The Surgical Oncologic Patient
Burden of Disease
The International Agency for Research on Cancer reported that the incidence of cancer in 2018 was estimated at 18.1 million new cases and 9.6 million deaths. The cumulative risk for developing cancer is 22% in men and 18.3% for women; 12.7% of men and 8.7% of women will die from the disease. Cancer incidence is predicted to continue to rise to 29.5% by 2040, and cancer deaths predicted to exceed 16 million by 2040 ; however, cancer survival has improved significantly over time. Death rates from all cancers in the United States from 2000 to 2014 decreased by 1.8% per year among men and by 1.4% per year among women. Comparing the periods between 1975 and 1977 and 2006 and 2012, the 5-year relative survival for cancer in all sites improved from 50.3% to 66.4%. The number of patients with malignancy admitted to the intensive care unit (ICU) ranges from 5% to 15%. In Scotland, approximately 5% of patients with solid organ tumors will be admitted to an ICU during the first 2 years of diagnosis. Surgical admissions will comprise 84%, of which 63% will be elective. At MD Anderson, a comprehensive cancer center in the United States, surgical ICU admissions represent 60% of all ICU admissions. The utilization rate of surgical patients requiring ICU care is approximately 18.9%. Head and neck surgery, and neurologic and gastrointestinal services represented 62% of the total ICU admissions.
Surgical Oncology Critical Care Outcomes
A large analysis of surgical ICU patients with cancer in West Scotland compared patients with and without cancer and found that surgical ICU patients with solid tumor diagnosis had an ICU mortality of 12.2% versus 16.8% ( P < 0.001) in those without cancer. The hospital mortality for those with cancer was 22.9% versus 28.1% ( P < 0.001). These patients were more likely to be admitted as a result of an elective admission to the ICU. For those patients with cancer requiring organ support, the ICU mortality was 18.6%, and the hospital mortality of those admitted emergently was 39.5%. A review of the Dutch National Intensive Care Evaluation (NICE) registry found that the most frequent underlying malignancies included colorectal cancer (25.6%), lung cancer (18.5%), and tumors of the central nervous system (14.3%). Reasons for admission included need for mechanical ventilation (24.8%) and vasopressors (20.7%). In the surgical intensive care population, oncologic patients have a favorable ICU and hospital mortality rate of 1.4% and 4.7%, respectively, following admission after elective surgery. Overall, those with cancer have comparable opportunity for recovery, and underlying malignancy should not disqualify acceptance to the surgical ICU. At MD Anderson, from 1994 to 2013 hospital mortality of surgical patients was 1.1% and 2.9% in the ICU.
ICU Resource Allocation and Admission Prioritization
In 2017 a consensus group representing the fields of critical care in cancer patients was assembled from German and Austrian Societies and made recommendations with regard to those with cancer. Three co-cohorts of patients were identified ( Box 39.1 ). Surgical patients, especially those receiving elective procedures, would most likely be encompassed in the category of receiving full ICU management.
Full-code ICU management (without limitations of ICU resources) should be offered to all critically ill cancer patients if long-term survival may be compatible with the general prognosis of the underlying malignancy.
Patients with poor performance status not eligible for further anticancer therapy, dying patients, as well as those rejecting critical care treatment should not be admitted to the ICU in general.
For patients not in categories 1 or 2, a time-limited ICU trials or predefined do-not-escalate decisions (e.g., do-not-intubate or do-not-attempt-resuscitation) may be adequate options.
Intensive care admission criteria and resource allocation must be adapted at the institution level. Local resources often dictate the provision of critical care; therefore the threshold for ICU admission at lower acuity organizations may be much lower. The Society of Critical Care Medicine guidelines for ICU admission, discharge, and triage provide a sustained review of considerations for the development of institutional and departmental policies regarding the appropriate level of care of patients referred to the ICU. Important concerns for making balanced decisions regarding patient assignment include patient interventions that can only be applied in the ICU, clinical expertise, patient condition, diagnosis, bed availability, evidence of stability, prognosis, and potential benefit. Matching the needs of the patient to the interventions and level of care available at an institution is often a fluid process. The Society of Critical Care Medicine has developed two tools to aid leadership and the practitioner in the allocation of resources and prioritization to available units. Table 39.1 provides a description of the components of care and how the level of care, patient needs, nursing-to-patient ratio, and potential interventions may align for the best interest of the patient.
|Level of Care||Level of Monitoring||Nursing-to-Patient |
|ICU||Continuous hemodynamic monitoring |
|1:1–1:2||External ventricular devices |
Continuous renal replacement
Mechanical circulatory support
Malingant or symptomatic arrhymia
|Intermediate||Every 2–4 h monitoring |
Frequent laboratory monitoring
|Telemetry||Continuous cardiac monitoring||<1:4||IV antiarrhythmic and vasodilator titration|
|Inpatient||Routine, every 4 h to every shift||<1:5||Additional evaluation |
Common Factors Affecting Patients Coming to the ICU
Patients arriving at the operating theater are likely to possess physiologic derangements secondary to their malignancy, or due to prior exposure to chemotherapy and newer targeted interventions including immuno- and biologic therapies. Frequently encountered factors complicating care include neutropenia and immunosuppression, thrombocytopenia, prothrombotic states, coagulopathic states, chemotherapeutic toxicities, malnutrition, and poor functional status.
Neutropenia and Immunosuppression
Neutropenia, defined as an absolute neutrophil count of less than 1500 cells per microliter, is a common complication of chemotherapy. Risk of infection increases with both severity and duration of neutropenia. , Among patients receiving myelosuppressive chemotherapy, the risk of developing neutropenic fever during the chemotherapy course is 13%–20%. , The Multinational Association of Supportive Care in Cancer (MASCC) index is a validated tool to identify a patient’s risk for complications associated with febrile neutropenia. The assessment is based on age, history, outpatient or inpatient status, clinical signs, severity of fever and neutropenia, and presence of medical comorbidities. Guidelines for prevention and treatment of neutropenic fever are outlined and supported by major societies.
Thrombocytopenia is a common complication in oncologic patients. In solid tumor patients receiving chemotherapy, the 3-month risk of developing chemotherapy-induced thrombocytopenia is 13% (platelet count <100 × 10 9 /L), 4% with grade 3 (25 to <50 × 10 9 /L), and 2% with grade 4 (<25 × 10 9 /L). Etiology of thrombocytopenia in cancer patients includes direct chemotherapy effects, splenomegaly, bone marrow infiltration, disseminated intravascular coagulation (DIC), thrombotic thrombocytopenic purpura, immune thrombocytopenia, infections, and medications. Patients receiving gemcitabine or carboplatin-based chemotherapy regimens tend to have increased incidence of thrombocytopenia. Several societies have recommended platelet count thresholds and targets for those patients undergoing elective surgery or invasive procedures ( Table 39.2 ). Preoperative consultation and postoperative involvement of hematologic specialists are key for diagnostic evaluation and selecting appropriate interventions that may include platelet transfusions, procoagulants, antifibrinolytics, and thrombopoietin receptor agonists. ,
|Platelet Count||Planned Procedure|
|>100 × 10 9 /L||Surgery on the brain or posterior eye|
|>50 × 10 9 /L||Major nonneuroaxial surgery |
Liver, renal, or transbronchial biopsy
|>20 × 10 9 /L||Central line placement |
Bronchoscopy with lavage
|>10 × 10 9 /L||Prophylaxis against spontaneous bleeding|
Cancer is a well-known prothrombotic state. Cancer patients have a 4 to 7-fold increased risk of venous thrombosis compared with the general population or patients without cancer. Incidence rate of venous thrombosis amongst all patients with cancer is estimated at 13 per 1000 person-years. Patients with cancer undergoing surgery are at increased risk of venous thromboembolism (VTE), the incidence of which is estimated at 1.3%–2.1% within 30 days with an annual prevalence of 4% following major cancer surgery. , Thrombotic events may manifest as VTE, arterial thrombosis, DIC, and thrombotic microangiopathy. In addition to standard patient risk factors for thrombosis, cancer-associated risk factors include site of cancer, stage of cancer, histology, time after diagnosis, surgical interventions, hospitalization, chemotherapy, exposure to vascular endothelial growth factor inhibitors, and the presence of central venous catheters. The American Society of Clinical Oncology (ASCO) updated guidelines for VTE in patients with cancer in 2019; these specifically address cancer patients undergoing surgery. The recommended approach includes unfractionated heparin or low-molecular-weight heparin. Risks of active bleeding, high bleeding risk, or other contraindications must be weighed with potential benefit. All cancer patients should be offered pharmacologic thromboprophylaxis to be started preoperatively and continued for at least 7 to 10 days. In patients undergoing major open or laparoscopic abdominal or pelvic surgery for cancer who have restricted mobility, obesity, history of VTE, or who have risk factors for VTE, a 4-week course of prophylaxis is recommended.
A state of hypocoagulopathy is less frequently encountered but can exist in the cancer patient. In addition to the more frequently encountered perioperative causes of coagulopathy, coagulation abnormalities secondary to hemorrhage, hemodilution, hemostatic factor consumption, exposure to anticoagulants and antiplatelet agents, renal disease, and liver disease, the oncologic patient may present with a variety of unique etiologies for the underlying cause of bleeding ( Box 39.2 ) . ,
Pathogenic tumor endothelium
Tumor-derived coagulation inhibitors
Treatment-related coagulation disorders
Postoperative coagulation disorders
A primary hyperfibrinolytic state associated with acute promyelocytic leukemia
Acquired hemophilia as is the case with factor VIII or
Acquired von Willebrand disease
DIC in cancer generally presents as a subacute consumptive state or as acute DIC. Chronic DIC is typically seen in solid organ tumors of the lung, breast, prostate, colon, and rectum, which comprise the most frequent primary tumors. Approximately 7%–29% of solid organ malignancies experience DIC, of which bleeding occurs in 59%, while thrombosis occurs in 34%. , Acute DIC can be seen in 29%–32% of non APL acute leukemia. , DIC in acute leukemia predominantly presents with bleeding as opposed to thrombosis. Management of DIC in cancer can require a complex and thoughtful approach and requires consideration of the thrombotic versus bleeding potential. Involvement of a hematologist will be worthwhile. For those who are bleeding or need an invasive procedure, Box 39.3 outlines an approach. Promyelocytic leukemia often presents with a particularly severe form of DIC associated with enhanced fibrinolysis. Fifty-three percent of those with acute promyelocytic leukemia (APL) will present with some manifestation of bleeding: bruising, epistaxis, abnormal menstrual bleeding, hematuria, hemoptysis, hematochezia, or melena. The coagulopathy associated with APL is particularly responsive to early induction therapy, thus mitigating bleeding risk. White blood cell count at presentation is the most significant predictor of early hemorrhagic death and early thrombo-hemorrhagic death. Use of recombinant thrombomodulin has been proposed as a rescue therapy for DIC. ,
Platelet transfusion to keep platelet count >300–50 × 10 9
Transfuse fresh frozen plasma or cryoprecipitate to maintain prothrombin time <3 s and fibrinogen >1.5 g/L
Vitamin K supplementation in case of deficiency
Antifibrinolytic treatment if excessive hyperfibrinolysis
Malnutrition is generally considered present by the American Society of Parenteral and Enteral Nutrition when at least two of six indicators are present ( Box 39.4 ). Generally, 26.6%–51% of oncologic patients will have nutritional impairment with 4.5%–9% overtly malnourished. , Malnutrition is consistently prevalent among hospitalized patients and present in approximately 40% of oncologic patients. , Given that the prevalence of malnutrition in hospitalized patients to admitted the ICU for more than 48 h is 70%, the burden of malnutrition in critically ill surgical oncology patients is likely to be substantial.
Insufficient energy intake
Loss of muscle mass
Loss of subcutaneous fat
Localized or generalized fluid accumulation that may sometimes mask weight loss
Diminished functional status as measured by hand grip strengths
Presence and degree of malnutrition can be affected by primary malignancy, metastases, prior and ongoing therapies, including chemotherapy and radiotherapy, prior surgeries, metabolic and mitochondrial derangements, procachexia cytokines and factors, homeostatic control in the central nervous system. , Severity of malnutrition was positively correlated with the stage of cancer.
For those patients receiving cancer therapies, the routine administration of total parenteral nutrition (TPN) is not recommended by the American Society of Parenteral and Enteral Nutrition (ASPEN) to all patients undergoing major cancer operations. These recommendations are due to the poor evidence to support the beneficial effects of parenteral and enteral nutrition in mortality and morbidity. In patients who are severely malnourished, there may be benefit with parenteral nutrition. , Additionally, there may be a role for nutrition support therapy beginning 7–14 days preoperatively in patients who have moderate and severe malnourishment. Recommendations for those patients who are critically ill and in their perioperative phase are not well delineated in the literature. The 2016 SCCM/ASPEN Guidelines for Nutrition Support Therapy in the Adult Critically Ill Patient provide a framework for those patients in the ICU. No guideline, however, can replace the multidisciplinary dialogue between the surgeon, intensivist, and dietitian while taking into consideration the patient and operation performed. Key recommendations include initiation of enteral feeding within 24–48 h, interventions to reduce aspiration, application of feeding protocols, and avoidance of gastric residual measurements in assessing tolerance.
Patients Requiring Monitoring
A common need for ICU admission following neurosurgery is the need for frequent neurologic monitoring and/or cerebral spinal fluid diversion via external ventricular drain. The level of intensive monitoring is considered necessary based on the rapidity of evolution of the physical exam and the need for rapid diagnostic evaluation and intervention required when a complication does arise. Challenges to the approach have occurred over time. Ziai et al. in 2003 retrospectively evaluated 158 consecutive patients having undergone brain tumor resection. Sixty-five (49%) patients required no interventions beyond postanesthetic care and frequent neurologic exams. Those who received ICU level interventions had the following: IV analgesics, IV anxiolytics, IV antihypertensives, hypertonic therapies, antiepileptics, treatment of bradycardias, mechanical ventilation, and vasopressors. Fifteen percent of patients required greater than 1 day of ICU stay after craniotomy for brain tumor resection.
The best predictors of ICU stay greater than 1 day include postoperative intubation, tumor severity score based on radiologic findings, and a fluid score composed of estimated blood losses, combined with administration of crystalloid, colloid, blood products, and hypertonic saline. Other indicators included surgery lasting longer than 7 h, a new postsurgical hemiparesis or lower cranial nerve deficit, and intraoperative use of vasopressor therapy for hypotension. The nursing skill set required for the management, as well as the concomitant diagnoses that preclude the need of an external ventricular device (EVD), will limit their use to the ICU at nearly all institutions. Methodical management with adherence to protocols and maintenance of procedural competencies can help to decrease the infectious complications associated with EVDs. Bundles of care for EVD management may include hair removal and skin preparation prior to insertion, catheter selection, aseptic technique, appropriate dressing and frequency of dressing changes, standardized techniques of EVD maintenance, uniform cerebrospinal fluid sampling procedures, limiting duration of catheter placement, recurring competencies, and surveillance for infection and complications. The Neurocritical Care Society 2016 evidence-based consensus statement regarding the insertion and management of external ventricular drains recommends a single dose of an antimicrobial at the time of EVD insertion for ventriculostomy-related infection prophylaxis and against the use of antimicrobials for the duration of EVD placement.
Potential complications encountered following resection of brain and spinal tumors are outlined in Box 39.5 .
Potential Post Craniotomy Complications
Cerebrospinal fluid leak
Delirium secondary to steroids or other
Dysphagia/vocal cord paralysis
Corneal abrasion/exposure risk
Postoperative respiratory failure
Deep venous thrombus
Medical morbidity and mortality from head and neck surgery are 5.65% and 2.98%, respectively. Characteristics associated with increased risk of 30-day serious complications include American Society of Anesthesiologists (ASA) score ≥4 and operating room time >6 h. In patients ≥90 years of age, risk factors for increasing 90-day mortality included elevated Adult Comorbidity Evaluation-27 score, preoperative dysphagia, and large resections. Complications unique to head and neck surgery that may require monitoring in the ICU are summarized in Box 39.6 . Reasons for postoperative ICU admission of head and neck cancer patients can be categorized into complications related to respiratory, cardiac, and wounds. Approximately 5% of head and neck cancer patients will have readmission within 30 days.
Superior laryngeal nerve injury
Recurrent laryngeal nerve injury
Acute hypocalcemia secondary to hypoparathyroidism
New tracheostomy associated bleeding, obstruction, and airway loss
Upper airway swelling and concern for compromised airway
Factors associated with unplanned readmissions to the hospital include diabetes, preoperative dyspnea at rest and with moderate exertion, long-term use of corticosteroids, disseminated cancer, and a contaminated wound. The most frequent diagnoses to result in 30-day unplanned hospital readmission include superficial incisional surgical site infection, deep incisional surgical site infection, organ or space surgical site infection, wound disruption, pneumonia, deep vein thrombosis, pulmonary embolism, urinary tract infection, stroke, sepsis, and septic shock.
Due to the high-risk nature of patients undergoing head and neck surgery, early elective tracheostomy is often incorporated into the primary surgery. Benefits of elective tracheostomy include decreased ventilator days, faster deescalation to floor status, decreased pneumonia, and decreased delirium. Those patients with suspected at-risk airways due to surgical trauma, bleeding, swelling, edema, hematoma, vocal cord paralysis, and tracheomalacia who do not receive elective tracheostomy must be approached with caution. The Difficult Airway Society provides a schema for the planning, preparation, performance, and follow up of such high-risk patients.
ICU utilization for head and neck surgery patients has dramatically evolved as clinical pathways of care have developed. Specialty specific floors with nurses specially educated in flap management, frequent vitals and flap checks, continuous oxygen saturation monitoring, and tracheostomy care have allowed for decreased ICU requirements. Avoiding ICU admission has in turn allowed for a reduction in length of hospital stay and a decline in hospital costs. ,
Postoperative delirium is common and is prevalent in approximately 17%–19% or patients undergoing head and neck cancer surgery. , Risk factors for delirium following head and neck cancer surgery include male sex, age >70 years, duration of surgery, history of hypertension, blood transfusions, tracheotomy, ASA physical status grade at least III, flap reconstruction, and neck dissection. Postoperative delirium often is poorly recognized and managed. The Society of Critical Care Medicine has incorporated a multifaceted approach in to their society guidelines ( Box 39.7 )
Critically ill adults should be regularly assessed for delirium using a valid tool
CAM-ICU, ICDSC, and ICU-7
Avoidance of Modifiable Risks
Benzodiazepines and blood transfusions
Avoiding Pharmacologic Prevention
Haloperidol, dexmedetomidine, β-hydroxy β-methylglutaryl-coenzyme A (HMG-CoA) reductase inhibitor, and ketamine have not been shown to prevent delirium in all critically ill adults
Dexmedetomidine for delirium in mechanically ventilated adults where agitation is precluding weaning/extubation
Not routinely using haloperidol or an HMG-CoA reductase inhibitor to treat delirium
Reorientation, cognitive stimulation, use of clocks
Sleep hygiene and minimizing nocturnal light and noise
Reduce sedation and improve wakefulness
Increase mobility and early rehabilitation
Avoid sensory deprivation with the use of hearing aids or eyeglasses
The role of electrocardiographic monitoring is often incorrectly applied and utilization questioned. In fact the American Board of Internal Medicine Foundation’s Choosing Wisely campaign highlighted the practice of continuous telemetry and recommended against the unnecessary use in the hospital. It is estimated that 20% of patients may receive telemetry for noncardiac indications. The clinical impact of telemetry in those deemed to have essential indications is in question as well. In the non-ICU, nonsurgical patient population physicians perceived cardiac telemetry helpful in only 12.6% of the patients and management decisions directly affected in only 7%.
The 2017 Standards for Electrocardiographic Monitoring in Hospital Settings recommends against the routine use of arrhythmia monitoring after noncardiac surgery for asymptomatic patients. Thoracic surgery and major vascular surgery patients represent two exceptions to this recommendation. Thoracic surgery patients are at increased risk for the development of atrial fibrillation, and for this reason 2–3 days of continuous electrographic monitoring is recommended. Atrial fibrillation rates can range from 3% to 30%, depending on the procedure performed. Vascular surgery patients often have concomitant atherosclerosis that justifies postoperative monitoring. Indications and evidence for continuous electrographic monitoring are outlined in the guideline, and common postoperative indications are listed in Box 39.8 .
Suspected coronary ischemia
Major cardiac interventions
Those with mechanical circulatory support
New or recurrent ventricular tachycardia, nonsustained ventricular tachycardia
Acute atrial fibrillation
Chronic atrial fibrillation postoperatively
Symptomatic bradycardia or new bradycardia postoperatively
Congenital or genetic arrhythmic syndromes
Intraoperative implantable cardioverter-defibrillator (ICD) shock
Acute decompensated heart failure
Potassium and magnesium abnormalities
Postoperative hemodynamic monitoring is generally not needed unless the patient has unstable hemodynamics, is at risk for unstable hemodynamics, or receiving vasopressors or inotropic agents. However, those requiring ongoing goal-directed resuscitation postoperatively will likely need hemodynamic monitoring. An excellent example of the need for ongoing hemodynamic monitoring to aid resuscitation and continued Enhanced Recovery After Surgery (ERAS) is hyperthermic intraperitoneal chemotherapy (HIPEC), which is used for the treatment of peritoneal surface malignancies. HIPEC can result in significant hemodynamic instability and heightened inflammatory response. , Patients receiving HIPEC can have great individual intraoperative fluid therapy needs, highlighting the need for personalized resuscitation. In a randomized controlled trial (RCT) of goal-directed versus standard fluid therapy, patients receiving hemodynamic-guided fluid therapy had reduced postoperative complications and hospital length of stay as compared with standard fluid therapy.
A frequently encountered component in head and neck as well as plastic surgery procedures is the creation of a free flap. Free flaps require frequent monitoring for observed complications and specific perioperative management. Potential complications postoperatively include vascular compromise, hematoma, surgical site infection, and wound dehiscence. Vascular compromise can be due to arterial thrombosis, venous thrombosis, hematoma, edema, recipient vessel disease, failure of anastomosis, and mechanical stress on the anastomosis. , In 990 consecutive free flaps, the overall thrombosis rate was 5.1%; 54% of thrombi occurred in the venous system, while 20% were arterial and 12% were combined artery and vein. In general, 80% of thrombi will occur within the first 48 h, highlighting the greatest yield for monitoring immediately postoperatively. Various forms of invasive and noninvasive monitoring may take place (see Box 39.9 ) and those with the most supportive evidence include Cook-Swartz implantable Doppler, near-infrared spectroscopy, laser Doppler flowmetry, quantitative fluorimetry, and digital smartphone applications.