ORIGINAL STUDIES

A Multicenter, Open Label, Double Tympanocentesis Study

of High Dose Cefdinir in Children With Acute Otitis Media
at High Risk of Persistent or Recurrent Infection
Adriano Arguedas, MD,* Ron Dagan, MD, Eugene Leibovitz, MD, Alejandro Hoberman, MD,
Michael Pichichero, MD, and Maria Paris, MD
Background: Given the relatively high prevalence of recurrent and
persistent acute otitis media (AOM) and the prominent etiologic role
of Streptococcus pneumoniae, especially penicillin-nonsusceptible
strains in children with these conditions, new alternative treatments
are desirable.
Methods: Children 6 months 4 years of age with AOM considered
to be at risk for recurrent or persistent infection received large
dosage cefdinir 25 mg/kg oral suspension once daily for 10 days.
Children were evaluated pretreatment (day 1), on therapy (days
4 6), end of therapy (days 1214) and at follow-up (days 2528).
All children had tympanocentesis at enrollment. In culture-positive
children, tympanocentesis was repeated after 35 days (days 4 6)
unless evidence of absence of middle ear effusion was documented.
Results: Of 447 children enrolled, 230 were clinically and bacteriologically evaluable (74% 2 years old or younger; 57% treated for
AOM in previous 3 months). Bacteriologic eradication, based on
repeat tympanocentesis on days 4 6, was achieved in 74% (170 of
230) of children; 76% (201 of 266) of AOM pathogens were
eradicated. Eradication of penicillin-susceptible, -intermediate and
-resistant S. pneumoniae was 91% (50 of 55), 67% (18 of 27) and
43% (10 of 23), respectively (P 0.001); eradication of H. influenzae was 72% (90 of 125). Overall clinical response at days 1214
was 83% (76 and 82% for children with S. pneumoniae and Haemophilus influenzae, respectively). Sustained clinical response at
days 2528 was 85%. Clinical response was 83% for culturepositive children versus 96% for culture-negative children at baseline tympanocentesis (P 0.001).
Conclusions: In this study of AOM among children at risk for
persistent or recurrent infection, large dose cefdinir resulted in an
overall successful clinical response at end of treatment of 83%. This
regimen was efficacious against penicillin-susceptible S. pneumoniae,
but effectiveness was markedly decreased against nonsusceptible
strains and was moderate for H. influenzae strains.

Accepted for publication October 28, 2005.

From *Instituto de Atencion Pediatrica, Neeman-ICIC, Hospital Nacional de
Ninos, San Jose, Costa Rica; the Pediatric Infectious Disease Unit, Soroka
University Medical Center, and the Faculty of Health Sciences, Ben Gurion
University of the Negev; Beer Sheva, Israel; the Childrens Hospital of
Pittsburgh, Pittsburgh, PA; the University of Rochester Medical Center,
Rochester, NY; and Abbott Laboratories, Abbott Park, IL.
Supported by a grant from Abbott Laboratories.
E-mail aarguedas@iped.net. Reprints not available.
Copyright 2006 by Lippincott Williams & Wilkins
ISSN: 0891-3668/06/2503-0211
DOI: 10.1097/01.inf.0000202138.12950.3c

Key Words: cefdinir, otitis media, recurrent, persistent

(Pediatr Infect Dis J 2006;25: 211218)

cute otitis media (AOM) is the most common bacterial

infection of childhood,1 with the majority (85%) of children experiencing their first episode by the time they are 12
months old.2 Results of prospective cohort studies3,4 and
national surveys5,6 suggest that 10 30% of children experience recurrent episodes of AOM during their first several
years of life or are therapeutic failures.7 This group of
children poses a challenge to clinicians because of the wellknown high risk of having an antimicrobial-resistant middle
ear fluid (MEF) pathogens.
The selection of an antibiotic for the treatment of
children with recurrent otitis media and otitis media therapeutic failure should be based on acceptable antimicrobial
activity against MEF pathogens and assurance that the selected agent achieves MEF concentrations above the minimal
inhibitory concentrations (MIC) of the target pathogens and,
in the case of -lactam agents, the antimicrobial concentration time above the MIC should be 40% of the dosing
interval time.8,9 Various microbiologic studies have shown
that children with recurrent or persistent AOM are at higher
risk of being infected with penicillin-nonsusceptible Streptococcus pneumoniae or -lactamase-producing pathogens
such as Haemophilus influenzae than children with uncomplicated AOM.8 14
Cefdinir is an extended spectrum cephalosporin with
in vitro activity against the major pathogens causing AOM,
including penicillin-susceptible (MIC90 0.12 g/mL) S.
pneumoniae, Streptococcus pyogenes (MIC90 0.06 g/mL),
-lactamase-producing strains of H. influenzae (MIC90 0.5
g/mL) and Moraxella catarrhalis (MIC90 0.25 g/mL).15,16
Based on the results of previous clinical trials, cefdinir (14
mg/kg/d) administered once or twice daily was approved for
the treatment of children with AOM.1722 Those trials were
conducted primarily in children with uncomplicated AOM,
with only a minority of children having a history of recurrent
infections and none of the evaluated children having refractory infections. Moreover, in the case of S. pneumoniae,
cefdinir at 14 mg/kg/d was approved by the Food and Drug
Administration only for the treatment of children with AOM
caused by penicillin-susceptible S. pneumoniae, because ev-

The Pediatric Infectious Disease Journal Volume 25, Number 3, March 2006

211

Arguedas et al

The Pediatric Infectious Disease Journal Volume 25, Number 3, March 2006

idence of clinical response in infections caused by penicillinnonsusceptible strains was lacking.

This double tympanocentesis study was therefore designed to evaluate the bacteriologic response during treatment, the clinical efficacy at end of treatment and the safety
of large dosage cefdinir (25 mg/kg/d) in children with AOM
at risk for recurrent or persistent infections, with special
emphasis on eradication rate of penicillin-nonsusceptible S.
pneumoniae isolates with this higher dose of cefdinir.

CHILDREN AND METHODS

Study Design
This study was a prospective, open label, multicenter
(17 sites in Latin America, the United States and Israel) study
that enrolled children with AOM between July 2002 and
April 2003. An Institutional Review Board or Ethics Committee at each participating center reviewed and approved the
protocol. Written informed consent was obtained from parents (or legal representatives) of each child before enrollment
into the study.

Children
The study population consisted of children between 6
months and 4 years of age with a clinical diagnosis of AOM
based on the presence of otalgia, ear fullness, decreased
hearing and/or purulent discharge caused by acute perforation
of the tympanic membrane, in association with at least 2 of
the following conditions: bulging tympanic membrane;
opacification of tympanic membrane; and/or purulent effusion behind the tympanic membrane corroborated by pneumatic otoscopy. The presence of MEF was established by
tympanometry or spectral gradient acoustic reflectometry
(SGAR) (Ear Check). To be eligible for study enrollment,
children must also have satisfied at least 2 of the following
criteria: 24 months of age or younger; antibiotic exposure for
AOM within the previous 3 months; day-care attendance or
siblings/household contacts 8 years of age or younger. Children who had either a ventilation tube or perforated tympanic
membrane for 24 hours before enrollment were excluded.

Antimicrobial Therapy
Children were treated in an open label manner with
cefdinir oral suspension (Omnicef; Abbott Laboratories) 25
mg/kg given once daily for 10 days.

Study Procedures
On day 1 (pretreatment), medical history, physical
examination with an otologic assessment and hematology,
chemistry and urine analyses were performed, and a sample
of MEF was obtained from the affected ear(s) by tympanocentesis. Direct culture was permitted for children with ruptured membrane and purulent discharge for 24 hours.
Clinical and otoscopic (pneumatic otoscopy and tympanometry or by SGAR) assessments were conducted at all
subsequent visits days 4 6 (on therapy), days 1214 (end of
treatment) and day 2528 (end of study, follow-up). Hematology, chemistry and urine testing was repeated at the endof-treatment visit and as needed thereafter if clinically significant abnormal results were detected at end of treatment.

212

Tympanocentesis was repeated at the on-therapy visit

in children who had AOM pathogens at baseline and persistent MEF confirmed by tympanometry or SGAR. An additional tympanocentesis was recommended if a child was
categorized as a clinical failure or relapse. At Childrens
Hospital of Pittsburgh, a repeat tympanocentesis at the ontherapy visit was performed only in those children with a
baseline nonsusceptible S. pneumoniae or H. influenzae and
with the following findings during the on-therapy visit: presence of MEF and otalgia plus mild bulging of the tympanic
membrane or moderate to severe bulging without otalgia.
Investigators evaluated children for adverse events at all
visits and categorized the adverse events by severity as mild,
moderate or severe and by potential relationship to study
drug.

Microbiology
Culture and susceptibility testing from MEF samples
were initially performed at local laboratories. Identified isolates were sent to a central laboratory (Covance Central
Laboratory, Indianapolis, IN) for confirmation of identification and in vitro antibiotic susceptibility testing. MICs were
determined by broth microdilution according to the guidelines of the Clinical and Laboratory Standards Institute.23,24
S. pneumoniae, H. influenzae, M. catarrhalis and S. pyogenes
were considered target pathogens, and in the case of H.
influenzae and M. catarrhalis strains, -lactamase production
was tested by the cefinase test. For S. pneumoniae, isolates
were considered susceptible to penicillin if MIC values were
0.06 mg/L; intermediate if penicillin MIC values were between 0.125 and 1.0 mg/L and resistant if MIC values were
2.0 mg/L.24 S. pneumoniae isolates were considered susceptible to cefdinir if the MIC value was 0.50 mg/L, intermediate
if the cefdinir MIC value was 1 mg/L and resistant if MIC value
was 2.0 mg/L.25 For H. influenzae, isolates with an MIC to
cefdinir of 1 g/mL were considered susceptible. For organisms other than H. influenzae and Streptococcus spp.,
isolates with an MIC to cefdinir of 1 g/mL were considered susceptible, isolates with an MIC of 2 g/mL were
considered intermediate and isolates with an MIC of 4
g/mL were considered resistant.

Outcome Measures
Bacteriologic response was assessed at the on-therapy
visit in bacteriologically evaluable and clinically/bacteriologically evaluable populations. Children were considered bacteriologically evaluable if at least 1 target pathogen (H. influenzae, S. pyogenes, S. pneumoniae or M. catarrhalis) was
isolated at baseline and a follow-up tympanocentesis or culture
of draining pus was performed at the on-therapy visit or if no
MEF could be aspirated at that time. To be considered clinically
and bacteriologically evaluable, children must have had (1)
completed at least 80% of their prescribed study drug or at least
3 days of study medication if the child was categorized as a
clinical failure, (2) a target pathogen must have been isolated
at baseline and (3) a clinical assessment at the end-oftreatment visit. Concurrent use of a systemic antimicrobial
agent was not allowed.
2006 Lippincott Williams & Wilkins

The Pediatric Infectious Disease Journal Volume 25, Number 3, March 2006

The primary efficacy end point was bacteriologic eradication rate at the on-therapy visit in the bacteriologically
evaluable population. Bacteriologic eradication rate was defined as the percentage of children with at least 1 target
pathogen isolated from the baseline culture that had a bacteriologic response of documented or presumed eradication for
the pretreatment pathogen(s) at the on-therapy visit. A child
was classified as presumed eradication if no material was
obtained (dry tap) at the time of the on-therapy tympanocentesis or if a second tympanocentesis was not performed
because of the absence of MEF.
Secondary endpoints included clinical response rate at
end of treatment and end of study among clinically and
bacteriologically evaluable children. Clinical response was
classified by the investigator as cure or failure at end of
treatment. Clinical cure was defined as complete (cure) or
partial (improvement) resolution of AOM signs and symptoms without the need for additional antimicrobials. A child
was considered a clinical failure if signs and symptoms of
AOM persisted after at least 72 hours of study drug and
additional antimicrobial therapy for AOM was required. Sustained cure and recurrence of AOM (child previously evaluated as cured or improved at end of treatment visit who
satisfied the criteria for failure between end of treatment and
end of study) were assessed at end of study.

Safety
Children who received at least 1 dose of study medication were included in the safety analysis. Child safety and
study drug tolerability were assessed at every visit. Treatment-related adverse events included those judged by the
investigators to be probably related or possibly related to the
study drug, and each event was categorized by the investigators as mild, moderate or severe. Diarrhea was defined as 3
watery stools/d or 2 watery stools/d for 2 consecutive days.
Any other modification of the childs stools, referred by the
parents, was classified as abnormal stools. Compliance with
the study regimen was verified by parental diaries and inspection of returned medication bottles at the end-of-treatment visit.

Cefdinir for Otitis Media

ables: geographic region (Israel, United States and Latin

America); age (24 months of age and younger or older than
24 months of age); baseline pathogens (single or multiple
baseline pathogens); and presence or absence of penicillinnonsusceptible S. pneumoniae. The level of statistical significance was set at P 0.05.

RESULTS
Patients
A total of 447 children received at least 1 dose of
cefdinir and were included in the safety analysis. The proportions of children treated in Latin America, Israel and the
United States were 68, 17 and 15%, respectively. Of these,
217 children were excluded from bacteriologic and bacteriologic/clinical efficacy populations, because of lack of a baseline pathogen (n 185) and children not returning for the
relevant visit (n 20 at the on-therapy visit and n 21 at
end of treatment, respectively). There were 230 children
evaluable at visit 2 for a bacteriologic assessment; and among
230 children with a baseline bacterial pathogen available for
visit 3 evaluation, 229 had a end of treatment clinical assessment (clinical and bacteriologically evaluable population).
Demographic and clinical characteristics of the clinically and bacteriologically evaluable population are presented
in Table 1. The majority (74%) of children were 24 months of
age or younger; more than one-half had bilateral infection
(56%) and more than one-half had received treatment of
AOM in the previous 3 months (57%). Children included in
Israel were younger (mean age, 13.0 months) than children
included in the United States (mean age, 17.5 months) and in
Latin America (mean age, 25.8 months) (P 0.001).

Microbiology
A total of 266 AOM pathogens were recovered from the
MEF of 230 bacteriologically and clinically evaluable children
at baseline (15.6% of children had a mixed infection). A total of
125 children had H. influenzae either alone (n 94) or with
S. pneumoniae (n 23), M. catarrhalis (n 3), S. pyogenes
(n 1), or S. pneumoniae and M. catarrhalis (n 4). A total

Statistical Analyses
This study included a single treatment of children with
AOM. For each efficacy variable, an overall rate and corresponding 95% confidence interval were calculated. Fishers
exact test was used to compare: child bacteriologic eradication rates and target pathogen eradication rates in children 24
months of age or younger versus 24 months of age or older;
pathogen eradication rates for -lactamase-positive and
-lactamase-negative strains of H. influenzae and for S.
pneumoniae in penicillin-susceptible versus nonsusceptible
strains; clinical cure rates at the end of treatment and end of
study in children infected with penicillin-susceptible versus
nonsusceptible strains of S. pneumoniae; and clinical cure
rate at the end of treatment in children with eradication versus
persistence of the target pathogen at the on-therapy visit. For
analyses greater that 2 2, the Cochran-Mantel-Haenszel
test was used. Multivariate analysis for bacteriologic and
clinical responses was performed with the following vari 2006 Lippincott Williams & Wilkins

TABLE 1. Demographic and Clinical Characteristics of

230 Clinically and Bacteriologically Evaluable Children
Demographic Characteristic

Observed Value

Mean age (mo)

No. of children 24 mo or younger
Gender ratio (M/F)
Mean weight (kg)
No. with siblings/household contact 8 yr or
younger
No. attending day care
No. of AOM infections in previous 12 mo
1
2
3 or more
No. treated for AOM in previous 3 mo
Prior vaccination
Conjugated Haemophilus influenzae type b
Conjugated Streptococcus pneumoniae

19.4 12.6*
171 (74)
1.4
10.9 3.0
167 (73)
96 (42)
73 (32)
61 (27)
96 (42)
130 (57)
198 (86)
32 (14)

*Mean SD.
Numbers in parentheses, percent.

213

The Pediatric Infectious Disease Journal Volume 25, Number 3, March 2006

Arguedas et al

of 109 children had S. pneumoniae either alone (n 81) or

with H. influenzae (n 23), M. catarrhalis (n 1) or H.
influenzae and M. catarrhalis (n 4). M. catarrhalis was
isolated from 23 children (10%), 15 alone and S. pyogenes
from 9 children (4%), 8 alone. Of the 118 H. influenzae
isolates and 19 M. catarrhalis isolates with -lactamase
results, 17 and 100%, respectively, were -lactamase-positive. The MIC90 of cefdinir for H. influenzae and M. catarrhalis was 0.5 and 0.25 g/mL, respectively.
Of the 105 S. pneumoniae isolates with susceptibility
results available, 27 (26%) were intermediately susceptible to
penicillin (MIC 0.121 g/mL) and 23 (22%) were fully
resistant to penicillin (MIC 2 g/mL). Twenty-one (42%)
of the 50 penicillin-nonsusceptible strains of S. pneumoniae
were susceptible to cefdinir. All of the 55 penicillin-susceptible strains of S. pneumoniae were susceptible to cefdinir,
and 5 of 27 (18.5%) penicillin intermediately resistant to S.
pneumoniae had an MIC to cefdinir of 0.12 g/mL. None
of the 23 penicillin-resistant S. pneumoniae isolates was
susceptible to cefdinir. Of the S. pneumoniae isolates with
susceptibility results available, 28% (29 of 105) were cefdinir-resistant.
There were regional differences in the penicillin-susceptible S. pneumoniae percentage rates; in Latin America,
70% (42 of 60) of the S. pneumoniae isolates were susceptible to penicillin, whereas in the United States 53% (8 of 15)
of isolates and in Israel only 17% (5 of 30) were penicillinsusceptible. There were no regional differences with other
target pathogens in terms of cefdinir susceptibility.
Bacteriologic Efficacy. Bacteriologic response was determined for 230 evaluable children from whom a target pathogen was isolated at baseline. The overall child bacteriologic
eradication rate (at days 4 6) was 74% (170 of 230) among
bacteriologic evaluable children, 70% for children 24 months
of age or younger and 86% for children older than 24 months
95% confidence interval (95% CI) for difference, 28.0,
5.7; P 0.01 (Table 2). The study target pathogen
eradication rate was 76% (201 of 266) 72% in children 24
months versus 86% in children older than 24 months of age

(95% CI for difference, 24.7, 4.0); P 0.02, including

72% (90 of 125) of baseline H. influenzae isolates, 74% (81
of 109) of baseline S. pneumoniae isolates, 92% (21 of 23) of
baseline M. catarrhalis isolates and 100% (9 of 9) of baseline
S. pyogenes isolates. Bacterial eradication of all organisms
was observed among 32% (8 of 25) of children with combined infections with S. pneumoniae and H. influenzae 5 of
6 with -lactamase-negative H. influenzae and penicillinsusceptible S. pneumoniae; 2 of 17 with -lactamase-negative H.
influenzae and penicillin-nonsusceptible (MIC 0.125 g/mL)
S. pneumoniae; and 1 of 2 with -lactamase-positive H.
influenzae and penicillin-nonsusceptible S. pneumoniae.
The overall bacteriologic eradication was also analyzed
by participating region; the rates were 83% (114 of 137) for
children from Latin America, 89% (34 of 38) for children
from the United States and 40% (22 of 55) for children from
Israel. Results from a multivariate analysis including all the
variables described in the statistical section (geographic region, age, single or multiple pathogens and presence or
absence of penicillin-nonsusceptible S. pneumoniae) demonstrated that the 2 variables associated with bacteriologic
failures were region (P 0.001) and the presence of a
penicillin-nonsusceptible S. pneumoniae (P 0.02). Although not statistically significant, the mean age of those children considered bacteriologic failures was lower in the Israeli
participants (mean age, 13.2 months) and the United States
participants (mean age, 17.5 months) than in the Latin America
children (mean age, 19.0 months) (P 0.06).
Among all H. influenzae isolates obtained at baseline
(alone or combined), eradication rates among -lactamasenegative and -positive strains were 68% (67 of 98) and 90%
(18 of 20), respectively (P 0.03). Bacteriologic eradication
was observed among 74% (87 of 118) of cefdinir-susceptible
strains and among 33% (1 of 3) a limited number of cefdinirresistant strains.
The overall eradication rate among S. pneumoniae
isolates (alone or combined with other pathogens) was 74%
(78 of 105). Bacteriologic eradication was 89% (65 of 74)

TABLE 2. Bacteriologic Response at the On Therapy Visit (Day 4 to 6) Among

Bacteriologically Evaluable Children
Eradication Rate

Bacteriologic response by child

Target pathogen eradication rate
Overall
Haemophilus influenzae
-Lactamase-positive
-Lactamase-negative
Streptococcus pneumoniae
Penicillin-susceptible
Penicillin-intermediate
Penicillin-resistant
Moraxella catarrhalis
-Lactamase-positive
-Lactamase-negative
Streptococcus pyogenes

95% CI for Difference

Age 24 mo

Age 24 mo

119/171 (70)*

51/59 (86)

28.0, 5.7

144/200 (72)
67/97 (69)
15/17 (88)
50/76 (66)
59/83 (71)
33/36 (92)
16/23 (70)
8/21 (38)
15/17 (88)
14/16 (88)

3/3 (100)

57/66 (86)
23/28 (82)
3/3 (100)
17/22 (77)
22/26 (85)
17/19 (89)
2/4 (50)
2/2 (100)
6/6 (100)
3/3 (100)

6/6 (100)

24.7, 4.0
30.0, 3.8
27.1, 3.6
32.0, 9.0
30.5, 3.4
14.3, 18.7
32.9, 72.0
82.7, 41.1
27.1, 3.6
28.7, 3.7

*Numbers in parentheses, percent.

Includes isolates that constituted single pathogen and mixed infections.

214

2006 Lippincott Williams & Wilkins

The Pediatric Infectious Disease Journal Volume 25, Number 3, March 2006

among cefdinir-susceptible strains and 42% (13 of 32) among

cefdinir-resistant strains.
S. pneumoniae eradication was also analyzed according
to the penicillin MIC. Bacterial eradication was observed in
91% (50 of 55) of penicillin-susceptible strains, in 67% (18 of
27) of penicillin-intermediate strains and in 43% (10 of 23) of
penicillin-resistant strains (P 0.001). When analyzing by
site, eradication rate of penicillin-susceptible S. pneumoniae
was 39 of 42 (93%), 8 of 8 (100%) and 3 of 5 (60%) in Latin
America, the United States and Israel, respectively. The
respective figures of S. pneumoniae eradication among penicillin-intermediate strains and penicillin-resistant strains
was: Latin America, 7 of 9 (78%) and 7 of 9 (78%), respectively; the United States, 4 of 4 (100%) and 2 of 3 (67%),
respectively; and Israel, 7 of 14 (50%) and 1 of 11 (9%),
respectively. Overall eradication rate in penicillin-nonsusceptible versus penicillin-susceptible S. pneumoniae strains was
56% (28 of 50) versus 91% (50 of 55), respectively (P
0.001). No significant differences between the study sites
were observed in term of bacteriologic eradication when
corrected for penicillin MIC.
Clinical Efficacy. Clinical cure rates for the clinically and
bacteriologically evaluable children are presented in Table 3.
The regional clinical cure rates at end of treatment among the
clinically and bacteriologically evaluable children were: 88%
(122 of 139) in Latin America; 68% (26 of 38) in the United
States; and 77% (41 of 53) in Israel. A multivariate analysis
demonstrated that the significant variable for these differences were the presence of a penicillin-nonsusceptible S.
pneumoniae isolates (P 0.048). Clinical cure rates for
children with pretreatment negative and positive cultures was
96 and 83%, respectively (P 0.001). Clinical cure data are
presented by pathogen also in Table 3.
The overall clinical cure rate for children with baseline
H. influenzae (alone or combined) was 82% (103 of 125) at
end of treatment and 68% (78 of 114) at end of study
(persistent clinical cure). The end of treatment clinical cure
rate among children with single H. influenzae infection was
82 of 93 (88%) versus 21 of 32 (66%) clinical cure rates
observed in children in which H. influenzae infections was
part of a mixed infection (P 0.004). The overall clinical

TABLE 3. Clinical Cure Rates Among Clinically and

Bacteriologically Evaluable Children
End of Therapy
(Day 1214)

End of Study
(Day 2528)

All children
Children 24 mo

189/229 (83)*
133/170 (78)

147/172 (85)
98/121 (81)

Baseline pathogen
Streptococcus pneumoniae
Penicillin-susceptible
Penicillin-intermediate
Penicillin-resistant
Haemophilus influenzae
Moraxella catarrhalis
Streptococcus pyogenes

82/108 (76)
48/54 (89)
16/27 (59)
14/23 (61)
103/125 (82)
19/24 (79)
8/9 (89)

Variable

67/104 (64)
42/52 (81)
13/27 (48)
10/23 (43)
78/114 (68)
14/21 (67)
7/8 (88)

*Numbers in parentheses, percent.

Includes isolates that caused single pathogen and mixed infections.

Includes 4 pathogens with no susceptibility results.

2006 Lippincott Williams & Wilkins

Cefdinir for Otitis Media

cure rate for children infected with -lactamase-negative and

-lactamase-positive isolates (alone and combined) was 82%
(80 of 89) and 80% (16 of 20), respectively, at end of
treatment and 66% (59 of 90) and 72% (13 of 18), respectively, at end of study. Clinical failure rates among children
infected with -lactamase-negative and -lactamase-positive
isolates alone was 11% (8 of 71) and 16% (3 of 18) (P
0.68), respectively, and the corresponding values observed in
children with combined infections (H. influenzae together
with other target pathogen) was 37% (10 of 27) and 50% (1
of 2), respectively. When analyzing the number of clinical
failures among children with single or combined H. influenzae infections, the corresponding clinical failure rates was
12% (11 of 89) and 38% (11 of 29) (P 0.005), respectively,
and the Odd ratio for a clinical failure among children with
combined infection versus children with single infection was
4.36 (95% CI, 1.6311.55). The clinical cure rate for those
infected with H. influenzae cefdinir-susceptible or -nonsusceptible isolates was 82% (97 of 118) versus 67% (2 of 3),
respectively, at end of treatment and 68% (73 of 118) versus
67% (2 of 3), respectively, at end of study.
The overall clinical cure rate among children with
baseline S. pneumoniae isolates (alone or combined) was
76% (82 of 108) at end of treatment and 64% (67 of 104) at
end of study. Clinical cure at end of treatment was observed
among 83% (59 of 71) of children infected with a cefdinirsusceptible strain and among 58% (18 of 31) of children
infected with a cefdinir-resistant strain (P 0.048).
Among children with baseline S. pneumoniae, clinical
cure at end of treatment and end of study was analyzed by
penicillin susceptibility. At end of treatment, the clinical cure
rate among children infected with penicillin-susceptible and
penicillin-nonsusceptible isolates was 89% (48 of 54) and
60% (30 of 50), respectively (P 0.0012). Maintained
clinical cure at the end of study was observed in 81% (42
of 52) and 46% (23 of 50) of children, respectively (P
0.001). The data were also analyzed by cefdinir susceptibility. The clinical cure at end of treatment was 60 of 73
(82%) for cefdinir-susceptible S. pneumoniae and 18 of 31
(58%) for cefdinir-nonsusceptible S. pneumoniae (P
0.008). At the end-of-study visit, the clinical cure rate was 52
of 71 (73%) for cefdinir-susceptible S. pneumoniae and 13 of
31 (42%) for cefdinir-nonsusceptible S. pneumoniae (p 0.004).
In children with baseline mixed infections with S. pneumoniae
and H. influenzae, clinical cure at end of treatment and end of
study was observed in 61% (17 of 28) and 57% (13 of 23) of
children, respectively.
Among the clinically and bacteriologically evaluable
children classified as a clinical failure at end of treatment visit
or relapse, 19 underwent a repeat tympanocentesis. A comparison of MICs and -lactamase results between baseline
and follow-up isolates suggested that in 11 of 19 (58%) had
a new infection and 8 of 19 (42%) had a relapse.
Correlation of Bacteriologic With Clinical Efficacy. Among
clinically and bacteriologically evaluable children with a
single pathogen isolated at baseline and a during-treatment
tympanocentesis performed at visit 2, 11% (17 of 156) of
children with bacterial eradication at the on-therapy visit
were clinical failures at end of treatment, and 29% (11 of 38)

215

The Pediatric Infectious Disease Journal Volume 25, Number 3, March 2006

Arguedas et al

of children with persistent isolates at the on-therapy visit

were assigned as clinical failures at the end-of therapy visit
(odds ratio, 3.3; 95% CI 1.41, 7.90; P 0.008).

Safety
The overall incidence of drug-related adverse events
was 52% (232 of 447), with abnormal stools (26%), diarrhea
(14%) and vomiting (9%) being the most commonly reported
events. Most drug-related adverse events were considered
mild or moderate, and they resolved rapidly without the need
for additional treatment. A minority (9.2%) of children required premature discontinuation of study drug because of a
drug-related adverse event(s) (ie, 3 because of vomiting and
1 because of allergic reaction, diarrhea, anxiety, increased
alanine aminotransferase/aspartate aminotransferase, somnolence/diarrhea and nausea/abdominal pain/vomiting/dizziness/headache). There were no serious adverse events that
were attributed to the study drug. Five (1%) children experienced a serious adverse event unrelated to study drug during
their participation in the study (2 each with pneumonia and 1
each with gastrointestinal disorder, gastroenteritis and accidental injury, all leading to hospitalization).

DISCUSSION
Although there are limited data regarding the penetration of cefdinir into the middle ear26 and therefore the
pharmacokinetic and pharmacodynamic characteristics of this
compound are difficult to predict, cefdinir has good in vitro
activity against most of the bacterial pathogens commonly
found in the MEF of children with otitis media, including
penicillin-susceptible S. pneumoniae isolates and -lactamase-producing H. influenzae strains.15,16 Cefdinir is not
approved for the treatment of children infected with penicillin-nonsusceptible S. pneumoniae isolates.26 The approval of
cefdinir was supported by the results of various comparative
clinical trials performed in children 6 months12 years of age
with uncomplicated AOM; clinical success reported with this
agent was similar to that of comparative agents.1722 Specifically end of treatment clinical success rates observed in
previous AOM trials with cefdinir at 14 mg/kg (73 80%)
were similar to the clinical success rates observed in our
study; furthermore, in one trial that included a baseline
tympanocentesis, with the use of cefdinir at 14 mg/kg/d, the
per pathogen clinical success was similar to the results
observed in the present study with a tendency towards lower
success rates in children with penicillin-nonsusceptible S.
pneumoniae than among children with penicillin-susceptible
S. pneumoniae.18
Recently the American Academy of Pediatrics and the
American Academy of Family Physicians recommended cefdinir as an alternative to amoxicillin for children with uncomplicated AOM who experienced non-type 1 penicillin
allergy, and a recent comparative, nontympanocentesis clinical trial suggested that 5 days of treatment with cefdinir (14
mg/kg divided twice daily) was comparable with a 10-day
course of low dose amoxicillin/clavulanate.21,27
Because of the need for antimicrobial alternatives for
treating AOM caused by penicillin-nonsusceptible S. pneumoniae and the lack of information regarding the value of

216

cefdinir in children at risk for infection caused by resistant

MEF pathogens, the present study was undertaken. Specifically our objective was to determine microbiologic and clinical responses to cefdinir at a higher dose (25 mg/kg once
daily) among children at risk for persistent or recurrent
infection, a group most likely to have severe disease and to be
difficult to treat.
Although we recognize that using a during-treatment
bacteriologic assessment as the primary analysis variable is
debatable, there is extensive evidence indicating that children
with during-treatment bacterial persistence have a higher rate
of clinical failure at the end-of-treatment visit and clinical
relapse during the posttreatment follow-up period.9,12 These
findings were also confirmed in this trial; the odds ratio for
end-of-treatment clinical failure was 3 times higher among
children in whom a MEF culture was still positive at the
on-treatment visit than in those with sterile MEF or no MEF.
H. influenzae was the most common pathogen observed
in this population, followed by S. pneumoniae, an observation
previously reported even from countries where pneumococcal
conjugate vaccines are not routinely used.8,9,28 In this study,
17% of H. influenzae strains produced -lactamase, and 48%
of the S. pneumoniae strains were penicillin-nonsusceptible.
Although regional differences were observed in the percentage of during-treatment bacteriologic persistence, a multivariate analysis demonstrated that these differences were attributed to the presence of penicillin-resistant S. pneumoniae in
both the clinical and bacteriologic response.
The overall bacteriologic eradication rate at the time of
the repeated tympanocentesis (on days 4 6 of treatment) was
76%, with the eradication rate lower in children 24 months of
age or younger than in children older than 24 months of age
(72% versus 86%, respectively; P 0.012). This age-related
difference in bacterial eradication was mostly related to the
higher proportion of penicillin- and cefdinir-nonsusceptible
S. pneumoniae causing infection in the younger age group,
because for penicillin- and cefdinir-susceptible S. pneumoniae, higher eradication rates were observed in both
groups. Eradication rates for M. catarrhalis (92%) and S.
pyogenes (100%) were excellent and similar to those seen
with other antimicrobial agents.12,29
In the case of H. influenzae, bacterial eradication was
lower among -lactamase-negative strains (68%) than among
-lactamase-positive strains (90%). The relative overall low
eradication rate for a third generation cephalosporin against
H. influenzae (72%), especially given at a high dose, was
unexpected and might have been related to the larger number
of children with combined infections in the group of children
with -lactamase-negative strains (27 children) than among
-lactamase-positive strains (2 children) (P 0.15), because
it was clearly not related to differences in age between
children with -lactamase-positive strains (mean age, 19.7
months; SD 15.68) and -lactamase-negative strains (mean
age, 19.1 month; SD 11.85) or with any other high risk factor
such as number of children with siblings 8 years old or
younger (65% versus 73%, respectively; P 0.62), number
of children attending day-care centers (41% versus 50%,
respectively; P 0.61), number of children with 3 otitis
media episodes in the previous 12 months (50% versus 44%,
2006 Lippincott Williams & Wilkins

The Pediatric Infectious Disease Journal Volume 25, Number 3, March 2006

respectively; P 0.84), number of children with antimicrobial treatment in the previous 3 months for an otitis media
episode (65% versus 59%, respectively; P 0.81). Furthermore the odds ratio for a clinical failure in the case of
children with combined infection versus children with single
H. influenzae infection was significant (odds ratio, 4.3).
The high cefdinir dose given in this study aimed at
better eradication of penicillin-nonsusceptible S. pneumoniae
than the standard 14 mg/kg daily dose. However, the eradication rate among penicillin-intermediate (67%) and penicillin-resistant S. pneumoniae (43%) was significantly lower
than the eradication observed among penicillin-susceptible
strains (91%) and the presence of penicillin-nonsusceptible S.
pneumoniae was the most important factor associated with
bacteriologic persistence and clinical failures. The relatively
low eradication rates among penicillin-nonsusceptible S.
pneumoniae isolates despite the higher cefdinir dose used in
this trial (25 mg/kg once a day) suggest a limitation for the
use of this agent for treating AOM episodes associated with
these strains and is in accordance with the current approval of
cefdinir for the treatment of otitis media children with penicillin-susceptible S. pneumoniae infections but not penicillinnonsusceptible S. pneumoniae infections.
Recently cefdinir MIC interpretative standard values
for S. pneumoniae were modified from the previous penicillin-based values.25 The bacteriologic and clinical results from
our study strongly suggest that the current Clinical and
Laboratory Standards Institute interpretation for cefdinirsusceptible (0.5 mg/L), intermediate (1 mg/L) and resistant
(2 mg/L) is probably not appropriate and that respective
values of 0.12, 0.50 and 1.0 mg/L for cefdinir-susceptible,
-intermediate and -resistant S. pneumoniae may be more
appropriate for middle ear infections.
The results of this clinical trial are consistent with those
of a recently presented pharmacokinetic study. Bowlware et
al30 showed that despite higher Cmax and area under the curve
values obtained with cefdinir 25 mg/kg versus 14 mg/kg once
daily, T MIC did not exceed 40% of the cefdinir dosing
interval for penicillin-nonsusceptible S. pneumoniae, predicting many bacteriologic and clinical failures. There is a clear
need to better characterize the pharmacokinetic and pharmacodynamic profile of cefdinir to fully evaluate whether other
dosing regimens can improve the bacteriologic outcome
against penicillin-nonsusceptible S. pneumoniae strains.
Among children in whom a follow-up tympanocentesis
was performed at the time of clinical failure or the relapse,
new pathogens were detected in 58%. This is consistent with
previous reports showing that most clinical recurrences are
associated with new infections.31 Similar to previous reports,
H. influenzae, alone or in combination, was the most common
pathogen detected at the time of the failure or relapse (63%,
12 of 19), and it was interesting to document lower clinical
success rates in children when H. influenzae was isolated at
baseline with other pathogens than when it was isolated as a
single pathogen (P 0.004).12,13,32
In this study, the high cefdinir dose of 25 mg/kg once
daily was generally well-tolerated with most of the adverse
events involving the gastrointestinal tract. Abnormal stools,
diarrhea and vomiting were the more frequent events and in
2006 Lippincott Williams & Wilkins

Cefdinir for Otitis Media

most of the cases were mild or moderate resolving without

the need for additional treatment. The total number of adverse
events, particularly gastrointestinal, was higher in the current
study than in previous studies with the lower daily dose of 14
mg/kg, and this may be related to a more active surveillance
for abnormal stools in our trial or to the increment in the total
cefdinir dose.1722
In conclusion, although both bacteriologic and clinical
results were excellent for penicillin-susceptible S. pneumoniae, M catarrhalis and S. pyogenes and intermediate for
H. influenzae, the reduced bacteriologic and clinical responses to other organisms, especially to penicillin-nonsusceptible S. pneumoniae, do not demonstrate a significant
improvement of the high dose cefdinir over the approved
daily dose of 14 mg/kg for the treatment of children with
AOM at increased risk of persistence or recurrence.

ACKNOWLEDGMENTS
We thank the following investigators and their study
teams for enrolling their children into the study: Jack B.
Anon, MD, Erie, PA; Edwin J. Asturias-Barnoya, MD, Guatemala City, Guatemala; Maria L. Avila-Aguero, MD, San
Jose, Costa Rica; Michelle Collins-Ogle, MD, Chapel Hill,
NC; Jesus M. Feris-Iglesias, MD, Santo Domingo, Dominican Republic; Nira Goldstein, MD, Brooklyn, NY; Richard V.
Guzzetta, MD, Clovis, CA; Candice E. Johnson, MD, PhD,
Denver, CO; Adib F. Rodriguez-Solares, MD, Guatemala
City, Guatemala; Andres I. Rosenblut Ratinoff, MD, Santiago, Chile; Xavier Saez-Llorens, MD, Panama, Republic of
Panama; Maria E. Santolaya de Pablo, MD, Santiago, Chile;
Kajori G. Thusu, MD, MBBS, Dinuba, CA.
We thank Todd A. Busman for his assistance in statistical analysis of the study results.
REFERENCES
1. Klein JO. Otitis media. Clin Infect Dis. 1994;19:823 833.
2. Block SL, Harrison CJ, Hedrick J, Tyler R, Smith A, Hedrick R.
Restricted use of antibiotic prophylaxis for recurrent acute otitis media
in the era of penicillin non-susceptible Streptococcus pneumoniae. Int
J Pediatr Otorhinolaryngol. 2001;61:47 60.
3. Daly KA, Brown JE, Lindgren BR, Meland MH, Le CT, Giebink GS.
Epidemiology of otitis media onset by six months of age. Pediatrics.
1999;103(6, pt 1):1158 1166.
4. Teele DW, Klein JO, Rosner B. Epidemiology of otitis media during the
first seven years of life in children in greater Boston: a prospective,
cohort study. J Infect Dis. 1989;160:8394.
5. Auinger P, Lanphear BP, Kalkwarf HJ, Mansour ME. Trends in otitis
media among children in the United States. Pediatrics. 2003;112:514
520.
6. Lanphear BP, Byrd RS, Auinger P, Hall CB. Increasing prevalence of
recurrent otitis media among children in the United States. Pediatrics.
1997;99:e1 e7. Available at: http//www.pediatrics.org/cgi/content/full/
99/3/e1. Accessed January 25, 2005.
7. Pichichero ME. Recurrent and persistent otitis media. Pediatr Infect Dis J.
2000;19:911916.
8. Arguedas A, Dagan R, Soley C, et al. Microbiology of otitis media in
Costa Rican children, 1999 through 2001. Pediatr Infect Dis J. 2003;
22:10631068.
9. Leibovitz E, Jacobs MR, Dagan R. Haemophilus influenzae: a significant
pathogen in acute otitis media. Pediatr Infect Dis J. 2004;23:11421152.
10. Pichichero ME, McLinn S, Aronovitz G, et al. Cefprozil treatment of
persistent and recurrent acute otitis media. Pediatr Infect Dis J. 1997;
16:471 478.

217

Arguedas et al

The Pediatric Infectious Disease Journal Volume 25, Number 3, March 2006

11. Gehanno P, NGuyen L, Derriennic M, Pichon F, Goehrs JM, Berche P.

Pathogens isolated during treatment failures in otitis. Pediatr Infect Dis J.
1998;17:885 890.
12. Dagan R, Hoberman A, Johnson C, et al. Bacteriologic and clinical
efficacy of high dose amoxicillin/clavulanate in children with acute otitis
media. Pediatr Infect Dis J. 2001;20:829 837.
13. Arrieta A, Arguedas A, Fernandez P, et al. High-dose azithromycin
versus high-dose amoxicillin-clavulanate for treatment of children with
recurrent or persistent acute otitis media. Antimicrob Agents Chemother.
2003;47:3179 3186.
14. Brook I, Gober AE. Antimicrobial resistance in the nasopharyngeal flora
of children with acute otitis media and otitis media recurring after
amoxicillin therapy. J Med Microbiol. 2005;54(pt 1):83 85.
15. Guay DRP. Pharmacodynamics and pharmacokinetics of cefdinir, an
oral extended spectrum cephalosporin. Pediatr Infect Dis J. 2000;19:
S141S146.
16. Sader HS, Fritsche TR, Mutnick AH, Jones RN. Contemporary evaluation of the in vitro activity and spectrum of cefdinir compared with other
orally administered antimicrobials tested against common respiratory
tract pathogens (2000 2002). Diag Microbiol Infect Dis. 2003;47:515
625.
17. Adler M, McDonal PJ, Trostmann U, Keyserling CH, Tack K. The
Cefdinir Otitis Media Study Group. Comparative safety and efficacy of
cefdinir vs. amoxicillin/clavulanate for treatment of suppurative in
children. Pediatr Infect Dis J. 2000;19:S166 S170.
18. Block SL, Hedrick JA, Kratzer J, Nemeth MA, Tack KJ. Five-day twice
daily cefdinir therapy for acute otitis media: microbiologic and clinical
efficacy. 2000;19(suppl):S153S158.
19. Block S, McCarty JM, Hedrick JA, Nemeth MA, Keyserling CH, Tack
K. The Cefdinir Otitis Media Study Group. Comparative safety and
efficacy of cefdinir vs. amoxicillin/clavulanate for treatment of suppurative acute otitis media in children. Pediatr Infect Dis J. 2000;
19(suppl):S159 S165.
20. Block S, Kratzer J, Nemeth MA, Tack K. Five-day cefdinir course vs.
ten-day cefprozil course for treatment of acute otitis media. Pediatr
Infect Dis J. 2000;19:S147S152.
21. Block SL, Busman TA, Paris MM, Bukofzer S. Comparison of five-day
cefdinir treatment with ten-day low dose amoxicillin/clavulanate treatment for acute otitis media. Pediatr Infect Dis J. 2004;23:834 838.

218

22. Block SL, Cifaldi M, Gu Y, Paris MM. A comparison of five days of

therapy with cefdinir or azithromycin in children with acute otitis media:
a multicenter, prospective, single-blind study. Clin Ther. 2005;27:786
794.
23. National Committee for Clinical Laboratory Standards. Performance
standard for dilution antimicrobial susceptibility tests. Approved standard, M7-A45. Wayne, PA: National Committee for Clinical Laboratory
Standards; 2000.
24. National Committee for Clinical Laboratory Standards. Performance
standard for dilution antimicrobial susceptibility testing: 12th informational supplement, M-100-S12. Wayne, PA: National Committee for
Clinical Laboratory Standards; 2002.
25. National Committee for Clinical Laboratory Standards. Performance
standard for dilution antimicrobial susceptibility testing: 12th informational supplement, M-100-S12. Wayne, PA: National Committee for
Clinical Laboratory Standards; 2002.
26. Omnicef Prescribing Information, Abbott Laboratories, c. 2004.
27. American Academy of Pediatrics. Clinical practice guidelines: diagnosis and management of acute otitis media. Pediatrics. 2004;113:
14511465.
28. Casey JR, Pichichero ME. Changes in frequency and pathogens causing
acute otitis media in 19952003. Pediatr Infect Dis J. 2004;23:824 828.
29. Leibovitz E, Piglansky L, Raiz S, Press J, Leiberman A, Dagan R.
Bacteriologic and clinical efficacy of one day vs. three day intramuscular
ceftriaxone for treatment of nonresponsive acute otitis media in children.
Pediatr Infect Dis J. 2000;19:1040 1045.
30. Bowlware KL, McCracken GH Jr, Hernandez LJ, Ghaffar F. Cefdinir
pharmacokinetics and tolerability in children receiving 25 mg/kg once
daily. Pediatr Infect Dis J. 2006;25:208 210.
31. Leibovitz E, Greenberg D, Piglansky L, et al. Recurrent acute otitis
media occurring within one month from completion of antibiotic therapy
relationship to the original pathogen. Pediatr Infect Dis J. 2003;22:209
215.
32. Arguedas A, Emparanza P, Schwartz RH, et al. A randomized, multicenter, double blind, double dummy trial of single dose azithromycin
versus high dose amoxicillin for treatment of uncomplicated acute otitis
media. Pediatr Infect Dis J. 2005;24:153161.

2006 Lippincott Williams & Wilkins