Renata Urban and Lee-may Chen
PREOPERATIVE RISK EVALUATION
A thorough preoperative history and physical should be taken from all patients undergoing surgery. Preoperative testing may include a complete blood count and chemistries with additional testing being based on the findings of the preoperative history, physical examination, indication for surgery, and planned procedures. Special attention needs to be paid to the preoperative and intraoperative issues that arise in the care of obese patients, cardiac patients, respiratory-compromised patients, and any other patients with significant medical comorbidities including patients with venous thromboembolism or malnutrition. Counseling and postoperative management of patients and their families will be influenced by the unique surgeries and conditions encountered in gynecologic oncology.
In general, preoperative evaluation and testing are stratified based on a patient’s comorbidities. All patients undergoing surgery for gynecologic cancer should undergo a thorough evaluation of other medical issues. Such evaluations will provide an individualized preoperative assessment. In addition, the identification of preoperative medical issues will allow these conditions to be medically optimized.
Many patients with gynecologic malignancies will be of older age. As a result, they often have other medical comorbidities. In 2007, the leading causes of death in the United States were heart disease, cancer, stroke, chronic lower respiratory disease, and accidents. Given the prevalence of coronary artery disease, diabetes, peripheral vascular occlusive disease, and obesity in our population, many patients will require some preoperative testing to assess their cardiopulmonary function in anticipation of anesthesia and surgery.
In healthy patients, the likelihood of an unrecognized medical condition that will cause undue surgical risk is low. A review of studies investigating routine preoperative laboratory evaluations with subsequent likelihood of postoperative complications demonstrated that only hematocrit, creatinine, and electrolytes provided a modest benefit to predict for postoperative complications. Preoperative tests should be selected judiciously, because the addition of unnecessary tests has been found to add a significant cost burden.1 Additionally, in patients who have had a recent laboratory evaluation, retesting will not likely lead to identification of new abnormalities. Our anesthesiologists recommend that preoperative laboratory tests be performed no more than 30 days before surgery to have an up-to-date baseline.
There is also little utility in screening electrocardiograms (ECGs) and chest radiographs (CXRs) in otherwise healthy patients. An abnormal preoperative ECG is not a useful predictor of postoperative cardiac complications, even in elderly patients. However, a preoperative ECG can be helpful as a baseline for comparison with postoperative ECG abnormalities. The 2007 American College of Cardiology/American Heart Association (ACC/AHA) guidelines on perioperative cardiac evaluation include a recommendation for a preoperative 12-lead resting ECG prior to intermediate-risk noncardiac or vascular surgery for patients with known cardiovascular disease, cerebrovascular disease, or peripheral artery disease.2 Intermediate-risk procedures include intraperitoneal and intrathoracic procedures, which are commonly performed in the surgical staging and treatment of patients with gynecologic malignancies. The ACC/AHA guidelines also recommend preoperative ECG in patients with other cardiac risk factors, such as diabetes, renal insufficiency, compensated or prior heart failure, or ischemic heart disease.
Even in the healthiest of patients, the preoperative evaluation of patients undergoing surgery for gynecologic cancer will typically include a CXR for staging. Such an evaluation can be helpful in the detection of subclinical pulmonary disease, which may affect intra-and postoperative respiratory function. In addition, the presence of a preoperative pleural effusion is associated with a decreased likelihood of achieving optimal surgical cytoreduction.
When the preoperative suspicion of malignancy is low, there is little evidence supporting the benefit of preoperative CXR regardless of age, unless there is a history of prior or current cardiopulmonary disease. In a meta-analysis of 21 studies investigating the routine use of preoperative CXR, only 0.1% of all CXRs performed led to a change in management. The American College of Physicians recommends preoperative CXR in patients with known cardiopulmonary disease and those older than 50 years of age undergoing upper abdominal/thoracic surgery.3 The AHA suggests a routine posteroanterior and lateral CXR prior to surgery in all patients with morbid obesity (body mass index [BMI] ≥ 40).4
Patients with suspected ovarian, fallopian tube, or peritoneal carcinomas are recommended to have an ultrasound and/or an abdominopelvic computed tomography (CT) scan. Preoperative imaging may also be of use in planning surgery, in order to appropriately counsel patients as to the extent of surgery and postoperative issues that may arise. In the management of ovarian cancer, certain features on CT scan have been associated with the feasibility of optimal cytoreduction. In 2 prospective studies, Bristow et al5 and Ferrandina et al6 both found that a predictive index incorporating features of peritoneal thickening, number of peritoneal implants, involvement of bowel mesentery, suprarenal para-aortic lymphadenopathy, omental extension to spleen and stomach, pelvic sidewall involvement, and/or hydroureter was accurate in the identification of patients unlikely to undergo optimal primary cytoreductive surgery. For patients with presumed ovarian or peritoneal cancer, a preoperative CT scan may allow for counseling of patients as to the likelihood that all disease can be surgically removed and potential selection of patients for primary chemotherapy.
In uterine cancer, the role of lymphadenectomy remains controversial. Histology and depth of myome-trial invasion have been associated with the likelihood of lymph node involvement. Unfortunately, imaging techniques have not been as reliable in the preoperative prediction of myometrial involvement or lymph node involvement. Positron emission tomography (PET)/-CT, CT scan, and Doppler ultrasound have not been found to be sensitive means to assess depth of myometrial involvement.7 However, magnetic resonance imaging (MRI) has been found to be sensitive in the assessment of cervical involvement8; preoperative knowledge of cervical involvement may indicate a need for radical hysterectomy, which in some series has been shown to improve outcome.9 MRI may also play a role when trying to determine whether a tumor is originating from the cervix or endometrium. With the advent of more minimally invasive surgery, preoperative imaging may help anticipate the presence of suspicious or bulky retroperitoneal disease.
MRI and PET/CT are commonly used in the preoperative assessment of cervical cancer. In the American College of Radiology Imaging Network 6651/Gynecologic Oncology Group (GOG) 183 series of early cervical cancer patients, MRI was found to be superior to CT scan in the evaluation of uterine body involvement, tumor size, and parametrial involvement.10 However, neither modality was accurate in the preoperative assessment of cervical stromal invasion. Although MRI has been demonstrated to have increased sensitivity compared with PET/CT in the preoperative assessment of patients with cervical cancer,11 a retrospective study correlating pathology outcome of 38 patients with stage IB/II cervical carcinoma demonstrated a negative predictive value of 92% for PET/CT scan.12 Another small prospective study found that PET/CT was superior to MRI in the preoperative detection of lymph node metastases in cervical cancer.13
In general, the risk of a perioperative coronary event following major gynecologic oncology surgeries is approximately 1% to 5%. Clinical evaluation of patients undergoing noncardiac surgery includes a review of systems to evaluate whether patients are at significant risk for coronary artery disease. The Goldman Cardiac Risk Index, which is based on 9 risk factors, and the subsequent Revised Cardiac Risk Index, which is based on 6 independent predictors of cardiac complications, are both only estimates of risk (Table 18-1).14 The 2007 ACC/AHA Guidelines recommend that “high-risk” cardiac patients, including those with unstable coronary syndromes, decompensated heart failure, significant arrhythmias, and severe valvular disease, undergo further evaluation.15 Patients deemed as being at “intermediate risk,” including those with factors described in Table 18-1, should undergo a clinical evaluation to determine the need for preoperative noninvasive cardiac testing with methods such as transthoracic echocardiogram to evaluate left ventricular function or dobutamine stress echocardiography.
Table 18-1 Revised Cardiac Risk Index
Preoperative heart failure can be an important determinant of postoperative cardiac complications. The ACC/AHA recommends that during the preoperative history and physical examination, an effort be made to assess for unrecognized heart failure.16 Impaired exercise tolerance, which can be a sign of heart failure, can also be a predictor of adverse postoperative cardiac outcome. A prior study of 600 patients undergoing noncardiac surgery showed that simple self-reported measures (eg, ability to walk or climb stairs) were significantly predictive for postoperative cardiac events. However, adequate exercise tolerance may also obviate the need for additional perioperative cardiac testing.
The use of perioperative β-blockade for prevention of coronary events was initially studied in cardiovascular surgery, with subsequent application for patients undergoing noncardiac surgery. The initiation of perioperative β-adrenergic receptor blockade (atenolol or metoprolol) has previously been recommended to decrease perioperative myocardial infarction and mortality. In a randomized controlled trial of 8000 patients undergoing noncardiac surgery, metoprolol therapy did reduce the risk of myocardial infarction, but actually increased the risk of perioperative death and stroke.17 The ACC/AHA has recommended that patients who are on β-blocker medications preoperatively be continued on the agent. For patients undergoing noncardiac surgery, only those who have existing coronary artery disease or 1 risk factor for coronary artery disease (as listed in Table 18-1) can be considered for perioperative β-blockers.18 Many patients do have indications for long-term β-blocker use including patients with known cardiac ischemia, and these patients may still be considered for initiation of β-blockade at the discretion of their primary care provided or cardiologist at least 2 weeks prior to surgery. Patients who are taking antihypertensive medications preoperatively should be continued on these drugs if possible, with careful follow-up of their blood pressure and heart rate because these are affected by perioperative pain and fluid management. Treatment with statins has also been associated with improved mortality after noncardiac surgery.
Any surgical procedure requiring intubation for general anesthesia increases the risk of pulmonary complications. The presence of an acute respiratory condition poses significant concerns in the perioperative patient. Acute infections should be treated before surgery in most nonemergent situations. Other patients with high-risk conditions, including asthma, bronchitis, emphysema, or smoking, should be optimized for their medical condition if possible. Preparing for surgery can also be a teachable moment to encourage a smoking patient to consider smoking cessation. However, prior case-control studies have suggested that a short period of smoking cessation may not abate and may actually increase the rate of pulmonary complications. Because a period of abstinence from smoking of 8 weeks or greater is not always possible prior to cancer surgery, awareness of an increased risk of pulmonary complications for smokers is necessary, even in the absence of chronic lung disease. Additionally, in the setting of a short period of smoking cessation, the evidence surrounding the increased risk is insufficient to dissuade patients from nicotine abstinence in the preoperative period.
Similar to the cardiac preoperative risk indices, pulmonary multifactorial risk indices have been developed and validated to identify patients at increased risk for postoperative pneumonia, so that appropriate respiratory interventions can be made. Age, poor functional status, upper abdominal surgery, general anesthesia, chronic obstructive pulmonary disease, transfusion, steroid use, and smoking all contribute to perioperative pulmonary risks. For patients with significant pleural effusions, consideration can be given to preoperative thoracentesis versus intraoperative chest tube placement to maximize pulmonary function during the time of surgery.
The majority of gynecologic oncology patients with diabetes will have insulin-resistant, or type 2, diabetes mellitus. However, patients with type 1 diabetes will also be encountered. With autoimmune destruction of the pancreatic islets, such patients have a complete lack of endogenous insulin production. Type 1 diabetics are susceptible to frank ketoacidosis. All diabetic patients are also at risk of metabolic and wound complications following surgery. Furthermore, patients with type 2 diabetes have a higher incidence of concomitant coronary atherosclerosis and are at risk for “silent ischemia.”19 Type 2 diabetics can also be at risk for hyperosmolar nonketotic acidosis in the setting of extreme hyperglycemia.
Prior to surgery, baseline glucose levels should be assessed in diabetic patients. Consideration can be given for a glycosylated hemoglobin (HbA1c) serum test. Elevated glucose values, as well as an abnormal HbA1c, are associated with an increased risk of wound infections.20 In addition, the medications and/or insulin used in management of diabetes should be recorded. For patients with evidence of poor glycemic control, aggressive management may include acute hospitalization and subcutaneous (or intravenous) insulin preoperatively. Patients taking oral hypoglycemic medications should be instructed to hold such medications(s) on the morning of surgery. For patients who require insulin and use long-acting insulin, one-third to one-half of their usual dose should be given the night prior to surgery. Scheduling diabetic patients for surgery earlier in the day may help minimize their risk of hypoglycemia while fasting.
Numerous medical conditions benefit from treatment with steroids, including patients with chronic obstructive pulmonary disease, asthma, and rheumatoid arthritis, and many organ transplantation survivors. As a result, some patients will be on chronic steroids prior to surgery. The ingestion of more than 20 mg of prednisone per day (or its equivalent) for ≥ 5 days leads to suppression of the hypothalamic-pituitary-adrenal (HPA) axis and subsequent inability of the adrenal gland to respond adequately to physiologic stress. Such adrenal suppression can result in hypotension and cardiovascular instability at the time of surgery. The use of 5 to 20 mg of prednisone a day is associated with variable suppression of the HPA axis. It is unclear whether high-dose steroids are necessary in the prevention of adrenal insufficiency. A summary of trials concluded that the use of a daily steroid dose (vs. a high dose of hydrocortisone) did not result in any difference in the incidence of perioperative hypotension or tachycardia.21 Weighing against the concern for perioperative adrenal crisis, it is important to note that the chronic use of high-dose steroids can be associated with impaired glycemic control and wound healing.
Chronic kidney disease is defined as a glomerular filtration rate of 60 mL/min, in the presence or absence of structural kidney disease. In 2010, there were estimated to be more than 600,000 patients with end-stage renal disease (ESRD) in the United States. In a large meta-analysis, patients with chronic kidney disease undergoing noncardiac surgery were found to have higher rates of cardiovascular events and perioperative death.22 Patients with ESRD on dialysis have significant fluid management issues and have been found to have increased perioperative complications, including bleeding, infections, and electrolyte abnormalities, particularly hyperkalemia. Although dialysis performed immediately prior to and after surgery has been associated with improved outcomes in patients undergoing cardiac surgery, there has been no such investigation in patients undergoing abdominal surgery. Common goals in patients with chronic kidney disease include a focus on intraoperative euvolemia to maintain renal perfusion. Coordination with nephrologists may help to optimize the timing of perioperative dialysis.
Given the improved care of patients with chronic liver disease and the advanced state of transplantation medicine, patients with chronic liver conditions may develop and require surgical intervention for staging of gynecologic malignancies. Patients with mild to moderate hepatitis, in the absence of cirrhosis, have no additional surgical risk. Cirrhotic patients are at significant risk of increased postoperative complications such as coagulopathy, hypoglycemia, hepatic decompensation with encephalopathy, and even death.23 In patients with large esophageal varices, consideration should be given to delaying laparotomy until variceal banding or shunting can be performed. Although the overall risk of surgery to the varices is unclear, minimally invasive surgery has been performed safely in patients with varices and splenomegaly; in a recent series of 52 laparoscopic procedures in patients with cirrhosis, 4% required conversion to laparotomy.24
The Child-Pugh classification of hepatic cirrhosis has been found to be predictive of surgical outcome, and such clarification should be made in conjunction with the patient’s hepatologist. Unfortunately, preoperative testing may not be helpful in assessing hepatic dysfunction, because transaminases may be normal even in the setting of cirrhosis.23 Thrombocytopenia, prolonged prothrombin time, and hypoalbuminemia may portend increased perioperative risk as well. Although cirrhotic patients often share findings of ascites and splenomegaly with ovarian cancer patients, superficial vascular skin changes such as spider telangiectasias are unique to cirrhotic patients.
Prior to surgery, many gynecologic oncology patients will have compromised nutrition. This can be due to prior chemotherapy and/or radiation, medical comorbidities, or the advanced nature of their disease. Perioperative nutritional assessment may help identify patients who are most likely to benefit from nutritional support. Preoperative weight loss should be quantitated, and the degree of malnutrition should be assessed. The presence of malnutrition has been demonstrated to be associated with prolonged hospitalization in gynecologic cancer patients,25 as well as poor postoperative outcome in other surgical specialties. Albumin, a serum protein marker produced by the liver, is a widely used indicator of malnutrition and has been shown in numerous studies to be associated with increased complications during the postoperative period,26 even when not associated with malnutrition cachexia. Extremely poor preoperative nutrition, as demonstrated by a prealbumin < 10 mg/dl, was shown to be significantly associated with intraoperative blood loss and perioperative morbidity in a series of more than 100 patients undergoing surgical cytoreduction.27 Patients with poor nutrition, in conjunction with complicated medical comorbidities, should be considered to be at increased risk of intensive care (ICU) unit admission.
Prior to surgery for gynecologic malignancies, patients should be made aware of possible complications in the postoperative period. Following surgery for ovarian cancer, 20% to 30% of patients will require admission to an ICU. The most common reasons for admission include respiratory support and fluid management. Consideration of such disposition in the preoperative period will also allow for appropriate resource allocation following surgery.
An important part of counseling prior to surgery for gynecologic cancer involves a discussion of postoperative sexual function and body image. Procedures that are unique to the surgical treatment of gynecologic malignancies are also associated with unique care issues in the intraoperative and postoperative period.
Patients may inquire as to the impact of cervical removal on sexual function; a prior randomized trial of supracervical versus total hysterectomy in benign gynecologic disease showed that there was no difference in postoperative sexual function.28 However, the extent of pelvic dissection in radical hysterectomy may alter postoperative sexual function. Issues with sexual function, such as lubrication and arousal, have been noted after radical hysterectomy for cervical carcinoma.29 In a series of 38 patients, the rate of postoperative sexual function was similar between those who underwent the procedure via laparotomy or laparoscopy.30 “Nerve-sparing” radical hysterectomy has been suggested to minimize rates of postoperative sexual dysfunction. Radical vulvectomy and pelvic exenteration can both affect body image and sexuality, yet both surgeries are performed with a goal of cure and prolongation of life. Counseling with a focus on psychosexual issues may also help in the adjustment period.
The rates of urinary tract dysfunction following radical hysterectomy are estimated to be between 50% and 75%. Patients should be made aware of the possibility of prolonged catheterization. The mode of catheterization can be either by transurethral or suprapubic catheterization.
Depending on the type of surgery and the extent of cancer involvement, patients may require either a temporary or permanent fecal ostomy. Preoperative discussion of the likelihood and nature of such diversion is essential for both the short-term and long-term adaptation of patients and families to such devices. Furthermore, preoperative marking for stoma placement can allow for marking in lying, sitting, and standing positions to determine the optimal place for stoma placement.
Preparing both the patient and involved family for home care coordination in the postoperative period is also helpful, particularly if the need for a skilled nursing facility is anticipated.
Infectious Disease Prophylaxis
Surgical Site Infection Prophylaxis
The majority of patients undergoing gynecologic oncology procedures will undergo a hysterectomy. Given the breach in the vaginal epithelium that occurs, such patients should receive preoperative antimicrobial prophylaxis. Table 18-2 lists possible antimicrobial regimens. Patients with a history of a hypersensitivity reaction to penicillins and/or cephalosporins are recommended to receive clindamycin or metronidazole, plus gentamicin or aztreonam or a quinolone.31 Antimicrobial prophylaxis should be given within 60 minutes prior to the surgical incision to ensure that appropriate tissue levels are present. Many gynecologic debulking procedures will also include intestinal resection; coverage of gram-negative and anaerobic bacterium must be incorporated. According to the Surgical Care Improvement Project, cefazolin may still be considered for preoperative antibiotic prophylaxis32; cefoxitin or cefotetan can be considered given the improved coverage of bowel anaerobes. Recently, a randomized trial comparing ertapenem to cefotetan in elective colorectal surgery showed that ertapenem was associated with a significantly decreased rate of surgical site infectious; it was, however, associated with an increased risk of Clostridium difficile–associated diarrhea.33
Table 18-2 Prophylactic Antimicrobial Regimens
The 2007 AHA guidelines incorporated major revisions to the groups of patients who will most benefit from endocarditis prophylaxis.34 Antimicrobial prevention of endocarditis is now only recommended in patients with prosthetic valves, a prior history of infective endocarditis, unrepaired cyanotic congenital heart disease, repaired congenital heart disease with prosthetic material, and valvular disease in a transplanted heart. Such patients should receive endocarditis prophylaxis prior to invasive genitourinary procedures with ampicillin and gentamicin, substituting vancomycin for ampicillin in penicillin-allergic patients.
Among cancer patients, those with gynecologic malignancies have the highest risk of thromboembolic disease. Prior to surgery, in addition to assessing the likelihood of malignant disease, patients should also be assessed for other risk factors for venous thromboembolism.35 Patients with known cancer, or those in whom the preoperative suspicion is high, should be considered for preoperative thromboprophylaxis with unfractionated heparin (UFH) or low molecular weight heparin 2 hours prior to surgery, as recommended by the 2008 American College of Chest Physician guidelines.36 Some gynecologic oncologists have expressed concern regarding preoperative administration of anticoagulants before major surgery, but continuous assessment for change in practice is indicated. Einstein et al37 reported that a recent change in protocol that included administration of UFH 1 to 2 hours prior to surgery led to a significant decrease in the rate of thromboembolic events. In the event that patients should be interested in the use of regional anesthesia or neuroaxial blockade, discussion should be made with the anesthesia team before administration of preoperative heparin. Continuation of pharmacologic prophylaxis may also be indicated in high-risk patients.
Some patients may have a pre-existing diagnosis of thromboembolism or atrial fibrillation or have a mechanical heart valve in place and be on anticoagulation therapy, prior to surgery. Because there is a risk of hypercoagulability with discontinuation of warfarin, patients at significantly high risk for thrombosis should be transitioned to intravenous or low molecular weight heparin before and after surgery.38 The international normalized ratio should decrease to less than 1.3 to 1.5 before elective surgery. Intravenous heparin should be stopped 6 hours prior to incision. Low molecular weight heparin should be stopped 12 hours prior to incision. Consultation with the patient’s hematologist or cardiologist is also indicated.
Elective surgery is traditionally to be avoided in the first month after acute venous thromboembolism given the significantly increased risk of perioperative complications. A temporary inferior vena cava (IVC) filter should be considered in situations of acute venous thromboembolism to minimize the incidence of perioperative pulmonary emboli or if the risk of bleeding on intravenous heparin is high. If planned in advance with interventional radiology, an IVC filter can be removed within 2 weeks after surgery, even after a patient has resumed anticoagulation. A recent retrospective study from Adib et al39 demonstrated not only that the use of perioperative IVC filters was feasible without a significant increase in the rate of recurrent venous thromboembolism or surgical complications, but also that surgery could be performed relatively soon after the placement of an IVC filter. After surgery, intravenous heparin should be restarted without bolus after at least 12 hours and potentially longer if there is continued concern for surgical bleeding. A temporary vena caval filter can be removed within 2 weeks after surgery, even after a patient has resumed anticoagulation.
For decades, mechanical bowel preparation has been included in the surgical preparation process; the goal of such preparation is to evacuate stool, allowing for improved visualization and reduction of endogenous intestinal bacteria. Increasing evidence has suggested that such preparations not only lead to an increase in anastomotic leaks, but also are associated with an increase in surgical site infections. This was shown in a meta-analysis involving 13 randomized trials as well as a 2009 Cochrane review.40,41 Magnesium citrate may also have risk in patients with renal impairment. Small volumes of polyethylene glycol may be considered if a bowel preparation is necessary.
In minimally invasive surgery, the use of mechanical bowel preparation may theoretically aid in surgical visualization through decompression of the bowels. However, in a series of patients undergoing gynecologic laparoscopy, the use of preoperative bowel preparation did not have a significant impact on the surgical field, operative difficulty, or operative time; however, preoperative discomfort was significantly elevated in the bowel preparation group.42 In open tumor debulking cases, bowel preparation may help to eliminate solid boluses of stool that may potentially confound intra-abdominal exploration for tumor resection.
A recent Cochrane review found that the use of oral and intravenous antibiotics, in the setting of colorectal surgery, was superior to intravenous antibiotics alone.43 This theoretical benefit is not seen in all series; furthermore, in a small series, the use of preoperative oral antibiotics was actually shown to be associated with an increased incidence of C difficile–associated diarrhea. The use of oral antibiotics may still be considered in gynecologic oncology patients who are likely to have intestinal surgery incorporated into surgery; however, given the gastrointestinal (GI) distress that may accompany oral antibiotics, parenteral antimicrobial prophylaxis may be preferred.
Surgery on the Obese Patient
The prevalence of obesity in the United States is increasing every year; 65% of the American population can now be classified as overweight (BMI ≥ 25) or obese (BMI ≥ 30). In addition to being at risk of multiple other malignances due to obesity, obese women are at particular risk of developing cancer of the endometrium. The practicing gynecologic oncologist is therefore extremely likely to operate upon patients with morbid obesity and should be familiar with the physiology and comorbidities that may be present.
The AHA recommends obtaining a preoperative 12-lead ECG and CXR in all morbidly obese patients prior to surgery.4 The AHA proposes additional testing when signs of right ventricular hypertrophy or left bundle branch block are seen on preoperative ECG, because these may be indicative of existing pulmonary hypertension and occult coronary artery disease, respectively. Obese patients with no risk factors for coronary heart disease, such as hypertension, heart failure, vascular disease, or pulmonary hypertension, may not require any further testing. However, patients with 3 risk factors or those with current coronary heart disease will likely require additional invasive testing with exertional cardiac testing. Exercise stress testing is an appropriate assessment of functional capacity and can be predictive of postoperative cardiovascular complications.44 If the patient’s functional capacity is poor or cannot be assessed due to extreme obesity, a dobutamine stress echocardiogram can be considered.
In the presence of morbid obesity, the sheer weight of the chest wall can lead to a restrictive lung physiology, leading to a decreased functional residual capacity and expiratory reserve volume.45 In addition, these patients may also have concomitant or unrecognized sleep apnea.4 If patients are on ambulatory continuous positive airway pressure, this should be continued in the hospital.
The GOG LAP 2 trial demonstrated that minimally invasive surgery in obese patients was feasible; although a higher BMI was associated with an increased likelihood of conversion to laparotomy, this randomized trial demonstrates the feasibility of this technique.46 In addition, a recent case-control study demonstrated that robotic-assisted laparoscopy in obese patients may also be feasible.47 Minimally invasive surgery in the obese population is associated with unique risks. Prolonged steep Trendelenburg positioning, combined with carbon dioxide pneumoperitoneum, will lead to increased airway pressure and decreased airway compliance. In addition, careful positioning is necessary to prevent pressure necrosis given extremes of body weight.48 Furthermore, prolonged surgical procedures may also increase the possibility of rhabdomyolysis. If this is a concern, a creatine kinase level may be obtained.
When minimally invasive approaches are not available or when the obese patient cannot tolerate the necessary positioning, the patient may be considered for a simultaneous panniculectomy, which may facilitate exposure during laparotomy. Such patients are at risk of wound breakdown, and wound infection rates following panniculectomy during gynecologic surgery have ranged from 3% to 33%. However, this procedure has been described in several series to be a beneficial addition to improve visualization in a morbidly obese patient; further, a long-term follow-up study of 42 such patients revealed that 91% of patients were pleased with their surgical outcome.49
CRITICAL CARE/POSTOPERATIVE EVALUATION
The ICU is an essential resource for the management of the most critically ill gynecologic cancer surgery patients. ICU utilization for gynecologic oncology patients ranges between 6% and 33%.50 A multivariate analysis of ovarian cancer patients admitted to the ICU for short (< 24 hours) versus longer stays found that the patients’ preoperative medical condition was less important than perioperative factors in utilization of ICU resources. Patients requiring bowel resection, placement of a pulmonary artery catheter, and ventilator dependence were most likely to require ICU care. Preoperative factors such as hypoalbuminemia and significantly elevated CA-125 have also been associated with an increased likelihood of extensive disease and need for ICU admission.26 Severity of illness by the Acute Physiology and Chronic Health Evaluation (APACHE) classification system has also been correlated with survival of critically ill gynecologic oncology patients.50 Most patients admitted to the ICU after gynecologic oncology surgery have a short critical care course, although in single-institution reports, the 30-day postoperative mortality rate ranges between 11% and 27%. Identification of patients who may be at greatest risk for needing ICU care allows for appropriate anesthesia and perioperative planning to optimize care and outcomes for the patients.
The ICU must be a collaborative setting, involving interactions between the surgeon, critical care–trained physicians, subspecialty consultants (including palliative care), nurses, respiratory therapists, nutritionists, physical therapists, and other support staff, all working together for the common goals of the patient. Although the surgeon may know the acute issues of the patient and her family best, the internist may be able to add perspective to goals of care, particularly if there are conflicts in the level of care being provided to critically ill patients who may actually be facing the end of life. Regardless of an “open” model, where the surgical team continues to be the primary service, or a “closed” model, where the ICU team becomes the primary service while in the ICU, the involvement of an ICU physician can help coordinate patient care and improve patient outcome.
The stresses and hemodynamic shifts related to gynecologic cancer surgery may have significant impact on the cardiovascular system. Cardiac events are estimated to account for more than 50% of perioperative deaths. Tachycardia is the most common hemodynamic abnormality associated with the postoperative period. This increase in heart rate both increases demand and decreases diastolic filling time; the subsequent imbalance between myocardial oxygen supply and demand results in ischemia. The incidence of myocardial infarction after noncardiac surgery in patients with ischemic heart disease is as high as 5% to 6%, with a peak incidence within 2 days of surgery.51 Postoperative hypotension and tachycardia may be related to a relative hypovolemic state. Despite intraoperative fluid resuscitation and tumor cytoreduction, patients with advanced gynecologic tumors may reaccumulate ascites and “third-space” fluid for 2 to 3 days after surgery. Fluid replacement may either be with crystalloid or colloid solutions; however, a low serum albumin and oncotic pressure may lead to continued intravascular depletion. Fluid resuscitation with goal-directed therapy is recommended; a colloid in the form of hetastarch can be used.52 After the second or third postoperative day, fluid mobilization begins; patients may require some assistance with diuretics if renal function is suboptimal or if clinically significant fluid overload is present.
Shock is a condition of inadequate tissue perfusion that may develop as a result of many conditions. Symptoms include hypotension, tachycardia, hypoxia, low urine output, and peripheral vasoconstriction. Depending on the underlying condition, there are various alternations seen on hemodynamic monitoring (Table 18-3). After surgery, the most common hemodynamic condition is hypovolemia. Treatment to improve oxygen delivery and decrease oxygen consumption should focus on volume replacement, keeping the patient warm, correcting coagulopathy, and controlling pain. Replacement of fluids can be done using either crystalloid or colloid solutions. Despite the hypoalbuminemia in many advanced ovarian cancer patients, albumin has not been consistently found to be superior to crystalloid in the resuscitation of hypovolemia.53 Intermittent fluid boluses, rather than continuous infusion, are preferable to allow for evaluation of response. Vasopressor agents may act through increasing cardiac output (inotrope or chronotrope) or by increasing systemic vascular resistance. Dobutamine and intermediate-dose dopamine (5-15 μg/kg/min) are β-selective agonists ideally used in patients with a history of heart failure and may act to improve myocardial contractility and improve cardiac output.
Table 18-3 Hemodynamic Parameters of Shock
Phenylephrine (Neo-Synephrine) and norepinephrine (Levophed) are vasoconstrictors that increase systemic vascular resistance and thus improve blood pressure. Both systemic and individual organ perfusion must be monitored in the pharmacologic management of shock. Hemorrhagic shock can be reversible with prompt replacement of circulating volume and oxygen delivery. Prolonged shock may trigger a cascade of local and systemic cytokines, resulting in a systemic inflammatory response syndrome, which may lead to multiple organ failure.
Sinus tachycardia is common in the postoperative ovarian cancer patient who may be volume depleted or experiencing pain. Treatment involves addressing the underlying physiologic condition. Supraventricular tachyarrhythmias are most commonly encountered and can include atrial fibrillation, atrial flutter, and other sinus node re-entrant tachycardias. Narrow complex tachycardias incorporate a wide variety of rhythm abnormalities, which may be difficult to distinguish on ECG. Initial management should focus on the hemodynamic impact of the rhythm disturbance by controlling the rapid ventricular rate, ruling out ischemia, and evaluating for underlying causes. Paroxysmal supraventricular tachycardia can be managed first with vagal maneuvers or adenosine for diagnosis. Atrial fibrillation or atrial flutter should be managed first by controlling the ventricular response rate. In a hemodynamically stable patient, calcium channel blockers, β-blockers, digoxin, or amiodarone may be used for rate control. Cardioversion may also override and reset the abnormal pulse generator using low doses of electrical energy. Although it may occur in 4% to 12% of patients undergoing noncardiac surgery, atrial fibrillation most commonly occurs within 3 days following surgery.
Regardless of the etiology, hypotensive patients may require central venous pressure monitoring, which may be helpful in assessing the patients’ volume status. However, practitioners should be aware of the infectious, mechanical, and thrombotic complications associated with such catheters. Pulmonary artery (Swan-Ganz) catheters are rarely needed in the absence of significant cardiac disease or pulmonary hypertension. Increasingly, echocardiography is becoming a less invasive modality to evaluate left ventricular function and vascular pathology during both the intraoperative and postoperative period.
Myocardial ischemia may manifest itself as angina/pain or nausea, or it may be asymptomatic with ECG changes only. Tachycardia and elevated catecholamines are a common response to surgery; thus, control of pain and anxiety may help to decrease the incidence of postoperative cardiac events. The classic symptom of chest pain may actually be absent or may be seen in association with other symptoms in female patients; in a review of 515 female patients with a recent myocardial infarction, only 30% reported chest pain, whereas 58% reported shortness of breath and 55% reported weakness. Prompt recognition and treatment allow for the prevention of myocardial infarction and other complications including dysrhythmias, congestive heart failure, and death. Immediate treatment of suspected cardiac ischemia includes administration of supplemental oxygen, nitrates to decrease myocardial demand by venodilation, and β-blockers to decrease heart rate and contractility. Serial ECGs and measurements of serum troponin I can be reflective of the extent of myocardial ischemia. In surgical patients, use of anticoagulation such as heparin must be considered with respect to their recent surgical procedure. Aspirin and antiplatelet drugs are also to be considered at the discretion of the managing surgeon and cardiologist. Transthoracic echocardiography may also evaluate myocardial function to determine the need for additional cardiology interventions.
At the end of surgery, most patients are awakened from anesthesia and extubated. In cases of large fluid resuscitation or prolonged surgery, extubation may be delayed until the recovery room or ICU. Laryngeal edema can increase the likelihood of airway obstruction; such edema can be exacerbated by prolonged Trendelenburg positioning. In addition, bowel edema may increase the intra-abdominal pressure, which in turn limits respiratory excursion and functional residual capacity.
Pulmonary hygiene is emphasized in the postoperative period and is most easily accomplished by encouraging incentive spirometry and early ambulation. Atelectasis can result in retention of bronchial secretions and in fevers and increases the likelihood of pneumonia. Aspiration pneumonitis without infection is treated by support. Pneumonia in the postoperative period is a common cause of respiratory compromise, a leading cause of nosocomial infection, and ICU deaths. Clinical risk factors for pneumonia include thoracic or upper abdominal surgery, history of respiratory disease, and a bedridden status resulting in atelectasis. Patients requiring mechanical ventilation for greater than 24 hours have a higher risk of nosocomial pneumonia.
Chronic obstructive pulmonary disease is usually smoking related and may range from mild to severe. Bronchodilators remain the mainstay of therapy, with a role for inhaled steroids and antibiotics in acute exacerbations. The ventilator management of patients with chronic obstructive pulmonary disease should be modified for the risk of hyperinflation. Because of reduced elasticity, alveoli are prone to overdistention with early airway collapse causing air trapping and inadvertent positive end-expiratory pressure (“autoPEEP”). Such positive airway pressure can accumulate, resulting in hyperinflation of the lungs, barotrauma, and hemodynamic compromise. Extrinsic PEEP can be helpful to decrease intra-alveolar pressure, and the use of a low ventilator sensitivity setting can allow for triggering breaths.
Although acute respiratory distress syndrome (ARDS) is unusual after gynecologic cancer surgery, ovarian cancer patients may be particularly susceptible. Predisposing causes include direct lung injury from aspiration or pneumonia, as well as indirect phenomena of massive fluid shifts, coagulopathy, transfusion, and sepsis. ARDS is defined as a condition with acute onset, bilateral infiltrates on CXR, pulmonary artery wedge pressure < 18 mm Hg in the absence of clinical evidence of left atrial hypertension, and a gradient of partial pressure of arterial oxygen (PaO2) to fractional inspired oxygen (FIO2) of < 200. A PaO2 to FIO2 ratio of < 300 is considered acute lung injury. Treatment consists of supportive care with ventilation and oxygenation, evaluation for underlying causes and nosocomial infection, and prevention of the development of multisystem organ failure. Because of the heterogenous process of ARDS, some areas of the lung may be normal while other areas may be poorly compliant. A positive pressure breath will go preferentially to normal lung; however, this may lead to overdistention of more compliant normal lung zones and subsequently greater stretch injury. Ventilator strategies include using lower tidal volumes and higher PEEP, thereby avoiding overdistension of normal alveoli. A phase III study by the Acute Respiratory Distress Syndrome Network demonstrated a 22% decrease in mortality using low tidal volumes when compared to the use of traditional volumes (6 mL/kg vs. 12 mL/kg).54 Restricting fluids may decrease pulmonary edema, but studies have been negative. The use of glucocorticoids or surfactant has not shown any pharmacologic advantage.
Mechanical ventilation supports respiration by delivering positive pressure though either volume control or pressure control. Intermittent mandatory ventilation delivers a set rate and volume, with unassisted spontaneous breaths allowed in between. In pressure-controlled ventilation, the volume delivered is determined by a preset level of pressure; the patient’s inspiratory effort triggers the ventilator to deliver the breath. Delivering PEEP or continuous positive airway pressure provides support to keep previously collapsed alveoli open and reduces the overall work of breathing. For patients intubated for airway management issues following surgery, the management of mechanical ventilation involves supportive care until a trial of weaning can be performed. The FIO2 should be weaned to minimize injury from oxygen radicals, and the peak inspiratory pressure should be kept low to minimize barotraumas. As long as oxygen and carbon dioxide exchange is adequate, the patient may be extubated once the patient has demonstrated adequate oxygenation and intact neurologic status.
Reintubation is most frequently a result of an unplanned extubation, often due to patient self-extubation. Risk factors for unplanned extubation include inadequate sedation and subsequent agitation, as well as physical restraints. In the setting of planned extubation following a prolonged intubation, a positive fluid balance in the previous 24 hours can be a risk factor for reintubation55; this is particularly pertinent to ovarian cancer patients, who often have a positive fluid balance due to reaccumulation of ascites and/or pleural effusions. Given that reintubation in the ICU setting is associated with worse posthospitalization outcome, patients requiring reintubation will benefit from a multidisciplinary care team involving anesthesiologists and critical care physicians. Of note, patients at risk for reintubation following planned extubation may be considered for noninvasive positive pressure ventilation (bilevel or continuous positive airway pressure), because this has been found to decrease rates of reintubation. Care for such patients should be done with communication with colleagues from anesthesia and critical care.
Fluid and Electrolyte Management
The daily fluid requirement for an average adult ranges between 2 and 3 L/d. Several formulas are available to estimate maintenance fluid requirements; a simple one includes 4 mL/kg/h for the first 10 kg of body weight, 2 mL/kg/h for the second 10 kg, and 1 mL/kg/h for each subsequent kilogram of weight. GI and urinary losses can be replaced with lactated Ringer’s solution or normal saline crystalloid solution (Tables 18-4 and 18-5). Postoperative considerations of fluid management must include insensible losses related to the surgical procedure in addition to the usual insensible losses associated with skin, lung, and fecal material. Weighing the patient daily is the best means of assessing total body fluid status. Supplemental intravenous fluids may be weaned off as the patient’s oral intake increases.
Table 18-4 Commonly Used Parenteral Solutions
Table 18-5 Gastrointestinal Fluid Content
The degree of fluid shifts into the extracellular “third space” reflects the severity of the surgical procedure. Postoperative sodium retention is a response to a decrease in extracellular volume. Fluid replacement during surgery should include 4 mL/kg/h of lactated Ringer’s solution or normal saline for a minimally traumatic procedure, 6 mL/kg/h for a moderately traumatic procedure, and 8 mL/kg/h for an extremely traumatic procedure such as an extensive debulking procedure for a patient with significant ascites. In a procedure where blood loss is significant, replacement should emphasize colloid or blood products. If cardiac and renal functions are normal, the retained fluid should mobilize back into the intravascular space 2 to 3 days after surgery. However, if cardiac function and/or renal function are impaired, inadequate clearance of this fluid may result in pulmonary edema or congestive heart failure. A patient who is slow to spontaneously diurese may only need a single dose of furosemide to facilitate the process.
Clinical evaluation of hypovolemia includes looking for signs of oliguria, supine hypotension, and ortho-static hypotension. Urine output of 0.5 mg/kg/h suggests adequate renal perfusion. Laboratory parameters may include hemoconcentration, azotemia, or low urinary sodium. A ratio of blood urea nitrogen to serum creatinine that is greater than 20 may suggest dehydration. In prerenal oliguria, urine sodium is low, due to increased sodium and water resorption. A fractional excretion of sodium of less than 1% is suggestive of a prerenal state that is best managed by volume expansion. This calculation may be less useful in patients who are elderly, have pre-existing renal disease, or have received diuretics. Even if the urine output is low while volume resuscitative efforts may be ongoing, continued urine production and a stable creatinine indicate adequate renal perfusion. Prevention of acute oliguria in the postoperative patient is best accomplished by recognition of prerenal events including blood loss, surgical trauma, and reaccumulation of ascites or effusions.
The goal of treating hypovolemia is to restore volume with fluid that is similar to the fluid that was lost. Patients with significant blood loss are managed by transfusion and administration of colloids intra- and postoperatively. Although blood products are required to treat anemia and coagulopathy, lactated Ringer’s solution and normal saline without added dextrose are the crystalloid solutions of choice for continued hypotension or shock. However, large volumes of lactated Ringer’s solution may result in hyperkalemia, whereas large volumes of normal saline may result in hyperchloremic acidosis. Intravenous albumin (250 mL of 5% concentration) may not be more effective than crystalloid solutions in the replacement of drained ascites and may eventually leak from the intravascular space into the peritoneal cavity.
Disorders of sodium concentration, hyponatremia and hypernatremia, reflect relative excesses or deficits of extracellular fluid. Pseudohyponatremia may occur with hyperproteinemia or hyperlipidemia, where protein or lipids displace water from plasma, producing a low plasma concentration of sodium. True hyponatremia (sodium < 136 mmol/L) may develop rapidly or chronically; acute hyponatremia can be associated with neurologic changes due to cerebral edema, whereas chronic hyponatremia may trigger compensatory mechanisms that then require slow correction. Serum sodium is frequently low in the postoperative patient with ovarian cancer. With the administration of crystalloid fluids during surgery, hyponatremia may be actually associated with increased total body sodium. In the immediate postoperative patient, there may still be a state of relative hypovolemia, which then results in secretion of antidiuretic hormone, which in turn preserves intravascular volume. Most patients with a serum sodium of greater than 125 mmol/L have few symptoms. A rapid decline of the sodium level to less than 130 mEq/L may result in mental status changes and even seizures; these patients will require rapid but controlled correction with hypertonic saline. An assessment of total body sodium may be made by measuring urine sodium and osmolarity. Although third spacing is ongoing during the immediate postoperative period, the urine sodium level is low (<10-15 mEq/L), and urine osmolarity is high (> 400 mOsm/kg). Treatment of hyponatremia with high serum osmolarity is directed toward the restriction of both sodium and water. Hypovolemic hyponatremic patients are treated with 0.9% saline. Hypervolemic hyponatremic patients are managed with fluid restriction, occasionally increasing free water excretion with the administration of diuretics. Hypernatremia is usually associated with a total fluid deficit and may also result in neurologic changes. It can result from pure water loss (eg, diabetes insipidus) or hypotonic sodium loss (eg, nasogastric drainage) but may also be iatrogenic from hypertonic sodium loading. Correction of hypovolemia must be performed slowly with hypotonic solutions to avoid precipitating cerebral edema or seizures.
Disorders of potassium homeostasis may be affected by the balance of intake versus excretion. Specifically, potassium can enter the body through oral or parenteral means and leaves through renal excretion, which may be affected by acid-base status. Potassium is the major intracellular cation—plasma potassium may reflect total body potassium poorly. At the time of surgery, adrenergic stress may mobilize potassium from the intravascular to the intracellular space. Additional GI fluid losses will produce hypokalemia, unless the potassium deficit is adequately replaced. Hypokalemia interferes with muscle contractility; thus ileus and generalized weakness are common manifestations of hypokalemia (potassium < 3.0 mmol/L). Significant hypokalemia is associated with an increased risk of cardiac arrhythmias and can predispose to digitalis toxicity. Treatment of hypokalemia involves potassium replacement, and this must be performed cautiously (10-20 mEq/h) to avoid hyperkalemic complications. Hyperkalemia may occur through excess ingestion or intravenous administration, but more frequently occurs through renal impairment. The most lethal manifestations of hyperkalemia are cardiac conduction abnormalities, including prolongation of the PR interval, decrease in P-wave amplitude, and widening of the QRS complex resulting in ventricular fibrillation or asystole. Cardiac effects are negligible when the potassium level is less than 6 mEq/L. Indications for treatment include the presence of ECG changes or when the serum concentration of potassium is greater than 7 mEq/L. Treatment of hyperkalemia involves eliminating exogenous potassium, reversing myocardial membrane hyperexcitability, and removing potassium from the body. Acute therapy includes administration of calcium, insulin and glucose, sodium bicarbonate, diuretics, or cation exchange resins (via the GI tract). Dialysis should be reserved for patients with renal failure or life-threatening hyperkalemia resistant to conventional treatment.
Approximately 50% of serum calcium is free, whereas the remainder is complexed, primarily to albumin. Hypoalbuminemia, as seen in many ovarian cancer surgery patients, alters total serum calcium concentration; clinical decisions should thus be based on ionized calcium levels. If ionized calcium cannot be measured, the total serum calcium can be corrected by adding 0.8 mg/dL for each 1.0 g/dL that the serum albumin is below 4.0 g/dL. In the perioperative patient receiving multiple transfusions, hypocalcemia may be attributable to chelation by citrate. Hyperphosphatemia may precipitate calcium or decrease intestinal absorption of calcium. Hypomagnesemia may suppress the production of parathyroid hormone. Clinical manifestations of hypocalcemia (ionized calcium < 0.7 mmol/L) include neuronal membrane irritability and tetany. Acute management includes replacement of calcium intravenously and correction of other electrolyte abnormalities. Hypercalcemia occurs most commonly with bone resorption, when calcium enters the extracellular volume more rapidly than can be excreted by the kidneys. Hypercalcemia (total serum calcium > 13 mg/dL or ionized calcium > 1.3 mmol/L) is less common in the perioperative gynecologic oncology patient; bone metastases, although uncommon in gynecologic oncology patients, should be considered in the differential of persistent hypercalcemia as well as a paraneoplastic syndrome.
Magnesium also plays an important role in neuronal conduction. Although hypomagnesemia is common in perioperative gynecologic oncology patients, symptoms are uncommon unless serum magnesium is less than 1.0 mg/dL, at which time patients may have symptoms of weakness, lethargy, muscle spasms, paresthesias, and depression. Causes of hypomagnesemia may include excessive losses through the GI tract or inability of the kidneys to conserve magnesium. The sodium–potassium pump is magnesium dependent; attempts to correct potassium deficits may not be successful unless the magnesium deficit is simultaneously corrected. Treatment is by replacing magnesium intravenously, with reduced doses given in patients with renal insufficiency. Most cases of hypomagnesemia are iatrogenic and will correct with urinary excretion.
Phosphate provides the primary energy bond in adenosine triphosphate, is an essential element of second messenger systems, and is a major component of cellular membranes and nucleic acids. Significant phosphate depletion also results in cellular energy depletion. Severe hypophosphatemia (< 1 mg/dL) usually indicates total body phosphate depletion and may manifest with paresthesias, muscle weakness, malaise, encephalopathy, seizures, and coma. Moderate hypophosphatemia (1-2.5 mg/dL) is usually attributable to renal losses and a decrease in GI absorption. Hypophosphatemic patients are often hypokalemic and hypomagnesemic. Intravenous administration of phosphorus should also be given cautiously to patients with renal dysfunction or hypocalcemia. Hyperphosphatemia in the postoperative patient is usually due to decreased renal excretion.
Prompt recognition and treatment of acid-base disturbances and electrolyte imbalances are important to the homeostasis of postoperative patients. A pH of 7.4 refers to the normal hydrogen concentration, which is maintained in balance with arterial partial pressure of carbon dioxide (PaCO2) and bicarbonate (HCO3–). Acidemia can result from either an increased PaCO2 or a decreased concentration. Alkalemia can result from either decreased PaCO2 or increased . Most of the time, the clinical process is mixed.
Metabolic acidosis results from a decrease in HCO3– (< 21 mEq/L) due to either loss of bicarbonate or accumulation of acid. Loss of HCO3– may be through diarrhea, biliary drainage, urinary diversion, or renal tubule losses—hyperchloremic metabolic acidosis is associated with a normal anion gap (calculated as: ). A high anion gap may be due to excess production of acids (lactic acidosis or ketoacidosis), increased retention of waste products (sulfate or phosphate), or ingestion of toxins (salicylic acid, ethylene glycol, or methanol). A compensatory response is seen through hyperventilation and a decrease in PaCO2. In the postoperative gynecologic cancer patient, metabolic acidosis may be most commonly due to GI fluid losses and renal failure. Treatment of metabolic acidosis should focus on correcting the underlying metabolic condition. In a mechanically ventilated patient, a compensatory hyperventilation should be included in ventilator settings. Administration of sodium bicarbonate or other alkalinizing agents should be reserved for severe acidemia.
Metabolic alkalosis results from either a loss of hydrogen (H+) or a gain in HCO3– (> 27 mEq/L). Loss of H+ may be through nasogastric suction or diuretic administration. The reabsorption of HCO3– in the distal renal tubules results in hypokalemia and hypovolemia, resulting in a so-called “contraction alkalosis.” Treatment of metabolic alkalosis therefore includes the replacement of volume and electrolytes. Fluid resuscitation with lactated Ringer’s solution may be advantageous versus normal saline because HCO3– can be generated from lactate.
Respiratory acidosis is characterized by hypercarbia , which occurs when ventilation is insufficient to eliminate carbon dioxide. Over time, the kidneys compensate by excreting H+ and retaining HCO3–. Postoperative patients are particularly at risk of respiratory depression with upper abdominal incisions or when receiving opiates for perioperative pain management. Supplemental oxygen minimizes the incidence of hypoxia, even when decreased ventilation increases the risk of hypercarbia. Severe respiratory acidosis is an indication for intubation and ventilatory support.
Respiratory alkalosis occurs with increased ventilation, causing hypocarbia . In a mechanically ventilated patient, overbreathing can also result in respiratory alkalosis. In the postoperative patient on the floor, pain, anxiety, central nervous system disease, and sepsis may all be causes of hyperventilation. An active approach may include interventions of sedation and reassurance if anxiety is the cause of hyperventilation.
Infectious Disease Issues
Postoperative fever is a common occurrence with potentially serious implications. Although infection must be considered, fever may also be a result of tissue inflammation. An accepted definition of fever in the postoperative period is a temperature elevation of greater than 38°C to 38.5°C in a 24-hour period. Reported incidences of postoperative fever range from 15% to 47%, with a source of infection identified in only 5% to 36% of patients and bacteremia identified in less than 3% of patients. A traditional “fever work-up” of complete blood count, urine culture, blood culture, and CXR may be an inefficient use of resources.
Following gynecologic surgery, the incidence of postoperative fever can be as high as 75%,56 with increased occurrence in the gynecologic oncology population. However, the actual incidence of infection is low. Patients undergoing radical pelvic procedures, including radical hysterectomy and debulking, were found in a recent series to have a higher rate of postoperative fever, but no increased incidence of infection.57 A separate series of 194 patients undergoing exploratory gynecologic surgery also found that surgery for malignancy, bowel resection, and higher postoperative fever or white blood cell count are associated with the presence of significant infection56 and should be considered for a thorough evaluation, including complete blood count, urine culture, blood culture, and CXR. However, empiric evaluation of all patients may not be cost effective. Within the first 3 days after surgery, postoperative fevers are often attributable to atelectasis. Clinical history is important to help assess the patient’s overall risk for postoperative infection. Length of surgery, blood loss, surgical contamination, pre-existing infection, nutritional status, immunocompromised state, and presence of malignancy all contribute to patient risk. Targeting cultures and laboratory tests to the population at highest risk helps to minimize excessive testing of low yield.
Foreign bodies can be a nidus for infection. Drains including Foley catheters, ureteral stents, and pelvic drains should be removed as soon as appropriate. Use of antibiotic-coated or antiseptic impregnated central venous catheters has been found to decrease the incidence of catheter colonization and catheter-related blood infections. Intra-abdominal and pelvic infections can be insidious in onset. Unfortunately, imaging studies may not always provide definitive evidence of infection, because the presence of free air on x-ray may persist for several days after abdominopelvic surgery. Furthermore an abscess may take time to consolidate before CT drainage can be undertaken.
Suspicion for intra-abdominal infection or systemic sepsis should prompt the initiation of broad-spectrum antibiotics. One-quarter of patients with clinical suspicion of sepsis will not be documented on microbiology; these patients have similar predisposing risk factors as those with positive cultures and also a similar risk for death. Antibiotics are essential for the management of septic shock but, unfortunately, are not sufficient for optimal treatment. Airway management, fluid management, and correction of hypoperfusion all contribute to maintaining organ function. Early recognition of systemic inflammatory response and organ dysfunction may identify patients who may benefit from recombinant activated protein C (drotrecogin alfa) but with a higher risk of significant bleeding. Serious bleeding occurred primarily in patients predisposed to bleeding, including patients with coagulopathy, severe thrombocytopenia (< 30,000/μL), GI bleeding, or trauma. However, in a prospective, randomized, multicenter trial, use of drotrecogin alfa in septic patients with an APACHE score < 25 or single-organ failure did not benefit from therapy; thus, the use in postsurgical patients must be considered carefully based on severity of illness.58
Selection of antimicrobial agents usually involves empiric and broad-spectrum agents. Wound infections are frequently due to staphylococcal or streptococcal organisms, which are frequently sensitive to penicillins or first-generation cephalosporins; clindamycin provides coverage of gram-positive organisms in patients with penicillin allergies. Intra-abdominal organisms such as Escherichia coli or Bacteroides fragilis may be treated with piperacillin with tazobactam, with the possible addition of an aminoglyco-side. More virulent gram-negative rods may respond better to a third- or fourth-generation cephalosporins, a β-lactam/β-lactamase inhibitor, or a carbapenem. Fungal organisms may require treatment with fluconazole or caspofungin. If non-albicans Candida is suspected, caspofungin may preferred over fluconazole. Treatment should be continued for 14 days after bacteremia or fungemia is cleared.
Antimicrobial resistance is a serious concern, with increasing resistance to several classes of antibiotics. Methicillin-resistant Staphylococcus aureus is increasingly recognized as a cause of nosocomial bloodstream infection. Vancomycin-resistant Enterococcus is another common and difficult to treat hospital-acquired infection. Limiting the use of drugs such as fluoroquinolones and increasing the treatment dose to reduce the risk of mutant selection are strategies that may aid in maintaining the antibiotic armamentarium.
When pneumonia is suspected, the timeline of symptoms and risk factors for pneumonia must be considered, because the microbiology of hospital-acquired pneumonia is significantly different than that of community-acquired pneumonia. Severely ill patients have a higher risk of being colonized with gram-negative bacilli, and the bacteria are often polymicrobial. The usual presentation is that of fever, sputum production, evidence of pulmonary consolidation on physical examination, and a localized infiltrate on x-ray. However, the clinical picture may vary, especially in critically ill patients, because fevers or leukocytosis may be from one of several sources and infiltrates on CXR may be confounded by atelectasis or malignant pleural effusions. Gram stain is more valid as an index of pulmonary infection because most cultures will reveal airway flora. The antibiotic used for surgical prophylaxis should be considered as one to which the organism may be resistant. Second- or third-generation cephalosporins are effective regimens, as are β-lactamase inhibitors or fluoroquinolones.
The perioperative setting in the hospital is associated with a high risk for acute renal failure. The overall incidence is highest in patients undergoing cardiac or vascular surgery, although patients with pre-existing renal disease, hypertension, cardiovascular disease, diabetes, and advanced age are all considered to be at higher risk. Elderly patients have a lower glomerular filtration rate and are more susceptible to volume depletion as well as other nephrotoxic insults. Mortality from acute renal failure remains high despite advancements in renal replacement therapy, perhaps because acute renal failure is often managed in the setting of multiorgan failure.
Traditionally, the evaluation of acute renal failure includes consideration of prerenal, intrarenal, and postrenal causes (Table 18-6). Prerenal injury to the kidney can be caused by ischemic or hypotensive insult, resulting in acute tubular necrosis. Full renal recovery can occur after several days, barring any further injury. Renal injury from nephrotoxic drugs is unusual in a healthy well-hydrated patient, but an older, volume-depleted woman with chronic renal insufficiency undergoing gynecologic cancer surgery is more vulnerable to renal injury. Attention should be paid to the use of nonsteroidal inflammatory drugs, amino-glycoside antibiotics, angiotensin-converting enzyme inhibitors, and angiotensin II receptor blockers, all of which may lead to renal injury. Radiographic contrast is also nephrotoxic, although this risk can be reduced with the administration of intravenous sodium bicarbonate or N-acetylcysteine before and after administering the contrast dye. Postrenal causes of renal failure include urinary retention and ureteral injury. Radical pelvic dissection in gynecologic cancer debulking can often result in bladder dysfunction. Although a thoracic-level epidural should not impede bladder function, a recent small series of female patients undergoing urologic surgery showed a high rate of postoperative urinary retention in the setting of thoracic epidural use.59 In addition, high doses of opiates for pain control may delay spontaneous voiding. If ureteral obstruction is suspected, a transient rise may be seen in the serum creatinine level. A renal ultrasound may confirm hydroureteronephrosis, but a nondilated collecting system does not exclude the possibility of obstruction. For patients with a high index of clinical suspicion, a CT urogram may evaluate the entire urinary tract and also identify any associated postsurgical findings. Isolated renal failure may recover, but given that renal failure is often caused by a syndrome of multiple organ failure associated with hemodynamic compromise, dialysis or hemofiltration may be necessary.
Table 18-6 Common Causes of Acute Renal Failure in the Surgical Setting
Prerenal acute renal failure can be reversible if renal perfusion is maintained and additional renal insults are avoided. In administering the patient’s medications, dosing of renally excreted drugs may require dose adjustment for creatinine clearance. Urinalysis, urine microscopy, and urine electrolytes may be helpful in the distinction between prerenal and intrarenal processes. In renal failure due to prerenal causes, the specific gravity will be high (> 1.020), urine sodium will be low (< 20 mmol/L), and the fractional excretion of sodium will be less than 1%. In the setting of acute tubular necrosis, granular or epithelial cell casts may be seen on the urinalysis. The treatment of acute renal failure is mainly supportive, with treatment of the underlying cause of renal failure and correction of fluid and electrolyte imbalances. Low-dose dopamine (1-5 μg/kg/min) has not been shown to be effective in protecting or improving renal function. Loop diuretics are commonly used for converting oliguric to nonoliguric renal insufficiency, although response may be only a demonstration of less severe kidney damage. In severe renal failure, large doses of diuretics may be required, and recent studies have suggested that a furosemide infusion may be more effective than bolus therapy.
When planning renal replacement therapy, management of volume and solute toxicity are equal goals. Intermittent hemodialysis is the most common approach for renal replacement therapy. Indications for dialysis in the acute setting include severe fluid overload, hyperkalemia, metabolic acidosis, and uremia. In a stable patient with acute renal failure, hemodialysis allows for the rapid removal of fluids and toxic metabolites. In a critically ill patient after surgery, hypotension from sepsis or multiorgan failure often precludes the use of intermittent hemodialysis. Continuous hemofiltration or hemodiafiltration allows for a slower rate of fluid removal and theoretically results in fewer hemodynamic shifts in the unstable ICU patient.60 In continuous venovenous hemofiltration, vascular access is achieved through a central vein, and anticoagulation with heparin is given to maintain the extracorporeal circuit. Despite these new technologies, however, renal replacement is only a means of support, and ultimate mortality reduction still requires recovery of the kidneys and other affected organs.
A cerebrovascular event is one of the most devastating neurologic complications of surgery; thankfully, the incidence of a stroke following noncardiac abdominal surgery is quite low and is estimated to be less than 1%.61 Patients with a history of prior stroke and/or transient ischemic attack can be at increased risk of postoperative stroke. Additional risk factors for stroke that are pertinent to gynecologic cancer patients are listed in Table 18-7. Nearly 50% of strokes occur 24 hours after surgery, with the remainder occurring ≥ 2 days postoperatively. In the immediate setting of an embolic stroke, the use of thrombolytic therapy up to 4.5 hours after the event has been found to improve functional outcomes.62 Although recent surgery is considered a contraindication to thrombolytic therapy, it has been performed safely in patients undergoing cardiac surgery. However, such an intervention should be emergently discussed with a neurologist.
Table 18-7 Risk Factors for Perioperative Stroke
Injury to peripheral nerves during surgery for gynecologic cancer may present with symptoms in the immediate or delayed postoperative period. The incidence of such injury is quite low; a recent prospective study of more than 600 patients undergoing gynecologic surgery found an overall rate of 1.8%, with the majority of injuries involving the femoral and lateral femoral cutaneous nerves.63 In general, the majority of such neuropathies will resolve. In a series of 1210 patients, the only patients to not achieve full recovery were those with an unrepaired nerve transection or an injury to the lumbosacral nerve plexus.
Delirium is a common event in the postoperative setting; the overall incidence following noncardiac surgery is approximately 10%, although studies from the various surgical specialties have estimated the incidence to be 3% to 70%.64 Given that many patients with gynecologic malignancies are elderly, they are particular susceptible to this issue. In contrast to dementia, which is often a chronic process, delirium is characterized by the acute onset of mental status change, accompanied by hallucinations, lack of orientation, and poor short-term memory. Risk factors for postoperative delirium include age, preoperative use of psychotropic medications, poor nutritional status, and preoperative impairment of cognitive and/or functional status. In addition, medications can precipitate delirium, as can metabolic disturbances, infection, dehydration, immobility, and malnutrition. In a recent prospective study of 103 patients on a gynecologic oncology service in which all patients underwent a pre- and postoperative Mini-Mental Status Examination, the overall incidence of postoperative delirium was 17.5%. Risk factors for delirium included age > 70 years, use of more than 5 medications, and the need for additional narcotics for pain.65 Interventions to minimize delirium are to orient the patient with cognitive stimulation, consider ambulation with minimal use of physical restraints, removing catheters and drains when medically appropriate, visual and hearing aids, and volume replacement for patients with dehydration. Undertreated pain can cause agitation, yet opiate use can precipitate delirium as well. Pain medications for modest surgical incisions can be started with acetaminophen, reserving opiates for breakthrough pain medications. Involving family members in postoperative care can help keep women recovering from ovarian cancer surgery oriented and active. In addition, the use of antipsychotic medications, as well as donepezil (Aricept), has been found to lessen the severity and time course of postoperative delirium.66
Red blood cells are transfused for anemia and to increase the oxygen-carrying capacity of the patient (Table 18-8). Fresh frozen plasma is separated from whole blood and frozen within 6 hours of collection to minimize loss of coagulation factors V and VIII. Fresh frozen plasma is indicated for coagulopathy from transfusion and correction of warfarin effect. Platelets are usually pooled from 6 units of whole blood and transfused to treat thrombocytopenia or deficits of platelet function.67 Typically, spontaneous bleeding is uncommon with platelet levels about 20,000/dL, although patients undergoing major operative procedures should generally have platelet counts of about 50,000/dL to 70,000/dL for effective hemostasis. Cryoprecipitated antihemophilic globulin is also pooled from whole blood and serves as a therapeutic source of fibrinogen. Although cryoprecipitate is rich in factor VIII, the treatment of choice for hemophiliacs currently is factor VIII concentrate. Current blood bank practice involves screening for disease transmission of syphilis, hepatitis B, hepatitis C, human immunodeficiency virus (HIV-1/2), human T-lymphotropic virus (HTLV-I/II), Chagas disease, and West Nile virus.
Table 18-8 Contents of Blood Products
Adverse reactions to transfusion may include febrile reactions that may be caused by recipient antibodies to leukocyte antigens reacting to leukocyte fragments in the transfused blood.67 These reactions are most commonly seen in patients with a history of multiple blood transfusions or pregnancies, which can stimulate the development of leukocyte antibodies. Allergic urticarial reactions are seen in approximately 1 in 100 transfusion recipients. This reaction is likely caused by foreign plasma proteins. Premedication with acetaminophen and diphenhydramine may help minimize febrile and allergic reactions. Fever is also the most frequent manifestation of acute hemolytic transfusion reactions. These reactions are most likely to occur when a group O patient is mistakenly transfused with group A, B, or AB blood. Symptoms may include fevers, chills, chest tightness, tachycardia, hypotension, and hemoglobinemia, with subsequent hemoglobin-uria and hyperbilirubinemia. If a transfusion reaction is suspected, the transfusion must be stopped and supportive measures undertaken. Fluid and diuretic therapy may be indicated, and blood specimens from the patient and the transfused blood product should be collected to confirm hemolysis. Delayed hemolytic reactions may occur in patients who have developed antibodies from prior transfusion and may not manifest until 4 to 8 days after transfusion. Such delayed reactions may not be detected as the red cell destruction occurs slowly, and they are diagnosed only with a decreasing hematocrit and positive direct antiglobulin (Coombs) test. Transfusion-related acute lung injury (TRALI) is a rare (1 in 1333 to 1 in 5000) complication of transfusion manifested by abrupt noncardiogenic pulmonary edema. This reaction is likely associated with the presence of donor antibodies reactive to recipient leukocyte antigens or with inflammatory mediators in stored blood components; it is most commonly seen after transfusion with fresh frozen plasma or whole blood–derived platelet concentrates.68Severe cases may require ventilator support but usually resolve within 72 hours. Donors who have been implicated in a case of TRALI are usually deferred from future blood donation.
The popularity of autologous and designated donor blood programs has increased while the estimated rates of transfusion-related infections have decreased. Autologous and designated donor programs vary by institution but have been proposed for procedures where the mean transfusion requirement exceeds 1 unit, where more than 10% of patients undergoing the procedure require transfusion, and where the anticipated blood loss is greater than 20% of the patient’s estimated blood volume. In most cases, however, it is probably not cost effective because surgical blood loss is difficult to predict.
The effects of anemia must be separated from those of hypovolemia. The likelihood of ongoing bleeding, the presence of underlying coagulopathy, and the preexisting cardiovascular condition of the patient should be considered prior to transfusion. Multiple studies comparing conservative versus liberal transfusion strategies in critically ill patients have demonstrated that red blood cell transfusion is an independent predictor of death, associated with increased risk of infection, organ dysfunction, and ARDS.69 In a nonbleeding patient with a hemoglobin concentration greater than 7.0 g/dL and without cardiac risk factors, the benefits of red blood cell transfusion are limited. An exception may be the patient who is going to receive chemotherapy soon after surgery.
Physical examination of the perioperative coagulopathic patient should include an assessment of bleeding from an anatomic site versus continued consumptive coagulopathy. In particular, there are large surfaces of tumor beds following debulking surgery in ovarian cancer where oozing may continue until the coagulation disorder is corrected. Inadequate surgical hemostasis may be considered if the patient’s condition appropriately responds to transfusion but then deteriorates again. If the physical examination findings suggest a hematoma in a self-limited space, the bleeding should be self-limited, and transfusion should be continued to support the self-limited process. If the physical examination is consistent with a bleeding pedicle (increasing tense abdomen, gross blood in an abdominal drain), re-exploration may be necessary to control hemostasis.
Deep venous thrombosis and pulmonary embolism continue to be the leading causes of postoperative morbidity and mortality in surgery. Gynecologic oncology patients are particularly at risk given the risk factors of malignancy, older age, longer surgical procedures, limited mobility after abdominal surgery, and frequent dissection around pelvic vessels, which are then prone to intimal injury. Clear cell ovarian cancers, in particular, are more commonly associated with venous thromboembolism when compared to other histologies.70 In a prospective study of gynecologic cancer patients undergoing major radical surgery, the incidence of deep venous thrombosis detected by 125I-fibrinogen scanning was 38%, the majority of which were evident within 24 hours of surgery. Only 10% of deep venous thromboses were clinically evident, and bilateral disease was present in 20% of cases. The clinical diagnosis of pulmonary embolism is equally inaccurate, with 70% of patients with fatal pulmonary embolism diagnosed at autopsy. A high index of suspicion is critical in the gynecologic oncology postoperative population, and prophylactic measures should be universally used.
Prophylactic interventions can decrease the incidence of deep venous thrombosis by 50% (16%-8%) and fatal pulmonary embolism by 75% (0.4%-0.1%).71 Consensus panel guidelines from the American College of Chest Physicians recommend routine prophylaxis of gynecologic surgery patients with low molecular weight heparin once to twice daily.35 Alternatively, low-dose UFH may be used, in combination with pneumatic compression devices (Table 18-9).
Table 18-9 American College of Chest Physicians Guidelines for Thromboprophylaxis in Gynecologic Oncology Patients
• Low molecular weight heparin daily or every 12 hours
• Low-dose unfractionated heparin 5000 units every 8 hours
• Intermittent pneumatic compression devices started before surgery and used continuously while the patient is not ambulating
• Alternative consideration: Combination of heparin and mechanical thromboprophylaxis, or fondaparinux
The duration of thromboprophylaxis should continue until discharge from the hospital and up to 28 days after discharge with low molecular weight heparin.
Adapted from Geerts WH, Bergqvist D, Pineo GF, et al. Prevention of venous thromboembolism: American College of Chest Physicians Evidence-Based Clinical Practice Guidelines (8th Edition). Chest. 2008;133(6 suppl):381S-453S.
Compared to UFH, low molecular weight heparins have a better anticoagulation profile, based on better bioavailability, longer half-life, dose-independent clearance, and decreased binding to plasma proteins and endothelial cells. They have less antithrombin activity, more anti–factor Xa activity, and less effect on partial thromboplastin time. They also have less platelet inhibition and do not increase microvascular permeability; therefore, fewer bleeding complications are seen with their use. Patients with renal impairment may have increased serum levels after enoxaparin administration, and its dosage should be adjusted downward when creatinine clearance is less than 30 mL/min.
Venous thromboembolic events may be minimally symptomatic or attributed to other perioperative processes. Classic clinical findings of deep venous thrombosis include leg swelling, pain, increased warmth, and erythema. Doppler ultrasonography is the most commonly used diagnostic procedure to diagnose deep venous thrombosis (Figure 18-1).
FIGURE 18-1. Diagnostic algorithm for evaluation of suspected postoperative deep venous thrombosis. CT, computed tomography; DVT, deep venous thrombosis; MRI, magnetic resonance imaging.
Particularly after pelvic surgery, the Doppler ultrasound is limited to evaluation of veins below the inguinal ligament. In the situation of a negative or indeterminate study with high clinical suspicion, contrast venography still remains the standard. Negative quantitative enzyme-linked immunosorbent assay D-dimer assays combined with negative noninvasive imaging have been found to have high negative predictive value in outpatients, but these results are not yet confirmed to have validity in cancer patients. A negative test in a high-risk patient does not exclude pulmonary embolism, and a positive D-dimer in a postoperative coagulopathic patient may not be specific either.
Pulmonary embolism is the most serious potential consequence of deep venous thrombosis. Classic clinical findings include hypoxia, chest pain, hemoptysis, shortness of breath, and tachycardia. In a critically ill postoperative ovarian cancer patient, the differential diagnosis is broad and may include fluid overload, pneumonia, effusion, atelectasis, or a cardiac event. The most common findings on CXR include atelectasis, infiltrate, and small effusion, all of which are nonspecific in a perioperative setting. Findings in ECG may include ST-segment or T-wave changes, but few patients will have electrical changes suggestive of right heart strain. An arterial blood gas may help determine the degree of hypoxia and hypercapnea, although a quarter of patients with acute pulmonary embolism will have a normal PaO2. A CT angiogram, or spiral CT, can reliably diagnose most clinically significant pulmonary emboli but is less sensitive in evaluating the subsegmental vessels (Figure 18-2). A ventilation/perfusion scan has the advantage of not using iodinated contrast but has the disadvantage of determining only indirect evidence of pulmonary embolism. In the Prospective Investigation of Pulmonary Embolism Diagnosis (PIOPED) study72, only 14% of patients had a high probability scan, whereas 77% of patients had an indeterminate, nondiagnostic study. Pulmonary angiography remains the standard of diagnosis for pulmonary embolism but is considered rather invasive by requiring catheterization of the main pulmonary vessels and requiring a considerable contrast dye, which is particularly significant in a patient population whose renal function may be compromised from intravascular depletion or acute tubular necrosis. For a critically ill patient in the ICU, transesophageal echocardiography may also be performed at the bedside to evaluate for pulmonary embolism. Although transesophageal echocardiography provides poor visualization of the left pulmonary and lobar arteries, it is able to identify right ventricular volume and pressure overload associated with pulmonary embolism.
FIGURE 18-2. Diagnostic algorithm for evaluation of suspected postoperative pulmonary embolism. CT, computed tomography; DVT, deep venous thrombosis; PE, pulmonary embolism; VTE, venous thromboembolism.
Clinically, the most practical examination may be spiral CT pulmonary angiography. Combining spiral CT of the chest with indirect CT venography may also exclude the need for lower extremity Doppler studies if either a pulmonary embolism or deep venous thrombosis is identified. Although a contrast dye load is still required, it is not as invasive and allows for evaluation of pelvic and lower extremity thrombi with greater than 90% sensitivity and specificity.
Heparin remains the primary therapy for postoperative patients with venous thromboembolic events. Failure to rapidly anticoagulate a patient with an acute event has been associated with late recurrence of thrombosis. However, the priority to anticoagulate a patient must be tempered with the knowledge of the patient’s postoperative condition and the potential for ongoing bleeding or coagulopathy. If an epidural catheter is in place, it should be removed before the initiation of anticoagulation if the patient is stable enough to wait 1 hour before starting heparin. The short half-life of intravenous UFH allows for rapid reversal of anticoagulation in a patient at high risk of bleeding or requiring an invasive procedure (Figure 18-3). Once the patient is hemodynamically stable and therapeutically anticoagulated with heparin, low molecular weight heparin or warfarin can be started for a course of long-term anticoagulation. Patients with active bleeding may benefit from an IVC filter, which may prevent new or recurrent pulmonary emboli. Although the placement of a filter reduces the acute risk of pulmonary embolism, the filter may still become occluded, and once stable, patients typically require lifelong anticoagulation to minimize their later risk of recurrent deep venous thrombosis.
FIGURE 18-3. Anticoagulation of acute deep venous thrombosis or pulmonary embolism with high bleeding risk.
Several meta-analyses of clinical trials comparing low molecular weight heparin to UFH for the treatment of patients with deep venous thrombosis have shown that low molecular weight heparin provides greater reduction of thrombus size, as well as a decrease in recurrent venous thromboembolism.73 Low molecular weight heparin also appears to be as safe and effective as UFH in the management of pulmonary embolism. The dose of low molecular weight heparin depends on the indications of prevention or treatment. A prophylactic dose of dalteparin is 5000 IU 10 to 12 hours before surgery and once daily after surgery. A prophylactic dose of enoxaparin is 40 mg 2 hours before surgery and once daily after surgery, whereas a therapeutic dose for anticoagulation is 1 mg/kg every 12 hours. Dalteparin was found to be comparable to tinzaparin in a randomized clinical study of treatment for deep vein thrombosis and pulmonary embolism. Monitoring is not required for prophylactic doses or in most patients for treatment, but measuring the anti–factor Xa level can be considered for patients with renal insufficiency or morbid obesity or patients who are refractory to therapy.
Arterial thrombotic events are rare occurrences in gynecologic cancer surgery patients; however, arterial emboli require prompt recognition as a vascular surgery emergency. Hypercoagulability, tissue trauma, and patient immobilization are risk factors that may exacerbate underlying vascular disease. Classic clinical signs include pain, pulselessness, paresthesia, pallor, and paralysis. Limb ischemia may lead to muscle necrosis and compartment syndrome within hours after onset. Prompt diagnosis and rapid revascularization allow the best opportunity for limb salvage and avoidance of amputation.
In general, GI prophylaxis against stress ulcer formation is recommended to patients in the ICU with coagulopathy, patients with a history of recent upper GI bleed, or those who have been ventilated for ≥ 48 hours.74 However, GI prophylaxis is often used inappropriately, as the frequency of GI bleeding outside of the ICU is quite low, and inappropriate use may result in increased rates of nosocomial pneumonia and C difficile–associated diarrheal illness. In the comparison of H2 antagonists versus proton pump inhibitors, a recent meta-analysis showed these medications to be equivalent as prophylaxis against stress ulcers.75
Following surgery for advanced gynecologic malignancies, a nasogastric tube is sometimes left in place to decompress the stomach and avoid stress on gastric vascular pedicles. Such patients may be at risk of aspiration of gastric contents and should receive acid-reducing medications.
Following gynecologic oncology surgery, intravenous and/or epidural administration of analgesic medications may be used for pain control. An intravenous patient-controlled analgesia machine is an effective means of delivering opioids to treat postoperative pain. Early oral analgesia in gynecologic oncology patients is also safe and efficacious, but this depends on the use of early postoperative feeding. Epidural analgesia is also an effective therapy for the management of pain after major abdominal surgery and has been associated with decreased intraoperative blood loss, fewer thromboembolic events, and early ambulation.
More recently a prospective randomized study from Memorial Sloan-Kettering Cancer Center compared perioperative patient-controlled epidural analgesia (morphine and bupivacaine) to postoperative intravenous patient-controlled analgesia (morphine) in open gynecologic surgery.76 Patients in the epidural arm had significantly less postoperative pain with no difference in time to discharge. In another nonrandomized prospective study, use of perioperative patient-controlled epidural analgesia (fentanyl and bupivacaine) did not improve pain management over postoperative patient-controlled analgesia (hydromorphone).77 However, in this study, patients selected for epidurals were more likely to have a diagnosis of cancer and more complicated surgical procedure. With no difference in time to ambulation, tolerating diet, and readiness for discharge, the authors suggest that patient selection for epidural may influence the beneficial effects of neuraxial blockade.
Patients should be cautioned on signs of wound infection, such as purulent drainage, erythema, or increased pain at the surgical site. Patients at increased risk of wound complications, such as the obese, diabetics, and the immunosuppressed, should take extra caution. In the setting of infection, hematoma, and/or seroma, large laparotomy incisions may open and require closure by secondary intention. Applying negative pressure to wounds, via the placement of a vacuum-assisted closure device, is a recent and novel means of debriding the wound while allowing for closure by secondary healing.78 Following surgery for vulvar and vaginal cancer, meticulous care of the surgical site is a key factor in postoperative wound healing; issues such as sitz baths and perineal care should be addressed at multiple times during the perioperative period. Such supportive care includes strict instructions to keep the surgical site dry with heat, blow dryers, and the use of gauze fluffs. When inguinofemoral lymphadenectomy is performed, the use of drains in the lymphatic space has been found to decrease the risk of wound hematoma, seroma, or infection; however, patients should have necessary arrangements for home drain management when they leave the hospital.
The wide variety of procedures performed for patients with gynecologic malignancies can include leaving patients with urostomies and/or temporary or permanent fecal diversion stomas. In the postoperative period, patients can benefit from education on avoiding complications of stomas, such as the use of barrier systems for skin protection. In addition, patients with ileostomies or ascending colostomies should receive information on fluid intake and diet in order to avoid long-term complications with fluid and electrolyte balance. Patients with colostomies should be encouraged to learn techniques to avoid constipation.79
Chronic lymphedema may occur following both pelvic and inguinal lymphadenectomy. In the setting of inguinal lymphadenectomy, early wound complications are not predictive of the incidence or severity of chronic lymphedema. Those with early and chronic lymphedema should be counseled on “complete decongestive therapy” (CDT), which includes manual lymphatic drainage, compression wraps and garments, skin care, and exercises to facilitate lymphatic drainage. Nurses and physicians may play critical roles in informing patients not only of the signs of lymph-edema, but also early techniques of CDT.
The clinical team involved in the disposition of a patient following surgery for gynecologic cancer can include not only physicians, but also case managers, social workers, and physical and occupational therapists. Given the complexities that can arise due to the various procedures performed for women with gynecologic cancers, patients may benefit from a period of subacute nursing care. Attention should also be paid to the postoperative follow-up plan in which the pathology and treatment plan will be discussed in order to smooth the transition between the acute postoperative period and subsequent outpatient oncology care.
Following open surgery, patients have traditionally been restricted from oral intake until they have signs of improved bowel function. However, several studies have demonstrated that early feeding after laparotomy may actually improve GI function and overall perioperative outcome.
Placement of a nasogastric tube intraoperatively does not necessarily mean that the patient continues with nasogastric intubation in the postoperative period. Patients frequently complain of discomfort from the tube, and although traditional surgeons argue that nasogastric decompression decreases gastric and intestinal distention, these benefits have not been proven in clinical trials. In a prospective study of 110 gynecologic oncology patients undergoing intra-abdominal surgery randomized to postoperative nasogastric tube or intraoperative orogastric tube, there was no difference in bowel complications, time to tolerating a regular diet, or length of hospital stay. A meta-analysis of 26 clinical trials including 3964 patients evaluating nasogastric decompression also found that patients managed without nasogastric tubes had significantly less febrile morbidity, atelectasis, and pneumonia.80 There was greater abdominal distension and vomiting, but this was not associated with any increase in complications (wound dehiscence, infection, or anastomotic leak) or length of stay.
Early oral feeding was first suggested to be safe in GI and colorectal surgery. Studying the safety and efficacy of early postoperative feeding in gynecologic oncology patients, 200 patients were randomized to clear liquids on postoperative day 1 versus nothing by mouth until passage of flatus. Time to development of bowel sounds (1.8 vs. 2.3 days, ), tolerance of clear liquids (1.2 vs. 3.5 days, ), and regular diet (2.3 vs. 4.2 days, ) and length of hospital stay (4.6 vs. 5.8 days, ) were significantly shorter in the early feeding group. These findings were confirmed by a separate prospective randomized study at Indiana University comparing clear liquids on postoperative day 1 with nothing by mouth until bowel sounds, flatus or bowel movement, or subjective hunger. The study group had a higher incidence of emesis but actually tolerated a regular diet 1 day earlier than the control group. In a recent retrospective review of 880 patients on a gynecologic oncology service at University of California, Irvine Medical Center, early feeding was well tolerated.81 In this series, only 44 patients (5%) required readmission; 11 of these patients were readmitted with a diagnosis of ileus or small bowel obstruction. A recent prospective study of gynecologic oncology patients undergoing intestinal resection found that early feeding was not associated with any increase in postoperative complications.82
Beyond its digestive capacities, the GI tract is recognized as being an immunologic barrier against infection. Once tolerating a regular diet, patients may tolerate oral analgesics earlier.
Evaluation of Malnutrition
Malnutrition in gynecologic cancer patients may result from the physical compression of the bowel from tumor masses, omental caking, and ascites, resulting in decreased oral intake or partial bowel obstruction. In addition, cancer cachexia may cause metabolic changes, including increased resting energy expenditure, increased anaerobic glycolysis, and a high turnover of glycerol and free fatty acids. Patients who are nutritionally depleted are at greater risk of surgical morbidity, including increased risk of infection, prolonged hospital stay, and mortality. Patients with low serum albumin level on admission (< 3.4 g/dL) to an acute care hospital have been observed to have greater than a 3-fold increased risk for mortality when compared to patients with normal serum albumin. Whereas albumin levels represent the long-term nutritional status, prealbumin levels may represent short-term status.83 Weight loss and diminished skinfold thickness are also indicators of malnutrition. Although many studies have demonstrated that nutritional support improved the nutritional parameters of patients, strong data are lacking to demonstrate improvement in clinically significant end points. Malnutrition is assessed more effectively in the preoperative setting, because nutritional parameters such as albumin may be altered from drainage of ascites and volume resuscitation and transferrin may be altered due to postoperative inflammation.
Total Parenteral Nutrition
Total parenteral nutrition may be considered in patients who are expected to remain without enteral feeding for a prolonged period of time, but the timing of nutrition support is also influenced by the hemodynamic stability of the patient in the postoperative period. The altered metabolic state of increased catecholamines, glucocorticoids, and glucagon favors the process of gluconeogenesis, glycogenolysis, and fatty acid oxidation. Nutritional support has become a standard of care for these patients, yet the length of tolerable starvation remains an unanswered question. Total parenteral nutrition is not recommended for patients with an intact and functional GI tract, but many gynecologic oncology patients requiring critical care may have also had bowel resection and can be expected to have a delayed return of bowel function. A meta-analysis of 26 prospective randomized trials including 2211 patients comparing total parenteral nutrition with (standard) oral diet and intravenous dextrose did not show any improvement in overall mortality of surgical or critically ill patients (risk ratio, 1.03; 95% confidence interval [CI], 0.81-1.31). The rate of major complications was lower among malnourished patients receiving total parenteral nutrition (risk ratio, 0.52; 95% CI, 0.31-0.91).84
Caloric requirements are assessed based on the patient’s stored reserves and body catabolism. The Harris-Benedict equation is based on a basal energy expenditure calculated using the age, sex, height, and weight of the individual, and then multiplied by a ratio for stress and activity (Figure 18-4). A reasonable estimate of energy in a critically ill adult patient is 25 to 30 kcal/kg. Carbohydrates compose 60% to 70% of nonprotein calories, whereas fats constitute 25% to 30% of calories and include essential fatty acids. Protein needs range from 1.2 to 2.0 g/kg per day. Fluids, electrolytes, vitamins, and trace elements complete the daily formulation of total parenteral nutrition (Table 18-10). H2-receptor antagonists and regular insulin are also compatible as additives in parenteral nutrition solutions. Central administration of parenteral nutrition is recommended due to hyperosmolarity and vein irritation. Concurrent administration of fat emulsion may reduce the amount of vein irritation.
Table 18-10 Standard Parenteral Nutrition Formulations per Liter
FIGURE 18-4. Total estimated calorie requirement is based on the basal energy expenditure and stress and injury factors.
Refeeding syndrome may occur in the feeding of severely malnourished patients. Intracellular incorporation of phosphate may result in severe hypophosphatemia and possibly respiratory failure. Potassium and magnesium shift intracellularly as well, resulting in hypokalemia and hypomagnesemia. Regular monitoring of laboratory profiles, including electrolytes, glucose, liver function, and lipid panels, can prevent metabolic complications of parenteral nutrition. Excess glucose can result in hyperglycemia and other hyperosmolar states. Excess lipids can result in hyper-lipidemia, and excess protein can worsen azotemia or encephalopathy. Albumin, prealbumin, and C-reactive protein may be monitored weekly. Adjustments to rates of additional maintenance fluids given should also be made accordingly.
A 24-hour urine urea nitrogen excretion can be collected after stable protein intake for 3 to 5 days to calculate the patient’s nitrogen balance. Nitrogen intake equals protein intake (in grams) divided by 6.25. Nitrogen output equals urine urea nitrogen plus insensible loses (constant of 3). A nitrogen balance of +4 to +5 g is required for anabolism; nutrition specialists should be involved to help optimize parenteral nutrition orders. Calorie counts may help quantitate the patient’s intake, and parenteral nutrition should be continued until the patient is taking at least 50% of calories through the enteric route.
Peripheral Parenteral Nutrition
Although small studies have suggested some benefit to peripheral parenteral nutrition when central venous access is not available, there is no evidence suggesting that peripheral parenteral nutrition is equivalent to total parenteral nutrition in providing nutritional support. To avoid thrombophlebitis, the concentrations of amino acids and dextrose are maintained at 3% and 10%, respectively.85 Solutions that can be infused through a peripheral vein must often be diluted into a large volume to meet daily caloric needs, which may have adverse effects in postoperative patients. To meet optimal nutritional needs, parenteral nutrition solutions require either a volume or concentration that is too high for infusion into a peripheral vein. Lipids may be added to provide nonprotein calories; however, maintaining the stability of a mixture of lipids with the lower concentration of glucose in peripheral solutions is difficult.
Compared to parenteral nutrition, enteral nutrition promotes decreased GI mucosal permeability and better wound healing. The enteral route is also associated with fewer metabolic disturbances and is less expensive than parenteral nutrition. Limitations in postoperative patients are its contraindication in patients with ileus and patients at risk for aspiration. Patients with short gut syndrome also may require parenteral supplementation until the remaining bowel adapts. Feeding into the small bowel may be preferable over gastric feeding; this avoids the risks of regurgitation and aspiration associated with delayed gastric emptying, particularly in an ovarian cancer surgery patient where gastric distention may compromise pedicles on the short gastric vessels from an infragastric omentectomy. However, bypassing the stomach does not necessarily increase the tolerance to feeds, and either approach to enteral feeding may be acceptable. Randomized trials of prokinetic agents have not shown them to be advantageous in improving postoperative ileus. In the immediate postoperative period, a critically ill, nutritionally depleted patient will more likely be considered for total parenteral nutrition, but enteral feeding should be considered after recovery of bowel function.
In the postoperative setting, appetite is often limited by the return of intestinal function. However, as neoadjuvant chemotherapy is increasingly incorporated into the treatment of gynecologic cancers, patients may have received cytotoxic treatment prior to surgery, which may also compromise their appetite. Megestrol acetate is commonly used and is effective in treating chemotherapy-induced anorexia. However, its use has also been associated with an increased risk for thromboembolic events; therefore, use in the postoperative period should be monitored closely. Corticosteroids such as dexamethasone have been found to have similar stimulatory effects on appetite, yet their use is also associated with mild agitation and insomnia, which can be difficult to tolerate in the postoperative period and may exacerbate delirium in elderly patients. Anabolic steroids have also been proposed.
1. St Clair CM, Shah M, Diver EJ, et al. Adherence to evidence-based guidelines for preoperative testing in women undergoing gynecologic surgery. Obstet Gynecol. 2010;116(3):694-700.
2. Fleisher LA, Beckman JA, Brown KA, et al. ACC/AHA 2007 Guidelines on Perioperative Cardiovascular Evaluation and Care for Noncardiac Surgery: executive summary: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines developed in collaboration with the American Society of Echocardiography, American Society of Nuclear Cardiology, Heart Rhythm Society, Society of Cardiovascular Anesthesiologists, Society for Cardiovascular Angiography and Interventions, Society for Vascular Medicine and Biology, and Society for Vascular Surgery. J Am Coll Cardiol. 2007;50(17):1707-1732.
3. Smetana GW, Lawrence VA, Cornell JE. Preoperative pulmonary risk stratification for noncardiothoracic surgery: systematic review for the American College of Physicians. Ann Intern Med. 2006;144(8):581-595.
4. Poirier P, Alpert MA, Fleisher LA, et al. Cardiovascular evaluation and management of severely obese patients undergoing surgery: a science advisory from the American Heart Association. Circulation. 2009;120(1):86-95.
5. Bristow RE, Duska LR, Lambrou NC, et al: A Model for predicting surgical outcome in patients with advanced ovarian carcinoma using computed tomography. Cancer. 89:1532-1540;2000.
6. Ferrandina G, Sallustio G, Fagotti A, et al. Role of CT scan-based and clinical evaluation in the preoperative prediction of optimal cytoreduction in advanced ovarian cancer: a prospective trial. Br J Cancer. 2009;101(7):1066-1073.
7. Ozdemir S, Celik C, Emlik D, et al. Assessment of myometrial invasion in endometrial cancer by transvaginal sonography, Doppler ultrasonography, magnetic resonance imaging and frozen section. Int J Gynecol Cancer. 2009;19(6):1085-1090.
8. Nagar H, Dobbs S, McClelland HR, et al. The diagnostic accuracy of magnetic resonance imaging in detecting cervical involvement in endometrial cancer. Gynecol Oncol. 2006;103(2):431-434.
9. Orezzoli JP, Sioletic S, Olawaiye A, et al. Stage II endometrioid adenocarcinoma of the endometrium: clinical implications of cervical stromal invasion. Gynecol Oncol. 2009;113(3):316-323.
10. Mitchell DG, Snyder B, Coakley F, et al. Early invasive cervical cancer: tumor delineation by magnetic resonance imaging, computed tomography, and clinical examination, verified by pathologic results, in the ACRIN 6651/GOG 183 Intergroup Study. J Clin Oncol. 2006;24(36):5687-5694.
11. Chung HH, Kang KW, Cho JY, et al. Role of magnetic resonance imaging and positron emission tomography/computed tomography in preoperative lymph node detection of uterine cervical cancer. Am J Obstet Gynecol. 2010;203(2):156:e1-5.
12. Boughanim M, Leboulleux S, Rey A, et al. Histologic results of para-aortic lymphadenectomy in patients treated for stage IB2/II cervical cancer with negative [18F]fluorodeoxyglucose positron emission tomography scans in the para-aortic area. J Clin Oncol. 2008;26(15):2558-2261.
13. Choi HJ, Roh JW, Seo SS, et al. Comparison of the accuracy of magnetic resonance imaging and positron emission tomography/computed tomography in the presurgical detection of lymph node metastases in patients with uterine cervical carcinoma: a prospective study. Cancer. 2006;106(4):914-922.
14. Ford MK, Beattie WS, Wijeysundera DN. Systematic review: prediction of perioperative cardiac complications and mortality by the revised cardiac risk index. Ann Intern Med. 2010;152(1):26-35.
15. 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(21):e169-e276.
16. Fleisher LA, Beckman JA, Brown KA, et al. ACC/AHA 2007 guidelines on perioperative cardiovascular evaluation and care for noncardiac surgery: executive summary: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (Writing Committee to Revise the 2002 Guidelines on Perioperative Cardiovascular Evaluation for Noncardiac Surgery). Anesth Analg. 2008;106(3):685-712.
17. Devereaux PJ, Yang H, Yusuf S, et al. Effects of extended-release metoprolol succinate in patients undergoing non-cardiac surgery (POISE trial): a randomised controlled trial. Lancet. 2008; 371(9627):1839-1847.
18. Fleischmann KE, Beckman JA, Buller CE, et al. 2009 ACCF/AHA focused update on perioperative beta blockade. J Am Coll Cardiol. 2009;54(22):2102-2128.
19. Scognamiglio R, Negut C, Ramondo A, et al. Detection of coronary artery disease in asymptomatic patients with type 2 diabetes mellitus. J Am Coll Cardiol. 2006;47(1):65-71.
20. Dronge AS, Perkal MF, Kancir S, et al. Long-term glycemic control and postoperative infectious complications. Arch Surg. 2006;141(4):375-380.
21. Marik PE, Varon J. Requirement of perioperative stress doses of corticosteroids: a systematic review of the literature. Arch Surg. 2008;143(12):1222-1226.
22. Mathew A, Devereaux PJ, O’Hare A, et al. Chronic kidney disease and postoperative mortality: a systematic review and meta-analysis. Kidney Int. 2008;73(9):1069-1081.
23. O’Leary JG, Yachimski PS, Friedman LS. Surgery in the patient with liver disease. Clin Liver Dis. 2009;13(2):211-231.
24. Cobb WS, Heniford BT, Burns JT, et al. Cirrhosis is not a contraindication to laparoscopic surgery. Surg Endosc. 2005;19(3):418-423.
25. Laky B, Janda M, Kondalsamy-Chennakesavan S, et al. Pretreatment malnutrition and quality of life: association with prolonged length of hospital stay among patients with gynecological cancer: a cohort study. BMC Cancer. 2010;10:232.
26. Diaz-Montes TP, Zahurak ML, Bristow RE. Predictors of extended intensive care unit resource utilization following surgery for ovarian cancer. Gynecol Oncol. 2007;107(3):464-468.
27. Geisler JP, Linnemeier GC, Thomas AJ, Manahan KJ. Nutritional assessment using prealbumin as an objective criterion to determine whom should not undergo primary radical cytoreductive surgery for ovarian cancer. Gynecol Oncol. 2007;106(1):128-131.
28. Kuppermann M, Summitt RL Jr, Varner RE, et al. Sexual functioning after total compared with supracervical hysterectomy: a randomized trial. Obstet Gynecol. 2005;105(6):1309-1318.
29. Cibula D, Velechovska P, Sláma J, et al. Late morbidity following nerve-sparing radical hysterectomy. Gynecol Oncol. 2010;116(3): 506-511.
30. Serati M, Salvatore S, Uccella S, et al. Sexual function after radical hysterectomy for early-stage cervical cancer: is there a difference between laparoscopy and laparotomy? J Sex Med. 2009;6(9):2516-2522.
31. ACOG Committee on Practice Bulletins. ACOG Practice Bulletin No. 104: antibiotic prophylaxis for gynecologic procedures. Obstet Gynecol. 2009;113(5):1180-1189.
32. Bratzler DW, Hunt DR. The surgical infection prevention and surgical care improvement projects: national initiatives to improve outcomes for patients having surgery. Clin Infect Dis. 2006;43(3):322-330.
33. Itani KM, Wilson SE, Awad SS, et al. Ertapenem versus cefotetan prophylaxis in elective colorectal surgery. N Engl J Med. 2006;355(25):2640-2651.
34. 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(15): 1736-1754.
35. ACOG Committee on Practice Bulletins. Practice Bulletin No. 84: prevention of deep vein thrombosis and pulmonary embolism. Obstet Gynecol. 2007;110(2 Pt 1):429-440.
36. Geerts WH, Bergqvist D, Pineo GF, et al. Prevention of venous thromboembolism: American College of Chest Physicians Evidence-Based Clinical Practice Guidelines (8th Edition). Chest. 2008;133(6 suppl):381S-453S.
37. Einstein MH, Kushner DM, Connor JP, et al. A protocol of dual prophylaxis for venous thromboembolism prevention in gynecologic cancer patients. Obstet Gynecol. 2008;112(5):1091-1097.
38. NCCN Clinical Practice Guidelines in Oncology (NCCN Guidelines TM) Venous Thromboembolic Disease. v.1.2010. Available at: http://www.nccn.org/professionals/physician_gls/f_guidelines.asp.
39. Adib T, Belli A, McCall J, et al. The use of inferior vena caval filters prior to major surgery in women with gynaecological cancer. BJOG. 2008;115(7):902-907.
40. Pineda CE, Shelton AA, Hernandez-Boussard T, et al. Mechanical bowel preparation in intestinal surgery: a meta-analysis and review of the literature. J Gastrointest Surg. 2008;12(11): 2037-2044.
41. Guenaga KK, Matos D, Wille-Jorgensen P. Mechanical bowel preparation for elective colorectal surgery. Cochrane Database Syst Rev. 2009(1):CD001544.
42. Muzii L, Bellati F, Zullo MA, et al. Mechanical bowel preparation before gynecologic laparoscopy: a randomized, single-blind, controlled trial. Fertil Steril. 2006;85(3):689-693.
43. Nelson RL, Glenny AM, Song F. Antimicrobial prophylaxis for colorectal surgery. Cochrane Database Syst Rev. 2009(1):CD001181.
44. Wijeysundera DN, Beattie WS, Austin PC, et al. Non-invasive cardiac stress testing before elective major non-cardiac surgery: population based cohort study. BMJ. 2010;340:b5526.
45. Poirier P, Giles TD, Bray GA, et al. Obesity and cardiovascular disease: pathophysiology, evaluation, and effect of weight loss: an update of the 1997 American Heart Association Scientific Statement on Obesity and Heart Disease from the Obesity Committee of the Council on Nutrition, Physical Activity, and Metabolism. Circulation. 2006;113(6):898-918.
46. Walker JL, Piedmonte MR, Spirtos NM, et al. Laparoscopy compared with laparotomy for comprehensive surgical staging of uterine cancer: Gynecologic Oncology Group Study LAP2. J Clin Oncol. 2009;27(32):5331-5336.
47. Seamon LG, Bryant SA, Rheaume PS, et al. Comprehensive surgical staging for endometrial cancer in obese patients: comparing robotics and laparotomy. Obstet Gynecol. 2009;114(1): 16-21.
48. Ebert TJ, Shankar H, Haake RM. Perioperative considerations for patients with morbid obesity. Anesthesiol Clin. 2006;24(3): 621-636.
49. Wright JD, Rosenbush EJ, Powell MA, et al. Long-term outcome of women who undergo panniculectomy at the time of gynecologic surgery. Gynecol Oncol. 2006;102(1):86-91.
50. Leath CA 3rd, Kendrick JE 4th, Numnum TM, et al. Outcomes of gynecologic oncology patients admitted to the intensive care unit following surgery: a university teaching hospital experience. Int J Gynecol Cancer. 2006;16(5):1766-1769.
51. Auerbach A, Goldman L. Assessing and reducing the cardiac risk of noncardiac surgery. Circulation. 2006;113(10):1361-1376.
52. Vanacker B. Anaesthetic issues in women undergoing gynaecological cytoreductive surgery. Curr Opin Anaesthesiol. 2009;22(3): 362-367.
53. Jacob M, Chappell D, Conzen P, et al. Small-volume resuscitation with hyperoncotic albumin: a systematic review of randomized clinical trials. Crit Care. 2008;12(2):R34.
54. The Acute Respiratory Distress Syndrome Network. Ventilation with lower tidal volumes as compared with traditional tidal volumes for acute lung injury and the acute respiratory distress syndrome. N Engl J Med. 2000;342(18):1301-1308.
55. Frutos-Vivar F, Ferguson ND, Esteban A, et al. Risk factors for extubation failure in patients following a successful spontaneous breathing trial. Chest. 2006;130(6):1664-1671.
56. de la Torre SH, Mandel L, Goff BA. Evaluation of postoperative fever: usefulness and cost-effectiveness of routine workup. Am J Obstet Gynecol. 2003;188(6):1642-1647.
57. Kendrick JE, Numnum TM, Estes JM, et al. Conservative management of postoperative fever in gynecologic patients undergoing major abdominal or vaginal operations. J Am Coll Surg. 2008;207(3):393-397.
58. Abraham E, Laterre PF, Garg R, et al. Drotrecogin alfa (activated) for adults with severe sepsis and a low risk of death. N Engl J Med. 2005;353(13):1332-1341.
59. Wuethrich PY, Burkhard FC, Panicker JN, Kessler TM. Effects of thoracic epidural analgesia on lower urinary tract function in women. Neurourol Urodyn. 2011;30(1):121-125.
60. Forni LG, Hilton PJ. Continuous hemofiltration in the treatment of acute renal failure. N Engl J Med. 1997;336(18):1303-1309.
61. Selim M. Perioperative stroke. N Engl J Med. 2007;356(7):706-713.
62. Lees KR, Bluhmki E, von Kummer R, et al. Time to treatment with intravenous alteplase and outcome in stroke: an updated pooled analysis of ECASS, ATLANTIS, NINDS, and EPITHET trials. Lancet. 2010;375(9727):1695-1703.
63. Bohrer JC, Walters MD, Park A, et al. Pelvic nerve injury following gynecologic surgery: a prospective cohort study. Am J Obstet Gynecol. 2009;201(5):531:e1-7.
64. Sieber FE. Postoperative delirium in the elderly surgical patient. Anesthesiol Clin. 2009;27(3):451-464.
65. McAlpine JN, Hodgson EJ, Abramowitz S, et al. The incidence and risk factors associated with postoperative delirium in geriatric patients undergoing surgery for suspected gynecologic malignancies. Gynecol Oncol. 2008;109(2):296-302.
66. Flinn DR, Diehl KM, Seyfried LS, et al. Prevention, diagnosis, and management of postoperative delirium in older adults. J Am Coll Surg. 2009;209(2):261-268.
67. American Red Cross. Practice Guidelines for Blood Transfusion. 2nd ed. Washington, DC: American Red Cross; 2007.
68. Silliman CC, Fung YL, Ball JB, Khan SY. Transfusion-related acute lung injury (TRALI): current concepts and misconceptions. Blood Rev. 2009;23(6):245-255.
69. Marik PE, Corwin HL. Efficacy of red blood cell transfusion in the critically ill: A systematic review of the literature. Crit Care Med. 2008;36(9):2667-2674.
70. Duska LR, Garret L, Henretta M, et al. When ‘never-events’ occur despite adherence to clinical guidelines: the case of venous thromboembolism in clear cell cancer of the ovary compared with other epithelial histologic subtypes. Gynecol Oncol. 2010;116(3):374-377.
71. Geerts WH, Heit JA, Clagett GP, et al. Prevention of venous thromboembolism. Chest. 2001;119(90010):132S-175S.
72. Stein PD, Fowler SE, Goodman LR, Gottschalk A, Halles CA, Hull RD, et al. PIOPED II Investigators: multidetector computed tomography for actue pulmonary embolism. N Engl J Med. 2006;354(22):2317-2327.
73. Siragusa S, Cosmi B, Piovella F, et al. Low-molecular-weight heparins and unfractionated heparin in the treatment of patients with acute venous thromboembolism: results of a meta-analysis. Am J Med. 1996;100(3):269-277.
74. Spirt MJ, Stanley S. Update on stress ulcer prophylaxis in critically ill patients. Crit Care Nurse. 2006;26(1):18-20, 22-28.
75. Lin PC, Chang CH, Hsu PI, et al. The efficacy and safety of proton pump inhibitors vs histamine-2 receptor antagonists for stress ulcer bleeding prophylaxis among critical care patients: a meta-analysis. Crit Care Med. 2010;38(4):1197-1205.
76. Ferguson SE, Malhotra T, Seshan VE, et al. A prospective randomized trial comparing patient-controlled epidural analgesia to patient-controlled intravenous analgesia on postoperative pain control and recovery after major open gynecologic cancer surgery. Gynecol Oncol. 2009;114(1):111-116.
77. Chen LM, Weinberg VK, Chen C, et al. Perioperative outcomes comparing patient controlled epidural versus intravenous analgesia in gynecologic oncology surgery. Gynecol Oncol. 2009; 115(3):357-361.
78. Schimp VL, Worley C, Brunello S, et al. Vacuum-assisted closure in the treatment of gynecologic oncology wound failures. Gynecol Oncol. 2004;92(2):586-591.
79. Doughty D. Principles of ostomy management in the oncology patient. J Support Oncol. 2005;3(1):59-69.
80. Cheatham ML, Chapman WC, Key SP, et al. A meta-analysis of selective versus routine nasogastric decompression after elective laparotomy. Ann Surg. 1995;221(5):469-476.
81. Chase DM, Lopez S, Nguyen C, et al. A clinical pathway for postoperative management and early patient discharge: does it work in gynecologic oncology? Am J Obstet Gynecol. 2008; 199(5):541:e1-7.
82. Minig L, Biffi R, Zanagnolo V, et al. Early oral versus “traditional” postoperative feeding in gynecologic oncology patients undergoing intestinal resection: randomized controlled trial. Ann Surg Oncol. 2009;16(6):1660-1668.
83. Devakonda A, George L, Raoof S, et al. Transthyretin as a marker to predict outcome in critically ill patients. Clin Biochem. 2008;41(14-15):1126-1130.
84. Keyland DK, MacDonald S, Keefe L, Drover JW. Total parenteral nutrition in the critically ill patient: a meta-analysis. JAMA. 1988;280(23):2013-2019.
85. Gura KM. Is there still a role for peripheral parenteral nutrition? Nutr Clin Pract. 2009;24(6):709-717.