Recurrent acute otitis media (AOM) is encountered in a subpopulation of 5 to 30% of all children with AOM. 1–3 These otitis media-prone children may experience four or more episodes of AOM and spend ≥7 months with middle ear effusion after AOM in the first year of life alone. Although a familial predisposition occurs in recurrent AOM, epidemiologic, environmental and immunologic risk factors have also been properly identified and include early male gender, early occurrence of infection, parental smoking, low socioeconomic status, day-care center attendance and immune abnormalities resulting from early or frequent exposure to middle ear pathogens. 1, 3–6
Whereas the microbiology of AOM is well-known, the bacteriologic correlates of recurrent AOM have been much less investigated. In the few studies published on this topic, the patients were followed for various periods of time (1 to 9 months) from the original AOM episode, the number of diagnostic tympanocenteses performed when recurrence developed was small and some children received antibiotic prophylaxis during the follow-up period. 7–11 In addition, although a higher prevalence of antibiotic-resistant Streptococcus pneumoniae and beta-lactamase-producing Haemophilus influenzae should be anticipated among AOM patients recently treated with antibiotics compared with patients not recently treated, 12–15 a systematic analysis of the bacteriologic spectrum of recurrent AOM during the past decade of increasing antibiotic resistance has not yet been performed.
The purpose of this study was to determine the etiology, the temporal occurrence of AOM during the follow-up period and the proportion of true bacteriologic relapses vs. new infections in children with clinical recurrence of AOM occurring within 3 to 4 weeks after completion of antibiotic therapy.
PATIENTS AND METHODS
Population and procedures
All children with clinical recurrence of AOM occurring within 1 month from successful completion of therapy for the initial AOM episode, who could be studied bacteriologically at recurrence, constituted the study population.
The initial baseline population included all the patients ages 3 to 36 months with AOM presenting at the Pediatric Emergency Room of the Soroka University Medical Center from January 1, 1995, through December 31, 1999, who were enrolled in various double tympanocentesis antibiotic efficacy studies performed by the Pediatric Infectious Disease Unit 16–25 and followed for at least 3 weeks after completion of therapy.
Acute otitis media was diagnosed if: (1) patients had symptoms and signs consistent with AOM (fever, irritability or tugging of the ears together with redness, bulging and blurring of tympanic membrane anatomic landmarks); (2) they had an acute illness of <7 days duration; (3) no spontaneous tympanic membrane perforation of >24 h was present; and (4) no tympanostomy tubes were present. An informed consent was signed by the parents at enrollment.
Tympanocentesis was performed at enrollment by an otolaryngologist, as described elsewhere. 16 After tympanocentesis the patients received one of the following antibiotic drugs for 1 to 10 days: azithromycin (10 mg/kg for 3 days or 10 mg/kg/day for the first day and 5 mg/kg/day for 4 additional days); ceftriaxone 50 mg/kg once daily (for 1 or 3 days); amoxicillin (50 or 80 mg/k/day for 10 days); cefaclor (40 mg/kg/day for 10 days); sanfetrinem (20 mg/kg twice daily for 10 days); trimethoprim-sulfamethoxazole (8 mg/kg/day of trimethoprim for 10 days) or amoxicillin/clavulanate (45/6.4 mg/kg/day or 90/6.4 mg/kg/day twice daily) for 10 days. 16–25 The patients initially treated with one or three doses of ceftriaxone or sanfetrinem presented with nonresponsive/recurrent AOM. 18, 21, 23
A second tympanocentesis was performed on Days 4 to 6 (72 to 120 h after initiation of therapy) in patients with positive initial middle ear fluid (MEF) culture. The patients with culture-negative MEF on Day 1 were followed without an additional tympanocentesis on Days 4 to 6. Tympanocentesis was performed in children who developed clinical recurrence at any time during the follow-up (before or after completion of therapy).
Bacteriologic eradication was defined by negative MEF culture on Days 4 to 6 if the original culture was positive.
Successful completion of therapy for the initial AOM episode was defined as bacteriologic eradication on Days 4 to 6 of therapy or no pathogen on MEF culture on Day 1 and clinical improvement at end of therapy.
Clinical recurrence was defined as a new AOM episode separated from the previous one by a period of at least 48 h of clinical resolution (defined as the presence of a noninflamed eardrum or absence of clinical symptomatology) after bacteriologic eradication on Days 4 to 6 of the initial AOM episode.
True bacteriologic relapse was defined as the presence in the MEF of an organism identical with that isolated before therapy as confirmed by serotype and by pulsed field gel electrophoresis (PFGE) for S. pneumoniae or by PFGE and beta-lactamase production for H. influenzae. When a mixed initial infection was followed by a recurrence caused by a single pathogen identical with one of the initial pathogens, the case was considered true bacteriologic relapse.
New infection was defined as any clinical recurrent AOM caused by an organism different than that isolated form the MEF at the initial AOM episode.
Patients were followed for a period of 3 to 4 weeks since completion of therapy (visits performed on Days 4 to 6, 11 to 14 and 24 to 30 since initiation of therapy). Patients treated with antibiotics for 10 days were followed for 3 weeks posttreatment only. Additional MEF cultures were obtained if clinical recurrence occurred. Bacterial isolates of the initial infection and the clinical recurrence were compared to determine whether the organisms were the same (true bacteriologic relapse) or newly acquired (new infection). The distribution of the clinical recurrent AOM cases during the follow-up (F/U) period was analyzed in two ways: (1) by relating the temporal occurrence of the recurrence to the real timing of completion of therapy (1, 3 or 5 days for azithromycin or ceftriaxone and 10 days for the other antibiotics administered); and (2) by considering Day 10 of antibiotic therapy as the uniform end-of-therapy day for all the various groups of patients receiving antibiotics for various periods of time.
MEF aspirates were sent to the Clinical Microbiology Laboratory of the Soroka University Medical Center and processed within 16 h, as described elsewhere. 16 Swabs of MEF aspirates were plated on Trypticase agar media containing 5% sheep blood and chocolate agar. The plates were incubated aerobically at 35°C for 48 h. Presumptive identification of S. pneumoniae was based on presence of alpha-hemolysis and inhibition of optochin by a positive slide agglutination test (Phadebact; Pharmacia Diagnostics, Uppsala, Sweden). The S. pneumoniae isolates were typed by the quellung reaction according to established procedures. 26
Identification of H. influenzae was based on Gram stain, growth on chocolate agar medium, failure to grow on Trypticase agar with added sheep blood and nutritional requirement of both hemin and nicotinamide adenine dinucleotide. Organisms that failed to agglutinate with polyvalent antisera to H. influenzae groups a, c-f and b (Phadebact) were considered untypable.
PFGE was performed on paired (initial and recurrent) bacterial isolates in which the pathogen recovered during clinical recurrence was phenotypically identical (identical serotype for S. pneumoniae and identical beta-lactamase production pattern for H. influenzae) with the original isolate recovered at the initial AOM episode.
Chromosomal DNA fragments were generated by Sma I digestion and prepared and analyzed as described previously. 27 A CHEF-DRIII apparatus (Bio-Rad Laboratories, Richmond, CA) was used for running the gels. Running conditions were 23 h at 11.3°C at 200 V ramped with initial forward time of 5 s and final forward time of 35 s. Gels were stained with ethidium bromide and photographed. Interpretation of strains relatedness on the basis of PFGE pattern was performed according to current consensus. 28
Significance of differences in proportions between true bacteriologic relapses and new infections during the study period was calculated using the chi square test for linear trend in proportions. P ≤ 0.05 was considered significant.
From a total of 1077 children with AOM initially enrolled during the study period and followed for 3 to 4 weeks after completion of the antibiotic therapy, 834 (77%) successfully completed the antibiotic regimen (Fig. 1). Documented clinical recurrence of AOM during the follow-up period occurred in 166 (20%) children; of these, 108 (13%) underwent tympanocentesis and MEF culture at the time of recurrence. In all other 58 cases, the patients were examined by their family physicians or otolaryngologists who made the diagnosis of AOM and decided on further treatment without performing a tympanocentesis.
The bacteriology of clinical recurrent AOM compared with the initial AOM episode is presented in Figure 2. A total of 111 pathogens were recovered at the initial AOM episode from 95 patients: 57 (51%) H. influenzae; 52 (47%) S. pneumoniae; 1 (1%) Moraxella catarrhalis; and 1 (1%) Streptococcus pyogenes. One hundred pathogens were recovered from 88 patients at clinical recurrence of AOM: 54 H. influenzae; 45 S. pneumoniae; and 1 M. catarrhalis.
Most episodes occurred during the first 2 weeks of follow-up; 39 (36%), 38 (35%), 21 (20%) and 10 (9%) early clinical recurrent AOM episodes occurred at Days 1 to 7, 8 to 14, 15 to 21 and 22 to 28 after completion of therapy, respectively; 71% of all early clinical recurrent AOM episodes occurred during the first 2 weeks after completion of therapy. A total of 37, 33, 21 and 9 pathogens were recovered at Days 1 to 7, 8 to 14, 15 to 21 and 22 to 28 after completion of therapy, respectively. No culture-positive clinical recurrent AOM episodes were recorded during the F/U week after completion of the shorter (1, 3 or 5 days) antibiotic therapies with azithromycin or intramuscular ceftriaxone. When Day 10 after initiation of therapy was considered uniformly as end of therapy for all the antibiotics used, the proportions for early clinical recurrent AOM episodes were 53 (49%), 32 (30%) and 23 (21%) for Days 1 to 7, 8 to 14 and 15 to 21, respectively, after Day 10. A total of 45, 33 and 22 pathogens were recovered at Days 1 to 7, 8 to 14 and 15 to 21, respectively, after Day 10.
Overall 70 pathogens (42 H. influenzae, 27 S. pneumoniae and 1 M. catarrhalis) were recovered during Days 1 to 14 after completion of therapy and 30 (18 S. pneumoniae and 12 H. influenzae) during Days 15 to 28 after completion of therapy. Pathogen recovery according to the timing of clinical recurrent AOM is shown in Figure 3. There were no statistically significant differences in the relative proportions between various pathogens or in the representation of MEF samples with bacterial growth compared with those without growth, when the four periods were compared.
PFGE was performed on 38 paired (initial and recurrent) bacterial isolates (23 H. influenzae and 15 S. pneumoniae) recovered at clinical recurrence of AOM, in which the pathogen recovered during clinical recurrence was phenotypically identical with the original isolate recovered at the initial AOM episode. The analysis revealed that 5 of 23 (22%) H. influenzae isolates looking phenotypically identical with the original ones were in fact genotypically different and represented a new infection. All S. pneumoniae isolates exhibiting the same serotype were also identical by PFGE.
Of all 108 clinically recurrent AOM episodes, 30 (28%) represented a true bacteriologic relapse: 13 (12%) with S. pneumoniae; 12 (11%) with H. influenzae; and 5 (5%) with both (Fig. 4).
We examined whether the population of true bacteriologic relapses differed as function of the timing of the clinical recurrence after completion of treatment (Fig. 5). The absolute numbers and proportions of episodes of true bacteriologic relapse from all cases of clinical recurrent AOM decreased as time elapsed since completion of treatment: 16 of 39 (41%), 10 of 38 (26%), 3 of 21 (14%) and 1 of 10 (10%) of all episodes of clinical recurrent AOM occurring during Days 1 to 7, 8 to 14, 15 to 21 and 22 to 28 after completion of treatment, respectively, were true bacteriologic relapses (P = 0.01). When the number of pathogens causing true bacteriologic relapse was analyzed, they represented 19 of 37 (51%), 12 of 33 (36%); 3 of 21 (14%) and 1 of 9 (11%) of all pathogens recovered during Days 1 to 7, 8 to 14, 15 to 21 and 22 to 28 after completion of treatment, respectively (P = 0.001). The respective rates for S. pneumoniae were 11 of 17 (65%), 3 of 10 (30%), 3 of 13 (23%) and 1 of 5 (20%) (P= 0.02) and for H. influenzae 8 of 19 (42%), 9 of 23 (39%), 0 of 8 (%) and 0 of 4 (0%) (P = 0.02). Whereas no significant differences were recorded in the proportion of true bacteriologic relapses caused by S. pneumoniae with time, all the true bacteriologic relapses with H. influenzae occurred during the first 14 days. When Day 10 after initiation of therapy was considered uniformly as end of therapy for all the antibiotics used, 20 of 45 (44%), 11 of 33 (33%) and 4 of 12 (33%) of the pathogens isolated on Days 1 to 7, 8 to 14 and 15 to 21 after completion of therapy were identical with the pathogens isolated at the initial AOM episode (P value not significant). The respective figures for S. pneumoniae were 12 of 19 (63%), 2 of 13 (15%) and 4 of 13 (31%) (P value not significant) and for H. influenzae 10 of 25 (40%), 7 of 20 (35%) and 0 of 9 (0%) (P = 0.05).
The dynamics of new acquisition/persistence of pathogens in cases of early clinical recurrent AOM in relation to the initial AOM episode are shown in Figure 6. S. pneumoniae was found at clinical recurrence only in 20 of 36 (54%) cases that grew S. pneumoniae alone at the initial AOM episode; 16 of 20 isolates were identical with the initial S. pneumoniae isolate. In the other 16 cases different pathogens or no growth were reported as replacing the original S. pneumoniae isolates. H. influenzae was found at clinical recurrence of AOM in 24 of 42 (57%) cases who grew H. influenzae alone at the initial AOM episode; 12 of 24 (50%) isolates were true bacteriologic relapse. The other 18 H. influenzae isolates were replaced by S. pneumoniae or sterile cultures. In the 15 cases when S. pneumoniae was isolated together with H. influenzae at the initial AOM episode, the replacement at recurrence was by 4 S. pneumoniae (1 true relapse), 10 H. influenzae (6 true relapses) and 4 sterile cultures. Thus, even when a case of pneumococcal AOM (as a single pathogen or as a mixed infection with another pathogen) had a relapse with S. pneumoniae, in 7 of 24 (29%) cases it was a new infection. Similarly of H. influenzae cases that had clinical recurrence with a H. influenzae, 16 of 34 (47%) were a new infection.
Our study is not the first to suggest that a substantial proportion of the clinical recurrences of AOM are in fact new infections. Halsted et al. 7 performed during 1964 to 1966 diagnostic tympanocenteses in 8 of 40 children with recurrence of AOM followed for a period of 9 months since the initial AOM episode and reported that at least 6 of cases of recurrent AOM were caused by a new pathogen. Liston et al. 8 analyzed in 1984 the bacteriology of recurrent AOM and the effect of sulfisoxazole prophylaxis in 31 children and found that H. influenzae predominated during prophylaxis whereas S. pneumoniae predominated in non-antibiotic-treated patients. In 17 of 28 isolates obtained from 8 children with consecutive recurrent AOM episodes, the infecting organisms belonged to the same species; however, 7 of them differed in serotype or biotype. Furthermore the average number of weeks between onset of recurrence in children with homologous strains was shorter (2.6 weeks) than in children with recurrent AOM caused by heterologous strains (5.7 weeks). 8
Barenkamp et al. 9 studied 30 children with recurrent AOM caused by H. influenzae by using the technique of outer membrane protein gel analysis and biotyping and reported that early recurrences were usually a result of relapse with the initial infecting organism whereas later recurrences were mostly caused by a new organism. Del Beccaro et al. 10 studied 13 of 17 children with recurrent AOM followed for a period of 1 month and reported that in 8 of 13 recurrences were due to a new pathogen. Carlin et al. 11 followed 103 patients with AOM for a period of 1 month and found that 36 (35%) developed recurrent AOM during this time interval; 75% of the patients in whom pathogens were recovered by tympanocentesis initially and at recurrence had an infection with a new pathogen.
Of the five previous studies reviewed above, four 7–10 did not determine the eradication of the pathogen causing the initial AOM episode. The fifth study 11 mixed patients in whom eradication of the initial pathogen was achieved with those without proof of initial pathogen eradication. Furthermore additional second line antibiotic therapy was administered for the initially bacteriologic and clinical failure cases.
In contrast to the previous studies, our study analyzed only cases of early clinical recurrences of AOM after effective antibiotic treatment, resulting in bacteriologic eradication of the pathogens causing the initial AOM episode or cases with initial negative cultures, excluding therefore the possibility that the recurrence of AOM represented in fact a persistent infection. Furthermore a comprehensive identification of the pathogens by determining their antibiogram profile, serotype (for S. pneumoniae), beta-lactamase production (for H. influenzae) and PFGE pattern was performed to determine whether the pathogens recovered at recurrence represented a new infection or a true bacteriologic relapse compared with those isolated at the initial AOM episode.
We did not type in our study multiple colonies, and carriage of multiple bacterial strains was previously reported with both H. influenzae and S. pneumoniae.29 Therefore it is possible that AOM was caused, in some instances, by more than one strain and also, at least theoretically, in some cases a true relapse occurred with a strain present initially that was not typed. However, it is unlikely that this occurred frequently. At least with S. pneumoniae, in the great majority of cases sterilization of the MEF was associated with disappearance from the nasopharynx of the same strain. 30, 31 We demonstrated recently that even in the nasopharynx, where it is more likely that multiple colonization might occur, compared with MEF, typing 3 or even 5 S. pneumoniae colonies will not increase significantly the chance of finding multiple colonization. 29 More than 25 colonies per sample must be typed to reliably rule out multiple colonization or infection, which becomes impossible with the currently applied methods.
The present study could not determine the real proportion of clinical and bacteriologic AOM recurrences during the 3 to 4 weeks after completion of antibiotic therapy, because some cases in which it occurred might not have been reported to us and others were treated by the family physicians without performing a tympanocentesis. In addition the F/U period after completion of therapy for the initial AOM episode was not uniform for all patients, varying between 3 and 4 weeks. Furthermore our study included both patients enrolled with an initial simple uncomplicated AOM episode (some of them experiencing their first AOM episode at enrollment) and patients with nonresponsive or persistent/recurrent AOM who are prone to recurrent AOM. However, we were able to clearly demonstrate that most of the cases of early clinical recurrent AOM were new infections and did not represent true bacteriologic relapses.
The American Food and Drug Administration guidelines for the evaluation of the efficacy of various antibiotic drugs in the treatment of AOM require an “end-of-therapy” (1 to 4 days after completion of antibiotic therapy) and a “test-of-cure” (up to 30 days since initiation of therapy) visits as clinical outcome endpoints. 32–34 The natural history of AOM, particularly in otitis-prone children, includes frequent recurrences of the disease in >one-third of cases, regardless of prior effective antibiotic therapy. 35 Therefore the traditional “test-of-cure” evaluation, by incorporating the recurrent AOM case occurring during the follow-up period, represents a less accurate evaluation of the clinical success of a drug used in the treatment of AOM and cannot be objectively used when comparing among various antibiotics used in the treatment of this disease. On the contrary the use of the “end-of-therapy” visit as the main clinical endpoint of the efficacy of an antibiotic drug in the therapy of AOM may be more representative of its true clinical efficacy. Therefore we suggest that the “end-of-therapy” visit should be established as the standard evaluation of the clinical efficacy of antibacterial agents in children with AOM, replacing the inaccurate “test-of-cure” evaluation after 28 to 30 days.
Although the number of true bacteriologic relapses of AOM decreased progressively as time elapsed since the initial AOM episode, the representation of S. pneumoniae and H. influenzae among the true bacteriologic relapses and the new infection cases was different. We have established that whereas S. pneumoniae may be present at any time during the F/U period among the true bacteriologic relapses of AOM, all cases of recurrent AOM caused by H. influenzae and occurring after Day 14 were new infections.
In conclusion this study showed that most early clinical recurrent AOM episodes (occurring 3 to 4 weeks after successful completion of antibiotic therapy) are in fact infections caused by new organisms. In addition if true bacteriologic relapses occur, the majority of cases are recorded close (first 14 days) to the completion of therapy for the initial AOM episode. Finally H. influenzae may only rarely cause true bacteriologic relapses after 14 days of completion of antibiotic treatment.
1. Faden H, Duffy L, Boeve M. Otitis media: back to basics. Pediatr Infect Dis J 1998; 17: 1105–13.
2. Pichichero ME. Recurrent and persistent otitis media. Pediatr Infect Dis J 2000; 19: 911–16.
3. Faden H. The microbiologic and immunologic basis for recurrent otitis media in children. Eur J Pediatr 2001; 160: 407–13.
4. Casselbrant ML, Mandel EM, Fall PA, et al. The heritability of otitis media: a twin and triplet study. JAMA 1999; 282: 2125–30.
5. Stahlberg MR, Ruuskanen O, Virolainen E. Risk factors for recurrent otitis media. Pediatr Infect Dis J 1986; 5: 30–2.
6. Alho OP. How common is recurrent acute otitis media
? Acta Otolaryngol (Stockh) 1997; (Suppl 529): 8–10.
7. Halsted C, Lepow ML, Balassanian N, Emmerich J, Wolinsky E. Otitis media: clinical observations, microbiology, and evaluation of therapy. Am J Dis Child 1968; 115: 542–51.
8. Liston TE, Foshee WS, MccLeskey FK. The bacteriology of recurrent otitis media and the effect of sulfisoxazole chemoprophylaxis. Pediatr Infect Dis 1984; 3: 20–4.
9. Barenkamp SJ, Shurin PA, Marchand CD, et al. Do children with recurrent Haemophilus influenzae
otitis media become infected with a new organism or reacquire the original strain? J Pediatr 1984; 105: 533–7.
10. Del Beccaro MA, Mendelman PM, Inglis AF, et al. Bacteriology of acute otitis media: a new perspective. J Pediatr 1992; 120: 81–4.
11. Carlin SA, Marchant CD, Shurin PA, et al. Early recurrences of otitis media: reinfection or relapse? J Pediatr 1987; 110: 20–5.
12. Block SL. Strategies for dealing with amoxicillin failure in acute otitis media. Arch Fam Med 1999; 8: 68–87.
13. Pichichero ME, Pichichero CL. Persistent acute otitis media: I. Causative pathogens. Pediatr Infect Dis J 1995; 14: 178–83.
14. Faden H, Bernstein J, Brodsky L, et al. Effect of prior antibiotic treatment on middle ear disease in children. Ann Otol Laryngol 1992; 101: 87–91.
15. Block SL, Harrison CJ, Hedrick JA, et al. Penicillin-resistant Streptococcus pneumoniae
in acute otitis media: risk factors, susceptibility patterns and antimicrobial management. Pediatr Infect Dis J 1995; 14: 751–9.
16. Dagan R, Abramson O, Leibovitz E, et al. Impaired bacteriologic response to oral cephalosporins in acute otitis media caused by pneumococci with intermediate resistance to penicillin. Pediatr Infect Dis J 1996; 15: 980–5.
17. Dagan R, Abramson O, Leibovitz E, et al. Bacteriologic response to oral cephalosporins: are established susceptibility breakpoints appropriate in the case of acute otitis media? J Infect Dis 1997; 176: 1253–9.
18. Leibovitz E, Piglansky L, Scandola M, et al. Sanfetrinem pharmacokinetics (plasma, middle ear fluid) in pediatric patients with acute otitis media [Abstract F-204]. In: 37th Interscience Conference on Antimicrobial Agents and Chemotherapy, Toronto, September 28 to October 1, 1997. Washington, DC: American Society for Microbiology, 1997: 181.
19. Dagan R, Piglansky L, Yagupsky P, et al. Bacteriologic response to acute otitis media: comparison between azythromycin, cefaclor and amoxicillin [Abstract K-103]. In: 37th International Conference on Antimicrobial Agents and Chemotherapy, Toronto, September 28 to October 1, 1997. Washington, DC: American Society of Microbiology, 1997: 346.
20. Leibovitz E, Raiz S, Piglansly L, et al. Resistance pattern of middle ear fluid isolates in acute otitis media recently treated with antibiotics. Pediatr Infect Dis J 1998; 17: 643–9.
21. Leibovitz E, Piglansky L, Raiz S, et al. Bacteriologic efficacy of a three-day intramuscular ceftriaxone regimen in non-responsive acute otitis media. Pediatr Infect Dis J 1998; 17: 1126–31.
22. Dagan R, Leibovitz E, Fliss DM et al. Bacteriologic efficacies of oral azithromycin and oral cefaclor in treatment of acute otitis media in infants and young children. Antimicrob Agents Chemother 2000; 44: 43–50.
23. Leibovitz E, Piglansky L, Raiz S, et al. The bacteriologic efficacy of one day versus
three day intramuscular ceftriaxone in the treatment of nonresponsive acute otitis media. Pediatr Infect Dis J 2000; 19: 1040–5.
24. Dagan R, Johnson C, McLinn S, et al. Bacteriological and clinical efficacy of amoxicillin/clavulanate vs.
azithromycin in acute otitis media. Pediatr Infect Dis J 2000; 19: 95–104.
25. Leiberman A, Leibovitz E, Piglansky L, et al. The bacteriologic and clinical efficacy of trimethoprim-sulfamethoxazole in the treatment of acute otitis media in children. Pediatr Infect Dis J 2001; 20: 260–4.
26. Austrian R. The quellung reaction, a neglected microbiologic technique. Mt. Sinai J. Med 1976; 43: 699–709.
27. Greenberg D, Mahenthiralingam E, Speert DP, Scheifele DW. Genotyping of penicillin-resistant invasive Streptococcus pneumoniae
obtained from Canadian children early in the resistance epidemic [Abstract Sa-183]. In: Infectious Diseases Society of America Meeting, Denver, CO, November 12 to 14, 1998. Alexandria, VA: Infectious Diseases Society of America, 1998.
28. Tenover FC, Arbeit RD, Goering RV, et al. Interpreting chromosomal DNA restriction patterns produced by pulsed-field gel electrophoresis: criteria for bacterial strain typing. J Clin Microbiol 1995; 33: 2233–9.
29. Huebner RE, Dagan R, Porath N, Wasas AD, Klugman KP. Lack of utility of serotyping multiple colonies for detection of simultaneous nasopharyngeal carriage of different pneumococcal serotypes. Pediatr Infect Dis J 2000; 19: 1017–9.
30. Dagan R, Leibovitz E, Cheletz G, Lieberman A, Porat N. Antibiotic treatment in acute otitis media promotes super-infection with resistant Streptococcus pneumoniae
carried before initiation of treatment. J Infect Dis 2001; 183: 880–6.
31. Eldan M, Leibovitz E, Piglansly L, et al. Predictive value of pneumococcal nasopharyngeal cultures for the assessment of nonresponsive acute otitis media in children. Pediatr Infect Dis J 2000; 19: 298–303.
32. Bluestone CD. Current therapy for otitis media and criteria for evaluation of new antimicrobial agents. Clin Infect Dis J 1992; 14 (Suppl 2): S197–203.
33. Beam TR, Gilbert DN, Kunin CM. General guidelines for the evaluation of anti-infective drug products. Clin Infect Dis 1992; 15 (Suppl 1): S5–32.
34. Chow AW, Hall CB, Klein JO, et al. Evaluation of new anti-infective drugs for the treatment respiratory infections. Clin Infect Dis 1992; 15 (Suppl 1): S62–88.
35. Leibovitz E, Piglansky L, Raiz S, et al. Bacteriological efficacy of gatifloxacin in the treatment of recurrent/non-responsive acute otitis media [Abstract 1558a]. In: 41th Interscience Conference on Antimicrobial Agents and Chemotherapy, Chicago, December 16 to 19, 2001. Washington, DC: American Society for Microbiology, 2001.