Download pdf - Connexin 26 mutations in nonsyndromic autosomal recessive hearing loss: speech and hearing rehabilitation

International Journal of Pediatric Otorhinolaryngology (2004) 68, 995—1005

Connexin 26 mutations in nonsyndromicautosomal recessive hearing loss: speech andhearing rehabilitation

Massimo Mesolella*, Gaetano Tranchino, Massimiliano Nardone,Sergio Motta, Vieri Galli

Department of Otolaryngology, University School of Medicine ‘‘Federico II,’’ Naples, Italy

Received 25 March 2003 ; received in revised form 18 February 2004; accepted 24 February 2004

KEYWORDSConnexin 26;Hereditary hearing loss;Non-syndromicrecessive deafness;Speech rehabilitation

Summary Objective: Hearing loss is the most common form of sensory impair-ment, with approximately one infant/1000 born with profound congenital deafness. Apre-lingual bilateral sensorineural hearing impairment poses a substantial problem asit negatively impacts on the subject’s ability to conduct a normal social life. The aimof the study was to observe, in a group of children affected by prelingual non-sydromicautosomal recessive hearing impairment:

(1) the role of the possible mutation of connexin 26 in the pathogenesis of thehearing loss;

(2) the audiological and clinical aspects of the hearing impairment;(3) therapy to be adopted for the different patients.

Methods: The study was carried out on 39 patients, 16 males and 23 females, agedbetween six and 17 years (mean 12 years), affected by non syndromic congenitaldeafness, presumably hereditary, referred to the out-patients audiology clinic for chil-dren of the Department of Otolaryngology of the ‘‘Federico II’’ University of Naples.Results: Our study conducted on 39 children with pre-lingual bilateral sensorineuralautosomal recessive deafness showed as follows:

(I) from a molecular perspective:• an incidence of 41% in the cases studied of mutations in the encoding of the

connexin 26 gene;• a prevalence in our case study of the 35delG mutation (69%).

(II) The characteristics of the hearing impairments in the children studied were ho-mogeneous, regardless of the presence or absence of a connexin 26 mutation:• the hearing impairment was pre-lingual bilateral sensorineural,• the impairment often involved mainly the high frequencies, but, especially in

the severe forms an involvement of all the frequencies was not rare;• the hearing impairments were symmetrical and non progressive in time.

*Corresponding author. Present address: Via Filangieri n. 72, 80121, Naples, Italy. Tel.: +39-081-415321; fax: +39-081-415321.E-mail address: [email protected] (M. Mesolella).

0165-5876/$ — see front matter © 2004 Elsevier Ireland Ltd. All rights reserved.doi:10.1016/j.ijporl.2004.02.015

996 M. Mesolella et al.

(III) the results of the application of prosthesis and thereafter rehabilitative languagetherapy are generally satisfactory but correlated of course to the severity of thehearing loss.

Conclusion: In conclusion, we hope that further developments in the research ongenetic hearing impairments will promptly result in advances in clinical practice.

© 2004 Elsevier Ireland Ltd. All rights reserved.

1. Introduction

In recent years there has been a marked improve-ment in diagnostic techniques used to identifyhearing impairments in children, in particular thoseinvolving objective audiometry; this has had a ma-jor clinical impact on the early identification ofhearing loss, on defining the degree and on deter-mining the site of the lesion.The results afforded with these, integrated with

the findings of conditioned child audiometry—whenit can reliably be performed—have made it possibleto intervene early on with the necessary rehabil-itation and prosthetic measures, with positive ef-fects on language learning and more generally onthe cognitive and psycho-affective development ofthe child with severe or profound deafness.Interesting progress has also been made in recent

years in the etiopathogenesis of hereditary deaf-ness thanks to the research into molecular genet-ics, which has much contributed to the knowledgeof the chromosome loci connected with the hear-ing function, to their alteration and to the relativerepercussions on the physiopathology of hearing [9].The epidemiological studies show that the inci-

dence in children of hearing impairments occurringin the pre-lingual stage is 1/1000 newborns per year,over 50% of which can be attributed to a geneticcause [7,21,22,24,25].Autosomal recessive deafness accounts for the

majority of the latter forms (70—80%), whereas theincidence is much lower for autosomal dominantforms (15—25%), for those linked to chromosome X(2—3%) and those of a mitochondrial origin (2—3%)[20,32].As regards the loci responsible for disease, for

the non syndromal forms the DFNB1 locus has beenmost extensively researched. First described in1994 by Guilford [11,12], it is epidemiologically re-sponsible for many forms of prelingual sensorineu-ral deafness, not only in different populations inthe Mediterranean [35], but also in other coun-tries such as the USA, Japan and China [3,17,19,23,37].Subsequent research [15,30] showed that a mu-

tation in the GJB2 gene is generally responsible for

severe or profound sensorineural deafness, whichcharacterises the first form of autosomal recessivedeafness described (DFNB1). This gene encodes theformation of the Connexin 26 (Cx 26) protein, amember of a large family of membrane proteins(Connexins) involved in the formation of the inter-cellular channels of the plasma membranes of manycells. These channels regulate the cell metabolism[1], the differentiation [34] and the transmission ofelectric impulses between the cells [10].Cx 26, presumably, is involved in a potassium

circulation pathway allowing potassium enter-ing the hair cells during sound mechanosensorytransduction to be removed to the stria vascularis[18], Mutations affecting Cx 26 presumably resultin aberrancies in potassium recirculation, subse-quently leading to cell death and deafness.Histochemical techniques have shown the pres-

ence of connexin 26 in the stria vascularis,basement membrane, limbus and in the spiralprominence of the cochlea [13,14].The aim of the study was to observe, in a group of

children affected by prelingual non-syndromic au-tosomal recessive hearing impairment:

• the role of the possible mutation of connexin 26in the pathogenesis of the hearing loss;

• the audiological and clinical aspects of the hear-ing impairment;

• therapy to be adopted for the different patients.

2. Materials and methods

The study was carried out on 39 patients, 16 malesand 23 females, aged between six and 17 years(mean 12 years), affected by non syndromic con-genital deafness, presumably hereditary, referredto the out-patients audiology clinic for children ofthe Department of Otolaryngology of the ‘‘FedericoII’’ University of Naples.All subjects underwent diagnostic evaluation,

which included:

(A) collection of anamnestic data;(B) general and otorhinolaryngoiatry examination;

Connexin 26 deafness 997

(C) audiological examinations comprising pure-toneaudiometry, acoustic admittance measure-ment, auditory brainstem response (ABR) andotoacoustic emissions;

(D) molecular gene testing;(E) C.T. of the ear;(F) The rehabilitative therapeutic options to be

adopted.

(A) personal and family anamnestic investigation:parents were questioned to ascertain the ab-sence of disorders and/or malformations ofthe ear, acquired pathologies of the ear andrisk factors (infections contracted during preg-nancy, birth weight less than 1500 g, neona-tal APGAR score <0—4 at 1min or <0—6 at5min, hyperbilirubinemia greater than 20mg%,administration of ototoxic medication, headtrauma with temporal bone involvement,neuro-degenerative disorders.

(B) General clinical and otorhinolaryngoiatry ex-aminations were carried out to exclude pos-sible malformations in organs or in othertracts.

(C) The audiological tests were selected accordingto the age of the child; they included both ob-jective (auditory brainstem evoked responses,acoustic admittancemeasurement) and subjec-tive audiometry (pure tone audiometry). Thesubjects were followed up for at least 3 years.In particular with:• pure tone audiometry. Using an AmplaidA315 apparatus we studied the type, mor-phology, severity and characteristics of thehearing loss: the following parameters wereconsidered:◦ type of hearing loss: conductive, sen-sorineural or mixed;

◦ morphology of the audiogram: the chartwas considered:(a) flat, when for the frequencies consid-

ered (250—500—1000—2000—4000Hz)the difference in the threshold valuewas lower or equal to 15 dB;

(b) ascending, when there was an averagedeficit of over 15 dB on the low fre-quencies (125—250—500) compared tothe high (2000—3000—4000Hz);

(c) descending, when there was an aver-age deficit of over 15 dB on the highfrequencies (2000—3000—4000Hz)compared to the low (250—500—1000Hz).

◦ severity of hearing loss: this was evaluatedon the basis of an average of the frequen-cies 500—1000—2000 pure threshold aver-

age (PTA); in particular four levels of hear-ing loss were considered:

(1) mild, with a hearing loss of less than 40 dBHL;

(2) moderate, from 40dB to 69 dB HL;(3) severe, from 70dB to 89 dB HL;(4) profound, with a hearing loss of over 90 dB

HL;The subjects with profound prelingual

bilateral sensorineural hearing loss weresubdivided into four groups, according tothe classification put forward by Boothroyd[2]:

◦ class I, subjects with PTA for the frequen-cies 500—1000—2000Hz, between 90 and95 dB HL;

◦ class II, subjects with PTA, for the frequen-cies 500—1000—2000Hz, between 96 and105 dB HL;

◦ class III, subjects with PTA for the frequen-cies 500—1000—2000Hz, between 106 and115 dB HL;

◦ class IV, subjects with PTA, for the frequen-cies 500—1000—2000Hz, equal to or over116 dB HL.

◦ symmetry of hearing loss: hearing thresh-olds were considered asymmetrical whenthere was a difference between the earsgreater or equal to 20 dB HL for the fre-quencies evaluated (500—1000—2000Hz);

◦ progression of hearing loss: hearing losswas considered progressive when therewas a worsening of the threshold value ofover 15 dB for the frequencies considered(500—1000—2000Hz) in the last year, and,therefore, year after year for the last 3years.–— Acoustic immittance measurement

was carried out using a Madsen Zodiac901;

–— A study of the auditory brainstemevoked responses (ABR) was carriedout using an Amplaid MK22, apparatus.We used the standard stimulation em-ployed at our institute [27], which in-volves the application of the followingparameters:

◦ test time: 10ms◦ cadence of the impulse: 21 pps◦ number of impulses: 2048◦ filtering: 100—2000Hz◦ type of impulse: click◦ intensity of impulse: from 120 dB SPLdown.

(D) the molecular evaluation: the 39 study patientsunderwent molecular gene testing to identify


the presence of mutations in the coding of theconnexin 26 gene.To carry out this test, 10ml of blood was

drawn from each patient, preserved in EDTA atroom temperature and dispatched within 48 hto the genetics laboratory for molecular evalu-ation.DNA sequencing was carried out on these

samples by polymerase chain reaction (PCR),according to the standard procedure in theliterature [33,37].On the basis of this the patients were divided

into two groups:Group A subjects with mutations in the con-

nexin 26 gene;Group B subjects without this mutation.

(E) A C.T. scan of the temporal bone excluded, inall cases, any bone malformation of the middleand/or inner ear.

(F) As regards the choice of therapy all subjectspromptly received a hearing aid, following thecriteria of dynamic prosthesization adopted atour institute [26,27], and started aggressivespeech therapy.

Dynamic prosthesisation is used to favour ini-tially the amplification of the low frequencies so asto favour a rapid educational program of the sub-ject and over time to put to better use the residualhearing capacity by personalising the hearing aid,implementing a more suitable or better adaptedprosthesis thanks to the acquisition of new infor-mation due to the better collaboration by the childas a consequence of his/her psycho-intellectualdevelopment and to the progress made in speechtherapy [26].In cases when the prosthesis offered poor re-

sults and the parents concurred, a multi channelcochlear implant was applied.The results of the prosthesis and language

therapy were considered satisfactory if they af-forded:

• recovery with hearing aid of the communicativeability with word discrimination capacity of over50% (the utterances were two-syllable words andthe intensity of the signal was 65 dB SPL in ab-sence of noise);

• as regards language development, the ability toconstruct phrases in a sentences using accuratesyntax, semantics and morphology.

After application of the prosthesis, language re-habilitation was carried out using the verbal-tonalmethod with the modifications introduced at our in-stitute [26].

For subjects who received cochlear implant,rehabilitation was carried out using a purely acu-pedica approach, the aim of which was to stimulateauditory speech perception as much as possible.

2.1. Statistical analysis

Statistical analysis was carried out using SPSS forWindows v. 6.0. Group data were expressed as per-centages and median (95% confidence intervals).Statistical comparisons between the two groupswere made using Student’s paired t-test. Signif-icance was set at P < 0.05. Multiple regressions,performed by forward stepwise comparisons, wereused to identify patients likely to exhibit a sat-isfactory speaking proficiency, considering eitheroverall the two groups or only the patients withprofound deafness.

3. Results

The results we achieved can be summarised withreference to:

(A) the findings of the molecular testing;(B) the data afforded by the audiological investi-

gation;(C) the results achieved with prosthesis and reha-

bilitation adopted.

(A) With regard to the molecular investigationand in particular to the incidence of mutations inthe Connexin 26 gene in the 39 subjects we ob-served:

• the presence of mutations in the GJB2 gene, alsocalled Connexin 26 in 16 subjects (41%), belongingto 14 families as there were two pairs of siblings;

• absence of these mutations in 23 patients (58%).

On the basis of the results of the molecular in-vestigation, the patients were subdivided into twogroups:

• Group A: 16 patients with mutations in the con-nexin 26 gene;

• Group B: the other 23 subjects with no such mu-tation.

Of the16 children with mutations in the con-nexin 26 gene, 11 (69%) were homozygotes forone mutation (35delG), 4 compound heterozygotesfor two mutations (35delG and another mutation);in particular, five mutations that have alreadybeen described, (35delG, 167delT, L90P, R184P,IVS1+ G>A).


In most cases we found a mutation characterisedby a deletion of a single base (35delG); in particularthe following mutations were found:

• 35delG 11 cases (69%);• 35delG-167delT 2 (13%);• 35delG-L90P 1 (6%);• 35delG-R184P 1 (6%);• 35delG-IVS1 +G>A 1 (6%).

(B) In relation to the audiological findings, theresults for the two groups showed:Group A: the 16 children with mutation in the

connexin 26 gene, as ascertained bymolecular eval-uation, all had bilateral sensorineural hearing loss(Fig. 1). More precisely:

Fig. 1 Group A: audiograms.

• pure tone audiometry showed:◦ with regard to the type of hearing loss, a bi-lateral sensorineural hearing impairment in allcases studied;

◦ as regards the morphology of the audiogram: aflat tone hearing impairment in five cases (31%)(Fig. 1: case nos. 5, 8, 11, 12, 15); a deficitmainly of the high frequencies in 11 cases (69%)(Fig. 1, case nos. 1, 2, 3, 4, 6, 7, 9, 10, 13, 14,16);

◦ as regards the severity, the hearing loss wasmoderate in 1 case (6%) (Fig. 1, case no.1),severe in 4 cases (25%) (Fig. 1, case nos. 2, 3,4, 5), profound in 11 (69%) (Fig. 1, case nos. 6,7, 8, 9, 10, 11, 12, 13, 14, 15,16);


◦ in 15 of the 16 cases (93%) the hearing deficitwas symmetrical (in both ears) (Fig. 1, case no.1 and from no. 3 to no.16).

◦ in all cases the hearing impairment was stablethroughout the study period;

• acoustic immittance measurement showed anormal tympanograms with absent stapedial re-flexes;

• the study of the auditory brainstem evoked re-sponses (ABR) showed the absence of waves in 15of the 16 cases; in one subject for whom audi-tory responses were recorded, pure tone audiom-etry (Fig. 1, case no. 1) showed a moderate sen-sorineural hearing loss:

• otoacoustic emissions were absent in all cases.

Group B: the remaining 23 children with no mu-tation in the connexin 26 gene, as ascertained bymolecular evaluation, were also affected by bilat-eral sensorineural hearing loss (Fig. 2). More pre-cisely:

• pure tone audiometry showed:◦ as regards the type of hearing loss, all casesunder study showed a bilateral sensorineuralhearing impairment;

◦ as regards the morphology of the audiometricchart, a flat tone deficit was found in one case(4.3%) (Fig. 2, case no. 23), and a loss mainlyin the high frequencies in 22 cases (95.7%)(Fig. 2, cases no. 1, 2, 3, 4, 5, 6, 7, 8, 9,10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21,22);

◦ as regards the severity, the hearing impair-ment was mild in 1 case (4.3%) (Fig. 2, caseno. 5), moderate in four cases (17.3%) (Fig. 2,case nos.3, 8, 15, 17), severe in 7 cases (31%)(Fig. 2, cases no. 2, 6, 7, 11, 14, 21, 23)and profound in 11 (47.4%) patients (Fig. 2,cases no. 1, 4, 9, 10, 12, 13, 16, 18, 19, 20,22);

◦ in 19 of the 23 cases (82.6%) the hearing losswas symmetrical, the other four cases (17.4%)(Fig. 2, cases no. 5, 9, 15, 17) were asymmet-rical;

◦ in this group the hearing impairment also re-mained stable throughout the study period.

• The acoustic immittance measurement showed inall cases a normal tympanograms with stapedialreflexes present in eight cases and absent in 15patients;

• The auditory brainstem evoked responses (ABR)showed a presence of wave V in four cases and theabsence of responses in the remaining 19 cases(Fig. 2, cases no. 1, 2, 4, 6, 7, 8, 9, 10, 11,12,13, 14, 16, 18, 19, 20, 21, 22, 23).

Onset data about the hearing parameters of boththe two groups did not show any significant differ-ence.(C) As regards the prosthetic rehabilitation re-

sults:Group A: The prosthetic and language rehabilita-

tion results achieved were in relation to the hearingimpairment:

• the subject with moderate hearing loss achievedgood results from an auditory and linguistic per-spective;

• the four subjects with severe hearing impairmentall achieved a good auditory and linguistic capac-ity;

• of the 11 patients with profound deafness:(a)one, of the first class according to Boothroydshowed good hearing and language therapy re-sults (Fig. 1, case no. 15);(b)eight subjects in the second class achievedunsatisfactory hearing results regardless of thetype of prosthesis used, except the subject whohad cochlear implant (Fig. 1, case no. 14); as re-gards the language therapy data, six of the eightpatients showed a satisfactory syntactic devel-opment (Fig. 1, case nos. 6, 7,10,12,14,16); theother two cases achieved unsatisfactory results(Fig. 1, case nos. 9, 11);(c)for the two patients with deafness of thethird class according to Boothroyd, the hear-ing and language therapy results were alto-gether unsatisfactory (Fig. 1, case nos. 8,13).

In short, satisfactory results were achieved:• in seven cases out of 16 (43.7%) regarding hear-ing; of these patients five had a moderate or se-vere hearing loss, one had profound hearing lossof the first class according to Boothroyd and one,with a hearing impairment of the second classaccording to Boothroyd, underwent cochlear im-plant.

• in 14 of 16 cases (87.5%) as regards the languagetherapy.

Group B: Also in this group the prosthetic andlanguage therapy results reflect the severity of thehearing impairment; in particular:

• from an auditory and linguistic viewpoint goodresults were achieved for:◦ the subject with mild hearing loss (Fig. 2, case5);

◦ the subjects with a moderate hearing impair-ment (Fig. 2, case nos. 3, 8, 15, 17);

◦ all seven subjects with severe hearing loss(Fig. 2, case nos. 2, 6, 7, 11, 14, 21, 23);


Fig. 2 Group B: audiograms.


the patients with profound deafness achievedless homogeneous results; in particular:

(a) of the three cases whose hearing impairmentwas of class 1 according to Boothroyd, oneshowed good hearing and language therapy re-sults (Fig. 2, case no. 22) whereas the othertwo achieved good results only for the acquisi-tion of grammatical and lexical structures;

(b) of the six subjects in class II, three had un-satisfactory hearing results (Fig. 2, case nos.4, 12, 18); as regards the language therapy re-sults five out of six achieved a good syntacticdevelopment (Fig. 2, case nos. 1, 4, 13, 18,20);

(c) the one patient with hearing impairment ofthe third class according to Boothroyd, showedaltogether unsatisfactory hearing and languagetherapy results (Fig. 2, case no. 16).

In short, satisfactory results were achieved:

• in 16 out of 23 cases (69.5%) as regards their hear-ing capacity; of these 12 had a mild, moderateor severe hearing impairment, one had profounddeafness of the first class according to Boothroyd,and three of the third class.

• The subject belonging to the second class ac-cording to Boothroyd, who underwent cochlearimplant, showed a good recovery of hearing andgood language development.

• 20 out of 23 cases (86.9%) showed good resultsfrom a language therapy viewpoint.

Hearing aid rehabilitative outcomes (hearing andspeaking skills) pointed out a noticeable improve-ment, even if not statistically significant, in pa-tients with connexin deficit.At the end of rehabilitation hearing skill ap-

peared to be the most important factor (P < 0.01)in achieving a satisfactory speaking proficiencywithin the two groups; in the patients with pro-found hearing loss a satisfactory speaking profi-ciency was achieved by the patients with a lowerdegree of hearing loss (P < 0.05).

4. Discussion

Increasing technological developments in medicinehave brought about significant progress in theknowledge of the etiopathogenesis of hereditarydeafness, as well as in the diagnostic techniquesand therapy implemented.Molecular analysis has made it possible to

identify genes involved in the auditory function

and established what gene mutations impact onhearing.However, because of the numerous chromosome

alterations implicated, the extraordinary pheno-typic heterogeneity and the marked inter and in-trafamily clinical variability, the knowledge in thisarea is still scanty.Our study was carried out on 39 patients with a

hereditary pre-lingual bilateral sensorineural, au-tosomal recessive hearing impairment.The aim of our study was to establish in our pa-

tients the incidence of cases with mutations in theconnexin 26 gene.Connexins are membrane proteins implicated in

the formation of gap junctions, which allow therapid transportation of ions (recycling of potassiumions back to the endolymph) and small moleculesbetween the cells in the organ of Corti, ensur-ing its function; alterations in these proteins in-duced by genetic factors are at present consid-ered an important cause of hereditary hearing loss[34].As regards the molecular analysis, we found

connexin 26 mutations in 16 patients (41%), i.e.in 11 (69%) cases the deletion of a single base(35delG) and in 5 (31%) cases two associated mu-tations (35delG associated respectively to 167delT,L90P, R184P and IVS1+ G>A). The IVS1+ G>A mu-tation, found in one patient, was determined bythe replacement of Guanine with Arginine on genereplication. This confirms the hypothesis expressedby other authors [10] that there are many differentgenes involved in the physiological mechanisms ofthe ear: this obscures the comprehension of theetiopathogenic mechanisms at the root of hearingimpairment.Present molecular tests can be considered spe-

cific but not always sufficiently sensitive, i.e. theycan identify mutations in a gene that is alreadyknown, but cannot rule out when negative otherabnormalities in unknown genes or unknown alter-ations in known genes, which are possible mutationsites responsible for hearing impairment.The reports in the literature are not concurrent

on the incidence of connexin 26 mutations: theyvary between 10—20 % [37] and 80% [6]; we foundan intermediate value of 41%.Various factors can account for this apparent dis-

cordance, but race is certainly an important factor.The 35delG mutation, which is present in relativelyhigh values in our case study (69%), has been mostfrequently found in the studies on white subjectsand in the Mediterranean populations [8], althougha marked variability can be found within the samepopulation. To this regard Estivill [6] reports apercentage of this mutation as 92% for the Italian


Fig. 3 Entity of hearing loss.

participants in his study and 63% for the Spanishparticipants.On the other hand, the 167delT mutation is more

frequent in the Ashkenazi Jews [5], the R143W mu-tation in African races [10], while the V271 and235delC mutations are more frequent in Asia [31].

As regards the characteristics of hearing loss,the tests we carried out allowed us to identify cer-tain peculiarities of hearing loss determined by con-nexin 26 gene mutations; more precisely, in thecase where this mutation is involved, there werepre-lingual bilateral sensorineural hearing impair-ments that:

• mainly involved the high frequencies (69%) butalso involved all the frequencies more or less tothe same extent (31%);

• were more frequently profound (69%), morerarely severe (25%), and only exceptionally mod-erate (6%);

• were usually symmetrical (93%);• did not show any progression over the years(100%).

These findings are essentially in accordance withthe data reported in the literature [4,6,16,28,29].

Fig. 4 Results after rehabilitative course.

A comparison of the characteristics of the hearingloss in the two groups (A and B) showed:

• a presence in all cases of a pre-lingual bilateralsensorineural hearing loss;

• a generally more severe hearing loss (Fig. 3) incases involving a connexin 26 mutation (group A);

• involvement mainly of the high frequencies inboth groups (69% for group A versus 95.7% forgroup B); however, in group A, the incidenceof hearing loss equally involving all frequen-cies was higher (31% in group A versus. 4.3%in group B); this was related to the greaterseverity of the hearing impairments found ingroup A;

• in both groups, the hearing loss, which wasgenerally symmetrical (93% in group A and82.6% in group B), showed no modification intime.

As regards the results of the prosthetic rehabili-tation therapy, satisfactory results were achieved:

• in group A:◦ in seven cases out of 16 (43.7%) for hearing ca-pacity;in 14 cases out of 16 (87.5%) for speechand language ability


• in group B:

◦ in 16 cases out of 23 (69.5%) for hearing capac-ity;

◦ in 20 patients out of 23 (86.9%) for speech andlanguage ability.

Therefore, the rehabilitation results show no sub-stantial differences between the two groups, butthe higher incidence of a more severe hearing im-pairment in group A had an impact on the prosthetichearing recovery, but not on the language therapyresults (Fig. 4).However we have observed that because hearing

loss from GJB2 is related to an insult to the cochleabut not to auditory system or acoustic nerve betterspeech and language ability is seen in children withGJB2-related deafness after cochlear implantation.

5. Conclusions

A pre-lingual bilateral sensorineural hearing impair-ment poses a substantial problem as it negativelyimpacts on the subject’s ability to conduct a nor-mal social life.Our study conducted on 39 children with

pre-lingual bilateral sensorineural autosomal re-cessive deafness showed as follows:

(I) from a molecular perspective:• an incidence of 41% in the cases studied ofmutations in the encoding of the connexin 26gene;a prevalence in our case study of the35delG mutation (69%): this is caused by adeletion of Guanine in position 35 and con-firms the data reported by other authors [36]on the highest frequency of this mutationamong white populations;

• one case with a mutation (IVS1 +G>A) char-acterised by the replacement of Guaninewith Alanine.It is noteworthy that despite the numerous

chromosomic loci and the diverse pathogeneticmechanisms involved, the characteristics ofthe relative hearing impairments showed nosubstantial differences in these hereditaryforms;

(II) The characteristics of the hearing impairmentsin the children studied were homogeneous, re-gardless of the presence or absence of a con-nexin 26 mutation:• the hearing impairment was pre-lingual bi-lateral sensorineural,

• the impairment often involved mainly thehigh frequencies, but, especially in the se-

vere forms an involvement of all the frequen-cies was not rare;

• the hearing impairments were symmetricaland non progressive in time.

(III) the results of the application of prosthesis andthereafter rehabilitative language therapy aregenerally satisfactory but correlated of courseto the severity of the hearing loss.

In cases with a connexin 26 mutation, the hear-ing loss was often more severe (69% profound hear-ing loss in group A versus 47.4% in group B) andhomogeneously involved all the frequencies, withprosthetic results naturally less satisfactory. This,however, did not influence the outcome of languagetherapy.Moreover the isolated insult to the cochlea cre-

ated by GJB2 allele variants allows for preservationof central cognitive function better speech and lan-guage ability is seen in children with GJB2-relateddeafness after cochlear implantation.These conditions might justify early cochlear im-

plant in subjects whose hearing impairment is as-sociated to a connexin 26 gene mutation.In general, early cochlear implantation is not in-

dicated because of the lack, in the first 18 monthsof life, of precise data on the hearing capacity ofthe subject with severe hearing loss; however, asthe connexin 26 mutation in general causes sub-stantial hearing loss, it may be an element forindicating cochlear implant in the light of theseconsiderations.In conclusion, identification of the cause of deaf-

ness in any individual still carries diagnostic, prog-nostic and therapeutic information that can be usedto improve medical and audiological care.

References

[1] J.F. Battey Jr., A genetic approach to understanding innerear function, J. Clin. Invest. 106 (2000) 1431—1432.

[2] Boothroyd et al., Clinical implication of Coclear impiantsin children, Ear Hear. 12 (1993) 81—89.

[3] K.A. Brown, A.H. Janjua, G. Karbani, G. Parry, A. Noble, G.Crockford, et al., Likage studies of non syndromic recessivedeafness (NSRD) in a family originating from Mirpur regionof Pakistan maps DFNB1 centromeric to D13S175, Hum.Mol. Gen. 5 (1996) 169—173.

[4] E.S. Cohn, P.M. Kelley, T.W. Fowler, M.P. Gorga, Clinicalstudies of families with hearing loss attributable to mu-tations in the Connexin 26 gene (GJB2/DFNB1), Pediatrics103 (1999) 546—551.

[5] Dong et al., Nonradioactive detection of the commonConnexin 26 167delT and 35delG mutations and frequen-cies among Ashkenazi Jews, Mol. Gene. Metab. 73 (2001)160—163.

[6] X. Estivill, P. Fortina, S. Surrey, Connexin-26 mutations insporadic and inherited sensorineural deafness, Lancet 351(1998) 394—398.


[7] B. Ewing, P. Green, Analysis of expressed sequence tagsindicates 35,000 human genes, Nat. Gen. 25 (2000) 232—234.

[8] P. Gasparini, High carrier frequency of the 35delG deafnessmutation in European populations, Eur. J. Human Genet.8 (2000) 19—23.

[9] R.J. Gorlin, H.V. Toriello, M.M. Cohen Jr., Hereditary Hear-ing Loss and its Syndrome, Oxford University Press, NewYork, 1995.

[10] K.M. Grundfast, N. Siparsky, D. Chuong, Genetics andmolecular biology of deafness, Update Otorhinolaringol.Clin. North Am. 33 (2000) 6.

[11] P. Guilford, H. Ayadi, S. Blanchard, A human gene respon-sible for neurosensory, non-syndromic recessive deafnessis a candidate homologue of the mouse sh-1 gene, Hum.Mol. Gen. 3 (1994) 989—993.

[12] P. Guilford, S. Ben Arab, S. Blanchard, A non-syndromicform of neurosensory, recessive deafness maps to the peri-centromeric region of chromosome 13q, Nature Genet. 6(1994) 28—32.

[13] A.I. Jun, T. Wyman, M.D. McGuirt, Temporal bone histo-pathology in Connexin 26 related hearing loss, Laryngo-scope 110 (2001) 269—275.

[14] D.P. Kelsell, J. Dunlop, H.P. Steven, Connexin 26 muta-tions in hereditary non-syndromic sensorineural deafness,Nature 387 (1997) 80—83.

[15] D.P. Kelsell, Connexin mutations associated with palmo-plantar keratoderma and profound deafness in a singlefamily, Eur. J. Human Genet. 8 (2000) 141—144.

[16] M.A. Kenna, B. Wu, A. Contanche, B.R. Korf, Connexin 26Studies in Patients With Sensorineural Hearing Loss, ArchOtolaryngol Head Neck Surg. 127 (2001) 1037—1042.

[17] Kudo et al., Novel mutations in the Connexin 26gene (GJB2) responsible for childhood deafness in theJapanese population, Am. J. Med. Genet. 90 (2000) 141—145.

[18] P.P. Lefebvre, R. Van De Water, Connexins, hearing anddeafness: clinical aspects of mutations in the connexin 26gene, Brain Res. Brain Res. Rev. 32 (2000) 159—162.

[19] X.Z. Liu, Mutations in connexin31 underlie recessive as wellas dominant non-syndromic hearing loss, Hum Mol Gen. 9(2000) 63—67.

[20] M.L. Marrazita, L.M. Ploughman, B. Rawlings, Genetic epi-demiological studies of early-onset deafness in the U.S.school-age population, Am. J. Med. Genet. 46 (1993) 486—491.

[21] A. Martini, A. Read, D. Stephens, Genetics and HearingImpairment, Whurr Publishers, London, 1996.

[22] A. Martini, M. Mazzoli, D. Stephens, A. Read, Definitions,Protocols and Guidelines in Genetic Hearing Impairment,Whurr Publishers, London, 2001.

[23] M.A. Maw, D.R. Allen-Powell, R.J. Goodey, The contributionof the DFNB1 locus to neurosensory deafness in a Caucasianpopulation, Am. J. Hum. Genet. 57 (1995) 629—635.

[24] McGuirt et al., Connexin 26 as a cause of hereditary hearingloss, Am. J. Audiol. 8 (1999) 93—100.

[25] N.E. Morton, Genetic epidemiology of hearing impairment,Ann. NY Acad. Sci. (1991) 16—31.

[26] G. Motta, V. Galli, Protesizzazione acustica, impianto co-cleare, terapia riabilitativa nel bambino con sordità pro-fonda, Acta Phoniatrica Latina 21 (1999) 1.

[27] Motta G., Audiologia protesica, Orl. Ped. (1992) III,3.[28] R.F. Mueller, A. Nehammer, A. Middleton, Congenital

non-syndromal sensorineural hearing impairment due toconnexin 26 gene mutations--molecular and audiologicalfindings, Int. J. Pediatr. Otorhinolaryngol. 50 (1999) 3—13.

[29] A. Murgia, E. Orzan, R. Polli, M. Martella, C. Vinanzi, Cx26 deafness: mutation analysis and clinical variability, J.Med. Genet. 36 (1999) 829—832.

[30] W.E. Nance, et al., Relation between choice of partnerand high frequency of connexin-26 deafness, Lancet 356(2000) 500—501.

[31] H.J. Park, S.H. Hahn, J.M. Chun, Connexin 26 mutationsassociated with nonsyndromic hearing loss, Laryngoscope110 (2000) 1535—1538.

[32] W. Reardon, H.R. Middleton-Price, S. Malcom, Clinical andgenetic heterogeneity in X linked deafness, Br. J. Audiol.26 (1992) 109—114.

[33] H. Roest Crollius, O. Jaillon, A. Bernot, C. Dasilva, L.Bouneau, C. Fischer, Estimate of human gene number pro-vided by genome-wide analysis using Tetraodon nigroviridisDNA sequence, Nat. Genet. 25 (2000) 235—238.

[34] H.M. Samy, Molecular biology of the auditory and vestibularsystem, Curr. Opin. Otolaryngol. 9 (2001) 310—313.

[35] D.A. Scott, M.L. Kraft, E.M. Stone, Connexin mutations andhearing loss, Nature 391 (1998) 32.

[36] L. Van Laer, A common founder for the 35delG GJB2gene mutation in connexin 26 hearing impairment, J. Med.Genet. 38 (2001) 515—518.

[37] S.A. Wilcox, et al., A simple PCR test to detect the common35delG mutation in the connexin 26 gene, Mol. diagn. 5(2000) 75—78.