3.1 Acute Otitis Media

3.1.1 ANTIBIOTIC TREATMENT

There is wide variation in the use of antibiotics between doctors in different countries, from as low as 31% of cases with AOM in the Netherlands to as high as 98% in Australia and the United States.¹⁴ Over a 30 year period the number of well conducted studies is small for such a common condition. There have only been eight trials of an acceptable standard and most of these trials suffer from a number of defects.^{17, 18, 36, 37, 38, 39, 40, 41} The number of children entered into these trials ranged from 142 to 536. Evidence level 1++,1+,1-

In the general practice-based studies, the numbers of cases entered per doctor ranged between four and 14.⁴² A typical GP will see about 20 children with AOM every year, so a notable proportion of children have been excluded. No information is available about patients who were excluded from these trials. Low recruitment rates indicate that the type of children entered into trials may only be those with mild to moderate symptoms and signs. This raises the issue of how to interpret results when applied to all children with symptoms and sign of AOM. Evidence level 1++,1+,1-

Another problem in assessing the evidence from clinical trials is that entry criteria vary considerably. Some studies based clinical diagnosis of AOM largely on the presence of acute earache and at least one abnormal eardrum, but two studies excluded children with perforated tympanic membranes and it was not clear if all children with bulging red drums were considered suitable for inclusion. Interpretation of the results is further complicated by the fact that five of the trials excluded children under two years of age.^{17, 18, 37, 38, 41} Only one trial has studied antibiotic treatment in children under two years of age, but even this study excluded infants who “needed antibiotics according to the doctor”.⁴⁰ Evidence level 1++,1+,1-

A meta-analysis of antibiotic versus placebo trials shows that antibiotics do not influence resolution of pain within 24 hours of presentation. At two to seven days after presentation, only 14% of children in control groups still have pain, although early use of antibiotics reduces the risk of pain by about 40%.⁴³ Antibiotics also reduce contralateral AOM but seem to have little influence on subsequent attacks of otitis media or deafness. Antibiotics are associated with a near doubling of the risk of vomiting, diarrhoea or rashes.¹⁷ Evidence level 1++,1+,1-

A study of predictors of poor outcome found that in children with AOM but without fever and vomiting, antibiotic treatment had little benefit.⁴⁴ The lack of antibiotic did not lead to a poor outcome. The simplest method to target the minority of children at higher risk of poor outcome would be to select for antibiotic treatment those children with systemic features (ie either high temperature or vomiting). Another study has found that antibiotic treatment may benefit infants and younger children with severe AOM.⁴⁵ Evidence level 1++,1+,1-

About 17 children with AOM would need to be treated with a broad spectrum antibiotic rather than no antibiotic treatment to avoid a clinical failure.⁴⁶ Evidence level 1++,1+,1-

An alternative to antibiotics is recolonisation with a streptococci which significantly diminishes the recurrence rate of AOM in susceptible children, but this is not currently a practical proposition in primary care in the UK.⁴⁷ Evidence level 1++,1+,1-

Antibiotics in comparison to placebo and observational treatment may have a modest benefit on symptom resolution and failure rates, as variously defined, in children over the age of two years with AOM. The available evidence on natural history of AOM shows that in studies with close follow up, very few episodes of mastoiditis or other suppurative complications are reported in children with AOM not initially treated with antibiotics.

Children diagnosed with acute otitis media should not routinely be prescribed antibiotics as the initial treatment.

3.1.2 DELAYED ANTIBIOTIC TREATMENT

In a delayed treatment trial, 315 children aged six months to 10 years were allocated to one of two treatment strategies: immediate antibiotic or delayed antibiotic (antibiotic to be collected at parents’ discretion after 72 hours if the child has not improved).⁴⁸ The outcome measures were symptom resolution, absence from nursery or school and paracetamol consumption. The main conclusions from this trial were that: Evidence level 1+

• immediate antibiotics provided symptomatic benefit mainly after the first 24 hours, when symptoms were already resolving
• immediate antibiotics increased the incidence of diarrhoea by 10%
• only 24% of the parents in the delayed prescription group used antibiotics
• a “wait and see” approach in the management of AOM is feasible and acceptable to most parents and results in a 76% reduction in the use of antibiotic prescriptions. Evidence level 1+

A large number of children with presumed AOM are seen by out-of-hours services. Given that follow up by the child’s own GP is unlikely to be delayed by more than a few hours, adequate analgesia plus a “wait and see” approach is reasonable as opposed to automatic recourse to antibiotic treatment.

Delayed antibiotic treatment (antibiotic to be collected at parents’ discretion after 72 hours if the child has not improved) is an alternative approach which can be applied in general practice.

3.1.3 CHOICE AND DURATION OF ANTIBIOTIC THERAPY

A large number of studies have established that, where organisms have been isolated from the middle ear, two organisms, Streptococcus pneumoniae and Haemophilus influenzae, are the principal aetiological agents in bacterial infection.⁴⁹ Occasionally Moraxella catarrhalis can be isolated.

With S. pneumoniae and H. influenzae, broad spectrum antibiotics such as amoxicillin, or amoxicillin with clavulanic acid, are the drugs of choice if an antibiotic is to be used. Cefaclor, cotrimoxazole, trimethoprim and erythromycin can be effective, but are less safe than amoxicillin.⁵⁰

The optimal duration of treatment is not known and varies worldwide, with 50% of GPs prescribing a five day course in the UK, and the majority of doctors who treat AOM in the Netherlands using a six to seven days’ duration of antibiotic therapy. In North America the standard duration of treatment is recommended as 10 days.

A Cochrane review of duration of treatment found that five days of antibiotic is an effective treatment of uncomplicated ear infections in children.⁵¹ The optimum duration of treatment for infants and very young children and for children with severe AOM, has yet to be established. Some UK based general practice studies have shown that short course treatment (two to three days’ antibiotic) at conventional or high dose levels is as effective as the traditional five day course in children aged three or older,^{52, 53} but in view of the small number of studies of two to three days’ treatment, the conventional five day course is recommended at dosage levels indicated in the British National Formulary.⁵⁴ Evidence level 1+

If an antibiotic is to be prescribed, the conventional five day course is recommended at dosage levels indicated in the British National Formulary.

3.1.4 DECONGESTANTS, ANTIHISTAMINES AND MUCOLYTICS

A Cochrane review of the efficacy of decongestant and antihistamine therapy for AOM examined a total of 13 RCTs published between 1993 and 2000, involving 2,569 patients. A meta-analysis of these studies was performed. For the combined control groups, healing rates at two weeks were high, with rates of persistent AOM <23%. No additional benefit was demonstrated in intervention subgroups. Only the combined decongestant and antihistamine treatment group demonstrated statistically lower rates of persistent AOM at the two week period. No benefit was found for other outcomes including early or late cure rates, symptom resolution, prevention of surgery or other complications. There was an increased risk of medication side effects for those receiving an intervention, which reached statistical significance for the “any medication” and decongestant groupings.⁵⁵ Evidence level 1++

Given the lack of benefit and increased risk of side effects, these data do not support the use of decongestant, antihistamine, or combined decongestant and antihistamine treatment in children with AOM. The small statistical benefit found in the combination medication group is of small clinical significance. No evidence to support the use of mucolytics for AOM was found.

Children with acute otitis media should not be prescribed decongestants or antihistamines.

3.1.5 ANALGESICS

One study of the efficacy of paracetamol for AOM has been identified.⁵⁶ The original study was flawed and relied on a parental pain observation scale. Recalculation from the original figures showed a statistically significant benefit for the use of paracetamol. Although non-steroidal anti-inflammatory drugs are frequently used by parents, caution should be exercised because of the side effect profile. Evidence level 1+

Parents should give paracetamol for analgesia but should be advised of the potential danger of overuse.

3.1.6 OILS

Two RCTs have been identified and both show no benefit of inserting oils in reducing pain in AOM.^{57, 58} Evidence level 1+

Insertion of oils should not be prescribed for reducing pain in children with acute otitis media.

3.1.7 HOMEOPATHY

There were no good quality trials identified in the treatment of AOM with homeopathy. One trial was identified as a randomised controlled study between antibiotic and homeopathic treatment.⁵⁹ This study claimed marginal benefits for the homeopathically treated group but was poorly constructed with limited randomisation and very unequal group sizes. Evidence level 1-

With a lack of robust evidence no recommendation can be made regarding the use of homeopathy in the treatment of AOM.

3.2 Otitis Media with effusion

3.2.1 ANTIBIOTIC TREATMENT

There is an extensive literature on the role of antibiotics in the management of OME but relatively few RCTs. Many of the available RCTs are comparison of one antibiotic with another rather than on the overall effectiveness of antibiotics on the condition. The evidence for the effectiveness of antibiotics is conflicting with some claiming substantial benefits and others not demonstrating benefit. Several meta-analyses of varying quality have been produced, again with conflicting conclusions.

On balance, the better conducted trials suggest short term benefit from antibiotics but this appears to be very short lived (two to four weeks). Two American systematic reviews suggest benefit at one month but there is no evidence of benefit beyond this.^{32, 60} A third meta-analysis of eight RCTs suggests no benefit from the particular antibiotic used.⁶¹ The particular antibiotic used does not seem material to this beneficial effect and the duration of treatment is also not relevant. The overall results suggest that there may be some benefit from antibiotics in the short term and the three reviews are not consistent. Evidence level 1-

This very common condition may be managed in a wide variety of ways ranging from observation to the prescription of relatively expensive antibiotics for long periods of time. The magnitude of the beneficial effect is small and the incidence of side effects including diarrhoea, skin rashes, allergy development, anaphylaxis and development of resistant strains of organism is considerable.

The short term benefits which appear to be scientifically demonstrable are not sufficient reason to recommend blanket prescription of antibiotics for this condition.

Children with otitis media with effusion should not be treated with antibiotics.

3.2.2 DECONGESTANTS, ANTIHISTAMINES AND MUCOLYTICS

One systematic review was identified which considered four RCTs dealing with antihistamines and decongestants but reporting on a heterogeneous patient group.³² A further RCT investigating the use of inhaled antihistamine in a Japanese population (age range: 5-38) was also examined.⁶² The studies considering intervention with antihistamines and/or decongestants showed no convincing benefits from the intervention on middle ear effusion clearance rate. Evidence level 1++,1+

With regard to mucolytic therapy, one systematic review comparing S-carboxymethylcysteine, its lysine salt or both versus placebo or no treatment was identified.⁶³ This review included trials with many confounding variables and concluded that there is no significant positive benefit of treatment compared to placebo. A more recent RCT considered the role of S-carboxymethylcysteine versus placebo in reducing the need for surgery in patients with persistent OME.⁶⁴ This study did not show sufficient evidence to promote the routine use of mucolytics. Evidence level 1+,1-

There is no evidence to support the routine use of antihistamines, decongestants or mucolytics in the management of OME, especially considering the potential adverse side effects.

Decongestants, antihistamines or mucolytics should not be used in the management of otitis media with effusion.

3.2.3 STEROIDS

One Cochrane review has been identified.⁶⁵ Trials considered were heterogeneous, with initial diagnostic criteria, intervention and outcome measures being variable. Four comparisons were undertaken: Evidence level 1+

Oral steroids versus control (placebo or non-intervention control)

The review identified three RCTs and concluded that there is no significant difference in improvement between the groups after two weeks of treatment. Evidence level 1+

Oral steroids plus antibiotics versus control plus antibiotic

Steroids combined with an antibiotic lead to a quicker resolution of OME in the short term. However, there is no evidence for long term benefit from treating hearing loss associated with OME with either oral or topical nasal steroids. Evidence level 1+

Intranasal steroid versus control

One study was included in the review and showed no benefit. Evidence level 1+

Intranasal steroid plus antibiotic versus control plus antibiotic or antibiotic alone

One RCT studied the effects of intranasal steroids in combination with antibiotics. This demonstrated an effect in clearing middle ear effusions at four and eight weeks, with a less marked effect at 12 weeks although at this time there was remaining evidence of improved middle ear pressure in those treated with intranasal steroids and antibiotics compared with either antibiotics alone or placebo nasal spray. Evidence level 1+

This systematic review compared its results with two previous reviews.^{32, 66} Although varying improvement rates were reported, all three reviews have concluded that they could not recommend the use of steroids in OME. Evidence level 1+

The use of either topical or systemic steroid therapy is not recommended in the management of children with otitis media with effusion.

3.2.4 AUTOINFLATION

One review that considered six RCTs has been identified.⁶⁷ The evidence from this review is conflicting, but does suggest that there may be some clinical benefit. However, young children may find autoinflation devices difficult to use and trials suggest that improvement is best if there is a high level of compliance. Evidence level 1-

In addition, the evidence is of poor quality as different methods of inflation were used, assessors were not blinded to the treatment, study numbers were small and the follow up period was short. Evidence level 1-

Autoinflation may be of benefit in the management of some children with otitis media with effusion.

3.2.5 HOMEOPATHY

There is very little high quality literature available on the role of homeopathy in the management of OME. One RCT comparing homeopathic and “standard care” for treatment of OME was identified.⁶⁸ The study size was small and randomisation was not concealed. No conclusive effects were demonstrated. Evidence level 1-

There is no evidence available to make any recommendations regarding the role of homeopathy in the management of OME.