TSH) in the pituitary, which in turn is regulated by secretion of thyrotropin-releasing hormone (TRH) in the hypothalamus. Secretion of TSH and TRH appears to be negatively regulated by T 4 and T 3 . Many investigators believe that all effects of thyroid hormones are mediated by T 3 and that T 4 functions only as a prohormone.
Because T 3 has greater biologic effect than does T 4 , one would expect the diagnosis of thyroid disorders to be based on levels of T 3 . However, this is not usually the case. The diagnosis of thyroid disease is confirmed by one of several biochemical measurements: levels of free T 4 or total serum concentrations of T 4 and the “free T 4 estimate.” This estimate is obtained by multiplying total T 4 (free and bound) by the thyroid-binding ratio (formerly called resin T 3 uptake) ( Table 32.5 ). Free T 4 can be accurately measured by many laboratories, this direct measurement of free T 4 obviates the need to account for changes in binding protein synthesis and affinity caused by other conditions. The T 3 -binding ratio measures the extra quantity of serum protein-binding sites. This measurement is necessary because thyroxine-binding globulin (TBG) levels are abnormally high during pregnancy, hepatic disease, and estrogen therapy (all of which would elevate the total T 4 level; Box 32.2 ). Reliable interpretation of measurements of the total hormone concentration in serum necessitates data on the percentage of bound hormone. The thyroid hormone–binding ratio test provides this information. In this test, iodine-labeled T 3 is added to a patient’s serum and is allowed to reach an equilibrium binding state. A resin is then added that binds the remaining radioactive T 3 . Resin uptake is greater if the patient has fewer TBG-binding sites. In normal patients, resin T 3 uptake (the thyroid hormone–binding ratio) is 25% to 35%. When serum TBG is elevated, the thyroid hormone–binding ratio is diminished (see Table 32.5 ). When serum TBG is diminished, as in nephrotic syndrome, in conditions in which glucocorticoids are increased, or in chronic liver disease, the thyroid hormone–binding ratio is increased.
|Examples of Normal Thyroid Status [CR]|
|FT 4 E||=||T 4||×||THBR||TSH|
|Normal||0.19 (0.12-0.25)||=||0.6 (0.4-0.9)||×||31% (25%-35%)||0.2 (0.2-0.8)|
|During use of oral contraceptives||0.19||=||1.3||×||15%||0.3|
|During use of corticosteroids||0.18||=||0.3||×||60%||0.3|
∗ FT 4 E is the free T 4 (thyroxine) estimate. It is usually obtained by multiplying the total T 4 concentration (the free amount and the amount bound to protein) by the thyroid hormone–binding ratio (THBR, formerly called the resin T 3 uptake). THBR is a measure of the bound thyroid hormone–binding protein. TSH is the thyroid-stimulating hormone secreted by the pituitary in the negative feedback loop. (TSH increases when FT 4 E is low in hypothyroidism.)
Conditions Increasing Serum Levels
Use of oral contraceptives
Use of estrogen
Chronic active hepatitis
Acute intermittent porphyria
Conditions Decreasing Serum Levels
Use of corticosteroids
The free T 4 estimate and the free T 3 estimate are frequently used as measures of a patient’s serum T 4 and T 3 hormone concentrations, respectively. To obtain these estimates, the concentration of total serum T 4 or total serum T 3 is multiplied by the measured thyroid hormone–binding ratio. Values of these two indices are normal in the event of a primary alteration in binding but not with an alteration in secretion of thyroid hormone.
Hyperthyroidism can be diagnosed by measuring levels of TSH after the administration of TRH. Although administering TRH normally increases TSH levels in blood, even a small increase in the T 4 or T 3 level in blood abolishes this response. Thus a subnormal or absent serum TSH response to TRH is a very sensitive indicator of hyperthyroidism. In one group of disorders involving hyperthyroidism, serum TSH levels are elevated in the presence of elevated levels of free thyroid hormone.
Measurement of the α-subunit of TSH has been helpful in identifying the rare patients who have a pituitary neoplasm and who usually have increased α-subunit concentrations. Some patients are clinically euthyroid in the presence of elevated levels of total T 4 in serum. Certain drugs, notably propranolol, glucocorticoids, and amiodarone, block the conversion of T 4 to T 3 and thereby elevate T 4 levels. Severe illness also slows this conversion, termed “sick thyroid” in a critical-illness setting. Levels of TSH are often high in situations in which the rate of conversion is decreased. In hyperthyroidism, cardiac function and responses to stress are abnormal; return of normal cardiac function parallels the return of TSH levels to normal.
Although hyperthyroidism is usually caused by the multinodular diffuse enlargement in Graves disease (also associated with disorders of the skin or eyes, or both), it can also occur with pregnancy, thyroiditis, thyroid adenoma, choriocarcinoma, or TSH-secreting pituitary adenoma. Five percent of women have thyrotoxic effects 3 to 6 months postpartum and tend to have recurrences with subsequent pregnancies. Major manifestations of hyperthyroidism are weight loss, diarrhea, warm and moist skin, weakness of large muscle groups, menstrual abnormalities, osteopenia, nervousness, jitteriness, intolerance to heat, tachycardia, cardiac arrhythmias, mitral valve prolapse, and heart failure. When the thyroid is functioning abnormally, the cardiovascular system is most at risk. When diarrhea is severe, the associated dehydration and electrolyte abnormalities should be corrected preoperatively. Mild anemia, thrombocytopenia, increased serum alkaline phosphatase, hypercalcemia, muscle wasting, and bone loss frequently occur in hyperthyroidism. Muscle disease usually involves the proximal large muscle groups; it has not been reported to cause respiratory muscle paralysis. In the apathetic form of hyperthyroidism (seen most commonly in persons >60 years old), cardiac effects predominate and include tachycardia, irregular heart rhythm, atrial fibrillation (in 10%), heart failure, and occasionally, papillary muscle dysfunction.
Although β-adrenergic receptor blockade can control the heart rate, its use is challenging in the setting of heart failure. However, a decreasing heart rate may improve heart-pumping function. Thus hyperthyroid patients who have fast ventricular rates and in heart failure, requiring emergency surgery, can be safely given short-acting β-blockers guided by clinical response. If slowing the heart rate with a small dose of esmolol (50 μg/kg) does not aggravate the heart failure, the physician should administer more esmolol, and titrate to effect. Antithyroid medications include propylthiouracil and methimazole, both of which decrease the synthesis of T 4 and may enhance remission by reducing TSH receptor antibody levels (the primary pathologic mechanism in Graves disease). Propylthiouracil also decreases the conversion of T 4 to the more potent T 3 . However, the literature indicates a trend toward preoperative preparation with propranolol and iodides alone. This approach is quicker (i.e., 7-14 days vs. 2-6 weeks); it shrinks the thyroid gland, as does the more traditional approach; it decreases conversion of the prohormone T 4 into the more potent T 3 ; and it treats symptoms but may not correct abnormalities in left ventricular function. Regardless of the approach, antithyroid drugs should be administered on a long-term basis and on the morning of the surgical procedure. If emergency surgery is necessary before the euthyroid state is achieved, if subclinical hyperthyroidism progresses without adequate treatment, or if hyperthyroidism is out of control intraoperatively, intravenous administration of esmolol, 50 to 500 μ/kg, could be titrated to restore a normal heart rate (assuming the absence of heart failure). In addition, intravascular fluid volume and electrolyte balance should be restored. However, administering propranolol or esmolol does not always prevent “thyroid storm.” No specific anesthetic drug is preferred for surgical patients who have hyperthyroidism.
A patient with a large goiter and an obstructed airway can be managed in the same way as any other patient with a problematic airway. In this type of case, reviewing CT scans of the neck preoperatively may provide valuable information regarding the extent of compression. Maintenance of anesthesia usually presents little difficulty. Postoperatively, extubation of the trachea should be performed under optimal circumstances for reintubation in the event that tracheomalacia (the tracheal rings have been weakened and the trachea collapses) developed.
Of the many possible postoperative complications including: nerve injury, bleeding, metabolic abnormalities, and thyroid storm (discussed in the next section); bilateral recurrent laryngeal nerve trauma and hypocalcemic tetany are the most feared. Bilateral recurrent laryngeal nerve injury (secondary to trauma or edema) causes stridor and laryngeal obstruction as a result of unopposed adduction of the vocal cords and closure of the glottic aperture. Immediate endotracheal intubation is required, usually followed by tracheostomy to ensure an adequate airway. Fortunately, Lahey Clinic records indicate that this rare complication occurred only once in more than 30,000 thyroid operations. Unilateral recurrent nerve injury often goes unnoticed because of compensatory overadduction of the uninvolved cord. However, we often test vocal cord function before and after this operation by asking the patient to say “e” or “moon.” Unilateral nerve injury is characterized by hoarseness, whereas aphonia characterizes bilateral nerve injury. Selective injury to the adductor fibers of both recurrent laryngeal nerves leaves the abductor muscles relatively unopposed, and pulmonary aspiration is a risk. Selective injury to the abductor fibers leaves the adductor muscles relatively unopposed, and airway obstruction can occur.
The intimate involvement of the parathyroid gland with the thyroid gland can result in inadvertent hypocalcemia during surgery for thyroid disease. Complications related to hypocalcemia are discussed in the later section on this disorder.
Because postoperative hematoma can compromise the airway, neck and wound dressings are placed in a crossing fashion (rather than vertically or horizontally) and should be examined for evidence of bleeding before a patient is discharged from the recovery room.
Thyroid storm is the name for the clinical diagnosis of a life-threatening illness in a patient whose hyperthyroidism has been severely exacerbated by illness or surgery. Thyroid storm is characterized by hyperthermia or pyrexia, tachycardia, and striking alterations in consciousness. Its clinical appearance is similar to malignant hyperthermia, pheochromocytoma, and neuroleptic malignant syndrome, further complicating the differential. [CR] No laboratory tests are diagnostic of thyroid storm, and the precipitating (nonthyroidal) cause is the major determinant of survival. Therapy can include blocking the synthesis of thyroid hormones by administering antithyroid drugs and the release of preformed hormone with iodine. Blocking the sympathetic nervous system symptoms with reserpine, α- and β-receptor antagonists, or α 2 drugs may be exceedingly hazardous and requires skillful management with constant monitoring of the critically ill patient.
Thyroid dysfunction, either hyperthyroidism or hypothyroidism, develops in more than 10% of patients treated with the antiarrhythmic agent amiodarone. Approximately 35% of the drug’s weight is iodine, and a 200-mg tablet releases approximately 20 times the optimal daily dose of iodine. This iodine can lead to reduced synthesis of T 4 or increased synthesis. In addition, amiodarone inhibits the conversion of T 4 to the more potent T 3 . These patients receiving amiodarone are in need of special attention preoperatively and intraoperatively, not just because of the arrhythmia that led to such therapy, but to ensure that no perioperative dysfunction or surprises result from unsuspected thyroid hyperfunction or hypofunction.
Hypothyroidism is a common disease that has been detected in 5% of a large population in Great Britain, in 3% to 6% of a healthy older population in Massachusetts, in 4.5% of a medical clinic population in Switzerland, and in 8.5% of a large Turkish population presenting to an anesthesiology preoperative clinic. [CR] The apathy and lethargy that often accompany hypothyroidism frequently delay its diagnosis, thus the perioperative period may be the first opportunity to spot many such hypothyroid patients. However, hypothyroidism is usually subclinical, serum concentrations of thyroid hormones are in the normal range, and only serum TSH levels are elevated. The normal range of TSH is 0.3 to 4.5 milliunits/L, and TSH values of 5 to 15 milliunits/L are characteristic of this entity. In such cases, hypothyroidism may have little or no perioperative significance. However, a retrospective study of 59 mildly hypothyroid patients found that more hypothyroid patients than control subjects required prolonged postoperative intubation (9 of 59 vs. 4 of 59), had significant electrolyte imbalances (3 of 59 vs. 1 of 59), and bleeding complications (4 of 59 vs. 0 of 59). Because only a few charts were examined, these differences did not reach statistical significance. In another study, overt hypothyroidism later developed in a high percentage of patients with a history of subclinical hypothyroidism.
Overt hypothyroidism is associated with slow mental functioning, slow movement, slow reflexes, dry skin, arthralgias, carpal tunnel syndrome, periorbital edema, intolerance to cold, depression of the ventilatory responses to hypoxia and hypercapnia, impaired clearance of free water with or without hyponatremia, slow gastric emptying, sleep apnea, and bradycardia. In extreme cases, cardiomegaly, heart failure, pericardial and pleural effusions can develop, often presenting as orthopnea, dyspnea, or general fatigue. Hypothyroidism is often associated with amyloidosis, which may produce an enlarged tongue, cardiac conduction abnormalities, and renal disease. Hypothyroidism decreases the anesthetic requirement slightly. The tongue may be enlarged in a hypothyroid patient even in the absence of amyloidosis, and such enlargement may hamper endotracheal intubation.
An increasing TSH level is the most sensitive indicator of failing thyroid function. Ideal preoperative and preprocedure management of hypothyroidism consists of restoring normal thyroid status: the physicians should consider administering the normal dose of levothyroxine the morning of the surgical procedure, even though these drugs have long half-lives (1.4-10 days). Reduced GI absorption of levothyroxine may occur with the coadministration of cholestyramine or aluminum hydroxide, iron, a high-bran meal, or sucralfate or colestipol. For patients in myxedema coma who require emergency surgery, liothyronine (T3 hormone) can be given intravenously (with fear of precipitating myocardial ischemia, however) while supportive therapy is undertaken to restore normal intravascular fluid volume, body temperature, cardiac function, respiratory function, and electrolyte balance. [CR]
In hypothyroidism, respiratory control mechanisms and renal fluid balance do not function normally, however, the response to hypoxia and hypercapnia, and clearance of free water normalize with thyroid replacement therapy. Drug metabolism is anecdotally reported to be slowed, and awakening times from sedatives are reported to be prolonged by hypothyroidism. However, few formal studies, and none in humans, of the pharmacokinetics and pharmacodynamics of sedatives or anesthetic drugs in this population have been published. These concerns disappear when thyroid function is normalized preoperatively. Addison disease (with its relative steroid deficiency) is more common in hypothyroidism, and some endocrinologists routinely treat patients with noniatrogenic hypothyroidism with stress doses of steroids perioperatively because both conditions are commonly caused by autoimmune responses. The possibility that this steroid deficiency exists should be considered if the patient becomes hypotensive perioperatively. Body heat mechanisms are inadequate in hypothyroid patients, so temperature should be monitored and maintained, especially in patients requiring emergency surgery. Because of an increased incidence of myasthenia gravis in hypothyroid patients, a peripheral nerve stimulator is used to guide judicious administration of muscle relaxants.
Thyroid Nodules and Carcinoma
More than 90% of thyroid nodules are benign, yet identifying malignancy in a solitary thyroid nodule is a difficult and important procedure. Male patients and patients with previous radiation therapy to the head and neck have an increased likelihood of malignant disease in their nodules. Often, needle biopsy and scanning are sufficient for the diagnosis, but occasionally excisional biopsy is needed. Papillary carcinoma accounts for more than 70% of all thyroid carcinomas. Simple excision of lymph node metastases appears to be as efficacious as radical neck procedures for the patient’s survival. Follicular carcinoma, which accounts for approximately 15% of thyroid carcinomas, is more aggressive, and has a less favorable prognosis.
Medullary carcinoma, the most aggressive form of thyroid carcinoma, is associated with a familial occurrence of pheochromocytoma, as are parathyroid adenomas. For this reason, a history should be obtained from patients with a surgical scar in the thyroid region so that the possibility of occult pheochromocytoma can be assessed and excluded.
Disorders of Calcium Metabolism
The three substances that regulate serum concentrations of calcium, phosphorus, and magnesium are parathyroid hormone (parathyrin, PTH), calcitonin, and vitamin D, which act on bone, kidney, gut, and their own receptors. Calcium excess in blood is caused by either malignant disease or hyperparathyroidism in more than 90% of patients. PTH stimulates bone resorption, inhibits renal excretion of calcium, and increases conversion to active vitamin D, three conditions that lead to hypercalcemia. Calcitonin can be considered an antagonist to PTH. Through its metabolites, vitamin D aids in the absorption of calcium, phosphate, and magnesium from the gut and facilitates the bone resorptive effects of PTH. Secretion of PTH is modulated through the calcium-sensing receptor on the cell surface of parathyroid cells. An increase in ionized calcium stimulates this receptor and thus causes a decrease in PTH secretion. Recognition of this effect has led to reevaluation of the therapy for hyperparathyroidism inasmuch as a drug upregulating this receptor’s sensitivity reduces PTH levels.
Hyperparathyroidism and Hypercalcemia
Primary hyperparathyroidism occurs in approximately 0.1% of the population, most commonly begins in the third to fifth decades of life, and occurs two to three times more frequently in women than in men. Primary hyperparathyroidism usually results from enlargement of a single gland, commonly an adenoma and very rarely a carcinoma. Hypercalcemia almost always develops.
Calcium is the chief mineral component of the body; it provides structure to the skeleton and performs key roles in neural transmission, intracellular signaling, blood coagulation, and neuromuscular functioning. Ninety-nine percent of the 1000 g of calcium present in the average human body is stored in the bone mineral reservoir. Fifty percent to 60% is bound to plasma proteins or is complexed with phosphate or citrate. The normal total serum calcium level is 8.6 to 10.4 mg/dL, as measured in most laboratories; though this value depends on the albumin level, noting a decline of 0.8 mg/dL for each 1 g/dL drop in albumin. Binding of calcium to albumin depends on pH: binding decreases with acidic pH and increases with alkaline pH. Serum calcium, not ionized calcium, decreases with reductions in albumin levels. Although ionized calcium is the clinically significant fraction, the cost and technical difficulties of stabilizing the electrodes used for measurement have limited the available assays. Nevertheless, PTH and vitamin D 3 work to keep the level stable within 0.1 mg/dL in any individual.
Many of the prominent symptoms of hyperparathyroidism are a result of the hypercalcemia that accompanies it. Regardless of the cause, hypercalcemia can produce any of a number of symptoms, the most prominent of which involve the renal, skeletal, neuromuscular, and GI system, including anorexia, vomiting, constipation, polyuria, polydipsia, lethargy, confusion, renal calculi, pancreatitis, bone pain, and psychiatric abnormalities. Free intracellular calcium initiates or regulates muscle contraction, release of neurotransmitters, secretion of hormones, enzyme action, and energy metabolism.
Nephrolithiasis occurs in 60% to 70% of patients with hyperparathyroidism. Sustained hypercalcemia can result in tubular and glomerular disorders, including proximal (type II) renal tubular acidosis. Polyuria and polydipsia are common complaints.
Skeletal disorders related to hyperparathyroidism are osteitis fibrosa cystica, simple diffuse osteopenia, and osteoporosis. The rate of bone turnover is five times higher in patients with hyperparathyroidism than in normal controls. Patients may have a history of frequent fractures or may complain of bone pain, especially in the anterior margin of the tibia.
Because free intracellular calcium initiates or regulates muscle contraction, neurotransmitter signaling, hormone secretion, enzyme action, and energy metabolism, abnormalities in these end organs are often symptoms of hyperparathyroidism. Patients may experience profound muscle weakness, especially in proximal muscle groups, as well as muscle atrophy. Depression, psychomotor retardation, and memory impairment may occur. Lethargy and confusion are frequent complaints.
Peptic ulcer disease is more common in these patients than in the rest of the population. Production of gastrin and gastric acid is increased. Anorexia, vomiting, and constipation may also be present.
Approximately one third of all hypercalcemic patients are hypertensive, but the hypertension usually resolves with successful treatment of the primary disease. Neither hypertension nor minimally invasive surgery seems to alter the perioperative risk associated with surgery in such patients in comparison with the usual hypertensive patients. Even octogenarians with asymptomatic hyperparathyroidism can be operated on without mortality and with morbidity no different from that in younger individuals, thus encouraging the use of parathyroidectomy as preventive therapy. Long-standing hypercalcemia can lead to calcifications in the myocardium, blood vessels, brain, and kidneys. Cerebral calcifications may cause seizures, whereas renal calcifications lead to polyuria that is unresponsive to vasopressin.
The most useful confirmatory test for hyperparathyroidism is radioimmunoassay for PTH. In fact, two changes have radically reduced anesthesia involvement in the care of patients with primary hyperparathyroidism. One change, the use of the calcimimetic drug class which modulates the calcium-sensitive PTH cell receptor and thereby decreases calcium levels, has been emphasized in older individuals. The other change is use of minimally invasive approaches after imaging procedures with just local anesthesia or a cervical plexus block—as with thyroidectomy. Most surgeons now performing minimally invasive parathyroid removal monitor PTH levels intraoperatively to determine whether the causative adenoma has been resected. The baseline PTH level should be determined before induction of anesthesia because even monitored anesthesia care increases PTH levels. In hyperparathyroid patients, hormone levels are abnormal for a given level of calcium. The level of inorganic phosphorus in serum is usually low, but it may be within normal limits. Alkaline phosphatase levels are elevated if considerable skeletal involvement is present.
Glucocorticoid administration reduces the level of calcium in blood in many other conditions that cause hypercalcemia, but not usually in primary hyperparathyroidism. In sarcoidosis, multiple myeloma, vitamin D intoxication, and some malignant diseases, all of which can cause hypercalcemia, administration of glucocorticoids may lower serum calcium levels through an effect on GI absorption. This effect occurs to a lesser degree in primary hyperparathyroidism.
Hypercalcemia may also occur as a consequence of secondary hyperparathyroidism in patients who have chronic renal disease. When phosphate excretion decreases as a result of decreased nephron mass, serum calcium levels fall because of deposition of calcium and phosphate in bone. Secretion of PTH subsequently increases, and this causes the fraction of phosphate excreted by each nephron to increase. Eventually, the chronic intermittent hypocalcemia of chronic renal failure leads to chronically high levels of serum PTH and hyperplasia of the parathyroid glands—one of the entities termed secondary hyperparathyroidism.
Symptomatic primary hyperparathyroidism in patients younger than 50 years or with serum calcium levels more than 1 mg/dL higher than the upper limit of normal, a 30% or greater reduction in the glomerular filtration rate (GFR), or severe bone demineralization is usually treated surgically. If the patient refuses surgery or if other illnesses render surgery inadvisable, medical management with the calcimimetic, cinacalcet, makes management much more feasible. The difficulty with such management is that the hyperfunctioning parathyroid glands secrete more hormone as the serum calcium concentration is lowered—as though the calcium set point for feedback regulation of PTH secretion had been raised. Blanchard and colleagues demonstrated that patients with “asymptomatic” primary hyperparathyroidism have clinical improvement of their symptoms postoperatively even after 1 year, noting younger patients and those with higher preoperative calcium levels show the best improvement.
Patients with moderate hypercalcemia who have normal renal and cardiovascular function present no special preoperative and preprocedure problems. The ECG can be examined preoperatively and intraoperatively for shortened PR or QT intervals ( Fig. 32.3 ). Because severe hypercalcemia can result in substantial hypovolemia, normal intravascular volume and electrolyte status should be evaluated and then restored before anesthesia and surgery.
Management of hypercalcemia preoperatively should include (even in urgent or emergency situations) treatment of the underlying cause, a frequent strategy in surgical patients with malignancy-associated hypercalcemia. Therapy preoperatively for both malignant and nonmalignant causes of hypercalcemia include aggressive volume repletion, with the addition of diuresis only if volume overload develops. Intravenous fluid infusion rates of 250 to 500 mL/h preoperatively are commonly used to maintain urine output greater than 200 mL/h. [CR] Careful monitoring during this time is needed to avoid administration of an excessive amount of intravenous fluids, as many patients may have compromised cardiac function. In the setting of fluid overload, diuresis with furosemide can be warranted; however, evidence for benefit is limited and mainly theoreteical. [CR] Other complications of these interventions include hypomagnesemia and hypokalemia.
In emergency situations, vigorous expansion of intravascular volume usually reduces serum calcium to a safe level (<14 mg/dL). Phosphate should be given to correct hypophosphatemia because it decreases calcium uptake into bone, increases calcium excretion, and stimulates breakdown of bone. Hydration, accompanied by electrolyte repletion mainly phosphate, suffices in the management of most hypercalcemic patients. Other measures to decrease reabsorption of bone include the bisphosphonates pamidronate sodium (90 mg intravenously) and zoledronate (4 mg intravenously). Case reports note success in the setting of extreme hypercalcemia (>20 mg/dL) correction with a low calcium bath dialysate. [CR]
Calcitonin lowers serum calcium levels through direct inhibition of bone resorption. It can decrease serum calcium levels within minutes after intravenous administration. Side effects include urticaria and nausea. It is so rapid acting that it can be used to reduce calcium levels while waiting for hydration and a bisphosphonate to take effect.
It is especially important to know whether the hypercalcemia has been chronic because serious cardiac, renal, or CNS abnormalities may have resulted.
Hypocalcemia (caused by hypoalbuminemia, hypoparathyroidism, hypomagnesemia, hypovitaminosis D, hungry bone syndrome after correction of hyperparathyroidism, anticonvulsant therapy, citrate infusion, or chronic renal disease) is not usually accompanied by a clinically evident cardiovascular disorder. The most common cause of hypocalcemia is hypoalbuminemia. In true hypocalcemia (i.e., when the free calcium concentration is low), myocardial contractility is affected, noting myocardial contractility varies directly with levels of blood ionized calcium. The clinical signs of hypocalcemia include clumsiness; convulsions; laryngeal stridor; depression; muscle stiffness; paresthesias; parkinsonism; tetany; Chvostek sign; dry and scaly skin, brittle nails, and coarse hair; low serum concentrations of calcium; prolonged QT intervals; soft tissue calcifications; and Trousseau sign.
Hypocalcemia delays ventricular repolarization, hence increasing the QTc interval (normal, 0.35-0.44 second). With electrical systole prolonged, the ventricles may fail to respond to the next electrical impulse from the sinoatrial node, with second-degree heart block resulting. Prolongation of the QT interval is a moderately reliable ECG sign of hypocalcemia, not for the population as a whole, but for the individual patient. Thus monitoring the QT interval as corrected for the heart rate is a useful, but not always accurate means of monitoring hypocalcemia in any individual patient (see Fig. 32.3 ). Heart failure may also occur with hypocalcemia, but is rare. Because heart failure in patients with coexisting heart disease is reduced in severity when calcium and magnesium ion levels are restored to normal, these levels may be normalized preoperatively in a patient with impaired exercise tolerance or signs of cardiovascular dysfunction; normalization can be achieved intravenously over a 15-minute period if absolutely necessary. Sudden decreases in blood levels of ionized calcium (as with chelation therapy) can result in severe hypotension.
Patients with hypocalcemia may have seizures. They may be focal, petit mal, or grand mal in appearance, often indistinguishable from such seizures in the absence of hypocalcemia. Patients may also have a type of seizure called cerebral tetany, which consists of generalized tetany followed by tonic spasms. Therapy with standard anticonvulsants is ineffective and may even exacerbate these seizures (by an anti–vitamin D effect), calcium must be repleted for treatment. In long-standing hypoparathyroidism, calcifications may appear above the sella; these calcifications represent deposits of calcium in and around small blood vessels of the basal ganglia. They may be associated with a variety of extrapyramidal syndromes.
The most common cause of acquired hypoparathyroidism is surgery of the thyroid or parathyroid glands. Other causes include autoimmune disorders, therapy with iodine-131, hemosiderosis or hemochromatosis, neoplasia, and granulomatous disease. Idiopathic hypoparathyroidism has been divided into three categories: an isolated persistent neonatal form, branchial dysembryogenesis, and autoimmune candidiasis related to multiple endocrine deficiency.
Pseudohypoparathyroidism and pseudopseudohypoparathyroidism are rare hereditary disorders characterized by short stature, obesity, rounded face, and shortened metacarpals. Patients with pseudohypoparathyroidism have hypocalcemia and hyperphosphatemia despite high serum levels of PTH. These patients have a deficient end-organ response to PTH as a result of abnormalities in G-protein function.
Because treatment of hypoparathyroidism is not surgical, hypoparathyroid patients who come to the operating room are those who require surgery for unrelated conditions. Their calcium, phosphate, and magnesium levels should be measured both preoperatively and postoperatively. Patients with symptomatic hypocalcemia may be treated with intravenous calcium gluconate preoperatively. Initially, 10 to 20 mL of 10% calcium gluconate may be given at a rate of 5 mL/min. The effect on serum calcium levels is of short duration, but a continuous infusion with 10 mL/min of 10% calcium gluconate in 500 mL of solution over a period of 6 hours helps keep serum calcium at adequate levels. For severe symptoms in emergent settings, 10 mL of 10% calcium chloride may be given over 10 minutes, followed by a 10% calcium gluconate infusion. Magnesium and phosphate levels may also require normalization to normalize cardiovascular and nervous system function.
The objective of therapy is to bring the symptoms under control before the surgical procedure and anesthesia. For patients with chronic hypoparathyroidism, the objective is to keep the serum calcium level in the lower half of the normal range. A preoperative and preprocedure ECG is useful for maintaining the QTc interval. The preoperative and preprocedure QTc value may be used as a guide to the serum calcium level if rapid laboratory assessment is not possible. Changes in the calcium level may alter the duration of muscle relaxation; thus careful monitoring and titration of muscle relaxation with a twitch monitor is necessary.
The intimate involvement of the parathyroid gland with the thyroid gland can result in unintentional hypocalcemia during surgery for diseases of either organ. Because of the affinity of their bones for calcium, this relationship is crucial in patients with advanced osteitis. Internal redistribution of magnesium, calcium, or both ions may occur (into “hungry bones”) after parathyroidectomy and may cause hypomagnesemia, hypocalcemia, or both. Because the tendency to tetany increases with alkalosis, hyperventilation should be avoided. The most prominent manifestations of acute hypocalcemia are distal paresthesias and muscle spasm (tetany). Potentially fatal complications of severe hypocalcemia include laryngeal spasm and hypocalcemic seizures. The clinical sequelae of magnesium deficiency include cardiac arrhythmias (principally ventricular tachyarrhythmias), hypocalcemic tetany, and neuromuscular irritability independent of hypocalcemia (tremors, twitching, asterixis, and seizures).
In addition to monitoring total serum calcium or ionized calcium postoperatively, one can test for the Chvostek and Trousseau signs. The Chvostek sign is a contracture of the facial muscles produced by tapping the ipsilateral facial nerve at the angle of the jaw, of note this sign can be elicited in up to 15% of patients who are not hypocalcemic, an attempt should be made to elicit this sign preoperatively to ensure that its appearance is meaningful. The Trousseau sign is elicited by applying a blood pressure cuff at a level slightly above the systolic level for a few minutes. The resulting carpopedal spasm, with contraction of the fingers and inability to open the hand, stems from the increased muscle irritability in hypocalcemic states, aggravated by ischemia produced by the blood pressure cuff.
Fifty percent of women who are older than 65 years sustain an osteoporotic fracture. (Because men are living longer, osteoporosis has become an increasing problem for them, too, and reports indicate a 15% per decade hip fracture rate for men >65 years old.) Men with COPD (even without steroid treatment) are at high risk for vertebral fractures. Furthermore, in either gender, each vertebral fracture is associated with up to a 10% decrease in lung capacity. Diagnosis and treatment of these conditions have increased with routine use of dual-energy x-ray absorptiometry or quantitative ultrasonography. Because T-scores and Z-scores were developed to relate changes in white postmenopausal women to those at age 21 years, care must be used in interpreting the results. Known risk factors for osteoporosis include age, relative lifetime estrogen deficiency (late menarche, amenorrhea, early menopause, nulliparity), deficiency of dietary calcium, tobacco use, increased aerobic exercise in combination with decreased weight-bearing exercise, decreased weight-bearing exercise by itself, use of soft drinks, and Asian or white ancestry. Although therapy for osteoporosis (use of biphosphates, bone mineral depositors, weight-bearing exercises, calcium, vitamin D, estrogen, and now designer estrogens that may be useful for men, such as raloxifene [Evista]) does not have major known implications for anesthesia care. Bone fractures in such patients have occurred on movement to and from an operating table, therefore these patients should be allowed to move and position themselves when possible. Recombinant PTH and calcitonin are also used, but again, no reports of perioperative interactions have been prominent.
Anterior Pituitary Hypersecretion
The anterior pituitary gland (or master endocrine gland) consists of five identifiable types of secretory cells and hormones produced: somatotrophs (GH), corticotrophs (ACTH), lactotrophs (prolactin), gonadotrophs (luteinizing hormone and follicle-stimulating hormone), and thyrotrophs (TSH). Secretion of these pituitary hormones is largely regulated by a negative-feedback loop by hypothalamic regulatory hormones and by signals that originate from the target site of pituitary action. Six hypothalamic hormones have been characterized: dopamine, the prolactin-inhibiting hormone; somatostatin, the GH release–inhibiting hormone; GH-releasing hormone; corticotropin-releasing hormone; gonadotropin-releasing hormone; and TRH. Most pituitary tumors (>60%) are hypersecretory and are classified according to the excess production of a specific anterior pituitary hormone.
The most common disorders of pituitary hypersecretion are those related to excesses of prolactin (amenorrhea, galactorrhea, and infertility), ACTH (Cushing syndrome), and GH (acromegaly). In addition to knowing the pathophysiologic processes of the disease involved, the anesthesiologist must determine whether the patient recently underwent air pneumoencephalography (almost obsolete, but still used rarely). If so, nitrous oxide should not be used to lessen the risk of intracranial hypertension from gas collection. CT or MRI of the sella has largely replaced neuroencephalography.
More than 99% of cases of acromegaly are attributable to pituitary adenoma (or use of recombinant human GH). Thus the primary treatment of acromegaly is transsphenoidal surgery (or withdrawal of drug) and symptomatic treatment of the carpal tunnel or other syndromes provoked. If the pituitary tumor is not totally removed, patients are often offered external pituitary irradiation. In the case of suprasellar extension, conventional transfrontal hypophysectomy is often performed. The dopaminergic agonist bromocriptine can lower GH levels, but the long-term follow-up with this drug is not favorable. Octreotide, a long-acting analogue of somatostatin, now given in depot form approximately once a month, produces effective palliation in 50% of patients. Other medical therapies such as pegvisomant or somatostatin analogues are also medications that have been tried before surgical intervention. In 2011, new guidelines were published with few changes to the available recommendations. However, the new guidelines reported some evidence that medication taken preoperatively may result in a better postoperative outcome.
Difficulty with endotracheal intubation should be anticipated in a patient with acromegaly; lateral neck radiographs or CT scans of the neck and direct or indirect visualization can identify patients with subglottic stenosis or an enlarged tongue, mandible, epiglottis, or vocal cords. If placement of an arterial line is necessary, a brachial or femoral site may be preferable to a radial site.
Anterior Pituitary Hypofunction
Anterior pituitary hypofunction results in deficiency of one or more of the following hormones: GH, TSH, ACTH, prolactin, or gonadotropin. No special preoperative and preprocedure preparation is required for a patient deficient in prolactin or gonadotropin; deficiency in GH, however, can result in atrophy of cardiac muscle, a condition that may necessitate preoperative and preprocedure cardiac evaluation. Nonetheless, anesthetic problems have not been documented in patients with isolated GH deficiency. Acute deficiencies are another matter.
Acute pituitary deficiency is often caused by bleeding into a pituitary tumor. In surgical specimens of resected adenomas, as many as 25% show evidence of hemorrhage. These patients often have acute headache, visual loss, nausea or vomiting, ocular palsy, disturbances of consciousness, fever, vertigo, or hemiparesis. In such patients, rapid transsphenoidal decompression should be accompanied by consideration of replacement therapy, including glucocorticoids and the treatment of increased intracranial pressure.
Obstetric anesthesiologists are often aware of these pituitary failure problems; Sheehan syndrome is the clinical manifestation of pituitary infarction associated with hypotension during or after obstetric hemorrhage. Conditions that strongly suggest this diagnosis are failure to start postpartum lactation, increasing fatigue, cold intolerance, and especially hypotension unresponsive to volume replacement and pressors; treatment is prompt hormone therapy. [CR]
Posterior Pituitary Hormone Excess and Deficiency
Secretion of vasopressin or antidiuretic hormone (ADH) is enhanced by increased serum osmolality or the presence of hypotension. Inappropriate secretion of vasopressin, without relation to serum osmolality, results in hyponatremia and fluid retention. This inappropriate secretion can result from the following: a variety of CNS lesions; drugs including nicotine, narcotics, tramadol, chlorpropamide, clofibrate, vincristine, vinblastine, and cyclophosphamide; and pulmonary infections, hypothyroidism, adrenal insufficiency, and ectopic production from tumors. Preoperative and preprocedure management of a surgical patient with inappropriate secretion of vasopressin includes appropriate treatment of the causative disorders and restriction of water. Occasionally, drugs that inhibit the renal response to ADH (e.g., lithium or demeclocycline) should be administered preoperatively to restore normal intravascular volume and electrolyte status.
Most of the clinical features associated with the syndrome of inappropriate secretion of antidiuretic hormone (SIADH) are related to hyponatremia and the resulting brain edema; such features include weight gain, weakness, lethargy, mental confusion, obtundation, and disordered reflexes and may culminate in convulsions and coma.
Investigators have recognized that up to 20% of long-distance runners have SIADH with increased vasopressin secretion. Because such people infrequently undergo surgical treatment of injuries, SIADH symptoms and laboratory evaluation may be routine for that group as well.
SIADH should be suspected in any patient with hyponatremia who excretes urine that is hypertonic relative to plasma. The following laboratory findings further support the diagnosis:
Urinary sodium greater than 20 mEq/L
Low serum levels of BUN, creatinine, uric acid, and albumin
Serum sodium lower than 130 mEq/L
Plasma osmolality lower than 270 mOsm/L
Urine hypertonic relative to plasma
Noting the response to water loading is a useful way of evaluating patients with hyponatremia. Patients with SIADH are unable to excrete dilute urine even after water loading. Assay of ADH in blood can confirm the diagnosis. Too vigorous treatment of chronic hyponatremia can result in disabling osmotic demyelination syndrome. The increase in serum sodium should not be greater than 1 mEq/L/h.
Patients with mild to moderate symptoms of water intoxication can be treated with restriction of fluid intake to approximately 500 to 1000 mL/day. Patients with severe water intoxication and CNS symptoms may need vigorous treatment consisting of intravenous administration of hypertonic saline solutions until symptoms resolve, followed by fluid restriction.
Treatment should be directed at the underlying problem. If SIADH is drug induced, use of the drug should be withdrawn. Inflammation should be treated with appropriate measures, and neoplasms should be managed by surgical resection, irradiation, or chemotherapy, whichever is indicated.
No drugs are available that can suppress release of ADH from the neurohypophysis or from a tumor. Phenytoin (Dilantin) and narcotic antagonists such as naloxone and butorphanol have some inhibiting effect on physiologic ADH release but are clinically ineffective in patients with SIADH. Drugs that block the effect of ADH on renal tubules include lithium, which is rarely used because its toxicity often outweighs its benefits, and demethylchlortetracycline in doses of 900 to 1200 mg/day. Demethylchlortetracycline interferes with the ability of the renal tubules to concentrate urine, thereby causing excretion of isotonic or hypotonic urine and lessening the hyponatremia. This drug can be used in ambulatory patients with SIADH when it is difficult to restrict fluids.
When a patient with SIADH comes to the operating room for any surgical procedure, fluids are managed by measuring volume status, when the clinical picture is unclear the use of arterial wave form analysis, central venous pressure, pulmonary artery pressure, or transthoracic and/or transesophageal echocardiography may be of value. Despite the common impression that SIADH is frequently seen in older patients in the postoperative period, studies have shown that the patient’s age and the type of anesthetic used have no bearing on the postoperative development of SIADH. It is common to see several patients in the neurosurgical ICU suffering from this syndrome. The diagnosis is usually one of exclusion. Patients with SIADH generally require only restriction of intravenous fluids, very rarely is hypertonic saline needed.
Lack of ADH, which results in diabetes insipidus, is caused by pituitary disease, brain tumors, infiltrative diseases such as sarcoidosis, head trauma (including trauma after neurosurgery), or lack of a renal response to ADH. The last can result from such diverse conditions as hypokalemia, hypercalcemia, sickle cell anemia, obstructive uropathy, and renal insufficiency. Preoperative or preprocedure treatment of diabetes insipidus consists of restoring normal intravascular volume by replacing urinary losses, administering desmopressin acetate (DDAVP) nasally, and giving daily fluid requirements intravenously.
Perioperative management of patients with diabetes insipidus is based on the extent of the ADH deficiency. Management of a patient with complete diabetes insipidus and a total lack of ADH does not usually present any major problem as long as the side effects of the drug are avoided and the presence of the condition is known preoperatively. Just before the surgical procedure, the patient is given the usual dose of DDAVP intranasally or an intravenous bolus of 100 milliunits of aqueous vasopressin, followed by a constant infusion of 100 to 200 milliunits/h. The dose is usually adjusted to permit the daily breakthrough polyuria that prevents the iatrogenic syndrome of SIADH. All the intravenous fluids given intraoperatively should be isotonic to reduce the risk of water depletion and hypernatremia. Plasma osmolality should be frequently measured, both intraoperatively and immediately postoperatively. If plasma osmolality rises much higher than 300 mOsm/L, hypotonic fluids can be administered; the rate of the intraoperative vasopressin infusion can be increased to greater than 200 milliunits/h.
For patients who have a partial deficiency of ADH, it is not necessary to use aqueous vasopressin perioperatively unless plasma osmolality rises to more than 300 mOsm/L. Nonosmotic stimuli (e.g., volume depletion) and the stress of surgery usually cause the release of large quantities of ADH perioperatively. Consequently, these patients require only frequent monitoring of plasma osmolality during this period.
Because of side effects, the dose of vasopressin should be limited to that necessary for control of diuresis. The oxytocic and coronary artery–constricting properties of vasopressin make this limit especially applicable to patients who are pregnant or have CAD.
Diseases Involving the Cardiovascular System
Analysis of the perioperative treatment of hypertension is important because of the prevalence of the condition (33.5% of adults aged 20 and over in the United States), the great risk in perioperative care of a hypertensive patient, and the high cost of unnecessary delays in surgical treatment. Numerous studies over the years have evaluated the impact of hypertension as one of the risk factors for cardiac morbidity. However, the need to delay surgery because of poorly controlled hypertension has been questioned. Weksler and colleagues studied 989 hypertensive patients who were treated on a long-term basis and who underwent noncardiac surgery with diastolic blood pressure between 110 and 130 mm Hg and no previous MI, unstable or severe angina pectoris, renal failure, pregnancy-induced hypertension, left ventricular hypertrophy, previous coronary revascularization, aortic stenosis, preoperative dysrhythmias, conduction defects, or stroke. The control group had their surgical procedures postponed and remained in the hospital for control of blood pressure, and the study patients received 10 mg of nifedipine intranasally. No statistically significant differences in postoperative complications were observed, thus suggesting that this subset of patients without significant cardiovascular comorbid conditions can proceed with surgery despite elevated blood pressure on the day of the operation.
Several studies have assessed the relationship between cardiovascular disease and preoperative hypertension. In a multicenter study of patients undergoing coronary artery bypass graft (CABG), the presence of isolated systolic hypertension was associated with a 30% increased incidence of perioperative cardiovascular complications when compared with normotensive individuals. Kheterpal and colleagues integrated data from their anesthesia information system (AIMS) and the American College of Surgeons National Surgical Quality Improvement Project (NSQIP) and found hypertension to be one of the independent predictors of events. Wax and colleagues used AIMS to identify independent predictors of troponin elevation or death, and independent predictors of adverse outcome included increased baseline systolic blood pressure (SBP), intraoperative diastolic blood pressure lower than 85 mm Hg, increased intraoperative heart rate, blood transfusion, and anesthetic technique, controlling for standard risk factors. A delay of surgery did not result in interval normalization of blood pressure.
Although preoperative blood pressure (both systolic and diastolic) is a significant predictor of postoperative morbidity, no data definitively establish whether preoperative treatment of hypertension reduces perioperative risk. Until a definitive study is performed, we recommend letting the weight of evidence guide preoperative treatment of a patient with hypertension. Such treatment would be based on three general beliefs: (1) the patient should be educated regarding the importance of lifelong treatment of hypertension, even isolated systolic hypertension; (2) perioperative hemodynamic fluctuations occur less frequently in treated than in untreated hypertensive patients (as demonstrated by Prys-Roberts and colleagues and confirmed by Goldman and Caldera and Mangano and associates ); and (3) hemodynamic fluctuations have some relation to morbidity. Kheterpal and colleagues demonstrated that patients who sustained a cardiac adverse event were more likely to experience an episode of mean arterial pressure lower than 50 mm Hg, an episode of 40% decrease in mean arterial pressure, and an episode of heart rate higher than 100 beats/min. The data of Pasternack and colleagues and Weksler and associates imply that rapid correction of blood pressure or prevention of increases in heart rate may be all that is needed. Sessler and colleagues (2018) studied 9765 patients in the POISE-II trial to assess the relationship between perioperative hypotension and a composite of MI and death within 30-days of surgery. Intraoperatively, the estimated average relative effect was 1.08 (98.3% confidence interval [CI], 1.03, 1.12; P < .001) per 10-minute increase in hypotension duration. The average relative effect odds ratio was 2.83 (98.3% CI, 1.26, 6.35; P = .002) in patients with hypotension during the subsequent 4 days of hospitalization. The Intraoperative Norepinephrine to Control Arterial Pressure (INPRESS) study was a multicenter, randomized, parallel-group clinical trial in adult patients ( n = 298) at increased risk of postoperative complications of individualized management strategy aimed at achieving a SBP within 10% of the reference value (i.e., patient’s resting SBP) or standard management strategy of treating SBP less than 80 mm Hg or lower than 40% from the reference value during and for 4 hours following surgery. [CR] Management targeting an individualized SBP, compared with standard management, reduced the risk of postoperative organ dysfunction. Taken together, these data suggest that maintenance of normal blood pressure is critical in patients with hypertension.
The INPRESS study demonstrated that preoperative data should be used to determine the individualized range of suitable arterial blood pressure values that are tolerable by a particular patient during and after a surgical procedure. Importantly, hypotension in patients at risk for a cerebrovascular event should be avoided. For example, the POISE (Perioperative Ischemic Evaluation) study demonstrated that short-term β-blocker administration resulted in an increased incidence of stroke and death that was associated with an increased rate of hypotension.
Preoperative Administration of All Antihypertensive Drugs
Continuation of all antihypertensive drugs preoperatively should be considered, except ACE inhibitors or angiotensin II antagonists, for which no clear consensus exists. Coriat and colleagues found that ACE inhibitors were associated with hypotension in 100% of patients during induction versus approximately 20% in whom ACE inhibitors were withheld on the morning of the surgical procedure. Bertrand and coworkers performed a prospective randomized study that demonstrated that more severe hypotensive episodes requiring vasoconstrictor treatment occurred after induction of general anesthesia in patients treated on a long-term basis with an angiotensin II antagonist and receiving the drug on the morning before the operation than in those in whom angiotensin II antagonists were discontinued on the day before the surgical procedure. Kheterpal and colleagues performed a propensity-matched analysis of 12,381 noncardiac surgical cases. Patients with long-term ACE inhibitor or angiotensin receptor blocker (ARB) and diuretic therapy showed more periods with a mean arterial blood pressure lower than 70 mm Hg, periods with a 40% decrease in SBP, periods with a 50% decrease in SBP, and vasopressor boluses than did patients receiving diuretic therapy alone. If these drugs are continued, vasopressin is the drug of choice for refractory hypotension. Investigators at the Cleveland Clinic evaluated 79,228 patients (9905 ACE inhibitor users [13%] and 66,620 [87%] non–ACE inhibitor users) who had noncardiac surgery between 2005 and 2009. These investigators did not find any association between use of ACE inhibitors and intraoperative or postoperative upper airway complications. ACE inhibitor use was not associated with in-hospital complications or increased 30-day mortality. Investigators of the VISION trial studied the relationship between withholding ACE inhibitors/angiotensin II receptor blockers and a primary composite outcome of all-cause death, stroke, or myocardial injury after noncardiac surgery at 30 days. Withholding ACE inhibitors/angiotensin II receptor blockers before major noncardiac surgery was associated with a lower risk of death and postoperative vascular events (150/1245 [12.0%] vs. 459/3557 [12.9%]; adjusted relative risk, 0.82; 95% CI, 0.70-0.96). In an accompanying editorial, London suggested that the current study does provide strong impetus for a randomized trial but does not warrant changes in local practice until such a trial is completed.
Ischemic Heart Disease
Preoperative evaluation of a patient with ischemic heart disease and a discussion of the AHA/ACC guidelines can be found in Chapters 31 and 54 . New guidelines were published in 2014 by both the AHA/ACC and the European Society of Cardiology as well as the Canadian Cardiovascular Society in 2017. This chapter will focus on the AHA/ACC guideline approach.
Role of Coronary Artery Bypass Graft or Percutaneous Coronary Interventions Before Noncardiac Surgical Procedures
Coronary revascularization may reduce the perioperative risk before noncardiac surgery, but the evidence suggests that it is limited to those with indications similar to the nonsurgical arena. The strongest retrospective evidence comes from the Coronary Artery Surgery Study registry, which enrolled patients from 1978 to 1981. Operative mortality in patients with CABG performed before noncardiac surgery was 0.9% but was significantly higher at 2.4% in patients without previous CABG. However, a 1.4% mortality rate was associated with the CABG procedure itself.
The benefit of percutaneous coronary intervention (PCI) before noncardiac surgery has also been examined in several cohort studies. Posner and colleagues used an administrative dataset of patients who underwent PCI and noncardiac surgery in Washington State. These investigators matched patients with coronary disease who were undergoing noncardiac surgery with and without previous PCI and looked at cardiac complications. In this nonrandomized design, Posner and colleagues noted a significantly lower rate of 30-day cardiac complications in patients who underwent PCI at least 90 days before the noncardiac surgery. However, PCI within 90 days of noncardiac surgery did not improve outcome. Although the explanation for these results is unknown, they may support the notion that PCI performed “to get the patient through surgery” may not improve perioperative outcome because cardiac complications may not occur in patients with stable or asymptomatic coronary stenosis. PCI may actually destabilize coronary plaque, which becomes manifest in the days or weeks after noncardiac surgery.
Godet and associates studied a cohort of 1152 patients after abdominal aortic surgery in which 78 patients underwent PCI. In the PCI group, the observed percentages of patients with a severe postoperative coronary event (9.0%; 95% CI, 4.4-17.4) or death (5.1% [95% CI, 2.0-12.5]) were not significantly different from the expected percentages (8.2% and 6.9%, respectively), which was confirmed by propensity analysis. PCI did not seem to significantly limit cardiac risk or death after aortic surgery.
Several randomized trials have addressed the value of testing and CABG or PCI, or both, in a subset of patients. McFalls and colleagues reported the results of a multicenter randomized trial in the VA Health System in which patients with documented CAD on coronary angiography, excluding those with left main CAD or a severely depressed ejection fraction (<20%), were randomized to CABG (59%), or percutaneous transluminal coronary angioplasty (PTCA; 41%) versus routine medical therapy. At 2.7 years after randomization, mortality in the revascularization group was not significantly different (22%) from that in the no-revascularization group (23%; Fig. 32.4 ). Within 30 days after the vascular operation, postoperative MI, defined by elevated troponin levels, occurred in 12% of the revascularization group and in 14% of the no-revascularization group ( P = .37). The authors suggested that coronary revascularization is not indicated in patients with stable CAD, and their results further support the lack of efficacy of PCI or CABG for single- or double-vessel disease before noncardiac surgery. However, in a follow-up analysis, Ward and coauthors reported improved outcome in the subset of patients who underwent CABG versus PCI.
Poldermans and colleagues randomized 770 patients about to undergo major vascular surgery and considered to have intermediate cardiac risk, defined as the presence of 1 or 2 cardiac risk factors, to either undergo further risk stratification with stress imaging or proceed directly to surgery. All patients received bisoprolol with a targeted heart rate of 60 to 65 beats/min initiated before and continued after the surgical procedure. The 30-day incidence of cardiac death and nonfatal MI was similar in both groups (1.8% in the no-testing group vs. 2.3% in the tested group). The conclusions of the authors were that further risk stratification in this group of patients considered to be at intermediate risk based on clinical history alone was unnecessary as long as perioperative β-blockers were used and that testing only delayed necessary vascular surgery. In a pilot study, Poldermans and associates tested patients with more than three risk factors; 101 (23%) showed extensive ischemia and were randomly assigned to revascularization ( n = 49) or no revascularization. Revascularization did not improve 30-day outcome; the incidence of the composite end-point was 43% versus 33% (odds ratio [OR], 1.4; 95% CI, 0.7-2.8; P = .30). In addition, no benefit during 1-year follow-up was observed after coronary revascularization (49% vs. 44%; OR, 1.2; 95% CI, 0.7-2.3; P = .48). Concern was expressed by Erasmus University (Rotterdam, the Netherlands) regarding the scientific integrity of studies led by Poldermans, as detailed in Erasmus MC Follow-up Investigation Committee: Report on the 2012 follow-up investigation of possible breaches of academic integrity, September 30, 2012 ( www.forbes.com/sites/larryhusten/2012/10/09/erasmus-medical-center-releases-final-report-on-cardiovascular-research-scandal/#675d592528ae ). The articles have not been retracted, but these data should be viewed with some skepticism. The authors of the 2014 AHA/ACCF Guidelines decided that the nonretracted decrease publications and/or other derivative studies by Poldermans that are relevant to the topic can only be cited in the text with a comment about the finding compared with the current recommendation but did not form the basis of that recommendation.
One issue in interpreting the results is that the length of time between coronary revascularization and noncardiac surgery most likely has an impact on its protective effect and potential risks. Back and coworkers studied 425 consecutive patients undergoing 481 elective major vascular operations at an academic VA Medical Center. Coronary revascularization was classified as recent (CABG < 1 year; PTCA < 6 months) in 35 cases (7%), prior (CABG >1 year and ≤5 years; PTCA >6 months and ≤2 years) in 45 cases (9%), and remote (CABG ≥ 5 years; PTCA ≥ 2 years) in 48 cases (10%). Outcomes in patients with previous PTCA were similar to those after CABG ( P = .7). Significant differences in adverse cardiac events and mortality were found among patients with CABG performed within 5 years or PTCA within 2 years (6.3%, 1.3%, respectively), individuals with remote revascularization (10.4%, 6.3%), and nonrevascularized patients stratified at high risk (13.3%, 3.3%) or intermediate/low risk (2.8%, 0.9%). The authors concluded that previous coronary revascularization (CABG < 5 years; PTCA < 2 years) may provide only modest protection against adverse cardiac events and mortality after major arterial reconstruction.
PCI using coronary stenting poses several special issues. Kaluza and associates reported on the outcome of 40 patients who underwent prophylactic coronary stent placement less than 6 weeks before major noncardiac surgery requiring general anesthesia. Among these patients, 7 MIs, 11 major bleeding episodes, and 8 deaths were noted. All deaths and MIs, as well as 8 of the 11 bleeding episodes, occurred in patients subjected to surgical procedures less than 14 days after stenting. Four patients died after undergoing surgical procedures 1 day after stenting. Wilson and colleagues reported on 207 patients who underwent noncardiac surgery within 2 months of stent placement. Eight patients died or suffered an MI, all of whom were among the 168 patients who had surgical procedures 6 weeks after stent placement. Vicenzi and coworkers studied 103 patients and reported that the risk of suffering a perioperative cardiac event was 2.11-fold greater in patients with recent stents (<35 days before surgery) than in those who underwent PCI more than 90 days before surgical procedures. Leibowitz and associates studied a total of 216 consecutive patients who underwent PCI within 3 months of noncardiac surgery (PTCA, 122; stent, 94). A total of 26 patients (12%) died, 13 in the stent group (14%), and 13 in the PTCA group (11%), a nonsignificant difference. The incidence of acute MI and death within 6 months was not significantly different (7% and 14% in the stent group and 6% and 11% in the PTCA group, respectively). Significantly more events occurred in the two groups when noncardiac surgery was performed within 2 weeks of PCI. Based on the accumulating data, elective noncardiac surgery after PCI, with or without stent placement, should be delayed for 4 to 6 weeks.
Drug-eluting stents may represent an even greater problem during the perioperative period based on case reports. Nasser and coauthors described two patients with in-stent thrombosis occurring 4 and 21 months after the implantation of sirolimus-eluting stents. Drug-eluting stents may represent an additional risk over a prolonged period (≤12 months), particularly if antiplatelet drugs are discontinued. One study demonstrated that although the frequency of major noncardiac surgery in the year after drug-eluting stent placement was more than 4%, the overall risk of adverse outcomes was less than previously reported when surgical procedures were performed months after drug-eluting stent placement. However, the risk was significantly increased in the week after major noncardiac surgery. A population-based study in Canada using administrative healthcare databases demonstrated that the earliest optimal time for elective surgery is 46 to 180 days after bare-metal stent implantation or more than 180 days after drug-eluting stent implantation. Hawn and colleagues used a national, retrospective cohort study of 41,989 VA and non-VA operations occurring in the 24 months after coronary stent implantation between 2000 and 2010. Among patients undergoing noncardiac surgery within 2 years of coronary stent placement, major adverse cardiac events were associated with emergency surgery and advanced cardiac disease but not stent type or timing of surgery beyond 6 months after stent implantation. The 2016 DAPT Guidelines ( Fig. 32.5 ) suggest continuing aspirin therapy in all patients with a coronary stent and discontinuing clopidogrel for as short a time interval as possible for patients with bare-metal stents less than 30 days or drug-eluting stents less than 6 months; with DAPT it can be discontinued. [CR]
Based upon the non-perioperative literature, there is a suggestion that holding clopidogrel for the traditional 8 days may actually increase risk associated with a hypercoagulable rebound suggesting a shorter period of time may be optimal. A recent cohort study suggests that withdrawal of antiplatelet agents greater than 5 days is associated with increased major adverse cardiac events. [CR]
Perioperative Risk Factors for Cardiac Morbidity and Mortality
A thorough history should focus on cardiovascular risk factors and symptoms or signs of unstable cardiac disease states, such as myocardial ischemia with minimal exertion, active CHF, symptomatic valvular heart disease, and significant cardiac arrhythmias. The presence of unstable angina is associated with a 28% incidence of perioperative MI. Such patients would benefit from delaying elective surgery to address their CAD. For those patients with chronic stable angina, exercise tolerance appears to be a good method of assessing perioperative risk.
In virtually all studies, the presence of active CHF has been associated with increased perioperative cardiac morbidity. In addition, multiple studies have demonstrated that reduced ejection fraction is associated with an increased incidence of perioperative cardiac events. Flu and colleagues performed echocardiography in patients undergoing vascular surgery and found that for open surgical procedures, asymptomatic systolic left ventricular dysfunction and asymptomatic diastolic left ventricular dysfunction were both associated with increased 30-day cardiovascular event rates (OR, 2.3; 95% CI, 1.4-3.6; and OR, 1.8; 95% CI 1.1 to 2.9, respectively) and long-term cardiovascular mortality (hazard ratio, 4.6; 95% CI 2.4-8.5; and hazard ratio, 3.0; 95% CI 1.5 to 6.0, respectively). In patients undergoing endovascular surgery ( n = 356), only symptomatic heart failure was associated with an increase in 30-day cardiovascular events and long-term cardiovascular mortality. These results suggest that stabilization of ventricular function and treatment of pulmonary congestion is prudent before elective surgery.
A recent MI has traditionally been an important predictor of perioperative risk. The more recent the MI, particularly within 3 to 6 months, the greater is the perioperative risk. However, like the Goldman Cardiac Risk Index, medicine has changed and outcomes are improved. The 2014 AHA/ACC Foundation (AHA/ACCF) guidelines advocate the use of 60 days as being high risk. After that time, further risk stratification depends on clinical symptoms.
For those patients without overt symptoms or a history of CAD, the probability of CAD varies with the type and number of atherosclerotic risk factors present. Diabetes accelerates the progression of atherosclerosis, which can frequently be silent, leading many clinicians to assume that diabetes is a CAD equivalent and treating patients as such. Diabetes is an independent risk factor for perioperative cardiac morbidity, and the preoperative treatment with insulin has been included in the Revised Cardiac Risk Index (RCRI). In attempting to determine the degree of the increased risk associated with diabetes, the treatment modality, duration of the disease, and other associated end-organ dysfunction should be taken into account.
Significant intraoperative factors that correlate with perioperative risk and that may be avoided or altered are (1) unnecessary use of vasopressors, (2) unintentional hypotension (this point is controversial, however, because some investigators have found that unintentional hypotension does not correlate with perioperative morbidity ), (3) hypothermia, (4) too low or too high a hematocrit, and (5) lengthy operations.
Significant intraoperative factors that correlate with perioperative morbidity and probably cannot be avoided are (1) emergency surgery and (2) thoracic or intraperitoneal surgery or above-the-knee amputations.
Several risk indices were developed in a prospective cohort study by Lee and associates. They studied 4315 patients 50 years old or older who were undergoing elective major noncardiac procedures in a tertiary care teaching hospital. The six independent predictors of complications included in a RCRI were high-risk type of surgery, history of ischemic heart disease, history of CHF, history of cerebrovascular disease, preoperative treatment with insulin, and preoperative serum creatinine greater than 2.0 mg/dL; increasing cardiac complication rates were noted with an increasing number of risk factors. The RCRI has become the standard tool in the literature for assessing perioperative cardiac risk in a given individual and has been used to direct the decision to perform cardiovascular testing and implement perioperative management protocols. It has been validated for both short-term and long-term cardiovascular outcomes. It has also been shown to predict long-term quality of life. Therefore, the RCRI can be used to help define both the short-term and long-term risks of cardiovascular disease in the surgical patient.
The American College of Surgeons NSQIP created a Surgical Risk Calculator from 525 participating hospitals and more than 1 million operations. This risk calculator uses the specific current procedural terminology code of the procedure being performed to enable procedure-specific risk assessment and includes 21 patient-specific variables (e.g., age, sex, body mass index, dyspnea, previous MI). From this input, it calculates the percentage of risk of a major adverse cardiac event, death, and eight other outcomes. Use of this risk calculator may offer the best estimation for surgery-specific risk of a major adverse cardiac event and death.
The American College of Surgeons NSQIP Myocardial Infarction and Cardiac Arrest risk prediction rule is more specific for cardiac complications. Using these definitions of outcome and chart-based data collection methods, the authors derived a risk index that was robust in the derivation and validation stages and appeared to outperform the RCRI (which was tested in the same dataset) in terms of discriminative power, particularly among patients undergoing vascular surgery.
A primary issue with all these indices is that a simple estimate of risk does not help in refining perioperative management for an individual patient. Therefore, the consultant must communicate the extent and stability of the patient’s CAD, rather than make a simple statement of risk classification.
The goal in providing anesthesia to patients with ischemic heart disease is to achieve the best preoperative condition obtainable by treating conditions that correlate with perioperative risk. The next step is to intraoperatively monitor for conditions that correlate with perioperative risk and avoid circumstances that lead to perioperative risk.
Preoperative and Preprocedure Therapy
The only way known to increase oxygen supply to the myocardium of patients with coronary artery stenosis is to maintain adequate diastolic blood pressure, hemoglobin concentration, and oxygen saturation. The main goals of anesthesia practice for these patients have been to decrease the determinants of myocardial oxygen demand, heart rate, ventricular wall tension, and contractile performance, and to improve plaque stabilization. Thus medical management designed to preserve all viable myocardial tissue may include the following:
Multiple studies have demonstrated improved outcome in patients given perioperative β-blockers, especially if heart rate is controlled, acknowledging the previously discussed concerns regarding the quality of the studies from the Erasmus group. 167a,b Subsequent studies demonstrated that β-blockers may not be effective if heart rate is not well controlled, or in lower risk patients. 167c-e The POISE trial was published in which 8351 high-risk β-blocker-naïve patients were randomized to high-dose continuous-release metoprolol versus placebo. [CR]
There was a significant reduction in the primary outcome of cardiovascular events associated with a 30% reduction in MI rate, but with a significantly increased rate of 30-day all-cause mortality and stroke. Several recent cohort studies continue to support the fact that high-risk patients on β-blockers were associated with improved outcome. A Canadian administrative dataset suggested that the perioperative morbidity would be higher if β-blockers were started within 7 days as compared to 8 days or greater. As part of the update to the current ACC/AHA Guidelines, an Evidence Review Committee was formed to independently review the data on perioperative β-blockade. Perioperative β-blockade started within 1 day or less before noncardiac surgery prevents nonfatal MI but increases risks of stroke, death, hypotension, and bradycardia. [CR] Without the controversial DECREASE studies, there are insufficient data on β-blockade started two or more days prior to surgery. Wallace and associates reported that perioperative β-blockade administered according to the Perioperative Cardiac Risk Reduction protocol is associated with a reduction in 30-day and 1-year mortality. [CR] Perioperative withdrawal of β-blockers is associated with increased mortality. The current ACCF/AHA Guidelines on perioperative β-blockade advocate that perioperative β-blockade is a Class I indication and should be used in patients previously on β-blockers. The new recommendations changed the recommendation from a Class IIa to IIb for patients undergoing vascular surgery who are at high cardiac risk owing to CAD or the finding of cardiac ischemia on preoperative testing ( Box 32.3 ).
β-Blockers should be continued in patients undergoing surgery who have been on β-blockers chronically. (Level of Evidence: B)
It is reasonable for the management of β-blockers after surgery to be guided by clinical circumstances, independent of when the agent was started. (Level of Evidence: B)
In patients with intermediate- or high-risk myocardial ischemia noted in preoperative risk stratification tests, it may be reasonable to begin perioperative β-blockers. (Level of Evidence: C)
In patients with three or more RCRI risk factors (e.g., diabetes mellitus, heart failure, coronary artery disease, renal insufficiency, cerebrovascular accident), it may be reasonable to begin β-blockers before surgery. (Level of Evidence: B)
In patients with a compelling long-term indication for β-blocker therapy but no other RCRI risk factors, initiating β-blockers in the perioperative setting as an approach to reduce perioperative risk is of uncertain benefit. (Level of Evidence: B)
In patients in whom β-blocker therapy is initiated, it may be reasonable to begin perioperative β-blockers long enough in advance to assess safety and tolerability, preferably more than 1 day before surgery. (Level of Evidence: B)
Class III: Harm
β-Blocker therapy should not be started on the day of surgery. (Level of Evidence: B)
RCRI, Revised cardiac risk index.
Vasodilation (with nitroglycerin or its “long-acting” analogues nitroprusside, hydralazine, or prazosin) to decrease ventricular wall tension may be beneficial, although currently no randomized trials support the prophylactic use of these agents. There are no data to support the routine use of pulmonary artery catheters and transesophageal echocardiography for this type of patient. The intraoperative management of patients with ischemic heart disease is discussed in further detail in Chapters 31 and 54 and in published guidelines.
Other medications. In POISE II, α-2 agonists were not shown to improve perioperative outcome. [CR] POISE II also evaluated the effectiveness of aspirin therapy in a cohort of patients without a recent stent. Administration of aspirin before surgery and throughout the early postsurgical period had no significant effect on the rate of a composite of death or nonfatal MI but increased the risk of major bleeding. [CR] Most recently, perioperative statins have been shown to improve cardiac outcome. Durazzo and colleagues published a randomized trial of 200 vascular surgery patients in which statins were started an average of 30 days prior to vascular surgery. [CR] A significant reduction in cardiovascular complications was demonstrated using this protocol. Le Manach and colleagues demonstrated that statin withdrawal greater than 4 days was associated with a 2.9 odds ratio of increased risk of cardiac morbidity in vascular surgery. [CR] The guidelines advocate continuing statin therapy in patients currently taking statins as a Class I indication. A multimodal approach to medical management should be taken in high-risk patients. There continues to be controversy regarding the optimal management of ACE inhibitors and ARBs. In the Veterans Administration, withholding ARB postoperatively is strongly associated with increased 30-day mortality, especially in younger patients, although residual confounding may be present. [CR] In the VISION trial, compared to patients who continued their ACE inhibitors/angiotensin II receptor blockers, the ACE/ARB users who withheld their agents in the 24 hours before surgery were less likely to suffer the primary composite outcome of all-cause death, stroke, or myocardial injury (adjusted relative risk, 0.82; 95% CI, 0.70 to 0.96; P = .01); and intraoperative hypotension (adjusted relative risk, 0.80; 95% CI, 0.72-0.93; P < .001). [CR] The current AHA/ACC Guidelines suggest that continuation of ACE inhibitors or angiotensin-receptor ARBs perioperatively is reasonable, but should be restarted as soon as reasonable. The new study questions this recommendation, but further randomized trials are needed.
Perioperative transfusion therapy is discussed in more detail in Chapter 49 . The FOCUS (Functional Outcomes in Cardiovascular Patients Undergoing Surgical Repair of Hip Fracture) trial was unable to demonstrate benefit in high-risk patients with hip fracture between a high and low transfusion trigger.
Valvular Heart Disease
Major alterations in the preoperative management of patients with valvular heart disease have been made regarding the use of anticoagulant therapy and are now based on the causes of the disease. Preoperative and intraoperative management of patients with valvular heart disease is discussed in Chapter 38, Chapter 67 .
The prognosis and the perioperative risk for patients with valvular heart disease depend on the stage of the disease. Although stenotic lesions typically progress faster than regurgitant lesions, regurgitant lesions from infective endocarditis, rupture of the chordae tendineae, or ischemic heart disease can be rapidly fatal. Left ventricular dysfunction is common in the late stage of valvular heart disease, both stenotic and regurgitant.
Preoperative maintenance of drug therapy can be crucial; for example, a patient with severe aortic stenosis can deteriorate rapidly with the onset of atrial fibrillation or flutter because the atrial contribution to left ventricular filling can be critical in maintaining cardiac output. One of the most serious complications of valvular heart surgery and of preoperative valvular heart disease is cardiac arrhythmia. Conduction disorders and long-term therapy with antiarrhythmic and inotropic drugs are discussed elsewhere in this chapter. The reader is referred to Chapter 78 and to other sources for discussion of the management of a child with congenital heart disease who is undergoing noncardiac surgery.
Preoperative Antibiotic Prophylaxis for Endocarditis
Patients who have any form of valvular heart disease, as well as those with intracardiac (ventricular septal or atrial septal defects) or intravascular shunts, should be protected against endocarditis at the time of a known bacteremic event. Endocarditis has occurred in a sufficiently significant number of patients with hypertrophic cardiomyopathy (subvalvular aortic stenosis, asymmetric septal hypertrophy) and mitral valve prolapse to warrant the inclusion of these two conditions in the prophylaxis regimen.
Bacteremia occurs after the following events: dental extraction, 30% to 80%; brushing of teeth, 20% to 24%; use of oral irrigation devices, 20% to 24%; barium enema, 11%; transurethral resection of the prostate (TURP), 10% to 57%; upper GI endoscopy, 8%; nasotracheal intubation, 16% (4 of 25 patients); and orotracheal intubation, 0% (0 of 25 patients). The most recent guidelines from the AHA consisted of an update in 2008 from the AHA/ACC on endocarditis in patients with valvular heart disease, with changes from the 2006 document shown in Table 32.6 .
|2006 VHD Guideline Recommendations||2008 VHD Focused Update Recommendations||Comments|
|Class I||Class IIa|
∗ This footnote is obsolete. Please see 2006 VHD Guideline 3 for footnote text, in Bonow RO, Carabello BA, Kanu C et al. ACC/AHA 2006 guidelines for the management of patients with valvular heart disease: a report of the American College of Cardiology/American Heart Association Task Force on practice guidelines (writing committee to revise the 1998 guidelines for the management of patients with valvular heart disease). Developed in collaboration with the Society of Cardiovascular Anesthesiologists: endorsed by the Society for Cardiovascular Angiography and Interventions and the Society of Thoracic Surgeons. Circulation . 2006;114:e84–e231.
Cardiac Valve Prostheses and Anticoagulant Therapy and Prophylaxis for Deep Vein Thrombosis
In patients with prosthetic valves, the risk of increased bleeding during a procedure in a patient receiving antithrombotic therapy must be weighed against the increased risk of thromboembolism caused by stopping the therapy. Common practice in patients undergoing noncardiac surgery with a mechanical prosthetic valve in place is cessation of anticoagulant therapy 3 days preoperatively. This time frame allows the international normalized ratio to fall to less than 1.5 times normal. The oral anticoagulants can then be resumed on postoperative day 1. Using a similar protocol, Katholi and colleagues found no perioperative episodes of thromboembolism or hemorrhage in 25 patients. An alternative approach in patients at high risk for thromboembolism is conversion to heparin during the perioperative period. The heparin can then be discontinued 4 to 6 hours preoperatively and resumed shortly thereafter. Current prosthetic valves may have a lower incidence of this complication, and the risk associated with heparin may outweigh its benefit in the perioperative setting. According to the AHA/ACC guidelines, heparin can usually be reserved for patients who have had a recent thrombus or embolus within 1 year, those with demonstrated thrombotic problems when previously off therapy, and those with more than three risk factors (atrial fibrillation, previous thromboembolism, hypercoagulable condition, and mechanical prosthesis). A lower threshold for recommending heparin should be considered in patients with mechanical valves in the mitral position, in whom a single risk factor would be sufficient evidence of high risk. Subcutaneous low-molecular-weight heparin offers an alternative outpatient approach. It is appropriate for the surgeon and cardiologist to discuss the optimal perioperative management for such a patient, including a review of the most recent guidelines. A new guideline publication was published in 2014.
Regional anesthetic techniques may be avoided, although this issue is controversial as many practitioners will use regional anesthesia in patients who are receiving prophylaxis for deep vein thrombosis. However, epidural hematoma has been associated with anticoagulant therapy in many reports. Large retrospective reviews of outcome after epidural or spinal anesthesia, or both, during or shortly before initiation of anticoagulant therapy with heparin have not reported neurologic dysfunction related to hematoma formation in any patient. This paucity of damaging epidemiologic evidence, although reassuring, does not reduce the need for frequent evaluation of neurologic function and a search for back pain in the perioperative period after regional anesthesia in any patient receiving any anticoagulation or antiplatelet. The risk of regional anesthesia concurrent with prophylaxis for deep vein thrombosis with heparin is greater with the use of low-molecular-weight heparin. Heparin-induced thrombocytopenia has been treated successfully with intravenous immunoglobulin. The American Society of Regional Anesthesia and Pain Management has issued a consensus statement on the use of regional anesthesia in anticoagulated patients. They suggest that the decision to perform spinal or epidural anesthesia or analgesia, and the timing of catheter removal in a patient receiving antithrombotic therapy should be made on an individual basis, with the small but definite risk of spinal hematoma weighed against the benefits of regional anesthesia for a specific patient.
It was, previously, determined that venous thromboembolism is so common in postoperative patients that almost 1% of postsurgical patients die of fatal pulmonary embolism ( Table 32.7 ). More recently, it has been estimated that venous thromboembolism is responsible for up to 10% of all hospital-related deaths. [CR] Because of this high mortality risk, prophylaxis against deep vein thrombosis has attained widespread acceptance; thus prophylaxis begins with 5000 units of heparin given subcutaneously 2 hours preoperatively. Other trials have shown equal effect with external pneumatic sequential compression devices. The most current recommendations are available from the American College of Chest Physicians for prophylaxis against venous thromboembolism in 2012.
|Type of Surgery||Incidence of|
|Deep Vein Thrombosis (%)||Proximal Deep Vein Thrombosis (%)||Fatal Pulmonary Embolism (%)|
|Age >40 years||10||<1||0.1|
|Age >60 years||10-40||3-15||0.8|
|Total hip replacement||40-80||10-20||1.0-5.0|
|Total knee replacement||60-70||20||1.0-5.0|
|Head, neck, chest wall||11|
|Acute myocardial infarction||30||6|
|Acute spine injury||60-100|
|Other bed bound||26|
Another problem that can arise is managing a pregnant patient with a prosthetic valve during delivery. It is recommended that warfarin be replaced by subcutaneous heparin during the peripartum period. During labor and delivery, elective induction of labor is advocated with discontinuance of all anticoagulant therapy, as indicated for the particular valve prosthesis (discussed earlier).
Auscultation of the prosthetic valve should be performed preoperatively to verify normal functioning. Abnormalities in such sounds warrant preoperative consultation and verification of functioning.
Cardiac Conduction Disturbances: Cardiac Arrhythmias
Bradyarrhythmias, especially if profound or associated with dizziness or syncope, are generally managed with pacemakers. However, chronic bifascicular block (right bundle branch block with a left anterior or posterior hemiblock or left bundle branch block with combined left anterior and posterior hemiblocks), even when only a first-degree heart block is present, can progress to complete heart block and sudden perioperative death on rare occasion. In six studies, less than 2% of the approximately 266 patients with bifascicular block progressed to complete heart block perioperatively. Conversely, these patients have a high 5-year mortality rate (160 of 554 patients, or 29%). Most of the deaths were related to tachyarrhythmias or ischemic events not usually preventable by traditional pacemakers. Thus the presence of a bifascicular block on the ECG should make the anesthesiologist worried about associated CAD or left ventricular dysfunction; an echocardiogram should be evaluated perioperatively. Nevertheless, these patients rarely have complete heart block perioperatively. Therefore, prophylactic preoperative insertion of temporary pacing wires for bifascicular block does not seem warranted; however, central access can be established in advance in the event that a temporary pacemaker needs to be inserted (most operating rooms do not rely on transthoracic pacing, although it may be attempted if available). The actual pacemaker equipment and appropriate personnel should be immediately available, and the equipment should be tested regularly, because symptomatic heart block does occur perioperatively in more than 1% of patients. One study appears to have confirmed this rate of at least 1% for patients undergoing cardiac surgery. One percent of patients in whom a pacing pulmonary artery catheter was not inserted preoperatively subsequently required pacing before cardiopulmonary bypass. By contrast, 19% of patients who had such a catheter in place underwent cardiac pacing before cardiopulmonary bypass. Predictors of the need for pacing included previous symptomatic bradyarrhythmia, a history of transient complete AV block, and aortic valve disease.
Older studies demonstrated that a rate of more than five PVCs per minute on preoperative examination correlates with perioperative cardiac morbidity. To the classic criteria for treating PVCs (the presence of R-on-T couplets, the occurrence of more than three PVCs per minute, and multifocality of PVCs) must be added frequent (>10/h over a 24-hour period) and repetitive ventricular beats. Electrophysiologic and programmed ventricular stimulation studies are being used to indicate and guide treatment of patients with ischemic heart disease or recurrent arrhythmias and survivors of out-of-hospital cardiac arrest. Although such patients are often treated with antiarrhythmic therapy, attention to their underlying condition should be a focus of our preoperative management. Long-term antiarrhythmic therapy is discussed in the last section of this chapter, on drug therapy. Torsades de pointes is an arrhythmia characterized by episodes of alternating electrical polarity such that the major vector of the QRS complex seems to alternate around an isoelectric line. The hallmark enabling differential diagnosis from ventricular tachycardia is the unusual response of this arrhythmia to commonly used antiarrhythmic drugs. In other words, the use of drugs that prolong the QT interval (e.g., quinidine, procainamide, disopyramide, some of the antihistamines, and the antipsychotic phenothiazines) may well make the arrhythmia more frequent or of longer duration. Reports of the sudden occurrence of torsades de pointes during surgical procedures have been rare in the anesthesia literature. Immediate therapy consists of the administration of magnesium or electrical cardioversion, followed by overdrive cardiac pacing or the administration of β-adrenergic agonists and discontinuation of drugs that prolong the QT interval.
Premature atrial contractions and cardiac rhythms other than sinus also correlate with perioperative cardiac morbidity. These arrhythmias may be more a marker of poor cardiovascular reserve than a specific cause of perioperative cardiac complications.
Preexcitation syndrome is the name for supraventricular tachycardias associated with AV bypass tracts. Successful treatment, which is predicated on an understanding of the clinical and electrophysiologic manifestations of the syndrome, consists of either catheter ablation techniques or surgery using preoperative and intraoperative techniques that avoid release of sympathetic and other vasoactive substances and therefore tachyarrhythmias. Anesthesia for electrophysiologic procedures is discussed in Chapter 55 .
Disorders of the Respiratory and Immune Systems
General Preoperative and Preprocedure Considerations
Pulmonary complications after procedures requiring anesthesia are as common as cardiovascular complications—even more common if venous thromboembolism is included. It has, relatively recently, been estimated that postoperative respiratory complications can occur in up to 80% of surgical patients, noting obesity, preexisting pulmonary disease, and advanced age are among the chief risk factors. [CR] Thus pulmonary complications are equally as important or more important to the patient and the health system in terms of morbidity, mortality, length-of-stay extension, and cost. Today there is an even greater appreciation of the effects of smoking and sleep apnea on perioperative and long-term care has increased. (Preoperative and preprocedure identification and perioperative care of patients with sleep apnea are discussed in the earlier section on obesity and in Chapter 58 .)
The main purpose of preoperative testing is to identify patients at risk for perioperative complications so that appropriate perioperative therapy can be instituted to foster return to functional status. Preoperative assessment can also establish baseline function and the feasibility of surgical intervention. Whereas numerous investigators have used pulmonary function tests to define inoperability or high-risk versus low-risk groups for pulmonary complications, few have been able to demonstrate that the performance of any specific preoperative or intraoperative measure, except perhaps smoking cessation and physical activity such as a walking program, reliably decreases perioperative pulmonary morbidity or mortality and improves patient outcomes. Because routine preoperative pulmonary testing and care are discussed extensively in Chapter 41 , the current discussion is limited to an assessment of the effectiveness of this type of care.
In fact, few randomized prospective studies indicate an outcome benefit of preoperative preparation. Stein and Cassara randomly allocated 48 patients to undergo preoperative therapy (cessation of smoking, administration of antibiotics for purulent sputum, and use of bronchodilating drugs, postural drainage, chest physiotherapy, and ultrasonic nebulizer) or no preoperative therapy. The no-treatment group had a mortality of 16% and morbidity of 60%, as opposed to 0% and 20%, respectively, for the treatment group. In addition, the treatment group spent an average of 12 postoperative days in the hospital as compared with 24 days for the 21 survivors in the no-treatment group.
Collins and colleagues prospectively examined the benefits of preoperative antibiotics, perioperative chest physiotherapy and therapy with bronchodilating drugs, and routine postoperative analgesia (morphine) on postoperative respiratory complications in patients with COPD. Of these therapies, only preoperative treatment with antibiotics had a beneficial effect.
Hulzebos and colleagues performed a single-center randomized trial of intensive inspiratory muscle training. Preoperative inspiratory muscle training reduced the incidence of postoperative pulmonary complications and the duration of postoperative hospitalization in patients at high risk of developing a pulmonary complication who were undergoing CABG surgery.
Warner and coworkers collected data retrospectively about smoking history and prospectively (concurrently) about pulmonary complications for 200 patients undergoing CABG. These investigators documented that 8 weeks or more of smoking cessation was associated with a 66% reduction in postoperative pulmonary complications. Smokers who stopped for less than 8 weeks actually had an increase (from 33% for current smokers to 57.1% for recent quitters) in the rate of one or more of the six complications surveyed: purulent sputum with pyrexia; need for respiratory therapy care; bronchospasm requiring therapy; pleural effusion or pneumothorax (or both) necessitating drainage; segmental pulmonary collapse, as confirmed by radiography; or pneumonia necessitating antibiotic therapy. Other investigators have found that both shorter and longer periods of cessation of smoking were needed before achieving cardiovascular and hematologic benefit. Bluman and associates performed a retrospective chart review of 410 patients undergoing noncardiac surgery at a VA hospital. Current smoking was associated with a nearly 6-fold increase in the risk of a postoperative pulmonary complication. Reduction in smoking within 1 month of surgery was not associated with a decreased risk for postoperative pulmonary complications. Nakagawa and coauthors also reported higher pulmonary complication rates in patients undergoing pulmonary surgery who quit within 4 weeks of surgery than in current smokers or those who had stopped smoking for more than 4 weeks. Wong and colleagues performed a systematic review of 25 studies of smoking cessation. At least 4 weeks of abstinence from smoking reduced respiratory complications, and abstinence of at least 3 to 4 weeks reduced wound healing complications. Short-term (<4 weeks) smoking cessation did not appear to affect the risk of postoperative respiratory complications.
Two randomized trials focused on smoking cessation. Wong and colleagues performed a prospective, multicenter, double-blind, placebo-controlled trial, in which 286 patients were randomized to receive varenicline or placebo. A perioperative smoking cessation intervention with varenicline increased abstinence from smoking 3, 6, and 12 months after elective noncardiac surgery with no increase in serious adverse events. Lee and colleagues randomized patients to a group receiving no specific smoking cessation intervention or to an intervention group that received (1) brief counseling by the preadmission nurse, (2) brochures on smoking cessation, (3) referral to the Canadian Cancer Society’s Smokers’ Helpline, and (4) a free 6-week supply of transdermal nicotine replacement therapy. All outcome assessors and caregivers on the operative day were blinded to group assignment. Smoking cessation occurred in 12 patients (14.3%) in the intervention group as compared with 3 patients (3.6%) in the control group (relative risk, 4.0; 95% CI, 1.2-13.7; P = .03). The overall rate of combined intraoperative and immediate postoperative complications was not significantly different between intervention and control groups. At follow-up 30 days postoperatively, smoking cessation was reported in 22 patients (28.6%) in the intervention group compared with 8 patients (11%) in controls (relative risk, 2.6; 95% CI, 1.2-5.5; P = .008).
When Skolnick and coworkers studied 602 children prospectively, exposure to passive smoking (as measured by urinary cotinine, the major metabolite of nicotine) correlated directly with airway complications. Children with the least exposure to passive smoke had the fewest complications. Secondhand smoke may be a model for particulate air pollution, which can have immediate and long-term effects in increasing lung dysfunction and inflammatory stimuli throughout the body.
Celli and associates performed a randomized prospective controlled trial of intermittent positive-pressure breathing (IPPB) versus incentive spirometry and deep-breathing exercises in 81 patients undergoing abdominal surgery. The groups exposed to a respiratory therapist (regardless of the treatment given) had more than a 50% lower incidence of clinical complications (30%-33% vs. 88%) and shorter hospital stays than did the control group. Thus this demonstrates that the outcome improves when any concern about lung function is shown by someone knowledgeable in maneuvers designed to clear lung secretions.
Bartlett and coworkers randomly assigned 150 patients undergoing extensive laparotomy to 1 of 2 groups. One group received preoperative instruction in and postoperative use of incentive spirometry (10 times/h). The other group received similar medical care but no incentive spirometry. Only 7 of 75 patients using incentive spirometry had postoperative pulmonary complications, as opposed to 19 of 75 in the control group. However, Lyager and colleagues randomly assigned 103 patients undergoing biliary or gastric surgery to receive either incentive spirometry with preoperative and postoperative chest physiotherapy or only preoperative and postoperative chest physiotherapy. No difference in the postoperative course or pulmonary complications was found between the groups. Other studies have shown a specific benefit (i.e., greater than that provided by routine care) for chest physiotherapy and IPPB. These studies are usually poorly controlled, not randomized, or retrospective in design (or any combination); these deficiencies probably substantially bias the results toward finding a benefit in reducing postoperative pulmonary complications. Although randomized prospective studies showed no benefit or actual harm from chest physiotherapy and IPPB on the resolution of pneumonia or postoperative pulmonary complications, the studies cited earlier and numerous retrospective studies strongly suggest that preoperative evaluation and treatment of patients with pulmonary disease actually decrease perioperative respiratory complications.
Meta-analyses have suggested a benefit of anesthetic and pain management with respect to respiratory outcomes. Rodgers and associates reviewed 141 trials involving 9559 patients who had been randomized to receive neuraxial blockade or general anesthesia. Overall mortality was significantly less frequent in the neuraxial blockade group (2.1% vs. 3.1%). The relative risk of pneumonia in the neuraxial blockade group was 0.61 (CI, 0.48-0.81), and the relative risk of respiratory depression was 0.41 (CI, 0.23-0.73). Further, Neuman and colleagues examined a retrospective cohort of 18,158 patients undergoing surgery for hip fracture in 126 hospitals in New York in 2007 and 2008. Patients receiving regional anesthesia experienced fewer pulmonary complications (359 [6.8%] vs. 1040 [8.1%]; P < .005). Regional anesthesia was associated with a lower adjusted odds of mortality (OR, 0.710; 95% CI, 0.541, 0.932; P = .014) and pulmonary complications (OR, 0.752; 95% CI, 0.637, 0.887; P < .0001) relative to general anesthesia. In subgroup analyses, regional anesthesia was associated with improved survival and fewer pulmonary complications among patients with intertrochanteric fractures but not among patients with femoral neck fractures.
Not all studies demonstrate beneficial effects of pharmacologic pretreatment. In afebrile outpatient American Society of Anesthesiologists (ASA) class I and II children with no lung disease or findings who underwent noncavitary, nonairway surgery lasting less than 3 hours, neither albuterol nor ipratropium premedication decreased adverse events.
Evaluation of dyspnea is especially useful. Boushy and coworkers found that grades of preoperative dyspnea correlated with postoperative survival. (Grades of respiratory dyspnea are provided in Table 32.8 .) Mittman demonstrated an increased risk of death after thoracic surgery, from 8% in patients without dyspnea to 56% in patients with dyspnea. Similarly, Reichel found that no patients died after pneumonectomy if they were able to complete a preoperative treadmill test for 4 minutes at the rate of 2 mph on level ground. Other studies have found that the history and physical examination of an asthmatic subject can also predict the need for hospitalization. Wong and colleagues found that the risk index correlated with postoperative pulmonary complications ( Table 32.9 ).
|0||No dyspnea while walking on a level surface at a normal pace|
|I||“I am able to walk as far as I like, provided I take my time”|
|II||Specific (street) block limitation (“I have to stop for a while after one or two blocks”)|
|III||Dyspnea on mild exertion (“I have to stop and rest while going from the kitchen to the bathroom”)|
|IV||Dyspnea at rest|
|I. EXPIRATORY SPIROGRAM|
|A. Normal (% FVC + [% FEV 1 /FVC] > 150)||0|
|B. % FVC + (% FEV 1 /FVC) = 100-150||1|
|C. % FVC + (% FEV 1 /FVC) <100||2|
|D. Preoperative FVC < 20 mL/kg||3|
|E. Post bronchodilator FEV 1 /FVC <50%||3|
|II. CARDIOVASCULAR SYSTEM|
|B. Controlled hypertension, myocardial infarction without sequelae for >2 years||0|
|C. Dyspnea on exertion, orthopnea, paroxysmal nocturnal dyspnea, dependent edema, congestive heart failure, angina||1|
|III. NERVOUS SYSTEM|
|B. Confusion, obtundation, agitation, spasticity, discoordination, bulbar malfunction||1|
|C. Significant muscular weakness||1|
|IV. ARTERIAL BLOOD GASES|
|B. Pa co 2 >50 mm Hg or PaO 2 <60 mm Hg on room air||1|
|C. Metabolic pH abnormality >7.50 or <7.30||1|
|V. POSTOPERATIVE AMBULATION|
|A. Expected ambulation (minimum, sitting at bedside) within 36 h||0|
|B. Expected complete bed confinement for ≥36 h||1|
Arozullah and associates developed the first validated multifactorial risk index for postoperative respiratory failure, defined as mechanical ventilation for more than 48 hours after surgical procedures, or reintubation and mechanical ventilation after postoperative extubation. In a prospective cohort study of 181,000 male veterans as part of the National Veterans Administration Surgical Quality Improvement Program, seven factors independently predicted risk ( Table 32.10 ). With increasing numbers of risk factors present, the rate of complications increased from 0.5% (class 1) to 26.6% (class 4). Arozullah and colleagues subsequently developed a risk index for postoperative pneumonia by using data on 160,805 patients undergoing major noncardiac surgery and validated the index by using data on an additional 155,266 patients. Patients were divided into five risk classes by using risk index scores ( Table 32.11 ). Pneumonia rates were 0.2% in patients with 0 to 15 risk points, 1.2% in those with 16 to 25 risk points, 4.0% in those with 26 to 40 risk points, 9.4% in those with 41 to 55 risk points, and 15.3% in those with more than 55 risk points.
|Variable||Odds Ratio (95% Confidence Interval)|
|Type of Surgery|
|Abdominal aortic aneurysm||14.3 (12.0-16.9)|
|Neurosurgery, upper abdominal, or peripheral vascular||4.21 (3.80-4.67)|
|Other surgery ∗||1.00 (reference)|
|Emergency surgery||3.12 (2.83-3.43)|
|Albumin < 30 g/L||2.53 (2.28-2.80)|
|Blood urea nitrogen > 30 mg/dL||2.29 (2.04-2.56)|
|Partially or fully dependent status||1.92 (1.74-2.11)|
|History of COPD||1.81 (1.66-1.98)|
|Preoperative Risk Factor||Point Value|
|Type of Surgery|
|Abdominal aortic aneurysm repair||15|
|Weight loss >10% in past 6 months||7|
|History of COPD||5|
|History of cerebrovascular accident||4|
|Blood Urea Nitrogen Level|
|<2.86 mmol/L (0.8 mg/dL)||4|
|7.85-10.7 mmol/L (22-30 mg/dL)||2|
|≥10.7 mmol/L (≥30 mg/dL)||3|
|Transfusion >4 units||3|
|Steroid use for chronic condition||3|
|Current smoker within 1 year||3|
|Alcohol intake >2 drinks/day in past 2 weeks||2|