meta analisis 1

1. Pediatr Blood Cancer 2014;61:1427–1433
Outpatient Versus Inpatient IV Antibiotic Management for Pediatric Oncology
Patients With Low Risk Febrile Neutropenia: A Randomised Trial
Lisa M. Orme,1
* Franz E. Babl,2,3,4
Chris Barnes,3,5
Peter Barnett,2,3,4
Susan Donath,3,4,6
and David M. Ashley7
INTRODUCTION
Febrile neutropenia (FN) is a frequent and potentially life
threatening complication of the intensive myelosuppressive
chemotherapy regimens commonly used to treat pediatric oncology
patients. Febrile neutropenic patients are at risk of bacterial sepsis
and prompt empiric administration of broad spectrum intravenous
antibiotics has reduced morbidity and mortality [1]. For many
patients and families, the occurrence of repeated FN episodes
requiring additional unpredictable admissions to hospital are
detrimental to patient/child and family quality of life (QOL)
during the cancer journey.
Acknowledging the heterogeneity of the FN population, the
development of risk stratification algorithms attempting to identify
patient subsets at low risk of bacterial sepsis has led to successful
reduced intensity, often outpatient management of FN, particularly
in the adult setting [2,3]. Whilst there is no consensus on risk
prediction in the pediatric setting [4], outpatient FN management
for a low-risk subset is increasingly utilised. In a number of
institutions studies have been conducted or practices have been
described using different criteria and a variable combination of oral
and parenteral and inpatient and outpatient elements. In general
these reports with a meta-analysis support the safety and efficacy of
this practice but lack of external validation limits the interpretation
or generalisation of the findings [5–11]. Advantages of outpatient
management include reduced cost burden and reduction in
nosocomial infections. It has long been speculated that outpatient
management would increase patient and family QOL [12,13]
however parents often express anxiety about supporting their ill
children with FN outside the acute hospital setting. Cefepime, a
fourth generation cephalosporin has been reported as a safe and
feasible choice for the treatment of childhood cancer patients with
FN [14–18]. Due to its broad coverage and long half life allowing
twice daily scheduling, it is a suitable candidate for outpatient FN
management.
We conducted a randomised controlled trial to determine the
effect on QOL of outpatient compared to inpatient management
using the same parenteral antibiotic regimen for paediatric
oncology patients with LRFN. Secondarily we wanted to assess
the adverse events profile of cefipime in this setting.
METHODS
Design and Setting
We conducted an unblinded randomised controlled trial
comparing inpatient and outpatient antibiotic management of fever
and neutropenia in children receiving chemotherapy for malignan-
cy. The study was conducted at Royal Children’s Hospital (RCH),
Melbourne, Australia, the single, urban, tertiary, children’s hospital
serving a pediatric population of 1.5 million. The study was
approved by the RCH ethics committee. The study was initiated
before registration with a Clinical Trials Registry was mandatory.
Background. Febrile neutropenia (FN) is a frequent, serious
complication of intensive pediatric chemotherapy regimens. The aim
of this trial was to compare quality of life (QOL) between inpatient
and outpatient intravenous antibiotic management of children and
adolescents with low risk febrile neutropenia (LRFN). Procedure. In
this randomised non-blinded trial, patients between 1 and 21 years
old, receiving low/moderate intensity chemotherapy were pre-
consented and, on presentation to emergency (ED) with FN satisfying
low risk criteria, randomised to either outpatient or inpatient care
with intravenous cefepime 50 mg/kg (12 hourly). All patients
continued antibiotics for at least 48 hours, until afebrile for 24 hours
and demonstrating a rising absolute neutrophil count 200/mm3
.
Several domains of QOL were examined by daily questionnaire.
Results. Eighty-one patients presented to ED with 159 episodes of
fever. Thirty-seven FN presentations involving 27 patients were
randomised to inpatient (18) and outpatient (19) management.
Combined QOL mean scores for parents were higher for the
outpatient group and scores for three specific parent variables
(keeping up with household tasks/time spent with partner/time spent
with other children) were higher among outpatients. There was no
difference in parent confidence/satisfaction in care between groups.
Patients scored better in the outpatient group overall and for sleep
and appetite. The mean length of fever was equivalent between
groups and there were no serious adverse events attributable to
cefepime or outpatient care. Conclusion. Outpatient cefepime
management of LRFN provided significant benefit to parents and
patients across several QOL domains and appeared both feasible and
safe. Pediatr Blood Cancer 2014;61:1427–1433.
# 2014 Wiley Periodicals, Inc.
Key words: antibiotic; fever; inpatient; neutropenia; outpatient; paediatric oncology
1
Children’s Cancer Centre, The Royal Children’s Hospital, Parkville,
Victoria, Australia; 2
Emergency Department, The Royal Children’s
Hospital, Parkville, Victoria, Australia; 3
Murdoch Children’s Research
Centre, The Royal Children’s Hospital, Parkville, Victoria, Australia;
4
University of Melbourne, Parkville, Victoria, Australia; 5
Department
of Hematology, The Royal Children’s Hospital, Parkville, Victoria,
Australia; 6
Clinical Epidemiology and Biostatistics Unit, The Royal
Children’s Hospital, Parkville, Australia; 7
Department of Medicine,
Deakin University, Barwon Health, Geelong, Victoria, Australia
Grant sponsor: Bristol Myers Squibb Noble Park North, Vic, Australia;
Grant sponsor: Murdoch Children’s Research Institute, Melbourne,
Australia; Grant sponsor: Victorian Government’s Operational Infra-
structure Support Program
Conflict of interest: Nothing to declare.
Correspondence to: Lisa M. Orme, Children’s Cancer Centre, The
Royal Children’s Hospital, Flemington Road, Parkville 3052, Victoria,
Australia. E-mail: lisa.orme@rch.org.au
Received 5 June 2013; Accepted 5 February 2014
C 2014 Wiley Periodicals, Inc.
DOI 10.1002/pbc.25012
Published online 6 March 2014 in Wiley Online Library
(wileyonlinelibrary.com).

2. Participants
Participants were patients with a malignancy receiving
chemotherapy managed by the paediatric oncology service at
RCH between July 2003 and July 2006. The unit sees
approximately 170 new malignant diagnoses per annum. All new
patients planned for chemotherapy during the study period and
those receiving chemotherapy at the time of study commencement
were considered for participation. Participants were identified in
oncology clinic during routine visits after initiation of chemothera-
py. Patients were considered eligible for the study if they were 1)
older than 12 months and younger than 21 years (to allow inclusion
of adolescent and young adult patients being treated on or as per a
paediatric clinical trial in line with institutional protocol) and 2)
resided within 1 hour travel time (or 30 km radius) of RCH with
appropriate transport and telephone and 3) a parent or legal
guardian able to give written informed consent prior to study entry.
Exclusion Criteria
Reasons for exclusion were: (1) acute myeloid leukaemia
(AML) or mature B-cell lymphoma, (2) patients in the induction
phase of acute lymphoblastic leukaemia (ALL) or receiving
predominantly high dose stem cell-supported chemotherapy, (3)
history of allergy to penicillins or cephalosporins or (4) families
who were non English speaking.
Consent
Written informed pre-consent was obtained from parents or
guardians in the oncology clinic prior to acute presentations with
fever and neutropenia and enrollment/randomisation. Parent
information statements were given to the families, which included
references to standard care for fever neutropenia episodes, risk
stratification, randomisation and home treatment with Home and
Community Care support.
Procedure
The procedure is depicted in Figure 1. Study patients presenting
to the RCH emergency department (ED) with fever and probable
neutropenia were promptly clinically assessed and risk stratified.
Fever was defined as one measurement of at least 38.5˚C or two
measurements of at least 38.0˚C in the preceding 4-hour period
(oral, axillary, or tympanic measurements). Probable neutropenia
was assumed if the patient had received a cycle of chemotherapy in
the previous 14 days, or at any time in ALL maintenance therapy.
Full blood examination (FBE), blood cultures (from central venous
line if present) and blood group and hold were immediately
performed as well as other investigations as clinically indicated. As
per institutional practice, to minimise procedural trauma, peripheral
blood cultures were not routinely performed for patients with
central venous access. Study patients were considered low risk if
they had (1) no signs of septic shock (including hypotension,
tachycardia, delayed capillary refill or rigors) nor (2), significant co
morbidities requiring inpatient monitoring or treatment including
clinical focus of infection, pain, mucositis, vomiting, diarrhoea, or
dehydration. Low risk study patients were given an initial dose of
intravenous cefipime 50 mg/kg while waiting for the absolute
neutrophil count to be confirmed as 0.5 109
/L. Randomisation
proceeded only after revision of study information by parent or legal
guardian with verbal re-consent to participate and approval from
senior ED medical staff after discussion with on-call oncology
consultant.
Participants were randomised to outpatient versus inpatient
treatment. Randomisation was prepared by the Clinical Epidemiol-
ogy and Biostatistics Unit at RCH, and was stratified by disease
(ALL or other diagnoses) and age (1–5 years or at least 6 years of
age). Within the strata, the randomisation was blocked using
randomly permuted block sizes.
All patients continued intravenous cefepime 50 mg/kg 12 hourly
unless positive blood cultures or medical complications dictated a
regimen change. Antipyretic agents were discouraged to avoid
masking fever. Full blood examination was performed daily. Blood
cultures in those with central venous catheters were performed daily
for the first 3 days and if febrile (38˚C) in the preceding 24 hours;
thereafter, and in patients with peripheral intravenous catheters
after 72 hours of treatment, patients were re-cultured if they were
febrile 38˚C in the preceding 24 hours. Additional investigations
were performed according to clinical need at the discretion of the
treating team.
Outpatients were observed in the ED for a minimum of 4 hours
in case of medical deterioration and revised risk status. After
discharge, Home and Community Care (HACC) nurses visited
twice daily to provide general review with vital sign monitoring,
cefipime administration and blood tests. HACC provided hospital
based community outreach services by general paediatric nurses
and is based at RCH. Parents of outpatients recorded tympanic
temperature 4–6 hourly. Blood product transfusions (in day
procedure centre) and outpatient medical reviews were arranged
as deemed appropriate. Outpatients were readmitted if they
developed (1) a positive blood culture and were still febrile 38˚
C in preceding 8 hours or grew an isolate requiring alternative
antibiotic not suitable for outpatient administration, (2) persistent
fever 38˚C for persisting longer than 72 hours after the first
antibiotic dose or recurrent fever prior to evidence of neutrophil
recovery, (3) a significant medical complication (4) a central venous
catheter complication or if there were (5) parental concerns, or (6)
physician concerns.
Patients randomised to the inpatient arm received routine
inpatient monitoring and care. In the absence of proven infection
Patient
Identification
Pre-consent
Fever Presentation
Clinical Assessment
Blood Cultures, FBE
Verbal Re-Consent
Risk Stratification
Enrollment
Randomisation
Inpatient
Daily QOL
Primary
Outcome
Secondary
Outcome
Outpatient
Days of Fever
Days of Treatment
Medical Course
Adverse Events
Recovery
Readmission
-
Baseline QOL
Daily QoL Daily QoL
Fig. 1. Study patient flow.
Pediatr Blood Cancer DOI 10.1002/pbc
1428 Orme et al.

3. requiring a specific treatment plan, antibiotic therapy was continued
in all patients for a minimum of 48 hours, until resolution of fever
for 24 hours and evidence of neutrophil recovery defined as a rising
absolute neutrophil count to at least 200/mm3
. Families were
telephoned approximately 7 days after treatment completion to
assess for delayed adverse events.
Definitions
Treatment intensity. ALL maintenance chemotherapy was
considered low intensity. High intensity chemotherapy courses such
as those requiring autologous stem cell reinfusion were an
exclusion to enrolment. All other chemotherapy courses were
considered moderate intensity.
Quality of life. Quality of life (QOL) was measured during the
enrolment episode by administering a QOL questionnaire and
recording responses from parents and patients (10 years of age).
This was administered at baseline, based on experience in the
previous 7 days and daily for each day of cefipime treatment, until
fever and neutropenia resolved or readmission of outpatients
occurred (Fig. 1). As there were no suitable validated tools for this
specific short term QOL assessment, a modified version of that used
by Close et al. [19] was used. This visual analogue scale assessed
seven patient variables and six parent variables. Using a linear
10 cm scale, parents (if patients 10 years) and patients 10 years
rated general well-being, mood, independence, appetite, concen-
tration, sleep and activity level of the patient. Parents also rated their
ability to keep up with home tasks, time spent with spouse and
siblings as well as level of anxiety, confidence in patient care and
satisfaction with care. Questions asked are shown in Figure 3.
Adverse events. All adverse events including drug toxicity
were assessed prospectively in both arms of the study according to
the NCI CTC Version 2.0 [20].
Data Management
Sample size. Indicative sample size calculations were framed
on the basis of results from a closely related randomised study of
chemotherapy in children with cancer [17]. This study showed that
home treatment led to increases of 25–40%, equivalent to about one
standard deviation, in mean score on six QOL scales. If the true
mean difference in a continuous QOL outcome score between the
two groups was 1 standard deviation, a study with seventeen
children in each group would have an 80% chance of producing a
statistically significant result (with a standard two group two-sided
t-test at P ¼ 0.05).
Data handling and analysis. All data were entered into an
Access software database (version 2003, Microsoft, Redmond,
WA). Statistical calculations were performed on Stata software
(version 9.0 Stata Corporation, College Station, TX). Descriptive
statistics were calculated for fever duration, treatment duration,
changes in antibiotics, readmission rate, number of microbiologi-
cally proven infections and number of infections with a clinical
focus. Median values are reported as median with interquartile
range (IQR). We used chi square tests for dichotomous variables
and t-tests for parametric variables. QOL scores for each group
were itemised and expressed as means and 95% confidence
intervals. QOL questionnaires were analysed (1) according to day
of FN treatment episode, as the mean score of all questions for that
day, and (2) according to each question, as the mean score for that
question on all days.
RESULTS
Participants were recruited to the study between July 2003 and
July 2006. A total of 81 pre- consented patients presented to the ED
with 159 episodes of fever. On six occasions the family did not
verbally re-consent to study participation. In 116 episodes, the
patient was ineligible at presentation for the following primary
reasons: (1) neutrophils 0.5 109
/L (32/116 patients (28%)); (2)
No documented fever according to the study criteria (24/116
(20%)); (3) medical co-morbidities including signs of sepsis
precluding low risk status (53/116 (46%)); 3. oncology consultant
decision in otherwise eligible patient (7/116 (6%)). A total of 37
presentations with FN involving 27 patients were randomised.
There were 18 inpatients and 19 outpatients (Fig. 2). There were no
patients over 18 years of age. Multiple FN episodes in the same
Readmissions 6
Refusal 6
Ineligible 116
Eligible Presentaons 43
Randomised 37 Presentaons
(Involving 27 Paents)
Fever Presentaons 159
Completed 19
Paents Pre-consented 81
Outpaent Presentaons 19 Inpaent Presentaons 18
Completed 18
Fig. 2. Participant flow
Pediatr Blood Cancer DOI 10.1002/pbc
Outpatient Management of Febrile Neutropenia 1429

4. A
Keep up with household tasks
Spend time with partner
Spend time with other children
Level of concern about child's condition
Confidence in ability to care for child
Satisfaction with overall care
as usual
as usual
as usual
very calm
very confident
very happy
not at all
not at all
not at all
very anxious
no confidence
very unhappy
0 1 2 3 4 5 6 7 8 9 10
Outpatients Inpatients
Error bars indicate standard deviations
QoL Scores on Parent Questionnaire
B
Q1
Q2
Q3
Q4
Q5
Q6
Well-being
Independence
Appetite
Interest or concentration
Mood
Sleep at night
Activity or energy level
very good
as usual
very good
as usual
very happy
as usual
most of the time
very poor
more clingy
very poor
not as usual
very upset
much worse
not at all
0 1 2 3 4 5 6 7 8 9 10
Outpatients Inpatients
Error bars indicate standard deviations
QoL Scores on Patient Questionnaire
Q1
Q2
Q3
Q4
Q5
Q6
Q7
Fig. 3. Quality of life bar graphs showing mean and standard deviation for each parent and patient variable for inpatients and outpatients.
Pediatr Blood Cancer DOI 10.1002/pbc
1430 Orme et al.

5. patient were allowed: one patient had five, one patient had four and
four patients had two FN episodes.
Overall 59% of randomised presentations were patients with
ALL and 78% patients were receiving moderate intensity
chemotherapy. Demographics were similar between groups and
are shown in Table I.
Mean length of treatment episodes and fever were not
significantly different between groups (Table II). Three patients
(2 inpatient, 1 outpatient) had microbiologically proven infection
and two in each group presented with or developed a clinical focus
of infection with details shown in Table III. The antibiotic regime
was changed in three inpatients, including change to acyclovir in
one patient. Antibiotic management was altered for the six
outpatients who were readmitted during the course of FN, including
change to acyclovir in one patient and addition of or change to
amphotericin in two (Tables II and III). There were no adverse
events due to outpatient management per se. There were no adverse
events attributable to cefipime.
Figure 3 shows the results of the responses to the parent and
patient questionnaires to assess QOL. Baseline questionnaires were
similar between the two groups. Parent questionnaires show a
higher level of QOL for outpatient care on Days 2, 3 and 4 overall,
and in several specific domains: parents were significantly more
able to keep up with household tasks, spend time with their partner
and spend time with their children. Parents showed a higher level of
concern about the child’s condition and a lower confidence in their
ability to care for the child in the outpatient group though these
differences were not statistically significant. Satisfaction levels for
TABLE I. Demographics by Randomisation Group
Total Inpatient Outpatient
Patients 27
Episodes 37 18 19
Age
Mean SD 6.38 3.87 7.19 4.66 5.62 2.87
(Range) (1.95–16.92) (1.95–16.92) (2.55–13.37)
Median (IQR) 5.38 (4.77) 5.42 (5.78) 5.10 (3.59)
Gender
Female 20 8 12
Male 17 10 7
Diagnosis
ALL 22 9 13
Solid tumour 15 9 6
Treatment intensity
Low 8 3 5
Moderate 29 15 14
Central venous access 13 14
Mean days since chemotherapy (n ¼ 29a
) 10.2 6.10 7.21 2.78
Mean hemoglobin g/L SD 94 22.2 88 18.6
Mean neutrophils 109
/L SD 0.08 0.14 0.09 0.14
Mean platelets 1012
/L SD 104 146 97 123
Platelet transfusion 7/18 3/19
PRBC transfusion 6/18 6/19
SD, standard deviation; IQR, interquartile range; PRBC, packed red blood cells. a
Excluding 8 of 37 patients with ALL on continuous maintenance
chemotherapy.
TABLE II. Secondary Outcomes
Inpatient Outpatient P value
Mean Tx episode days (range) SD 4.8 (2–11) 2.6 6.3 (3–13) 3.0 0.13 (t-test)
Median Tx episode days (IQR) 4.0 (3.0) 5.0 (5.0)
Mean fever days (range) 2.4 (1–6) 1.5 3.5 (1–12) 2.7 0.12 (t-test)
Median fever days (IQR) 2.0 (2.0) 3.0 (2.0)
Microbiologically proven infection 2 1
Clinical focus of infection 2 2
Other medical complication 2 2
Change to antibiotic regimen 3 6
Blood products 9/18 50% 9/19 43% 0.99 (chi sq)
Readmission N/A 6.0
Tx, treatment; SD, standard deviation; IQR, interquartile range.
Pediatr Blood Cancer DOI 10.1002/pbc
Outpatient Management of Febrile Neutropenia 1431

6. overall care were similar in both in- and outpatient groups. Patient
questionnaire responses trended higher across all domains in the
outpatient group including well being, independence, interest/
concentration, energy level though only differences in sleep (7.7 vs.
5.6, P ¼ 0.01) and appetite (6.2 vs. 4.4, P ¼ 0.05) were significant.
When analysed by patients rather than episodes of FN, outcome
data were similar.
DISCUSSION
This study demonstrates that home treatment for a low risk
febrile neutropenic children offered superior QOL for patients and
families by several of the measures studied. Outpatient care
provided less disruption to family life. Patients had better appetites
and slept better at home. Families were not significantly more
concerned about the care of their child compared to inpatients and
satisfaction regarding overall care was similar across the two
groups.
Provision of outpatient FN care was feasible within the hospital
studied and provided benefit to the hospital in terms of bed
utilisation. Cost effectiveness of outpatient management has been
previously documented in the literature [21–23]. There were no
significant adverse events attributed to participation in the study; in
particular none ascribed to outpatient FN management. The trend
towards longer fever and episode length in the outpatient group
perhaps reflects the 24-hour clock of multidisciplinary care and
decision making for the inpatient group, versus intermittent
conservative nurse/parent based outpatient care. Cefipime was
not associated with any significant adverse events. Our study had a
relatively high readmission rate relative to several other stud-
ies [6,9,12] likely due to differences in risk stratification and
procedure. These studies were assessing efficacy of outpatient
management whereas our study, focused on QOL, was conservative
in terms of readmission criteria.
While neither family nor patient/child preferences were
assessed, this study supports that, given the choice to receive
treatment in their home environment, a number of patients and
families are willing to assume shared responsibility of care during a
FN episode and would choose to do so. Other authors who have
surveyed parents’ and healthcare professionals’ preferences for
outpatient oral antibiotic management, relative to inpatient
parenteral management, have found that 71% of healthcare
professionals and 53% of parent respondents would choose oral
outpatient management [24]. Parents who preferred outpatient oral
management placed higher importance on comfort and lower
importance on fear/anxiety. Using a threshold technique researchers
were able to ascertain that in general an initial decision for
outpatient management could be overturned if there would be 3
clinic visits per week or 15% chance of readmission. Subse-
quently, from the same group in Toronto, Cheng et al. [25] surveyed
anticipated health-related quality of life (HRQL) and preference
among parents and patients relative to four different (hypothetical)
FN management alternatives: entire inpatient IV, early discharge,
outpatient oral and outpatient IV therapy [25]. Outpatient IV
therapy and early discharge were associated with better HRQL
scores than inpatient management and outpatient oral management.
Similarly, about 50% of parents preferred inpatient management
and factors associated with preference for outpatient management
included higher family income and higher current QOL status of the
child. Quezada et al. [26] highlighted that Outpatient FN
management was only suitable for a minority of cases and
described a number of barriers to outpatient FN alternatives,
including significant medical co-morbidities, language, distance
between residence and medical facility and lack of interest [26].
There is evidence highlighting discrepancy between parent and
patient/child preferences; this needs consideration in further
studies [27]. Outpatient LRFN management is desired by some
and a legitimate alternative for health services to incorporate.
However, decision making around management of LRFN episodes
will continue to involve a complex array of family, social, medical,
local health system and resource factors. The current study
augments a recent review by Teuffel and Sung [28] which suggests
TABLE III. Focus of Infection, Complications, Microbiology and Management Changes
Age, diagnosis Clinical Microbiology Management Admission
Inpatient
11y, ALL Clinically stable BCX:
Streptococcus mitis
Vancomycin D2-D6 Discharge
on teicoplanin
N/A
7y, HL Pneumonia/hypotension IV fluid boluses, vancomycin/timentin/
gentamicin D4-10
N/A
9y, ES Lip ulcer HSV1 Acyclovir D3-D6 N/A
15y, OS Mucositis IV Morphine D3-D7 N/A
3y, ALL Reaction to red blood cell transfusion N/A
Outpatient
7y, ALL Well BCX Candida
species D2
Amphotericin D6-D12 D6-D12
3y, ALL Unwell, prolonged/recurrent fever,
positive fluid balance, lobar
pneumonia with effusion
Vancomycin D2/
Amphotericin D10-D13
D2-D13
4y, ALL Perianal fissure, constipation Timentin/gentamicin D3 D3-D5
5y, RMS Varicella sibling, recurrent fever Zoster immune globulin/acyclovir D3 D3-D7
6y, ALL Perianal fissure, recurrent fever Timentin/gentamicin D6 D6-D10
3y, ALL Afebrile, dental abscess Timentin/gentamicin D8 D8-12
D, days; Y, years; IV, intravenous; BCX, blood culture; HSV, 1 herpes simplex virus type 1; N/A, not applicable.
Pediatr Blood Cancer DOI 10.1002/pbc
1432 Orme et al.

7. that further work should aim to examine the effectiveness of
outpatient FN management through large cohort studies which may
also assess the use of decision tools which are intended to facilitate
ambulatory approaches [28].
There are a number of limitations to our current study. First, the
number of study participants was limited and as such the ability to
understand safety, feasibility and risk stratification is limited. Only
23% of the fever presentations were deemed eligible for
randomisation, and this has limited the depth to which we have
been able to explore the effect on QOL. Findings may therefore
reflect an overall bias towards an ultra low risk FN subset. While
we tracked and recorded most reasons for ineligibility of patients at
the time of the presentation to the ED, we did not track and record
reasons for ineligibility and refusals of patients at initial pre-
consent when they were identified in oncology clinic. We therefore
cannot provide a full picture of the patients who were eligible,
approached and either refused or were ineligible, versus the
patients enrolled. Upon presentation to the emergency, in 6% of
potentially eligible episodes, patients were deemed unsuitable for
participation and randomisation after discussion with the oncology
consultant. A standard fever criterion was used for presentation to
the ED regardless of whether measurement was axillary, oral or
tympanic measurements, allowing inconsistencies in presentation
and enrollment. Both of these factors influenced the study sample
although any effect of this on the outcomes obtained should have
been neutralised by randomisation. Multiple presentations for the
same patient were allowed and this applied to six patients. To
check for the possibility that this introduced bias, an analysis was
done allowing all patients only one FN episode; the results were
similar. We did not assess culturally and linguistically diverse
families and their experience may be very different from the study
population. The majority of families were represented in survey
participation by the mother; fathers may have had different views.
Finally, the survey questions chosen were not part of a validated
QOL tool; however, no other suitable tool for this study design
existed.
CONCLUSION
In this RCT in pediatric patients with FN, we examined QOL
factors for outpatient versus inpatient management. Outpatient
management of low risk FN episodes provided significant benefit to
families by several measures of QOL. Although a small study,
outpatient cefepime management appeared feasible and safe. This
study supports the development of a clinical guideline for selected
families within a tertiary pediatric setting. It requires robust
interdepartmental collaboration and significant, sustained commit-
ment to a hospital based community care program. Such a guideline
should be developed in alignment with broader strategies for
consensus regarding risk stratification and management of pediatric
FN.
ACKNOWLEDGMENTS
We thank the participating patients and families. We thank
oncology and ED medical and nursing staff for their support. We
thank Lucy Holland for her time in preparing the manuscript for
final submission.
REFERENCES
1. Viscoli C. Prophylaxis and empirical therapy for infection in cancer patients. In: Mandell GL, Bennett JE,
Dolin R, editors. Principles and practice of infectious diseases. Philladelphia: Elsevier Churchill
Livingstone; 2005. 3442–3462 p.
2. Innes H, Marshall E. Outpatient management for febrile neutropenia. Curr Opin Oncol 2007;19:294–298.
3. Klastersky J, Paesmans M, Rubenstein EB, et al. The multinational association for supportive care in
cancer risk index: A multinational scoring system for identifying low-risk febrile neutropenic cancer
patients. J Clin Oncol 2000;18:3038–3051.
4. Phillips RS, Lehrnbecher T, Alexander S, et al. Updated systematic review and meta-analysis of the
performance of risk prediction rules in children and young people with febrile neutropenia. PLoS ONE
2012;7:e38300.
5. Gupta A, Swaroop C, Argwala S, et al. Randomized controlled trial comparing oral amoxicillin-
clavulanate and ofloxacin with intravenous ceftriaxone and amikacin as outpatient therapy in pediatric
low-risk febrile neutropenia. J Pediatr Hematol Oncol 2009;31:635–641.
6. Paganini H, Gomez S, Ruvinsky S, et al. Outpatient, sequential, parenteral-oral antibiotic therapy for
lower risk febrile neutropenia in children with malignant disease: A single-center, randomized, controlled
trial in Argentina. Cancer 2003;97:1775–1780.
7. Paganini H, Rodriguez-Brieshcke T, Zubizarretta P, et al. Oral ciprofloxacin in the management of
children with cancer with lower risk febrile neutropenia. Cancer 2001;91:1563–1567.
8. Paganini HR, Sarkis CM, De Martino MG, et al. Oral administration of cefixime to lower risk febrile
neutropenic children with cancer. Cancer 2000;88:2848–2852.
9. Petrilli AS, Dantas LS, Campos MC, et al. Oral ciprofloxacin vs. intravenous ceftriaxone administered in
an outpatient setting for fever and neutropenia in low-risk pediatric oncology patients: Randomized
prospective trial. Med Pediatr Oncol 2000;34:87–91.
10. Santolaya ME, Alvarez AM, Aviles CL, et al. Early hospital discharge followed by outpatient
management versus continued hospitalization of children with cancer, fever, and neutropenia at low risk
for invasive bacterial infection. J Clin Oncol 2004;22:3784–3789.
11. Teuffel O, Ethier MC, Alibhai SM, et al. Outpatient management of cancer patients with febrile
neutropenia: A systematic review and meta-analysis. Ann Oncol 2011;22:2358–2365.
12. Mullen CA, Petropolous D, Roberts WM, et al. Outpatient treatment of fever and neutropenia for low risk
paediatric cancer patients. Cancer 1999;86:126–134.
13. Mustafa MM, Aguino VM, Pappo A, et al. A pilot study of outpatient management of febrile neutropenic
children with cancer at low risk of bacteremia. J Pediatr 1996;128:847–849.
14. Ariffin H, Ai CL, Lee CL, et al. Cefepime monotherapy for treatment of febrile neutropenia in children. J
Paediatr Child Health 2006;42:781–784.
15. Chastagner P, Plouvier E, Eyer D, et al. Efficacy of cefepime and amikacin in the empiric treatment of
febrile neutropenic children with cancer. Med Paediatr Oncol 2000;34:306–306.
16. Chuang YY, Hung IJ, Yang CP, et al. Cefepime versus ceftazidime as empiric monotherapy for fever and
neutropenia in children with cancer. Pediatr Infect Dis J 2002;21:203–209.
17. Mustafa MM, Carlson L, Tkaczewski I, et al. Comparative study of cefepime versus ceftazidime in the
empiric treatment of paediatric cancer patients with fever and neutropenia. Paediatr Infect Dis J
2001;20:362–369.
18. Corapçioglu F, Sarper N. Cefepime versus cetazidime þ amikacin as empirical therapy for febrile
neutropenia in children with cancer: A prospective randomized trial of the treatment efficacy and cost.
Pediatr Hematol Oncol 2005;22:59–70.
19. Close P, Burkey E, Kazac A, et al. A prospective, controlled evaluation of home chemotherapy for
children with cancer. Pediatrics 1995;95:896–900.
20. National Cancer Institute, Common Toxicity Criteria Version 2.0 Fact Sheet. National Cancer Institute
home page. Published 1998 http://ctep.cancer.gov/protocolDevelopment/electronic_applications/docs/
ctcv20_4-30-992.pdf Published 30 April 1999. Accessed 15 June 2012.
21. Klaassen RJ, Allen U, Doyle JJ. Randomized placebo-controlled trial of oral antibiotics in paediatric
oncology patients at low-risk with fever and neutropenia. J Pediatr Hematol Oncol 2000;22:405–411.
22. Teuffel O, Amir E, Alibhai SM, et al. Cost-effectiveness of outpatient management for febrile neutropenia
in children with cancer. Pediatrics 2011;127:e279–e286.
23. Wiernikowski JT, Rothney M, Dawson S, et al. Evaluation of a home intravenous antibiotic program in
paediatric oncology. Am J Pediatr Hematol Oncol 1991;13:144–147.
24. Sung L, Feldman BM, Schwamborn G, et al. Inpatient versus outpatient management of low-risk
paediatric febrile neutropenia: Measuring parents’ and healthcare professionals’ preferences. J Clin
Oncol 2004;22:3922–3929.
25. Cheng S, Teuffel O, Ethier MC, et al. Health-related quality of life anticipated with different management
strategies for paediatric febrile neutropaenia. BJC 2011;105:606–611.
26. Quezada G, Sunderland T, Chan KW, et al. Medical and nonmedical barriers to outpatient treatment of
fever and neutropenia in children with cancer. Pediatr Blood Cancer 2007;48:273–277.
27. Upton P, Lawford J, Eiser C. Parent-child agreement across child health-related quality of life
instruments: A review of the literature. Qual Life Res 2008;17:895–913.
28. Teuffel O, Sung L. Advances in management of low-risk febrile neutropenia. Curr Opin Pediatr
2012;24:40–45.
Pediatr Blood Cancer DOI 10.1002/pbc
Outpatient Management of Febrile Neutropenia 1433