This article describes the diagnostic techniques needed in the approach to otitis externa in the dog with particular emphasis on the correct identification of microbes causing secondary and perpetuating features of the disease. The common organisms are described, along with the numbers and features that determine whether treatment is necessary, with the emphasis on the correct use of the microscope and cytological interpretation.
Statistics from Altmetric.com
Steve Shaw graduated from the Royal Veterinary College in 1987 and then worked for four years in general practice, before taking a dermatology residency at the Animal Health Trust in 1991. He was awarded the RCVS Certificate in Small Animal Dermatology in 1992. In 2000 he was awarded his PhD entitled ‘The immune response in canine atopy: hypersensitivity to house dust mites (Dermatophagoides species)’. From 2001 to 2006, he was head of dermatology at the Animal Health Trust and then underwent a further residency at Rutland House Referrals from 2006 to 2009. Since 2009, he has worked as an independent dermatologist and in 2011 was awarded the RCVS Diploma in Veterinary Dermatology and is an RCVS Recognised Specialist in Veterinary Dermatology. He is widely involved in veterinary education for veterinary students and general practitioners, and is the national dermatology adviser for the Guide Dogs for the Blind Association.
EAR disease nearly always becomes complicated by bacterial and yeast secondary infection, to the point that we often simplify our approach to otitis by ignoring the predisposing, primary and perpetuating factors until we encounter treatment failure or disease recurrence. Secondary bacterial and Malassezia infections have marked effects on the ear, causing increased inflammation resulting in pruritus and pain and inciting hypersensitivity. Chronic infections play a key role in causing irrevocable changes that perpetuate disease and may result in the need for surgery. Prompt and accurate diagnosis allows treatment to minimise these effects and in this context we must not allow chronic mild signs to go untreated.
This article will describe the microorganisms important in otitis and their identification and concentrate on their correct diagnosis through correct sampling, staining and microscopic examination of cytology samples because cytology is the most important single technique in assessing the involvement of pathogens in the ear. The broader approach to the ear case, the interpretation of culture and susceptibility results and treatment choices are discussed elsewhere in this supplement.
Micro-organisms in the ear canal
Small numbers of bacteria are normal in the canine ear; both commensal and transient organisms are described, particularly Staphylococcus pseudintermedius, and other cocci as well as rods. These bacteria serve similar roles to those on skin more generally, occupying environmental niches that would otherwise be suitable for pathological bacterial growth and acting on lipids to produce free fatty acids, which help create an antimicrobial environment. Ear cerumen contains immunoglobulins (IgA, IgG and IgM) and it is likely that these are accompanied by a wide variety of other substances deleterious to bacteria and yeasts, such as antimicrobial peptides (eg, cathelicidins). These chemicals are produced by sebaceous and ceruminous glands as well as epithelial cells lining the ear canal. The directional growth of cells towards the opening of the ear, and differences in the secretions between deep and superficial parts of the canal, also add to the defence mechanisms by providing a ‘conveyor belt’ for wax and pathogens to the external meatus (Tabacca and others 2011).
In early ear inflammation there is an increase in ceruminous secretions that are more watery, which may help control bacterial proliferation. Later on in the disease process, hyperplasia of both sebaceous and ceruminous glands is less helpful and the movement of cerumen along the canal is reduced. Such hyperplasia is particularly seen in cocker spaniels (Angus and others 2002). Stenosis, accompanied by fibrosis and papular to nodular hyperplasia of the walls of the canal, allows further microbial growth that perpetuates disease (Huang and others 2009).
Failure of the normal balance between host and commensal organisms means that normal bacteria, such as S pseudintermedius, increase in numbers – in atopic dermatitis there is significantly higher carriage in the ears (Bannoehr and Guardabassi 2012). There may be little inflammatory response initially, but as bacterial numbers increase along with toxin production and the formation of biofilms through quorum sensing (Boyen and others 2009), inflammatory cells are seen in the canal. Mixed infection is common early on, but as the ear becomes more diseased, and often under the influence of veterinary or owner interventions, bacteria, usually those more resistant to antibacterial agents, proliferate and monoculture often results. At this point, S pseudintermedius, Pseudomonas aeruginosa, Proteus mirabilis, Escherichia coli, Corynebacterium species, Enterococcus species, and Streptococcus species are reported; as the otitis becomes more chronic, P aeruginosa predominates. Common organisms are shown in Table 1.
Common organisms of ear disease may be resistant to a wide range of antibacterial agents due to local selection pressure; for this reason there can be markedly different susceptibility data from each ear (Oliveira and others 2008). Meticillin-resistant staphylococci (MRS) including S pseudintermedius and Staphylococcus schleiferi (Misic and others 2015) are seen in ear disease. But it is Pseudomonas species that are constitutively resistant to a large number of antimicrobial treatments and that will often gain further resistance on treatment that is often associated with antibacterial resistance in the ear.
Yeasts and fungi
Fungal organisms, particularly Malassezia pachydermatis, form part of the commensal flora of the canine ear. In common with commensal bacteria, early inflammation allows an increase in the number of yeasts, which provokes a variable inflammatory response with similar effects to those seen in bacterial otitis in some cases. Malassezia species associated with otitis in the dog are shown in Table 1. Identification of the species in yeast infections is not commonly undertaken and susceptibility testing is not widely available. Most yeasts are lipid dependent and will not grow on blood agar, but M pachydermatis will grow on non-lipid-supplemented media and this may result in over-reporting of this organism when specific fungal culture has not been undertaken (Crespo and others 2002).
Other fungi, including Aspergillus species have been rarely reported to cause otitis. As filamentous organisms are not adapted to grow in the ear, a careful examination for the underlying causes, especially those that cause immunosuppression, is mandatory (Coyner 2010, Ghibaudo and Peano 2010).
Mechanisms of inflammation
Bacteria and yeasts cause disease through a variety of different mechanisms. Toxin production is important in initiating innate and non-specific immune responses, but many organisms, especially in allergic dogs, may also induce specific responses and contribute to and exacerbate the primary hypersensitivity (Chen and others 2002, Hofer and others 1999). Direct damage may also occur as a result of enzymes, such as collagenase, fibrinolysin, elastase, caseinase and gelatinase produced by bacteria such as P aeruginosa (Bergan 1981). This may cause local tissue damage and weaken or cause rupture of the ear drum as well as contributing to hearing loss in otitis media/interna cases (Stenqvist and Anniko 2004).
When evaluating an animal with otitis, a holistic approach is necessary. Include a detailed history, such as previous medications and the compliance of dog and owner. A general and dermatological examination will provide valuable clues as to what to expect in the ears, and is discussed in detail on pages 2 to 6 of this issue.
Examination of the pinna and outer meatus is possible in the vast majority of dogs, and any discharge, ulceration and swelling may be noted at this point. Samples taken from here may be representative of the infection present in the canal, but should not be relied on. Some dogs have almost normal pinnae and vertical canals and yet have severe disease in the horizontal canal. For this reason, otoscopic examination should be performed, but consideration should be taken to both the behaviour of the dog and the risk of causing unnecessary pain. A sterile cone and speculum must be used. Within the practice, measures should be in place to avoid cross-contamination between patients. The best way to ensure this is to have sufficient cones available for a busy clinic. The cone should be washed clean first and then either sterilised in an autoclave or via a proprietary endoscopic equipment cleaning system. Some studies have shown marked contamination of ear speculae in veterinary clinics with only 60 per cent of ear cones being sterile (Kirby and others 2010).
The nature of the aural discharge can hint as to the likely organism. Typically, dog cerumen is brown with a greasy appearance, and a little cerumen is normal in the canal (Fig 1). In yeast and staphylococcal infections, the discharge becomes lighter and looks moist. In Pseudomonas infection, the discharge is often green, but may also be tarry, black and mucoid, whereas in other rod infections the discharge tends to be creamy and yellow. For some, the odour is also helpful, with yeasts giving a sweet, biscuit-like aroma and many bacterial infections giving a more metallic smell.
In all cases of otitis, discharge should be collected for cytology (Angus 2004). A cotton bud should be gently inserted into the ear and rotated to collect material (Fig 2). In the less tractable patient it is often sensible to also take a swab for culture and use that to place a small amount of material on a sterile slide. Alternatively, a gloved finger can be used with little danger of causing pain in a moving patient, but this may not provide a deep sample. Unfortunately, there may be a considerable difference in the number and type of cells and micro-organisms depending on the level of sampling. Many owners will use tissue or cotton wool to wipe and clean the external surface of the ear and to a certain extent the vertical canal. Home treatments such as ear cleaners may affect an apparent cure and remove pertinent cytology, where further down the ear, pus and waxy debris and marked bacterial and/or yeast proliferation continue. Ideally, ears should not be cleaned for at least 48 hours before examination to help provide good cytology, but for bacterial culture longer periods may be needed. As cleaning may help examination of the ear canal, this may then need to be performed in the consulting room. If there is a risk that superficial medications or erroneous cytological material may be collected from the outer part of the ear canal, a swab can be guarded in a large catheter cover until placed deep in the ear, preferably through an otoscopic speculum.
The cytology sample should be placed on the slide immediately. The cotton tip of the bud should be gently rolled onto the slide trying not to smear the material – which might produce artefacts – and then air dried before staining (Mendelsohn and others 2006). Large lumps of wax should be removed and smeared onto another slide to leave an even layer of material for staining on the primary slides. These additional slides can then be used if necessary. Heat fixation is not necessary (Toma and others 2006) and open flames are often associated with increased risks in a practice laboratory.
Staining using modified rapid Romanowsky stains, such as Diff-Quik or Rapi-Diff is ideal. The three-step staining method of methanol, eosin and methylene blue can be used as for haematology samples when the sample is purulent and will adhere to the slide. When the sample contains a large amount of lipid this method is less suitable as large amounts of the sample may be lost. Using just the methylene blue step provides adequate staining for diagnosis in most cases and preserves more of the sample (Toma and others 2006). Using a staining rack or draining board, one or all of the stains can be used with less loss of material and the added advantage that pus and wax debris are not left in the stain pots. In all cases, the rapid preparation of these materials allows patient-side testing and can direct the choice of treatment for most clinical effect and to effect good antimicrobial stewardship (Beco and others 2013).
Cytology slides should be reviewed at a low power and then the clinician should focus on areas of interest under higher power (Fig 3). To be confident that bacteria are present, view the slide using the x 100 objective and immersion oil. In very thick smears, cellular detail and organisms cannot be seen and the edges of the material may be most useful. Having a microscope of adequate quality that has been serviced and cleaned, as well as knowing how to obtain optimal images is essential. A comfortable seat and suitable bench for the microscope make the whole process easier. Figs 3 and 4 show how to set up a modern microscope. Fig 5 shows the condenser diaphragm (a control that is often not used correctly) may prevent micro-organisms from being seen, even in excellent samples.
In normal ears, squames, anuclear, large flattened keratinocytes should predominate (see Figs 6 and 7). There will be small numbers of Malassezia and cocci as discussed previously. When inflammation becomes apparent – seen most often as erythema and then later as swelling and ulceration – neutrophilic exudation occurs, and when these cells are damaged the discharge becomes sticky, forming typical pus. Neutrophils may appear intact or show signs of degeneration. In highly purulent samples, only nuclear streaming may remain. When nuclear streaming or ghost cells are present, the observer should not be overconfident in deciding on the original cell type. When many poorly preserved cells, are seen the possibility of missing neoplasia remains. In some cases of hypersensitivity or parasitic disease, and in many forms of inflammation in the cat, eosinophils will be noted. In severe ulcerative disease, macrophages may also be present, although these are uncommon in samples from the ear.
In all otitis externa cases, micro-organisms may be seen extra- or intracellularly. When extracellular organisms are noted, they should be classified as normal or as pathogens. Typically, a few Malassezia (perhaps three or four per high-power field) and occasional bacterial cocci are noted on normal squames. When numbers increase, even before there is an inflammatory response typified by neutrophils, the numbers should be considered potentially pathological; however, cleaning rather than antimicrobial therapy might be most appropriate at this point. When organisms are seen within cells, whether eosinophils, neutrophils or macrophages, a pathological infection is present. Typically, a progression from normality to severe pathology would comprise several stages in which a Malassezia otitis develops into a mixed infection, followed by the introduction of rods, and, ultimately, Pseudomonas otitis would remain. The cytology findings of these stages are shown in Figs 6 to ⇑⇓⇓10.
Other findings on cytology include artefacts such as cotton strands from swabs or cleaning materials, and, very commonly, oil droplets from topically applied treatments. These can make cytology very difficult and reduce the power of detection considerably. Stain debris can also cause confusion (Fig 11), as can organisms contaminating stains (Duffield and others 2015). Stain debris itself can be avoided by regularly changing the methylene blue stain and also, when used infrequently, by filtering the stain through filter paper. It is important to clean the staining pot/Coplin jar when stain has been deposited on the walls as this encourages further stain deposition.
Culture and susceptibility testing
Cytology and clinical information is used to decide whether culture and susceptibility testing is indicated. Sampling considerations for cytology are equally appropriate for culture. Fine-tipped swabs are easier to use through otoscope cones. Swabs for culture and susceptibility should be sent in charcoal transport medium.
Common reasons to consider culturing are shown in Box 1. Whereas cytology gives a clear numerical evaluation of the organism present, bacteriology allows a more or less precise identification of the species involved and of the antimicrobial susceptibility for a range of antimicrobials chosen by the laboratory to be suitable.
Indications for culture and susceptibility testing in otitis
▪ Rods and/or cocci seen. The most suitable antibiotic treatments can only be chosen if the organisms are known.
▪ Marked purulent discharge without organisms being noted. Possibility of finding a pathogen that is relevant, but also may grow organisms that are irrelevant clinically.
▪ Pyogranulomatous inflammation. Organisms are often difficult to see with cytology in this uncommon ear inflammation, so culture is obligatory.
▪ In the event of treatment failure regardless of the organisms found, culture should help direct secondary and further treatments.
▪ If there is a suspicion of meticillin-resistant Staphylococcus species (MRS). The dog or cat has previously suffered MRS or has been in-contact with MRS ‘carriers’ or MRS is common in the hospital environment.
▪ If considering video otoscopy or ear flush for diagnosis or treatment in a bacterial otitis. In the event of an adverse event following these procedures, systemic antimicrobials may be required.
The breakpoints at which organisms are determined to be either susceptible or resistant to antimicrobials are determined based on the use of oral and other forms of systemic antimicrobials. Diagnostic laboratories acknowledge this and often make the comment that antimicrobials should be chosen on the basis of clinical knowledge. In fact, both laboratory and clinical knowledge is needed to decide on the best antibacterial agent. Factors such as whether the organism identified can ever be susceptible to the agent, whether a particular agent is denatured by pus or affected by acid cleaners or other topical agents need to be applied in combination with the laboratory results. These will direct the clinician to the correct agent even when the organism is not susceptible to the agent at the systemic minimum inhibitory concentration (MIC). In this context, it should be remembered that ear treatments fail for a variety of reasons, including choice of antibacterial agent, compliance and the need for adequate cleaning in many cases.
In an ideal world, culture would be performed in all cases, but in those presenting for the first time, and where rods are not present pre-treatment, cultures can be obtained retrospectively by putting the swab in the fridge until the recheck after seven days. This technique can also be used when the cytology cannot be appraised immediately as charcoal transport medium bacteriology swabs are inexpensive.
Bacteria and yeast infections are a major component of otitis externa. Observation of clinical signs and a client's history will give much information, but cytological examination with, if necessary, culture and susceptibility testing, is crucial for quality care.
Competing interests The author has in the past 10 years, been employed by Novartis Animal Health, and consulted for, lectured and advised Elanco, Vetruus, Zoetis, Dechra, Merial, Fort Dodge, ALK and Schering Plough – in all cases as an independent veterinary dermatologist.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.