Intensive Care of the Cancer Patient

Published on 07/03/2015 by admin

Filed under Critical Care Medicine

Last modified 07/03/2015

Print this page

rate 1 star rate 2 star rate 3 star rate 4 star rate 5 star
Your rating: none, Average: 0 (0 votes)

This article have been viewed 2824 times


Intensive Care of the Cancer Patient

Chapter Outline













The annual estimated incidence of new invasive cancers in the United States in 2012 exceeds 1.6 million, with greater than 570,000 deaths.1 Long-term remissions and control of advanced cancer are being achieved with targeted therapy, and new immunotherapy agents in many malignancies.26 We are entering an era of “personalized” oncologic care in which treatments are prescribed based on the profile of mutated or overexpressed genes in the tumor specimen. For the treatment of metastatic malignancies, enhanced success has come from the ability to deliver chemotherapy, radiation therapy, immunotherapy, or combination regimens with increased dose intensity. Progress in supportive care and intensive care medicine has allowed oncologists to treat their patients aggressively and support them despite the toxicities inherent in dose-intense treatment modalities. A greater understanding of the mechanisms for these toxicities has also improved care and patient outcomes.

This chapter describes specific oncologic clinical entities and cancer treatment toxicities that require intensive care management. The discussion of these problems is organized according to organ system because cancers are best understood as systemic diseases that can directly or indirectly influence all organ systems. The pathophysiology that gives rise to these clinical circumstances is increasingly understood and is used as the basis for treatment recommendations. Aspects unique to biologic therapies and bone marrow transplantation are treated in separate sections. The ethical aspects of treating cancer patients in the intensive care unit (ICU) are also discussed.

Metabolic and Endocrine Complications

Endocrine syndromes associated with malignancies have been described for many years and some clinically significant endocrinopathies are induced by immunotherapy agents, such as ipilimumab, currently used in the treatment of advanced melanoma.4,7 These problems may manifest as solitary laboratory derangements, such as hypercalcemia or hyperphosphatemia, or can present as clinical syndromes, such as Cushing syndrome in small cell lung cancer. Metabolic disorders can also arise as a consequence of cancer treatment. This is most often seen with chemotherapy for rapidly growing tumors such as leukemias or lymphomas. Abrupt changes in metabolic variables have also been observed after interleukin 2 (IL-2)-based immunotherapy and the rapid in vivo expansion of lymphocytes.8 The most common of these clinical entities are tumor lysis syndrome (TLS), hypercalcemia, oncogenic osteomalacia, syndrome of inappropriate secretion of antidiuretic hormone (SIADH), adrenal failure, pheochromocytoma, tumor-induced hypoglycemia, and chemotherapy-induced metabolic disturbances.

Tumor Lysis Syndrome

Case reports of metabolic and electrolyte abnormalities after chemotherapy for rapidly growing tumors such as Burkitt’s lymphoma and leukemias were first published in the 1950s.9 Cadman and his colleagues in the 1970s proposed a mechanism that linked these metabolic observations.10 More recently, TLS has been observed in patients with solid tumors and has been observed in patients receiving immunotherapy, such as IL-2, sunitinib, imatinib, and rituximab.1114 TLS after treatment for solid tumors is relatively rare.

TLS is characterized by hyperuricemia, hyperkalemia, hyperphosphatemia, and hypocalcemia.10 Electrolyte abnormalities can appear as soon as 6 hours after chemotherapy administration and can persist for 5 to 7 days after treatment. The hyperuricemia comes from the massive release of intracellular nucleic acids and their metabolism by xanthine oxidase into uric acid. Urate crystals can form in the renal collecting ducts and result in oliguric and anuric renal failure. Similarly, potassium and phosphate are released from lysing tumor cells, and renal excretion of these intracellular ions is impaired by hyperuricemia. Serum calcium levels drop from ectopic calcium deposition; this becomes more likely as the calcium-phosphorus product increases. Calcium deposition is favored by a calcium-phosphorus product greater than 60 mg2/dL2 and becomes severe when the product is more than 75 mg2/dL2. The clinical manifestations of TLS depend on which electrolyte derangement predominates. Tetany, confusion, hypotension, dysrhythmias, and sudden death have been reported with TLS. The most effective management approach for this syndrome is to anticipate its occurrence and intervene prospectively. Patients at greatest risk for TLS are those with a diagnosis of a rapidly growing lymphoma or leukemia with high blast counts, and pretreatment levels of lactate dehydrogenase greater than 1500 U/dL and uric acid greater than 10 mg/dL. Pretreatment azotemia is also a poor prognostic sign. Azotemia may be exacerbated by uric acid nephropathy, which is more common with uric acid levels exceeding 20 mg/dL. It is unlikely that TLS will occur in patients at risk who do not develop metabolic changes within 48 hours after receiving chemotherapy. Guidelines for prophylaxis and treatment of TLS are given in Box 80.1. It is important to note that rasburicase poses an oxidative stress and can induce hemolytic anemia in patients with glucose-6-phosphate dehydrogenase (G6PD) deficiency.


Hypercalcemia is the most common metabolic abnormality occurring in cancer patients. Approximately 10% to 20% of all cancer patients have hypercalcemia at some point in their course. The clinical symptoms of hypercalcemia are nonspecific and include lethargy, confusion, nausea, and anorexia. Often the clinical symptoms in cancer patients may be subtle because the onset of hypercalcemia is gradual. The mechanism that underlies all cases of cancer-related hypercalcemia is increased calcium resorption from bone due to enhanced osteoclast activity mediated through receptor activator for nuclear factor κB ligand (RANKL).15 This resorption increase can be due to local action of tumor in bone or to the production of bone-resorbing hormones and cytokines by tumor cells remote from bone. Normally, increased circulating calcium results in decreased parathyroid hormone (PTH) production. When PTH levels decrease, bone resorption and renal tubular reabsorption of calcium decline. Low PTH levels cause a decrease in vitamin D production; thus, gut absorption of calcium is lowered. Although PTH levels are suppressed in cancer patients with hypercalcemia, the destructive action of tumor deposits in bone or the action of tumor-produced hormones on bone maintains high calcium resorption rates. This is accomplished through osteoclast activation and proliferation from factors produced by the tumor, such as interleukin 1 (IL-1), tumor necrosis factor (TNF), prostaglandin E2, granulocyte-macrophage colony-stimulating factor (GM-CSF), transforming growth factor-α, platelet-derived growth factor, and PTH-related peptides.1620 Malignancies that commonly cause hypercalcemia include multiple myeloma, breast carcinoma, epidermoid lung carcinoma, and renal cell carcinoma. Hypercalcemia in lymphoma and leukemia is probably not associated with PTH-related peptide, but rather with the overproduction of activated vitamin D.20,21

Cardiac complications and renal dysfunction are the most serious end-organ effects of hypercalcemia. Electrocardiogram (ECG) changes include prolongation of the PR and QRS intervals and shortening of the QT interval. Bradydysrhythmias and bundle branch blocks become more frequent with serum calcium levels greater than 16 mg/dL. These may progress to complete heart block and asystole. Renal dysfunction may also occur because increased serum calcium induces a diuretic effect, which can cause moderate to severe dehydration and prerenal azotemia. This process can result in acute tubular necrosis if untreated.

Management for any symptomatic hypercalcemic patient should begin with intravenous hydration, which may increase renal blood flow and enhance calciuresis.22 Renal excretion of calcium can be enhanced with furosemide diuresis, although no randomized trials exist to support its use in hypercalcemia. These measures should be viewed as temporizing steps until definitive treatment has been implemented. The bisphosphonates zoledronic acid, pamidronate, alendronate, etidronate, and clodronate have been shown to be highly effective in the long-term treatment of hypercalcemia of malignancy. These agents work by binding to the hydroxyapatite in bone and preventing calcium resorption, although they may also have much more complicated effects on the cell cycle and bone turnover.23,24 A commonly used bisphosphonate regimen is a single dose of pamidronate (60-90 mg intravenously over 2 to 4 hours) or zoledronic acid (4 mg intravenously over 15 minutes). Doses may be repeated in 3 to 4 days if the calcium does not decline. In addition, therapy directed at controlling the tumor should be implemented. RANKL inhibitors are now available for clinical use (denosumab) that inhibit osteoclast activity induced by malignancy and may be more potent in inhibiting bone resorption compared to bisphosphonates.25,26 Gallium nitrate can be tried in patients with hypercalcemia unresponsive to bisphosphonates.27 Calcitonin, glucocorticoids, or mithramycin can also be tried in patients unresponsive to first-line therapies, although these therapies are no longer commonly used owing to potential renal injury. Dialysis may be necessary if renal compromise is severe. Treatment recommendations are summarized in Table 80.1.

Table 80.1

Agents Used for the Management of Hypercalcemia

Drug Dosage
Pamidronate 90 mg IV over 2 hours
Zoledronic acid 4 mg IV over 15 minutes
Gallium nitrate 200 mg/m2 by continuous infusion for 5 days
Calcitonin 400 IU SQ every 8 hours
Mithramycin 25 µg/kg IV once or twice per week

IV, intravenously; SQ, subcutaneously.

Syndrome of Inappropriate Secretion of Antidiuretic Hormone

SIADH is associated with carcinoid tumors, myeloma, lymphoma, and carcinomas originating in the lung, prostate, esophagus, head and neck, adrenal gland, and pancreas. Cerebral metastasis from any tumor can also give rise to SIADH. Clinical findings are mainly neurologic, often do not correspond precisely to serum sodium levels, and range from mild confusion to coma and seizures. Because the clinical findings are secondary to water intoxication and hyponatremia, treatment involves water restriction (500 mL/day) and control of the primary tumor. More aggressive treatment should be started with 0.9% or 3% saline solution and furosemide diuresis for patients with neurologic deficits from hyponatremia. The rate of intravenous fluid supplementation should be adjusted based on the urinary excretion of sodium and potassium. Correction of severe hyponatremia (sodium less than 125 mEq/L) should take place over 7 to 10 days. Too rapid a correction can lead to serious neurologic sequelae such as central pontine myelinosis. Patients who fail to respond to water restriction can be treated with demeclocycline (600 to 1200 mg daily), which blocks the peripheral action of antidiuretic hormone (ADH). There is also a new class of selective vasopressin V2-receptor antagonists that compete with native vasopressin, resulting in an increase of free water clearance. Tolvaptan (15-60 mg daily) can be used in patients who do not respond to demeclocycline.

Adrenal Failure

Cancers of the lung, breast, kidney, stomach, and pancreas and melanoma are the tumors that metastasize most often to the adrenal glands. It is estimated that more than 90% of adrenal tissue must be destroyed before clinical manifestations of adrenal insufficiency appear. The clinical signs and symptoms of hypoadrenalism include weakness, gastrointestinal complaints, postural hypotension, dehydration, and electrolyte disturbances. The typical electrolyte profile is hyponatremia, hyperkalemia, and a mild anion gap acidosis.

Adrenal failure has also been observed in patients receiving ipilimumab, an antibody that blocks the effects of an inhibitory protein in T cells known as CTLA-4 (cytotoxic T-lymphocyte antigen 4) and is used in the treatment of advanced melanoma.4 Hypoadrenalism from ipilimumab is usually secondary to panhypopituitarism induced by T-cell–mediated hypophysitis.28 These patients can present with headache and visual changes from pituitary swelling in addition to the clinical findings of hypoadrenalism as reviewed earlier. Evaluation of these patients should include a contrast-enhanced brain magnetic resonance imaging (MRI), and measurement of serum levels of cortisol, adrenocorticotropic hormone (ACTH), and thyroid-stimulating hormone (TSH).

The diagnosis of hypoadrenalism can be made with a cosyntropin stimulation test. Plasma cortisol levels are obtained before cosyntropin injection (0.25 mg intravenously) and 30 and 60 minutes after injection. A normal response is an increase in plasma cortisol of at least 7.0 mg/dL in 60 minutes. If the cortisol response to cosyntropin stimulation is suboptimal, physiologic doses of glucocorticoids should be administered twice a day (cortisone acetate, 25 mg every morning and 12.5 mg every evening). Mineralocorticoid supplementation is required in some patients (fludrocortisone, 0.05-0.1 mg daily). If the diagnosis of adrenal insufficiency is highly suspected, treatment should begin immediately after completion of the cosyntropin test. For patients in adrenal crisis with circulatory collapse, hydrocortisone should be given at stress doses (100 mg intravenously every 8 hours). This dose of hydrocortisone should also be adequate to supplement mineralocorticoid-deficient patients. Patients receiving glucocorticoids as part of their chemotherapy regimen (usually lymphoma or myeloma patients) or for treatment of brain tumors or metastases, may already have adrenal suppression and require stress-dose glucocorticoid replacement during episodes of neutropenic sepsis or elective surgery.


Pheochromocytoma is most commonly associated with the multiple endocrine neoplasia syndrome. The clinical features of this tumor are related to episodic catecholamine release and include hypertension, severe headache, cardiac dysrhythmias, pallor, perspiration, and rarely, hypotension. Patients can also present with a multisystem crisis characterized by encephalopathy, hyperpyrexia, and hemodynamic instability.29 Diagnosis is made by measuring urinary catecholamine metabolites. An elevated vanillylmandelic acid is accurate approximately 90% of the time in making the diagnosis.30 Patients with borderline catecholamine levels can often be diagnosed with the clonidine suppression test.31 Localization of pheochromocytomas can be difficult because the tumors can arise anywhere between the base of the brain and the scrotum and can be multicentric. MRI and computed tomography (CT) are helpful in visualizing adrenal abnormalities. Nuclear medicine studies with m-[111I]iodobenzylguanidine (MIBG) can be used if the CT scan is negative. MIBG scans are sensitive and specific for detecting ectopic adrenal medullary tissue.32 Positron emission tomography (PET) imaging and diffusion-weighted MRI can detect sites of disease not apparent on CT or MIBG and can aid in surgical planning.3335 Surgical extirpation of the tumor is the only effective treatment. Preoperative control of catechol secretion is necessary and can be attained with long- or short-acting α-adrenergic blockade (phenoxybenzamine 10 mg orally two or three times daily or doxazosin 2-16 mg orally daily). A comparison of preoperative management strategies using long- versus short-acting α-antagonists showed no difference in long-term outcome after surgery, although the incidence of intraoperative hypertension was greater with short-acting medications like doxazosin, terazosin, and prazosin.36 Tachycardia can be controlled with beta blockers, but these should be started only after phenoxybenzamine. Patients in hypertensive crisis can be managed with α-methyltyrosine or calcium channel blockers such as nifedipine or nicardipine.3739

Tumor-Induced Hypoglycemia

Functional endocrine tumors can give rise to a variety of clinical syndromes. Most of these problems can be managed outside the ICU; however, tumor-induced hypoglycemia can cause serious consequences including coma, seizures, and focal neurologic deficits. A number of different mechanisms can give rise to hypoglycemia. Autonomous insulin production is most commonly associated with islet cell tumors, whereas production of insulin-like growth factors (IGF-1 or IGF-2) is seen with non–islet cell tumors.40 Slow-growing mesenchymal tumors such as leiomyosarcoma, mesothelioma, and fibrosarcoma are the most common non–islet cell tumors that cause hypoglycemia.

Treatment should be focused on control of the tumor. Insulinomas are often benign and can be cured by surgical removal. For unresectable malignancies, hypoglycemic episodes can often be reduced with supportive measures such as dietary modification with frequent meals. Insulinomas may respond to diazoxide, an inhibitor of insulin secretion. Glucagon infusions may be beneficial in some patients.41

Chemotherapy-Induced Metabolic Disturbances

A number of chemotherapy drugs can cause potentially severe electrolyte disturbances. Cyclophosphamide is associated with hyponatremia from SIADH. Vinca alkaloids such as vinorelbine and vinblastine also cause SIADH. Cisplatin and carboplatin can cause renal tubular defects resulting in hypokalemia and hypomagnesemia, which can be severe enough to require intravenous replacement. Mithramycin lowers serum calcium by a mechanism that is thought to involve inhibition of the effect of PTH on osteoclasts. Although mithramycin can be used for the treatment of hypercalcemia, it can also cause hypocalcemia in patients with normal serum calcium. Cetuximab, a humanized murine antibody directed against the epidermal growth factor receptor (EGFR) and used to treat colon carcinoma and head and neck cancer is associated with severe and symptomatic hypomagnesemia from inappropriate urinary excretion.42 Cetuximab may interact with EGFR in the loop of Henle blocking resorption of magnesium and causing secondary hypokalemia and hypocalcemia. Abiraterone, a CYP17 inhibitor of androgen biosynthesis used in men with advanced prostate cancer, also increases adrenal mineralocorticoid synthesis resulting in clinically significant hypokalemia and decreases glucocorticoid synthesis necessitating concurrent administration of prednisone. Everolimus, an oral inhibitor of the mammalian target of rapamycin (mTOR), used in the management of advanced renal cancer and neuroendocrine tumors, can induce severe hyperglycemia requiring insulin.43,44 The mechanism through which mTOR inhibitors cause hyperglycemia is not fully understood but may involve decreased insulin secretion, direct toxicity to pancreatic β-cells, or impaired suppression of hepatic glucose production.45

Cardiac Complications in Cancer Patients

Cardiac dysfunction in cancer patients can be secondary to direct mechanical effects of the tumor on the heart, pericardium, or great vessels. Certain chemotherapy drugs, immunotherapy agents, and radiation can also cause treatment-related cardiac problems. Clinical entities that require emergent and chronic management are discussed.

Superior Vena Cava Syndrome

Obstruction of blood flow through the superior vena cava (SVC) can be caused by fibrosis, thrombosis, external compression, or invasion of the vessel by tumor. SVC syndrome can also be caused by thrombus secondary to a central venous access device, which is now a common fixture of oncologic care. Malignancies that involve the mediastinum, such as lung carcinoma and lymphoma, are the most common causes of this syndrome. Facial and upper extremity edema, facial plethora, headache, and tachypnea are the most common clinical presentations. Collateral venous channels may be found on the chest or abdomen. Death from SVC syndrome is rare, but life-threatening respiratory compromise and elevated intracranial pressure can occur. Therapy for SVC syndrome depends on the underlying malignancy; thus, a biopsy is mandatory for optimal management of these patients. If lymphoma or small cell lung carcinoma is the cause of SVC syndrome, initiation of the appropriate chemotherapy regimen can rapidly shrink the mediastinal mass and is the treatment of choice. For tumors not responsive to chemotherapy, radiation therapy given with high initial fractions (3 to 4 Gy/day) can provide symptomatic relief in more than 80% of patients.46 Thrombolysis has only been studied in catheter-associated SVC syndrome and is effective in this setting.47 Endovascular stents can restore patency of the SVC in approximately 50% of patients and can result in significant palliation.48 Improvement is often evident within 72 hours and the patency of occluded endovascular stents can sometimes be restored with angioplasty.

Cardiac Tamponade

Although primary or metastatic tumors of the heart can decrease cardiac output by impairing ventricular outflow, the most frequent causes of cardiac tamponade in cancer patients are metastatic tumors of the breast and lung, and melanoma, lymphomas, and leukemias. Tamponade may occur through either encasement of the heart by tumor or production of a malignant pericardial effusion. The clinical manifestations of tamponade include decreased exercise tolerance, shortness of breath, and cough. Voltage may be decreased on ECG with a pulsus alternans pattern present. Muffled heart tones, a pericardial rub, or an increased paradoxical pulse (i.e., decrease in systolic blood pressure on inspiration exceeds 10 mm Hg) may be present on physical examination. Echocardiography is extremely useful in confirming the diagnosis of tamponade if it is suspected on physical examination. Diastolic collapse of the right atrium or right ventricle on echocardiogram is an indicator of hemodynamic compromise.49,50 Swan-Ganz catheterization may be helpful to confirm the presence of significant tamponade.

Pericardiocentesis for relief of tamponade is indicated emergently when echocardiographic or clinical evidence of hemodynamic compromise is present. Intravenous infusions of normal saline at high flow rates (100 to 500 mL/hour) may be required to support the patient until a drainage procedure is performed. Although rapid reversal of cardiac filling problems can be accomplished by this procedure, a long-term solution is required. Creation of a pericardial window can prevent the reaccumulation of fluid in more than 90% of patients.51 Sclerosing agents such as tetracycline and bleomycin have also been used to prevent reaccumulation of pericardial fluid.52 Sclerosants may be instilled into the pericardial space after adequate drainage has been accomplished and appear to have a success rate comparable to pericardial window procedures. Some centers also perform pericardial windows using video-assisted thoracoscopy, but there have been no prospective randomized studies showing superior clinical outcomes for subxiphoid versus video-assisted thoracoscopy approaches.53

Treatment-Induced Cardiac Dysfunction

A number of chemotherapy medicines have cardiac toxicities that can be life threatening. Cumulative doses of doxorubicin greater than 450 mg/m2 are associated with an increased risk for congestive heart failure (CHF). Heart damage from this drug is thought to be from an iron-dependent generation of free radicals, which secondarily cause oxidative damage to lipid membranes and intracellular organelles.52 This toxicity can present acutely or months after drug administration. It is more prevalent in older patients and those with a history of coronary artery disease, hypertension, tobacco abuse, or chest radiation therapy. Initial management with diuretics, digoxin, and angiotensin blockers is usually of benefit, but heart failure can be progressive. Liposomal encapsulation of doxorubicin or the use of dexrazoxane to prevent oxygen-derived free radical formation, may diminish the cardiac toxicities of this agent.5456 Other anthracyclines, such as mitoxantrone and epirubicin, may have a lower incidence of CHF. Furthermore, weekly low-dose boluses or continuous-infusion methods of doxorubicin administration appear to reduce the incidence of clinically significant heart damage. Although anthracycline-induced cardiac damage has generally been considered irreversible, some studies suggest that some improvement in cardiac function may occur with aggressive medical management.57 Paclitaxel, which is commonly used in ovarian, breast, and lung carcinomas, is associated with bradydysrhythmias.57,58 Ventricular tachycardia, myocardial infarction, and cardiac ischemia have also been reported. Cyclophosphamide, which is commonly used in breast cancer, lymphoma, and stem cell transplant conditioning regimens, is associated with sporadic instances of CHF, which may be severe and occurs within a few days of cyclophosphamide administration, especially at high doses. Hemorrhagic myocarditis with myonecrosis was seen on autopsy specimens from these patients. These events appear unrelated to cumulative dose or method of administration. CHF from ifosfamide has also been reported.59 CHF is usually seen approximately 2 weeks after high doses of the drug and appears more frequently in patients with concurrent renal insufficiency. Medical management successfully reverses the heart failure in most patients. CHF is also associated with trastuzumab, an anti-HER2 (human epidermal growth factor receptor 2) antibody used commonly in the management of certain forms of breast carcinoma. The incidence of CHF after trastuzumab in a large randomized trial was between 3% and 4%, and was more common in patients with antecedent cardiac disease, older patients, and those having diminished ejection fraction (EF) after anthracycline-containing chemotherapy.60 Most patients with trastuzumab-induced cardiac dysfunction have improved symptoms with appropriate medical management for CHF and discontinuation of trastuzumab.

Serial echocardiography studies have been used to assess cardiac toxicities in patients receiving chemotherapy for many years, but subtle alterations in myocardial function can be missed with standard assessment of EF. A decrease in longitudinal strain assessed by Doppler measurements of tissue velocity at baseline and repeated at 3 months after starting anthracyclines or trastuzumab is a more sensitive predictor of cardiac dysfunction than EF61,62 and may be useful in identifying patients in need of medical management before significant symptoms occur.

Radiation therapy delivered to the chest for the treatment of Hodgkin’s disease, lung malignancies, breast cancer, or other neoplasms can result in a number of cardiac toxicities, including radiation pericarditis with tamponade, myocardial fibrosis, and premature coronary artery disease. The toxic effects of radiation therapy are secondary to microvessel fibrosis and may take up to 20 years to appear.63 After mantle-field radiation therapy, the risk of fatal myocardial infarction is more than three times greater than in age-matched control subjects, although mantle-fields are rarely used currently to treat patients with lymphoma. Nervertheless, it is difficult to treat the mediastinum without also treating the heart.

Pulmonary Complications in Cancer Patients

Many of the same mechanical issues that influence cardiac dysfunction are also pertinent to pulmonary problems with an underlying neoplasm. Chemotherapy and radiation therapy can also cause lasting and sometimes fatal pulmonary complications. Many of these conditions are difficult to diagnose and can be confused with other clinical entities, such as opportunistic infections. Indeed, pneumonias are the most common pulmonary disorder requiring intensive care. A number of acute and chronic pulmonary presentations are discussed.

Lymphangitic Tumor Involvement

Interstitial lung processes in cancer patients may be due to a variety of infectious insults but can also be caused by direct lymphangitic spread of the tumor. The symptoms of lymphangitic involvement are nonspecific and include dyspnea, nonproductive cough, and hypoxemia. Pulmonary hypertension and cor pulmonale can also be present. The diagnosis can be established by video-assisted thoracoscopic biopsy or transbronchial biopsy. Pulmonary microvascular cytologic specimens obtained with a wedged pulmonary artery catheter may be a less invasive way to make the diagnosis of lymphangitic carcinomatosis.64 The prognosis of this condition is generally poor, with a life expectancy of 1 to 6 months. Appropriate systemic treatment should be implemented when the site of the primary malignancy is diagnosed.

Treatment-Induced Pulmonary Dysfunction

A number of chemotherapy agents and radiation therapy can cause pneumonitis leading to chronic pulmonary fibrosis. The chemotherapeutic agents most likely to cause this problem are bleomycin and mitomycin, but other alkylators, nitrosoureas, antimetabolites, gemcitabine, taxanes, and vinca alkaloids can cause pulmonary dysfunction. Inhibitors of the mTOR pathway such as everolimus and temsirolimus used in the treatment of advanced renal cancer can also induce severe pneumonitis.44,65 Erlotinib, an inhibitor of the phosphorylation of EGFR used in the treatment of advanced lung cancer, can also induce severe and irreversible pneumonitis.66 The underlying mechanisms for lung injury induced by these agents are not fully understood, but likely involves oxygen-derived free radical toxicity67 and dysregulation of leukocyte apoptosis regulated by TNF-receptor family members. TRAIL (TNF-related apoptosis-inducing ligand) has shown promise in preclinical models of bleomycin lung injury and may also have antitumor properties.68 We advocate that cancer patients in need of supplemental oxygen should receive the lowest possible fractional concentration of oxygen that produces a hemoglobin oxygen saturation of greater than 90%. Irreversible lung damage can occur if excessive oxygen is administered to patients receiving bleomycin or radiation therapy. Clinical assessment is crucial to patient management because there are no sensitive or accurate tests to predict the onset or course of bleomycin-induced pulmonary toxicity. The resting diffusion capacity has been used, but is suboptimal to follow patients.69 Treatment recommendations are based on the recognition of three distinct clinical entities:

Diffuse Interstitial Pneumonitis

The differential diagnosis of diffuse pulmonary infiltrates in cancer patients is large. Infectious causes include bacterial, viral, fungal, and protozoal pathogens. Noninfectious causes for diffuse pulmonary infiltrates are neoplasm, autoimmune disease, cardiac failure, leukostasis, pulmonary hemorrhage, and radiation- or chemotherapy-induced pneumonitis. Making a diagnosis on clinical grounds is difficult because the radiographic and physical examination findings are virtually indistinguishable among these diverse causative entities. Performing an open-lung biopsy is often the only way to confirm a diagnosis; however, empiric treatment may result in equally good patient outcomes. A randomized study compared immediate open-lung biopsy followed by therapy directed at the diagnosis versus empiric antibiotics alone without biopsy to treat diffuse pulmonary infiltrates in cancer patients.70 The antibiotic regimen included trimethoprim-sulfamethoxazole (20 mg/kg/day intravenously) and erythromycin (30 mg/kg/day intravenously, divided into four daily doses). A broad-spectrum antibiotic was added if the patient was neutropenic at the time of diagnosis. There was no significant difference in the outcome for these patients; however, those who received an open lung biopsy had a greater complication rate. Empiric antibiotics are appropriate initial management for diffuse interstitial infiltrates, but patients who do not improve after 4 days of empiric therapy should receive open-lung biopsy.

The decision to place a patient with a cancer diagnosis on ventilator support is often controversial for medical staff and families. It is generally recognized that such patients have a poor prognosis, with a mortality rate approaching 80%. A large multicenter trial prospectively examined prognostic variables for cancer patients requiring ventilatory support.71

Buy Membership for Critical Care Medicine Category to continue reading. Learn more here