Medical Evaluation of the Patient Undergoing Surgery

From WiserWiki

[edit] Medical Evaluation of the Patient Undergoing Surgery

James J. Heffernan

Robert A. Witzburg

Gerald W. Smetana

Consultation practice is an important component of many physicians' professional activity. Up to 30% of the clinical activity of general internists and up to 50% of the patient care activities of selected medical subspecialists may be devoted to consultation. The continued growth of managed care may increase the role of primary care physicians on surgical services. Denial of preoperative hospital days by health insurers has already shifted preoperative evaluation of patients undergoing elective surgery to the outpatient setting, sometimes to specialized preoperative clinics and sometimes to the clinical practice sites of primary care physicians.

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 13-1). The consultant's role in perioperative care focuses on those medical factors that may increase the risks of anesthesia and surgery, with the formulation of a management plan to minimize risk.

Continued advances in the techniques of anesthesia and surgery have made surgical death very uncommon. In the United States, expected perioperative mortality (during surgery or within the first 48 hours postoperatively) is less than 0.03% in healthy adults younger than 70 years undergoing elective surgery. Of all perioperative deaths, 10% to 15% occur during anesthesia induction, 30% to 40% during surgery, and the remaining 45% to 60% during the subsequent 48 hours. This broad temporal distribution of perioperative mortality dictates the need for medical vigilance well into the recovery phase.

Systematic assessment of surgical risk requires the identification of anesthesia-related, procedure-related, practitioner-related, and, especially, patient-related factors. No one method of anesthesia appropriate for major procedures is inherently safer than another. The selection of anesthetic technique for a given patient, procedure, surgeon, and anesthetist is highly individualized and remains the primary responsibility of the anesthesiologist rather than of the consultant.

Certain procedures carry a higher risk, such as craniotomy and cardiovascular surgery, whereas others carry a lower risk, such as herniorrhaphy, cystoscopy, cervical dilation and curettage, and ophthalmologic surgery. The importance of comorbid disease in determining surgical risk may outweigh the nature of the procedure in predicting outcome. Any procedure conducted as an emergency carries an increased risk—as much as twice the expected mortality for similar surgery performed electively.

Several protocols have been developed for estimating an overall, integrated risk profile for patients subjected to general anesthesia and surgery. The traditional and most widely utilized classification system is that of the American Society of Anesthesiologists (ASA), which can assist in risk stratification (Table 13-1). This schema is predicated on a global assessment of the patient's medical problems and functional status. Clinical studies have demonstrated that a patient's functional status, especially exercise capacity, is an excellent and independent predictor of surgical outcome, most likely as a reflection of overall cardiopulmonary reserve. Patients in higher risk classes generally require more intensive preoperative evaluation and more vigilant perioperative management.

Table 13-1 Physical Status Scale of the American Society of Anesthesiologists

[edit] PREOPERATIVE EVALUATION OF THE HEALTHY PATIENT

Healthy patients may undergo surgery with an extremely low risk of morbidity and mortality. Patient- and procedure-related factors contribute more to risk than anesthetic considerations. Perioperative mortality in ASA class I patients is less than 0.03%. Recent surveys suggest an even lower risk. In a study of patients undergoing ambulatory surgery, there were no deaths among 14,609 ASA class I patients.^[1] Since the risk is extremely low in healthy patients, extensive preoperative evaluation does not often bring to light factors that increase risk. Furthermore, most abnormal findings are false-positives. One may identify a higher risk subset of healthy patients through the use of a screening questionnaire, a focused clinical evaluation, and selective laboratory testing.

The primary goal of preoperative evaluation is to identify previously unrecognized illness that may have an impact on surgical risk. Preoperative screening questionnaires correctly identify most patients who are at higher than average risk. With one such questionnaire, negative answers to all questions predicted fitness for surgery in 96% of cases, as determined by the clinical evaluation of 10 staff anesthesiologists (Box 13-2).

Poor exercise capacity identifies patients at higher risk for overall morbidity, as well as cardiac and pulmonary complications. Patients who are unable to walk at least two blocks on level ground are at higher risk.

[edit] Screening Blood Tests

Most routine preoperative laboratory tests are performed out of habit and are not justified by evidence in the medical literature. These tests are costly in the aggregate and may lead to patient worry and unnecessary delay of surgery. Physicians can usually predict abnormal results based on a patient's known medical problems; routine preoperative laboratory testing rarely influences management. In a study of 2000 patients, only 0.22% of routine preoperative blood tests were of potential surgical significance.^[2] None of these abnormalities were recognized and acted on prior to surgery and there were no adverse consequences. Other authors have reported similar results.

Suggested routine blood tests include a pregnancy test for women who may be pregnant, hematocrit if expected major blood loss, and a serum creatinine measurement if hypotension is likely or the patient is over age 50 years. One should not obtain serum glucose, urinalysis, liver function tests, electrolytes, or tests of hemostasis routinely in the preoperative evaluation of the healthy patient.

Other preoperative laboratory screenings should be driven by the history of underlying illness, with directed laboratory evaluation in the setting of known or suspected renal impairment; liver disease; metabolic disease, especially diabetes mellitus; hemorrhagic or thromboembolic diatheses; or treatment with medications known to cause metabolic derangements, especially diuretics.

Results of tests performed in the preceding 4 months may be safely substituted for preoperative screening tests if the previous results were normal and there is no obvious clinical indication for retesting.

[edit] Electrocardiogram

Patients with certain abnormal findings on electrocardiogram (ECG) have an increased risk of postoperative cardiac complications. These abnormalities include Q waves of indeterminate age, more than 5 premature ventricular contractions per minute, atrial ectopy, and a rhythm other than sinus. These findings are more often present in patients who are older, have known cardiac disease, or have diabetes. Clinically important ECG findings in patients under age 45 are rare. We support the recommendations first proposed by Goldberger and O'Konski (Box 13-3).

[edit] Chest Radiograph

The frequency of abnormal chest radiographs increases with age. The reported rates of abnormal preoperative chest radiographs vary from 1.2% to 52.9% among patients in different clinical series. Most abnormal studies, however, do not influence perioperative management. In a meta-analysis including a total of 14,390 routine chest radiographs, there were only 140 unexpected abnormal studies and only 14 studies that influenced management.^[3] Recent data suggest that patients undergoing abdominal surgery who have abnormal chest radiographs have a higher likelihood of postoperative pulmonary complications. Reasonable criteria for obtaining a preoperative chest radiograph include age greater than 60 years or suspicion of cardiac or pulmonary disease on the basis of history and examination.

[edit] CARDIOVASCULAR RISK ASSESSMENT AND MODIFICATION

[edit] Risk Assessment

Cardiac complications are the leading cause of morbidity and mortality in the perioperative period. All requests for preoperative medical evaluation should include an assessment of cardiac risk. In most cases, a careful history and physical examination are sufficient to assess risk.

In 1977, Goldman and colleagues first outlined the principal cardiac risk factors in their groundbreaking report of a multifactorial cardiac risk index.^[4] This study demonstrated that a myocardial infarction within the previous 6 months and congestive heart failure at the time of surgery were the two most important predictors of cardiac risk. Among possible findings of congestive heart failure, only jugular venous distention and an S3 were correlated with an increased risk. Other independent risk factors in the multivariate analysis included age greater than 70 years, rhythm other than sinus, frequent premature ventricular beats, significant aortic stenosis, and poor general medical status as defined by several metabolic parameters. In addition, there was a higher risk with thoracic, abdominal, aortic, and all emergency procedures. One can assign a total number of points based on the strength of each factor in the multivariate analysis. Patients are grouped into one of four risk categories. This index has been repeatedly validated prospectively and correctly estimates risk among patients undergoing nonvascular surgery. Factors that do not increase risk in the Goldman analysis include stable hypertension, stable angina, diabetes, elevated cholesterol, and cigarette smoking.

Limitations of the Goldman index include an underestimation of risk in vascular surgical procedures, absence of unstable angina or angina at low workload as risk factors, and an underappreciation of the risk associated with significant aortic stenosis. Detsky and colleagues developed a modified index that addresses the last two of these concerns (Table 13-2).

Table 13-2 Modified Cardiac Risk Index✢

✢Class I = 0 to 15 points; class II = 20 to 30 points; class III = more than 30 points.

†Canadian Cardiovascular Society classification of angina: 0 = asymptomatic; I = angina with strenuous exercise; II = angina with moderate exertion; III = angina with walking one to two level blocks or climbing one flight of stairs or less at a normal pace; IV = inability to perform any physical activity without development of angina.

In general, patients who are able to perform at least Canadian Cardiovascular Society class II activities without limitation or cardiac symptoms have an average risk of postoperative cardiac complications. Examples of class II activities include the ability to walk two blocks on level ground or to carry two bags of groceries up one flight of stairs. Exercise testing is no more predictive of cardiac complications than a reliable history of exercise tolerance and should not be a routine preoperative test.

Recently the American College of Cardiology (ACC) and the American Heart Association (AHA) proposed a new cardiac risk assessment guideline grounded in an evidence-based review of the available medical literature.^[5] This guideline includes an assessment of functional capacity, clinical predictors of risk, and procedure-related risks. These factors are incorporated into a complex algorithm designed to identify those patients who may proceed directly to surgery and those who will benefit from preoperative cardiac testing. This algorithm has been validated in a prospective study of high-risk patients.

In 1997, the American College of Physicians (ACP) published a cardiac risk assessment guideline, also developed after evidence-based review of the literature^[6][7](Figs. 13-1 and 13-2). Practicing physicians will likely find this guideline easier to use than the ACC/AHA proposal. According to the ACP guideline, all patients are stratified by the Detsky modified risk index. For risk class II or III patients, strategies for risk reduction are individualized based on the nature of the risk. For risk class I patients undergoing vascular surgery, one applies additional clinical risk variables to identify patients who would benefit from further study (Table 13-3). Low-risk patients may proceed directly to surgery. We recommend the use of this guideline for preoperative cardiac risk stratification.

Figure 13-1 Suggested algorithm for the risk assessment and management of patients at low or intermediate risk for perioperative cardiac events, usually myocardial infarction and death. Boxed phrases indicate recommended actions. The italicized words beside the boxes indicate the level of evidence supporting the recommendation. If no italicized word is present, no evidence exists for or against use. DTI, Dipyridamole thallium imaging; DSE, dobutamine stress echocardiography. (Modified from American College of Physicians: Guidelines for assessing and managing the perioperative risk from coronary artery disease associated with major noncardiac surgery, Ann Intern Med 127:309-312, 1997.)

Figure 13-2 Suggested algorithm for the management of patients at high risk for perioperative cardiac events. Boxed phrases indicate recommended actions. The italicized words below the boxes indicate the level of evidence supporting the recommendation. If no italicized word is present, no evidence exists for or against use. AHA, American Heart Association; CHF, congestive heart failure. (Modified from American College of Physicians: Guidelines for assessing and managing the perioperative risk from coronary artery disease associated with major noncardiac surgery, Ann Intern Med 127:309-312, 1997.)

Table 13-3 Low-Risk Variables

✢Eagle KA, et al: Combining clinical and thallium data optimizes preoperative assessment of cardiac risk before major vascular surgery, Ann Intern Med 110:859-866, 1989.

†Vanzetto G, et al: Additive value of thallium single-photon emission computed tomography myocardial imaging for prediction of perioperative events in clinically selected high risk patients having abdominal aortic surgery, Am J Cardiol 77:143-148, 1996.

In 1999, Lee and colleagues proposed a revised cardiac index based on a prospective study of 4315 patients undergoing elective major noncardiac procedures.^[8] In their multivariate analysis of derivation and validation cohorts, different risk factors emerged than in previous analyses. Six factors equally predicted risk and were assigned one point each in the revised index. These factors included high-risk surgery (intraperitoneal, intrathoracic, or suprainguinal vascular), ischemic heart disease, history of congestive heart failure, history of cerebrovascular disease, insulin therapy for diabetes, or preoperative serum creatinine > 2 mg/dl. Postoperative cardiac complication rates were 0.4%, 0.9%, 7%, and 11% among patients with no, one, two, or ≥ three factors, respectively. The revised cardiac risk index outperformed the original Goldman index, the Detsky modified index, and the ASA classification in the prediction of cardiac risk in the validation cohort. Whether physicians will use this index in lieu of other established indices in the initial estimation of cardiac risk remains to be determined.

[edit] Strategies to Reduce Perioperative Cardiac Risk

Risk reduction strategies follow logically from the well-known risk factors. Physicians should delay elective surgery until at least 6 months after a myocardial infarction. Six to 12 weeks following myocardial infarction, semielective surgery may be considered after a cardiac functional assessment such as an exercise test. One should diurese patients in congestive heart failure until there is no pulmonary congestion but not so aggressively as to lead to orthostasis. There is no evidence to support routine antidysrhythmic therapy for patients whose risk is based on the presence of atrial or ventricular ectopy. One should evaluate all patients with aortic stenosis that is severe on physical examination or symptomatic. Each metabolic abnormality in the general medical status risk category should be corrected if possible. Finally, one may consider a less ambitious procedure for patients at high risk with few other opportunities for risk reduction.

The most important recent observation in this field is that the perioperative use of atenolol reduces risk among patients at high risk.^[9] Eligible patients were those with known coronary artery disease or with at least two of the following risk factors: age greater than 65 years, hypertension, current smoking, cholesterol greater than 240 mg/dl, and diabetes. The authors excluded patients with congestive heart failure, bronchospasm, or third-degree heart block. Thirty minutes before surgery and again immediately after surgery, the study group received 5-mg intravenous atenolol, followed in 5 minutes by a second infusion if tolerated clinically by the systolic blood pressure (>100 mm Hg) and the pulse (>55). Similarly as tolerated, the patient received 50-to 100-mg oral atenolol daily until discharge, up to a maximum of 7 days. One-year all-cause mortality was 3% in the atenolol group and 14% in the placebo group. The ACP guideline recommends perioperative atenolol for patients who meet the criteria of this study.

Recently, another study confirmed the benefit of perioperative β-blockers in high-risk patients undergoing major vascular surgery.^[10] Eligible patients had at least one cardiac risk factor and a positive dobutamine echocardiogram. Study patients received oral bisoprolol beginning at least 1 week before surgery and continuing for 30 days after surgery. If unable to take oral medications in the immediate perioperative period, intravenous metoprolol was used instead to keep the heart rate below 80 beats per minute. The combined incidence of cardiac death and nonfatal myocardial infarction was 34% in the standard-care group and 3.4% in the bisoprolol group. This dramatic risk reduction was similar in magnitude to that previously reported in the atenolol study.

No prospective trials have evaluated the potential benefit of coronary revascularization prior to noncardiac surgery. In a retrospective review of the Coronary Artery Surgery Study (CASS) experience, the risk of coronary bypass surgery among patients with stable angina approximately equaled the reduction of risk in subsequent noncardiac surgery. There are no studies of percutaneous transluminal coronary angioplasty (PTCA) prior to noncardiac surgery. Both the ACP and AHA/ACC guidelines recommend preoperative coronary revascularization only for those patients who have an indication independent of the need for noncardiac surgery. The data suggest that perioperative β-blockade is more likely to reduce cardiac morbidity than preoperative coronary revascularization.

[edit] PULMONARY RISK ASSESSMENT AND MODIFICATION

Postoperative pulmonary complications are as frequent as cardiac complications and are an important component of overall perioperative risk. Important complications include prolonged mechanical ventilation, pneumonia, atelectasis, bronchospasm, and exacerbation of chronic lung disease. In recent years, studies have identified the principal risk factors for pulmonary complications. These may be divided into patient-related and procedure-related risk factors. (Pulmonary embolism, which has different risk factors, is discussed later.)

Patient-related risk factors include smoking, poor general health status, and chronic obstructive pulmonary disease (COPD). Current cigarette smokers have a twofold to sixfold higher risk of pulmonary complications than nonsmokers. Patients who stop smoking for less than 8 weeks prior to surgery have an unexpectedly higher risk of pulmonary complications than current smokers. Physicians should therefore encourage patients to stop smoking for at least 8 weeks prior to elective surgery.

Patients with poorer general health and functional capacity are at higher risk. A high Goldman cardiac-risk index, ASA class greater than 2, and poor exercise capacity are all predictors of higher risk. Advanced age and obesity are not significant risk factors when one controls for comorbid conditions.

Well-controlled asthma does not increase the risk of postoperative pulmonary complications significantly. Patients with asthma should be treated aggressively to reduce airflow obstruction. Patients should achieve a peak flow of at least 80% of predicted or of their personal best. Corticosteroids, if necessary, may be safely used to achieve this goal without an increase in the risk of surgical site infections.

Patients with COPD have a threefold to fivefold increase in postoperative pulmonary complications. This risk may be reduced by the same treatment one would use for an exacerbation of COPD in the nonoperative setting. Studies support a beneficial effect of multimodality treatment including bronchodilators, chest physical therapy, smoking cessation, and corticosteroids. Patients with symptomatic COPD should receive inhaled ipratropium and supplemental inhaled β-agonists. Although not all patients with COPD respond to systemic corticosteroids, one should consider this treatment if the patient is not at his or her best personal baseline and symptoms are not reduced optimally with other therapies. Physicians should use antibiotics only if respiratory infection is suspected; they do not reduce risk in unselected patients with COPD prior to surgery.

The most important predictor of risk of pulmonary complications is the surgical site. Upper abdominal and thoracic procedures carry the highest risk as a result of splinting and diaphragmatic dysfunction. The average risk of pulmonary complications in upper abdominal, lower abdominal, and thoracic surgeries is 25%, 10%, and 20%, respectively. Pulmonary complications occur in fewer than 1% of patients undergoing laparoscopic cholecystectomy. There is a low risk for all surgeries outside of the chest and abdomen. Other surgical risk factors include surgery lasting longer than 3 hours, the use of general anesthesia, and the use of the long-acting neuromuscular blocker pancuronium.

Physical examination findings that identify patients at higher risk include decreased breath sounds, dullness to percussion, wheezes, rhonchi, and a prolonged expiratory phase.

Despite decades of study, the role of pulmonary function testing prior to surgery remains controversial. Although these studies are clearly valuable prior to lung resection, their predictive value prior to other surgeries is debated. Authors who support the routine use of spirometry suggest a higher risk for patients with a forced expiratory volume in one second (FEV₁) or forced vital capacity (FVC) of less than 70% of the predicted value, or an FEV₁ to FVC ratio of less than 65%.

Several recent studies have shown no increased risk of postoperative pulmonary complications among patients with abnormal spirometry. A recent case-control study found clinical variables to be more predictive of pulmonary complications than spirometry.^[11] These variables were abnormal chest examination, abnormal chest radiograph, a high Goldman score, and a high score on the Charlson comorbidity index. There was no difference in the frequency of abnormal spirometry between the complication and control groups.

Other authors have reported acceptable risks of perioperative morbidity in patients with COPD and abnormal spirometry. Because clinical evaluation may be more predictive of complications than spirometry, and risks are acceptable even in high-risk patients as defined by spirometry, spirometric results should not be used to deny surgery.

Physicians should employ risk reduction strategies for patients undergoing upper abdominal or thoracic surgery who have other risk factors for complications. Postoperative strategies include lung expansion maneuvers, such as chest physical therapy or deep-breathing exercises, and pain control by either postoperative epidural analgesia or intercostal nerve blocks. Lung expansion maneuvers reduce the risk by one half. There have been mixed reports of the benefit of pain control. Strategies to reduce the risk of postoperative pulmonary complications are summarized in Box 13-4.

[edit] PREVENTION OF BACTERIAL ENDOCARDITIS

Patients with valvular heart disease, hypertrophic cardiomyopathy, most congenital cardiac lesions, and systemic-pulmonary shunts should receive endocarditis prophylaxis before certain surgical or dental procedures. The risk of endocarditis in a given patient depends on the type and severity of the underlying cardiac condition, the specific nature of the surgical or dental procedure to be performed, and its association, if any, with bacteremia (Box 13-5). α-Hemolytic streptococci (Streptococcus viridans) and related organisms are the leading cause of endocarditis following dental and respiratory tract procedures, whereas enterococci pose the greatest risk of endocarditis in patients undergoing genitourinary, gynecologic, or gastrointestinal procedures. The standard of care has shifted toward a single preoperative dose of one or two clinically appropriate antibiotics (Table 13-4).

Table 13-4 Endocarditis Prophylaxis Regimens

[edit] HEMODYNAMIC MONITORING

Substantial reductions in perioperative cardiovascular complications and mortality have been noted among patients undergoing noncardiac surgery over the past 40 years. Perioperative mortality among patients over 80 years of age has dropped from 20% in 1960 to approximately 6%,^[12] and perioperative cardiac reinfarction within 3 months of initial myocardial infarction has dropped from 30% to 5.8%.^[13] These reductions are probably due in part to pulmonary artery catheter monitoring, although there are no prospective, randomized trials assessing the benefit of pulmonary artery catheter monitoring among general surgery patients. Several trials have demonstrated significant benefit among peripheral vascular surgery patients. Reasonable criteria for selecting patients in whom perioperative central hemodynamic monitoring appears warranted include recent (within 6 months) myocardial infarction; clinically apparent congestive heart failure, or suspected congestive heart failure in a patient requiring emergency surgery; severe aortic or mitral stenosis; planned major vascular surgical procedure in the setting of coronary heart disease; pulmonary edema of unknown etiology; and patients scheduled for coronary revascularization if they have any of the following: (1) recent myocardial infarction, (2) poor left ventricular function, (3) severe/unstable angina, or (4) high-grade left main coronary artery stenosis.

[edit] PERIOPERATIVE DRUG THERAPY

Questions regarding the manipulation of ongoing drug treatment constitute some of the most common and difficult issues in perioperative consultation. Often, the literature does not support clear, forceful articulation of guidelines, and the internist must tailor management to individual circumstances. In general, three factors must be considered in managing each situation: (1) Have the indications and dosages for the specific drug been clearly defined? (2) What are the likely anesthetic and surgical complications/interactions? (3) Is a clinically important withdrawal syndrome likely, and how can it be safely managed?

Most drugs used in the management of chronic medical illness should be continued through the perioperative period. For minor procedures the patient can receive an oral dose with sips of water on the morning of surgery and another dose later in the day as indicated. For more difficult surgery with a prolonged recovery period, long-acting or parenteral substitutes for chronic therapy may be needed.

[edit] PROPHYLACTIC ANTIMICROBIAL THERAPY

Elective surgery should be delayed until preexisting bacterial infections are treated. Bacterial colonization or infection of the urinary tract is relatively common, especially among individuals with a history of urinary tract infection, in those with indwelling catheters, in those with atonic bladders or obstructive uropathy, and among the elderly. A preoperative screening urinalysis and, when the urinalysis is abnormal, a culture with antibiotic sensitivities, are appropriate in such patients. To reduce the likelihood of perioperative urosepsis, treatment should be initiated with an appropriate antibiotic before surgery in patients with evidence of bacterial colonization or infection of the urinary tract, especially if urinary tract instrumentation is necessary.

Acute exacerbations of chronic bronchitis should be treated to the point of resolution or substantial improvement before surgery. Subsequent perioperative flares must be assumed to be caused by bacteria resistant to whatever antibiotic might have been employed preoperatively.

Antimicrobial agents may be given perioperatively to prevent infection of normally sterile tissues by direct contamination during a surgical procedure. Such therapy is appropriate when the likelihood of infection is great or the sequelae of infection are severe, when relatively well-tolerated drugs are effective against the likely organisms, and when available evidence demonstrates a reduction in perioperative morbidity as a result of prophylaxis.

Most clinical trials have focused on clean-contaminated procedures such as vaginal hysterectomy and colorectal surgery or unusual and extensive clean surgery such as cardiothoracic procedures and hip arthroplasty. However, limited recent data suggest that antibiotic prophylaxis may also be cost-effective for some commonly performed, relatively simple clean procedures such as inguinal herniorrhaphy and certain types of breast surgery. Recent trials have also highlighted the efficacy of very short courses of therapy; a single dose of antibiotic immediately before surgery, followed by no more than one dose postoperatively, is as effective as longer regimens and less likely to be associated with toxicity or the development of resistant organisms.

Table 13-5 lists the common surgical procedures for which antibiotic prophylaxis is recommended. The specific regimen must be individualized to the patient and the local microbial environment.

Table 13-5 Perioperative Wound Infection Prophylaxis

✢Poorly substantiated by literature, but the standard practice.

†Based on recent data; no uniform practice yet established.

‡High risk; advanced age, infection or inflammation, biliary tract obstruction, jaundice.

§High risk; emergency, premature rupture of membranes, obesity.

∥High risk; open wound, reexploration, microsurgery.

¶May be beneficial in centers with high infection rates.

[edit] THROMBOEMBOLISM PROPHYLAXIS

The surgical environment often presents circumstances that fulfill Virchow's triad: stasis, intimal injury, and a hypercoagulable state. However, some patients and procedures present a particularly high risk. Specific factors predisposing patients to thromboembolic complications include advanced age, prolonged anesthesia or surgery, extended perioperative immobilization, paralysis, malignancy, history of venous disease or thromboembolism, and premenopausal exogenous estrogen use (Table 13-6).

Table 13-6 Classification of Level of Risk of Venous Thromboembolism (VTE)

✢Example: uncomplicated minor surgery in patients < 40 yr with no clinical risk factors.

†Examples: any surgery (major and minor) in patients 40 to 60 yr but no additional risk factors; major surgery in patients < 40 yr but no additional risk factors; minor surgery in patients with risk factors.

‡Examples: major surgery in patients > 60 yr without additional risk factors; major surgery in patients 40 to 60 yr who have additional risk factors; patients with MI and medical patients with risk factors.

§Examples: major surgery in patients > 40 yr plus prior VTE or malignant disease or hypercoagulable state; patients with elective major lower extremity orthopedic surgery, or hip fracture, or stroke, or multiple trauma, or spinal cord injury.

There is general agreement that patients at moderate or high risk for perioperative thromboembolism should receive prophylaxis. Available data from a number of well-conducted clinical studies have demonstrated substantial reductions in thromboembolic complications with a variety of regimens in different surgical settings. The American College of Chest Physicians (ACCP) publishes evidence-based recommendations for treating and preventing thromboembolic disease every 3 years.^[14] Guidelines, by category of surgery and perceived underlying risk, for thromboembolism prophylaxis in the surgical patient are summarized in Box 13-6. Although there are convincing data supporting an independent risk of thromboembolic disease from postmenopausal estrogens, the ACCP does not include this as a specific predisposing risk factor.

[edit] MANAGEMENT OF SELECTED CLINICAL PROBLEMS IN PATIENTS UNDERGOING SURGERY

[edit] Hypertension

Patients with hypertension experience greater blood pressure lability during anesthesia, surgery, and the perioperative period than do normotensive individuals. This lability manifests primarily in hypertensive episodes, although interventions to support blood pressure (fluid challenge or adrenergic agents) are more common among hypertensive patients than those with normal blood pressure taking no medications. Among hypertensive patients this lability is largely independent of preoperative blood pressure control and appears to be mediated primarily by transient sympathetic overactivity following noxious stimuli, such as laryngoscopy before endotracheal intubation.^[15] Sympathetic overactivity may also occur on reversal of anesthesia as a result of pain, hypoxia, or hypercarbia. Acute elevations in blood pressure may occur intraoperatively from reflex vasoconstriction as a result of vascular or visceral traction. Blood pressure elevations 2 to 3 days postoperatively may result from recruitment of third-spaced fluid. Perioperative hemorrhagic complications, generally within the operative field, are believed to be more common among patients with poor preoperative blood pressure control (diastolic blood pressure 110 mm Hg or higher). Available data suggest no increase in risk for stable preoperative diastolic blood pressure values <110 mm Hg.^[16] Of note, perioperative use of β-blockers among patients at increased risk for cardiovascular complications has been associated with decreased mortality.^[9][10] On the other hand, an unexpected finding in a randomized trial of cardiac valve replacement patients receiving a dihydropyridine Ca⁺⁺ channel blocker vs. placebo was an excess in major postoperative bleeding among those treated with nimodipine.^[17] See Box 13-7 for management points on hypertension.

[edit] Diabetes Mellitus

Diabetes mellitus is the most common endocrine problem encountered among surgical patients. Patients with diabetes require surgery more often than do nondiabetic individuals, and surgery among diabetic patients is more often associated with complications, especially adverse cardiovascular events and perioperative infections. As many as half the diabetic individuals in the United States are undiagnosed, and it is not uncommon for a patient's diabetes to first come to medical attention when he or she presents for evaluation of a surgical problem. Although surgical and perioperative mortality and morbidity are increased in patients with diabetes, most such patients fare extremely well when appropriately supported through surgery. Several clinical series have demonstrated a modest advantage of intravenously infused insulin over the use of intermediate-acting insulin injected subcutaneously in patients undergoing surgery. In practice, insulin infusion is generally reserved for surgical patients perceived to be at unusually high risk or those who are in ketoacidosis and require emergent surgery. Diabetic patients undergoing coronary artery bypass surgery, however, generally require an insulin infusion to achieve reasonable glucose control intraoperatively and in the immediate postoperative period.

Objectives in the management of the diabetic patient undergoing surgery include (1) characterization of the status of the patient's diabetes, (2) preoperative optimization of glucose control, (3) identification of preexisting complications, treating those amenable to immediate intervention and anticipating problems with others, (4) provision of adequate insulin and substrate for cellular function during surgery, striving for smooth, moderate blood glucose control (150 to 250 mg/dl) and avoiding hypoglycemic and hyperglycemic peaks and ketoacidosis, (5) resumption of appropriate diabetic maintenance therapy and nutritional support postoperatively, and (6) aggressive postoperative monitoring of potential problems arising from the chronic complications of diabetes. See Box 13-8 for management points on diabetes mellitus.

[edit] Thyroid Disease

Hypothyroidism is quite prevalent in the general population, especially among older individuals. Patients who are euthyroid when taking replacement l-thyroxine are at no increased surgical risk. Patients with mild to moderate hypothyroidism who have not been treated should generally be made euthyroid before elective surgery. These patients, however, and even those who are severely myxedematous, can be safely managed throughout urgently required surgery. Special attention must be paid to (1) the mode and rate of treatment of the underlying hypothyroidism, (2) monitoring for hypoglycemia and hyponatremia, (3) exaggerated responses to anesthetic and other agents in the setting of an often profound hypometabolic rate, (4) avoidance of cold stress and hypothermia in the operative and postoperative setting, and (5) anticipated hypoventilation and its implications for assisted ventilation.

Untreated hyperthyroidism poses a more serious risk to the operative patient than does either treated or untreated hypothyroidism. The predominant risks of general anesthesia and surgery in thyrotoxic patients are (1) the predisposition toward dysrhythmias and high-output cardiac failure engendered by striking increases in metabolic rate, (2) the unpredictable level and duration of response to various medications, including anesthetic agents, and (3) the induction of thyroid storm, with its substantial associated morbidity and 5% to 10% mortality. Although the symptoms and signs of hyperthyroidism can be ameliorated in thyrotoxic patients, thyroid storm cannot be prevented until the patient is rendered euthyroid for several months. Indeed, in older clinical reviews surgery was the most common precipitant of this endocrine emergency. Accordingly, elective surgery should be postponed until a known hyperthyroid patient has been made and maintained euthyroid for 3 months.

Thyrotoxic patients who need urgent or emergent surgery require extremely close monitoring and aggressive management. A combination of thyroid-blocking agents, sympatholytics, and stress-dose corticosteroids is generally employed. See Box 13-9 for management points on thyroid disease.

[edit] The Patient Taking Corticosteroids

Primary adrenal insufficiency (Addison's disease) is quite uncommon, as is secondary adrenal insufficiency on the basis of an intrinsic deficit of corticotropin. Adrenal suppression by exogenous corticosteroid administration is, on the other hand, quite prevalent. Patients who receive potentially suppressive doses of exogenous corticosteroids are most often those with severe dermatoses, inflammatory rheumatologic conditions,hematologic or lymphoproliferative disorders, or asthma or other forms of COPD. Individuals who have received more than 40-mg prednisone per day, or its equivalent, for more than 1 week in the preceding year are at risk of adrenal suppression. Smaller doses of oral corticosteroids administered for longer periods, and even the inhaled corticosteroids employed now as mainline therapy in the management of reactive airways disease, may also result in adrenal suppression. Of note in this regard is the recent report of a small randomized trial of preoperative patients maintained on oral prednisone with evidence of adrenal suppression on cosyntropin stimulation test. Patients all received their usual prednisone dose and were randomized to either supplemental stress doses of hydrocortisone in saline, or saline alone. No differences were noted in perioperative hypotension or tachycardia.^[18] Due to small sample size, the power of this study is limited. Alternate-day administration of relatively low doses of oral corticosteroids (less than 20-mg prednisone equivalent) reduces the risk of suppression. Long-term treatment with very low doses of corticosteroids (less than 5 mg per day prednisone equivalent) is also associated with a low likelihood of adrenal suppression.

Until more data are available, the goal of perioperative management for patients with known or suspected adrenal suppression remains the provision of sufficient exogenous corticosteroid to match the maximal physiologic output caused by the stress of surgery and anesthesia. This has been established at 300-to 400-mg cortisol equivalent in the first 24 hours after general anesthesia and major surgery. Hydrocortisone sodium succinate or hydrocortisone sodium phosphate should be employed preferentially to achieve combined glucocorticoid and mineralocorticoid effects. Protocols using the intramuscular route of administration have demonstrated variable effectiveness. Repeated intravenous boluses or combined bolus and infusion protocols are the most reliable means of supporting patients through surgery. Up to 40% of bolus-administered hydrocortisone may be lost in the urine. See Box 13-10 for management points on the patient taking corticosteroids

[edit] The Elderly Patient

Changing demographics combined with advances in anesthetic and surgical technique have led to a steady increase in minor and major surgery for elderly patients. Although the elderly currently constitute less than 15% of the population, they account for 20% to 40% of all surgery, 50% of emergency procedures, and 75% of overall surgical mortality. Age itself appears to account for little, if any, of the excess mortality seen in elderly surgical patients. More important variables include the nature and severity of underlying disease(s), and the type of surgery, especially if performed as an emergency procedure. Age is, however, an independent risk factor for cardiovascular morbidity, including congestive heart failure, myocardial infarction, and stroke. Therefore the commonly utilized schema for quantifying perioperative risk, described earlier, includes an adjustment for age alone. There are no data to support the routine use of perioperative invasive hemodynamic monitoring in the elderly, although such monitoring can be carried out safely when indicated on the basis of underlying disease. As in younger patients, an inadequate cardiovascular response to exercise is an important predictor of risk and constitutes an indication for additional evaluation, aggressive perioperative management, and/or reconsideration of the surgical decision. A recent systematic review of medical issues related to the management of hip fracture has broad applicability.^[19]

Delirium is a special risk among seniors undergoing general anesthesia and major surgery, and may occur in as many as 30% to 50% of older surgical patients. Individuals with preexisting cognitive impairment are at special risk, as are those over age 70, or those with a history of alcohol or other substance abuse and those undergoing certain categories of operation.^[20] Many medications employed in the perioperative setting are associated with delirium, especially anticholinergics, sedatives, and narcotics, especially meperidine. Delirium may also be the mode of presentation of major medical insults, such as myocardial infarction or sepsis. See Box 13-11 for management points on the elderly patient.

[edit] Liver Disease

Patients with serious underlying liver disease tolerate general anesthesia and surgery poorly. All anesthetic regimens, including epidural and spinal anesthesia, reduce hepatic blood flow. Older halogenated hydrocarbon anesthetic gases (halothane and methoxyflurane) carry an independent risk of hepatic toxicity. Isoflurane, in wide use at present, is not a direct hepatotoxin but does cause systemic vasodilatation and a reduction in cardiac output, with resultant risk of impaired hepatic blood flow.

The quantifiable risk of surgery in patients with underlying liver disease correlates with the extent of hepatic functional impairment. This is generally manifested by readily identifiable factors: encephalopathy, cutaneous stigmata of chronic hepatic dysfunction, the presence and severity of ascites, elevation in serum bilirubin, hypoalbuminemia, and coagulopathy. Hepatic synthetic dysfunction may complicate acute processes (viral, alcoholic, or drug-induced hepatitis, shock liver) or chronic liver diseases (cirrhosis, chronic viral or autoimmune hepatitis). Surgical and perioperative mortality among patients with cirrhosis undergoing major vascular surgery (excluding portal-systemic shunt surgery) ranges from less than 10% in those with well-compensated livers to greater than 75% in those with evidence of severe derangements in hepatic function. Elevations in hepatic transaminases alone do not predict an adverse outcome from anesthesia and surgery, but must be interpreted in context, and may be an early marker of severe hepatic injury after shock, trauma, or toxic insult, or in severe viral hepatitis.

Preoperative efforts for a patient with evidence of hepatic dysfunction must be directed at optimizing nutritional and volume status, including reduction of significant ascites, correcting coagulopathy, anticipating and treating encephalopathy, and correcting electrolyte disturbances. Common operative complications include excessive hemorrhage and exaggerated responses to sedatives, narcotic analgesics, intravenous induction agents, and neuromuscular blocking agents. Postoperative complications include worsening of hepatic function, hemorrhage in the operative field and from the gastrointestinal tract, encephalopathy, impaired wound healing, pneumonia, volume and electrolyte derangements, acute tubular necrosis, and hepatorenal syndrome. See Box 13-12 for management points on liver disease.

[edit] Renal Disease

Special precautions are necessary in patients with acute renal failure (ARF), chronic renal failure (CRF), or the nephrotic syndrome. Among such patients physiologic derangements relevant to surgery include impairment of fluid and electrolyte balance, acidosis, increased risk from hypotension and potential nephrotoxins, exaggerated responses to neuromuscular blocking agents eliminated through renal means, anemia, a hemorrhagic diathesis from platelet dysfunction, hypercoagulability in the setting of the nephrotic syndrome, increased risk of infection, and impaired wound healing.

Azotemic patients handle volume loads poorly, and such patients are at risk for both dehydration and volume overload with perioperative stresses that would be well compensated in individuals with normal renal function. Volume overload, in particular, may result in the rapid development of hypertension, pulmonary edema, or peripheral edema despite normal cardiac function. Hypoalbuminemia from the nephrotic syndrome enhances the risk of total body volume overload and confounds estimation of intravascular volume status.

Potassium handling is impaired in patients with severe renal dysfunction but may also be quite deranged among those with only modest degrees of azotemia in the clinical setting of distal tubulointerstitial disease or the hyporenin- hypoaldosterone syndrome, most commonly noted in diabetic patients. Operative stresses, muscle injuries, burns, gastrointestinal or traumatic hemorrhage, and perioperative infections markedly enhance catabolism, with the production of enormous endogenous potassium loads, whereas transfused blood contributes an added exogenous potassium burden. Metabolic acidoses and further derangements in calcium-phosphate balance may also be engendered by the same processes that promote perioperative hyperkalemia.

A variety of diagnostic pharmaceuticals and medications may compound preexisting renal abnormalities in the perioperative setting. Those that pose the greatest risk to azotemic surgical patients include iodinated radiocontrast agents and aminoglycoside antibiotics. Other medications that carry substantial nephrotoxic risk, but which are less often indicated perioperatively, include nonsteroidal antiinflammatory drugs, chemotherapeutic agents, amphotericin B, and angiotensin-converting enzyme inhibitors. Substitution of interventions with less potential nephrotoxicity should be considered when feasible; if a potential nephrotoxin is strongly indicated on clinical grounds, pretreatment optimization of volume status and close monitoring of renal function are necessary.

Human recombinant erythropoietin improves the anemia associated with renal failure, but in the setting of surgery such patients often require transfusion to achieve or maintain an optimal hematocrit, generally above 30%. Dialysis itself may also improve the anemia somewhat. Clinically significant platelet dysfunction associated with uremia is generally manifested by an abnormal bleeding time. This disorder often does not fully correct despite dialysis but generally responds promptly to some combination of arginine vasopressin, cryoprecipitate, and red cell transfusion. Estrogens have also been found to be effective in controlling the bleeding that is associated with uremia.

Hemodialysis provides the means of optimizing volume, electrolyte balance, acid-base status, and to some extent the hemostasis defect associated with renal failure. Dialysis is generally best timed to occur on the days preceding and following surgery. Patients with the nephrotic syndrome have an increased risk of venous thromboembolism, which is compounded by surgery, and require appropriate prophylaxis (see Table 13-6). See Box 13-13 for management points on renal disease.

[edit] The Pregnant Patient

Pregnant women generally tolerate nonobstetric surgery without significantly increased morbidity or mortality. Putative risks from exposure of the fetus to general anesthetic agents can be minimized by avoiding all but emergent surgery during the periods of organogenesis (first trimester) and central nervous system myelination (third trimester); indeed, the risk to the fetus of nonemergent maternal surgery performed in the second trimester has been shown to be extremely low. The risk of fetal loss or of premature delivery increases substantially with increasing severity of the underlying surgical problem, with intraoperative shock, hypoxemia, or acidosis, and with postoperative complications. Trauma and acute abdominal processes, such as appendicitis and cholecystitis, are the most common clinical conditions necessitating urgent or emergent surgery among pregnant women. Abdominal processes may confound or be confounded by obstetric complications. The possibility of pregnancy must be considered in any woman of childbearing potential who requires surgery.

Evaluation and management of the pregnant woman undergoing nonobstetric surgery require knowledge of the major physiologic changes associated with pregnancy. Blood pressure drops in early to midgestation; relatively modest blood pressure elevations near term, especially after the twenty-eighth week, may presage the obstetric emergency of preeclampsia. Dyspnea, pedal edema, and a third heart sound are common findings in late stages of normal pregnancies. Also late in gestation, uterine compression of the vena cava and aorta may compromise both maternal cardiac output and placental perfusion when a pregnant woman is maintained in supine recumbency, a position also associated with arterial hypoxemia. The normal respiratory pattern of pregnancy is one of hyperventilation (Pco₂ 25 to 30 mm Hg) with metabolic compensation (HCO₃ 17 to 22 mEq/L), factors that must be considered in pregnant patients who require intubation and mechanical ventilation. In normal pregnancies blood and plasma volumes increase substantially, generally with little clinical consequence, although the volume of distribution of many medications may be affected. Venous stasis, from increased filling of capacitance vessels and extrinsic compression of pelvic veins by the uterine contents, and the hypercoagulability associated with pregnancy result in a marked increase in the risk of thromboembolic disease, a situation compounded by trauma or surgery. Delayed gastric emptying and relaxation of the lower esophageal sphincter during pregnancy lead to an increased perioperative risk of aspiration pneumonitis. Dilation of the urinary collecting system predisposes to retention, bacterial colonization, and infection. The glomerular filtration rate increases 30% to 50% in normal pregnancies, with an associated increase in the clearance of many drugs.

Pregnant women undergoing nonobstetric surgery often require adjunctive medications, such as antibiotics, in the perioperative period. Many medications can be safely administered to a pregnant woman without adversely affecting the fetus: heparin, most β-lactam antibiotics, erythromycin, methyldopa, hydralazine, most asthma medications, and others. In many instances, especially with newer drugs, the fetal risk is unknown. A moderate number of common medications are known to be fetotoxic or result in malformations; these should be employed during pregnancy only when strongly indicated and when suitable alternatives do not exist or cannot be used because of other reasons, such as drug allergy. Medications that are problematic in pregnancy include aminoglycoside, tetracycline, quinolone, and sulfonamide antimicrobials; warfarin; angiotensin-converting enzyme inhibitors; phenytoin; barbiturates; high-dose aspirin; histamine₂-blockers; and a wide range of chemotherapeutic agents. See Box 13-14 for management points on the pregnant patient.

[edit] Seizure Disorders

New-onset seizures in the perioperative setting generally relate to metabolic derangements, such as hypoglycemia, severe hyponatremia, and malignant hyperthermia; unanticipated alcohol or sedative-hypnotic withdrawal; toxic medication effects, as with enflurane, high doses of local anesthetic agents, or meperidine against a backdrop of monoamine oxidase inhibitor use; or a structural problem, such as an intracranial tumor or abscess, but especially after trauma, with central nervous system hemorrhage, cerebral contusion, or penetrating brain injury. Patients with established seizure disorders that are well controlled with anticonvulsant medications generally tolerate surgery well, without neurologic sequelae. Indeed, although inhalation induction of anesthesia and emergence are risk periods for increased seizure activity, barbiturate coma and general anesthesia are accepted therapies for refractory status epilepticus. Maintenance of oral anticonvulsant therapy in the perioperative period may not be possible, and alternate dosage forms or different agents may be necessary. Prophylaxis of seizure activity in the neurologically injured patient is somewhat controversial, although there is consensus on treating certain categories of such patients (see below). See Box 13-15 for management points on seizure disorders.

[edit] The Alcoholic Patient

There are 10 million to 20 million alcoholics and problem drinkers in the United States. These individuals require surgery at a greater rate than the nondrinking public on the basis of the medical complications of alcohol (and usually concurrent tobacco abuse), but especially as a result of trauma associated with alcohol use—motor vehicle and other accidents, burns, and interpersonal violence. Complications of surgery in alcohol-abusing patients may relate to (1) metabolic derangements, especially of intermediary metabolism and of potassium, calcium, and magnesium homeostasis, (2) organ system dysfunction, especially of the cardiac, hematologic, hepatic, and central nervous systems, and (3) the withdrawal state. Treatment is predicated on rapid identification and treatment of correctable derangements and anticipatory treatment of alcohol withdrawal states. See Box 13-16 for management points on the alcoholic patient.

[edit] Narcotic-, Sedative-Hypnotic–, and Cocaine-addicted Patients

Patients addicted to narcotics require special attention to analgesics and other medications in the perioperative period. Narcotic tolerance and metabolism are generally enhanced in such patients. Those addicted at the time of surgery require administration of narcotic agents sufficient to suppress withdrawal and control pain, most often achieved by the administration of an appropriate methadone regimen to suppress withdrawal supplemented by appropriate narcotic agonists. Narcotic antagonists and mixed agonist-antagonists must be avoided to preclude precipitating acute narcotic withdrawal. Coordination of this aspect of care with the anesthesiologist is clearly important, since administration of narcotic antagonists and mixed agonist-antagonists in the operative setting is common. Narcotic-addicted patients scheduled to undergo elective surgery may be offered a withdrawal regimen before surgery.

Patients who have been maintained on long-term sedative hypnotics, most commonly benzodiazepines, may experience a withdrawal reaction similar to that associated with withdrawal from alcohol, including seizures and frank delirium, in the postoperative setting. Onset of symptoms may be quite delayed in patients addicted to agents with longer half-lives. Treatment is predicated on identification of the risk of withdrawal, observation for minor symptoms, and administration of appropriate cross-tolerant agents.

Cocaine-addicted individuals carry to surgery an enhanced risk of cardiovascular complications, including blood pressure lability, cardiac ischemic events, dysrhythmias, stroke, and vascular compromise of other organs. This risk appears to persist for at least 2 weeks after last use of cocaine. Elective surgery should be delayed accordingly. Optimum prophylaxis and treatment of these complications in cocaine-addicted individuals who require emergency surgery are not known, but α- andalpha;-and β-adrenergic blockers, labetalol, and chlorpromazine have been used in individual cases. Individuals abusing amphetamines in the preoperative period may experience reactions similar to those associated with cocaine as well as frank psychosis perioperatively. See Box 13-17 for management points on narcotic-, sedative-, hypnotic-, and cocaine-addicted patients.

[edit] REFERENCES