Auditory Spatial Deficits in the Early Stage of Ischemic Cerebral Stroke
jska-Grymaj1o, PhD,† and Dariusz Ga˛secki, PhD† Tomasz Przewo
zny, PhD,* Anna Go

Background: Clinical research, together with computed tomography/magnetic resonance imaging findings, proves that ischemic stroke (IS) that damages auditory pathways can cause hearing loss and impairment of higher auditory processes such as sound localization. The goal of the study was to find possible correlations between the IS risk factors, ischemic lesion volume and localization, neurologic status, and the sound localization capability in acute IS patients. Methods: We consecutively enrolled 61 IS patients into the study. The control group consisted of 60 healthy volunteers. All neuro-otological evaluations were performed up to 30 days from the incidence of stroke. All the subjects underwent the horizontal minimum audible angle test (HMAAT) and standard tonal and speech audiometric assessments. Results: HMMAT results were significantly worse in the IS patients and were present in 82.0% of the patients. There were more patients with unilateral disturbances than with bilateral ones (54.1% versus 27.9%). It was the characteristics of the ischemic lesions that correlated strongly with the sound localization deterioration, that is, their bilateral (the 90
azimuth, P 5 .018; the 180
, P 5 .002), multiple (the 45
, P 5.020; the 180
, P 5.007; the 225
, P 5.047), and lacunar character (the 90
, P 5 .015; the 225
, P 5 .042). Differences in the types of HMAAT results were significant for lesions in the frontal and the temporal lobe (P 5 .018 and P 5.040). In addition, worse sound localization ability was more common in patients with poor speech discrimination and the bilateral sensorineural hearing loss. We have not found statistically significant correlations for other analyzed factors such as the cortical/subcortical character of the lesions, the patients’ neurologic status, and cerebrovascular risk factors. Conclusions: Sound localization impairment is common in IS patients and it is the multiple, bilateral, and lacunar character of the ischemic lesions that seems to be strongly positively correlated with the disturbance of the sound localization ability. Key Words: Ischemic stroke—lacunar stroke—sound localization impairment—spatial hearing. Ó 2015 by National Stroke Association

Ischemic cerebral stroke can cause not only various otological symptoms such as the sensorineural hearing loss, tinnitus, or vertigo but also central auditory processing

From the *Department of Otolaryngology, Medical University of Gda
nsk, Gda
nsk; and †Department of Neurology, Medical University of Gda
nsk, Gda
nsk, Poland. Received February 3, 2015; revision received April 14, 2015; accepted May 1, 2015. Address correspondence to Tomasz Przewo
zny, PhD, Department of Otolaryngology, Medical University of Gda
nsk, 17, Smoluchowskiego Str., 80-214 Gda
nsk, Poland. E-mail: [email protected]. 1052-3057/$ - see front matter Ó 2015 by National Stroke Association http://dx.doi.org/10.1016/j.jstrokecerebrovasdis.2015.05.001

disorders. One of the central auditory processing disorders is the impairment of sound localization ability.1-11 This type of impairment was reported in different types of brain tissue damage, especially in case of ischemic lesions, both of the right and left cerebral hemisphere.12-21 Anatomic structures responsible for hearing and sound localization are organized on different central nervous system (CNS) levels and spread from the brainstem to the cortex. Thus, auditory dysfunction in stroke patients varies depending on the localization and volume of ischemic lesions damaging the auditory pathways. The nuclei of the trapezoid corpus, nuclei of the lateral lemniscus, superior nucleus of the oliva, and the nucleus

Journal of Stroke and Cerebrovascular Diseases, Vol. -, No. - (---), 2015: pp 1-12

1


T. PRZEWOZNY ET AL.

2

of the inferior colliculus of the tectal lamina constitute the anatomic basis of sound localization in the mechanism of interaural time delay and intensity.22 Pathologic vascular or neoplastic processes of these regions, although rare, impair sound localization substantially because they restrict the broad stream of information going to the upper levels of the CNS.10,15,23 Functional magnetic resonance imaging (fMRI) studies proved that tasks demanding sound localization in healthy subjects cause activation of lower part of the inferior parietal lobule and posterior parts of the middle and inferior frontal gyri, bilaterally.24-26 In addition, it was proved that sound localization process activates the planum temporal and parietotemporal operculum,27,28 and it seems that it is the temporal lobe that is most important in the sound localization on the cortical level.29 Thus, most of research focuses on the cortical–subcortical lesions of this region,2,4,8,10,12-21,30,31 and on the correlations between the impairment of sound localization and lesions in the right, left, or both hemispheres. The goal of this study was to find components of the ischemic cerebral stroke that correlate with the most profound impairment of the sound localization process.

Materials and Methods Stroke Patients Between May 2006 and January 2011, we consecutively enrolled 61 ischemic stroke (IS) patients into the study–33 men and 28 women admitted to the Adults’ Neurology Department of the Medical University of Gda
nsk, Poland. The average age of the patients was 56.2 6 17.3 years (range, 21-80 years). All subjects were right-handed. All neuro-otologic evaluations were performed during the early stage after the incidence of stroke (up to 30 days, average, 10 6 7 days). The diagnosis of stroke was based on the World Health Organization criteria in patients with neurologic symptoms lasting at least 24 hours. The exclusion criteria were: age older than 80 years, patients with previous history of stroke (but not transient ischemic attack), serious general state, dementia, neurodegenerative disorders, other previously identified neurologic diseases, patients without logical verbal contact due to aphasia, psychotic symptoms, visual spatial neglect syndrome tested with the line bisection test and the nonverbal shape cancellation task,32,33 conductive or mixed type hearing loss, asymmetric sensorineural hearing loss, and history of ear surgery. According to the Food and Drug Administration classification of the hearing loss (the pure middle tone average [PMTA] for values at 500, 1000, 2000, and 4000 Hz in the better ear), 40 subjects had hearing within normal limits and 21 presented the sensorineural hearing loss (17 mild and 4 moderate).

Control Group The control group consisted of 60 age-matched subjects, 28 men and 32 women. The average age of the group was 53.1 6 19.2 years (range, 21-80 years), and it consisted of healthy volunteers (mainly students and members of the hospital medical staff). The exclusion criteria for the control group were previously identified neurologic diseases, diabetes, circulatory insufficiency, alcoholism, smoking, use of medications affecting the CNS, history of noise exposure at work, hearing disorders including the conductive and the mixed type hearing loss, and history of ear surgery. All subjects underwent otological and neurologic examination. According to the Food and Drug Administration classification, 3 subjects had sensorineural mild hearing loss. All of them were aged older than 60 years, and their hearing loss was because of cochlear presbycusis.

Neurologic Examination and Localization of Ischemic Lesions Initial quantitative assessment of the patients was performed on the first day of the hospitalization, no more than 48 hours from the incidence of stroke. Full neurologic examination was performed in each of the IS patients. All of the IS patients received standard stroke unit treatment. During the hospitalization in the Department of Neurology, laboratory tests, electrocardiogram, chest X-ray, and ultrasound of the carotid and vertebral arteries were performed. In certain cases, echocardiogram and internal medicine and cardiological or vascular surgery consultations were necessary. The National Institutes of Health Stroke Scale (NIHSS) was used to examine patients neurologically.34 History of hypertension, coronary heart disease, diabetes, and hyperlipidemia was taken. The presence and the localization of ischemic lesions were identified by MRI (3.0 T). On the basis of the previously mentioned measures, patients were divided into stroke etiological subgroups: large-artery atherosclerosis, cardioembolism, small-artery disease, other etiologies, and mixed etiologies. Depending on the lateralization of ischemic lesions, they were divided into the right-sided, the left-sided, and the bilateral ones. In addition, localization of the lesions was further described as of (1) the frontal lobe; (2) the temporal lobe; (3) the parietal lobe; (4) the occipital lobe; (5) the midbrain; (6) the diencephalon; and (7) the cerebellum. This classification does not include lesions localized at the frontoparietal and the temporo– parieto–occipital border. In such cases, lesions were classified to 1 of the 7 groups on the basis of the localization of the biggest part of the lesion. To localize the lesions properly, radiologic tables of anatomy by Weir et al35 were used. Furthermore, the lesions were divided according to the classification by Ruff et al19 into the lesions of the right anterior hemisphere, the right posterior hemisphere, the left anterior hemisphere, the left posterior hemisphere, the cerebellum, and of the brainstem. According

AUDITORY SPATIAL DEFICITS IN STROKE

to this classification, anterior quadrants include frontal, frontocentral, and fronto-temporal regions and posterior quadrants include parietal, parieto-occipital, and parietotemporal regions. Finally, the lesions were divided according to the depth of their localization as (1) cortical lesions, (2) subcortical lesions (white matter and the basal ganglia: the striatum and the thalamus), and (3) mixed type lesions and their number–single and multiple ischemic lesions.

Hearing Tests Tonal Audiometry The standard tonal audiometry in a soundproof booth was performed in all the study subjects. Signals were generated by calibrated clinical audiometers: Midimate 622, manufactured by Madsen Electronics (Otometrics, Copenhagen, Denmark). The equipment had corrections for standard hearing level–ISO-389-1: 1998 for the air conduction and ISO-389-3: 1994 for the bone conduction. For the air conduction testing, the electrical signal was generated by the audiometer coupled with TDH-39P headphones. For bone conduction testing, the audiometer was coupled to a radioear B-71 bone-conduction vibrator (New Eagle, PA). The audiometric values are averages of values at 500, 1000, 2000, and 4000 Hz. The classification of the Food and Drug Association for the degree of hearing loss was used.36 We excluded patients from the study if they presented asymmetrical hearing loss–if the difference between the ears was more than 15 dB HL (hearing level). If the hearing threshold for frequencies 500-10002000-4000 Hz was more than 20 dBHL in one ear, the unilateral hearing loss was diagnosed, if in both ears, bilateral hearing loss (with the differences #15 dB HL between the ears in both situations). Speech Audiometry The monosyllabic consonant–vowel–consonant test in Polish was used, consisting of 20 words per list.37 Recorded words were presented in the free field. Signals were transmitted by loudspeakers C115, 8 Ohms impedance, manufactured by Martin Audio (London, Great Britain) and amplified by amplifier PA 210 manufactured by Madsen Electronics (GN Otometrics, Taastrup, Denmark) at 65 dB sound pressure level (SPL) in noise at a speech-to-noise ratio of 110 dB. The speech discrimination score was established, that is, the percentage of properly received signals. Measures below 70% were recognized as the poor speech discrimination. Horizontal Minimum Audible Angle Test (HMAAT) For the estimation of the horizontal minimum audible angle the wide-band noise bursts were used between 80 and 12.500 Hz. The signal was presented in 2 sets of 1-second bursts, separated by 2.5 seconds with rise–fall

3

times of 50 ms. Finally, a 4.5-second pause finished the cycle. The acoustic pressure level of these bursts was 85 dB SPL. Signals were reproduced from CD player and after amplification transmitted to the loudspeaker attached to a metal arm installed to the ceiling of the study room. The laser pointer was attached to this arm, which showed its displacement every 1
on the scaled table installed on the ceiling. The distance from the loudspeaker to the subject’s head was fixed at 50 cm. During the test, the subject was seated on the metal armchair with his head immobilized by a metal head holder and was blind-folded. The measurement was made in the free field in the horizontal plane at 8 equally spaced angles (every 45
) around the head for the azimuths: 0
, 45
, 90
, 135
, 180
, 225
, 270
, and 315
, and each time, the average angular value was calculated. For the 0
azimuth the loudspeaker was positioned in front, for the 45
azimuth on the right side, for the 180
azimuth behind, and for the 270
azimuth on the left side of the tested person’s head. The test was started from the 0
azimuth onward (counter-clockwise) and finished on the 315
azimuth. At each of all the 8 azimuths measurements were made with a movement of the sound source to the right and to the left, and the final result was the arithmetic mean value of both of these results, as in the method proposed by Mills.38 Each test was started by a short training at the 0
azimuth with eyes open and then closed. The examined person was instructed to inform verbally ‘‘from two places versus from one place’’ after the second signal was generated in another point of space. That value of angles was noted in a special form as the minimum audible angle (MAA) for each azimuth (on the right, on the left, and final MAA). The MAA was measured at each of the 8 azimuths. The upper limit of the reference range was set at the 95th percentile of the results received in the control group divided into 6 age subgroups (21-30; 31-40; 41-50; 51-60; 61-70; and 71-80 years). Results obtained in the tested group were considered incorrect if they exceeded the reference range for at least one of the azimuths. In case of incorrect MAA value measured for at least one of the 45
, 90
, and 135
azimuths the result was considered as the right-sided unilateral abnormal result of the test, and for the 225
, 270
, and 315
azimuths the result was considered as the left-sided unilateral abnormal result. In case of incorrect MAA values measured for both: at least one of the 45
, 90
, and 135
azimuths and for the 225
, 270
, and 315
azimuths, the result was considered as the bilateral abnormal result. In case of an incorrect MAA value measured only for the 0
and 180
azimuths the result was considered as normal. Finally, the results were reported as follows: normal; the unilateral incorrect result on the side of the ischemic lesion; the unilateral incorrect result opposite to the side of the ischemic lesion; the unilateral incorrect result with bilateral ischemic lesions; the bilateral incorrect result with unilateral lesion and the bilateral incorrect result with bilateral lesions.

4

Table 1. Mean HMAAT results for the IS patients and the controls and the univariate analysis of the risk factors for the spatial hearing deficits Azimuths Groups

0


45


90


135


180


225


270


315


IS patients, mean (
) Controls, mean (
) P value Variables Type of HMAAT resulty,z Number of ischemic lesionsjj,{ Lateralization of ischemic lesionsz,# Lobar stroke localizationz,** Quadrants of hemispheres affected by the ischemic lesion(s)z,yy Depth of the lesionsz,zz Stroke etiologyz,xx Neurologic statusz,jjjj Hypertension{,{{ Vascular risk factorsz,## Speech discrimination{,*** Type of hearing lossz,yyy

11.4 6 7.5 4.9 6 2.3 ,.001*

11.7 6 8.7 7.9 6 4.4 .037*

18.8 6 10.9 11.0 6 4.8 ,.001*

10.7 6 7.7 9.1 6 4.5 .630

10.2 6 7.3 7.1 6 2.8 .046*

11.1 6 7.5 7.4 6 2.6 .022*

17.7 6 10.5 10.0 6 4.4 ,.001*

10.5 6 8.0 8.0 6 4.3 .275

.012x .318 .240 .768 .145

,.001x .020x .095 .766 .758

,.001x .447 .018x .853 .860

.001x .772 .216 .814 .893

.005x .007x .002* .150 .144

.005x .047x .779 .828 .757

.002x .135 .308 .443 .395

.001x .198 .196 .471 .327

.500 .066 .885 .683 .852 .009x .208

.375 .015x .973 .455 .676 .011x .006x

.996 .085 .792 .456 .489 .012x .201

.494 .129 .543 .425 .765 .001x .058

.374 .439 .521 .753 .643 ,.001x .109

.869 .042x .419 .878 .538 .064x .165

.525 .076 .248 .305 .982 .024x .206

.984 .164 .207 .939 .980 .006x .163


T. PRZEWOZNY ET AL.

Abbreviations: ANOVA, analysis of variance; HMAAT, horizontal minimum audible angle test; IS, ischemic stroke. *P value is for comparison between the controls and the IS patients; Mann–Whitney U test. yOrdered as normal, the unilateral incorrect result on the side of the ischemic lesion, the unilateral incorrect result opposite to the side of the ischemic lesion, the unilateral incorrect result in case of bilateral ischemic lesions, the bilateral incorrect result with unilateral ischemic lesion, and the bilateral incorrect result with bilateral ischemic lesions. zANOVA rank Kruskal–Wallis test. xP value is for comparison between subgroups of tested factors, statistical significance for P , .05; ANOVA rank Kruskal–Wallis test. jjOrdered as the single lesion or multiple ischemic lesions. {Mann–Whitney U test. # Ordered as the right-sided, left-sided, or bilateral lesion/s. **Ordered as the right frontal lobe, the right temporal lobe, the right parietal lobe, the right occipital lobe, the right part of the midbrain, the right part of the brainstem and the cerebellum, the left frontal lobe, the left temporal lobe, the left parietal lobe, the left occipital lobe, the right part of the midbrain, the left part of the brainstem, and the cerebellum. yyOrdered as the right anterior hemisphere, the right posterior hemisphere, the left anterior hemisphere, the left posterior hemisphere, the cerebellum, and the brainstem. zzOrdered as the cortical, subcortical, or mixed lesions. xxOrdered as large-artery atherosclerosis, small-artery disease, cardioembolism, and other etiologies. jjjjAssessed with use of the National Institutes of Health Stroke Scale (ordered as 1-5 pt., 6-10 pt., 11-15 pt., .15 pt.). {{Ordered as hypertension present and no hypertension. ## Ordered as hypertension, coronary heart disease, hyperlipidemia, and diabetes. ***Ordered as speech discrimination higher/lower than 70% of words in the free field with 65 dB sound pressure level level, speech-to-noise ratio 10 dB. yyyOrdered as normal hearing, the unilateral hearing loss, and the bilateral symmetrical hearing loss.

AUDITORY SPATIAL DEFICITS IN STROKE

5

Statistics The statistical analysis was carried out with the MannWhitney U test for independent groups and the analysis of variance Kruskal–Wallis test. Values of P less than .05 were considered as statistically significant. Statistical analysis was performed using STATISTICA, version 7.1, StatSoft, Inc. (2005).

Ethical Approval The study was approved by the regional independent ethics committee (NKEB/32/2006). All the patients and the control subjects provided written, informed consent for involvement in the study.

Results The angular values of HMAAT for each of the azimuths in the IS patients and the control group are presented in Table 1 and Figure 1. Statistically significant differences were found for all of the azimuths apart from the 135
and 315
azimuth. Fifty patients (82.0%) had incorrect HMAAT results. The unilateral incorrect result was found in 33 patients (54.1%), with 17 cases (27.9%) on the side of the lesion, 12 cases (19.7%) opposite to the side of the lesion and 4 cases (6.6%) with bilateral lesions. The bilateral incorrect result was found in 17 patients (27.9%)–12 subjects (19.7%) with unilateral lesion and 5 subjects (8.2%) with bilateral lesions. Differences between the results of the patients of these subgroups were statistically significant for all of the azimuths. The highest HMAAT results were present in patients with bilateral lesions and bilateral sound localization disturbances. Slightly better results were present in subjects with unilateral ischemic lesion and bilateral sound localization disturbances, but still the results were much worse than those in the control group (Fig 2). Single ischemic lesion was found in 44 subjects (72.1%) and multiple lesions were found in 17 (27.9%). Patients with multiple lesions had worse sound localization in comparison with the single lesion patients. Statistically significant differences were present for the 45
, 180
i 225
azimuths. Right-sided lesions were observed in 30 cases (49.2%), left-sided in 22 (36.1%), and bilateral in 9 (14.8%). Worse results were found in case of the right-sided lesions, especially at the 90
and 270
azimuth; however, the highest HMAAT values were found in case of the bilateral lesions. The differences were statistically significant for the 90
and 180
azimuth. The localization of all the ischemic lesions taken together was as follows: frontal right, 8 lesions (7.7%); frontal left, 8 (7.7%); temporal right, 16 (15.4%); temporal left, 10 (9.6%); parietal right, 15 (14.4%); parietal left, 16 (15.4%); occipital right, 5 (4.8%); occipital left, 2 (1.9%);

Figure 1. HMAAT results in the IS patients and the controls. Black line– average values for the controls, red dotted line–average values found in the IS patients. Abbreviations: HMAAT, horizontal minimum audible angle test; IS, ischemic stroke; L, left; R, right. (Color version of figure is available online.)

midbrain right, 6 (5.8%); midbrain left, 6 (5.8%); brainstem/cerebellum right, 8 (7.7%); and brainstem/cerebellum left, 4 lesions (3.8%). The global number of lesions was 104 due to the fact that many of them could be assigned to more than one anatomic localization. The differences between the HMAAT results for the previously mentioned localizations were statistically insignificant but the worst sound localization was found in case of lesions localized in the left frontal lobe. Other localizations presented lower values of HMAAT (Table 1, Fig 2). We also correlated the types of the incorrect HMAAT results with number of lesions present in the previously mentioned localizations (Table 2). The incorrect HMAAT results were most abundant in case of lesions in the parietal and the temporal lobes (25 and 24 lesions, respectively). The statistically significant differences between the number of lesions in a specific lobe (localization) depending on the type of the HMAAT result was found for the frontal and the temporal lobe. The most common type of the HMAAT result found in case of the frontal lobe injury was the bilateral incorrect result with bilateral lesions. Similarly, it was the bilateral incorrect result but with unilateral lesion that was most common for the injury of the temporal lobe. According to the quadrant classification by Ruff et al,19 13 (17.6%) patients had right anterior hemisphere stroke, 22 (29.7%), right posterior hemisphere stroke; 14 (18.9%), left anterior hemisphere stroke; 14 (18.9%), left posterior hemisphere stroke; and 11 (17.2%), brainstem/cerebellar stroke. The highest HMAAT were found in the leftsided anterior stroke, similarly to the situation with the lobular classification, however, the differences were not statistically significant (Table 1, Fig 4)

6


T. PRZEWOZNY ET AL.

Figure 2. Differences in HMAATresults depending on the (A) type of HMAATresult; (B) number of ischemic lesions; (C) lateralization of the ischemic lesions; and (D) anatomic localization of the lesion. Abbreviations: BIR, bilateral incorrect result; HMAAT, horizontal minimum audible angle test; IL, ischemic lesion(s); UIR, unilateral incorrect result; L, left; R, right.

The ischemic lesions were cortical in 8 patients (13.1%), cortical–subcortical (mixed) in 13 (21.3%), and subcortical in 40 patients (65.6%). Results of the sound localization test did not differ significantly statistically between the cortical and the subcortical localizations. Thirty-nine (63.9%) patients presented with IS due to the large-artery atherosclerosis, 11 (18.0%) the small-artery disease (lacunar) stroke, 8 (13.1%) the cardioembolic stroke, and 3 (4.9%) developed stroke of other etiology. The highest HMAAT values were found in the subgroup of the small-artery disease stroke. However, the differences were statistically significant only for the 45
and 225
azimuth (Table 1, Fig 2). The mean NIHSS result on admission was 7.3 6 2.4 pts. Patients were divided in 4 subgroups according to the NIHSS results: 1-5 (22; 36.1%), 6-10 (30; 49.2%), 11-

15 (6; 9.8%), and more than 15pt. (3; 4.9%). The higher was the NIHSS result, the higher were the MAA values, particularly in case of the left-sided azimuths (Table 2, Figs 3 and 4). Forty-four (72.1%) of the IS patients reported chronic systemic disease in anamnesis. Thirty-eight patients (62.3%) had hypertension, 25 (41.0%) coronary heart disease, 10 (16.4%) diabetes, and 11 (18.0%) dyslipidemia. None of the cardiovascular risk factors correlated with higher HMAAT values more than the others (Fig 5). Speech discrimination results were strong differentiating factors. Poor speech discrimination was found in 5 patients (8.2%) (3 with sensory aphasia) and they had significantly statistically worse HMAAT results in all of the azimuths apart from 225
.

AUDITORY SPATIAL DEFICITS IN STROKE

7

Table 2. Correlations between the HMAAT results and the localization of ischemic lesions Localization of ischemic lesions

FL

TL

Normal result (%) Unilateral incorrect HMAAT result On the side of the ischemic lesion (%) Opposite to the side of the ischemic lesion (%) In case of bilateral ischemic lesions (%) Bilateral incorrect HMAAT result In case of the unilateral ischemic lesion (%) In case of bilateral ischemic lesions (%) All P value

3 (16.7)

2 (11.1)

1 (5.0) 2 (11.8)

PL

OL

M

B/C

All

6 (33.3) 2 (11.1)

2 (11.1)

3 (16.7)

18 (100.0)

5 (25.0) 5 (29.4)

7 (35.0) 2 (10.0) 4 (23.5) 1 (5.9)

2 (10.0) 2 (11.8)

3 (15.0) 3 (17.6)

20 (100.0) 17 (100.0)

1 (11.1)

3 (33.3)

4 (44.4) 0 (.0)

1 (11.1)

0 (.0)

9 (100.0)

3 (13.6)

9 (40.9)

3 (13.6) 2 (9.1)

4 (18.2)

1 (4.5)

22 (100.0)

6 (33.3) 2 (11.1) 16 (15.4) 26 (25.0) .018y .040y

7 (38.9) 0 (.0) 31 (29.8) 7 (6.7) .455 .886

1 (5.6) 2 (11.1) 12 (11.5) 12 (11.5) .883 .886

18 (100.0) 104 (100.0)* .854

Abbreviations: ANOVA, analysis of variance; B/C, brainstem/cerebellum; FL, frontal lobe; HMAAT, horizontal minimum audible angle test; M, midbrain; OL, occipital lobe; PL, parietal lobe; TL, temporal lobe. *The global number of lesions was 104 due to the fact that many of them could be assigned to more than one anatomic localization. yP value is for comparison between subgroups of HMAAT results for lobar localizations; statistical significance for P , .05; ANOVA rank Kruskal–Wallis test.

Normal bilateral hearing was found in 27 (44.3%) IS patients, unilateral hearing loss in 13 (21.3%), and bilateral symmetrical hearing loss in 21 (34.4%). Sound localization ability was worse in patients with symmetrical bilateral hearing loss in comparison with other IS patients and the controls; however, the differences were statistically significant only for the 45
azimuth. (Table 1, Fig 4). Figure 5 presents HMAAT results and MRI findings in a 56-year-old patient with bilateral symmetric mild sensorineural hearing loss and poor speech discrimination in the seventh day after the incident of stroke.

Discussion Various methods and testing signals have been used in studies on sound localization ability in patients with cerebral injury. There are 2 main strategies, which are most commonly used. The first one is based on sound localization in the free acoustic field, where signals are emitted through loudspeakers (that is HMAAT),9,15,19,31,39 the second one is based on the interaural time difference and interaural intensity difference, where signals are emitted through headphones.12,15,40,41 Results on sound localization ability has been different in the same patient tested with these 2 methods with higher sensitivity of HMAAT.15 Thus, in the presented study, we used the HMAAT method; however, with one mobile loudspeaker instead of a few stationary ones that are more commonly used.19,31 Level of acoustic stimulation differs strongly between various studies and ranges from 54 dB SPL18 to 85-100 dB SPL.15 We used highlevel acoustic stimulation of 85 dB SPL to prevent differences between patient with the sensorineural hearing loss and those with normal hearing.

Furthermore, we used broadband noise to prevent a problem of standing waves that appear during tonal stimuli and impede results. In sound localization studies, some of the researchers have used tonal stimuli of 1000 Hz12,19,42 that activate only narrow part of the basilar membrane of the organ of Corti. It seems more accurate to use broadband signals in the free field sound localization tests because such signals activate broad range of the cochlear nerve endings and vast parts of the auditory cortex. Another methodological problem of the sound localization testing in patients with vascular cerebral damage is the spatial neglect.18,43 There are 2 factors that cause improper sound localization. One of them is the lowered discrimination ability and the other is the spatial neglect. To trace down the patients only with auditory discrimination deficit, we excluded patients who presented spatial neglect. An important factor that could influence our results was the short period of time (up to 30 days) from the incidence of stroke during which we performed the sound localization test. In most of the studies, this test was conducted much later–months to years from stroke, when reparative and scarring processes could have already caused reprogramming of the auditory neuronal network9,40,44 Klingon et al45 have suspected that sound localization is a specific acquired function, and thus, this ability could be recovered because of healing of ischemic lesions. We tried to test our patients as early as possible (depending on the neurologic and general state of the patients), and this may be the reason for high percentage (over 80%) of incorrect results in this study. Hearing loss reduces spatial auditory stimulation. However, although the conductive hearing loss always leads to disturbance of sound localization ability, the

8


T. PRZEWOZNY ET AL.

Figure 3. Differences in HMAAT results depending on the (A) quadrants of hemispheres affected by the ischemic lesion(s); (B) depth of the lesions; (C) stroke etiology; and (D) neurologic status. Abbreviations: HMAAT, horizontal minimum audible angle test; NIHSS, National Institutes of Health Stroke Scale.

disturbance is not always present in patients with the sensorineural hearing loss.15 Results of the studies on the influence of the sensorineural hearing loss on sound localization are not unanimous. Sanchez-Longo and Forster9 recommended that subjects with difference greater than 20 dB between the 2 ears should not be tested. Other authors sustained this opinion.10,40 However, some researchers have not taken hearing loss under consideration although they performed their studies on elderly patients with vascular brain injury.30,46 We have decided to exclude patients with bilateral asymmetrical hearing loss from our study. And the influence of the symmetrical sensorineural hearing loss on sound localization was small in our study. The worst sound localization ability was present in patients with bilateral hearing loss in comparison with normal hearing patients and those with unilateral hearing loss and the

control group in the 90
and 270
azimuth; however, the differences were statistically significant only for the 45
azimuth (the right-sided azimuth). If we assume that sound localization disturbance is typically contralateral to the damaged hemisphere,4 this finding could suggest the prevalence of the sound localization centers of the left hemisphere in patients with hearing loss. However, in the whole study group, with nearly half of patients presenting normal hearing, worse results were found in cases of the right hemisphere damage, and for the 90
azimuth– a right-sided azimuth, ipsilateral to the lesion. This finding could indicate that both hemispheres interact in the sound localization process,41 and that vascular damage can cause the auditory pathways to reorganize themselves in a more complex manner. Some scientists exclude patients with presbycusis from their studies,19,23 others accept patients with this

AUDITORY SPATIAL DEFICITS IN STROKE

9

Figure 4. Differences in the HMAAT results depending on the presence of (A) hypertension; (B) vascular risk factors; (C) speech discrimination disorder (poor discrimination was diagnosed when a patient recognized less than 70% of words in the free field with 65 dB SPL level, S/N ratio 10 dB); and (D) different types of hearing loss. Abbreviations: HMAAT, horizontal minimum audible angle test; S/N, speech-to-noise ratio; SPL, sound pressure level.

type of hearing loss.9,40,43 The exclusion of patients with presbycusis causes younger people to be included, and obviously, these have better sound localization ability.19 It has been reported that subjects with presbycusis, which is the high frequency bilateral sensorineural hearing loss, were significantly less accurate in sound localization in the median sagittal plane from the normal hearing subjects.47 In our study, the tested and the control group had wide age range (21-80 years) and in both, the average age was above 50 years. Thus, both groups included subjects with presbycusis, however, with a small advantage of IS group that could increase the number of subjects with sound localization problems. Our study confirmed influence of the disturbed speech discrimination on HMAAT results although only 8.2% of

the patients had this deficit. H€ ausler et al15 reported similar results of their study. The ability of understanding speech is disturbed in aphasia, but it depends also on the hearing threshold of human speech (.5-2 kHz) and age.48 There were 3 subjects with aphasia among our patients with speech discrimination score below 70%. It is not clear whether sound localization is disturbed in such cases because of damage of the posterior part of the upper temporal gyrus where speech analysis takes place or also of other parts of the brain because sound localizations seem to activate both hemispheres.41 The research on the influence of the low speech discrimination ability on sound localization is difficult because of the lack of accurate verbal contact with the patients. Nonverbal tests, which are yet unavailable, would be necessary to conduct such research properly.

10


T. PRZEWOZNY ET AL. Figure 5. The HMAAT results and MRI findings in a 56-year-old patient, with bilateral symmetric mild sensorineural hearing loss and poor speech discrimination in the seventh day after the incident of stroke. (A) The arrows indicate multiple, bilateral, and lacunar ischemic lesions. (B) Black line–95th percentile values for the age-matched subgroup of controls, red dotted line–results of the patient; (bilateral incorrect result for azimuths 45
, [40
]; 90
, [30
]; 180
, [21
]; 225
, [16
]; and 270
, [17
]. Abbreviations: HMAAT, horizontal minimum audible angle test; MRI, magnetic resonance imaging; L, left; R, right. (Color version of figure is available online.)

Many studies on the influence of cerebral injury on the sound localization ability have focused only on the unilateral localization disturbances, contralateral to the injury.9,18,21,45 There are only few reports on bilateral disturbances caused by unilateral injury,43,49 and bilateral disturbances, due to bilateral injury have been reported in single cases found among larger case series.9 In our study, we found the worst HMAAT results in patients with bilateral ischemic lesions and the bilateral sound localization disturbance, similarly as other authors.50 We found only slightly better results in the patients with unilateral lesion and bilateral localization disturbance and the patients with bilateral lesions and unilateral localization disturbance. In the study by Pinek et al,18 patients with the left posterior brain injury presented profound disturbance in the horizontal and vertical planes of both of the hearing hemispaces, particularly in the anterofrontal region. Interestingly, the average result for the right-sided azimuths in our patients with contralateral lesions was similar to the average result in the control group. We found disturbances mainly in the left-sided azimuths. This suggests the dominant role of the right-sided lesions over the left-sided ones in the disruption of proper sound localization, which is in concordance with reports of other authors.19,41,43,49 Lacunar infarcts are associated with pathology of deep perforating arterioles and are rarely fatal.51 These small and often multiple lesions affect the white matter, deep gray matter of the hemispheres, and the brainstem and present with several, often discrete, neurologic syndromes.52 However, diffusion MRI tractography studies reported diffused (up to 2 cm) abnormalities remote from the lacune present in the affected tract.53 Thus, although lacunar lesions are small, if they are multiple, they can cause substantial disruption of pathways responsible for cognitive functions with the sound localization ability among them. In our study, lacunar lesions correlated more than other stroke types with the deterioration of sound localization, with significant differences

for the contralateral 45
and 225
azimuths. Thus, sound localization disturbances due to multiple, bilateral lacunar lesions could be included in the vast definition of the vascular cognitive impairment.54 It is worth noticing that neuroimaging studies performed on admission in many of the lacunar stroke patient present a preexisting load of lacunar lesions. These lesions often remain clinically silent. It is the index stroke that either leads to a new sound localization abnormality or worsening of a pre-existing abnormality. Further studies are required to find whether sound localization abnormalities are present in patients with ‘‘silent’’ lacunar lesion that are silent only because sound localization impairment is difficult to test at the bedside. Some authors recognize sound localization disturbances contralateral to the temporal lobe lesions as typical, and findings of our study seem to confirm this hypothesis.9,15,21 We found this kind of result in 1 of 5 of our patients, who had lesions mainly in the temporal and the parietal lobe. Similarly, it was the temporal and the parietal lobes that were most commonly affected in cases of bilateral incorrect results in patients with unilateral lesions. From these 2 lobar localizations, it was the temporal one for which the differences between the HMAAT results were statistically significant, which may indicate its dominant role in the sound localization process. The importance of these 2 anatomic localizations has been proved in fMRI studies and electrical neuroimaging analyses on healthy subjects,27-29,55 and studies on patients with brain injury.56 Spierer et al41 used stereophonic tests to assess sound localization ability in patients with brain injury (14 of the 25 patients had right-sided ischemic lesions and 9 left-sided ones, mainly in the temporal and parietal lobes). Analysis of the anatomic localization of the lesions and neuropsychological deficits suggest the presence of a binaural sound localization system primarily supported by the right hemisphere, and the right temporal lobe seemed to play the paramount role in the sound

AUDITORY SPATIAL DEFICITS IN STROKE

11

41

localization process. This could be additionally supported by observations on ipsilateral temporal lobe injury of the dominant hemisphere (that is the left hemisphere in right-handed patients), which did not cause sound localization disturbance.57 Another factor influencing the ability to localize sounds properly could be the individual variability of neural pathways. It was proved in magnetoencephalographic studies that substantial difference in precision of sound localization could be because of the different brain geometry in individual subjects.58 Disturbance of sound localization ability in case of lesions localized in other than the temporal and the parietal lobes although difficult to explain, was showed in cases of injuries of the frontal lobe and the thalamus.59 The explanation given was that the injured lobe pressed the temporal lobe or that the disturbance was because of blockage of the cerebrospinal fluid flow.9 fMRI studies on stroke patients suggest also a possible role of the inferior frontal and precentral and postcentral areas of the left hemisphere.60 In our study, in case of bilateral localization disturbance and bilateral lesions, higher percentage of patients in comparison with the other types of the HMAAT results, had frontal lobe lesions. It seems that the role of the frontal lobe in the sound localization process becomes apparent only in cases of bilateral brain injury.25-26

Conclusions In the presented study, we analyzed various anatomic and clinical factors of the sound localization ability in IS patients. It seems that ischemic brain injury in the early stages after stroke can cause disturbance of sound localization of one or both hemispaces in more than 80% of the patients. It was the multiple, bilateral, and lacunar character of the ischemic lesions that correlated strongly with the disturbance of the sound localization ability.

References 1. Doyle KJ, Fowler C, Starr A. Audiologic findings in unilateral deafness resulting from contralateral pontine infarct. Otolaryngol Head Neck Surg 1996;114:482-486. 2. Earnest MP, Monero PA, Parnell PR. Cortical deafness. Demonstration of the pathologic anatomy by CT scan. Neurology 1977;27:1172-1175. 3. Hariri MA, Lakshmi MV, Larner S, et al. Auditory problems in elderly patients with stroke. Age Ageing 1994; 23:312-316. 4. H€ausler R, Levine RA. Auditory dysfunction in stroke. Acta Otolaryngol 2000;120:689-703. 5. Lee H. Sudden deafness related to posterior circulation infarction in the territory of the nonanterior inferior cerebellar artery: frequency, origin, and vascular topographical pattern. Eur Neurol 2008;59:302-306. 6. Lee H. Recent advances in acute hearing loss due to posterior circulation ischemic stroke. J Neurol Sci 2014; 338:23-29. 7. Martines F, Dispenza F, Gagliardo C, et al. Sudden sensorineural hearing loss as prodromal symptom of anterior

8.

9.

10.

11.

12.

13.

14.

15.

16. 17.

18.

19.

20.

21. 22. 23.

24.

25.

26.

27.

inferior cerebellar artery infarction. ORL J Otorhinolaryngol Relat Spec 2011;73:137-140. Musiek FE, Charette L, Morse D, et al. Central deafness associated with a midbrain lesion. J Am Acad Audiol 2004;15:133-151. Sanchez-Longo LP, Forster FM. Clinical significance of impairment of sound localization. Neurology 1958; 8:119-125. Sonoda S, Mori M, Goishi A. Pattern of localisation error in patients with stroke to sound processed by a binaural sound space processor. J Neurol Neurosurg Psychiatry 2001;70:43-49. Tabuchi S. Auditory dysfunction in patients with cerebrovascular disease. ScientificWorldJournal 2014; 2014:261824. Bisiach E, Cornacchia L, Sterzi R, et al. Disorders of perceived auditory lateralization after lesions of the right hemisphere. Brain 1984;107:37-52. Efron R, Crandall PH, Koss B, et al. Central auditory processing. III. The ‘cocktail party’ effect and anterior temporal lobectomy. Brain Lang 1983;19:245-263. Engelien A, Silbersweig D, Stern E, et al. The functional anatomy of recovery from auditory agnosia. A PET study of sound categorization in a neurological patient and normal controls. Brain 1995;118:1395-1409. H€ ausler R, Colburn S, Marr E. Sound localization in subjects with impaired hearing. Spatial-discrimination and interaural-discrimination tests. Acta Otolaryngol Suppl 1983;400:1-62. Klingon GH, Bontecou DC. Localization in auditory space. Neurology 1966;16:879-886. Poirier P, Lassonde M, Villemure JG, et al. Sound localization in hemispherectomized patients. Neuropsychologia 1994;32:541-553. Pinek B, Duhamel JR, Cav
e C, et al. Audio-spatial deficits in humans: differential effects associated with left versus right hemisphere parietal damage. Cortex 1989;25: 175-186. Ruff RM, Hersh NA, Pribram KH. Auditory spatial deficits in the personal and extrapersonal frames of reference due to cortical lesions. Neuropsychologia 1981;19: 435-443. Walsh EG. An investigation of sound localization in patients with neurological abnormalities. Brain 1957; 80:222-250. Wortis SB, Pfeffer AZ. Unilateral auditory-spatial agnosia. J Nerv Ment Dis 1948;108:181-186. Moore JK, Osen KK. Anatomy and physiology of binaural hearing. Audiology 1991;30:125-134. Furst M, Aharonson V, Levine RA, et al. Sound lateralization and interaural discrimination. Effects of brainstem infarcts and multiple sclerosis lesions. Hear Res 2000; 143:29-42. Maeder PP, Meuli RA, Adriani M, et al. Distinct pathways involved in sound recognition and localization: a human fMRI study. Neuroimage 2001;14:802-816. Owen AM, Evans AC, Petrides M. Evidence for a twostage model of spatial working memory processing within the lateral frontal cortex: a positron emission tomography study. Cereb Cortex 1996;6:31-38. Owen AM, Stern CE, Look RB, et al. Functional organization of spatial and non spatial working memory processing within the human lateral frontal cortex. Proc Natl Acad Sci U S A 1998;95:7721-7726. Warren JD, Zielinski BA, Green GG, et al. Perception of sound-source motion by the human brain. Neuron 2002; 34:139-148.

12 28. Warren JE, Wise RJ, Warren JD. Sounds do-able: auditorymotor transformations and the posterior temporal plane. Trends Neurosci 2005;28:636-643. 29. Karnath HO, Ferber S, Himmelbach M. Spatial awareness is a function of the temporal not the posterior parietal lobe. Nature 2001;411:950-953. 30. Altman J, Balanov L, Delgin V. Effects of unilateral disorder of the brain hemispheric functions in man on directional hearing. Neuropsychologia 1979;17:295-301. 31. Zatorre RJ, Penhune VB. Spatial localization after excision of human auditory cortex. J Neurosci 2001;21: 6321-6328. 32. Mesulam M. Functional anatomy of attention and neglect: from neurons to network. In: Karnath HO, Milner AD, Vallar G, eds. The cognitive and neural bases of spatial neglect. New York: Oxford University Press 2002:33-45. 33. Schenkenberg T, Bradford DC, Ajax ET. Line bisection and unilateral visual neglect in patients with neurologic impairment. Neurology 1980;30:509-517. 34. Adams HP Jr, Davies PH, Leira EC, et al. Baseline NIH Stroke Scale score strongly predicts outcome after stroke. A report of Trial of ORG 10172 in Acute Stroke Treatment (TOAST). Neurology 1999;53:126-131. 35. Weir J, Abrahams PH, Belli AM, et al. An imaging atlas of human anatomy. Wroc1aw: Urban & Partner 2005 (in Polish). 36. Peng S.C. Hearing aids: the basic information you need to know. FDA basics webinar. May 23, 2012. Available at: http://www.fda.gov/downloads/aboutfda/transparency/ basics/ucm305102.pdf. Accessed May 23, 2012. 37. Pruszewicz A, Demenko G, Richter L, et al. New articulation lists for speech audiometry. Part II. Otolaryngol Pol 1994;48:56-62 (in Polish). 38. Mills AW. On the minimum audible angle. J Acoust Soc Am 1958;30:237-246. 39. Sanchez Longo LP, Forster FM, Auth TL. A clinical test for sound localization and its applications. Neurology 1957;7:655-663. 40. Clarke S, Bellmann A, Meuli RA, et al. Auditory agnosia and auditory spatial deficits following left hemispheric lesions: evidence for distinct processing pathways. Neuropsychologia 2000;38:797-807. 41. Spierer L, Bellmann-Thiran A, Maeder P, et al. Hemispheric competence for auditory spatial representation. Brain 2009;132:1953-1966. 42. Zakrzewski A. Clinical test for the acuity of directional hearing. Bulletin de la Societe des Amis des Science et des Letters de Pozna
n, Serie C 1960;10:7-14 [Article in English]. 43. Bisiach E, Luzzatti C, Perani D. Unilateral neglect, representational schema and consciousness. Brain 1979; 102:609-618.


T. PRZEWOZNY ET AL. 44. Bellmann A, Meuli R, Clarke S. Two types of auditory neglect. Brain 2001;124:676-687. 45. Klingon GH, Bontecou DC. Auditory field localization. Trans Am Neurol Assoc 1964;89:210-211. 46. Ferber S, Karnath HO. Parietal and occipital lobe contributions to perception of straight ahead orientation. J Neurol Neurosurg Psychiatry 1999;67:572-578. 47. Rakerd B, Vander Velde TJ, Hartmann WM. Sound localization in the median sagittal plane by listeners with presbycusis. J Am Acad Audiol 1998;9:466-479. 48. Maukari SZ, Wahata NH, Mazian R. Effects of ageing and hearing thresholds on speech perception in quiet and in noise perceived in different locations. Korean J Audiol 2014;18:112-118. 49. Schankweiler DP. Performance of brain-damaged patients of two tests of sound localization. J Comp Physiol Psychol 1961;54:375-381. 50. Yamada K, Kaga K, Uno A, et al. Comparison of interaural time and intensity difference discrimination in patients with temporal lobe lesions. Acta Otolaryngol Suppl 1997;532:135-137. 51. Bailey EL, Smith C, Sudlow CL, et al. Pathology of lacunar ischemic stroke in humans: a systematic review. Brain Pathol 2012;22:583-591. 52. Ghandehari K, Izadi ZM. Clinical evaluation of 625 lacunar syndrome patients. Turk J Med Sci 2009; 39:607-612. 53. Reijmer YD, Freeze WM, Leemans A, et al, Utrecht Vascular Cognitive Impairment Study Group. The effect of lacunar infarcts on white matter tract integrity. Stroke 2013;44:2019-2021. 54. O’Brien JT, Erkinjuntti T, Reisberg B, et al. Vascular cognitive impairment. Lancet Neurol 2003;2:89-98. 55. Spierer L, Murray MM, Tardif E, et al. The path to success in auditory spatial discrimination: electrical neuroimaging responses within the supratemporal plane predict performance outcome. Neuroimage 2008;41:493-503. 56. Karnath HO. Spatial orientation and the representation of space with parietal lobe lesions. Philos Trans R Soc Lond B Biol Sci 1997;352:1411-1419. 57. Gazzaniga MS, Glass AV, Sarno MT, et al. Pure word deafness and hemispheric dynamics: a case history. Cortex 1973;9:136-143. 58. Larson E, Maddox RK, Lee AK. Improving spatial localization in MEG inverse imaging by leveraging intersubject anatomical differences. Front Neurosci 2014; 8:330. 59. Hugdahl K, Wester K, Asbjørnsen A. Auditory neglect after right frontal lobe and right pulvinar thalamic lesions. Brain Lang 1991;41:465-473. 60. Z€ undorf IC, Karnath HO, Lewald J. The effect of brain lesions on sound localization in complex acoustic environments. Brain 2014;137:1410-1418.