Abstract
Purpose
The aims of this study were to follow recovery during the first 6 months after acute unilateral vestibular loss (AUVL) and to determine predictors for self-rated remaining symptoms.
Materials and methods
Forty-two subjects were included less than 10 days after AUVL. Static and dynamic clinical balance tests, visual analogue scales, University of California Los Angeles Dizziness Questionnaire, Dizziness Beliefs Scale, European Quality of Life questionnaire, Dizziness Handicap Inventory, and Hospital Anxiety and Depression Scale were performed at inclusion and at 7 follow-ups over 6 months. Subjects rated their symptoms on visual analogue scales daily at home. Videonystagmography was performed in the acute stage and after 10 weeks.
Results
Decrease of symptoms and improvement of balance function were larger during the first compared with the latter part of the follow-up period. Visual analogue scale ratings for balance problems were higher than those for dizziness. A prediction model was created based on the results of 4 tests in the acute stage: standing on foam with eyes closed, standing on 1 leg with eyes open, visual analogue scale rating of vertigo at rest, and European Quality of Life questionnaire rating of health-related quality of life. The prediction model identified subjects at risk of having remaining symptoms after 6 months with a sensitivity of 86% and a specificity of 79%.
Conclusions
Recovery mainly takes place during the first weeks after AUVL. Subjects rate more balance problems than dizziness. Self-rated remaining symptoms after 6 months may be predicted by clinical balance tests and subjective ratings in the acute stage.
1
Introduction
Acute unilateral vestibular loss (AUVL) causes a sudden onset of vertigo, spontaneous nystagmus, nausea, vomiting, disequilibrium, and impaired balance. Vestibular function can recover by complete or, most frequently, partial return of the peripheral vestibular function itself. However, recovery after loss of vestibular function is also a result of vestibular compensation processes including multiple mechanisms of both neural and behavioral plasticity . There is moderate to strong evidence that vestibular rehabilitation therapy is an effective management for unilateral peripheral dysfunction .
The acute stage of severe vertigo after AUVL lasts for only a few days, but 30% to 50% of patients complain of long-lasting symptoms, for example, dizziness or disequilibrium provoked by head movements, stress, or darkness, when asked after 1 to 8 years . In this study, the term dizziness is used to describe all types of illusions of movements, including vertigo, because this is in agreement with the Swedish language terminology in the measures used.
The purposes of this study were to follow the recovery during the first 6 months after AUVL and to determine predictors for self-rated remaining symptoms.
2
Material and methods
2.1
Subjects and study design
Forty-nine consecutive subjects diagnosed as having AUVL at the ear, nose, and throat departments in 2 Swedish hospitals fulfilled the criteria for the study and were asked to participate. Inclusion criteria were (1) sudden onset of vertigo and nausea less than 10 days ago and (2) absent or greater than 20% reduced vestibular caloric response in 1 ear, as confirmed using the videonystagmography testing. Besides caloric tests, videonystagmography testing included spontaneous nystagmus as a sign of the vestibular loss and also ocular pursuit, saccades, positional testing, and visual suppression of the caloric responses to unveil possible central neurologic disturbance. Exclusion criteria for the study were (1) central neurologic symptoms, (2) associated auditory symptoms, or (3) symptoms of other vertigo disease.
Forty-two subjects (23 men, 19 women; mean age, 54 years; age range, 30–73 years) accepted to participate. Subjects were diagnosed within 8 days (average, 3 days; range, 0–8 days) after symptom onset. They were enrolled in the study and initially examined by a physiotherapist within 9 days (average, 4 days; range, 1–9 days) after symptom onset. The subjects stayed in the hospital for an average of 2 days (range, 0–6 days). In the acute stage, some subjects had vestibular suppressive medication for 1 or 2 days. Of the 42 subjects, 7 were initially treated with prednisone (high dose for 5 days and tapering off for 3 days). At inclusion, subjects received instructions from a physiotherapist of a home-based vestibular rehabilitation therapy exercise program, which are to be performed twice a day. All subjects were asked to register the number of exercise sessions and the degree of symptoms in a diary at home each day during the first 4 weeks after inclusion in the study, and each week from the 5th to the 10th week. They also returned for follow-up evaluations after 1, 2, 3, 4, 7, 10 weeks, and 6 months. Videonystagmography testing was performed in the acute stage and at the 10-week follow-up ( Fig. 1 ).
Seven subjects rejected participation because of long distance to travel (n = 4), going on holiday for a long period (n = 2), or lack of time (n = 1). They did not differ from the included subjects regarding age, sex, or asymmetry of vestibular caloric response in the acute stage.
Of the 42 included subjects, 2 did not want to continue their participation after the initial examination or 1-week follow-up because of other disease not related to AUVL (n = 1) and lack of time (n = 1), respectively. The remaining 40 subjects completed the 6-month follow-up.
2.2
Measured variables
At inclusion, at home, and at the follow-ups, different instruments were used to measure balance function, self-rated degree of and consequences of dizziness and balance problems, and vestibular caloric response ( Fig. 1 ).
2.2.1
Static clinical balance tests
The length of time the subjects managed to stand with their eyes open or closed was measured without shoes in the following positions: the Romberg test with the feet together, the sharpened Romberg test with the nondominant foot on a line right in front of the dominant foot , normal standing on a foam cushion, and standing on 1 leg, both right and left. The tasks were timed using a digital stopwatch until the subjects moved their feet from the given position, touched the wall, opened their eyes on the eyes-closed tests, or reached the maximum time of 30 seconds. This performance of the tests has previously been described in detail and evaluated regarding test-retest reliability (intraclass correlation coefficient [ICC], 0.56–0.92) and intertester reliability (ICC, 0.98–1.00) in subjects with remaining symptoms after AUVL .
2.2.2
Dynamic clinical balance tests
Subjects performed a tandem walk of 15 steps both forward and backward without shoes at a self-chosen speed on a line 1.5 cm wide. The number of incorrect steps was counted. A correct step was defined as a step touching the line and with heel and toe not visibly separated .
When walking in a figure of 8 , the subjects walked twice without shoes in a figure of 2 circles with double lines (inner diameter of 1.5 m and outer diameter of 1.8 m) put together to form an “8.” The space between the lines was 15 cm. Subjects walked with a self-chosen step length and at a speed of 1 step per second given by a metronome. Steps outside or touching the line were counted as incorrect.
We have described and evaluated these performances of the dynamic clinical balance tests regarding test-retest reliability (ICC, 0.90–0.96) and intertester reliability (ICC, 0.96–0.99) in subjects with remaining symptoms after AUVL .
2.2.3
Visual analogue scales
Subjects were asked to rate their degree of symptoms on visual analog scale (VAS) both in the diary at home (each evening during the first 4 weeks after inclusion in the study and at the end of each week from the fifth to the tenth week) and at the follow-up evaluations. Three scales ranging from no symptoms (0 mm) to worst possible symptoms (100 mm) were used to rate the levels of perceived dizziness during movement, perceived dizziness at rest, and balance problems. The scales have been evaluated regarding test-retest reliability (ICC 0.93–0.96) in subjects with remaining symptoms after AUVL .
2.2.4
The University of California Los Angeles Dizziness Questionnaire
The University of California Los Angeles Dizziness Questionnaire (UCLA-DQ) asks subjects to characterize dizziness with regard to frequency, intensity, impact on daily activities, impact on quality of life, and fear of dizziness. The subjects chose 1 of 5 response alternatives for each item, giving 1 (least severe) to 5 (most severe) points each, and a total score ranging from 5 to 25. If the subject does not have dizziness at all, a 0 point is given. We have translated the scale into Swedish following the method of back translation and evaluated test-retest reliability (ICC, 0.89) in subjects with remaining symptoms after AUVL .
2.2.5
Dizziness Beliefs Scale
The Dizziness Beliefs Scale measures agreement, on a 5-point scale ranging from strongly agree to strongly disagree, with 17 beliefs about potential negative consequences of dizziness. The Dizziness Beliefs Scale has been used and found to predict handicap for patients with dizziness and disequilibrium . We have translated the scale into Swedish following the method of back translation and evaluated regarding test-retest reliability (ICC, 0.90) in subjects with remaining symptoms after AUVL .
2.3
European Quality of Life questionnaire EQ-5D
The European Quality of Life questionnaire EQ-5D (EuroQol) is a widely used, reliable, and valid measure of health-related quality of life. The first part of the instrument is a self-reported description of the subject’s health state using a 5-dimensional classification. It contains 5 items, each with 3 response choices. The answers are converted into a score ranging up to 1.00, indicating high health-related quality of life. The second part is a self-rated valuation of the subject’s health state using a vertical VAS in the form of a thermometer ranging from 0 (worst imaginable state of health) at the bottom to 100 (best imaginable state of health) at the top. The instrument has been evaluated regarding test-retest reliability (ICC, 0.86–0.97) in subjects with remaining symptoms after AUVL .
2.3.1
Dizziness Handicap Inventory
The Dizziness Handicap Inventory (DHI) is a widely used 25-item scale designed to assess self-perceived dizziness-associated disability and handicap. A “yes” response gives 4 points; a “sometimes,” 2 points; and a “no,” 0 point. The total score ranges from 0, suggesting no handicap, to 100, indicating a severe level of self-perceived handicap. The questionnaire has been evaluated regarding test-retest reliability (ICC, 0.95) in subjects with remaining symptoms after AUVL .
2.3.2
Hospital Anxiety and Depression Scale
The Hospital Anxiety and Depression Scale (HADS) assesses nonsomatic symptoms of anxiety and depression. It contains 14 items, each with 4 response choices giving 0 to 3 points. Scores on the 2 subscales for anxiety and depression range from 0 (no sign of anxiety or depression) to 21 (maximum level of anxiety or depression). Scores between 8 and 10 indicate possible, and scores greater than 10 probable, anxiety and depression . The scales have been evaluated regarding test-retest reliability (ICC, 0.86) in subjects with remaining symptoms after AUVL .
2.3.3
Videonystagmography
Videonystagmography is a test of the vestibulooculomotor system, including the function of the semicircular canal system of the vestibular organ. In caloric testing, irrigation of the external auditory canal with 30°C and 44°C water is used to determine the excitability of the individual horizontal canals and thus whether and to what extent they are functioning . In this study, values of greater than 20% asymmetry between the right and the left labyrinth were considered pathological.
2.4
Vestibular rehabilitation therapy
All subjects received practical and written instructions from a physiotherapist of a home-based vestibular rehabilitation therapy exercise program, with 7 exercises to be performed twice a day . The exercise instructions were quick horizontal and vertical eye movements, fixating a stationary target while moving the head, standing with feet together and eyes closed, walking on a line, and moving the head to look to the right and left and up and down while walking. All subjects were also encouraged to take walks outdoors. The subjects were instructed to, when possible, increase the difficulty of the exercises by increasing the velocity of head and body movements, increasing the duration of the exercises, narrowing the base of support, standing on a soft surface, and combining walking on a line with head movements. The subjects were instructed to continue with each exercise as long as it was difficult to perform or provoked symptoms. They were asked to register the number of exercise sessions (none/1/2 or more) in the diary each evening during the first 4 weeks after inclusion in the study and at the end of each week from the 5th to the 10th week. According to the diaries, the subjects performed exercises 89 times on average (SD, 40; range, 21–140) during the 70 days the diary lasted.
2.5
Statistical analyses
Three trials were performed for each of the static and dynamic balance tests. If a maximum value (30 seconds or no incorrect steps, respectively) was reached in the first or second trial, no additional trials were performed, and the maximum value was assigned to each remaining trial as in previous studies . The mean value of 3 trials was used for analysis.
Medians and interquartile range are presented because most variables were not normally distributed. To analyze change over time of variables on VAS, we used a rank-invariant nonparametric method developed by Svensson . For all other variables, Fisher exact test was used for comparisons between groups for categorical variables and Mann-Whitney U test for quantitative variables. For comparisons within groups, Wilcoxon signed rank test was used. P < .05 was considered significant. These analyses were computed using Statistica 8 computer software (StatSoft, Inc, Tulsa, OK).
As a dichotomous dependent variable in the predictor analysis, substantial self-rated remaining symptoms reported at 6 months and no self-rated remaining symptoms at 6 months were chosen. The 2 groups were defined according to the first 2 questions of the UCLA-DQ as in a previous study . Subjects in the group with substantial remaining symptoms reported at least occasional (at least 2 points) and at least mild (at least 2 points) symptoms according to the UCLA-DQ. Subjects in the group with no remaining symptoms reported no symptoms at all according to the UCLA-DQ. Those 2 groups were considered to be truly different regarding remaining symptoms. Subjects with very mild remaining symptoms (rarely and/or very mild symptoms according to UCLA-DQ) were not included in this part of the analysis.
2.6
Ethics
The Regional Ethics Committee for Human Research of the Faculty of Health Sciences, Linköping University approved the study (approval 02-470, 021217), and informed consent from all participants was obtained after the nature of the study was explained.
2
Material and methods
2.1
Subjects and study design
Forty-nine consecutive subjects diagnosed as having AUVL at the ear, nose, and throat departments in 2 Swedish hospitals fulfilled the criteria for the study and were asked to participate. Inclusion criteria were (1) sudden onset of vertigo and nausea less than 10 days ago and (2) absent or greater than 20% reduced vestibular caloric response in 1 ear, as confirmed using the videonystagmography testing. Besides caloric tests, videonystagmography testing included spontaneous nystagmus as a sign of the vestibular loss and also ocular pursuit, saccades, positional testing, and visual suppression of the caloric responses to unveil possible central neurologic disturbance. Exclusion criteria for the study were (1) central neurologic symptoms, (2) associated auditory symptoms, or (3) symptoms of other vertigo disease.
Forty-two subjects (23 men, 19 women; mean age, 54 years; age range, 30–73 years) accepted to participate. Subjects were diagnosed within 8 days (average, 3 days; range, 0–8 days) after symptom onset. They were enrolled in the study and initially examined by a physiotherapist within 9 days (average, 4 days; range, 1–9 days) after symptom onset. The subjects stayed in the hospital for an average of 2 days (range, 0–6 days). In the acute stage, some subjects had vestibular suppressive medication for 1 or 2 days. Of the 42 subjects, 7 were initially treated with prednisone (high dose for 5 days and tapering off for 3 days). At inclusion, subjects received instructions from a physiotherapist of a home-based vestibular rehabilitation therapy exercise program, which are to be performed twice a day. All subjects were asked to register the number of exercise sessions and the degree of symptoms in a diary at home each day during the first 4 weeks after inclusion in the study, and each week from the 5th to the 10th week. They also returned for follow-up evaluations after 1, 2, 3, 4, 7, 10 weeks, and 6 months. Videonystagmography testing was performed in the acute stage and at the 10-week follow-up ( Fig. 1 ).
Seven subjects rejected participation because of long distance to travel (n = 4), going on holiday for a long period (n = 2), or lack of time (n = 1). They did not differ from the included subjects regarding age, sex, or asymmetry of vestibular caloric response in the acute stage.
Of the 42 included subjects, 2 did not want to continue their participation after the initial examination or 1-week follow-up because of other disease not related to AUVL (n = 1) and lack of time (n = 1), respectively. The remaining 40 subjects completed the 6-month follow-up.
2.2
Measured variables
At inclusion, at home, and at the follow-ups, different instruments were used to measure balance function, self-rated degree of and consequences of dizziness and balance problems, and vestibular caloric response ( Fig. 1 ).
2.2.1
Static clinical balance tests
The length of time the subjects managed to stand with their eyes open or closed was measured without shoes in the following positions: the Romberg test with the feet together, the sharpened Romberg test with the nondominant foot on a line right in front of the dominant foot , normal standing on a foam cushion, and standing on 1 leg, both right and left. The tasks were timed using a digital stopwatch until the subjects moved their feet from the given position, touched the wall, opened their eyes on the eyes-closed tests, or reached the maximum time of 30 seconds. This performance of the tests has previously been described in detail and evaluated regarding test-retest reliability (intraclass correlation coefficient [ICC], 0.56–0.92) and intertester reliability (ICC, 0.98–1.00) in subjects with remaining symptoms after AUVL .
2.2.2
Dynamic clinical balance tests
Subjects performed a tandem walk of 15 steps both forward and backward without shoes at a self-chosen speed on a line 1.5 cm wide. The number of incorrect steps was counted. A correct step was defined as a step touching the line and with heel and toe not visibly separated .
When walking in a figure of 8 , the subjects walked twice without shoes in a figure of 2 circles with double lines (inner diameter of 1.5 m and outer diameter of 1.8 m) put together to form an “8.” The space between the lines was 15 cm. Subjects walked with a self-chosen step length and at a speed of 1 step per second given by a metronome. Steps outside or touching the line were counted as incorrect.
We have described and evaluated these performances of the dynamic clinical balance tests regarding test-retest reliability (ICC, 0.90–0.96) and intertester reliability (ICC, 0.96–0.99) in subjects with remaining symptoms after AUVL .
2.2.3
Visual analogue scales
Subjects were asked to rate their degree of symptoms on visual analog scale (VAS) both in the diary at home (each evening during the first 4 weeks after inclusion in the study and at the end of each week from the fifth to the tenth week) and at the follow-up evaluations. Three scales ranging from no symptoms (0 mm) to worst possible symptoms (100 mm) were used to rate the levels of perceived dizziness during movement, perceived dizziness at rest, and balance problems. The scales have been evaluated regarding test-retest reliability (ICC 0.93–0.96) in subjects with remaining symptoms after AUVL .
2.2.4
The University of California Los Angeles Dizziness Questionnaire
The University of California Los Angeles Dizziness Questionnaire (UCLA-DQ) asks subjects to characterize dizziness with regard to frequency, intensity, impact on daily activities, impact on quality of life, and fear of dizziness. The subjects chose 1 of 5 response alternatives for each item, giving 1 (least severe) to 5 (most severe) points each, and a total score ranging from 5 to 25. If the subject does not have dizziness at all, a 0 point is given. We have translated the scale into Swedish following the method of back translation and evaluated test-retest reliability (ICC, 0.89) in subjects with remaining symptoms after AUVL .
2.2.5
Dizziness Beliefs Scale
The Dizziness Beliefs Scale measures agreement, on a 5-point scale ranging from strongly agree to strongly disagree, with 17 beliefs about potential negative consequences of dizziness. The Dizziness Beliefs Scale has been used and found to predict handicap for patients with dizziness and disequilibrium . We have translated the scale into Swedish following the method of back translation and evaluated regarding test-retest reliability (ICC, 0.90) in subjects with remaining symptoms after AUVL .
2.3
European Quality of Life questionnaire EQ-5D
The European Quality of Life questionnaire EQ-5D (EuroQol) is a widely used, reliable, and valid measure of health-related quality of life. The first part of the instrument is a self-reported description of the subject’s health state using a 5-dimensional classification. It contains 5 items, each with 3 response choices. The answers are converted into a score ranging up to 1.00, indicating high health-related quality of life. The second part is a self-rated valuation of the subject’s health state using a vertical VAS in the form of a thermometer ranging from 0 (worst imaginable state of health) at the bottom to 100 (best imaginable state of health) at the top. The instrument has been evaluated regarding test-retest reliability (ICC, 0.86–0.97) in subjects with remaining symptoms after AUVL .
2.3.1
Dizziness Handicap Inventory
The Dizziness Handicap Inventory (DHI) is a widely used 25-item scale designed to assess self-perceived dizziness-associated disability and handicap. A “yes” response gives 4 points; a “sometimes,” 2 points; and a “no,” 0 point. The total score ranges from 0, suggesting no handicap, to 100, indicating a severe level of self-perceived handicap. The questionnaire has been evaluated regarding test-retest reliability (ICC, 0.95) in subjects with remaining symptoms after AUVL .
2.3.2
Hospital Anxiety and Depression Scale
The Hospital Anxiety and Depression Scale (HADS) assesses nonsomatic symptoms of anxiety and depression. It contains 14 items, each with 4 response choices giving 0 to 3 points. Scores on the 2 subscales for anxiety and depression range from 0 (no sign of anxiety or depression) to 21 (maximum level of anxiety or depression). Scores between 8 and 10 indicate possible, and scores greater than 10 probable, anxiety and depression . The scales have been evaluated regarding test-retest reliability (ICC, 0.86) in subjects with remaining symptoms after AUVL .
2.3.3
Videonystagmography
Videonystagmography is a test of the vestibulooculomotor system, including the function of the semicircular canal system of the vestibular organ. In caloric testing, irrigation of the external auditory canal with 30°C and 44°C water is used to determine the excitability of the individual horizontal canals and thus whether and to what extent they are functioning . In this study, values of greater than 20% asymmetry between the right and the left labyrinth were considered pathological.
2.4
Vestibular rehabilitation therapy
All subjects received practical and written instructions from a physiotherapist of a home-based vestibular rehabilitation therapy exercise program, with 7 exercises to be performed twice a day . The exercise instructions were quick horizontal and vertical eye movements, fixating a stationary target while moving the head, standing with feet together and eyes closed, walking on a line, and moving the head to look to the right and left and up and down while walking. All subjects were also encouraged to take walks outdoors. The subjects were instructed to, when possible, increase the difficulty of the exercises by increasing the velocity of head and body movements, increasing the duration of the exercises, narrowing the base of support, standing on a soft surface, and combining walking on a line with head movements. The subjects were instructed to continue with each exercise as long as it was difficult to perform or provoked symptoms. They were asked to register the number of exercise sessions (none/1/2 or more) in the diary each evening during the first 4 weeks after inclusion in the study and at the end of each week from the 5th to the 10th week. According to the diaries, the subjects performed exercises 89 times on average (SD, 40; range, 21–140) during the 70 days the diary lasted.
2.5
Statistical analyses
Three trials were performed for each of the static and dynamic balance tests. If a maximum value (30 seconds or no incorrect steps, respectively) was reached in the first or second trial, no additional trials were performed, and the maximum value was assigned to each remaining trial as in previous studies . The mean value of 3 trials was used for analysis.
Medians and interquartile range are presented because most variables were not normally distributed. To analyze change over time of variables on VAS, we used a rank-invariant nonparametric method developed by Svensson . For all other variables, Fisher exact test was used for comparisons between groups for categorical variables and Mann-Whitney U test for quantitative variables. For comparisons within groups, Wilcoxon signed rank test was used. P < .05 was considered significant. These analyses were computed using Statistica 8 computer software (StatSoft, Inc, Tulsa, OK).
As a dichotomous dependent variable in the predictor analysis, substantial self-rated remaining symptoms reported at 6 months and no self-rated remaining symptoms at 6 months were chosen. The 2 groups were defined according to the first 2 questions of the UCLA-DQ as in a previous study . Subjects in the group with substantial remaining symptoms reported at least occasional (at least 2 points) and at least mild (at least 2 points) symptoms according to the UCLA-DQ. Subjects in the group with no remaining symptoms reported no symptoms at all according to the UCLA-DQ. Those 2 groups were considered to be truly different regarding remaining symptoms. Subjects with very mild remaining symptoms (rarely and/or very mild symptoms according to UCLA-DQ) were not included in this part of the analysis.
2.6
Ethics
The Regional Ethics Committee for Human Research of the Faculty of Health Sciences, Linköping University approved the study (approval 02-470, 021217), and informed consent from all participants was obtained after the nature of the study was explained.
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
