The Effect of Chronic Otitis Externa-Media on Brainstem Auditory Evoked Potentials in Dogs

Besaltı O., Y.S. Sırın, Z. Pekcan: The Effect of Chronic Otitis Externa-Media on Brainstem Auditory Evoked Potentials in Dogs. Acta Vet. Brno 2008, 77: 615-624. The objective of the study was to present normative Brainstem Auditory Evoked Potentials (BAEP) data elicited by both air-conducted clicks (30–100 dBHL) and bone-conducted clicks (100–130 dBHL) for healthy dogs, and dogs with chronic otitis externa-media (COEM). The data were analyzed to acquire reference values in 55 normal dogs and to estimate the degree of hearing impairment associated with the disease. Fifty-four dogs with COEM were divided into two degrees of severity. The first one included dogs with severe (n = 16), and the second with moderate COEM (n = 38). The recognisability wave V was found higher than the others at every stimulation level. Wave VII was not seen by the bone-conducted click at all stimulation levels, and also in cases with severe COEM by the air-conducted click. Recognisability of wave V was seen more than 50% at 60 dBHL in severe COEM. This ratio was seen at 30 dBHL in moderate COEM. In healthy dogs, the latencies of waves (I–VII) were compared between the left and the right side, and there were no significant differences between each side except for a shorter latency of wave VI. In conclusion, COEM caused increasing latency rather than total deafness, and BAEP can be suggested as an ancillary diagnostic tool in ear disorders. Ear disease, hearing loss, BAEP, latency, canine Chronic otitis externa is characterized by severe inflammation, with irreversible hyperplasic changes of the ear canal and the surrounding tissues, and mineralization (with subsequent ossification) of the auricular and annular cartilage (Bellah 1997). Chronic inflammation and accumulation of exudates within the canal eventually lead to the disruption of the tympanic membrane. The infection invades the middle ear (otitis media), and may eventually extend to the inner ear (otitis interna), and sometimes the facial nerve is involved (facial paralysis) (Stickney et al. 2004). Although profound hearing loss is most commonly related to neurological causes, chronic otitis externa-media (COEM) can also cause a partial hearing loss (Eger and Lindsay 1997; Krahwinkel et al. 1993). It was demonstrated that total ear canal ablation combined with bulla osteotomy, which is the standard approach to the end stage of COEM, may be the cause of a significant loss of hearing (McAnulty et al. 1995). Deafness can be classified into conductive or sensory neural deafness according to the involved anatomical structures. Conductive hearing loss occurs with the occlusion of the ear canal, thickening or disruption of the tympanic membrane, replacement of the air within the middle ear with fluid, stiffening or damage to the ossicular chain, and loss of compliance of the vestibular or round windows (Eger and Lindsay 1997). Brainstem Auditory Evoked Potential (BAEP) is an accurate, repeatable and noninvasive method in evaluation of the deafness in dogs and cats. It provides valuable information in the treatment protocol and prognosis, and also differentiates conductive ACTA VET. BRNO 2008, 77: 615-624; doi:10.2754/avb200877040615 Address for correspondence: O. Besaltı, DVM, PhD Ankara University Faculty of Veterinary Medicine Department of Surgery, Diskapi, Ankara, TURKEY Phone: +90 535 264 0546 Fax: +90 312 316 4472 E-mail: besalti@hotmail.com http://www.vfu.cz/acta-vet/actavet.htm and sensory neural deafness (Sims 1988). This test can be carried out by bone-conducted clicks and air-conducted clicks. By comparing the results of air and bone conduction BAEP tests, an audiologist can determine whether the hearing loss is sensory, neural, conductive or mixed. Brainstem auditory evoked potentials also provide information about which part of the anatomical structures related to hearing is affected (Cox 2002; Sims 1988). The aim of this study was to present the effects of COEM on hearing, and provide standard values for BAEP on physiological recordings. Materials and Methods Fifty-five dogs of different age, sex and breed without any history or clinical signs of ear disorders, but anaesthetized for the aim of different elective surgical procedures were subjected to recordings for obtaining physiological values for the study. The cases with COEM that were diagnosed by otoscopic and radiographic examination were included prospectively. Fifty-four dogs met the criteria to be included in the study. Thirty-eight dogs had moderate COEM (stenotic canal and perforated, bulged, cloudy or opaque appearance in colour, and/ or thickened tympanic membrane) and 16 dogs had end stage COEM (ear canal narrowing or total obliteration, unable to visualize tympanic membrane). BAER were recorded under general anaesthesia which was induced using 0.05 mg/kg atropine sulphate SC (Biofarma, Turkey), 2 mg·kg-1 xylazine hydrochlorure IM, (Alfazyne, Izmir, Turkey) and 15 mg·kg-1 ketamine hydrochloride IM (Alfamine, Izmir, Turkey) combination. The study was approved by the Ethics Commitee of the Faculty of Veterinary Medicine, Ankara University. Recording BAER was recorded by Medelec Synergy 5 channel EMG/ EP device (Oxford Instrument-UK). Headset and ear tips for air-conducted clicks (Oxford, S1021) and bone conductor for bone-conducted clicks (Oxford, 017C080) were used for all cases except for dogs with a totally obliterated ear canal in which only bone conductor was used. Ear tips were inserted into both ear canals, and bone conductor was placed against the mastoid process of the temporal bone for delivering the stimulus. Subdermal stainless steel monopolar needle electrodes (Oxford, 017K103) were placed at the vertex (reference) and just rostral (active) to the base of each ear and the ground electrode was placed over the frontal sinus along the midline (Plate I, Fig.1). The impedance of the electrodes was checked to ensure that it was less than 2 kilo ohm. While stimulating one ear, the other was masked by white noise for the purpose of recording each ear individually. Impulses generated in the first 10 milliseconds, and each click was filtered (100 Hz-3 KHz). The tested ear was stimulated at 100 dBHL to 30 dBHL in air-conducted clicks and at 130 dBHL to 100 dBHL in bone-conducted clicks by the decreasing (decrement of) 10 dBHL for each season at a click rate of 11 pps. At each level an average of 1500 clicks/stimuli were used. Identification of the waves was carried out as reported before (Sanders and Bagley 2003; Sims 1988). Briefly, wave I is the first and wave V has a large amplitude and characteristic negative trough following the positive peak. Wave III and IV are seen either merged with or separate from each other and their amplitudes are smaller. The other waves were estimated in turn, and latencies were measured from the positive peak point of the wave. Statistical Analysis Latency values were compared with one-way ANOVA for healthy, severe COEM and moderate COEM cases. The latencies were analyzed in all three groups with Tukey test. The differences between right and left sides were compared using paired t-test. Results Brain stem auditory evoked responses were recorded from 110 normal ears of 55 dogs. Ages ranged from 1 to 11 years (mean 4.22). The breeds included were cross-breed (n = 30), Poodle (n = 6), German Shepherd (n = 4), Turkish Anatolian Sheep Dog (n = 4), Cocker Spaniel (n = 4), Labrador (n = 3), Rottweiler (n = 2), Siberian Husky (n = 1) and Pointer (n = 1). Moderate COEM was diagnosed in 60 ears of 38 dogs, and ages ranged from 2 to 14 years (mean 7.54). Twenty-two dogs had bilateral and 16 dogs had monolateral moderate COEM. The contralateral ears had mild inflammation or were healthy. The breeds were Cocker Spaniel (n = 11), Poodle (n = 8), German Shepherd (n = 4), cross breed (n = 4), Boxer (n = 3), Dalmatian (n = 1), Labrador (n = 3), Golden Retriever (n = 1), Turkish Anatolian Sheep Dog (n = 1), Pointer (n = 1) and Pit Bull Terrier (n = 1). Severe COEM was diagnosed in 28 ears of 16 dogs, and mean ages were 11.6 years old. Twelve dogs had bilateral and 4 had monolateral severe COEM. Breed dispersion was Cocker Spaniel (n = 8), Poodle (n = 6), Labrador (n = 1) and Pointer (n = 1). Facial 616

Chronic otitis externa is characterized by severe inflammation, with irreversible hyperplasic changes of the ear canal and the surrounding tissues, and mineralization (with subsequent ossification) of the auricular and annular cartilage (Bellah 1997).Chronic inflammation and accumulation of exudates within the canal eventually lead to the disruption of the tympanic membrane.The infection invades the middle ear (otitis media), and may eventually extend to the inner ear (otitis interna), and sometimes the facial nerve is involved (facial paralysis) (Stickney et al. 2004).Although profound hearing loss is most commonly related to neurological causes, chronic otitis externa-media (COEM) can also cause a partial hearing loss (Eger and Lindsay 1997;Krahwinkel et al. 1993).It was demonstrated that total ear canal ablation combined with bulla osteotomy, which is the standard approach to the end stage of COEM, may be the cause of a significant loss of hearing (McAnulty et al. 1995).
Deafness can be classified into conductive or sensory neural deafness according to the involved anatomical structures.Conductive hearing loss occurs with the occlusion of the ear canal, thickening or disruption of the tympanic membrane, replacement of the air within the middle ear with fluid, stiffening or damage to the ossicular chain, and loss of compliance of the vestibular or round windows (Eger and Lindsay 1997).
Brainstem Auditory Evoked Potential (BAEP) is an accurate, repeatable and noninvasive method in evaluation of the deafness in dogs and cats.It provides valuable information in the treatment protocol and prognosis, and also differentiates conductive and sensory neural deafness (Sims 1988).This test can be carried out by bone-conducted clicks and air-conducted clicks.By comparing the results of air and bone conduction BAEP tests, an audiologist can determine whether the hearing loss is sensory, neural, conductive or mixed.Brainstem auditory evoked potentials also provide information about which part of the anatomical structures related to hearing is affected (Cox 2002;Sims 1988).
The aim of this study was to present the effects of COEM on hearing, and provide standard values for BAEP on physiological recordings.

Materials and Methods
Fifty-five dogs of different age, sex and breed without any history or clinical signs of ear disorders, but anaesthetized for the aim of different elective surgical procedures were subjected to recordings for obtaining physiological values for the study.The cases with COEM that were diagnosed by otoscopic and radiographic examination were included prospectively.Fifty-four dogs met the criteria to be included in the study.Thirty-eight dogs had moderate COEM (stenotic canal and perforated, bulged, cloudy or opaque appearance in colour, and/ or thickened tympanic membrane) and 16 dogs had end stage COEM (ear canal narrowing or total obliteration, unable to visualize tympanic membrane).BAER were recorded under general anaesthesia which was induced using 0.05 mg/kg atropine sulphate SC (Biofarma, Turkey), 2 mg•kg -1 xylazine hydrochlorure IM, (Alfazyne, Izmir, Turkey) and 15 mg•kg -1 ketamine hydrochloride IM (Alfamine, Izmir, Turkey) combination.The study was approved by the Ethics Commitee of the Faculty of Veterinary Medicine, Ankara University.

Recording
BAER was recorded by Medelec Synergy 5 channel EMG/ EP device (Oxford Instrument-UK).Headset and ear tips for air-conducted clicks (Oxford, S1021) and bone conductor for bone-conducted clicks (Oxford, 017C080) were used for all cases except for dogs with a totally obliterated ear canal in which only bone conductor was used.Ear tips were inserted into both ear canals, and bone conductor was placed against the mastoid process of the temporal bone for delivering the stimulus.Subdermal stainless steel monopolar needle electrodes (Oxford, 017K103) were placed at the vertex (reference) and just rostral (active) to the base of each ear and the ground electrode was placed over the frontal sinus along the midline (Plate I, Fig. 1).The impedance of the electrodes was checked to ensure that it was less than 2 kilo ohm.While stimulating one ear, the other was masked by white noise for the purpose of recording each ear individually.Impulses generated in the first 10 milliseconds, and each click was filtered (100 Hz-3 KHz).The tested ear was stimulated at 100 dBHL to 30 dBHL in air-conducted clicks and at 130 dBHL to 100 dBHL in bone-conducted clicks by the decreasing (decrement of) 10 dBHL for each season at a click rate of 11 pps.At each level an average of 1500 clicks/stimuli were used.
Identification of the waves was carried out as reported before (Sanders and Bagley 2003;Sims 1988).Briefly, wave I is the first and wave V has a large amplitude and characteristic negative trough following the positive peak.Wave III and IV are seen either merged with or separate from each other and their amplitudes are smaller.The other waves were estimated in turn, and latencies were measured from the positive peak point of the wave.

Statistical Analysis
Latency values were compared with one-way ANOVA for healthy, severe COEM and moderate COEM cases.The latencies were analyzed in all three groups with Tukey test.The differences between right and left sides were compared using paired t-test.
Severe COEM was diagnosed in 28 ears of 16 dogs, and mean ages were 11.6 years old.Twelve dogs had bilateral and 4 had monolateral severe COEM.Breed dispersion was Cocker Spaniel (n = 8), Poodle (n = 6), Labrador (n = 1) and Pointer (n = 1).Facial paralysis was diagnosed beside severe COEM in three dogs ipsilaterally and in one dog bilaterally by neurological examination and needle EMG of the involved facial muscle.
The BAEP appears as a series of I-VII vertex positive peaks (Plate I, Fig. 2).The recognisability ratio of the waves at different stimulation levels is illustrated graphically in Figs 3a-f, 4a-f).

Healthy dogs
By increasing the signal intensity, the percentage recognisability of wave I, II and III increased.Wave IV appeared lower than 50% at all stimulation intensity levels.In contrast, wave V recognised at the highest amplitude at all stimulation intensity levels.Waves VI and VII were seen in a higher percentage between 50-90 dBHL.The latency values for both air-conducted clicks and bone-conducted clicks are presented in Tables 1 and 2. Significant difference at 70 dBHL of air-conducted click records was seen as: The latency of wave I and II, III in healthy ears was smaller than in moderate and/or severe COEM.The latency of wave V (on the left side) and VI (on the right side) in healthy ears was smaller than in severe COEM.When the latency values were compared regarding the left or right side, there were no significant differences except for wave VI which was smaller on the left side than on the right.
According to this classification there was no hearing problem in 55 normal dogs in which BAER was recognized at 30 dBHL.In addition, r e c o g n i s a b i l i t y of wave V at all stimulus intensities in a higher degree means that it can be used for determining the threshold, as reported before (Cox 2002;Sims 1988).In this study, severe COEM cases had moderate hearing problems.In these cases the recognisability ratio of wave V with more than 50% was seen at 60 dBHL intensity.In moderate COEM there was a non-significant hearing loss, from which it was deducted that the recognisability rate was more than 50% at 30 dBHL.
The presence of wave VII whose generator is uncertain in healthy dogs and its absence in severe COEM cases was found interesting.As mentioned before for these cases, having higher hearing thresholds and the absence of wave VII seem to be related.However, before making a clear conclusion, more cases with severe COEM should be investigated.Another interesting finding related to wave VII was its absence at all stimulus intensities in boneconducted click records.Another conclusion for wave VII can be deducted as conductive structure related to the hearing having a role in the generation of this wave.Since the generator of wave VII is an auditory reflexion (Sims 1988(Sims , 2003)), in severe COEM cases this reflexion is prevented.
The latency of all waves increased with the severity of otitis, although the prolongation of times was only significant in advanced cases (Eger and Lindsay 1997).In BAER evoked by air-conducted clicks, latency of wave I increased in both moderate and severe COEM cases compared to healthy dogs in our study.For wave II on the left side, and for wave III on both sides, for wave V which was recognized at approximately all stimulus intensities, the latency was increased in both moderate and severe COEM compared to healthy dogs.These findings are similar to Eger and Lindsay's report (1997).
The sound waves from the external ear canal are transmitted through the tympanic membrane through the ossicles, which articulate to amplify sound and transmit vibrations to the fluid-filled inner ear.On the surface of the organ of Corti, the inner and outer hair cells lie that connect with the spiral ganglion for transmission to the cochlear nerve (Cook 2004).There is a general agreement that wave I of the BAEP is generated by acoustic nerve activity (Sims 1988).Significant differences of latency, which was lower in severe COEM than normal wave I at 110 dBHL of bone-conducted click can be explained as follows: In severe COEM cases ear canals are mostly obstructed due to chronic inflammation, and sound waves can not be transmitted through the external ear canal.By the bone-conducted click ear canal, tympanic membrane, and perhaps the ossicles vibrations are bypassed, and the receptors in the organ of Corti detect this vibration and the acoustic nerve is stimulated by this impulse.This is the possible cause of the decreased latency of wave I of BAEP in severe COEM.
In a study of BAEP in 58 healthy dogs, 4-5 vertex positive waves were explained and latency values at 60 dBHL were I: 1.49 ± 0.13 ms; II: 2.32 ± 0.14 ms; III: 3.01 ± 0.25 ms; IV: 4.22 ± 0.27 ms and V: 5.55 ± 0.37 ms (Bodenhamer et al. 1985).In our study 7 vertex positive waves were identified at different recognisability rates, and also different latency values.The differences are related to the different explanation of wave V.
In conclusion, reference values for both latency and recognisability rate for BAEP in healthy dogs were obtained.Chronic otitis externa-media causes an increase in latency rather than total deafness.Even in severe COEM, dogs have the ability to hear, but to a lesser extent than healthy ones.BAER may be suggested as an ancillary diagnostic tool in ear disorders.Sensory neural deafness associated with COEM can also be differentiated by this test.

Fig. 3 .
Fig. 3.The recognisability of wave I (a, b), V (c, d) and VII (e, f) recorded by air conducted clicks at different stimulus intensities in normal (N), moderate (M) and severe (S) COEM.

Fig. 3 .f
Fig. 3.The recognisability of wave I (a, b), V (c, d) and VII (e, f) recorded by air conducted clicks at different stimulus intensities in normal (N), moderate (M) and severe (S) COEM.
Fig. 1.Configuration of recording procedures for BAEP

Table 1 .
Summary of latency data from air conducted clicks (±SD)

Table 2 .
Summary of latency data from bone conducted clicks (±SD)