eBMJ - O'Neill 319 (7213): 833

Clinical review

Clinical evidence

Acute otitis media

Paddy O'Neill, general practitioner.

Norton Medical Centre, Stockton on Tees TS20 1AN

100674.3120@compuserve.com

This review of the effects of treatment for otitis media and of the effects of
preventive interventions is one of over 60 chapters included in the first
issue of Clinical Evidence, which is published by the BMJ Publishing Group.
Future issues of Clinical Evidence will cover myringotomy; a separate chapter
in issue 1 contains information on otitis media with effusion.

Key messages

We found limited evidence from one RCT that non-steroidal anti-inflammatory
drugs are more effective than placebo in relieving pain in children with acute
otitis media.

Evidence on the effectiveness of antibiotics is conflicting; we found no clear
evidence favouring a particular antibiotic for acute otitis media.

One systematic review of RCTs has found greater immediate benefit but no
difference in long term outcome with short (5 days) rather than longer courses
of antibiotics.

One systematic review of RCTs has found that long term antibiotic prophylaxis
has a modest effect in preventing recurrences of acute otitis media, but the
questions of which antibiotic to use, for how long, and how many episodes of
acute otitis media justify treatment have not yet been adequately evaluated 

Definition:

Otitis media is inflammation in the middle ear. Subcategories include acute
otitis media, otitis media with effusion (also known as "glue ear"), recurrent
acute otitis media, and chronic suppurative otitis media. Acute otitis media
presents with systemic and local signs and has a rapid onset. The persistence
of an effusion beyond three months without signs of infection defines otitis
media with effusion, whereas chronic suppurative otitis media is characterised
by continuing inflammation in the middle ear giving rise to otorrhoea and a
perforated tympanic membrane.

Incidence/prevalence:

Acute otitis media is a common condition with a high morbidity and low
mortality. In the United Kingdom about 30% of children aged under 3 years
visit their general practitioner with acute otitis media each year, and 97%
receive antimicrobial treatment. About 1 in 10 children will have an episode
of acute otitis media by 3 months of age. It is the most common reason for
outpatient antimicrobial treatment in the United States.1

Aetiology:

The most common bacterial causes for acute otitis media are Streptococcus
pneumoniae, Haemophilus influenzae, and Moraxella catarrhalis. The incidence
of penicillin resistant S pneumoniae has risen, but rates vary between
countries. The most important risk factors for poor outcome are young age and
attendance at day care centres such as nursery schools. Others include white
race, male sex, and history of enlarged adenoids, tonsillitis, and asthma.
Other factors that may make a poor outcome more likely include multiple
previous episodes, bottle feeding, a history of ear infections in parents or
siblings, and use of a soother or pacifier. The evidence for the effect of
environmental tobacco smoke is controversial.1

Prognosis:

In about 80% of children the condition resolves without antibiotic treatment
in about three days. Complications are rare but include hearing loss,
mastoiditis, meningitis, and recurrent attacks.1

Aims:

To reduce the severity and duration of pain and other symptoms, to prevent
complications, and to minimise adverse effects of treatment. Outcomes: Pain
control, which can be assessed by proxy measures in infants such as parental
observation of distress or crying and use of analgesics; incidence of
complications such as deafness (usually divided into short term and long term
hearing loss), recurrent attacks of otitis media, mastoiditis, and meningitis;
resolution of otoscopic appearances; incidence of adverse effects of
treatment.

Methods:

Search of the Cochrane Library, Medline, and Embase, using validated search
string, July 1998. We gave priority to relevant systematic reviews. When more
than one was found both were included for comparison. Where systematic reviews
were unavailable, randomised controlled trials (preferably double blind) were
used.

Treating acute otitis media

Likely to be beneficial:

Non-steroidal anti-inflammatory drugs

Trade off between benefits and harms:

Antibiotics

Unknown effectiveness:

Paracetamol

Preventing recurrent otitis media

Likely to be beneficial:

Long term antibiotic prophylaxis

Question: What are the effects of treatments?

Option: Analgesia

References

We found limited evidence from one RCT that non-steroidal anti-inflammatory
drugs are more effective than placebo in relieving pain.2 The trial found no
significant difference between paracetamol and placebo, but the dosing regimen
may not have been optimal and the trial may have been too small.

Benefits

We found no systematic review. We found one double blind multicentre RCT
comparing thrice daily treatment with ibuprofen, paracetamol, or placebo for
48 hours in 219 children aged 1-6 years with otoscopically proved acute otitis
media.2 All children received antibiotic treatment with ceflacor. The
proportions of children still experiencing pain by the second day were 7% with
non-steroidal anti-inflammatory drugs, 10% with paracetamol, and 25% with
placebo. Ibuprofen was significantly more effective than placebo (P<0.01) but
paracetamol was not (P value not given).There was no significant difference
between placebo and active treatments for other outcomes (appearance of the
tympanic membrane, rectal temperature, and parental assessment).

Harms

All treatments were equally well tolerated.

Comment

The trial may have been too small to detect a significant difference between
paracetamol and placebo. The lack of significant difference may also be
explained by the use of a three times daily regimen, since paracetamol is
usually given four times daily. The evidence from this trial may be further
limited because the assessment of the child's pain relief was based on
parental observation, using a scale of 0 or 1.

Option: Antibiotics

References

Evidence from systematic reviews of RCTs is conflicting. The more recent and
inclusive review suggests that antibiotics reduce the proportion of children
still in pain at 2-7 days and reduce the risk of developing contralateral
acute otitis media, but that they have no immediate beneficial effect in terms
of reduced pain within 24 hours, and no long term effect in terms of rates of
subsequent attacks or deafness at one month. Rates of adverse effects are
almost doubled in children receiving antibiotics compared with placebo. There
is no clear evidence favouring a particular antibiotic.

Benefits

Versus placebo or no treatment: We found two systematic reviews. The first
identified four RCTs of antibiotics versus placebo or no treatment in 535
children aged 4 months to 18 years with acute otitis media.3 Co-intervention
with analgesics and other means of relieving symptoms was allowed in most
trials. Rate of treatment success (absence of all presenting symptoms and
signs at around 7-14 days after treatment started) was significantly greater
for active treatment (absolute risk 13.7% v 81%, absolute risk reduction
67.3%; no confidence interval given). This means that seven children would
need to be treated with antibiotics for one additional child to achieve
complete resolution of signs and symptoms, or that six of every seven children
with acute otitis media either do not need or will not respond to antibiotic
treatment (number needed to treat 7, no confidence interval given).

The second systematic review identified six RCTs comparing early use of
antibiotics versus placebo in children aged 7 months to 15 years with acute
otitis media.4 Most trials did not state the time interval between onset of
symptoms and starting treatment: the two that did gave intervals of 1-24 hours
and about 30 hours. Data from the three trials that reported pain outcomes 24
hours after presentation (n=633) suggested that antibiotic treatment had no
effect (61% of children receiving antibiotics were pain free v 60% receiving
placebo). Data from all six trials showed that, two to seven days after
presentation (when only 14% of children receiving placebo still had pain), the
proportion of children still in pain was significantly lower among those
receiving antibiotics (relative risk reduction compared with placebo 41%, 95%
confidence interval 15% to 60%, absolute risk reduction 5.6%), as was the rate
of contralateral acute otitis media (relative risk reduction compared with
placebo 43%; 9% to 64%). There was no significant difference in the rate of
subsequent attacks of acute otitis media (absolute risk reduction 0.1%; 4% to
4%) or deafness at one month (absolute risk reduction 2.3%; 6% to 11%). This
means that 20 children would need to be treated with antibiotics early to
prevent one additional child from experiencing pain at 2-7 days after
presentation (number needed to treat 20; 13% to 46%).5

Versus each other:

We found one systematic review, which identified 33 RCTs of antibiotics in
children aged 4 months to 18 years with acute otitis media (n=5400).3 Compared
with placebo or no treatment, the rate of treatment success (absence of all
presenting signs and symptoms of acute otitis media at around 7-14 days after
treatment was started) was significantly higher with penicillin (increase in
absolute risk 15.7%; 4.7% to 26.7%), ampicillin/amoxicillin (increase in
absolute risk 12.9%; 6.8% to 19%), and for any antibiotic (increase in
absolute risk 13.7%; 8.2% to 19.2%). No significant differences were found
between antimicrobial agents in rate of treatment success at 7-14 days or of
middle ear effusion at 30 days.

Harms

The first review gave no information on adverse events.3 In the second review,
antibiotics were associated with a near doubling of the risk of vomiting,
diarrhoea, or rashes (odds ratio 1.97; 1.19 to 3.25).4

Comment

The first review3 excluded two placebo controlled trials that were included in
the second,4 on the basis that they included myringotomy as part of treatment.
This may have biased results in favour of antibiotic treatment and may explain
the lower number needed to treat given in the first review.

Option:

Short versus longer courses of antibiotics

References

A systematic review of RCTs found increased risk of relapse or reinfection
around 10 days but no difference in long term outcome with short courses (5
days) rather than longer courses of antibiotics.6

Benefits

We found one systematic review, which identified 32 RCTs of antibiotic
treatment in children aged 4 weeks to 18 years with acute otitis media.6
Treatment failure, relapse, or reinfection at an early evaluation (8-19 days)
were significantly more likely with shorter courses of antibiotics (5 days)
than with longer courses (8-10 days) (summary odds ratio compared with longer
courses 1.52; 1.17 to 1.98). However, by 20-30 days there were no significant
differences between treatment groups (1.22, 0.98 to 1.54; absolute risk
reduction 2.3%, 0.2% to 4.9%).

Harms

An RCT of amoxicillin plus clavulanate potassium in 868 children aged between
2 months and 12 years reported protocol defined diarrhoea in 26.7% of children
receiving three times daily treatment for 10 days, compared with 9.6% in
children receiving twice daily treatment for 10 days (P<0.0001), and 8.7% in
children receiving twice daily treatment for five days (P<0.0001).7 No P value
was quoted for the comparison between 10 day and 5 day twice daily treatments.
The trial made no mention of other adverse effects such as rash.

Comment

The five day treatment group did not receive a placebo on days 6-10, which may
have biased the results.

Question: What are the effects of preventive interventions? 

Option: Long term antibiotic treatment

References

One systematic review of RCTs has found that long term antibiotic prophylaxis
has a modest effect in preventing recurrences of acute otitis media.8 The
questions of which antibiotic to use, for how long, and how many episodes of
acute otitis media justify treatment have not yet been adequately evaluated.

Benefits

Versus placebo: We found one systematic review, which identified 33 RCTs
comparing antibiotics versus placebo to prevent recurrent otitis media and
otitis media with effusion.8 Nine of the trials (n=945) looked only at
recurrent otitis media. It was not clear from the review which of the studies
referred only to children; four either included the word "children" in the
title or appeared in paediatric journals. Most studies defined recurrent
otitis media as at least three episodes of acute otitis media in six months.
The most commonly used antibiotics were amoxicillin, co-trimoxazole, and
sulfamethoxazole, given for three months to two years. All nine studies showed
a lower rate of recurrence with antibiotic treatment, although in seven the
difference was not significant. Pooled results showed an absolute risk of 0.08
recurrences per patient per month for active treatment compared with 0.19 for
placebo (absolute risk reduction 0.11 episodes per month; 0.03 to 0.19). This
is a small effect favouring antibiotics, meaning that nine children with
recurrent otitis media would need to be treated for a month to prevent one
additional acute episode (number needed to treat 9, confidence interval not
quoted). Choice and duration of antibiotic: The same systematic review found
no significant difference in rate of recurrence between antibiotics.8 Greater
treatment effect was seen with treatment lasting more than six months, but the
confidence intervals overlapped (absoulte risk of recurrence with courses <6
months 0.21, 0.07 to 0.49; with courses >6 months 0.04, 0.01 to 0.09).

Harms

No evidence was presented on the harmful effects of prophylactic treatment.

Comment

None.

References 

1. Froom J, Culpepper L, Jacobs M, DeMelker RA, Green LA, van Buchem L, et al.
Antimicrobials for acute otitis media? A review from the International Primary
Care Network. BMJ 1997; 315: 98-102[Full Text].

2. Bertin L, Pons G, d'Athis P, Duhamel JF, Mandelonde C, Lasfargues G, et al.
A randomized double blind multicentre controlled trial of ibuprofen versus
acetaminophen and placebo for symptoms of acute otitis media in children.
Fundam Clin Pharmacol 1996; 10: 387-392[Medline].

3. Rosenfeld RM, Vertrees JE, Carr J, Cipolle RJ, Uden DL, Gieink GS, et al.
Clinical efficacy of antimicrobial drugs for acute otitis media: meta-analysis
of 5400 children from thirty-three randomised trials. J Pediatr 1994; 124:
355-367[Medline].

4. Del Mar C, Glasziou P, Hayem M. Are antibiotics indicated as initial
treatment for children with acute otitis media? A meta-analysis. BMJ 1997;
314: 1526-1529[Abstract/Full Text].

5. Johansen HK, Gøtzschw PC, Osborne JE. Antibiotics as initial treatment for
children with acute otitis media. BMJ 1997; 315: 879.

6. Kozyrskj AL, Hildes Ripstein E, Longstaffe SEA, Wincott JL, Sitar DS,
Klassen TP, et al. Treatment of acute otitis media with a shortened course of
antibiotics: a meta-analysis. JAMA 1998; 279: 1736-1742[Medline].

7. Hoberman A, Paradise JL, Burch DJ, Valinski WA, Hedrick JA, Aronovitz GH,
et al. Equivalent efficiency and reduced occurrence of diarrhoea from a new
formulation of amoxycillin/clavulanate potassium (Augmentin) for treatment of
acute otitis media in children. Pediatr Infect Dis J 1997; 16:
463-470[Medline].

8. Williams RL, Chalmers TC, Stange KC, Chalmers FT, Bowin SJ. Use of
antibiotics in preventing recurrent acute otitis media and in treating otitis
media with effusion: a meta-analytic attempt to resolve the brouhaha. JAMA
1993; 270: 1344-1351[Medline]. [Published erratum appears in JAMA
1994;271:430.]