January 1978 The Journal o f P E D I A T R I C S
Aerobic and anaerobic bacteriology of acute otitis media in children Tympanocentesis and aerobic and anaerobic cultivation of the middle-ear fluid obtained was performed through one or both tympanic membranes of 62 children with acute otitis media. Aerobic bacteria alone, predominantly pneumoeoecus and Hemophilus influenzae, were isolated.from 57% of patients," anaerobic organisms alone, most commonly Propionibacterium acnes and Peptococcus, from 15%. Thirteen percent yielded mixtures of aerobes and anaerobes, and several had multiple aerobic or anaerobic agents. The isolation of only anaerobic bacteria, supported in some patients by gram-stain observations o f middle-ear fluid, suggest that these bacteria, especially Peptococcus, may occasionally play a direct role in acute otitis media.
I t z h a k Brook, M . D . , Irvine, Calif., B a s c o m F. Anthony, M . D . , and
Sydney M. Finegold, M.D.,* L o s A n g e l e s , Calif.
T H E B AC T ER I O L O G Y of acute otitis media in children is well defined; the most common pathogens are Streptococcus pneumoniae and Hemophilus influenzae. 1 Anaerobic bacteria, however, have not been related to acute otitis media, in spite of their presence in the oropharynx and their association with chronic otitis media3 -4 We undertook the present study to re-examine the bacteriology of acute otitis media in children, employing both anaerobic and aerobic methodology.
MATERIALS AND M E T H O D S Sixty-two children with the diagnosis of acute otitis media who presented to the Pediatric Acute Care Clinic at Harbor General Hospital between February 1 and March 31, 1976, were studied. Their ages ranged from four months to eight years (mean 21/12years). The criteria for consideration for the study were the presence of acute febrile illness, inflamed but intact tympanic membrane with decreased or absent motility, absence of other apparent foci of infection, and lack of prior antibiotic From the Department of Medicine and Anaerobic Bacteriology Laboratory, Wadsworth Hospital Center, Veterans Administration, and the Department o f Pediatrics, Harbor General Hospital Campus, UCLA School of Medicine. *Reprint address: WadsworthHospital Center, Veterans Administration, Wilshire and Sawtelle Birds., Los Angeles, CA 90073.
therapy. Informed, written parental consent was obtained for participation in the study. The external ear canal was cleansed of cerumen with a blunt curet when indicated, but no disinfectant preparation of the canal or tympanic membrane was attempted. Tympanocentesis was done with an intravenous cannUla set (Medicut, Sherwood Medical Instruments, Inc., St. Louis Mo.), that consisted of an 18-gauge needle attached to a 2 ml syringe and covered by a plastic cannula. The needle was bent slightly and the cannula was slipped forward to cover the bevel. Adjacent to the tympanic membrane, the cannula was retracted, the membrane was entered in the posterior, inferior quadrant, and negative pressure was applied through the syringe. No contact was made by the needle tip With the speculum or auditory canal. The aspirated fluid was diluted in 0.5 ml of thioglycolate broth (Difco Laboratories, Detroit, Mich.) and inoculated into aerobic and anaerobic media. The former included sheep blood agar and chocolate agar plates, which were incubated at 37~ under 5% COs and examined at 24 and 48 hours. Anaerobic media included prereduced vitamin Ki-enriched brucella blood agar s and thioglycolate broth, they were inoculated, placed at 37~ under anaerobic conditions (GasPak jars, BBL, Cockeysville, Md.) and examined at 48 and 96 hours. The thioglycolate broth was incubated for 14 days. Anaerobes were identified by techniques previously described?
VoL 92, No. 1, pp. 13-15
Brook, A nthony, and Finegold
The Journal of Pediatrics January 1978
Table I. Bacterial cultures of 62 cases of acute otitis media
No. of patients Patients with aerobic organisms only
Streptococcus pneumoniae Strep. pneumoniae and H. influenzae H. influenzae H. influenzae and Staph. aureus H. parainfluenzae Staph. epidermidis Staph. aureus
10 3 10
1 1 4 1
Group D streptococci Alpha hemolytis streptococci (not group D)
Subtotal Patients with anaerobic organisms only
Peptococcus Peptococcus and Propionibacterium acnes Propionibacterium acnes
3 1 5
Subtotal 9 Patients with mixed aerobic and anaerobic organisms
Peptocoecus and Staph. aureus and Group D streptococci Peptococcus and Strep. pneumoniae and H. influenzae Peptococcus and Eubacterium lentum and H. influenzae Propionibacterium acnes and Strep.
pneumoniae Propionibacterium acnes and Staph. aureus Propionibacterium acnes and H. influenzae Subtotal Total of all patients with positive cultures Patients with negative cultures
1 1 1 3 1 1 8 52 l0
Aerobic bacteria were identified by the staff of the clinical microbiology laboratory at Harbor General Hospital using standard techniques. Gram- and Wright-stained preparations were m a d e of all aspirates. The patients were managed by pediatric house officers. Most were treated for seven to ten days with oral medications, most c o m m o n l y ampicillin, and then were re-examined and re-evaluated. RESULTS One hundred fifteen middle-ear aspirates were obtained from the 62 patients. In 53 patients bilateral tympanocentesis was performed, and in three patients repeated aspirates were done within 10 to 14 days when
there was persistence or recurrence of symptoms. The bacteriologic results from initial, pretreatment tympanocentesis are summarized in Table I. Aerobic bacteria only were isolated in 35 patients (57%) and anaerobic organisms only from nine (15%). Mixed aerobic and anaerobic isolates were recovered from eight children (13%), and cultures were negative in ten (16%). Thus aerobes were recovered from almost 70% and anaerobes from 27% of the children studied. Streptococcus pneumoniae was isolated from 17 patients (27%). In ten cases, pneumococci were recovered in pure culture, in three instances with Hemophilus influenzae, and from one patient with both H. influenzae and peptococci. H. influenzae was isolated from 17 patients (27%) and 11. parainfluenzae in one case. Staphylococcus epidermidis was recovered in four instances. Staphylococcus aureus was isolated in pure culture in one case, and in one case was mixed with Propionibacterium acnes and in another with group D streptococci and Propionibacterium acnes. Pseudomonas aeruginosa was isolated in pure culture in two cases, one of which was a case of recurrent otitis media. Anaerobic organisms were isolated from 17 patients (27%) (Table I). Propionibacterium were identified in ten patients (16%), in pure culture in five, and in mixed culture in five cases. All of these isolates were identified as Propionibacterium acnes except one which was identified as Propionibacterium avidum. Peptococcus was cultured from seven patients (1 t%), in three as the only bacterial isolates and in two others with other anaerobes. One of these was identified as Peptococcus variabilis. In one patient Peptococcus was isolated with Eubacterium lentum from one ear, and H. influenzae alone was recovered from the other ear. Observations of the Gram- and Wright=stained preparations of the ear aspirate of 30 patients revealed polymorphonuclear leukocytes, most often with pneumococcal or Hemophilus infections, but also with other isolates, including two of three patients with Peptococcus in pure culture. Gram-stain confirmation of the presence of gram-positive anaerobic cocci was established in two instances. DISCUSSION The aerobic organisms recovered in this study are consistent with previous studies o f acute otitis m e d i a ? Streptococcus pneumoniae and H. influenzae were the major pathogens; they were isolated in more than 50% of the patients. Anaerobic bacteria, however, in the middle-ear fluid of children with acute infection, to our knowledge, have not
Volume 92 Number 1
Bacteriology of otitis media
been reported. The recovery o f gram-positive anaerobic organisms as the only isolate in some patients and in combination with other bacteria in others suggests a possible role for anaerobes in acute otitis media. Peptococci have been isolated from serious pleuropulmonary infections in adults and are part o f the normal oropharyngeal flora. ~' 7 Propionibacterium acnes, on the other hand, is a component of the skin flora and is rarely a pathogen? -1~ Despite our precautions to avoid contamination, it is possible that some isolations of anaerobic as well as aerobic bacteria represented contamination of middle-ear aspirate by the external ear flora. Further studies, with appropriate anaerobic cultures, of middle-ear fluid and of the flora of the external auditory canal, should cast some light on this possibility. Although the anaerobic bacteria described here are relatively avirulent in healthy individuals, the association of acute otitis media with bacteria of relatively low pathogenicity is well known. Moreover, anaerobes may conceivably explain the finding of negative bacterial cultures obtained in m a n y cases of acute otitis media, since viruses and mycoplasma do not appear to play a significant role. 11 With the recent developments in anaerobic bacteriology, the processing and identification of these organisms has become feasible. Their possible role in acute otitis media warrants further study.
1. Howie VM, Ploussard JH, and Sloyer JL: Natural history of otitis media, Ann Otol Rhinol Laryngol 85:18, 1976. 2. Pliot I, and Pearlman SJ: Studies in fusiform bacilli and spirochetes. V. Occurrence in otitis media chronica, J Infect Dis 33A:139, 1923. 3. Palva T, and Hallstrom O: Bacteriology of chronic otitis media. Results of analysis from ear canal and from the operative cavity, Arch Otolaryngol 82:359, 1965. 4. Palva T, Karja J, Palva A, and Raunis V: Bacteria in chronic ear infection, pre- and postoperative evaluation, Pract OtoRhino-Laryngol 31:30, 1969. 5. Sutter VL, Vargo VL, and Finegold, SM: Wadsworth anaerobic bacteriology manual, ed 2, Los Angeles, 1975, UCLA Extension Divison. 6. Bartlett JG, and Finegold SM: Anaerobic pleuropulmonary infections, Medicine 51:413, 1972, 7. Bartlett JG, Gorbach SL, and Finegold SM: The bacteriology of aspiration pneumonia, Am J Med 56:202, 1974. 8. Dunkle LM, Brotherton TJ, and Feigin RD: Anaerobic infections in children: A prospective study, Pediatrics 57:311, 1976. 9. Thirumoorthi MC, Keen BM, and Dajani AS: Anaerobic infection in children: a prospective survey, J Clin Microbiol 3:318, 1976. 10. Everett ED, Eickhoff TC, and Simon RH: Cerebrospinal fluid shunt infections with anaerobic diphtheroids, J Neurosurg 44:580, 1976. 11. Klein JO, and Teele DW: Isolation of viruses and mycoplasma from middle-ear effusions: A review, Ann Otol Rhinol Laryngol 85:140, 1976.
THrS is a provocative study of the possible pathogenic role of anaerobic bacteria in otitis media of infancy and childhood. The authors correctly point out that their data are inconclusive and that additional studies are warranted. The fact that anaerobes alone were recovered from cultures of 15% of patients with otitis media is consonant with the observations of others that bacteria usually considered nonpathogenic and part of the normal
nasopharyngeal flora can be isolated in 10 to 20% of individuals with acute middle-ear disease. 1 4 It is possible that the role of these organisms in the pathogenesis of acute otitis media could be assessed in an experimental animal model such as that established in chinchilla by Giebink and associates? A major shortcoming is the failure of the investigators to disinfect the canal and m e m b r a n e before tympanocen-
Vol. 92, No. 1, pp. 15-16