|Year : 2020 | Volume
| Issue : 3 | Page : 122-126
Speech recognition in noise in patients with type II diabetes
Somayeh Falahzadeh1, Sima Tajik2, Faezeh Azadi3, Farnoosh Farjadi3
1 Communication Disorder Research Center, Department of Audiology, School of Rehabilitation Sciences, Isfahan, Iran
2 Department of Audiology, School of Rehabilitation, Babol University of Medical Sciences, Mazandaran, Iran
3 Department of Audiology, Isfahan University of Medical Sciences, Isfahan, Iran
|Date of Submission||17-Sep-2019|
|Date of Decision||15-May-2020|
|Date of Acceptance||13-Jul-2020|
|Date of Web Publication||22-Dec-2020|
Ms. Somayeh Falahzadeh
Room No. 345, School of Rehabilitation Sciences, Isfahan University of Medical Sciences, Hezarjerib St., Isfahan
Source of Support: None, Conflict of Interest: None
Context: The chronic diabetes is associated with damage to the sensory and cognitive regions brain. The central auditory system is susceptible to the damage caused by high glucose level. Aims: Since the healthy auditory system plays an important role in communication, this study examined speech recognition in noise performance of these people so as to better identify the harmful impacts of diabetes on the auditory processing. Settings and Design: This cross-sectional comparative study compares the speech recognition in noise performance of 30 diabetic patients and 30 normal individuals aged 30–55 years with quick speech in noise (Q-SIN) test. Subjects and Methods: All people had normal hearing and the speech recognition performance in silent, the Persian version of the Q-SIN test was used. Statistical Analysis Used: Results of signal-to-noise ratio loss (SNR loss) and recognition of words at different SNR levels were analyzed with Chi-square test and independent t-test in two groups. Results: There was a significant difference between diabetic patients and normal individuals in mean of SNR loss (P < 0.05). The comparison of word recognition scores in each SNR showed no significant difference in 25, 20 SNRs between the two groups (P > 0.05), but the performance of diabetic patients was weaker in 15, 10, 5, 0 SNR (P < 0.05). Conclusions: In the absence of hearing loss, the diabetic patients have a significant speech perception disorder, especially at lower levels of SNRs, compared to normal people of the same age. Impaired speech comprehension in the presence of a competitive message can result from the damage to central auditory processing as a result of diabetes.
Keywords: Diabetes, quick speech in noise test, speech in noise
|How to cite this article:|
Falahzadeh S, Tajik S, Azadi F, Farjadi F. Speech recognition in noise in patients with type II diabetes. Indian J Otol 2020;26:122-6
| Introduction|| |
Diabetes is a heterogeneous group of metabolic disorders resulting from defects in insulin secretion, insulin activity or both, which is characterized by chronic hyperglycemia. The effects of diabetes include long-term damage, malfunction, and various organ defects, especially the eyes, kidneys, heart, and blood vessels., The diabetes is important mostly due to its high prevalence and numerous complications. The World Health Organization has announced diabetes as an underlying epidemic due to its growing prevalence. An increase in the global prevalence of diabetes in adults from about 285 million in 2010 to about 552 million by 2030 is due to population aging, lifestyle changes as well as economic development, and increased obesity. It is estimated that Iran will rank second in the Middle East after Pakistan up to 2030 considering the growing annual prevalence of diabetes. The prevalence of diabetes in different regions of Iran is estimated to be between 7% and 16%. However, diabetes is an untreatable disease and attempts should be made to prevent its chronic complications so as to control it. Previous studies showed different results regarding diabetes and hearing impairment., The most commonly finding is bilateral sensorineural hearing loss at high frequencies, the lower amplitude of auditory evoked emissions, increasing the latency of auditory brainstem response (ABR) waves.,,, Researchers state that auditory changes are due to neuropathy, angiopathy, or the combination of both agents. Increased thickness of the capillary wall of striavascularis, basal membrane vascular involvement, significant reduction of external hair cells, decreased ganglion cells, and demyelination of the nerve sheath VIII has been observed in diabetic patients., The relationship between diabetes and auditory/vestibular dysfunction has been going on for more than a century. However, no cause-effect relationship has been established, and there is still controversy over the findings of diabetes audiology and pathology., According to the results, it seems necessary to carry out further research on the understanding of the complications of diabetes on the structures of the peripheral and central auditory system.
Signal speech varies rapidly. Speech comprehension is the process by which sounds are heard, interpreted, and then understood. Speech plays an important role in human communication and impaired speech processing can disrupt communication between individuals. On the other, understanding real-world speech does not occur in pristine acoustic environment, rather in the presence of background noise. The capacity of the auditory system to process complex sounds in the presence of background noise is essential for the proper speech perception. This process is even challenging for adults with normal hearing and cognitive abilities.,, People with impaired speech perception in noise performance complain of auditory fatigue, auditory without meaning perception, discomfort with the presence of background noise and the inability to understand conversations.
Central auditory processing plays an important role in speech comprehension. Neuronal degeneration has been reported in several studies in people with Type II diabetes. The speech discrimination scores on silence and noise in diabetic patients with a normal pure tone audiometry (PTA) threshold are lower than in the control group, with a greater difference in speech in noise (SIN). Abnormalities have also been shown in the central auditory pathways of diabetics. However, the pathogenesis is still unclear.
Henriques and Costa warned that PTA thresholds and words assessment alone were not appropriate for a credible and extensive identification of communication abilities; rather, recognition tests in silence and noise would allow direct assessment of the communication capabilities of individuals, so, they are valuable assessment tools for analyzing hearing abilities in situations similar to everyday auditory experiences. One of these tests includes quick SIN (Q-SIN). Q-SIN is a quick test to quantify the auditory ability in noise so as to recognize the sentence in an open set. To do the assessment, the signal-to-noise ratio (SNR) loss is used. A set of sentences are provided concurrently with the babble noise, and the person must repeat the sentence. The Persian version of the Q-SIN test was developed by Tahaei and Sameni in 2008. Each list contains 6 sentences with 5 key words per sentence, which is presented with the voice of a female speaker in a 4-speacker babble noise and six SNRs. Noninvasive behavioral tests can identify defects easily and can be used to treat them.
Although neuropathy, one of the most common complications in diabetes, affects up to 50% of patients, little attention has been paid to the neurological side effects of diabetes, including auditory neuropathy and central auditory pathways. These defects can lead to hearing problems, even if there are normal hearing thresholds. Early diagnosis of changes in auditory sensitivity and nerve function in diabetic patients can provide incentives for adopting measures to control and monitor the complications of the disease and a better quality of life for diabetic patients. Because few studies have examined the behavioral consequences of neurological complications in patients with diabetes, Therefore, the aim of the present study was to compare quick speech perception in noise performance of people with Type II diabetes and normal people by calculating SNR loss.
| Subjects and Methods|| |
The present descriptive-analytical and cross-sectional comparative study was performed on 30 adult patients with Type II diabetes and 30 normal adults. Patients were selected using convenience sampling method and referred from Isfahan University of Medical Sciences Research Centre to the Audiology Clinic of the Faculty of Rehabilitation Sciences. The control group was matched in terms of age and gender. Inclusion criteria for diabetic group include suffering from diabetes for at least 2 years, patients aged 30–55 years, hearing thresholds of <25 dB within the frequency range of 250–8000 Hz, normal tympanogram, normal speech recognition performance in silent, no history of etiological, cognitive disorders, head trauma, neurology (such as epilepsy and multiple sclerosis), and lack of taking sedatives 48 h before the test. After completing the consent form, the required information was recorded in the patient history form, and then the Persian version of Edinburgh Handedness Inventory was completed to ensure the right handedness of the individuals. The test completion procedure was then explained to participants. Initially, individuals underwent audiometry examination with an otoscope, a 226-Hz tympanometry, an air conduction audiometry test at frequencies of 250–8000 Hz, and bone conduction audiometry test at frequencies of 250–4000 Hz, speech discrimination score test in silent. Q-SIN test was performed in case of meeting the inclusion criteria. The test was conducted in an acoustic chamber with an actuator recorded on a compact disc and using a laptop connected to the standard AC40 dual-channel audiometer and a standard earphone. The stimulators consist of a set of sentences and multi-speakers babble noise that are at played to the both ears at the intensity level of 60 dB simultaneously; in other words, there is a competitive relationship between noise and signal. The SNR decreases from 25 to 0 in 5-dB steps in each list. Each individual is required to fully repeat each sentence. In order to ensure the patient's correct perception, five sentences were presented in the form exercises. The total number of correct words was recorded and SNR loss calculated using the following formula: Total number of correct words–27.5 (dB) = SNR loss. The SNR loss is obtained from the total score of five sentences. Kolmogorov–Smirnov test was used to investigate normal distribution, Chi-square test to investigate gender distribution in two groups, independent t-test to compare mean SNR loss and word recognition score in each SNR between normal individuals and diabetic patients.
| Results|| |
Data analysis was performed using SPSS software (version 22; IBM, Armonk, NY, USA) and at the significant level of P < 0.05. A total of 23 diabetic women and 7 men with a mean and standard deviation (SD) 46.17 ± 4.45 years and 24 normal women and 6 normal men aged 30–60 years with mean ± SD of 42.9 ± 7.25 participated in the present study. Chi-square test did not show a significant difference between the two groups in terms of gender distribution (P = 0.7). T-test also, showed no significant difference between the two groups in terms of age distribution (P = 0.05). Results of comparing mean and SD of SNR loss showed a significant difference between diabetic (1.20 ± 0.79) and normal subjects (2.13 ± 0.55) (P = 0.00). Another indicator studied was the comparison of the word recognition sore in each SNR between the normal and diabetic groups. The word recognition scores for each SNR in the normal and diabetic groups are shown in [Table 1]. There was no significant difference between groups within 20–25 SNRs (P = 0.3) and as SNRs decreased to 15 (P = 0.00), 10 (P = 0.01), 5 (P = 0.03), and 0 (P = 0.009), the sentences recognition performance of diabetics becomes weaker. As shown in [Figure 1], at high levels of the SNR, the normal and diabetic group scores overlap. As the SNR decreases, the diabetic person's score decreases compared to the normal person.
|Table 1: The word recognition scores for each signal-to-noise ratio in the normal and diabetic groups|
Click here to view
|Figure 1: The word recognition sore in each signal-to-noise ratio between the normal and diabetic groups|
Click here to view
| Discussion|| |
The main objective of the present study was to evaluate the effect of Type II diabetes on the speech perception ability in patients with chronic diabetes history. To this end, the speech perception ability in noise of people with and without diabetes was compared of using Q-SIN test in Isfahan, Iran. Results of comparing the SNR loss revealed that normal people have a better speech perception performance than diabetic ones. These results are consistent with the previous findings of Bajaj et al. who showed long-term complications of Type II diabetes with reduced speech perception in noise. Frisina et al. also showed a significant difference between elderly diabetic and healthy people using hearing in noise noise test. AlJasser et al. provided evidence of significant impairment of SIN performance in Type I diabetics patients without increasing PTA thresholds. Silva et al. also showed a 20% reduction in the speech perception ability (speech perception threshold) of words in silent and noise among individuals with Type I diabetics.
The second objective of the present study was to evaluate the effect of diabetes on speech perception at different SNR levels. The results revealed no significant difference between normal and diabetic patients in terms of speech perception at high SNR levels (SNR = 25, 20) (P > 0.05), but as SNR levels decreased (SNR = 15, 10, 5, 0), the diabetic people exhibited a weaker performance than the normal ones. In the study of AlJasser et al., Type I diabetic patients showed significantly higher SNRs than healthy controls in the speech comprehension test in isolated and colocated spatial noise.
Overall, diabetic people seem to have a significant speech perception disorder, especially at lower SNR levels than normal people of the same age. To understand the effect of diabetes on auditory processing, it is vital to recognize the neurophysiological relationships of these processes. Microangiopathy in diabetics gradually results in neuropathy, which leads to various defects such as delayed nerve conduction, central auditory processing, and so on. These defects will have adverse effects on quality of life. Magnetic resonance imaging has showed several lesions in the pons and thalamus area of diabetic patients. An increasing latency of ABR waves was also reported in at least 64% of diabetic patients. This latency indicates delayed transmission of auditory stimuli in the auditory pathway at the brain stem and the midbrain as well as neuropathy at these levels in diabetic people. The early stages of separation of the auditory stream occur subcortically and timing signals necessary for speech perception and auditory stream segregation in brainstem are maintained using neural synchrony. Therefore, the presence of neuropathy could explain the result of the present study and decreased Q-SIN performance in diabetic patients., Weaker speech recognition in situations with adverse SNRs compared to high SNRs may in part be due to the negative effects of noise on neural synchrony that results in distorted speech expressions at the cortical and subcortical levels.
Hyperglycemia causes extensive tissue damage, especially, in the myelin sheath and other neurological components in the peripheral nervous system, cranial, and vestibular nerves. Exposure to glucose, even for short periods, can lead to anatomical and physiological destruction of the auditory system. Chronic hyperglycemic environments cause pathophysiological changes in the nervous system, oxidative stress, and decreased Na/K/ATPase activity.
The results of microscopic examination of temporal bone in diabetic patients also showed the spinal ganglion atrophy, reduced number of nerve fibers and spinal ganglion cells, demyelination of the nerve VIII, degenerative changes in ventral cochlear nucleus , dorsal cochlear nucleus cells, superior olive complex, medial geniculate body, and varying degrees of degenerative changes in both temporal lobes.
On the other hand, when the peripheral and central auditory system is inadequate, the higher activity level in the prefrontal cortex (PFC) facilitates the auditory perception in noise. Although its precise mechanism is not known yet, a larger and more active PFC successfully blocks inappropriate information from the peripheral system and facilitates the identification process. Another justification based on PFC's role is attention and memory. The results of brain imaging indicate that the thickness and volume of the PFC, which is strongly linked with the attention and memory region, is related with speech perception in noise ability. In addition, attention deficits have been associated with the inability to understand SIN. Therefore, it can be concluded that higher levels of cognitive function can affect bottom-up paths as up-down and improve the signal quality. Patients with Type II diabetes have further deficiencies in their cognitive impairment, including verbal memory, attention, processing speed, and reduced executive function of the frontal lobe. AlJasser et al. also attributed some of the problems of speech comprehension in the noise of these patients to nonfundamental defects, including cognitive defects, which may lead to changes in performance in the real world.
However, it should be noted that the results of the present study can only be (partly) correlated with brain processing, because there are few articles about higher auditory potentials among people with diabetes. In sum, the results of a subjective test alone cannot be definitively stated. It is advisable to use a set of tests to detect hearing impairment. Early diagnosis and intervention of auditory problems can provide a better communication quality in the daily life of these people. An annual cardiovascular, eye, and nephrology assessment is considered as a “standard of care” for diabetic people. The auditory assessment, including the peripheral and central auditory systems should also be added to this list.
The limitation of the present study is the lack of distinction between diabetic with and without neuropathy, ignoring the effect of hemoglobin A1c and the duration of the disease on the results of speech perception in noise. Furthermore, in the present study, the effect of cognitive factors has not been considered. It is suggested that the relationship between these factors and speech perception in the noise of diabetic patients be examined in another studies.
| Conclusion|| |
The results of the current study revealed impaired sentence recognition in undesirable auditory conditions in people with Type II diabetes, which may be due to the involvement of various structures of the central auditory nervous system. A PTA test alone is not suitable for detecting speech impairment in diabetic patients. Simple and fast central auditory behavioral tests that are similar to real-world performance seem more appropriate.
This article is the result of a research project with the Ethic Code No. IR.MUI.REC.1395.2179 approved by University of Medical Sciences and Health Services, Isfahan, Iran.
We would like to thank Mr. Arash Najimi, Ph.D. in Health Education and Health Promotion from the Department of Medical Education of Isfahan University of Medical Sciences, for their valuable guidance in this study.
Financial support and sponsorship
This study was supported by Communication Disorders Research Center, Isfahan University of Medical Sciences, Isfahan, Iran.
Conflicts of interest
There are no conflicts of interest.
| References|| |
Silva BC, Mantello EB, Freitas MC, Foss MC, Isaac ML, Anastasio AR. Speech perception performance of subjects with type I diabetes mellitus in noise. Braz J Otorhinolaryngol 2017;83:574-9.
Ugur AK, Kemaloglu YK, Ugur MB, Gunduz B, Saridogan C, Yesilkaya E, et al
. Otoacoustic emissions and effects of contralateral white noise stimulation on transient evoked otoacoustic emissions in diabetic children. Int J Pediatr Otorhinolaryngol 2009;73:555-9.
Olfatifar M, Karami M, Shokri P, Mehdi HS. Prevalence of chronic complications and related risk factors of diabetes in patients referred to the diabetes center of Hamedan Province. Sci J Hamadan Nurs Midwifery Fac 2017;25:69-4.
Larejani B, Zahedi F. Epidemiology of diabetes mellitus in Iran. J Diabetes Metab Disord 2001;1:1-8.
Diniz TH, Guida HL. Hearing loss in patients with diabetes mellitus. Braz J Otorhinolaryngol 2009;75:573-8.
Frisina ST, Mapes F, Kim S, Frisina DR, Frisina RD. Characterization of hearing loss in aged type II diabetics. Hear Res 2006;211:103-13.
Díaz de León-Morales LV, Jáuregui-Renaud K, Garay-Sevilla ME, Hernández-Prado J, Malacara-Hernández JM. Auditory impairment in patients with type 2 diabetes mellitus. Arch Med Res 2005;36:507-10.
Bajaj G, Puthuchery S, Bhat J, Ranjan R. Effect of type II diabetes on speech perception in noise. Int J Innov Res Dev 2014;3:50-4.
Sasso FC, Salvatore T, Tranchino G, Cozzolino D, Caruso AA, Persico M, et al
. Cochlear dysfunction in type 2 diabetes: A complication independent of neuropathy and acute hyperglycemia. Metabolism 1999;48:1346-50.
Konrad-Martin D, Austin DF, Griest S, McMillan GP, McDermott D, Fausti S. Diabetes-related changes in auditory brainstem responses. Laryngoscope 2010;120:150-8.
AlJasser A, Uus K, Prendergast G, Plack CJ. Subclinical auditory neural deficits in patients with type 1 diabetes mellitus. Ear Hear 2020;41:561-75.
Muhumina KM. Brainstem Responses to Speech Sounds in Individuals with Type 2 Diabetes Melitus Undergoing Insulin Therapy with Poor Glycaemic Control. Kerala University; 2019.
Falahzadeh S, Khosravi S, Rohani S. Systematic review of auditory behavioral tests with speech stimuli. Iran J Rehab Med 2019;8:35-47.
Wong PC, Ettlinger M, Sheppard JP, Gunasekera GM, Dhar S. Neuroanatomical characteristics and speech perception in noise in older adults. Ear Hear 2010;31:471-9.
Song JH, Skoe E, Banai K, Kraus N. Training to improve hearing speech in noise: Biological mechanisms. Cereb Cortex 2012;22:1180-90.
Song JH, Skoe E, Banai K, Kraus N. Perception of speech in noise: Neural correlates. J Cogn Neurosci 2011;23:2268-79.
Aarabi S, Jarollahi F, Badfar S, Hoseinabadi R, Ahadi MJ, Research V. Speech perception in noise mechanisms. AVR 2016;25:221-6.
Khalili M, Fatahi J, Hajiabolhassan F, Tahaei AA, Jalaei Sh. Test-retest reliability and list equivalency of the Persian quick speech in noise test. Mod Rehabil 2010;3:16-21.
Sharifinik M, Tajik S, Mohammadkhani G, Jalaie S. Comparison of gaps in noise test (gin) in adults with normal and conductive hearing loss. JRSR 2013;9:726-34.
Sanju HK, Kumar P. Annual audiological evaluations should be mandatory for patients with diabetes. Hear Rev 2016;23:14.
Makishima K, Tanaka K. Pathological changes of the inner ear and central auditory pathway in diabetics. Ann Otol Rhinol Laryngol 1971;80:218-28.
Kodl CT, Seaquist ER. Cognitive dysfunction and diabetes mellitus. Endocr Rev 2008;29:494-511.
Tajik S, Adel Ghahraman M, Tahaie AA, Hajiabolhassan F, Jalilvand Karimi L, Jalaie SH. Deficit of auditory temporal processing in children with dyslexia-dysgraphia. AVR 2012;21:76-83.