- Email: [email protected]
Microbiology and antibiotic therapy of subperiosteal orbital abscess in children with acute ethmoiditis
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
T
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 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 b
A R T I C L E I N F O
A B S T R A C T
Keywords: Acute sinusitis Subperiosteal orbital abscess Antibiotics Microbiology Pediatrics
Objective: The objective of this study was to investigate the microbiological cultures and the management of acute ethmoiditis complicated by subperiosteal orbital abscess (SPOA) in a pediatric population. Methods: The medical records of children under 18 years old was performed in a tertiary referral pediatric center from January 2009 to April 2017. Clinical examination, computed tomography scans, medical and surgical treatments were reviewed and compared to other studies in literature. 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 ethmoiditis frequently occur. It accounts for 21% of pediatric antibiotic prescriptions [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 hematogenous dissemination [2,3]. Sometimes, this complication can be favored 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 cerebral 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 children 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 ethmoiditis 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: [email protected] (A. Coudert), [email protected] (S. Ayari-Khalfallah), [email protected] (P. Suy), [email protected] (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 0165-5876/ © 2018 Elsevier B.V. All rights reserved.
International Journal of Pediatric Otorhinolaryngology 106 (2018) 91–95
A. Coudert et al.
antibiotic use is discussed. 2. Material and methods This retrospective study was performed in a tertiary referral pediatric 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. 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 CTscans were reviewed for SPOA width and length, and collection in sinuses. Our surgery indications, based on previous study [15,16], were the following ones: (1) abscess width more than 5 mm or extended to the optic nerve, (2) oculomotor disorder, (3) absence of symptoms improvement after 48–72 h of intravenous antibiotics, (4) severe clinical complications such as epidural empyema, loss of visual acuity or cavernous sinus thrombophlebitis. The surgical therapy was recorded and separated into three categories: an external approach (EA), a transnasal endoscopic approach (TEA) and a combined external and transnasal approach (CA) to drain the SPOA. Since our department was the ENT emergency center for the Rhone department, the surgery was performed by the ENT surgeon on call who could be a fellow. The choice between the EA, TEA or the CA was decided by the surgeon on call, taking into consideration the abscess characteristics, the anatomical conditions but also his personal training and skills in pediatric transnasal endoscopic surgery. During surgery, sinus secretions and the pus from the SPOA were systematically sampled and carried out, immediately to the laboratory, in a plastic tube containing a sterile swab. The way of detecting and identifying bacteria was based on culturing using different culture media with control of the nutrients and culture conditions (temperature, air supply, O2, light, blood, pH …), enumeration and isolation of presumptive colonies with study of phenotypic characteristics completed if needed by genotypic characteristics. The phenotypic method included biotyping (growth requirement, environmental conditions, antibiotic resistance, cell morphology …), and identification by mass spectrometry. Mass spectrometry is an analytical technique in which chemical compounds are ionized into charged molecules and ratio of their mass to charge (m/z) is measured. Identification of microbes is done by either comparing the peptide mass fingerprint (PMF) of unknown organism with the PMFs contained in the database or by matching the masses of biomarkers of unknown organism with the proteome database [17]. The genotypic method used molecular techniques to identify bacteria by doing DNA or RNA analysis of the bacterium's genome. The system used in our center was the Vitek®MS system (by bioMérieux France). If the mass spectrophotometry was not contributory, a universal PCR detection method was used. The Ethics Committee of the Hospices Civils de Lyon in France approved the study (Number 17-145) and all patients gave written informed consent.
Fig. 1. Evaluation and treatment of 129 children with acute ethmoiditis.
More than 50% of children with an operated SPOA had at least one ophthalmologic trouble at the beginning. The most constant sign was ophthalmoplegia (71% cases). In 50% of cases, we recorded proptosis and/or diplopia. The mean age of operated children with an oculomotor dysfunction was nine years and six months whereas for the others without oculomotor trouble it was four years and six months. On the CT-scan, the mean width of SPOA was 6 mm (range 3–12 mm). Twenty-one children were treated with an external approach, six with a transnasal endoscopic technique and seven with a combined approach. An example of a young child with a SPOA drained by an external approach is given in Fig. 2. Among the operated children, seventeen (50%) had a previous antibiotic treatment before hospitalization (amoxicillin-clavulanate, or pristinamycin, or a third-generation cephalosporin, or josamycin). The average duration of this antimicrobial therapy was four days. Only two children had a sterile surgical sample despite a first medical therapy during 48 h. 3.1. Microbiologic cultures (Table 1) Cultures were obtained by pus sample during the surgical intervention. Samples were contributive in 91% of cases. The most frequent encountered bacteria was Streptococcus spp which was found in more than 60% of cases. Furthermore, the other identified species were the anaerobic bacteria (12%), Staphylococcus aureus (12%) and Haemophilus influenzae (9%). There was only one case of methicillin-resistant Staphylococcus aureus (MRSA). The age of children with anaerobic cultures ranged from 3 years to 15 years. All children older than nine years had a Streptococcus intermedius (which belongs to milleri group) in their culture. After surgery, thirty-two children received two intravenous antibiotics (third-generation cephalosporin and metronidazole) for a mean duration of four days. Only two children had a prolonged intravenous antimicrobial treatment. The first one had a MRSA, he received intravenous clindamycin and vancomycin during ten days. The second one had a subdural empyema which required intravenous third-generation cephalosporin and metronidazole for fifteen days. The average hospital stay was 6.5 days (range 3–16 days). Then after hospitalization, each child had an oral antibiotic during about 8.5 days (from 7 to 15 days). The most frequently used antibiotic was amoxicillin-clavulanate (76% cases). Because of allergy, some children had pristinamycin, or clindamycin and metronidazole. All children were cured without sequelae at the end of the antibiotic treatment. Moreover we did not notice any recurrence of ethmoiditis with our follow-up. The mean follow-up length was 85.8 days after the surgery, with a range of 9 days to 5.3 years.
3. Results One hundred and twenty-nine children were hospitalized in our center for acute ethmoiditis between 2009 and 2017. Among them, 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 drainage. For the others, a medical treatment was sufficient. Clinical data and patient characteristics are summarized in Fig. 1. Before surgery, each child underwent an oculomotor examination.
4. Discussion Many studies have been conducted over the past decades on microbiology of SPOA complicating sinusitis. In a recent literature review 92
International Journal of Pediatric Otorhinolaryngology 106 (2018) 91–95
A. Coudert et al.
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 external 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 different 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. explained that despite the high resolution of the CT-scan, it was often impossible to differentiate a real SPOA pus collection from an inflammatory 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 association of two intravenous antibiotics: (1) a third generation cephalosporin 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 bacteriological samples, it seems to be reasonable to propose only an intravenous monotherapy by amoxicillin-clavulanate if there is no
by Brook in 2016, the rate of SPOA due to sinusitis was estimated to occur in about 5% of hospitalized patients [18]. Bacteriological identification of SPOA is a fundamental factor in order to adapt the appropriate intravenous antibiotic followed by an oral treatment. One of the most relevant microbiological samples comes from direct sinus aspiration or by drainage of a subperiosteal orbital abscess [19]. Other sampling, like nasal aspiration, can easily lead to bacteriological contamination 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 abscess in children. In Table 1, we compared our results with microbiological 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 difference 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 different species: Streptococcus anginosus, Streptococcus intermidius, and Streptococcus constellatus. These commensal bacteria can be virulent and lead to profound abscess formation. This microbiological evolution is a direct consequence of vaccinations 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 oropharyngeal 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
94
milleri group pneumoniae pyogenes Other viridans others methicillin-sensitive S. aureus methicillin-resistant S. aureus coagulase negative influenzae parainfluenzae
Moraxella catarrhalis Acinetobacteria corynebacterium Pseudomonas aeruginosa Eikenella corredens Klebsiella pneumoniae Anaerobic bacteria Fusobacterium necrophorum Peptostreptococcus spp prevotella propionibacterium spp Bacteroides spp others Normal oropharyngeal flora Others Sterile culture Polymicrobial culture
Haemophilus spp
Staphylococcus spp
Streptococcus spp
Microorganisms isolated from SPOA
Pena [20] (2013)
N = 59 7 0 2 0 0 10 14 0 4 0 0 0 1 3 0 0 0 0 0 0 0 0 10 12 not noticed
Our study (2017)
N = 34 8 4 7 0 2 3 1 0 3 0 0 0 0 0 0 0 0 1 1 0 3 0 0 3 1 0 0 0 5 0 1 2 4 2 0 2 0 0 0 0 0 0 0 0 0 0 0 1 4 3
N = 18
(2012)
Eviatar [21]
0 2 1 8 0 9 7 4 0 0 0 1 0 2 2 0 1 1 2 0 0 0 4 2 7
N = 30
(2011)
Huang [22]
Table 1 Literature review of microbiological culture results of subperiosteal abscess complicating acute ethmoiditis in children.
(2007)
Sinclair [24]
0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 2 0
N=3 10 3 0 0 7 3 0 0 2 1 0 0 0 0 0 0 0 0 0 0 0 0 0 13 0
N = 39
Number of cultures
(2011)
Soon [23]
7 3 0 0 3 6 0 2 1 0 0 0 0 2 0 0 1 1 1 1 3 2 3 2 not noticed
N = 23
(2006)
Oxford [1]
0 0 2 3 2 3 0 0 2 0 1 0 0 2 0 0 3 0 0 2 0 0 0 8 7
N = 28
(2006)
Nageswaran [14]
1 0 0 0 1 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 1
N=2
(2004)
Hermann [25]
2/0,7% 18/6,1% 46/15,5% 19/6,4%
3/1% 2/0,7% 1/0,3% 9/3% 2/0,7% 22/7,4%
33/11,1% 12/4,1% 12/4,1% 16/5,4% 15/5,1% 35/11,8% 24/8,1% 10/3,4% 15/5,1%
TOTAL/% (micro-organisms)
A. Coudert et al.
International Journal of Pediatric Otorhinolaryngology 106 (2018) 91–95
International Journal of Pediatric Otorhinolaryngology 106 (2018) 91–95
A. Coudert et al.
meningeal risk. If there is a meningeal risk, the association of two antibiotics, a third generation cephalosporin and metronidazole, is still indicated at a meningeal dose because these drugs proved their efficiency 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 efficient 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 resistant 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 underestimated because the oculomotor examination in the smallest children can be a real challenge. In these cases, some children could have recovered only with medical treatment. Finally, the surgical approach was not totally homogenous in our study because of the difference 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 complications.
[3]
[4] [5]
[6] [7] [8]
[9] [10] [11] [12] [13]
[14] [15]
[16]
[17]
[18]
5. Conclusion
[19] [20]
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 pathogens remain rare in bacteriological samples in Europe. Based on our experience and the bacterial ecology in our region, the following recommendations on the antibiotic use are being proposed: without any meningeal risk, intravenous amoxicillin-clavulanate may be good enough 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.
[21]
[22]
[23] [24]
[25]
[26] [27]
Conflict of interest
[28]
None.
[29] [30]
References
[31]
[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
[32]
95
abscess secondary to acute sinusitis in children, Int. J. Pediatr. Otorhinolaryngol. 70 (2006) 1853–1861. 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. G.H. Garcia, G.J. Harris, Criteria for nonsurgical management of subperiosteal abscess of the orbit, Ophthalmology 107 (2000) 1454–1458. 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. J.R. Chandler, D.J. Langenbrunner, E.F. Stevens, The pathogenesis of orbital complications in acute sinusitis, Laryngoscope 80 (1970) 1414–1428. V.E.S. Tan, Pediatric subperiosteal orbital abscess secondary to acute sinusitis: a 5year review, Am. J. Otolaryngol. Head Neck Med. Surg. 32 (2011) 62–68. 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. S. Fakhri, K. Pereira, Endoscopic management of orbital abscesses, Otolaryngol. Clin. 39 (5) (2006) 1037–1047 viii. J. Bedwell, N.M. Bauman, Management of pediatric orbital cellulitis and abscess, Curr. Opin. Otolaryngol. Head Neck Surg. 19 (6) (2011) 467–473. V.A. Epstein, R.C. Kern, Invasive fungal sinusitis and complications of rhinosinusitis, Otolaryngol. Clin. 41 (3) (2008) 497–524. I. Ketenci, Y. Unlu, A. Vural, et al., Approaches to subperiosteal orbital abscesses, Eur. Arch. Oto-Rhino-Laryngol. 270 (2013) 1317–1327. J.T. Ryan, D.A. Preciado, N. Bauman, et al., Management of pediatric orbital cellulitis in patients with radiographic findings of subperiosteal abscess, Otolaryngol. Head Neck Surg. 140 (6) (2009) 907–991. S. Nageswaran, C.R. Woods, D.K. Benjamin, et al., Orbital cellulitis in children, Pediatr. Infect. Dis. J. 25 (2006) 695–699. 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. 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. 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. 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. A. Jain, P.A. Rubin, Orbital cellulitis in children, Int. Ophthalmol. Clin. 41 (2001) 71–86. M.T. Pena, D. Preciado, M. Orestes, et al., Orbital complications of acute sinusitis, JAMA Otolaryngol. Head Neck Surg. 139 (3) (2013) 223–227. E. Eviatar, T. Lazarovitch, H. Gavriel, The correlation of microbiology growth between subperiosteal orbital abscess and affected sinuses in young children, Am. J. Rhinol. Allergy 26 (6) (November–December 2012). 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. V.T.E. Soon, Pediatric subperiosteal orbital abscess secondary to acute sinusitis: a 5year review, Am. J. Otolaryngol. Head Neck Med. Surg. 32 (2011) 62–68. 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. B.W. Herrmann, J.W. Forsen, Simultaneous intracranial and orbital complications of acute rhinosinusitis in children, Int. J. Pediatr. Otorhinolaryngol. 68 (2004) 619–625. M. Miller, H.A. Cook, E.Y. Furuya, et al., Staphylococcus aureus in the community: colonization versus infection, PLoS One 4 (2009) e6708. J.C. Liao, M.L. Durand, M.J. Cunningham, Sinogenic orbital and subperiosteal abscesses: microbiology and methicillin-resistant Staphylococcus aureus incidence, Otolaryngol. Head Neck Surg. 143 (2010) 392–396. J.C. Liao, G.J. Harris, Subperiosteal abscess of the orbit: evolving pathogens and the therapeutic protocol, Ophthalmology 122 (2015) 639–647. I. Brook, The role of anaerobic bacteria in sinusitis, Clin. Anaerobe 12 (2006) 5–12. G.J. Harris, Subperiosteal abscess of the orbit: older children and adults require aggressive treatment, Ophthalmic Plast. Reconstr. Surg. 17 (2001) 395–397. M.H. Mair, T. Geley, W. Judmaier, et al., Using orbital sonography to diagnose and monitor treatment of acute swelling of the eyelids in pediatric patients, Am. J. Roentgenol. 179 (2002) 1529–1534. I. Brook, Antimicrobial treatment of anaerobic infections, Expert. Opin. Pharmacother. 12 (2011) 1691–1707.
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
T
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 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 b
A R T I C L E I N F O
A B S T R A C T
Keywords: Acute sinusitis Subperiosteal orbital abscess Antibiotics Microbiology Pediatrics
Objective: The objective of this study was to investigate the microbiological cultures and the management of acute ethmoiditis complicated by subperiosteal orbital abscess (SPOA) in a pediatric population. Methods: The medical records of children under 18 years old was performed in a tertiary referral pediatric center from January 2009 to April 2017. Clinical examination, computed tomography scans, medical and surgical treatments were reviewed and compared to other studies in literature. 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 ethmoiditis frequently occur. It accounts for 21% of pediatric antibiotic prescriptions [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 hematogenous dissemination [2,3]. Sometimes, this complication can be favored 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 cerebral 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 children 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 ethmoiditis 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: [email protected] (A. Coudert), [email protected] (S. Ayari-Khalfallah), [email protected] (P. Suy), [email protected] (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 0165-5876/ © 2018 Elsevier B.V. All rights reserved.
International Journal of Pediatric Otorhinolaryngology 106 (2018) 91–95
A. Coudert et al.
antibiotic use is discussed. 2. Material and methods This retrospective study was performed in a tertiary referral pediatric 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. 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 CTscans were reviewed for SPOA width and length, and collection in sinuses. Our surgery indications, based on previous study [15,16], were the following ones: (1) abscess width more than 5 mm or extended to the optic nerve, (2) oculomotor disorder, (3) absence of symptoms improvement after 48–72 h of intravenous antibiotics, (4) severe clinical complications such as epidural empyema, loss of visual acuity or cavernous sinus thrombophlebitis. The surgical therapy was recorded and separated into three categories: an external approach (EA), a transnasal endoscopic approach (TEA) and a combined external and transnasal approach (CA) to drain the SPOA. Since our department was the ENT emergency center for the Rhone department, the surgery was performed by the ENT surgeon on call who could be a fellow. The choice between the EA, TEA or the CA was decided by the surgeon on call, taking into consideration the abscess characteristics, the anatomical conditions but also his personal training and skills in pediatric transnasal endoscopic surgery. During surgery, sinus secretions and the pus from the SPOA were systematically sampled and carried out, immediately to the laboratory, in a plastic tube containing a sterile swab. The way of detecting and identifying bacteria was based on culturing using different culture media with control of the nutrients and culture conditions (temperature, air supply, O2, light, blood, pH …), enumeration and isolation of presumptive colonies with study of phenotypic characteristics completed if needed by genotypic characteristics. The phenotypic method included biotyping (growth requirement, environmental conditions, antibiotic resistance, cell morphology …), and identification by mass spectrometry. Mass spectrometry is an analytical technique in which chemical compounds are ionized into charged molecules and ratio of their mass to charge (m/z) is measured. Identification of microbes is done by either comparing the peptide mass fingerprint (PMF) of unknown organism with the PMFs contained in the database or by matching the masses of biomarkers of unknown organism with the proteome database [17]. The genotypic method used molecular techniques to identify bacteria by doing DNA or RNA analysis of the bacterium's genome. The system used in our center was the Vitek®MS system (by bioMérieux France). If the mass spectrophotometry was not contributory, a universal PCR detection method was used. The Ethics Committee of the Hospices Civils de Lyon in France approved the study (Number 17-145) and all patients gave written informed consent.
Fig. 1. Evaluation and treatment of 129 children with acute ethmoiditis.
More than 50% of children with an operated SPOA had at least one ophthalmologic trouble at the beginning. The most constant sign was ophthalmoplegia (71% cases). In 50% of cases, we recorded proptosis and/or diplopia. The mean age of operated children with an oculomotor dysfunction was nine years and six months whereas for the others without oculomotor trouble it was four years and six months. On the CT-scan, the mean width of SPOA was 6 mm (range 3–12 mm). Twenty-one children were treated with an external approach, six with a transnasal endoscopic technique and seven with a combined approach. An example of a young child with a SPOA drained by an external approach is given in Fig. 2. Among the operated children, seventeen (50%) had a previous antibiotic treatment before hospitalization (amoxicillin-clavulanate, or pristinamycin, or a third-generation cephalosporin, or josamycin). The average duration of this antimicrobial therapy was four days. Only two children had a sterile surgical sample despite a first medical therapy during 48 h. 3.1. Microbiologic cultures (Table 1) Cultures were obtained by pus sample during the surgical intervention. Samples were contributive in 91% of cases. The most frequent encountered bacteria was Streptococcus spp which was found in more than 60% of cases. Furthermore, the other identified species were the anaerobic bacteria (12%), Staphylococcus aureus (12%) and Haemophilus influenzae (9%). There was only one case of methicillin-resistant Staphylococcus aureus (MRSA). The age of children with anaerobic cultures ranged from 3 years to 15 years. All children older than nine years had a Streptococcus intermedius (which belongs to milleri group) in their culture. After surgery, thirty-two children received two intravenous antibiotics (third-generation cephalosporin and metronidazole) for a mean duration of four days. Only two children had a prolonged intravenous antimicrobial treatment. The first one had a MRSA, he received intravenous clindamycin and vancomycin during ten days. The second one had a subdural empyema which required intravenous third-generation cephalosporin and metronidazole for fifteen days. The average hospital stay was 6.5 days (range 3–16 days). Then after hospitalization, each child had an oral antibiotic during about 8.5 days (from 7 to 15 days). The most frequently used antibiotic was amoxicillin-clavulanate (76% cases). Because of allergy, some children had pristinamycin, or clindamycin and metronidazole. All children were cured without sequelae at the end of the antibiotic treatment. Moreover we did not notice any recurrence of ethmoiditis with our follow-up. The mean follow-up length was 85.8 days after the surgery, with a range of 9 days to 5.3 years.
3. Results One hundred and twenty-nine children were hospitalized in our center for acute ethmoiditis between 2009 and 2017. Among them, 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 drainage. For the others, a medical treatment was sufficient. Clinical data and patient characteristics are summarized in Fig. 1. Before surgery, each child underwent an oculomotor examination.
4. Discussion Many studies have been conducted over the past decades on microbiology of SPOA complicating sinusitis. In a recent literature review 92
International Journal of Pediatric Otorhinolaryngology 106 (2018) 91–95
A. Coudert et al.
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 external 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 different 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. explained that despite the high resolution of the CT-scan, it was often impossible to differentiate a real SPOA pus collection from an inflammatory 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 association of two intravenous antibiotics: (1) a third generation cephalosporin 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 bacteriological samples, it seems to be reasonable to propose only an intravenous monotherapy by amoxicillin-clavulanate if there is no
by Brook in 2016, the rate of SPOA due to sinusitis was estimated to occur in about 5% of hospitalized patients [18]. Bacteriological identification of SPOA is a fundamental factor in order to adapt the appropriate intravenous antibiotic followed by an oral treatment. One of the most relevant microbiological samples comes from direct sinus aspiration or by drainage of a subperiosteal orbital abscess [19]. Other sampling, like nasal aspiration, can easily lead to bacteriological contamination 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 abscess in children. In Table 1, we compared our results with microbiological 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 difference 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 different species: Streptococcus anginosus, Streptococcus intermidius, and Streptococcus constellatus. These commensal bacteria can be virulent and lead to profound abscess formation. This microbiological evolution is a direct consequence of vaccinations 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 oropharyngeal 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
94
milleri group pneumoniae pyogenes Other viridans others methicillin-sensitive S. aureus methicillin-resistant S. aureus coagulase negative influenzae parainfluenzae
Moraxella catarrhalis Acinetobacteria corynebacterium Pseudomonas aeruginosa Eikenella corredens Klebsiella pneumoniae Anaerobic bacteria Fusobacterium necrophorum Peptostreptococcus spp prevotella propionibacterium spp Bacteroides spp others Normal oropharyngeal flora Others Sterile culture Polymicrobial culture
Haemophilus spp
Staphylococcus spp
Streptococcus spp
Microorganisms isolated from SPOA
Pena [20] (2013)
N = 59 7 0 2 0 0 10 14 0 4 0 0 0 1 3 0 0 0 0 0 0 0 0 10 12 not noticed
Our study (2017)
N = 34 8 4 7 0 2 3 1 0 3 0 0 0 0 0 0 0 0 1 1 0 3 0 0 3 1 0 0 0 5 0 1 2 4 2 0 2 0 0 0 0 0 0 0 0 0 0 0 1 4 3
N = 18
(2012)
Eviatar [21]
0 2 1 8 0 9 7 4 0 0 0 1 0 2 2 0 1 1 2 0 0 0 4 2 7
N = 30
(2011)
Huang [22]
Table 1 Literature review of microbiological culture results of subperiosteal abscess complicating acute ethmoiditis in children.
(2007)
Sinclair [24]
0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 2 0
N=3 10 3 0 0 7 3 0 0 2 1 0 0 0 0 0 0 0 0 0 0 0 0 0 13 0
N = 39
Number of cultures
(2011)
Soon [23]
7 3 0 0 3 6 0 2 1 0 0 0 0 2 0 0 1 1 1 1 3 2 3 2 not noticed
N = 23
(2006)
Oxford [1]
0 0 2 3 2 3 0 0 2 0 1 0 0 2 0 0 3 0 0 2 0 0 0 8 7
N = 28
(2006)
Nageswaran [14]
1 0 0 0 1 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 1
N=2
(2004)
Hermann [25]
2/0,7% 18/6,1% 46/15,5% 19/6,4%
3/1% 2/0,7% 1/0,3% 9/3% 2/0,7% 22/7,4%
33/11,1% 12/4,1% 12/4,1% 16/5,4% 15/5,1% 35/11,8% 24/8,1% 10/3,4% 15/5,1%
TOTAL/% (micro-organisms)
A. Coudert et al.
International Journal of Pediatric Otorhinolaryngology 106 (2018) 91–95
International Journal of Pediatric Otorhinolaryngology 106 (2018) 91–95
A. Coudert et al.
meningeal risk. If there is a meningeal risk, the association of two antibiotics, a third generation cephalosporin and metronidazole, is still indicated at a meningeal dose because these drugs proved their efficiency 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 efficient 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 resistant 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 underestimated because the oculomotor examination in the smallest children can be a real challenge. In these cases, some children could have recovered only with medical treatment. Finally, the surgical approach was not totally homogenous in our study because of the difference 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 complications.
[3]
[4] [5]
[6] [7] [8]
[9] [10] [11] [12] [13]
[14] [15]
[16]
[17]
[18]
5. Conclusion
[19] [20]
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 pathogens remain rare in bacteriological samples in Europe. Based on our experience and the bacterial ecology in our region, the following recommendations on the antibiotic use are being proposed: without any meningeal risk, intravenous amoxicillin-clavulanate may be good enough 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.
[21]
[22]
[23] [24]
[25]
[26] [27]
Conflict of interest
[28]
None.
[29] [30]
References
[31]
[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
[32]
95
abscess secondary to acute sinusitis in children, Int. J. Pediatr. Otorhinolaryngol. 70 (2006) 1853–1861. 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. G.H. Garcia, G.J. Harris, Criteria for nonsurgical management of subperiosteal abscess of the orbit, Ophthalmology 107 (2000) 1454–1458. 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. J.R. Chandler, D.J. Langenbrunner, E.F. Stevens, The pathogenesis of orbital complications in acute sinusitis, Laryngoscope 80 (1970) 1414–1428. V.E.S. Tan, Pediatric subperiosteal orbital abscess secondary to acute sinusitis: a 5year review, Am. J. Otolaryngol. Head Neck Med. Surg. 32 (2011) 62–68. 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. S. Fakhri, K. Pereira, Endoscopic management of orbital abscesses, Otolaryngol. Clin. 39 (5) (2006) 1037–1047 viii. J. Bedwell, N.M. Bauman, Management of pediatric orbital cellulitis and abscess, Curr. Opin. Otolaryngol. Head Neck Surg. 19 (6) (2011) 467–473. V.A. Epstein, R.C. Kern, Invasive fungal sinusitis and complications of rhinosinusitis, Otolaryngol. Clin. 41 (3) (2008) 497–524. I. Ketenci, Y. Unlu, A. Vural, et al., Approaches to subperiosteal orbital abscesses, Eur. Arch. Oto-Rhino-Laryngol. 270 (2013) 1317–1327. J.T. Ryan, D.A. Preciado, N. Bauman, et al., Management of pediatric orbital cellulitis in patients with radiographic findings of subperiosteal abscess, Otolaryngol. Head Neck Surg. 140 (6) (2009) 907–991. S. Nageswaran, C.R. Woods, D.K. Benjamin, et al., Orbital cellulitis in children, Pediatr. Infect. Dis. J. 25 (2006) 695–699. 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. 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. 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. 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. A. Jain, P.A. Rubin, Orbital cellulitis in children, Int. Ophthalmol. Clin. 41 (2001) 71–86. M.T. Pena, D. Preciado, M. Orestes, et al., Orbital complications of acute sinusitis, JAMA Otolaryngol. Head Neck Surg. 139 (3) (2013) 223–227. E. Eviatar, T. Lazarovitch, H. Gavriel, The correlation of microbiology growth between subperiosteal orbital abscess and affected sinuses in young children, Am. J. Rhinol. Allergy 26 (6) (November–December 2012). 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. V.T.E. Soon, Pediatric subperiosteal orbital abscess secondary to acute sinusitis: a 5year review, Am. J. Otolaryngol. Head Neck Med. Surg. 32 (2011) 62–68. 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. B.W. Herrmann, J.W. Forsen, Simultaneous intracranial and orbital complications of acute rhinosinusitis in children, Int. J. Pediatr. Otorhinolaryngol. 68 (2004) 619–625. M. Miller, H.A. Cook, E.Y. Furuya, et al., Staphylococcus aureus in the community: colonization versus infection, PLoS One 4 (2009) e6708. J.C. Liao, M.L. Durand, M.J. Cunningham, Sinogenic orbital and subperiosteal abscesses: microbiology and methicillin-resistant Staphylococcus aureus incidence, Otolaryngol. Head Neck Surg. 143 (2010) 392–396. J.C. Liao, G.J. Harris, Subperiosteal abscess of the orbit: evolving pathogens and the therapeutic protocol, Ophthalmology 122 (2015) 639–647. I. Brook, The role of anaerobic bacteria in sinusitis, Clin. Anaerobe 12 (2006) 5–12. G.J. Harris, Subperiosteal abscess of the orbit: older children and adults require aggressive treatment, Ophthalmic Plast. Reconstr. Surg. 17 (2001) 395–397. M.H. Mair, T. Geley, W. Judmaier, et al., Using orbital sonography to diagnose and monitor treatment of acute swelling of the eyelids in pediatric patients, Am. J. Roentgenol. 179 (2002) 1529–1534. I. Brook, Antimicrobial treatment of anaerobic infections, Expert. Opin. Pharmacother. 12 (2011) 1691–1707.