The Cochrane Corner

Antibiotics for Otitis Media with Effusion

in Children
Anne G. M. Schilder, MD, PhD1,2, David H. Darrow, MD, DDS3, and
Richard M. Rosenfeld, MD, MPH4

Sponsorships or competing interests that may be relevant to content are disclosed at the end of this article.

Abstract
The Cochrane Corner is a quarterly section in the journal
that highlights systematic reviews relevant to otolaryngology
head and neck surgery, with invited commentary to aid clinical
decision making. This installment features a Cochrane Review,
Antibiotics for Otitis Media with Effusion in Children, that
does not support the routine use of antibiotics for treating
otitis media with effusion in children. Although children treated
with antibiotics had higher rates of effusion resolution, there
was no impact on hearing levels or the need for tympanostomy tubes.
Keywords
systematic review, antibiotic therapy, otitis media with
effusion
Received April 9, 2013; accepted April 10, 2013.

beneficial impact of antibiotic therapy on resolution

of otitis media with effusion (OME) in children was
first demonstrated through systematic review of randomized controlled trials (RCTs) in 1992.1 Despite this finding, concerns of drug-induced bacterial resistance and the
apparent short-term nature of the benefit obtained led to a recommendation against routine treatment of OME in children
with antibiotics, as monotherapy or in combination with a corticosteroid.2 Since this recommendation, however, additional
RCTs have been published, and the method for evidence
synthesis has matured. The high prevalence of OME mandates
that all clinicians who manage children base treatment decisions on high-quality evidence with the lowest risk of bias, a
task that is facilitated by the information in this review.

Antibiotics for Otitis Media with Effusion in

Children, by van Zon A, van der Heijden
GJ, van Dongen TMA, Burton MJ, and
Schilder AGM3
Disclaimer
This is an abstract of a Cochrane Review published in the
Cochrane Library 2012 Issue 9 (see www.thecochranelibrary.com

Otolaryngology
Head and Neck Surgery
148(6) 902905
American Academy of
OtolaryngologyHead and Neck
Surgery Foundation 2013
Reprints and permission:
sagepub.com/journalsPermissions.nav
DOI: 10.1177/0194599813488520
http://otojournal.org

for information). Cochrane Reviews are regularly updated as

new evidence emerges and in response to feedback, and the
Cochrane Library should be consulted for the most recent
version of the review.

Background
Otitis media with effusion (OME) is characterized by an
accumulation of fluid in the middle ear behind an intact
tympanic membrane, without the symptoms or signs of
acute infection. In approximately one in three children with
OME, however, a bacterial pathogen is identified in the
middle ear fluid. In most cases, OME causes mild hearing
impairment of short duration. When experienced in early
life and when episodes of (bilateral) OME persist or recur,
the associated hearing loss may be significant and have a
negative impact on speech development and behavior. Since
most cases of OME will resolve spontaneously, only children with persistent middle ear effusion and associated hearing loss potentially require treatment. Previous Cochrane
reviews have focused on the effectiveness of ventilation
tube insertion, adenoidectomy, autoinflation, antihistamines,
decongestants, and oral and topical intranasal steroids in
OME. This review focuses on the effectiveness of antibiotics in children with OME.

Objectives
To assess the effects of antibiotics in children up to 18
years with OME.

Search Methods
We searched the Cochrane Ear, Nose and Throat Disorders
Group Trials Register; the Cochrane Central Register of
1
Department of Otolaryngology, University College of London, United
Kingdom
2
University Medical Center of Utrecht, Utrecht, the Netherlands
3
Department of Otolaryngology, Eastern Virginia Medical School, Norfolk,
Virginia, USA
4
Department of Otolaryngology, State University of New York, Downstate
Medical Center, Brooklyn, New York, USA

Corresponding Author:
Richard M. Rosenfeld, MD, MPH, Department of Otolaryngology, State
University of New York, Downstate Medical Center, 339 Hicks St,
Brooklyn, NY 11201, USA.
Email: richrosenfeld@msn.com

Downloaded from oto.sagepub.com at IMSS on May 26, 2015

Schilder et al

903

Controlled Trials (CENTRAL); PubMed; EMBASE;

CINAHL; Web of Science; BIOSIS Previews; Cambridge
Scientific Abstracts; ICTRP and additional sources for published and unpublished trials. The date of the search was 22
February 2012.

Selection Criteria
Randomized controlled trials comparing oral antibiotics
with placebo, no treatment or therapy of unproven effectiveness. Our primary outcome was complete resolution of
OME at two to three months. Secondary outcomes included
resolution of OME at other time points, hearing, language
and speech, ventilation tube insertion and adverse effects.

Data Collection and Analysis

Two authors independently extracted data using standardized data extraction forms and assessed the quality of the
included studies using the Cochrane Risk of bias tool.
We presented dichotomous results as risk differences as
well as risk ratios, with their 95% confidence intervals. If
heterogeneity was greater than 75% we did not pool data.

Main Results
We included 23 studies (3027 children) covering a range of
antibiotics, participants, outcome measures, and time points
of evaluation. Overall, we assessed the studies as generally
being at low risk of bias.
Our primary outcome was complete resolution of OME at
two to three months. The differences (improvement) in the
proportion of children having such resolution (risk difference
(RD)) in the five individual included studies ranged from 1%
(RD 0.01, 95% CI 20.11 to 0.12; not significant) to 45% (RD
0.45, 95% CI 0.25 to 0.65). Results from these studies could
not be pooled due to clinical and statistical heterogeneity.
Pooled analysis of data for complete resolution at more
than six months was possible, with an increase in resolution
of 13% (RD 0.13, 95% CI 0.06 to 0.19).
Pooled analysis was also possible for complete resolution at
the end of treatment, with the following increases in resolution
rates: 17% (RD 0.17, 95% CI 0.09 to 0.24) for treatment for
10 days to two weeks, 34% (RD 0.34, 95% CI 0.19 to 0.50)
for treatment for four weeks, 32% (RD 0.32, 95% CI 0.17 to
0.47) for treatment for three months, and 14% (RD 0.14, 95%
CI 0.03 to 0.24) for treatment continuously for at least six
months.
We were unable to find evidence of a substantial
improvement in hearing as a result of the use of antibiotics
for otitis media with effusion; nor did we find an effect on
the rate of ventilation tube insertion. We did not identify
any trials that looked at speech, language, and cognitive
development or quality of life.
Data on the adverse effects of antibiotic treatment
reported in six studies could not be pooled due to high heterogeneity. Increases in the occurrence of adverse events
varied from 3% (RD 0.03, 95% CI 20.01 to 0.07; not significant) to 33% (RD 0.33, 95% CI 0.22 to 0.44) in the individual studies.

Authors Conclusions
The results of our review do not support the routine use of
antibiotics for children up to 18 years with otitis media with
effusion. The largest effects of antibiotics were seen in children treated continuously for four weeks and three months.
Even when clear and relevant benefits of antibiotics have
been demonstrated, these must be balanced against the
potential adverse effects when making treatment decisions.
Immediate adverse effects of antibiotics are common and
the emergence of bacterial resistance has been causally
linked to the widespread use of antibiotics for common conditions such as otitis media.

Comments on Cochrane Review

Comments by Darrow
The rationale for treating OME with antibiotics is that it
may be preceded by bacterial infection, and bacterial pathogens are cultured in approximately one-third of middle ear
effusions (MEEs).4 Furthermore, bacterial DNA and messenger RNA have been detected by polymerase chain reaction in MEEs,5 and bacterial biofilms and live bacteria have
also been demonstrated with high frequency.6,7
In their review of 23 studies involving 3027 children,
van Zon and colleagues3 noted that choice of antibiotic,
duration of therapy, outcome measures, and times of assessment varied considerably. Interestingly, they did not discuss
the homogeneity of the medical history that preceded enrollment in the included studies, particularly the duration of the
presence of fluid. This information would be important
since, as the authors themselves observe, newly detected
OME (of unknown prior duration) spontaneously resolves
within 3 months in only 28% of children, whereas spontaneous resolution of OME following a known episode of
acute otitis media (AOM) is around 74%.8
The authors established a primary outcome measure of
complete resolution of OME at 2 to 3 months. Two studies
in which antibiotics were given for 2 weeks demonstrated
resolution rates of 1% and 30%, whereas 3 studies in which
antibiotics were given for 3 months found resolution in 15%
to 45%. The reviewers determined that results from the
included studies could not be pooled due to heterogeneity.
When they were used for 1 or 3 months, antibiotics resulted
in 34% and 32% improvements in resolution, respectively,
by the end of treatment (Table 1).
A recommendation against routinely treating OME with
antibiotics, even when most analyses (Table 1) show a statistically significant benefit, must be based on adverse events, benefits that lack clinical importance, or both. Although 2
included studies demonstrated minimal improvement in mean
speech recognition thresholds, effusion-related hearing loss can
still be associated with reduced responsiveness in the classroom, elevated television volumes at home, and reduced quality of life.9-11 Furthermore, no studies examined tympanic
membrane retraction or atrophy, or hearing loss or complications associated with these changes. Finally, the risk of adverse
effects such as skin rash, diaper dermatitis, diarrhea, and

Downloaded from oto.sagepub.com at IMSS on May 26, 2015

904

OtolaryngologyHead and Neck Surgery 148(6)

Table 1. Summary of systematic review outcomes and pooled analyses.

Outcome
OME resolution at 2 to 3 mo
OME resolution at more than 6 mo
OME resolution after 2-wk treatment
OME resolution after 4-wk treatment
OME resolution after 3-mo treatment
OME resolution after 6-mo treatment
New AOM onset at 4 to 8 wk
New AOM onset at more than 6 mo
Need for ventilation tubes
Tympanic membrane sequelae
Adverse effects

No. of
Studies

No. of
Children

Outcome with
Placebo or No Drug,
Mean (Range), %

Absolute Rate
Differencea (95% CI), %

5
6
6
4
2
2
5
2
1
1
6

738
1213
533
150
203
1157
203
60
103

17 (6-39)
29 (0-40)
16 (6-42)
16 (0-26)
27 (10-43)
18 (0-36)
20 (5-47)
51 (43-58)
35
27
3 (0-6)

13 (6 to 19)
17 (9 to 24)
34 (19 to 50)
32 (17 to 47)
14 (3 to 24)
25 (29 to 22)
220 (240 to 0)
27 (233 to 18)
216 (231 to 21)

Adapted from van Zon et al.3 Abbreviations: AOM, acute otitis media; CI, confidence interval; OME, otitis media with effusion; , not stated.
a
Absolute increase (%) in outcomes attributable to antibiotic therapy, beyond what occurred in the control group (placebo or no drug); P \ .05 when the
95% CI does not contain zero.

vomiting was less than 10% in 5 of the 6 included studies.

The authors cite studies outside their analysis to establish a
higher rate of adverse effects.
Although a conclusion that routine use of antibiotics for
treating OME is not justified may be true, it is not entirely
supported by this study, which reveals definite improvements
in OME resolution with antibiotics while failing to demonstrate a significant risk of adverse effects. In addition, the
authors seem to impose their own judgment that resolution of
hearing loss due to OME is not an important goal of treatment, while overlooking other potential advantages of OME
clearance such as reductions in tympanic membrane sequelae,
otalgia, and occurrences of acute otitis media (AOM).
If in fact there is some advantage to using antibiotics for
OME with a low risk of relatively mild complications, a
brief course of medical therapy is a reasonable alternative to
surgery for a child who is struggling at school or developing
structural changes in the eardrum.2 The results may be further improved when antibiotics are given concurrently with
corticosteroids.2 Of course, many children with OME
referred for tympanostomy tubes have already been treated
with multiple courses of antibiotics; for such children, the
risk of adverse effects due to additional antibiotics certainly
outweighs the likelihood of improvement with additional
therapy, underscoring the importance of determining prior
therapy and duration of effusion. Ultimately, the decision to
treat OME with antibiotics is made between a family and
their physician, based on how the persisting effusion is
affecting the family and child and their willingness to
accept the risk of adverse effects.

Comments by Schilder
Dr Darrow raises a number of important issues pertinent to
this review and to managing OME, in general, and in particular the duration of OME at baseline as a potential predictor

of resolution of middle ear fluid. Although resolution rates

for OME following an episode of AOM are known to be
higher than for newly detected OME, it is largely unknown
whether this is also true for newly detected OME of short
duration vs persistent OME. Duration of OME at baseline
varied highly from any to more than 3 months across the
included trials. For the primary outcome complete resolution of OME at two to three months, however, 4 of the 5
(n = 439) trials included children with OME persisting for a
minimum of 2 or 3 months, and 1 trial included children
with OME for more than 1 month. Therefore, we could not
study whether the benefits of antibiotics differed in those
with OME of short duration or following AOM from those
with persistent OME.
I agree with Dr Darrow that the results of most trials
included in the review point toward a small to modest beneficial effect of antibiotics on the resolution of OME. This
would, at first glance, support the use of antibiotics in
OME. However, in clinical practice, it is not the middle ear
effusion as such we aim to treat; rather, the aim is to alleviate, or prevent, the associated hearing loss and potential detrimental effects on quality of life and speech and language
development. We were unable to find sufficient evidence to
support beneficial effects of antibiotics in these domains. As
the focus of the trials included in this review was OME,
newly detected AOM was reported in only 2 trials (n = 202)
in which children received continuous antibiotics or placebo
for 6 months. Antibiotic treatment reduced the proportion of
children experiencing new AOM episodes by 20% (95%
confidence interval, 0.37-0.95; number needed to treat = 5).
None of these studies included parental reported adverse
effects.
Although this Cochrane view does not support the routine
use of antibiotics in children with OME, I do agree that in
some children, its benefits, as shown in some of our analyses,

Downloaded from oto.sagepub.com at IMSS on May 26, 2015

Schilder et al

905

may be worth the potential downsides of treatment. In particular, as a possible alternative to tube insertion, a single course
of antibiotics may be something the clinician and parent
decide to try. We hope this Cochrane Review answers some of
the questions around this common condition and at the same
time provides a rich source of information for future studies.
This Cochrane Corner breaks with the tradition of comments by Mr Martin J. Burton in his role as Coordinating
Editor of the Cochrane ENT Disorders Group. Since June
2012, Martin and I have shared this role as joint
Coordinating Editors. I feel privileged to join my friends
and colleagues in the United States as a contributor to this
journal.

Comments by Rosenfeld
Having published the first meta-analysis of antibiotics for
OME more than 2 decades ago,1 I was pleased to note that
this new and more rigorous analysis reached the same conclusion: despite a modest impact on effusion resolution, routine antibiotic therapy is unjustified because of potential
adverse events. Interest in conducting RCTs of antibiotic
efficacy for OME waned substantially after our study
appeared in 1992, with only 8 of the 23 articles in the current review published thereafter. Nonetheless, the role of
antibiotics in managing OME is still debated, as evidenced
by the somewhat contrasting viewpoints offered in this
Cochrane Corner.
In pondering the role of medical management, in general,
of OME, it helps to recall that otitis media is primarily a
eustachian tube problem. In the words of my mentor,
Charles Bluestone, a childs eustachian tubes are too short,
too floppy, too horizontal, and dont work. Antibiotics, of
course, have no lasting impact on eustachian tube function,
and the modest improvements after therapy are often transient, persisting only until the childs next upper respiratory
infection, AOM, or reflux of nasopharyngeal secretions into
the middle ear. Otitis media remains the ultimate occupational hazard of early childhood, the only real cures
being maturation of the childs eustachian tubes and
immune system.
A key role of Cochrane Reviews is to arm clinicians with
raw material to facilitate evidence-based health care decisions. This review succeeds admirably in that task, leaving
it to clinicians and caregivers to decide if the benefits of
antibiotic therapy outweigh the potential adverse events and
likelihood of OME relapse. In most cases, I would argue
they do not, but sparing and judicious use of antibiotics in
children with OME who are symptomatic or become surgical candidates would seem reasonable.

Author Contributions
Anne G. M. Schilder, writer; David H. Darrow, writer; Richard M.
Rosenfeld, concept, writer, critical revision.

Disclosures
Competing interests: Dr Schilder received a grant for a study on
the microbiology of acute tympanostomy tube otorrhea from
GlaxoSmithKline.
Sponsorships: None.
Funding source: None.

References
1. Rosenfeld RM, Post JC. Meta-analysis of antibiotics for the
treatment of otitis media with effusion. Otolaryngol Head
Neck Surg. 1992;106:378-386.
2. Rosenfeld RM, Culpepper L, Doyle KJ, et al. Clinical practice
guideline: otitis media with effusion. Otolaryngol Head Neck
Surg. 2004;130:S95-S118.
3. van Zon A, van der Heijden GJ, van Dongen TMA, Burton
MJ, Schilder AGM. Antibiotics for otitis media with effusion
in children. Cochrane Database Syst Rev. 2012;(9):CD009163.
4. Poetker DM, Lindstrom DR, Edmiston CE, Krepel CJ, Link TR,
Kerschner JE. Microbiology of middle ear effusions from 292
patients undergoing tympanostomy tube placement for middle ear
disease. Int J Pediatr Otorhinolaryngol. 2005;69:799-804.
5. Rayner MG, Zhang Y, Gorry MC, Chen Y, Post JC, Ehrlich
GD. Evidence of bacterial metabolic activity in culturenegative otitis media with effusion. JAMA. 1998;279:296-299.
6. Daniel M, Imtiaz-Umer S, Fergie N, Birchall JP, Bayston R.
Bacterial involvement in otitis media with effusion. Int J
Pediatr Otorhinolaryngol. 2012;76:1416-1322.
7. Hall-Stoodley L, Hu FZ, Gieseke A, et al. Direct detection of
bacterial biofilms on the middle-ear mucosa of children with
chronic otitis media. JAMA. 2006;296:202-211.
8. Rosenfeld RM, Kay D. Natural history of untreated otitis
media. Laryngoscope. 2003;113:1645-1657.
9. Lee J, Witsell DL, Dolor RJ, Stinnett S, Hannley M. Quality
of life of patients with otitis media and caregivers: a multicenter study. Laryngoscope. 2006;116:1798-1804.
10. Bellussi L, Mandala` M, Passa`li FM, Passa`li GC, Lauriello M,
Passali D. Quality of life and psycho-social development in
children with otitis media with effusion. Acta Otorhinolaryngol
Ital. 2005;25:359-364.
11. Facione N. Quality of life issues in chronic otitis media with
effusion: parameters for future study. Int J Pediatr
Otorhinolaryngol. 1991;22:167-179.

Downloaded from oto.sagepub.com at IMSS on May 26, 2015