Eikenella corrodens in head and neck infections

Eikenella corrodens in head and neck infections

Journal of Infection (2007) 54, 343e348 www.elsevierhealth.com/journals/jinf Eikenella corrodens in head and neck infections Tsuyoshi Udaka a, Nobua...

131KB Sizes 6 Downloads 91 Views

Journal of Infection (2007) 54, 343e348

www.elsevierhealth.com/journals/jinf

Eikenella corrodens in head and neck infections Tsuyoshi Udaka a, Nobuaki Hiraki a, Teruo Shiomori a, Hiroshi Miyamoto b, Takeyuki Fujimura a, Tsuyoshi Inaba a, Hideaki Suzuki a,* a

Department of Otorhinolaryngology, School of Medicine, University of Occupational and Environmental Health, 1-1 Iseigaoka, Yahatanishi-ku, Kitakyushu 807-8555, Japan b Department of Microbiology, School of Medicine, University of Occupational and Environmental Health, 1-1 Iseigaoka, Yahatanishi-ku, Kitakyushu 807-8555, Japan Accepted 6 June 2005 Available online 7 September 2006

KEYWORDS Eikenella corrodens; Head and neck; Infection; Tonsil; Streptococcus milleri group

Summary Background: Eikenella corrodens (E. corrodens) is a Gram-negative facultative anaerobic bacillus that originally was thought to be an attenuated and indigenous bacterium. In recent years, a number of reports have documented that E. corrodens can be a potential pathogen not only in immunocompromised patients but also in hosts with normal immunity. We herein study E. corrodens infections of the head and neck encountered in our department. Methods: Twenty-two consecutive patients treated in our department for E. corrodens infections of the head and neck were retrospectively analyzed. Microbial specimens were subjected to light microscopic examination, aerobic culture using chocolate and sheep blood agar media, and anaerobic culture using Brucella HK agar medium. Cultured bacteria were subjected to antimicrobial susceptibility tests by means of the broth microdilution method. Results: There were 16 males and 6 females with an average age of 29.9 years. Two patients had malignancy, while the other patients had no particular risk factors or underlying diseases. Infected sites were the ear in 6 cases, pharynx in 12 cases (tonsil in 10 cases), paranasal sinuses in 3 cases, and salivary gland in 1 case. Seventeen patients suffered polymicrobial infections. Staphylococcus and Streptococcus were the most frequently detected pathogens coexisting with E. corrodens, and mixed infections of E. corrodens and Streptococcus milleri group bacteria were prone to form abscesses. Isolated E. corrodens was susceptible to third-generation cephems (MIC90 Z 0.15e0.25 mg/ml), carbapenems (MIC90 & 0.15 mg/ml), and new quinolones (MIC90 & 0.15 mg/ml), and resistant to oxacillin (MIC90 > 8 mg/ml), cefazolin (MIC90 > 4 mg/ml), macrolides (MIC90 Z 4e8 mg/ml), and clindamycin (MIC90 > 4 mg/ml). Conclusions: E. corrodens infections of the head and neck occur most frequently in the tonsil even in hosts with normal immunity. Coexistence with Streptococcus milleri group bacteria and the use of ineffective antibiotics can be exacerbating factors. First-choice drugs for E. corrodens infections should be third-generation cephems, carbapenems, or new quinolones. ª 2006 The British Infection Society. Published by Elsevier Ltd. All rights reserved.

* Corresponding author. Tel.: þ81 93 691 7448; fax: þ81 93 601 7554. E-mail address: [email protected] (H. Suzuki). 0163-4453/$30 ª 2006 The British Infection Society. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.jinf.2005.06.015

344

Introduction Eikenella corrodens (E. corrodens) is a Gram-negative facultative anaerobic bacillus1 that was first isolated by Eiken in 1958 from human saliva.2 In primary culture, E. corrodens grows under anaerobic conditions only, but survives in aerobic conditions after subcultivation.1 This bacterium absolutely requires 5e10% of carbon dioxide at the beginning of primary culture, and hemin must be supplemented for it to grow under aerobic conditions.3 Moreover, it takes several days for E. corrodens to form macroscopic colonies, which are generally small in size and are often hidden by colonies of other organisms.3 E. corrodens is, therefore, usually overlooked during routine bacteriological examinations, and special laboratory precautions and techniques are necessary to isolate this bacterium.4 E. corrodens was originally thought to be an attenuated and indigenous bacterium in the human oral cavity, gastrointestinal tract, genitalia, etc.,5 and did not arouse much interest until the 1960s. However, since the 1970s, a number of reports have documented that E. corrodens can be a potential pathogen of various infective diseases not only in immunocompromised patients but also in hosts with normal immunity.6e8 It is also interesting that E. corrodens enhances the virulence of coexisting bacteria such as Streptococcus species in experimental animals.9,10 Although there are sporadic case reports describing E. corrodens infections of the head and neck, systematic analysis of the significance of this pathogen in the field of otorhinolaryngology has not been performed yet. Here, we present 22 cases with E. corrodens infections encountered in our department during the past 7 years, and summarize their clinical characteristics and potential problems upon treatment.

Materials and methods The subjects were 22 consecutive patients with E. corrodens infections of the head and neck who were treated in the Department of Otorhinolaryngology of the University of Occupational and Environmental Health between January 1997 and February 2004. Smear samples taken from the foci of infections were Gram-stained and observed under a light microscope at a magnification of 1000. Two different methods were performed for bacterial cultures. For aerobic cultures, the samples were inoculated on chocolate and sheep blood agar media, and incubated under a humidified atmosphere containing 5% CO2 at 37  C for 48 h. For anaerobic culture, the samples were inoculated on Brucella HK agar medium and incubated under a mixture of 90% N2, 5% CO2, and 5% H2 at 37  C for 2e7 days. To semi-quantify the number of bacteria in the samples, they were serially diluted by 10 and cultured as described above. Bacterial colonies were picked up and examined using a Gram-negative bacillus identification card (VITEK; BioMe ´rieux Co., Tokyo, Japan) to determine the species of bacteria. Cultured bacteria were subjected to antimicrobial susceptibility tests by means of the broth microdilution method using Dryplate-Eiken disks (Eiken Chemical Co., Tokyo, Japan) in accordance with the guidelines of the

T. Udaka et al. National Committee for Clinical Laboratory Standards.11 A bacterium was judged to be pathogenic if it met two criteria: (a) smear preparations showed the coexistence of phagocytes with the bacterium,12 and (b) there were estimated to be 107/ml or more of the bacterium in the primary culture.13,14

Results Profiles of the patients are summarized in Table 1. The average age was 29.9 years; 11 patients were under 15 years of age. There were 16 males and 6 females. Infected sites were the ear in 6 cases, pharynx in 12 cases (tonsil in 10 cases), paranasal sinuses in 3 cases, and salivary gland in 1 case (Table 2). The clinical courses of infection were acute in 9 cases and chronic in 13 cases. Twelve patients underwent surgical treatment while the other 10 were treated conservatively. Two patients had undergone radical surgery for laryngeal carcinomas (patients 9 and 10), while the other patients had no particular risk factors or underlying diseases. The bacteria co-isolated with E. corrodens are listed in Table 3. E. corrodens was detected as a single pathogen in 5 patients, whereas the other 17 suffered polymicrobial infections. The isolates other than E. corrodens were Staphylococcus in 11, Streptococcus in 7, Pseudomonas in 2, Moraxella in 2, Corynebacterium in 2, Haemophilus in 2 cases, and others. Streptococcus milleri group bacteria, such as Streptococcus constellatus and Streptococcus intermedius, were co-isolated in 5 cases (patients 6, 8, 11, 16, and 17), and 3 of them exhibited abscess formation (patients 6, 8, and 17). There were 4 cases with abscess formation (patients 6, 8, 13, and 17), and 3 of them had mixed infections of E. corrodens and Streptococcus milleri group bacteria (patients 6, 8, and 17). Table 4 shows the MICs of antimicrobial agents for isolated E. corrodens. The isolates were susceptible to third-generation cephems (MIC90 Z 0.15e0.25 mg/ml), carbapenems (MIC90 & 0.15 mg/ml), and new quinolones (MIC90 & 0.15 mg/ml); moderately susceptible to benzylpenicillin (MIC90 Z 1 mg/ml), ampicillin (MIC90 Z 1 mg/ml), and minocycline (MIC90 Z 1 mg/ml); and resistant to oxacillin (MIC90 > 8 mg/ml), cefazolin (MIC90 > 4 mg/ml), macrolides (MIC90 Z 4e8 mg/ml), and clindamycin (MIC90 > 4 mg/ ml). Of the 4 cases with abscess formation, 3 were initially given clindamycin (patients 8, 13, and 17), but clinical improvement was obtained only after surgical drainage.

Discussion Joshi reviewed 24 patients with E. corrodens infections of the lower respiratory tract.5 He found that 20 of the patients (83%) had some underlying disease, such as diabetes mellitus, malignancy, steroid use, alcoholism, cerebrovascular accidents, polymyosis, or chronic obstructive lung disease, any of which might induce several adverse factors: immunosuppression, propensity to pulmonary aspiration, and poor mucociliary clearance. However, in the present study only 2 patients (9%) had malignancy, and the other 20 did not have such underlying diseases. To the best of our knowledge, 29 cases with E. corrodens infections of

E. Corrodens in head and neck Table 1

Profiles of patients with E. corrodens infections of the head and neck in our department

Patient Age and sex (year) 1 2 3 4 5 6 7 8 9

10

11 12 13 14 15 16 17 18 19 20 21 22

345

Disease (microbial specimen) Underlying Treatment disease

11, M (1997) Cholesteatoma (otorrhea) 30, M (1997) Acute sinusitis (pus from middle meatus) 14, M (1998) Acute parotitis (pus from Stensen’s duct) 6, M (1999) Cholesteatoma (otorrhea) 11, M (1999) Cholesteatoma (otorrhea) 11, M (1999) Congenital aural fistula (pus from abscess) 13, M (1999) Acute otitis media (otorrhea) 44, M (1999) Peritonsillar abscess (pus in abscess) 77, M (1999) Postoperative wound infection (pus from pharyngeal fistula) 82, M (1999) Postoperative wound infection (pus from pharyngeal fistula) 4, M (2000) Chronic tonsillitis (tonsillar tissue) 10, M (2000) Chronic otitis media (otorrhea) 11, F (2000) Peritonsillar abscess (pus in abscess) 45, M (2000) Acute tonsillitis (pus from crypt) 7, F (2001) Chronic tonsillitis (tonsillar tissue) 29, F (2001) Acute tonsillitis (pus from crypt) 48, M (2001) Paranasal sinus abscess (pus in abscess) 53, F (2001) Acute tonsillitis (pus from crypt) 34, F (2002) Chronic tonsillitis (tonsillar tissue) 92, F (2003) Acute sinusitis (pus from middle meatus) 6, M (2003) Chronic tonsillitis (tonsillar tissue) 20, M (2003) Chronic tonsillitis (tonsillar tissue)

Co-isolated pathogen

None

Clean up

None

Cefmetazole, drainage

None

Ceftazidime

None None

Clean up Clean up

None

Laryngeal cancer

Cefcapene pivoxil drainage Ofloxacin MSSA, Moraxella catarrhalis Ceftazidime, clindamycin Streptococcus intermedius surgical drainage Ceftazidime, MRSA, Pseudomonas aeruginosa debridement, irrigation

Laryngeal cancer

Ceftazidime, debridement, irrigation

MRSA, Staphylococcus sp., Pseudomonas aeruginosa

None

Tonsillectomy

None

Irrigation

Streptococcus intermediu[s], Klebsiella oxytoca Staphylococcus sp.

None None

Ceftazidime, clindamycin None surgical drainage Imipenem None

None

Tonsillectomy

MSSA, Haemophilus influenzae

None

Cefdinir

Streptococcus constellatus

None None

Ceftazidime, clindamycin Streptococcus intermedius, surgical drainage Haemophilus parainfluenzae Cefpirome, clindamycin None

None

Tonsillectomy

MSSA

None

Clarithromycin

None

None

Tonsillectomy

Streptococcus sp.

None

Tonsillectomy

MSSA, Streptococcus sanguis

None None

the head and neck have been reported in the English-language literature,3,6e8,15e26 and only one of those cases had underlying diseases. These lines of evidence indicate that E. corrodens in the upper airway, as opposed to the lower respiratory tract, often exerts its pathogenicity even in hosts with normal immunity. There were 10 patients (45%) with tonsillar infections in the present study, suggesting a close relationship between E. corrodens and tonsillar infections. In contrast, only 3 tonsillar infections (10%) are included in the 29 previously reported cases of E. corrodens infections of the head and

MSSA, Moraxella catarrhalis, Corynebacterium jeikeium Stenotrophomonas maltophilia, Enterobacter cloacae None MSSA Staphylococcus sp. Corynebacterium xerosis Streptococcus constellatus

neck.18,22,25 Microbial specimens of tonsillar infections can be obtained by several different methods: swabbing the surface of the tonsil, collecting debris or purulent material from the tonsillar crypts, aspiration of the pus from abscesses, and most directly, dissecting a piece of tissue from surgically resected tonsils. A discrepancy between the microflorae of the surface and those of the core of the tonsil has been reported.27,28 Surow et al. demonstrated that swab cultures from the pharyngeal wall do not reliably reflect pathogens in the tonsil core.27 Blook et al. reported that the anaerobic microorganisms most

346

T. Udaka et al.

Table 2 Summary of infected sites and microbial specimens Infection Disease site

Microbial specimen

Ear

Otorrhea Otorrhea Otorrhea Pus from fistula

3 1 1 1

Tonsillar tissue Pus from crypt Pus in abscess Pus from pharyngeal fistula Pus from middle meatus Pus in abscess

5 3 2 2

Pus from Stensen’s duct

1

Nose

Cholesteatoma Acute otitis media Chronic otitis media Congenital aural fistula Chronic tonsillitis Acute tonsillitis Peritonsillar abscess Postoperative wound infection Acute sinusitis

Other

Paranasal sinus abscess Acute parotitis

Pharynx

No. of cases

Total

2 1

22

frequently isolated by swab cultures from the inflamed tonsil were Bacteroides sp., Fusobacterium nucleatum, Eubacterium sp., and anaerobic Gram-positive cocci, whereas E. corrodens was rarely isolated.28 On the other hand, Khun et al. examined the tonsillar tissues of surgical specimens from 15 tonsillectomized patients with recurrent tonsillitis and isolated E. corrodens in 8 cases (53%).29 These observations suggest that E. corrodens may be harbored deep in the tonsil and thereby overlooked in surface swab culture. In the present study, microbial specimens were obtained from the core of the tonsil in most cases with tonsillar infections: tonsillar tissues from 5 tonsillectomized patients and pus from 2 patients with peritonsillar abscesses (Table 2). Meanwhile, no tonsillectomized cases

Table 3 Summary of pathogens co-isolated with E. corrodens Co-isolated pathogen

No. of isolates

Staphylococcus aureus Staphylococcus sp. Streptococcus intermedius Streptococcus constellatus Pseudomonas aeruginosa Moraxella catarrhalis Streptococcus sanguis Streptococcus sp. Haemophilus influenzae Corynebacterium xerosis Corynebacterium jeikeium Klebsiella oxytoca Haemophilus parainfluenzae Stenotrophomonas maltophilia Enterobacter cloacae Total

8 3 3 2 2 2 1 1 1 1 1 1 1 1 1 29

are included in the 29 previously reported cases of E. corrodens infections of the head and neck,3,6e8,15e26 partially explaining why tonsillar infections are seemingly minor. Seventy-seven percent of the cases (17/22) in the present study had polymicrobial infections. Staphylococcus and Streptococcus were the most frequently detected bacteria coexisting with E. corrodens (Table 3). The role of E. corrodens in mixed infections with Staphylococcus is unclear. Meanwhile, Brooks et al. reported that larger abscesses were formed in rabbits by inoculation with a mixture of E. corrodens and Streptococci than by inoculation with either bacterium alone.9 They stressed that the enhanced virulence resulted from the coexistence of these two pathogens. Within the Streptococcus genus, Streptococcus milleri group bacteria, such as Streptococcus constellatus, Streptococcus intermedius, and Streptococcus anginosus are known to be aggressive pathogens in the head and neck, with a propensity for abscess formation.30 In particular, Streptococcus constellatus and Streptococcus intermedius have recently been reported to be extremely pathogenic in combination with E. corrodens: Young et al.10 reported that exponential growth of Streptococcus constellatus and Streptococcus intermedius, in mixed culture with E. corrodens, occurred within 6 h after inoculation, whereas it occurred after 25 h without E. corrodens. In the present study, 3 of the 5 patients with mixed infections of E. corrodens and Streptococcus milleri group bacteria exhibited abscess formation. In addition, 3 of the 4 cases with abscess formation had mixed infections of E. corrodens and Streptococcus milleri group bacteria. These results imply that a strong synergy might exist between the two bacteria, thereby enhancing their virulency. E. corrodens has been known to show an unusual antimicrobial susceptibility pattern. It has been reported that E. corrodens is susceptible to antibiotics such as benzylpenicillin, ampicillin, third-generation cephems, and chloramphenicol; relatively resistant to aminoglycosides and erythromycin; and resistant to clindamycin, metronidazole, and penicillinase-resistant penicillin.3,9,31 According to our results of antimicrobial susceptibility tests shown in Table 4, the first-choice drugs for E. corrodens infection should be third-generation cephems, carbapenems, or new quinolones. It is intriguing that, as shown in Table 4, E. corrodens was resistant to clindamycin, which is usually effective against anaerobic bacteria and is often used for the treatment of severe head and neck infections. Of the 4 present cases with abscess formation, 3 were initially given clindamycin, but clinical improvement was obtained only after surgical drainage. Brooks et al.9 documented that the growth of E. corrodens was enhanced rather than inhibited when clindamycin was added to the culture medium. We must, therefore, be cautious in using this antibiotic if an E. corrodens infection is suspected. Although some authors have reported the efficacy of erythromycin for E. corrodens infections,9 the present results demonstrated the resistance of E. corrodens against macrolide antibiotics (Table 4). The ineffectiveness of macrolides in the present study may be attributable, at least in part, to macrolide therapy, which is prevalent as a treatment for chronic upper and lower airway diseases in Japan and may thus produce macrolide-resistant bacteria.32 Patients with

E. Corrodens in head and neck Table 4

347

Antimicrobial susceptibility of E. corrodens Antimicrobial agent

Penicillin

I cephem III cephem

Carbapenem Macrolide Lincomycin Tetracycline New quinolone

Benzylpenicillin Ampicillin Oxacillin Cefazolin Cefotaxime Cefpodoxime Cefditoren Cefdinir Imipenem Panipenem Erythromycin Clarithromycin Clindamycin Minocycline Tosufloxacin Levofloxacin

MICs (mg/ml) MIC50

MIC90

Range

No. of isolates tested

1 0.5 >8 4 &0.063 0.15 &0.063 &0.063 &0.063 &0.063 4 4 >4 0.5 &0.063 &0.063

1 1 >8 >4 0.15 0.25 0.15 0.25 0.15 &0.063 8 4 >4 1 &0.063 0.15

&0.063e2 0.25e8 2e>8 1e>4 &0.063e0.15 &0.063e0.25 &0.063e0.15 &0.063e0.25 &0.063e0.15 &0.063 0.25e>16 1e4 1e>4 1e2 &0.063 &0.063e0.15

22 22 12 16 15 16 10 6 6 10 20 6 22 22 6 16

MIC50 and MIC90 represent MICs that inhibit the growth of 50 and 90% of isolates, respectively.

chronic infective diseases would have repeated chances of receiving various antibiotics, leading to shifts in the composition of oral and pharyngeal microflorae and to the selection of E. corrodens. This may be one of the factors that drives E. corrodens, by nature an attenuated and indigenous bacterium, to acquire pathogenicity in hosts with normal immunity.

Conclusion E. corrodens infections of the head and neck encountered in our department were retrospectively studied. The vast majority of the patients were hosts with normal immunity. The tonsil was the most frequent site of infection. Staphylococcus and Streptococcus were the most frequently detected coexisting bacteria, and mixed infections of E. corrodens and S. milleri group bacteria were prone to form abscesses. The antimicrobial susceptibility indicated that first-choice drugs for E. corrodens infection should be third-generation cephems, carbapenems, or new quinolones. Further studies are needed to more clearly understand the characteristics of E. corrodens infections of the head and neck.

References 1. Jacson FL, Goodman YV. Transfer of the facultatively anaerobic organism Bacteroides corrodens Eiken to a new genus, Eikenella. Int J System Bacteriol 1972;22:73e7. 2. Eiken M. Studies on an anaerobic, rod-shaped, gram-negative microorganism: Bacteroides corrodens sp. Acta Pathol Microbiol Scand 1958;43:404e16. 3. Tami TA, Parker GS. Eikenella corrodens. an emerging pathogen in head and neck infections. Arch Otolaryngol 1984;110:752e4. 4. Dorff GJ, Jackson LJ, Rytel MW. Infections with Eikenella corrodens: a newly recognized human pathogen. Ann Intern Med 1974;80:305e9.

5. Joshi N. Pleuropulmonary infections caused by Eikenella corrodens. Rev Infect Dis 1991;13:1207e12. 6. Dudly JP, Goldstein EJC, George WL, Bock BV, Kirby BD, Finegold SM. Sinus infection due to Eikenella corrodens. Arch Otolaryngol 1978;104:462e3. 7. DeMello FJ, Leonard MS. Eikenella corrodens: a new pathogen. Oral Surg 1979;47:401e4. 8. Jones JL, Romig RA. Eikenella corrodens: a pathogen in head and neck infections. Oral Surg Oral Med Oral Pathol 1979;48: 501e5. 9. Brooks GF, O’Donoghue JM, Rissing JP, Soapes K, Smith JW. Eikenella corrodens. a recently recognized pathogen: infections in medical-surgical patients and in association with methylphenidate abuse. Medicine 1974;53:325e42. 10. Young KA, Allsker RP, Hardie JM, Whiley RA. Interactions between Eikenella corrodens and ‘Streptococcus millerigroup’ organisms: possible mechanisms of pathogenicity in mixed infections. Antonie Van Leeuwenhoek 1996;69:371e4. 11. National Committee for Clinical Laboratory Standards. Methods for antimicrobial susceptibility testing of anaerobic bacteria, 3rd ed. Approved standard. National Committee for Clinical Laboratory Standards; 1993. 12. Forbes BA, Sahm DF, Weissfeld AS. Role of microscopy in the diagnosis of infectious diseases. In: Forbes BA, Sahm DF, Weissfeld AS, editors. Bailey & Scott’s Diagnostic Microbiology. Mosby; 2002. p. 119e35. 13. Monroe PW, Muchmore HG, Felton FG, Pirtle JK. Quantitation of microorganisms in sputum. Appl Microbiol 1969;18:214e20. 14. Wilson MJB, Martin DE. Quantitative sputum culture as a means of excluding false positive reports in the routine microbiology laboratory. J Clin Pathol 1972;25:697e700. 15. Appelbaum PC. Thyroid abscess associated with Eikenella corrodens in a 7-year-old child. Clin Pediatr (Phila) 1982;21: 241e2. 16. Vichyanond P. Eikenella corrodens as a cause of thyroid abscess. Am J Dis Child 1983;137:971e3. 17. Reynold MA, Hart CA, Harris F, Taitz LS. Anaerobes in acute otitis media. J Infect 1985;10:262e4. 18. Drake AF, Wolf GT, Fischer JJ. Eikenella corrodens as a cause of recurrent and persistent infections of the head and neck. Am J Otolaryngol 1986;7:426e30.

348 19. Tveteras K, Kristensen S, Bach V, Ravlo O. Eikenella corrodens: a recently recognized pathogen in head and neck infections. J Laryngol Otol 1987;101:592e4. 20. Queen JS, Clegg HW, Council JC, Morton D. Acute suppurative thyroiditis caused by Eikenella corrodens. J Pediatr Surg 1988; 23:359e61. 21. Cheng AF, Man DW, French GL. Thyroid abscess caused by Eikenella corrodens. J Infect 1988;16:181e5. 22. Zgheib A, el Allaf D, Demonty J, Rorive G. Intrathoracic infections with bacteraemia due to Eikenella corrodens as a complication of peritonsillar abscesses: report of a case and review of the literature. Acta Clin Belg 1992;47:124e8. 23. Hemady R, Zimmerman A, Katzen BW, Karesh JW. Orbital cellulitis caused by Eikenella corrodens. Am J Ophthalmol 1992;114:584e8. 24. Assefa D, Dalitz E, Handrich W, Leitz R, Braun W, Michalski H. Septic cavernous sinus thrombosis following infection of ethmoidal and maxillary sinuses: a case report. Int J Pediatr Otorhinolaryngol 1994;29:249e55. 25. Knudsen TD, Simko EJ. Eikenella corrodens: an unexpected pathogen causing a persistent peritonsillar abscess. Ear Nose Throat J 1995;74:114e7.

T. Udaka et al. 26. Heymann WR, Drezner D. Submandibular abscess caused by Eikenella corrodens. Cutis 1997;60:101e2. 27. Surow JB, Handler SD, Telian SA, Fleisher GR, Baranak CC. Bacteriology of tonsil surface and core in children. Laryngoscope 1989;99:261e6. 28. Blook I, Yocum P, Shah K. Surface vs core-tonsillar aerobic and anaerobic flora in recurrent tonsillitis. JAMA 1980;244:1696e8. 29. Kuhn JJ, Brook I, Waters CL, Church LW, Bianchi DA, Thompson DH. Quantitative bacteriology of tonsils removed from children with tonsillitis hypertrophy and recurrent tonsillitis with and without hypertrophy. Ann Otol Rhinol Laryngol 1995;104:646e52. 30. Han JK, Kerschner JE. Streptococcus milleri: an organism for head and neck infections and abscess. Arch Otolaryngol Head Neck Surg 2001;127:650e4. 31. Hoyler SL, Antony S. Eikenella corrodens: an usual cause of severe paraneumonic infection and empyema in immunocompetent patients. J Natl Med Assoc 2001;93:224e9. 32. Suzuki H, Ikeda K. Mode of action of long-term low-dose macrolide therapy for chronic sinusitis in the light of neutrophil recruitment. Curr Drug Targets Inflamm Allergy 2002;1: 117e26.