Support Care Cancer (2013) 21:1321–1327
DOI 10.1007/s00520-012-1668-4
ORIGINAL ARTICLE
Symptom to door interval in febrile neutropenia:
perspective in India
Sapna Oberoi & Amita Trehan & R. K. Marwaha &
Deepak Bansal
Received: 27 July 2012 / Accepted: 19 November 2012 / Published online: 15 December 2012
# Springer-Verlag Berlin Heidelberg 2012
Abstract
Purpose Febrile neutropenia (FN) is an oncological emer-
gency to be treated within an hour. In a developing country,
patients are often unable to reach hospital speedily. Our aim
was to determine the symptom to door interval (SDI) in
febrile neutropenic children with acute lymphoblastic leu-
kaemia [ALL] and to identify factors resulting in delay.
Methods All consecutive children of ALL (< 14 years) pre-
senting with FN over a period of 1 year were evaluated.
Data for demographics, clinical details, phase of therapy,
profile of caregivers, travelling time, SDI, reasons for delay,
modes of transport, complications, invasive bacterial infec-
tions (IBI), length of hospital stay and outcome were
recorded.
Results Among 320 FN episodes, median SDI (in hours)
was 24 (IQR 8, 36). SDI during intensive phases was
significantly less as compared to nonintensive phases 12
(IQR 6, 24) and 24 (IQR 24, 48) (p<0.001). Children on
induction phase reported to hospital at earliest [median
8 (IQR 4, 12)], while those on maintenance phase came late
[median 36 (24, 48)]. Median travelling time was 15 min
(IQR 15, 25) for patients on intensive phase and was
180 min (IQR 60, 285) for those on nonintensive phase
(p<0.001). Ingestion of acetaminophen at home (30 %),
inability to realise the gravity of the situation (27 %), un-
awareness of parents (9 %) and nonavailability of transport
(12 %) were the most common reasons for delay. No sig-
nificant association of SDI was seen with complications,
IBI, duration of hospital stay and mortality (p>0.05).
S. Oberoi : A. Trehan (*) : R. K. Marwaha : D. Bansal
Division of Pediatric Hematology–Oncology,
Advanced Pediatric Centre,
Postgraduate Institute of Medical Education and Research,
Chandigarh 160012, India
e-mail: trehanamita@hotmail.com
Conclusions Considerable time lag was seen between onset
of symptoms and reaching hospital. Health education and
establishment of shared care are urgent needs in countries
where tertiary care facilities are limited.
Keywords Symptom to door interval . Delay . Febrile
neutropenia . Childhood cancer . Acute lymphoblastic
leukaemia . India . Shared care
Introduction
Febrile neutropenia is a frequently encountered complica-
tion of therapy in children with cancer. It remains a major
cause of morbidity and mortality especially in countries with
limited resources [1–3]. Morbidity ranges from increased
duration of hospital stay, increased cost of therapy and delay
in chemotherapy to life threatening complications [4–6].
Prompt administration of broad spectrum antibiotics is rec-
ommended in this subset of patients. As per the recent
guidelines on febrile neutropenia by the Infectious Diseases
Society of America, every patient with febrile neutropenia
should receive empirical antibiotic therapy within 2 h of
presentation to hospital [7]. Few centres/groups advise that
door to antibiotic interval should be even less than 60 min
[8]. These recommendations are based on studies which
have shown that systematic time bound management of
febrile neutropenic or a septic patient significantly decreases
mortality [9–12]. It has also been seen that length of hospital
stay correlates positively with door to antibiotic interval
[13].
Studies have evaluated door to antibiotic interval in chil-
dren with febrile neutropenia to assess the factors responsi-
ble for delay and to develop strategies to decrease this time
interval [14–17]. However, no studies have been done to
gauge the time taken from onset of symptoms to reach
1322
hospital, i.e. symptom to door (SDI) in these patients. The
relation of symptom to door, travel time and door to antibi-
otic interval is depicted in Fig. 1. SDI may be much longer
than door to antibiotic interval, adding onto delay in antibiotic
administration. This would be of importance particularly in
countries with limited resources. Identifying this interval
and factors responsible for delay would help institutions to
develop strategies/policies to decrease the symptom to anti-
biotic interval in this vulnerable population. The aim of our
study was to determine the symptom to door interval in
children with acute lymphoblastic leukaemia (ALL) present-
ing with febrile neutropenia and to identify the reasons for
delay. The secondary aim was to see association of SDI with
complications, length of hospital stay, invasive bacterial
infections and mortality.
Patients and methods
This was a prospective study done over a period of 1 year
(Jan 2010–Dec 2010). All children <14 years with ALL,
presenting with febrile neutropenia during this period, were
enrolled. An informed consent was taken from the parents/
guardian within 24 h of admission. The study was approved
by the Institute Ethics Committee. Patient details were en-
tered in a predesigned proforma at admission and during
hospital stay. The demographic and clinical variables en-
tered included age, sex, education status of parents, socio-
economic status of family, income of the family, phase of
treatment, symptom to door interval, travelling time, cause
for the delay, modes of travel, degree of temperature, signs
and symptoms indicative of an infectious focus, number of
episodes per patient, complications, invasive bacterial infec-
tions (IBI), length of hospital stay and outcome. Episodes
occurring during the intensive phases were analysed sepa-
rately from those on nonintensive phases to avoid bias, as
majority of patients stayed near hospital during intensive
phases.
Definitions
Febrile neutropenia was defined in the presence of a single
oral/axillary temperature of ≥38.3 °C (101 °F) or a temper-
ature of ≥38.0 °C (100.4 °F) for ≥1 h along with absolute
Fig. 1 Line diagram depicting
relation between symptom to
door, travel time and door to
antibiotic interval
Support Care Cancer (2013) 21:1321–1327
neutrophil count <500/mm3 or <1,000/mm3 with expected
decline to <500/mm3 within the next 2 days. Socioeconomic
status was classified as per modified Kuppuswamy scale, as
is recommended in our country [18]. Induction, BFM con-
solidation and intensification phases of therapy were la-
belled as intensive phase while CNS consolidation and
maintenance phase were considered as a nonintensive phase.
Travelling time was defined as the time taken to travel to
reach to the hospital. Symptom to door interval was calcu-
lated as the time taken from onset of fever to the time of
reporting to the hospital. The time of onset of fever and
initiation of travel was based on parents’ memory when they
were questioned at the hospital. Inability to understand the
gravity of the illness was considered when the parent/guard-
ian was aware of the need to report to the hospital immedi-
ately but came late for no definite reason. A parent was
considered as ‘unaware’ when he/she did not totally com-
prehend the importance of reporting early and its possible
consequence. A child was considered to have invasive bac-
terial infection if one of the following criteria was met: (1)
occurrence of bacteremia, defined as one or more blood
cultures positive for a bacterial pathogen, and/ or (2) a
positive bacterial culture obtained from a usually sterile site
(indwelling catheter, urine, CSF) or (3) in the absence of a
positive culture, clinical and laboratory findings strongly
suggestive of a sepsis syndrome were present. Complica-
tions were included under the following headings: enceph-
alopathy, congestive heart failure, bleeding manifestations
(mucosal bleeds) severe sepsis, septic shock, multi organ
dysfunction syndrome, metabolic disturbances, ICU admis-
sion, pneumonia requiring invasive or noninvasive ventila-
tion, renal failure, neutropenic enterocolitis and other
complications which were considered serious and clinically
significant.
Unit demographics and protocol
Our Institute is one of the few tertiary care university hos-
pitals in north India treating childhood cancers. Our catch-
ment area extends to a radius of 500–700 km. Patients
coming from remote areas often have to travel by foot to
reach the nearest bus/train station. In our country, train
travel is free for oncology patients. Bus services are free in
some states. No transport service is provided by the hospital.
Support Care Cancer (2013) 21:1321–1327
Ambulance services are limited. Public transport is available
during the daytime with restricted or no service during the
night. Few patients would have their own cars, and cabs are
expensive. In some of the remote districts, it is difficult to
find even a basic hospital and/or a paediatrician within easy
reach. Treatment costs are not borne by the hospital in our
setting. Chemotherapy, antibiotics, disposables, etc. are not
supplied by the hospital. University government hospital
services (as ours) have a nominal charge for admission and
investigations. Health insurance is available to a minority.
Some jobs provide for medical reimbursement. A few state
governments provide a lump sum to children with cancer.
Finances for drugs are also available through government
relief funds and nongovernmental organisations. In our unit,
around 65 % patients do get partial financial support for
drugs. However, all nonmedical expenditure is borne by the
family.
Newly diagnosed ALL patients are treated as per the
guidelines of MRC UKALL 2003, and relapsed patients
are treated as per MRC-R2 protocol. During the intensive
phases of therapy, families are counselled and advised to
stay in or around the hospital area, so that they can reach the
hospital within less than half hour in case of an emergency.
Outstation patients are mostly housed in a patient hostel
within the hospital premises. Some take a flat on rent in
the city. After the intensive phases of therapy, patients return
to their homes (irrespective of distance) and visit the clinic
by appointment. Parents are counselled at the time of diag-
nosis and subsequently at every hospital visit to report to the
paediatric oncology ward/day care when febrile or with any
worrying problem. Patients are also advised to get their
blood counts done in their area when they have fever.
Laboratories doing routine investigations are available in
some places and do not charge astronomically. Some
patients phone the ward/day care/medical personnel with
the blood count report for further advice. The ward is
manned 24/7 by a resident and registrar. All patients with
febrile neutropenia are managed with uniform institutional
febrile neutropenia guidelines consistent with the recom-
mendations of the Infectious Diseases Society of America,
modified for the local setting [7].
Statistical analysis
Baseline clinical and laboratory variables were summarised
using descriptive statistics. Correlation between two
continuous variables was calculated by Pearson or
Spearman correlation coefficient. Comparison of skewed
quantitative variables between two or more groups (with
or without complications, died versus survived) was
done by Mann–Whitney U test or Kruskal–Wallis test.
Categorical variables between groups were compared by
chi-square test. Predictor variables found significantly
1323
related to outcome (p value <0.05) on univariate analysis
were further subjected to multivariate analysis using linear
mixed model with random effects. Statistical analyses were
performed using SPSS software (version 19). All the tests
were two-tailed and p value <0.05 was taken as significant.
Results
A total of 320 febrile neutropenic (FN) episodes occurring
in 176 patients of ALL were studied. There were two chil-
dren who died en route to hospital during the time period of
this study. One was in the BFM consolidation phase and
another in the maintenance phase. This information was
communicated to the team telephonically. These were not
included in the analysis as full details are not available. Of
the 176, 173 were newly diagnosed and three were relapsed
cases. The number of FN episodes per patient was as fol-
lows: one episode in 89 (50.6 %), two episodes in 45 cases
(25.5 %), three episodes in 27 (15.3 %) and four episodes in
15 (8.5 %). The baseline characteristics of all febrile neu-
tropenic episodes are mentioned in Table 1. The number of
febrile neutropenic episodes per patient was not significantly
different in the three socioeconomic strata (p00.54). Median
travelling time in the whole cohort was half hour. This was
15 min [IQR 15, 25 (range 15–360)] for patients on inten-
sive phases of therapy as they were staying close to the
hospital/in the hospital premises. This time was significantly
less when compared to patients on nonintensive phase of
therapy where the travelling time was 180 min [IQR 60, 285
(range 15–600)] (p<0.001). Median duration of symptoms
prior to seeking medical care was 24 h (IQR 8, 36). The
median symptom to door interval in the episodes which
occurred during intensive phases of therapy was 12 h [IQR
6, 24; range 1–72 h]. This was significantly different from
the median SDI of 24 h [IQR 24, 48 (range 2–168)] seen
during nonintensive phase of therapy (p<0.001). The rea-
sons for reaching late as per parents and modes of transport
used during intensive and nonintensive phases of therapy
are given in Table 2.
Complications seen included metabolic disturbances,
pneumonia requiring oxygen support, severe sepsis, severe
mucosal bleeds, NEC, encephalopathy and renal failure in
16, 12, 8, 6, 3, 2 and 0.3 % of the episodes, respectively.
Eighteen patients had septic shock at admission, 14 of
whom had decompensated shock requiring inotropic sup-
port. The median SDI in these patients was 13 h [IQR 7.5,
39; range 2–72]. Of the 76 episodes of invasive bacterial
infection, proven invasive bacterial infection was present in
62 (81.6 %) episodes. Thirty seven of the isolates (57.8 %)
were gram-positive and 27 (42.2 %) were gramnegative.
The most common gram-positive organism was Staphylococ-
cus aureus, seen among 18 (25.7 %) isolates. Escherichia coli
1324 Support Care Cancer (2013) 21:1321–1327

Table 1 Demographic, clinical and laboratory characteristics of febrile Table 2 Reasons for the reaching late and modes of transport during
neutropenic episodes (n0320) episodes of febrile neutropenia (n0320)

Parameter Value Reasons for reaching late Intensive phases Nonintensive phases
1. Defervescence due 29 32
Number of episodes 320 to intake of acetaminophen
Age (years) 5 (4, 7) 2. Inability to 25 29
Male 228 (71 %) comprehend situation
3. Unawareness of 15 4
Education of the caregiver parents/guardian
<High school 89 (27.8 %) 4. Chemotherapy 13 0
≥High school 231 (72.2 %) induced fever
5. Financial constrains 7 10
Monthly income of the family
6. Nonavailability 5 20
Indian rupees 6,000 (4,000, 12,000)
of transport
US dollars 112 (75, 225) 7. Unknown 6 5
Socioeconomic status Modes of transport Intensive phases Nonintensive phases
Upper 40 (12.5 %) 1. Travel by foot 61 7
Middle 200 (62.5 %) 2. Rickshawa 21 3
Lower 80 (25 %) 3. Public bus 7 74
Number of episodes/patient/socioeconomic strata 4. Auto rickshawb 4 7
Upper 1.8 5. Personal/private car 7 9
Middle 1.8
Lower 1.9 Values are in percentage
a
Travelling time (min) 30 (15, 207) Rickshaw is a two-wheeled passenger cart attached to a bicycle
b
Symptom to door interval (h) Auto rickshaw is a limited-capacity passenger vehicle on three wheels
≤1 4 (1 %)
2–12 133 (42 %) septic shock was higher as compared to others (22 vs 4.6 %,
13–24 96 (30 %) p00.002).
>24 87 (27 %) During intensive phases of therapy, no correlation of SDI
Phase of the therapy was observed with travelling time (Spearman’s r00.16)
Induction 45 (14 %) (Fig. 2a), age (r00.02), income of the family (r0−0.21)
BFM consolidation 21 (7 %) and degree of temperature (r0−0.09). Univariate analysis
CNS consolidation 12 (4 %)
of SDI was performed with the various patient character-
Intensification 97 (30 %)
istics. Significant association was observed with the differ-
Maintenance 145 (45 %)
ent blocks of intensive therapy (p00.03) (Table 3). Patients
on induction phase of therapy had least SDI [median 8 (IQR
Fever at presentation ≥39 °C 168 (52 %)
4, 12)], whereas those on BFM consolidation and delayed
Clinical focus of infection at admission 89 (28 %)
intensification had similar symptom door interval [median
Complications 73 (23 %)
12 (IQR 6, 24)] (p00.03) (Fig. 3).
Invasive bacterial infection 76 (24 %)
During nonintensive phases, no correlation of SDI was
Length of hospital stay (days) 6 (4, 9)
seen with travelling time (Spearman’s r00.36) (Fig. 2b), age
Mortality 18 (6 %)
(r00.02), income of the family (r0−0.12) and degree of
Values are median (IQR) and n (percent) temperature (r0−0.03). During nonintensive chemotherapy,
sex, socioeconomic status and the different blocks of non-
intensive chemotherapy had significant association with
was the most common gram-negative isolate. Median dura- SDI (Table 4). Girls, children from upper and middle socio-
tion of hospital stay was 6 days (range 1–40). In the 320 economic strata and patients on CNS consolidation phase
episodes, 302 (94.3 %) patients were discharged and 18 died had significantly lower SDI (p00.04, 0.04 and 0.02). Mul-
(5.6 %). Six deaths occurred in delayed intensification, five tivariate analysis was performed on the parameters found to
during induction, three in consolidation and four in the be significant on univariate analysis, i.e. sex, socioeconomic
maintenance phase of therapy. The causes of death were status and phase of therapy. Maintenance phase of chemo-
pneumonia in nine (2.8 %), neutropenic enterocolitis in therapy was the only independent parameter associated with
four (1.2 %), severe bacterial sepsis in three (1 %), fungal a prolonged symptom to door interval (p00.019).
sinusitis in one (0.3 %) and intracranial haemorrhage in Univariate analysis done to look at the association of
one (0.3 %). Mortality in the patients who presented with SDI with complications, IBI, duration of hospitalisation
Support Care Cancer (2013) 21:1321–1327 1325

Table 3 Univariate analysis of putative risk factors with symptom to

door interval in febrile neutropenic episodes occurring during intensive
phases of therapy (n0163)

Parameter Symptom to door interval P value

(h) [median (IQR)]

Sexa
Male (110) 12 (16, 24) 0.33
Female (53) 18 (4, 24)
Education status of the caregivera
Less than high school (45) 12 (5.5, 24) 0.97
≥High school (118) 12 (6, 24)
a
Socioeconomic status
Upper and middle (119) 10 (6, 20) 0.199
Lower (44) 12 (6, 24)
Focus of infection at admissiona
Yes (42) 9 (4.7, 18.5) 0.44
No (121) 12 (6, 24)
Phase of chemotherapya 18 (4, 12)
Induction (45) 12 (6, 24) 0.03
BFM consolidation (21) 12 (6, 24)
Delayed intensification (97)
Complicationsb
Yes (42) 10 (4, 19.5) 0.48
No (121) 12 (6, 24)
Invasive bacterial infectionsb
Yes (40) 18 (4, 16) 0.22
No (123) 12 (6, 24)
Outcomeb
Discharge (150) 24 (6,24) 0.39
Death (13) 6 (5, 21)
a
SDI is the outcome of interest
b
SDI is the primary variable. Complications, invasive bacterial infec-
tions and outcome are the outcomes of interest
Fig. 2 Correlation between travelling time (in minutes) and symptom
to door interval (in hours) a during intensive phases of therapy (n=163)
(Spearman’s r00.16) b during nonintensive phases of therapy (n=157)
(Spearman’s r00.36)

and mortality during both intensive and nonintensive

phases did not reveal any significant association (p>0.05)
(Tables 3 and 4).

Discussion

Our data was collected prospectively from a large sam-

ple of homogenous patient population in a developing
country setup. It is an attempt to depict the actual
scenario that exists in resource-limited countries, where
symptom to door interval is prolonged. It highlights the
causes responsible for the delay in SDI and higher rates
of complications, IBI and mortality seen in developing Fig. 3 Box and whisker plot of symptom to door interval (in hours)
countries. during different phases of chemotherapy (n0320)
1326 Support Care Cancer (2013) 21:1321–1327

Table 4 Univariate analysis of putative risk factors with symptom to thereby meaning that distance from the hospital is not re-
door interval in febrile neutropenic episodes occurring during
sponsible for the delay. Patients on induction reached earli-
nonintensive phases of therapy (n0157)
est, with a median SDI of 8 h, which is far behind the goal of
Parameter Symptom to door interval P value reaching within minutes. The most common reason for
(h) [median (IQR)] delay was false reassurance secondary to defervescence
Sexa
resulting from intake of acetaminophen. This occurred de-
spite counselling sessions, where patients are advised to
Male (118) 36 (24, 48) 0.04
Female (39) 24 (12, 48) avoid intake of acetaminophen at the onset of fever. The
Education status of the caregivera second widespread reason was the lack of understanding of
Less than high school (44) 30 (24, 66) 0.14 the urgency in febrile neutropenia by the parents/guardians.
≥High school (113) 24 (18, 48) This was seen essentially during the maintenance phase. We
Socioeconomic status a
found that the general feeling of the caregivers during main-
Upper and middle (121) 24 (18, 48) 0.043 tenance therapy is of security, especially so when the ANC
Lower (36) 36 (24, 72) at the last clinic visit was appropriate, resulting in parents
Focus of infection at admissiona waiting for a second spike of fever prior to reporting to the
Yes (47) 36 (24, 48) 0.18 medical personnel. Child being ‘well’ was another reason
No (110) 24 (8, 48) cited by parents to await another spike of fever. The third
Phase of chemotherapya most common reason for delay during nonintensive phases
CNS consolidation (12) 13 (6, 42) 0.02 of therapy was nonavailability of public transport at night,
Maintenance (145) 36 (24, 48)
especially in the remote areas, with parents being unable to
Complicationsb
afford private transport.
Yes (31) 24 (12, 48) 0.44
Our results point towards the need for repetitive counsel-
No (126) 27 (24, 48)
ling sessions, giving written instructions regarding febrile
Invasive bacterial infectionsb
neutropenia to patients and local physicians, providing an
Yes (36) 42 (19.5, 48) 0.29
No (121) 24 (24, 48) antibiotic kit and developing a 24-h help line. This is a big
Outcomeb challenge in developing countries where patient to doctor
Discharge (152) 24 (13, 54) 0.68 ratio is very high. Our unit caters to around 350 new paedi-
Death (5) 26 (24, 48) atric oncology patients per year and is manned by three
consultants and three registrars. We do not have a dedicated
a
SDI is the outcome of interest counsellor in our unit. Counselling is done mainly by the
b
SDI is the primary variable. Complications, invasive bacterial infec- doctors and nurses at admission and discharge. Issues re-
tions and outcome are the outcomes of interest
garding poor connectivity and inadequate public transport
need to be addressed. University hospitals need to coordi-
Our results show that symptom to door interval in febrile nate with doctors/hospital in smaller and remote areas for a
neutropenic patients of ALL is prolonged. Children during system of ‘shared care’. We need to work towards the
induction phase of therapy are brought to the hospital early, improvement of medical facilities in the smaller towns and
while those in maintenance phase reach late. Travelling time villages for our oncology patients. If a patient is assured of
has no correlation with symptom to door interval. The two medical care in his/her town, there is a higher probability of
main causes for reaching late were administration of acet- seeking medical advice instantly. Arranging workshops for
aminophen at home and incomprehension of the possible the local physicians may help to tune them to the manage-
gravity of febrile neutropenia by the parents. Complications, ment of fever in immunocompromised patients. Providing
invasive bacterial infections and mortality were higher as an antibiotic kit to every patient with instructions to local
compared to the developed world. However, no impact of physicians would help them to treat these children at the
SDI was seen on complications, invasive bacterial infec- earliest. Patients living in very remote areas without access
tions, length of hospital stay and mortality. to local physician need to be provided with an oral antibiotic
The goal of reaching the cancer centre within minutes of to be administered when the child develops fever prior to
onset of fever was not met in nearly all of our patients being able to leave home and reach the nearest health care
including those staying close to the hospital, as is evident facility. As financial constraints were responsible for delay
by only 1 % reporting within ≤1 h. Studies available have in nearly one fifth of the patients, there is a need that
focused on delay after reaching the hospital and not on SDI. universal health insurance is made mandatory and free
Most studies available are from developed nations where transport is provided for the sick children.
SDI is unlikely to be a hindrance for immediate medical We failed to document the association of SDI with
care. Travelling time did not have any correlation with SDI, complications, length of hospital stay, invasive bacterial
Support Care Cancer (2013) 21:1321–1327
infections and mortality. This could be explained by three
reasons. Firstly, we did not look at the door to antibiotic
interval in our study, which could have played a role in
negating the effect of SDI in these parameters. Secondly,
nearly 99 % patients reached beyond the targeted time, so
the effect of coming early on morbidity could not be
demonstrated in our study. The patients who presented with
septic shock had a mortality of 22 % which was significantly
higher than other patients. Median SDI in these patients was
13 h. In severe sepsis, each hour of delay in antimicrobial
administration over the ensuing 6 h of onset of hypotension
has been shown to be associated with an average decrease in
survival of 7.6 % [19]. It is possible that these patients may
have survived if they had reached hospital early. Finally, the
parents might have tended to bring ‘sick-looking’ febrile
children earlier than ‘well-looking’ febrile children, and
so these ‘sick’ children got treatment earlier than ‘well’
children. This may have nullified the effect on SDI on
the outcome.
In conclusion, symptom to door interval is prolonged in
our setup. The majority reached beyond the targeted time to
the hospital. Health education on supportive care and estab-
lishing a system of shared care with local physicians is the
need of the hour.
Conflict of interest None
References
1. Bakhshi S, Padmanjali KS, Arya LS (2008) Infections in child-
hood acute lymphoblastic leukemia: an analysis of 222 febrile
neutropenic episodes. Pediatr Hematol Oncol 25:385–392
2. Gupta S, Bonilla M, Gamero M et al (2011) Microbiology and
mortality of pediatric febrile neutropenia in El Salvador. J Pediatr
Hematol Oncol 33:276–280
3. Raje NS, Rao SR, Iyer RS et al (1994) Infection analysis in acute
lymphoblastic leukemia: a report of 499 consecutive episodes in
India. Pediatr Hematol Oncol 11:271–280
4. Kuderer NM, Dale DC, Crawford J et al (2006) Mortality,
morbidity, and cost associated with febrile neutropenia in adult
cancer patients. Cancer 106:2258–2266
1327
5. Michels SL, Barron RL, Reynolds MW et al (2012) Costs associ-
ated with febrile neutropenia in the US. PharmacoEconomics
30:809–823
6. Hakim H, Flynn PM, Knapp KM, Srivastava DK, Gaur AH (2009)
Etiology and clinical course of febrile neutropenia in children with
cancer. J Pediatr Hematol Oncol 31(9):623–629
7. Freifeld AG, Bow EJ, Sepkowitz KA et al (2011) Clinical practice
guideline for the use of antimicrobial agents in neutropenic
patients with cancer: 2010 update by the Infectious Diseases
Society of America. Clin Infect Dis 52:e56–e93
8. Corey AL, Snyder S (2008) Antibiotics in 30 minutes or less for
febrile neutropenic patients: a quality control measure in a new
hospital. J Pediatr Oncol Nurs 25:208–212
9. Zuckermann J, Moreira LB, Stoll P et al (2008) Compliance with a
critical pathway for the management of febrile neutropenia and
impact on clinical outcomes. Ann Hematol 87:139–145
10. Battleman DS, Callahan M, Thaler HT (2002) Rapid antibiotic
delivery and appropriate antibiotic selection reduce length of
hospital stay of patients with community-acquired pneumonia: link
between quality of care and resource utilization. Arch Intern Med
162:682–688
11. Rivers E, Nguyen B, Havstad S et al (2001) Early goal-directed
therapy in the treatment of severe sepsis and septic shock. N Engl J
Med 345:1368–1377
12. Townsend SR, Schorr C, Levy MM et al (2008) Reducing mortal-
ity in severe sepsis: the Surviving Sepsis Campaign. Clin Chest
Med 29:721–733, x
13. Sammut SJ, Mazhar D (2012) Management of febrile neutropenia
in an acute oncology service. QJM 105:327–336
14. Burry E, Punnett A, Mehta A et al (2011) Identification of educa-
tional and infrastructural barriers to prompt antibiotic delivery in
febrile neutropenia: a quality improvement initiative. Pediatr
Blood Cancer 59:431–435. doi:10.1002/pbc.23418
15. Ben-Yakov MES, Kiss A, J-Schull M (2010) CBC testing before
physician assessment delays antibiotics in emergency febrile
neutropenia patients. Emergencias 22:429–434
16. van Vliet M, Potting CM, Sturm PD et al (2011) How prompt is
prompt in daily practice? Earlier initiation of empirical antibacte-
rial therapy for the febrile neutropenic patient. Eur J Cancer Care
(Engl) 20:679–685
17. Amado VM, Vilela GP, Queiroz A Jr et al (2011) Effect of a quality
improvement intervention to decrease delays in antibiotic delivery
in pediatric febrile neutropenia: a pilot study. J Crit Care 26(103):
e109–e112
18. Mishra D, Singh HP (2003) Kuppuswamy's socioeconomic status
scale—a revision. Indian J Pediatr 70:273–274
19. Kumar A, Roberts D, Wood KE et al (2006) Duration of hypoten-
sion prior to initiation of effective antimicrobial therapy is the
critical determinant of survival in human septic shock. Crit Care
Med 34:1589–1596
Copyright of Supportive Care in Cancer is the property of Springer Science & Business Media B.V. and its
content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's
express written permission. However, users may print, download, or email articles for individual use.