Increasing incidence of invasive aspergillosis in pediatric hematology oncology patients over the last decade: a retrospective single centre study.

Abstract

There is scanty information about invasive aspergillosis (IA) in the pediatric population. A review of IA at Hospital Infantil Universitario Niño Jesús between 1996 and 2006 was undertaken to analyze incidence, risk factors, and treatment response. Twenty patients were diagnosed with probable or proven IA during the study period, with a cumulative incidence of 1.96%. Incidence was higher in hematopoietic stem cell transplantation (HSCT) recipients: 2.26% (3.5% in allogeneic HSCT and 1.2% in autologous HSCT). A significative increase in IA incidence was observed along the study period (P=0.013), although this increase did not reach signification if only proven cases were compared (P=0.058). Most patients presented multiple risk factors for IA (87% more than 1, and 47% more than 3). The most frequently described risk factor was chemotherapy (90%), after by long-term neutropenia (90%), and corticotherapy (75%). Main locations of the infection were pulmonary (8 patients), cutaneous (3 patients) and intestinal (3 patients). Six patients presented disseminated IA. Initial response to treatment was 55%, although 3 of these cases had a subsequent episode. Global antifungal response, at the end of the follow-up, was 45%. IA-related mortality was 55%. Global mortality was 90%. Only 2 patients (isolated cutaneous IA cases) survived. Seven patients died due to their underlying malignant disease without active fungal disease. Incidence of IA in oncology children is increasing, and in adults. In our experience, IA is a marker of poor outcome even for patients who initially respond to antifungal treatment.

Full text

Page 1

Increasing Incidence of Invasive Aspergillosis in Pediatric
Hematology Oncology Patients Over the Last Decade
A Retrospective Single Centre Study
Pedro M. Rubio, MD,* Julia´n Sevilla, MD,w Marta Gonza ´lez-Vicent, MD,w
Alvaro Lassaletta, MD,w Manuel Cuenca-Estrella, MD,z Miguel A. Dı´az, MD,w
Susana Riesco, MD,* and Luis Madero, MDw
Summary: There is scanty information about invasive aspergillosis
(IA) in the pediatric population. A review of IA at Hospital Infantil
Universitario Nin ˜ o Jesu ´ s between 1996 and 2006 was undertaken
to analyze incidence, risk factors, and treatment response. Twenty
patients were diagnosed with probable or proven IA during the
study period, with a cumulative incidence of 1.96%. Incidence was
higher in hematopoietic stem cell transplantation (HSCT) recipi-
ents: 2.26% (3.5% in allogeneic HSCT and 1.2% in autologous
HSCT). A significative increase in IA incidence was observed along
the study period (P=0.013), although this increase did not reach
signification if only proven cases were compared (P=0.058). Most
patients presented multiple risk factors for IA (87% more than 1,
and 47% more than 3). The most frequently described risk factor
was chemotherapy (90%), after by long-term neutropenia (90%),
and corticotherapy (75%). Main locations of the infection were
pulmonary (8 patients), cutaneous (3 patients) and intestinal
(3 patients). Six patients presented disseminated IA. Initial
response to treatment was 55%, although 3 of these cases had a
subsequent episode. Global antifungal response, at the end of the
follow-up, was 45%. IA-related mortality was 55%. Global
mortality was 90%. Only 2 patients (isolated cutaneous IA cases)
survived. Seven patients died due to their underlying malignant
disease without active fungal disease. Incidence of IA in oncology
children is increasing, and in adults. In our experience, IA is a
marker of poor outcome even for patients who initially respond to
antifungal treatment.
Key Words: aspergillosis, invasive fungal infection, mould,
filamentous fungi, voriconazole
(J Pediatr Hematol Oncol 2009;31:642–646)
I
suffering from systemic fungal infections. Among these
infections, filamentous fungi and particularly invasive
aspergillosis (IA) have an increasing role.1
Information about IA in children is scarce. Most
information about this age group is extracted from studies
focused on adult or at least general population.2–7Only a
few specific reports from pediatric units have been
mmunocompromised patients, especially those with pro-
longed and profound neutropenia, are prone to be
published, and most of them suggest that there may be
differences according to age in the clinical features and
outcome of IA.8–16
The rising incidence and mortality of systemic fungal
infections along the last decade entails for a more urgent
need of specific pediatric information.1,2
We present a retrospective analysis of IA in pediatric
patients at the Hospital Infantil Universitario Nin ˜ o Jesu ´ s.
Our purpose is firstly to describe the epidemiologic, clinical,
and microbiologic features of this disease in children.
Secondly, response to treatment and toxicities are also
described, as information of new antifungals in combina-
tion with amphotericin in children is also limited.17
MATERIALS AND METHODS
Population Selection
Cases of invasive filamentous fungal infection (IFFI)
were identified through a search in the discharge reports
and the Pathology and Microbiology reports of the
institution between January 1, 1996 and December 31,
2006.
Discharge reports of the Hospital Infantil Universitario
Nin ˜ o Jesu ´ s were selected when at least 1 of the diagnoses
was: aspergillosis, zygomicosis, and other opportunistic
mycoses (Alternaria, Dreschlera, and Fusarium). In pathol-
ogy reports a search was made for the following terms:
aspergillosis, Aspergillus fumigatus, and fungi. Microbiol-
ogy reports were investigated for mold isolations. Finally,
the enzyme linked immunosorbent assay for the detection
of galactomannan (GM) (Bio-Rad Platelia Aspergillus) was
performed in our unit since January 2004; therefore positive
results among years between 2004 and 2006 were also
included.
Sample Selection
Medical records of the patients identified were
thoroughly reviewed to assess IFFI diagnosis. Diagnostic
criteria for IA in immunocompromised patients have been
reported previously.18Only those patients with probable or
proven IFFI defined by the European Organization for
Research and Treatment of Cancer (EORTC) criteria were
selected for study (more information at www.eortc.be).18
Samples from patients diagnosed with IFFI by pathology
records were studied by Polymerase Chain Reaction (PCR)
assay for Aspergillus identification.19
Copyrightr2009 by Lippincott Williams & Wilkins
Received for publication November 14, 2008; accepted May 4, 2009.
From the *Oncohematologı´a Pedia ´ trica Hospital La Paz, Madrid;
wOncohematologı´a Pedia ´ trica Hospital Infantil Universitario Nin ˜ o
Jesu ´ s; and zMycology Reference Laboratory Centro Nacional de
Microbiologı´a, Instituto de Salud Carlos III.
Reprints: Julia ´ n Sevilla, MD, Oncohematologı´a Pedia ´ trica Hospital
Nin ˜ o Jesu ´ s, Mene ´ ndez Pelayo 65, Madrid 28009, Spain (e-mail:
jsevilla.hnjs@salud.madrid.org).
ORIGINAL ARTICLE
642|www.jpho-online.comJ Pediatr Hematol Oncol?Volume 31, Number 9, September 2009

Page 2

Data Collection and Analysis
Data from the medical records of the selected patients
were divided in the following categories: demographics, risk
factors, treatment, and outcome.
Demographical details analyzed were: sex, age at
diagnosis, underlying disease, type of infection (probable/
proven), method for diagnosis, date of diagnosis, localiza-
tion, and fungal species. Risk factors for IA considered
were: chemotherapy, neutropenia (neutrophil count below
1000?106/L), corticotherapy, HSCT, immunosuppressive
treatment, and graft versus host disease.
Regarding treatment and outcome, we reviewed the
drugs used, adverse events, and response, of all IA episodes
(primary and relapses). From each patient we recorded the
status at the end of the follow-up: alive or dead, cause of death
when applicable, and evolution of the underlying disease.
Incidence for hematology-oncology patients was re-
lated to admissions in the oncology-hematology ward and
to new diagnoses of malignant diseases, to compare our
results with those previously reported in other series.
However, risk factors and overall response and survival
were analyzed in all IA cases.
The series was divided in 2 consecutive groups to
compare the incidence between the first period and the last
5 years of the study period. A w2test was used to compare
incidence between these 2 periods.
Treatment Response
Response to therapy was evaluated as ‘‘complete
response’’ when all the symptoms and signs of fungal
infection disappeared. ‘‘Partial response’’ was considered
when symptoms disappeared and signs clearly improved.
‘‘Stable disease’’ was considered when treatment achieved
no progression of the fungal infection. And ‘‘no response’’,
when the fungal infection progressed.
RESULTS
Diagnosis (Table 1)
Within the 11-year period between January 1996 and
December 2006, 20 patients with at least 1 probable or
proven IFFI episode were found among the clinical records
of our institution.
There were 12 patients diagnosed of proven IFFI. Five
were identified as proven IA by positive culture from a
sample obtained by a sterile procedure (4 from affected
tissue and 1 from blood culture). Contamination of culture
samples was excluded by repeated studies. Seven patients
were identified by pathology findings: 5 biopsies and 2
necropsies. All positive pathology findings (septated hyphae
in tissue samples) were confirmed as IA by PCR assay for
Aspergillus.19,20,21In all proven IFFI cases A. fumigatus was
amplified by PCR. A. flavus was also amplified by PCR, in 2
cases with A. flavus isolation in culture.
Eight patients met EORTC criteria for probable IA.18
Diagnosis was supported in 6 patients by positive serum
GM tests. Aspergillus detection in bronchial aspiration
culture provided diagnosis in one patient, and in another
one by positive cytology (Table 1). The possible bias on
incidence of the late availability of GM tests (since 2004) is
discussed further below. Aspergillus spp was therefore
isolated in culture in 6 patients (5 proven and 1 probable
be IA). They were identified as A. flavus in 2 of them
(previously discussed), and A. fumigatus in 1. Species
identification of the 3 remaining isolates was not achieved.
TABLE 1. Diagnosis and Outcome
Diagnosis
Case TypeMethod DateLocationSpeciesResponse TreatmentUD Status ExitusCause
1
2
3
Proven
Proven
Proven
Necropsy
Culture
Culture
2001
2005
2004
Disseminated
Lung
Disseminated
A-No
A-No
A-Yes
B-No
A-Yes
B-Yes
A-No
A-Yes
LAB
LAB, V, C Remission
LAB, V, C MRD
LAB, V, C
LAB, V, C Relapse
V, C
LAB
LAB
LAB, V
LAB, C
LAB
LAB, V, C Remission
Remission Yes
Yes
Yes
IA
IA
IA
A. spp.
A. flavus
4Probable GMN2005Lung YesUD
5
6
7
8
9
Proven
Proven
Proven
Proven
Proven
Probable GMN
Necropsy
Biopsy
Culture
Biopsy
Biopsy
2004
1999
2002
2004
2006
2004
Intestinal
Skin
Disseminated
Lung
Skin
Intestinal
Active
Relapse
Relapse
Remission
Remission
Yes
Yes
Yes
Yes
No
Yes
IA
UD
UD
IA
IA
Toxicity
A. fumigatus A-Yes
A-No
A-Yes
A-Yes10
Graft failure
Remission11
12
13
14
15
16
Probable Cytology (BA)
ProvenCulture
Probable GMN
Proven Biopsy
Probable GMN
Proven Biopsy
1996
2003
2005
2002
2005
2006
Lung
Disseminated
Disseminated
Intestinal
Lung
Disseminated
A-No
A-Yes
A-No
A-No
A-Yes
A-No
AB
LAB, V, C Remission
LAB, V, C Remission
LAB
V
LAB, V, C Remission
Yes
Yes
Yes
Yes
Yes
Yes
IA
Toxicity
IA
IA
UD
IA
A. spp.
Remission
Relapse
Graft failure
Remission
Relapse
17
18
19
Proven
Probable GMN
Probable GMN
Culture2004
2005
2006
Skin
Lung
Lung
A. flavusA-Yes
A-No
A-Yes B-Yes LAB, V, C Remission
V, C
C-No V, C
A-YesLAB
LAB, V
LAB
No
Yes
Yes
IA
UD
IA
20 Probable Culture (BA)2002Lung A. spp.Active YesUD
AB indicates amphotericin B; C, caspofungin; IA, invasive aspergillosis; LAB, lypidic/lyposomal amphotericin B; UD, underlying disease; V, voriconazole.
J Pediatr Hematol Oncol?Volume 31, Number 9, September 2009Invasive Aspergillosis in Children
r2009 Lippincott Williams & Wilkins
www.jpho-online.com|643

Page 3

Demographics and Incidence
Demographic characteristics of the patients are shown
in Table 2. Ages at the time of diagnosis ranged from 1 to
19 years old, median 10 years. Distribution by sex was 1:1.
Nineteen out of the 20 identified patients with IA suffered
from malignant diseases.
During the study period 966 children were diagnosed in
our Hospital of malignant diseases. There were 8717 admis-
sions in the Oncology ward. IA incidence in this group was 218
cases per 1000 admissions, or 1.96% patients. There were 243
patients diagnosed with acute lymphoblastic leukemia (ALL),
and 63 with acute nonlymphoblastic leukemia/myelodysplasia/
myeloproliferative diseases. These data yield an IA incidence
of 2.88% for patients diagnosed with ALL (7/243), and 11.1%
for those diagnosed with acute nonlymphoblastic leukemia/
myelodysplasia/myeloproliferative diseases (7/63). Global IA
incidence in patients with hematologic malignancies was
4.57% (14/306).
Among HSCT recipients, IA incidence increased
to 2262 cases per 1000 HSCT. It was higher in the
allogeneic setting (35.2 cases per 1000 allogeneic HSCT),
than in autologous procedures (12.3 per 1000 autologous
HSCT).
An increase of accumulated annual incidence was
observed along the 11 years researched. Between 1996 and
2001 there were 3 IA cases, with an incidence of 0.61 per
1000 admissions. In the second period, there were 16 cases,
with an incidence of 4.19 per 1000 admissions. If this
analysis is performed considering IA per new oncology
patients, the incidence is 0.63% (3 cases in 475 patients
between 1996 and 2001), and 3.25% (16 cases in 491
patients between 2002 and 2006). The difference between
the first and the second period is statistically significative
both considering cases per admissions (P=0.0138) and
cases per patients (P<0.001).
Since 2004, diagnosis of probable IA has improved
owing to routine GM detection. To avoid this possible bias
in incidence, we have compared incidences of proven IA.
Accumulated incidence was from 0.52 cases per 1000
admissions between 1996 and 2001 period to 2.04 per
1000 admissions between 2002 and 2006. The difference
found, even with a small sample, is near signification
(P=0.0580), showing a clear positive tendency.
Risk Factors (Table 2)
Almost all patients (95%) had a compromised immune
function owing to underlying oncohematologic diseases.
Eighteen patients (90%) had received chemotherapy before
IA was suspected, and 16 (80%) received corticosteroids.
Eleven patients (55%) had relapsing malignancies. Ten
patients (50%) were treated with immunomodulating
agents. In general, no chemoprophylaxis for fungal infec-
tion is performed in our oncology patients, except those
undergoing hematopoietic transplantation, in which pro-
phylaxis with fluconazole is routinely administered.
Thirteen patients (65%) were HSCT recipients. These
patients were isolated in rooms with high efficiency
particulate air filters. Six had been diagnosed with graft
versus host disease.
Owing to chemotherapy, 9 patients (45%) presented
neutropenia at the time of diagnosis with a median duration
of 24 days (range: 9 to 91d). Moreover, 18 patients (90%)
had suffered from long-term neutropenia (10 or more days)
in the 3 months before diagnosis.
Hospital stay at the time of IA diagnosis ranged from
14 to 105 days (median 34d). All patients were receiving
TABLE 2. Demographics and Risk Factors
TreatmentNeutropenia
Case Sex Year
Age
(y)
Underlying
Condition ChemotherapyHSCTSteroids ISDiagnosis
?3mo
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
No
Yes
GVHD
1
2
3
4
5
6
7
8
9
Male
Female
Female
Male
Male
Female
Male
Male
Female
Female
2001
2005
2004
2005
2004
1999
2002
2004
2005
2004
19
10
11
15
13
10
6
8
2
14
AML
AML
AML
AML
OS
CML Ph+
ALL
ALL
Esophagitis
ES
Yes
Yes
Yes
Yes
Yes
Yes
Yes
Yes
No
Yes
Auto and Allo Yes
No
No
No
No
Allo (2)
No
Auto
No
Auto
CsA, MP
No
No
No
No
CsA
No
No
No
CsA, Et,
Rit
CsA, Mtx,
ATG
No
MP, CsA
No
83d
9d
23d
No
91d
No
24d
No
No
L
NA
NA
NA
NA
S, L, I
NA
NA
NA
NA
No
Yes
No
Yes
Yes
Yes
Yes
No
Yes 10
11Male19961 ALL YesAlloYes NoYes S, L, I
12
13
Female
Female
2003
2005
14
9
ALL
AML
Yes
Yes
No
Auto and Allo Yes
YesNo
No
Yes
Yes
NA
Chronic
diffuse
NA
S
L
NA
No
NA
NA
14
15
16
17
18
19
20
Male
Male
Female
Male
Male
Female
Female
2002
2005
2005
2004
2005
2005
2002
19
7
7
4
4
13
13
Wilms
ALL
ALL
ALL
AML
Biph AL
GBL
Yes
Yes
Yes
Yes
Yes
Yes
No
Auto (2)
Auto and Allo Yes
Allo
No
Allo
No
No
YesNo
Rit, FK
CsA, Mtx
No
CsA, Mtx
No
No
No
No
14d
34d
18d
31d
No
Yes
Yes
Yes
Yes
Yes
Yes
No
Yes
Yes
Yes
No
Yes
AL indicates acute leukemia; ALL, acute lymphoblastic leukemia; Allo, allogeneic HSCT; AML, acute myeloblastic leukemia; ATG, anti-tymocite
globuline; Auto, autologous HSCT; Biph AL, biphenotypical acute leukemia; CML, chronic myeloid leukemia; CsA, cyclosporin A; ES, Ewing sarcoma; Et,
etanercept; FK, tacrolimus; GBL, glioblastoma; GVHD, graft versus host disease (I, intestinal; L, liver; S, skin); HSCT, hematopoietic stem cell
transplantation; MP, mercaptopurine; Mtx, methotrexate; NA, nonapplicable; OS, osteosarcoma; Rit, rituximab.
Rubio et al
J Pediatr Hematol Oncol?Volume 31, Number 9, September 2009
644|www.jpho-online.com
r2009 Lippincott Williams & Wilkins

Page 4

treatment with broad spectrum antibiotics when the
diagnosis was made. Although chemoprophylaxis for
Aspergillus infection was not performed routinely in the
Oncology ward, at the time of IA diagnosis, 15 patients
(75%) had already started empirical Amphotericin B
treatment at a dosage of at least 1mg/kg/day.
Clinical Presentation
Three patients were diagnosed of cutaneous IA with
negative extension studies. All of them were successfully
treated with lyposomal B-amphotericin (LAB). One of
them died due to a relapse of her underlying disease.
Eight patients were diagnosed of pulmonary IA
(2 proven and 6 probable). In those to whom computed
tomography could be performed (4), we found IA-specific
images (nodules, halo sign, and cavitation). The main
reason why the other patients did not have a computed
tomography was severe clinical condition that did not allow
performing the study. Six of those patients were treated
with combined therapy adding voriconazole or caspofungin
to LAB. Only 2 patients received monotherapy with
conventional B amphotericin, and with voriconazole,
respectively. It is noticeable the frequency of IA relapse in
this group. Two patients cured from the infection,
presented a second IA episode, and 1 of them even a third
one. All patients with pulmonary IA died, although in 4 of
them there were no signs of IA at the time of exitus.
There were 3 patients with intestinal IA diagnosis.
Two of them presented with acute sepsis with abdominal
participation and required emergency surgery. Both died in
the subsequent 24 hours. A third patient was treated with 3
drugs (LAB, voriconazole, and caspofungin). She died 6
months after diagnosis with no IA signs.
Finally, 6 patients met EORTC criteria for dissemi-
nated IA (5 probable, 1 proven). All of them had the lungs
as one of the organs affected. All of them were receiving
LAB before diagnosis; 5 of them received combined therapy
thereafter. There were no survivors in this group. Four of
them died as a direct consequence of IA.
Global Response to Treatment (Table 1)
Initial response to therapy was evaluated as positive
(complete response, partial response) in 11 of the 20 patients
in their first infectious episode (55%). Most patients received
LAB in monotherapy for treatment. Seventeen patients
received amphotericin B, with a mean total duration of
48 days (7 to 159). Sixteen of these patients received the
liposomal formulation; 2 of these also received the lipid
formulation at some point of the treatment. Only 1 patient
received conventional amphotericin B. Dosage of liposomal
amphotericin B or the lipid formulation ranged from 1 to
5mg/kg/day in 1 daily doses. Switching to combined therapy
with caspofungin and voriconazole was the most frequent
alternative (7 cases). Three patients (15%) presented sub-
sequent IA episodes. Of these, one did not respond to
treatment, another one responded to combined therapy with
caspofungin and LAB, and the remaining one responded
initially to the same treatment but presented a third episode
without response. Nine children (45%) had no signs of
Aspergillus infection at the end of the follow-up.
Global mortality in our series is 90% (18/20). The
survivors were 2 localized cutaneous IA cases. Ten patients
(55.6%) died as a direct consequence of IA. Eight patients
(44.4%) died because of their underlying disease: relapse
(6 patients) or treatment toxicity (2 patients). Of these
8 patients, only 1 had an active IA when he died, although
the cause of death was a relapse and not the infection.
Survival in patients who initially responded to IA treatment
was only 22%.
Four patients (20%) had their treatment modified
because of adverse drug events. Three of these patients
received combined therapy with 3 drugs (voriconazole,
LAB, and caspofungin). We recorded 2 episodes of renal
toxicity owing to LAB, 1 episode of visual disturbances
caused by voriconazole, and 1 episode of liver toxicity
associated to caspofungin. In all these cases toxicities
reverted with temporary withdrawal of the responsible
drug, which could be reintroduced later without any new
adverse event recorded.
DISCUSSION
In the last years, an increasing incidence of IA has
been reported in oncology patients. This is applicable,
based on our own data, to oncology children and for adults.
We have described in our cases a significant IA incidence
increase (0.63% vs. 3.25%, P<0.001). The practice of GM
detection since 2004 is responsible for most of the probable
IA diagnoses and might be the cause of part of this increase.
Our sample is not large enough to prove this increase in
proven IA, although we have found a clear tendency
towards a real increase incidence.
PCR-based diagnosis has shown important results for
detection of IA.19,20,21In our patients PCR analysis has
been of great use to confirm the isolation of Aspergillus in
affected tissues and to clarify the species.
Our results are comparable to those found in the
literature. Abbasi et al found an incidence similar to ours in
a study between 1965 and 1999.8Zaoutis et al, in 2006,
reported 666 cases of IA found among 152.231 immuno-
compromised children, yielding an incidence much lower
than ours. This epidemiologic review was extracted from a
poblational database, which does not specify diagnosis
criteria or type of infection. This makes it difficult to
compare Zaoutis data with ours. Other pediatric series
recently published describe the characteristics of the
infection in children, but do not report IA incidence.10,14
Adult series report higher incidences.2
We have found some disparity in literature concerning
only patients with hematologic malignancies. Our incidence
between 1996 and 2006 is almost double than reported by
Crassard et al between years 1986 and 2000 (1.5% in ALL
patients and 5.35% in ANLL).14This is consistent with an
increase in incidence in late years: only 1 of our 14 IA cases
in acute leukemia was diagnosed before 2001. In the other
hand, Groll et al published in 1999 a prospective study,
finding a global IA incidence of 6.8% in patients with
hematologic malignancies (3.79% for ALL and 28% for
ANLL), clearly higher than ours.15
Pediatric HSCT recipients are considered a group of
high risk for IA. Several reports have found a superior
incidence in these patients. A recent review of IA in
pediatric HSCT recipients set the incidence of probable and
proven IA in 3.4%.11Hovi et al reported in a retrospective
pediatric study an incidence of 5% (8% allogeneic and
3% autologous),12but found lower figures in a more recent
prospective study (1.02%).16Our data are similar to these
pediatric series with a global incidence of 2.26% IA in
HSCT (3.52% allogeneic; 1.23% autologous). Adult series
offer higher incidences in both autologous (3.8%) and
allogeneic (14% to 17%) settings.5–7
J Pediatr Hematol Oncol?Volume 31, Number 9, September 2009Invasive Aspergillosis in Children
r2009 Lippincott Williams & Wilkins
www.jpho-online.com|645

Page 5

There are certain host characteristics frequently
associated with IA. In our series these factors (chemo-
therapy, neutropenia, and steroids) presented in similar
percentages as in the literature.7It is of great interest the
accumulation of risk factors in each patient. Although in
EORTC recommendations only 1 host criterion is required
for IA diagnosis,1884% of our patients met more than 1,
and 47% more than 3. The presence of a single risk factor
does not seem to be enough to make pediatric patients
prone to Aspergillus infection.
Response to treatment in children diagnosed with IA
is poor. Several series yield responses between 25% and
64%.14–15In our series, 55% patients presented an initial
positive response, and 45% had no signs of IA at the end of
the follow-up. Use of new drugs and/or combined therapy
could improve the response. In a Canadian study that reports
a response rate of 25%, amphotericin-B monotherapy was the
rule, whereas 55% of our patients were treated with combined
therapy.10Crassard et al, who report a similar response to
ours, also used more than 1 drug in 75% patients.14We lack
enough clinical data to recommend the use of combined
treatment as first line of therapy. As reported in Table 1 9 IA
episodes were treated with a single drug, with 4 positive
responses (44%); and 15 episodes were treated with multiple
drugs, with 9 responses (60%). In most cases it was used in
patients already treated with amphotericin-B when they were
diagnosed with the fungal infection.17Toxicity is similar or
lower than reported in adults.22–24
Global mortality (owing or not owing directly to IA)
in our patients was 90%. Three significant pediatric studies
offer mortality rates between 74% and 87%.8,10,14Mortal-
ity due directly to IA was 55%. The severity of the
underlying disease in patients diagnosed with IA (11
relapses and 13 HSCT among 20 patients) not only
enhances the risk for IA infection, but it is also an
independent marker of poor outcome. Thus, survival is
low even in patients cured from the IA.9One exception is
isolated cutaneous affectation, with a better prognosis both
in our series (response to treatment 100% and overall
survival 66%), and in the literature.10
In conclusion, IA in children has a lower incidence
than in adults, even though, according to the literature and
our findings, it has increased in late years. Risk factors for
developing IA in pediatric population are similar to those
described in adults, although in these patients association of
multiple risk factors is the rule. Children with IA have a
very unfavorable prognosis. High mortality is mainly owing
to progression of the infection, but the severe underlying
malignant disease of these patients plays an important role
in the prognosis.
REFERENCES
1. McNeil MM, Nash SL, Hejjeh RA, et al. Trends in mortality
due to invasive mycotic diseases in the United States, 1980 to
1997. Clin Infect Dis. 2001;33:641–647.
2. Sanz MA, Jarque I, Salavert M, et al. Epidemiology of invasive
fungal infections due to Aspergillus spp and Zygomycetes. Clin
Microbiol Infect. 2006;12:2–6.
3. Cornet M, Fleury L, Maslo C, et al. Epidemiology of invasive
aspergillosis in France: a six-year multicenter survey in the
greater Paris area. J Hosp Infect. 2002;51:288–296.
4. Yuen KY, Woo PCY, Ip MSM, et al. Stage-specific
manifestations of mold infections in bone marrow transplant
recipients: risk factors and clinical significance of positive
concentrated smears. Clin Infect Dis. 1997;25:37–42.
5. Kruger W, Russmann B, Kroger N, et al. Early infections in
patients undergoing bone marrow or blood stem cell trans-
plantation: a 7-year single centre investigation of 409 cases.
Bone Marrow Transplant. 1999;23:589–597.
6. Fukuda T, Boeckh M, Cater RA, et al. Risks and outcomes of
invasive fungal infections in recipients of allogeneic hematopoie-
tic stem cell transplants after nonmyeloablative conditioning.
Blood. 2003;102:827–833.
7. Marr KA, Carter RA, Crippa F, et al. Epidemiology and
outcome of mould infections in hematopoietic stem cell
transplant recipients. Clin Infect Dis. 2002;34:909–917.
8. Abbasi S, Shenep JL, Hughes WT, et al. Aspergillosis in
children with cancer: a 34-year experience. Clin Infect Dis.
1999;29:1210–1219.
9. Steinbach WJ. Pediatric aspergillosis. Disease and treatment
differences in children. Pediatr Infect Dis J. 2005;24:358–364.
10. Walmsley S, Devi S, King S, et al. Invasive Aspergillus
infections in a pediatric hospital: a ten-year review. Pediatr
Infect Dis J. 1993;12:673–682.
11. Kobayashi R, Kaneda M, Sato T, et al. Evaluation of risk
factors for invasive fungal infection after allogeneic stem cell
transplantation in pediatric patients. J Pediatr Hematol Oncol.
2007;29:786–791.
12. Hovi L, Saarinen-Pihkala UM, Vettenranta K, et al. Invasive
fungal infections in pediatric bone marrow transplant recipi-
ents: single center experience of 10 years. Bone Marrow
Transplant. 2000;26:999–1004.
13. Zaoutis TE, Heydon K, Chu JH, et al. Epidemiology,
Outcomes, and Costs of Invasive Aspergillosis in Immuno-
compromised Children in the United States, 2000. Pediatrics.
2006;117:711–716.
14. Crassard N, Hadden H, Piens MA, et al. Invasive aspergillosis
in a paediatric haematology department: a 15-year review.
Mycoses. 2008;51:109–116.
15. Groll AH, Kurz M, Scheneider W, et al. Five-year-survey of
invasive aspergillosis in paediatric cancer centre. Epidemiology,
management and long-term survival. Mycoses. 1999;42:431–442.
16. Hovi L, Saxen H, Saarinen-Pihkala UM, et al. Prevention and
Monitoring of Invasive Fungal Infections in Pediatric Patients
with Cancer and Hematologic Disorders. Pediatr Blood
Cancer. 2007;48:28–34.
17. Walsh TJ, Anaissie EJ, Denning DW, et al. Treatment
of aspergillosis: clinical practice guidelines of the Infec-
tious Diseases Society of America. Clin Infect Dis. 2008;46:
327–360.
18. Ascioglu S, Rex JH, de Pauw B, et al. Defining opportunistic
invasive fungal infections in immunocompromised patients
with cancer and hematopoietic stem cell transplants: an
international consensus. Clin Infect Dis. 2002;34:7–14.
19. Cuenca-Estrella M, Meije Y, Diaz-Pedroche C, et al. Value
of serial quantification of fungal DNA by a real-time PCR-
based technique for early diagnosis of invasive Aspergillosis in
patientswith febrileneutropenia.
2009;47:379–384.
20. White PL, Linton CJ, Perry MD, et al. The evolution and
evaluation of a whole blood polymerase chain reaction assay for
the detection of invasive aspergillosis in hematology patients in a
routine clinical setting. Clin Infect Dis. 2006;42:479–486.
21. Verweij PE, Klont RR, Donnelly JP. Validating PCR for
detecting invasive aspergillosis. Br J Haematol. 2004;127:
235–236.
22. Eiden C, Peyrie ` re H, Cociglio M, et al. Adverse effects of
voriconazole: analysis of the French Pharmacovigilance
Database. Ann Pharmacother. 2007;41:755–763.
23. Herbrecht R, Denning DW, Patterson TF, et al. Voriconazole
versus amphotericin B for primary therapy of invasive
aspergillosis. N Engl N Med. 2002;347:408–415.
24. Walsh TJ, Teppler H, Donowitz GR, et al. Caspofungin versus
liposomal amphotericin B for empirical antifungal therapy in
patients with persistent fever and neutropenia. N Engl J Med.
2004;351:1391–1402.
J ClinMicrobiol.
Rubio et al
J Pediatr Hematol Oncol?Volume 31, Number 9, September 2009
646|www.jpho-online.com
r2009 Lippincott Williams & Wilkins