The Appropriate Neutrophil Mediated Inflammatory Response

Published Date: 02 Nov 2017

INTRODUCTION

Cancer patients, who are receiving cytotoxic antineoplastic therapy sufficient to harmfully affect myelopoiesis and the integrity of the gastrointestinal mucosa, are at high risk for invasive infection due to the translocation of colonizing bacteria and/or fungi across intestinal mucosal surfaces.

Since neutropenic patients typically lack the appropriate neutrophil-mediated inflammatory response 1, physical findings of exudate, fluctuation, ulceration or fissure, local heat, swelling, and regional adenopathy are all less prevalent in the neutropenic patient1. Thus, fever might be the earliest and only sign of a severe underlying infection 2.

With the increasing use of myelo-suppressive agents in the treatment of neoplastic and nonneoplastic diseases, the increased rate of infection in patients with neutropenia has been clearly established 3. However, many of these commonly fatal infections go unrecognized until autopsy 4.

Therefore, in order to avoid unfortunate outcomes such as sepsis and possibly death, the early recognition of neutropenic fever and prompt initiation of empiric systemic antibacterial therapy is crucial. Risk assessment for serious complications is also crucial in patients with febrile neutropenia, because this assessment will guide the management, including the need for hospitalization, IV antibiotics, and the length of stay 2.

The definitions of neutropenia and fever, the risk assessment and diagnostic evaluation are reviewed here. Treatment and prophylaxis of infection in febrile neutropenic patients are also discussed here.

DEFINITIONS

Fever:

Many definitions of fever as a sign of infection in patients with neutropenia have been proposed. Carl Wunderlich proposed in 1868, that the mean normal body temperature was 37�C (98.6�F) with an upper limit of normal of 38�C (100.4�F), above this limit, fever was defined 5,6.

Although it has been observed that there is a variety of normal body temperatures, according to a survey of 270 medical professionals, 75 percent of subjects reported that normal body temperature is 37�C (98.6�F)5,7.

While, a study of members of the British Society for Hematology concerning their institutional definitions of fever identified ten definitions of fever, varying from a single temperature >37.5�C to either a single temperature >39�C or two successive temperatures >38.4�C 5,6.

Despite these beliefs, the mean oral temperature was reported as 36.8�0.4�C (98.2�0.7�F) with a range of 35.6�C (96.0�F) to 38.2�C (100.8�F), after observation of 148 healthy adults aged between 18 and 40 years 6.

The definition of fever in neutropenic patients, according to The Infectious Diseases Society of America, was proposed as a single oral temperature of >38.3�C (101�F) or a temperature of >38.0�C (100.4�F) sustained for >1 hour2. It has been approved to use this definition of fever in neutropenic patients9�11.

Neutropenia:

The definition of neutropenia differs from institution to institution, but it is generally defined as an absolute neutrophil count (ANC) <1500 cells/microliter.

The absolute neutrophil count (ANC) is defined as the product of the white blood cell count (WBC) and the percentage of polymorphonuclear cells (PMNs) and bands:

ANC = WBC (cells/microliter) x percent (PMNs + bands) � 100

Based upon the level of ANC, neutropenia is categorized as mild, moderate or severe. An absolute neutrophil count between 1000 and 1500/microL corresponds to mild neutropenia. While, an ANC between 500 and 1000/microL corresponds to moderate neutropenia. Severe neutropenia is usually defined as an ANC <500 cells/microL, or an ANC that is expected to drop to <500 cells/microL over the next 48 hours2.

As the neutrophil count drops below 500 cells/microL, the risk of clinically serious infection increases and is higher in those with a prolonged duration of neutropenia (>7 days). However, the risk is also correlated to the sufficiency of the marrow store pool of granulocytes.

Two terms, leukopenia and granulocytopenia are often used interchangeably with neutropenia, although they are somewhat different:

� Leukopenia is defined as a low total white blood cell count that may be due to any cause such as lymphopenia and/or neutropenia; yet, almost all leukopenic patients are neutropenic since the amount of neutrophils is so much higher than the amount of lymphocytes.

� Granulocytopenia is defined as a reduced absolute number of all circulating cells of the granulocyte series (neutrophils, eosinophils, and basophils); yet, almost all granulocytopenic patients are neutropenic since the amount of neutrophils is so much higher than the amount of eosinophils and basophils.

� Agranulocytosis is defined as the absence of granulocytes, but the term is often inaccurately used to denote severe neutropenia.

RISK CATEGORIES

Risk Factors

The risk factors for the development of neutropenic fever can be divided into three sub-categories including patient-related, disease-related and anti-cancer treatment-related predictors.

Patient-related predictors include: age = 65 years, female sex, high body surface area, poor performance status based upon preexisting active comorbidities (e.g., cardiovascular, pulmonary, renal, endocrine, etc.) and poor nutritional status12�19.

Disease-related predictors include: Elevated lactate dehydrogenase (LDH) in patients with lymphoreticular diseases, bone marrow failure due to replacement of hematopoietic tissue by anomalous tissue which is know as myelophthisis19, lymphopenia20,21 and advanced stage of the underlying malignancy 13,16,20�22

Treatment-related predictors include: administration of the planned dose-intensity of high-dose chemotherapy regimens15,20�23 and failure to administer prophylactic hematopoietic growth factor support to patients receiving high-risk regimens14,21.

Risk of Serious Complications

This risk assessment is essential to determine the management of patients, including the need for hospital admission, IV antibiotics, and length of stay. High-risk patients require hospital admission for IV antibiotics, and often a prolonged length of stay. In contrast, low-risk patients may be treated as outpatients with oral antibiotic after a short period of observation or hospitalization.

Definitions of high-risk and low-risk patients:

The Infectious Diseases Society of America (IDSA) and National Comprehensive Cancer Network (NCCN) use different definitions in their guidelines:

� Low-risk patients are those who are expected to have neutropenia (absolute neutrophil count [ANC] <500 cells/microL) for =7 days, those with no comorbid conditions or evidence of significant renal or hepatic impairment2, and most patients receiving chemotherapy for solid tumors. This group has been considered to be at low-risk for complications requiring hospitalization or prolonging hospitalization after being well studied by randomized trials.

� High-risk patients are those expected to have neutropenia (ANC <500 cells / microliter) for > 7 days. While, neutropenic febrile patients with comorbid conditions or evidence of significant renal or hepatic impairment are considered high risk, independently of the duration of neutropenia. Some experts have identified patients at high risk as those who are expected to have profound neutropenia (ANC = 100 cells / microliter) for > 7 days on the basis of experience that these patients are more likely to have serious and potentially fatal complications2,24. Nevertheless, formal studies to clearly distinguish patients with a neutrophil count <500 cells / microliter and those with a neutrophil count = 100 cells / microliter are missing.

Some studies combine these groups to define high-risk patients.

Deep prolonged neutropenia (ANC = 100 cells / microliter expected to last > 7 days) is more likely to occur in the pre-transplant hematopoietic cell transplantation (allogeneic in particular) and in patients receiving induction chemotherapy for acute leukemia.

Risk based on underlying disease

Patients who suffer from neutropenia after induction chemotherapy for acute myelogenous leukemia or as part of the conditioning regimen for allogeneic hematopoietic stem cell transplantation (HCT) are at a high-risk for serious infections.

Other factors that are considered as high-risk status include gastrointestinal and oral mucositis, uncontrolled cancer, chronic obstructive pulmonary disease, advanced age and poor functional status.

Patients receiving consolidation chemotherapy for leukemia or undergoing autologous HCT may also have long periods of neutropenia, but seem to be at somewhat lower risk, especially if they received prophylactic hematopoietic growth factors.

In contrast, patients with solid tumors are mostly at low risk for serious infections.

Guidelines

An assessment of risk (high versus low-risk) for medical complications related to neutropenic fever should be obtained at the initial assessment of neutropenic fever episode. The Infectious Diseases Society of America (IDSA), the European Society for Medical Oncology (ESMO), the National Comprehensive Cancer Network (NCCN) as well as the American Society of Clinical Oncology (ASCO)2,24,25 , has recommended this.

The IDSA and ASCO defined high-risk neutropenic patients as those who are expected to have profound neutropenia (ANC = 100 cells / microliter) for > 7 days or those with evidence of current comorbidities or hepatic or renal dysfunction2,24 . The National Comprehensive Cancer Network (NCCN) has used similar criteria for definition, but also includes a category of intermediate risk [21].

Multinational Association of Supportive Care in Cancer (MASCC) risk index that can be used as an alternative to clinical criteria, is a validated tool to assess the risk of medical complications associated with neutropenic fever (calculator 2)26�29 .

IDSA Risk assessment:

The Infectious Diseases Society of America (IDSA) has established the following criteria for the definition of high risk or low risk patients with neutropenic fever 2:

High-risk febrile neutropenic patients are defined as having one or more of the following criteria:

? Profound neutropenia (ANC = 100 cells / microliter) expected to last for > 7 days.

? Proof of current comorbidities, such as (but not limited to):

? Hemodynamic instability

? Oral mucositis limiting swallowing or gastrointestinal tract mucositis causing severe diarrhea

? Gastrointestinal symptoms such as abdominal pain, nausea and vomiting or diarrhea

? Changes in neurological status or mental appearance of new onset

? Intravascular catheter infection

? New pulmonary infiltrates or hypoxia

? Underlying chronic lung disease

? Signs of hepatic insufficiency (serum transaminase> 5 times normal) or renal insufficiency (creatinine clearance <30 mL / min)2

Low-risk febrile neutropenic patients are expected to have a relatively short duration of neutropenia for 7 days or less, with an absolute neutrophil count (ANC) <500 cells / microliter and have no comorbid conditions or evidence of significant renal or hepatic impairment2 . In addition, most patients taking chemotherapy for solid tumors are considered at low-risk for complications.

Patients with evidence of severe sepsis (sepsis syndrome in end organ dysfunction) should be considered at high risk and managed as in-patients with initial intravenous antibacterial empirical treatment. While, patients with signs of septic shock should be managed in an intensive care unit based upon goal-oriented therapy30 .

NCCN risk assessment:

The National Comprehensive Cancer Network (NCCN) has developed certain criteria to classify patients as high risk or low risk, which must be performed during the initial evaluation [21].

High-risk febrile neutropenic patients are those having one or more of the following criteria: [21]

? Development of fever during hospitalization

? Evidence of significant medical comorbidity or the presence of clinical instability

? Expected profound prolonged neutropenia (ANC = 100 cells / microliter expected to last> 7 days)

? Hepatic insufficiency (serum transaminase> 5 times normal) or renal insufficiency (creatinine clearance <30 mL / min)

? Any patient with leukemia not in complete remission, or any non-leukemic patient with signs of disease progression after more than two courses of chemotherapy.

? Any complex infection such as pneumonia at clinical presentation

? Alemtuzumab (antineoplastic agent) in the last two months

? Grade 3 or 4 mucositis

? MASCC risk index score <21

Low-risk febrile neutropenic patients are those who do not meet any of the criteria for high-risk described above and meet most of the criteria as follows [21]:

? Development of fever in outpatients

? No acute comorbid illness requiring hospitalization and close monitoring

? Expected short duration of severe neutropenia (ANC = 100 cells / microliter should last for 7 days or less)

? Good performance status (Eastern Cooperative Oncology Group [ECOG] 0-1

? No hepatic or renal insufficiency

? MASCC risk index score of = 21 risk index

Intermediate risk neutropenic patients are defined as those meeting one or more of following criteria: [21]

? Patients undergoing autologous HCT

? Lymphoma

? Chronic lymphocytic leukemia

? Multiple Myeloma

? Patients receiving purine analogue therapy

? The expected duration of neutropenia is 7 to 10 days

For patients at intermediate risk, the NCCN recommends consideration of fluoroquinolone prophylaxis.

Multinational Association of Supportive Care in Cancer (MASCC) score:

As an alternative to the IDSA and NCCN risk assessments described above, the MASCC risk index is validated for assessing the risk of medical complications associated with febrile neutropenia.

Using the MASCC risk index, the following features are assessed and given a weighted score2,26 :

? Burden of disease (clinical condition of the patient at the time of presentation with neutropenic fever):

? No symptoms or mild symptoms (5 points)

? Moderate symptoms (3 points)

? Severe symptoms or dying (0 point)

? No hypotension (systolic blood pressure> 90 mmHg) (5 points)

? No chronic obstructive pulmonary disease COPD (4 points)

? Solid tumor or hematologic malignancy without prior history of fungal infections (4 points)

? No dehydration that requires parenteral fluids (3 points)

? Ambulatory status at the time of the onset of neutropenic fever syndrome (3 points)

? Age <60 years (2 points)

The highest possible score is 26.

Patients with a score = 21 are considered to be at low risk of serious medical complications, and for whom outpatient treatment with oral empirical antimicrobial can be safe and effective29 .

While, patients with a score <21 are considered to be at high risk for serious medical complications, and therefore should be treated as inpatients.

The MASCC risk index has classified 98% of patients as low-risk and 86% as high risk with a sensitivity and specificity of 95%, and positive and negative predictive value of 98, and 86 percent, respectively 28.

Patients with complicated infections have been reclassified as high risk for serious medical complications, which further increased the predictive value of the model. Complicated infections include non-necrotizing skin or soft tissue infection (SSTI) of >5 cm diameter, necrotizing SSTI of any size, grade 2 oral mucositis, sepsis syndrome or the presence of a visceral site of infection. [28]. The classification error rate has been 10 to 29 percent. [4]

In addition, the MASCC risk index can predict the probability of death as follows:27

? Score = 15: 29 %

? Score = 15 but <21: 9 %

? Score = 21: 2 %

The MASCC risk index has been criticized for the lack of a standard definition of this criterion "the burden of febrile neutropenia", which could be a source of confusion2, or it could be interpreted differently by different clinicians. I addition, the MASCC risk index does not include the duration of neutropenia as a criterion, though it is considered as an important predictor of risk2.

The MASCC risk index has been also criticized because it was developed using heterogeneous patient populations; thus, it might not function optimally in all populations. For example, in a retrospective study of patients with solid tumors who seemed to be clinically stable, the MASCC risk index had a low sensitivity to detect complications (36 percent)31. The low sensitivity was likely to be attributed to the fact that patients were all outpatients, and the rates of hypotension, dehydration and invasive fungal infections were low; hence, only three criteria were present to distinguish prognosis.

The serious medical complications are provided by the MASCC risk index as follows 26:

? Hypotension defined as systolic blood pressure <90 mmHg or the need for vasopressor support to maintain blood pressure

? Respiratory failure defined as arterial oxygen pressure <60 mmHg on room air or the need for mechanical ventilation

? The admission to ICU

? Disseminated intravascular coagulation

? Presence of confusion, delirium, or altered mental status

? The development of congestive heart failure documented by chest imaging and requiring treatment

? Bleeding diathesis sufficient to require a blood transfusion

? Electrocardiogram changes or arrhythmias requiring treatment

? Renal failure sufficient to require an investigation and / or treatment with IV fluids, dialysis, or other intervention

? Other complications judged serious and clinically significant by the health care team

All patients who were treated with systemic antineoplastic therapy six weeks prior to a systemic inflammatory response syndrome (SIRS) are assumed to have neutropenic sepsis syndrome until proven otherwise. SIRS is defined by the presence of two or more of the following conditions: temperature >38�C or <36�C, heart rate> 90/minute, respiratory frequency > 20/minute, PaCO2 <32 mmHg32.

Patients presenting with altered mental status, hypotension, hypoxia, oliguria or any other sign of new organ impairment must be managed emergently for severe sepsis.

Risk Of Treatment Failure

The risk of failure to respond to initial empirical antibacterial therapy is a composite outcome to be considered clinicians. Treatment failure is proposed if one or more of the following events occur within 30 days after the start of treatment 33,34:

? Persistence, progression or recurrence of signs of infection

? Modification of the initial empirical antibacterial treatment

? Readmission to the hospital for outpatients

? Death

Patients with documented clinical or microbiological infections are more likely to be at risk for treatment failure, clinical or microbiological than for unexplained neutropenic fever (39 against 18 percent33.

High-risk patients are more likely to be at risk for treatment failure than those with low risk. For example, patients with hematologic malignancies have a higher percentage of treatment failure than those with solid tumors (44 against 18 percent) 33.

Observations have shown that among all febrile neutropenic patients at low risk of medical complications, adult patients at higher risk for treatment failure than children with 16% against 5% respectively34.

PREVENTION

In order to prevent neutropenic fever and infectious complications in patients at increased risk, the administration of an antimicrobial drug should be used as a prophylaxis.

Antibacterial prophylaxis

Pseudomonas aeruginosa and other gram-negative bacilli is the target of the antibacterial prophylaxis, because these pathogens are virulent and may cause life-threatening infections.

Indications

The beneficial effect on clinical outcomes has been sought from the administration of prophylactic antibacterial agents. The fluoroquinolones, levofloxacin (500 mg orally once daily) and ciprofloxacin (500 mg orally twice daily) have been the most studied antibacterial agents. Levofloxacin in particular is preferred in patients at increased risk for oral mucositis-related Streptococcus viridans infection 2. Results have been mixed with respect to effectiveness and have incited concern about toxicities and antibacterial resistance35�37. A systematic monitoring of the prevalence of fluoroquinolone resistance among gram-negative bacilli should be done, at the intitutions that use fluoroquinolone prophylaxis.

Based upon the available data, high-risk neutropenic patients defined by those who are expected to have an absolute neutrophil count <500 cells/microL for more than 7 days and who do not have a contraindication to receiving a fluoroquinolone, are recommended to receive fluoriquinolone prophylaxis. While, intermediate-risk patients are sometimes given an antibacterial prophylaxis, a decision made on a case-by-case basis.

Fluoroquinolone should be used with caution in patients at risk of a prolonged QT interval particularly in those who may require other QT prolonging agents, such as voriconazole.

In addition, the potential to promote resistance among gram-negative and gram-positive should be considered when deciding whether to give a fluoroquinolone prophylaxis or not37. Concerns about the possibility of increasing the risk of Clostridium difficile infection has also been present, though this has not been proven in neutropenic patients receiving fluoroquinolone prophylaxis37. The use of prophylactic agents in institutions and geographic areas where the levels of resistance to fluoroquinolones are high is less likely to be efficient36,38.The use of antibacterial prophylaxis varies from one center to another with some centers avoiding such practices.

For most patients with chemotherapy-induced neutropenia expected to be of short duration particularly patients with solid tumors, the use of antibacterial prophylaxis is not recommended.

Timing

The ideal timing for the initiation and cessation antibacterial prophylaxis has not been sufficiently studied2. Many clinicians begin anti-bacterial prophylaxis, the first day of chemotherapy or the day after the administration of the last dose of chemotherapy cycle. Antibacterial prophylaxis is usually withheld when neutropenia resolves, or when empirical antibacterial regimen is initiated for patients who become febrile during neutropenia.

Antifungal prophylaxis

Among cancer patients and HCT recipients, a high rate of life-threatening invasive fungal infections such as candidemia has been observed since the late 1980s, which incited interest in antifungal prophylaxis for patients receiving chemotherapy.

Antiviral prophylaxis

Influenza

Annual immunization with an inactivated influenza vaccine is recommended for all cancer patients undergoing treatment2. The influenza vaccine is generally administered >2 weeks before the initiation of chemotherapy or, when circumstances dictate, between chemotherapy cycles and at least seven days after the last cycle. However, the best timing for such immunization has not been established2. All family members and other close contacts should get annual immunization too.

HSV and VZV

Reactivation of herpes simplex virus (HSV-1 and HSV-2) and varicella-zoster virus (VZV) occur commonly in HCT recipients who are not receiving prophylaxis and are important causes of morbidity. However, reactivation of both HSV and VZV infections can be effectively prevented with antiviral prophylaxis.

Antiviral prophylaxis with acyclovir (400 mg orally three to four times daily or 800 mg orally twice daily) or valacyclovir (500 mg orally once or twice daily) is recommended in all patients who are seropositive for HSV and who are undergoing allogeneic HCT or induction chemotherapy for acute leukemia39. Antiviral prophylaxis with acyclovir or valacyclovir is also recommended in all HCT recipients who are seropositive for VZV. Based upon randomized trials, benefits of antiviral prophylaxis in these populations have been demonstrated; thus, recommended41.

CMV

CMV prophylaxis is indicated for HCT recipients because they are at are at significant risk for reactivation. In contrast, prophylaxis is not indicated in patients with chemotherapy-induced neutropenia, because it does not occur commonly.

Hepatitis B

Antiviral prophylaxis should be considered for the following categories of patients and should be sustained for at least six months after the completion of chemotherapy40 :

? Patients receiving chemotherapy who have a previous history of hepatitis B virus infection, due to the risk of reactivation and hepatic failure.

? Patients with elevated circulating hepatitis B DNA or detectable levels of circulating hepatitis B surface antigen (HBsAg)

? Patients with a previous history of infection with detectable levels of antibody to HBsAg or to hepatitis B core antigen.

This has been demonstrated to be able to reduce the risk of reactivation from 24 to 53 percent to 0 to 5 percent.

Colony stimulating factors

Granulocyte colony stimulating factors (CSFs) have been widely evaluated for prophylactic use following the administration of intensive cytotoxic chemotherapy when neutropenia is expected (primary prophylaxis). CSFs have been also evaluated for their prophylactic use during retreatment after a previous cycle of chemotherapy that caused neutropenic fever (secondary prophylaxis), and have been shown to minimize the extent and duration of severe chemotherapy-induced neutropenia in afebrile patients (afebrile neutropenia). Their use is not recommended in febrile chemotherapy-induced neutropenia2. However, prophylactic use of granulocyte CSFs has not been shown to have an effect on survival in most clinical situations.

Primary Prophylaxis

Primary prophylaxis denotes the use of granulocyte CSFs during the first cycle of myelosuppressive chemotherapy in order to prevent neutropenic complications. The goal of primary prophylaxis is to decrease the incidence of neutropenic fever and the need for hospitalization, to maintain dose-dense or dose-intense chemotherapy strategies that have survival benefits.

Updated 2010 guidelines from the European Organization for Research and Treatment of Cancer (EORTC), the Infectious Diseases Society of America (IDSA), consensus-based guidelines from the National Comprehensive Cancer Network (NCCN), and The 2006 guidelines from the American Society of Clinical Oncology (ASCO), all recommend primary prophylaxis when the expected incidence of neutropenic fever is over 20 percent, to reduce the need for hospitalization for antibiotic therapy2,22,41. These recommendations are based upon randomized trials that have shown that primary prophylaxis was cost effective when the risk of neutropenic fever with a specific regimen was over 20 percent42,43 .

In contrast, guidelines recommend against the routine use of granulocyte CSFs for primary prophylaxis in adult patients receiving chemotherapy regimens with an anticipated low probability (<10 percent) of neutropenic fever2,22,41 .

However, when the anticipated risk of neutropenic fever is between 10 and 20 percent, the decision of primary prophylaxis should be individualized and may be appropriate in a number of clinical settings in which patients are at risk or increased complications22,41 :

� Age >65 years

� Preexisting neutropenia

� More advanced cancer

� Poor performance and/or nutritional status

� Renal or hepatic impairment

� In the case of epithelial ovarian cancer

� Extensive pre-chemotherapy surgery, particularly if it included a bowel resection.

In patients receiving concomitant chemo radiotherapy for either head and neck cancer or lung cancer, the use of granulocyte CSFs has been associated with adverse outcomes, therefore, it better be avoided.

Despite the lack of comparative data from randomized controlled trials, that could recommend one CSF over the other for prophylaxis of infection during neutropenia, practically most institutions use G-CSF.

Secondary Prophylaxis

Secondary prophylaxis denotes to the use of a granulocyte CSF in subsequent chemotherapy cycles after a prior cycle has caused neutropenic fever. Secondary prophylaxis with CSFs reduces the risk of reccurence of neutropenic fever by approximately one-half 44.

ASCO and EORTC guidelines recommend that secondary prophylaxis with granulocyte CSFs be limited to patients for whom primary prophylaxis was not given and who experience a neutropenia-related complication from a prior cycle of chemotherapy if neutropenic fever would prevent the administration of full dose chemotherapy and if reduced dose intensity might affect treatment outcome22,41 .

Timing

G-CSF and GM-CSF therapy is usually initiated 24 to 72 hours after cessation of chemotherapy and is frequently continued until the absolute neutrophil count reaches 5000 to 10,000/microliter. A reasonable alternative is continuation until clinically adequate neutrophil recovery.

MANAGEMENT

Initial Assessment

Since, fever might be the first and only sign of infection in a neutropenic patient, its occurrence should be considered a medical emergency. Therefore, empiric broad-spectrum antibacterial therapy should be started immediately after obtaining blood cultures and before the completion of any other investigation. The Infectious Diseases Working Party of the German Society of Hematology and Oncology and the Northern Ireland Cancer Network has recommended this.

It has been recommended that empiric broad-spectrum antibacterial therapy should be administered within one hour of presentation for all patients suffering from neutropenic fever at presentation.

Diagnostic Approach

At presentation, a detailed history and physical examination should be done, as well as a complete laboratory, microbiologic and imaging work-up for all febrile neutropenic patients.

The table below summarizes the diagnostic approach to patients with febrile neutropenia.

Item Purpose Comments

Directed history Look for suspicious sites of infection Detection of infection-related symptoms

Detailed physical exam (digital rectal exam to be avoided) Identifies infectious sites and guides biological and imaging studies Since, the neutrophil response is muted, pus won�t be found; erythema or pain might guide to infection; chest exam might be normal even with pneumonia; abdominal tenderness might refer to neutropenic enterocolitis; perianal tenderness might guide to gram-negative or anaerobic infection

CBC with differential Identifies the intensity of neutropenia The likelihood of serious infections is inversely proportional to the initial neutrophil count; daily counts estimate the prognosis

Creatinine, electrolytes, liver function tests Identifies current comorbidities Guides the selection and posology of antibacterial agent(s) and allows serial monitoring of toxicities

2 sets of blood culture (peripheral and from central venous catheter) and antibiogram Identifies bacteremia Allows modification of antibiotic therapy if necessary

Cultures and gram stains of samples from suspicious sites of infection Identifies the cause of infection Bacterial and fungal stains can be useful

Imaging studies Identifies the site of infection CT scans are usually more helpful than plain radiographs that might be normal during profound neutropenia

Empiric Therapy

The goal of empiric therapy is to cover the spectrum of the most likely and most virulent pathogens that can cause serious or life-threatening infection in

neutropenic patients2. Despite the frequency of infection with gram-positive bacteria, it is critical to cover with broad-spectrum antibiotics for gram-negative organisms because these pathogens are virulent and associated with sepsis. In addition, antibiotics should be bactericidal.

The patient�s history, allergies, symptoms and signs, current antibiotic use and culture data, knowledge of institutional nosocomial infection and susceptibility patterns should direct initial antibiotic selection.

Low-risk patients

Low-risk patients with neutropenic fever can be given oral antibiotic as outpatient after a brief period of observation or hospitalization. It is recommended that these patients be monitored for at least four hours after the initial dose of antibiotics before discharge. The recommended initial empiric oral antibacterial regimen is a combination of a fluoroquinolone (2,24 750 mg orally twice daily or levofloxacin 750 mg orally once daily) and a beta-lactam agent such as amoxicillin-clavulanic acid (500 mg/125 mg orally three times daily)2,24.

Since, fluoroquinolone is the only oral antibiotic that covers P. aeruginosa, patients receiving fluoroquinolone prophylaxis should not receive a fluoroquinolone-based initial empiric antibacterial therapy but one of the IV regimens available for high-risk patients that provide coverage against P. aeruginosa2,25.

Outpatients receiving oral antibiotic therapy should be able to access medical care 24 hours daily, seven days per week, and should be able to reach the hospital within one hour if their clinical condition worsens. Persistent fever or new signs of infection require hospitalization for IV antibacterial therapy2,24.

High-risk patients

As recommended by the Infectious Diseases Society of America (IDSA), the initial empiric antibacterial therapy for high-risk patients with neutropenic fever involves a monotherapy that targets P. aeruginosa. Anti-pseudomonal agents that can be used are cefepime (2 g IV every eight hours), imipenem (500 mg IV every six hours), meropenem (1 g IV every eight hours), ceftazidime (2 g IV every eight hours) or piperacillin-tazobactam (4.5 g IV every six hours) 2.

However, because of rising rates of resistance, ceftazidime is avoided at most centers.

The routine addition of Vancomycin and other agents that target gram-positive cocci to the initial antibacterial regimen is not recommended. In contrast, antibiotics with gram-positive coverage should be added for patients who are suspected to have a central venous catheter-related infection, soft tissue or skin infection, pulmonary infection and those who are hemodynamically unstable.

For patients presenting with complications or antimicrobial resistance, broad-spectrum antibiotics should be used to cover the most likely organisms including resistant bacteria (gram negative and gram positive), anaerobic bacteria, and fungi.

The following algorithm summarizes the initial management of neutropenic fever2.

Empiric Therapy For Persistent Fever

Recommendations are against the modification of the initial antibiotic regimen for clinically stable patients with persistent fever alone. Nevertheless, modifications to the initial regimen should be sought for patients at risk for infection with resistant organisms, for patients who are or become clinically or hemodynamically unstable, and for patients with positive blood cultures suggestive of a resistant infection.

Patients with persistent fever after the start of empiric antibiotics should be reassessed for possible sources of infection after two to four days and after four and more days. Management algorithms for the reevaluation of neutropenic patients with persistent fever are as follows2 :

For patients who are still febrile four to seven days after a an appropriate therapy with broad-spectrum antibiotics and who have no documented infectious source, the empiric addition of an antifungal agent is recommended2

Amphotericin B deoxycholate, caspofungin, voriconazole, or itraconazole are reasonable choices for empiric antifungal therapy in patients with chemotherapy-induced neutropenia, as recommended by the 2010 IDSA guidelines.

Among these agents, the appropriate choice depends upon which fungi are most likely causing infection, as well as toxicity profiles2:

? Caspofungin (or another echinocandin) is favored for patients with persistent fever, not receiving antifungal prophylaxis and with no apparent site of infection, because Candida species are the most frequent cause in these patients.

? Voriconazole or a lipid formulation of amphotericin B is favored for patients with persistent fever and nodular pulmonary infiltrates or pulmonary nodules suggestive of an invasive mold infection45.

? Patients receiving anti-mold prophylaxis and still suffering from persistent fever, another class of antifungal agent with anti-molds coverage should be administered for empiric therapy.

Catheter Removal

Based upon observational studies, in addition to antibacterial therapy, central venous catheter (CVC) removal has been recommended for patients with positive bloodstream culture suggestive of a CVC infection caused by Staphylococcus aureus, Pseudomonas aeruginosa, Candida species, fungi other than Candida species, or rapidly growing nontuberculous mycobacteria.

CVC removal is also recommended in patients with complicated infections such as tunnel infection, port pocket infection, persistent bloodstream infection despite > 3 days of appropriate antibiotics, etc. 2

Methods and Materials

This is an epidemiologic, descriptive, prospective study conducted from September 2011 till December 2012. All patients who are more than 18 years old who were admitted to the hematology oncology department of Notre Dame de Secours Jbeil with the diagnosis of chemotherapy-induced neutropenic fever were included and neutropenic patients who were hospitalized and developed fever during their hospital stay were also included.

Neutropenic fever was defined as follows:

1. ANC <1500 cells/microL.

2. A single oral temperature of >38.3�C (101�F) or

3. A temperature of >38.0�C (100.4�F) sustained for >1 hour.

Patients who didn�t meet these criteria were excluded.

Thus, a total of 51 patients who were diagnosed by specialists of having febrile neutropenia were included.

The data set included the information on age, sex, type of cancer, WBC counts with differential counts at presentation, days since last cycle of chemotherapy, site of infection, days until recovery.

Patients� evaluation was initially done in the clinic or in the emergency room and a risk assessment was done at presentation using the MASCC risk index score.

A comprehensive history, complete physical examination, and all required radiological and laboratory investigations were done on all patients.

Based upon the latest guidelines for the treatment of neutropenic fever, all patients were treated initially with broad-spectrum intravenous empiric antibiotics. These antibiotics were later adjusted and modified based on culture results. The addition of vancomycin and amphotericin B was established also based upon the guidelines.

Patients were reevaluated 2-4 days after fever and 4 days after fever and antibiotics were modified or added when necessary.

All patients were treated in the oncology hematology unit while neutropenic patients with septic shock were managed in the intensive care unit.

Antibiotics were discontinued 48 hours after afebrile status.

Transfusion of blood products was done as per requirement of individual patient.

Study Variable

Outcome of neutropenic fever, discharge/death

Independent Variable

Age and ANC at presentation were collected as continuous and later classified as >50 and <50 for age and an arbitrary cutoff was later created for ANC. AMC at presentation was also collected as continuous.

Statistical Analysis

Results were expressed by means � standard deviations (SD) and percentages for normally distributed continuous variables. Groups� comparisons were realized using Pearson Chi-square and Student t test according to the type and distribution of variables.

Level of significance was defined as a two-tailed p-value = 0.05. Statistical analysis was realized using IBM SPSS 20.0 (SPSS Inc., Chicago,IL,USA).

Results

During the study period, a total of 51 patients with diagnosed neutropenic fever were identified and followed. The mean age was 55.67 (24-83) years.

Patients who were >50 years of age were slightly more than patients who were <50 years of age with 29 (56.9%) and 22 (43.1%) respectively. As for gender distribution, there were 35 (68.6%) females and only 16 (31.4%) males.

There was almost equal distribution of solid and hematological malignancies with 26 (51%) patients with solid tumor and 25 (49%) patients with hematological malignancies.

More patients 28 (54.9%) were ambulatory and developed febrile neutropenia while 23 (45.1%) patients were diagnosed during their hospital stay. Using MASCC index score, there were 27 (52.9%) patients with low risk and 24 (47.1%) patients with high risk.

The mean ANC at presentation was 210.98 cells / microliter (0-1400), mean AMC at presentation was 165.80 cells / microliter (0-732).

The mean number of days since last cycle of chemotherapy was 12.08 (3-60) days. There were 33 (64.7%) patients who developed febrile neutropenia more than 7 days since last chemotherapy.

The demographics and clinical characteristics of study patients are shown in Table 1.

Table 1. Baseline Dermographic and Clinical Caracterestics of Patients with Febrile Neutropenia.

Characteristics No. Of Patients

Mean age, years 56

Sex, Female/male 35/16

Types of cancer (%)

HEMATOLOGICAL MALIGNANCIES 25(49%)

a. Lymphoma 5(9.8%)

b. Leukemia 19(37.3%)

c. Multiple Myeloma 1(2%)

SOLID TUMOR 26(51%)

a. Sarcoma 2(3.9%)

b. Head & Neck 4(7.8%)

c. Lung 3(5.9%)

d. Breast 6(11.8%)

e. Genitourinary 3(5.9%)

f. Gastrointestinal 7(13.7%)

g. Thymoma 1(2%)

Mean WBC at presentation

Mean ANC at presentation 210.98 � 322.175

ANC < 100/mm3, No. (%) 28(54.9%)

ANC b/w 100-500/mm3, No. (%) 19(37.3%)

ANC b/w 500-1000/mm3, No. (%) 1(1.9%)

ANC b/w 1000-1500/mm3, No. (%) 3(5.9%)

Mean Monocyte at Presentation 165.80� 156.669

Mean No. of Days since last cycle

Occurrence of Neutropenia

Outpatient/Inpatient

MASCC Index score

Low risk/High risk 12.08 � 11.343

28/23

27/24

Most patients who recovered and were discharged were <50 years of age, whereas all patients who died were >50 years (P-value of 0.02).

There were 32 females who were discharged and only 3 died. While, 13 males were discharged and 3 died. Gender distribution had no statistical significance.

There was no significant difference between patients with solid or hematological tumor in terms of prognosis.

Mean ANC at presentation was 229.51 cells/mm3 in patients who had recovered and were discharged and 72 cells/mm3 in patients who died during their hospital stay (P value 0.27). There were 28 (54.9%) patients with ANC less 100 cells/mm3, 10(19.6%) between 100-300 cells/mm3 and 13 (25.5%) greater than 300 cells/mm3.

Twelve (23.5%) patients had ANC recovery in 1-3 days, seventeen (33.3%) patients showed ANC recovery in 4-7 days. While, sixteen (31.4%) patients needed more than 7 days till ANC recovery out of which two patients (3.9%) died. Six (11.8%) patients had no ANC recovery out of which four (7.8%) died. The overall mortality was 6 (11.8%) patients. Forty-five (88.2%) patients had recovered and were discharged, out of which twenty-nine patients had ANC recovery in less than 7 days (p-value <0.001).

Patients who were discharged had mean monocyte of 181.49 cells/mm3 , whereas patients who died had mean monocyte count of 48.17 cell/mm3 (P value = 0.05). Table 2

Table 2. Predictor of Poor Outcome in Patients Admitted with Febrile Neutropenia.

Variable Discharged Death P-value

Age (years)

>50 23 6

<50 22 0 0.02

Gender

Male 13 3

Female 32 3 0.3

Type of Cancer

Solid 23 3

Hematological 22 3 0.1

ANC at Presentation

Mean 229.51 72 0.27

>300 13 0

100-300 9 1

<100 23 5 0.25

Monocyte at Presentation

Mean 181.49 48.17 0.05

Days To ANC Recovery

No Recovery 2 4

1-3 Days 12 0

4-7 Days 17 0

Beyond 7 Days 14 2 0

Accurate Antibiotic

Yes 36 4

No 9 2 0.46

All patients were assessed as high or low risk at presentation but there were no correlation between categories of risk and the type of cancer.

Most of patients with solid tumor had ANC recovery in less than 7 days while most of patients with hematological malignancies had ANC recovery in more than 7 days or had no recovery (p-value 0.04).

There was no significant difference between patients with solid or hematological malignancies in terms of number of days until afebrile status. Most of patients (56.9%) had persistent fever for less than 4 days. Table 3

Table 3. Risk Prognosis Depending On Type Of Cancer

Cancer Type P-value

Solid Hematological

MASCC

Low 15 12

High 11 13 0.49

Days Until recovery

No recovery 2 4

1-3 10 2

4-7 9 8

>7 5 11 0.04

Days until afebrile status

<2 11 9

3-4 7 3

>4 6 13

No recovery 2 0 0.1

As for documented sites of infection, most of patients 22(43%) had mouth & pharynx infection, 13 (25.5%) patients had skin & soft tissue infection 10 (19.6%) patients had respiratory tract infection, 14 (27.5%) had GI infection.

There were 19 (37.3%) patients with bloodstream infection out of which 4 patients had both peripheral and central line positive blood cultures.

Eleven patients had positive peripheral blood culture out of which seven were nosocomial infections. Twelve patients had positive central line blood culture out of which nine were nosocomial infections.

There were only 2 (3.9%) patients with documented genito-urinary infection out of which one was a nosocomial infection. Table 4

Table 4. Documented Site of Infection in Patients with Febrile Neutropenia

Site Of Infection No. Of Patient (%)

Mouth & Pharynx 22 (43%)

Skin & Soft Tissue 13 (25.5%)

Respiratory Tract 10 (19.6%)

GI 14 (27.5%)

GU 2 (3.9%)

Bloodstream 19 (37.3%)

Peripheral 11

Central 12

Discussion

In patients who were admitted with the diagnosis of neutropenic fever, our study showed that age was a predictor of poor outcome. Patients younger than 50 years of age had better prognosis than patients older than 50 years. This is because older patients had more co morbidities, which has been also associated with worse outcomes and higher mortality rate, as it has been previously studied (Kuderer et al., 2006).

The study population was heterogeneous in terms of diagnosis, categories of risk and treatment. The age distribution was equal in the study group but all patients who died were more than 50 years of age. There were more females than males in the study group though mortality rate had a male predominance but this was statistically insignificant. This was probably due to the fact that men had much more comorbid conditions.

The overall mortality was 11.8%, which is comparable with international studies (Berghmans et al., 2002) and cause of death was attributed to severe sepsis or septic shock.

Previous studies (Oguz et al, 2006) have shown that a low monocyte count after chemotherapy is linked to neutropenia while monocytosis indicates neutrophil recovery. Our study also demonstrated that a low absolute monocyte count at presentation is a predictor of a poor prognosis. Hence, patients with a high monocyte count at presentation had significantly better outcomes than those with a low monocyte count. This parameter can be used in addition to other variables as an assessment at presentation to predict the course of the patient. In addition, the administration or avoidance of GCSF to patients with neutropenic fever can be decided based upon the monocyte count.

The majority of patients (33.3 %) had ANC recovery in 4 to 7 days. We noticed that ANC recovery was better in Solid tumors than in hematological malignancies probably due to the longer time needed for the bone marrow to recover in treatment of acute leukemia and in lymphomas. Similar observations have been described in the literature in terms of recovery patterns (Maher et al., 1997; Carbonero et al., 2001) studies.

ANC recovery was associated with the length of hospital stay and hence delayed recovery contributed to longer duration of hospitalization. In our study 22 (43.1%) patients needed more than 7 days until ANC recovery and hence had longer hospital stay. These results suggest that age, the absolute monocyte count at presentation and days until ANC recovery were independent indicators of patients� outcome and length of stay. However, ANC at presentation and hence the degree of neutropenia at presentation was not a significant predictor of outcome. We noticed that relatively young patients recovered and were discharged more than the older patients; similarly older patients with higher monocytes at presentation and faster ANC recovery also recovered quickly and had better outcomes.

Patients with hematological malignancies are usually considered to be high-risk patients. In our study we compared the type of cancer with the MASCC risk index score, which is calculated for risk assessment. We observed that there was no correlation between patients with hematologic malignancies and patients with a high MASCC score. This is probably because the MASCC risk index score is based upon clinical observations and comorbidies not taking into consideration the duration of hospitalization or ANC recovery.

We also observed that the number of days until fever recovery had no correlation with the type of cancer. This is because the duration of fever is linked to the type of infection, the organism that has affected the patient and the accuracy of the treatment that is being used.

In the population that has been studied, mouth and pharynx were the most frequent sites of infection; this is due to the chemotherapy that affects the mucous membranes and makes them vulnerable to fungal infections.

A large number of patients (37.3%) had a positive blood culture and most of these blood cultures revealed germs that were nosocomial, which emphasizes on the necessity of the hygiene during our medical practice and the necessity of an early recovery because the majority of these patients had a long period of hospitalization.

To the best of our knowledge, our study is the first epidemiologic study done in Lebanon regarding neutropenic fever and its outcomes. The management of neutropenic fever is very challenging as it might be misdiagnosed or not accurately treated in patients whose immune system is attenuated. Thus, our study gave an idea of how neutropenic fever is being managed in one of the university hospital in Lebanon and how much our practice is comparable to the newest guidelines regarding this topic. Lebanon is considered one of the underdeveloped countries, however 78,4% of our patients were treated based upon recommendations and guidelines and the remaining 21.6% were treated with empiric antibiotics that have accurate coverage but had antibiotic modification or addition not as recommended, hence were considered as not having accurate antibiotics. This is why the accuracy of treatment showed no significance in terms of prognosis. Among studies that dealt with similar subjects, this study is characterized by the risk assessment that was done at the time of admission, which helped classify patients into low and high risk and guided the management.

However, since our study is prospective, a large number of patients is usually recommended, so the relatively small number of patients that were included had some limitations. In addition, we didn�t manage to collect data about the number of patients who had prophylactic antibiotics or prophylactic GCSF. We propose to collect data about prophylaxis in addition to risk assessment at hospital admission. Another limitation of our study is that this is a single center study and our results even though they were pertinent they cannot be generalized.

Conclusion

In conclusion, our study demonstrated that the overall mortality rate is comparable to international centers and that recommendations and guidelines were respected in approximately 80 % of cases.

Our study also demonstrated that age, monocyte at presentation and days until ANC recovery were important predictor of poor outcome. It revealed that delayed ANC recovery extended the duration of hospitalization and adversely affected the outcome.

This study also revealed that most of the infections in hospitalized patients were nosocomial infections and thereby resistant organisms, which needs more attention in our daily medical practice.

We propose in order to evaluate the cost effectiveness of the use of GCSF and to guide the management, to assess the monocyte count in addition to risk classification at hospital admission.