Premium
Is type 2 diabetes mellitus associated with alterations in hearing? A systematic review and meta‐analysis
Author(s) -
Akinpelu Olubunmi V.,
MujicaMota Mario,
Daniel Sam J.
Publication year - 2014
Publication title -
the laryngoscope
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.181
H-Index - 148
eISSN - 1531-4995
pISSN - 0023-852X
DOI - 10.1002/lary.24354
Subject(s) - medicine , confidence interval , diabetes mellitus , meta analysis , incidence (geometry) , hearing loss , audiology , auditory brainstem response , pure tone audiometry , type 2 diabetes , audiometry , endocrinology , physics , optics
Objectives/Hypothesis The aim of this study was to systematically and quantitatively review the available evidence on the effects of type 2 diabetes mellitus on hearing function. Data Sources and Review Methods Eligible studies were identified through searches of eight different electronic databases and manual searching of references. Articles obtained were independently reviewed by two authors using predefined inclusion criteria to identify eligible studies. Meta‐analysis was performed on pooled data using Cochrane's Review Manager. Results Eighteen articles fulfilled the inclusion criteria. Hearing loss (HL) was defined by all studies as pure tone average greater than 25dB in the worse ear. The incidence of HL ranged between 44% and 69.7% for type 2 diabetics, significantly higher than in controls (OR 1.91; 95% confidence interval 1.47–2.49). The mean PTA (pure tone audiometry) thresholds were greater in diabetics than in controls for all frequencies [test or overall effect Z = 3.68 , P = 0.0002]. Auditory brainstem response (ABR) wave V latencies were also statistically significantly longer in diabetics when compared to control groups [OR 3.09, 95% CI 1.82– 4.37, P < 0.00001]. Conclusions Type 2 diabetic patients had significantly higher incidence for at least the mild degree of HL when compared with controls. Mean PTA thresholds were greater in diabetics for all frequencies but were more clinically relevant at 6000 and 8000 Hz. Prolonged ABR wave V latencies in the diabetic group suggest retro‐cochlear involvement. Age and duration of DM play important roles in the occurrence of DM‐related HL. Level of Evidence 3a. Laryngoscope , 124:767–776, 2014