Acta Oto-Laryngologica. 2013; 133: 1242–1249
ORIGINAL ARTICLE
Acta Otolaryngol Downloaded from informahealthcare.com by University of Texas at Austin on 11/25/14 For personal use only.
Correlation analysis of genotypes, auditory function, and vestibular size in Chinese children with enlarged vestibular aqueduct syndrome
FEI-FAN ZHAO*, LAN LAN*, DA-YONG WANG, BING HAN, YUE QI, YALI ZHAO, LIANG ZONG, QIAN LI & QIU-JU WANG Department of Otorhinolaryngology/Head and Neck Surgery, Chinese People’s Liberation Army Institute of Otolaryngology, Chinese People’s Liberation Army General Hospital, Beijing, China
Abstract Conclusion: In children with enlarged vestibular aqueduct syndrome (EVAS), their hearing was more related to genotype than VA size, and VA size was related to genotype. Objective: To study genotypes of the SLC26A4 gene, types and levels of hearing loss, and vestibular aqueduct (VA) size in children with EVAS. Methods: A total of 271 children with nonsyndromic sensorineural hearing loss and EVA underwent SLC26A4 gene screening. According to genotype typing, the phenotypes including pure tone average (PTA), distribution of subjects, and diameters of the external aperture and middle portion of the VA, were compared by t test or Pearson’s c2 tests. Further, divided by the dilated level of the VA, subject distribution in different hearing loss levels was compared by Pearson’s c2 test. Results: In all, 66 types of mutations were identified and 2 were novel (c.665G >T and c.1639G >A). Biallelic genotype was found in 207 subjects, monoallelic in 56, and no mutation in 8. The hearing loss was more stable in the subjects with monoallelic mutation than in other genotype groups. An air–bone gap was more frequently found in subjects with biallelic missense mutations than in other groups. The patients with no mutation had the most slightly enlarged VA. There was no dominant correlation between hearing loss level and VA size, and between VA size and different genotypes.
Keywords: SLC26A4, nonsyndromic sensorineural hearing loss, vestibular aqueduct
Introduction Valvassori and Clemis first described the association between enlargement of the vestibular aqueduct (EVA) and sensorineural hearing impairment (SHI) and suggested the term enlarged vestibular aqueduct syndrome (EVAS, MIM 600791). It was considered to be one of the most common forms of deafness and was estimated to account for about 10–20% of all hereditary hearing impairment (HI) [1]. Molecular etiology research mapped the locus to 7q31, SLC26A4 gene. The audiological features of patients with EVAS have been described in detail: the hearing can be normal or impaired from mild to profound, the hearing loss can be from fluctuant to progressive, and conductive hearing loss in low frequencies is present
in some patients. The radiological features have been reported to range from simple EVA to abnormality complexes comprising deformed cochlea, narrow internal auditory canal, and large vestibular canal. To date, more than 240 mutations have been found in the encoding region of the SLC26A4 gene in patients with EVAS. The mutation spectra are distinct in different populations and regions. Many investigators have attempted to discover the correlations between genotypes and phenotypes. Some conclude that EVA is usually associated with profound hearing loss [2]. Some studies suggested that it was the number of mutations, but not the type, that influenced the phenotype, which meant biallelic mutations were more frequently found in patients with bilateral EVA [3].
Correspondence: Qiu-Ju Wang MD PhD, Department of Otorhinolaryngology/Head and Neck Surgery, Chinese People’s Liberation Army Institute of Otolaryngology, Chinese People’s Liberation Army General Hospital, Beijing, China. Fax: +86 10 55499143. E-mail: [email protected] *These authors contributed equally to this work.
(Received 7 May 2013; accepted 30 June 2013) ISSN 0001-6489 print/ISSN 1651-2251 online 2013 Informa Healthcare DOI: 10.3109/00016489.2013.822555
Genotype and phenotype correlations in children with EVAS Respecting the special mutation spectrum with a predominant mutation (c.919-2A >G) in Chinese patients with EVAS, it is necessary to study the correlations between genotype and phenotype. In this paper, we describe the mutation spectrum and clinical, audiological, radiological, and genetic features of a group of 271 children with EVAS. Material and methods
1243
change in threshold of 15 dB or more at any frequency or a 20 dB threshold change in the ABR, with a follow-up of at least 3 months. Stable state was defined as a decline in the audiometric thresholds of less than 5 dB at more than three frequencies over a follow-up period of at least 3 months. Based on pure tone audiometry at each frequency, a symmetrical audiogram was defined if the difference between the left and right ears was less than 10 dB at more than three frequencies.
Acta Otolaryngol Downloaded from informahealthcare.com by University of Texas at Austin on 11/25/14 For personal use only.
Subject recruitment Subjects with nonsyndromic SHI diagnosed with EVAS by temporal bone CT scan were sequentially accrued from the otolaryngology department and genetics unit from 2003 to 2008. Enlargement of the VA was defined when its antero-posterior diameter was 1.5 mm or greater, as measured via CT scans midway between the outer aperture and common crus. Informed consent, blood samples, and clinical evaluations were obtained from all of the participants according to the protocols approved by the Institutional Review Board of the Ethics Committee of the Chinese People’s Liberation Army General Hospital. The personal information comprised a complete history, physical examinations, radiological examinations, and audiometric examinations. A total of 271 subjects with EVAS from 16 provinces underwent SLC26A4 gene screening. The follow-up time was from 11 months to 4 years, and the mean was 19 months. Audiological examination and grouping criteria In 271 subjects that underwent SLC26A4 gene screening, 266 subjects underwent pure tone or behavioral audiometry, acoustic impedance admittance measurements using a Madsen 622-type pure-tone audiometer in a soundproof room. Auditory brainstem response (ABR) was determined using a Smart EP-type evoked potential instrument (Intelligent Hearing Systems, USA) in an electric shielding room, as were the distortion product otoacoustic emissions (DPOAEs); the DPOAEs of 95 children were measured. The infants’ auditory steady-state responses (ASSRs) were evaluated to estimate the hearing threshold. According to the guidelines of European working group, Gendeaf recommendations, average thresholds in the range of 21–40 dB on frequencies from 0.5 to 4 kHz (PTA0.5–4kHz) were defined as mild hearing impairment (HI), in the range of 41–70 dB as moderate HI, in the range of 71–95 dB as severe HI, and > 95 dB as profound HI. According to auditory examination, HI was described as stable or fluctuating. Fluctuation of hearing was defined as a
CT temporal bone measurements Among 266 subjects who underwent SLC26A4 gene screening and the complete set of auditory evaluations, detailed measurement by high-resolution computed tomography (HRCT) was applied in 152 patients. Measurements were made with images enlarged 10–15 times with the use of current workstation software (Synapse; Fuji Film Medical Systems, Stamford, CT, USA). The width of the VA was measured at the operculum and the midpoint. Measurements were taken using the electronic calipers and recorded in millimeters. The largest opercular width or midpoint width on CT image was used for analyses. SLC26A4 gene screening Genomic DNA was extracted using the phenol extraction method. All of the 21 exons of the SLC26A4 gene, including the flanking sequences, were amplified by the polymerase chain reaction (PCR) using genomic DNA, and the primers were designed using online PRIMER 3.0 software. The PCR amplified products were purified with Millipore plate and then sequenced with an ABI 3730 Sequencer (Applied Biosystems). Sequence data were analyzed by aligning with the National Center for Biotechnology Information reference sequence of SLC26A4 (NT_007933) using DNAStar 5.0 and BioEdit software. Mutations or polymorphisms were identified per the reference sequence. Statistical analysis The correlations of data were investigated using the Pearson correlation coefficient, r. As metering data, PTA was compared by Student’s t test. For all analyses, a two-sided p value of 0.05 was considered significant. All statistical calculations were performed using SPSS for Windows, version 13.0 (Lead Technologies Inc., Charlotte, NC, USA). Using the HI classification of mild, moderate, severe, and profound, Pearson’s c2 testing was
1244
F.-F. Zhao et al. 1991C > T (2) 1990G > A (2)
281C > T (5)
1
2
3
4
916dupG (4)
1336C > T (5)
589G > A (3)
907G > C (3)
1318A > T (2)
5
7
349delC (2)
6
8
9
626G > A (2) 665G > T (2) 919-2A > G (266)
10
11
12
1975G > C (16) 1692duA (3)
13
14
15
16
2027T > A (10) 2168A > G (57)
17
18
1174A > T (9) 1343C > T (4)
1586T > G (2)
1226G > A (10)
1594A > C (3) 1707 + 5G > A (4)
19
20
21
1746delG (2)
Acta Otolaryngol Downloaded from informahealthcare.com by University of Texas at Austin on 11/25/14 For personal use only.
1229C > T (11)
Figure 1. The allelic frequencies of mutations in Chinese children with enlarged vestibular aqueduct (EVA). This indicates the distribution of mutations in 21 exons of the SLC26A4 gene. The 26 mutations were present more than twice in the homozygous state or in the heterozygous state combined with other mutations, and the numbers in brackets are the allelic frequencies in 271 subjects.
performed to determine whether the frequency distribution of HI was randomized in different genotype groups (biallelic mutations, monoallelic mutation, and no mutation). Then, Fisher’s exact probability testing with 2 2 contingency tables of appropriately dichotomized data was performed to determine the most frequent class of HI in each group.
Results Identification of mutations and genotype typing In all, 66 types of mutations were found in 271 subjects, and 2 were novel. There were 24 mutations that appeared more than twice (Figure 1) and 42 with allelic frequency of only once (data not shown). The 66 mutations comprised missense mutations, nonsense mutations, and frameshift and splice site mutations. Two heterozygotes carried the genotype of c.665G >T and one heterozygote carried c.1639A >G. These two mutations were not found in dbSNP and not reported. c.665G >T caused replacement of glycine at position 222 with valine (p.G222V), and c.1639A >G caused replacement of isoleucine at position 547 with valine (p.I547V). The SIFT software predicted the functional changes of p.G222V to be damaging and p.I547V to be tolerated. Both the sequences are evolutionarily highly conserved in the pendrin protein of humans, mice, rats, rhesus, dog, chick, opossum, zebrafish, and X_tropicalis based on the University of California Santa Cruz Genome Bioinformatics (http://genome.ucsc.edu). Among 271 subjects, there were 207 with biallelic mutations in the SLC26A4 gene. These 207 subjects were further classified into those with 2 alleles of truncating mutations (n = 82), with 2 alleles of heterozygous truncating mutations (n = 91), and with 2 alleles of missense mutations (n = 31). There were
56 monoallelic mutation carriers, and no mutation was found in 8 subjects. Results of audiological and radiological examinations In most subjects, hearing loss level was unsymmetrical; the better laterality was taken as the reference, and hearing thresholds were evaluated in 266 subjects by PTA or auditory stable stimulated reaction (ASSR). The average hearing thresholds of 32 subjects with mild hearing loss, 73 subjects with moderate hearing loss, 102 subjects with severe hearing loss, and 59 subjects with profound hearing loss were 30, 67, 84, and 103 dB, respectively. The mean value of the average PTA was compared among biallelic, monoallelic, and no mutation groups. Among the 266 subjects, the width of the VA was measured at the operculum and the midpoint on axial CT scan in 152 subjects. The midpoint mean was 2.3 mm, ranging from 1 to 5.8 mm, and the operculum mean was 3.5 mm, ranging from 1.9 to 8 mm. EVA with normal cochlea was found in 109 subjects and EVA combined with Mondini’s deformity in the cochlea was found in 43 subjects. Correlation between genotype and hearing impairment There was no significant difference of PTA mean in genotype groups (all p values > 0.05). The hearing loss was more stable in the subjects with monoallelic mutation than in other groups (p < 0.05). A total of 76 subjects were tested by ASSR. An air–bone gap was found in 149 of 190 subjects who underwent pure tone audiometry, ranging from 5 to 95 dB, and the mean was 55 dB. The air–bone gap was more frequent in subjects with biallelic missense mutations than those with biallelic truncating, heterozygous truncating, or monoallelic mutation (all p values < 0.05). p values were calculated by c2 or Fisher’s exact test for
Numerical value
Genotype and phenotype correlations in children with EVAS Cases with fluctuating hearing 100 90 80 70 42 60 50 40 30 40 20 10 0
Stable hearing
51 38 40
Numerical value
100 90 80 70 60 50 40 30 20 10 0
17
3 4
Subjects with no Subjects with Subjects with mutations biallelic missense monoallelic mutations mutations
Descending audiogram
Ascending audiogram
4
5
3
67
62 15
20
Subjects with biallelic truncating mutations
Subjects with heterozygous truncating mutations
Cases with air-bone gap 100 90 80 70 60 50 40 30 20 10 0
18 13
Subjects with heterozygous truncating mutations
Cases with flat audiogram
Numerical value
Acta Otolaryngol Downloaded from informahealthcare.com by University of Texas at Austin on 11/25/14 For personal use only.
Subjects with biallelic truncating mutations
1245
1 26 4
41 11
1 6
Subjects with Subjects with Subjects with no mutations biallelic missense monoallelic mutations mutations
Not with air-bone gap
Cases tested by ASSR
23 24 16 17
16 7 52
41
Subjects with biallelic truncating mutations
11 1 19
Subjects with Subjects with heterozygous biallelic missense truncating mutations mutations
32
2 5
Subjects with no Subjects with mutations monoallelic mutations
Figure 2. Comparison of hearing impairment levels among different genotype groups. The x-axis shows the genotype typing, and the yaxis shows the numbers of subjects in each genotype group, composed of numbers of patients with distinct hearing characteristics, which included stable or fluctuating hearing (a), audiogram shapes (b), and the presence of an air–bone gap (c).
categorical variables and ANOVA for continuous variables (Figure 2). Correlation between genotype and VA size The widths were compared according to different genotypes, and the midpoint diameter and operculum diameter of VA in subjects with no mutation were both narrower than other groups by Fisher’s test (all p values < 0.05) (Figure 3). Correlation between VA size and hearing impairment The distribution of phenotypes of hearing loss had no significant difference in different groups divided by diameters of VA (Figure 4).
Discussion The current radiographic diagnostic criterion for EVA is a VA larger than 1.5 mm at the midpoint, although Cincinnati criteria proposed the definition as VAs with midpoint width greater than 1.0 mm or opercular width greater than 2.0 mm [4]. Further, the subjects carrying biallelic mutations in the SLC26A4 gene have Pendred syndrome, i.e. usually hereditary hearing loss and large VA, sometimes goiter after adolescence. The mutation of this gene can cause two autosomal recessive disorders, nonsyndromic hereditary hearing loss DFNB4 (MIM 600791) and Pendred syndrome (MIM 274600), and the distinction is the presence or absence of goiter. The clinical distinction between Pendred syndrome and DFNB4 can
1246
F.-F. Zhao et al. Midpoint diameter of VA > 2.3 mm
Midpoint diameter of VA > 2.3 mm
Total amount
30
28
27
25
22
21
19
20
16
15 8
10
6
5
5 0
0 Subjects with heterozygous truncating mutations
Subjects with Subjects with biallelic missense monoallelic mutations mutations
Total amount
30 25
Subjects with mutations no mutations
Operculum diameter of VA > 3.5 mm
Operculum diameter of VA > 3.5 mm 35
30 25 21
22
20
17
18
15 10
6
8
5
5
0
0 Subjects with biallelic truncating mutations
Total amount
Acta Otolaryngol Downloaded from informahealthcare.com by University of Texas at Austin on 11/25/14 For personal use only.
Subjects with biallelic truncating mutations
45 40 35 30 25 20 15 10 5 0
Normal cochlear 42
Subjects with heterozygous truncating mutations
Subjects with Subjects with biallelic missense monoallelic mutations mutations
Subjects with mutations no mutations
Hypoplastic cochlear
30 24 13
13 8
11 6
5 0
Subjects with biallelic truncating mutations
Subjects with heterozygous truncating mutations
Subjects with Subjects with biallelic missense monoallelic mutations mutations
Subjects with mutations no mutations
Figure 3. Comparison of size of vestibular aqueduct (VA) among distinct genotype groups. The x-axis shows the genotype typing and the yaxis shows the numbers of subjects in each genotype group, composed of numbers of patients with distinct radiological characteristics, which included midpoint diameter of VA (a), operculum diameter of vestibular diameter (b), and cochlea formation (c).
be difficult to make because the penetrance of goiter phenotype is variable and not usually evident until adolescence. Although some authors still prefer them to be distinct entities [5], more and more investigators regard them as identical. The goiter was seldom reported in Chinese subjects. The possible reason was probably the iodine policy of ‘Universal Salt Iodination, USI’ in China. The daily iodine intake of adult Chinese subjects is about 220–1960 mg and is much higher than that in Europe [6]. The incidence of goiter correlates with the iodide intake was 1/14 in Japan and 8/101 in Taiwan. Iodine deficiency causes goiter, and sufficient iodine intake may make up the deficit.
The frequency of biallelic SLC26A4 mutations in deaf patients varied among different populations, as 3.5–5% mainly in Caucasian populations [7] and around 8.8–12.5% in Middle Eastern and Asian patients [8,9]. Biallelic mutation genotype of SLC26A4 accounted for 30% in nonsyndromic hearing loss with EVA and 60% in subjects with Pendred syndrome [10]. The ratio was 76.4% (207/271) in the present study. SLC26A4 gene displays a scattering mutation spectrum with more than 240 types of mutation being reported in the world [1]. In the Caucasoid populations, three frequent mutations, L236P (16%), T416P (15%), and IVS8+1G >A (14%), account for nearly half of all PDS mutant
Genotype and phenotype correlations in children with EVAS Cases with fluctuating hearing 80 70 60 50 40 25 30 20 10 19 0 1.5−2 (44)
1247
Stable hearing
41
18
27 2−3 (68)
8
11 3
3−4 (26)
> 4 (14)
Acta Otolaryngol Downloaded from informahealthcare.com by University of Texas at Austin on 11/25/14 For personal use only.
Mid-portion diameter (mm) Cases with flat audiogram
Cases with flat audiogram
Ascending audiogram
70 3
60 50 40
4 16
30 20
19
10 0
1
35 5 1.5−2 (44)
14
6
2−3 (68)
3−4 (26)
1 12 1 > 4 (14)
Mid-portion diameter (mm) Cases with air-bone gap 80 70 60 50 40 30 20 10 0
Not with air-bone gap
Cases tested by ASSR
14 8 11 6
46
12 3 11
2 2 12
2−3 (68)
3−4 (26)
> 4 (14)
27 1.5−2 (44)
Mid-portion diameter (mm) Figure 4. Comparison of levels of hearing impairment among distinct VA size groups. The x-axis shows the vestibular size typing by midpoint diameter, and the y-axis shows the numbers of subjects in different size groups, composed of numbers of patients with distinct hearing characteristics, which included stable or fluctuating hearing (a), audiogram shapes (b), and the presence or absence of an air–bone gap (c).
alleles [11]. p.H723R was a common allele in the Korean population, while p.L676Q was common in the Mongolian population [1]. H723R accounted for 53% of PDS mutations in Japanese subjects [12]. This was the second most frequent mutation in the Chinese population, which was present mostly in heterozygous and homozygous c.919-2A >G forms. However, the incidence of this mutation was found to be low in Western countries and has been identified in only four families [13]. This difference suggested the existence of a common ancestor (founder effect) rather than a mutational hot spot [1]. Two novel mutations were found in the present study. The mutation c.665G >T (p.G222V) was predicted to be damaging to pendrin by SIFT software, while the effect of p.I547V (c.1639A >G) on
pendrin was tolerated. In consideration of both the sequences involving the two mutations being highly conserved in multiple species, both changes were supposed to jeopardize the protein function. Moreover, c.665G >T was found in two unrelated subjects, while c.1639A >G was found only once. Thus, based on the combined evidence, c.665G >T was thought to be a mutation, while c.1639A >G was more like a variation. Some studies showed that different kinds of mutations could produce different phenotypes [14]. But there were at least 19 shared mutations between Pendred syndrome, and autosomal recessive hearing loss associated with EVA has been reported [15]. Wu et al. found hearing loss in subjects with the biallelic genotype to be more severe than in subjects
Acta Otolaryngol Downloaded from informahealthcare.com by University of Texas at Austin on 11/25/14 For personal use only.
1248
F.-F. Zhao et al.
with the monoallelic genotype [16]. The present study also showed no distinct difference of hearing loss level in different genotypes, not only among subjects with biallelic mutation genotype, monoallelic, and no mutation, but also among subjects with biallelic truncating mutation genotype, biallelic heterozygous truncating, and biallelic missense. However, we found that the subjects with monoallelic missense mutations showed more stable hearing loss than those with other genotypes, which could be explained by the true genotype effect or the interference of unstable hearing status because of young ages. Some studies found a significant difference in mutation distribution between patients with EVA and those with Mondini’s deformity. In addition, bilateral malformations are also more frequently associated with the Mondini’s deformity phenotype group as compared with the EVA phenotype [17], but most studies did not show correlations between hearing loss and aqueduct size [18]. In addition, there were more than 31 shared mutations [8,19]. We found that there was no significant difference in the distribution of hearing loss level among distinct genotypes. Also, the midpoint diameter and operculum diameter of VA in subjects with no mutation were narrower than other genotype groups, which implied an effect of SLC26A4 gene mutations on VA size. The air–bone conduction gap in low frequencies was reported to be discovered in subjects with EVAS. A large VA may act as a third mobile window in the inner ear, resulting in an air–bone gap at low frequencies [20]. Our data showed that 149 of 190 subjects who underwent pure tone audiometry had an airbone gap, but no difference in distribution of subjects with air–bone gap was identified according to the dilated level of the VA. In addition, the air–bone gap was more frequent in subjects with biallelic missense mutations than in those with other genotypes. This finding implies that the air–bone conduction gap was more related to genotypes than the VA size. This result suggests that the membranous structures in the VA should be explored rather than bone structures. In other words, the endolymphatic sac and the intracellular ion concentration in the formation of the air–bone gap should not be ignored. Although we found some correlations between genotype and phenotype, this is far from proving direct correlation between the mutation genotype and hearing loss level or EVA size. Acknowledgments We thank the patients and their families for their cooperation during this work. This work was supported by grants from the National Natural Science
Foundation of China, Major Project, No. 81120108009 and National Natural Science Foundation of Youth Science Foundation No.81100719. Declaration of interest: The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the paper.
References [1] Park HJ, Shaukat S, Liu XZ, Hahn SH, Naz S, Ghosh M, et al. Origins and frequencies of SLC26A4 (PDS) mutations in east and south Asians: global implications for the epidemiology of deafness. J Med Genet 2003;40:242–8. [2] Harnsberger HR, Dahlen RT, Shelton C, Gray SD, Parkin JL. Advanced techniques in magnetic resonance imaging in the evaluation of the large endolymphatic duct and sac syndrome. Laryngoscope 1995;105:1037–42. [3] King KA, Choi BY, Zalewski C, Madeo AC, Manichaikul A, Pryor SP, et al. SLC26A4 genotype, but not cochlear radiologic structure, is correlated with hearing loss in ears with an enlarged vestibular aqueduct. Laryngoscope 2009;120:384– 9. [4] Boston M, Halsted M, Meinzen-Derr J, Bean J, Vijayasekaran S, Arjmand E, et al. The large vestibular aqueduct: a new definition based on audiologic and computed tomography correlation. Otolaryngol Head Neck Surg 2007;136:972–7. [5] Pryor SP, Madeo AC, Reynolds JC, Sarlis NJ, Arnos KS, Nance WE, et al. SLC26A4/PDS genotype-phenotype correlation in hearing loss with enlargement of the vestibular aqueduct (EVA): evidence that Pendred syndrome and non-syndromic EVA are distinct clinical and genetic entities. J Med Genet 2005;42:159–65. [6] Delange F. Iodine deficiency in Europe and its consequences: an update. Eur J Nucl Med Mol Imaging 2002; 29:S404–16. [7] Madden C, Halsted M, Meinzen-Derr J, Bardo D, Boston M, Arjmand E, et al. The influence of mutations in the SLC26A4 gene on the temporal bone in a population with enlarged vestibular aqueduct. Arch Otolaryngol Head Neck Surg 2007;133:162–8. [8] Wang QJ, Zhao YL, Rao SQ, Guo YF, Yuan H, Zong L, et al. A distinct spectrum of SLC26A4 mutations in patients with enlarged vestibular aqueduct in China. Clin Genet 2007;72: 245–54. [9] Wu CC, Lin SY, Su YN, Fang MY, Chen SU, Hsu CJ. Preimplantation genetic diagnosis (embryo screening) for enlarged vestibular aqueduct due to SLC26A4 mutation. Audiol Neurootol 2010;15:311–17. [10] Pera A, Villamar M, Vinuela A, Gandia M, Meda C, Moreno F, et al. A mutational analysis of the SLC26A4 gene in Spanish hearing-impaired families provides new insights into the genetic causes of Pendred syndrome and DFNB4 hearing loss. Eur J Hum Genet 2008;16: 888–96. [11] Campbell C, Cucci RA, Prasad S, Green GE, Edeal JB, Galer CE, et al. Pendred syndrome, DFNB4, and PDS/ SLC26A4 identification of eight novel mutations and possible genotype-phenotype correlations. Hum Mutat 2001;17: 403–11. [12] Usami S, Abe S, Weston MD, Shinkawa H, Van Camp G, Kimberling WJ. Non-syndromic hearing loss associated with
Genotype and phenotype correlations in children with EVAS
[13]
[14]
Acta Otolaryngol Downloaded from informahealthcare.com by University of Texas at Austin on 11/25/14 For personal use only.
[15]
[16]
enlarged vestibular aqueduct is caused by PDS mutations. Hum Genet 1999;104:188–92. Van Hauwe P, Everett LA, Coucke P, Scott DA, Kraft ML, Ris-Stalpers C, et al. Two frequent missense mutations in Pendred syndrome. Hum Mol Genet 1998;7:1099–104. Choi BY, Stewart AK, Madeo AC, Pryor SP, Lenhard S, Kittles R, et al. Hypo-functional SLC26A4 variants associated with nonsyndromic hearing loss and enlargement of the vestibular aqueduct: genotype-phenotype correlation or coincidental polymorphisms? Hum Mutat 2009;30:599–608. Tsukamoto K, Suzuki H, Harada D, Namba A, Abe S, Usami S. Distribution and frequencies of PDS (SLC26A4) mutations in Pendred syndrome and nonsyndromic hearing loss associated with enlarged vestibular aqueduct: a unique spectrum of mutations in Japanese. Eur J Hum Genet 2003; 11:916–22. Wu CC, Lu YC, Chen PJ, Yeh PL, Su YN, Hwu WL, et al. Phenotypic analyses and mutation screening of the
[17]
[18]
[19]
[20]
1249
SLC26A4 and FOXI1 genes in 101 Taiwanese families with bilateral nonsyndromic enlarged vestibular aqueduct (DFNB4) or Pendred syndrome. Audiol Neurootol 2009; 15:57–66. Azaiez H, Yang T, Prasad S, Sorensen JL, Nishimura CJ, Kimberling WJ, et al. Genotype-phenotype correlations for SLC26A4-related deafness. Hum Genet 2007;122:451–7. Gopen Q, Zhou G, Whittemore K, Kenna M. Enlarged vestibular aqueduct: review of controversial aspects. Laryngoscope 2011;121:1971–8. Iwasaki S, Tsukamoto K, Usami S, Misawa K, Mizuta K, Mineta H. Association of SLC26A4 mutations with clinical features and thyroid function in deaf infants with enlarged vestibular aqueduct. J Hum Genet 2006;51:805–10. Merchant SN, Nakajima HH, Halpin C, Nadol JB Jr, Lee DJ, Innis WP, et al. Clinical investigation and mechanism of air-bone gaps in large vestibular aqueduct syndrome. Ann Otol Rhinol Laryngol 2007;116:532–41.
ORIGINAL ARTICLE
Acta Otolaryngol Downloaded from informahealthcare.com by University of Texas at Austin on 11/25/14 For personal use only.
Correlation analysis of genotypes, auditory function, and vestibular size in Chinese children with enlarged vestibular aqueduct syndrome
FEI-FAN ZHAO*, LAN LAN*, DA-YONG WANG, BING HAN, YUE QI, YALI ZHAO, LIANG ZONG, QIAN LI & QIU-JU WANG Department of Otorhinolaryngology/Head and Neck Surgery, Chinese People’s Liberation Army Institute of Otolaryngology, Chinese People’s Liberation Army General Hospital, Beijing, China
Abstract Conclusion: In children with enlarged vestibular aqueduct syndrome (EVAS), their hearing was more related to genotype than VA size, and VA size was related to genotype. Objective: To study genotypes of the SLC26A4 gene, types and levels of hearing loss, and vestibular aqueduct (VA) size in children with EVAS. Methods: A total of 271 children with nonsyndromic sensorineural hearing loss and EVA underwent SLC26A4 gene screening. According to genotype typing, the phenotypes including pure tone average (PTA), distribution of subjects, and diameters of the external aperture and middle portion of the VA, were compared by t test or Pearson’s c2 tests. Further, divided by the dilated level of the VA, subject distribution in different hearing loss levels was compared by Pearson’s c2 test. Results: In all, 66 types of mutations were identified and 2 were novel (c.665G >T and c.1639G >A). Biallelic genotype was found in 207 subjects, monoallelic in 56, and no mutation in 8. The hearing loss was more stable in the subjects with monoallelic mutation than in other genotype groups. An air–bone gap was more frequently found in subjects with biallelic missense mutations than in other groups. The patients with no mutation had the most slightly enlarged VA. There was no dominant correlation between hearing loss level and VA size, and between VA size and different genotypes.
Keywords: SLC26A4, nonsyndromic sensorineural hearing loss, vestibular aqueduct
Introduction Valvassori and Clemis first described the association between enlargement of the vestibular aqueduct (EVA) and sensorineural hearing impairment (SHI) and suggested the term enlarged vestibular aqueduct syndrome (EVAS, MIM 600791). It was considered to be one of the most common forms of deafness and was estimated to account for about 10–20% of all hereditary hearing impairment (HI) [1]. Molecular etiology research mapped the locus to 7q31, SLC26A4 gene. The audiological features of patients with EVAS have been described in detail: the hearing can be normal or impaired from mild to profound, the hearing loss can be from fluctuant to progressive, and conductive hearing loss in low frequencies is present
in some patients. The radiological features have been reported to range from simple EVA to abnormality complexes comprising deformed cochlea, narrow internal auditory canal, and large vestibular canal. To date, more than 240 mutations have been found in the encoding region of the SLC26A4 gene in patients with EVAS. The mutation spectra are distinct in different populations and regions. Many investigators have attempted to discover the correlations between genotypes and phenotypes. Some conclude that EVA is usually associated with profound hearing loss [2]. Some studies suggested that it was the number of mutations, but not the type, that influenced the phenotype, which meant biallelic mutations were more frequently found in patients with bilateral EVA [3].
Correspondence: Qiu-Ju Wang MD PhD, Department of Otorhinolaryngology/Head and Neck Surgery, Chinese People’s Liberation Army Institute of Otolaryngology, Chinese People’s Liberation Army General Hospital, Beijing, China. Fax: +86 10 55499143. E-mail: [email protected] *These authors contributed equally to this work.
(Received 7 May 2013; accepted 30 June 2013) ISSN 0001-6489 print/ISSN 1651-2251 online 2013 Informa Healthcare DOI: 10.3109/00016489.2013.822555
Genotype and phenotype correlations in children with EVAS Respecting the special mutation spectrum with a predominant mutation (c.919-2A >G) in Chinese patients with EVAS, it is necessary to study the correlations between genotype and phenotype. In this paper, we describe the mutation spectrum and clinical, audiological, radiological, and genetic features of a group of 271 children with EVAS. Material and methods
1243
change in threshold of 15 dB or more at any frequency or a 20 dB threshold change in the ABR, with a follow-up of at least 3 months. Stable state was defined as a decline in the audiometric thresholds of less than 5 dB at more than three frequencies over a follow-up period of at least 3 months. Based on pure tone audiometry at each frequency, a symmetrical audiogram was defined if the difference between the left and right ears was less than 10 dB at more than three frequencies.
Acta Otolaryngol Downloaded from informahealthcare.com by University of Texas at Austin on 11/25/14 For personal use only.
Subject recruitment Subjects with nonsyndromic SHI diagnosed with EVAS by temporal bone CT scan were sequentially accrued from the otolaryngology department and genetics unit from 2003 to 2008. Enlargement of the VA was defined when its antero-posterior diameter was 1.5 mm or greater, as measured via CT scans midway between the outer aperture and common crus. Informed consent, blood samples, and clinical evaluations were obtained from all of the participants according to the protocols approved by the Institutional Review Board of the Ethics Committee of the Chinese People’s Liberation Army General Hospital. The personal information comprised a complete history, physical examinations, radiological examinations, and audiometric examinations. A total of 271 subjects with EVAS from 16 provinces underwent SLC26A4 gene screening. The follow-up time was from 11 months to 4 years, and the mean was 19 months. Audiological examination and grouping criteria In 271 subjects that underwent SLC26A4 gene screening, 266 subjects underwent pure tone or behavioral audiometry, acoustic impedance admittance measurements using a Madsen 622-type pure-tone audiometer in a soundproof room. Auditory brainstem response (ABR) was determined using a Smart EP-type evoked potential instrument (Intelligent Hearing Systems, USA) in an electric shielding room, as were the distortion product otoacoustic emissions (DPOAEs); the DPOAEs of 95 children were measured. The infants’ auditory steady-state responses (ASSRs) were evaluated to estimate the hearing threshold. According to the guidelines of European working group, Gendeaf recommendations, average thresholds in the range of 21–40 dB on frequencies from 0.5 to 4 kHz (PTA0.5–4kHz) were defined as mild hearing impairment (HI), in the range of 41–70 dB as moderate HI, in the range of 71–95 dB as severe HI, and > 95 dB as profound HI. According to auditory examination, HI was described as stable or fluctuating. Fluctuation of hearing was defined as a
CT temporal bone measurements Among 266 subjects who underwent SLC26A4 gene screening and the complete set of auditory evaluations, detailed measurement by high-resolution computed tomography (HRCT) was applied in 152 patients. Measurements were made with images enlarged 10–15 times with the use of current workstation software (Synapse; Fuji Film Medical Systems, Stamford, CT, USA). The width of the VA was measured at the operculum and the midpoint. Measurements were taken using the electronic calipers and recorded in millimeters. The largest opercular width or midpoint width on CT image was used for analyses. SLC26A4 gene screening Genomic DNA was extracted using the phenol extraction method. All of the 21 exons of the SLC26A4 gene, including the flanking sequences, were amplified by the polymerase chain reaction (PCR) using genomic DNA, and the primers were designed using online PRIMER 3.0 software. The PCR amplified products were purified with Millipore plate and then sequenced with an ABI 3730 Sequencer (Applied Biosystems). Sequence data were analyzed by aligning with the National Center for Biotechnology Information reference sequence of SLC26A4 (NT_007933) using DNAStar 5.0 and BioEdit software. Mutations or polymorphisms were identified per the reference sequence. Statistical analysis The correlations of data were investigated using the Pearson correlation coefficient, r. As metering data, PTA was compared by Student’s t test. For all analyses, a two-sided p value of 0.05 was considered significant. All statistical calculations were performed using SPSS for Windows, version 13.0 (Lead Technologies Inc., Charlotte, NC, USA). Using the HI classification of mild, moderate, severe, and profound, Pearson’s c2 testing was
1244
F.-F. Zhao et al. 1991C > T (2) 1990G > A (2)
281C > T (5)
1
2
3
4
916dupG (4)
1336C > T (5)
589G > A (3)
907G > C (3)
1318A > T (2)
5
7
349delC (2)
6
8
9
626G > A (2) 665G > T (2) 919-2A > G (266)
10
11
12
1975G > C (16) 1692duA (3)
13
14
15
16
2027T > A (10) 2168A > G (57)
17
18
1174A > T (9) 1343C > T (4)
1586T > G (2)
1226G > A (10)
1594A > C (3) 1707 + 5G > A (4)
19
20
21
1746delG (2)
Acta Otolaryngol Downloaded from informahealthcare.com by University of Texas at Austin on 11/25/14 For personal use only.
1229C > T (11)
Figure 1. The allelic frequencies of mutations in Chinese children with enlarged vestibular aqueduct (EVA). This indicates the distribution of mutations in 21 exons of the SLC26A4 gene. The 26 mutations were present more than twice in the homozygous state or in the heterozygous state combined with other mutations, and the numbers in brackets are the allelic frequencies in 271 subjects.
performed to determine whether the frequency distribution of HI was randomized in different genotype groups (biallelic mutations, monoallelic mutation, and no mutation). Then, Fisher’s exact probability testing with 2 2 contingency tables of appropriately dichotomized data was performed to determine the most frequent class of HI in each group.
Results Identification of mutations and genotype typing In all, 66 types of mutations were found in 271 subjects, and 2 were novel. There were 24 mutations that appeared more than twice (Figure 1) and 42 with allelic frequency of only once (data not shown). The 66 mutations comprised missense mutations, nonsense mutations, and frameshift and splice site mutations. Two heterozygotes carried the genotype of c.665G >T and one heterozygote carried c.1639A >G. These two mutations were not found in dbSNP and not reported. c.665G >T caused replacement of glycine at position 222 with valine (p.G222V), and c.1639A >G caused replacement of isoleucine at position 547 with valine (p.I547V). The SIFT software predicted the functional changes of p.G222V to be damaging and p.I547V to be tolerated. Both the sequences are evolutionarily highly conserved in the pendrin protein of humans, mice, rats, rhesus, dog, chick, opossum, zebrafish, and X_tropicalis based on the University of California Santa Cruz Genome Bioinformatics (http://genome.ucsc.edu). Among 271 subjects, there were 207 with biallelic mutations in the SLC26A4 gene. These 207 subjects were further classified into those with 2 alleles of truncating mutations (n = 82), with 2 alleles of heterozygous truncating mutations (n = 91), and with 2 alleles of missense mutations (n = 31). There were
56 monoallelic mutation carriers, and no mutation was found in 8 subjects. Results of audiological and radiological examinations In most subjects, hearing loss level was unsymmetrical; the better laterality was taken as the reference, and hearing thresholds were evaluated in 266 subjects by PTA or auditory stable stimulated reaction (ASSR). The average hearing thresholds of 32 subjects with mild hearing loss, 73 subjects with moderate hearing loss, 102 subjects with severe hearing loss, and 59 subjects with profound hearing loss were 30, 67, 84, and 103 dB, respectively. The mean value of the average PTA was compared among biallelic, monoallelic, and no mutation groups. Among the 266 subjects, the width of the VA was measured at the operculum and the midpoint on axial CT scan in 152 subjects. The midpoint mean was 2.3 mm, ranging from 1 to 5.8 mm, and the operculum mean was 3.5 mm, ranging from 1.9 to 8 mm. EVA with normal cochlea was found in 109 subjects and EVA combined with Mondini’s deformity in the cochlea was found in 43 subjects. Correlation between genotype and hearing impairment There was no significant difference of PTA mean in genotype groups (all p values > 0.05). The hearing loss was more stable in the subjects with monoallelic mutation than in other groups (p < 0.05). A total of 76 subjects were tested by ASSR. An air–bone gap was found in 149 of 190 subjects who underwent pure tone audiometry, ranging from 5 to 95 dB, and the mean was 55 dB. The air–bone gap was more frequent in subjects with biallelic missense mutations than those with biallelic truncating, heterozygous truncating, or monoallelic mutation (all p values < 0.05). p values were calculated by c2 or Fisher’s exact test for
Numerical value
Genotype and phenotype correlations in children with EVAS Cases with fluctuating hearing 100 90 80 70 42 60 50 40 30 40 20 10 0
Stable hearing
51 38 40
Numerical value
100 90 80 70 60 50 40 30 20 10 0
17
3 4
Subjects with no Subjects with Subjects with mutations biallelic missense monoallelic mutations mutations
Descending audiogram
Ascending audiogram
4
5
3
67
62 15
20
Subjects with biallelic truncating mutations
Subjects with heterozygous truncating mutations
Cases with air-bone gap 100 90 80 70 60 50 40 30 20 10 0
18 13
Subjects with heterozygous truncating mutations
Cases with flat audiogram
Numerical value
Acta Otolaryngol Downloaded from informahealthcare.com by University of Texas at Austin on 11/25/14 For personal use only.
Subjects with biallelic truncating mutations
1245
1 26 4
41 11
1 6
Subjects with Subjects with Subjects with no mutations biallelic missense monoallelic mutations mutations
Not with air-bone gap
Cases tested by ASSR
23 24 16 17
16 7 52
41
Subjects with biallelic truncating mutations
11 1 19
Subjects with Subjects with heterozygous biallelic missense truncating mutations mutations
32
2 5
Subjects with no Subjects with mutations monoallelic mutations
Figure 2. Comparison of hearing impairment levels among different genotype groups. The x-axis shows the genotype typing, and the yaxis shows the numbers of subjects in each genotype group, composed of numbers of patients with distinct hearing characteristics, which included stable or fluctuating hearing (a), audiogram shapes (b), and the presence of an air–bone gap (c).
categorical variables and ANOVA for continuous variables (Figure 2). Correlation between genotype and VA size The widths were compared according to different genotypes, and the midpoint diameter and operculum diameter of VA in subjects with no mutation were both narrower than other groups by Fisher’s test (all p values < 0.05) (Figure 3). Correlation between VA size and hearing impairment The distribution of phenotypes of hearing loss had no significant difference in different groups divided by diameters of VA (Figure 4).
Discussion The current radiographic diagnostic criterion for EVA is a VA larger than 1.5 mm at the midpoint, although Cincinnati criteria proposed the definition as VAs with midpoint width greater than 1.0 mm or opercular width greater than 2.0 mm [4]. Further, the subjects carrying biallelic mutations in the SLC26A4 gene have Pendred syndrome, i.e. usually hereditary hearing loss and large VA, sometimes goiter after adolescence. The mutation of this gene can cause two autosomal recessive disorders, nonsyndromic hereditary hearing loss DFNB4 (MIM 600791) and Pendred syndrome (MIM 274600), and the distinction is the presence or absence of goiter. The clinical distinction between Pendred syndrome and DFNB4 can
1246
F.-F. Zhao et al. Midpoint diameter of VA > 2.3 mm
Midpoint diameter of VA > 2.3 mm
Total amount
30
28
27
25
22
21
19
20
16
15 8
10
6
5
5 0
0 Subjects with heterozygous truncating mutations
Subjects with Subjects with biallelic missense monoallelic mutations mutations
Total amount
30 25
Subjects with mutations no mutations
Operculum diameter of VA > 3.5 mm
Operculum diameter of VA > 3.5 mm 35
30 25 21
22
20
17
18
15 10
6
8
5
5
0
0 Subjects with biallelic truncating mutations
Total amount
Acta Otolaryngol Downloaded from informahealthcare.com by University of Texas at Austin on 11/25/14 For personal use only.
Subjects with biallelic truncating mutations
45 40 35 30 25 20 15 10 5 0
Normal cochlear 42
Subjects with heterozygous truncating mutations
Subjects with Subjects with biallelic missense monoallelic mutations mutations
Subjects with mutations no mutations
Hypoplastic cochlear
30 24 13
13 8
11 6
5 0
Subjects with biallelic truncating mutations
Subjects with heterozygous truncating mutations
Subjects with Subjects with biallelic missense monoallelic mutations mutations
Subjects with mutations no mutations
Figure 3. Comparison of size of vestibular aqueduct (VA) among distinct genotype groups. The x-axis shows the genotype typing and the yaxis shows the numbers of subjects in each genotype group, composed of numbers of patients with distinct radiological characteristics, which included midpoint diameter of VA (a), operculum diameter of vestibular diameter (b), and cochlea formation (c).
be difficult to make because the penetrance of goiter phenotype is variable and not usually evident until adolescence. Although some authors still prefer them to be distinct entities [5], more and more investigators regard them as identical. The goiter was seldom reported in Chinese subjects. The possible reason was probably the iodine policy of ‘Universal Salt Iodination, USI’ in China. The daily iodine intake of adult Chinese subjects is about 220–1960 mg and is much higher than that in Europe [6]. The incidence of goiter correlates with the iodide intake was 1/14 in Japan and 8/101 in Taiwan. Iodine deficiency causes goiter, and sufficient iodine intake may make up the deficit.
The frequency of biallelic SLC26A4 mutations in deaf patients varied among different populations, as 3.5–5% mainly in Caucasian populations [7] and around 8.8–12.5% in Middle Eastern and Asian patients [8,9]. Biallelic mutation genotype of SLC26A4 accounted for 30% in nonsyndromic hearing loss with EVA and 60% in subjects with Pendred syndrome [10]. The ratio was 76.4% (207/271) in the present study. SLC26A4 gene displays a scattering mutation spectrum with more than 240 types of mutation being reported in the world [1]. In the Caucasoid populations, three frequent mutations, L236P (16%), T416P (15%), and IVS8+1G >A (14%), account for nearly half of all PDS mutant
Genotype and phenotype correlations in children with EVAS Cases with fluctuating hearing 80 70 60 50 40 25 30 20 10 19 0 1.5−2 (44)
1247
Stable hearing
41
18
27 2−3 (68)
8
11 3
3−4 (26)
> 4 (14)
Acta Otolaryngol Downloaded from informahealthcare.com by University of Texas at Austin on 11/25/14 For personal use only.
Mid-portion diameter (mm) Cases with flat audiogram
Cases with flat audiogram
Ascending audiogram
70 3
60 50 40
4 16
30 20
19
10 0
1
35 5 1.5−2 (44)
14
6
2−3 (68)
3−4 (26)
1 12 1 > 4 (14)
Mid-portion diameter (mm) Cases with air-bone gap 80 70 60 50 40 30 20 10 0
Not with air-bone gap
Cases tested by ASSR
14 8 11 6
46
12 3 11
2 2 12
2−3 (68)
3−4 (26)
> 4 (14)
27 1.5−2 (44)
Mid-portion diameter (mm) Figure 4. Comparison of levels of hearing impairment among distinct VA size groups. The x-axis shows the vestibular size typing by midpoint diameter, and the y-axis shows the numbers of subjects in different size groups, composed of numbers of patients with distinct hearing characteristics, which included stable or fluctuating hearing (a), audiogram shapes (b), and the presence or absence of an air–bone gap (c).
alleles [11]. p.H723R was a common allele in the Korean population, while p.L676Q was common in the Mongolian population [1]. H723R accounted for 53% of PDS mutations in Japanese subjects [12]. This was the second most frequent mutation in the Chinese population, which was present mostly in heterozygous and homozygous c.919-2A >G forms. However, the incidence of this mutation was found to be low in Western countries and has been identified in only four families [13]. This difference suggested the existence of a common ancestor (founder effect) rather than a mutational hot spot [1]. Two novel mutations were found in the present study. The mutation c.665G >T (p.G222V) was predicted to be damaging to pendrin by SIFT software, while the effect of p.I547V (c.1639A >G) on
pendrin was tolerated. In consideration of both the sequences involving the two mutations being highly conserved in multiple species, both changes were supposed to jeopardize the protein function. Moreover, c.665G >T was found in two unrelated subjects, while c.1639A >G was found only once. Thus, based on the combined evidence, c.665G >T was thought to be a mutation, while c.1639A >G was more like a variation. Some studies showed that different kinds of mutations could produce different phenotypes [14]. But there were at least 19 shared mutations between Pendred syndrome, and autosomal recessive hearing loss associated with EVA has been reported [15]. Wu et al. found hearing loss in subjects with the biallelic genotype to be more severe than in subjects
Acta Otolaryngol Downloaded from informahealthcare.com by University of Texas at Austin on 11/25/14 For personal use only.
1248
F.-F. Zhao et al.
with the monoallelic genotype [16]. The present study also showed no distinct difference of hearing loss level in different genotypes, not only among subjects with biallelic mutation genotype, monoallelic, and no mutation, but also among subjects with biallelic truncating mutation genotype, biallelic heterozygous truncating, and biallelic missense. However, we found that the subjects with monoallelic missense mutations showed more stable hearing loss than those with other genotypes, which could be explained by the true genotype effect or the interference of unstable hearing status because of young ages. Some studies found a significant difference in mutation distribution between patients with EVA and those with Mondini’s deformity. In addition, bilateral malformations are also more frequently associated with the Mondini’s deformity phenotype group as compared with the EVA phenotype [17], but most studies did not show correlations between hearing loss and aqueduct size [18]. In addition, there were more than 31 shared mutations [8,19]. We found that there was no significant difference in the distribution of hearing loss level among distinct genotypes. Also, the midpoint diameter and operculum diameter of VA in subjects with no mutation were narrower than other genotype groups, which implied an effect of SLC26A4 gene mutations on VA size. The air–bone conduction gap in low frequencies was reported to be discovered in subjects with EVAS. A large VA may act as a third mobile window in the inner ear, resulting in an air–bone gap at low frequencies [20]. Our data showed that 149 of 190 subjects who underwent pure tone audiometry had an airbone gap, but no difference in distribution of subjects with air–bone gap was identified according to the dilated level of the VA. In addition, the air–bone gap was more frequent in subjects with biallelic missense mutations than in those with other genotypes. This finding implies that the air–bone conduction gap was more related to genotypes than the VA size. This result suggests that the membranous structures in the VA should be explored rather than bone structures. In other words, the endolymphatic sac and the intracellular ion concentration in the formation of the air–bone gap should not be ignored. Although we found some correlations between genotype and phenotype, this is far from proving direct correlation between the mutation genotype and hearing loss level or EVA size. Acknowledgments We thank the patients and their families for their cooperation during this work. This work was supported by grants from the National Natural Science
Foundation of China, Major Project, No. 81120108009 and National Natural Science Foundation of Youth Science Foundation No.81100719. Declaration of interest: The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the paper.
References [1] Park HJ, Shaukat S, Liu XZ, Hahn SH, Naz S, Ghosh M, et al. Origins and frequencies of SLC26A4 (PDS) mutations in east and south Asians: global implications for the epidemiology of deafness. J Med Genet 2003;40:242–8. [2] Harnsberger HR, Dahlen RT, Shelton C, Gray SD, Parkin JL. Advanced techniques in magnetic resonance imaging in the evaluation of the large endolymphatic duct and sac syndrome. Laryngoscope 1995;105:1037–42. [3] King KA, Choi BY, Zalewski C, Madeo AC, Manichaikul A, Pryor SP, et al. SLC26A4 genotype, but not cochlear radiologic structure, is correlated with hearing loss in ears with an enlarged vestibular aqueduct. Laryngoscope 2009;120:384– 9. [4] Boston M, Halsted M, Meinzen-Derr J, Bean J, Vijayasekaran S, Arjmand E, et al. The large vestibular aqueduct: a new definition based on audiologic and computed tomography correlation. Otolaryngol Head Neck Surg 2007;136:972–7. [5] Pryor SP, Madeo AC, Reynolds JC, Sarlis NJ, Arnos KS, Nance WE, et al. SLC26A4/PDS genotype-phenotype correlation in hearing loss with enlargement of the vestibular aqueduct (EVA): evidence that Pendred syndrome and non-syndromic EVA are distinct clinical and genetic entities. J Med Genet 2005;42:159–65. [6] Delange F. Iodine deficiency in Europe and its consequences: an update. Eur J Nucl Med Mol Imaging 2002; 29:S404–16. [7] Madden C, Halsted M, Meinzen-Derr J, Bardo D, Boston M, Arjmand E, et al. The influence of mutations in the SLC26A4 gene on the temporal bone in a population with enlarged vestibular aqueduct. Arch Otolaryngol Head Neck Surg 2007;133:162–8. [8] Wang QJ, Zhao YL, Rao SQ, Guo YF, Yuan H, Zong L, et al. A distinct spectrum of SLC26A4 mutations in patients with enlarged vestibular aqueduct in China. Clin Genet 2007;72: 245–54. [9] Wu CC, Lin SY, Su YN, Fang MY, Chen SU, Hsu CJ. Preimplantation genetic diagnosis (embryo screening) for enlarged vestibular aqueduct due to SLC26A4 mutation. Audiol Neurootol 2010;15:311–17. [10] Pera A, Villamar M, Vinuela A, Gandia M, Meda C, Moreno F, et al. A mutational analysis of the SLC26A4 gene in Spanish hearing-impaired families provides new insights into the genetic causes of Pendred syndrome and DFNB4 hearing loss. Eur J Hum Genet 2008;16: 888–96. [11] Campbell C, Cucci RA, Prasad S, Green GE, Edeal JB, Galer CE, et al. Pendred syndrome, DFNB4, and PDS/ SLC26A4 identification of eight novel mutations and possible genotype-phenotype correlations. Hum Mutat 2001;17: 403–11. [12] Usami S, Abe S, Weston MD, Shinkawa H, Van Camp G, Kimberling WJ. Non-syndromic hearing loss associated with
Genotype and phenotype correlations in children with EVAS
[13]
[14]
Acta Otolaryngol Downloaded from informahealthcare.com by University of Texas at Austin on 11/25/14 For personal use only.
[15]
[16]
enlarged vestibular aqueduct is caused by PDS mutations. Hum Genet 1999;104:188–92. Van Hauwe P, Everett LA, Coucke P, Scott DA, Kraft ML, Ris-Stalpers C, et al. Two frequent missense mutations in Pendred syndrome. Hum Mol Genet 1998;7:1099–104. Choi BY, Stewart AK, Madeo AC, Pryor SP, Lenhard S, Kittles R, et al. Hypo-functional SLC26A4 variants associated with nonsyndromic hearing loss and enlargement of the vestibular aqueduct: genotype-phenotype correlation or coincidental polymorphisms? Hum Mutat 2009;30:599–608. Tsukamoto K, Suzuki H, Harada D, Namba A, Abe S, Usami S. Distribution and frequencies of PDS (SLC26A4) mutations in Pendred syndrome and nonsyndromic hearing loss associated with enlarged vestibular aqueduct: a unique spectrum of mutations in Japanese. Eur J Hum Genet 2003; 11:916–22. Wu CC, Lu YC, Chen PJ, Yeh PL, Su YN, Hwu WL, et al. Phenotypic analyses and mutation screening of the
[17]
[18]
[19]
[20]
1249
SLC26A4 and FOXI1 genes in 101 Taiwanese families with bilateral nonsyndromic enlarged vestibular aqueduct (DFNB4) or Pendred syndrome. Audiol Neurootol 2009; 15:57–66. Azaiez H, Yang T, Prasad S, Sorensen JL, Nishimura CJ, Kimberling WJ, et al. Genotype-phenotype correlations for SLC26A4-related deafness. Hum Genet 2007;122:451–7. Gopen Q, Zhou G, Whittemore K, Kenna M. Enlarged vestibular aqueduct: review of controversial aspects. Laryngoscope 2011;121:1971–8. Iwasaki S, Tsukamoto K, Usami S, Misawa K, Mizuta K, Mineta H. Association of SLC26A4 mutations with clinical features and thyroid function in deaf infants with enlarged vestibular aqueduct. J Hum Genet 2006;51:805–10. Merchant SN, Nakajima HH, Halpin C, Nadol JB Jr, Lee DJ, Innis WP, et al. Clinical investigation and mechanism of air-bone gaps in large vestibular aqueduct syndrome. Ann Otol Rhinol Laryngol 2007;116:532–41.
