Congenital sensorineural deafness
The earliest studies of deafness in animals were in the Dalmatian in the last century (Rawitz, 1896); most studies have been performed with Dalmatians or white cats. Deafness does not develop in dogs and cats until the first few weeks of life, with normal functional development occurring to that point (Pujol & Hilding, 1973). Studies in our laboratory have shown that Dalmatians do not go deaf until weeks 3-4 after birth. The histologic pattern that occurs in most dog breeds and white cats is known as cochleo-saccular, or Scheibe, type of end organ degeneration. The deafness results from initial degeneration of the stria vascularis, followed by collapse of Reissner's membrane and the cochlear duct, degeneration of the hair cells of the organ of Corti, and collapse of the saccule. Secondary loss of spiral ganglion cells is also seen at later stages (Anderson et al., 1968; Bosher & Hallpike, 1965; Hudson & Ruben, 1962; Igarashi et al., 1972; Johnsson et al., 1973; Lurie, 1948; Mair, 1973, 1976; Suga & Hattler, 1970). The cochlear hair cell and spiral ganglion cell loss is permanent in mammals. Histologic studies of deaf Dalmatians have shown that the degeneration begins as early as one day after birth, and is clearly evident histologically by four weeks (Johnsson et al., 1973). Degeneration begins in the middle coil of the cochlea, followed by the basal then apical coils (Anderson et al., 1968). The cause of the strial degeneration is not known, but there is an observed absence of melanocytes in the strial tissue of many deaf animals (Savin, 1965; Steel et al., 1987). The function of melanocytes in normal stria is not known but appears to tie in with hair pigment associations with deafness (Steel & Barkway, 1989; Carlisle et al., 1990)). Most melanocytes originate in the neural crest (Weston, 1969), so the absence of strial melanocytes could reflect either a failure of migration from the neural crest or a failure of differentiation after arrival. In the Doberman, and probably other dog breeds not carrying the merle or piebald pigment genes, the deafness results from direct loss of cochlear hair cells without any antecedent effects on the stria vascularis (Wilkes & Palmer, 1992).
It has been shown that the auditory cortex of deaf Dalmatians is grossly reduced in size (Ferrara & Halnan, 1983), leading the authors to the suggestion that the origin of deafness in the breed was central rather than peripheral. Although not reported, it is likely that other CNS structures in the auditory pathway were also smaller than in hearing animals. However, it is well known from classical studies that kittens whose eyelids were kept sealed after birth failed to develop normal CNS visual structures, demonstrating that normal sensory input is necessary for the full development and maintenance of these structures (Hubel et al, 1977). As a result, the findings in the Dalmatian are undoubtedly a reflection of a similar pathophysiological process. These CNS changes in deaf dogs have been used to justify euthanasia on the basis of having an "abnormal" brain, but neurologically the brain function of deaf animals is normal except for the loss of auditory function.
Conductive deafness may result from developmental defects affecting the ossicles, such as fusion, from failure of the ear canal to completely open after birth, or otosclerosis, but these events have not been documented in dogs or cats and are probably rare. Congenital tympanic membrane absence occasionally occurs, but does not produce deafness. Conductive deafness is most often a result of chronic otitis externa and media, where stenosis and eventual occlusion of the external canal results, or impaction from excess cerumen accumulation. Chronic otitis externa may ultimately result in mineralization and ossification of the external ear canal, requiring lateral ear resection (Elkins et al., 1981) or other remedies. However, hearing function can be maintained, or even regained, after procedures as extreme as total ear canal ablation with lateral bulla osteotomy (Payne et al., 1989; Krahwinkel et al., 1993).
Ototoxic agents may cause hearing loss or deafness by direct effects on cochlear and/or vestibular hair cells, or may cause damage to the stria vascularis with secondary hair cell loss (Miller, 1985). Ototoxicity in humans is frequently accompanied by tinnitus, a high pitched ringing in the ears. Ototoxicity in dogs and cats may likewise be accompanied by behavior suggesting the presence of similar sensory phenomena. Over 180 compounds and classes of compounds have been identified as ototoxic (Govaerts et al., 1990; Mansfield, 1990; Pickrell et al., 1993). Many of those most likely to be seen in veterinary practice are listed in Table II; it must be noted that not all are equally toxic. In some cases the ototoxic effects are reversible if caught early, such as with salicylates, but in most instances the deficit is permanent by the time of detection. The best recognized, and perhaps most frequent, agents of ototoxicity are the aminoglycoside antibiotics, especially gentamicin (Govaerts et al., 1990). Drugs within this group are also nephrotoxic, and vary in their toxicity to the auditory, vestibular, and renal systems. Gentamicin and streptomycin are most toxic to the vestibular system, while neomycin, kanamycin, tobramycin, and amikacin are most toxic to the cochlea; netilmicin is thought to be least toxic (Govaerts et al., 1990). However, since gentamicin and neomycin are the most used aminoglycosides in veterinary practice, especially as topical otic agents, they are the drugs most likely to produce cochlear ototoxicity in companion animals.
The mechanism of toxicity of aminoglycosides is unclear, but the pathology includes a progression from basal coil outer hair cells, to more apical outer hair cells, followed by inner hair cells; strial changes are concurrent with, or precede outer hair cell changes (Govaerts et al, 1990). Although the effect is often ascribed to concentration of the drug in the perilymph, the most current evidence points to binding of the drug to glycosaminoglycans of the stria vascularis and disruption of phosphoinositide metabolism (Govaerts et al., 1990). It has been reported that serum gentamicin levels must exceed a 2 µg/ml threshold level for over 10 days to produce toxicity (Sande & Mandell, 1990). Early ototoxic effects of gentamicin may be reversible by calcium administration (Pickrell et al., 1993).
Route of administration may affect ototoxicity, with systemic exposure providing better access of drugs to the cochlea than topical administration in ears with intact tympanic membranes. However, tympanum rupture frequently accompanies otitis externa, increasing access of drugs to the oval and round windows of the cochlea, through which absorption occurs. As a result, care must be exercised in topical drug application when visualization of the tympanic membrane is not possible, and when possible it is advisable to monitor hearing function with BAER recordings when high concentrations and long treatment courses of gentamicin or similar agents are employed. Attempts by us to produce ototoxicity in dogs with both intact and ruptured tympanic membranes, using typical clinical treatment protocols and assessing toxicity with BAER recordings, were unsuccessful for both gentamicin (Strain et al., 1995) and chlorhexidine (Merchant et al., 1993). This suggests, but does not guarantee, that topical application of drugs at recommended levels can generally be assumed to be safe. Age, concurrent infection, anesthesia, or pre-existing cochlear damage may potentiate drug ototoxicity, and repeated courses of antibiotic treatment may produce cumulative effects that are initially clinically inapparent.
Presbycusis is the decline in hearing associated with various types of auditory system dysfunction that accompany aging, and cannot be accounted for by ototraumatic, genetic, or pathological conditions (Schukneckt, 1955; Willott, 1991). Presbycusis can be classified into four types of pathology: sensory, neural, strial, and cochlear conductive (Schuknecht & Gacek, 1993). The pathologic change in most dogs and cats appears to be sensorineural, although decreased tympanum and ossicle joint articulation flexibility can potentially contribute. Presbycusis is common in geriatric dogs (Knowles et al., 1988, 1989), but prevalence rates or other related data are not available. Although it is a progressive disorder, owners usually report an acute onset because of the ability of the animal to compensate for hearing loss until nearly complete deafness occurs. Hearing aids have been successfully utilized in dogs with some residual auditory function, but not all dogs will tolerate the presence of the ear plug (Marshall, 1990). The primary determinant of the success of hearing aids is the ability of the owner to train the animal to accept the presence of a foreign body in the ear canal. Because of this training requirement and the sensitivity of the cat to ear contact it is unlikely that these devices would be successful in cats. There is no known way of retarding the progression of the deafness. In humans, men are affected more at high frequencies, while women are affected more at low frequencies (Jerger et al., 1993), but it is not known if similar patterns hold in dogs or cats.
Noise-induced hearing loss or deafness can be temporary or permanent (Peterson, 1980). Temporary increases in hearing threshold occur after brief exposure to intense sounds (over 100 dB), with gradual recovery of function occurring over periods ranging from minutes to two weeks. Noise- induced hearing loss is thought to result from either disarrangement or breakage of hair cell cilia (Flottorp, 1990), but can also result from damage to the tympanum and ossicles. Continuous or repeated exposure to noise results in a progressive loss of hair cells and a corresponding deafness. In humans the greatest hearing loss is at the middle frequency range near 4000 Hz, but progresses to higher and lower frequencies with continued exposure (Peterson, 1980). Dogs used to hunt with firearms, like their human companions, may develop noise-induced hearing loss (personal observation).
Hearing loss or deafness may also result from anoxia, anesthesia, trauma, or infections, such as otitis interna and meningitis. On occasion, interactive effects may be expected to produce loss or deafness when the individual causes would have insufficient alone to produce an effect (Pickrell et al, 1993).
| Aminoglycoside antibiotics | Non-aminoglycoside antibiotics | Miscellaneous agents |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| Diuretics |
|
|
|
|
|
|
|
|
|
Antiseptics |
|
| Antineoplastic agents |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|