Comparing Treatment Choice in Acute Otitis Externa Between Primary and Secondary Care

Otitis externa (OE) is a common condition with a prevalence around 1% in the United Kingdom [1]. Its hallmark is ear discharge with other symptoms including itchiness, otalgia and hearing loss due to ear auditory canal (EAC) occlusion. The main differential is acute otitis media (AOM), typically occurring in children. Bacterial AOM can be managed with oral antibiotics if there are no other complicating factors. Conversely, the mainstay of treatment for OE is topical antibiotics (usually drops or spray) for 7–14 days [2, 3]. Despite this, a recent survey demonstrated GPs commonly prescribe oral antibiotics for OE, with consultations for OE in primary care increasing 25% in the last 15 years [4]. Across the United Kingdom, OE is the most referred pathology to ENT emergency clinics, comprising around 29% of referrals [5]. Indications for referral to an emergency clinic generally include cases of persistent symptoms despite optimal management, or patients requiring additional intervention, for example an otowick in a stenosed EAC to aid topical antibiotic delivery or debris requiring microsuction. The aim of this study was to review patients referred to our ENT emergency clinic (secondary care) for OE to compare how they were managed in primary and secondary care. All patients seen with OE in the ENT emergency clinic at a district general hospital between 1 September 2021 and 31 August 2022 were retrospectively analysed. All patients > 16 years referred from primary care were included. Referrals from any other pathway, for other diagnoses, or with incomplete/missing information were excluded. Data collected included age, gender, treatment given in primary care and in emergency clinic, additional treatment/service in emergency clinic (e.g. ear microsuction or otowick insertion) and number of face-to-face follow-ups after first clinic review. Sub-group analyses were conducted based on initial treatment in primary care to see if this affected management in the emergency clinic. Basic descriptive statistics (mean, standard deviations, percentages) and statistical analysis with Fisher's Exact Test and independent t-tests were used where applicable with a significance level of 0.05. The study was prospectively registered with the hospital audit department, and no ethical approval was required as per the Declaration of Helsinki as an anonymous, retrospective analysis. A total of 236 patients were seen in the emergency clinic for OE during the study period, with eighty-seven included after exclusion of 139 patients (55 referred from non-primary care sources, 51 diagnosed/treated as fungal OE, 17 having incomplete/missing documentation, 16 being < 16 years). From the included patients, the majority were female (n = 53, 60.1%) and > 70 years old (n = 30, 34.5%, range 17–94 years). In primary care, the most common treatment was topical antibiotics (n = 25, 28.7%), followed by combined oral and topical antibiotic treatment (n = 20, 23.0%) and oral antibiotics alone (n = 12, 13.8%). The most prescribed antibiotic was topical Otomize (neomycin/dexamethasone/acetic acid) (n = 14, 56.0%). No antibiotics were given in two cases (2.30%). No treatment was documented in 28 cases (32.2%). In the ENT emergency clinic, rates of topical antibiotics were higher: 88.5% (n = 77). The most prescribed antibiotic was topical Cilodex (ciprofloxacin/dexamethasone) (n = 48, 62.3%). Seventeen patients had their treatment continued (19.5%), all of which were topical antibiotics. Oral antibiotics were only given in one case (1.15%) (a continuation of oral ciprofloxacin commenced in primary care). No other oral antibiotics were continued from primary care, and no new prescriptions of oral antibiotics were given from secondary care. Combined topical and oral treatment was given in four cases (4.6%), and no further treatment was given in five cases (5.7%) (Figure 1 and Table 1). No data regarding swabs (being performed or results) were available from either setting. Patients in the emergency clinic received an additional intervention on their first visit in 70.1% (n = 61) of cases. These included ear microsuction (n = 39, 44.8%), otowick insertion (n = 14, 16.1%) or both (n = 8, 9.20%). Ten patients did not require any follow-up (11.5%). Of the 77 patients followed up (88.5%), the majority had no further treatment (n = 41, 53.3%). Twenty-seven patients (35.1%) had a change in antibiotic, all to a different topical antibiotic. Thirty-six patients were only followed up once (41.4%), with the remaining being followed up more than once (n = 41, 47.1%) (Table 2). In terms of the sub-group analysis, when looking at patients who required microsuction or otowick insertion, there were no statistically significant differences between patients who had no antibiotics (p = 1.00 for microsuction, p = 1.00 for otowick), oral antibiotics alone (p = 0.291, p = 0.659) or oral and topical antibiotics (p = 0.559, p = 0.176), compared to those who only received topical antibiotics. Although the mean number of follow-ups in the emergency clinic was lower in those patients who only received topical antibiotics in primary care (1.64), this was not statistically significant compared to those who were treated with oral antibiotics (1.90, p = 0.510) or oral and topical antibiotics (1.95, p = 0.410) in primary care (Table 3). The results from this study highlight the difference in OE management comparing primary and secondary care. Topical antibiotics were the most prescribed treatment in both settings in keeping with guidelines. However, rates of topical antibiotics were much higher in secondary care (88.5% vs. 28.7%), with oral antibiotics more favoured in primary care (13.8% vs. 1.2%). Although existing data suggests there is no difference in OE cure rate amongst different topical antibiotics [3], there was an observed difference in this study in topical antibiotic preference in primary (Otomize) and secondary care (Cilodex). Compared to the existing literature, studies demonstrated patients from primary care either mostly received no treatment [5] or topical antibiotics [6] prior to referral. with oral antibiotic use between 11% and 16%, similar to this study (13.8%). In contrast to our findings, a systematic review concluded the rate of topical antibiotic prescriptions for OE in primary care was high (77%–95%). However, when examining this more closely, it was noted that this rate was lower in patients attending emergency clinics, suggesting a possible link between lack of topical therapy and the need for further intervention [7]. Several reasons may explain these observed differences. One could be lack of guideline awareness, with a study demonstrating improved rates of topical antibiotic prescriptions from 50% to 60% after letters of encouragement for their use were sent to GPs [8]. Another is the diagnostic uncertainty in treating patients with otorrhoea, particularly differentiating OE and AOM. Primary care doctors have less time per consultation (10 min vs. 15-20 min in our clinic), and less specialist equipment to aid diagnosis. Furthermore, patients who attend emergency clinics are filtered by primary care and thus are more likely to present with established OE rather than vaguer symptoms of ‘ear pain’ or ‘discharge’ requiring work-up. Around 30% of patients attending our emergency clinic did not require any additional services besides topical antibiotics, suggesting management in primary care without onwards referral could have been achieved. Despite this, there were no statistically significant differences in terms of microsuction, otowick or mean number of follow-ups with patients who received ‘sub-optimal’ treatment (either no antibiotic, oral antibiotic or oral and topical antibiotics together) compared to those receiving topical antibiotics in primary care. Although conclusions related to the ‘no antibiotic group’ are likely skewed with the small sample size (n = 2), this finding is unexpected. This may be explained by patients referred to the emergency clinic likely representing a subset of more complex patients inherently requiring additional service (e.g., microsuction, otowick or further follow-up) irrespective of their initial treatment. Another explanation is the availability of more specialist equipment (i.e. microscope and suction) allows for procedures such as microsuction to occur in most consultations as a ‘standard practice’ to optimise patient management. From the patients who had follow-up from their initial appointment (88.5%), only 36 patients (46.7%) required further treatment. This may suggest minimising follow-up could better optimise healthcare resources. A potential alternative could be utilisation of patient-initiated follow-up or telephone clinics [9, 10]. There are a few limitations to this study, the first due to documentation. In 32% of cases, there was no documentation of any prior treatment given before presentation to the emergency clinic. It is unclear whether such patients had no treatment or if given treatment was undocumented. Additionally, some patients had only ‘topical antibiotics’ documented, but not stating the antibiotic type or duration. Furthermore, the lack of microbiology swab data may also have been due to an inability to access patients' GP notes. This highlights the need for a multidisciplinary approach with microbiology to ensure patients are receiving optimal therapy as per the NICE CKS guidance, where ear swabs are recommended to be considered in treatment failure [2]. It may follow that treatment failure could be defined as requiring referral to secondary care, and therefore ear swabs may benefit all patients attending the emergency clinic to guide treatment. The availability of microscopes in such clinics can also allow for swabs to be taken in a safer manner. Furthermore, it is unclear whether decisions to prescribe treatments in primary care were implemented independently or after advice from ENT on-call doctors. These limitations could be overcome by creating an online centralised emergency clinic referral pathway integrated into popular UK primary care computer systems. This would ease the referral burden of primary care doctors faced with difficult OE cases and standardise the information received through a proforma. Despite these limitations, our study includes patients encompassing a wide geographical area, reflecting the most up-to-date analysis of OE management comparing primary and secondary care. Topical antibiotics are less commonly prescribed in primary care, and while this may represent a potential avenue of optimising initial treatment, it does not seem to influence subsequent management in emergency clinics. Future studies can therefore build on this work with a prospective methodology over multiple centres to help see if this is a true effect and draw more robust conclusions through larger data sets. K.K.G. and M.W. designed the work. J.Y., C.Y. and M.N. acquired and analysed the data. K.K.G. and J.Y. drafted the manuscript. All authors revised and approved the final manuscript. K.K.G. agreed to be accountable for all aspects of the work. The authors have nothing to report. The study was prospectively registered with the hospital audit department. Adhering to the guiding principles of the Declaration of Helsinki this was an anonymous, retrospective analysis of case notes and investigations and as such no ethical approval was sought or required. The authors declare no conflicts of interests. The data that support the findings of this study are available from the corresponding author upon reasonable request.