4,106–108 Patients must be able to tolerate oral
medications. In general, patients considered for oral therapy must be without
microbiologic or clinical evidence of infection (other than fever), clinically stable,
and closely observed. Patients with severe (ANC < 100 cells/μL), or lasting
anticipated ≥7 days, or those receiving prior fluoroquinolone prophylaxis would not
be eligible. This would also exclude patients with any of the following: serious
comorbidities, inpatient acquisition of infection, uncontrolled malignancy,
pneumonia, recent HSCT, dehydration, hypotension, chronic lung disease, abnormal
liver (>3× upper limits of normal) or renal function (serum creatinine >2 mg/dL), or
signs and symptoms lasting longer than 7 days.
4 An international collaborative study
established and validated a risk scoring system in adults that incorporated these
principles to identify low-risk patients for whom oral therapy may be an option.
Cefixime has been used effectively as an alternative regimen in low-risk pediatric
patients initially receiving IV therapy,
109 but there is not adequate experience to
recommend it at this time. It also lacks activity against Pseudomonas spp.
Fluoroquinolone-containing regimens have been the mainstay of most oral regimens
used in this setting. Patients already receiving fluoroquinolones as prophylaxis
should be excluded from receiving this regimen. Oral ciprofloxacin in combination
with amoxicillin–clavulanate (both administered every 8 hours) is useful in low-risk
adult patients with febrile neutropenia
111 and is the most commonly used oral
regimen. Oral levofloxacin may be used in place of ciprofloxacin, and oral
clindamycin may be used in place of amoxicillin–clavulanate if the patient is allergic
t o β-lactam antibiotics. More recently, moxifloxacin has been evaluated in this
patient population and demonstrates efficacy comparable to combination regimens
113 However, although moxifloxacin permits
simplification of the regimen, it lacks activity against Pseudomonas spp. Therefore,
its use should be restricted to patients at low risk of pseudomonal infections.
Low-risk patients with adequate home support (e.g., access to emergency
facilities, phone access) and the desire for home treatment may be treated as an
outpatient (with either IV or oral therapy). Therapy is usually initiated in a clinic or
4 Monitoring in the outpatient setting should include either home
nursing or office visits daily for 3 days to review progress and to screen for
problems. Patients who are stable and responding after the initial observation period
may continue to receive monitoring by phone contact.
Continued administration of antimicrobials in the outpatient setting initiated as an
inpatient has been suggested in a subset of low-risk patients. The criteria used to
define eligibility for outpatient therapy are generally similar to those established and
previously discussed for oral therapy. Therefore, continued outpatient administration
of parenteral antibiotics can be considered in a subset of low-risk patients with close
In general, intravenous monotherapy with an antipseudomonal β-lactam (such as an
antipseudomonal cephalosporin, carbapenem, or β-lactam and β-lactamase
combination) is advocated as initial empiric therapy for most neutropenic cancer
4 There are no convincing data to support one choice over the others as
Antipseudomonal Cephalosporins
Ceftazidime monotherapy has been found to be as effective as combination therapy.
However, in some of these trials, the efficacy of ceftazidime in patients with
documented staphylococcal and streptococcal infections was suboptimal. Select
clinical trials have empirically added antistaphylococcal coverage with a
glycopeptide (e.g., vancomycin) and have shown improved outcomes in these
patients. In addition, pathogens (particularly gram-negative pathogens) that produce
either type 1 β-lactamase or ESBL (e.g., K. pneumoniae) are not likely to respond to
ceftazidime monotherapy. These organisms are more likely with prolonged
hospitalization or receipt of antimicrobial therapy. Therefore, local in vitro
susceptibilities of common gram-negative pathogens should be examined before the
routine use of ceftazidime as monotherapy.
Cefepime is a fourth-generation cephalosporin with an FDA-approved indication
for monotherapy for empiric management of infection in patients with febrile
neutropenia. The potential advantage of this agent compared with third-generation
cephalosporins is its low affinity for major chromosomally mediated β-lactamases.
Compared with ceftazidime, cefepime has more potent activity in vitro against select
gram-positive bacteria (methicillin-susceptible Staphylococcus species, viridans
streptococci, and S. pneumoniae). Similar to ceftazidime, numerous randomized,
comparative studies have evaluated the role of cefepime as monotherapy in both
adults and children with febrile neutropenia. The improved gram-positive activity
(relative to ceftazidime) may decrease the empiric need for vancomycin in some
patients. However, this advantage is less likely in institutions with a high rate of
MRSA because cefepime and other cephalosporins (with the exception of
ceftaroline) are inactive against this pathogen. Although a meta-analysis published in
2007 suggested increased all-cause mortality in neutropenic patients treated
115 subsequent analysis by the FDA concluded no such
differences existed when compared with control patients.
The carbapenems are a unique class of antibiotics with broad-spectrum activity
against numerous gram-positive and gram-negative bacteria, including anaerobes. In
addition, carbapenems may be used in the setting in which the patient is at increased
risk of ESBL-producing gram-negative pathogens (such as K. pneumoniae or E. coli).
Imipenem (in combination with the dehydropeptidase inhibitor cilastatin) and
meropenem are two currently available agents in this class that have been studied as
monotherapy for febrile neutropenic patients.
Clinical outcomes with imipenem–cilastatin monotherapy have been comparable
with those of the β-lactam plus aminoglycoside combinations. Although effective,
imipenem–cilastatin has generally been associated with a higher incidence of nausea
and vomiting compared with ceftazidime or meropenem.
generally dose-related (3–4 g/day) and associated with the rate of IV administration.
Therefore, dosages of 2 g/day (divided every 6 hours) are generally given to patients
Meropenem monotherapy for febrile neutropenia has been evaluated in both adults
and children, and the results have supported the value of meropenem as empiric
monotherapy for use in febrile neutropenic patients. Meropenem may have
advantages over imipenem–cilastatin in the treatment of central nervous system
infections (owing to less associated seizure activity).
Although doripenem possesses comparable microbiologic activity, studies
evaluating its administration in this population are lacking. Although ertapenem
possesses microbiologic activity comparable to that of the other carbapenems, it
lacks in vitro activity against Acinetobacter species and Pseudomonas spp., including
P. aeruginosa. Considering this lack of activity, ertapenem would not be appropriate
as empiric therapy for febrile neutropenic patients. Agents with increased anaerobic
activity (including carbapenems) might be used as empiric therapy in cases in which
an intra-abdominal infection is suspected.
Routine carbapenem use may be associated with increased drug acquisition cost
(relative to cephalosporins) and increased potential for development of carbapenem
resistance. Therefore, many institutions have elected to reserve the carbapenems for
patients who have failed to respond to prior empiric therapy, have a history of
infections with pathogens resistant to third- and fourth-generation cephalosporins, are
clinically unstable, or have need for expanded anaerobic coverage.
β-Lactam/β-Lactamase Inhibitors
Randomized trials have compared piperacillin–tazobactam monotherapy to various
antibiotics. Published trials have documented that piperacillin–tazobactam is not
inferior to cefepime in this patient population.
119 Although experience with this
agent as monotherapy is limited relative to that for the antipseudomonal carbapenems
and antipseudomonal cephalosporins, it is considered to be an adequate choice for
4 Higher doses of piperacillin–tazobactam (i.e., 3.375 g IV every 4
hours or 4.5 g IV every 6 hours in adult patients with normal renal function) should
be used because of the risk of pseudomonal infections.
interfere with the galactomannan assay used in the diagnosis of select IFIs, including
INITIAL ANTIMICROBIAL COMBINATIONS (EXCLUDING
QUESTION 1: B.L., a 13-year-old boy, presented with a 3-week history of “always being tired” and a
experienced a fever (102°F) and chills. The ANC was 48 cells/μL. SCr and BUN were 1.0 and 15 mg/dL,
What is the role of this combination in the empiric management of febrile neutropenia? Are there any
differences in efficacy between these combinations?
Before the introduction of third- and fourth-generation cephalosporins and
carbapenems, the empiric use of antibacterial combinations in febrile neutropenic
cancer patients was favored because these regimens offered a broader spectrum of
activity against bacteria commonly infecting cancer patients.
combinations offered the potential for an additive or synergistic effect.
However, infections in neutropenic patients have shifted largely from gram-negative
to gram-positive pathogens, for which these traditional combination regimens have
limited efficacy. Despite this concern, some clinicians continue to favor antibiotic
combinations as initial empiric therapy, especially in patients who are clinically
It is unknown whether combination therapy prevents the emergence of
Until the 1980s, most febrile neutropenic patients were treated with two-drug
combination regimens that contained an aminoglycoside (gentamicin, tobramycin, or
amikacin) plus a β-lactam antibiotic, such as antipseudomonal penicillin, or an
antipseudomonal third-generation cephalosporin. This combination is one of the most
established empiric treatment regimens for the management of febrile neutropenia.
Numerous studies have been conducted to evaluate the efficacy of combination
therapy of an aminoglycoside with an antipseudomonal cephalosporin (ceftazidime
3 Antipseudomonal penicillins plus β-lactamase inhibitors (in
combination with an aminoglycoside) would also be considered a comparable
regimen. A better outcome was demonstrated with full-course amikacin plus
ceftazidime than with a short course (3 days) of amikacin plus a full course of
122 However, the longer course of the aminoglycoside is likely to be
associated with more toxicity.
Because carbapenem antibiotics are more frequently evaluated as monotherapy in
this patient population, there are limited studies examining the combination of an
aminoglycoside with either imipenem–cilastatin or meropenem. However, in one
such evaluation, the combination of imipenem–cilastatin plus amikacin was found to
be superior to imipenem–cilastatin monotherapy.
123 Carbapenems (in combination
with aminoglycosides and vancomycin) have been recommended as empiric initial
therapy for patients who are clinically unstable.
Aminoglycosides have generally been considered the backbone of combination
regimens because of their potential for bactericidal action against various bacteria.
However, the addition of an aminoglycoside is associated with increased costs for
therapeutic monitoring. The benefit of an aminoglycoside for empiric therapy has not
been consistently demonstrated.
In addition, there is an increased potential for
the development of nephrotoxicity and ototoxicity.
124 This concern is especially
relevant in patients receiving concomitant nephrotoxins (such as cisplatin and
CASE 75-2, QUESTION 2: Seven days into therapy, despite rehydration, B.L.’s SCr and BUN rose to 2.0
g/dL and 45 mg/dL, respectively. Because B.L. has nephrotoxicity (believed to be secondary to the
Are these regimens as effective as aminoglycoside-containing regimens?
As previously stated, ciprofloxacin has been studied in combination with other
antibacterials (either an aminoglycoside or a β-lactam, including an antipseudomonal
penicillin) as initial empiric therapy for treatment of suspected infection in febrile
125–128 Ciprofloxacin in combination with clindamycin may also
be useful in patients unable to receive β-lactam-containing combinations owing to the
presence of immediate-type hypersensitivity reactions.
were associated with increased gram-positive infections in ciprofloxacin-treated
patients. In addition, the in vitro activity of ciprofloxacin against P. aeruginosa has
declined significantly (to <70% in many institutions). Therefore, if used as part of
combination therapy, ciprofloxacin should be combined with an antimicrobial with
favorable in vitro activity against this pathogen. Alternatively, the combination of
aztreonam and vancomycin has been recognized as reasonable empiric combination
therapy in patients with immediate-type hypersensitivity reactions to penicillin.
Use of initial empiric therapy with combination antibiotic therapy is generally
reserved for patients who are clinically unstable (hypotension, tachycardia,
tachypnea, mental status changes, etc.). In such settings, intravenous therapy with an
antipseudomonal β-lactam is generally combined with an aminoglycoside and
vancomycin. Systemic antifungals (such as fluconazole or an echinocandin) may also
be added in such settings, especially in patients not receiving antifungal prophylaxis.
CASE 75-2, QUESTION 3: B.L. is begun on a two-drug regimen of ceftazidime and vancomycin. What is
the rationale for adding vancomycin to the regimen?
As previously discussed, gram-positive bacteria are important pathogens in these
patients. Because cephalosporins lack activity against methicillin-resistant
staphylococci, vancomycin is often added to empiric regimens. A growing proportion
of S. aureus infections are methicillin resistant (as many as 60% in some institutions).
However, widespread use of vancomycin may be one factor associated with the rise
have been reported. Vancomycin use has also been associated with a modest increase
in the incidence of nephrotoxicity.
Primarily because of differences in the measured endpoints, the need for
vancomycin as initial empiric therapy continues to be debated.
febrile neutropenic patients with cancer had more rapid resolution of fever, fewer
days of bacteremia, and a lower frequency of treatment failure when vancomycin was
added to an initial regimen of antipseudomonal penicillin plus an
131 Similarly, the addition of vancomycin to ceftazidime showed
improved results compared with ceftazidime alone or with a three-drug
132 However, other studies have concluded that mortality was not
increased when vancomycin therapy was delayed.
staphylococcal (generally coagulase-negative staphylococci) infections is generally
considered to be low (<4%) during the first 48 hours after the onset of fever. In
contrast, the mortality associated with viridans streptococcal infections is higher
among patients who are not initially treated with vancomycin.
viridans streptococci are either resistant or tolerant to penicillin.
vancomycin would represent a reasonable alternative treatment of such infections, or
in patients allergic to penicillins.
There continues to be considerable debate about whether vancomycin should be
included in the initial empiric regimen for febrile neutropenic patients.
routine use of empiric vancomycin should be discouraged.
institutions frequently isolating invasive gram-positive bacterial pathogens (e.g.,
those caused by viridans streptococci), vancomycin should be considered in the
initial empiric regimen. In addition, patients at highest risk of serious, invasive
infections with gram-positive organisms should be considered for initial vancomycin
fluoroquinolone or TMP–SMX prophylaxis, those with prior history of colonization
with β-lactam-resistant pneumococci or MRSA or with gram-positive bacteria in
blood cultures (before identification and susceptibility testing), and those with sepsis
without an identified pathogen should also be considered for vancomycin therapy.
The addition of vancomycin to an aminoglycoside-containing regimen should be done
with caution because data support an increased risk of aminoglycoside-induced
nephrotoxicity in patients receiving these agents concomitantly with vancomycin.
More recently, vancomycin has been shown to increase the risk of nephrotoxicity in
patients receiving piperacillin–tazobactam.
Alternatives to Vancomycin with Activity Against Gram-positive Pathogens
Options exist for the treatment of invasive gram-positive infections. Linezolid is an
oxazolidinone that can be administered IV or orally. A randomized, double-blind
trial comparing the use of linezolid with vancomycin for empiric therapy
demonstrated comparable safety and efficacy.
thrombocytopenia and secondary neutropenia, especially when given for prolonged
periods (i.e., >14 days according to the product’s package insert). Considering the
reduced bone marrow reserve in cancer chemotherapy patients, these adverse events
are of particular concern. In addition, it is currently not recommended for patients
role in this patient population would be as a treatment for resistant or refractory
gram-positive infections (such as MRSA or VRE). Tedizolid, an oxazolidinone for
IV and oral therapy, has more recently been approved in the US. Although advantages
over linezolid include a potential reduction in hematologic adverse events and drug
interactions, data are lacking to support tedizolid use in this population. In addition,
tedizolid is not currently recommended in patients with neutropenia.
Quinupristin–dalfopristin is available for IV administration only, and concerns
about potential drug interactions and patient tolerability (including myalgias and
arthralgias) limit its use as empiric therapy in this setting. Daptomycin provides
potent in vitro activity against many multidrug-resistant gram-positive pathogens
(including VRE and MRSA). Compared to historical controls treated with
vancomycin, daptomycin use in cancer patients with gram-positive catheter-related
bacteremia was associated with earlier and overall improved response.
Daptomycin requires IV therapy and should not be used for the treatment of
pneumonia. Tigecycline possesses activity in vitro against both MRSA and VRE (in
addition to many gram-negative and anaerobic pathogens). However, it lacks activity
against P. aeruginosa and therefore would not be considered a viable option for
empiric monotherapy. Telavancin is a new lipoglycopeptide with potent activity in
vitro against select gram-positive pathogens (including MRSA). However,
experience to date in this population has not been published. Ceftaroline is a
cephalosporin with activity against MRSA recently approved for the treatment of
complicated skin and skin structure infections and community-acquired pneumonia.
However, reports of the use of ceftaroline in this patient population are sparse. In
addition, ceftaroline has infrequently been associated with neutropenia, most notably
at higher doses and/or for prolonged courses of therapy.
for the use of oritavancin and dalbavancin (also recently approved for IV therapy of
MRSA infections) in these patients. Therefore, these alternative agents (daptomycin
and linezolid) are generally reserved for situations in which vancomycin is
inappropriate (because of resistance or intolerance).
In the case of B.L., empiric use of vancomycin is not warranted based on the
previous discussion, and it should be discontinued unless cultures indicate the need
ANTIBIOTIC DOSING, ADMINISTRATION, AND
Intermittent versus Continuous versus Prolonged
Infusion of Intravenous Antibiotics
β-Lactam antibiotics exhibit time-dependent (i.e., concentration-independent)
pharmacodynamic activity, and in vitro models of infection suggest that prolonged
exposure of bacteria to drug concentrations above the minimum inhibitory
concentration is linked with improved bacterial killing and survival. On the basis of
these observations and the poor prognosis of neutropenic cancer patients with
bacteremia, noncomparative, open-label trials were conducted to evaluate the role of
continuous infusions of β-lactams (i.e., ceftazidime) in the empiric treatment of
suspected infection in cancer patients.
149–151 Additional studies in patients with
febrile neutropenia have also been performed with meropenem suggesting a potential
benefit of prolonged infusion over traditional administration.
method of administration would require an IV line dedicated for continuous drug
administration and limit B.L.’s ability to receive intermittent tobramycin infusions
unless additional IV ports or lines are available for use. In contrast to continuous
infusions, prolonged infusions (i.e., 3–4 hours) of select β-lactams (notably
carbapenems, third- or fourth-generation cephalosporins, or piperacillin–tazobactam)
have shown promise in a variety of pharmacodynamic models of infection
(specifically with elevated minimum inhibitory concentrations) and tend to show
better outcomes in observational studies. However, this has not been verified in
prospective randomized, controlled trials. Prolonged infusions of β-lactams also
have the potential to reduce the total dose and therefore result in drug acquisition cost
savings. Although only limited clinical data exist to support such a strategy, such
evaluations have not been tested specifically in the setting of infections in the
Consolidated Interval (Extended-interval or “Oncedaily”) Aminoglycoside Dosing
neutropenic patients such as B.L.?
Because of the concentration-dependent pharmacodynamic properties of
aminoglycosides and the convenience of administration, studies have been conducted
to describe both the pharmacokinetic properties and efficacy of consolidated dosing
of aminoglycosides in animals and in neutropenic patients. Pharmacokinetic studies
156 have not revealed pharmacokinetic
differences when compared with other populations. Several clinical studies have
included consolidated aminoglycoside dosing for amikacin, gentamicin, and
157 However, most of the studies in this population were not designed to
evaluate differences between consolidated aminoglycoside dosing compared with
similar regimens using intermittent dosing. In general, the various studies suggest that
consolidated dosing is as effective and possibly less nephrotoxic than traditional
dosing. Therefore, consolidated dosing of aminoglycosides appears reasonable in
empiric therapy of neutropenic patients.
HOST FACTORS INFLUENCING RESPONSE TO
CASE 75-2, QUESTION 6: What factors may have influenced B.L.’s clinical response to antimicrobial
The most important prognostic determinants of a favorable outcome in patients
with neutropenia and infection are the recovery of the granulocyte count and (for the
patient with infection) proper selection of antimicrobial therapy. Patients with
profound, persistent neutropenia (<100 cells/μL that does not rise during therapy or
an initial ANC of 100–500 cells/μL that declines during therapy) respond to
antibiotics less favorably than patients whose bone marrow recovers. The initial
granulocyte count appears to be less important than the trend toward granulocyte
recovery. Although other evidence of bone marrow recovery (such as the absolute
phagocyte count, absolute monocyte count, or reticulocyte fraction) may precede the
ANC target of 500 cells/μL by several days, they are not as widely used clinically.
The site of infection also influences outcome. Septic shock and pneumonia are
associated with high mortality in bacteremic neutropenic patients.
MODIFYING INITIAL EMPIRIC ANTIBIOTIC
The need for modification of the initial empiric therapy is dependent on the risk
group (i.e., low vs. high risk), establishment of an infection site or causative
pathogen, response to initial therapy, and clinical stability. In the absence of
worsening of clinical status or onset of new signs and symptoms of infection, 3 to 5
days of empiric treatment is generally required to determine initial efficacy after
initiation of empiric antibiotics.
4 Defervescence in patients with hematologic
malignancies, as well as HSCT recipients, may be delayed (up to 5 days). During
this time, daily assessments should include history and physical examinations, review
of laboratory results, assessment of response, and evaluation of any antibiotic-related
or the patient’s condition deteriorates.
Premature discontinuation of antibiotics may predispose these patients to
recrudescence of bacterial infection and increase the risk of infection-related
morbidity and mortality. Cancer patients with unexplained fever who became
afebrile after empiric antibiotics
were randomly assigned to continue or discontinue antibiotic therapy after 7
158 The patients whose neutropenia resolved had no infectious sequelae
regardless of whether antibiotics were continued or discontinued. However, for
persistently neutropenic patients randomly assigned to continue or discontinue
antibiotic therapy until their ANC was greater than 500 cells/μL, the percentages of
patients remaining febrile without infections complications were 94% and 41%,
respectively. Therefore, initial empiric therapy in responding patients without an
identified source should be continued until the patient’s ANC is >500 cells/μL and
increasing, and the patient is well and afebrile for at least 24 hours. Patients who are
afebrile and stable but whose neutropenia persists may be considered for
continuation with oral antibiotics (such as ciprofloxacin plus
amoxicillin/clavulanate). Patients whose initial empiric regimen appropriately
included vancomycin should continue receiving such therapy. In contrast, those
initiated inappropriately on vancomycin, or in whom cultures, diagnosis or condition
does not support continued use should have vancomycin therapy discontinued.
CASE 75-3, QUESTION 2: On day 3, M.H.’s temperature is normal (97.6°C). However, two sets of blood
As previously stated, modification of initial empiric therapy should be made based
on culture results as well as site-specific signs and symptoms of infection (Table 75-
1). For example, anaerobic coverage (often with a carbapenem or piperacillin–
tazobactam if not used for initial monotherapy) may be expanded in the setting of
abdominal pain. Metronidazole or oral vancomycin should be initiated in the setting
of diarrhea when C. difficile is suspected.
4 Vesicular lesions may be suggestive of
viral infections (such as HSV or VZV) and may respond to the addition of
14 Suspected IV catheter infections should be managed with catheter
removal (whenever possible), and this as well as skin and skin structure infections
should be managed with treatment directed at MRSA.
coverage should be consistent with published guidelines for the treatment of health
care-associated infections (see Chapter 67, Respiratory Tract Infections).
Comments
Post a Comment
اكتب تعليق حول الموضوع