Impact of Tumarkin attacks on complaints and work ability in Ménière’s disease

2.1 Study design and participants

Permission was obtained from the Finnish Ménière Federation (FMF; Suomen Meniere-liitto) to analyse the registry data that the FMF had collected from their members. The data was based on an extensive questionnaire on symptoms related to MD. According to Finnish law, this kind of survey study conducted in association with patient organizations, which involves anonymous data, does not require ethical approval. The FMF has 1,646 personal members. The e-mail address was missing among 704 members. Hence, the FMF sent the electronic survey to 952 members. Four email reminders were sent for those who did not respond to the initial request to complete the survey. In total, 602 persons responded (i.e., 63.2% response rate) to the survey. The mean age of respondents was 56.7 years (range25 to 75 years). The duration of the disease was on average 12.4 years (range 0.5 to 35 years). The respondents included 477 (79.2) females and 125 (20.8%) males, respecting the gender distribution of FMF.

2.2 Data collection

The data was gathered using an electronic questionnaire sent via the Internet. The questionnaire was used for assessing the symptoms and consequences of the disorder (see Appendix). The questionnaire was evaluated by 12 FMF board members for content appropriateness, and revisions were made based on their suggestions. The questions focused on impact, social economic aspects, and complaints of MD, specifically related to Tumarkin attacks. In this study, the Tumarkin attack was defined as a short attack of vertigo or postural instability that was not associated with head movement. When questioning about T-LoC in connection with Tumarkin attack, we asked: “During an attack of sudden loss of balance (i.e., Tumarkin attack) have you ever experienced a loss of consciousness? Here we mean attack related to loss of consciousness, not related to any head injury due to falling or any type of accident.” Following were the response options provided: (1) I do not have Tumarkin attacks; (2) I have Tumarkin attacks but have not lost consciousness; (3) Yes, in my opinion I have been unconscious during a Tumarkin attack; (4) Yes, external person has eye witnessed that I have been unconscious during a Tumarkin attack. In addition, there were a few disease and impact specific assessments using a mixture of open-ended and closed questions. In structured (i.e., closed) questions, the impact of the disorder was rated on a four-step scale from no impact to severe impact.

2.3 Data analysis

Descriptive statistics were explored. Non-parametric tests such as Chi square test, Mann-Whitney U test and Kruskal Wallis H test were used. Student’s t test was used to study the difference between continuous variables among participants with and without Tumarkin attacks. To further explore the association between Tumarkin attacks and demographic details and also complaints of the disease, we used logistic regression analysis. In stepwise regression modelling for binary variables, we used p-value of 0.10 to enter into the model. However, a p-value of 0.05 was used for a statistically significant interpretation. Answers to open-ended question were analysed using the qualitative content analysis; also, the frequency of occurrence of each category was noted.

3.1 Prevalence and occurrence of Tumarkin attacks

From 602 respondents, 295 reported that they have had Tumarkin attacks (i.e., 49 %). The duration of Tumarkin attacks is often short (only a few seconds) among 278 subjects. However, 31 respondents reported that the duration was a minute or longer. The frequency of the attacks was as follows: less than once a week in 216, weekly in 67 and daily in 23 respondents. The attacks occurred sporadically among 262 respondents and in cluster among 39 respondents. The length of the MD or age of the respondents did not explain the occurrence of Tumarkin attacks.

3.2 Complaints associated with Tumarkin attacks

A chi square test was used to compare the association among complains and occurrence of Tumarkin attacks in people with MD. Results showed that several complaints were more prominent among respondents with Tumarkin attacks when compared to participants without Tumarkin attacks (see Table 1). Also, a t test showed no significant difference in age [Mean of 57.1 years and 56.6 years; t = 0.474, p = 0.635] and also duration of disease [Mean of 12.0 years and 12.5 years; t = 0.362, p = 0.718] among respondents with and without Tumarkin attacks.

In stepwise logistic regression analysis, we explored the complaints that could predict Tumarkin attacks in MD (see Table 2). The model was highly significant (p < 0.001) and explained 16.3% of the variability of Tumarkin attacks. In the model, the factors of vertigo (Odds ratio = 3.591) and balance problems (Odds ratio = 4.186) were statistically significant, but the variable of temporary moving spots in the visual field (Odds ratio = 1.535) was not statistically significant.

The Tumarkin attacks had led to tripping among 133 persons and falling among 92 patients. From the fallers due to the Tumarkin attack, 53 persons suffered injuries:38 had bruises, 12 had fractures, and 3 suffered severe back injuries. One person had multiple injuries due to a car crash; additionally, among two persons, the partner sitting beside them prevented the car accident that could have been caused by Tumarkin attack. In addition, respondents with Tumarkin attacks experienced various other problems with 19 to 42% respondents experiencing significant impact (see Table 3).

From the 295 respondents, 45 reported that they had experienced T-LoC in connection with a Tumarkin attack. T-LoC among 23 respondents was witnessed, and 22 respondents self-reported the event. Factors such as age, gender, and duration of MD in patients with Tumarkin attacks did not differ among those with and without T-LoC. During rotatory vertigo attack, none of the patients reported an attack of T-LoC associated with vertigo.

When querying participation in the working life, of the 602 respondents, 242 (i.e., 40.2%) were working full-time, 65 (i.e., 10.8) were working part-time, 252 (i.e., 41.9%) were retired either by age or medical reasons, and the remaining 15 (i.e., 2.5%) were on sick leave. Also, from the 602 subjects, 86 (i.e., 14.3%) were retired because of MD. In addition, in 21 respondents (i.e., 3.5%) the sick pensions were rejected, and in 4 persons the decision was pending. A Kruskal-Wallis test showed that the respondents with Tumarkin attacks had more frequently applied pension than those not having Tumarkin attacks (Chi-square 3.90, p = 0.048; see Table 4), but when evaluating their own concept or willingness to search for sick-pension, the difference was not statistically significant (Chi-square 3.16, p = 0.076; see Table 5).

The disease related complaints between MD retired persons and those not retired because of MD were evaluated using the Mann-Whitney U-test. The results suggest that retired persons with MD, the age was somewhat higher (Z = 2.218, p = 0.027) and also had a longer course of MD (Z = 4.620, p < 0.001) when compared to non-retired persons. Marital status did not have an impact on retirement nor on hearing difficulties. The balance difficulties (Z = 5.283, p < 0.001), vertigo (Z = 3.516, p < 0.001), problems working on a PC-screen (Z = 5.539, p < 0.001) and Tumarkin attacks (Z = 2.724, p < 0.001) were significantly more problematic among retired persons with MD when compared to non-retired persons with MD. Tumarkin attacks were significantly more prevalent among retired persons with MD (Z = 3.590, p < 0.001). Also, among retired persons with MD, the Tumarkin attacks were related to social isolation (Z = 2.876, p < 0.004).

We also asked the subjects to rate their ability to work on scale of 0 to 100. On average, the mean quality of life on a VAS scale was 61.9 and 70.4 in participants with those with and without Tumarkin attacks respectively (see Fig. 1). The student’s t-test showed statistically significant differences in self-rated work ability scores between non-Tumarkin and Tumarkin groups (t = 3.828, p < 0.001).

OPEN IN VIEWER

Fig. 1

Distribution of self-rated work ability on Euro-Qolin Visual Analogue Scale among Ménière’s disease participants with and without Tumarkin attacks.

When the MD-retirement was modelled in logistic regression analysis, we included all persons less than 65 years of age. In the regression model (see Table 6), 4 variables were included and the model was statistically significant (p < 0.001). The variables including balance problems (Odds ratio = 3.7), problems with PC-screen (Odds ratio = 3.4), and T-LoC attacks (Odds ratio = 2.5) were statistically significant, but fatigue (Odds ratio = 2.1) was not statistically significant (Odds ratio = 1.521). However, the model could only explain 14.1% of variability, indicating that there are other factors that could help decide early retirement by MD.

4.1 Consequences

Postural stability related problems are often prominent in people with MD [26]. Regarding the rating of postural stability, the current study showed that only 55% of the patients with Tumarkin attacks considered their postural stability to be significantly impaired when compared to 35% among patients without Tumarkin attacks. The consequences of Tumarkin attacks could be severe as bruises in 33% of these patients. Exercise is commonly recommended and used to improve postural stability in MD. Kentala et al. [12] demonstrated that among participants of FMF, that exercise has ranked high in self-help methods used by the participants, and that 50% of them specified either general (35%) or specific exercises (15%) to alleviate their balance problems. The vestibular rehabilitation would definitely be beneficial for gait instability, but as in severe cases Tumarkin attacks appeared without any warning and led to falls with or without T-LoC. It is difficult to believe that vestibular rehabilitation would prevent injuries caused by the attacks. Ratdke et al. [31] reported that the 12-months prevalence of orthostatic dizziness among patients with or without vestibular complaints was 10.9%. Moreover, there is a strong longitudinal relationship between prevalent orthostatic intolerance and injury-related deaths [6]. Other therapeutic methods are needed to prevent falls in MD.

Patients with disabling MD typically experience circular vection when looking at moving pictures on a TV-screen, such as watching cars passing by and even queuing in shops [19]. Disturbing or conflicting visual orientation causes patients to postural instability as an expression of vestibular distrust. We have earlier shown that the visual complaints correlate with T-LoC associated with MD [28, 29]. The patients experienced visual field changes as loss of part of the visual field and moving spots in the visual field. These changes in vision seem to be generated by the vestibulo-ocular reflex originating from the stimulation of the otolith afferent [13]. We also interpret that the high prevalence of headaches among patients depending on vestibular conflict is due to the enhanced muscle tone needed to improve stability.

Tumarkin attacks can occur at any time during the course of MD as seen in our previous study [9]. In exceptional cases, it may even be the initial manifestation [2]. Ballester et al. [1] indicated among elderly patients with MD, the drop attacks were highly prevalent (about 40%). Although in the present study we did not observe age dependency between the falls, we have noticed that in elderly homes, there exists a high prevalence of vestibular problems among fallers, and in 10% of the elderly, vertigo attacks were identical with those of MD [37]. The etiology of the drop attacks is in question as they are prone to be misdiagnosed as the symptoms and signs were not consistent with classical MD [21]. In an audit, Kwong and Pimlott [15] showed that 46% of elderly, who came for primary consultation for vertigo, were either diagnosed improperly or their symptoms were considered to be simply symptomatic of their natural aging process. Furthermore, a study conducted by Lawson et al. [16] on 59 patients with benign paroxysmal vertigo showed that 59% were referred to Fall and Syncope Unite instead of ENT. The authors concluded that making a diagnosis of inner ear originated vertigo, especially in elderly, may be difficult, as this group frequently has more than one type of symptom related to dizziness and even symptoms from multiple co-morbidities and coexistent cardiovascular disorders. We emphasize that while evaluating T-LoC or a sudden fall in a patient with MD, the possibility of a Tumarkin attack should be considered as a differential diagnosis. Our evidence shows that when these patients entered the emergency unit for a differential diagnosis, a Tumarkin attack was not even considered at a tertiary hospital level [29].

Fatigue is a subjective sensation of weakness or tiredness or lack of energy and is a common complaint in the general healthy population. van’t Leven et al. [38] found in a postal survey that short-term fatigue affects 4.9% and chronic fatigue 30.5% of the general population. Community and primary care studies estimate that 5 to 45% of the population has reported fatigue as a debilitating symptom, and that these rates increase with age [14]. Most physical illnesses are associated with fatigue, although it is often viewed as one of the least important symptoms by physicians [20]. In neurologic diseases, prevalence of fatigue is even more elevated than would be expected only on the basis of age and disability [14]. Levo et al. [19] showed 70% of people with MD reported fatigue and 30% were moderately or severely exhausted. In another study on people with MD, the personality trait, attitude of the subject, and anxiety were associated with fatigue and are likely to predispose the person to changes in vitality [18]. In the present study, we observed that MD specific complaints leading to changes in vitality were mobility and balance disturbances between attacks, intensity of vertigo attacks, and hearing loss, all of which are more prominent among patients with Tumarkin attacks. Our findings also confirm the outcome of a study of Perez-Fernandez et al. [24] who observed that patients who suffered Tumarkin attacks were more disabled and experienced more autonomic nervous symptoms as well as fatigue. As fatigue is a common complaint in MD and especially in its severe form, it should be recognized and included in therapeutic procedures.

Calzadaet et al. [5] evaluated the otolith membrane of patients with sudden drop attacks and described that all patients with Tumarkin attacks showed evidence of a disrupted utricularotolithic membrane, thus making the patients liable for even small pressure changes and suggesting that the underlying pathophysiology in sudden drop attack results from injury to the otolithic membrane of the saccule and utricle. We also found that problems such as T-LoC, gait problems and visual complaints were less prevalent among patients without Tumarkin attacks, indicating that the role of semicircular canals may not be as important in provoking these complaints as the otolithic system has proven to be. Our findings support the theory that the reason for T-LoC involves locating the malfunctioning otolith organ, which disturbs the vestibular sympathetic reflex originating from the otoliths and provokes the paradoxical vaso-vagal attack leading to T-LoC [28, 29]. In previous studies we have shown that in MD, 4 to 6% of patients have experienced T-LoC [28, 29]. The animal experiments indicate that T-LOC arises by activation of the vestibular sympathetic reflex [29, 32]. We interpret that the activation of the vestibular sympathetic reflex controlling the circulatory homeostasis of the body becomes erroneous in patients with MD. The sympathetic tone is lost and the baroreflex feedback is inhibited leading to T-LoC in connection with Tumarkin attacks.

4.2 Work problems

In the present study, the consequences of Tumarkin attacks on work were significant. Tumarkin attacks reduced social interaction and working ability and interfered with the well-being of the participants. We noticed that 14% of the participants were retired because of MD. The rate of persons on sick-retirement was significantly greater among patients with Tumarkin attacks, and 35% reported that they had problems managing their daily work. The Tumarkin attacks per se and T-LoC attacks were not the only determinant of limiting work ability, but the associated factors of otolithic disturbance; for example, visual problems due to working on a PC screen, poor balance, and fatigue all proved to be factors in this regard. A recent study showed that MD patients show significantly higher levels of motion sickness susceptibility, more headaches and migraines, increased visual display dizziness, increased syncope and work impact when compared to people with other vestibular conditions [7]. A study by Soto-Varela [34] demonstrated that Tumarkin attacks were a significant factor in increasing the severity of the impact of MD. Also, Perez-Fernandes et al. [24] showed that patients who suffer Tumarkin attacks are more disabled and experience more severe and more frequent autonomic symptoms. The authors considered that treatment must be carefully planned to be as conservative as possible in terms of hearing, and that psychiatric and/or psychological treatment must always be considered as an adjuvant therapy. Our study findings confirm the observations discussed above.

There are, however, limited possibilities to alleviate the complaints with psychiatric and/or psychological treatment as the basis of complaints are delivered by vestibular maladjustment. Recently we have introduced a web-based portal for peer support and physical training to improve understanding and provide trust for the patients with MD, and this approach seem to alleviate the psychological maladjustment of the subject [27]. However, if the medical condition is not optimally treated, the peer support or psychological guidance may not be effective. There is abundant data in the literature that Tumarkin attacks can be abolished with vestibular nerve sectioning or vestibular organ ablation [3]. To control the impact of Tumarkin attacks and/or postural instability, intratympanically applied gentamicin has been recommended [10, 23]. The outcome of the efficacy of gentamicin in people suffering Tumarkin attacks varies between 60 to 100% [10, 40]. The gentamicin ablation of the otolith organ was not effective in some patients with T-LoC [29], probably due to individual variability of the tightness of the oval window membrane and annular ligament [41]. The endolymphatic sac surgery or the use of vestibular suppressants were not effective in preventing Tumarkin attacks [9].

4.3 Study limitations

It is important to be aware of the study limitations in order to ensure that the study findings are interpreted with caution. First, the study used a survey design in which the participants are required to report the Tumarkin attacks that they have had in the past. This may have introduced some recall bias. Second, the questionnaire used has not been validated for its interpretation from a larger sample. Also, we do not have a control group to gauge the accuracy of the questions. This may have introduced some interpretation bias. Third, participants were recruited through a patient organization. Hence, the study sample may not be representative of the outpatient clinical population. Fourth, the diagnostic accuracy of MD participants in the current study is difficult to confirm. To evaluate the accuracy, we have previously conducted an in-depth analysis of the complaint, indicating that few people in FMF had probable MD and most of them had definite MD based on the criteria of the American Academy of Otolaryngology – Head and Neck Surgery (AAO-HNS) [33]. Also, there is some confusion among physicians concerning the diagnosis of MD versus migraine as the differentiation between MD and vestibular migraine is extremely difficult to identify [22]. It is likely that people with vestibular migraine may have been included in the current study and that could account for errors in some of the results, although we do not have evidence of this. Both vestibular migraine and MD have endolymphatic hydrops and the response to migraine therapy may discriminate between MD and vestibular migraine [22]. For these reasons, the findings of the current exploratory study should be considered tentative and should be confirmed with more robust study designs.