International Journal of Pediatric Otorhinolaryngology 106 (2018)

91–95

Contents lists available at ScienceDirect

International Journal of Pediatric Otorhinolaryngology

journal homepage: www.elsevier.com/locate/ijporl

Microbiology and antibiotic therapy of subperiosteal orbital abscess in

children with acute ethmoiditis
A. Couderta,b,∗ , S. Ayari-Khalfallaha,b, P. Suya,b, E. Truya,b,c,d
a
Service d’ORL Pédiatrique, Hôpital Femme Mère Enfants, Centre Hospitalier et Universitaire, Lyon, France
b
Service d’ORL, Hôpital Edouard Herriot, Centre Hospitalier et Universitaire, Lyon, France
c
Université de Lyon, Lyon, France
d
INSERM, U1028, CNRS, UMR5292, Lyon Neuroscience Research Center, IMPACT Team, Lyon, France

AR T IC L E I N F O AB S T R AC T

Keywords: Objective: The objective of this study was to investigate the microbiological cultures and the management of
Acute sinusitis acute ethmoiditis complicated by subperiosteal orbital abscess (SPOA) in a pediatric population.
Subperiosteal orbital abscess Methods: The medical records of children under 18 years old was performed in a tertiary referral pediatric center
Antibiotics from January 2009 to April 2017. Clinical examination, computed tomography scans, medical and surgical
Microbiolog treatments were reviewed and compared to other studies in literature.
y Pediatrics Results: One hundred and twenty-nine children were hospitalized for acute ethmoiditis. Among them, forty eight
were complicated by SPOA. The mean age of these children were 7 years (range 10 months–16 years). Thirtyfour
underwent surgical drainage; for the others the medical treatment was sufficient. Microbiological samples were
obtained during the surgical intervention and were contributive in 91% of cases. Streptococcus spp was the most
frequently encountered bacteria (60% of cases). We also found anaerobic bacteria (12%), and Staphylococcus
aureus (12%). 94% of children received two intravenous antibiotics (a third-generation cephalosporin and
metronidazole) for a mean duration of four days. Then the oral treatment was based on amoxicillin-clavulanate
during about 8.5 days. All children were cured without sequelae.
Conclusions: For five years Streptococcus milleri, Staphylococcus spp and anaerobic bacteria are on the rise in acute
ethmoiditis complicated by SPOA. That is why antibiotics must be adapted to these bacteria even in children
under ten years old.

1. Introduction
During the first life decade, acute sinusitis and especially ethmoi-
ditis frequently occur. It accounts for 21% of pediatric antibiotic pre-
scriptions [1]. Orbital infection is the most frequent complication of
ethmoiditis and can arise 91% of sinusitis complications in children [1].
The spread of infection from the ethmoid sinus to the periorbital space
can occur by eroding the lamina papyracea or through the hemato-
genous dissemination [2,3]. Sometimes, this complication can be fa-
vored by a congenital dehiscence of the lamina papyracea. The spread
of this infection can lead to a subperiosteal orbital abscess (SPOA). The
progression of SPOA may result in serious complications such as cere-
bral abscess, cavernous sinus thrombosis and permanent visual loss
[4,5].
Contrast-enhanced paranasal sinus computed tomographic scan
(CT-scan) is a very useful method to diagnose and to classify orbital
complications in Chandler's classification [6]. SPOA seems to represent
the most common orbital complication of sinusitis in children [7] and
requires active management. The optimal management of SPOA is still
controversial. Indeed, the choice of the treatment (medical treatment
versus surgery) is central to the debate as well as the type of surgical
approach in case of surgery. Some physicians favor immediate surgical
drainage while others recommend initial medical treatment keeping
surgery for non-responders [8,9]. It seems that smaller abscesses in
young children [8,10–12] are suitable to medical treatment with close
observation [13,14]. Oxford and McClay [2] reported that older chil-
dren with SPOA were also successfully managed with medical therapy.
In all cases, antibiotic therapy is necessary, first probabilistic and then
adapted to the identified germ if bacteriological samples were carried
out.
The main objectives of the present study were to review all children
who were referred to our center for SPOA secondary to acute ethmoi-
ditis and to assess the current bacteriology of drained SPOA. Our results
were compared to those of past studies. Finally, the impact on the

∗
Corresponding author. Service d’ORL Pédiatrique, Hôpital Femme Mère Enfant, 59 Boulevard Pinel, 69677 BRON, France.
E-mail addresses: aurelie.coudert@chu-lyon.fr (A. Coudert), sonia.ayari-khalfallah@chu-lyon.fr (S. Ayari-Khalfallah), paul.suy@chu-lyon.fr (P. Suy), eric.truy@chu-
lyon.fr (E. Truy).

https://doi.org/10.1016/j.ijporl.2018.01.021
Received 4 November 2017; Received in revised form 8 January 2018; Accepted 10 January 2018
Available online 02 February 2018
0165-5876/ © 2018 Elsevier B.V. All rights reserved.
A. Coudert et International Journal of Pediatric Otorhinolaryngology 106 (2018)
al. 91–95

antibiotic use is discussed.

2. Material and methods

This retrospective study was performed in a tertiary referral pe-

diatric center from January 2009 to April 2017. All children under 18
years old were included in the study if they presented an evident
clinical ethmoiditis, with a sinus CT-scan showing a SPOA (stage III of
Chandler's classification). In the CT-scan, SPOA was defined as a central
low-density region with ring enhancement in the orbital region. The
exclusion criteria were a chronic rhinosinusitis, and/or an intraorbital
abscess on the CT-scan.
Fig. 1. Evaluation and treatment of 129 children with acute ethmoiditis.
Charts were evaluated for age, gender, physical exam findings (with
an oculomotor exam and a neurological assessment), CT-scans, surgical
procedure, culture results, antibiotic treatment and follow up. The CT- More than 50% of children with an operated SPOA had at least one
scans were reviewed for SPOA width and length, and collection in si- ophthalmologic trouble at the beginning. The most constant sign was
nuses. Our surgery indications, based on previous study [15,16], were ophthalmoplegia (71% cases). In 50% of cases, we recorded proptosis
the following ones: (1) abscess width more than 5 mm or extended to and/or diplopia. The mean age of operated children with an oculomotor
the optic nerve, (2) oculomotor disorder, (3) absence of symptoms dysfunction was nine years and six months whereas for the others
improvement after 48–72 h of intravenous antibiotics, (4) severe clin- without oculomotor trouble it was four years and six months.
ical complications such as epidural empyema, loss of visual acuity or On the CT-scan, the mean width of SPOA was 6 mm (range
cavernous sinus thrombophlebitis. 3–12 mm). Twenty-one children were treated with an external ap-
The surgical therapy was recorded and separated into three cate- proach, six with a transnasal endoscopic technique and seven with a
gories: an external approach (EA), a transnasal endoscopic approach combined approach. An example of a young child with a SPOA drained
(TEA) and a combined external and transnasal approach (CA) to drain by an external approach is given in Fig. 2.
the SPOA. Since our department was the ENT emergency center for the Among the operated children, seventeen (50%) had a previous an-
Rhone department, the surgery was performed by the ENT surgeon on tibiotic treatment before hospitalization (amoxicillin-clavulanate, or
call who could be a fellow. The choice between the EA, TEA or the CA pristinamycin, or a third-generation cephalosporin, or josamycin). The
was decided by the surgeon on call, taking into consideration the ab- average duration of this antimicrobial therapy was four days. Only two
scess characteristics, the anatomical conditions but also his personal children had a sterile surgical sample despite a first medical therapy
training and skills in pediatric transnasal endoscopic surgery. during 48 h.
During surgery, sinus secretions and the pus from the SPOA were
systematically sampled and carried out, immediately to the laboratory,
3.1. Microbiologic cultures (Table 1)
in a plastic tube containing a sterile swab. The way of detecting and
identifying bacteria was based on culturing using diff erent culture
Cultures were obtained by pus sample during the surgical inter-
media with control of the nutrients and culture conditions (tempera-
vention. Samples were contributive in 91% of cases. The most frequent
ture, air supply, O2, light, blood, pH …), enumeration and isolation of
encountered bacteria was Streptococcus spp which was found in more
presumptive colonies with study of phenotypic characteristics com-
than 60% of cases. Furthermore, the other identified species were the
pleted if needed by genotypic characteristics. The phenotypic method
anaerobic bacteria (12%), Staphylococcus aureus (12%) and Haemophilus
included biotyping (growth requirement, environmental conditions,
influenzae (9%). There was only one case of methicillin-resistant
antibiotic resistance, cell morphology …), and identification by mass
Staphylococcus aureus (MRSA). The age of children with anaerobic
spectrometry. Mass spectrometry is an analytical technique in which
cultures ranged from 3 years to 15 years. All children older than nine
chemical compounds are ionized into charged molecules and ratio of
years had a Streptococcus intermedius (which belongs to milleri group) in
their mass to charge (m/z) is measured. Identification of microbes is
their culture.
done by either comparing the peptide mass fingerprint (PMF) of un-
After surgery, thirty-two children received two intravenous anti-
known organism with the PMFs contained in the database or by
biotics (third-generation cephalosporin and metronidazole) for a mean
matching the masses of biomarkers of unknown organism with the
duration of four days. Only two children had a prolonged intravenous
proteome database [17]. The genotypic method used molecular tech-
antimicrobial treatment. The first one had a MRSA, he received in-
niques to identify bacteria by doing DNA or RNA analysis of the bac-
travenous clindamycin and vancomycin during ten days. The second
terium's genome. The system used in our center was the Vitek®MS
one had a subdural empyema which required intravenous third-gen-
system (by bioMérieux France). If the mass spectrophotometry was not
eration cephalosporin and metronidazole for fifteen days. The average
contributory, a universal PCR detection method was used.
hospital stay was 6.5 days (range 3–16 days).
The Ethics Committee of the Hospices Civils de Lyon in France ap-
Then after hospitalization, each child had an oral antibiotic during
proved the study (Number 17-145) and all patients gave written in-
about 8.5 days (from 7 to 15 days). The most frequently used antibiotic
formed consent.
was amoxicillin-clavulanate (76% cases). Because of allergy, some
children had pristinamycin, or clindamycin and metronidazole. All
3. Results
children were cured without sequelae at the end of the antibiotic
treatment. Moreover we did not notice any recurrence of ethmoiditis
One hundred and twenty-nine children were hospitalized in our
with our follow-up. The mean follow-up length was 85.8 days after the
center for acute ethmoiditis between 2009 and 2017. Among them,
surgery, with a range of 9 days to 5.3 years.
forty eight (37%) were complicated by SPOA. The age of these children
with SPOA ranged from 4 months to 16 years, with a mean age of 7
years. From these SPOA, thirty-four (71%) underwent surgical drai- 4. Discussion
nage. For the others, a medical treatment was sufficient. Clinical data
and patient characteristics are summarized in Fig. 1. Many studies have been conducted over the past decades on mi-
Before surgery, each child underwent an oculomotor examination. crobiology of SPOA complicating sinusitis. In a recent literature review

92
A. Coudert et
al.
by Brook in 2016, the rate of SPOA due to sinusitis was estimated to
occur in about 5% of hospitalized patients [18]. Bacteriological iden-
tification of SPOA is a fundamental factor in order to adapt the ap-
propriate intravenous antibiotic followed by an oral treatment. One of
the most relevant microbiological samples comes from direct sinus as-
piration or by drainage of a subperiosteal orbital abscess [19]. Other
sampling, like nasal aspiration, can easily lead to bacteriological con-
tamination and is not advisable.
Since the last ten years, few studies determined microorganism
species responsible for complicated acute ethmoiditis in the pediatric
population. Our study is part of the largest studies in the literature that
specifically focused on the bacteriology of operated subperiosteal ab-
scess in children. In Table 1, we compared our results with micro-
biological culture results of SPOA in other studies [1,14,20–25]. In our
study, we found three main bacteria: Streptococcus (60%), Staphylococus
aureus (12%) and anaerobic bacteria (12%). This trend is not totally
consistent with other discussed articles. In fact, some authors found in
their studies a majority of Staphylococcus spp (47% for Pena et al., 41%
for Eviatar et al., and 39% for Huang et al.). This diff erence could be
due to the diversity of the bacterial ecology between Europe and other
continents. Moreover, it is well known that, in general population, S.
aureus colonizes about 25% of nasal vestibules [26]. Our low rate of
Staphylococcus spp can be explained by our direct surgical sampling in
the abscess that allowed to avoid contamination [27,28].
In all combined studies, the two main microorganisms were
Staphylococcus aureus (20%) and Streptococcus milleri group (11%). The
S. milleri group (also called S. anginosus group) is part of oropharyngeal
flora and includes three diff erent species: Streptococcus anginosus,
Streptococcus intermidius, and Streptococcus constellatus. These com-
mensal bacteria can be virulent and lead to profound abscess formation.
This microbiological evolution is a direct consequence of vaccina-
tions against Streptococcus pneumoniae and Haemophilus influenzae. As a
matter of fact before the seven-valent pneumococcal conjugate vaccine,
invasive S. pneumoniae was widely spread in head and neck infections.
Since generalized vaccination other pathogens have been amplified
such as S. milleri and S. aureus [20–22].
Anaerobic bacteria represent 7.4% of pathogens in all literature
studies. As Streptococcus milleri group, anaerobic group belongs to or-
opharyngeal flora. They included Fusobacterium spp, Peptostreptococcus
spp, Propionibacterium spp and gram negative bacilli (Prevotella,
Bacteroides spp). They are probably underestimated because cultures
are specific and they can be mixed with other oral flora, especially with
93
International Journal of Pediatric Otorhinolaryngology 106 (2018)
91–95
Fig. 2. Acute ethmoiditis complicated by
SPOA. A: before surgery/B and C: axial and
coronal CT scan showing a left medial
SPOA/D: during ex-
ternal surgery/E: after surgery with a drain.
a swab sampling [29]. Contrary to other studies, we found a high
percentage of anaerobic bacteria (12%) because we systematically used
diff erent culture media which allowed identification of even aerobic
and anaerobic bacteria. Moreover, we noticed few sterile sampling (8%)
contrary to the high rate in literature (15.5%).
According to many authors, children under ten years are more likely
to be infected by Streptococcus pneumoniae or Staphylococcus aureus
whereas children over ten-fifteen years are rather contaminated by
polymicrobial pathogens [12,30]. Furthermore, after eight years,
anaerobic bacteria are more frequently encountered because of dental
infections [29]. Therefore, antibiotic treatment must be appropriate to
target these bacteria. Nevertheless in our study, we managed to find a
lot of anaerobic bacteria in children under ten years old (60% of
anaerobic bacteria before ten years, 40% after ten years old).
Surprisingly, despite the width of SPOA on the CT scan, we did not
find any pus in our surgery sample in four cases. In these cases, the
average size of SPOA was 6.5 mm (3–9 mm). In 2009, Ryan et al. ex-
plained that despite the high resolution of the CT-scan, it was often
impossible to diff erentiate a real SPOA pus collection from an in-
flammatory collection such as a phlegmon [13]. However, some authors
consider that orbital sonography is better than CT-scan to distinguish a
pus collection and a phlegmon. But, its achievement can be painful and
difficult in case of acute ethmoiditis in the child. This examination is
currently not a gold standard [31].
Ethmoiditis is a therapeutic emergency. A probabilistic intravenous
antibiotic therapy must be started as soon as possible without waiting
for bacteriological results. In fact, a large part of patients with SPOA
can just improve with a medical treatment based on Penicillin [12,18].
In case of clinical severity, antibiotics must be introduced before the
surgical drainage and the bacterial samples. We noticed in our study
that samples were rarely sterile because of a previous antibiotic therapy
(only 8% of sterile samples among 50% of treated children before
hospitalization). This low rate of sterile samples can't only be explained
by the direct sampling into the SPOA, but also by the identification
technique in our laboratory.
In our medical center, we used to propose at the beginning the as-
sociation of two intravenous antibiotics: (1) a third generation cepha-
losporin to cover Streptococcus spp, Staphylococcus aureus, Haemophilus
spp, and anaerobic positive gram bacilli, (2) metronidazole to cover
anaerobic negative gram bacilli. Nevertheless, with regard to bacter-
iological samples, it seems to be reasonable to propose only an in-
travenous monotherapy by amoxicillin-clavulanate if there is no
 al.
A. Coudert et
Table 1
Literature review of microbiological culture results of subperiosteal abscess complicating acute ethmoiditis in children.

Microorganisms isolated from SPOA Our Pena [20] Eviatar [21] Huang [22] Soon [23] Sinclair [24] Oxford [1] Nageswaran [14] Hermann [25] TOTAL/% (micro-
study organisms)
(2017) (2013) (2012) (2011) (2011) (2007) (2006) (2006) (2004)

Number of
cultures
N = 34 N = 59 N = 18 N = 30 N=3 N = 39 N = 23 N = 28 N=2

Streptococcus spp milleri group 8 7 0 0 0 10 7 0 1 33/11,1%

pneumoniae 4 0 0 2 0 3 3 0 0 12/4,1%
pyogenes 7 2 0 1 0 0 0 2 0 12/4,1%
Other viridans 0 0 5 8 0 0 0 3 0 16/5,4%
others 2 0 0 0 0 7 3 2 1 15/5,1%
Staphylococcus spp methicillin-sensitive S. aureus 3 10 1 9 0 3 6 3 0 35/11,8%
methicillin-resistant S. aureus 1 14 2 7 0 0 0 0 0 24/8,1%
94

coagulase negative 0 0 4 4 0 0 2 0 0 10/3,4%

Haemophilus spp influenzae 3 4 2 0 0 2 1 2 0 15/5,1%
parainfluenzae 0 0 0 0 0 1 0 0 0
Moraxella catarrhalis 0 0 2 0 0 0 0 1 0 3/1%
Acinetobacteria corynebacterium 0 0 0 1 1 0 0 0 0 2/0,7%
Pseudomonas 0 1 0 0 0 0 0 0 0 1/0,3%
aeruginosacorredens
Eikenella 0 3 0 2 0 0 2 2 0 9/3%
Klebsiella pneumoniae 0 0 0 2 0 0 0 0 0 2/0,7%

91–95
International Journal of Pediatric Otorhinolaryngology 106 (2018)
Anaerobic bacteria Fusobacterium necrophorum 0 0 0 0 0 0 0 0 1 22/7,4%
Peptostreptococcus spp 0 0 0 1 0 0 1 3 0
prevotella 1 0 0 1 0 0 1 0 0
propionibacterium spp 1 0 0 2 0 0 1 0 0
Bacteroides spp 0 0 0 0 0 0 1 2 0
others 3 0 0 0 0 0 3 0 0
Normal flora 0 0 0 0 0 0 2 0 0 2/0,7%
oropharyngeal
Others 0 10 1 4 0 0 3 0 0 18/6,1%
Sterile culture 3 12 4 2 2 13 2 8 0 46/15,5%
Polymicrobial culture 1 not noticed 3 7 0 0 not noticed 7 1 19/6,4%
A. Coudert et
al.
meningeal risk. If there is a meningeal risk, the association of two an-
tibiotics, a third generation cephalosporin and metronidazole, is still
indicated at a meningeal dose because these drugs proved their effi-
ciency by passing through the blood brain barrier.
After forty eight hours of apyrexia and obtaining bacteriological
results, we can switch to an available oral antibiotic during at least
eight days. We advocate amoxicillin plus clavulanate which are effi-
cient on all bacteria usually encountered. Metronidazole alone must be
avoided because it is only efficient on gram negative bacteria and 83%
of gram positive bacteria are resistant to metronidazole. Pristinamycin
must also be avoided because about 30% of Streptococcus spp are re-
sistant to this antibiotic. Fluoroquinolones should not be routinely used
as first line agents in children under eighteen years old, except in
specific conditions for which there is no alternative [32]. This protocol
is valid for ethmoiditis which are only treated medically. In the case of
surgical treatment, antibiotics must be adapted to the bacteriological
samples.
Our study has some limitations. First, the retrospective study is
dependent on medical record strictness. Then, among drained abscess,
more than half of the children had an oculomotor problem. This
number is highly significant and we can wonder if it is not under-
estimated because the oculomotor examination in the smallest children
can be a real challenge. In these cases, some children could have re-
covered only with medical treatment. Finally, the surgical approach
was not totally homogenous in our study because of the diff erence of
surgeons' experience. In fact, the surgical drainage was realized during
night emergencies in more than 50% of cases, with surgeons who were
not always trained for endoscopic approach in children. This probably
explains the high number of external surgery in our study. However,
regardless of the surgical technique used, we did not notice any com-
plications.
5. Conclusion
Streptococcus spp and Staphylococcus spp are the two main bacteria
responsible for subperiosteal abscess complicating acute ethmoiditis in
children. For five years, S. milleri are on the rise and can be virulent.
The antibiotic treatment should be adapted to these bacteria, and
especially to anaerobic bacteria which can be encountered in children
also under ten years. Nevertheless even if MRSA increase, these pa-
thogens remain rare in bacteriological samples in Europe. Based on our
experience and the bacterial ecology in our region, the following re-
commendations on the antibiotic use are being proposed: without any
meningeal risk, intravenous amoxicillin-clavulanate may be good en-
ough at a first line treatment. With a meningeal risk, a third generation
cephalosporin must be associated with metronidazole. Then the oral
relay should be adapted to bacteriological sampling results and the use
of amoxicillin-clavulanate is proposed if the samples are sterile.
Conflict of interest
None.
References
[1] L.E. Oxford, J. McClay, Complications of acute sinusitis in children,
Otolaryngol.
Head Neck Surg. 133 (2005) 32–37.
[2] L.E. Oxford, J. McClay, Medical and surgical management of subperiosteal
orbital
International Journal of Pediatric Otorhinolaryngology 106 (2018)
91–95
abscess secondary to acute sinusitis in children, Int. J. Pediatr.
Otorhinolaryngol. 70 (2006) 1853–1861.
[3] I. Brook, Microbiology and antimicrobial treatment of orbital and
intracranial complications of sinusitis in children and their management,
Int. J. Pediatr. Otorhinolaryngol. 73 (2009) 1183–1186.
[4] G.H. Garcia, G.J. Harris, Criteria for nonsurgical management of
subperiosteal abscess of the orbit, Ophthalmology 107 (2000) 1454–
1458.
[5] M. Sulte´sz, Z. Csa´ka´nyi, T. Majoros, et al., Acute bacterial rhinosinusitis
and its complications in our pediatric otolaryngological department between
1997 and
2006, Int. J. Pediatr. Otorhinolaryngol. 73 (2009) 1507–1512.
[6] J.R. Chandler, D.J. Langenbrunner, E.F. Stevens, The pathogenesis of orbital
com- plications in acute sinusitis, Laryngoscope 80 (1970) 1414–1428.
[7] V.E.S. Tan, Pediatric subperiosteal orbital abscess secondary to acute
sinusitis: a 5- year review, Am. J. Otolaryngol. Head Neck Med. Surg. 32
(2011) 62–68.
[8] C.L. Brown, S.M. Graham, M.C. Griffin, et al., Pediatric medial subperiosteal
orbital abscess: medical management where possible, Am. J. Rhinol. 18 (5)
(2004)
321–327.
[9] S. Fakhri, K. Pereira, Endoscopic management of orbital abscesses,
Otolaryngol.
Clin. 39 (5) (2006) 1037–1047 viii.
[10] J. Bedwell, N.M. Bauman, Management of pediatric orbital cellulitis and
abscess, Curr. Opin. Otolaryngol. Head Neck Surg. 19 (6) (2011) 467–473.
[11] V.A. Epstein, R.C. Kern, Invasive fungal sinusitis and complications of
rhinosinu- sitis, Otolaryngol. Clin. 41 (3) (2008) 497–524.
[12] I. Ketenci, Y. Unlu, A. Vural, et al., Approaches to subperiosteal orbital
abscesses, Eur. Arch. Oto-Rhino-Laryngol. 270 (2013) 1317–1327.
[13] J.T. Ryan, D.A. Preciado, N. Bauman, et al., Management of pediatric orbital
cel-
lulitis in patients with radiographic findings of subperiosteal abscess,
Otolaryngol. Head Neck Surg. 140 (6) (2009) 907–991.
[14] S. Nageswaran, C.R. Woods, D.K. Benjamin, et al., Orbital cellulitis in
children, Pediatr. Infect. Dis. J. 25 (2006) 695–699.
[15] F. Rubin, S. Pierrot, M. Lebeton, et al., Drainage of subperiosteal orbital
abscesses complicating pediatric ethmoiditis: comparison between external
and transnasal approaches, Int. J. Pediatr. Otorhinolaryngol. 77 (2013) 796–
802.
[16] F. Tabarino, M. Elmaleh-Bergès, S. Quesnel, et al., Subperiosteal orbital
abscess: volumetric criteria for surgical drainage, Int. J. Pediatr.
Otorhinolaryngol. 79 (2015) 131–135.
[17] N. Singhal, M. Kumar, P.K. Kanaujia, et al., MALDI-TOF mass spectrometry:
an emerging technology for microbial identification and diagnosis, Front.
Microbiol. 6 (2015) 791.
[18] I. Brook, Microbiology and choice of antimicrobial therapy for acute
sinusitis complicated by subperiosteal abscess in children, Int. J. Pediatr.
Otorhinolaryngol.
84 (2016) 21–26.
[19] A. Jain, P.A. Rubin, Orbital cellulitis in children, Int. Ophthalmol. Clin. 41
(2001)
71–86.
[20] M.T. Pena, D. Preciado, M. Orestes, et al., Orbital complications of acute
sinusitis, JAMA Otolaryngol. Head Neck Surg. 139 (3) (2013) 223–227.
[21] E. Eviatar, T. Lazarovitch, H. Gavriel, The correlation of microbiology growth
be-
tween subperiosteal orbital abscess and aff ected sinuses in young children,
Am. J. Rhinol. Allergy 26 (6) (November–December 2012).
[22] S.F. Huang, T.J. Lee, Y.S. Lee, et al., Acute rhinosinusitis–related orbital
infection in pediatric patients: a retrospective analysis, Ann. Otol. Rhinol.
Laryngol. 120 (3) (2011 Mar) 185–190.
[23] V.T.E. Soon, Pediatric subperiosteal orbital abscess secondary to acute
sinusitis: a 5- year review, Am. J. Otolaryngol. Head Neck Med. Surg. 32
(2011) 62–68.
[24] C.F. Sinclair, R.G. Berkowitz, Prior antibiotic therapy for acute sinusitis in
children
and the development of subperiosteal orbital abscess, Int. J.
Pediatr. Otorhinolaryngol. 71 (2007) 1003–1006.
[25] B.W. Herrmann, J.W. Forsen, Simultaneous intracranial and orbital
complications of acute rhinosinusitis in children, Int. J. Pediatr.
Otorhinolaryngol. 68 (2004)
619–625.
[26] M. Miller, H.A. Cook, E.Y. Furuya, et al., Staphylococcus aureus in the
community:
colonization versus infection, PLoS One 4 (2009) e6708.
[27] J.C. Liao, M.L. Durand, M.J. Cunningham, Sinogenic orbital and
subperiosteal ab- scesses: microbiology and methicillin-resistant
Staphylococcus aureus incidence, Otolaryngol. Head Neck Surg. 143 (2010)
392–396.
[28] J.C. Liao, G.J. Harris, Subperiosteal abscess of the orbit: evolving pathogens
and the therapeutic protocol, Ophthalmology 122 (2015) 639–647.
[29] I. Brook, The role of anaerobic bacteria in sinusitis, Clin. Anaerobe 12 (2006)
5–12. [30] G.J. Harris, Subperiosteal abscess of the orbit: older children and
adults require
 A. Coudert et International Journal of Pediatric Otorhinolaryngology 106 (2018)
al. aggressive treatment, Ophthalmic Plast. Reconstr. Surg. 17 (2001) 395–397. 91–95
[32] I. Brook, Antimicrobial treatment of anaerobic infections, Expert. Opin.
[31] M.H. Mair, T. Geley, W. Judmaier, et al., Using orbital sonography to Pharmacother. 12 (2011) 1691–1707.
diagnose and monitor treatment of acute swelling of the eyelids in pediatric
patients, Am. J. Roentgenol. 179 (2002) 1529–1534.

95