Clendaniel, Richard A. PT, PhD
Numerous studies have documented the efficacy of exercises to alleviate the symptoms and physical limitations associated with the loss of vestibular function.1–4 The exercise approaches have generally fallen into 1 of 2 categories: adaptation or habituation exercises. The adaptation exercises are based on the demonstrated ability of the vestibular system to modify the magnitude of the vestibulo-ocular reflex (VOR) in response to a given input (head movement). The adaptation of the VOR has been demonstrated in individuals with normal vestibular function and those with unilateral vestibular hypofunction.5,6 One of the signals that induces adaptation of the VOR is retinal slip combined with head movement.7 This is the basis for what have traditionally been considered adaptation exercises. These exercises require the individual to perform rapid, active head rotations while watching a visual target, with the stipulation that the target remains in focus during the head movements.8 If the target is stationary, then the exercises are referred to as ×1 viewing exercises. If the target is moving in the opposite direction of the head movement, then these exercises are referred to as ×2 viewing exercises. Although these exercises have been shown to improve dynamic visual acuity (DVA), the actual mechanism behind this improvement is not known.9 As such, it may be more appropriate to refer to these exercises as gaze stability (GS) exercises.
In contrast with adaptation exercises, habituation exercises are based on the idea that repeated exposure to a provocative stimulus (eg, head movements) will lead to a reduction of the motion-provoked symptoms.3,10 This is thought to occur due to long-term changes within the nervous system, and there is clinical evidence indicating that the habituation exercises can lead to long-term changes in symptoms.11 The actual neural mechanism behind the effectiveness of the habituation exercises is not understood.
Although there is support for the 2 different intervention approaches, there are no studies, to date, that compare the 2 interventions in terms of changes in symptoms, motion sensitivity, and DVA. The preliminary results of a study designed to investigate this issue in individuals with an identified unilateral vestibular loss is reported. The underlying hypotheses for this study are that (1) individuals who perform GS exercises will have a greater improvement in DVA compared with those who perform the habituation exercises and (2) individuals who perform habituation exercises will have a greater reduction in their motion sensitivity compared with those who perform the GS exercises.
Participants in this study were recruited from the clinical practice of the author. All participants had a documented unilateral vestibular hypofunction based on surgical history, caloric test results, or clinical examination, and had abnormal clinical DVA test findings.12,13 Individuals were excluded from the study if they had a central nervous system (CNS) disease or cause of dizziness, orthopaedic problems that precluded performance of the exercises, if they were legally blind, or if they had dementia. Informed consent was obtained, and the institutional review board for human subjects protection at Duke University approved all aspects of the study.
The computerized DVA testing apparatus was similar to devices reported previously in the literature.14 An optotype (the letter E) was displayed on 25-inch, high-resolution computer monitor; the orientation and size of the optotype was under computer control. The orientation of the optotype was randomly altered, and the size of the optotype was progressively decreased in 0.1 MAR (logarithm of the minimal angle of resolution) increments. For each optotype size, 5 orientations of the optotype were presented. Static visual acuity was determined first. When the participant was unable to correctly identify all trials at a given acuity level, the test was stopped, and the number of missed optotypes was recorded. For determination of DVA, the optotype was presented every time the horizontal head velocity was between 120 and 180 degrees per second (determined with a rate sensor). Within each trial, the optotype was presented 5 times before the participant was forced to determine its orientation. When the participant was unable to correctly identify all trials at a given acuity level, the test was stopped. The participant's DVA was determined by subtracting the number of optotypes missed under the static visual acuity test from the number of optotypes missed under the dynamic component of the test. The DVA test was performed under both active (self-generated head movements) and passive (examiner-generated head movements) conditions. The test was performed separately for ipsilesional and contralesional head rotation. Because the ipsilesional DVA test result was the item of interest in this study, the contralesional passive DVA test was always run first to account for any learning effects that might occur. This method of DVA testing has been shown to be reliable with good sensitivity and specificity.14
The Dizziness Handicap Inventory (DHI) was administered to measure the participant's perception of how the symptoms are interfering with their lives.15 This tool is used routinely in clinical settings and has been shown to have greater sensitivity to change than global health surveys.16 Motion-provoked dizziness was measured using the Motion Sensitivity Test (MST). The MST measures the perceived intensity and duration of symptoms provoked by 16 rapid changes in head or body position. An overall score, the motion sensitivity quotient (MSQ), is determined from the results of each of the movements. The MST has been shown to be both a reliable and valid measure of motion-provoked dizziness.17
Participants were randomly assigned to either the GS exercise group, or the habituation exercise group using a program that generated a randomized block assignment to group based on the order in which the participant was enrolled in the study. The participants in the GS group performed a series of exercises designed to improve GS during head movements (Table 1) and balance and gait exercises. The participants in the habituation group performed a series of exercises designed to decrease their sensitivity to head movements (Table 1), and balance and gait exercises similar to those performed by the GS group. The exercises used in each group are used routinely in clinical practice and are often referred to as either adaptation exercises (the GS group), or habituation exercises (the habituation group). The participants were instructed to perform the exercises 3 times daily over a 6-week period. The participants returned to the laboratory once per week for clinical assessment and progression of the exercise program. The exercise progression was based on clinical experience treating patients with acute and subacute vestibular disorders. The exercise program was devised so that the participants could successfully complete each phase of the program. Static and DVA tests, MST, and the DHI were administered at the start of the program and at the end of the 6-week intervention period. The author performed all testing and treatment interventions.
Given the preliminary nature of this report, between-group statistical analyses were not performed. With the small sample size, pre- to posttreatment changes across all participants were assessed with the Wilcoxon signed rank test. Based on the initial studies of the DHI, an 18-point change is considered to be statistically significant.15 This value was based on test-retest results of a small sample (n = 14) and raises the question of how to measure improvement in individuals whose pretreatment DHI score is <18. Despite these limitations, the DHI changes of each individual will be compared with this 18-point value. Although there have been demonstrated changes in DVA testing with treatment, neither a minimal clinically significant difference nor statistically significant change scores have been reported for DVA testing.9 Earlier work by Herdman et al14 documented the normal values of the DVA by age. In subsequent studies, they refer to a reference value for documenting individual improvement in DVA.9 This is reported to be a change of 6 or more optotypes (personal communication). The DVA changes of each participant were compared with these values. To our knowledge, there are no reports have determined significant clinical or statistical change scores for the MSQ.
Eight participants with unilateral vestibular hypofunction have been enrolled in the study to date. Seven of the participants completed the study. One participant was unable to complete the study because of having to return to work. The mean age of the participants was 43.9 years (range, 27-65 years). Six of the 7 participants were female. The time since onset of the symptoms ranged from 4 weeks to 6 months, and the causes of the unilateral vestibular hypofunction included vestibular nerve resection, labyrinthectomy, and vestibular neuronitis. Three of the participants were enrolled in the GS group and 4 were enrolled in the habituation group. These data are summarized in Table 2. Compliance with the exercise program (based on weekly exercise logs) averaged 69.7% (range, 34%-90%).
Taken as a group, the participants demonstrated improvements in DHI scores, MSQ score, and both active and passive ipsilesional DVA. The DHI scores improved with intervention (pretreatment mean = 55.71, SD = 19.34; posttreatment mean = 8.0, SD = 7.30). This change was significant based on the Wilcoxon signed rank test (P < .01). The number of optotypes missed during the active ipsilesional DVA also decreased (pretreatment mean = 14.7, SD = 9.35, posttreatment mean = 8.3, SD = 9.37), and this change was also significant (P < .05). In a similar manner, the number of optotypes missed during the passive ipsilesional DVA decreased (pretreatment mean = 14.3, SD = 10.0, posttreatment mean = 8.7, SD = 9.71), which was also a significant change (P < .05). There was a marked reduction in the measure of motion sensitivity posttreatment (pretreatment mean = 25.21; posttreatment mean = 1.85), but because of the small sample size and the large variability, this was not statistically significant.
Because the number of participants was small, between-group statistical comparisons could not be performed. Observation of the individual data points for the different measures, however, is illuminating because there are clear trends in the preliminary data. Figure 1A and B demonstrate the individual changes in DHI and MSQ, respectively, by intervention group across the 6-week intervention session. For the DHI, all participants demonstrated a reduction in their scores (Fig. 1A). The 3 participants in the GS group each had decreases in the DHI score of >18 points. Three of the 4 participants in the habituation group also demonstrated decreases in their DHI scores of >18 points. The individual who did not demonstrate an 18-point change had a 10-point decrease from 22 to 12. Regardless of intervention group, the majority of the participants had improvements in DHI that are considered to be meaningful based on current understanding of the test metrics.15
Because of symptom severity at the time of the initial assessment, not all participants were able to complete the MST. Three of the 5 participants who were able to complete this test demonstrated a reduction in the MSQ score (Fig. 1B). For 2 of the individuals in the habituation group, there was a marked reduction in their motion sensitivity from values >40 to values <2. The other individual in the habituation group had a pretreatment MSQ score less than 10 and demonstrated a slight increase in MSQ score; the posttreatment value, however, was still <10. For the 2 participants in the GS group, the MSQ score decreased from 10.98 to 0.73 in 1 individual and increased from 0.05 to 0.44 in the other. For the first individual, this represents a change from a motion sensitive condition to one with little motion sensitivity. For the other individual, there was little motion sensitivity either pre- or posttreatment.
The individual changes in ipsilesional active and passive DVA by intervention group are depicted in Figure 2A and B. For the majority of the participants, there was an improvement in both active and passive DVA. For the participants in the GS intervention group, the results were mixed. For participant S1, even though the clinical DVA test result was abnormal, the computerized DVA test result was normal. There was an improvement of 2 optotypes during the passive DVA; however, this is not a significant change. Participant S3 had a markedly abnormal active and passive DVA at intial testing, neither of which showed improvement with the exercise intervention. There were significant improvements in the active and passive DVA for participant S4. The active DVA improved by 11 optotypes and returned to a normal level. The passive DVA improved by 8 optotypes, but did not return to a normal level.
Surprisingly, there were substantial reductions in the number of missed optotypes for the participants in the habituation intervention group. Participant S2 demonstrated an improvement of 16 optotypes for the active DVA and an improvement of 8 optotypes for the passive DVA. This also represents a change from an abnormal pretreatment DVA to a normal result posttreatment. Participant S5 had an improvement of 8 optotypes for the active DVA, and an improvement of 15 optotypes for the passive DVA. Although this is a large magnitude of change, this individual did not attain normal levels for either active or passive DVA posttreatment. The third participant in the habituation group, S7, did not demonstrate a significant in improvement in active DVA and did not attain normal levels for this test condition. For the passive DVA, however, there was an improvement of 3 optotypes, which improved performance to a normal level. Because participant S8 was too symptomatic to perform the active DVA tests at the pretreatment assessment, only the passive DVA was measured. There was an improvement of 4 optotypes with the passive DVA, which represents a change from an abnormal DVA pretreatment to a normal DVA posttreatment.
The improvements observed across participants were not unexpected based on previous studies.2–4 Both intervention groups demonstrated similar improvements in the DHI score, which suggests that there was a decrease in the impact of the symptoms on the participants' lives and that the type of exercise intervention was not an important factor. Participant S3 showed unusual findings; there was a significant decrease in the DHI score, yet no change in either active or passive DVA. In addition, the MSQ score was <1 on both the pre- and posttreatment assessments. According to the self-report exercise log, participant S3 was compliant with the exercise program (73%). A possible explanation for the disparity between the DVA test results and DHI scores is that prior to beginning the exercise program this participant avoided movements and activities for fear of symptom provocation, which would be reflected in the pretreatment DHI score. With initiation of the exercise program, the participant may have discovered that he was able to move and function without symptoms (as reflected by the low MSQ scores), which would lead to decreases in the DHI scores. It is not clear why participant S3 demonstrated no improvement in DVA. It could be as a result of improper performance of the exercises at home or some other factor. Exercise intervention in the treatment of vestibular hypofunction has been shown to not be beneficial in all cases.18 In that study, the percentage of participants who felt the rehabilitation program was of benefit was greater than the percentage of participants who demonstrated improvements in gait measures, which is consistent with the disparity between physiologic measures and patient-reported outcome measures observed in participant S3.
Given the small number of participants in this preliminary report, one must be cautious to not overinterpret the results. So, the present results demonstrate to somewhat unexpected results that bear further discussion and exploration. The first result is the reduction of the MSQ score in the GS group. This intervention group did not perform exercises designed specifically to alleviate their sensitivity to large amplitude, rapid head movements. The 1 participant in this group who had a pretreatment MSQ score >10 had a marked reduction in the MSQ score to a value <1. Although there are several possible explanations for this, one must keep in mind that of the 2 participants in the GS group who completed the MST, only 1 had measurable motion sensitivity. One possible explanation for the reduction in motion sensitivity is that the GS and balance exercises, as well as the participant's daily activities, included sufficient provocative stimuli to cause habituation of the motion-provoked symptoms. Another possible explanation for the reduction in the motion sensitivity is that the GS intervention led to adaptation of the vestibular system. This adaptation would result in resolution of the sensory mismatch among vestibular, visual, and somatosensory inputs. With the loss of the sensory mismatch, which is thought to produce the symptoms of motion sensitivity, there would be no motion-provoked dizziness. The efficacy of the GS exercises to reduce motion sensitivity needs to be confirmed with continued enrollment of participants in the study.
The most interesting finding in the study is the improvement in the active and passive ipsilesional DVA in the participants performing the habituation exercises. This improvement was seen across all participants in the habituation intervention group, who did not perform exercises designed to improve their visual acuity during head movements. The improvements in both the GS and habituation groups are similar (except for 1 participant in the GS group who showed little to no improvement). There are several possible explanations for this improvement in DVA in the habituation intervention group. One, it is possible that the balance exercises and the participants' daily activities provided sufficient stimuli to induce improvements in DVA. However, in a study by Herdman et al,9 in which the participants performed either gaze stabilization and balance exercises, or a placebo (saccadic eye movements) and balance exercises, the group performing the placebo and balance exercises showed no improvement in their DVA measures. Another possible explanation is that the head movements used in the habituation intervention enabled, or facilitated, the central compensations that resulted in the improvements in DVA. It may be that the head movements pose a challenge (sensory mismatch) to the CNS, which then attempts to resolve the challenge. As the CNS learns to compensate for the sensory mismatch, it may also learn strategies for improving visual acuity during head movements. The common factor in the exercises performed by both the GS and habituation groups is the head movement, which may be the enabling factor in the compensation process. A similar result was described by Cohen and Kimball.19 These investigators demonstrated improvements in measures of ataxia, as well as static and dynamic postural stability, in individuals with unilateral vestibular hypofunction after a rehabilitation program consisting solely of habituation exercises. This explanation, that head movement enables the compensation process, is also supported by the clinical observation that those individuals with a vestibular, who are active and move, tend to improve (decreased symptoms and increased functional levels) without intervention.
The actual mechanisms underlying the improved DVA for either intervention are not known. Analysis of the eye movements during the DVA test both pre- and postintervention may help elucidate these mechanisms. These data are currently being collected but are beyond the scope of this report.
Gaze stabilization and habituation exercises have previously been shown to decrease symptoms of dizziness and increase function in individuals with vestibular disorders. The preliminary results of this study indicate that both exercise interventions lead to a reduction in the self-report measure of the impact of symptoms on the ability to function, a decrease in the sensitivity to movements, and an improvement in the ability to see clearly during head movements. Continued investigation is needed to determine whether these results will hold, to determine whether there are different effects of the 2 interventions, and to determine the mechanisms of improved visual acuity.
1.Cawthorne T. The physiological basis for head exercises. J Chart Soc Physiother. 1944;30:106–107.
2.Horak FB, Jones-Rycewicz C, Black FO, Shumway-Cook A. Effects of vestibular rehabilitation on dizziness and imbalance. Otolaryngol Head Neck Surg. 1992;106:175–180.
3.Shepard NT, Telian SA, Smith-Wheelock M, Raj A. Vestibular and balance rehabilitation therapy. Ann Otol Rhinol Laryngol. 1993;102(3 Pt 1):198–205.
4.Herdman SJ, Clendaniel RA, Mattox DE, Holliday MJ, Niparko JK. Vestibular adaptation exercises and recovery: acute stage after acoustic neuroma resection. Otolaryngol Head Neck Surg. 1995;113:77–87.
5.Pfaltz CR. Vestibular compensation. Physiological and clinical aspects. Acta Otolaryngol. 1983;95:402–406.
6.Szturm T, Ireland DJ, Lessing-Turner M. Comparison of different exercise programs in the rehabilitation of patients with chronic peripheral vestibular dysfunction. J Vestib Res. 1994;4:461–479.
7.Shelhamer M, Tiliket C, Roberts D, Kramer PD, Zee DS. Short-term vestibulo-ocular reflex adaptation in humans. II. Error signals. Exp Brain Res. 1994;100:328–336.
8.Herdman SJ. Exercise strategies for vestibular disorders. Ear Nose Throat J. 1989;68:961–964.
9.Herdman SJ, Schubert MC, Das VE, Tusa RJ. Recovery of dynamic visual acuity in unilateral vestibular hypofunction. Arch Otolaryngol Head Neck Surg. 2003;129:819–824.
10.Telian SA, Shepard NT, Smith-Wheelock M, Kemink JL. Habituation therapy for chronic vestibular dysfunction: preliminary results. Otolaryngol Head Neck Surg. 1990;103:89–95.
11.Clement G, Tilikete C, Courjon JH. Retention of habituation of vestibulo-ocular reflex and sensation of rotation in humans. Exp Brain Res. 2008;190:307–315.
12.Halmagyi GM, Curthoys IS. A clinical sign of canal paresis. Arch Neurol. 1988;45:737–739.
13.Longridge NS, Mallinson AI. The dynamic illegible E-test. A technique for assessing the vestibulo-ocular reflex. Acta Otolaryngol. 1987;103:273–279.
14.Herdman SJ, Tusa RJ, Blatt P, Suzuki A, Venuto PJ, Roberts D. Computerized dynamic visual acuity test in the assessment of vestibular deficits. Am J Otol. 1998;19:790–796.
15.Jacobson GP, Newman CW. The development of the Dizziness Handicap Inventory. Arch Otolaryngol Head Neck Surg. 1990;116:424–427.
16.Enloe LJ, Shields RK. Evaluation of health-related quality of life in individuals with vestibular disease using disease-specific and general outcome measures. Phys Ther. 1997;77:890–903.
17.Akin FW, Davenport MJ. Validity and reliability of the Motion Sensitivity Test. J Rehabil Res Dev. 2003;40:415–421.
18.Krebs DE, Gill-Body KM, Parker SW, Ramirez JV, Wernick-Robinson M. Vestibular rehabilitation: useful but not universally so. Otolaryngol Head Neck Surg. 2003;128:240–250.
19.Cohen HS, Kimball KT. Decreased ataxia and improved balance after vestibular rehabilitation. Otolaryngol Head Neck Surg. 2004;130:418–425.