Chronic Conductive Hearing Loss Is Associated With Speech Intelligibility Deficits in Patients With Normal Bone Conduction Thresholds

Abstract

Objectives: The main objective of this study is to determine whether chronic sound deprivation leads to poorer speech discrimination in humans.

Design: We reviewed the audiologic profile of 240 patients presenting normal and symmetrical bone conduction thresholds bilaterally, associated with either an acute or chronic unilateral conductive hearing loss of different etiologies.

Results: Patients with chronic conductive impairment and a moderate, to moderately severe, hearing loss had lower speech recognition scores on the side of the pathology when compared with the healthy side. The degree of impairment was significantly correlated with the speech recognition performance, particularly in patients with a congenital malformation. Speech recognition scores were not significantly altered when the conductive impairment was acute or mild.

Conclusions: This retrospective study shows that chronic conductive hearing loss was associated with speech intelligibility deficits in patients with normal bone conduction thresholds. These results are as predicted by a recent animal study showing that prolonged, adult-onset conductive hearing loss causes cochlear synaptopathy.

Figure 1:. Ages, observation spans, and visit intervals for subjects in the five groups.

Means (±SEMs) are shown for each parameter.

Figure 2:. Mean hearing sensitivity in the affected ear for each cohort.

Mean Air-conduction (AC) and Bone-conduction (BC) thresholds on the CHL side of each cohort, color coded according to etiologies: three chronic types of CHL (Cong.: Congenital malformations of the external/middle ear; COM: Chronic Otitis Media; Chole.: Cholesteatoma) and two acute types of CHL group (AOM: Acute Otitis Media; OME: Secretory Otitis Media). Error bars are for SEMs

Figure 3:. Individual PTAs, by bone and air conduction, for the affected sides vs. contralateral ears.

Box and whiskers plots of AC- and BC-threshold PTAs (500, 1000 and 200 Hz) for each subject from each group. As defined in key, two degrees of hearing loss were considered for the CHL ears: this color coding convention will be carried forward in the remaining Figures.

Figure 4:. Change in bone conduction thresholds over the obserbation span.

Rate of PTA shift in each ear was computed over the entire observation period from first to last visit. As shown in the key, an ensemble average was computed for each group (black circles) as well as separate averages for each PTA group on the affected sides. Error bars are for SEMs

Figure 5:. Word recognition scores as a function of CHL and etiology.

For all conditions, WR scores were averaged, and error bars are for SEMs. Statistical significance of the post-hoc analysis (Steel-Dwass test for multiple comparisons) is indicated (*p<0.05).

Figure 6:. Predictibility of word recognition scores as a function of degree of CHL and etiology.

No statistically significant relationship was found between BC thresholds and WR score in any cohort (A-C). However, significant correlations were observed in the affected ear between AC PTAs and WR scores for all chronic CHL groups (Congenital malformations of the external/middle ear (D), Chronic Otitis Media and Cholesteatoma (E)). The same relationship did not reach statistical significance in CHL (F). The linear regression is shown when the coefficient correlation was significant. For all conditions, scores were obtained at the last visit. Statistical significance is indicated: *p<0.05; ***p<0.001.

Figure 7:. Effect of hearing loss configuration on word recognition scores in patients presenting a chronic CHL.

No statistically significant difference in WR scores was found from patient presenting a chronic CHL with different hearing loss configurations. U; Upward sloping, D; Downward sloping; F: Flat.