Simon Raiz

Bacteriologic and clinical efficacy of high dose amoxicillin for therapy of acute otitis media in children

Background. High dose (70 to 90 mg/kg/day) amoxicillin is recommended as first line therapy of acute otitis media (AOM) in geographic areas where drug-resistant Streptococcus pneumoniae is prevalent. Information on the bacteriologic efficacy of high dose amoxicillin treatment for AOM is limited. Objectives. To evaluate the bacteriologic and clinical efficacy of high dose amoxicillin as first line therapy in AOM. Methods. In a prospective study 50 culture-positive patients ages 3 to 22 months (median, 9 months; 77% <1 year) were treated with high dose amoxicillin (80 mg/kg/day three times a day for 10 days) No antibiotics were administered 72 h before enrollment. Twenty-four (48%) patients presented with their first episode of AOM. Middle ear fluid was cultured by tympanocentesis at enrollment and on Days 4 to 6 of therapy. Additional middle ear fluid cultures were obtained if clinical relapse occurred. Bacteriologic failure was defined by positive cultures on Days 4 to 6 and clinical failure by no change or worsening of AOM signs and symptoms and requirement for additional antibiotics during therapy and/or at end of therapy. Patients were followed until Day 28 ± 2. Susceptibility to penicillin and amoxicillin was measured by E-test. Results. Sixty-five organisms were recovered at enrollment:Haemophilus influenzae (38), Streptococcus pneumoniae (24), Streptococcus pyogenes (2) and Moraxella catarrhalis (1). Eighteen (75%) S. pneumoniae were nonsusceptible to penicillin (MIC > 0.1 &mgr;g/ml). All 24 S. pneumoniae isolates had amoxicillin MIC ≤ 2.0 &mgr;g/ml. Thirteen (34%) of the 38 H. influenzae were beta-lactamase producers. Eradication was achieved in 41 (82%) patients for 54 of 65 (83%) pathogens: 22 of 24 (92%) S. pneumoniae, 21 of 25 (84%) beta-lactamase-negative H. influenzae, 8 of 13 (62%) beta-lactamase-positive H. influenzae, 2 of 2 S. pyogenes and 1 of 1 M. catarrhalis. Seven organisms not initially present were isolated on Days 4 to 6 in 5 patients: 3 beta-lactamase-positive H. influenzae; 1 beta-lactamase-negative H. influenzae; 2 S. pneumoniae; and 1 M. catarrhalis. In total 14 of 50 (28%) patients failed bacteriologically on Days 4 to 6 (persistence + new infection), of whom 9 (64%) had beta-lactamase-positive H. influenzae. Three (33%) of the 9 patients with bacteriologic failure (2 beta-lactamase-positive H. influenzae, 1 S. pneumoniae) failed also clinically on Days 4 to 6. Conclusions. The predominant pathogens isolated from children with AOM failing high dose amoxicillin therapy were beta-lactamase-producing organisms. Because its overall clinical efficacy is good, high dose amoxicillin is still an appropriate choice as first line empiric therapy for AOM, followed by a beta-lactamase-stable drug in the event of failure.

Resistance pattern of middle ear fluid isolates in acute otitis media recently treated with antibiotics.

BACKGROUND Little information is available about the effect of antibiotic treatment on the prevalence and MIC of the subsequently isolated pathogens in cases of acute otitis media (AOM) failing a course of antibiotic therapy. This information is important, particularly regarding the effectiveness of the oral antibiotics used in children failing initial therapy. PATIENTS AND METHODS One hundred eighty-one children with culture-positive AOM were prospectively studied between October, 1995, and July, 1996. Sixty-three (35%) patients received various antibiotics for variable periods during the 14 days preceding enrollment. RESULTS A total of 94 Streptococcus pneumoniae (Pnc) and 113 Haemophilus influenzae (Hi) were isolated. Thirty-eight Pnc and 35 Hi were isolated in the 63 patients with recently treated AOM. Pnc as a single isolate was more prevalent in patients recently treated with antibiotics (27 of 63, 43%) than among those not recently treated (32 of 118, 27%, P = 0.047). The MIC50 values of penicillin, cefaclor and cefuroxime axetil for Pnc were significantly higher in the pneumococci isolated from patients recently treated than among those isolated from patients not recently treated with antibiotics (0.38, 3 and 0.75 microg/ml vs. 0.094, 0.38 and 0.12 microg/ml, respectively). Seventy-nine percent of Pnc isolates in the recently treated group had MIC for penicillin of >0.1 microg/ml vs. only 47% in those not recently treated (P < 0.05). The respective figures for MIC >0.5 microg/ml of cefaclor were 79% vs. 41% for the recently treated and not recently treated groups (P < 0.001); cefuroxime MIC >0.5 microg/ml was found in 61 and 25%, respectively (P = 0.001). CONCLUSIONS Pneumococcus is more prevalent in AOM after a recent antibiotic treatment, and the MIC of the commonly used beta-lactam drugs for Pnc is considerably higher in this setting. In view of our data, the use of oral cephalosporins like cefaclor or cefuroxime as second line drugs in the treatment of unresponsive AOM, particularly in regions where resistant PNC is prevalent, should be reconsidered.

Near-Elimination of Otitis Media Caused by 13- Valent Pneumococcal Conjugate Vaccine (PCV) Serotypes in Southern Israel Shortly After Sequential Introduction of 7-Valent/13-Valent PCV

BACKGROUND Otitis media (OM) is common in early childhood. Streptococcus pneumoniae caused approximately 30%-60% of episodes before the pneumococcal conjugate vaccine (PCV) era. The 7-valent PCV (PCV7) was introduced to the Israeli National Immunization Plan in July 2009, and was gradually replaced by the 13-valent PCV (PCV13) starting in November 2010. We aimed at assessing the impact of PCV7/PCV13 sequential introduction on pneumococcal and overall OM necessitating middle ear fluid culture in children aged <2 years in southern Israel. METHODS This was a prospective, population-based, active surveillance. Our medical center is the only one in the region, enabling incidence calculation. All pneumococcal episodes submitted for culture between July 2004 and June 2013 were included. Three subperiods were defined: pre-PCV, PCV7, and PCV13. RESULTS Overall, 6122 OM episodes were recorded, and 1893 were pneumococcal. Compared with the pre-PCV period, OM caused by PCV7 plus serotype 6A and the 5 additional PCV13 serotypes (5VT : 1, 3, 5, 7F, 19A) decreased by 96% and 85%, respectively (incidence rate ratios [IRRs], 0.04 [95% confidence interval {CI}, .02-.08] and 0.15 [95% CI, .07-.30], respectively) in a 2-step pattern: In the PCV7 period, only OM caused by PCV7 + 6A serotypes was decreased; in the PCV13 period, 5VT OM rates decreased, along with an additional PCV7 + 6A OM reduction. A nonsignificant increase in non-PCV13 serotype OM was observed (IRR, 1.07 [95% CI, .72-1.58]). In total, 77% and 60% reductions of all-pneumococcal and all-cause OM incidences, respectively, were observed. CONCLUSIONS A substantial 2-step reduction of pneumococcal OM rates, with near-elimination of PCV13 disease, was observed shortly after PCV7/PCV13 introduction.

Acute mastoiditis in Southern Israel: a twelve year retrospective study (1990 through 2001)

Background. Acute mastoiditis is a serious complication of acute otitis media (AOM) and has been increasingly reported in the last decade. Objectives. To report the experience accumulated with acute mastoiditis at the Soroka University Medical Center, Beer-Sheva, Israel, in a period of increasing antimicrobial resistance with Streptococcus pneumoniae. Patients and methods. We reviewed the records of all children with acute mastoiditis hospitalized from 1990 through 2001. Acute mastoiditis was diagnosed when one or more of the physical signs of mastoiditis (swelling, erythema, tenderness of the retroauricular area and anteroinferior displacement of the auricle) were diagnosed in the presence of concomitant or recent (≤4 weeks) AOM. Results. One hundred sixteen episodes of acute mastoiditis occurred in 101 children age 2 months to 14 years (median, 25 months; 19% <1 year old). The average yearly incidence was 6.1 cases per 100 000 population <14 years old, with a significant increase in the number of cases during the study period. Acute mastoiditis was the first evidence of AOM in 10 (10%) patients. Fever >38°C and >15 000 WBC/mm3 were present in 67 and 43% of cases, respectively. Irritability, retroauricular swelling, redness and protrusion of the auricle occurred more commonly in patients <3 years old (79, 90, 84 and 76%vs. 28, 42, 45 and 30%, respectively, in patients ≥3 years old; P < 0.002). Computed tomography scans were performed in 54 of 116 (47%) cases and revealed bone destruction in 38 (70%). Periosteal abscess and lateral sinus vein thrombosis were diagnosed in 8 and 2 patients, respectively. Simple mastoidectomy was done in 32 of 116 (28%) cases, after no response to intravenous antibiotics was observed; ventilation tubes were inserted in 12 patients. Cultures were obtained at admission in 83 (72%) episodes (71 by tympanocentesis and 12 from ear discharges). Overall 43 pathogens were isolated: 34 at admission; 14 at surgery; and 5 at both occasions. The most commonly isolated organisms were S. pneumoniae (14 of 43, 33%; 4 penicillin-nonsusceptible), Streptococcus pyogenes (11 of 43, 26%), nontypable Haemophilus influenzae (6 of 43, 14%), Pseudomonas aeruginosa (5 of 43, 12%) and Escherichia coli (4 of 43, 9%). All S. pneumoniae organisms were isolated between 1996 and 2001. Conclusions. (1) The incidence of acute mastoiditis in children in Southern Israel is greater than that reported in the literature; (2) a significant increase in the number of cases was recorded during the study period; (3) the clinical picture of acute mastoiditis was more severe in infants and young children; (4) the pathogen distribution in acute mastoiditis differs from that of AOM with significantly higher rates of S. pyogenes and lower rates of nontypable H. influenzae recovery; (5) penicillin-nonsusceptible S. pneumoniae played only a minor role in the etiology of acute mastoiditis in Southern Israel.

Bacteriologic and clinical efficacy of one day vs. three day intramuscular ceftriaxone for treatment of nonresponsive acute otitis media in children.

BACKGROUND One dose of intramuscular ceftriaxone has been recently licensed in the United States for the treatment of acute otitis media. However, data regarding the bacteriologic and clinical efficacy of this regimen in the treatment of nonresponsive acute otitis media are incomplete. OBJECTIVES To determine the bacteriologic and clinical efficacy of a 1-day 50-mg/kg vs. a 3-day 50-mg/kg/day intramuscular ceftriaxone regimen in the treatment of nonresponsive acute otitis media in children. PATIENTS AND METHODS In an open, prospective study 109 patients ages 3 to 36 months with culture-proved, nonresponsive acute otitis media were randomized to receive 1 (n = 49) or 3 (n = 60) 50-mg/kg/day intramuscular ceftriaxone doses, respectively. Middle ear fluid was aspirated for culture by tympanocentesis on the day of enrollment (Day 1); a second tympanocentesis with middle ear fluid culture was performed on Days 4 to 5. Additional middle ear fluid cultures were obtained if clinical relapse occurred after completion of therapy. Bacteriologic failure was defined by positive cultures on Days 4 to 5. Patients were followed until Day 28 after completion of therapy. Susceptibility of the middle ear pathogens was measured by E-test. RESULTS Organisms recovered (n = 133) were Streptococcus pneumoniae (30 and 35 isolates for the 1-day and 3-day treatment group, respectively), Haemophilus influenzae (26 and 38, respectively) and Moraxella catarrhalis (n = 4). Of the 30 S. pneumoniae isolated from the 1-day group, 27 (90%) and 6 (20%) were nonsusceptible to penicillin and ceftriaxone, respectively; 9 of 27 (33%) were fully resistant to penicillin. Thirty-four (97%) and 6 (17%) of the 35 S. pneumoniae isolated from the 3-day group were nonsusceptible to penicillin and ceftriaxone, respectively; 16 of 34 (47%) were fully resistant to penicillin. Bacterial eradication of all H. influenzae and penicillin-susceptible S. pneumoniae was achieved in both treatment groups. Bacterial eradication of 14 of 27 (52%) and 33 of 34 (97%) penicillin-nonsusceptible S. pneumoniae was achieved in the 1-day and 3-day group, respectively. Seven (50%) of the 14 patients from the 2 groups who did not achieve bacterial eradication did not improve clinically on Days 4 to 5 and required additional ceftriaxone treatment. CONCLUSION The 3-day intramuscular ceftriaxone regimen was significantly superior to the 1-day intramuscular ceftriaxone regimen in the treatment of nonresponsive acute otitis media caused by penicillin-resistant S. pneumoniae.

Bacteriologic and clinical efficacy of trimethoprim-sulfamethoxazole for treatment of acute otitis media.

Background. Trimethoprim-sulfamethoxazole (T/S) has often been used as first and second line of treatment for acute otitis media (AOM). Because of the increasing resistance of Streptococcus pneumoniae and Haemophilus influenzae to T/S, we undertook the present study to investigate the bacteriologic and clinical efficacy of this drug in AOM. Methods. Fifty-four culture-positive evaluable patients ages 3 to 32 months with AOM were treated with T/S 4/20 mg/kg in two divided daily doses for 10 days. Middle ear fluid (MEF) was cultured at enrollment (Day 1) and on Days 4 and 5 after initiation of treatment. Additional MEF cultures were obtained if clinical relapse occurred. Clinical failure was determined when the symptoms and signs of AOM did not improve or recurred during therapy. Bacteriologic failure was defined by positive culture on Days 4 and 5, or negative on Days 4 and 5 but positive again before the end of treatment. Patients were followed until Day 28 ± 2. Results. A total of 67 organisms were isolated from MEF specimens of the 54 study patients:S. pneumoniae, 24;H. influenzae, 40; and Streptococcus pyogenes, 3. Fifteen (63%) of 24 S. pneumoniae were nonsusceptible to T/S (trimethoprim MIC, >0.5 &mgr;g/ml), of which 10 (67%) were highly resistant to T/S (trimethoprim MIC, ≥4.0 &mgr;g/ml). Twelve (30%) of 40 H. influenzae and all 3 S. pyogenes isolates were nonsusceptible to T/S (MIC ≥ 4.0 &mgr;g/ml). Bacteriologic eradication occurred in 9 of 9 (100%) and 27 of 27 (100%) T/S-susceptible S. pneumoniae and H. influenzae, respectively, vs. 4 of 15 (27%) and 6 of 12 (50%) T/S-nonsusceptible S. pneumoniae and H. influenzae, respectively (P < 0.001). The 3 patients with S. pyogenes failed bacteriologically. Nine new organisms, not initially isolated, emerged during treatment, 7 of which (77%) were resistant to T/S. Altogether bacteriologic failure (organisms not eradicated plus newly emerged) occurred in 29 (53%) of 54 patients. Clinical failures occurred in 8 (15%) of 54 patients, and in 7 of these 8 cases the clinical failures occurred in those with bacteriologic failures. Ten patients relapsed clinically after completion of treatment and in 8 of them tympanocentesis for MEF culture was performed. Six of these 8 cultures were positive, and the initial pathogen was isolated in 4 of 6 (67%). Conclusions. A high bacteriologic failure rate as well as a considerable clinical failure rate occurred among patients with AOM treated with T/S. We believe that T/S is no longer an appropriate empiric choice for the treatment of AOM in regions where high T/S resistance among respiratory pathogens is reported.

Bacteriologic efficacy of a three-day intramuscular ceftriaxone regimen in nonresponsive acute otitis media

OBJECTIVE To determine the bacteriologic efficacy of ceftriaxone in nonresponsive acute otitis media in children. METHODS In a prospective study 92 patients ages 3 to 36 months (median, 11 months) with culture-proved nonresponsive acute otitis media were studied from January, 1995, through August, 1997. The patients were treated with intramuscular ceftriaxone (50 mg/kg/l/day) for 3 days. Middle ear fluid was aspirated for culture by tympanocentesis on day of enrollment (Day 1); a second tap was performed on Days 4 to 10. Additional middle ear fluid cultures were obtained if clinical relapse occurred. Bacteriologic failure was defined by positive culture on Days 4 to 10. Patients were followed until Day 17+/-2. Susceptibility was measured by E test. RESULTS The main drugs administered before enrollment were amoxicillin (38%), amoxicillinclavulanate (25%) and cefaclor (20%). Organisms recovered (n=105) were: Haemophilus influenzae, 54; Streptococcus pneumoniae, 47; Moraxella catarrhalis, 2; and Streptococcus pyogenes, 2. Thirty-four (72%) of the 47 S. pneumoniae isolates were intermediately resistant to penicillin (MIC 0.1 to 1.0 microg/ml), but all were susceptible to ceftriaxone (MIC < 0.5 microg/ml). Bacteriologic eradication was achieved in 100 of 105 (95%) cases: 54 of 54 (10O%) H. influenzae, 43 of 47 (92%) S. pneumoniae, 1 of 2 (50%) M. catarrhalis and 2 of 2 (100%) S. pyogenes. Bacteriologic success (with no relapse) occurred in 13 of 13 (100%) penicillin-susceptible S. pneumoniae vs. 28 of 34 (82%) S. pneumoniae intermediately resistant to penicillin (4 cases of bacteriologic failure and 2 cases of relapse). CONCLUSION A 3-day intramuscular ceftriaxone regimen is efficacious for the treatment of nonresponsive acute otitis media. The optimal duration of treatment in cases of nonresponsive acute otitis media and whether ceftriaxone is efficacious for the treatment of nonresponsive otitis media caused by S. pneumoniae highly resistant to penicillin is yet to be determined.

Predictive value of pneumococcal nasopharyngeal cultures for the assessment of nonresponsive acute otitis media in children

Background. Nonresponsive acute otitis media (NR‐AOM) is reported in >10% of children with AOM treated with antibiotics. Drug‐resistant Streptococcus pneumoniae is currently considered the leading cause of antibiotic failures in AOM. Nasopharyngeal colonization with S. pneumoniae was found to increase significantly during episodes of AOM. Objectives. To investigate the nasopharyngeal colonization with S. pneumoniae during NR‐AOM and compare it with that found in AOM not recently treated with antibiotics (NT‐AOM); to assess the predictive value of nasopharyngeal pneumococcal cultures results for the bacteriologic assessment of NR‐AOM. Materials and methods. Patients age 3 to 48 months with NT‐AOM and NR‐AOM were prospectively studied. Simultaneous nasopharyngeal cultures for S. pneumoniae and middle ear fluid cultures were obtained at enrollment. Antibiotic susceptibility testing was performed in all S. pneumoniae isolates. Penicillin and ceftriaxone MICs for S. pneumoniae were determined by E‐test. The sensitivity, specificity and positive and negative predictive values of positive or negative nasopharyngeal cultures for the presence of S. pneumoniae in middle ear fluid were calculated. Results. We studied 362 and 217 children with NT‐AOM and NR‐AOM, respectively. Of the children with NT‐AOM and NR‐AOM, 95 and 97%, respectively, were younger than 2 years of age. S. pneumoniae was isolated in the nasopharynx of 66 and 58% of children with NT‐AOM and NR‐AOM, respectively. Penicillin‐nonsusceptible S. pneumoniae was isolated more frequently from the nasopharynx of patients with NR‐AOM than from those with NT‐AOM (84%vs. 47%;P < 0.01). Antibiotic susceptibility patterns were similar for S. pneumoniae isolates recovered from the nasopharynx and those from the middle ear fluid in both NT‐AOM and NR‐AOM. A positive nasopharyngeal culture had only little predictive value for the presence of S. pneumoniae in middle ear fluid (41 and 51% for NT‐AOM and NR‐AOM, respectively). However, the negative predictive value of nasopharyngeal cultures for recovery of S. pneumoniae in NR‐AOM was high and significantly higher in NR‐AOM than in NT‐AOM (91%vs. 78%, respectively;P = 0.009). The negative predictive value of nasopharyngeal cultures for recovery of antibiotic‐resistant S. pneumoniae was 95 and 93% in NT‐AOM and NR‐AOM, respectively. Conclusions. A significantly higher nasopharyngeal colonization rate with antibiotic‐resistant S. pneumoniae was found in patients with NR‐AOM than in those with NT‐AOM. Negative nasopharyngeal culture for antibiotic‐resistant S. pneumoniae practically rules out its presence in middle ear fluid.

Can acute otitis media caused by Haemophilus influenzae be distinguished from that caused by Streptococcus pneumoniae

Background. Previous limited data suggest that acute otitis media (AOM) caused by Streptococcus pneumoniae can present as a more severe disease than that caused by Haemophilus influenzae or Moraxella catarrhalis, as expressed by both tympanic membrane and systemic findings. Objectives. To evaluate the severity of disease and impact of various pathogens, age, disease history and previous antibiotic therapy in children with AOM by using a comprehensive clinical/otologic score. Patients and methods. The study group consisted of 372 children ages 3 to 36 months with AOM seen at the pediatric emergency room during 1996 through 2001. All patients had tympanocentesis and middle ear fluid culture performed at enrollment. Clinical status was determined by a clinical/otologic score evaluating severity (0 = absent to 3 = severe) of tympanic membrane findings (redness and bulging) and patient’s fever, irritability and ear tugging. Maximal severity score was 15. Results. There were 138 (37%) H. influenzae, 76 (21%) S. pneumoniae, 64 (17%) mixed infections (H. influenzae +S. pneumoniae) and 94 (25%) culture-negative cases. The overall clinical/otologic score was higher in culture-positive than in culture-negative patients (9.27 ± 2.75 vs. 8.38 ± 3.08, P = 0.01). When analyzed by age groups, this difference was significant only for the youngest age group (3 to 6 months, P = 0.05). The severity scores for AOM caused by H. influenzae and S. pneumoniae were significantly higher than in the culture-negative AOM when tympanic membrane redness and bulging were analyzed separately. No differences were recorded in clinical/otologic scores between different pathogens (9.49 ± 2.86, 9.03 ± 2.72 and 9.09 ± 2.54 for H. influenzae, S. pneumoniae and H. influenzae +S. pneumoniae, respectively). The mean clinical/otologic score was higher in culture-positive than in culture-negative patients without relationship to previous antibiotic treatment or number of previous AOM episodes. Conclusions. (1) The clinical/otologic score of culture-positive young infants was higher than that of culture-negative infants; (2) the severity of tympanic membrane redness and bulging were the most indicative factors discriminating between a bacterial and nonbacterial etiology of AOM; and (3) the use of a clinical/otologic score could not discriminate among various bacterial etiologies of AOM.

Bacteriologic and clinical efficacy of oral gatifloxacin for the treatment of recurrent/nonresponsive acute otitis media: an open label, noncomparative, double tympanocentesis study

Background. Gatifloxacin is an 8-methoxyfluoroquinolone with good activity against respiratory pathogens. Objectives. To document the bacteriologic and clinical efficacy of gatifloxacin in recurrent/nonresponsive acute otitis media (AOM). Methods. One hundred sixty patients 6 to 48 months of age with recurrent/nonresponsive AOM received gatifloxacin suspension (10 mg/kg once daily for 10 days). Recurrent AOM was defined as ≥3 AOM episodes during the previous 6 months or ≥4 AOM episodes during the previous 12 months. Nonresponsive AOM was defined as AOM occurring ≤14 days after completing antibiotic treatment or not improving after ≥48 h of therapy. Middle ear fluid (MEF) obtained by tympanocentesis pretreatment (Day 1) and 3 to 5 days after initiation of treatment (Days 4 to 6) was cultured. Additional MEF cultures were obtained if clinical failure or recurrence of AOM occurred. Bacteriologic failure was defined by culture-positive MEF during treatment. Patients were followed until Days 22 to 28. Susceptibility was determined by broth microdilution. Results. One hundred twenty-eight (80%) patients completed treatment, and 32 discontinued the study prematurely (adverse events, 17; lost to follow-up, 10; consent withdrawal, 3; and laboratory abnormalities, 2). From 89 patients (median age, 1 year; median number of prior AOM episodes, 4; range, 0 to 12), 121 pathogens were recovered:Haemophilus influenzae, 74 (61%); Streptococcus pneumoniae, 36 (30%); Moraxella catarrhalis, 9 (7%); and Streptococcus pyogenes, 2 (2%). The 36 S. pneumoniae isolates were susceptible to gatifloxacin (MIC50 0.25 &mgr;g/ml); 26 of 36 (72%) were penicillin-nonsusceptible (15 fully resistant). All 74 H. influenzae isolates were susceptible to gatifloxacin (MIC ≤ 0.03 mg/ml). Fourteen of 74 (19%) and 9 of 9 (100%) H. influenzae and M. catarrhalis isolates, respectively, produced beta-lactamase. Bacteriologic eradication was achieved for 118 of 121 (98%) pathogens: 74 of 74 H. influenzae; 34 of 36 (94%) S. pneumoniae; 9 of 9 M. catarrhalis; and 1 of 2 S. pyogenes. Clinical improvement/cure at end of treatment was seen in 103 of 114 (90%) clinically evaluable patients. Clinical recurrence of AOM after completion of therapy occurred in 31 patients. Of the 27 recurrent AOM cases in which tympanocentesis was performed, there were 16 (59%) new infections, 4 (15%) culture-negative results and only 7 (26%) true bacteriologic relapses. Adverse events were recorded in 21 of 160 (13%) patients: vomiting, 16; diarrhea, 3; maculopapular rash, 2. No articular adverse events were recorded. Conclusion. Gatifloxacin is efficacious and safe for the treatment of recurrent/nonresponsive AOM.