Use of the auditory brainstem response testing in the clinical evaluation of the patients with diabetes mellitus

Journal of the Neurological Sciences 181 (2000) 29–32 www.elsevier.com / locate / jns

Use of the auditory brainstem response testing in the clinical evaluation of the patients with diabetes mellitus a, b c a a Yildirim Bayazit *, Mustafa Yilmaz , Yalc¸in Kepekc¸i , Semih Mumbuc¸ , Muzaffer Kanlikama a

Department of Ear Nose Throat, Faculty of Medicine, Gaziantep University, Kolejtepe, Gaziantep, Turkey b Department of Neurology, Faculty of Medicine, Gaziantep University, Kolejtepe, Gaziantep, Turkey c Department of Internal Medicine, Faculty of Medicine, Gaziantep University, Kolejtepe, Gaziantep, Turkey Received 3 January 2000; received in revised form 4 August 2000; accepted 8 August 2000

Abstract The objective of the study was to assess whether a relationship exists between the auditory brain stem response (ABR) results and diabetes mellitus with and without complications. In the clinical and audiometry laboratory settings, diabetic patients with and without complications (retinopathy and / or nephropathy) were examined using ABR testing, and the results were interpreted for their applicability in clinical practice. Fifty-nine patients with diabetic retinopathy or nephropathy (study group) and 20 diabetic patients without any known diabetic complication (control group) were assessed with audiometry and ABR testing. ABR revealed that the absolute latencies and interwave intervals of the waves I through V were prolonged significantly in the study group when compared to the control group. The amplitudes of waves I through V were diminished in the study group when compared to the control group, but a statistical significance was present only for wave V amplitude. Quantitative (wave I to wave V amplitude ratio) and qualitative analyses of the ABR waves showed abnormal waveforms in the study and control groups by 55.2 and 27.6%, respectively. There is a brain stem neuropathy in diabetes mellitus which can be assessed with ABR testing. The likelihood of encountering a diabetic complication increases as the ABR results become abnormal.  2000 Elsevier Science B.V. All rights reserved. Keywords: Diabetes mellitus; Retinopathy; Nephropathy; Auditory brainstem responses

1. Introduction Auditory brainstem response (ABR) testing can be used to record evoked potential waves generated by the auditory pathway. By means of ABR, it is possible assess the integrity of neural brainstem generators. The origin of wave I is the proximal cochlear nerve, the wave II is of distal cochlear nerve origin, and III is of lower brainstem origin while the source of the waves IV and V is in the higher brainstem. The amplitude ratio of wave I–V is a precise indicator of retrocochlear or lower brainstem pathology [1,2]. Complications of diabetes mellitus occur due mostly to

microangiopathy and neuropathy. This is true for the central nervous system involvement and possibly for the retinopathy and nephropathy. In diabetic patients, abnormalities in the ABR waves can be seen, and ABR testing has a role in the assessment of diabetic neuropathy [3,4]. It was also suggested that multimodal evoked potential evaluation was a sensitive and early diagnostic approach to the study of diabetic complications [5]. In this study, we presented the results of ABR changes in diabetic patients, and assessed whether a relationship existed between ABR changes and diabetic complications, retinopathy and nephropathy. 2. Materials and methods

*Corresponding author. Tel.: 190-342-336-5400; fax: 190-342-3365505. E-mail address: [email protected] (Y. Bayazit).

Fifty nine patients who were diagnosed as having complications related to diabetes mellitus constituted the

0022-510X / 00 / $ – see front matter  2000 Elsevier Science B.V. All rights reserved. PII: S0022-510X( 00 )00400-7

Y. Bayazit et al. / Journal of the Neurological Sciences 181 (2000) 29 – 32

30

study group. There were 40 male and 19 female patients, with a mean age of 47 years. Thirty four had insulindependent diabetes mellitus (IDDM) and 25 had noninsulin-dependent diabetes mellitus (NIDDM). Thirty eight patients had diabetic retinopathy, and 11 had microalbuminurea that was considered to be a predictor of the diabetic nephropathy, and ten had both retinopathy and microalbuminurea. The control group consisted of 20 diabetic patients without any known complication of diabetes such as retinopathy or nephropathy. There were 13 males and seven female patients in the control group, with a mean age of 51, of whom 11 had IDDM and nine had NIDDM. All patients in the study and control groups had been taking medical treatment for diabetes at least for 5 years. According to the past history, and physical and neurological examinations, none of the patients had neurological disease, cerebrovascular accident or cranial nerve palsy. Otoscopic examination of the ears was made by the otolaryngologist, and the patients who had conductive type hearing loss or otitis media with effusion were excluded. The patients whose blood chemistry results, including serum creatinine levels, were consistent with renal failure also were not included in the study. Each patient underwent audiometric examination including pure tone averages at the frequencies of 500, 1000, 2000 and 4000 Hz, speech discrimination scores and tympanometry as well as ABR testing. Patients with moderate or profound hearing loss were not included in the study. ABR results were interpreted for absolute wave latency of the waves I, III and V, and for interpeak latency of the waves I–III, III–V and I–V. Amplitudes of waves I, III and V were calculated. Mann–Whitney U-test in SPSS statistical package was used to perform statistical analysis of the ABR results. Morphological analysis of the ABR waves was performed both qualitatively and quantitatively, and expressed as normal or abnormal wave morphology. The ratio of amplitudes of wave I and wave V was calculated for quantitative assessment, and a wave I to wave V amplitude ratio less than 1 was considered normal wave morphology,

while when this ratio was above 1 it was considered abnormal wave morphology. Further, a qualitative evaluation of wave morphology was performed by subjective assessment of the otolaryngologist. The qualitative (subjective) assessment was made as follows: the responses were recorded ipsilaterally and repeated twice to ensure waveform consistency in a sound isolated room and impedance ,3 Vat. In the case of inability to obtain consistent waveforms at the first two recordings, additional recordings were performed. When more than two recordings were needed in order to obtain the actual waveform, this condition was considered to be an indicator of abnormal wave morphology. An ABR system was used with (Intelligent Hearing System, Multichannel Evoked Potential System, Miami, FL) filters 100–3000 Hz, repetitions 1024, stimulus intensity 80 dB HL and stimulation rate 11 rarefaction clicks per second.

3. Results Audiometry revealed that the pure tone averages of the patients in the study and control groups were 39 and 33 dB, respectively. ABR revealed that the absolute latencies of the waves I, III and V were prolonged significantly in the study group when compared to the control group. The interpeak latencies of the waves I–III, III–V and I–V were longer in the former than the latter group (Table 1). The amplitudes of the waves I through V were diminished in the study group when compared to the control group, but a statistical significance was present only for wave V amplitude while the other waves were not significantly different (Table 2). The analysis of ABR (both quantitative and qualitative) waves showed abnormal wave morphology in 55.2 and 27.6% of the patients in the study and control groups, respectively.

Table 1 Latency of the auditory brainstem response waves in diabetic patients with and without complications a Subjects

Absolute latency (ms)

Interpeak latency (ms)

I

III

V

I–III

III–V

I–V

DM1C Mean S.D.

2.10 0.32

4.65 0.43

6.92 0.46

2.55 0.38

2.27 0.40

4.82 0.47

DM Mean S.D.

1.87 0.33

3.91 0.51

6.11 0.35

2.04 0.30

2.10 0.37

4.24 0.41

Statistical result (Mann–Whitney U-test)

P,0.001 z526.44

P,0.001 z526.43

P,0.001 z526.49

P,0.001 z526.49

P.0.05 –

P,0.001 z524.52

a

DM, diabetes mellitus; C, complication; S.D., standard deviation; S, significant; NS, not significant.

Y. Bayazit et al. / Journal of the Neurological Sciences 181 (2000) 29 – 32 Table 2 Amplitude of the auditory brainstem response waves in diabetic patients with and without complications Subjects

Wave amplitude (mV) I

III

V

DM1C Mean S.D.

0.36 0.19

0.51 0.20

0.39 0..20

DM Mean S.D.

0.40 0.15

0.46 0.23

0.55 0.26

Statistical result (Mann–Whitney U-test)

P.0.05 –

P.0.05 –

P,0.001 z527.74

4. Discussion Diabetes can lead to multisystem disorders or complications that are likely mediated through angiopathy. However, the pathophysiological process leading to complications of diabetes such as retinopathy, nephropathy and peripheral or central neuropathy has not been understood entirely. The microvascular disease in diabetes is central to its complications, but metabolic control of diabetes does not reverse the ABR abnormalities [6]. Furthermore, it was reported that there was no relationship between the duration and level of control of diabetes and the central diabetic neuropathy that was diagnosed on ABR testing [7]. Diabetic patients should be evaluated individually, and those who are susceptible or at risk to develop complications, should be identified early in the course of the disease. That is, early awareness of the complication before it has been clinically evident, is essential in order to treat it appropriately. At this point, the function of the ABR testing in the clinical evaluation of diabetes needs to be illuminated more precisely. The wave I delay on ABR testing reflects a hearing loss or middle ear damage. The patients in this study had a mild hearing loss. This can cause a delay in the wave I latency. But, since there was not moderate or severe hearing loss, this delay can also be attributed to cochlear nerve involvement by the diabetes that could decrease the nerve conduction velocity. Diabetes can cause recruitment through cochlear involvement. But in such a case we would expect a diminished I–V interval accompanying the wave I latency. However, there was a wave I–V interval prolongation on ABR testing which could suggest a retrocochlear or brainstem involvement [8]. Therefore, the prolongations of intervals between the waves I through V disclosed the existence of a neuropathy at the level of the brain stem in the diabetics with or without complications. Additionally, the high rate of abnormal wave morphology in diabetic patients enhanced our results that a retrocochlear or brainstem neural involvement was present. We did not search for the ABR changes in the subtypes

31

of diabetes (IDDM or NIDDM) separately. The patients in this study had either IDDM or NIDDM. Therefore, the overall results of both types of diabetes were presented together. Nevertheless, similar ABR testing results were reported previously in patients with IDDM [9]. Thus, it would be reasonable to postulate that it was not the type of diabetes (insulin dependent or non-dependent), but rather the presence of complication might be the mechanism involved in the worsening of the ABR results. However, in order to clarify whether type of diabetes has an influence on the ABR, further study is required. There is much evidence that ABR results become abnormal in DM. The existence of more ABR abnormalities in diabetic retinopathy as compared to diabetes without retinopathy was previously shown [10]. In this study, high rates of ABR abnormalities were also found when diabetes was accompanied with the complications. In other words, the brainstem neuropathy as noted on ABR testing is even worse in diabetes with complications than without complication. It was shown that ABR impairment had been high in diabetics with cardiovascular autonomic failure [11]. This supports our contention that the ABR latencies are elongated when diabetes is complicated. It also was shown that prolonged ABR latencies could be detected in the absence of clinical abnormalities in the visual and hearing systems [12]. This is an important result, because ABR may have a predictive value for the diabetes-related complications. In other words, existence of a complication may be suspected when ABR abnormalities (prolonged absolute wave latencies and interwave intervals, and abnormal wave morphology) are noted in a diabetic patient. Better understanding of the predictive value of ABR in relation with the diabetic complications as they are at the subclinical stage needs further investigation. In conclusion, the results indicate a brainstem neuropathy in DM which can be assessed with ABR testing. As the ABR is more severely impaired, the likelihood of encountering a diabetic complication increases. The ABR abnormality is high when a clinically evident diabetic complication is present. In the clinical practice, ABR testing can be used as a complementary tool to the periodic laboratory examinations of diabetic patients who are likely to develop complications. The use ABR testing as a predictor or diagnostic tool for diabetic complications should be supported with further research, and ABR standards should be established for this purpose. Understanding the quantitative relationship between DM complication ABR requires further research as well.

References [1] Musiek F, Kibbe D, Rackliffe L, Weider D. The ABR I to V amplitude ratio in normal, cochlear and retrocochlear ears. Ear Hear 1984;5:52–5.

32

Y. Bayazit et al. / Journal of the Neurological Sciences 181 (2000) 29 – 32

[2] Musiek F. ABR in eight nerve and brainstem disorders. Am J Otol 1982;3:243–8. [3] Martini A, Comacchio F, Magnavita V. Auditory brainstem and middle latency evoked responses in the clinical evaluation of diabetes. Diabet Med 1991;8:74–7. [4] Zieger D, Hubinger A, Gries FA. Changes in brain stem auditory revoked potentials during insulin induced hypoglycemia in type I diabetic patients. Diabet Med 1991;8:805–11. [5] Pietravalle P, Morano S, Cristina G et al. Early complications in type 1 diabetes: central nervous system alterations preceded kidney abnormalities. Diabetes Res Clin Pract 1993;21:143–54. [6] Virtaniemi J, Kuusisto J, Karjalainen L et al. Improvement of metabolic control does not normalize auditory brain stem latencies in subjects with insulin dependent diabetes mellitus. Am J Otolaryngol 1995;16:172–6. [7] Buller N, Shvili Y, Laurian N et al. Delayed brain stem auditory evoked responses in diabetic patients. J Laryngol Otol 1988;102:857–60.

[8] Bayazit Y. Neurovascular decompression for tinnitus. J Neurosurg 1998;89:1072. [9] Virtaniemi J, Laakso M, Karja J et al. Auditory brain stem latencies in type I (insulin dependent) diabetic patients. Am J Otolaryngol 1993;14:413–8. [10] Bayazit Y, Bekir N, Gungor K et al. The predictive value of auditory brainstem responses for diabetic retinopathy. Auris Nasus Larynx 2000;27:219–22. [11] Martini A, Comcchio F, Fedele D. Auditory brain stem evoked responses in the clinical evaluation and follow up of insulin dependent diabetic subjects. Acta Otolaryngol Stockh 1987;103:620–7. ¨ U, Tezic T. Brain stem auditory evoked [12] Akinci A, Deda G, Karagol potential, visual evoked potential and nerve conduction velocity and their relation with HbA1c and beta 2 microglobulin in children with insulin dependent diabetes mellitus. Turk J Pediatr 1994;36:279–87.