The following is the published abstract and the expanded poster content presented at the May, 1996 meeting of the American College of Veterinary Internal Medicine in San Antonio, Texas:
Strain, G.M. and B.L. Tedford. 1996. Deafness prevalence and associated phenotypic markers in dog breeds with high risk. Proc. 14th Ann. Vet. Med. Forum, ACVIM 14:772. San Antonio, May, 1996. Also: J. Vet. Internal Med. 10:190, 1996.
DEAFNESS PREVALENCE AND ASSOCIATED PHENOTYPIC MARKERS IN DOG BREEDS WITH HIGH RISK. G.M. Strain and B.L. Tedford, School of Veterinary Medicine, Louisiana State University, Baton Rouge, LA.
Hereditary congenital sensorineural deafness presents in dog breeds carrying the piebald or merle hair coat pigment genes; a small number of these breeds have a high prevalence of deafness (Strain, B Vet J 152(1), 1996). The purpose of this investigation was to document the prevalence in these breeds and identify phenotypic markers positively or negatively associated with deafness. Data were collected for Dalmatians (N=4,596), Bull Terriers (N=516), English Setters (N=394), English Cocker Spaniels (N=413), and Australian Cattle Dogs (N=72). Testing was performed at LSU, specialty and all-breed dog shows, and sponsored hearing clinics. Dalmatian data include those (N=1,031) reported in Strain et al (J Vet Int Med 6:175, 1992).
Prevalence data (% unilaterally deaf/% deaf) were: Dal - 21.96/7.99; BT - 10.08/ 0.97; ES - 13.45/2.79; ECS - 6.30/1.69; ACD - 8.33/4.17. No significant sex differences in deafness prevalence were seen. Differences between left-uni and right-uni were not seen. No significant coat color differences were seen in Dal (black, liver), ES (blue, orange, tri), ECS (11 colors), or ACD (4 colors). However, significant (p < 0.001, 2) differences were seen for BT: whites had greater deafness (16.85/1.80) than coloreds (2.03/0.00) (p < 0.001).
Dals with a patch had significantly less deafness (8.46/1.99)than those without (23.36/8.43) (p < 0.001); lack of definite criteria for a patch in the other breeds did not allow assessment. Dals with blue irises had significantly more deafness (32.62/18.24) than those with brown irises (20.99/6.67) (p < 0.001); no differences were seen between dogs with one (N=339) or two (N=125) blue eyes. Insufficient numbers of blue-eyed dogs were seen in other breeds for analysis. Dals missing pigment on the tapetum had greater deafness (p < 0.001). Dals with one or both parents uni or deaf (all combinations combined) had significantly more deafness (30.29/11.30) than Dals with bi/bi parents (21.26/5.91) (p < 0.001). Even dogs with bi/uni parents had greater deafness (28.84/11.38) than those from bi/bi parents (p < 0.001). Similar associations with parental hearing were seen for ECS, but parent hearing status data were insufficient for analysis in the other breeds.
Congenital sensorineural deafness has been reported or observed in a large number of dog breeds (Table 1). For nearly all of these breeds the deafness is pigment-associated, relating to the genes (piebald, extreme piebald, merle) responsible for white pigment in the hair coat. Deafness may affect one or both ears, but partial hearing loss in an ear does not appear to occur in this disorder. Unilateral deafness is thought to reflect incomplete expression of the defect. This deafness is thought to have a hereditary component, since prevalence is higher in the offspring of unilaterally deaf dogs than of bilaterally hearing dogs. Current findings indicate that deafness in these dogs actually occurs approximately 3 weeks after birth, when cochlear hair cells die secondary to degeneration of the stria vascularis in the cochlear duct. The strial degeneration is in turn thought to result from the absence of vascular melanocytes (see Strain 1996 for review).
In an effort to better understand the relationship between congenital sensorineural deafness and phenotypic traits we have collected data on deafness prevalence, pigmentation markers, and gender from dogs of breeds with a high prevalence of deafness while documenting the hearing status for owners and breeders. Statistical correlations between hearing status, gender, and pigmentation markers were then evaluated. This report supplements and expands upon a previous study of Dalmatians (Strain et al, 1992). Data are reported here for Dalmatian, Bull Terrier, English Setter, English Cocker Spaniel, and Australian Cattle Dog breeds; insufficient data was available for other affected breeds for analysis.
Data are reported for Dalmatians (N=4,596), Bull Terriers (N=525), English Setters (N=394), English Cocker Spaniels (N=420), and Australian Cattle Dogs (N=72). Numbers of tested dogs from other affected breeds were of insufficient size for analysis. Brainstem auditory evoked potential testing was by accepted techniques, taking into consideration animal age (Strain et al, 1991) and utilizing bone stimulation where appropriate (Strain et al, 1993). Testing was performed at LSU, at specialty and all-breed dog shows, and at sponsored hearing clinics. Data collected for each dog are shown in Table 2. Tapetal pigmentation presence was evaluated by ophthalmoscope; pigmentation was considered to be present if any pigmentation was present. Deafness in adult animals was always considered to be of the hereditary form even if owner experience suggested that hearing had been present at some earlier date. Sire and dam hearing status data were accepted only when BAEP testing was known to have been performed. Data on the presence of a patch in the hair coat was only evaluated for Dalmatians due to inconsistent definitions for other breeds.
Statistical analysis of data was performed with the Statistical Analysis System's (SAS) FREQ procedure with the CHISQ option (chi-square test of homogeneity - based on the likelihood ratio chi- square). Associations between observed hearing status and individual phenotypic markers (i.e., coat color, iris color, etc.) were compared to their expected values. The hypothesis of no general association was rejected if the chi-square test indicated significance at a probability level p < 0.05, and general association was concluded. Results are reported as significantly associated, rather than significantly correlated, since the chi-square statistic was used instead of a correlation coefficient to determine significance.
| Gender: | M | F | ||
| Coat color: | Dalmatian: | black | liver | |
| Bull Terrier: | white | colored | ||
| English Setter: | blue | orange | tri colored | |
| English Cocker Spaniel: | blue roan | orange roan | liver roan | |
| liver tri | solid black | solid red | ||
| solid golden | black & white | orange & white | ||
| blue, roan & tan | black, white & tan | solid black & tan | ||
| Australian Cattle Dog: | blue, black & tan | blue & tan | ||
| red | blue | |||
| Patch (Dal only): | present | absent | ||
| Iris color: | brown | blue | ||
| Tapetal color: | present | absent | ||
| Hearing status: | bilateral | unilateral | deaf | |
| Sire hearing status: | bilateral | unilateral | deaf | |
| Dam hearing status: | bilateral | unilateral | deaf |
1. Deafness prevalence data for the five breeds of this study are shown in Table 3 (updated from this presentation).
2. No significant sex differences in deafness prevalence were seen in any breed.
3. Among unilaterally deaf dogs, there were no significant differences in deafness prevalence between left and right ears.
4. No statistically significant coat color differences for deafness prevalence were seen for Dalmatian (black, liver), English Setter (blue, orange, tri colored), English Cocker Spaniel (11 colors), or Australian Cattle Dog (red, blue, blue & tan, blue, black & tan).
5. Statistically significant coat color differences for deafness prevalence were seen for Bull Terriers: whites had greater deafness (16.85% uni/1.80% deaf, N=279) than coloreds (2.03% uni/0.00% deaf, N=246) (p < 0.001, 2).
6. Dalmatians with a patch had significantly less deafness (8.46% uni/1.99% deaf, N=402) than those without (23.36% uni/8.43% deaf, N=4,152) (p < 0.001).
7. Dalmatians with blue irises had significantly more deafness (32.62% uni/ 18.24% deaf, N=466) than those with brown irises (20.99% uni/6.67% deaf, N=3,912) (p < 0.001); no differences were seen between dogs with one (N=339) or two (N=125) blue eyes. Insufficient numbers of blue-eyed dogs were seen in other breeds for analysis.
8. Dalmatians missing pigment on the tapetum of one or both eyes had greater deafness (30.62% uni/14.87% deaf, N=) than those with pigmented tapeta (21.79% uni/7.16% deaf, N=2,487) (p < 0.001).
9. Dalmatians with one or both parents uni or deaf (all combinations combined) had significantly more deafness (30.29% uni/11.30% deaf, N=690) than Dalmatians with bi/bi parents (21.26% uni/5.91% deaf, N=2,131) (p < 0.001). Even dogs with bi/uni parents had greater deafness (28.84% uni/11.38% deaf, N=756) than those from bi/bi parents (p < 0.001). Similar associations with parental hearing were seen for English Cocker Spaniels, but parent hearing status data were insufficient for analysis in the other breeds.
1. Deafness prevalence in the Dalmatian is approximately twice that seen in other affected breeds where reliable data are available.
2. No deafness prevalence differences by sex or ear were seen.
3. No coat color differences in deafness prevalence within breeds were seen except for the Bull Terrier, where whites had greater deafness than coloreds.
4. Pigmentation traits in Dalmatians were consistent with an association between strong expression of the extreme piebald gene and deafness: dogs with blue irises or missing tapetal pigmentation were more likely to be deaf (piebald suppression of eye pigmentation), while dogs with a patch were less likely to be deaf (failure of piebald to suppress the underlying coat color).
5. The increased prevalence of deafness in dogs with at least one deaf ear among its parents compared to dogs from parents with four good ears supports a hereditary factor in pigment-associated congenital sensorineural deafness, consistent with similar syndromes in humans, cats, mice, mink, and other species.
6. Pigment-associated congenital sensorineural deafness continues to be a significant disorder in dog breeds with the piebald, extreme piebald, and merle genes. Reduction in prevalence can currently only result from selective breeding away from affected animals, including those with unilateral deafness, and animals with a history of high numbers of affected offspring. Hearing registries, such as one recently established by the English Setter Association of America and one proposed by the Dalmatian Club of America, will provide the information base for progress in this area.
Supported by grants from the National Institute on Deafness and Other Communication Disorders, National Institutes of Health; the American Kennel Club; the Dalmatian Clubs of America; and by private donors.
Strain, G.M., B.L. Tedford, and R.M. Jackson. 1991. Postnatal development of the brainstem auditory-evoked potential in dogs. Am. J. Vet. Res. 52:410-415.
Strain, G.M., M.T. Kearney, I.J. Gignac, D.C. Levesque, H.J. Nelson, B.L. Tedford, and L.G. Remsen. 1992. Brainstem auditory evoked potential assessment of congenital deafness in Dalmatians: associations with phenotypic markers. J. Vet. Internal Med. 6: 175-182.
Strain, G.M., K.D. Green, A.C. Twedt, and B.L. Tedford. 1993. Brain stem auditory evoked potentials from bone stimulation in dogs. Am. J. Vet. Res. 54: 1817-1821.
Strain, G. M. 1996. Aetiology, prevalence, and diagnosis of deafness in dogs and cats. Brit. Vet. J. 152:17-36.