 |
|
Speech Recognition Scores Related to Age and Degree of Hearing Impairment in DFNA2/KCNQ4 and DFNA9/COCH
Steven J. H. Bom, MD;
Els M. R. De Leenheer, MD;
François X. Lemaire, MD;
Martijn H. Kemperman, MD;
Wim I. M. Verhagen, PhD;
Henri A. M. Marres, PhD;
Henricus P. M. Kunst, PhD;
Robbert J. H. Ensink, PhD;
Arjan J. Bosman, PhD;
Guy Van Camp, PhD;
Frans P. M. Cremers, PhD;
Patrick L. M. Huygen, PhD;
Cor W. R. J. Cremers, PhD
Arch Otolaryngol Head Neck Surg. 2001;127:1045-1048.
ABSTRACT
Objective To analyze the relationship between pure-tone hearing threshold and
speech recognition performance in DFNA2/KCNQ4 and
DFNA9/COCH, 2 types of high-frequency nonsyndromic
hearing impairment.
Design Case series with cross-sectional analysis of phoneme recognition scores
related to age and hearing level.
Setting University hospital.
Patients Forty-five members of 4 separate families, all carrying 1 of 3 different
mutations in the KCNQ4 gene at the DFNA2 locus (1p34);
42 members of 7 separate families, all carrying the same Pro51Ser mutation
in the COCH gene at the DFNA9 locus (14q12-q13).
Results The deterioration of speech recognition dropped to a 90% score at a
higher level of hearing impairment (pure-tone-average at 1, 2, and 4 kHz)
in DFNA2-affected patients (65 dB) than in DFNA9-affected patients (46 dB).
Conclusion At similar levels of hearing impairment, DFNA2/KCNQ4-affected patients showed better speech recognition performance than
DFNA9/COCH-affected patients.
INTRODUCTION
AUTOSOMAL dominant nonsyndromic types of hereditary sensorineural hearing
impairment can be identified by genetic linkage and mutation analysis. The
corresponding chromosomal loci are genetically designated DFNA, followed by
a number in order of discovery (DFN = deafness, A = autosomal dominant inheritance).1 One locus may harbor 1 or more disease-causing genes.
The discovery of such genes and their function may enhance our understanding
of the pathophysiology of the inner ear. Concurrently, clinical studies are
necessary to relate the latter to the resulting phenotype.
Clinical studies on DFNA2/KCNQ4-affected families
and DFNA9/COCH-affected families have demonstrated
fairly similar high-frequency sensorineural hearing impairment and progression
(S.J.H.B., unpublished data, 2000)2-16
In addition to sensorineural hearing loss, DFNA9-affected patients also develop
vestibular failure.6-16
The DFNA9-affected families studied in the Netherlands and Flanders all have
the same mutation in the COCH gene, corresponding
to a P51S substitution in the expressed protein, cochlin.12, 17-18
The function of cochlin is still unknown. In the 4 Dutch DFNA2-affected families,
3 different mutations (W276S, G321S, L274W) of the KCNQ4 gene were found.19-20 KCNQ4 encodes a potassium (K+) channel that
is predominantly expressed in the basolateral membrane of cochlear hair cells.21-22
The present study focuses on the relationship between speech recognition
performance on the one hand and age and pure-tone hearing threshold on the
other hand in 2 different types of nonsyndromic, autosomal dominant, sensorineural
hearing impairment. The 2, DFNA2/KCNQ4 and DFNA9/COCH, are both characterized by progressive, predominantly
high-frequency hearing impairment.
PATIENTS AND METHODS
This study compared speech recognition data between patients with DFNA2/KCNQ4-affected patients and DFNA9/COCH-affected patients. Speech recognition data for patients with DFNA2
were obtained from 45 carriers of a mutation in the KCNQ4 gene. There were 30 patients from 2 families with a W276S mutation
(G.V.C., unpublished data, 2000),2, 5, 19
10 patients with a G321S mutation,3, 19
and 5 patients with a L274W mutation.4, 20
Speech recognition data for patients with DFNA9 were obtained from 42 carriers
of the P51S mutation in the COCH gene, from 6 Dutch6-11,17
families and 1 Flemish family (S.J.H.B., unpublished data, 2000).13, 18
Audiometry was performed according to common clinical standards. For
speech recognition, standard monosyllabic Dutch word lists were presented
at either ear.23 Performance-intensity curves
relating to the phoneme recognition score were analyzed for the right ear
only. The last-visit speech reception threshold (SRT) (in decibels sound pressure
level [dB SPL] for 50% phoneme score) and the maximum phoneme recognition
score (percentage correct), at which the pure-tone-threshold at 1, 2, and
4 kHz could be measured, were used.
Plots of SRT vs pure-tone-average at 1, 2, and 4 kHz (PTA1,2,4
kHz) were used with linear regression analysis to check for the reliability
of the phoneme recognition scores. The Chauvenet criterion24
and the residual SD were used to identify and exclude outlying values.
Because a comparison of speech recognition scores between the patient
groups may be complicated by underlying differences in sensorineural hearing
threshold, the thresholds at each frequency for "modal-age" patients between
the groups were compared, using the t test. These
patients were selected from each group by requiring their age to be within
the limits of the percentiles P37.5 and P62.5 of the
corresponding age distribution.
Nonlinear regression analysis of the maximum phoneme recognition score
on log(age) and on log(PTA1,2,4 kHz) was performed using the Prism
PC version 3.02 program (GraphPad, San Diego, Calif). Cross-sectional and
individual longitudinal performance-age (percentage recognition vs age) and
performance-impairment plots (percentage recognition vs PTA1,2,4 kHz) were fitted with a sigmoidal dose-response function with a variable
slope13: Y = {100%/[1
+ 10(logX90 - logX)xHillSlope-log9]},
where Y is the phoneme score, X is either age or PTA1,2,4 kHz, X90 is the value of X where Y = 90%, and HillSlope is the slope factor
on a log scale of X. The fitted values of X90 and HillSlope were used to test
between curves relating to the patient groups, using the t test (with the Welch correction if the Bartlett test detected unequal
variances).
To simplify the results and allow for additional testing, "local average"
slope (ie, on a linear scale) for X>X90 was obtained by using a linear regression line as an
approximation of the corresponding part of the fitted sigmoidal curve. Slope
was the deterioration rate in the performance-age plot, whereas it was the
deterioration gradient in the performance-impairment plot. X.90 was the onset age for the performance-age plot and
onset level for the performance-impairment plot. Regression lines were compared
between the groups using analysis of covariance (ANCOVA) to find out whether
slopes and intercepts were significantly different. Again, Chauvenet's criterion24 was used in combination with the residual SD to detect
outlying values.
Individual longitudinal data were available in 18 DFNA2/KCNQ4-affected patients and 23 DFNA9/COCH-affected
patients. Analyses constituted plotting of serial phoneme recognition scores
against age and PTA1,2,4 kHz, and comparing these to the curves
fitted to the corresponding cross-sectional data. A 5% lower normal limit
established at the Nijmegen Otorhinolaryngology department (P.L.M.H., unpublished
data, 2000) was used to see whether there were scores that raised suspicion
of retrocochlear dysfunction.
The phenotype of the DFNA2/KCNQ4-affected patients
may be influenced by the nature of the mutation that is present in the family.
In addition, even for patients carrying the same mutation, differences in
other genes (genetic background) may also influence the phenotype. It was,
therefore, checked whether there were significant differences in score behavior
between either the different DFNA2/KCNQ4-affected
families or the different DFNA9/COCH-affected families.
RESULTS
In each group, the SRT showed an excellent correlation with the corresponding
PTA1,2,4 kHz (n = 33-44 after exclusion of 3 outlying values; r = 0.8-0.9, residual SD = 9-13 dB). This corresponds fairly
well with the residual SD found by Bosman and Smoorenburg.24
Mean pure-tone thresholds for a representative, "modal-age" selection from
each group, that is, cases with P37.5<age<P62.5
(percentiles in the frequency distribution of age), generally did not differ
significantly at any frequency.
Figure 1 shows cross-sectional
performance vs age plots (A-B) and combined performance-impairment plots (C).
Onset age was 34 years in DFNA2/KCNQ4-affected patients
and 43 years in DFNA9/COCH-affected patients. The
local average deterioration rate in DFNA2-affected patients was 0.3% per year,
while in DFNA9-affected patients, it was 1.8% per year (Figure 1A-B). There was no significant difference in HillSlope between the plotted curves in Figure 1C. However, the local average deterioration gradients were
significantly different (0.5% per decibel vs 1.2% per decibel). The latter
finding is related to the significant difference in onset level that can be
detected (65 vs 46 dB), as well as the observation that there is a clear separation
between the data points (Figure 1C). In other words, at a given level of impairment, DFNA2/KCNQ4-affected patients showed higher scores than DFNA9/COCH-affected patients.
|
|
|
|
Speech recognition score against age (A [DFNA2-affected patients]
and B [DFNA9-affected patients]) and pure-tone average (PTA) at 1, 2, and
4 kHz expressed in decibels hearing level (dB HL) (C [DFNA2 vs DFNA9]) in
DFNA2/KCNQ4-affected patients (circles) and DFNA9/COCH-affected patients (triangles). Smaller symbols indicate
outlying values.
|
|
|
Individual longitudinal analyses produced fairly similar results to
the cross-sectional analysis (data not shown). Four of the DFNA9/COCH-affected patients had scores below the normal limit that would
have been suggestive of retrocochlear dysfunction. Some of these scores renormalized
during follow-up. None of the DFNA2/KCNQ4-affected
patients had any scores suggestive of retrocochlear dysfunction. No significant
differences in score behavior between either the different DFNA2/KCNQ4-affected families or the different DFNA9/COCH-affected families were found (data not shown).
COMMENT
The marked difference in speech recognition performance as related to
sensorineural hearing level between DFNA2/KCNQ4 and
DFNA9/COCH, both characterized by predominantly high-frequency
sensorineural hearing impairment, is an appealing finding in this study. Better
recognition scores in the DFNA2-affected patients are remarkable, especially
in light of recent findings that the mouse homologue of KCNQ4 is abundantly expressed in the central auditory pathways.25 KCNQ4 is thought to play
a role in the K+ recycling pathway of the inner ear.22, 26
Three of the 4 studied families with DFNA2/KCNQ4-affected
patients had different mutations, but fairly similar speech recognition. The
L274W and W276S mutations produce changes in the pore region of the expressed
K+ channel protein, whereas the G321S mutation exerts an effect
just outside the pore region, however, apparently to a similar phenotypic
effect.20
On the other hand, the DFNA9/COCH-affected
patients had poorer speech recognition scores compared with age and sensorineural
hearing level (Figure 1C). The American
DFNA9-affected patients, with a V66G mutation within the COCH gene,27-28 however,
showed an even greater drop in recognition scores.14
The latter had earlier onset (at age 20 years) and anacusis at around age
45 years. Combined speech performance-impairment plots of the recognition
scores (not shown) are suggestive of poorer scores in the American V66G carriers.
However, our longitudinal analyses also disclosed the existence of temporarily
poor scores in some of our DFNA9/COCH-affected patients.
Such poor scores may have been related to Ménière-like paroxysms.11-12
Histopathologic findings reported for 1 American COCH/V66G–mutation carrier comprised general destruction of the
cochlear and vestibular sensory elements, including hair cells and dendrites
(cochlea, crista, and macula), as well as accumulation of an acellular substance
(glycosaminoglycan) throughout the labyrinth.15
These findings were similar to those reported previously in DFNA9/COCH-affected patients.16, 29
In chicken, COCH expression was found in fairly similar
places where the deposits were found in human patients.28
It has been postulated that "strangulation" of cochlear and vestibular nerve
endings occurs.15-16,29
Alternatively, the possibility was suggested that normal fibrillogenesis is
disrupted by an excess in microfibrillar substance, which results in degradation
of collagens and extracellular matrix components.15
It is also possible that cochlin, which is expressed in the stroma underlying
the sensory structures of the inner ear,28
has a role in ion homeostasis, for example, recycling of K+ ions
from the hair cells to the endolymph.12
In the present study, DFNA2/KCNQ4-affected
patients seemed to have better speech recognition scores than DFNA9/COCH-affected patients (Figure 1C); the difference could not be explained by underlying
differences in pure-tone thresholds. Cochlear KCNQ4
expression was initially thought to be confined to the outer hair cells.21 However, recent findings in the rat have shown that
it is also expressed in inner hair cells and the spiral ganglion.22 The strongest KCNQ4 expression,
that is, in normally hearing animals, was found in inner hair cells in the
lower cochlear turns and in outer hair cells in the upper turns. Thus DFNA2/KCNQ4-related high-frequency sensorineural hearing impairment,
which is associated with primary dysfunction of the lower cochlear turns,
might be attributed to a lack of expression of K+ channels, especially
in inner hair cells. Relative sparing of function of outer hair cells in the
lower turns, thus preserving their function as "cochlear preamplifier" in
fine tuning mechanisms,30 might account for
the better speech recognition in DFNA2/KCNQ4-affected
patients. On the other hand, the poor recognition scores in DFNA9/COCH-affected patients might be explained by the generalized, histopathologic,
vestibulocochlear changes,15-16,28-29
already mentioned earlier, and, in part, by its Ménièriform
features.11-12
AUTHOR INFORMATION
Accepted for publication February 7, 2001.
This study was supported in part by the Heinsius Houbolt Foundation,
Wassenaar, the Netherlands, and the Nijmegen KNO-Research Foundation (Dr Cor
Cremers).
We thank the family members who participated in this study.
Drs Bom and De Leenheer contributed equally to this work.
Corresponding author: Steven J. H. Bom, MD, Department of Otorhinolaryngology,
University Medical Centre St Radboud, PO Box 9101, 6500 HB Nijmegen, the Netherlands
(e-mail: S.Bom{at}kno.azn.nl).
From the Departments of Otorhinolaryngology (Drs Bom, De Leenheer,
Kemperman, Marres, Kunst, Ensink, Bosman, Huygen, and C.W.R.J. Cremers) and
Human Genetics (Dr F.P.M. Cremers), University Medical Centre St Radboud,
and Department of Neurology, Canisius-Wilhelmina Hospital (Dr Verhagen) Nijmegen,
the Netherlands; Department of Otorhinolaryngology, Head and Neck Surgery,
University Hospital Leuven, Leuven, Belgium (Drs Lemaire and C.W.R.J. Cremers);
and Department of Medical Genetics, University of Antwerp, Belgium (Dr Van
Camp).
REFERENCES
| |
1. Van Camp G, Smith RJH. Hereditary Hearing Loss Homepage. Available at:
http://dnalab-www.uia.ac.be/dnalab/hhh/index.html.
Accessed December 2000.
2. Marres H, van Ewijk M, Huygen P, et al. Inherited nonsyndromic hearing loss: an audiovestibular study in a
large family with autosomal dominant progressive hearing loss related to DFNA2. Arch Otolaryngol Head Neck Surg. 1997;123:573-577.
ABSTRACT
3. Kunst H, Marres H, Huygen P, et al. Nonsyndromic autosomal dominant progressive sensorineural hearing loss:
audiologic analysis of a pedigree linked to DFNA2. Laryngoscope. 1998;108:74-80.
FULL TEXT
|
ISI
| PUBMED
4. Ensink RJH, Huygen PLM, Van Hauwe P, Coucke P, Cremers CWRJ, Van Camp G. Another family with progressive sensorineural hearing impairment linked
to the DFNA2 region. Eur Arch Otorhinolaryngol. 2000;257:62-67.
FULL TEXT
| PUBMED
5. De Leenheer EMR, Huygen PLM, Coucke PJ, Admiraal RJC, Van Camp G, Cremers CWRJ. Longitudinal and cross-sectional phenotype analysis in a new, large
Dutch DFNA2/KCNQ4 family. Ann Otol Rhinol Laryngol. In press.
6. Bom SJH, Kemperman MH, De Kok YJM, et al. Progressive cochleovestibular impairment caused by a point mutation
in the COCH gene at DFNA9. Laryngoscope. 1999;109:1525-1530.
FULL TEXT
|
ISI
| PUBMED
7. Verhagen WIM, Huygen PLM, Joosten EMG. Familial progressive vestibulocochlear dysfunction. Arch Neurol. 1988;45:766-768.
ABSTRACT
8. Verhagen WIM, Huygen PLM, Theunissen EJJM, Joosten EMG. Hereditary vestibulo-cochlear dysfunction and vascular disorders. J Neurol Sci. 1989;92:55-63.
FULL TEXT
|
ISI
| PUBMED
9. Verhagen WIM, Huygen PLM, Joosten EMG. Familial progressive vestibulocochlear dysfunction [letter]. Arch Neurol. 1991;48:262.
ISI
| PUBMED
10. Verhagen WIM, Huygen PLM, Bles W. A new autosomal dominant syndrome of idiopathic progressive vestibulo-cochlear
dysfunction with middle-age onset. Acta Otolaryngol (Stockh). 1992;112:899-906.
PUBMED
11. Verhagen WIM, Bom SJH, Huygen PLM, Fransen E, Van Camp G, Cremers CWRJ. Familial progressive vestibulocochlear dysfunction caused by a COCH mutation (DFNA9). Arch Neurol. 2000;57:1045-1047.
FREE FULL TEXT
12. Fransen E, Verstreken M, Verhagen WIM, et al. High prevalence of symptoms of Menière's disease in three families
with a mutation in the COCH gene. Hum Mol Genet. 1999;8:1425-1429.
FREE FULL TEXT
13. Lemaire FX, Feenstra L, Huygen PL, et al. Progressive late-onset sensorineural hearing loss and vestibular impairment
with vertigo. Otol Neurotol. In press.
14. Halpin C, Khetarpal U, McKenna M. Autosomal-dominant progressive sensorineural hearing loss in a large
North American family. Am J Audiol. March 1996;5:105-111.
FREE FULL TEXT
15. Khetarpal U. DFNA9 is a progressive audiovestibular dysfunction
with a microfibrillar deposit in the inner ear. Laryngoscope. 2000;110:1379-1384.
FULL TEXT
|
ISI
| PUBMED
16. Khetarpal U, Schuknecht HF, Gacek RR, Holmes LB. Autosomal dominant sensorineural hearing loss: pedigrees, audiologic
findings, and temporal bone findings in two kindreds [review]. Arch Otolaryngol Head Neck Surg. 1991;117:1032-1042.
ABSTRACT
17. de Kok YJM, Bom SJH, Brunt TM, et al. A Pro51Ser mutation in the COCH gene is associated
with late onset autosomal dominant progressive sensorineural hearing loss
with vestibular defects. Hum Mol Genet. 1999;8:361-366.
FREE FULL TEXT
18. Fransen E, Verstreken M, Bom SJH, et al. A common ancestor for COCH-related cochleovestibular
(DFNA9) patients in Belgium and the Netherlands, all bearing the P51S mutation. J Med Genet. 2001;38:61-64.
FREE FULL TEXT
19. Coucke PJ, Van Hauwe P, Kelley PM, et al. Mutations in the KCNQ4 gene are responsible
for autosomal dominant deafness in four DFNA2 families. Hum Mol Genet. 1999;8:1321-1328.
FREE FULL TEXT
20. Van Hauwe P, Coucke PJ, Ensink RJ, Huygen P, Cremers CW, Van Camp G. Mutations in the KCNQ4 K+ channel
gene, responsible for autosomal dominant hearing loss, cluster in the channel
pore region. Am J Med Genet. 2000;93:184-187.
FULL TEXT
|
ISI
| PUBMED
21. Kubisch C, Schroeder BC, Friedrich T, et al. KCNQ4, a novel potassium channel expressed
in sensory outer hair cells, is mutated in dominant deafness. Cell. 1999;96:437-446.
FULL TEXT
|
ISI
| PUBMED
22. Beisel KW, Nelson NC, Delimont DC, Fritzsch B. Longitudinal gradients of KCNQ4 expression
in spiral ganglion and cochlear hair cells correlate with progressive hearing
loss in DFNA2. Brain Res Mol Brain Res. 2000;82:137-149.
PUBMED
23. Bosman AJ, Smoorenburg GF. Intelligibility of Dutch CVC syllabes and sentences for listeners with
normal hearing and with three types of hearing impairment. Audiology. 1995;34:260-284.
ISI
| PUBMED
24. Documenta Geigy. Scientific Tables. 5th ed. Basel, Switzerland: S Karger AG; 1959:47.
25. Kharkovets T, Hardelin J, Safieddine S, et al. KCNQ4, a K+ channel mutated in
a form of dominant deafness, is expressed in the inner ear and the central
auditory pathway. Proc Natl Acad Sci U S A. 2000;97:4333-4338.
FREE FULL TEXT
26. Van Hauwe P, Coucke P, Van Camp G. The DFNA2 locus for hearing impairment: two genes regulating K+ ion
recycling in the inner ear. Br J Audiol. 1999;33:285-289.
ISI
| PUBMED
27. Manolis EN, Yandavi N, Nadol JB Jr, et al. A gene for non-syndromic autosomal dominant progressive postlingual
sensorineural hearing loss maps to chromosome 14q12-13. Hum Mol Genet. 1996;5:1047-1050.
FREE FULL TEXT
28. Robertson NG, Lu L, Heller S, et al. Mutations in a novel cochlear gene cause DFNA9, a human nonsyndromic
sensorineural deafness with vestibular dysfunction. Nat Genet. 1998;20:299-303.
FULL TEXT
|
ISI
| PUBMED
29. Khetarpal U. Autosomal dominant sensorineural hearing loss: further temporal bone
findings. Arch Otolaryngol Head Neck Surg. 1993;119:106-108.
ISI
| PUBMED
30. Nobili R, Mammano F, Ashmore J. How well do we understand the cochlea [review]? Trends Neurosci. 1998;21:159-167.
FULL TEXT
|
ISI
| PUBMED
RELATED ARTICLE
Archives of Otolaryngology–Head & Neck Surgery Reader's Choice: Continuing Medical Education
Arch Otolaryngol Head Neck Surg. 2001;127(9):1146-1148.
FULL TEXT
THIS ARTICLE HAS BEEN CITED BY OTHER ARTICLES
Audiometric Characteristics of a Dutch Family Linked to DFNA15 With a Novel Mutation (p.L289F) in POU4F3
Pauw et al.
Arch Otolaryngol Head Neck Surg 2008;134:294-300.
ABSTRACT
| FULL TEXT
A Dutch Family With Hearing Loss Linked to the DFNA20/26 Locus: Longitudinal Analysis of Hearing Impairment
Kemperman et al.
Arch Otolaryngol Head Neck Surg 2004;130:281-288.
ABSTRACT
| FULL TEXT
|
