Perioperative and Critical Care
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.
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
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.
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.
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