- Email: [email protected]
Pediatric recurrent acute mastoiditis: Risk factors and insights into pathogenesis
International Journal of Pediatric Otorhinolaryngology 111 (2018) 142–148
Contents lists available at ScienceDirect
International Journal of Pediatric Otorhinolaryngology journal homepage: www.elsevier.com/locate/ijporl
Pediatric recurrent acute mastoiditis: Risk factors and insights into pathogenesis
T
Józef Mierzwińskia,∗, Justyna Tyraa, Karolina Habera, Maria Drelaa, Anna Sinkiewiczb, Michael David Puricellic a Department of Otolaryngology, Audiology and Phoniatrics, Pediatric Cochlear Implant Center, Children's Hospital of Bydgoszcz, Poland, Chodkiewicza 44, 85-667, Bydgoszcz, Poland b Department of Phoniatrics and Audiology, University Hospital of Bydgoszcz, Poland c Paparella Ear, Head & Neck Institute, 701 25th Ave. S. Suite 200, Minneapolis, MN, 55454, United States
A R T I C LE I N FO
A B S T R A C T
Keywords: Child Mastoiditis Otitis media Recurrence Operative therapy Mastoidectomy
Objectives: Recurrent acute mastoiditis is repeatedly reported in the literature, but data to understand the pathogenesis, update treatment recommendations and inform future trials are sparse due to the infrequency of the disease. Methods: A retrospective chart review from 2001 to 2016 was conducted including 73 children treated for acute mastoiditis. A follow-up survey was attempted for each patient. Bacteriology, method of treatment, hospital course, complications, and otologic history were analyzed. A chi-squared test, Fisher's exact test and MannWhitney U test compared recurrent acute mastoiditis to single acute mastoiditis cases. Additionally, a comprehensive PubMed search and review of world literature addressing recurrent pediatric acute mastoiditis was performed for comparative purposes. Results: Among 73 children with acute mastoiditis, six (8%) experienced recurrent acute mastoiditis. Streptococcus pneumoniae was the only bacteria isolated in this group. History of recurrent acute otitis media (> 4 per year) prior to the first episode of acute mastoiditis was identified in 24% with single episode of acute mastoiditis and 83% with recurrent mastoiditis (p < 0.05). Fewer intracranial/intratemporal complications were identified among recurrent mastoiditis patients (p < 0.05). In a group of patients treated with more extensive surgical communication during mastoidectomy for primary acute mastoiditis (wide mastoidectomy with broad attic exposure and posterior tympanotomy) no recurrence was observed. Conclusion: We identify multiple risk factors associated with recurrence and provide early data supporting anatomic predisposition to the development of recurrent acute mastoiditis. More aggressive opening between the mastoid cavity and middle ear may prevent recurrent acute mastoiditis episodes.
1. Introduction Acute mastoiditis (AM) is a middle ear infection that extends to the mastoid cells and involves the mastoid bone, leading to periostitis and/ or osteitis. The major hypothesized cause of AM is lack of communication between the tympanic and mastoid cavities, preventing pus drainage through the Eustachian tube and/or tympanic membrane and leading to mastoid osteolysis and abscess formation. Although many studies have examined the etiology and treatment options for pediatric acute mastoiditis, little has been published about recurrent acute mastoiditis (rAM). Primary AM has a reported incidence of 0.6–4.8 per 100,000 per year [1,2,3]. Recurrent episodes of mastoiditis are rare and account for
∗
Corresponding author. E-mail address: [email protected] (J. Mierzwiński).
https://doi.org/10.1016/j.ijporl.2018.06.001 Received 4 February 2018; Received in revised form 31 May 2018; Accepted 2 June 2018 Available online 06 June 2018 0165-5876/ © 2018 Elsevier B.V. All rights reserved.
only a few percent of all AM cases, but are constantly present in major series of AM reported in the world literature, what makes them much more common that it is generally thought. rAM was first reported in 1910 b y Mygind, who found that 4% of AM patients experienced recurrence [4]. William House reported recurrence rates of up to 20% in the pre-antibiotic era [5]. In the post-antibiotic era, rates of rAM range from 2 to 18%, with most series finding frequency of 5–10%. Variable treatments have been successfully applied ranging from antibiotics alone; myringotomy and tube placement with antibiotics; and mastoidectomy with tympanostomy tube placement and antibiotics (Table 1). Despite recognition of rAM as an entity, there has been little published regarding the mechanism of development or factors that might predict or prevent rAM.
32/798 4%
5/78 6%
22/434
Groth et al. [6]
Lahav et al. [7]
Gavriel et al. [8]
143 No data
8/188 4%
1/57 2%
11/223 5%
Luntz et al. [16]
1/1 IV antibiotics, myringotomy and tube insertion, mastoidectomy No data
4/11 surgical treatment 7/11 conservative treatment 2/2 antibiotics
3/3 mastoidectomy
1/2 no data 1/2 incision and drainage
2/11 18%
3/79 4% 11/134 8% 2/29 7%
No data
2/5 aspiration 3/5 mastoidectomy 5/5 IV antibiotics and myringotomy 20/22 IV antibiotics and myringotomy 2/22 mastoidectomy
24/32 mastoidectomy 8/32 IV antibiotics and myringotomy
5/6 mastoidectomy 1/6 conservative treatment
First episode management of recurrent cases
3/43 7%
Harley et al. [15]
Glynn et al. [14]
Khafif et al. [13]
Petersen [12]
Zawawi (post-cochlear implant AM) [9] -review including Migirov study [10] Migirov (post-cochlear implant AM) [10] Quesnel et al. [11]
6/73 8%
Present study
5.1%
Number of rAM/AM cases
Author
Table 1 Comparison of reported cases of pediatric recurrent acute mastoiditis.
None
None
1/1 myringotomy and tube insertion No data
None
2 cases
None
None
None
None
2/2 mastoidectomy
No data
No data
No data
1/2 no data 1/2 IV antibiotics
2/3 medical treatment 1/3 antibiotics and tympanostomy
21/22 conservative treatment
No data
Second episode: 1-Streptococcus pneumoniae 1- no growth First episode: 1-Haemophilus influenzae
No data
Second episode: 1-Streptococcus pneumoniae 1-Anaerobes 4-No growth No data
No data
First episode: 6- Streptococcus pneumoniae 9- Pseudomonas aeruginosa 3-non-typable Haemophilus 15- anaerobes 4- no growth Second episode: 3-Streptococcus pneumoniae 5- Pseudomonas aeruginosa 2- non-typable Haemophilus 5- anaerobes 11- no growth No data
1/1 conservative treatment
None
No data
1/6 remastoidectomy 5/6 IV antibiotics and myringotomy
2/32 remastoidectomy 26/32 antibiotics and myringotomy 4/32 mastoidectomy 4/5 antibiotics 1/5 aspiration
1/22 mastoidectomy
First episode: 1-Streptococcus pneumoniae 4-Negative 1-Unknown Second episode: 4-Negative 1-Streptococcus pneumoniae 1-Candida albicans First episode: Pneumococci/Streptococci: 16/3
Isolated pathogen
1/1 remastoidectomy, posterior tympanotomy
Third episode management
1/6 medical 5/6 mastoidectomy, posterior tympanotomy
Second episode management
11/11
No data
No data
No data
(continued on next page)
None
None
None
None
None
None
No data
No data
2
3
No data
No data
None
No data
None
2/32
None
1
5/6
No data
Number of postcochlear implant patients
Number with history of recurrent acute otitis media
J. Mierzwiński et al.
International Journal of Pediatric Otorhinolaryngology 111 (2018) 142–148
International Journal of Pediatric Otorhinolaryngology 111 (2018) 142–148
2. Materials and methods
None
None
None
None
We sought to identify factors associated with rAM to inform the pathogenesis of the condition and provide support for interventions that might be utilized to prevent rAM in the pediatric age group. None
Number of postcochlear implant patients
J. Mierzwiński et al.
No data No data None 1/1 mastoidectomy and myringotomy 1/1 myringotomy
No data No data None No data
The study was completed in accordance with all procedures performed in studies involving human participants and in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards. A retrospective medical record review of all children treated for AM between 2001 and 2016 in the Department of Pediatric Otolaryngology, Audiology and Phoniatrics at the Children's Hospital of Bydgoszcz, Poland was performed. AM was defined clinically with typical symptoms including postauricular swelling, erythema, and tenderness requiring hospitalization and intensive treatment. rAM was defined as an episode of AM more than 4 weeks after a previous episode and in the absence of symptoms or findings to suggest persistent infection between episodes. This is in accordance with previous studies [6]. A new episode of AM or readmission to the hospital within 4 weeks was considered persistent AM. Patients with bilateral AM were considered a single case. Patients with AM symptoms lasting longer than 60 days without serious acute worsening were considered chronic and were excluded. Other exclusion criteria include AM resulting from an oncological process, chronic ear infection, cholesteatoma, or fibrous dysplasia. Data including the treatment modality applied, disease course, etiology, and preventative factors were collected. Children with second and/or third episodes of AM (rAM) were compared to solitary AM cases. The Children's Hospital of Bydgoszcz, Poland is the major tertiary referral center for complex pediatric cases in the region and patients suffering from AM or rAM would be expected to be referred to our facility. On admission we used sterile swab-sticks to collect a specimen of the ear discharge. In cases of surgical intervention, the pus was collected during surgery. All patients were treated according to a management algorithm. Empirically chosen IV antibiotics, typically third generation cephalosporins were frequently combined with clindamycin. If the clinical course was satisfactory, such treatment was sufficient. Patients with an unfavorable course of disease after 24–48 h qualified for subsequent treatment, comprising adjunctive drainage procedures such as myringotomy/tympanostomy or mastoidectomy to ensure adequate drainage of the mastoid and middle ear cavities. If subperiosteal abscess was diagnosed, the patient qualified for mastoidectomy. The only exception was AM after cochlear implantation wherein the abscess was lanced instead of performing mastoidectomy. During the study period, a change in surgical mastoidectomy technique occurred based upon the common finding of attic obstruction during mastoidectomy. The technique was changed from simple, cortical, mastoidectomy to wide mastoidectomy with broad attic exposure and posterior tympanotomy. Once implemented, all subsequent cases were treated with this approach. Follow-up data regarding otologic history was pursued via a survey that was sent to all patients with diagnosed AM during the study period. Categorical data were statistically analyzed using a chi-square test, Fisher's exact test and Mann-Whitney U test.
Abdel-Aziz et al. [21]
Nussinovitch et al. [20]
Migirov et al. [19]
Rosen et al. [18]
4/64 6% 2/49 4% 2/86 2% 1/19 5%
No data
No data No data None
No data No data 1/1 conservative treatment
1/4 surgical treatment 3/4 no data 2/2 reoperated
No data No data None 5/5 non-defined further management following protocol 5/167 3% Psarommatis et al. [17]
3/5 mastoidectomy 2/5 IV antibiotics, myringotomy/ tympanostomy 1/4 mastoidectomy 3/4 no data 2/2 surgical treatment
Third episode management Number of rAM/AM cases Author
Table 1 (continued)
First episode management of recurrent cases
Second episode management
Isolated pathogen
Number with history of recurrent acute otitis media
2.1. Clinical investigation
2.2. Literature assessment A systematic review of all articles published prior to April 2018 was performed through a comprehensive and stepwise search of the PubMed electronic database. The articles had to pertain to the occurrence of rAM in children. 144
International Journal of Pediatric Otorhinolaryngology 111 (2018) 142–148
J. Mierzwiński et al.
The investigation strategy was based on an advanced search with the following additional filters: language, English; age, birth through 18 years; included text words (in all fields), recurrent, new episode, second episode, second surgery, relapse, recurrence, recurring. Each phrase was combined with operator AND with textword mastoiditis (in all fields). We included publications discussing children 18 years or younger treated for AM after cochlear implants. We excluded articles reporting on patients over 18 years of age and those in which the age group was not identified. We also excluded articles with suspected duplicate data. Finally, commentaries, conference abstracts, letters, and replies were not considered eligible. Some papers were missed in this search strategy; however, these papers appear in this article. These reports were additionally chosen from the reference lists of the analyzed papers. The above-mentioned criteria with the additional age filter for children identified 130 English-language studies. All abstracts from the 130 results obtained from the database search were assessed. In case of doubt concerning study content, papers underwent additional screening. The full text of 24 articles were carefully assessed. Thirteen articles were excluded because of duplicate, irrelevant, or insufficient data. Five articles were added from reference screening. Sixteen articles were ultimately included for analysis and review comparison as summarized in Table 1.
Table 3 Patients with rAM - microbiological examination results.
Streptococcus pneumoniae Streptococcus pyogenes Pseudomonas aeruginosa Haemophilus influenzae Candida albicans Escherichia coli Acinetobacter baumannii Enterococcus faecalis Staphylococcus aureus Moraxella catarrhalis Negative
28 13 3 3 2 1 1 1 1 1 29
33.7% 15.7% 3.6% 3.6% 2.4% 1.2% 1.2% 1.2% 1.2% 1.2% 35%
Third episode:
1
Negative
None
2 3
Streptococcus pneumoniae Negative Negative
Negative Negative
4 5
Negative Negative
6
Unknown
Negative Streptococcus pneumoniae Candida albicans
None Streptococcus pneumoniae None None None
Negative Otologic History
Episodes of Acute otitis media but not Recurrent Acute otitis Media
Recurrent Acute Otitis Media (> 4 per year)
14 (48%)
8 (28%)
7 (24%)
1 (17%)
–
5 (83%)
4. Treatment RAM and primary AM were successfully treated with multiple different therapies. Treatment decisions are made on a case-by-case basis with reliance upon treatment algorithm as referenced in the methods section. In our series, the majority, 89% required procedural intervention. In 10 patients (12%) tympanostomy/myringotomy was performed, 8 patients (10%) underwent mastoidectomy only, 56 (67%) mastoidectomy with tympanostomy. 9 patients (11%) received pharmacological treatment only. All patients receiving surgical therapy were also treated with systemic antibiotics. Among those receiving mastoidectomy (64), 51 received cortical mastoidectomy only. The attic area, was usually found to be obliterated with granulation tissue, impairing communication between the middle ear and mastoid cavity. For the subsequent 13 mastoidectomies for AM, a modified approach was taken more extensive surgical communication during mastoidectomy (wide mastoidectomy with broad attic exposure and posterior tympanotomy). There were no surgical complications using either of the approaches. The follow-up for more aggressive treatment vs standard mastoidectomy is shorter, however it ranges from 20 months to 5 years in patients with posterior tympanotomy and 5–17 years in standard mastoidectomy group. We have noticed in the follow-up, that in patients where mastoidectomy was required - primary treatment of AM with the more complete surgical approach including attic exposure and posterior tympanotomy prevented recurrent episodes of AM, in this group no rAM occurred. In a group of 51 patients who undergone simple mastoidectomies – 5 rAM occurred. However, the sample is too small for statistical analysis. rAM occurred in 6 patients. The time from the first AM to second episode ranged from 6 weeks to 3 years.
Table 2 Microbiological examination results. (%)
Second episode:
Acute Mastoiditis (n = 67) Recurrent Acute Mastoiditis (n = 6)
From 2001 to 2017, 83 ears in 73 children were diagnosed with acute mastoiditis. Three cases were excluded per the methodological criteria. Seventy three cases were primary AM (35 boys, 38 girls). Sex predominance was not statistically significant (p > 0.05). Six patients (8%) developed a single rAM episode, five in boys and one in a girl. One patient developed two episodes of rAM. The mean hospitalization duration among AM patients was 11.34 days, compared to a mean hospitalization time among rAM patients of 10.71 days, this was not statistically significant (p > 0.05). Microbiological examination results are shown in Table 2 and Table 3. The predominant pathogen in AM group was Streptococcus pneumoniae (p < 0.05). In a group of patients with rAM (N = 6), the only pathogen isolated in one patient at first AM episode was Streptococcus pneumoniae. Four cultures were negative. At second episode Streptococcus pneumoniae and Candida albicans were isolated. Four cultures were negative again (Table 3). Streptococcus pneumoniae was the only bacteria isolated in this group, but the sample was too small to perform statistical analysis. The follow-up questionnaire achieved a response rate of 48% (Table 4). AM was often the first episode of otologic disease. Recurrent acute otitis media (> 4 episodes per year preceding AM) is far more common among patients with rAM than those with isolated AM (p < 0.05) (Table 4). The positive predictive value of recurrent acute otitis media for rAM in this series is 41.7%.
Number of ears (N = 83)
First episode:
Table 4 Otologic history prior to initial AM episode. Response was received from 35 of 73 patients. History of recurrent acute otitis media (AOM) prior to initial acute mastoiditis significantly predicts rAM (p < 0.05).
3. Results
Isolated pathogen
No.
- One rAM patient previously treated with simple mastoidectomy achieved resolution of recurrence with IV antibiotics alone. - One patient successfully treated with antibiotics initially, underwent more aggressive mastoidectomy to treat recurrence. - Four patients with rAM previously treated with simple 145
International Journal of Pediatric Otorhinolaryngology 111 (2018) 142–148
J. Mierzwiński et al.
remained unchanged as well as the number of mastoidectomies [25]. On the other hand, Tawfik et al. found that introduction of two pneumococcal vaccines did not decrease rates of hospitalization for pediatric AM between 2000 and 2012. Between 2009 and 2012 in children aged 0–2 years of age declining hospitalization rates, possibly reflecting the protective benefit of the 13-valent pneumococcal vaccine were observed [26]. It will be interesting to see whether pneumococcal vaccination will decrease the incidence of AM and rAM in our country after introduction into the national immunization programme, but it is still too early to draw any conclusions. In our whole group of children with rAM at first episode of AM 80% of samples were negative, at second episode (rAM) 60% (Table 3). All published studies pertaining to the microbiology rAM shows that only some percent of such cultures exhibit bacterial growth [8,11]. In study by Gavriel et al. at first episode of AM 11% of samples were negative, at second episode 42% [8]. Quesnel in his study described 8 a.m. recurrences with 44% of negative samples [11]. Potential higher cultures negativity in our rAM group may be due to the fact, that almost all the children have received intensive, sometimes multiple antibiotic therapy prior to hospitalization of children for AM. Management of AM included administering empiric intravenous antibiotics shortly after the admission in every single case. Patients with an unfavorable course of disease after 24–48 h qualified for subsequent treatment, comprising adjunctive drainage procedures such as myringotomy/tympanostomy or mastoidectomy. However, in our series, the majority of patients, 89% required procedural intervention. In 16 studies with reported cases of rAM only in 5 of them results of the microbiologic growth in rAM group were described (Table 1).
mastoidectomy underwent more aggressive mastoidectomy for rAM and 3 of them had no further episodes. One remaining patient who received a second cortical mastoidectomy for recurrence developed a second recurrence of AM. This was treated with more complete surgical opening and silastic sheeting inserted in the attic and facial recess. The most common complication in the AM group was sigmoid sinus thrombosis (5%), followed by epidural abscess, petrositis, and facial paresis (1% each). Subperiosteal abscess was present in 22% of AM patients (16 cases) and in 33% of rAM patients (2 cases). There were no intratemporal/ intracranial complications associated with rAM compared to 9.6% rate in AM (p < 0.05). There was no progression of intracranial complications after treatment initiation. There were no patient deaths in the series. 5. Discussion This report is the first to our knowledge to identify features that predict recurrent acute mastoiditis and implement a treatment that reduces rAM episodes. The 8% recurrence rate identified in this series falls within the expected frequency reported by others in the post-antibiotic era [6,10,20]. 5.1. Isolation of some pathogens at first episode of AM is considered a prognostic factor related to recurrence The predominant bacterial species present in all AM children during the study period was Streptococcus pneumoniae (33.7% of cases) (p < 0.05). Other authors have reported similar observations and rates [2,12,16,22,23]. The only bacteria isolated from rAM cases in our patients was Streptococcus pneumoniae. This differs from Gavriel et al. who found that recurrence is most common among patients with Pseudomonas aeruginosa, anaerobic bacteria, and polymicrobial infection cultured from an AM infection [8]. Since the most frequent bacteria was Streptococcus pneumoniae, theoretically the vaccination should prevent acute otitis media episodes and reduce the number of AM and rAM in the pediatric population. Since January 1, 2017 the vaccination against S. pneumoniae has been obligatory in Poland. Our study included all children treated for AM between 2001 and 2016, when mass vaccination had not been obligatorily implemented. According to Polish demographic data, the total Polish population includes about 6 million children (5.771.400 people aged 0–14 years). PCV7 was registered in Poland in 2001, but in fact the vaccine became available in 2005 [24]. PCV10 was registered in April 2009 [24]. Antipneumococcal conjugated vaccines were recommended, but only voluntary. According to data provided by Pfizer Poland, in the years 2005, 2006, 2007, 2008 and 2009, 4000, 41.230, 153.986, 224.825 and 268.833 doses of PCV7, respectively were sold. Since April 2009 54.033 doses of PCV10 have been sold [data from GlaxoSmithKline Poland] [24]. As we can see from the data above, not many children were vaccinated in general population. In our group of 73 children only few children have received voluntary vaccines, making this group too small for comparative purposes. The more - vaccines received by our patients were not homogenous. Pneumococcal conjugate vaccines available at that time contained different serotypes - PCV7 or PCV10 making this statistics even more confusing. However there are several literature data on pneumococcal vaccination and the rate of mastoiditis in pediatric population. According to Cavel et al. the incidence of AM is lower in the post-PCV13 period compared to the pre-PCV period. The rate of AM complications
5.2. Recurrent episodes of acute otitis media are considered another predictive factor of recurrence For many patients, the initial AM presentation was the first manifestation of otologic disease. Among those with history of recurrent acute otitis media prior to the initial AM case, there is a significantly greater risk of developing rAM in the future [16]. This is consistent with the current understanding of AM wherein acute otitis media progresses to AM. As evidenced by our series and described by others, there is not always a sustained connection between the middle ear cleft and the mastoid in the setting of infection so once infection occurs, intervention might need to address the middle ear and mastoid cavities as two anatomically separated spaces [27]. Some have even hypothesized resolution of acute otitis media in the context of persistent acute mastoiditis. The end of the window of risk for rAM episodes has yet to be well characterized. Gavriel reported a median duration between the first and second episode of only 2 months [8]. We found a longer range of 6 weeks up to 3 years. While, some guidelines suggest that expectant management is appropriate for select acute otitis media, we feel that the greatly elevated risk of rAM among patients with history of AM warrants consideration of earlier and more aggressive treatment of subsequent acute otitis media with antibiotics. 5.3. Surgical treatment is also discussed as suggested contributing factor in rAM Partly through the series, a change in surgical approach from simple mastoidectomy to more aggressive initial surgical approach with broad attic exposure and posterior tympanotomy was conducted to address anatomic obstruction at the attic. Theoretically the surgical connections can be secondarily obliterated by granulation tissue and bone prior to subsequent episodes of AM, however, none of the patients undergoing this procedure at initial mastoidectomy had AM recurrence. In a group of patients treated for the first episode of AM with simple mastoidectomy only – 5 recurrences were observed. The approach is 146
International Journal of Pediatric Otorhinolaryngology 111 (2018) 142–148
J. Mierzwiński et al.
6. Conclusions
consistent with the proposal of broad mastoidectomy including facial recess exposure by Linder [28]. However, the samples of patients are too small to perform the statistical analysis that will allow us to compare groups with and without posterior tympanotomy performed to determine if creating large communication with additional posterior tympanotomy indeed prevented rAM. We assume that this approach could be protective against rAM, but future studies with larger groups will permit stronger conclusions. Others, including Groth et al. suggested that mastoidectomy for AM may be a predisposing factor for rAM [6]. Luntz et al. contest this opinion, finding that a history of recurrent otitis media prior to a first episode of AM is a predisposing factor for rAM, but not treatment mode [16]. Vasbotn et al. emphasized that AM patients treated with mastoidectomy have a more severe course of disease than other AM patients [2]. In a 10-year observation period, Khafif et al. collected a group of 134 patients with AM [13]. They observed recurrent episodes in 11 patients and found that the recurrence rate was not related to the mode of treatment. Lahav reported recurrence in seven of 78 rAM patients; four of these had undergone mastoidectomy, one received medical treatment, and three underwent abscess aspiration [7]. We think that the recurrence rate in the mastoidectomy group may be higher than that in the medical treatment group, however, they are two different populations. AM requiring surgical intervention represents a more advanced state of infection and includes children in whom medical treatment was ineffective. Therefore, implication that initial surgery is a causative factor for recurrence is not appropriate. Our series includes also one patient with prior cochlear implantation. The patient was successfully treated with lancing the site. Others report that mastoiditis occurs in about 1% of patients after cochlear implant (CI) recipients. It has been shown that second and third AM episodes may also occur in this patient group. Zawawi reported AM in three out of 43 CI patients; Migirov reported AM in two out of 11 [9,10]. Migirov analyzed 234 children with CI; 116 of these underwent surgery via the transmastoid mastoidectomy, posterior tympanotomy approach and 118 via the suprameatal approach [10]. Post-CI AM developed in 11 cases, all of them in the group receiving mastoidectomy with posterior tympanotomy approach. In two cases Migirov observed AM recurrence, which was treated with incision and abscess drainage. The finding that patients are more vulnerable to AM after mastoidectomy than after suprameatal surgery might be extrapolated to indicate that children are more vulnerable to AM after mastoidectomy. We contest this supposition, however, as the surgeon typically secondarily obliterates the facial recess during the cochlear implantation procedure and the additional communication pathway is closed again.
- AM occurs with incidence of 0.6–4.8 per 100,000 per year. Among these cases of AM, rAM accounts for 2–18%. - Streptococcus pneumoniae is the most common pathogen isolated in our series of AM and the only bacteria isolated from cases of rAM. Most children were unvaccinated against S. pneumoniae species. - History of recurrent acute otitis media prior to the first episode of AM is a significant risk factor for development of rAM. - Wider surgical connection of the mastoid and middle ear with attic exposure and posterior tympanotomy was protective against subsequent episodes of AM in this series, but future studies with larger groups of patients will permit stronger conclusions. - Patients with rAM present with fewer intratemporal and intracranial complications than those with AM. Conflicts of interest The authors disclose no conflicts of interest. Formatting of funding sources This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors. Declarations of interest None. Consent for publication and ethics approval The study was completed in accordance with all procedures performed in studies involving human participants and in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards. Consent for publication: written informed consent has been obtained from the parents of the patients and from patients (if regarded as competent). Bioethics committee permission number KB 582/2017. Bioethics Committee of the Nicolaus Copernicus University in Toruń functioning at Collegium Medicum in Bydgoszcz. Authors' contributions
5.4. Another suggested contributing factor in rAM is immunodeficiency secondary to IG4-related disease [29,30,31]
- Study concept and design - Józef Mierzwiński - Acquisition of data: Justyna Tyra, Karolina Haber, Michael David Puricelli - Analysis and interpretation of data: Józef Mierzwiński, Anna Sinkiewicz, Justyna Tyra, Karolina Haber, Michael David Puricelli - Drafting of the manuscript: Józef Mierzwiński, Anna Sinkiewicz, Justyna Tyra, Karolina Haber, Michael David Puricelli - Critical revision of the manuscript for important intellectual content: Józef Mierzwiński - Statistical analysis: Justyna Tyra - Administrative, technical, and material support: Anna Sinkiewicz, Karolina Haber, Michael David Puricelli - Study supervision: Józef Mierzwiński
Specific investigation into the immune status of our patients with rAM was not made, however, this may be considered in patients with rAM. Finally, we find significantly fewer intratemporal and extratemporal complications among patients with rAM. We believe that rAM cases are susceptible to such complications and theorize that the reduction could be secondary to increased awareness and earlier presentation. Nevertheless, secondary episodes are associated with prolonged hospitalization and effort should be made to reduce the frequency whenever possible. There are several limitations to the data presented. Most significantly, AM and rAM are rare conditions and there is limited sample size of patients available. Future multi-institutional collaboration would be expected to yield greater numbers and permit stronger conclusions. We expect follow-up within our institution but recognize that patients could have been seen elsewhere biasing our findings. We attempted to mitigate this by secondarily contacting the patients.
Acknowledgements The authors would like to express thanks to Prof. Leszek Knopik for his invaluable help in statistical analysis. 147
International Journal of Pediatric Otorhinolaryngology 111 (2018) 142–148
J. Mierzwiński et al.
References [1] P. Homøe, K. Kværner, J.R. Casey, R.A.M.J. Damoiseaux, T.M.A. van Dongen, H. Gunasekera, R.G. Jensen, E. Kvestad, P.S. Morris, H.M. Weinreich, Panel 1: epidemiology and diagnosis, Otolaryngol. Head Neck Surg. 156 (2017) S1–S21. [2] F.S. Vassbotn, O.G. Klausen, O. Lind, P. Moller, Acute mastoiditis in a Norwegian population: a 20 year retrospective study, Int. J. Pediatr. Otorhinolaryngol. 62 (2002) 237–242. [3] D.A. Van Zuijlen, A.G. Schilder, F.A. Van Balen, A.W. Hoes, National differences in incidence of acute mastoiditis: relationship to prescribing patterns of antibiotics for acute otitis media? Pediatr. Infect. Dis. J. 20 (2001) 140–144. [4] H. Mygind, The subperiosteal abscess of the mastoid region, Ann. Otol. Rhinol. Laryngol. 19 (1910) 529–540. [5] H.P. House, Otitis media; a comparative study of the results obtained in therapy before and after the introduction of the sulfonamide compounds, Arch. Otolaryngol. 43 (1946) 371–378. [6] A. Groth, F. Enoksson, J. Stalfors, K. Stenfeldt, M. Hultcrantz, A. Hermansson, Recurrent acute mastoiditis - a retrospective national study in Sweden, Acta Otolaryngol. 132 (2012) 1275–1281. [7] J. Lahav, O. Handzel, R. Gertler, M. Yehuda, D. Halperin, Postauricular needle aspiration of subperiosteal abscess in acute mastoiditis, Ann. Otol. Rhinol. Laryngol. 114 (2005) 323–327. [8] H. Gavriel, R.A. Eta, E. Eviatar, The role of bacteriologic studies in predicting recurrent mastoiditis in 456 patients, Otol. Neurotol. 37 (2016) 1059–1062. [9] F. Zawawi, I. Cardona, O.V. Akinpelu, S.J. Daniel, Acute mastoiditis in children with cochlear implants: is explantation required? Otolaryngol. Head Neck Surg. 151 (2014) 394–398. [10] L. Migirov, A. Yakirevitch, Y. Henkin, R. Kaplan-Neeman, J. Kronenberg, Acute otitis media and mastoiditis following cochlear implantation, Int. J. Pediatr. Otorhinolaryngol. 70 (2006) 899–903. [11] S. Quesnel, M. Nguyen, S. Pierrot, P. Contencin, Y. Manach, V. Couloigner, Acute mastoiditis in children: a retrospective study of 188 patients, Int. J. Pediatr. Otorhinolaryngol. 74 (2010) 1388–1392. [12] C.G. Petersen, T. Ovesen, C.B. Pedersen, Acute mastoidectomy in a Danish county from 1977 to 1997-operative findings and long-term results, Acta Otolaryngol. Suppl. 543 (2000) 122–126. [13] A. Khafif, D. Halperin, I. Hochman, R. Gertler, I. Poria, D. Shindel, G. Marshak, Acute mastoiditis: a 10-year review, Am. J. Otolaryngol. 19 (1998) 170–173. [14] F. Glynn, L. Osman, M. Colreavy, H. Rowley, T.P.O. Dwyer, A. Blayney, Acute mastoiditis in children: presentation and long term consequences, J. Laryngol. Otol. 122 (2008) 233–237. [15] E.H. Harley, T. Sdralis, R.G. Berkowitz, Acute mastoiditis in children: a 12-year retrospective study, Otolaryngol. Head Neck Surg. 116 (1997) 26–30. [16] M. Luntz, A. Brodsky, S. Nusem, J. Kronenberg, G. Keren, L. Migirov, D. Cohen,
[17]
[18] [19] [20]
[21] [22]
[23]
[24]
[25]
[26]
[27]
[28] [29]
[30] [31]
148
S. Zohar, A. Shapira, D. Ophir, G. Fishman, G. Rosen, V. Kisilevsky, I. Magamse, S. Zaaroura, H.Z. Joachims, D. Goldenberg, Acute mastoiditis-the antibiotic era: a multicenter study, Int. J. Pediatr. Otorhinolaryngol. 57 (2001) 1–9. I.M. Psarommatis, C. Voudouris, K. Douros, P. Giannakopoulos, T. Bairamis, C. Carabinos, Algorithmic management of pediatric acute mastoiditis, Int. J. Pediatr. Otorhinolaryngol. 76 (2012) 791–796. A. Rosen, D. Ophir, G. Marshak, Acute mastoiditis: a review of 69 cases, Ann. Otol. Rhinol. Laryngol. 95 (1986) 222–224. L. Migirov, A. Yakirevitch, J. Kronenberg, Mastoid subperiosteal abscess: a review of 51 cases, Int. J. Pediatr. Otorhinolaryngol. 69 (2005) 1529–1533. M. Nussinovitch, R. Yoeli, K. Elishkevitz, I. Varsano, Acute mastoiditis in children: epidemiologic, clinical, microbiologic, and therapeutic aspects over past years, Clin. Pediatr. 43 (2004) 261–267. M. Abdel-Aziz, H. El-Hoshy, Acute mastoiditis: a one year study in the pediatric hospital of Cairo university, BMC Ear Nose Throat Disord. 10 (2010) 1. A. Groth, F. Enoksson, A. Hermansson, M. Hultcrantz, J. Stalfors, K. Stenfeldt, Acute mastoiditis in children in Sweden 1993-2007-no increase after new guidelines, Int. J. Pediatr. Otorhinolaryngol. 75 (2011) 1496–1501. L.H.Y. Pang, M.S. Barakate, T.E. Havas, Mastoiditis in a paediatric population: a review of 11 years experience in management, Int. J. Pediatr. Otorhinolaryngol. 73 (2009) 1520–1524. A. Skoczyńska, E. Sadowy, K. Bojarska, J. Strzelecki, A. Kuch, A. Gołbiewska, I. Waśko, M. Foryś, M. van der Linden, W. Hryniewicz, The current status of invasive pneumococcal disease in Poland, Vaccine 29 (2011) 2199–2205. O. Cavel, R. Tauman, E. Simsolo, D. Yafit, E. Reindorf-Kfir, O. Wasserzug, O. Unger, O. Handzel, G. Fishman, Y. Oestreicher-Kedem, A. DeRowe, Changes in the epidemiology and clinical features of acute mastoiditis following the introduction of the pneumococcal conjugate vaccine, Int. J. Pediatr. Otorhinolaryngol. 104 (2018) 54–57. K.O. Tawfik, S.L. Ishman, M.E. Tabangin, M. Altaye, J. Meinzen-Derr, D.I. Choo, Pediatric acute mastoiditis in the era of pneumococcal vaccination, Laryngoscope 128 (6) (2018 Jun) 1480–1485. P. Schachern, M.M. Paparella, S. Sano, S. Lamey, Y. Guo, A histopathological study of the relationship between otitis media and mastoiditis, Laryngoscope 101 (1991) 1050–1055. T.E. Linder, H.R. Briner, T. Bischoff, Prevention of acute mastoiditis: fact or fiction? Int. J. Pediatr. Otorhinolaryngol. 56 (2000) 129–134. V. Deshpande, N.A. Zane, S. Kraft, J.H. Stone, W.C. Faquin, Recurrent mastoiditis mimics IgG4 related disease: a potential diagnostic pitfall, Head Neck Pathol. 10 (3) (2016 Sep) 314–320. V. Deshpande, Y. Zen, J.K. Chan, E.E. Yi, Y. Sato, et al., Consensus statement on the pathology of IgG4-related disease, Mod. Pathol. 25 (2012) 1181–1192. A. Sebastian, Choroba związana z IgG4 – czy to całkiem nowa choroba? Biul. Sekc. Młodych Reumatol. (2014) PTR 10/2014.
Contents lists available at ScienceDirect
International Journal of Pediatric Otorhinolaryngology journal homepage: www.elsevier.com/locate/ijporl
Pediatric recurrent acute mastoiditis: Risk factors and insights into pathogenesis
T
Józef Mierzwińskia,∗, Justyna Tyraa, Karolina Habera, Maria Drelaa, Anna Sinkiewiczb, Michael David Puricellic a Department of Otolaryngology, Audiology and Phoniatrics, Pediatric Cochlear Implant Center, Children's Hospital of Bydgoszcz, Poland, Chodkiewicza 44, 85-667, Bydgoszcz, Poland b Department of Phoniatrics and Audiology, University Hospital of Bydgoszcz, Poland c Paparella Ear, Head & Neck Institute, 701 25th Ave. S. Suite 200, Minneapolis, MN, 55454, United States
A R T I C LE I N FO
A B S T R A C T
Keywords: Child Mastoiditis Otitis media Recurrence Operative therapy Mastoidectomy
Objectives: Recurrent acute mastoiditis is repeatedly reported in the literature, but data to understand the pathogenesis, update treatment recommendations and inform future trials are sparse due to the infrequency of the disease. Methods: A retrospective chart review from 2001 to 2016 was conducted including 73 children treated for acute mastoiditis. A follow-up survey was attempted for each patient. Bacteriology, method of treatment, hospital course, complications, and otologic history were analyzed. A chi-squared test, Fisher's exact test and MannWhitney U test compared recurrent acute mastoiditis to single acute mastoiditis cases. Additionally, a comprehensive PubMed search and review of world literature addressing recurrent pediatric acute mastoiditis was performed for comparative purposes. Results: Among 73 children with acute mastoiditis, six (8%) experienced recurrent acute mastoiditis. Streptococcus pneumoniae was the only bacteria isolated in this group. History of recurrent acute otitis media (> 4 per year) prior to the first episode of acute mastoiditis was identified in 24% with single episode of acute mastoiditis and 83% with recurrent mastoiditis (p < 0.05). Fewer intracranial/intratemporal complications were identified among recurrent mastoiditis patients (p < 0.05). In a group of patients treated with more extensive surgical communication during mastoidectomy for primary acute mastoiditis (wide mastoidectomy with broad attic exposure and posterior tympanotomy) no recurrence was observed. Conclusion: We identify multiple risk factors associated with recurrence and provide early data supporting anatomic predisposition to the development of recurrent acute mastoiditis. More aggressive opening between the mastoid cavity and middle ear may prevent recurrent acute mastoiditis episodes.
1. Introduction Acute mastoiditis (AM) is a middle ear infection that extends to the mastoid cells and involves the mastoid bone, leading to periostitis and/ or osteitis. The major hypothesized cause of AM is lack of communication between the tympanic and mastoid cavities, preventing pus drainage through the Eustachian tube and/or tympanic membrane and leading to mastoid osteolysis and abscess formation. Although many studies have examined the etiology and treatment options for pediatric acute mastoiditis, little has been published about recurrent acute mastoiditis (rAM). Primary AM has a reported incidence of 0.6–4.8 per 100,000 per year [1,2,3]. Recurrent episodes of mastoiditis are rare and account for
∗
Corresponding author. E-mail address: [email protected] (J. Mierzwiński).
https://doi.org/10.1016/j.ijporl.2018.06.001 Received 4 February 2018; Received in revised form 31 May 2018; Accepted 2 June 2018 Available online 06 June 2018 0165-5876/ © 2018 Elsevier B.V. All rights reserved.
only a few percent of all AM cases, but are constantly present in major series of AM reported in the world literature, what makes them much more common that it is generally thought. rAM was first reported in 1910 b y Mygind, who found that 4% of AM patients experienced recurrence [4]. William House reported recurrence rates of up to 20% in the pre-antibiotic era [5]. In the post-antibiotic era, rates of rAM range from 2 to 18%, with most series finding frequency of 5–10%. Variable treatments have been successfully applied ranging from antibiotics alone; myringotomy and tube placement with antibiotics; and mastoidectomy with tympanostomy tube placement and antibiotics (Table 1). Despite recognition of rAM as an entity, there has been little published regarding the mechanism of development or factors that might predict or prevent rAM.
32/798 4%
5/78 6%
22/434
Groth et al. [6]
Lahav et al. [7]
Gavriel et al. [8]
143 No data
8/188 4%
1/57 2%
11/223 5%
Luntz et al. [16]
1/1 IV antibiotics, myringotomy and tube insertion, mastoidectomy No data
4/11 surgical treatment 7/11 conservative treatment 2/2 antibiotics
3/3 mastoidectomy
1/2 no data 1/2 incision and drainage
2/11 18%
3/79 4% 11/134 8% 2/29 7%
No data
2/5 aspiration 3/5 mastoidectomy 5/5 IV antibiotics and myringotomy 20/22 IV antibiotics and myringotomy 2/22 mastoidectomy
24/32 mastoidectomy 8/32 IV antibiotics and myringotomy
5/6 mastoidectomy 1/6 conservative treatment
First episode management of recurrent cases
3/43 7%
Harley et al. [15]
Glynn et al. [14]
Khafif et al. [13]
Petersen [12]
Zawawi (post-cochlear implant AM) [9] -review including Migirov study [10] Migirov (post-cochlear implant AM) [10] Quesnel et al. [11]
6/73 8%
Present study
5.1%
Number of rAM/AM cases
Author
Table 1 Comparison of reported cases of pediatric recurrent acute mastoiditis.
None
None
1/1 myringotomy and tube insertion No data
None
2 cases
None
None
None
None
2/2 mastoidectomy
No data
No data
No data
1/2 no data 1/2 IV antibiotics
2/3 medical treatment 1/3 antibiotics and tympanostomy
21/22 conservative treatment
No data
Second episode: 1-Streptococcus pneumoniae 1- no growth First episode: 1-Haemophilus influenzae
No data
Second episode: 1-Streptococcus pneumoniae 1-Anaerobes 4-No growth No data
No data
First episode: 6- Streptococcus pneumoniae 9- Pseudomonas aeruginosa 3-non-typable Haemophilus 15- anaerobes 4- no growth Second episode: 3-Streptococcus pneumoniae 5- Pseudomonas aeruginosa 2- non-typable Haemophilus 5- anaerobes 11- no growth No data
1/1 conservative treatment
None
No data
1/6 remastoidectomy 5/6 IV antibiotics and myringotomy
2/32 remastoidectomy 26/32 antibiotics and myringotomy 4/32 mastoidectomy 4/5 antibiotics 1/5 aspiration
1/22 mastoidectomy
First episode: 1-Streptococcus pneumoniae 4-Negative 1-Unknown Second episode: 4-Negative 1-Streptococcus pneumoniae 1-Candida albicans First episode: Pneumococci/Streptococci: 16/3
Isolated pathogen
1/1 remastoidectomy, posterior tympanotomy
Third episode management
1/6 medical 5/6 mastoidectomy, posterior tympanotomy
Second episode management
11/11
No data
No data
No data
(continued on next page)
None
None
None
None
None
None
No data
No data
2
3
No data
No data
None
No data
None
2/32
None
1
5/6
No data
Number of postcochlear implant patients
Number with history of recurrent acute otitis media
J. Mierzwiński et al.
International Journal of Pediatric Otorhinolaryngology 111 (2018) 142–148
International Journal of Pediatric Otorhinolaryngology 111 (2018) 142–148
2. Materials and methods
None
None
None
None
We sought to identify factors associated with rAM to inform the pathogenesis of the condition and provide support for interventions that might be utilized to prevent rAM in the pediatric age group. None
Number of postcochlear implant patients
J. Mierzwiński et al.
No data No data None 1/1 mastoidectomy and myringotomy 1/1 myringotomy
No data No data None No data
The study was completed in accordance with all procedures performed in studies involving human participants and in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards. A retrospective medical record review of all children treated for AM between 2001 and 2016 in the Department of Pediatric Otolaryngology, Audiology and Phoniatrics at the Children's Hospital of Bydgoszcz, Poland was performed. AM was defined clinically with typical symptoms including postauricular swelling, erythema, and tenderness requiring hospitalization and intensive treatment. rAM was defined as an episode of AM more than 4 weeks after a previous episode and in the absence of symptoms or findings to suggest persistent infection between episodes. This is in accordance with previous studies [6]. A new episode of AM or readmission to the hospital within 4 weeks was considered persistent AM. Patients with bilateral AM were considered a single case. Patients with AM symptoms lasting longer than 60 days without serious acute worsening were considered chronic and were excluded. Other exclusion criteria include AM resulting from an oncological process, chronic ear infection, cholesteatoma, or fibrous dysplasia. Data including the treatment modality applied, disease course, etiology, and preventative factors were collected. Children with second and/or third episodes of AM (rAM) were compared to solitary AM cases. The Children's Hospital of Bydgoszcz, Poland is the major tertiary referral center for complex pediatric cases in the region and patients suffering from AM or rAM would be expected to be referred to our facility. On admission we used sterile swab-sticks to collect a specimen of the ear discharge. In cases of surgical intervention, the pus was collected during surgery. All patients were treated according to a management algorithm. Empirically chosen IV antibiotics, typically third generation cephalosporins were frequently combined with clindamycin. If the clinical course was satisfactory, such treatment was sufficient. Patients with an unfavorable course of disease after 24–48 h qualified for subsequent treatment, comprising adjunctive drainage procedures such as myringotomy/tympanostomy or mastoidectomy to ensure adequate drainage of the mastoid and middle ear cavities. If subperiosteal abscess was diagnosed, the patient qualified for mastoidectomy. The only exception was AM after cochlear implantation wherein the abscess was lanced instead of performing mastoidectomy. During the study period, a change in surgical mastoidectomy technique occurred based upon the common finding of attic obstruction during mastoidectomy. The technique was changed from simple, cortical, mastoidectomy to wide mastoidectomy with broad attic exposure and posterior tympanotomy. Once implemented, all subsequent cases were treated with this approach. Follow-up data regarding otologic history was pursued via a survey that was sent to all patients with diagnosed AM during the study period. Categorical data were statistically analyzed using a chi-square test, Fisher's exact test and Mann-Whitney U test.
Abdel-Aziz et al. [21]
Nussinovitch et al. [20]
Migirov et al. [19]
Rosen et al. [18]
4/64 6% 2/49 4% 2/86 2% 1/19 5%
No data
No data No data None
No data No data 1/1 conservative treatment
1/4 surgical treatment 3/4 no data 2/2 reoperated
No data No data None 5/5 non-defined further management following protocol 5/167 3% Psarommatis et al. [17]
3/5 mastoidectomy 2/5 IV antibiotics, myringotomy/ tympanostomy 1/4 mastoidectomy 3/4 no data 2/2 surgical treatment
Third episode management Number of rAM/AM cases Author
Table 1 (continued)
First episode management of recurrent cases
Second episode management
Isolated pathogen
Number with history of recurrent acute otitis media
2.1. Clinical investigation
2.2. Literature assessment A systematic review of all articles published prior to April 2018 was performed through a comprehensive and stepwise search of the PubMed electronic database. The articles had to pertain to the occurrence of rAM in children. 144
International Journal of Pediatric Otorhinolaryngology 111 (2018) 142–148
J. Mierzwiński et al.
The investigation strategy was based on an advanced search with the following additional filters: language, English; age, birth through 18 years; included text words (in all fields), recurrent, new episode, second episode, second surgery, relapse, recurrence, recurring. Each phrase was combined with operator AND with textword mastoiditis (in all fields). We included publications discussing children 18 years or younger treated for AM after cochlear implants. We excluded articles reporting on patients over 18 years of age and those in which the age group was not identified. We also excluded articles with suspected duplicate data. Finally, commentaries, conference abstracts, letters, and replies were not considered eligible. Some papers were missed in this search strategy; however, these papers appear in this article. These reports were additionally chosen from the reference lists of the analyzed papers. The above-mentioned criteria with the additional age filter for children identified 130 English-language studies. All abstracts from the 130 results obtained from the database search were assessed. In case of doubt concerning study content, papers underwent additional screening. The full text of 24 articles were carefully assessed. Thirteen articles were excluded because of duplicate, irrelevant, or insufficient data. Five articles were added from reference screening. Sixteen articles were ultimately included for analysis and review comparison as summarized in Table 1.
Table 3 Patients with rAM - microbiological examination results.
Streptococcus pneumoniae Streptococcus pyogenes Pseudomonas aeruginosa Haemophilus influenzae Candida albicans Escherichia coli Acinetobacter baumannii Enterococcus faecalis Staphylococcus aureus Moraxella catarrhalis Negative
28 13 3 3 2 1 1 1 1 1 29
33.7% 15.7% 3.6% 3.6% 2.4% 1.2% 1.2% 1.2% 1.2% 1.2% 35%
Third episode:
1
Negative
None
2 3
Streptococcus pneumoniae Negative Negative
Negative Negative
4 5
Negative Negative
6
Unknown
Negative Streptococcus pneumoniae Candida albicans
None Streptococcus pneumoniae None None None
Negative Otologic History
Episodes of Acute otitis media but not Recurrent Acute otitis Media
Recurrent Acute Otitis Media (> 4 per year)
14 (48%)
8 (28%)
7 (24%)
1 (17%)
–
5 (83%)
4. Treatment RAM and primary AM were successfully treated with multiple different therapies. Treatment decisions are made on a case-by-case basis with reliance upon treatment algorithm as referenced in the methods section. In our series, the majority, 89% required procedural intervention. In 10 patients (12%) tympanostomy/myringotomy was performed, 8 patients (10%) underwent mastoidectomy only, 56 (67%) mastoidectomy with tympanostomy. 9 patients (11%) received pharmacological treatment only. All patients receiving surgical therapy were also treated with systemic antibiotics. Among those receiving mastoidectomy (64), 51 received cortical mastoidectomy only. The attic area, was usually found to be obliterated with granulation tissue, impairing communication between the middle ear and mastoid cavity. For the subsequent 13 mastoidectomies for AM, a modified approach was taken more extensive surgical communication during mastoidectomy (wide mastoidectomy with broad attic exposure and posterior tympanotomy). There were no surgical complications using either of the approaches. The follow-up for more aggressive treatment vs standard mastoidectomy is shorter, however it ranges from 20 months to 5 years in patients with posterior tympanotomy and 5–17 years in standard mastoidectomy group. We have noticed in the follow-up, that in patients where mastoidectomy was required - primary treatment of AM with the more complete surgical approach including attic exposure and posterior tympanotomy prevented recurrent episodes of AM, in this group no rAM occurred. In a group of 51 patients who undergone simple mastoidectomies – 5 rAM occurred. However, the sample is too small for statistical analysis. rAM occurred in 6 patients. The time from the first AM to second episode ranged from 6 weeks to 3 years.
Table 2 Microbiological examination results. (%)
Second episode:
Acute Mastoiditis (n = 67) Recurrent Acute Mastoiditis (n = 6)
From 2001 to 2017, 83 ears in 73 children were diagnosed with acute mastoiditis. Three cases were excluded per the methodological criteria. Seventy three cases were primary AM (35 boys, 38 girls). Sex predominance was not statistically significant (p > 0.05). Six patients (8%) developed a single rAM episode, five in boys and one in a girl. One patient developed two episodes of rAM. The mean hospitalization duration among AM patients was 11.34 days, compared to a mean hospitalization time among rAM patients of 10.71 days, this was not statistically significant (p > 0.05). Microbiological examination results are shown in Table 2 and Table 3. The predominant pathogen in AM group was Streptococcus pneumoniae (p < 0.05). In a group of patients with rAM (N = 6), the only pathogen isolated in one patient at first AM episode was Streptococcus pneumoniae. Four cultures were negative. At second episode Streptococcus pneumoniae and Candida albicans were isolated. Four cultures were negative again (Table 3). Streptococcus pneumoniae was the only bacteria isolated in this group, but the sample was too small to perform statistical analysis. The follow-up questionnaire achieved a response rate of 48% (Table 4). AM was often the first episode of otologic disease. Recurrent acute otitis media (> 4 episodes per year preceding AM) is far more common among patients with rAM than those with isolated AM (p < 0.05) (Table 4). The positive predictive value of recurrent acute otitis media for rAM in this series is 41.7%.
Number of ears (N = 83)
First episode:
Table 4 Otologic history prior to initial AM episode. Response was received from 35 of 73 patients. History of recurrent acute otitis media (AOM) prior to initial acute mastoiditis significantly predicts rAM (p < 0.05).
3. Results
Isolated pathogen
No.
- One rAM patient previously treated with simple mastoidectomy achieved resolution of recurrence with IV antibiotics alone. - One patient successfully treated with antibiotics initially, underwent more aggressive mastoidectomy to treat recurrence. - Four patients with rAM previously treated with simple 145
International Journal of Pediatric Otorhinolaryngology 111 (2018) 142–148
J. Mierzwiński et al.
remained unchanged as well as the number of mastoidectomies [25]. On the other hand, Tawfik et al. found that introduction of two pneumococcal vaccines did not decrease rates of hospitalization for pediatric AM between 2000 and 2012. Between 2009 and 2012 in children aged 0–2 years of age declining hospitalization rates, possibly reflecting the protective benefit of the 13-valent pneumococcal vaccine were observed [26]. It will be interesting to see whether pneumococcal vaccination will decrease the incidence of AM and rAM in our country after introduction into the national immunization programme, but it is still too early to draw any conclusions. In our whole group of children with rAM at first episode of AM 80% of samples were negative, at second episode (rAM) 60% (Table 3). All published studies pertaining to the microbiology rAM shows that only some percent of such cultures exhibit bacterial growth [8,11]. In study by Gavriel et al. at first episode of AM 11% of samples were negative, at second episode 42% [8]. Quesnel in his study described 8 a.m. recurrences with 44% of negative samples [11]. Potential higher cultures negativity in our rAM group may be due to the fact, that almost all the children have received intensive, sometimes multiple antibiotic therapy prior to hospitalization of children for AM. Management of AM included administering empiric intravenous antibiotics shortly after the admission in every single case. Patients with an unfavorable course of disease after 24–48 h qualified for subsequent treatment, comprising adjunctive drainage procedures such as myringotomy/tympanostomy or mastoidectomy. However, in our series, the majority of patients, 89% required procedural intervention. In 16 studies with reported cases of rAM only in 5 of them results of the microbiologic growth in rAM group were described (Table 1).
mastoidectomy underwent more aggressive mastoidectomy for rAM and 3 of them had no further episodes. One remaining patient who received a second cortical mastoidectomy for recurrence developed a second recurrence of AM. This was treated with more complete surgical opening and silastic sheeting inserted in the attic and facial recess. The most common complication in the AM group was sigmoid sinus thrombosis (5%), followed by epidural abscess, petrositis, and facial paresis (1% each). Subperiosteal abscess was present in 22% of AM patients (16 cases) and in 33% of rAM patients (2 cases). There were no intratemporal/ intracranial complications associated with rAM compared to 9.6% rate in AM (p < 0.05). There was no progression of intracranial complications after treatment initiation. There were no patient deaths in the series. 5. Discussion This report is the first to our knowledge to identify features that predict recurrent acute mastoiditis and implement a treatment that reduces rAM episodes. The 8% recurrence rate identified in this series falls within the expected frequency reported by others in the post-antibiotic era [6,10,20]. 5.1. Isolation of some pathogens at first episode of AM is considered a prognostic factor related to recurrence The predominant bacterial species present in all AM children during the study period was Streptococcus pneumoniae (33.7% of cases) (p < 0.05). Other authors have reported similar observations and rates [2,12,16,22,23]. The only bacteria isolated from rAM cases in our patients was Streptococcus pneumoniae. This differs from Gavriel et al. who found that recurrence is most common among patients with Pseudomonas aeruginosa, anaerobic bacteria, and polymicrobial infection cultured from an AM infection [8]. Since the most frequent bacteria was Streptococcus pneumoniae, theoretically the vaccination should prevent acute otitis media episodes and reduce the number of AM and rAM in the pediatric population. Since January 1, 2017 the vaccination against S. pneumoniae has been obligatory in Poland. Our study included all children treated for AM between 2001 and 2016, when mass vaccination had not been obligatorily implemented. According to Polish demographic data, the total Polish population includes about 6 million children (5.771.400 people aged 0–14 years). PCV7 was registered in Poland in 2001, but in fact the vaccine became available in 2005 [24]. PCV10 was registered in April 2009 [24]. Antipneumococcal conjugated vaccines were recommended, but only voluntary. According to data provided by Pfizer Poland, in the years 2005, 2006, 2007, 2008 and 2009, 4000, 41.230, 153.986, 224.825 and 268.833 doses of PCV7, respectively were sold. Since April 2009 54.033 doses of PCV10 have been sold [data from GlaxoSmithKline Poland] [24]. As we can see from the data above, not many children were vaccinated in general population. In our group of 73 children only few children have received voluntary vaccines, making this group too small for comparative purposes. The more - vaccines received by our patients were not homogenous. Pneumococcal conjugate vaccines available at that time contained different serotypes - PCV7 or PCV10 making this statistics even more confusing. However there are several literature data on pneumococcal vaccination and the rate of mastoiditis in pediatric population. According to Cavel et al. the incidence of AM is lower in the post-PCV13 period compared to the pre-PCV period. The rate of AM complications
5.2. Recurrent episodes of acute otitis media are considered another predictive factor of recurrence For many patients, the initial AM presentation was the first manifestation of otologic disease. Among those with history of recurrent acute otitis media prior to the initial AM case, there is a significantly greater risk of developing rAM in the future [16]. This is consistent with the current understanding of AM wherein acute otitis media progresses to AM. As evidenced by our series and described by others, there is not always a sustained connection between the middle ear cleft and the mastoid in the setting of infection so once infection occurs, intervention might need to address the middle ear and mastoid cavities as two anatomically separated spaces [27]. Some have even hypothesized resolution of acute otitis media in the context of persistent acute mastoiditis. The end of the window of risk for rAM episodes has yet to be well characterized. Gavriel reported a median duration between the first and second episode of only 2 months [8]. We found a longer range of 6 weeks up to 3 years. While, some guidelines suggest that expectant management is appropriate for select acute otitis media, we feel that the greatly elevated risk of rAM among patients with history of AM warrants consideration of earlier and more aggressive treatment of subsequent acute otitis media with antibiotics. 5.3. Surgical treatment is also discussed as suggested contributing factor in rAM Partly through the series, a change in surgical approach from simple mastoidectomy to more aggressive initial surgical approach with broad attic exposure and posterior tympanotomy was conducted to address anatomic obstruction at the attic. Theoretically the surgical connections can be secondarily obliterated by granulation tissue and bone prior to subsequent episodes of AM, however, none of the patients undergoing this procedure at initial mastoidectomy had AM recurrence. In a group of patients treated for the first episode of AM with simple mastoidectomy only – 5 recurrences were observed. The approach is 146
International Journal of Pediatric Otorhinolaryngology 111 (2018) 142–148
J. Mierzwiński et al.
6. Conclusions
consistent with the proposal of broad mastoidectomy including facial recess exposure by Linder [28]. However, the samples of patients are too small to perform the statistical analysis that will allow us to compare groups with and without posterior tympanotomy performed to determine if creating large communication with additional posterior tympanotomy indeed prevented rAM. We assume that this approach could be protective against rAM, but future studies with larger groups will permit stronger conclusions. Others, including Groth et al. suggested that mastoidectomy for AM may be a predisposing factor for rAM [6]. Luntz et al. contest this opinion, finding that a history of recurrent otitis media prior to a first episode of AM is a predisposing factor for rAM, but not treatment mode [16]. Vasbotn et al. emphasized that AM patients treated with mastoidectomy have a more severe course of disease than other AM patients [2]. In a 10-year observation period, Khafif et al. collected a group of 134 patients with AM [13]. They observed recurrent episodes in 11 patients and found that the recurrence rate was not related to the mode of treatment. Lahav reported recurrence in seven of 78 rAM patients; four of these had undergone mastoidectomy, one received medical treatment, and three underwent abscess aspiration [7]. We think that the recurrence rate in the mastoidectomy group may be higher than that in the medical treatment group, however, they are two different populations. AM requiring surgical intervention represents a more advanced state of infection and includes children in whom medical treatment was ineffective. Therefore, implication that initial surgery is a causative factor for recurrence is not appropriate. Our series includes also one patient with prior cochlear implantation. The patient was successfully treated with lancing the site. Others report that mastoiditis occurs in about 1% of patients after cochlear implant (CI) recipients. It has been shown that second and third AM episodes may also occur in this patient group. Zawawi reported AM in three out of 43 CI patients; Migirov reported AM in two out of 11 [9,10]. Migirov analyzed 234 children with CI; 116 of these underwent surgery via the transmastoid mastoidectomy, posterior tympanotomy approach and 118 via the suprameatal approach [10]. Post-CI AM developed in 11 cases, all of them in the group receiving mastoidectomy with posterior tympanotomy approach. In two cases Migirov observed AM recurrence, which was treated with incision and abscess drainage. The finding that patients are more vulnerable to AM after mastoidectomy than after suprameatal surgery might be extrapolated to indicate that children are more vulnerable to AM after mastoidectomy. We contest this supposition, however, as the surgeon typically secondarily obliterates the facial recess during the cochlear implantation procedure and the additional communication pathway is closed again.
- AM occurs with incidence of 0.6–4.8 per 100,000 per year. Among these cases of AM, rAM accounts for 2–18%. - Streptococcus pneumoniae is the most common pathogen isolated in our series of AM and the only bacteria isolated from cases of rAM. Most children were unvaccinated against S. pneumoniae species. - History of recurrent acute otitis media prior to the first episode of AM is a significant risk factor for development of rAM. - Wider surgical connection of the mastoid and middle ear with attic exposure and posterior tympanotomy was protective against subsequent episodes of AM in this series, but future studies with larger groups of patients will permit stronger conclusions. - Patients with rAM present with fewer intratemporal and intracranial complications than those with AM. Conflicts of interest The authors disclose no conflicts of interest. Formatting of funding sources This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors. Declarations of interest None. Consent for publication and ethics approval The study was completed in accordance with all procedures performed in studies involving human participants and in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards. Consent for publication: written informed consent has been obtained from the parents of the patients and from patients (if regarded as competent). Bioethics committee permission number KB 582/2017. Bioethics Committee of the Nicolaus Copernicus University in Toruń functioning at Collegium Medicum in Bydgoszcz. Authors' contributions
5.4. Another suggested contributing factor in rAM is immunodeficiency secondary to IG4-related disease [29,30,31]
- Study concept and design - Józef Mierzwiński - Acquisition of data: Justyna Tyra, Karolina Haber, Michael David Puricelli - Analysis and interpretation of data: Józef Mierzwiński, Anna Sinkiewicz, Justyna Tyra, Karolina Haber, Michael David Puricelli - Drafting of the manuscript: Józef Mierzwiński, Anna Sinkiewicz, Justyna Tyra, Karolina Haber, Michael David Puricelli - Critical revision of the manuscript for important intellectual content: Józef Mierzwiński - Statistical analysis: Justyna Tyra - Administrative, technical, and material support: Anna Sinkiewicz, Karolina Haber, Michael David Puricelli - Study supervision: Józef Mierzwiński
Specific investigation into the immune status of our patients with rAM was not made, however, this may be considered in patients with rAM. Finally, we find significantly fewer intratemporal and extratemporal complications among patients with rAM. We believe that rAM cases are susceptible to such complications and theorize that the reduction could be secondary to increased awareness and earlier presentation. Nevertheless, secondary episodes are associated with prolonged hospitalization and effort should be made to reduce the frequency whenever possible. There are several limitations to the data presented. Most significantly, AM and rAM are rare conditions and there is limited sample size of patients available. Future multi-institutional collaboration would be expected to yield greater numbers and permit stronger conclusions. We expect follow-up within our institution but recognize that patients could have been seen elsewhere biasing our findings. We attempted to mitigate this by secondarily contacting the patients.
Acknowledgements The authors would like to express thanks to Prof. Leszek Knopik for his invaluable help in statistical analysis. 147
International Journal of Pediatric Otorhinolaryngology 111 (2018) 142–148
J. Mierzwiński et al.
References [1] P. Homøe, K. Kværner, J.R. Casey, R.A.M.J. Damoiseaux, T.M.A. van Dongen, H. Gunasekera, R.G. Jensen, E. Kvestad, P.S. Morris, H.M. Weinreich, Panel 1: epidemiology and diagnosis, Otolaryngol. Head Neck Surg. 156 (2017) S1–S21. [2] F.S. Vassbotn, O.G. Klausen, O. Lind, P. Moller, Acute mastoiditis in a Norwegian population: a 20 year retrospective study, Int. J. Pediatr. Otorhinolaryngol. 62 (2002) 237–242. [3] D.A. Van Zuijlen, A.G. Schilder, F.A. Van Balen, A.W. Hoes, National differences in incidence of acute mastoiditis: relationship to prescribing patterns of antibiotics for acute otitis media? Pediatr. Infect. Dis. J. 20 (2001) 140–144. [4] H. Mygind, The subperiosteal abscess of the mastoid region, Ann. Otol. Rhinol. Laryngol. 19 (1910) 529–540. [5] H.P. House, Otitis media; a comparative study of the results obtained in therapy before and after the introduction of the sulfonamide compounds, Arch. Otolaryngol. 43 (1946) 371–378. [6] A. Groth, F. Enoksson, J. Stalfors, K. Stenfeldt, M. Hultcrantz, A. Hermansson, Recurrent acute mastoiditis - a retrospective national study in Sweden, Acta Otolaryngol. 132 (2012) 1275–1281. [7] J. Lahav, O. Handzel, R. Gertler, M. Yehuda, D. Halperin, Postauricular needle aspiration of subperiosteal abscess in acute mastoiditis, Ann. Otol. Rhinol. Laryngol. 114 (2005) 323–327. [8] H. Gavriel, R.A. Eta, E. Eviatar, The role of bacteriologic studies in predicting recurrent mastoiditis in 456 patients, Otol. Neurotol. 37 (2016) 1059–1062. [9] F. Zawawi, I. Cardona, O.V. Akinpelu, S.J. Daniel, Acute mastoiditis in children with cochlear implants: is explantation required? Otolaryngol. Head Neck Surg. 151 (2014) 394–398. [10] L. Migirov, A. Yakirevitch, Y. Henkin, R. Kaplan-Neeman, J. Kronenberg, Acute otitis media and mastoiditis following cochlear implantation, Int. J. Pediatr. Otorhinolaryngol. 70 (2006) 899–903. [11] S. Quesnel, M. Nguyen, S. Pierrot, P. Contencin, Y. Manach, V. Couloigner, Acute mastoiditis in children: a retrospective study of 188 patients, Int. J. Pediatr. Otorhinolaryngol. 74 (2010) 1388–1392. [12] C.G. Petersen, T. Ovesen, C.B. Pedersen, Acute mastoidectomy in a Danish county from 1977 to 1997-operative findings and long-term results, Acta Otolaryngol. Suppl. 543 (2000) 122–126. [13] A. Khafif, D. Halperin, I. Hochman, R. Gertler, I. Poria, D. Shindel, G. Marshak, Acute mastoiditis: a 10-year review, Am. J. Otolaryngol. 19 (1998) 170–173. [14] F. Glynn, L. Osman, M. Colreavy, H. Rowley, T.P.O. Dwyer, A. Blayney, Acute mastoiditis in children: presentation and long term consequences, J. Laryngol. Otol. 122 (2008) 233–237. [15] E.H. Harley, T. Sdralis, R.G. Berkowitz, Acute mastoiditis in children: a 12-year retrospective study, Otolaryngol. Head Neck Surg. 116 (1997) 26–30. [16] M. Luntz, A. Brodsky, S. Nusem, J. Kronenberg, G. Keren, L. Migirov, D. Cohen,
[17]
[18] [19] [20]
[21] [22]
[23]
[24]
[25]
[26]
[27]
[28] [29]
[30] [31]
148
S. Zohar, A. Shapira, D. Ophir, G. Fishman, G. Rosen, V. Kisilevsky, I. Magamse, S. Zaaroura, H.Z. Joachims, D. Goldenberg, Acute mastoiditis-the antibiotic era: a multicenter study, Int. J. Pediatr. Otorhinolaryngol. 57 (2001) 1–9. I.M. Psarommatis, C. Voudouris, K. Douros, P. Giannakopoulos, T. Bairamis, C. Carabinos, Algorithmic management of pediatric acute mastoiditis, Int. J. Pediatr. Otorhinolaryngol. 76 (2012) 791–796. A. Rosen, D. Ophir, G. Marshak, Acute mastoiditis: a review of 69 cases, Ann. Otol. Rhinol. Laryngol. 95 (1986) 222–224. L. Migirov, A. Yakirevitch, J. Kronenberg, Mastoid subperiosteal abscess: a review of 51 cases, Int. J. Pediatr. Otorhinolaryngol. 69 (2005) 1529–1533. M. Nussinovitch, R. Yoeli, K. Elishkevitz, I. Varsano, Acute mastoiditis in children: epidemiologic, clinical, microbiologic, and therapeutic aspects over past years, Clin. Pediatr. 43 (2004) 261–267. M. Abdel-Aziz, H. El-Hoshy, Acute mastoiditis: a one year study in the pediatric hospital of Cairo university, BMC Ear Nose Throat Disord. 10 (2010) 1. A. Groth, F. Enoksson, A. Hermansson, M. Hultcrantz, J. Stalfors, K. Stenfeldt, Acute mastoiditis in children in Sweden 1993-2007-no increase after new guidelines, Int. J. Pediatr. Otorhinolaryngol. 75 (2011) 1496–1501. L.H.Y. Pang, M.S. Barakate, T.E. Havas, Mastoiditis in a paediatric population: a review of 11 years experience in management, Int. J. Pediatr. Otorhinolaryngol. 73 (2009) 1520–1524. A. Skoczyńska, E. Sadowy, K. Bojarska, J. Strzelecki, A. Kuch, A. Gołbiewska, I. Waśko, M. Foryś, M. van der Linden, W. Hryniewicz, The current status of invasive pneumococcal disease in Poland, Vaccine 29 (2011) 2199–2205. O. Cavel, R. Tauman, E. Simsolo, D. Yafit, E. Reindorf-Kfir, O. Wasserzug, O. Unger, O. Handzel, G. Fishman, Y. Oestreicher-Kedem, A. DeRowe, Changes in the epidemiology and clinical features of acute mastoiditis following the introduction of the pneumococcal conjugate vaccine, Int. J. Pediatr. Otorhinolaryngol. 104 (2018) 54–57. K.O. Tawfik, S.L. Ishman, M.E. Tabangin, M. Altaye, J. Meinzen-Derr, D.I. Choo, Pediatric acute mastoiditis in the era of pneumococcal vaccination, Laryngoscope 128 (6) (2018 Jun) 1480–1485. P. Schachern, M.M. Paparella, S. Sano, S. Lamey, Y. Guo, A histopathological study of the relationship between otitis media and mastoiditis, Laryngoscope 101 (1991) 1050–1055. T.E. Linder, H.R. Briner, T. Bischoff, Prevention of acute mastoiditis: fact or fiction? Int. J. Pediatr. Otorhinolaryngol. 56 (2000) 129–134. V. Deshpande, N.A. Zane, S. Kraft, J.H. Stone, W.C. Faquin, Recurrent mastoiditis mimics IgG4 related disease: a potential diagnostic pitfall, Head Neck Pathol. 10 (3) (2016 Sep) 314–320. V. Deshpande, Y. Zen, J.K. Chan, E.E. Yi, Y. Sato, et al., Consensus statement on the pathology of IgG4-related disease, Mod. Pathol. 25 (2012) 1181–1192. A. Sebastian, Choroba związana z IgG4 – czy to całkiem nowa choroba? Biul. Sekc. Młodych Reumatol. (2014) PTR 10/2014.