Spontaneous vertigo and headache: Endolymphatic hydrops or migraine? (Original Article).

Abstract

We undertook a study to assess whether patients who had both spontaneous vertigo and headache would respond to treatment for endolymphatic hydrops rather than treatment for migraine. We also attempted to discover if there were any individual characteristics that might predict which patients with an equivocal history and symptoms would be more likely to have either of the two conditions as opposed to the other. All patients were initially treated for endolymphatic hydrops with sodium restriction and increased water intake. Response to treatment was assessed by three tools: the Dizziness Handicap Inventory, a symptom severity scale, and a disability scale. Patients who did not respond to treatment for endolymphatic hydrops were switched to treatment for migraine. At the first follow-up period, 23 patients were available for analysis by telephone survey. According to data obtained by the three assessment tools, 14 patients (60.9%) improved after initial therapy and six (26.1%) improved following subsequent migraine treatment; the remaining three patients (13.0%) did not comply with their initial treatment regimen, and they showed no improvement. Although no statistically significant conclusions can be drawn because of the small sample size, there were trends to suggest that certain demographic, clinical, and objective-testing characteristics might predict which patients are likely to have one of these two conditions as opposed to the other. We conclude that a stepwise approach to treatment, beginning with therapy for endolymphatic hydrops, is an appropriate strategy. Also, the fact that no improvement was seen among the three patients who did not comply suggests that improvement is indeed the result of treatment rather than simply a function of the passage of time.

Introduction

Patients with vertigo often describe symptom complexes that do not fit neatly into textbook diagnostic categories. In particular, a diagnosis of endolymphatic hydrops can be uncertain in those patients who have a history of migraine, aura, and/or features of migraine during vertigo attacks. In this article, we describe our study of 46 patients with equivocal vertiginous symptoms who were initially treated for endolymphatic hydrops.

Patients and methods

During the 18-month period between Nov. 1, 1997, and April 30, 1999, our balance center evaluated 770 patients who had symptoms of vertigo. We obtained each patient's medical history (including a full headache history) and a detailed description of vertigo symptoms. Our assessment included an audiogram, balance testing, and an examination by an otolaryngologist or neurotologist. Site-of-lesion testing--either brainstem evoked response or contrast-enhanced magnetic resonance imaging of the internal auditory canals--was performed when appropriate. Laboratory studies included measurements of FTA-ABS, thyroid-stimulating hormone, and complete blood count.

Balance testing included electronystagmography, the Hallpike maneuver, water calorics, and recordings of spontaneous, positional, and post-head-shaking nystagmus. Rotation tests included assessment of phase, gain, and symmetry at six frequencies of sinusoidal rotation (from 0.01 through 0.64 Hz at 50[degrees]/sec peak velocity), high-peak-velocity sinusoids (0.32 Hz at 250[degrees]/sec peak velocity), rotational step tests, and visual-vestibular fixation and enhancement tests. Computerized oculomotility tests included assessments of multifrequency sinusoidal pursuit from 0.2 through 0.7 Hz, full-field optokinetics, semirandom saccades (both conjugate and individual eye recordings), lateral and vertical stationary gaze, and antisaccades. Among the postural control tests were observations of casual gait, the modified Fukuda stepping test, (1) and either computerized dynamic posturography or the modified Clinical Test of Sensory Integration in Balance. (2) All recordings of eye movement were made with standard electro-oculographic electrodes.

The degrees of symptom severity and disability were assessed by the use of three tools: the Dizziness Handicap Inventory (DHI), (3) a 0-to-10-point symptom severity scale (response to the question, On a scale from 0 to 10--where 0 is no dizziness or balance problems at all and 10 is the worst dizziness or balance symptoms you can possibly imagine--what number would you give yourself for the way you've been feeling lately?), and the 0-to-5-point disability scale (0 = no disability; 5 = severe, long-term disability) described by Shepard et al. (4) Each patient was counseled by the lead author (T.E.B.), who explained the differential diagnoses and the progression of treatment from therapy for endolymphatic hydrops to therapy for migraine-associated vertigo. Patients were also given written educational materials.

Based on the results of these batteries of tests, we gave 46 patients (6.0%) a differential diagnosis of endolymphatic hydrops versus migraine-associated vertigo. (5) The history and physical examination of these patients did not suggest any of the proposed etiologies for endolymphatic hydrops (e.g., allergy or autoimmune processes), nor was the audiogram suggestive of classic Meniere's disease.

We initially treated these 46 patients for endolymphatic hydrops with a strict diet that featured a low intake of sodium (1.5 to 2 g/day) and a high intake of fresh water (24 to 32 oz/day). A few patients also received diuretic drug therapy at the discretion of the referring physician. Our treatment goal was to stabilize body-fluid levels and thereby prevent fluctuations in levels of endolymphatic contents. Endolymphatic balance is presumed to help control spontaneous vertigo attacks. (6,7)

Patients whose symptoms were not controlled by treatment for endolymphatic hydrops after several months were referred to the neurology department for a migraine evaluation and possible treatment with a prophylactic migraine medication such as verapamil, propranolol, or amitriptyline. (8) Vestibular rehabilitation therapy was prescribed as an adjunctive treatment for those patients who experienced significant unsteadiness and/or motion-provoked symptoms between vertigo attacks, provided that their attacks occurred no more frequently than every 6 weeks. (9) Measurements of symptom severity and disability scores were taken several times during the course of treatment.

After treatment had been in place for a minimum of 4 months, we attempted to reach each patient by telephone to obtain DHI, symptom severity, and disability data (mean length of treatment before contact: 11 mo). Six months after each telephone contact, a second telephone survey was conducted to obtain data on symptom severity and disability. We then made comparisons of pre- and post-treatment status.

Results

First follow-up. We were successful in contacting 23 of the 46 patients during our initial telephone survey. According to their scores on the various assessment tools, 14 of the 23 patients (60.9%) experienced improvement in their condition following treatment for endolymphatic hydrops alone (EH group) (figure 1). Six patients (26.1%) had not improved with treatment for endolymphatic hydrops but did subsequently experience an alleviation of symptoms after they had been referred to a neurologist and prescribed treatment for migraine-associated vertigo (MAV group). The remaining three patients (13.0%) said they had not complied with their endolymphatic hydrops diet or followed up with their referring physician (noncompliant group). None of these three patients experienced any improvement; in fact, their dizziness worsened dramatically.

Dizziness. According to a comparison of mean DHI scores before treatment and at the first follow-up, those in both the EH and MAV groups experienced an alleviation of dizziness--reductions of 42 and 23%, respectively (table 1). Those in the noncompliant group experienced a 90% increase in mean DHI score.

Symptom severity. Similar findings were observed in all three groups in terms of symptom severity (table 1). The EH and MAV groups experienced mean decreases of 54 and 44%, respectively, in symptom severity scores, while the noncompliant group had a decrease of only 5%.

Disability. Those in both the EH and MAV groups registered substantial declines in mean disability scores (figure 2). The mean disability score was unchanged in the noncompliant group.

Second follow-up. We telephoned each of the 20 compliant patients again 6 months after each initial post-treatment survey, and we were successful in contacting 11 patients in the EH group and two in the MAV group. Mean symptom severity (table 1) and disability (figure 3) scores continued to show improvement over pretreatment levels.

The EH and MAV groups were not large enough to statistically evaluate group differences, but several trends became apparent. Notable differences were observed in terms of age, family history of migraine, history of head trauma, the duration of the vertigo attacks, and the onset of symptoms (table 2). The results of objective tests suggested trends toward group differences with respect to the presence of static nystagmus, postural control abnormalities, caloric weakness, rotation asymmetry, and the rotation phase/time constant (figure 4). There was no incidence of significant asymmetrical hearing loss in either the EH or MAV group; all audiograms revealed either normal sensitivity or bilateral sensorineural hearing loss consistent with presbyacusis.

Discussion

The results of our study suggest that for patients who have an equivocal differential diagnosis of endolymphatic hydrops versus migraine-associated vertigo, a stepwise approach to treatment is appropriate. This method entails initial treatment for endolymphatic hydrops followed by treatment for migraine-associated vertigo for those who do not respond. The fact that the three noncompliant patients in our study did not experience any improvement suggests that treatment is indeed effective and that improvement is not simply a function of the passage of time.

The trends we noted in the differences between the EH and MAV groups might serve as diagnostic guides. Although these trends could not be statistically validated because of the small initial sample and the large degree of attrition inherent in this type of study, our results do suggest that there are demographic and clinical factors that might merit further investigation as possible diagnostic predictors (table 2). For example, these trends suggest that patients with migraine-associated vertigo are younger and have much shorter vertigo attacks than those with endolymphatic hydrops. Patients with migraine-associated vertigo might also be more likely to have a history of head injury.

The results of objective testing also suggest possible predictive trends (figure 4). Patients with endolymphatic hydrops might be more likely to have spontaneous, positional, and post-head-shaking nystagmus; abnormal postural control; and caloric weakness. Those with migraine-associated vertigo might be more likely to exhibit rotation abnormalities.

Further study involving a larger group of patients is warranted. Given the small percentage of patients (6.0%) in our center who were candidates for this study and the large number lost to follow-up as the study progressed, a multi-institutional study might be required in order to recruit a sufficient number of patients. Those patients who do not comply with treatment might be used as a de facto control group.

References

(1.) Shepard NT, Shepard NP, Boismier T. Fukuda stepping test: Test performance and criteria for abnormal. Presented at the 18th biennial meeting of the Barany Society; Uppsala, Sweden; 1994.

(2.) El-Kashlan HK, Shepard NT. Asher AM, et al. Evaluation of clinical measures of equilibrium. Laryngoscope 1998:108:311-9.

(3.) Jacobson GP. Newman CW. The development of the Dizziness Handicap Inventory. Arch Otolaryngol Head Neck Surg 1990;116:424-7.

(4.) Shepard NT. Telian SA, Smith-Wheelock M. Habituation and balance retraining therapy. A retrospective review. Neurol Clin 1990;8:459-75.

(5.) Johnson GD. Medical management of migraine-related dizziness and vertigo. Laryngoscope 1998:108:1-28.

(6.) Santos PM. Hall RA, Snyder JM, et al. Diuretic and diet effect on Meniere's disease evaluated by the 1985 Committee on Hearing and Equilibrium guidelines. Otolaryngol Head Neck Surg 1993:109:680-9.

(7.) Baloh RW. Honrubia v. Clinical Neurophysiology of the Vestibular System. 2nd ed. Philadelphia: F.A. Davis, 1990:214-9.

(8.) Bikhazi P. Jackson C. Ruckenstein MJ. Efficacy of antimigrainous therapy in the treatment of migraine-associated dizziness. Am J Otol 1997:18:350-4.

(9.) Shepard NT, Telian SA. Practical Management of the Balance Disorder Patient. San Diego: Singular Publishing, 1996.

[Graph omitted]

[Graph omitted]

[Graph omitted]

[Graph omitted]

Figure 1.

Chart illustrates the number (%) of patients who experienced various
outcomes among the 23 patients who were available for assessment at the
first follow-up.


Responded to MAV treatment     n = 6  (26.1%)
Responded to EH treatment      n = 14 (60.9%)
Did not comply with treatment  n = 3  (13.0%)

Note: Table made from pie chart
Table 1.

Mean changes (%) in DHI and symptom scale scores pre- and post-treatment

                    EH group  MAV group  Noncompliant group

DHI (first            -42        -23            +90 (*)
follow-up)
Symptom severity      -54        -44             -5
(first follow-up)
Symptom severity      -47        -42
(second follow-up)

(*)Dizziness Handicap Index scores nearly doubled in the noncompliant
group.
Table 2

Demographic and clinical Characteristics of the two groups

                               EH group              MAV group
                               (n = 14)               (n = 6)

Age range (mean)          13 to 75 yr (52.8)     12 to 55 yr (30.7)
No. female (%)                10 (71.4)               5 (83.3)
No. w/family history           5 (35.7)               3 (50.0)
of migraine (%)
No. w/history of head          4 (28.6)               3 (50.0)
trauma (*) (%)
Duration of vertigo       2 to 72 hr (23.4)       1 to 12 hr (5.3)
attack (mean)
Time from first attack  3 mo to 20 yr (5.5 yr)  2 mo to 5 yr (20 mo)
to DDx (mean)

(*)Includes barotrauma and cranial surgery.