ABSTRACT: The employment of best practices is essential for minimizing the disease burden of cancer and optimizing patient outcomes. However, areas of clinical confusion remain and patients continue to receive inappropriate therapies. The aim of this article is to present key evidence-based strategies for initial assessment, early management, and prevention of bacterial infection in patients with common solid tumors. First, the need for astute clinical assessment, including risk assessment and symptom-based diagnostic testing, will be discussed. Then, the selection of therapeutics for early intervention will be outlined, along with key nursing considerations regarding monitoring for toxicities. The efficacy of innovations, such as antimicrobial central venous catheters, will also be reviewed. Recommendations are based on practice guidelines from the National Comprehensive Cancer Network, Infectious Diseases Society of America, Healthcare Infection Control Practices Advisory Committee, Centers for Disease Control and Prevention, Surviving Sepsis Campaign, and Oncology Nursing Society.
Early and appropriate management of infection in the patient with cancer is critical to optimizing patient outcomes. Practices that are not recommended by current guidelines continue to be used frequently, however. Examples include routine fluoroquinolone prophylaxis, reactive use of colony-stimulating factors, and inappropriate selection of anti-infectives for empiric therapy in low-risk patients. The purpose of this article is to summarize pharmacologic and nonpharmacologic management and prevention strategies for bacterial infection in adult patients with solid tumors. For prevention and management strategies for fungal and viral processes in both solid tumors and hematologic malignancies, readers are referred to guidelines from the National Comprehensive Cancer Network (NCCN), Infectious Diseases Society of America (IDSA), and Centers for Disease Control and Prevention (CDC).
Patients with cancer are at increased risk of infection and at greater risk of complications from infectious processes. Multiple factors escalate their susceptibility: the absence of granulocytes; disruption of integumentary, mucosal, and mucociliary barriers; foreign devices such as vascular catheters; nutritional deficiencies; and pre-existing or newly acquired comorbidities.[2,3] Timely and effective management of infection increases the likelihood of timely receipt of anticancer therapies, decreased disease burden and symptom distress, and enhanced quality of life. In this article, we will focus first on initial assessment and interventions, and will conclude with a summary of strategies for prevention of infection.
Early management of an infectious process is of utmost importance in the patient with cancer, to decrease the risk of sepsis and related complications. The NCCN defines neutropenia as either (1) an absolute neutrophil count (ANC) < 500/ L, or (2) an ANC < 1,000/ L and a predicted decline to 500/ L over the next 48 hours. Fever is defined as a single temperature of 38.3°C (about 101°F) or higher orally or of 38.0°C (100.4°F) or higher lasting longer than 1 hour in the absence of an obvious cause.
Signs and symptoms of an infection may be muted or absent when neutropenia is present. Fever remains an early sign, however. Approximately 40% of patients who become febrile have an established or occult infection, whereas 57% of patients with cancer in a large national observational study did not have a documented source of infection. Initial evaluation of a patient with fever should begin with a site-specific history and complete physical examination, with a focus on determining the potential site and causative organisms of infection, plus the patient's risk of developing a related complication (Figure 1). High-risk patients include those with the following conditions:
• Clinically unstable at presentation;
• Significant medical comorbidity;
• Anticipated prolonged severe neutropenia (100 cells/ L for 7 days);
• Hepatic insufficiency (aminotransferase levels five times ULN [upper limit of normal]);
• Renal insufficiency (creatinine clearance < 30 mL/min);
• Uncontrolled or progressive cancer;
• Pneumonia or other complex infection on presentation;
• Oral mucositis grade 3–4; and/or
• Multinational Association for Supportive Care in Cancer (MASCC) Risk Index score < 21.
These high-risk patients should be referred to a hospital for intravenous anti-infective therapy and close monitoring for response. Low-risk patients, for whom treatment at an ambulatory clinic may be appropriate, include those with:
• No associated acute comorbid illness that requires inpatient treatment or close observation;
• Anticipated short duration of severe neutropenia (100 cells/ L for < 7 days);
• Good performance status (ECOG [Eastern Cooperative Oncology Group] score of 0–1);
• No hepatic insufficiency;
• No renal insufficiency; and/or
• MASCC Risk Index score 21.
In predicting risk in febrile neutropenic cancer patients, the MASCC Risk Index assigns weighted numerical values to variables such as burden of illness (signs or symptoms at presentation), hypotension, chronic obstructive pulmonary disease (COPD), solid tumor or hematologic malignancy, history of previous fungal infection, dehydration, outpatient status, and age. The score predicts the probability that a patient will experience serious complications. It may also be used as a guide to determine whether the patient is a candidate for outpatient management with oral antibiotics.
Initial laboratory tests should include a complete blood count with differential analysis, platelets, blood urea(Drug information on urea) nitrogen, creatinine, electrolytes, total serum bilirubin, and liver-associated enzymes. Chest radiographs should be taken for all patients with respiratory signs or symptoms. A source of infection may not be evident in neutropenic patients with pulmonary infection, however. A computerized tomography (CT) scan of the chest may be more effective in detecting a source of infection.
Culture specimens should be collected during or immediately after completing the history and physical examination. It is essential that the samples be collected prior to the initiation of antibiotic therapy; there is concern regarding the potential for lower yields from cultures if antibiotics are provided prior to culture collection. Two sets of blood cultures should be obtained. (Each set consists of two bottles: one aerobic sample and one anaerobic sample.) If the patient has a vascular access device (VAD), the NCCN panel recommends obtaining one sample from the catheter and one sample peripherally. This may be helpful in determining whether the VAD is the source of infection based on the differential time to positivity. Alternatively, both sets can be obtained peripherally or through the catheter. The positive predictive value of catheter cultures is lower than that of peripheral cultures. If a patient is suspected to have a catheter-related bloodstream infection (CRBSI) and a blood sample cannot be drawn from a peripheral vein, the IDSA recommends that two or more blood samples be drawn from different catheter lumens. It is unclear whether blood samples should be drawn from all catheter lumens in such cases. The need for samples from both peripheral and central sites also remains unclear at this time. The NCCN panel consensus is that the volume of blood for culture is the most important aspect of blood culturing. The recommended volume is 20–30 mL.
The NCCN panel states that routine cultures of the urine, stool, rectum, anterior nares, and oropharynx are rarely helpful in the absence of lesions or clinical signs and symptoms. In the presence of diarrhea, however, stools should be tested for the presence of Clostridium difficile. Especially in the winter months and during outbreaks, diarrheal stools should be tested for rotavirus and norovirus, as part of the differential diagnosis. A urinalysis and urine culture should be done in the presence of symptoms of urinary tract infection; microbiologic and pathologic evaluations should be performed for new and undiagnosed skin lesions; and viral cultures and rapid viral antigen testing of nasopharyngeal secretions can be useful in evaluating respiratory viral symptoms. It must be noted, however, that inability to mount an effective inflammatory response may decrease the reliability of some laboratory tests. One study from the 1970s found that 40% of neutropenic patients with pneumonia had a normal chest radiograph and only 8% of patients had sputum Gram stains revealing polymorphonuclear cells. In more recent research investigators have emphasized that while pneumonia in febrile neutropenic patients needs to be diagnosed as early as possible, conventional chest radiographs often fail to detect lung infiltrates at an early stage in these patients, “meaning that a normal chest radiograph must be interpreted with caution,” and recommending thoracic computed tomography (CT) scans in this setting.