Otitis externa (OE) is een diffuse ontsteking van de huid van de gehoorgang met pijn, jeuk, afscheiding, schilfering, roodheid of zwelling, eventueel in combinatie met gehoorverlies (NHG-standaard, 2014).

Er kan onderscheid gemaakt worden in een primaire vorm en een secundaire vorm. Primair heeft zijn oorzaak in een bacteriële ontsteking van de gehoorganghuid zonder andere bijkomende pathologie. Secundair houdt in dat de gehoorgang ontsteekt bij aanwezigheid van andere pathologie (bijv. chronische suppuratieve otitis media (CSOM), trommelvliesperforatie, cholesteatoom, occlusie door hoortoestel, obstruerende osteomata en/of exostosen, keratosis obturans en/of huidziekten (zoals eczeem en psoriasis)).

Wanneer OE minder dan 12 weken aanhoudt of minder dan 3 keer recidiveert in 12 weken bij volwassenen, wordt gesproken van acute otitis externa (AOE). In deze acute fase kunnen er verschillende topische interventies overwogen worden om de acute otitis externa te behandelen bij volwassenen. Het is momenteel echter onduidelijk welke topische behandeling het meest effectief is en wanneer voor welke behandeling gekozen moet worden om acute otitis externa te behandelen en mogelijk te voorkomen dat deze overgaat in een chronische otitis externa (COE).

Indien adequate primaire behandeling faalt (therapieresistent) dienen andere differentiaaldiagnoses te worden geëvalueerd. Hieronder vallen: otomycose, secundaire vormen van OE en osteomyelitis schedelbasis – maligne otitis externa.

Dry ear (resolution of symptoms) (critical)

Pain (critical)

Quality of life, wearing a hearing aid, hearing loss (important)

Description of studies

The systematic review by Kaushik (2010) reported the effectiveness of different interventions, such as antibiotic, steroid, and antiseptic treatment for acute otitis externa. RCTs in line with the research question of the systematic review, evaluating ear cleaning, topical medication, or systematic therapy for treatment of acute otitis externa were eligible for inclusion. Topical treatments consisted of drops, sprays or ointments. The following combinations were used in the trials included in this review: antiseptic, antibiotic, steroid, antibiotic/steroid, antiseptic/steroid, antiseptic/antibiotic/steroid, antibiotic/steroid/antifungal, antiseptic/astringent. RCTs investigating complicated acute otitis externa, otitis externa secondar to otitis media, or chronic supportive otitis media, chronic otitis externa, fungal otitis externa (otomycosis), eczematous otitis externa, viral otitis externa, and furunculosis were excluded from the systematic review. Electronic searches were performed in the Cochrane Ear, Nose and Throat Disorders Group Trials Register, CENTRAL, PubMed, EMBASE, CINAHL, Web of Science, BIOSIS Previews, Cambridge Scientific Abstracts, mRCT and additional sources, to identify relevant studies up to 6 January 2009. Primary outcomes were: resolution of symptoms and/or signs. Secondary outcomes were: eradication of pathogenic ear canal bacteria (determined by cultures), the recurrence of symptoms, and complications from treatment. Data of 19 published randomized controlled trials, including a total of 3382 participants (children and adults) were included. Three meta-analyses were performed in the systematic review. Subgroup analyses to explore whether heterogeneity can be explained were performed for age, ear cleaning, co-interventions, and methodological quality. For this literature review, we only included the studies performed in adults (N=9). Important limitations of the current review were that results were based on odds ratios of single trials with small to modest sample sizes, that only three meta-analyses were performed, and that the findings might not be generalizable to primary care.

Results

Dry ear (critical)

The outcome measure dry ear (resolution of acute otitis media) was reported by 7 trials included in the review from Kaushik (2010). The results are reported separately per comparison, meta-analyses were not performed.

Quinolone versus non-quinolone antibiotic

One trial included in Kaushik (2010) (Sabater, 1996) compared a quinolone-antibiotic drop with a non-quinolone (aminoglycoside antibiotic drop) with a follow-up at day 8. In total, 26 out of 30 (87%) in patients treated with a quinolone drop were cured at 7 to 9 days, versus 19 out of 24 (79%) in patients treated with a non-quinolone drop. The OR was 1.71 (95%CI 0.4 to 7.23), in favour of the quinolone drop. This difference is considered clinically relevant.

Antibiotic/steroid versus antiseptic

Two trials included in Kaushik (2010) (Slack, 1987; van Balen, 2003) compared an antiseptic with an antibiotic/steroid. The steroid components were hydrocortisone and dexamethasone, respectively. The antiseptic components were boric acid and acetic acid, respectively. The antibiotic component was the same in both trials: polymyxin B + neomycin.

The clinical cure rate at 7 to 9 days reported by Slack (1987) was 6 out of 9 (67%) in patients treated with an antiseptic, versus 3 out of 7 (43%) in patients treated with an antibiotic/steroid. The OR was 2.67 (95%CI 0.35 to 20.51), in favour of an antiseptic. This difference is considered clinically relevant.

The clinical cure rate at 7 to 9 days reported by van Balen (2003) was 19 out of 65 (29%) in patients treated with an antiseptic, versus 31 out of 73 (42%) in patients treated with an antibiotic/steroid. The OR was 0.56 (95%CI 0.28 to 1.14), in favour of an antibiotic/steroid. This difference is considered clinically relevant.

Antibiotic/steroid versus antiseptic/steroid

One trial included in Kaushik (2010) (van Balen, 2003) compared an antibiotic/steroid with an antiseptic/steroid. The study compared a non-quinolone antibiotic/steroid with an antiseptic/steroid drop (polymyxin B + neomycin + dexamethasone versus acetic acid + triamcinolone). The clinical cure rate at 7 to 9 days reported by van Balen (2003) was 29 out of 61 (48%) in patients treated with an antiseptic/steroid, versus 31 out of 73 (42%) patients treated with an antibiotic/steroid. The OR was 1.23 (95%CI 0.62 to 2.43), in favour of an antibiotic/steroid. This difference is not considered clinically relevant.

Antibiotic/steroid versus antibiotic

Two trials included in Kaushik (2010) (Mosges, 2007; Mosges, 2008) compared an antibiotic/steroid versus antibiotic treatment. Mosges (2008) compared an antibiotic/steroid drop with an antibiotic drop. Mosges (2007) used medicated wicks and compared an antibiotic ointment with an antibiotic/steroid.

In Mosges (2008), the steroid component was dexamethasone, the antibiotic component was neomyxin+polymyxin B. The clinical cure rate at day 14 to 20 was 64 out of 164 (39%) in patients treated with an antibiotic versus 84 out of 164 (51%) in patients treated with an antibiotic/steroid. The OR was 0.61 (95%CI 0.39 to 0.94), in favour of an antibiotic/steroid. This difference is considered clinically relevant.

In Mosges (2007) the antibiotic ointment was polymyxin B and bacitracin, the steroid component was hydrocortisone. A medicated gauze strip was inserted on day 1 and removed by the patient after 24 hours. Ointment was then applied twice daily. Another medicated gauze strip was inserted on day three to five and again removed 24 hours later by the patient. Ointment was then applied twice daily until day 10. The authors did not report results by binary outcomes. There were no significant differences between the groups in terms of clinical cure rate. Whether the differences are clinically relevant remains unknown.

Antiseptic/steroid versus antiseptic

One trial included in Kaushik (2010) (van Balen, 2003) compared an antiseptic/steroid with an antiseptic. The antiseptic component was acetic acid, the steroid component was triamcinolone. The clinical cure rate at 7 to 9 days reported by van Balen (2003) was 29 out of 61 (48%) in patients treated with an antiseptic/steroid, versus 19 out of 65 (29%) in patients treated with an antiseptic. The OR was 2.19 (95%CI 1.05 to 4.57), in favour of an antiseptic/steroid. This difference is considered clinically relevant.

Antibiotic/steroid versus steroid alone

One trial included in Kaushik (2010) (Tsikoudas, 2002) compared an antibiotic/steroid with a steroid-only drop. The steroid component was betamethasone, the antibiotic component was neomycin. Tsikoudas (2002) did not report results by binary outcomes. Instead, assessment scores were compared across groups at day 11, using p-values. No statistical difference was demonstrated between the antibiotic/steroid group and the steroid group. Whether the differences are clinically relevant remains unknown.

Antiseptic/antibiotic/steroid versus antiseptic

One trial included in Kaushik (2010) (Johnston, 2006) compared an antiseptic/antibiotic/steroid spray with an antiseptic spray. The antiseptic component was glacial acetic acid, the antibiotic component was neomycin, and the steroid component was dexamethasone. The clinical cure rate at 2 weeks was 12 out of 32 (37.5%) in patients treated with an antiseptic, versus 15 out of 21 (71%) in patients treated with an antiseptic/antibiotic/steroid. The OR is 0.24 (95%CI 0.07 to 0.79), in favour of antiseptic/antibiotic/steroid treatment. The difference is clinically relevant.

Pain (critical)

The outcome measure pain was not reported in the systematic review from Kaushik (2010).

Quality of life (important)

The outcome measure quality of life was not reported in the systematic review from Kaushik (2010).

Wearing hearing aid (important)

The outcome measure wearing a hearing aid was not reported in the systematic review from Kaushik (2010).

Hearing (loss) (important)

The outcome measure hearing loss was not reported in the systematic review from Kaushik (2010).

Level of evidence of the literature

Dry ear (critical)

The level of evidence comes from RCT’s and therefore starts high. The level of evidence was downgraded because of study limitations of the included trials (risk of bias, -1), and a very small number of included patients and/or because the confidence intervals overlap with the boundaries for clinical relevance (imprecision, -2). The level of evidence is therefore very low.

The level of evidence could not be graded for the outcome measures pain, quality of life, wearing a hearing aid, and hearing (loss) as these outcome measures were not reported by the included systematic review.

A systematic review of the literature was performed to answer the following question:

‘What are the (un)favorable effects of topical treatment of acute otitis externa in adults?’

P (Patiënten) = adults (≥ 18 y) with acute otitis externa

I (Interventie) = topical treatment (antiseptic, antibiotic, with or without corticosteroids, aeration, drying, decongesting, cleaning, rinsing with vinegar, ear toilet)

C (Comparison) = mutual comparison of topical treatments

O (Outcomes) = dry ear, quality of life, pain, wearing hearing aid, hearing (loss)

Relevant outcome measures

The guideline development group considered resolution of complaints (dry ear and pain) as critical outcome measure for decision making; and quality of life, wearing a hearing aid, and hearing (loss) as important outcome measures for decision making.

The working group did not define the outcome measures a priori, but used the definitions used in the studies.

The working group defined the following minimal clinically (patient) important differences:

• Dichotomous outcome measures: 25% (RR < 0.8 or RR > 1.25).
• Quality of life: 10%
• Pain: 2 points on VAS scale (10%)

 Search and select (Methods)

A MEDLINE search from 1966 through July 2005 was performed on PubMed with the terms “otitis externa” (MeSH term) and “swimmer’s ear.” Titles and abstracts unrelated to AOE were excluded, leaving 240 articles that were collated under these headings: risk factors (n _ 30), microbiology (n _ 24), pharmacologic intervention (n _ 118), other interventions (n _ 17), epidemiology and practice patterns (n _ 14), potential harms (n _ 30), and otomycosis (n _ 9). Citations and abstracts were distributed to all group members to assist in formulating and prioritizing evidence-based statements.

In January 2010 an additional targeted MEDLINE search was performed through 2006 to supplement the initial broad search. This resulted in 165 new articles, of which the RCT’s (18) and systematic reviews (5) were selected. The search strategy is depicted under the tab Methods. Two systematic reviews were included based on title and abstract screening (Kaushik, 2010 and Rosenfeld, 2006).

In February 2022 an additional search is performed. Database Medline (via OVID) and Embase were searched with relevant search terms between 1 January 2010 and 24 February 2022. The detailed search strategy is depicted under the tab Methods. The systematic literature search resulted in 332 hits. Studies were selected based on the following criteria: systematic reviews, randomized controlled trials, and observational studies on the effects of topical treatment of acute otitis externa. No studies were included based on title and abstract screening. However, from the previous search, the systematic review from Kaushik was included, the systematic review from Rosenfeld (2006) did not add any evidence.

Results

One systematic review was included in the analysis of the literature. Important study characteristics and results are summarized in the evidence tables. The assessment of the risk of bias is summarized in the risk of bias tables.