514793

research-article2013
JPOXXX10.1177/1043454213514793Journal of Pediatric Oncology Nursing XX(X)Cox and Bradford

Article
Journal of Pediatric Oncology Nursing

Management of Febrile Neutropenia in

2014, Vol. 31(1) 28­–33
© 2013 by Association of Pediatric
Hematology/Oncology Nurses
Pediatric Oncology Across Queensland, Reprints and permissions:
sagepub.com/journalsPermissions.nav

Australia: A Retrospective Review on DOI: 10.1177/1043454213514793

jpo.sagepub.com

Variations Between Locations

Anita Cox, BSc (Hons), RGN, RSCN, Dip (ANC Pediatric Oncology)1, and
Natalie Bradford, RN, BNursing, MPH2

Abstract
Febrile neutropenia (FN) is a common complication in pediatric oncology with intravenous antibiotics being given
routinely for decades. This study aimed to compare the management of FN in children in different locations across
Queensland, Australia. FN episodes were identified from 4 settings: tertiary oncology outpatient department (OD),
tertiary emergency department (ED), regional ward (RW), and regional emergency department (RED) between July
2009 and June 2011. Retrospective data were extracted from medical records, collated, and then analyzed to identify
differences in outcomes attributable to location. Sixty-one episodes of FN were reviewed: 5 cases from OD, 28 from
ED, 19 from RW, and 5 from RED. Statistically significant differences occurred in the time taken for medical review
for cases depending on location of presentation. Patients who presented to the ED or the RW were more likely to
be seen within 30 minutes than in other locations (P = .014), and patients who presented to the tertiary hospital
in either the OD or ED were more likely to commence antibiotics within 120 minutes of presentation (P = .023).
Antibiotics were commenced within 60 minutes (the gold standard) on only 10 occasions. Despite education input
emphasizing the importance of early commencement of antibiotics, this study identified that this objective was not met
in the majority of cases. Further study is needed to look at reasons for the delay in beginning treatment for pediatric
oncology patients in Queensland and what measures may assist with improving the time from presentation with FN
to antibiotic administration across all settings.

Keywords
febrile neutropenia, Pediatric oncology, patient management

Background demonstrates that appropriate antibiotic therapy should

Febrile neutropenia (FN) is a common complication in
patients undergoing chemotherapy to treat cancer. It
remains the most common cause of nonplanned hospital-
ization, occurring at unpredictable times during the
patients’ treatment regime (Lyman et al., 2010; Segel &
Halterman, 2008). Definitions of both fever and neutro-
penia differ across the oncology supportive care litera-
ture; however, general consensus defines fever as greater
than 38.5°C on 1 occasion, or greater than 38°C on 2
occasions less than an hour apart, and neutropenia as an
absolute neutrophil count of <1.0 × 109/L. FN is a poten-
tially life-threatening condition that requires prompt
medical intervention (Livadiotti et al., 2012; Lyman et
al., 2010). Baltic, Schlosser, and Bedell (2002) state that
although the term prompt is not defined in numbers of
minutes in biomedical literature, the literature does indi-
cate that the sooner antibiotics are initiated, the greater
the likelihood of a positive clinical outcome. Evidence
be administered as early as possible, preferably within the
first hour of recognition of severe sepsis or shock
(Dellinger et al., 2008). In febrile neutropenic patients,
signs of sepsis may be subtle and not always identifiable
at presentation. Any delay in care could result in increased
morbidity and mortality.
For decades empiric treatment with intravenous broad-
spectrum antibiotics has been the treatment of choice for
FN in pediatric oncology patients. Mortality from FN in
the 1970s was approximately 30% to 40%, but in the late
1
Royal Children’s Hospital, Brisbane, Queensland, Australia
2
The University of Queensland, Centre for Online Health, Royal
Children’s Hospital, Brisbane
Corresponding Author:
Anita Cox, BSc (Hons), RGN, RSCN, Dip (ANC Pediatric Oncology),
Royal Children’s Hospital, Level 1 South Tower, Herston Road,
Brisbane, Queensland 4029, Australia.
Email: anitajcox@gmail.com

Downloaded from jpo.sagepub.com at PENNSYLVANIA STATE UNIV on March 5, 2016

Cox and Bradford 29
1990s this had fallen to just 1% (Orudjev & Lange, 2002).
This dramatic improvement in the outcome of FN has
been attributed to more aggressive initial management of
FN, improvements in antibiotics, increased use of central
venous access devices, and effective supportive care.
Standard treatment for FN patients is emergency hospi-
talization and empirical broad-spectrum intravenous anti-
microbial therapy until resolution of fever, neutropenia,
and signs of infection (Brack et al., 2012). There remains,
however, a lack of a standardized approach to the man-
agement of FN. An Australasian prospective audit in
2005 identified 18 different antibiotic regimes imple-
mented as first-line therapy across 127 FN episodes in 9
centers, highlighting the lack of published national guide-
lines for this group of patients (Chamberlain, Smibert,
Skeen, & Alvaro, 2005).
The Queensland Pediatric Haematology Oncology
Network (QPHON) was established in 2006 to coordinate
care for patients across the whole state of Queensland and
Northern New South Wales and is located within the
QCCC (Queensland Children’s Cancer Centre) at the
major pediatric tertiary hospital in Queensland. Through
education, coordination, and communication, QPHON’s
aim is to improve clinical care and optimize management
and health outcomes for this client group, as well as cre-
ate a comprehensive integrated service. Since 2007, nurs-
ing Regional Case Manager (RCM) roles have been
implemented in 10 regional sites and lead pediatricians
for the pediatric oncology patients have been identified.
The development of these roles alongside an extensive
education program and development of resources for
staff enables patients to return home (or closer to home)
to receive parts of their therapy in Shared Care Units
(SCUs) at the local regional hospital, while still having
their care overseen by the consultants at the tertiary
hospital.
Patients undergoing treatment for leukemia and cancer
at the QCCC can present with FN in a number of loca-
tions: (a) the tertiary hospital’s oncology outpatient
department (OD), (b) the tertiary hospital’s emergency
department (ED), (c) the regional hospital pediatric ward
(RW), or (d) the regional hospital’s emergency depart-
ment (RED). Guidelines for the administration of antibi-
otics have been developed by QCCC and are circulated to
each of the regional SCU around Queensland. The guide-
lines do not specify any timing for administration but list
the antibiotics of choice and their doses as well as any
changes to treatment based on the clinical state of the
patient (see Figure 1).
Anecdotal reports from families, staff, and the authors’
own clinical practice indicate that at times patients do not
get assessed and treated as promptly as they should and
treatment may be delayed in starting.
Downloaded from jpo.sagepub.com at PENNSYLVANIA STATE UNIV on March 5, 2016
The aim of this study was to formally review and com-
pare the management of FN in children who were treated
in different locations throughout Queensland.
Methods
A retrospective review of patients’ medical records was
undertaken to compare the interventions and outcomes
for patients treated for FN in different locations. This
study was approved by the RCH HREC, Ref Number
HREC/12/QRCH/46.
Eligibility and Selection of Cases
Patients receiving treatment for cancer or leukemia who
were treated for febrile neutropenia between July 1, 2009,
and June 30, 2011, were eligible for review. The study
objective was to review 15 records of FN episodes in each
of the 4 groups: (a) OD, (b) ED, (c) RW, and (d) RED, a
total of analysis of 60 FN episodes. Patients were eligible
to be included in the review more than once as it is the FN
episode that was of interest and not the patient. The 2-year
window was chosen as the regional shared care sites have
fewer patients presenting with FN. Assistance from the
coding department generated a list of all the patients admit-
ted during the 2-year time period for FN. Charts were
reviewed for children treated for FN at the tertiary hospital
site. Each of the 10 SCUs across Queensland were invited
to be involved in the study, and the RCM was asked to
assist with data collection from each regional site’s medi-
cal records. One email reminder was sent out to the SCUs
requesting their participation in the data collection.
Data Collection
A report form was created to collect the same information
from all sites. This form captured information regarding
when and where the patients presented for treatment, who
was involved in the care and coordination of the patients,
what treatment was received, and the patients’ response
to the treatment. The form was piloted at the tertiary hos-
pital to ensure it was user friendly and that the correct
information was collected. The form was also sent to
each shared care site contact (RCM) for comment.
Each shared care site in Queensland was then sent a
package containing a cover letter about the study, a copy
of the ethics approval, and also a number of the report
forms on which to collect the data.
Data Analysis
Data were entered into an Excel spreadsheet and sum-
marized using descriptive statistics (counts, frequencies,
percentages). Categories were developed for data
30 Journal of Pediatric Oncology Nursing 31(1)

Figure 1.  Queensland’s Pediatric Hematology Oncology Network’s febrile neutropenia guidelines.

Downloaded from jpo.sagepub.com at PENNSYLVANIA STATE UNIV on March 5, 2016

Cox and Bradford 31
pertaining to time to enable statistical analysis across
categories. Fisher’s exact test was used to assess the sig-
nificance of differences attributed to location. All statis-
tical analysis was undertaken using Stata, Version 12.
Results
Forms were returned from 5 of the 10 SCU sites. Data
were collected for 61 episodes of FN across the 4 groups:
9 presentations were to OD, 28 presentations to ED, 19
presentations to RW, and 5 presentations to RED.
Patient Characteristics
There was an even distribution across female (n = 31,
51%) and male (n = 30, 49%) patients. There was a
range of diagnoses represented including leukemia (n =
48, 80%) and solid tumors (n = 13, 20%). Febrile neu-
tropenia occurred across all ages; there was no obvious
pattern regarding the FN episode in relation to the time
since diagnosis with cancer (median = 11 months; range
= 3 weeks to 3 years). The standard antibiotic regime
was commenced in 84% of cases. Appropriate reasons
were given for alternative antibiotic regimes in the
remainder of cases (eg, patient had recently been on a
certain antibiotic, patient had shown a previous reaction
to standard antibiotics). Appropriate tests were ordered
for each patient depending on the individual patient’s
condition (FBC, blood cultures, electrolytes, urine,
chest x-ray, and nasopharangeal aspirate). Patients were
treated for a mean of 5 days on intravenous antibiotics
(range = 1-18).
Patient Management
Table 1 presents the differences in management of treat-
ment for febrile neutropenia attributable to the location of
presentation. Due to lack of documentation in patient
medical records, not all data could be collected for all
patients. Additionally, patients were excluded from anal-
ysis when there were delays in treatment based on advice
from the treating oncologist at the tertiary hospital. There
were statistically significant differences (P = .014) in the
time to medical review depending on place of presenta-
tion. Patients were more likely to be seen in under 30
minutes if they presented to ED (87%) or RED (60%)
compared with patients presenting to OD (25%) or RW
(26%). There were also statistically significant differ-
ences (P = .023) in the time taken to commence antibiot-
ics depending on place of presentation. Only 2 locations
recorded time of administration of antibiotics occurring
in less than 60 minutes—RW (16%) and ED (25%). There
was no difference observed between groups regarding the
number of days needed to treat the FN episode or
Downloaded from jpo.sagepub.com at PENNSYLVANIA STATE UNIV on March 5, 2016
difference seen between patients with tunneled central
lines or in-dwelling implanted catheters.
There were also concerns regarding the administration
of the second prescribed antibiotic. The gold standard is
60 minutes following the administration of the first anti-
biotic; however, there were cases where the second
administration was administered up to 255 minutes later.
As these data were not consistently available for all sites,
no analysis could be undertaken.
Discussion
The aim of the study was to review and compare the man-
agement of FN in children who were treated in different
locations. Although all the patients were correctly
assessed and treated with the correct drug regime, the
time to drug administration exceeds the recommended
time allocation across all locations. This would indicate
that the educational input to date developed by QPHON
is not sufficient to ensure best practice for this group of
patients. This small study has provided baseline informa-
tion from which further studies are needed to identify
ways to improve the time to antibiotic administration for
FN in pediatric patients who are undergoing treatment for
cancer. Significant differences in the management and
treatment of FN were identified that are attributable to the
differences in location of care. Due to poor documenta-
tion, particularly at the tertiary hospital, there were miss-
ing data regarding the time to medical review from 29%
of the surveys. Despite this, our analysis identified that
the differences in those reported are statistically signifi-
cant and therefore of concern. Medical reviews occurred
later than 1 hour after presentation in 16% of cases. In the
RW less than 30% (n = 5) of patients were seen medically
within half an hour, although almost 75% (n = 14) were
seen within an hour of presenting with FN. Patients seen
in RED appeared to be reviewed in a timely manner but
commencement of their antibiotics was significantly
delayed (up to 299 minutes in 1 case).
The time to start antibiotics is associated with the
place of presentation for treatment for FN. Possible rea-
sons for the differences could be attributed to staffing lev-
els, experience of the staff available, understanding of the
urgency for prompt treatment, or lack of knowledge about
resources available to assist with decisions for treatment
of these patients. To facilitate best practice clinical care
for patients with FN, QPHON has provided extensive
education, written guidelines, and flow charts to support
decision making for assessment and treatment. These
resources are available in all RED and RW. This study
suggests that these efforts are not sufficient to ensure best
practice. The gold standard is for patients presenting with
FN to commence intravenous antibiotics within 60 min-
utes of presentation; this occurred in only 10 cases in this
32 Journal of Pediatric Oncology Nursing 31(1)

Table 1.  Outcomes for Management of Febrile Neutropenia Treatment by Location.

OD, N = 9 (15%), ED, N = 28 RW, N = 19 RED, N = 5 Significance Test,

Variables n (%) (46%) (31%) (8%) Fisher’s
Time to start antibiotics P = .023
(minutes)
 0-59 — 7 (25) 3 (16) —  
 60-119 5 (63) 16 (57) 7 (37) —  
 120-179 2 (25) 4 (14) 5 (26) 2 (40)  
 180-239 1 (13) — 2 (11) 1 (20)  
 240-299 — — 1 (5) 2 (40)  
 300+ — 1 (4) 1 (5) —  
  Missing [excluded] [1] — — [1]  
Time to medical review P = .014
(minutes)
 0-29 1 (25) 13 (87) 5 (26) 3 (60)  
 30-59 2 (50) 1 (7) 9 (47) 2 (40)  
 60-89 1 (25) 1 (7) 2 (9) —  
 90+ — — 3 (15) —  
  Missing [excluded] [1] [13] — —  
Time to admit (minutes) P = .702
 0-59 — — 17 (100) 1 (25)  
 60-119 1 (20) — — —  
 120-179 1 (20) 7 (33) — 1 (25)  
 180+ 3 (60) 14 (7) — 2 (50)  
  Missing [excluded] [4] [7] [1] [1]  
Number of days on IVAbs P = .439
 0-3 5 (56) 11 (39) 10 (53) 5 (100)  
 4-6 2 (22) 9 (32) 6 (32) —  
 7+ 2 (21) 8 (29) 3 (16) —  
IV access P = .724
 Portacath 6 (67) 18 (64) 9 (47) 4 (80)  
  Central line 3 (33) 7 (25) 7 (37) 1 (20)  
 Peripheral — 3 (10) 1 (5) —  
  Peripheral + central — — 2 (11) —  

Note: OD = oncology outpatient department; ED = tertiary hospital’s emergency department; RW = regional hospital pediatric ward; RED = the
regional hospital’s emergency department; IVAb = intravenous antibiotics.

study period and only in the RW and tertiary hospital ED
settings.
There were also concerns regarding the administration
of the second antibiotic. This issue was further compli-
cated when transfer of patients occurred from emergency
departments to a ward area and appears to be a factor con-
tributing to the delay. The time of transfer to wards was
very difficult to track in some charts as documentation
was very poor. In some cases, it was not clear what ward
the patient had been transferred to or at what time, and
numerous parts of the medical record had to be reviewed
to track patient movement.
It should be noted, however, that in this study delayed
antibiotic administration was not associated with nega-
tive patient outcomes or an increased hospital length of
stay. It is acknowledged that most patients returning to
regional areas are likely to have been in a less intensive
phase of treatment. However, best practice, and safety
and quality principles should apply throughout the patient
journey and the time lag between presentation and treat-
ment for some patients has to be addressed.
There is a need for further investigation to understand
where and why the delays occur between medical review
and commencement of treatment.
Conclusion
In this small retrospective chart review of the manage-
ment and treatment of FN for children with cancer in dif-
ferent hospital settings, significant issues regarding the
delay in treatment attributable to location of care was
identified. Reviewing medical records that track the

Downloaded from jpo.sagepub.com at PENNSYLVANIA STATE UNIV on March 5, 2016

Cox and Bradford 33
patient journey for treatment of FN assisted with identify-
ing gaps in service provision and areas where the quality
of care delivered to patients could be improved.
Targeted interventions need to be identified and imple-
mented in all areas where pediatric oncology patients
may present. Minimum standards and key performance
indicators need to be established and regular evaluation
of care needs to be specific enough to ensure that children
are getting safe, efficient, timely treatment regardless of
where they present. Documentation of care needs to be
detailed, accurate, and timely from all clinicians who are
caring for the patient and be clearly accessible in the
medical records. Although approaches to FN have
changed little over the past 30 years, improvements in
efficiency and consistency could have a significant
impact on the quality of life for the patient and family as
well as on the cost of care.
Acknowledgment
The authors would like to acknowledge the ongoing support for
the Sporting Chance Cancer Foundation and their commitment
to improving services for children with cancer in Queensland.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with
respect to the research, authorship, and/or publication of this
article.
Funding
The author(s) received no financial support for the research,
authorship, and/or publication of this article.
References
Baltic, T., Schlosser, E., & Bedell, M. K. (2002). Neutropenic
fever: One institution’s quality improvement project to
decrease time from patient arrival to initiation of antibiotic
therapy. Clinical Journal of Oncology Nursing, 6, 337-340.
Downloaded from jpo.sagepub.com at PENNSYLVANIA STATE UNIV on March 5, 2016
Brack, E., Bodmer, N., Simon, A., Leibundgut, K., Kühne, T.,
Niggli, F. K., & Ammann, R. A. (2012). First-day step-
down to oral outpatient treatment versus continued standard
treatment in children with cancer and low-risk fever in neu-
tropenia. A randomized controlled trial within the multi-
center SPOG 2003 FN study. Pediatric Blood & Cancer,
59, 423-430.
Chamberlain, J., Smibert, E., Skeen, J., & Alvaro, F. (2005).
Prospective audit of treatment of pediatric febrile neutrope-
nia in Australasia. Journal of Pediatrics and Child Health,
41, 598-603.
Dellinger, R. P., Levy, M. M., Carlet, J. M., Bion, J., Parker,
M. M., Jaeschke, R., . . . Beale, R. (2008). Surviving sep-
sis campaign: International guidelines for management
of severe sepsis and septic shock: 2008. Intensive Care
Medicine, 34(1), 17-60.
Livadiotti, S., Milano, G. M., Serra, A., Folgori, L., Jenkner,
A., Castagnola, E., . . . Zanazzo, G. (2012). A survey on
hematology-oncology pediatric AIEOP centers: prophy-
laxis, empirical therapy and nursing prevention procedures
of infectious complications. Haematologica, 97, 147-150.
Lyman, G. H., Michels, S. L., Reynolds, M. W., Barron, R.,
Tomic, K. S., & Yu, J. (2010). Risk of mortality in patients
with cancer who experience febrile neutropenia. Cancer,
116, 5555-5563.
Orudjev, E., & Lange, B. J. (2002). Evolving concepts of man-
agement of febrile neutropenia in children with cancer.
Medical and Pediatric Oncology, 39, 77-85.
Segel, G. B., & Halterman, J. S. (2008). Neutropenia in pediat-
ric practice. Pediatrics in Review, 29(1), 12-24.
Author Biographies
Anita Cox, BSc (Hons), RGN, RSCN, Dip (ANC Pediatric
Oncology), has more than 20 years of experience in pediatric
oncology across two countries (the United Kingdom and
Australia) and a variety of settings from tertiary hospitals to
regional units.
Natalie Bradford, RN, BNursing, MPH, is a nurse researcher
with the University of Queensland’s Centre for Online Health,
where she coordinates the Raelene Boyle Outreach Program.