European Journal of
Eur J Pediatr (1990) 149 : 560-564
Pediatrics
9 Springer-Verlag1990
Acute mastoiditis: clinical, microbiological, and therapeutic aspects D. Nadal 1, P. Herrmann 2, A. Baumann 1, and A. Fanconi 1 Departments of 1Pediatrics, 20torhinolaryngology, University of Ztirich, Kinderspital, Steinwiesstrasse 75, CH-8032 Ziirich, Switzerland Received June 12, 1989 / Accepted October 31, 1989
Abstract. The charts of 73 children (31 girls, 42 boys) aged 4 months to 14 years ( m e a n 4.5 years) with acute mastoiditis m a n a g e d during a 16-year period were reviewed. O f the patients 36% were less than 24 months old. Retro-auricular swelling was described in 63 of the 73 children, tenderness in 59, e r y t h e m a in 58, and protrusion of the auricle in 45. A pathological tympanic m e m b r a n e was noted in 33% of the patients and fever in only 29%. A p a r t from local inflammation, the most frequent complaints and symptoms were otalgia (n = 42), recent u p p e r respiratory tract infections (n = 22), and fever alone (n = 22). A subperiosteal abscess was found in 36 patients, and CNS involvement in 5. Nearly half of the patients (48%) were on antibiotic therapy at admission. The isolation rates in bacterial cultures f r o m subperiosteal aspirates (81%) and f r o m mastoid mucosa (68%) were considerably higher than f r o m blood cultures (14%) and were not influenced by previously administered antibiotics. Pneumococci (9/32) and Staphylococcus epidermidis (6/32) were the agents most often isolated. The incidence of the bacteria isolated f r o m patients pre-treated with antibiotics differed f r o m the incidence in patients not previously treated. In 24 patients (33%) the lesion healed with antibiotic therapy without mastoid surgery. Myringotomy and the insertion of a ventilation tube is indicated initially, if acute otitis media with effusion is found. In the absence of a subperiosteal abscess and of CNS involvement, a 48-hour trial of intravenous antibiotic therapy, directed also against staphylococci, is justified before mastoid surgery is considered.
Key words: Acute mastoiditis - Subperiosteal abscess Central nervous system - Microbiology - T r e a t m e n t
h y p e r a e m i a and o e d e m a of the mucosal lining of the pneumatized cells with subsequent accumulation of serous and then purulent exudate, lead to acute mastoiditis [2]. Otitis media is the most frequent diagnosis in paediatric practice [1]. Although most cases of suppurative otitis media are associated with inflammation of the mastoid cells, clinical mastoiditis has b e c o m e a rare disorder in the last decades [15]. M a n a g e m e n t of acute mastoiditis remains controversial [4, 5, 7, 8, 10]. Prospective studies focusing on the most adequate treatment modality are h a m p e r e d by the relatively rare occurrence of the disease, and decisions for the m a n a g e m e n t of acute mastoiditis have to rely on retrospective studies. H o w e v e r , there are few comprehensive studies reviewing treatment during the antibiotic era [4, 5, 8, 9, 11]. The present report on 73 children treated for acute mastoiditis at the D e p a r t m e n t s of Paediatrics and of Otorhinolaryngology of the University of Ziirich during a period of 16 years aims to fill this gap.
Patients and methods The hospital records of all children less than 16 years of age discharged with the diagnosis of acute mastoiditis between June 1971 and May 1988 at the Departments of Paediatrics and of Otorhinolaryngology of the University of Zurich were reviewed. This period was chosen because computerized registration of the patients was available back to June 1971. The diagnosis "acute mastoiditis" was made if both of the following criteria were met: 1. Presence of at least one of the physical signs retroauricular swelling, erythema, with or without tenderness, and protrusion of the auricle. 2. Evidence of co-existent or recent otitis media. A mastoid subperiosteal abscess was diagnosed if the retro-auricular swelling fluctuated. In the absence of these signs, cases with surgical findings revealing acute mastoiditis were also included.
Introduction The mastoid b o n e consists of a single cell at birth, the antrum. Soon thereafter, pneumatization of the b o n e occurs. The resulting mastoid air cells are lined with modified respiratory mucosa. All cells are anatomically interconnected with the antrum which, in turn, is connected to the middle ear by a small channel, the aditus ad antrum. Inflammation of the middle ear can thus, after
Offprint requests to: D. Nadal
Results The charts of 75 children were reviewed. Seventy-three children aged 4 months to 14 years (mean age 4 years and 6 months) matched our definition of acute mastoiditis. M o r e than one third of the patients (n = 26) were less than 24 months old. Boys (n = 42) were slightly m o r e often affected than girls (n = 31), with similar age distribution. The left and the right mastoid were involved
561 with equal frequency (37 vs. 36). In both children excluded from the study the diagnosis of mastoiditis was based on radiological findings alone. A synopsis of the history and the findings on admission of the 73 children is given in Table 1. A history of otological symptoms including otalgia, retro-auricular swelling or erythema, and otorrhoea was reported by all but two patients or their parents. These two had an acute onset of retroauricular swelling on the day of admission. Pain was the sole otological complaint in 19 patients, swelling in 11, otorrhoea in 8, and retro-auricular erythema in 3 patients. The duration of these symptoms was 1 day in 20 patients (27%), 2 days in 13 patients (17%), and 3 or 4 days in 10 patients (13%). An additional 20 patients (27%) had otological symptoms 6 to 14 days prior to admission and the remaining 10 patients (13%) 15 days or more prior to hospitalization. Symptoms of recent or actual upper respiratory tract infection (22 patients) or fever alone (20 patients) were more frequently reported than nausea and vomiting (n = 7), headaches (n = 6), anorexia (n = 5), vertigo (n = 4), hearing loss (n = 4), or diarrhoea (n = 4). The duration of these symptoms prior to admission was 7 days or less in 21 patients, between 8 and 14 days in 15, and 15 days or more in 6 patients.
Table 1. History and findings on admission in 73 children with acute mastoiditis Features
Present (n)
History Otalgia Retro-auricular swelling Retro-auricular erythema Otorrhoea
42 34 19 13
Upper respiratory tract infection Fever alone (> 38.5~
22 20
Antibiotic treatment in the last 2 weeks
43
Previous ipsilateral otitis media once recurrent Previous surgical procedure on ipsilateral ear
15 11 8
Clinicalfindings Retro-auricular swelling tenderness erythema Abnormal tympanic membrane Protrusion of the auricle Swellingof regional lymph nodes Otorrboea Normal body temperature (_ 38.5~
63 59 58 48 45 33 23 31 21
Laboratory findings Erythrocyte sedimentation rate > 20 mm/b (n = 51) Leukocytosis (> 15 x 109/1) (n = 70)
43 35
Antibiotic therapy had been started in 35 patients (48%) 1 to 28 days prior to admission (mean 6 days). Eight other children had received antibiotics during the last 2 weeks prior to hospitalization, but treatment was discontinued at least 3 days before admission. Previous episodes of acute otitis media were reported by 26 patients or their parents, in 11 of these the episodes had recurred. In eight patients one or more surgical procedures had been performed on the ipsilateral ear: myringotomy and insertion of tympanostomy tubes (n = 5), mastoidectomy (n = 2), incision of a subperiosteal abscess (n = 2), removal of an epidermoid cyst (n = 1). The most frequent local finding was retro-auricular swelling followed by tenderness, erythema, and protrusion of the auricle. Retro-anricular swelling, tenderness, and erythema were found simultaneously in 50 (68%) patients. Forty-eight patients had a pathological tympanic membrane. Hyperaemia was most frequent (n = 35), followed by perforation (n = 23), bulging (n = 17), thickening (n = 16), retraction (n = 12), and blurred landmarks (n = 10). A diagnosis of co-existent acute otitis media was made in 35 patients (48%). Enlarged regional lymph nodes were noted in 33 patients: cervical (n = 25), angular (n = 20), nuchal (n = 7) and retro-auricular (n = 7). Fever over 38.5~ was noted in less than one third of the patients. Lumbar puncture was performed in eight patients with meningeal signs on admission, three of them disclosed pleocytosis. Two of these three had positive bacterial cultures of CSF (Table 2). The patient with pleocytosis without bacterial growth in his CSF had cholesteatoma with lateral sinus thrombosis and an intra- and extracranial abscess. One of the two patients with proven bacterial meningitis had additional labyrinthitis. Signs of CNS involvement were also noted in two further patients. Both had increased intracranial pressure: one with abducens paresis and concomitant subperiosteal abscess, the other with consecutive otitic hydrocephalus. Forty-two patients had conventional radiological examination, 34 (80%) yielded abnormal findings. These included clouding of the mastoid (n = 28), absent pneumatization (n =18), and sclerosis (n = 9). Computed tomography was performed in two patients because of suspected intracranial complication. An intracranial abscess was visualized in one of these two cases. The other case revealed no pathological finding. The erythrocyte sedimentation rate was determined in 51 patients and was found to be elevated (>_ 20 mm) in 43 (84%). Serum C-reactive protein was measured in ten patients; five had values higher than 90 mg/1 (range 91159), three slightly elevated (range 14-37), and two norreal (< i mg/1). A white blood cell count was performed in 70 patients disclosing leukocytes >15 • 109/1 in 35 (50%), 10-15 x 109/1 in 20 (29%), and < 10 • 109/1 in 15 (21%). Bacterial cultures were taken in 54 patients (77 samples) (Table 2) including 30 patients (43 samples) under oral antibiotic therapy. Cultures of retro-auricular abscesses yielded organisms in 81% (13/16), of mastoid
562 Table 2. Results of bacterial cultures in 54 children with acute mastoiditis Isolates
Patients (n=32)
Streptococcuspneumoniae Staphylococcus epidermidis Staphylococcus aureus Streptococcus pyogenes Pseudomonas aeruginosa Streptococcus viridans Haemophilus influenzae Peptostreptococcus spp. Propionibacterium spp. Escherichia coli
Samples (total number) Blood (n=21)
Mastoid mucosa (n = 22)
Retroauricular aspirate (n = 16)
Myfingotomy asp~ate (n = 10)
CSF
5 4 2 2 2b
4 2 2 2 1 -
1 1 1 1 -
1 -
1
-
-
-
-
9 6 4 4 3 2
(3) a (3) (2) (1) (2) (1)
2 -
1
(1)
.
1
(1)
-
-
1
.
-
-
1 1
(1)
.
.
.
(n=8)
1
.
. 1
.
a Numbers in parenthesis denote the number of patients on antibiotic therapy at the time of sampling b Cultures in one patient with chronic granulomatous disease grew in addition Candida albicans CSF = cerebrospinal fluid Table 3. Management of 73 children with acute mastoiditis Initial management
Patients Total (n=73)
IVAB alone IVAB plus myringotomy IVAB plus mastoidectomy with or without myringotomy
Delayed surgery (n=6)
S. aureus, and P. aeruginosa n = 2 respectively, and S. pyogenes n = 1) by means of the disk diffusion test. None of the S. epidermidis isolates was found to be methicillin-
Myringo- Mastoidtomy ectomy (n = 2) (n = 4)
resistant. There was no difference between the organisms isolated in the 1970s versus those in the late 1980s. Only one of the 73 patients had an underlying disorder. Cultures of the mastoid mucosa of this patient with chronic granulomatous disease grew Candida albicans and S.
26a
2
2
viridans.
8
0
2
39b
0
0
IVAB denotes intravenous antibiotics a Including one patient with central nervous system (CNS) involvement u Including 33 patients with initial subperiosteal abscess, one patient with CNS involvement, and three patients with both
mucosa in 68% (15/22), of middle ear aspirates in 40% (4/10), and blood cultures in 14% (3/21). The isolated bacteria were Streptococcus pneumoniae, Staphylococ-
cus epidermidis, S. aureus, S.pyogenes, Pseudomonas aeruginosa, S. viridans, Peptostreptococcus spp., Propionibacterium app., and Escherichia coli. Haernophilus influenzae type b and S. epidermidis proved to be the cause of bacterial meningitis in two cases. There was no significant difference between the isolation rate of cultured material of patients with or without previous antibiotic therapy. However, there was a higher incidence of isolation of S. pneumoniae and S. pyogenes in patients who had not previously taken antibiotics (6/24 and 3/24, respectively) compared to children previously treated with adequate antibiotics (3/30 and 0/30, respectively). In 7 of the 15 pre-treated patients with positive cultures, antimicrobial testing revealed that the isolated bacteria were resistant to the antibiotic prescribed (S. epidermidis,
The management and outcome of the 73 children are summarized in Table 3. All patients received different regimens of intravenous antibiotics on the day of admission for at least 3 days (mean duration 5 days). In most cases antibiotic therapy was continued orally afterwards. In addition, myringotomy alone or in combination with mastoidectomy was performed in 47 patients (64%). The reason for mastoid surgery in 37 of these 47 patients was a retro-auricular subperiosteal abscess (n = 33) or CNS involvement (n = 1), or both (n = 3). Mastoid surgery was also performed in 4 further patients of these 47 because of suspected subperiosteal abscess. The six remaining patients had myringotomy because of acute otitis media with effusion. Delayed surgery was performed in six patients: delayed myringotomy was done in two patients after the retro-auricular finding had improved, but middle ear effusion persisted; 2 of the 26 children initially treated with intravenous antibiotics alone and two of the eight children who had initial myringotomy required mastoidectomy because of insufficient local improvement 2 days or more after admission. A p a r t from the four patients who underwent delayed mastoid surgery, all patients improved or recovered from their local symptoms within 48 h. All symptoms had resolved after 4 days in 46 patients (63%), after i week in 59 (80%), and after 2 weeks in 70 (96%). The five children with CNS involvement were initially operated ex= cept one child who presented with meningitis but no sub-
563 periosteal abscess. All of the 73 patients healed without sequelae. Discussion Clinical acute mastoiditis, once a common complication of otitis media [15], has become a rare entity. A marked decrease in the incidence of acute mastoiditis was observed when antibiotics were introduced [15]. This retrospective study analyses 73 patients seen during a 16-year period at our hospitals with a mean of 5800 paediatric inpatients yearly. A similar low incidence was found in the few other retrospective studies, focusing on acute mastoiditis [4, 5, 7, 8, 10-12]. However, our study centre is a university hospital and milder cases of mastoiditis might not have been referred and may thus be underrepresented in our series. The definition of "acute mastoiditis" has not been uniform in former studies: while some authors failed to list their criteria for patient selection [10, 15], others included only patients with simultaneous acute otitis media [4, 8], or patients with both retro-auricular physical findings and radiological abnormalities [5], and still other investigators required that acute mastoiditis be confirmed at surgery [11]. The comparison of the results is therefore difficult and conclusions are limited. In this context we would like to emphasize that mastoiditis may progress and become clinically manifest at a time when infection in the middle ear has already resolved [1]. Thus, simultaneous acute otitis media should not be a mandatory finding for the diagnosis of acute mastoiditis. Also, diagnosis should not be based on radiographs. We and others [8] found that radiography may be misleading. Only in very rare occasions will mastoiditis be masked and escape clinical recognition [6, 8]. Acute mastoiditis affected all paediatric age groups, but more than one third of the children (36%) were younger than 24 months. This corresponds to the findings in another study [8] and reflects the age distribution of children with otitis media [1]. In our series nearly half of the patients were on oral antibiotics at admission. The reported percentage of patients on antibiotic regimens developing acute mastoiditis ranges from 36%-71% [5, 7, 8, 11-13]. This raises the question whether treatment failure was due to low compliance and/or inadequate antibiotic therapy. In 7 of our 15 patients who had been treated with oral antibiotics prior to admission and in whom bacteriological workup revealed growth, the isolates were resistant to the prescribed antibiotics. Treatment failure in the remaining 8 of the 15 patients could, beside low compliance, have been due to a moderate or inadequate dosage of the antibiotics [14] or due to an insufficient penetration of the drug into the inflamed tissue [10]. The latter could explain our finding of similar isolation rates of bacteria in pre-treated and not treated patients. However, in a recent study including only patients with surgically proven mastoiditis some purulent discharges were found to have been sterilized by the pre-operative treatment [11]. Although acute mastoiditis is a complication of acute otitis media, the incidence of the bacteria isolated in
both entities is not identical. The predominant causative agents of acute otitis media are S. pneumoniae and H. influenza, followed by Branhamella catarrhalis, S. pyogens, and S. aureus [3]. Enterobacteria and anaerobes are less frequent. In acute mastoiditis S. pneumoniae is also the most common pathogen and the spectrum of the other agents is similar, but, as shown by the present and former studies [4, 5, 8, 9, 11, 12], H. influenzae is only rarely isolated. The reason for this distinct microbiological pattern is not clear. Antibiotic therapy before the onset of acute mastoiditis might be one factor. We and others [11] have demonstrated that the frequency of S. pneumoniae and S. pyogenes isolation was considerably lower in patients previously treated with antibiotics than in patients without previous treatment (in the present study: 10% vs 25% and 0% vs 12.5%, respectively). Also infections with H. influenzae are less apt to progress to more severe degrees of mastoiditis with subperiosteat abscess formation affording mastoidectomy. The isolation rate of bacteria from blood cultures is extremely low in acute mastoiditis (in the present study 14%). Thus, the incidence of the bacteria isolated might not correspond to the true frequency, since cultures were mainly derived from a subperiosteal abscess or mastoid mucosa. In contrast to other investigators, we found a considerable number of cases in whom S. epidermidis (6/32 patients with positive cultures) was isolated from subperiosteal abscess aspirates or mastoid mucosa. Since the specimens were taken under sterile conditions in the operating room, contamination seems unlikely. None of the isolates was found to be methicillin-resistant. We postulate that S. epidermidis which is one of the predominant pathogens in chronic otitis media with effusion [3] should be included in the list of the main agents causing acute mastoiditis. The retrospective design of our study does not allow treatment recommendations with certainty. However, the fact that 24 of our patients (33%), none with subperiosteal abscess but one with meningitis, recovered with intravenous antibiotics without mastoid surgery provides clear evidence for mastoidectomy not being obligate in the management of acute mastoiditis. Remarkably, none of our patients showed progression of the infection or intracranial or meningeal spread while on intravenous antibiotics, including the four patients who required delayed mastoid surgery because of insufficient clinical improvement. A similar experience was reported by other investigators [4, 5]. Thus, in children with acute mastoiditis without subperiosteal abscess or CNS involvement, we would recommend as initial treatment a 48 h trial of intravenous antibiotics. These should be directed against the major pathogens of acute otitis media and, according to our findings, also against staphylococci. The combination amoxicillirdclavulanic acid would fit these requirements. In addition, myringotomy should be performed and ventilation tubes inserted if acute otitis media with effusion is present. Mastoid surgery should be considered first if the antibiotic therapy is not followed by clinical improvement within 2 days or if a subperiosteal abscess is diagnosed. In the presence of neurological symptoms high resolution computed tomography is indicated
564 to d e t e r m i n e the e x t e n t of the process a n d to p l a n surgery. T h e t i m e to switch to oral antibiotics will d e p e n d o n the p a t i e n t ' s c o n d i t i o n . It is i m p o r t a n t that p a e d i a trician a n d otologist c o l l a b o r a t e in the m a n a g e m e n t of acute mastoiditis.
References 1. Bluestone CD (1983) Otitis media in children: to treat or not to treat? N Engl J Med 306:1399-1404 2. Bluestone CD, Klein JO (1983) Mastoiditis. In: Bluestone CD, Stool SE (eds) Pediatric otolaryngology. Saunders, Philadelphia, pp 546-552 3. Giebink GS (1989) The microbiology of otitis media. Pediatr Infect Dis [Suppl] 8:$18-$20 4. Ginsburg CM, Rudoy R, Nelson JD (1980) Acute mastoiditis in infants and children. Clin Pediatr 19: 549-553 5. Hawkins DB, Dru D, House JW, Clark RW (1983) Acute mastoiditis in children: a review of 54 cases. Laryngoscope 93 : 568-572
6. Holt GR, Gates GA (1983) Masked mastoiditis. Laryngoscope 93 : 1034-1037 7. Holt GR, Young WC (1981) Acute coalescent mastoiditis. Otolaryngol Head Neck Surg 89 : 317-321 8. Ogle JW, Lauer BA (1986) Acute mastoiditis. Diagnosis and complications. Am J Dis Child 140:1178-1182 9. Palva T, Virtanen H, M~ikinen H (1985) Acute and latent mastoiditis in children. J Laryngol Otol 99:127-136 10. Pfaltz CR, Griesemer C (1984) Complications of acute middle ear infections. Ann Otol Rhinol Laryngol [Suppl] 112:133-137 11. Prellner K, Rydell R (1986) Acute mastoiditis. Influence of antibiotic treatment on the bacterial spectrum. Acta Otolaryngol 102: 52-56 12. Rosen A, Ophir D, Marshak G (1986) Acute mastoiditis: a review of 69 cases. Ann Otol Rhinol Laryngol 95 : 222-224 13. Rubin JS, Wei WI (1985) Acute mastoiditis: a review of 34 patients. Laryngoscope 95 : 963-965 14. Tudor RB (1963) Mastoiditis in children. A review of 77 cases. Lancet 83 : 56-58 15. Zoller H (1972) Acute mastoiditis and its complications: a changing trend. South Med J 65 : 477-480
Eur J Pediatr (1990) 149 : 560-564
Pediatrics
9 Springer-Verlag1990
Acute mastoiditis: clinical, microbiological, and therapeutic aspects D. Nadal 1, P. Herrmann 2, A. Baumann 1, and A. Fanconi 1 Departments of 1Pediatrics, 20torhinolaryngology, University of Ztirich, Kinderspital, Steinwiesstrasse 75, CH-8032 Ziirich, Switzerland Received June 12, 1989 / Accepted October 31, 1989
Abstract. The charts of 73 children (31 girls, 42 boys) aged 4 months to 14 years ( m e a n 4.5 years) with acute mastoiditis m a n a g e d during a 16-year period were reviewed. O f the patients 36% were less than 24 months old. Retro-auricular swelling was described in 63 of the 73 children, tenderness in 59, e r y t h e m a in 58, and protrusion of the auricle in 45. A pathological tympanic m e m b r a n e was noted in 33% of the patients and fever in only 29%. A p a r t from local inflammation, the most frequent complaints and symptoms were otalgia (n = 42), recent u p p e r respiratory tract infections (n = 22), and fever alone (n = 22). A subperiosteal abscess was found in 36 patients, and CNS involvement in 5. Nearly half of the patients (48%) were on antibiotic therapy at admission. The isolation rates in bacterial cultures f r o m subperiosteal aspirates (81%) and f r o m mastoid mucosa (68%) were considerably higher than f r o m blood cultures (14%) and were not influenced by previously administered antibiotics. Pneumococci (9/32) and Staphylococcus epidermidis (6/32) were the agents most often isolated. The incidence of the bacteria isolated f r o m patients pre-treated with antibiotics differed f r o m the incidence in patients not previously treated. In 24 patients (33%) the lesion healed with antibiotic therapy without mastoid surgery. Myringotomy and the insertion of a ventilation tube is indicated initially, if acute otitis media with effusion is found. In the absence of a subperiosteal abscess and of CNS involvement, a 48-hour trial of intravenous antibiotic therapy, directed also against staphylococci, is justified before mastoid surgery is considered.
Key words: Acute mastoiditis - Subperiosteal abscess Central nervous system - Microbiology - T r e a t m e n t
h y p e r a e m i a and o e d e m a of the mucosal lining of the pneumatized cells with subsequent accumulation of serous and then purulent exudate, lead to acute mastoiditis [2]. Otitis media is the most frequent diagnosis in paediatric practice [1]. Although most cases of suppurative otitis media are associated with inflammation of the mastoid cells, clinical mastoiditis has b e c o m e a rare disorder in the last decades [15]. M a n a g e m e n t of acute mastoiditis remains controversial [4, 5, 7, 8, 10]. Prospective studies focusing on the most adequate treatment modality are h a m p e r e d by the relatively rare occurrence of the disease, and decisions for the m a n a g e m e n t of acute mastoiditis have to rely on retrospective studies. H o w e v e r , there are few comprehensive studies reviewing treatment during the antibiotic era [4, 5, 8, 9, 11]. The present report on 73 children treated for acute mastoiditis at the D e p a r t m e n t s of Paediatrics and of Otorhinolaryngology of the University of Ziirich during a period of 16 years aims to fill this gap.
Patients and methods The hospital records of all children less than 16 years of age discharged with the diagnosis of acute mastoiditis between June 1971 and May 1988 at the Departments of Paediatrics and of Otorhinolaryngology of the University of Zurich were reviewed. This period was chosen because computerized registration of the patients was available back to June 1971. The diagnosis "acute mastoiditis" was made if both of the following criteria were met: 1. Presence of at least one of the physical signs retroauricular swelling, erythema, with or without tenderness, and protrusion of the auricle. 2. Evidence of co-existent or recent otitis media. A mastoid subperiosteal abscess was diagnosed if the retro-auricular swelling fluctuated. In the absence of these signs, cases with surgical findings revealing acute mastoiditis were also included.
Introduction The mastoid b o n e consists of a single cell at birth, the antrum. Soon thereafter, pneumatization of the b o n e occurs. The resulting mastoid air cells are lined with modified respiratory mucosa. All cells are anatomically interconnected with the antrum which, in turn, is connected to the middle ear by a small channel, the aditus ad antrum. Inflammation of the middle ear can thus, after
Offprint requests to: D. Nadal
Results The charts of 75 children were reviewed. Seventy-three children aged 4 months to 14 years (mean age 4 years and 6 months) matched our definition of acute mastoiditis. M o r e than one third of the patients (n = 26) were less than 24 months old. Boys (n = 42) were slightly m o r e often affected than girls (n = 31), with similar age distribution. The left and the right mastoid were involved
561 with equal frequency (37 vs. 36). In both children excluded from the study the diagnosis of mastoiditis was based on radiological findings alone. A synopsis of the history and the findings on admission of the 73 children is given in Table 1. A history of otological symptoms including otalgia, retro-auricular swelling or erythema, and otorrhoea was reported by all but two patients or their parents. These two had an acute onset of retroauricular swelling on the day of admission. Pain was the sole otological complaint in 19 patients, swelling in 11, otorrhoea in 8, and retro-auricular erythema in 3 patients. The duration of these symptoms was 1 day in 20 patients (27%), 2 days in 13 patients (17%), and 3 or 4 days in 10 patients (13%). An additional 20 patients (27%) had otological symptoms 6 to 14 days prior to admission and the remaining 10 patients (13%) 15 days or more prior to hospitalization. Symptoms of recent or actual upper respiratory tract infection (22 patients) or fever alone (20 patients) were more frequently reported than nausea and vomiting (n = 7), headaches (n = 6), anorexia (n = 5), vertigo (n = 4), hearing loss (n = 4), or diarrhoea (n = 4). The duration of these symptoms prior to admission was 7 days or less in 21 patients, between 8 and 14 days in 15, and 15 days or more in 6 patients.
Table 1. History and findings on admission in 73 children with acute mastoiditis Features
Present (n)
History Otalgia Retro-auricular swelling Retro-auricular erythema Otorrhoea
42 34 19 13
Upper respiratory tract infection Fever alone (> 38.5~
22 20
Antibiotic treatment in the last 2 weeks
43
Previous ipsilateral otitis media once recurrent Previous surgical procedure on ipsilateral ear
15 11 8
Clinicalfindings Retro-auricular swelling tenderness erythema Abnormal tympanic membrane Protrusion of the auricle Swellingof regional lymph nodes Otorrboea Normal body temperature (_ 38.5~
63 59 58 48 45 33 23 31 21
Laboratory findings Erythrocyte sedimentation rate > 20 mm/b (n = 51) Leukocytosis (> 15 x 109/1) (n = 70)
43 35
Antibiotic therapy had been started in 35 patients (48%) 1 to 28 days prior to admission (mean 6 days). Eight other children had received antibiotics during the last 2 weeks prior to hospitalization, but treatment was discontinued at least 3 days before admission. Previous episodes of acute otitis media were reported by 26 patients or their parents, in 11 of these the episodes had recurred. In eight patients one or more surgical procedures had been performed on the ipsilateral ear: myringotomy and insertion of tympanostomy tubes (n = 5), mastoidectomy (n = 2), incision of a subperiosteal abscess (n = 2), removal of an epidermoid cyst (n = 1). The most frequent local finding was retro-auricular swelling followed by tenderness, erythema, and protrusion of the auricle. Retro-anricular swelling, tenderness, and erythema were found simultaneously in 50 (68%) patients. Forty-eight patients had a pathological tympanic membrane. Hyperaemia was most frequent (n = 35), followed by perforation (n = 23), bulging (n = 17), thickening (n = 16), retraction (n = 12), and blurred landmarks (n = 10). A diagnosis of co-existent acute otitis media was made in 35 patients (48%). Enlarged regional lymph nodes were noted in 33 patients: cervical (n = 25), angular (n = 20), nuchal (n = 7) and retro-auricular (n = 7). Fever over 38.5~ was noted in less than one third of the patients. Lumbar puncture was performed in eight patients with meningeal signs on admission, three of them disclosed pleocytosis. Two of these three had positive bacterial cultures of CSF (Table 2). The patient with pleocytosis without bacterial growth in his CSF had cholesteatoma with lateral sinus thrombosis and an intra- and extracranial abscess. One of the two patients with proven bacterial meningitis had additional labyrinthitis. Signs of CNS involvement were also noted in two further patients. Both had increased intracranial pressure: one with abducens paresis and concomitant subperiosteal abscess, the other with consecutive otitic hydrocephalus. Forty-two patients had conventional radiological examination, 34 (80%) yielded abnormal findings. These included clouding of the mastoid (n = 28), absent pneumatization (n =18), and sclerosis (n = 9). Computed tomography was performed in two patients because of suspected intracranial complication. An intracranial abscess was visualized in one of these two cases. The other case revealed no pathological finding. The erythrocyte sedimentation rate was determined in 51 patients and was found to be elevated (>_ 20 mm) in 43 (84%). Serum C-reactive protein was measured in ten patients; five had values higher than 90 mg/1 (range 91159), three slightly elevated (range 14-37), and two norreal (< i mg/1). A white blood cell count was performed in 70 patients disclosing leukocytes >15 • 109/1 in 35 (50%), 10-15 x 109/1 in 20 (29%), and < 10 • 109/1 in 15 (21%). Bacterial cultures were taken in 54 patients (77 samples) (Table 2) including 30 patients (43 samples) under oral antibiotic therapy. Cultures of retro-auricular abscesses yielded organisms in 81% (13/16), of mastoid
562 Table 2. Results of bacterial cultures in 54 children with acute mastoiditis Isolates
Patients (n=32)
Streptococcuspneumoniae Staphylococcus epidermidis Staphylococcus aureus Streptococcus pyogenes Pseudomonas aeruginosa Streptococcus viridans Haemophilus influenzae Peptostreptococcus spp. Propionibacterium spp. Escherichia coli
Samples (total number) Blood (n=21)
Mastoid mucosa (n = 22)
Retroauricular aspirate (n = 16)
Myfingotomy asp~ate (n = 10)
CSF
5 4 2 2 2b
4 2 2 2 1 -
1 1 1 1 -
1 -
1
-
-
-
-
9 6 4 4 3 2
(3) a (3) (2) (1) (2) (1)
2 -
1
(1)
.
1
(1)
-
-
1
.
-
-
1 1
(1)
.
.
.
(n=8)
1
.
. 1
.
a Numbers in parenthesis denote the number of patients on antibiotic therapy at the time of sampling b Cultures in one patient with chronic granulomatous disease grew in addition Candida albicans CSF = cerebrospinal fluid Table 3. Management of 73 children with acute mastoiditis Initial management
Patients Total (n=73)
IVAB alone IVAB plus myringotomy IVAB plus mastoidectomy with or without myringotomy
Delayed surgery (n=6)
S. aureus, and P. aeruginosa n = 2 respectively, and S. pyogenes n = 1) by means of the disk diffusion test. None of the S. epidermidis isolates was found to be methicillin-
Myringo- Mastoidtomy ectomy (n = 2) (n = 4)
resistant. There was no difference between the organisms isolated in the 1970s versus those in the late 1980s. Only one of the 73 patients had an underlying disorder. Cultures of the mastoid mucosa of this patient with chronic granulomatous disease grew Candida albicans and S.
26a
2
2
viridans.
8
0
2
39b
0
0
IVAB denotes intravenous antibiotics a Including one patient with central nervous system (CNS) involvement u Including 33 patients with initial subperiosteal abscess, one patient with CNS involvement, and three patients with both
mucosa in 68% (15/22), of middle ear aspirates in 40% (4/10), and blood cultures in 14% (3/21). The isolated bacteria were Streptococcus pneumoniae, Staphylococ-
cus epidermidis, S. aureus, S.pyogenes, Pseudomonas aeruginosa, S. viridans, Peptostreptococcus spp., Propionibacterium app., and Escherichia coli. Haernophilus influenzae type b and S. epidermidis proved to be the cause of bacterial meningitis in two cases. There was no significant difference between the isolation rate of cultured material of patients with or without previous antibiotic therapy. However, there was a higher incidence of isolation of S. pneumoniae and S. pyogenes in patients who had not previously taken antibiotics (6/24 and 3/24, respectively) compared to children previously treated with adequate antibiotics (3/30 and 0/30, respectively). In 7 of the 15 pre-treated patients with positive cultures, antimicrobial testing revealed that the isolated bacteria were resistant to the antibiotic prescribed (S. epidermidis,
The management and outcome of the 73 children are summarized in Table 3. All patients received different regimens of intravenous antibiotics on the day of admission for at least 3 days (mean duration 5 days). In most cases antibiotic therapy was continued orally afterwards. In addition, myringotomy alone or in combination with mastoidectomy was performed in 47 patients (64%). The reason for mastoid surgery in 37 of these 47 patients was a retro-auricular subperiosteal abscess (n = 33) or CNS involvement (n = 1), or both (n = 3). Mastoid surgery was also performed in 4 further patients of these 47 because of suspected subperiosteal abscess. The six remaining patients had myringotomy because of acute otitis media with effusion. Delayed surgery was performed in six patients: delayed myringotomy was done in two patients after the retro-auricular finding had improved, but middle ear effusion persisted; 2 of the 26 children initially treated with intravenous antibiotics alone and two of the eight children who had initial myringotomy required mastoidectomy because of insufficient local improvement 2 days or more after admission. A p a r t from the four patients who underwent delayed mastoid surgery, all patients improved or recovered from their local symptoms within 48 h. All symptoms had resolved after 4 days in 46 patients (63%), after i week in 59 (80%), and after 2 weeks in 70 (96%). The five children with CNS involvement were initially operated ex= cept one child who presented with meningitis but no sub-
563 periosteal abscess. All of the 73 patients healed without sequelae. Discussion Clinical acute mastoiditis, once a common complication of otitis media [15], has become a rare entity. A marked decrease in the incidence of acute mastoiditis was observed when antibiotics were introduced [15]. This retrospective study analyses 73 patients seen during a 16-year period at our hospitals with a mean of 5800 paediatric inpatients yearly. A similar low incidence was found in the few other retrospective studies, focusing on acute mastoiditis [4, 5, 7, 8, 10-12]. However, our study centre is a university hospital and milder cases of mastoiditis might not have been referred and may thus be underrepresented in our series. The definition of "acute mastoiditis" has not been uniform in former studies: while some authors failed to list their criteria for patient selection [10, 15], others included only patients with simultaneous acute otitis media [4, 8], or patients with both retro-auricular physical findings and radiological abnormalities [5], and still other investigators required that acute mastoiditis be confirmed at surgery [11]. The comparison of the results is therefore difficult and conclusions are limited. In this context we would like to emphasize that mastoiditis may progress and become clinically manifest at a time when infection in the middle ear has already resolved [1]. Thus, simultaneous acute otitis media should not be a mandatory finding for the diagnosis of acute mastoiditis. Also, diagnosis should not be based on radiographs. We and others [8] found that radiography may be misleading. Only in very rare occasions will mastoiditis be masked and escape clinical recognition [6, 8]. Acute mastoiditis affected all paediatric age groups, but more than one third of the children (36%) were younger than 24 months. This corresponds to the findings in another study [8] and reflects the age distribution of children with otitis media [1]. In our series nearly half of the patients were on oral antibiotics at admission. The reported percentage of patients on antibiotic regimens developing acute mastoiditis ranges from 36%-71% [5, 7, 8, 11-13]. This raises the question whether treatment failure was due to low compliance and/or inadequate antibiotic therapy. In 7 of our 15 patients who had been treated with oral antibiotics prior to admission and in whom bacteriological workup revealed growth, the isolates were resistant to the prescribed antibiotics. Treatment failure in the remaining 8 of the 15 patients could, beside low compliance, have been due to a moderate or inadequate dosage of the antibiotics [14] or due to an insufficient penetration of the drug into the inflamed tissue [10]. The latter could explain our finding of similar isolation rates of bacteria in pre-treated and not treated patients. However, in a recent study including only patients with surgically proven mastoiditis some purulent discharges were found to have been sterilized by the pre-operative treatment [11]. Although acute mastoiditis is a complication of acute otitis media, the incidence of the bacteria isolated in
both entities is not identical. The predominant causative agents of acute otitis media are S. pneumoniae and H. influenza, followed by Branhamella catarrhalis, S. pyogens, and S. aureus [3]. Enterobacteria and anaerobes are less frequent. In acute mastoiditis S. pneumoniae is also the most common pathogen and the spectrum of the other agents is similar, but, as shown by the present and former studies [4, 5, 8, 9, 11, 12], H. influenzae is only rarely isolated. The reason for this distinct microbiological pattern is not clear. Antibiotic therapy before the onset of acute mastoiditis might be one factor. We and others [11] have demonstrated that the frequency of S. pneumoniae and S. pyogenes isolation was considerably lower in patients previously treated with antibiotics than in patients without previous treatment (in the present study: 10% vs 25% and 0% vs 12.5%, respectively). Also infections with H. influenzae are less apt to progress to more severe degrees of mastoiditis with subperiosteat abscess formation affording mastoidectomy. The isolation rate of bacteria from blood cultures is extremely low in acute mastoiditis (in the present study 14%). Thus, the incidence of the bacteria isolated might not correspond to the true frequency, since cultures were mainly derived from a subperiosteal abscess or mastoid mucosa. In contrast to other investigators, we found a considerable number of cases in whom S. epidermidis (6/32 patients with positive cultures) was isolated from subperiosteal abscess aspirates or mastoid mucosa. Since the specimens were taken under sterile conditions in the operating room, contamination seems unlikely. None of the isolates was found to be methicillin-resistant. We postulate that S. epidermidis which is one of the predominant pathogens in chronic otitis media with effusion [3] should be included in the list of the main agents causing acute mastoiditis. The retrospective design of our study does not allow treatment recommendations with certainty. However, the fact that 24 of our patients (33%), none with subperiosteal abscess but one with meningitis, recovered with intravenous antibiotics without mastoid surgery provides clear evidence for mastoidectomy not being obligate in the management of acute mastoiditis. Remarkably, none of our patients showed progression of the infection or intracranial or meningeal spread while on intravenous antibiotics, including the four patients who required delayed mastoid surgery because of insufficient clinical improvement. A similar experience was reported by other investigators [4, 5]. Thus, in children with acute mastoiditis without subperiosteal abscess or CNS involvement, we would recommend as initial treatment a 48 h trial of intravenous antibiotics. These should be directed against the major pathogens of acute otitis media and, according to our findings, also against staphylococci. The combination amoxicillirdclavulanic acid would fit these requirements. In addition, myringotomy should be performed and ventilation tubes inserted if acute otitis media with effusion is present. Mastoid surgery should be considered first if the antibiotic therapy is not followed by clinical improvement within 2 days or if a subperiosteal abscess is diagnosed. In the presence of neurological symptoms high resolution computed tomography is indicated
564 to d e t e r m i n e the e x t e n t of the process a n d to p l a n surgery. T h e t i m e to switch to oral antibiotics will d e p e n d o n the p a t i e n t ' s c o n d i t i o n . It is i m p o r t a n t that p a e d i a trician a n d otologist c o l l a b o r a t e in the m a n a g e m e n t of acute mastoiditis.
References 1. Bluestone CD (1983) Otitis media in children: to treat or not to treat? N Engl J Med 306:1399-1404 2. Bluestone CD, Klein JO (1983) Mastoiditis. In: Bluestone CD, Stool SE (eds) Pediatric otolaryngology. Saunders, Philadelphia, pp 546-552 3. Giebink GS (1989) The microbiology of otitis media. Pediatr Infect Dis [Suppl] 8:$18-$20 4. Ginsburg CM, Rudoy R, Nelson JD (1980) Acute mastoiditis in infants and children. Clin Pediatr 19: 549-553 5. Hawkins DB, Dru D, House JW, Clark RW (1983) Acute mastoiditis in children: a review of 54 cases. Laryngoscope 93 : 568-572
6. Holt GR, Gates GA (1983) Masked mastoiditis. Laryngoscope 93 : 1034-1037 7. Holt GR, Young WC (1981) Acute coalescent mastoiditis. Otolaryngol Head Neck Surg 89 : 317-321 8. Ogle JW, Lauer BA (1986) Acute mastoiditis. Diagnosis and complications. Am J Dis Child 140:1178-1182 9. Palva T, Virtanen H, M~ikinen H (1985) Acute and latent mastoiditis in children. J Laryngol Otol 99:127-136 10. Pfaltz CR, Griesemer C (1984) Complications of acute middle ear infections. Ann Otol Rhinol Laryngol [Suppl] 112:133-137 11. Prellner K, Rydell R (1986) Acute mastoiditis. Influence of antibiotic treatment on the bacterial spectrum. Acta Otolaryngol 102: 52-56 12. Rosen A, Ophir D, Marshak G (1986) Acute mastoiditis: a review of 69 cases. Ann Otol Rhinol Laryngol 95 : 222-224 13. Rubin JS, Wei WI (1985) Acute mastoiditis: a review of 34 patients. Laryngoscope 95 : 963-965 14. Tudor RB (1963) Mastoiditis in children. A review of 77 cases. Lancet 83 : 56-58 15. Zoller H (1972) Acute mastoiditis and its complications: a changing trend. South Med J 65 : 477-480
