- Email: [email protected]
Histopathological morphometric study of cochleosaccular dysplasia in Dalmatian dogs
International Journal of Pediatric Otorhinolaryngology 74 (2010) 934–938
Contents lists available at ScienceDirect
International Journal of Pediatric Otorhinolaryngology journal homepage: www.elsevier.com/locate/ijporl
Histopathological morphometric study of cochleosaccular dysplasia in Dalmatian dogs§,§§ Andre L.L. Sampaio b,c,*, Elizabeth Paine c, Patricia A. Schachern a,c, Carolyn Sutherland a, Sebahattin Cureoglu a,c, Carlos A.C.P. Olivieira b, Michael M. Paparella a,c,d a
Department of Otolaryngology, Otitis Media Research Center, University of Minnesota, Minneapolis, MN, United States Brasilia University Medical School - Capes, Fulbright Scholarship, Brasilia, Brazil c International Hearing Foundation, Minneapolis, MN, United States d Paparella Ear Head and Neck Institute, Minneapolis, MN, United States b
A R T I C L E I N F O
A B S T R A C T
Article history: Received 14 February 2010 Received in revised form 17 May 2010 Accepted 18 May 2010 Available online 12 June 2010
Objective: To analyze temporal bones of deaf Dalmatian dogs from 5 days after birth to adulthood to better understand the pathogenesis of cochleosaccular dysplasia. Methods: This is an experimental animal histopathological temporal bone study that included two groups of temporal bones. Group I consisted of 41 temporal bones from deaf Dalmatian dogs and group II of 25 temporal bones from 15 ‘‘normal’’ aged-matched, hearing Black Labradors. Morphometric analysis included: stria vascularis and spiral ligament area measurements, and cell counts of spiral ganglion, Scarpa’s ganglion, and hair cells of saccular macula. Results: The following findings were significantly less in deaf Dalmatian group compared to hearing Labradors: (1) cellular area of the stria vascularis in all cochlear turns; (2) cellular area of spiral ligament in the inferior part of the basal turn; (3) cellular density of spiral ganglion cells within segments III and IV; (4) number of Scarpa’s ganglion cells; and (5) density of saccular hair cells types I and II. A borderline negative correlation was found between average density of spiral ganglion cells of segments III and IV and age in group I. Young deaf animals showed some cochlear hair cells, however in adult dogs all hair cells were replaced by supporting cells. Conclusion: General pattern of cochleosaccular dysplasia is variable, even when only one etiology, the genetic one, is involved. The gradual degeneration of inner ear elements in the cochleosaccular degeneration might indicate that early intervention might be crucial to stop the progression of cochleosaccular dysplasia. ß 2010 Elsevier Ireland Ltd. All rights reserved.
Keywords: Cochleosaccular dysplasia Dalmatian dog Histopathology
1. Introduction Congenital hearing impairment is responsible for 50–60% of cases of profound early-onset deafness, a condition that affects about 4–11 of 10,000 births [1]. Cochleosaccular dysplasia, a malformation of the membranous labyrinth is considered the most common cause of profound congenital hearing impairment, and accounts for approximately 70% of these cases [1,2]. Typical pathological findings are limited to the phylogenetically younger
§ This work was supported in part by the International Hearing Foundation, The Starkey Hearing Foundation and Hubbard Broadcasting Foundation, MN, USA. §§ Presented as a poster in the Annual Meeting of the AAO-HNS, NY, September, 2004 and as an oral presentation in the 8th Conference on Cholesteatoma and Ear Surgery, Turkey, July 2008. * Corresponding author at: SQN 205 Bloco B Apartamento 506, 70843020 - Asa Norte, Brası´lia, Distrito Federal, Brazil. Tel.: +55 61 30375484; fax: +55 61 34433397. E-mail address: [email protected] (Andre L.L. Sampaio).
0165-5876/$ – see front matter ß 2010 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.ijporl.2010.05.020
parts of the labyrinth: the cochlea and saccule. We recently described a series of temporal bones from subjects with cochleosaccular dysplasia with ages ranging from newborn to 86 years [1,2]. In our series, histopathological findings were quite variable, perhaps reflecting the different etiologies involved in the origin of this inner ear abnormality. It is well established that genetic deafness occurs in many animals, including mice [3], cats [4], minks [5] and dogs [6–13]. In mice, deafness is combined with a locomotive disturbance of various types and grades. The deaf mouse displays no less than 11 variants of such disorders. In cats and dogs, deafness is often combined with pigmentation anomalies, however hereditary deafness also occurs in breeds without pigmentation anomalies, including Dalmatians. Deafness in Dalmatians appears mostly isolated as in humans, making this breed, a supposedly, good model for hereditary deafness. Although inherited deafness of Dalmatian dogs has been recognized since the last century [6], the mode of inheritance has not been totally determined. Several authors have reported
A.L.L. Sampaio et al. / International Journal of Pediatric Otorhinolaryngology 74 (2010) 934–938
histopathologic findings of the inner ear of deaf Dalmatians [6–13]. The histopathological patterns vary, as do those in humans with cochleosaccular dysplasia. Human cochleosaccular dysplasia contains congenital genetic and acquired etiologies such as rubella. Therefore we decided to look at the Dalmatian dogs as a genetic model for this problem in order to remove other confounding factors’ effects. Despite some reports in the literature, to our knowledge, no study has been performed that compares temporal bone morphometric findings of Dalmatian dogs of different ages. The purpose of the present study was to analyze a series of temporal bones of deaf Dalmatian dogs from 5 days after birth to adulthood to better understand the characteristics and the progression pathogenesis of genetic cochleosaccular dysplasia.
935
hearing. The histopathologic inclusion criteria for the cochleosaccular pattern were normal histologic appearance of the middle ear, utricle and semicircular canals and cochlear bone structures with some level of dysplasia of the cochlea and saccule as pointed out by Scheibe. 2.2. Hair cells Organ of Corti and hair cells were identified as present or absent. 2.3. Reissner’s membrane It was evaluated as normal, hydropic or collapsed.
2. Materials and methods 2.4. Stria vascularis measurement In this paper, we refer the pathology consistent with the cochleosaccular pattern as ‘‘cochleosaccular dysplasia’’ to disordered development causing congenital deafness. Temporal bones were obtained from the collection at the University of Minnesota, Minneapolis, MN. This study was submitted and approved by the Committee on Animal Research of the University of Minnesota. Regional Dalmatian breeders were contact in the past decades 1970s and 1980s in order to obtain animals with a high suspicion of deafness based on their behavior. The animals were deeply anesthetized and intravitally perfused by Heidenhein´s Susa solution. Temporal bones were harvested, fixed in a solution of 10% formalin, defatted in graded series of ethanol, decalcified in 5% trichloroacetic acid, neutralized with 5% sodium sulfate, dehydrated in graded concentrations of ethanol and embedded in celloidin. Specimens were cut at a section thickness of 20 mm in the horizontal plane from superior to inferior. Every 10th section was mounted on a glass slide, and stained with hematoxylin and eosin for light microscopy. 2.1. Temporal bone selection Due to ethic concerns, and following the University of Minnesota protocols for animal studies, only one animal for each age was selected for this investigation. Two groups of temporal bones were selected. Group I included 41 temporal bones (2 from each of 17 deaf adult Dalmatian dogs and only the right side from each of 7 deaf puppies). The puppies were sacrificed at ages of 5, 6, 7, 9 14, 21 and 28 days. These animals came from two litters of Dalmatian dogs obtained from mating the same deaf bitch with the same deaf male. Among these puppies, five were females and two males. The 17 adult deaf Dalmatian dogs were sacrificed at different ages from 6 to 261 weeks (5 years). The interval between the ages was 4 weeks until the first year and 1 year until 261 weeks. Among these animals, 10 were male and 7 female. Only specimens from animals with normal pigmentation status, a strong familial history of deafness and no history of ototoxic drugs use or ear disorders were included. Group II consisted of 25 temporal bones from 15 ‘‘normal’’ aged-matched, hearing Black Labradors (5 puppies contributed with only the right temporal bone). Deafness in this breed has not been described. No animal in this group had a history of otologic disease or ototoxic drug use. The puppies were sacrificed at ages of 1, 2, 3, 4 and 6 weeks. There were 3 males and 2 females among the puppies. The adult hearing Labradors were sacrificed every 2 months until the first year, and every year until 260 weeks. Six animals were female and 4 males. The hearing of the animals up to 1 month of age was assessed by simple hearing procedures, such as, whistle and hand clap and by the observation of their behavior. The hearing of the pups was not tested and the term deaf, as used here, refers to the presence of changes in the cochlea that could, definitely, be responsible for nonfunctional
Morphometric measurements of the area of the stria vascularis were made in all turns of the cochlea at the mid-modiolar level and 200 mm above and below. Averages of the areas were used for statistical analysis. All observed cystic-like structures or concretions were excluded from the total area of the stria. 2.5. Spiral ligament measurement Area counts of the spiral ligaments were made in all turns of the cochlea at the same level. The amount of spiral ligament tissue was quantified by determining the areas of its cut surfaces and the averages of the areas were statistically analyzed. 2.6. Spiral ganglion cells The average density of spiral ganglion cells within Rosenthal’s canal was determined. The canal was divided into five segments. All nuclei were counted in each section. Tissue sections that tangentially cut through the outer edge of the spiral canal were identified and plotted on a vertical line. All points were joined to form a spiral line, which represented the outer edge of the spiral canal. Cellularity was considered as the number of nuclei/crosssectional area of Rosenthal’s canal. Averages of cellularity were statistically compared for each segment in the two groups of dogs. 2.7. Scarpa’s ganglion cells Counts of Scarpa’s ganglion cells were obtained by the technique outlined by Richter [14]. Nucleoli were counted in every stained section (i.e., in every 10th section) at a magnification of 400. Because the saccule is innervated by the superior and inferior division of Scarpa’s ganglion, the total number of cells was used for statistical comparisons between the groups. Raw cell counts were corrected for double counting of nucleoli split between 2 sections by the formula of Abercrombie [14]: Hi = hi t\(t + d) where Hi is the corrected ganglion cell count, hi is the raw number of ganglion cells counted, t is the section thickness and d is the diameter of the nucleolus. The number of counted cells multiplied by the above correction factor and then by 10 (to account for unstained sections) gave the total count. The diameter of the nucleoli was measured in all cells of the section that showed the largest number of neurons for each ear. 2.8. Saccule The assessment of hair cells of the saccular macula was performed as described by Merchant [15]. Differential interference contrast (Nomarski) microscopy was used to examine the specimens. Counts were made in only those areas where the plane of
936
A.L.L. Sampaio et al. / International Journal of Pediatric Otorhinolaryngology 74 (2010) 934–938
section was perpendicular to the surface of the sensory epithelium. Counts were made at the mid-modiolar level and 200 mm above and bellow. Only hair cells with nuclei were counted. Counts were expressed in terms of density (i.e., number of hair cells per surface area). Surface area was determined by multiplying the thickness of the section by the distance along the endolymphatic surface of the epithelium where the count was made. Average density of saccular hair cells was used for statistical analysis. The raw hair cell density counts were corrected for double counting using the same Abercrombie formula. The nuclear diameter of 100 hair cells was also measured. 2.9. Statistical analysis For the statistical analysis, the temporal bones of both groups were evaluated in a pool of ears. The Mann–Whitney test was used to compare temporal bones from normal and deaf dogs, temporal bones from the male and female animals within the same group and the right and left temporal bones within the same group. Significance was defined as p < 0.05. Pearson’s linear correlation coefficient was employed to investigate the effects of age on the area of the stria vascularis and the spiral ligament, the number of spiral ganglion cells and Scarpa’s ganglion cell count and saccular hair cell density. Correlation was defined as moderate when the coefficient was 0.51 to 0.75 and high when the coefficient was 0.76 to 1. No correlation was assumed when the coefficient was less than 0.50. Count reliability of over 95% was obtained by two authors and in count/recount comparisons by the same individual.
Fig. 1. In the upper basal turn of this left temporal bone from a 3-week-old puppy, we show the main findings of cochleosaccular dysplasia. O, organ of Corti represented by some supporting cells; T, deformed tectorial membrane; Sv, atrophic stria vascularis. RM, Collapsed Reissner’s Membrane. H&E. Light microscopy 20.
was totally collapsed. The collapse of Reissner’s membrane started at its attachment at limbus followed by total collapse of the membrane. 3.3. Stria vascularis
3. Results The young dogs were born with immature inner ear structures in both groups. The outer hair cell nucleus was positioned at the central part of the cells. The tunnel of Corti was filled by amorphous material and was a virtual structure. There was an aggregation of cells with epithelial characteristics close to the inner sulcus. 3.1. Cochlear hair cells In group I, although young animals had a few cochlear outer hair cells (5, 6, 7 and 9 days), these hair cells were totally missed after the 14th day and were replaced by supporting cells in older dogs (21st day and older). The total loss of outer hair cells was preceded by the collapse of the Reissner’s membrane. The outer hair cells disappeared first and were more susceptible than inner hair cells. The 28th day dog had the inner hair cell. 3.2. Reissner’s membrane The area of scala media decreased gradually by the collapse of Reissner’s membrane in group I starting in the 9th day dog (Fig. 1). At 1 week of age, Reissner’s membrane started to collapse from the attachment at the limbus. At age 14 days, Reissner’s membrane
Stria vascularis structural abnormalities were found in the marginal layer in young animals (5, 7, 9, 14 days). The typical digital projections upon the capillaries were missed in group I, and the marginal cells were totally replaced by basal layer cells in older animals (21 days and older). This finding happened in parallelism with the collapse of Reissner’s membrane and the loss of hair cells. Averages of cellular areas (SD) of the stria vascularis and spiral ligament in all 5 cochlear turns in groups I and II are shown in Table 1. Average areas of the stria vascularis and spiral ligament were not significantly different (p > 0.05) between right and left ears or genders of the animals in either group I or II. Average cellular area of the stria vascularis in group I was significantly lower than group II in all cochlear turns (p < 0.0001) (Fig. 1). 3.4. Spiral ligament There were no morphological abnormalities in the spiral ligament in both groups. Average cellular area of spiral ligament was lower in group I than in group II only in the inferior part of the basal turn (p = 0.03). Differences in other cochlear turns were not statistically significant (p > 0.05). There was no statistical correlation between the areas of the stria vascularis, the spiral ligament, and the age of the animals in either group.
Table 1 Averages of the cellular areas (SD) of the stria vascularis and spiral ligament in all 5 cochlear turns in groups I and II. Cochlear turns
Structure Stria vascularis
Basal inferior Basal superior Middle inferior Middle superior Apical
Spiral ligament
Group I
Group II
Group I
3030 (913) 2459 (629) 2468 (619) 2241(565) 2372 (730)
7804 6441 4789 3735 3718
12,9201 90,239 79,742 55,786 50,289
Stria vascularis and spiral ligament areas in m2.
(1416) (1239) (1200) (893) (769)
Group II (30,947) (14,609) (16,186) (11,491) (11,652)
137,242 90,570 82,073 59,484 51,724
(2377) (15,436) (12,098) (8683) (6756)
A.L.L. Sampaio et al. / International Journal of Pediatric Otorhinolaryngology 74 (2010) 934–938
937
Table 2 Averages of cellular densities of spiral ganglion cells (SD) within the five segments of Rosenthal’s canal in groups I and II. Segments of the Rosenthal canal
Group I
Segment Segment Segment Segment Segment
I II III IV V
14.1 18.3 16.3 15.1 17.2
II (4.6) (4.9) (6.0) (7.7) (6.1)
13.9 19.9 19.5 19.7 18.3
(4.0) (2.8) (2.8) (2.6) (2.8)
Cellular densities in number of cells/0.01 mm2.
3.5. Spiral ganglion cells No young dogs showed apparent loss of spiral ganglion cells. The apparent neuronal losses started by the end of the first year (Fig. 3). Average of the cellular density of spiral ganglion cells (SD) within the five segments of Rosenthal’s canal is shown in Table 2. No statistical differences were found between the sides of the temporal bones in groups I and II or between the gender of the animals (p > 0.05). Average cellular density of spiral ganglion cells within segments III and IV were significantly lower in group I than group II (p = 0.009 and 0.002 respectively). Differences in the rest of the segments were not significant (p > 0.05) (Fig. 2). A moderate negative correlation was found between average density of spiral ganglion cells of segments III and IV and age in group I (Pearson coefficient 0.59 and 0.54 respectively). No correlation with age was found in the other segments of group I or in any of the segments of group II. 3.6. Scarpa’s ganglion cells
Fig. 3. The left saccule of 3-week-old puppy where there is a collapsed saccular wall (S) with amorphous deposit on the atrophic maculae (C). I, Hair cell type I; II, Hair cell type II. H&E. Nomarski microscopy, 40.
Table 3 Average density of saccular types I and II hair cells (SD) in groups I and II. Hair cell type
No morphological abnormality was found in the groups. There was a statistically significant difference in the average number of Scarpa’s ganglion cells; 5911 (1284) in group I compared to 6478 (760) in group II (p = 0.022). Differences in comparisons of side or gender within groups I and II were not statistically significant (p > 0.05). No correlation with age was found in either group. 3.7. Saccule The saccular structure was normal until 14th day dog and started collapsing during the 3rd week. The older dogs showed
Group I
I II Total
II 8.2 (5.2) 6.9 (4.0)
15.1
23.8 (6.0) 23.1 (3.8) 46.9
Cellular densities in number of cells/0.01 mm2.
several level of saccular membrane collapse and otolitic membrane deformity. Average density of saccular types I and II hair cells (SD) in both groups is shown in Table 3. There was no significant difference between the genders or right and left ears in either group (p > 0.05). Average density of saccular types I and II hair cells was significantly lower in group I than in group II (p < 0.0001). No correlation with age was found in either group. 4. Discussion
Fig. 2. This left temporal bone from the 44-week-old dog shows a general loss of spiral ganglion cells in Rosenthal’s canal pointed by arrows. H&E. Light microscopy, 10.
Quantitative evaluation of inner ear structures has been poorly studied in the cochleosaccular histopathologic pattern of deafness. In humans, to our knowledge, we described for the first time a series of temporal bones showing the quantitative aspects of the inferior part of the labyrinth in cochleosaccular dysplasia [1,2]. Two previous studies described histopathological quantitative analysis of the inner ear in Dalmatian dogs. Mair [10] analyzed the spiral ganglion cells of 10 Dalmatian dogs and Branis and Burda [16] reported the results of cochlear reconstruction and hair cell counts of one deaf Dalmatian dog. In none of the previous studies did the authors compare the results to normal dogs and the conclusions were limited since the effects of aging on the inner ear were not considered. We found the areas of the stria vascularis to be significantly lower in all cochlear turns in group I compared to group II. Atrophy of the stria vascularis was reported in all previous studies of cochleosaccular dysplasia in Dalmatian dogs [6–13,16]. The stria vascularis
938
A.L.L. Sampaio et al. / International Journal of Pediatric Otorhinolaryngology 74 (2010) 934–938
median area showed a gradient from the base to the apex of the cochlea in group II. This well-known gradient has been reported in classic morphometric studies of the stria vascularis [17]. This gradient was not observed in group I. This might suggest that the apex was preserved during the development of the disease or that the stria vascularis never developed entirely in all cochlear turns. Group I showed a significantly lower spiral ligament area in the inferior portion of the basal turn than group II. The spiral ligament has been described as normal in previous histopathological and morphometric studies in cochleosaccular dysplasia in humans [1] and in Dalmatian dogs [6–13,16] however, none of these studies performed morphometric analysis of the cellular area. The reduction of the spiral ligament area, only in the lower basal turn, might emphasize the direction of development of the disease, from base to apex, in this species. Interestingly, Reissner’s membrane collapse occurred gradually, starting from the attachment at the limbus at 1 week of age. By the second week, Reissner’s membrane was totally collapsed. Then the complete loss of hair cells is continued and ended up with the loss of spiral ganglion cells in adult animal. Quantitative studies of spiral ganglion cells in Dalmatian dogs with the cochleosaccular histopathologic pattern by Mair [10], Igarashi et al. [9], Anderson et al. [11] and Lurie [13] reported a severe loss of cells and some level of atrophy in the remaining cells. Hudson and Ruben [8] described the ganglion as normal. We found a significant loss of spiral ganglion cells in segments III and IV in group I. In reconstruction of Rosenthal’s canal, segments III and IV in dogs are radially correlated to the upper basal and inferior medial turns. Spiral ganglion cell loss secondary to degeneration of the end organ has been reported in Dalmatian dogs [10]. Webster [18], working with mutant rats, reported a normal count of spiral ganglion cells in the apex of the cochlea, in spite of loss in the rest of the cochlear turns. He suggested the relative preservation of the spiral ganglion cells in the apex of the cochlea resulted from the remaining hair cells in the apex in these animals. Our findings and Webster’s observation support the thesis of relative preservation of the cochlear apex during the development of cochleosaccular dysplasia. Spoendlin [19] studied the cochlear neuroanatomy of several mammalian species and concluded that hair cells from the middle cochlear turn have the highest innervation density within the cochlea. This specialized cochlear region might be very vulnerable to pathogenic agents. Pujol et al. [3] reported very early abnormalities in the spiral ganglion cells in mutant rats with cochleosaccular dysplasia. The authors attributed these findings to the primary action of a gene on spiral ganglion cells, rather than the effect of secondary degeneration. A moderate negative correlation was found between the average density of spiral ganglion cells of segments III and IV and age in group I. However, the coefficients were close to the cutoff point ( 0.59 and 0.54 respectively), and the role of age interfering with the number of spiral ganglion cells in these segments of Rosenthal’s canal can not be clearly assumed. These observations might perhaps suggest a direct effect of the genetic disorder on ganglion cells rather then ganglion cell loss being a result of neuroepithelial degeneration as previously assumed. This new finding might have some implication in the cases of cochlear implantation due to deafness secondary to genetic causes. A quantitative study of Scarpa’s ganglion cells has not been performed previously in dogs having cochleosaccular dysplasia. Igarashi et al. [9] and Hudson and Ruben [8] described the ganglion as being morphologically normal in the dogs studied. We found Scarpa’s ganglion cells significantly lower in group I than in group II. The damage to the end vestibular organ (lack of hair cells in the saccule) might be responsible for the loss of Scarpa’s ganglion cells in our specimens, because of the retrograde mechanism of degeneration. However, the loss of Scarpa’s ganglion cells secondary to the loss of the end organ is not clear in other species.
Furthermore, we did not find correlation between the loss of the neurons and the ages of the animals. We found types I and II hair cells of the saccule to be significantly lower in group I than in group II. Deol [20] in the piroute, shaker-1 and waltzer have shown that saccular morphologic abnormalities developed after the changes in the cochlea. Igarashi et al. [9] reported that changes in the cochlea were more severe than in the saccule in a Dalmatian puppy specimen. They suggested that the saccule, an older structure, is more resistant than the cochlea, the youngest inferior part. Our findings go against these previous reports. In conclusion, this quantitative histologic examination of heredity deafness in Dalmatian dogs shows that the general pattern of cochleosaccular dysplasia is variable, even when only one etiology, the genetic one, is involved and it depends on the age of the animal studied. The most important histopathological findings develop during the first weeks. Nevertheless, the velocity of the process seems to be particular to each animal. Gradual degeneration of inner ear elements, starting in Reissner’s membrane, then complete hair cells loss in parallel with the stria degeneration, collapse of saccular wall and ending up with loss of spiral ganglion cells might indicate that early intervention in the immediate postnatal period might be very crucial to stop the progression of cochleosaccular dysplasia due to the genetic causes.
Acknowledgment The authors would like to thank Capes-Fulbright exchange program. References [1] A.L.L. Sampaio, S. Coreoglu, P. Schachern, et al., Cochleosaccular dysplasia: a morphometric and histopathologic study in a series of temporal bones, Otol. Neurotol. 25 (2004) 530–535. [2] Shin Kariya, S. Cureoglu, A.L.L. Sampaio, et al., Quantitative study of vestibular sensory epithelium in cochleosaccular dysplaisa, Otol. Neurotol. 26 (2005) 495–499. [3] R. Pujol, A. Shnerson, M. Lenoir, et al., Early degeneration of sensory and ganglion cells in the inner ear of mice with uncomplicated genetic deafness (dn): preliminary observation, Hear. Res. 12 (1983) 463–466. [4] M. Rebillard, G. Rebillard, R. Pujol, Variability of hereditary in the white cat I, Physiol. Hear. Res. 5 (1981) 179–187. [5] L.Z. Saunders, The histopathology of congenital deafness in white mink, Pathol. Vet. 2 (1965) 256–263. [6] G.B. Rawiz, Gehirn eines weissen Hundes mit blauen Augen, Morphol. Arb. 6 (1896) 545–553. [7] F. Altmann, Histologic picture of inherited nerve deafness in man and animals, Arch. Otolaryngol. 51 (1950) 852–890. [8] W.R. Hudson, R.J. Ruben, Hereditary deafness in the Dalmatian dog, Arch. Otorhinolaryngol. 75 (1962) 213–219. [9] M. Igarashi, B.R. Alford, R. Saito, et al., Inner ear abnormalities in dogs, Ann. Otol. 81 (1972) 249–255. [10] I.W.S. Mair, Hereditary deafness in Dalmatian dog, Arch. Otorhinolaryngol. 212 (1976) 1–14. [11] H. Anderson, B. Hericson, R.G. Lundquist, et al., Genetic hearing impairment in the Dalmatian dog, Acta. Otolaryngol. (Stockh.) 232 (1968) 1–32. [12] L.G. Johnson, J.E. Hawkins Jr., A.A. Muraski, et al., Vascular anatomy and pathology of the cochlea in Dalmatian dogs, in: A.D.J. De Lorenzo (Ed.), Vascular Disorders and Hearing Defects, University Park Press, Baltimore, 1974, p. 249. [13] M.H. Lurie, The membranous labyrinth in congenitally deaf Collie and Dalmatian dog, Laryngoscope 58 (1948) 279–287. [14] E. Richter, Quantitative study of human Scarpa’s ganglion and vestibular sensory epithelia, Acta Otolaryngol. 90 (1980) 199–208. [15] S.N. Merchant, A method for quantitative assessment of vestibular otopathology, Laryngoscope 109 (1999) 1560–1569. [16] M. Branis, H. Burda, Inner ear structure in deaf and normally hearing Dalmatian dog, J. Comp. Pathol. 95 (1985) 295–299. [17] P.A. Santi, B.N. Lakhani, C. Bingham, The volume density of cells and capillaries of the normal stria vascularis, Hear. Res. 11 (1983) 7–22. [18] D.B. Webster, The spiral ganglion and cochlear nuclei of deafness mice, Hear. Res. 18 (1985) 19–27. [19] H. Spoendlin, Neuroanatomy of the cochlea, in: Z.E. Terhardt (Ed.), Facts and Models in Hearing, 1st ed., Springer, Berlin/Heidelberg/New York, 1974, p. 18. [20] M.S. Deol, The anatomy and development of mutants Pirouette, shaker-1 and waltzer in the mouse, Proc. R. Soc. (Biol.) 145 (1956) 206–213.
Contents lists available at ScienceDirect
International Journal of Pediatric Otorhinolaryngology journal homepage: www.elsevier.com/locate/ijporl
Histopathological morphometric study of cochleosaccular dysplasia in Dalmatian dogs§,§§ Andre L.L. Sampaio b,c,*, Elizabeth Paine c, Patricia A. Schachern a,c, Carolyn Sutherland a, Sebahattin Cureoglu a,c, Carlos A.C.P. Olivieira b, Michael M. Paparella a,c,d a
Department of Otolaryngology, Otitis Media Research Center, University of Minnesota, Minneapolis, MN, United States Brasilia University Medical School - Capes, Fulbright Scholarship, Brasilia, Brazil c International Hearing Foundation, Minneapolis, MN, United States d Paparella Ear Head and Neck Institute, Minneapolis, MN, United States b
A R T I C L E I N F O
A B S T R A C T
Article history: Received 14 February 2010 Received in revised form 17 May 2010 Accepted 18 May 2010 Available online 12 June 2010
Objective: To analyze temporal bones of deaf Dalmatian dogs from 5 days after birth to adulthood to better understand the pathogenesis of cochleosaccular dysplasia. Methods: This is an experimental animal histopathological temporal bone study that included two groups of temporal bones. Group I consisted of 41 temporal bones from deaf Dalmatian dogs and group II of 25 temporal bones from 15 ‘‘normal’’ aged-matched, hearing Black Labradors. Morphometric analysis included: stria vascularis and spiral ligament area measurements, and cell counts of spiral ganglion, Scarpa’s ganglion, and hair cells of saccular macula. Results: The following findings were significantly less in deaf Dalmatian group compared to hearing Labradors: (1) cellular area of the stria vascularis in all cochlear turns; (2) cellular area of spiral ligament in the inferior part of the basal turn; (3) cellular density of spiral ganglion cells within segments III and IV; (4) number of Scarpa’s ganglion cells; and (5) density of saccular hair cells types I and II. A borderline negative correlation was found between average density of spiral ganglion cells of segments III and IV and age in group I. Young deaf animals showed some cochlear hair cells, however in adult dogs all hair cells were replaced by supporting cells. Conclusion: General pattern of cochleosaccular dysplasia is variable, even when only one etiology, the genetic one, is involved. The gradual degeneration of inner ear elements in the cochleosaccular degeneration might indicate that early intervention might be crucial to stop the progression of cochleosaccular dysplasia. ß 2010 Elsevier Ireland Ltd. All rights reserved.
Keywords: Cochleosaccular dysplasia Dalmatian dog Histopathology
1. Introduction Congenital hearing impairment is responsible for 50–60% of cases of profound early-onset deafness, a condition that affects about 4–11 of 10,000 births [1]. Cochleosaccular dysplasia, a malformation of the membranous labyrinth is considered the most common cause of profound congenital hearing impairment, and accounts for approximately 70% of these cases [1,2]. Typical pathological findings are limited to the phylogenetically younger
§ This work was supported in part by the International Hearing Foundation, The Starkey Hearing Foundation and Hubbard Broadcasting Foundation, MN, USA. §§ Presented as a poster in the Annual Meeting of the AAO-HNS, NY, September, 2004 and as an oral presentation in the 8th Conference on Cholesteatoma and Ear Surgery, Turkey, July 2008. * Corresponding author at: SQN 205 Bloco B Apartamento 506, 70843020 - Asa Norte, Brası´lia, Distrito Federal, Brazil. Tel.: +55 61 30375484; fax: +55 61 34433397. E-mail address: [email protected] (Andre L.L. Sampaio).
0165-5876/$ – see front matter ß 2010 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.ijporl.2010.05.020
parts of the labyrinth: the cochlea and saccule. We recently described a series of temporal bones from subjects with cochleosaccular dysplasia with ages ranging from newborn to 86 years [1,2]. In our series, histopathological findings were quite variable, perhaps reflecting the different etiologies involved in the origin of this inner ear abnormality. It is well established that genetic deafness occurs in many animals, including mice [3], cats [4], minks [5] and dogs [6–13]. In mice, deafness is combined with a locomotive disturbance of various types and grades. The deaf mouse displays no less than 11 variants of such disorders. In cats and dogs, deafness is often combined with pigmentation anomalies, however hereditary deafness also occurs in breeds without pigmentation anomalies, including Dalmatians. Deafness in Dalmatians appears mostly isolated as in humans, making this breed, a supposedly, good model for hereditary deafness. Although inherited deafness of Dalmatian dogs has been recognized since the last century [6], the mode of inheritance has not been totally determined. Several authors have reported
A.L.L. Sampaio et al. / International Journal of Pediatric Otorhinolaryngology 74 (2010) 934–938
histopathologic findings of the inner ear of deaf Dalmatians [6–13]. The histopathological patterns vary, as do those in humans with cochleosaccular dysplasia. Human cochleosaccular dysplasia contains congenital genetic and acquired etiologies such as rubella. Therefore we decided to look at the Dalmatian dogs as a genetic model for this problem in order to remove other confounding factors’ effects. Despite some reports in the literature, to our knowledge, no study has been performed that compares temporal bone morphometric findings of Dalmatian dogs of different ages. The purpose of the present study was to analyze a series of temporal bones of deaf Dalmatian dogs from 5 days after birth to adulthood to better understand the characteristics and the progression pathogenesis of genetic cochleosaccular dysplasia.
935
hearing. The histopathologic inclusion criteria for the cochleosaccular pattern were normal histologic appearance of the middle ear, utricle and semicircular canals and cochlear bone structures with some level of dysplasia of the cochlea and saccule as pointed out by Scheibe. 2.2. Hair cells Organ of Corti and hair cells were identified as present or absent. 2.3. Reissner’s membrane It was evaluated as normal, hydropic or collapsed.
2. Materials and methods 2.4. Stria vascularis measurement In this paper, we refer the pathology consistent with the cochleosaccular pattern as ‘‘cochleosaccular dysplasia’’ to disordered development causing congenital deafness. Temporal bones were obtained from the collection at the University of Minnesota, Minneapolis, MN. This study was submitted and approved by the Committee on Animal Research of the University of Minnesota. Regional Dalmatian breeders were contact in the past decades 1970s and 1980s in order to obtain animals with a high suspicion of deafness based on their behavior. The animals were deeply anesthetized and intravitally perfused by Heidenhein´s Susa solution. Temporal bones were harvested, fixed in a solution of 10% formalin, defatted in graded series of ethanol, decalcified in 5% trichloroacetic acid, neutralized with 5% sodium sulfate, dehydrated in graded concentrations of ethanol and embedded in celloidin. Specimens were cut at a section thickness of 20 mm in the horizontal plane from superior to inferior. Every 10th section was mounted on a glass slide, and stained with hematoxylin and eosin for light microscopy. 2.1. Temporal bone selection Due to ethic concerns, and following the University of Minnesota protocols for animal studies, only one animal for each age was selected for this investigation. Two groups of temporal bones were selected. Group I included 41 temporal bones (2 from each of 17 deaf adult Dalmatian dogs and only the right side from each of 7 deaf puppies). The puppies were sacrificed at ages of 5, 6, 7, 9 14, 21 and 28 days. These animals came from two litters of Dalmatian dogs obtained from mating the same deaf bitch with the same deaf male. Among these puppies, five were females and two males. The 17 adult deaf Dalmatian dogs were sacrificed at different ages from 6 to 261 weeks (5 years). The interval between the ages was 4 weeks until the first year and 1 year until 261 weeks. Among these animals, 10 were male and 7 female. Only specimens from animals with normal pigmentation status, a strong familial history of deafness and no history of ototoxic drugs use or ear disorders were included. Group II consisted of 25 temporal bones from 15 ‘‘normal’’ aged-matched, hearing Black Labradors (5 puppies contributed with only the right temporal bone). Deafness in this breed has not been described. No animal in this group had a history of otologic disease or ototoxic drug use. The puppies were sacrificed at ages of 1, 2, 3, 4 and 6 weeks. There were 3 males and 2 females among the puppies. The adult hearing Labradors were sacrificed every 2 months until the first year, and every year until 260 weeks. Six animals were female and 4 males. The hearing of the animals up to 1 month of age was assessed by simple hearing procedures, such as, whistle and hand clap and by the observation of their behavior. The hearing of the pups was not tested and the term deaf, as used here, refers to the presence of changes in the cochlea that could, definitely, be responsible for nonfunctional
Morphometric measurements of the area of the stria vascularis were made in all turns of the cochlea at the mid-modiolar level and 200 mm above and below. Averages of the areas were used for statistical analysis. All observed cystic-like structures or concretions were excluded from the total area of the stria. 2.5. Spiral ligament measurement Area counts of the spiral ligaments were made in all turns of the cochlea at the same level. The amount of spiral ligament tissue was quantified by determining the areas of its cut surfaces and the averages of the areas were statistically analyzed. 2.6. Spiral ganglion cells The average density of spiral ganglion cells within Rosenthal’s canal was determined. The canal was divided into five segments. All nuclei were counted in each section. Tissue sections that tangentially cut through the outer edge of the spiral canal were identified and plotted on a vertical line. All points were joined to form a spiral line, which represented the outer edge of the spiral canal. Cellularity was considered as the number of nuclei/crosssectional area of Rosenthal’s canal. Averages of cellularity were statistically compared for each segment in the two groups of dogs. 2.7. Scarpa’s ganglion cells Counts of Scarpa’s ganglion cells were obtained by the technique outlined by Richter [14]. Nucleoli were counted in every stained section (i.e., in every 10th section) at a magnification of 400. Because the saccule is innervated by the superior and inferior division of Scarpa’s ganglion, the total number of cells was used for statistical comparisons between the groups. Raw cell counts were corrected for double counting of nucleoli split between 2 sections by the formula of Abercrombie [14]: Hi = hi t\(t + d) where Hi is the corrected ganglion cell count, hi is the raw number of ganglion cells counted, t is the section thickness and d is the diameter of the nucleolus. The number of counted cells multiplied by the above correction factor and then by 10 (to account for unstained sections) gave the total count. The diameter of the nucleoli was measured in all cells of the section that showed the largest number of neurons for each ear. 2.8. Saccule The assessment of hair cells of the saccular macula was performed as described by Merchant [15]. Differential interference contrast (Nomarski) microscopy was used to examine the specimens. Counts were made in only those areas where the plane of
936
A.L.L. Sampaio et al. / International Journal of Pediatric Otorhinolaryngology 74 (2010) 934–938
section was perpendicular to the surface of the sensory epithelium. Counts were made at the mid-modiolar level and 200 mm above and bellow. Only hair cells with nuclei were counted. Counts were expressed in terms of density (i.e., number of hair cells per surface area). Surface area was determined by multiplying the thickness of the section by the distance along the endolymphatic surface of the epithelium where the count was made. Average density of saccular hair cells was used for statistical analysis. The raw hair cell density counts were corrected for double counting using the same Abercrombie formula. The nuclear diameter of 100 hair cells was also measured. 2.9. Statistical analysis For the statistical analysis, the temporal bones of both groups were evaluated in a pool of ears. The Mann–Whitney test was used to compare temporal bones from normal and deaf dogs, temporal bones from the male and female animals within the same group and the right and left temporal bones within the same group. Significance was defined as p < 0.05. Pearson’s linear correlation coefficient was employed to investigate the effects of age on the area of the stria vascularis and the spiral ligament, the number of spiral ganglion cells and Scarpa’s ganglion cell count and saccular hair cell density. Correlation was defined as moderate when the coefficient was 0.51 to 0.75 and high when the coefficient was 0.76 to 1. No correlation was assumed when the coefficient was less than 0.50. Count reliability of over 95% was obtained by two authors and in count/recount comparisons by the same individual.
Fig. 1. In the upper basal turn of this left temporal bone from a 3-week-old puppy, we show the main findings of cochleosaccular dysplasia. O, organ of Corti represented by some supporting cells; T, deformed tectorial membrane; Sv, atrophic stria vascularis. RM, Collapsed Reissner’s Membrane. H&E. Light microscopy 20.
was totally collapsed. The collapse of Reissner’s membrane started at its attachment at limbus followed by total collapse of the membrane. 3.3. Stria vascularis
3. Results The young dogs were born with immature inner ear structures in both groups. The outer hair cell nucleus was positioned at the central part of the cells. The tunnel of Corti was filled by amorphous material and was a virtual structure. There was an aggregation of cells with epithelial characteristics close to the inner sulcus. 3.1. Cochlear hair cells In group I, although young animals had a few cochlear outer hair cells (5, 6, 7 and 9 days), these hair cells were totally missed after the 14th day and were replaced by supporting cells in older dogs (21st day and older). The total loss of outer hair cells was preceded by the collapse of the Reissner’s membrane. The outer hair cells disappeared first and were more susceptible than inner hair cells. The 28th day dog had the inner hair cell. 3.2. Reissner’s membrane The area of scala media decreased gradually by the collapse of Reissner’s membrane in group I starting in the 9th day dog (Fig. 1). At 1 week of age, Reissner’s membrane started to collapse from the attachment at the limbus. At age 14 days, Reissner’s membrane
Stria vascularis structural abnormalities were found in the marginal layer in young animals (5, 7, 9, 14 days). The typical digital projections upon the capillaries were missed in group I, and the marginal cells were totally replaced by basal layer cells in older animals (21 days and older). This finding happened in parallelism with the collapse of Reissner’s membrane and the loss of hair cells. Averages of cellular areas (SD) of the stria vascularis and spiral ligament in all 5 cochlear turns in groups I and II are shown in Table 1. Average areas of the stria vascularis and spiral ligament were not significantly different (p > 0.05) between right and left ears or genders of the animals in either group I or II. Average cellular area of the stria vascularis in group I was significantly lower than group II in all cochlear turns (p < 0.0001) (Fig. 1). 3.4. Spiral ligament There were no morphological abnormalities in the spiral ligament in both groups. Average cellular area of spiral ligament was lower in group I than in group II only in the inferior part of the basal turn (p = 0.03). Differences in other cochlear turns were not statistically significant (p > 0.05). There was no statistical correlation between the areas of the stria vascularis, the spiral ligament, and the age of the animals in either group.
Table 1 Averages of the cellular areas (SD) of the stria vascularis and spiral ligament in all 5 cochlear turns in groups I and II. Cochlear turns
Structure Stria vascularis
Basal inferior Basal superior Middle inferior Middle superior Apical
Spiral ligament
Group I
Group II
Group I
3030 (913) 2459 (629) 2468 (619) 2241(565) 2372 (730)
7804 6441 4789 3735 3718
12,9201 90,239 79,742 55,786 50,289
Stria vascularis and spiral ligament areas in m2.
(1416) (1239) (1200) (893) (769)
Group II (30,947) (14,609) (16,186) (11,491) (11,652)
137,242 90,570 82,073 59,484 51,724
(2377) (15,436) (12,098) (8683) (6756)
A.L.L. Sampaio et al. / International Journal of Pediatric Otorhinolaryngology 74 (2010) 934–938
937
Table 2 Averages of cellular densities of spiral ganglion cells (SD) within the five segments of Rosenthal’s canal in groups I and II. Segments of the Rosenthal canal
Group I
Segment Segment Segment Segment Segment
I II III IV V
14.1 18.3 16.3 15.1 17.2
II (4.6) (4.9) (6.0) (7.7) (6.1)
13.9 19.9 19.5 19.7 18.3
(4.0) (2.8) (2.8) (2.6) (2.8)
Cellular densities in number of cells/0.01 mm2.
3.5. Spiral ganglion cells No young dogs showed apparent loss of spiral ganglion cells. The apparent neuronal losses started by the end of the first year (Fig. 3). Average of the cellular density of spiral ganglion cells (SD) within the five segments of Rosenthal’s canal is shown in Table 2. No statistical differences were found between the sides of the temporal bones in groups I and II or between the gender of the animals (p > 0.05). Average cellular density of spiral ganglion cells within segments III and IV were significantly lower in group I than group II (p = 0.009 and 0.002 respectively). Differences in the rest of the segments were not significant (p > 0.05) (Fig. 2). A moderate negative correlation was found between average density of spiral ganglion cells of segments III and IV and age in group I (Pearson coefficient 0.59 and 0.54 respectively). No correlation with age was found in the other segments of group I or in any of the segments of group II. 3.6. Scarpa’s ganglion cells
Fig. 3. The left saccule of 3-week-old puppy where there is a collapsed saccular wall (S) with amorphous deposit on the atrophic maculae (C). I, Hair cell type I; II, Hair cell type II. H&E. Nomarski microscopy, 40.
Table 3 Average density of saccular types I and II hair cells (SD) in groups I and II. Hair cell type
No morphological abnormality was found in the groups. There was a statistically significant difference in the average number of Scarpa’s ganglion cells; 5911 (1284) in group I compared to 6478 (760) in group II (p = 0.022). Differences in comparisons of side or gender within groups I and II were not statistically significant (p > 0.05). No correlation with age was found in either group. 3.7. Saccule The saccular structure was normal until 14th day dog and started collapsing during the 3rd week. The older dogs showed
Group I
I II Total
II 8.2 (5.2) 6.9 (4.0)
15.1
23.8 (6.0) 23.1 (3.8) 46.9
Cellular densities in number of cells/0.01 mm2.
several level of saccular membrane collapse and otolitic membrane deformity. Average density of saccular types I and II hair cells (SD) in both groups is shown in Table 3. There was no significant difference between the genders or right and left ears in either group (p > 0.05). Average density of saccular types I and II hair cells was significantly lower in group I than in group II (p < 0.0001). No correlation with age was found in either group. 4. Discussion
Fig. 2. This left temporal bone from the 44-week-old dog shows a general loss of spiral ganglion cells in Rosenthal’s canal pointed by arrows. H&E. Light microscopy, 10.
Quantitative evaluation of inner ear structures has been poorly studied in the cochleosaccular histopathologic pattern of deafness. In humans, to our knowledge, we described for the first time a series of temporal bones showing the quantitative aspects of the inferior part of the labyrinth in cochleosaccular dysplasia [1,2]. Two previous studies described histopathological quantitative analysis of the inner ear in Dalmatian dogs. Mair [10] analyzed the spiral ganglion cells of 10 Dalmatian dogs and Branis and Burda [16] reported the results of cochlear reconstruction and hair cell counts of one deaf Dalmatian dog. In none of the previous studies did the authors compare the results to normal dogs and the conclusions were limited since the effects of aging on the inner ear were not considered. We found the areas of the stria vascularis to be significantly lower in all cochlear turns in group I compared to group II. Atrophy of the stria vascularis was reported in all previous studies of cochleosaccular dysplasia in Dalmatian dogs [6–13,16]. The stria vascularis
938
A.L.L. Sampaio et al. / International Journal of Pediatric Otorhinolaryngology 74 (2010) 934–938
median area showed a gradient from the base to the apex of the cochlea in group II. This well-known gradient has been reported in classic morphometric studies of the stria vascularis [17]. This gradient was not observed in group I. This might suggest that the apex was preserved during the development of the disease or that the stria vascularis never developed entirely in all cochlear turns. Group I showed a significantly lower spiral ligament area in the inferior portion of the basal turn than group II. The spiral ligament has been described as normal in previous histopathological and morphometric studies in cochleosaccular dysplasia in humans [1] and in Dalmatian dogs [6–13,16] however, none of these studies performed morphometric analysis of the cellular area. The reduction of the spiral ligament area, only in the lower basal turn, might emphasize the direction of development of the disease, from base to apex, in this species. Interestingly, Reissner’s membrane collapse occurred gradually, starting from the attachment at the limbus at 1 week of age. By the second week, Reissner’s membrane was totally collapsed. Then the complete loss of hair cells is continued and ended up with the loss of spiral ganglion cells in adult animal. Quantitative studies of spiral ganglion cells in Dalmatian dogs with the cochleosaccular histopathologic pattern by Mair [10], Igarashi et al. [9], Anderson et al. [11] and Lurie [13] reported a severe loss of cells and some level of atrophy in the remaining cells. Hudson and Ruben [8] described the ganglion as normal. We found a significant loss of spiral ganglion cells in segments III and IV in group I. In reconstruction of Rosenthal’s canal, segments III and IV in dogs are radially correlated to the upper basal and inferior medial turns. Spiral ganglion cell loss secondary to degeneration of the end organ has been reported in Dalmatian dogs [10]. Webster [18], working with mutant rats, reported a normal count of spiral ganglion cells in the apex of the cochlea, in spite of loss in the rest of the cochlear turns. He suggested the relative preservation of the spiral ganglion cells in the apex of the cochlea resulted from the remaining hair cells in the apex in these animals. Our findings and Webster’s observation support the thesis of relative preservation of the cochlear apex during the development of cochleosaccular dysplasia. Spoendlin [19] studied the cochlear neuroanatomy of several mammalian species and concluded that hair cells from the middle cochlear turn have the highest innervation density within the cochlea. This specialized cochlear region might be very vulnerable to pathogenic agents. Pujol et al. [3] reported very early abnormalities in the spiral ganglion cells in mutant rats with cochleosaccular dysplasia. The authors attributed these findings to the primary action of a gene on spiral ganglion cells, rather than the effect of secondary degeneration. A moderate negative correlation was found between the average density of spiral ganglion cells of segments III and IV and age in group I. However, the coefficients were close to the cutoff point ( 0.59 and 0.54 respectively), and the role of age interfering with the number of spiral ganglion cells in these segments of Rosenthal’s canal can not be clearly assumed. These observations might perhaps suggest a direct effect of the genetic disorder on ganglion cells rather then ganglion cell loss being a result of neuroepithelial degeneration as previously assumed. This new finding might have some implication in the cases of cochlear implantation due to deafness secondary to genetic causes. A quantitative study of Scarpa’s ganglion cells has not been performed previously in dogs having cochleosaccular dysplasia. Igarashi et al. [9] and Hudson and Ruben [8] described the ganglion as being morphologically normal in the dogs studied. We found Scarpa’s ganglion cells significantly lower in group I than in group II. The damage to the end vestibular organ (lack of hair cells in the saccule) might be responsible for the loss of Scarpa’s ganglion cells in our specimens, because of the retrograde mechanism of degeneration. However, the loss of Scarpa’s ganglion cells secondary to the loss of the end organ is not clear in other species.
Furthermore, we did not find correlation between the loss of the neurons and the ages of the animals. We found types I and II hair cells of the saccule to be significantly lower in group I than in group II. Deol [20] in the piroute, shaker-1 and waltzer have shown that saccular morphologic abnormalities developed after the changes in the cochlea. Igarashi et al. [9] reported that changes in the cochlea were more severe than in the saccule in a Dalmatian puppy specimen. They suggested that the saccule, an older structure, is more resistant than the cochlea, the youngest inferior part. Our findings go against these previous reports. In conclusion, this quantitative histologic examination of heredity deafness in Dalmatian dogs shows that the general pattern of cochleosaccular dysplasia is variable, even when only one etiology, the genetic one, is involved and it depends on the age of the animal studied. The most important histopathological findings develop during the first weeks. Nevertheless, the velocity of the process seems to be particular to each animal. Gradual degeneration of inner ear elements, starting in Reissner’s membrane, then complete hair cells loss in parallel with the stria degeneration, collapse of saccular wall and ending up with loss of spiral ganglion cells might indicate that early intervention in the immediate postnatal period might be very crucial to stop the progression of cochleosaccular dysplasia due to the genetic causes.
Acknowledgment The authors would like to thank Capes-Fulbright exchange program. References [1] A.L.L. Sampaio, S. Coreoglu, P. Schachern, et al., Cochleosaccular dysplasia: a morphometric and histopathologic study in a series of temporal bones, Otol. Neurotol. 25 (2004) 530–535. [2] Shin Kariya, S. Cureoglu, A.L.L. Sampaio, et al., Quantitative study of vestibular sensory epithelium in cochleosaccular dysplaisa, Otol. Neurotol. 26 (2005) 495–499. [3] R. Pujol, A. Shnerson, M. Lenoir, et al., Early degeneration of sensory and ganglion cells in the inner ear of mice with uncomplicated genetic deafness (dn): preliminary observation, Hear. Res. 12 (1983) 463–466. [4] M. Rebillard, G. Rebillard, R. Pujol, Variability of hereditary in the white cat I, Physiol. Hear. Res. 5 (1981) 179–187. [5] L.Z. Saunders, The histopathology of congenital deafness in white mink, Pathol. Vet. 2 (1965) 256–263. [6] G.B. Rawiz, Gehirn eines weissen Hundes mit blauen Augen, Morphol. Arb. 6 (1896) 545–553. [7] F. Altmann, Histologic picture of inherited nerve deafness in man and animals, Arch. Otolaryngol. 51 (1950) 852–890. [8] W.R. Hudson, R.J. Ruben, Hereditary deafness in the Dalmatian dog, Arch. Otorhinolaryngol. 75 (1962) 213–219. [9] M. Igarashi, B.R. Alford, R. Saito, et al., Inner ear abnormalities in dogs, Ann. Otol. 81 (1972) 249–255. [10] I.W.S. Mair, Hereditary deafness in Dalmatian dog, Arch. Otorhinolaryngol. 212 (1976) 1–14. [11] H. Anderson, B. Hericson, R.G. Lundquist, et al., Genetic hearing impairment in the Dalmatian dog, Acta. Otolaryngol. (Stockh.) 232 (1968) 1–32. [12] L.G. Johnson, J.E. Hawkins Jr., A.A. Muraski, et al., Vascular anatomy and pathology of the cochlea in Dalmatian dogs, in: A.D.J. De Lorenzo (Ed.), Vascular Disorders and Hearing Defects, University Park Press, Baltimore, 1974, p. 249. [13] M.H. Lurie, The membranous labyrinth in congenitally deaf Collie and Dalmatian dog, Laryngoscope 58 (1948) 279–287. [14] E. Richter, Quantitative study of human Scarpa’s ganglion and vestibular sensory epithelia, Acta Otolaryngol. 90 (1980) 199–208. [15] S.N. Merchant, A method for quantitative assessment of vestibular otopathology, Laryngoscope 109 (1999) 1560–1569. [16] M. Branis, H. Burda, Inner ear structure in deaf and normally hearing Dalmatian dog, J. Comp. Pathol. 95 (1985) 295–299. [17] P.A. Santi, B.N. Lakhani, C. Bingham, The volume density of cells and capillaries of the normal stria vascularis, Hear. Res. 11 (1983) 7–22. [18] D.B. Webster, The spiral ganglion and cochlear nuclei of deafness mice, Hear. Res. 18 (1985) 19–27. [19] H. Spoendlin, Neuroanatomy of the cochlea, in: Z.E. Terhardt (Ed.), Facts and Models in Hearing, 1st ed., Springer, Berlin/Heidelberg/New York, 1974, p. 18. [20] M.S. Deol, The anatomy and development of mutants Pirouette, shaker-1 and waltzer in the mouse, Proc. R. Soc. (Biol.) 145 (1956) 206–213.