Berek Novak's Gyn 2019. Chapter 25 Preoperative Evaluation and Postoperative Management

 CHAPTER 25

Preoperative Evaluation and Postoperative Management

KEY POINTS

1 The preoperative evaluation should be complete and thorough, taking into account

the essential aspects of the patient’s general medical condition and prior surgical

history. The risks, benefits, and potential complications of the surgical procedure

should be discussed with the patient, including the most frequent complications of

the particular surgical procedure. Alternative management, if any, should be

presented.

2 A calculated body mass index (BMI) can be used as a surrogate marker for nutritional

status.

3 Careful and meticulous fluid and electrolyte management is essential for all patients

undergoing major surgical procedures.

4 Although satisfactory analgesia is easily achievable with available methods, patients

continue to suffer unnecessarily from postoperative pain.

5 Prophylactic antibiotics should be employed judiciously. Prompt identification of

perioperative infections and their specific treatments are critical to minimize the

impact of this common morbidity.

6 Enhanced recovery protocols are comprehensive and include preoperative,

intraoperative, and postoperative elements. Adherence to these protocols has

reduced hospital length of stay, postoperative complications, and health care costs.

7 Early in the clinical course, a postoperative small bowel obstruction may exhibit

signs and symptoms similar to those of an ileus. Initial conservative management as

outlined for the treatment of ileus is appropriate.

8 Because pulmonary embolism is the leading cause of death following gynecologic

surgical procedures, the use of prophylactic venous thromboembolism regimens is

an essential part of management. In patients at moderate risk, intermittent pneumatic

compression (IPC) during and after gynecologic surgery reduces the incidence of

deep venous thrombosis (DVT) on a level similar to that of low-dose or low–

molecular-weight heparin. A combination of mechanical (IPC) and pharmacologic

prophylaxis is recommended for high-risk patients.

9 Patients who are predisposed to cardiovascular, respiratory, and endocrine illnesses

must be thoroughly evaluated preoperatively. Coronary artery disease and chronic

obstructive pulmonary disease (COPD) are major risk factors for patients

undergoing abdominal surgery. Patients with hypertension should receive

medication to control their blood pressure before surgery. Perioperative management

1310of medical complications must be prompt and meticulous.

The successful outcome of gynecologic surgery is based on thorough evaluation,

careful preoperative preparation, and attentive postoperative care. This chapter

discusses the general perioperative management of patients undergoing major

gynecologic surgery and addresses specific medical problems that could

complicate the surgical outcome.

MEDICAL HISTORY AND PHYSICAL EXAMINATION

[1] Gynecologic surgery should be undertaken only after gaining a thorough

understanding of a patient’s medical history and performing a complete physical

examination.

1. The medical history should include detailed questions to identify any

medical illnesses that might be aggravated by surgery or anesthesia.

Coronary artery disease, diabetes, pulmonary diseases, and obesity are the

most common causes of postoperative complications.

2. Medications currently being taken (including nonprescription drugs) and

those discontinued within the month before surgery should be recorded.

Information about the use of “alternative therapies,” herbs, and vitamins

should be elicited (1,2). Specific instructions must be given to the patient

regarding the need to discontinue any medication before surgery (e.g., aspirin,

antiplatelet agents, diuretics, hormone replacement, or oral contraceptives),

and those medications that should be continued (e.g., β-blockers, α2-agonists,

statins, H2 blockers, and proton pump inhibitors). Collaboration with the

anesthesiologist in making decisions about continuing preoperative

medications is essential. In the case of herbal medications, there is no evidence

that herbal medications improve surgical outcomes and many may increase

complications (Table 25-1) (3). It is recommended that all herbal medications

be discontinued at least a week before surgery.

3. The patient should be questioned regarding known allergies to

medications (e.g., sulfa and penicillin), foods, or environmental agents

(latex).

4. Previous surgical procedures, and the patient’s course following those

surgical procedures, should be reviewed to identify and protect against

potential complications. The patient should be asked about specific

complications, such as excessive bleeding, wound infection, venous

thromboembolism, peritonitis, or bowel obstruction. Prior pelvic surgery

should alert the gynecologist to the possibility of distorted surgical anatomy

1311such as adhesions or ureteral stricture. In such cases, it may be prudent to

identify any pre-existing abnormality by performing computed tomography

(CT) or other imaging. Many patients may not be entirely clear about the

extent of the previous surgical procedure or the details of intraoperative

findings. Therefore, operative notes from previous procedures should be

obtained and reviewed.

5. Family history may identify familial traits that might complicate planned

surgery. A family history of excessive intra- or postoperative bleeding,

venous thromboembolism, malignant hyperthermia, and other potentially

inherited conditions should be sought.

6. The review of systems should be detailed to identify any coexisting

medical or surgical conditions. Inquiry about gastrointestinal and urologic

function is particularly important before undertaking pelvic surgery because

many gynecologic diseases involve adjacent nongynecologic viscera. The

patient may have less serious symptoms, which could be corrected along with

performing the primary surgery (e.g., stress urinary incontinence, fecal soiling,

symptomatic cystocele).

7. Although many women undergoing gynecologic surgical procedures are

otherwise healthy, with pathology identified only on pelvic examination,

other major organ systems should not be neglected in the physical

examination. Identification of abnormalities, such as a heart murmur,

pulmonary compromise, hernia, or osteoarthritis of hips or knees should lead

the surgeon to obtain additional testing and consultation to minimize intra- and

postoperative complications.

Table 25-1 Potential Effects of Common Herbal and Dietary Supplements

Herb/Dietary

Supplement

Potential Perioperative Effect

Aconite Potential ventricular arrhythmias

Aloe May potentiate thiazides

Black cohosh May potentiate hypotensive effects

Danshen May cause bleeding

Dong quai May cause bleeding

Echinacea Allergic reactions; decreased effectiveness of immunosuppressants

Ephedra/ma Risk of myocardial ischemia and stroke from tachycardia and

1312huang hypertension; ventricular arrhythmias with halothane; long-term use

may cause intraoperative hemodynamic instability; life-threatening

interaction with monoamine oxidase inhibitors, anesthesia, potential

for withdrawal

Garlic May increase bleeding risk

Ginkgo May increase bleeding risk

Ginseng Lowers blood sugar and may increase bleeding risk

Kava May increase the sedative effect of anesthetics (an association

between kava use and fatal hepatotoxicity has been reported)

Licorice May cause hypertension and hypokalemia

Senna May cause electrolyte imbalance

St. John’s wort Induction of cytochrome p450 enzyme; excessive sedation and

delayed emergence from general anesthesia; potential serotonin

syndrome if used in combination with other serotonergic agents

Valerian Excessive sedation and delayed emergence from general anesthesia;

benzodiazepine-like acute withdrawal

Yerba mate May cause hypertension or hypotension and excess sympathetic

nervous system stimulation

LABORATORY EVALUATION

“Routine” preoperative laboratory testing of healthy women is to be discouraged

as abnormal results are infrequent and are rarely of consequence in the surgical or

anesthetic management of the patient (4). Despite well-established guidelines,

approximately 90% of patients undergo unnecessary testing in a major university

medical center (5). The selection of appropriate preoperative laboratory studies

should depend on the type of the anticipated surgical procedure and the patient’s

medical status. The Revised Cardiac Risk Index (RCRI) can be used to predict the

risk of cardiac complications after noncardiac surgery. These clinical risk factors

include cerebrovascular disease, congestive heart failure (CHF), creatinine level

>2 mg/dL, and insulin-dependent diabetes.

Chest x-ray:

New or unstable cardiovascular or pulmonary signs or symptoms

At risk for pulmonary complications

Electrocardiogram:

1313Signs or symptoms of cardiovascular disease

High-risk surgery (risk of perioperative cardiac event >5%)

Intermediate-risk surgery (risk of perioperative cardiac event 1% to

5% and one RCRI clinical risk factor)

Complete blood count:

Major surgery

Patients with conditions with an increased risk for anemia

Renal function:

Recognized renal or cardiovascular disease

Coagulation studies (activated partial thromboplastin time [APTT],

prothrombin time [PT], platelet count):

Not recommended routinely

Patients with a history of bleeding or liver disease

Patients taking anticoagulants

History or examination concerning for an underlying coagulation

disorder

Urinalysis:

Not recommended routinely; may be considered, given symptoms or

history

Consider for invasive urologic procedures (6)

Tumor markers:

In general, tumor markers are used to follow patients with known

ovarian cancers. They are not considered “diagnostic.” However,

an elevated tumor marker in a patient with a “suspicious” pelvic

mass may be helpful in determining whether the patient should be

referred to a gynecologic oncologist (7).

Imaging of adjacent organ systems should be undertaken in individual cases as

follows:

1. CT urography is helpful to delineate ureteral patency and course, especially in

the presence of a pelvic mass, gynecologic cancer, or congenital müllerian

anomaly. A CT urogram is not of value in the evaluation of most patients

undergoing pelvic surgery.

2. Upper endoscopy, colonoscopy, barium enema, or upper gastrointestinal

studies with small bowel assessment may be of value in evaluating some

patients before undergoing pelvic surgery. Because of the proximity of the

female genital tract to the lower gastrointestinal tract, the rectum and sigmoid

colon may be involved with benign (endometriosis or pelvic inflammatory

disease) or malignant gynecologic conditions. Conversely, a pelvic mass could

have a gastrointestinal origin such as a diverticular abscess or a mass of

1314inflamed small intestines (Crohn disease) or, rarely, a gastric or pancreatic

carcinoma. Any patient with gastrointestinal symptoms should be further

evaluated.

3. Other imaging studies, including ultrasonography, CT scanning, or magnetic

resonance imaging (MRI), may be useful in selected patients, for example, to

evaluate a pelvic mass. There are several scoring systems that can be used in

the evaluation of a pelvic ultrasound that may suggest the increased likelihood

of ovarian cancer (8).

PREOPERATIVE DISCUSSION AND INFORMED CONSENT

[1] The preoperative discussion should include a description of the surgical

procedure, its expected outcome, and risks and is the basis for obtaining

signed informed consent (9,10). Informed consent is an educational process for

the patient and her family and fulfills the need to convey information in

understandable terms. The items listed in Table 25-2 should be discussed, and,

after each item, the patient and family should be invited to ask questions.

Documentation of the discussion is an important component of the patient’s

record that the physician should always include with the preprinted surgical

consent form.

Following are components of the informed consent process:

1. A discussion of the nature and the extent of the disease process should

include an explanation in lay terms of the significance of the disease or

condition. Printed materials, computer-based learning programs, and videos

may assist in this process. The patient’s competency to understand the

discussion and written consent should be assessed. If the patient speaks a

different language, a qualified interpreter should be present and the presence

of the interpreter documented.

2. The goals of proposed surgery should be discussed in detail. Some

gynecologic surgical procedures are performed purely for diagnostic purposes

(e.g., dilation and curettage, cold knife conization, diagnostic laparoscopy),

whereas most are aimed at correcting a specific problem. The extent of the

surgery should be outlined, including which organs will be removed. Most

patients like to be informed regarding the type of surgical incision and the

estimated duration of anesthesia.

3. The expected outcome of the surgical procedure should be explained. If

the procedure is being performed for diagnostic purposes, the outcome will

depend on surgical or pathologic findings that are not known before surgery.

When treating an anatomic deformity or disease, the expected success of the

1315operation should be discussed, and the potential for failure of the operation

(e.g., failure of tubal sterilization or the possibility that stress urinary

incontinence may not be alleviated, or may recur). When treating cancer, the

possibility of finding advanced disease and the potential need for adjunctive

therapy (e.g., postoperative radiation therapy or chemotherapy) should be

mentioned. Other issues of importance to the patient include discussion of loss

of fertility or loss of ovarian function. These issues should be raised by the

physician to ensure that the patient adequately understands the

pathophysiology that may result from the surgery and to allow her to express

her feelings regarding these issues. Unanticipated findings at the time of

surgery should be mentioned. For example, if the ovaries are unexpectedly

found to be diseased, the best surgical judgment may be that they should be

removed.

4. The risks and potential complications of the surgical procedure should be

discussed, including the most frequent complications of the particular

surgical procedure. For most major gynecologic surgeries, the risks include

intra- and postoperative hemorrhage, postoperative infection, venous

thromboembolism, injury to adjacent viscera, and wound complications. The

patient should understand that minimally invasive surgery, despite small skin

incisions, may have many of the same risks of injury or complications as does

“open” surgery. Given the potential for transfusion of blood products, it should

be clarified whether the patient would object to receiving a transfusion. Preexisting medical problems (e.g., diabetes, obesity, chronic obstructive

pulmonary disease [COPD], coronary artery disease) result in additional risks

and should be reviewed with the patient. Measures that will be taken to reduce

the risk of complications should be described (e.g., prophylactic antibiotics,

bowel preparation, venous thromboembolism prophylaxis).

5. The usual postoperative course should be discussed in enough detail to

allow the patient to understand what to expect in the days following

surgery. Information regarding the need for a suprapubic catheter, prolonged

central venous monitoring, or an intensive care stay helps the patient accept

her postoperative course and avoids surprises that may be disconcerting to the

patient and her family. The expected duration of the recovery period, in and

out of the hospital, should be outlined.

6. The surgeon should describe others who will be involved with the surgical

procedure (residents, assistants) and their roles in the patient’s care. Any

conflict of interest should be disclosed.

7. Alternative methods of therapy should be discussed, including medical

management or other surgical approaches. The potential risks and benefits

of alternative treatments should be discussed.

13168. The patient should have an understanding of the outcome of the disease.

Table 25-2 Outline of Key Points of the Preoperative Informed Consent Discussion

1. The nature and extent of the disease process

2. The extent of the actual operation proposed and the potential modifications of the

operation, depending on intraoperative findings

3. The anticipated benefits of the operation, with a conservative estimate of successful

outcome

4. The risks and potential complications of the surgery

5. Alternative methods of therapy and the risks and results of those alternative methods

of therapy

6. The results likely, if the patient is not treated

GENERAL CONSIDERATIONS

Nutrition

Young patients undergoing elective gynecologic surgery have adequate

nutritional stores and, for the most part, do not require nutritional support. All

patients should have a nutritional assessment, especially elderly patients and

those undergoing gynecologic cancer surgery or other major gynecologic

procedures in which a prolonged postoperative recovery is expected.

Nutritional status should be reassessed at regular intervals postoperatively until

the patient successfully returns to a regular diet.

A nutritional assessment includes a careful history and physical examination,

which are the most useful, reliable, and cost-effective methods of determining a

patient’s nutritional status. In particular, information about recent weight loss,

dietary history, fad diets, extreme exercise, or anorexia or bulimia should be

elucidated. Physical evidence of malnutrition, including temporal wasting, muscle

wasting, ascites, and edema, should be noted. Accurate height and weight

measurements should be obtained and an ideal body weight, percentage ideal

body weight, and percentage usual body weight may be calculated. A variety of

techniques were developed to determine a patient’s nutritional state; however,

many methods lack clinical utility outside of a research setting. Anthropometric

measurements of skin-fold thickness and arm-muscle circumference provide an

estimate of total body fat and lean muscle mass. Nutritional screening

assessments, such as the Mini Nutritional Assessment and the Prognostic

1317Nutritional Index, have been described to evaluate a patient’s nutritional status

(11).

[2] The calculated body mass index (BMI) can be used as a surrogate marker for

nutritional status. BMI is calculated as body weight in kilograms divided by

the height in square meters. A BMI less than 22 increases the risk of

malnutrition, and a BMI less than 19 gives clear evidence of malnutrition

(12). The degree of malnutrition can in part be determined by serum

concentrations of albumin, transferrin, and prealbumin. The levels of these serum

proteins are greatly influenced by the patient’s level of hydration. Prealbumin has

the shortest half-life, at 2 to 3 days, and levels of this protein are depressed very

early in comparison with serum transferrin and albumin, which have half-lives of

8 and 20 days, respectively (14). Serum albumin is a substitute for the Prognostic

Nutritional Index, which is a time-consuming calculation, in assessing

malnutrition in women with gynecologic malignancies (15). Hypoalbuminemia,

characterized by an albumin of less than 3.5 g/dL, is correlated with

morbidity, mortality, and increased postoperative complication rates in data

from the National Surgical Quality Improvement Program (13). Proteinenergy malnutrition, as characterized by hypoalbuminemia, can result in poor

wound healing, increased risk for infection, functional decline, increased

morbidity and mortality.

Decisions regarding the need for nutritional support should be based on several

individualized factors. These factors include the patient’s prior nutritional state,

the anticipated length of time in which the patient will not be able to eat, the

extent of surgery, and the likelihood of complications. The nutritional assessment

should determine whether the cause of the malnutrition is increased enteral loss

(malabsorption, intestinal fistula), decreased oral intake, increased nutritional

requirements as a result of hypermetabolism (sepsis, malignancy), or a

combination of these factors. Severe malnutrition, if not corrected, can further

complicate the postoperative problem by causing altered immune function,

chronic anemia, impaired wound healing, and eventually multiple organ system

failure and death.

There is evidence that preoperative nutritional support may improve

postoperative outcomes in patients with significant pre-existing malnutrition

(16,17). According to the American Society for Parenteral and Enteral

Nutrition (ASPEN) guidelines, evidence-based medicine supports the use of

preoperative nutritional support for 7 to 14 days in severely malnourished

patients undergoing major nonemergent gastrointestinal surgery (14).

Preoperative oral supplementation may reduce infectious complications and

shorten hospital stays, however the effect on mortality is unclear (18). ASPEN

guidelines do not support the routine use of parenteral nutritional support in the

1318immediate postoperative period for patients undergoing major gastrointestinal

surgery; however, the guidelines do indicate a role for nutritional support

postoperatively in patients in whom oral intake will be inadequate for 7 to 10 days

(14).

Antimicrobial Prophylaxis in Gynecologic Surgery

Gynecologic procedures often involve breaching the reproductive and

gastrointestinal tracts, which harbor endogenous bacteria capable of causing

polymicrobial infections in the postoperative period (Table 25-3). Despite great

advances in aseptic technique and drug development, bacterial

contamination of the operative site and postoperative infections are a

frequent part of the practice of gynecologic surgery. Prevention of these

surgical complications includes using proper aseptic technique, minimizing tissue

trauma, minimizing the amount of foreign material in the surgical site, controlling

diabetes, avoiding immunologic suppression, maximizing tissue oxygenation,

draining blood and serum from the surgical site, and using prophylactic

antibiotics. [5] Antibiotic prophylaxis is given with the belief that antibiotics

enhance the immune mechanisms in host tissues that resist infections by killing

the bacteria that inoculate the surgical site during surgery (19).

Table 25-3 Bacteria Indigenous to the Lower Genital Tract

Lactobacillus Enterobacter agglomerans

Diphtheroids Klebsiella pneumoniae

Staphylococcus aureus Proteus mirabilis

Staphylococcus epidermidis Proteus vulgaris

Streptococcus agalactiae Morganella morganii

Streptococcus faecalis Citrobacter diversus

α-Hemolytic streptococci Bacteroides species

B. disiens

B. fragilis

B. melaninogenicus

Group D streptococci

Peptostreptococci

Peptococcus

Clostridium

1319Gaffkya anaerobia

Escherichia coli

Fusobacterium

Enterobacter cloacae

Infections in the skin or pelvis that result from gynecologic surgery (e.g.,

parametritis, cuff cellulitis, pelvic abscess) typically are polymicrobial in nature.

These infections are complex and often involve gram-negative rods, grampositive cocci, and anaerobes. Antibiotic prophylaxis should be sufficiently broad

to cover these potential pathogens (Table 25-4) (20).

The timing of antimicrobial prophylaxis is important. There is a relatively

narrow window of opportunity for affecting outcomes (21). In the United

States, it is customary to give antimicrobial prophylaxis shortly before or during

the induction of anesthesia. Data revealed that a delay of 3 hours or more

between the time of bacterial inoculation (i.e., skin incision) and

administration of antibiotics may result in ineffective prophylaxis. Evidence

indicates that for prophylaxis, one dose of antibiotic is appropriate. When

the surgical procedure proceeds longer than 1 to 2 times the half-life of the

drug or blood loss is greater than 1.5 L, additional intraoperative doses of

antibiotics should be administered to maintain adequate levels of medication

in serum and tissues (22). There are no data to support the continuation of

prophylactic antimicrobial agents into the postoperative period for routine

gynecologic procedures.

Cephalosporins emerged as the most important class of antimicrobial

agents for prophylaxis. These drugs have a broad spectrum and relatively low

incidence of adverse reactions. Cefazolin (1 g) appears to be widely used in the

United States by gynecologic surgeons because of its relatively low cost and long

half-life (1.8 hours). Other cephalosporins such as cefoxitin, cefotaxime, and

cefotetan are commonly used for prophylaxis. These agents appear to have a

broader spectrum of activity against anaerobic bacteria and are appropriate

selections when colorectal resections are possible, such as during a debulking

surgery for ovarian cancer. For the majority of gynecologic procedures, there is

little evidence that a clinically relevant distinction exists between cefazolin and

the other agents. Obese patients, defined as having a BMI greater than 35 or

weight greater than 100 kg, should receive 2 g of cefazolin to achieve appropriate

blood and tissue antibiotic concentrations (20).

Table 25-4 Antibiotic Prophylaxis Regimens by Procedure

1320Procedure Antibiotic Dose

Hysterectomy

Urogynecology

procedures, including

those involving mesh

Cefazolina

Clindamycinc plus

gentamicin or

quinoloned or

aztreonam

Metronidazolec plus

gentamycin or

quinoloned

1 g or 2 g IVb 600 mg IV 1.5 mg/kg IV

400 mg IV 1 g IV 500 mg IV 1.5

mg/kg IV 400 mg IV 1 g IV

Hysterosalpingogram

or Chromotubation

Doxycyclinee 100 mg orally, twice daily for 5 days

Induced

abortion/dilation and

evacuation

Doxycycline

Metronidazole

100 mg orally 1 hr before procedure

and 200 mg orally after procedure

500 mg orally twice daily for 5 days

aAlternatives include cefotetan, cefoxitin, cefuroxime, or ampicillin–sulbactam.

b

A 2-g dose is recommended in women with a body mass index greater than 35 or weight

greater than 100 kg or 220 lb.

cAntimicrobial agents of choice in women with a history of immediate hypersensitivity to

penicillin.

d

Ciprofloxacin or levofloxacin or moxifloxacin.

eIf patient has a history of pelvic inflammatory disease or procedure demonstrates dilated

fallopian tubes. No prophylaxis is indicated for a study without dilated tubes.

IV, intravenously.

Adapted from Antibiotic prophylaxis for gynecologic procedures. American College of

Obstetricians and Gynecologists Practice Bulletin No. 104, May 2009.

Antimicrobial prophylaxis, although usually beneficial, is not without risk.

Anaphylaxis is the most life-threatening complication from antibiotic use.

Allergic reactions to penicillins are reported in 0.7% to 8% courses of

treatment (23). While first-generation cephalosporins have cross-reactivity to

penicillins, data indicate that it is safe to administer second- or third-generation

cephalosporins to women who report a history of adverse reactions to penicillins

as the cross-allergy is negligible (24).

A single dose of broad-spectrum antibiotics can result in

pseudomembranous colitis, caused by Clostridium difficile. Diarrhea may

develop in as many as 15% of hospitalized patients treated with β-lactam

antibiotics (25). These gastrointestinal complications from antibiotics may cause

1321serious morbidity in the surgical patient, and the surgeon should be able to

recognize and manage these problems.

Not all gynecologic surgery patients need to receive prophylactic

antibiotics. The surgeon should choose agents to cover procedures based on

available data, thereby avoiding the potential for adverse reactions and

minimizing the unnecessary use of antibiotics, which may contribute to increased

rates of antimicrobial resistance. In patients with cephalosporin allergies or

anaphylaxis to penicillin, other drugs or combinations should be chosen to

provide adequate prophylactic coverage. Antimicrobial prophylaxis options for

common gynecologic procedures are presented in Table 25-3. Antibiotic

prophylaxis is not indicated for diagnostic or operative laparoscopy, exploratory

laparotomy, or diagnostic or operative hysteroscopy, including endometrial

ablation, intrauterine device insertions, endometrial biopsy, or urodynamics (20).

Subacute Bacterial Endocarditis Prophylaxis

It was thought that women who had severe valvular disease or other cardiac

conditions required antibiotic prophylaxis prior to genitourinary (GU) or

gastrointestinal (GI) procedures in order to prevent bacterial endocarditis as a

result of the transient bacteremia provoked by the surgery. After reviewing the

pertinent evidence-based literature, the American Heart Association (AHA)

issued revised guidelines in 2007 stating that antibiotic prophylaxis was not

necessary solely to prevent endocarditis in patients undergoing GI or GU

procedures, including hysterectomy (26). For those at highest risk of infective

endocarditis undergoing high-risk procedure, antibiotics should still be considered

(Table 25-5).

Postoperative Infections

Infections are a major source of morbidity in the postoperative period. Risk

factors for infectious morbidity include the absence of perioperative antibiotic

prophylaxis, contamination of the surgical field from infected tissues or from

spillage of large bowel contents, an immunocompromised host, poor nutrition,

chronic and debilitating severe illness, poor surgical technique, and pre-existing

focal or systemic infection. Sources of postoperative infection can include the

lung, urinary tract, surgical site, pelvic sidewall, vaginal cuff, abdominal wound,

and sites of indwelling intravenous catheters. Early identification and treatment of

infection will result in the best outcome for these potentially serious

complications.

Table 25-5 Recommendations for Prophylaxis of Bacterial Endocarditis

1322Highest-Risk

Patients

Agents Regimen (Within 30–60 min of

Starting Procedure)

Standard regimen Amoxicillin 2 g PO

Ampicillin 2 g IM or IV

or

Cefazolin or

ceftriaxone

1 g IM or IV

Cephalexin 2 g

Penicillin-allergic

(oral)

Cephalexin 2 g

Clindamycin 600 mg

Azithromycin or

clarithromycin

500 mg

Penicillin-allergic

(non-oral)

Cefazolin or

ceftriaxone

1 g IM or IV

Clindamycin 600 mg IM or IV

IM, intramuscularly; IV, intravenously.

Derived from Wilson W, Taubert KA, Gewitz M, et al. Prevention of infective

endocarditis: guidelines from the American Heart Association: a guideline from the

American Heart Association Rheumatic Fever, Endocarditis, and Kawasaki Disease

Committee, Council on Cardiovascular Disease in the Young, and the Council on Clinical

Cardiology, Council on Cardiovascular Surgery and Anesthesia, and the Quality of Care

and Outcomes Research Interdisciplinary Working Group. Circulation 2007;116:1736–

1754.

Although infectious morbidity is a frequent complication of surgery, the

incidence of infections can be decreased by the appropriate use of simple

preventive measures. Systemic antibiotic prophylaxis and meticulous surgical

technique will help decrease the incidence of postoperative pelvic and abdominal

infections in these patients. Blood and necrotic tissue are excellent media for the

growth of aerobic and anaerobic organisms. Care should be taken to obtain

hemostasis to prevent postoperative hematomas. Antibiotic therapy, rather than

prophylaxis, should be initiated during surgery in patients who have frank intraabdominal infection or pus. Elective surgical procedures should be postponed in

patients who have preoperative infections. In an epidemiologic study conducted

1323by the Centers for Disease Control and Prevention (CDC), the incidence of

nosocomial surgical infections ranged from 4.3% in community hospitals to 7%

in municipal hospitals (27). Data confirmed this, with an incidence of 2% to 5%

(28). Urinary tract infections accounted for approximately 40% of these

nosocomial infections. Infections of the skin and wound accounted for

approximately one-third of the infections, and respiratory tract infections

accounted for approximately 16%. In patients who had any type of infection

before surgery, the risk of infection at the surgical wound site increased fourfold.

Rates of infection were higher in older patients, in patients with increased length

of surgery, and in those with increased length of hospital stay before surgery. The

relative risk was three times higher in patients with a community-acquired

infection before surgery. These community-acquired infections included

infections of the urinary and respiratory tracts.

Historically, the standard definition of febrile morbidity for surgical

patients was the presence of a temperature higher than or equal to 100.4°F

(38°C) on two occasions at least 4 hours apart during the postoperative

period, excluding the first 24 hours. Other sources defined fever as two

consecutive temperature elevations greater than 101.0°F (38.3°C) (29,30).

Febrile morbidity is estimated to occur in as many as one-half of patients; it is

often self-limited, resolves without therapy, and is usually noninfectious in origin

(31). Overzealous evaluations of postoperative fever, especially during the early

postoperative period, are time consuming, expensive, and sometimes

uncomfortable for the patient (31). The value of 101.0°F is more discriminatory

than 100.4°F to distinguish an infectious cause from an inconsequential

postoperative fever in the benign gynecologic patient.

The assessment of a febrile surgical patient should include a review of the

patient’s history with regard to risk factors. The history and the physical

examination should focus on the potential sites of infection (Table 25-6). The

examination should include inspection of the pharynx, a thorough pulmonary

examination, percussion of the kidneys to assess for costovertebral angle

tenderness, inspection and palpation of the abdominal incision, examination of

sites of intravenous catheters, and an examination of the extremities for evidence

of deep venous thrombosis (DVT) or thrombophlebitis. In gynecologic patients,

an appropriate workup may include inspection and palpation of the vaginal cuff

for signs of induration, tenderness, or purulent drainage. A pelvic examination

should be performed to identify a mass consistent with a pelvic hematoma or

abscess and to look for signs of pelvic cellulitis.

Patients with fever in the early postoperative period should have an

aggressive pulmonary toilet, including incentive spirometry (30). [5] Fever

within the first 48 hours of surgery is likely to be cytokine related; thus if fever

1324persists beyond 48 hours postoperatively, additional laboratory and radiologic

data may be obtained. The evaluation may include complete and differential white

blood cell counts and a urinalysis. In one study, results from fever workups

included positive blood cultures in 9.7% of patients, a positive urine culture in

18.8%, and a positive chest x-ray in 14%. These data support the need for a

tailored workup based on the patient’s clinical picture (32). Blood cultures can be

obtained but will most likely be of little yield unless the patient has fever at the

time of collection. In patients with costovertebral angle tenderness, renal

ultrasound or CT urogram may be indicated to rule out the presence of ureteral

damage or obstruction from surgery, particularly in the absence of laboratory

evidence of urinary tract infection. Patients who have persistent fevers without a

clear localizing source should undergo CT scanning of the abdomen and pelvis to

rule out the presence of an intra-abdominal abscess. If fever persists in patients

who had gastrointestinal surgery, a barium enema or upper gastrointestinal studies

with small bowel assessment may be indicated late in the course of the first

postoperative week to rule out an anastomotic leak or fistula.

Table 25-6 Posthysterectomy Infections

Operative Site Nonoperative Site

Vaginal cuff Urinary tract

Pelvic cellulitis Asymptomatic bacteriuria

Pelvic abscess Cystitis

Intraperitoneal Respiratory

Adnexa Atelectasis

Cellulitis Pneumonia

Abscess Vascular

Abdominal incision Phlebitis

Cellulitis Septic pelvic thrombophlebitis

Simple Pyelonephritis

Progressive bacterial synergistic

Necrotizing fasciitis

1325Myonecrosis

Urinary Tract Infections

Historically, the urinary tract was the most common site of infection in surgical

patients (33). A significant decrease in urinary tract infections was noted after the

institution of perioperative use of prophylactic antibiotics. The incidence of

postoperative urinary tract infection in gynecologic surgical patients not receiving

prophylactic antibiotics is thought to be as high as 40%, and even a single dose of

perioperative prophylactic antibiotic decreases the incidence of postoperative

urinary tract infection to as low as 4% (34,35).

Symptoms of a urinary tract infection may include urinary frequency, urgency,

and dysuria. In patients with pyelonephritis, patients may exhibit systemic

symptoms such as headache, malaise, nausea, and vomiting. A urinary tract

infection is diagnosed on the basis of microbiology and is defined as the growth

of 105 organisms cultured per milliliter of a voided specimen urine. Most

infections are caused by coliform bacteria, with Escherichia coli being the most

frequent pathogen. Other pathogens include Klebsiella, Proteus, and Enterobacter

species. Staphylococcus organisms are the causative bacteria in fewer than 10%

of cases.

Despite the high incidence of urinary tract infections in the postoperative

period, few of these infections are serious. Most are confined to the lower urinary

tract, and pyelonephritis is a rare complication (36). Catheterization of the urinary

tract, either intermittently or continuously with the use of an indwelling catheter,

is implicated as a main cause of urinary tract contamination (37). More than 1

million catheter-associated urinary tract infections occur yearly in the United

States, and catheter-associated bacteria remain the most common etiology of

gram-negative bacteremia in hospitalized patients. Bacteria adhere to the

surface of urinary catheters and grow within bile films, which appear to protect

embedded bacteria from antibiotics, making treatment less effective. The use of

urinary tract catheters should be minimized. An indwelling catheter should be

removed or replaced in a patient undergoing treatment for catheter-related

infections.

The treatment of urinary tract infection includes hydration and antibiotic

therapy. Infections may be considered complicated or uncomplicated which

helps providers dictate the type and length of therapy. Infections are considered

complicated if they occur in patients with poorly controlled diabetes,

immunosuppression, pregnancy, kidney disease, urinary obstruction, anatomic

abnormality of the genitourinary tract, recent hospitalization, or those with an

indwelling catheter, stent, or nephrostomy tube (38). Commonly prescribed and

1326effective antibiotics include penicillin, sulfonamides, cephalosporins,

fluoroquinolones, and nitrofurantoin (39). The choice of antibiotic should be

based on knowledge of the susceptibility of organisms cultured at a particular

institution. In some institutions, for example, more than 40% of E. coli strains are

resistant to ampicillin. For uncomplicated urinary tract infections, an antibiotic

that has good activity against E. coli should be given in the interim while awaiting

results of the urine culture and sensitivity data.

Patients who have a history of recurrent urinary tract infections, those with

chronic indwelling catheters (Foley catheters or ureteral stents), and those who

have urinary conduits should be treated with antibiotics that will be effective

against the less common urinary pathogens such as Klebsiella and Pseudomonas.

Chronic use of fluoroquinolones for prophylaxis is not advised because these

agents are notorious for inducing antibiotic-resistant strains of bacteria.

Pulmonary Infections

The respiratory tract is an uncommon site for infectious complications in

gynecologic surgical patients. This low incidence is probably a reflection of the

young age and good health status of gynecologic patients in general. In acute care

facilities, pneumonia is a frequent hospital-acquired infection, particularly in

elderly patients (40). Risk factors include extensive or prolonged atelectasis,

preexistent COPD, severe or debilitating illness, central neurologic disease

causing an inability to clear oropharyngeal secretions effectively, nasogastric

suction, and a prior history of pneumonia (40,41). In surgical patients, early

ambulation and aggressive management of atelectasis are the most important

preventive measures.

A significant percentage (40% to 50%) of cases of hospital-acquired

pneumonia is caused by gram-negative organisms (33). These organisms gain

access to the respiratory tract from the oral pharynx. Gram-negative colonization

of the oral pharynx is increased in patients in acute care facilities and is associated

with the presence of nasogastric tubes, pre-existing respiratory disease,

mechanical ventilation, and tracheal intubation (42). The use of antimicrobial

drugs seems to significantly increase the frequency of colonization of the oral

pharynx with gram-negative bacteria.

A thorough lung examination should be included in the assessment of all

febrile surgical patients. In the absence of significant lung findings, chest

radiography is probably of little benefit in patients at low risk for postoperative

pulmonary complications. In patients with pulmonary findings or with risk factors

for pulmonary complications, chest radiography should be performed. A sputum

sample should be obtained for Gram stain and culture. The treatment should

include postural drainage, aggressive pulmonary toilet, and antibiotics. The

1327antibiotic chosen should be effective against gram-positive and gram-negative

organisms. In patients who are receiving assisted ventilation, the antibiotic

spectrum should include drugs that are active against Pseudomonas organisms.

Phlebitis

Historically, intravenous catheter–related infections were common; the reported

incidence is 25% to 35% in the 1980s (43). Because the incidence of catheterrelated phlebitis increases significantly after 72 hours, intravenous catheters

should be changed after 72 to 96 hours with special care to examine the site and

dressing daily (44).

The catheter should be removed if there is any associated pain, redness, or

induration. Phlebitis can occur even with close surveillance of the intravenous

site. In one study, more than 50% of the cases of phlebitis became evident more

than 12 hours after discontinuation of intravenous catheters (45). Less than onethird of patients had symptoms related to the intravenous catheter site 24 hours

before the diagnosis of phlebitis.

Phlebitis can be diagnosed based on the presence of fever, pain, redness,

induration, or a palpable venous cord. Occasionally, suppuration will be present.

Phlebitis is usually self-limited and resolves within 3 to 4 days. The treatment

includes application of warm, moist compresses and prompt removal of any

catheters from the infected vein. Antibiotic therapy with antistaphylococcal

agents should be instituted for catheter-related sepsis. Excision or drainage of an

infected vein rarely is necessary.

Wound Infections

The results of a prospective study of more than 62,000 wounds were revealing in

regard to the epidemiology of wound infections (46). The wound infection rate

varied markedly, depending on the extent of contamination of the surgical field.

The wound infection rate for clean surgical cases (infection not present in the

surgical field, no break in aseptic technique, no viscus entered) was lower than

2%, whereas the incidence of wound infections with dirty, infected cases was

40% or higher. Preoperative showers with hexachlorophene slightly lowered the

infection rate for clean wounds, whereas preoperative shaving of the wound site

with a razor increased the infection rate. The wound infection rate increased with

the duration of preoperative hospital stay and with the duration of surgery.

Incidental appendectomy increased the risk of wound infection in patients

undergoing clean surgical procedures. The study concluded that the incidence of

wound infections could be decreased by short preoperative hospital stays,

hexachlorophene showers before surgery, minimizing shaving of the wound

site, use of meticulous surgical technique, decreasing operative time as much

1328as possible, bringing drains out through sites other than the wound, and

dissemination of information to surgeons regarding their wound infection

rates. A program instituting these conclusions led to a decrease in the clean

wound infection rate from 2.5% to 0.6% over an 8-year period. The wound

infection rate in most gynecologic services is lower than 5%, reflective of the

clean nature of most gynecologic operations.

The symptoms of wound infection often occur late in the postoperative

period, usually after the fourth postoperative day, and may include fever,

erythema, tenderness, induration, and purulent drainage. Wound infections

that occur on postoperative days 1 through 3 are generally caused by

streptococcal and Clostridia infections. The management of wound infections is

mostly mechanical and involves opening the infected portion of the wound above

the fascia, with cleansing and debridement of the wound edges as necessary.

Wound care, consisting of debridement and dressing changes two to three times

daily, will promote growth of granulation tissue, with gradual filling in of the

wound defect by secondary intention. Clean, granulating wounds can often be

secondarily closed with good success, shortening the time required for complete

wound healing.

Pelvic Cellulitis

Vaginal cuff cellulitis is present in most patients who undergo hysterectomy.

It is characterized by erythema, induration, and tenderness at the vaginal cuff. A

purulent discharge from the apex of the vagina may be present. The cellulitis is

often self-limited and does not require any treatment. Fever, leukocytosis, and

pain localized to the pelvis may accompany severe cuff cellulitis and most often

signifies extension of the cellulitis to adjacent pelvic tissues. In such cases, broadspectrum antibiotic therapy should be instituted with coverage for gram-negative,

gram-positive, and anaerobic organisms. If purulence at the vaginal cuff is

excessive or if there is a fluctuant mass noted at the vaginal cuff, the vaginal cuff

should be gently probed and opened with a blunt instrument. The cuff can be left

open for dependent drainage or, alternatively, a drain can be placed into the lower

pelvis through the cuff and removed when drainage, fever, and symptoms in the

lower pelvic region have resolved.

Intra-Abdominal and Pelvic Abscess

The development of an abscess in the surgical field or elsewhere in the abdominal

cavity is an uncommon complication after a gynecologic surgery. It is likely to

occur in contaminated cases in which the surgical site is not adequately drained or

as a secondary complication of hematomas. The causative pathogens in patients

who have intra-abdominal abscesses are usually polymicrobial in nature. The

1329aerobes most commonly identified include E. coli, Klebsiella, Streptococcus,

Proteus, and Enterobacter. Anaerobic isolates are common, usually from the

Bacteroides group. These pathogens arise mainly from the vaginal tract but can be

derived from the gastrointestinal tract, particularly when the colon was entered at

the time of surgery.

Intra-abdominal abscess may be difficult to diagnose. The evolving clinical

picture is often one of persistent febrile episodes with a rising white blood cell

count. Findings on abdominal examination may be equivocal. If an abscess is

located deep in the pelvis, it may be palpable by pelvic or rectal examination. CT

scanning is more sensitive and specific than ultrasonography for diagnosing

intra-abdominal abscesses and often is the radiologic procedure of choice.

Standard therapy for intra-abdominal abscess is evacuation and drainage

combined with appropriate administration of antibiotics. Abscesses located

low in the pelvis, particularly in the area of the vaginal cuff, can be reached

through a vaginal approach. In many patients, the ability to drain an abscess by

placement of a drain percutaneously under CT guidance obviated the need for

surgical exploration. With CT guidance, a pigtail catheter is placed into an

abscess cavity via transabdominal, transperineal, transrectal, or transvaginal

approaches. The catheter is left in place until drainage output decreases.

Transperineal and transrectal drainage of deep pelvic abscesses are successful in

90% to 93% of patients, obviating the need for surgical management (47,48). For

those patients in whom radiologic drainage is not successful, surgical

exploration and evacuation are indicated. The standard approach to initial

antibiotic therapy is to employ broad-spectrum parenteral antibiotic therapy with

both aerobic and anaerobic coverage (49). There are several acceptable regimens

including piperacillin–tazobactam, or clindamycin with ceftriaxone, or

metronidazole with ceftriaxone, or ertapenem, or ticarcillin–clavulanate, or

aztreonam with clindamycin (for those who have a penicillin allergy) (49).

Parenteral antibiotics can be transitioned to oral therapy when clinical

improvement is noted (WBC improving, afebrile for 48 hours, and abscess is

shrinking). Oral antibiotics should be tailored to culture results.

Necrotizing Fasciitis

Necrotizing fasciitis is an uncommon infectious disorder, affecting roughly

1,000 patients per year (50). This disease process is characterized by a rapidly

progressive bacterial infection that involves the subcutaneous tissues and fascia

while characteristically sparing underlying muscle. Systemic toxicity is a frequent

feature of this disease, as manifested by the presence of dehydration, septic shock,

disseminated intravascular coagulation, and multiple organ system failure.

There are two forms of necrotizing fasciitis, type I and type II. Type I

1330infections are caused by mixed aerobic and anaerobic bacteria and occur in

patients with risk factors such as recent surgery, diabetes, or immune

compromise. Type II infections are a result of β-hemolytic streptococci and can

occur in healthy patients who have a predisposing injury, such as a skin laceration

(51). Bacterial enzymes such as hyaluronidase and lipase released in the

subcutaneous space destroy the fascia and adipose tissue and induce a liquefactive

necrosis. Noninflammatory intravascular coagulation or thrombosis subsequently

occurs. Intravascular coagulation results in ischemia and necrosis of the

subcutaneous tissues and skin. Subcutaneous spread of up to 1 inch per hour can

be seen, often with little effect on the overlying skin (52). Late in the course of

the infection, destruction of the superficial nerves produces anesthesia in the

involved skin. The release of bacteria and bacterial toxins into the systemic

circulation can cause septic shock, acid–base disturbances, and multiple organ

impairment.

The diagnosis is often difficult to make early in the disease course. Most

patients with necrotizing fasciitis develop erythema, edema, and pain, which

in the early stages of the disease are disproportionately greater than that

expected from the degree of cellulitis present and characteristically extend

beyond the border of erythema (51). Late in the course of the infection, the

involved skin may be anesthetized secondary to necrosis of superficial nerves.

Temperature abnormalities, both hyperthermia and hypothermia, are concomitant

with the release of bacterial toxins and with bacterial sepsis (52). The involved

skin is initially tender, erythematous, and warm. Edema develops, and the

erythema spreads diffusely, fading into normal skin, characteristically without

distinct margins or induration. Subcutaneous microvascular thrombosis induces

ischemia in the skin, which becomes cyanotic and blistered. As necrosis develops,

the skin becomes gangrenous and may slough spontaneously (52). Most patients

will have leukocytosis and acid–base abnormalities. Subcutaneous gas may

develop, which can be identified by palpation and radiography. The finding of

subcutaneous gas by radiography is often indicative of clostridial infection,

although it is not a specific finding and may be caused by other organisms. These

organisms include Enterobacter, Pseudomonas, anaerobic streptococci, and

Bacteroides, which, unlike clostridial infections, spare the muscles underlying the

affected area.

Successful management of necrotizing fasciitis involves early recognition,

immediate initiation of resuscitative measures (including correction of fluid,

acid–base, electrolyte, and hematologic abnormalities), aggressive surgical

debridement and redebridement as necessary, and broad-spectrum

intravenous antibiotic therapy (52). During surgery, the incision should be

made through the infected tissue down to the fascia. An ability to undermine the

1331skin and subcutaneous tissues with digital palpation often will confirm the

diagnosis. Excision of infected, necrotic tissue should extend toward the

periphery of the affected tissue until well-vascularized, healthy, resistant tissue is

reached at all margins. The remaining affected tissue must be excised. The wound

can be packed and sequentially debrided as necessary until healthy tissue is

displayed at all margins. Hyperbaric oxygen therapy may be of some benefit,

particularly in patients for whom culture results are positive for anaerobic

organisms (53). Retrospective nonrandomized studies demonstrated that the

addition of hyperbaric oxygen therapy to surgical debridement and antimicrobial

therapy appear to significantly decrease wound morbidity and overall mortality in

patients with necrotizing fasciitis (53). After the initial resuscitative efforts and

surgical debridement, the primary concern is the management of the open wound.

Allograft and xenograft skin can be used to cover open wounds, thus decreasing

heat and evaporative water loss.

Significant improvement in wound healing may be aided by a vacuum-assisted

closure (VAC) method that uses a subatmospheric pressure technique (54). In

situations in which spontaneous closure is not likely, the VAC device may permit

the development of a suitable granulation bed and prepare the tissue for graft

placement, thereby increasing the probability of graft survival. Skin flaps can be

mobilized to help cover open wounds when the infection resolves and granulation

begins.

Postoperative Gastrointestinal Complications

Ileus

Following open abdominal or pelvic surgery, most patients experience some

degree of intestinal ileus. The exact mechanism by which this arrest and

disorganization of gastrointestinal motility occurs is unknown, but it appears to be

associated with the opening of the peritoneal cavity and is aggravated by

manipulation of the intestinal tract, intestinal resection, and prolonged surgical

procedures. Infection, peritonitis, opioids, and electrolyte disturbances may result

in ileus. For most patients undergoing common gynecologic operations, the

degree of ileus is minimal, and gastrointestinal function returns relatively rapidly,

allowing the resumption of oral intake within 12 hours of surgery. Ileus is

reported in 3% of patients undergoing total abdominal hysterectomy (55). Patients

who have intolerance for oral intake, persistently diminished bowel sounds,

abdominal distension, absence of flatus, nausea, or vomiting require further

evaluation. Patients with symptoms of ileus or small bowel obstruction who

underwent minimally invasive surgery are a different matter. Minimally

invasive surgery should result in a daily improvement in GI function. An

1332“ileus” in the case of minimally invasive surgery more likely represents GI

injury, which should be evaluated immediately with a CT scan using GI

contrast.

Ileus is usually manifested by abdominal distension with nausea and/or

vomiting and should be evaluated by physical examination. Pertinent points of the

abdominal examination include assessment of the quality of bowel sounds and

palpation in search of distension, tympany, masses, diffuse tenderness, or

rebound. The possibility that the patient’s signs and symptoms may be associated

with a more serious intestinal obstruction or intestinal complication (such as a

perforation) must be considered. Pelvic examination should be performed to

evaluate the possibility of a pelvic abscess or hematoma that may contribute to the

ileus. Abdominal radiography to evaluate the abdomen in the flat (supine)

position and in the upright position will aid in the diagnosis of an ileus. The

most common radiographic findings include dilated loops of small and large

bowel and air–fluid levels while the patient is in the upright position. Air

should be present in the colon or rectum. In the postoperative gynecology

patient, especially in the upright position, the abdominal x-ray may show free air.

This common finding following surgery lasts 7 to 10 days in some instances and

is not necessarily indicative of a perforated viscus in most patients.

The initial management of a postoperative ileus is aimed at gastrointestinal

tract decompression and maintenance of appropriate intravenous replacement

fluids and electrolytes.

1. The patient should be made NPO status (nothing by mouth) with

intravenous (IV) fluids and electrolytes. If nausea and vomiting persist, a

nasogastric tube should be used to evacuate the stomach of its fluid and

gaseous contents. Continued nasogastric suction removes swallowed air,

which is the most common source of gas in the small bowel.

2. Laboratory evaluation should include CBC, serum electrolytes, creatinine

and blood urea nitrogen (BUN), and liver functions.

3. Fluid and electrolyte replacement must be adequate to keep the patient

well hydrated and in metabolic balance. Significant amounts of third-space

fluid loss occur in the bowel wall, the bowel lumen, and the peritoneal cavity

during the acute episode. Gastrointestinal fluid losses from the stomach may

lead to metabolic alkalosis and depletion of other electrolytes. Careful

monitoring of serum chemistry levels and appropriate replacement are

necessary.

4. Most cases of severe ileus begin to improve over a period of several days.

This improvement is recognizable by a reduction in the abdominal distension,

return of normal bowel sounds, and passage of flatus or stool. Repeat

1333abdominal radiographs should be obtained as necessary for further monitoring.

5. When the gastrointestinal tract function appears to have returned to

normal, the nasogastric tube may be removed and a liquid diet instituted.

If a patient shows no evidence of improvement during the first 48 to 72

hours of medical management, other causes of ileus should be sought. Such

cases may include ureteral injury, peritonitis from pelvic infection, unrecognized

gastrointestinal tract injury with peritoneal spill, or fluid and electrolyte

abnormalities such as hypokalemia. With persistent ileus, the use of water-soluble

upper gastrointestinal contrast studies (CT scan with oral contrast) may assist in

distinguishing an ileus from a small bowel obstruction. In the latter case, a

“transition point” is usually identified.

Small Bowel Obstruction

Obstruction of the small bowel following major, open gynecologic surgery

occurs in approximately 1% to 2% of patients (56,57). The most common

cause of small bowel obstruction is adhesions to the operative site. If the small

bowel becomes adherent in a twisted position, partial or complete obstruction

may result. Less common causes of postoperative small bowel obstruction include

entrapment of the small bowel into an incisional hernia or an unrecognized defect

in the small bowel or large bowel mesentery (internal hernia). [7] Early in its

clinical course, a postoperative small bowel obstruction may exhibit signs

and symptoms similar to those of ileus. Initial conservative management as

outlined for the treatment of ileus is appropriate. Because of the potential for

mesenteric vascular occlusion and resulting ischemia or perforation, worsening

symptoms of abdominal pain, progressive distension, fever, leukocytosis, or

acidosis should be evaluated carefully because immediate surgery may be

required.

In most cases of small bowel obstructions following gynecologic surgery, the

obstruction is only partial and the symptoms usually resolve with conservative

management.

1. Further evaluation after several days of conservative management may be

necessary. Evaluation of the small intestines with an upper gastrointestinal

study, or a CT scan with small bowel assessment is appropriate. In most cases,

complete obstruction is not documented, although a narrowing (“transition

point”) or tethering of the segment of small bowel may indicate the site of the

problem.

2. Laboratory evaluation should include CBC, serum electrolytes, and serum

lactate.

13343. Further conservative management with nasogastric decompression and

intravenous fluid and electrolyte replacement may allow time for bowel

wall edema or torsion of the mesentery to resolve.

4. Serial and radiographic examinations should be performed to exclude

possible bowel ischemia or perforation.

5. If resolution is prolonged and the patient’s nutritional status is marginal,

the use of TPN may be necessary.

6. Conservative medical management of postoperative small bowel

obstruction usually results in complete resolution. If persistent evidence of

small bowel obstruction remains after full evaluation and an adequate trial of

medical management, exploratory laparotomy may be necessary to manage the

obstruction. Risk factors for operative management include persistent

abdominal pain, distension, fever at 48 hours, and CT findings of high-grade

obstruction (58). In most cases, lysis of adhesions is all that is required,

although a segment of small bowel that is badly damaged or extensively

sclerosed from adhesions may require resection and reanastomosis.

Colonic Obstruction

Postoperative colonic obstruction following surgery for most gynecologic

conditions is exceedingly rare. It is usually associated with a pelvic malignancy,

which in most cases was known at the time of the initial operation. Advanced

ovarian carcinoma is the most common cause of colonic obstruction in

postoperative gynecologic surgery patients, and it is caused by extrinsic

impingement on the colon by the pelvic malignancy. Intrinsic colonic lesions may

be undetected, especially in a patient with some other benign gynecologic

condition. When colonic obstruction is manifested by abdominal distension and

abdominal radiography reveals a dilated colon and enlarging cecum, further

evaluation of the large bowel is required by water-soluble enema or colonoscopy.

Dilation of the cecum to more than 10 to 12 cm in diameter as viewed by

abdominal radiography requires immediate evaluation and surgical

decompression by performing colectomy or colostomy. Surgery should be

performed as soon as the obstruction is documented. Conservative management

of colonic obstruction is not appropriate because the complication of colonic

perforation has an exceedingly high mortality rate. In patients who are too ill to

undergo surgery, the interventional radiologist may be able to place a cecostomy

tube or the gastroenterologist may place a colonic stent (59).

Diarrhea

Episodes of diarrhea often occur following abdominal and pelvic surgery as the

gastrointestinal tract returns to its normal function and motility. Prolonged and

1335multiple episodes may represent a pathologic process such as impending small

bowel obstruction, colonic obstruction, or pseudomembranous colitis. Excessive

amounts of diarrhea should be evaluated by abdominal radiography and stool

samples tested for the presence of ova and parasites, bacterial culture, and C.

difficile toxin. Proctoscopy and colonoscopy may be advisable in severe cases.

Evidence of intestinal obstruction should be managed as outlined. Infectious

causes of diarrhea should be managed with the appropriate antibiotics and fluid

and electrolyte replacement. C. difficile–associated pseudomembranous colitis

may result from exposure to any antibiotic. Discontinuation of these antibiotics

(unless they are needed to treat another severe infection) is advisable, along with

the institution of appropriate therapy. Oral vancomycin is the preferred initial

treatment for severe C. difficile colitis (oral metronidazole is an alternative

treatment for moderate cases) (60).

Therapy should be continued until the diarrhea abates, and several weeks of

oral therapy may be required to obtain complete resolution of the

pseudomembranous colitis.

Fistula

Gastrointestinal fistulas are rare complications of gynecologic surgery. They are

most often associated with malignancy, prior radiation therapy, intestinal

resection with anastomosis, or surgical injury to the large or small bowel that was

improperly repaired or unrecognized. Signs and symptoms of gastrointestinal

fistula are often similar to those of small bowel obstruction or ileus, except that

fever is usually a more prominent component of the patient’s symptoms. When

fever is associated with gastrointestinal dysfunction postoperatively, evaluation

should include early assessment of the gastrointestinal tract to confirm its

continuity. When fistula is suspected, the use of water-soluble gastrointestinal

contrast material is advised to avoid the complication of barium peritonitis.

Evaluation with abdominal pelvic CT scan may assist in identification of a fistula

and associated abscess. Recognition of an intraperitoneal gastrointestinal leak

or fistula formation usually requires immediate surgery, unless the fistula

has drained spontaneously through the abdominal wall or vaginal cuff.

An enterocutaneous fistula arising from the small bowel and draining

spontaneously through the abdominal incision (with no evidence of

intraperitoneal infection) may be managed successfully with medical

therapy. Therapy should include nasogastric decompression, replacement of

intravenous fluids, TPN, and appropriate antibiotics to treat an associated mixed

bacterial infection. If the infection is under control and there are no other signs of

peritonitis, the surgeon may consider allowing potential resolution of the fistula

over a period of up to 2 weeks. Some authors suggested the use of somatostatin to

1336decrease intestinal tract secretion and allow earlier healing of the fistula. In some

cases, the fistula will close spontaneously with this mode of management. If the

enterocutaneous fistula does not close with conservative medical management,

surgical correction with resection, bypass, or reanastomosis will be necessary.

A rectovaginal fistula that occurs following gynecologic surgery is usually the

result of surgical trauma that may have been predisposed by the presence of

extensive adhesions and scarring in the rectovaginal septum associated with

endometriosis, pelvic inflammatory disease, or pelvic malignancy. A small

rectovaginal fistula may be managed with a conservative medical approach,

in the hope that decreasing the fecal stream will allow closure of the fistula. A

small fistula that allows continence except for an occasional leak of flatus may be

managed conservatively until the inflammatory process in the pelvis resolves. At

that point, usually several months later, correction of the fistula is appropriate.

Large rectovaginal fistulas for which there is no hope of spontaneous closure are

best managed by performing a diverting colostomy followed by repair of the

fistula after inflammation resolves. After the fistula closure is healed and deemed

successful, the colostomy may be reversed.

Thromboembolism

Risk Factors

[8] DVT and pulmonary embolism are largely preventable, yet significant,

complications in postoperative patients. The magnitude of this problem is

relevant to the gynecologist, because 40% of all deaths following gynecologic

surgery are directly attributed to pulmonary emboli, and it is the most

frequent cause of postoperative death in patients with uterine or cervical

carcinoma (61,62).

The causal factors of venous thrombosis were first proposed by Virchow in

1858 and include a hypercoagulable state, venous stasis, and vessel endothelial

injury. Risk factors include major surgery; advanced age; nonwhite race;

malignancy; history of DVT, lower extremity edema, or venous stasis changes;

presence of varicose veins; being overweight; a history of radiation therapy; and

hypercoagulable states, such as factor V Leiden, pregnancy, and use of oral

contraceptives, estrogens, or tamoxifen. Intraoperative factors associated with

postoperative DVT include increased anesthesia time, increased blood loss, and

the need for transfusion in the operating room. It is important to recognize these

risk factors and to provide the appropriate level of venous thrombosis prophylaxis

(63–66). Caprini has created a scoring system that attributes different “points” for

each risk factor that when totaled define a level of risk (67). The American

College of Chest Physicians (ACCP) has embraced the Caprini Risk Score to

1337define the recommended intensity of prophylaxis for an individual patient (68).

The levels of thromboembolism risk are listed in Table 25-7.

Prophylactic Methods

A number of prophylactic methods significantly reduced the incidence of DVT,

and a few studies included a large enough patient population to show a reduction

in fatal pulmonary emboli (69). The ideal prophylactic method would be

effective, free of significant side effects, well accepted by the patient and nursing

staff, widely applicable to most patients, and inexpensive.

Low-Dose Heparin

The use of small doses of subcutaneously administered heparin for the prevention

of DVT and pulmonary embolism is the most widely studied of all prophylactic

methods. More than 25 controlled trials demonstrated that heparin given

subcutaneously 2 hours preoperatively and every 8 to 12 hours

postoperatively is effective in reducing the incidence of DVT. The value of

low-dose heparin in preventing fatal pulmonary emboli was established by a

randomized, controlled multicenter international trial, which demonstrated a

significant reduction in fatal postoperative pulmonary emboli in general

surgery patients receiving low-dose heparin every 8 hours postoperatively

(69). Trials of low-dose heparin in gynecologic surgery patients showed a

significant reduction in postoperative DVT.

Although low-dose heparin is considered to have no measurable effect on

coagulation, most large series noted an increase in the bleeding complication rate,

especially a higher incidence of wound hematoma (70). Although relatively rare,

thrombocytopenia is associated with low-dose heparin use and was found in 6%

of patients after gynecologic surgery (70). If patients remain on low-dose heparin

for more than 4 days, it is reasonable to check their platelet count to assess the

possibility of heparin-induced thrombocytopenia.

Low–Molecular-Weight Heparin

Low–molecular-weight heparins (LMWH) are fragments of heparin that vary in

size from 4,500 to 6,500 Da. When compared with unfractionated heparin,

LMWH have more anti-Xa and less antithrombin activity, leading to less effect on

partial thromboplastin time and possibly leading to fewer bleeding complications

(71). An increased half-life of 4 hours results in increased bioavailability when

compared with unfractionated heparin. The increase in half-life of LMWH allows

the convenience of once-a-day dosing.

Table 25-7 Thromboembolism Risk Stratification (Caprini Risk Score)

13381 Point

Age 41–60 yrs

Minor surgery

BMI >25 kg/m2

Swollen legs

Varicose veins

Pregnancy or postpartum state

History of unexplained or recurrent abortions (>3)

Oral contraceptive use or hormone replacement

Sepsis (<1 mo)

Serious lung disease, including pneumonia (<1 mo)

Abnormal pulmonary function

Congestive heart failure

History of inflammatory bowel disease

Medical patient at bed rest

2 Points

Age 61–74 yrs

Major open surgery (>45 min)

Laparoscopic surgery (>45 min)

Malignancy

Confined to bed (>72 hrs)

Immobilizing cast

Central venous access

3 Points

Age >74 yrs

History of VTE

Family history of VTE

Congenital or acquired thrombophilias (i.e., Factor V Leiden, anticardiolipin antibodies,

elevated serum homocysteine, prothrombin 20210A)

Heparin-induced thrombocytopenia

5 Points

Stroke (<1 mo)

Elective arthroplasty

Hip, pelvis, or leg fracture

Acute spinal cord injury (<1 mo)

Randomized controlled trials have compared LMWH with unfractionated

heparin in patients undergoing gynecologic surgery. In all studies, there was

1339a similar incidence of DVT. Bleeding complications were similar between the

unfractionated heparin and LMWH groups (72). A meta-analysis of general

surgery and gynecologic surgery patients from 32 trials indicated that daily

LMWH administration is as effective as unfractionated heparin in DVT

prophylaxis without any difference in hemorrhagic complications (73).

Mechanical Methods

Stasis in the veins of the legs occurs while the patient is undergoing surgery and

continues postoperatively for varying lengths of time. Stasis occurring in the

capacitance veins of the calf during surgery, plus the hypercoagulable state

induced by surgery, are the prime factors contributing to the development of acute

postoperative DVT. Prospective studies of the natural history of postoperative

DVT showed that the calf veins are the predominant site of thrombi and that most

thrombi develop within 24 hours of surgery (74).

Although probably of only modest benefit, reduction of stasis by short

preoperative hospital stays and early postoperative ambulation should be

encouraged for all patients. Elevation of the foot of the bed, raising the calf above

heart level, allows gravity to drain the calf veins and should further reduce stasis.

Graduated Compression Stockings

Controlled studies of graduated compression stockings are limited but do suggest

modest benefit when they are carefully fitted (75,76). Poorly fitted stockings may

be hazardous to some patients who develop a tourniquet effect at the knee or midthigh (62). Variations in human anatomy do not allow perfect fit of all patients to

available stocking sizes. The simplicity of graduated compression stockings and

the absence of significant side effects are probably the two most important

reasons that they are often included in routine postoperative care. Compared to

thigh length stockings, calf-high stockings appear to offer the same degree of

venous thromboembolism protection (77).

Intermittent Pneumatic Compression

The largest body of literature dealing with the reduction of postoperative venous

stasis deals with intermittent compression of the leg by pneumatically inflated

sleeves placed around the calf or leg during intra- and postoperative periods. [8]

Various pneumatic compression devices and leg sleeve designs are available, and

the literature has not demonstrated superiority of one system over another. Calf

compression, during and after gynecologic surgery, significantly reduces the

incidence of DVT on a level similar to that of low-dose heparin. In addition to

increasing venous flow and pulsatile emptying of the calf veins, intermittent

pneumatic compression (IPC) appears to augment endogenous fibrinolysis, which

1340may result in lysis of very early thrombi before they become clinically significant

(78).

The duration of postoperative external pneumatic compression differed in

various trials. External pneumatic compression may be effective when used in the

operating room and for the first 24 hours postoperatively in patients at moderate

risk who will ambulate on the first postoperative day (78,79).

External pneumatic compression used in patients undergoing major

surgery for gynecologic malignancy reduced the incidence of postoperative

venous thromboembolic complications by nearly threefold, but only if calf

compression was applied intraoperatively and for the first 5 postoperative days

(80,81). Patients with gynecologic malignancies may remain at risk for a longer

period than general surgical patients because of stasis and hypercoagulable states;

therefore, these patients appear to benefit from longer use of IPC.

Table 25-8 VTE Prophylactic Regimens Based on an Individual’s Caprini Risk Score

Score

0: Early ambulation.

1–2: Low risk: Graded compression stockings and/or intermittent pneumatic

compression.

3–4: Moderate risk: Intermittent pneumatic compression or low-dose heparin or low–

molecular-weight heparin.

>4: High risk: Intermittent pneumatic compression and low-dose heparin or low–

molecular-weight heparin. Consider prolonged prophylaxis for 28 days.

Intermittent pneumatic leg compression has no significant side effects or risks

and is considered slightly more cost-effective when compared with

pharmacologic methods of prophylaxis (82). Compliance in wearing the leg

compression while in bed is of utmost importance, and the patient and nursing

staff should be educated to the proper regimen for maximum benefit (83).

[8] Using the Caprini Risk Score, the ACCP guidelines recommend different

prophylactic regimens based on an individual patient’s risk score (Table 25-8).

Continued prophylaxis for 28 days with low–molecular-weight heparin has

been shown to significantly reduce the incidence of VTE at 30 and 90 days

postoperatively in patients undergoing “open” abdominal procedures for

malignancy (84). There have been no trials specific to gynecologic surgery that

have evaluated prolonged prophylaxis, although this strategy seems reasonable,

especially for gynecologic cancer patients undergoing “open” procedures.

1341Guidelines do not distinguish the risks of minimally invasive gynecologic

surgery (MIGS) from “open surgery.” Some have argued that MIGSs are much

lower-risk surgical procedures and no prophylaxis is necessary (85). Others have

shown that despite the use of IPC, there remains approximately 2% incidence of

VTE in patients undergoing more complex gynecologic surgery (86).

Management of Postoperative Deep Venous Thrombosis and Pulmonary Embolism

Because pulmonary embolism is the leading cause of death following

gynecologic surgical procedures, identification of high-risk patients and the

use of prophylactic venous thromboembolism regimens are essential parts of

management (61,62,87).

The early recognition of DVT and pulmonary embolism and immediate

treatment are critical. Most pulmonary emboli arise from the deep venous

system of the leg following gynecologic surgery; the pelvic veins are also a

known source of fatal pulmonary emboli.

The signs and symptoms of DVT of the lower extremities include pain, edema,

erythema, and prominent vascular pattern of the superficial veins. These signs and

symptoms are relatively nonspecific; 50% to 80% of patients with these

symptoms will not have DVT (88). Conversely, approximately 80% of patients

with symptomatic pulmonary emboli have no signs or symptoms of thrombosis in

the lower extremities (89). Because of the lack of specificity when signs and

symptoms are recognized, additional tests should be performed to establish the

diagnosis of DVT.

Diagnosis

Doppler Ultrasound

B-mode duplex Doppler imaging is the most common technique for the

diagnosis of symptomatic venous thrombosis, especially when it arises in the

proximal lower extremity. With duplex Doppler imaging, the femoral vein can

be visualized and clots may be seen directly (90). Compression of the vein with

the ultrasound probe tip allows assessment of venous collapsibility; the presence

of a thrombus diminishes vein wall collapsibility. Doppler imaging is less

accurate when evaluating the calf and the pelvic veins.

Venography

Although venography is considered the “gold standard” technique for diagnosis of

DVT, other diagnostic studies are accurate when performed by a skilled

technologist and, in nearly all patients, may replace the need for contrast

venography. Venography is moderately uncomfortable, requires the intravenous

injection of a contrast material that may cause allergic reaction or renal injury,

1342and may result in phelebitis in approximately 1% of patients (91). If the results of

noninvasive imaging are normal or inconclusive and the clinician remains

concerned given clinical symptoms, venography should be performed to obtain a

definitive answer.

Magnetic Resonance Venography

In addition to having a sensitivity and specificity comparable to venography,

magnetic resonance venography (MRV) may detect thrombi in pelvic veins that

are not imaged by venography (92). The primary drawback to MRV is the time

involved in examining the lower extremity and pelvis and the expense of this

technology.

Treatment

Deep Venous Thrombosis

The treatment of postoperative DVT requires the immediate institution of

anticoagulant therapy. Treatment may be with either unfractionated heparin

or LMWH, followed by 3 to 6 months of oral anticoagulant therapy with

warfarin (Coumadin). The new oral antithrombins and factor Xa inhibitors are

options for secondary prophylaxis.

Unfractionated Heparin

After venous thromboembolism is diagnosed, unfractionated heparin (or low–

molecular-weight heparins, see below) should be initiated to prevent proximal

propagation of the thrombus and allow physiologic thrombolytic pathways to

dissolve the clot. An initial bolus of 80 U/kg is given intravenously, followed by a

continuous infusion of 1,000 to 2,000 U/hr (18 U/kg/hr). Heparin dosage is

adjusted to maintain APTT levels at a therapeutic level 1.5 to 2.5 times the

control value. Initial APTT should be measured after 6 hours of heparin

administration and the dose adjusted as necessary. Patients having subtherapeutic

APTT levels in the first 24 hours have a 15-fold increased risk of recurrent VTE

when compared to patients who are adequately anticoagulated. Patients should be

managed aggressively using intravenous heparin to achieve prompt

anticoagulation. A weight-based nomogram is helpful in achieving a therapeutic

APTT level (93).

Low–Molecular-Weight Heparin

Two LMWH preparations (enoxaparin and dalteparin) were effective in the

treatment of venous thromboembolism and have a cost-effective advantage

over intravenous heparin in that they may be administered in the outpatient

setting. The dosages used in treatment of thromboembolism are unique and

1343weight adjusted according to each LMWH preparation. Because LMWH has a

minimal effect on APTT, serial laboratory monitoring of these levels is not

necessary. Similarly, monitoring of anti-Xa activity (except in difficult cases or

those with renal impairment) is not of significant benefit in a dose adjustment of

LMWH. The increased bioavailability associated with LMWH allows for twice-aday dosing, potentially making outpatient management an option for a subset of

patients. A meta-analysis involving more than 4,000 patients from 22 trials

suggests that LMWH is more effective, safer, and less costly when compared

with unfractionated heparin in preventing recurrent thromboembolism (94).

Following initial treatment with either unfractionated heparin (UFH) or

LMWH, continued secondary prophylaxis/treatment is usually achieved by an

oral agent. In patients with active cancer, continued treatment with LMWH has

been found to be superior (95).

Oral anticoagulant (warfarin) administration may be started on the first

day of heparin infusion. The international normalized ration (INR) should be

monitored daily until a therapeutic level is achieved (2 to 3 times normal

value). The change in the INR resulting from warfarin administration often

precedes the anticoagulant effect by approximately 2 days, during which time

low-protein C levels are associated with a transient hypercoagulable state.

Therefore, heparin should be administered until the INR is maintained in a

therapeutic range for at least 2 days, confirming proper warfarin dose.

Intravenous heparin may be discontinued in 5 days if an adequate INR level is

established.

Oral inhibitors of thrombin or factor Xa may be used in place of warfarin

(96). Advantages of these agents include faster uptake and there is no need for

ongoing monitoring. Disadvantages include the fact that there is no agent to

“reverse” the anticoagulant effect (in the case of urgent surgery or bleeding), cost,

and the inconvenience of twice-a-day dosing (patient compliance). Safety and

efficacy in pregnancy have not been established. Therapeutic dosing of these

different agents is shown in Table 25-9.

Pulmonary Embolism

Many of the signs and symptoms of pulmonary embolism are associated with

other, more commonly occurring pulmonary complications following surgery.

The classic findings of pleuritic chest pain, hemoptysis, shortness of breath,

tachycardia, and tachypnea should alert the physician to the possibility of a

pulmonary embolism. Many times the signs are subtle and may be demonstrated

only by a persistent tachycardia or a slight elevation in the respiratory rate.

Patients suspected of pulmonary embolism should be evaluated initially by chest

x-ray, electrocardiography (ECG), and arterial blood gas assessment. Any

1344evidence of abnormality should be further evaluated by a spiral CT scan of the

chest or a ventilation–perfusion lung scan. A high percentage of lung scans may

be interpreted as “indeterminate.” In this setting, careful clinical evaluation and

judgment are required to decide whether pulmonary arteriography should be

performed to document or exclude the presence of a pulmonary embolism.

Table 25-9 Therapeutic Dosing of Non–Vitamin K Antagonist Oral Anticoagulants

Dosing—Typical initial doses in those with normal renal function are:

Rivaroxaban, 15 mg twice daily (for the first 3 weeks)

Apixaban, 10 mg twice daily (for first 7 days)

Edoxaban, 60 mg once daily (and 30 mg once daily in patients with a creatinine 30–

50 mL/min or a body weight below 60 kg)

Dabigatran, 150 mg twice daily

The treatment of pulmonary embolism is as follows:

1. Immediate anticoagulant therapy, identical to that outlined for the treatment of

DVT, should be initiated.

2. Respiratory support, including oxygen and bronchodilators, and an intensive

care setting, if necessary.

3. Although massive pulmonary emboli are usually quickly fatal, rarely

pulmonary embolectomy is successful.

4. Pulmonary artery catheterization (PAC) with the administration of

thrombolytic agents bears further evaluation and may be important in patients

with massive pulmonary embolism.

5. Vena cava interruption (filter) may be necessary in situations in which

anticoagulant therapy is ineffective in the prevention of rethrombosis and

repeated embolization from the lower extremities or pelvis. A vena cava filter

may be inserted percutaneously above the level of the thrombus and caudad to

the renal veins. In most cases, anticoagulant therapy is sufficient to prevent

repeat thrombosis and embolism and to allow the patient’s own endogenous

thrombolytic mechanisms to lyse the pulmonary embolus.

ENHANCED RECOVERY AFTER SURGERY (ERAS)

Rationale for ERAS

As surgical technology has been evolving, so have concepts for perioperative

1345care. Traditional perioperative management has included nothing by mouth (nil

per os, NPO) after midnight, liberal intraoperative fluids, and dependence on

opioid pain medications for postoperative pain. However, in the 1990s, European

physicians began to better understand the association between perioperative

physiologic stress and postoperative morbidity (97). Stress increases

immunosuppression, hypoxia, insulin resistance, and catabolism (98). An optimal

perioperative management strategy would target each of these factors to minimize

morbidity, and in the early 2000s, the initial enhanced recovery protocol was

crafted (99). ERAS was first embraced by colorectal physicians but has been

spreading throughout disciplines (100).

[6] Enhanced recovery protocols are comprehensive and address elements from

the patient’s initial arrival in the office with her chief complaint until she leaves

the hospital following surgery. Preoperatively, its components include

preoperative counseling, carbohydrate loading, the use of liberal antiemetics

(including preoperative steroids), and avoidance of prolonged fasting.

Intraoperatively, there is a focus on normothermia, balanced fluids, avoidance of

drains, a minimally invasive approach, multimodal pain medications, including

regional anesthesia when able. Following surgery, the aims include early

mobilization and feeding, prompt discontinuation of intravenous fluids,

multimodal pain medications, timely discontinuation of urinary catheters, and

regional anesthesia (99–101). Some important components of an ERAS protocol

will be highlighted below.

Perioperative Diet

With respect to perioperative nutrition, there are important considerations for an

ERAS protocol. There is no evidence regarding prolonged preoperative fasting

and aspiration risk (102), and fasting is known to increase insulin resistance

which increases perioperative morbidity. Given the paucity of data regarding

“NPO after midnight,” eating ad lib until 6 hours preoperatively followed by

consumption of clear liquids until 2 hours preoperatively is recommended by the

American Society of Anesthesiology (103). Another unique feature of an ERAS

protocol is preoperative carbohydrate loading. Drinks containing 12.5%

maltodextrin given 2 to 3 hours preoperatively have been shown to improve

insulin resistance and decrease hospital stay (104,105). Practically, the

recommendation is for the patient to ingest a 12-oz Gatorade 2 to 3 hours

preoperatively. Postoperatively the aim is early feeding, including for patients

who have undergone colorectal surgery, as there has not been an association in

increased anastomotic leak (106). With respect to postoperative fluids, the goal is

to discontinue as soon as possible or at the latest, on postoperative day 1.

1346Bowel Preparation

While some surgeons may recommend a bowel preparation preoperatively, there

has not been definitive evidence of benefit but there are areas of concern with

their use (107). Bowel preparations are implicated in patient dissatisfaction,

electrolyte disturbances, and dehydration. Therefore, ERAS protocols recommend

avoidance of bowel preparation.

Management of Nausea and Vomiting

Postoperative nausea and vomiting are unfortunately common in patients. A

multimodal, prophylactic approach utilizing medications in at least two antiemetic

classes has been employed. Available agents include NK-1 antagonists,

corticosteroids, 5-HT3 antagonists, butyrophenones, antihistamines,

phenothiazines, and anticholinergics (108). If nausea is present postoperatively,

an antiemetic of a different class should be used. A scopolamine patch can be

considered as it has been shown to improve postoperative nausea and vomiting

(109).

Perioperative Fluids

General Principles

Fluid and Electrolyte Maintenance Requirements

The body adjusts to higher and lower volumes of intake by changes in plasma

tonicity. Alterations in plasma tonicity induce adjustments in circulating

antidiuretic hormone (ADH) levels, which ultimately regulate the amount of

water retained in the distal tubule of the kidney. In the preoperative and the early

postoperative periods, it is usually necessary to replace only sodium and

potassium. Chloride is automatically replaced, concomitant with sodium and

potassium, because chloride is the usual anion used to balance sodium and

potassium in electrolyte solutions. There are various commercially available

solutions containing 40 mmol of sodium chloride, with smaller amounts of

potassium, calcium, and magnesium, designed to meet the requirements of a

patient who is receiving 3 L of intravenous fluids per day. The daily

requirement can be met by any combination of intravenous fluids. For

example, 2 L of D5 (5% dextrose)/0.45 normal saline (7 mEq sodium chloride

each), supplemented with 20 mEq of potassium chloride, followed by 1 L of

D5W (5% dextrose in water) with 20 mEq of potassium chloride, would

suffice.

Fluid and Electrolyte Replacement

1347Fluid and electrolyte losses beyond the daily average must be replaced by

appropriate solutions. The choice of solutions for replacement depends on the

composition of the fluids lost. Often, it is difficult to measure free water loss,

particularly in patients who have high losses from the lungs, skin, or the

gastrointestinal tract. Weighing these patients daily can be very useful. Up to 300

g of weight loss daily can be attributable to weight loss from catabolism of

protein and fat in the patient who is taking nothing by mouth (110). Any loss

beyond this level represents fluid loss, which should be replaced accordingly.

Patients with a high fever can have increased pulmonary and skin loss of

free water, sometimes in excess of 2 to 3 L per day. These losses should be

replaced with free water in the form of D5W. Perspiration typically has one-third

the osmolarity of plasma and can be replaced with D5W or, if the loss is

excessive, with D5/0.25 normal saline.

Patients with acute blood loss need replacement with appropriate isotonic

fluid or blood or both. There is a wide range of plasma volume expanders,

including albumin, dextran, and hetastarch solutions, that contain large molecularweight particles (<50 kDa molecular weight). These particles are slow to exit the

intravascular space, and about one-half of the particles remain after 24 hours.

Controversy exists over the ideal strategy for intravascular volume replacement

(111). A systematic review of 25 randomized clinical trials demonstrated

preserved renal function and reduced intestinal edema in surgical patients

receiving hyperoncotic albumin solutions, as compared with control fluids (112).

Meta-analyses on the use of human albumin and crystalloids versus colloids in

fluid resuscitation did not show a benefit in mortality rates (113,114). Caution is

required in interpreting results from these pooled controlled trials because

mortality outcome was not the end point of most of the studies, and publication

bias is a limitation. Possible side effects with synthetic colloid solutions include

adverse effects on hemostasis, severe anaphylactic reactions, and impairment of

renal function (111). These solutions are expensive and for most cases, simple

replacement with 0.9 normal saline or lactated Ringer solution will suffice.

One-third of the volume of lactated Ringer solution or normal saline typically will

remain in the intravascular space and the remainder goes to the interstitium.

Appropriate replacement of gastrointestinal fluid loss depends on the

source of fluid loss in the gastrointestinal tract. Gastrointestinal secretions

beyond the stomach and up to the colon are typically isotonic with plasma, with

similar amounts of sodium, slightly lower amounts of chloride, slightly alkaline

pH, and more potassium (in the range of 10 to 20 mEq/L). Under normal

conditions, stool is hypotonic. However, under conditions of increased flow (i.e.,

severe diarrhea), stool contents are isotonic with a composition similar to that of

the small bowel contents. Gastric contents are typically hypotonic, with one-third

1348the sodium of plasma, increased amounts of hydrogen ion, and low pH.

Postoperative Fluid and Electrolyte Management

Several hormonal and physiologic alterations in the postoperative period may

complicate fluid and electrolyte management. The stress of surgery induces an

inappropriately high level of circulating ADH. Circulating aldosterone levels are

increased, especially if sustained episodes of hypotension occurred either intra- or

postoperatively. The elevated levels of circulating ADH and aldosterone make

postoperative patients prone to sodium and water retention.

Total body fluid postoperative volume may be altered significantly. First, 1

mL of free water is released for each gram of fat or tissue that is catabolized and,

in the postoperative period, several hundred milliliters of free water are released

daily from tissue breakdown, particularly in the patient who has undergone

extensive intra-abdominal dissection and who is restricted from ingesting food

and fluids by mouth. This free water is often retained in response to the altered

levels of ADH and aldosterone. Second, fluid retention is further enhanced by

third spacing, or sequestration of fluid in the surgical field. The development of

an ileus may result in an additional 1 to 3 L of fluid per day being

sequestered in the bowel lumen, bowel wall, and peritoneal cavity.

In contrast to renal sodium homeostasis, the kidney lacks the capacity for

retention of potassium. In the postoperative period, the kidneys continue to

excrete a minimum of 30 to 60 mEq/L of potassium daily, irrespective of the

serum potassium level and total body potassium stores (115). If this potassium

is not replaced, hypokalemia may develop. Tissue damage and catabolism during

the first postoperative day usually result in the release of sufficient intracellular

potassium to meet the daily requirements. Beyond the first postoperative day,

potassium supplementation is necessary.

Correct maintenance of fluid and electrolyte balance in the postoperative

period starts with the preoperative assessment, with emphasis on establishing

normal fluid and electrolyte parameters before surgery. Postoperatively, close

monitoring of daily weight, urine output, serum hematocrit, serum electrolytes,

and hemodynamic parameters will yield the necessary information to make

correct adjustments in crystalloid replacement. The normal daily fluid and

electrolyte requirements must be met and any unusual fluid and electrolyte losses,

including those from the gastrointestinal tract, lungs, or skin, must be

compensated. After the first few postoperative days, third-space fluid begins to

return to the intravascular space, and ADH and aldosterone levels revert to

normal. The excess fluid retained perioperatively is mobilized and excreted

through the kidneys, and exogenous fluid requirements decrease. Patients with

inadequate cardiovascular or renal reserve are prone to fluid overload during this

1349time of third-space reabsorption, especially if intravenous fluids are not

appropriately reduced.

The most common fluid and electrolyte disorder in the postoperative

period is fluid overload. Fluid excess can occur concomitantly with normal or

decreased serum sodium. Large amounts of isotonic fluids are usually infused

intraoperatively and postoperatively to maintain blood pressure and urine output.

Because the infused fluid is often isotonic with plasma, it will remain in the

extracellular space. Under such conditions, serum sodium will remain within

normal levels. Fluid excess with hypotonicity (decreased serum sodium) can

occur if large amounts of isotonic fluid losses (e.g., blood and gastrointestinal

tract) are inappropriately replaced with hypotonic fluids. The predisposition

toward retention of free water in the immediate postoperative period compounds

the problem. An increase in body weight occurs concomitantly with the fluid

expansion. In the patient who is not allowed anything by mouth, catabolism

should induce a daily weight loss as great as 300 g per day. The patient who is

gaining weight in excess of 150 g per day is in a state of fluid expansion. Simple

fluid restriction will correct the abnormality. When necessary, diuretics can be

used to increase urinary excretion.

States of fluid dehydration are uncommon but will occur in patients who have

large daily losses of fluid that are not replaced. Gastrointestinal losses should be

replaced with the appropriate fluids. Patients with high fevers should be

given appropriate free water replacement, because up to 2 L per day of free

water can be lost through perspiration and hyperventilation. Although these

increased losses are difficult to monitor, a reliable estimate can be obtained by

monitoring body weight.

Tenets of Fluid Management in ERAS

[3] The ideal fluid management for a perioperative patient is euvolemia. In order

to achieve this, ERAS protocols consist of minimizing crystalloids and

prioritizing the use of colloids and vasopressors when needed. When crystalloids

are used, balanced fluids (such as lactated Ringers) are preferred. Many times this

approach is goal-oriented, utilizing measurements of stroke volume to guide

resuscitation. There is evidence that patients have improved postoperative

outcomes when utilizing either hypovolemic or goal-oriented fluid paradigms

(116).

Postoperative Pain Management

General Principles

[4] Although satisfactory analgesia is easily achievable with available

1350methods, patients continue to suffer unnecessarily from postoperative pain.

Studies consistently show that 25% to 50% of patients suffer moderate to

severe pain in the postoperative period (117,118). There are several reasons for

the existing inadequacies in pain management. First, patient expectations of pain

relief are low and they are not aware of the extent of analgesia that they should

expect. Second, there is a lack of formal physician training in pain management.

Third, attitudes continue to be influenced by the common misconception that the

use of narcotics in the postoperative period results in opioid dependence.

Many effective pain control paradigms exist. The patient-controlled

analgesia (PCA) technique, which allows patients to self-administer small

doses of narcotic on demand, allows titration of measured boluses of narcotic

as needed to relieve pain. This technique can provide a more thorough

analgesia with maintenance of steady-state drug concentrations. These

devices eliminate the delay between the onset of pain and the administration

of analgesic agents, a common problem inherent with on-demand analgesic

orders in busy hospital wards. PCA has excellent patient acceptance. Compared

with conventional intramuscular injections, serum narcotic levels have

significantly lower variability in patients using PCA (119). Patients using PCA

have improved analgesia, a lower incidence of postoperative pulmonary

complications, and less confusion than those given intramuscular narcotics (120).

The total dose of narcotic used is lower with PCA than with conventional

intramuscular depot injection.

The use of PCA does not eliminate the adverse side effects of narcotics.

Potentially life-threatening respiratory depression is seen in as many as 0.5% of

patients using PCA. The use of a continuous narcotic infusion in addition to

demand dosing is associated with a fourfold increase in respiratory depression.

Elderly patients and those with pre-existing respiratory compromise are at risk for

respiratory depression (119).

In summary, use of PCA shortens the time between the onset of pain and

the administration of pain medication, provides more continuous access to

analgesics, and permits an overall steadier state of pain control.

Epidural and Spinal Analgesia

Anesthetics and narcotics administered either in the epidural space or

intrathecally are among the most potent analgesic agents available; the efficacy of

these agents is greater than that provided by intravenous PCA techniques. These

drugs can be administered in several ways, including a single-shot dose given by

epidural or intrathecal injection, intermittent injection given either on schedule or

on demand, and continuous infusion.

Because of the risk of central nervous system infections and headaches,

1351intrathecal administration is usually limited to a single dose of narcotic, local

anesthetic, or both. In comparison with epidural administration, duration of action

for a single dose is increased via the intrathecal route as a result of the high

concentrations of drug attained in the cerebrospinal fluid. The risk of central

nervous system and respiratory depression, and systemic hypotension, is

increased. The low doses of opioids required for intrathecal analgesia are

sufficient to be associated with an increased risk of respiratory depression (121).

Some investigators warn against the use of intrathecal spinal analgesia outside the

intensive care setting.

Epidural administration is the preferred approach and provides extended

(>24 hours) pain control during the postoperative period. Relative

contraindications are the presence of coagulopathy, sepsis, and hypotension. Both

anesthetic and narcotic agents are used with excellent efficacy. Among the

anesthetic agents, bupivacaine is the most popular, providing excellent analgesia

with minimal toxicity. Epidural analgesia is most suited for pain control in the

lower abdomen and extremities. Potential adverse effects of epidural anesthetic

agents include urinary retention, motor weakness, hypotension, and central

nervous system and cardiac depression. In contrast to anesthetic agents, opioids

offer excellent analgesia without accompanying sympathetic blockade. Epidural

opioids tend to have a much longer duration of action, and hypotension is a rare

complication. Compared with epidural anesthetics, there is a higher incidence of

nausea and vomiting, respiratory depression, and pruritus (122).

Compared with analgesics administered intramuscularly or intravenously,

epidural analgesia is associated with improved postoperative pulmonary function,

a lower incidence of pulmonary complications, a decrease in postoperative venous

thromboembolic complications (most likely secondary to earlier ambulation),

fewer gastrointestinal side effects, a lower incidence of central nervous system

depression, and shorter convalescence (122). A systematic review concluded

that continuous epidural anesthesia is more effective than intravenous opioid

PCA in reducing postoperative pain for up to 72 hours after abdominal

surgery (123). Severe respiratory depression, which occurs in less than 1% of

patients, is the most serious potential complication. A lower incidence of

respiratory depression occurs with the more lipophilic drugs such as fentanyl,

which is quickly absorbed within the spinal cord and is less likely to diffuse to the

central nervous system respiratory control centers. Pruritus, nausea, and urinary

retention are common but can be managed easily and usually are of little clinical

significance. Cost is perhaps the main and most limiting drawback of epidural

analgesia.

Close monitoring by nursing staff is required for safe administration of

epidural analgesia. An intensive care setting is not necessary. Epidural

1352analgesics can be administered safely in a hospital ward setting under close

nursing supervision, using respiratory monitoring with hourly ventilatory checks

during the first 8 hours of epidural analgesia.

Nonsteroidal Anti-Inflammatory Drugs

Current therapeutic strategies for perioperative pain control are largely dependent

on multimodal therapy with opioid analgesics and nonsteroidal anti-inflammatory

drugs (NSAIDs). The nonselective NSAID ketorolac is a potent drug that can be

given orally or parenterally. Ketorolac has a slightly slower onset of activity

than fentanyl but has an analgesic potency comparable to morphine. The

theoretical advantages of NSAIDs over opioids include absence of respiratory

depression, lack of abuse potential, decreased sedative effects, decreased nausea,

early return of bowel function, and faster recovery. In clinical studies, ketorolac is

found to have analgesic effects similar to those of morphine in postoperative

orthopedic patients and, when used in conjunction with PCA, significantly

reduced opioid requirements (124,125). Depending on the type of surgery,

ketorolac has an opioid dose-sparing effect of a mean of 36% and improves

analgesic control of moderate to severe pain 24 hours postoperatively (126). In

the obstetric population, intravenous ketorolac is effective in reducing

postoperative narcotic use after cesarean delivery (127). Although the U.S. Food

and Drug Administration has not approved ketorolac for use during lactation, it

was quantified in breast milk and has lower levels than ibuprofen (128).

Potential adverse effects associated with the use of NSAIDs include an

increased risk of renal compromise (particularly in patients suffering from acute

hypovolemia), gastrointestinal side effects, hypersensitivity reactions, and

bleeding. The effects of ketorolac on bleeding are inconsistent. Studies of

ketorolac on healthy volunteers showed transient increases in bleeding time and

decreases in platelet aggregation, but these changes were not clinically significant

(129). A retrospective cohort study showed increased risk of gastrointestinal and

operative site bleeding in elderly patients receiving high doses of ketorolac,

between 105 and 120 mg per day. Increased risk for all gastrointestinal bleeding

was associated with the use of ketorolac for more than 5 days (130). Controlled

prospective studies did not show a significant increase in blood loss in patients

who receive NSAIDs perioperatively. Ketorolac may be associated with elevated

rates of acute renal failure when therapy exceeds 5 days (131). A meta-analysis of

the use of postoperative NSAIDs in patients with normal preoperative renal

function showed a clinically insignificant reduction in renal function (132). These

agents should be used with extreme care, if at all, in patients with asthma, because

5% to 10% of adult patients with asthma are sensitive to aspirin and other NSAID

preparations.

1353With the advantages of less gastrointestinal toxicity and a lack of antiplatelet

effects, selective cyclooxygenase-2 (COX-2) inhibitors are a valuable option in

perioperative pain management (133). Although evidence exists showing an

increased risk of serious cardiovascular events associated with COX-2 inhibitors,

short-term use of these agents in the perioperative setting can be considered in

low-risk patients without existing cardiovascular disease (134–139).

In addition to NSAIDs, other adjuvant analgesics are being explored to

minimize opioid use and the accompanying side effects, which can delay

recovery. Capsaicin is a nonnarcotic that promotes release of substance P, a

neurotransmitter for pain and heat, which initially results in a burning sensation,

but eventually leads to substance P depletion and a reduction in pain. It is

available in topical and injectable preparations. Ketamine blocks centrally located

N-methyl-d-aspartate pain receptors, and at low subanesthetic doses can reduce

central sensitization caused by surgery and prevent opioid-induced hyperalgesia.

At higher doses, ketamine is associated with hallucinations, dizziness, nausea, and

vomiting. Gabapentin and pregabalin are nonnarcotics that prevent the release of

excitatory neurotransmitters that relay pain signals. They reduce opioid

requirements and are effective antihyperalgesic agents (139).

Tenets of Pain Control in ERAS

Postoperative pain control is an important focus of ERAS protocols. Opioids are

implicated in slowing gastrointestinal function, increasing nausea and vomiting,

and at times creating patient confusion. All of these factors directly inhibit

postoperative progress. Therefore, a successful pain medication approach needs to

be multimodal and not primarily rely on opioids. ERAS protocols include

preoperative medications which often include NSAIDs, GABA agonists, and/or

muscle relaxants. Many protocols include the use of regional anesthesia for pain

control, especially in the context of a laparotomy. With the use of an epidural,

postoperative pain control and gastrointestinal function have been found to be

improved versus the use of an opioid PCA (140). The use of epidural analgesia

has been associated with difficulty ambulating and urinary retention, which goes

against the goals of an ERAS protocol (141). After incision closure, use of local

bupivacaine is encouraged. Other important postoperative pain medications

include the use of NSAIDs (i.e., ketorolac).

Drains

Prophylactic drains have been used since the 1800s in an effort to reduce

perioperative fluid collections or identify bleeding or anastomotic leaks. Studies

have found that drains provide low sensitivity in detecting a leak (142). There is

1354limited data regarding prophylactic drain use in pelvic surgery and data is

inconsistent. Routine use of prophylactic drains is not recommended, but can be

considered in the setting of patient risk factors.

Similarly, nasogastric tubes had commonly been utilized in the postoperative

period in order to protect anastomoses. Studies have found no clear benefit of

prophylactic nasogastric tube use. A meta-analysis of approximately 5,000

patients revealed that patients without nasogastric tubes demonstrated less time to

bowel function and a shorter hospital stay. The rate of anastomotic leak was no

different between the two groups (143). Routine nasogastric tube use is

discouraged.

Laxatives

To encourage expeditious return of bowel function, laxatives are often used.

While available data does not demonstrate a clear benefit, there are few if any

adverse effects with their use. Therefore, laxatives may be used in the

postoperative period.

Outcomes of ERAS Protocols

[5] Given the purposeful design of enhanced recovery protocols, outcomes have

demonstrated significant improvement in many areas of perioperative care. In

meta-analyses, ERAS protocols have been associated with an average decrease in

length of hospital stay of 1.14 days without a concomitant increase in readmission

(144). Similarly, complication rates were decreased by half (145). Studies have

shown that as compliance to a protocol increases, mortality decreases (146).

Fitting with decreased perioperative complications and reduced readmissions,

Canadian ERAS protocols have been associated with health care cost savings up

to $7,000 per patient (147). Another benefit of an ERAS protocol is that patients

report improved quality of life and decreased pain scores (148,149).

A challenge of enhanced recovery protocols is the need for a highly dynamic

and cooperative atmosphere as teamwork and standardization are crucial to the

success of such a protocol (141). This includes the patient herself, the surgeon,

the anesthesiologist, and the nurses, dietitians, trainees, and hospital

administrators. While this is not an easy undertaking, it is of utmost importance,

given the broad benefits seen in perioperative care with initiation of and

adherence to an enhanced recovery protocol.

MANAGEMENT OF MEDICAL PROBLEMS

Correction of Existing Fluid and Electrolyte Abnormalities

1355Correct maintenance of fluid and electrolyte balance in the postoperative

period starts with the preoperative assessment, with emphasis on establishing

normal fluid and electrolyte parameters before surgery. Preoperative

electrolyte abnormalities can pose a diagnostic challenge. The correct diagnosis

and therapy are contingent on a correct assessment of total body fluid and

electrolyte status. After taking a detailed history, initial evaluation should include

an assessment of hemodynamic, clinical, and urinary parameters to determine the

overall level of hydration and the fluid status of the extracellular fluid

compartment. The patient with good skin turgor, moist mucosa, stable vital signs,

and good urinary output is well hydrated. The patient with orthostasis, sunken

eyes, parched mouth, and decreased skin turgor has extracellular volume

contraction.

The laboratory workup for patients with pre-existing fluid problems

should include assessment of blood hematocrit, serum chemistry, glucose,

BUN and creatinine, urine osmolarity, and urine electrolyte levels. Serum

osmolarity is calculated by the following equation:

2[Na+] + glucose (mg/dL)/18 + BUN (mg/dL)/2.8

Normal serum osmolarity is typically 290 to 300 mOsm. The

BUN:creatinine ratio is typically 10:1 but will rise to a ratio of greater than

20:1 under conditions of extracellular fluid contraction. Under conditions of

extracellular fluid deficit, urine osmolarity will typically be high (>400

mOsm), whereas urine sodium concentration is low (<15 mEq/L), indicative

of an attempt by the kidney to conserve sodium. Under conditions of

extracellular fluid excess or in cases of renal disease, in which the kidney has

impaired ability to retain sodium and water, urine osmolarity will be low and

urine sodium will be high (>30 mEq/L).

Specific Electrolyte Disorders

Hyponatremia

Because sodium is the major extracellular cation, shifts in serum sodium

levels are usually inversely correlated with the hydration state of the

extracellular fluid compartment. The pathophysiology of hyponatremia is

usually expansion of body fluids leading to excess total body water (115,150).

Symptomatic hyponatremia usually does not occur until the serum sodium is

below 120 to 125 mEq/L. The severity of the symptoms (nausea, vomiting,

lethargy, seizures) is related more to the rate of change of serum sodium than to

the actual serum sodium level.

Hyponatremia secondary to inappropriate secretion of ADH can occur with

1356head trauma, pulmonary or cerebral tumors, and states of stress. The abnormally

elevated ADH results in excess water retention. Treatment includes water

restriction and, if possible, correction of the underlying cause. Hyponatremia in

the form of extracellular fluid excess can be seen in patients with renal or cardiac

failure and in conditions such as nephrotic syndrome, in which total body salt and

water are increased, with a relatively greater increase in the latter. Administration

of hypertonic saline to correct the hyponatremia would be inappropriate in this

setting. In addition to correcting the underlying disease process, treatment should

include, water restriction with diuretic therapy.

Inappropriate replacement of body salt losses with water alone will result

in hyponatremia. This situation will typically occur in patients who lose large

amounts of electrolytes secondary to vomiting, nasogastric suction, diarrhea, or

gastrointestinal fistulas, and who received replacement with hypotonic solutions.

Simple replacement with isotonic fluids and potassium will usually correct the

abnormality. Rarely, rapid correction of the hyponatremia is necessary, in which

case hypertonic saline (3%) can be administered. Hypertonic saline should be

administered very cautiously to avoid a rapid shift in serum sodium, which will

induce central nervous system dysfunction.

Hypernatremia

Hypernatremia is an uncommon condition that can be life-threatening if

severe (serum sodium greater than 160 mEq/L). Extracellular fluid deficit

results in a hyperosmolar state, which can cause disorientation, seizures,

intracranial bleeding, and death. The causes include excessive extrarenal water

loss, which can occur in patients who have a high fever, who have diabetes

insipidus, either central or nephrogenic, and who have iatrogenic salt loading. The

treatment involves correction of the underlying cause and replacement with free

water either by the oral route or intravenously with D5W. As with severe

hyponatremia, marked hypernatremia should be corrected slowly, no faster than

10 mEq per day, unless the patient is symptomatic from severe acute

hypernatremia (151).

Hypokalemia

Hypokalemia may be encountered preoperatively in patients with significant

gastrointestinal fluid loss (prolonged emesis, diarrhea, nasogastric suction,

intestinal fistulas) or marked urinary potassium loss secondary to renal

tubular disorders (renal tubular acidosis, acute tubular necrosis,

hyperaldosteronism, prolonged diuretic use). The symptoms associated with

hypokalemia include neuromuscular disturbances, ranging from muscle weakness

to flaccid paralysis, and cardiovascular abnormalities, including hypotension,

1357bradycardia, arrhythmias, and enhancement of digitalis toxicity. These symptoms

rarely occur unless the serum potassium level is less than 3 mEq/L. The treatment

is potassium replacement. Oral therapy is preferable in patients who are on an oral

diet. If necessary, potassium replacement can be given intravenously in doses that

should not exceed 10 mEq per hour.

Hyperkalemia

Hyperkalemia is encountered infrequently in preoperative patients. It is

usually associated with renal impairment, adrenal insufficiency, or

potassium-sparing diuretic use. Marked hyperkalemia (potassium >7 mEq/L)

can result in bradycardia, ventricular fibrillation, and cardiac arrest. The treatment

chosen depends on the severity of the hyperkalemia and whether there are

associated cardiac abnormalities detected with ECG. Calcium gluconate (10 mL

of a 10% solution), given intravenously, can quickly offset the toxic effects of

hyperkalemia on the heart. One ampule each of sodium bicarbonate and D5W,

with or without insulin, will cause a rapid shift of potassium into cells. Over the

longer term, cation exchange resins such as sodium polystyrene sulfate

(Kayexalate), taken orally or by enema, will bind and decrease total body

potassium. Hemodialysis is reserved for emergent conditions in which other

measures are not sufficient or have failed (151).

Postoperative Fluid and Electrolyte Management

Several hormonal and physiologic alterations in the postoperative period may

complicate fluid and electrolyte management. The stress of surgery induces an

inappropriately high level of circulating ADH and aldosterone levels, making

postoperative patients prone to sodium and water retention.

Total body fluid postoperative volume may be altered significantly. First,

free water is released from tissues postoperatively, particularly in the patient who

has undergone extensive intra-abdominal dissection and who is restricted from

oral hydration. This free water is often retained in response to the elevated levels

of ADH and aldosterone. Second, fluid retention is further enhanced by third

spacing of fluid in the surgical field. The development of an ileus may result in

an additional 1 to 3 L of fluid per day being sequestered in the bowel lumen,

bowel wall, and peritoneal cavity. After the first few postoperative days, thirdspace fluid begins to return to the intravascular space, and ADH and aldosterone

levels revert to normal. The excess fluid retained perioperatively is mobilized and

excreted through the kidneys, and exogenous fluid requirements decrease.

In contrast to renal sodium homeostasis, the kidney lacks the capacity for

retention of potassium. In the postoperative period, the kidneys continue to

excrete a minimum of 30 to 60 mEq/L of potassium daily, irrespective of the

1358serum potassium level and total body potassium stores (115). If this potassium

is not replaced, hypokalemia may develop. Tissue damage and catabolism during

the first postoperative day usually result in the release of sufficient intracellular

potassium to meet the daily requirements. Beyond the first postoperative day,

potassium supplementation is necessary.

Patients with inadequate cardiovascular or renal reserve are prone to fluid

overload during this time of third-space reabsorption, especially if intravenous

fluids are not appropriately reduced.

The most common fluid and electrolyte disorder in the postoperative

period is fluid overload. An increase in body weight occurs concomitantly with

the fluid expansion. The patient who is gaining weight in excess of 150 g per day

is in a state of fluid expansion. Simple fluid restriction will correct the

abnormality. Diuretics can be used to increase urinary excretion, where

appropriate.

States of fluid dehydration are uncommon but will occur in patients who have

large daily losses of fluid that are not replaced. Gastrointestinal losses should be

replaced with the appropriate fluids. Patients with high fevers should be

given appropriate free water replacement, because up to 2 L per day of free

water can be lost through perspiration and hyperventilation. Although these

increased losses are difficult to monitor, a reliable estimate can be obtained by

monitoring body weight.

Postoperative Acid–Base Disorders

A variety of metabolic, respiratory, and electrolyte abnormalities in the

postoperative period can result in an imbalance in normal acid–base homeostasis,

leading to alkalosis or acidosis. Proper fluid and electrolyte replacement and

maintenance of adequate tissue perfusion will help prevent or correct most acid–

base disorders that occur during the postoperative period.

Alkalosis

The most common acid–base disorder encountered in the postoperative

period is alkalosis (115). Alkalosis is usually of no clinical significance and

resolves spontaneously. While respiratory alkalosis usually results from

hyperventilation caused by excitation of the central nervous system, numerous

metabolic derangements can result in metabolic alkalosis, including posttraumatic

transient hyperaldosteronism, nasogastric suction, infusion of bicarbonate during

blood transfusions in the form of citrate, and use of diuretics. Alkalosis can be

corrected with removal of the inciting cause and the correction of extracellular

fluid and potassium deficits (Table 25-10). Full correction usually can be safely

achieved over 1 to 2 days.

1359Marked alkalosis (serum pH >7.55) can result in serious cardiac

arrhythmias or central nervous system seizures. Myocardial excitability is

particularly pronounced with concurrent hypokalemia. Under such conditions,

fluid and electrolyte replacement may not be sufficient to rapidly correct the

alkalosis, requiring treatment with acetazolamide or an acidifying agent.

Table 25-10 Causes of Metabolic Alkalosis

Disorder Source of Alkali Cause of Renal HCO

Retention

Gastric alkalosis

Nasogastric suction Gastric mucosa ↓↓ ECF, ↓ K

Vomiting

Renal alkalosis

Diuretics Renal epithelium ↓ ECF, ↓ K

Respiratory acidosis and

diuretics

↓ ECF, ↓ K, ↑ PCO2

Exogenous base NaHCO3, Na citrate, Na

lactate

Coexisting disorder of ECF, K,

PaCO2

↓ ECF, extracellular fluid depletion; ↓ K, potassium depletion; ↑↑ PCO2, carbon dioxide

retention; NaHCO3, sodium bicarbonate; PaCO2, partial pressure of carbon dioxide,

arterial.

Acidosis

Respiratory acidosis results from carbon dioxide retention in patients who

have hypoventilation. Metabolic acidosis is less common during the

postoperative period but can be serious because of its effect on the

cardiovascular system, including decreased myocardial contractility, a

propensity for hypotension, and refractoriness of the fibrillating heart to

defibrillation (115).

Metabolic acidosis results from a decrease in serum bicarbonate levels. The

proper workup includes a measurement of the anion gap:

Anion gap = (Na+ + K+) − (Cl− + HCO3−) = 10 to 14 mEq/L (normal)

The anion gap is composed of circulating protein, sulfate, phosphate, citrate,

1360and lactate.

With metabolic acidosis, the anion gap can be increased or normal. An

increase in circulating acids will consume and replace bicarbonate ion, increasing

the anion gap. Postoperatively, conditions of poor tissue perfusion will increase

circulating lactic acid, severe diabetes or starvation will increase ketoacids, and

renal dysfunction will increase circulating sulfates and phosphates (152). The

diagnosis can be established via a thorough history and measurement of serum

lactate (normal <2 mmol/L), serum glucose, and renal function parameters.

Metabolic acidosis in the face of a normal anion gap is usually the result of excess

chloride and decreased bicarbonate that can be seen in patients who underwent

saline loading, those with severe diarrhea or renal tubular acidosis. A summary of

the various causes of metabolic acidosis is shown in Table 25-11.

The treatment of metabolic acidosis depends on the cause. In patients with

lactic acidosis, restoration of tissue perfusion is imperative. This state can be

accomplished through cardiovascular and pulmonary support as needed, oxygen

therapy, and aggressive treatment of systemic infection wherever appropriate.

Ketosis from diabetes can be corrected gradually with insulin therapy. Ketosis

resulting from chronic starvation or from lack of caloric support postoperatively

can be corrected with nutrition. In patients with normal anion gap acidosis,

bicarbonate lost from the gastrointestinal tract should be replaced, excess chloride

administration can be curtailed, and, where necessary, a loop diuretic can be used

to induce renal clearance of chloride. Dilutional acidosis can be corrected with

mild fluid restriction.

Table 25-11 Causes of Metabolic Acidosis

High Anion Gap Normal Anion Gap

Hyperkalemic Hypokalemic

Uremia Hyporeninism Diarrhea

Ketoacidosis Primary adrenal failure Renal tubular acidosis

Lactic acidosis NH2Cl Ileal and sigmoid bladders

Aspirin Sulfur poisoning Hyperalimentation

Paraldehyde Early chronic renal failure

Methanol Obstructive uropathy

Ethylene glycol

1361Methylmalonic aciduria

NH2Cl (chloramine)

Adapted from Narins RG, Lazarus MJ. Renal system. In: Vandam LD, ed. To Make the

Patient Ready for Anesthesia: Medical Care of the Surgical Patient. 2nd ed. Menlo Park,

CA: Addison Wesley; 1984:67–114.

Bicarbonates should not be given unless serum pH is lower than 7.2 or

severe cardiac complications secondary to acidosis are present. Close

monitoring of serum potassium levels is mandatory. Under states of acidosis,

potassium will exit the cell and enter the circulation. The patient with a normal

potassium concentration and metabolic acidosis is actually depleted of

intracellular potassium. Treatment of the acidosis without potassium replacement

will result in severe hypokalemia with its associated risks.

Endocrine Disease

The three most frequent endocrine disorders that occur in patients undergoing

gynecologic surgery are diabetes mellitus (DM), thyroid disease, and adrenal

abnormalities. The pathophysiology of these disorders aids in understanding the

effects of surgery on patients with these problems.

Diabetes Mellitus

DM is a complicated medical disorder of glucose metabolism that is related to a

lack of production of, or resistance to, insulin. According to the American

Diabetes Association, 14.9 million American women, or 11.7% of all women

older than 18 years, suffer from diabetes (153). Approximately 50% of

individuals with DM will require surgery during their lives (154). The direct

effect of DM on the end organs determines the risk of surgery, rather than the

type or duration of surgery or the management of the condition itself. The current

criteria for diagnosis of diabetes include (155):

1. Polyuria, polydipsia, or unexplained weight loss with a random nonfasting

glucose of ≥200 mg/dL, or

2. Fasting glucose ≥126 mg/dL (in which fasting is defined as no food intake for

8 hours), or

3. Two-hour oral glucose tolerance test of 75 g, with serum glucose ≥200 mg/dL,

or

4. Hemoglobin A1c ≥6.5%.

1362Confirmation of the diagnosis requires repeating the same test on a different

day or concordant results of two different tests simultaneously.

Patients with DM have elevated risk of coronary artery disease and

experience exaggerated hyperglycemia perioperatively. With this in mind,

goals of the preoperative assessment and perioperative management are to ensure

glycemic control and metabolic homeostasis and to anticipate complications

related to DM.

Preoperative Risk Assessment

Preoperative risk assessment for diabetes should begin with a review of systems.

Nocturia, polyuria, polydipsia, glucosuria, obesity, previous gestational diabetes,

ethnicity, and family history are relevant aspects of the history.

Type 1 (insulin-dependent) diabetes, or type 2 diabetes (noninsulin-dependent),

should be established because the preoperative risk and perioperative

management of each differs. The patient’s routine glucose management strategies,

glucose levels, medications, and baseline hemoglobin A1c should be assessed

(156). The presence of end-organ complications of diabetes should be

documented.

Cardiovascular risk resulting from large- and small-vessel arterial

occlusive disease should be a focus in the preoperative setting. A careful

history and physical examination should be performed to determine the presence

or absence of coronary artery or cerebral vascular disease (154). Diabetic patients

are less likely to have symptoms of coronary disease compared to their

nondiabetic counterparts. For patients with inducible ischemia, coronary artery

disease, or multiple clinical risk factors for heart disease, perioperative β-

blockade should be considered, with initiation and careful titration of medication

several weeks prior to surgery (157,158). Further evidence of end-organ disease

can manifest as retinal or renal dysfunction. If diabetic nephropathy is present,

imaging studies using contrast dye should be avoided. If a contrast study must be

performed, adequate hydration before and after the procedure is essential, and oral

metformin should be withheld for 24 to 48 hours after the procedure.

Diabetes is associated with increased perioperative infections (159).

Preoperative evaluation should include examination of the skin and urine

sediment to detect asymptomatic infection. Wound infections, skin infections,

pneumonia, and urinary tract infections account for two-thirds of the

postoperative complications in patients with diabetes (160). There is a known

predisposition for patients with DM to have increased colonization by methicillinresistant Staphylococcus aureus, increased infections by gram-negative and

staphylococcal bacteria, and an increased incidence of gram-negative and group B

streptococcal sepsis (161,162). Seven percent of individuals with diabetes will

1363have postoperative gram-negative sepsis, a rate approximately seven times higher

than that of the nondiabetic population. These complications occur more often in

patients with poor glucose control, probably caused by impaired leukocyte

function in the presence of hyperglycemia (163,164). Individuals with DM have

an increased risk of wound dehiscence and wound infection, possibly related to an

impaired immune function, with changes in phagocytosis, cell-mediated

immunity, and intracellular bactericidal activity (165).

The traditional goal for glucose control perioperatively is to maintain the

glucose level below 200 mg/dL (156,160). Significant debate continues

regarding whether strict glycemic control below 110 mg/dL may be

beneficial in critically ill patients (166,167). Perioperative hyperglycemia

(>250 mg/dL) is associated with increased susceptibility to infection and poor

wound healing. The American Diabetes Association endorses a perioperative

glucose target range of 80 to 180 mg/dL (168).

For patients with diabetes controlled with diet alone or oral medications, oral

hypoglycemic agents should be discontinued when the patient ceases oral intake

of food, and hyperglycemic episodes in the perioperative period are treated with

sliding-scale regular insulin if blood sugar levels exceed 200 mg/dL (156,160).

Insulin-dependent and type 1 diabetes pose more difficult problems. These

patients require a basal rate of insulin and a baseline intake of glucose. Type

1 diabetics are at risk of developing diabetic ketoacidosis whether or not they

are eating (156). Preoperatively, the goals include avoiding ketoacidosis and

hypoglycemia, and, to a lesser extent, hyperglycemia. Traditionally,

approximately one-third to one-half of the patient’s usual daily dose of

intermediate-acting insulin is given subcutaneously the morning of surgery.

Omit any short-acting insulin without oral intake. An infusion of 5%

dextrose should be given while being restricted from oral intake. Additional

regular insulin can be administered in the operating room as needed

(154,156). If patients are normally on a continuous insulin infusion, they may

continue at their usual infusion rate. There is no single regimen that is superior for

the intraoperative management of type 1 diabetic patients and consultation with

endocrine and anesthesia colleagues can be helpful in managing these complex

regimens.

Postoperative Management

Postoperative monitoring of patients with DM includes careful monitoring of

serum glucose levels to avoid severe hypoglycemia or hyperglycemia. If an

intravenous insulin regimen is used, blood glucose levels must be checked

every 1 to 2 hours. If a sliding-scale insulin administration is used, blood

glucose should be checked and documented approximately every 6 hours

1364until the patient is eating and stable on her preoperative regimen. The serum

glucose level should be maintained at less than 250 mg/dL, and ideally below

140 mg/dL when fasting and below 180 mg/dL with random draws (169). For

type 2 diabetics, oral hypoglycemics can be restarted when the patient resumes

eating, except with metformin, which requires normal renal and liver function

(156).

Thyroid Syndromes

Thyroid dysfunction should be suspected in any patient with a history of

hyperthyroidism, use of thyroid replacement medication or antithyroid

medication, prior thyroid surgery, or radioactive iodine therapy.

Hyperthyroidism

Diffuse toxic goiter (Graves disease) is the most common cause of

hyperthyroidism and results from abnormal stimulation of the thyroid gland by

antithyroid antibodies. Other causes of hyperthyroidism include multinodular

goiter, excess thyroid hormone, or thyroiditis. Any signs or symptoms suggestive

of weight loss, tachycardia, atrial fibrillation, goiter, or proptosis should initiate a

more extensive laboratory evaluation of thyroid function. Total thyroxin, free

triiodothyronine (T3), free thyroxin (T4), and thyroid-stimulating hormone (TSH)

tests are useful in diagnosis. In hyperthyroidism, the free T4 level will be

elevated, and the TSH level will be suppressed (170). A new diagnosis of

hyperthyroidism necessitates postponement of elective surgery until adequate

treatment with antithyroid medication is received because of the risk of thyroid

storm. Thyroid storm is associated with mortality of up to 40% (171). Stable

thyroid conditions do not require any special preoperative treatments or tests.

Ideally, a euthyroid state should be maintained for 3 months before elective

surgery. In emergent situations, β-blockers can be used to counter

sympathomimetic drive such as palpitations, diaphoresis, and anxiety. Antithyroid

medications such as propylthiouracil (PTU) or radioactive iodine do not render

patients euthyroid quickly enough for urgent surgery. Radioactive iodine requires

6 to 18 weeks to establish a euthyroid state (170). When thyroid dysfunction is

corrected and maintained for several months, elective surgery can proceed

without additional perioperative monitoring. Antithyroid medications should be

resumed with return of bowel function. If a prolonged delay in resumption of oral

intake is encountered, PTU and methimazole can be administered rectally (172).

When time does not permit establishment of a euthyroid state

preoperatively, oral administration of PTU and a b-blocker can be

implemented for 2 weeks before surgery, and with careful monitoring,

optimal results can be achieved (173). Alternatively, oral b-blockers,

1365glucocorticoids, and sodium iopanoate can be used for 5 days, followed by

surgery on day 6 (172). In the emergent setting, close monitoring of the patient

for tachycardia, arrhythmias, and hypertension is necessary. b-Blockers can

control these symptoms until definitive therapy can be initiated after recovery

from surgery.

Any signs suggestive of the development of thyroid storm—including

hemodynamic instability, tachycardia, arrhythmias, hyperreflexia, diarrhea, fever,

delirium, or CHF—mandate transfer to an intensive care setting for optimal

monitoring and management in consultation with a medical endocrinologist. Such

thyroid instability can be triggered by underlying infection, which requires

diagnosis and treatment to facilitate management of this medical emergency. The

mortality rate from thyroid storm is reportedly between 10% and 75% (172).

Treatment of thyroid storm consists of b-blockers, thioamides, iodine, iodinated

contrast agents, and corticosteroids (160). Aspirin should not be given for fever

in the patient with thyroid storm because it may interfere with the protein binding

of T4 and T3, resulting in increased free serum concentrations (160).

Hypothyroidism

The incidence of hypothyroidism is approximately 1% in the adult population,

and 5% in adults older than 50 years (170). In women older than 60 years, the

incidence of hypothyroidism may approach 6% (171). Hypothyroidism is 10

times more common in women than in men (170). Many such cases are

secondary to previous antithyroid therapy (radioactive iodine or thyroidectomy)

for hyperthyroidism. The most common primary cause of hypothyroidism is

Hashimoto thyroiditis, an autoimmune condition (170). A history of lethargy,

cold intolerance, lassitude, weight gain, fluid retention, constipation, dry skin,

hoarseness, periorbital edema, and brittle hair can be indicative of inadequate

thyroid function. In this setting, physical findings of increased relaxation phase of

deep tendon reflexes, cardiomegaly, pleural or pericardial effusions, or peripheral

edema should stimulate further investigation of thyroid function by assessment of

TSH and free T4 levels. Hypothyroidism decreases cardiac output by 30% to 50%

as a result of decreased stroke volume and heart rate (174). Hyponatremia may be

associated with hypothyroidism because of the inability of the kidneys to excrete

water (174). When elective surgery is planned for severely hypothyroid patients,

surgery should be postponed until thyroid replacement therapy is initiated (170).

In patients with mild or moderate hypothyroidism, the delay of surgery is

controversial (170).

For young patients with mild to moderate hypothyroidism, a starting dose of

1.6 μg/kg of thyroid hormone replacement can be given. In elderly patients,

thyroxin dosage (0.025 mg once a day) should be given with interval dose

1366increases every 4 to 6 weeks until the patient is euthyroid (171). Dosage levels

can ultimately be titrated against TSH levels. In severely hypothyroid patients

requiring urgent or emergent surgery, intravenous T3 or T4 may be given, along

with intravenous corticosteroids to avoid consequences of unrecognized adrenal

insufficiency (160,170).

In the immediate postoperative setting, T4 therapy can be held for 5 to 7 days

while waiting for return of bowel function because the half-life of circulating T4

is approximately 5 to 9 days (173). If more than 5 to 7 days of decreased bowel

function are expected, T4 can be given by the intramuscular or intravenous route

at approximately 80% of the oral dose (174).

Adrenal Insufficiency

Adrenal insufficiency may result in catastrophic postoperative complications,

including death. The most common cause of adrenal insufficiency in the surgical

patient is secondary to the exogenous use of corticosteroids. The physician should

ascertain whether a patient used exogenous steroids for asthma (including inhaled

steroids), malignant conditions, arthritis, or irritable bowel syndrome. The type of

steroid use, the route, the dose, the duration, and the temporal relationship to the

timing of the surgical procedure must be determined. The type of surgical

procedure and its associated stress should be taken into consideration. The use of

high doses of exogenous steroids for prolonged periods can cause circulatory

collapse, and they have adverse effects on wound healing and

immunocompetence.

The daily replacement dose of cortisol is approximately 5 to 7.5 mg of

prednisone. Suppression of the hypothalamic–pituitary–adrenal (HPA) axis

by exogenous steroids for more than a few weeks may produce relative

adrenal insufficiency. When systemic steroids are used for longer periods,

adrenal insufficiency may persist for up to 1 year. Short courses of low-dose oral

steroids (<5 mg of prednisone in a single morning dose for any duration of time,

alternate-day dosing of short-acting glucocorticoids, and any dose of

corticosteroids given for less than 3 weeks) are not thought to cause clinically

significant suppression of the HPA axis (160). More than 1,500 mcg daily for

inhaled corticosteroids (or 750 mcg daily of fluticasone) or more than 2 g per day

of class I topical glucocorticoids may cause suppression (175). If either the dose

or duration of glucocorticoid administration exceeds the preceding regimen,

biochemical tests are recommended to preoperatively evaluate the function of the

adrenal gland. The easiest and safest test to assess HPA function is the

cosyntropin stimulation test. Cosyntropin, a synthetic analog of

adrenocorticotropic hormone, is given in a dose of 250 μg intravenously, and a

1367blood sample is collected 30 minutes after the injection and assayed for plasma

cortisol. A plasma cortisol value of greater than 18 to 20 μg/dL indicates adequate

adrenal function (160,170). If the history regarding exogenous steroid use is

unclear, the cosyntropin stimulation test should be considered as a preoperative

test to determine whether the patient will need perioperative glucocorticoid

coverage. The amount of glucocorticoid replacement should be equivalent to the

normal physiologic response to surgical stress (Table 25-12) (176).

For minor surgical stress, such as colonoscopy, the glucocorticoid target is

approximately 25 mg of hydrocortisone equivalent on the day of the procedure

(176). For moderate surgical stress, for example, open cholecystectomy, the

glucocorticoid target is 50 to 75 mg of hydrocortisone equivalent on the day of

the procedure and tapered quickly for 1 to 2 days. The patient should receive her

normal daily dose preoperatively, followed by 50 mg of hydrocortisone

intravenously administered intraoperatively. For major surgical stress, such as

liver resection, the glucocorticoid target range is 100 to 150 mg hydrocortisone

equivalent on the day of the procedure, tapering rapidly over the next 1 to 2 days

to the usual dosage (176). The patient should receive her normal daily dose

preoperatively.

Administration of high-dose steroids should be stopped as soon as possible

postoperatively because they can inhibit wound healing and promote

infection. Hypertension and glucose intolerance can develop. When a prolonged

or involved procedure is performed and longer steroid use is necessary, careful

tapering may be required. The previously recommended approach was to halve

the dose of hydrocortisone on a daily basis until a dose of 25 mg is reached.

Eliminating one daily dose each day until the drug is stopped was considered the

safest method of withdrawal; no consensus on the timing or duration of steroid

tapering exists. Addison disease is uncommon but should be considered in the

differential diagnosis if the patient develops perioperative hypotension after

steroids are withdrawn. In addition to blood and isotonic fluid replacement, a

“stress” dose of steroids should be given if adrenal insufficiency is suspected and

sepsis and hypovolemia are excluded.

Cardiovascular Diseases

The incidence of perioperative cardiovascular complications decreased markedly

as a result of improvements in preoperative detection of high-risk patients,

preoperative preparation, and surgical and anesthetic techniques (177).

Preoperative Evaluation

Cardiovascular outcomes are related to baseline cardiovascular risk. Thus,

1368the goal of a preoperative cardiac evaluation is to determine the presence of

heart disease, its severity, and the potential risk to the patient during the

perioperative period. Every patient should be questioned about symptoms of

cardiac disease including chest pain, dyspnea on exertion, peripheral edema,

wheezing, syncope, claudication, or palpitations. Prescriptions for

antihypertensive, anticoagulant, antiarrhythmic, antilipid, or antianginal

medications may be the only indication of cardiac problems.

On physical examination, the presence of findings such as hypertension,

jugular venous distension, laterally displaced point of maximum impulse,

irregular pulse, third heart sound, pulmonary rales, heart murmurs,

peripheral edema, or vascular bruits should prompt a more thorough

evaluation. Laboratory evaluation of patients with known or suspected heart

disease should include a blood count and serum chemistry analysis. Patients with

heart disease tolerate anemia poorly. Serum sodium and potassium levels are

particularly important in patients taking diuretics and digitalis. BUN and

creatinine values provide information on renal function and hydration status.

Assessment of blood glucose levels may detect undiagnosed DM.

Coronary Artery Disease

[9] Coronary artery disease is a major risk factor for patients undergoing

abdominal surgery. In an adult population without a prior history of

myocardial infarction, the incidence of myocardial infarction following

surgery is 0.1% to 0.7% (178).

Because of the high mortality and morbidity associated with perioperative

myocardial infarction, considerable effort is made to predict perioperative cardiac

risk. Risk assessment is stratified into three major categories: (i) clinical

predictors, (ii) functional capacity, and (iii) surgery-specific risk (179). Clinical

predictors of increased perioperative cardiac risk include high-risk surgical

procedures, history of ischemic heart disease, history of CHF, history of transient

ischemic attack or stroke, preoperative insulin therapy, and preoperative serum

creatinine levels greater than 2.0 mg/dL. These factors are incorporated into the

RCRI (Table 25-13) (180), a prospectively validated tool that places patients into

one of four risk classes. Depending on the number of risk factors, the risk of

major cardiac events (myocardial infarction, cardiac arrest, pulmonary edema,

and complete heart block) ranges from 0.5% to 9.1% (Table 25-14).

Table 25-13 Major Cardiac Event Rates by the Revised Cardiac Risk Index

1369The patient’s functional status is assessed by a thorough history (Table 25-15),

and self-reported exercise tolerance can be used to predict perioperative risk,

based on a system of metabolic equivalents (METS) (181). Surgery-specific risk

is subdivided into high-risk procedures (emergent major operations, aortic and

vascular procedures, and prolonged surgical procedures associated with large

fluid shifts or blood loss), intermediate-risk procedures (intraperitoneal and

intrathoracic), and low-risk procedures (endoscopic, breast surgery, and

ambulatory procedures). Patients with any clinical risk factor and poor

functional capacity (METS <4) undergoing more than low-risk nonemergent

surgery should undergo preoperative testing, based on the AHA guidelines

(182).

Table 25-14 Clinical Predictors of Increased Perioperative Cardiovascular Risk

Major

Unstable coronary syndromes: acute (≤7 days) or recent (7 days–30 days) MI, unstable

or severe angina

Decompensated congestive heart failure

Significant arrhythmias (high-grade AV block, symptomatic ventricular arrhythmias,

supraventricular arrhythmias with uncontrolled ventricular rate)

Severe valvular disease

Intermediate

History of cerebrovascular disease

Prior ischemic cardiac disease

Compensated or prior congestive heart failure

Diabetes mellitus

1370Renal insufficiency

Minor

Advanced age

Abnormal ECG (LVH, LBBB, ST–T abnormalities)

Rhythm other than sinus

Uncontrolled systemic hypertension

MI, myocardial infarction; AV, atrioventricular; ECG, electrocardiogram; LVH, left

ventricular hypertrophy; LBBB, left bundle branch block.

Adapted from Fleisher LA, Beckman JA, Brown KA, et al. 2009 ACCF/AHA focused

update on perioperative beta blockade incorporated into the ACC/AHA 2007 guidelines on

perioperative cardiovascular evaluation and care for noncardiac surgery: a report of the

American College of Cardiology Foundation/American Heart Association Task Force on

Practice Guidelines. Circulation 2009;120:e169–e276.

Table 25-15 Functional Capacity Assessment From Clinical History

Excellent

Carry 24 lb up eight steps

Carry objects that weigh 80 lb

Outdoor work (shovel snow, spade soil)

Recreation (ski, basketball, squash, handball, jog, or walk 5 mph)

Moderate

Have sexual intercourse without stopping

Walk at 4 mph on ground level

Outdoor work (garden, rake, weed)

Recreation (roller skate, dance)

Poor

Shower and dress without stopping

Basic housework

1371Walk 2.5 mph on level ground

Recreation (golf, bowl)

Adapted from Mehta RH, Bossone E, Eagle KA. Perioperative cardiac risk assessment

for noncardiac surgery. Cardiologia 1999;44:409–418.

Extent of preoperative testing depends on the level of perioperative cardiac

risk. Patients who have dyspnea of unknown origin, current or past heart failure,

prior cardiomyopathy, or with any of the above factors and no cardiac assessment

in the past 12 months should consider echocardiogram testing preoperatively

(182). Exercise stress testing before surgery can identify patients who have

ischemic heart disease not apparent at rest but are at high risk for

perioperative myocardial infarction and cardiac mortality, up to 25% and

18.5%, respectively (183). The exercise stress test must be selectively applied to

a high-risk population because its predictive value depends on the prevalence of

the disease.

Exercise stress testing is limited in some patients who cannot exercise because

of musculoskeletal disease, pulmonary disease, or severe cardiac disease.

Dipyridamole-thallium scanning may be used to overcome the limitations of

exercise stress testing. This study has a high degree of sensitivity and specificity

but a low positive predictive value (182,184).

Dobutamine stress echocardiography is another test to evaluate cardiac risk in

patients who are unable to exercise. This method identifies regional cardiac wall

motion abnormalities after dobutamine infusion. Positive and negative predictive

values are similar to those of dipyridamole-thallium testing for a perioperative

event (185). Coronary angiography should be considered only in patients who

have an indication for angiography independent of the planned surgery, such as

patients with acute coronary syndromes, unstable angina, angina refractory to

medical therapy, or high-risk results on noninvasive testing.

Preoperative testing should be used discriminately in intermediate-risk patients.

Controversy exists regarding the accuracy of these tests to provide prognostic

information beyond what is obtained from clinical risk stratification for

nonvascular procedures. Diagnostic testing should not lead to unnecessary

additional testing or harmful delays in surgery. The American College of

Cardiology and the AHA present a detailed algorithm that incorporates risk

factor stratification to guide clinicians to proceed directly to surgery, to delay

surgery and obtain preoperative noninvasive testing, or to attempt risk

factor modification (182).

Nearly two-thirds of postoperative myocardial infarctions occur during

1372the first 3 days postoperatively (189). Prevention, early recognition, and

treatment are important because myocardial infarctions that occur in the

postoperative period have mortality rates of up to 25% and are associated with

increased rates of cardiovascular death in the 6 months following surgery (186).

Postoperative tachycardia and increased preload are the most important causes of

ischemia because both conditions decrease oxygen supply to the myocardium

while simultaneously increasing myocardial oxygen demand. Tachycardia limits

the diastolic time, when the coronary arteries are perfused, which decreases the

volume of oxygen available to the myocardium. Increased preload increases the

pressure exerted by the myocardial wall on the arterioles within it, thus decreasing

myocardial blood flow.

Other factors associated with perioperative myocardial ischemia include

physiologic responses to the stress of intubation, intravenous or arterial line

placement, emergence from anesthesia, pain, and anxiety. These stresses result in

catecholamine stimulation of the cardiovascular system, resulting in increased

heart rate, blood pressure, and contractility, which may induce or worsen

myocardial ischemia. Loss of intravascular volume because of third spacing of

fluids or postoperative hemorrhage can induce ischemia (187).

Postoperative myocardial infarction is often difficult to diagnose. Chest

pain, which is present in 90% of nonsurgical patients with myocardial infarction,

may be present in only 50% of patients with postoperative infarction because

myocardial pain may be masked by coexisting surgical pain and the use of

analgesics (178). The presence of arrhythmia, CHF, or hypotension may indicate

infarction and should prompt a thorough cardiac investigation and

electrocardiographic monitoring. Measurement of creatinine kinase (CK-MB)

isoenzyme and troponin T levels are the most sensitive and specific indicators

of myocardial infarction, and assessments should be obtained for all patients

suspected of myocardial infarction (186).

Despite the high incidence of silent myocardial infarction, routine use of

postoperative ECG for all patients with cardiovascular disease is

controversial. Many patients will exhibit P-wave changes that spontaneously

resolve and do not represent ischemia or infarction. Conversely, patients with

proven myocardial infarctions may show few, if any, ECG abnormalities. The

American College of Cardiology and the AHA advise consideration of

postoperative surveillance via ST-segment monitoring for myocardial infarction

in patients with known or suspected coronary artery disease (179). In a review of

over 2,400 patients, the sensitivity of predicting postoperative cardiac events was

55% to 100%, specificity 37% to 85%, positive predictive value 7% to 57%, and

negative predictive value 89% to 100% (188). If routine screening of

asymptomatic patients is desired, ECG should be performed 24 hours following

1373surgery because significant ECG changes that occur immediately postoperatively

will persist for 24 hours. It is prudent to continue serial ECG assessments for at

least 3 days postoperatively.

Postoperative management of patients with coronary artery disease is

based on maximizing delivery of oxygen to the myocardium and decreasing

myocardial oxygen utilization. Most patients benefit from supplemental oxygen

in the postoperative period, although special care should be exercised in patients

with COPD. Oxygenation can easily be monitored by pulse oximetry. Anemia is

detrimental because of loss of oxygen-carrying capacity and resultant tachycardia

and should, therefore, be carefully corrected in high-risk patients. Although

transfusion criteria are not absolute, all patients with a hemoglobin less than 6

mg/dL, and hemoglobin of 6 to 10 mg/dL with significant cardiac risk factors

should be offered blood transfusion (189).

Patients with coronary artery disease may benefit from pharmacologic

control of hyperadrenergic states that result from increased postoperative

catecholamine production. β-Blockers decrease heart rate, myocardial

contractility, and systemic blood pressure, all of which are increased by

adrenergic stimulation. The Perioperative Ischemic Evaluation (POISE) trial, a

randomized controlled trial of metoprolol versus placebo, enrolling 8,000 patients

undergoing noncardiac surgery, revealed a reduction in cardiovascular death,

myocardial infarction, and cardiac arrest. There was an increased risk of stroke

and total mortality (190). The AHA provides the following guidelines: Continue

β-blocker therapy in patients on baseline β-blockers for treatment of cardiac

disease. Consider initiating and titrating β-blockers in patients with coronary

artery disease or high cardiac risk (as defined by the presence of more than one

clinical risk factor) who are undergoing intermediate-risk surgery (158). Therapy

should be initiated at least 1 week before surgery to allow for proper titration. The

timing and optimal duration of β-blocker therapy remain an area of uncertainty.

Nevertheless, for patients already on β-blocker therapy, they should continue it

perioperatively because abrupt withdrawal results in a rebound hyperadrenergic

state.

Prophylactic use of other agents such as nitroglycerin and calcium channel

blockers remains controversial, as data did not show a consistent benefit

toward reducing risk of ischemic cardiac events. Nitroglycerin may cause

hypotension, which may worsen cardiac status (158).

Congestive Heart Failure

Patients with CHF face a substantially increased risk of myocardial

infarction during and after surgery (191). The postoperative development of

pulmonary edema may be associated with a high mortality rate, especially if it

1374occurs in the setting of cardiac ischemia (192,193). To prevent severe

postoperative complications, CHF must be corrected preoperatively. The

signs and symptoms of CHF are listed in Table 25-16 and should be assessed

based on preoperative history and physical examination. Patients who are able to

perform usual daily activities without developing CHF are at limited risk of

perioperative heart failure.

Table 25-16 Signs and Symptoms of Congestive Heart Failure

1. Presence of an S3 gallop

2. Jugular venous distension

3. Lateral shift of the point of maximal impulse

4. Lower extremity edema

5. Basilar rales

6. Increased voltage on electrocardiogram

7. Evidence of pulmonary edema or cardiac enlargement on chest radiograph

8. Tachycardia

Treatment usually relies on aggressive diuretic therapy, although care must be

taken to avoid dehydration, which may result in hypotension during the induction

of anesthesia. Hypokalemia can result from diuretic therapy and is especially

deleterious to patients who are taking digitalis. In addition to diuretics and

digitalis, treatment often includes the use of preload- and afterload-reducing

agents. Optimal use of these drugs and correction of CHF may be aided by

consultation with a cardiologist. It is preferable to continue the usual regimen of

cardioactive drugs throughout the perioperative period.

Postoperative CHF frequently results from excessive administration of

intravenous fluids and blood products. Other common postoperative causes are

myocardial infarction, systemic infection, pulmonary embolism, and cardiac

arrhythmias. The cause of postoperative heart failure must be diagnosed because,

to be successful, treatment should be directed simultaneously to the underlying

cause. The most reliable method of detecting CHF is chest radiography, in which

the presence of cardiomegaly or evidence of pulmonary edema is a helpful

diagnostic feature.

Acute postoperative CHF frequently manifests as pulmonary edema.

Treatment of pulmonary edema may include the use of intravenous furosemide,

1375supplemental oxygen, morphine sulfate, and elevation of the head of the bed.

ECG, arterial blood gas, serum electrolytes, and renal function chemistry

measurements should be obtained expediently. If the patient’s condition does not

improve rapidly, she should be transferred to an intensive care unit.

Arrhythmias

Nearly all arrhythmias found in otherwise healthy patients are asymptomatic

and of limited consequence. In patients with underlying cardiac disease,

however, even brief episodes of arrhythmias may result in significant cardiac

morbidity and mortality.

Preoperative evaluation of arrhythmias by a cardiologist and anesthesiologist is

important because many anesthetic agents and surgical stress contribute to the

development or worsening of arrhythmias. Patients with heart disease have an

increased risk of arrhythmias, most commonly ventricular arrhythmias.

Patients taking antiarrhythmic medications before surgery should continue taking

those drugs during the perioperative period.

Patients with first-degree atrioventricular (AV) block or asymptomatic Mobitz I

(Wenckebach) second-degree AV block require no preoperative therapy. A

pacemaker is appropriate in patients with symptomatic Mobitz II second- or thirddegree AV block before elective surgery (194). In emergency situations, a pacing

pulmonary artery catheter can be used. When performing gynecologic surgery

on patients with pacemakers, it is preferable to place the electrocautery unit

ground plate on the leg to minimize interference by preventing the

pacemaker generator from sitting within the electrocautery circuit and to

maximize distance from the pacemaker device. If possible, use of bipolar

cautery devices is recommended rather than monopolar devices. In patients

with a demand pacemaker in place, the pacemaker should be converted

preoperatively to the fixed-rate (or asynchronous) mode. Patients should be

monitored continuously intraoperatively with both telemetry and continuous

pulse oximeter. Close coordination with anesthesia and cardiology is

imperative. Patients with an implantable cardioverter defibrillator device

should have their device programmed off prior to surgery and

reprogrammed postoperatively (158).

Surgery is not contraindicated in patients with bundle branch blocks or

hemiblocks (195). Perioperative mortality rates are not increased by bundle

branch block. Complete heart block rarely develops during noncardiac surgical

procedures in patients with conduction system disease. The presence of a left

bundle branch block may indicate the presence of aortic stenosis, which can

increase surgical mortality if it is severe.

1376Valvular Heart Disease

Although there are many forms of valvular heart disease, two types—aortic

and mitral stenosis—primarily are associated with significantly increased

operative risk. Patients with significant aortic stenosis appear to be at greatest

risk, which is increased in the presence of atrial fibrillation, CHF, or coronary

artery disease. Significant stenosis of aortic or mitral valves should be repaired

before elective gynecologic surgery.

Severe valvular heart disease usually is evident during physical exertion.

Common findings in such patients are listed in Table 25-17. The classic history

presented by patients with severe aortic stenosis includes exercise dyspnea,

angina, and syncope, whereas symptoms of mitral stenosis are paroxysmal and

effort dyspnea, hemoptysis, and orthopnea. Most patients have a remote history of

rheumatic fever. Severe stenosis of either valve is considered to be a valvular area

of less than 1 cm2, and diagnosis can be confirmed by echocardiography or

cardiac catheterization.

Table 25-17 Signs and Symptoms of Valvular Heart Disease

Aortic Stenosis

1. Systolic murmur at right sternal border, which radiates into carotids

2. Decreased systolic blood pressure

3. Apical heave

4. Chest radiograph with calcified aortic ring, left ventricular enlargement

5. Electrocardiogram with high R waves, depressed T waves in lead I, and precordial

leads

Mitral Stenosis

1. Precordial heave

2. Diastolic murmur at apex

3. Mitral opening snap

4. Suffused face and lips

5. Chest radiograph with left atrial dilation

6. Electrocardiogram with large P waves and right-axis deviation

1377Patients with valvular abnormalities are subdivided by the AHA into risk

groups for the development of subacute bacterial endocarditis following surgery.

Patients in the highest-risk groups should receive prophylactic antibiotics

immediately preoperatively to prevent subacute bacterial endocarditis (Table

25-5). As defined by the AHA, only patients with prosthetic cardiac valves,

congenital heart disease, and cardiac transplantation who develop cardiac

valvulopathy should receive perioperative endocarditis prophylaxis (26). All other

patients do not require antibiotics for subacute bacterial endocarditis prophylaxis.

Routine prophylaxis for GI or GU procedures is not recommended. Only in cases

of a known infection of the GI or GU tract should antibiotics coverage for

enterococcus with amoxicillin or ampicillin or vancomycin be provided.

Patients with aortic and mitral stenosis tolerate sinus tachycardia and other

tachyarrhythmias poorly. In patients with aortic stenosis, sufficient levels of

digitalis should be provided to correct preoperative tachyarrhythmias, and

propranolol may be used to control sinus tachycardia. Patients with mitral valve

stenosis often have atrial fibrillation and, if present, digitalis should be used to

reduce rapid ventricular response.

Patients with mechanical heart valves usually tolerate surgery well (196). These

patients should receive antibiotic prophylaxis (Table 25-5). If the patient is taking

aspirin therapy, it should be discontinued 1 week before the procedure and

restarted as soon as it is considered safe by the surgeon. Patients with a bileaflet

aortic valve with no risk factors (atrial fibrillation, previous thromboembolism,

left ventricular dysfunction, a hypercoagulable state, older-generation

thrombogenic valve) generally do not require anticoagulation bridging. Warfarin

should be stopped 72 hours prior to the procedure and resumed 24 hours after the

procedure. In contrast, patients with a mechanical aortic valve and any abovementioned risk factor, or a mechanical mitral valve, should be bridged with

intravenous unfractionated heparin or subcutaneous low–molecular-weight

heparin when the INR falls below 2. Anticoagulation should be stopped 6 to 8

hours (if using heparin drip) or 12 hours (if using low–molecular-weight heparin)

before the surgery. Bridging anticoagulation should be restarted postoperatively

when the patient’s bleeding risk is low and can be stopped when the INR reaches

therapeutic levels (197).

In the postoperative period, patients with mitral stenosis should be

carefully monitored for pulmonary edema because they may not be able to

compensate for the amount of intravenous fluid administered during surgery.

Prevention of tachycardia is important, as it may lead to pulmonary edema.

Patients with mitral stenosis frequently have pulmonary hypertension and

decreased airway compliance. They may require more pulmonary support and

therapy postoperatively, including prolonged mechanical ventilation.

1378For patients with significant aortic stenosis, it is imperative that a sinus

rhythm be maintained during the postoperative period. Even sinus

tachycardia can be deleterious because it shortens the diastolic time. Bradycardia

less than 45 beats per minute should be treated with atropine. Supraventricular

dysrhythmias may be controlled with verapamil or direct-current cardioversion.

Particular attention should be provided to the maintenance of proper fluid status,

digoxin levels, electrolyte levels, and blood replacement.

Hypertension

Patients with controlled essential hypertension have no increased risk of

perioperative cardiac morbidity or mortality. Patients with concomitant heart

disease are at elevated risk and should be completely evaluated by a cardiologist

preoperatively. Laboratory studies should include an ECG, chest radiography,

blood count, urinalysis, and serum electrolytes and creatinine measurement.

Antihypertensive medications should be continued perioperatively. β-Blockers

should be continued, parenterally if necessary, to avoid rebound tachycardia,

hypercontractility, and hypertension. Clonidine may cause significant rebound

hypertension, if withdrawn acutely. Angiotensin-converting enzyme inhibitor

agents and angiotensin II receptor antagonists are associated with increased

intraoperative hypotension and perioperative renal dysfunction possibly resulting

from a hypovolemic state. It may be prudent to withhold these agents on the

morning of surgery and resume them postoperatively when good renal function

and euvolemia are confirmed (158).

[9] Patients with diastolic pressures higher than 110 mm Hg or systolic

pressures higher than 180 mm Hg should receive medication to control their

hypertension before surgery. β-Blockers may be particularly effective agents for

treatment of preoperative hypertension (179).

Postoperative hypertension is usually treated parenterally because

gastrointestinal absorption may be diminished, and transdermal absorption can be

erratic in patients who are cold and rewarming. Commonly used parenteral

antihypertensives are listed in Table 25-18.

Table 25-18 Common Parenteral Antihypertensives

1379Perioperative Antiplatelet Agents

Increasing numbers of patients undergo coronary revascularization procedures,

otherwise known as coronary artery bypass grafting, or percutaneous coronary

intervention, typically stent placement. With the evolution of bare-metal and

drug-eluting stents, perioperative management of cardiovascular thrombotic risk

versus perioperative bleeding and mortality is challenged. Given that drug-eluting

stents generally require 12 months’ treatment with dual agent aspirin and

thienopyridine (i.e., clopidogrel), the American College of Cardiology and

American Heart Association (ACC/AHA) guidelines recommend avoiding

elective surgery within 12 months of drug-eluting stent placement. When surgery

cannot wait, the ACC/AHA recommends continuation of aspirin perioperatively,

discontinuation of thienopyridine 5 days prior to surgery, and resuming it as soon

as possible postoperatively. Ultimately the risk of perioperative morbidity

secondary to bleeding must be weighed against the risk of repeat thrombosis and

cardiovascular morbidity and mortality. If a patient requires new placement of a

cardiac stent and noncardiac surgery in the following 12 months, placement of a

bare-metal stent is recommended rather than a drug-eluting stent, as these require

only 4 to 6 weeks of dual-antiplatelet therapy. Again, aspirin should be continued

perioperatively. Following a newly placed bare-metal stent, noncardiac surgery

should be scheduled at least 30 to 45 days after the stent placement to decrease

cardiac morbidity (179).

Hemodynamic Monitoring

Hemodynamic monitoring is integral to the perioperative management of patients

with cardiovascular and pulmonary diseases. The desire for the quantitative

estimate of cardiac function resulted in the development of bedside PAC.

Before the development of the pulmonary artery catheter, central venous

pressure (CVP) measurement was used to assess intravascular volume status and

cardiac function. To measure the CVP, a catheter is placed in the central venous

system, most frequently the superior vena cava, allowing an estimation of right

atrial pressure. If the pulmonary vascularity and left ventricular function are

normal, the CVP accurately reflects the left ventricular end-diastolic pressure

(LVEDP). The LVEDP reflects cardiac output or systemic perfusion and was

considered the standard estimator of left ventricular pump function. If right

ventricular function is normal, the CVP accurately reflects intravascular volume.

When left or right ventricular dysfunction is present, measurement of

pulmonary artery occlusion pressure can be used to accurately assess volume

status and cardiovascular function. A baloon-tipped PAC, known as a Swan-Ganz

catheter, is an invasive tool used to improve detection in cardiovascular function

changes over clinical observation (198).

1380The distal lumen of the catheter, which is beyond the balloon occluding the

pulmonary artery, measures left atrial pressure (LAP) and, in the absence of

mitral valvular disease, LAP approximates LVEDP. Pulmonary–capillary

wedge pressure (PCWP) equals the LAP, which equals LVEDP and is

normal at 8 to 12 mm Hg. Because the standard pulmonary artery catheter has

an incorporated thermistor, thermodilution studies can be performed to determine

cardiac output. Knowledge of the cardiac output is helpful in establishing

cardiovascular diagnoses. For example, a patient with hypotension, low-to-normal

wedge pressure, and a cardiac output of 3 L per minute is most likely

hypovolemic. The same patient with a cardiac output of 8 L per minute is

probably septic with resultant low systemic vascular resistance.

The effect of PAC use on patient outcome is controversial. Several large trials

did not confer a definite benefit to PAC use. A randomized controlled trial of

1,994 high-risk (American Society of Anesthesiologists class III or IV risk)

patients aged 60 or older undergoing urgent or elective major noncardiac surgery

compared outcomes of those who underwent PAC placement versus standard

care. Results analyzed by a blinded assessor revealed no benefit of PAC over

standard care with central venous catheters (199). Another randomized controlled

trial enrolling 65 intensive care units in the United Kingdom found no difference

in mortality among critically ill patients managed with or without a PAC (200). A

meta-analysis of 13 randomized controlled trials of PAC use found no significant

difference in mortality rates and an increased use of inotropes and intravenous

vasodilators (201). Routine preoperative use of PAC in noncardiac surgery

patients is no longer indicated. Use of PAC in critically ill patients

postoperatively remains controversial.

1381Hematologic Disorders

The presence of hematologic disorders, although uncommon in gynecologic

patients, significantly affects operative morbidity and mortality and should be

considered routinely in preoperative evaluation. Preoperative assessment should

include consideration of anemia, platelet and coagulation disorders, white blood

cell function, and immunity.

Anemia

Moderate anemia is not in itself a contraindication to surgery because it can be

corrected by transfusion. If possible, surgery should be postponed until the cause

of the anemia can be identified and the anemia corrected without resorting to

transfusion. By tradition, anesthetic and surgical practice recommended a

hemoglobin level of greater than 10 g/dL or a hematocrit of greater than 30%.

Data suggest a lower tolerance for pre- and intraoperative transfusion threshold to

improve intra- and postoperative morbidity and mortality (202). It remains that no

universal “transfusion threshold” is agreed upon, but that a hematocrit of less than

24% should prompt strong consideration (203). The circulating blood volume

provides oxygen-carrying capacity and tissue oxygenation. Individual tolerance of

anemia depends on overall physical fitness and cardiovascular reserve.

Maintenance of adequate tissue perfusion requires an increase in cardiac output as

hemoglobin concentration falls (189). In the healthy patient, oxygen delivery is

unchanged when it falls below 7 g/dL (204). In contrast, a patient with ischemic

heart disease will not tolerate anemia as well (205). The presence of cardiac,

pulmonary, or other serious illness justifies a more conservative approach to the

management of anemia. Patients with long-standing anemia may have normal

blood volume levels and tolerate surgical procedures well. There is no evidence

that mild to moderate anemia increases perioperative morbidity or mortality

(205).

Autologous blood donation is advocated as a safer alternative for the patient;

however, the use of preoperative autologous blood donation has come under

scrutiny (206). Preoperative autologous blood may lead to more liberal blood

transfusion, iatrogenic anemia, volume overload, and bacterial

contamination (207). Preoperative autologous blood donation is poorly cost

effective (208). The National Heart, Lung, and Blood Institute does not

recommend collection of autologous blood for procedures with a likelihood of

transfusion less than 10%, such as uncomplicated abdominal and vaginal

hysterectomies (209).

Platelet and Coagulation Disorders

1382Surgical hemostasis is provided by platelet adhesion to injured vessels, which

plugs the opening as the coagulation cascade is activated, resulting in the

formation of fibrin clots. Functional platelets and coagulation pathways are

necessary to prevent excessive surgical bleeding. Platelet dysfunction is

encountered preoperatively more frequently than coagulation disorders.

Platelets may be deficient in both number and function. The normal peripheral

blood count is 150,000 to 400,000 per mm3, and the normal life span of a platelet

is approximately 10 days. Although there is no clear-cut correlation between the

degree of thrombocytopenia and the presence or amount of bleeding, several

generalizations can be made. If the platelet count is higher than 100,000/mm3 and

the platelets are functioning normally, there is little chance of excessive bleeding

during surgical procedures. Patients with a platelet count higher than 75,000/mm3

almost always have normal bleeding times, and a platelet count higher than

50,000/mm3 is probably adequate. A platelet count lower than 20,000/mm3

often will be associated with severe and spontaneous bleeding. Platelet counts

higher than 1,000,000/mm3 are often, paradoxically, associated with bleeding.

In patients with immune destruction of platelets, human leukocyte antigen

(HLA)–matched donor-specific platelets may be required to prevent rapid

destruction of transfused platelets. If surgery can be postponed, a hematology

consultation should be obtained to identify and treat the cause of the platelet

abnormality.

Abnormally low platelet counts result from either decreased production or

increased consumption of platelets. Although there are numerous causes of

thrombocytopenia, most are exceedingly uncommon. Decreased platelet

production may be drug induced and is associated with the use of sulfonamides,

cinchona alkaloids, thiazide diuretics, NSAIDs, gold salts, penicillamine,

anticonvulsants, and heparins (210). Decreased platelet count is a feature of

several diseases, including vitamin B12 and folate deficiency, aplastic anemia,

myeloproliferative disorders, renal failure, and viral infections. Inherited

congenital thrombocytopenia is extremely rare. More commonly,

thrombocytopenia results from immune destruction of platelets by diseases such

as idiopathic thrombocytopenia purpura and collagen vascular disorders.

Consumptive thrombocytopenia is a feature of disseminated intravascular

coagulation, which is encountered most frequently in conjunction with sepsis

or malignancy in the preoperative population.

Platelet dysfunction most often is acquired, but may be inherited. Occasionally,

a patient with von Willebrand disease, the second most common inherited

disorder of coagulation, may be encountered in the preoperative setting. Platelet

dysfunction is more difficult to diagnose than abnormalities of platelet count. A

1383history of easy bruising, petechiae, bleeding from mucous membranes, or

prolonged bleeding from minor wounds may signify an underlying abnormality of

platelet function. Such dysfunction can be identified with the help of a bleeding

time, but full characterization of the underlying etiology should be carried out

with hematologic consultation. If at all possible, surgery should be postponed

until therapy is instituted.

Disorders of the coagulation cascade often are diagnosed through a personal or

family history of excessive bleeding during minor surgery, childbirth, or menses.

Many women with menorrhagia are referred for surgical intervention and require

a thorough preoperative evaluation for possible inherited disorders of hemostasis,

such as factor VIII (hemophilia), factor IX (Christmas disease), factor XI

deficiencies, and von Willebrand disease. Von Willebrand disease is the most

common hereditary bleeding disorder, with prevalence in the general population

of roughly 1%. Seventy percent to 90% of women with von Willebrand disease

have menorrhagia (211). Identified women can be treated effectively with

desmopressin nasal spray, avoiding unanticipated or excessive bleeding during

surgery (212). In the absence of a genetic diagnosis, the diagnosis of von

Willebrand disease is difficult and involves a combination of clinical and

laboratory assessments, including von Willebrand factor antigen and von

Willebrand factor functional activity or ristocetin cofactor assay (212).

Physiologic fluctuations occur with von Willebrand factor levels, requiring repeat

testing and consultation or referral to a hematologist. It is recommended that

women presenting with menorrhagia without obvious pelvic abnormalities should

be routinely screened for inherited bleeding disorders before undergoing invasive

procedures.

There are few commonly prescribed drugs that affect coagulation factors, the

exceptions being warfarin and heparin. Disease states that may be associated with

decreased coagulation factor levels are primarily liver disease, vitamin K

deficiency (secondary to obstructive biliary disease, intestinal malabsorption, or

antibiotic reduction of bowel flora), and disseminated intravascular coagulation.

Preoperative laboratory screening for coagulation deficiencies is controversial.

Routine screening is not warranted in patients who do not have historical

evidence of a bleeding problem (213). Patients who are seriously ill or who will

be undergoing extensive surgical procedures should undergo testing

preoperatively to determine PT, partial thromboplastin time, fibrinogen level, and

platelet count.

Blood Component Replacement

Packed red blood cells (PRBCs), which may be stored for several weeks, are used

for most postoperative transfusions. Most clotting factors are stable for long

1384periods. The exceptions are factors V and VIII, which decrease to 15% and 50%

of normal, respectively. Most hematologic problems observed in the

postoperative period are related to perioperative bleeding and blood component

replacement. Although the primary cause of the bleeding is usually lack of

surgical hemostasis, other factors, including deranged coagulation, may

compound the problem. Such coagulopathy can result from massive transfusion

(less than one blood volume) and is thought to be caused by dilution of platelets

and labile coagulation factors by platelet- and factor-poor PRBCs, fibrinolysis,

and disseminated intravascular coagulation.

A review in Transfusion questioned the traditional practice of limiting blood

component replacement in massive transfusion. Summarizing 14 articles and

encompassing nearly 4,600 patients, the conclusions note a decrease in all-cause

mortality with more liberal transfusion of platelets and fresh frozen plasma (FFP)

(214).

A task force for the American Society of Anesthesiologists recommended

critical values for replacement in patients with massive transfusion and

microvascular bleeding (189):

1. Platelet transfusion usually is indicated for counts less than 50,000/mm3

(with intermediate platelet counts, i.e., 50,000/mm3 to 100,000/mm3, the

transfusion of platelet concentrates should be based on the risk of more

significant bleeding).

2. FFP therapy is indicated if the prothrombin or APTT values exceed 1.5

times the normal value.

3. Cryoprecipitate transfusion is indicated if fibrinogen concentrations

decrease to less than 80 to 100 mg/dL.

Cryoprecipitate transfusions are recommended for prophylaxis in nonbleeding

perioperative patients with fibrinogen deficiencies or von Willebrand disease

refractory to desmopressin acetate and bleeding patients with von Willebrand

disease (212).

Donor blood is stored in the presence of citrate, which chelates calcium to

prevent clotting, increasing the theoretical risk of hypocalcemia following

massive transfusion. Citrate is metabolized at a rate equivalent to 20 U of blood

transfused per hour; thus, routine supplementation of calcium is unnecessary.

Close monitoring of calcium levels is required in patients with hypothermia, liver

disease, or hyperventilation because citrate metabolism may be slowed. Hepatic

metabolism of citrate to bicarbonate can result in metabolic alkalosis following

transfusion, resulting in subsequent hypokalemia, despite the high level of

extracellular potassium in stored blood.

1385Pulmonary Disease

In patients undergoing abdominal surgery, anesthesia-related airway irritation,

splinting of breathing related to pain, and immobilization lead to several

pulmonary physiologic changes, including a decrease in the functional residual

capacity (FRC), an increase in ventilation–perfusion mismatching, and impaired

mucociliary clearance of secretions from the tracheobronchial tree (215). Risk

factors for postoperative pulmonary complications include the following

(Table 25-19) (216):

Upper abdominal or thoracic, or abdominal aortic aneurysm surgery

Surgical procedure time longer than 3 hours

COPD

Smoking within 2 months of surgery

Use of pancuronium for general anesthesia

New York Heart Association Class II pulmonary hypertension

General anesthesia

Emergency surgery

Poor nutrition (serum albumin <3.5 mg/dL or BUN <8)

Healthy, asymptomatic patients do not require preoperative pulmonary

function testing, arterial blood gas measurements, or chest x-rays. Most

patients with abnormal chest x-rays have history or physical examination findings

suggestive of pulmonary disease.

Table 25-19 Predictors of Postoperative Pulmonary Complicationsa

Parameter Value

Maximal breathing capacity <50% predicted

FEV1 <1 L

FVC <70% predicted

FEV1/FVC <65% predicted

Pao2 <60 mm Hg

PaCO2 >45 mm Hg

aComplication defined as atelectasis or pneumonia.

1386FEV, forced expiration volume; FVC, forced vital capacity; Pao2, partial pressure of

oxygen, arterial; PaCO2, partial pressure of carbon dioxide, arterial.

Adapted from Blosser SA, Rock P. Asthma and chronic obstructive lung disease. In:

Breslow MJ, Miller CJ, Rogers MC, eds. Perioperative Management. St. Louis, MO:

Mosby; 1990:259–280.

Asthma

Asthma affects approximately 22 million individuals in the United States,

including 6% of children (217). It is characterized by a history of episodic

wheezing, physiologic evidence of reversible obstruction of the airways either

spontaneously or following bronchodilator therapy, and pathologic evidence of

chronic inflammatory changes in the bronchial submucosa.

Multiple stimuli are noted to precipitate or exacerbate asthma, including

environmental allergens or pollutants, respiratory tract infections, exercise, cold

air, emotional stress, nonselective β-adrenergic blockers, and NSAIDs.

Management of asthma includes removal of the inciting stimuli and use of

appropriate pharmacologic therapy to relax the airways and alleviate

inflammation (217). The optimal therapy for asthma involves managing the acute

symptoms and long-term management of the inflammatory component of the

disease.

Pharmacotherapy of Asthma

The treatment of asthma is divided into long-term control modalities and shortterm modalities. Recognizing the underlying pathophysiology in asthma as an

inflammatory condition, inhaled corticosteroids are the cornerstone for

maintenance therapy. Onset of action is slow (several hours), and up to 3

months of steroid therapy may be required for optimal improvement of bronchial

hyperresponsiveness. Even with acute bronchospasm, steroid treatment can

enhance the beneficial effect of β-adrenergic treatment. During acute

exacerbations of asthma, a short course of oral steroids, in addition to inhaled

steroids, may be necessary. For adults with chronic asthma, only a minority will

require chronic oral steroid therapy. Patients taking oral steroids should receive

intravenous steroid support in the form of 100 mg of hydrocortisone IV every 8

hours perioperatively, sharply tapering within 24 hours of surgery to avoid

adrenal insufficiency. Other long-term control therapies include (217):

1. Leukotriene modifiers (i.e., montelukast): These agents interfere with

leukotrienes, substances released from mast cells, eosinophils, and basophils

important in the inflammatory response.

2. Cromolyn sodium: Highly active in the treatment of seasonal allergic asthma

1387in children and young adults. It is usually not as effective in older patients or

in patients in whom asthma is not allergic in nature.

3. Immunomodulators: Omalizumab, a monoclonal antibody to immunoglobulin

E (IgE), has gained support for prevention and may be an important adjunct in

symptomatic suppression.

4. Long-term β2-adrenergic agonists (i.e., salmeterol): Although not appropriate

for single-agent management or use in mild disease, they are an important

adjunct in the suppression of symptoms in patients with significant disease.

5. Methylxanthines (i.e., theophylline): These were relegated to third-line status

in the management of asthma. Theophylline toxicity can develop when other

drugs such as ciprofloxacin, erythromycin, allopurinol, Inderal, or cimetidine

are concomitantly administered. Therapeutic serum levels must be monitored

closely.

β2-Adrenergic agonists remain the first-line drugs for acute asthma

attacks. These drugs, inhaled four to six times daily, rapidly relax smooth muscle

in the airways and are effective for up to 6 hours. Studies of β2-agonists in

chronic asthma failed to show any influence of these agents on the inflammatory

component of asthma. β2-Agonists are recommended for short-term relief of

bronchospasm (“rescue inhalers”) or as first-line treatment for patients with very

infrequent symptoms or symptoms provoked solely by exercise (217).

Anticholinergic agents are weak bronchodilators that work via inhibition of

muscarinic receptors in the smooth muscle of the airways. The quaternary

derivatives, such as ipratropium bromide (Atrovent), are available in an inhaled

form that is not absorbed systemically. Anticholinergic drugs may provide

additional benefit in conjunction with standard steroid and bronchodilator therapy

but should not be used as single-agent therapy because they do not inhibit mast

cell degranulation, do not have any effect on the late response to allergens, and do

not have an anti-inflammatory effect (217).

Perioperative Management of Asthma

In patients with asthma, elective surgery should be postponed whenever

possible until pulmonary function and pharmacotherapeutic management

are optimized. Recent guidelines of the American Academy of Allergy, Asthma

and Immunology recommend three interventions to reduce perioperative

pulmonary complications related to asthma (217):

1. Review the patient’s asthmatic control, including the need for oral steroids.

2. Optimize the patient’s symptomatic control through the use of long-acting

1388pharmacotherapy, including oral steroids if needed.

3. For patients on oral steroid therapy within 6 months of therapy or for patients

using large doses of inhaled corticosteroids, consider perioperative stress dose

steroids.

For mild asthma, the use of inhaled β-adrenergic agonists preoperatively may

be all that is required. For chronic asthma, optimization of steroid therapy will

greatly decrease alveolar inflammation and bronchiolar hyperresponsiveness.

Inhaled β2-agonists should be added to therapy as needed for further control of

asthma. Each drug prescribed should be used in maximal dosage before adding an

additional agent. Preoperative treatment with combined corticosteroids and an

inhaled β2-adrenergic agonist for a 5-day period may decrease the risk of

postoperative bronchospasm in patients with asthma (218). For patients

undergoing emergent surgery who have significant bronchoconstriction, a

multimodal approach should be instituted, including aggressive bronchodilator

inhalation therapy and intravenous steroid therapy. The role of spirometry,

outside of cardiothoracic procedures, has limited value in predicting postoperative

pulmonary complications and should be limited to the confirmation of

undiagnosed obstructive pulmonary disease (219).

Chronic Obstructive Pulmonary Disease

[9] COPD is the greatest risk factor for the development of postoperative

pulmonary complications. COPD encompasses both chronic bronchitis and

emphysema, disease entities that often occur in tandem. Cigarette smoke is

implicated in the pathogenesis of both, and any treatment plan must include

cessation of smoking (220). Chronic bronchitis is defined as the presence of

productive cough on most days for at least 3 months per year and for at least 2

successive years (221). It is characterized by chronic airway inflammation and

excessive mucus production. The histologic changes of emphysema include

destruction of alveolar septa and distension of airspaces distal to terminal alveoli.

The destruction of alveoli results in air trapping, loss of pulmonary elastic recoil,

collapse of the airways in expiration, increased work of breathing, and significant

ventilation–perfusion mismatching (221). The impaired ability to cough

effectively and clear secretions predisposes patients with COPD to atelectasis and

pneumonia in the postoperative period.

Patients with COPD and a history of heavy smoking account for most

postoperative pulmonary complications in gynecologic surgical patients. The

severity of COPD can be determined preoperatively via a thorough history and

physical examination. According to recommendations regarding the use of

preoperative pulmonary function tests by the American College of Physicians,

1389these should be reserved for individuals in whom COPD is suspected, but

unconfirmed (219,222). Typically, patients with COPD demonstrate impaired

expiratory air flow, manifested by diminished FEV1, forced vital capacity (FVC).

The preoperative preparation for patients at risk for postoperative

pulmonary complications should include cessation of smoking for as long as

possible preoperatively. One to 2 weeks of cessation decreases sputum volume.

Two months of smoking abstinence is required to significantly lower the risk of

postoperative pulmonary complications (215). Longer periods of abstinence can

be counseled in patients undergoing elective surgery.

In patients with severe COPD, maximum improvement in airflow limitation

can be achieved with a therapeutic trial of high-dose oral corticosteroids followed

by a 2-week trial of high-dose inhaled steroid (beclomethasone 1.5 mg per day or

the equivalent) in addition to inhaled bronchodilator therapy. Ideally, oral and

inhaled steroid therapy should be initiated 1 to 2 weeks preoperatively. Inhaled

steroids, in particular, address the inflammatory component of COPD. Oral

steroid therapy initiated preoperatively should be maintained throughout the

perioperative period and then tapered postoperatively. β-Adrenergic agonist

therapy can be initiated at least 72 hours preoperatively and is beneficial in

patients who demonstrate either clinical or spirometric improvement on

bronchodilators.

Patients with COPD and an active bacterial infection suggested by

purulent sputum should undergo a full course of antibiotic therapy before

surgery. The antibiotic used should cover the most likely etiologic organisms,

Streptococcus pneumoniae and Haemophilus influenzae. In any patient with acute

upper respiratory infection, surgery should be delayed if possible. The use of

antibiotics to sterilize the sputum in the absence of evidence of an acute infection

should be avoided because this practice may lead to bacterial resistance.

Aggressive pulmonary toilet, including incentive spirometry, chest physical

therapy, and continuous positive airway pressure devices, reduced the risk of

perioperative pulmonary complications in patients undergoing upper abdominal

surgery, many of which can be instituted preoperatively (222).

Postoperative Pulmonary Management

Atelectasis

Atelectasis accounts for more than 90% of all postoperative pulmonary

complications. The pathophysiology involves a collapse of the alveoli, resulting

in ventilation–perfusion mismatching, intrapulmonary venous shunting, and a

subsequent drop in the PaO2. Collapsed alveoli are susceptible to superimposed

infection, and if managed improperly, atelectasis will progress to pneumonia.

1390Despite the decrease in PaO2, the partial pressure of carbon dioxide (PCO2)

remains unaffected unless atelectatic changes progress to large volumes of the

lung or pre-existing lung disease is present.

Auscultation of the chest may reveal decreased breath sounds at the bases or

dry rales upon inspiration. Percussion of the posterior thorax may suggest

elevation of the diaphragm. Radiologic findings include the presence of horizontal

lines or plates on posteroanterior chest x-rays, occasionally with adjacent areas

containing hyperinflation. These changes are most pronounced during the first 3

postoperative days.

Therapy for atelectasis should be aimed at expanding the alveoli and increasing

the FRC. The most important maneuvers are those that promote maximal

inspiratory pressure, which is maintained for as long as possible. It can be

achieved with aggressive supervised use of incentive spirometry, deep breathing

exercises, coughing, and in some cases, the use of positive expiratory pressure

with a mask (continuous positive airway pressure). Oversedation should be

avoided, and patients should be encouraged to ambulate and change positions

frequently.

Cardiogenic (High-Pressure) Pulmonary Edema

Cardiogenic pulmonary edema can result from myocardial ischemia,

myocardial infarction, or from intravascular volume overload, particularly

in patients who have low cardiac reserve or renal failure. The process usually

begins with an increase in the fluid in the alveolar septa and bronchial vascular

cuffs, ultimately seeping into the alveoli. Complete filling of the alveoli impairs

secretion and production of surfactant. Concomitant with alveolar flooding, there

is a decrease in lung compliance, impairment of the oxygen diffusion capacity,

and an increase in the arteriolar–alveolar oxygen gradient. Ventilation–perfusion

mismatching in the lung results in a decrease in the PaO2, resulting eventually in

decreased oxygenation of the tissues and impairment of cardiac contractility.

Symptoms may include tachypnea, dyspnea, wheezing, and use of the

accessory muscles of respiration. Clinical signs may include distension of the

jugular veins, peripheral edema, rales upon auscultation of the lungs, and an

enlarged heart. Radiographic findings may include the presence of bronchiolar

cuffing and increased interstitial fluid markings extending to the periphery of the

lung. The diagnosis can be confirmed with the use of central hemodynamic

monitoring, which will denote an elevated CVP and, more specifically, an

elevation in the PCWP.

The patient’s volume status should be evaluated thoroughly. In addition,

myocardial ischemia or infarction should be ruled out by performing ECG and

analyzing cardiac enzyme levels. The management of cardiogenic pulmonary

1391edema includes oxygen support, aggressive diuresis, and afterload reduction to

increase the cardiac output. In the absence of myocardial infarction, an inotropic

agent may be used. Mechanical ventilation should be reserved for cases of acute

respiratory failure.

Noncardiogenic Pulmonary Edema (Adult Respiratory Distress Syndrome)

In contrast with cardiogenic pulmonary edema, in which alveolar flooding is

a result of an increase in the hydrostatic pressure of the pulmonary

capillaries, alveolar flooding in patients with adult respiratory distress

syndrome (ARDS) is the result of an increase in pulmonary capillary

permeability. The primary pathophysiologic process is one of damage to the

capillary side of the alveolar–capillary membrane. This damage results in rapid

movement of fluid containing high concentrations of protein from the capillaries

to the pulmonary parenchyma and alveoli. Lung compliance decreases and

oxygen diffusion capacity is impaired, resulting in hypoxemia. If not managed

aggressively, respiratory failure may result. The causes of ARDS include shock,

sepsis, multiple red blood cell transfusions, aspiration injury, inhalation

injury, pneumonia, pancreatitis, disseminated intravascular coagulation, and

fat emboli (244). Twenty-eight-day mortality is reported between 25% and 40%,

with overall mortality as high as 70% (223). Irrespective of the cause, which

should be identified and treated if possible, the evolving clinical picture and

management are very similar.

Clinically, ARDS passes through several stages. Initially, patients develop

tachypnea and dyspnea with no remarkable findings on clinical evaluation or on

chest x-ray. Chest x-rays eventually reveal bilateral diffuse pulmonary infiltrates.

As lung compliance becomes impaired, FRC, tidal volume, and vital capacity

decrease. The PaO2 decreases and, characteristically, increases only marginally

with oxygen supplementation. An attempt should be made to maintain the arterial

oxygen level above 90%. This may be achievable initially by administering

oxygen by mask. For patients with severe hypoxemia, endotracheal intubation

with positive-pressure ventilation should be instituted.

Hemodynamic monitoring is invaluable and should be initiated early in the

course of the disease process in the appropriate intensive care unit setting.

Patients with any evidence of fluid overload should receive aggressive diuresis,

whereas others may require fluid resuscitation for maintenance of tissue perfusion

while the PCWP is maintained below 15 mm Hg. Pulmonary wedge pressure may

be falsely elevated when PEEP is being applied. The goal of management is to

maintain the lowest PCWP, with acceptable cardiac output and blood

pressure. In the setting of hypotension and oliguria, inotropic support with

dopamine or dobutamine or both is helpful.

1392With aggressive management, particularly if the inciting cause is identified

and treated, ARDS can be reversed during the first 48 hours with few

sequelae. After the first 48 hours, progression of the ARDS will cause lung

damage that may leave residual pulmonary fibrosis. The long-term outcome is

usually apparent within the first 10 days, at which time approximately half of

patients are weaned from ventilatory support or have died (223).

Renal Disease

The need for surgical intervention in patients with renal impairment resulted in

the development of a very specialized medical approach to their care. Precautions

are necessary to compensate for the kidneys’ impaired ability to regulate fluids

and electrolytes and excrete metabolic waste products. Equally important are the

unique problems that develop in patients with chronic renal impairment, including

an increased risk of sepsis, coagulation defects, impaired immune function and

wound healing, and a propensity to develop specific acid–base abnormalities.

Special consideration must be given to a variety of different medications,

anesthetic agents, and numerous hematologic and nutritional factors that are

important in the successful surgical care of patients with renal insufficiency.

Management of fluid levels and cardiovascular hemodynamics in patients

with acute or chronic renal impairment is paramount. Intravascular fluid

volume changes that lead to hypertension or hypotension are very common in

these patients and often are difficult to manage secondary to autonomic

dysfunction, acidosis, and other problems that are inherent to the underlying

kidney disease. Dialysis-dependent patients are especially susceptible to

postoperative complications, and invasive monitoring should be considered to

help guide fluid replacement and avoid volume overload.

Co-management of dialysis-dependent patients with their nephrologists is

imperative to reduce morbidity and mortality in the perioperative period. For

those patients with end-stage renal disease (ESRD) on hemodialysis, dialysis is

frequently recommended the day prior to surgery to allow patients to be as

euvolemic as possible prior to their procedure. Patients who perform peritoneal

dialysis may increase their regimen during the week preceding surgery to offset

potential delays in resumption of peritoneal dialysis postoperatively. Dialysis

should be resumed according to the patient’s normal schedule postoperatively. A

short-lived but rather significant fall in the number of platelets occurs during

dialysis; in addition, heparin is used in hemodialysis equipment to prevent

clotting. Because of these factors and concerns about postoperative bleeding,

dialysis is usually avoided during the first 12 to 24 hours following surgery. If

patients require significant intraoperative volume repletion or blood transfusion,

they are at an increased risk for volume overload and earlier dialysis may be

1393necessary to prevent significant volume overload.

The major hematologic concern in patients with chronic renal

insufficiency is the increased incidence of bleeding. These bleeding problems

are secondary to abnormal bleeding times and, in particular, disorders of platelet

secretion, platelet aggregation, and decreased von Willebrand factor activity

(224). Anemia, which is common in patients with renal insufficiency, should be

treated preoperatively with erythropoiesis-stimulating agents to allow patients to

enter surgery at their target hemoglobin concentration. Checking a bleeding time

is no longer recommended prior to surgery in patients with ESRD as a normal

bleeding time does not predict the safety of surgical procedures. Optimizing

preoperative hemoglobin, appropriate dialysis timing, and the use of

desmopressin, cryoprecipitate, or conjugated estrogens can reduce uremic

bleeding (225,226).

Impaired kidney function causes phosphate retention by the kidney and

impaired vitamin D metabolism. Therefore, hypocalcemia is common in patients

with renal insufficiency, but tetany and other signs of hypocalcemia are relatively

uncommon because metabolic acidosis increases the level of ionized calcium.

Oral phosphate binders, such as aluminum hydroxide (1 to 2 g per meal), and

dietary phosphate restriction (1 g per day) is the usual treatment for

hypocalcemia–hyperphosphatemia in patients with renal insufficiency. In chronic

situations, because of central nervous system toxicity associated with elevated

aluminum levels, it is preferable to treat hypocalcemia–hyperphosphatemia with

large doses of calcium carbonate (6 to 12 g per day) rather than with the standard

aluminum-containing antacids (227).

Approximately 20% of patients with renal insufficiency exhibit clinical

evidence of protein-calorie malnutrition. Vitamin deficiencies, most notably

with water-soluble vitamins, occur with dialysis. Nutritional disturbances in

patients with chronic renal insufficiency arise secondary to deficiencies in protein

intake, and studies show that, in patients with chronic renal insufficiency, their

kidneys are hyperfiltrating (228). Postoperatively, both protein and caloric intake

may need to be increased dramatically to meet catabolic demands in surgical

patients. As much as 1.5 g/kg of protein and 45 kcal/kg of calories may be needed

(228). Cardiovascular disease is very common in patients with chronic kidney

disease and complications are the leading cause of mortality in patients with

ESRD (229). Careful preoperative cardiovascular risk assessment should be done

for patients with ESRD with a focus on optimization preoperatively to reduce

cardiovascular events in the perioperative period.

Patients with chronic renal disease have an altered ability to excrete drugs and

are prone to significant metabolic derangements secondary to the altered

bioavailability of many commonly used medications. Because of this, and the

1394effect of dialysis on drug pharmacokinetics, the gynecologic surgeon and

nephrologist must be aware of the lowered metabolism and bioavailability of

narcotics, barbiturates, muscle relaxants, antibiotics, and other drugs that require

renal clearance. Pain control in the perioperative period can be achieved with a

variety of agents including acetaminophen, tramadol, and certain opiates. Nonopioids are always preferred, and acetaminophen can be used without dose

adjustment (230). Morphine should be avoided in patients with renal dysfunction

while fentanyl or hydromorphone can be used if opiates are necessary (230,231).

Liver Disease

Management of perioperative problems in gynecologic patients with liver disease

requires a comprehensive understanding of normal liver physiology and the

pathophysiology underlying diseases of the liver that may complicate surgery or

recovery. Patients with liver disease often have numerous complicated

problems involving nutrition, coagulation, wound healing, encephalopathy,

and infection.

Table 25-20 Child’s Classification of Liver Dysfunction

History and Physical Examination

Patients with a history of alcohol abuse, drug use, hepatitis, jaundice, blood

product exposure, or a family member with liver disease should undergo

biochemical evaluation. During the physical examination, note should be made

of any jaundice, signs of muscle wastage, ascites, right upper quadrant tenderness,

palmar erythema, or hepatomegaly.

Laboratory Testing

The biochemical profile (alkaline phosphatase, calcium, lactate dehydrogenase,

bilirubin, serum glutamic–oxaloacetic transaminase, cholesterol, uric acid,

phosphorous, albumin, total protein, and glucose) is not useful for routine

preoperative evaluation. Mild abnormalities can result in further extensive testing

1395that requires consultation, delays in surgery, and increased cost without net

benefit. A possible exception is selected use of biochemical testing when the

history or physical examination reveals abnormalities. Patients with known liver

disease should undergo albumin and bilirubin testing using the Child’s risk

classification (Table 25-20). This system was originally designed to predict

mortality following portosystemic shunt surgery. It divides patients into three

classes of severity based on five easily assessed clinical parameters. Measurement

of PT may be helpful in patients with significant histories of liver disease. If a

history of hepatitis is ascertained, the patient should be tested for serum

aminotransferase, alkaline phosphatase, bilirubin, albumin levels, and PT.

Serologic documentation of hepatitis is important.

Drug Metabolism

Patients with altered liver function should be carefully monitored because of

the prolonged action of many medications used during surgery. In addition to

impaired metabolism, hypoalbuminemia decreases drug binding, which alters

serum levels and biliary clearance rates. The degree of hepatic metabolism varies

greatly, depending on the type of medication considered.

Determination of Operative Risk

Although it is well known that acute hepatobiliary damage results in increased

morbidity and mortality in the surgical patient, estimating the operative risk in

patients with hepatic dysfunction is difficult based on the history and physical

examination. The most accurate method for risk assessment of surgery in

patients with hepatic dysfunction is Child’s classification (Table 25-20).

Using this system, accurate assessment of morbidity and mortality can be directly

related to the degree of liver dysfunction (232). The Child’s classification is

useful for patients undergoing a variety of different types of abdominal surgery.

Data show operative mortalities of 10%, 30%, and 82% for each of the three

Child’s classifications, respectively, while other data called this into question with

operative mortality of 2%, 12%, and 12% (233,234). The major cause of

perioperative death was often sepsis. This classification correlated significantly

with postoperative complications such as bleeding, renal failure, wound

dehiscence, and sepsis. Another method for determining operative risk in patients

with cirrhosis is the Model for End-Stage Liver Disease (MELD), which takes

into account the patient’s PT, bilirubin, and creatinine with various iterations used

to better predict perioperative morbidity and mortality (235). Originally designed

to predict outcomes in cirrhotic patients undergoing the transjugular intrahepatic

portosystemic shunt (TIPS) procedure, it was further studied to include patients

undergoing other surgical procedures. In patients with a MELD score greater than

139615, elective surgery should be deferred (236).

Acute Viral Hepatitis

Acute viral hepatitis poses an increased risk of operative complications and

perioperative mortality and is a contraindication for elective surgery (237).

Elective surgery should be delayed for approximately 1 month after the results of

all biochemical tests have returned to normal (238). In patients with ectopic

pregnancy, hemorrhage, or bowel obstruction secondary to malignancy, surgical

intervention must take place before normalization of serum transaminase levels

(237). In these situations, the perioperative morbidity (12%) and mortality (9.5%)

rates are much higher than when they are performed under ideal situations (239).

Chronic Hepatitis

Chronic hepatitis is a group of disorders characterized by inflammation of

the liver for at least 6 months. The disease is divided by morphologic and

clinical criteria into chronic persistent hepatitis and chronic active hepatitis. A

liver biopsy is usually required to establish the extent and type of injury. The

surgical risk in these patients correlates most closely with the severity of disease.

The risk of surgery in patients with asymptomatic or mild disease is minimal in

contrast to a significant risk for those patients who have symptomatic chronic

active hepatitis. Elective surgery is contraindicated in symptomatic patients, and

nonelective surgery is associated with significant morbidity (238).

Asymptomatic carriers of the hepatitis B virus (HBV; individuals who test

positive for the HBV surface antigen) are not at increased risk for

postoperative complications in the absence of elevated aminotransferase

levels and liver inflammation.

Alcoholic Liver Disease

Alcoholic liver disease encompasses a spectrum of diseases including fatty liver,

acute alcoholic hepatitis, and cirrhosis. Elective surgery is not contraindicated in

patients with fatty liver because liver function is preserved. If nutritional

deficiencies are discovered, they should be corrected before elective surgery.

Acute alcoholic hepatitis is characterized on biopsy by hepatocyte edema,

polymorphonuclear leukocyte infiltration, necrosis, and the presence of Mallory

bodies. Elective surgery in these patients is contraindicated (240). Abstinence

from alcohol for approximately 6 to 12 weeks along with clinical resolution of the

biochemical abnormalities are recommended before surgery is considered. Severe

alcoholic hepatitis may persist for several months despite abstinence and, if any

question of continued activity exists, a liver biopsy should be repeated (241). In

cases of urgent or emergent surgery on patients with alcohol dependence,

1397administration of tapered doses of benzodiazepine is appropriate as prophylaxis

against alcohol withdrawal.

Cirrhosis

Cirrhosis is an irreversible liver lesion characterized histologically by

parenchymal necrosis, nodular degeneration, fibrosis, and a disorganization of

hepatic lobular architecture. The most serious complication of cirrhosis is portal

venous hypertension, which ultimately leads to bleeding from esophageal varices,

ascites, and hepatic encephalopathy. Conventional liver biochemical test results

correlate poorly with the degree of liver impairment in patients with cirrhosis.

Hepatic dysfunction may be somewhat quantified by low albumin levels and

prolonged PTs.

Surgical risk is increased in patients with advanced liver disease, although

it is substantially greater in emergency surgery than in elective surgery.

Perioperative mortality correlates with the severity of cirrhosis and can be

estimated through the use of the Child’s classification (Table 25-20) and the

Model for End-Stage Liver Disease (MELD) score. In patients with Child’s class

A cirrhosis, surgery can usually be performed without significant risk, whereas in

patients with Child’s class B or C, surgery poses a major risk and requires careful

preoperative consideration. Preoperative preparation should include the following

measures: (i) optimizing nutritional status by enteral and parenteral nutrition and

supplementation with vitamin B1, (ii) correcting coagulopathy with

administration of FFP or cryoprecipitate or both, (iii) minimizing pre-existing

encephalopathy, (iv) preventing sepsis from spontaneous bacterial peritonitis by

administering prophylactic antibiotic therapy, and (v) optimizing renal function

and carefully correcting electrolyte abnormalities (242). Meticulous preoperative

preparation focused on correcting abnormalities associated with advanced liver

disease may improve surgical outcomes (243).

Nhận xét