Abstract
The objective of the study was to examine migrainous vertigo prospectively by means of a diary. We included 146 patients with at least one migraine attack per month. All patients underwent a semistructured interview, completed questionnaires on depression, anxiety and quality of sleep and kept a diary covering detailed information on headache, vertigo and dizziness over a period of 30 days. A completed diary was returned by 116 patients (79.5%). Based on the diary migrainous vertigo (MV) was diagnosed in 18 patients (15.5%) and non-migrainous vertigo or dizziness (non-MV) in 35 patients (30.2%). MV was present on 65 of 3477 patient days (1.9%) and non-MV on 145 days (4.2%). MV occurred more often on days with headache (P < 0.001). Its median duration was 3 h and it lasted longer on days with headache than on days without headache (P < 0.001). The most prominent specific feature of MV was head motion intolerance. Patients with MV showed anxiety more often (P < 0.001) and tended to have worse quality of sleep and higher depression scores. In conclusion, vertigo and dizziness are frequent symptoms in migraineurs. The 1-month prevalence of MV is 16% and that of non-MV 30% in patients with at least one migraine attack per month. Frequency of MV is higher and duration longer on days with headache. MV is a risk factor for co-morbid anxiety.
Introduction
Migraine and vertigo are very common in the general population with life-time prevalences of 16% and 7%, respectively (1). The chance of coincidence is about 3–4%. It is a well-known fact, however, that the association of migraine and vertigo is more than pure coincidence. Vertigo related to migraine is termed migrainous vertigo (MV) showing a life-time prevalence of 1% in the general population, with female preponderance, but no relation to age (2). Diagnostic criteria for MV were proposed by Neuhauser et al. (3) and allow for a standardised nomenclature. These diagnostic criteria also apply to the recently introduced term of ‘vestibular migraine’ (4), that is now widely used as a synonym for MV in order to allow a clear separation from non-vestibular dizziness associated with migraine. Rotational vertigo, illusory self or object motion, positional vertigo and head motion intolerance are considered to be typical of MV. Symptoms such as light-headedness, drowsiness and a feeling of unsteadiness are unspecific and due to non-vestibular dizziness (4). Possible pathophysiological mechanisms of MV include cortical spreading depression, an abnormal activity of central and peripheral vestibular neurons and an involvement of the connection between trigeminal and vestibular nuclei (5). Up to now, no underlying genetic defect has been found (6).
The diagnosis of MV poses a challenge to the clinician, because various dizziness and vertigo syndromes show an increased prevalence in migraineurs (7–9). In addition, there are no MV-specific neuro-otological findings to support the diagnosis (9,10). To complicate matters further, MV is not temporally associated with any certain phase of the migraine attack and may even occur without headache (3). Sometimes, only accompanying migrainous symptoms such as photophobia or phonophobia are pointing to a connection between vertigo and migraine. Besides, vertigo may occur by chance in migraineurs without fulfilling the diagnostic criteria for MV. The International Headache Society (IHS) accepts the inter-relation of migraine and vertigo only in the context of basilar-type migraine (BM) regarding vertigo as an aura symptom of the posterior circulation (11). In addition to vertigo, the diagnosis of BM requires at least one other aura symptom (such as dysarthria, tinnitus, hypacusis, visual symptoms bilaterally in both hemifields or bilateral paresthesia). Less than 10% of migraine patients suffering from migraine-related vertigo or dizziness fit into this concept (12). A study on clinical manifestations and audiovestibular function in 15 patients with BM, 30 with definite MV, and 32 with probable MV suggested that BM is the most severe form followed by definite MV, and probable MV as its mildest form (13).
MV must be differentiated from other vestibular disorders such as Meniere’s disease and benign paroxysmal vertigo (BPPV). Hearing loss is a frequent symptom of Meniere’s disease and the most reliable feature to differentiate it from MV (14). BPPV represents the most common recurrent vestibular syndrome in migraineurs. In contrast to BPPV, positional vertigo typical of MV lasts as long as head position is maintained.
So far, only a few studies on MV have been prospective and used explicit inclusion and exclusion criteria (2,3). Based on the criteria of Neuhauser et al. (3), the prospectively evaluated prevalence of MV was 7% among patients of a dizziness clinic and 9% in a migraine clinic group.
The aim of the present study was to examine MV prospectively for the first time by means of a diary in order to assess further its prevalence and characteristics and to investigate its relation to impact of headache, quality of sleep, depression and anxiety.
Subjects and methods
In 2002, we recruited patients with migraine through articles in the two most popular newspapers in Eastern Austria and performed a comprehensive diary study analysing a wide spectrum of factors related to headache in migraineurs (15). Recently, we published a 30-month follow-up examination focusing on the course of headache (16). In this study, we present the findings on vertigo and dizziness recorded at follow-up. The study was approved by the local ethics committee and all patients gave written informed consent.
Inclusion and exclusion criteria have been published previously (15). The patients were not informed about the specific interest in vertigo and dizziness before the inclusion visit in order to avoid selection bias. In summary, patients aged >18 years were included, with at least one migraine attack per month fulfilling the criteria of migraine without aura or migraine with aura according to the second edition of the International Classification of Headache Disorders (ICHD-II) (11). We included patients with co-existing tension-type headache (TTH), but excluded those with probable migraine, medication overuse headache, any other primary or secondary headache and severe other diseases. In addition, we excluded patients with known other vestibular disorders and those with continuous vertigo or dizziness.
Included patients underwent a semi-structured interview, assessing biographical data, life-style, general medical history and characteristics of headache, vertigo and dizziness (17). Following the interview, patients were asked to fill out four validated questionnaires: Headache Impact Test (HIT-6) (18), Pittsburgh Sleep Quality Index (PSQI) (19), Self-rating Depression Scale (SDS) (20), and Self-rating Anxiety Scale (SAS) (20).
The HIT-6 allows quantification of the impact of headache on daily life and consists of six questions. Each has five possible answers: never (6 points), rarely (8 points), sometimes (10 points), frequently (11 points) and always (13 points) and, therefore, offers a total score of minimum 36 to maximum 78. A severe impact on quality of life is presumed of scores 60 and higher (18).
The PSQI measures the quality and patterns of sleep in the preceding 4 weeks. It differentiates ‘poor’ from ‘good’ sleep by measuring seven areas: subjective sleep quality, sleep latency, sleep duration, habitual sleep efficiency, sleep disturbances, use of sleeping medication, and day-time dysfunction over the last 4 weeks. Scoring of answers is based on a 0–3 Likert scale, whereby 3 reflects the negative extreme. A total score of 5 or more indicates a ‘poor’ sleeper (19).
The SDS is a 20-item self-report measure of the symptoms of depression. Subjects rate each item according to how they felt during the preceding 7 days. Item responses are ranked from 1–4. The sum of the 20 items produces a score ranging between 20–80, a value of >40 suggests clinically relevant depression (21).
The SAS measures affective and somatic symptoms of an anxiety disorder. The structure of the SAS is like the one of the SDS. It also consists of 20 questions, which refer to the past 7 days. The result of the SAS is obtained by summing up the scores. A cut-off value of >40 suggests the presence of a clinically relevant anxiety disorder (20).
Finally, patients were provided with a paper-pencil diary for 30 days and asked to return it filled out on a second visit. The diary covered questions regarding the presence and characteristics of headache, vertigo and dizziness and had to be filled in every night. The questions about headache characteristics strictly followed the ICHD-II criteria of migraine and tension-type headache and covered duration, laterality (unilateral or bilateral), quality (pulsating, pressing or other) and intensity (mild, moderate or severe) as well as the absence and presence of aggravation by routine physical activity, nausea, vomiting, photophobia, phonophobia and aura symptoms. The questions about the characteristics of vertigo and dizziness followed the criteria of Neuhauser et al. (3) and covered vertigo, illusionary motion, head motion intolerance, light-headedness, unsteadiness and drowsiness. In addition, the patients were instructed about the specific meaning of vertigo (a sensation of spinning or turning when stationary, ‘like being on a roundabout’), illusionary motion (a perception of movement in surrounding objects) and head motion intolerance (an experience of imbalance, spinning or turning when moving the head).
One of the authors (CW) was responsible for classifying each single headache attack and each episode of vertigo or dizziness recorded in the diaries. The evaluation was blinded in two ways. First, CW did not see patients and was not involved in performing interviews. Second, for classification of headache and vertigo an anonymised SPSS data set was used. Migraine without aura, migraine with aura and probable migraine were classified as migraine according to ICHD-II (11). Headaches not fulfilling these criteria were classified as non-migrainous headache. Episodes of vertigo were classified as MV, if they fulfilled the diagnostic criteria proposed by Neuhauser et al. (3). All other episodes of vertigo and dizziness were summarised as non-migrainous vertigo or dizziness (non-MV).
Statistical analysis
All data were analysed with SPSS for Windows v.15.0. For comparing MV, non-MV and absence of vertigo and dizziness as well as for comparing migraine, non-migrainous headache and absence of headache we used Kruskal–Wallis tests and Wilcoxon tests, respectively, for continuous variables and chi-squared tests for categorial variables. For analysing the relation of precipitants to types of vertigo and dizziness, we calculated Pearson correlation coefficients. To adjust for multiple comparisons, Bonferroni correction was performed online using SISA online statistical analysis <http://www.quantitativeskills.com/sisa/calculations/bonfer.htm> following the suggestions of Sankoh et al. (22) Considering correlations between the variables, the adjusted P-value was set to 0.005.
Results
A total of 153 patients completed the semistructured interview and the questionnaires. Seven were excluded because they did not fulfil the ICHD-II criteria of migraine without aura or migraine with aura and 116 patients (79.5%) returned a completed diary.
Biographical data
Biographical data of 146 patients with migraine seen at the first visit
Patients with and without diary
Patients who completed the diary were older than non-completers (46.4 ± 11.4 vs 37.8 ± 11.1 years; P < 0.001) and their time since onset of headache was longer (25.0 ± 13.2 vs 16.3 ± 9.4 years; P < 0.001). In contrast, gender, frequency and duration of headache, HIT-6 score and prevalence of vertigo and dizziness did not differ in the two groups.
Prevalence of headache, vertigo and dizziness
In a total of 3477 patient-days recorded by 116 patients, headache was present on 960 days (27.6%) and vertigo or dizziness on 210 days (6.0%). Headache was classified as migraine on 572 days (16.5%), as non-migrainous headache on 360 days (10.4%). Headache could not be classified on 28 days (0.8%). MV occurred on 65 days (1.9%) in 18 patients and non-MV on 145 days (4.2%) in 35 patients. Ten patients had episodes of MV as well as non-MV. Accordingly, the one-month prevalence of MV was 15.5% and that of non-MV 30.2%.
Subjects with MV recorded 5.6 ± 6.7 days with vertigo or dizziness (range [R] 1–28, lower quartile [Q1] 1.0, median [M] 3.5, upper quartile [Q3] 6.5), 3.5 ± 4.0 days with MV (R 1–15, Q1 1.0, M 2.0, Q3 3.0), and 2.1 ± 3.5 days with non-MV (R 0–13, Q1 0.0, M 1.0, Q3 2.25). Subjects experiencing exclusively non-MV recorded 3.1 ± 3.1 days with vertigo or dizziness (R 1–12, Q1 1.0, M 2.0, Q3 3.0).
The frequency of MV and non-MV was not related to migraine frequency. Patients with ≤3 migraine-days recorded 0.3 ± 0.8 days with MV (R 0–3, Q1 0.0, M 0.0, Q3 0.0) and 0.9 ± 2.1 days with non-MV (R 0–12, Q1 0.0, M 0.0, Q3 1.0) and patients with ≥4 attacks recorded 0.8 ± 2.6 days with MV (R 0–15, Q1 0.0, M 0.0, Q3 0.0) and 1.5 ± 2.9 days with non-MV (R 0–13, Q1 0.0, M 0.0, Q3 2.0). The trend towards an increase of MV and non-MV in patients with frequent migraine was statistically not significant (P = 0.48 and P = 0.17, respectively).
On headache-free days, MV was reported in 1.0% and non-MV in 1.8% of days. On days with non-migrainous headache, MV was present in 1.4% and non-MV in 6.4% of days. On days with migraine, MV was present in 6.1% and non-MV in 13.1% of days. These increases in prevalence were statistically significant for MV (P < 0.001) and non-MV (P < 0.001).
Duration of vertigo and dizziness
The median duration of any episode of vertigo or dizziness was 4 h (R, 30 s to 24 h, Q1 1 h, Q3 9 h). The duration of MV and non-MV did not differ (P = 0.16), but was significantly shorter on headache-free days than on days with non-migrainous headache and migraine, respectively (Figures 1 and 2). On days with both MV and migraine, MV lasted longer than headache. On days with both non-MV and migraine, non-MV lasted shorter than headache. These differences, however, did not remain statistically significant after Bonferroni correction (Figures 1 and 2).
Duration of MV (light-grey) and headache (dark-grey) in 116 patients completing a 30-day diary. NonMig, non-migrainous; HA, headache. P-values in bold indicate statistical significance after Bonferroni correction. Duration of non-MV (light-grey) and headache (dark-grey) in 116 patients completing a 30-day diary. NonMig, non-migrainous; HA, headache. P-values in bold indicate statistical significance after Bonferroni correction.
Characteristics of vertigo and dizziness
Characteristics of MV and non-MV in 116 patients with migraine completing a 30-day diary
For calculating percentages, missing values were excluded.
P-values in bold indicate statistical significance after Bonferroni correction.
Headache and associated symptoms on days with vertigo or dizziness
On days with vertigo or dizziness, migraine was present on 109 days (52.4%), non-migrainous headache on 28 days (13.5%), and headache was absent on 71 days (34.1%). Nausea, photophobia, phonophobia and aura symptoms were recorded on 33, 68, 69, and 42 days, i.e. on 15.9%, 33.0%, 33.5%, and 20.2% of the days with vertigo or dizziness. Further details are given in Table 2.
Occurrence and precipitation of vertigo and dizziness
Patients recorded one or more situations related to the occurrence of vertigo or dizziness on 152 days (74.5%). As triggers, they recorded head movement on 86 days (45.5%), getting up on 82 days (43.4%) and other triggers on 58 days (29.6%). Occurrence during sitting or lying quietly was recorded on 24 days (12.8%). As expected, head movement was related to head motion intolerance (Pearson correlation coefficient (CC) 0.39; P < 0.001). In addition, head movement showed a weak negative correlation with light-headedness (CC –0.15; P = 0.036), which did not remain statistically significant after Bonferroni correction. Similarly, weak correlations of both getting up and occurrence during sitting or lying quietly to drowsiness (CC 0.18; P = 0.016 and CC 0.19; P = 0.009, respectively) did not remain statistically significant. Other triggers recorded by the patients were related positively to rotatory vertigo (CC 0.23; P = 0.001), and negatively to drowsiness (CC –0.26; P < 0.001). A weak correlation of other triggers and head motion intolerance (CC 0.15; P = 0.042) was statistically not significant. Further details are given in Table 2.
Impact of headache, quality of sleep, depression, and anxiety
Impact of headache, quality of sleep, depression and anxiety in 116 patients completing a 30 days diary. Comparison of patients with MV, exclusively non-MV and free of vertigo or dizziness
For calculating percentages, missing values were excluded.
P-values in bold indicate statistical significance after Bonferroni correction.
Discussion
In this diary study, we analysed more than 3000 patient-days recorded by 116 subjects with at least one migraine attack per month. Participants were recruited via articles in newspapers not stating our specific interest in vertigo and dizziness. Age and gender were typical of studies in migraine. Variety in social backgrounds, educational levels and living standards was broad. We did not include children and adolescents and, therefore, we cannot comment on the relation of vertigo and migraine in these age groups.
In our patients, the 1-month prevalence of MV was 16% and that of non-MV was 30%. These high prevalence rates reflect that vertigo and dizziness are important features in the daily medical management of migraineurs. Comparing patients with ≤3 and ≥4 migraine-days per month did not show a statistically significant difference in days with MV or non-MV. This lack of statistical significance may be explained by small patient numbers, however. The life-time prevalence found by Neuhauser et al. (3) was 9% in a specialised migraine clinic and 1% in the general population; the 12-month prevalence was 0.89% in the general population. The difference between the life-time prevalence of 9% in a specialised migraine clinic and the 1-month prevalence of 16% in our study is quite prominent and may be due to methodological differences such as use of a daily diary. The patients in our study had to pay more attention to vertigo and dizziness, as they had to record presence/absence and characteristics every night. In addition, patients were instructed how to fill in the diary during a clinic visit, whereas in other studies data were collected via telephone interviews (2,3). Performing a retrospective study in migraineurs, Vukovic et al. (23) found a life-time prevalence of MV of 23.2%. Using a neuro-otological database Cha et al. (24) reported MV in 87% of the patients, 70% of them had definite and 30% had probable MV. The high prevalence of MV in this study may be explained by the selection of patients, who had to experience at least two attacks of spontaneous rotational vertigo not exclusively triggered by head movement (24).
In accordance with previous findings (3,23), the occurrence of MV in the present study was mostly, but not necessarily, linked to migraine and head motion intolerance represented the most prominent feature of MV (2,10). This also supports results recently published by Celebisoy et al. (10) who reported head motion intolerance in 71.4% of MV patients during vertiginous episodes. Former studies on ‘migraine-related vertigo’ revealed similar results (25,26) without using generally accepted diagnostic criteria.
Duration of MV proved to be quite variable (2,3) and, therefore, has not been regarded as a diagnostic criterion (5). In contrast, vertiginous episodes in BM require a duration of 5–60 min according to ICHD-II (11). The wide range in duration of MV was confirmed in the present study by means of a diary. Duration between a few seconds and one day and median attack duration of 4 h in our study is by far beyond the duration of migraine aura. Reported attack duration of more than 24 h mainly comes from studies that did not use explicit diagnostic criteria (14,26). Interestingly, the duration of MV was significantly shorter on headache-free days.
The prevalence of headache and associated symptoms on days with MV did not differ significantly from days with non-MV. Recording associated symptoms such as photophobia or phonophobia is of great diagnostic importance as presence of these symptoms during vertigo allows to diagnose MV in the absence of migraine headache.
Our study was not designed to assess treatment of vertigo or dizziness in migraineurs. Recommendations for treatment of MV (based on the criteria proposed by Neuhauser et al. (3) include zolmitriptan studied in a randomised, placebo-controlled trial (27) for acute attacks and topiramate assessed in an open trial (28) for prophylactic treatment. In addition, drugs used in migraine prophylaxis showed a beneficial effect in MV in some case reports (12,29). Recently, precipitation or exacerbation of headache was reported with disappearance of MV after orally administered triptans (30). To sum up, it may be said that little is known about effective treatment of MV and further research is needed.
The burden of migraine is the result of reduced quality of life (31) and co-morbid conditions such as sleep disturbances (32), depression and anxiety (33,34). The question arises whether vertigo and dizziness have additional impact. Furman and colleagues (33) argued that the co-existence of migraine and vertigo exposes patients to even more psychological stress. They defined a new syndrome named ‘migraine anxiety related dizziness’ (33). Studying psychiatric co-morbidity in different organic vertigo syndromes, Eckhardt-Henn et al. (34) reported that the most evident association existed between depression or anxiety and MV. In the present study, a trend towards decreased quality of sleep and higher prevalence of depression in patients with MV failed to reach the level of statistical significance after correction for multiple testing. In contrast, the association of anxiety and MV was statistically highly significant thus supporting previous findings (34–36).
The strength of our study is the inclusion of volunteers recruited via articles in newspapers not stating our special interest in vertigo and dizziness, thus avoiding the selection bias of specialised headache or vertigo centres. For the first time, we analysed MV by means of a daily diary, which enabled us to analyse more than 3000 patient days by diagnosing each episode of headache and each episode of vertigo or dizziness according to clear diagnostic criteria.
Study limitations
Our study has several limitations. First, it was not population-based and inclusion required at least one migraine attack per month. Accordingly, our data cannot be transferred to the general population. Second, exclusion of patients with other vestibular disorders was based on medical history and the patients did not undergo clinical neurological or otological examinations. Finally, patients were not clinically investigated during episodes of vertigo or dizziness and all findings are based on the patients’ self-reports.
Conclusions
Vertigo and dizziness are frequent symptoms in migraineurs. The 1-month prevalence of MV is 16% and that of non-MV 30% in patients with at least one migraine attack per month. Frequency of MV is higher and duration longer on days with headache and MV is a risk factor for co-morbid anxiety.
Footnotes
Acknowledgements
The authors thank Drs Panteha Fathinia, Renate Massl and Ania Prajsnar for their contribution in performing the patient visits and data input and Sonja Schobesberger for editing the English text.