Long-Term Effects of Intervention on Vestibular Migraine: A Preliminary Study

Introduction

Vestibular migraine (VM) is one of the most common causes of episodic dizziness, ¹ with a prevalence of 4.3% to 29.3% depending on diagnostic criteria and population. ² Intervention includes medication for preventive or rescue use, with 91% of patients receiving a prescription after initial evaluation. ³ However, variability exists in medications prescribed and in treatment outcomes.^4-6 Interventions also frequently include lifestyle modifications, in conjunction with medication or in isolation.^1,7-9

Although most interventions report improvement for their participants, the follow-up time periods vary across and within investigations.^8,10 Measuring treatment effects at similar follow-up points and for longer periods of time is important to ensure benefit is maintained. The purpose of this study was to determine long-term effects of intervention types on dizziness in patients with VM. Based on prior evidence, we hypothesized that groups receiving medication plus lifestyle modifications would demonstrate the greatest reduction in dizziness symptoms over the long-term.

Methods

Participants

Following Institutional Review Board approval (#200680), we recruited patients diagnosed with VM and who were provided lifestyle modifications as part of their treatment, either in isolation or along with medication (preventive or rescue). Chart review was completed to identify English-speaking adults 18 years and older who met inclusion criteria. Forty-one patients meeting the criteria were contacted and 23 agreed to participate. Participant information is shown in Table 1.

OPEN IN VIEWER

Table 1.

Demographic, Medication, and Dizziness Handicap Inventory (DHI) Results for Each Participant and Group.

	n	Mean age/(% female)	Pre-DHI (SE)/severity category	Post-DHI (SE)/severity category	Change in DHI
All participants	23	44.9/(95.7)	47 (5.5) moderate	37 (5.9) moderate	10
Preventive + lifestyle modification	9	47.4/(100)	53 (8.5) Moderate to severe	48 (9) Moderate	5
Amitriptyline (10 mg)		57	36 moderate	30 mild	6
Amitriptyline (50 mg)		56	64 severe	92 severe	−28 a
Topiromate (25 mg)		38	20 mild	20 mild	0
Nortriptyline (10 mg) + diazepam (2 mg)		32	44 moderate	54 severe	−10
Nortriptyline (50 mg) + diazepam (2 mg)		64	48 moderate	48 mod	0
Nortriptyline (50 mg), gabapentin (100 mg)		41	30 mild	28 mild	2
Topiromate (200 mg) + diazepam (2 mg)		35	92 severe	88 severe	4
Nortriptyline (50 mg) + sumatriptan (50 mg) + diazepam (2 mg)		42	48 moderate	22 moderate	26 a
Topiromate (25 mg), venlafaxine (75 mg), galcanezumab (120 mg/ml)		62	92 severe	52 moderate	40 a
Rescue + lifestyle modification	6	47.0/(100)	51 (11.3) moderate	32 (10) mild	19 a
Diazepam (2 mg)		49	60 severe	22 mild	38 a
Diazepam (2 mg)		50	12 none	14 none	−2
Sumatriptan (50 mg)		40	36 moderate	26 mild	10
Diazepam (2 mg) + sumatriptan (50 mg)		54	38 moderate	60 severe	−22 a
Diazepam (2 mg) + sumatriptan (100 mg)		52	94 severe	74 severe	20 a
Diazepam (2 mg) + sumatriptan (100 mg)		35	56 severe	22 mild	32 a
Lifestyle modification only	8	40.8/(87.5)	38 (9.7) moderate	25 (10.8) mild	13
None		52	18 mild	8 none	10
None		49	68 severe	8 none	60 a
None		45	18 mild	16 mild	2
None		30	10 none	12 none	−2
None		27	26 mild	12 none	14
None		53	90 severe	86 severe	4
None		25	54 severe	60 severe	−6
None		45	26 mild	0 none	26 a

Standard error is shown for group average DHI data.

a

Indicates groups and participants with a ≥18 point change in DHI. DHI severity categories are from Kinney et al. ¹¹

Participants were divided into 3 groups based on their intervention: Preventive + Lifestyle Modification (PLM), Rescue + Lifestyle Modification (RLM), and Lifestyle Modification Only (LMO). Table 1 lists medications and dosages. The PLM group took daily medication. Six participants in this group only had 1 preventive medication throughout the course of the study, whereas 2 participants used more than 1 preventive medication, discontinuing 1 prior to starting the next. Four participants in the PLM group were also prescribed diazepam or sumatriptan for rescue needs. The RLM group only used diazepam and/or sumatriptan as needed. The LMO group did not take any medication for symptoms of VM. All participants were provided the lifestyle modification intervention detailed in Roberts et al. ⁹

Results

Pre- and post-intervention scores for the DHI are shown for each participant and group in Table 1. Mean pre-intervention DHI scores were moderate-severe for the PLM group and moderate for the RLM group. The mean total DHI score was also in the moderate severity category for the LMO group but was borderline moderate. An 18-point change is often used as the cut-off for significant change when total score is considered. ¹² Overall, the mean total score decreased 10 points from 47 to 37 points. The PLM group changed by only 5 points and the LMO group changed by 13 points. However, the RLM group changed by 18.6 points, indicating a significant improvement for this group and intervention using the DHI total score consideration.

Across all participants, 7 individuals had a ≥18 point improvement in DHI total score. Two were in the PLM group, 3 in the RLM group, and 2 in the LMO group. These results are expressed as percentages in Figure 1. Only 2 participants had ≥18 point worsening in symptoms. One was in the PLM group and 1 was in the RLM group; no participant had worse dizziness post-intervention in the LMO group. These results indicate the RLM group was the most successful using this criterion with 50% of participants improving by ≥18 points and only 1 participant experiencing worse dizziness.

OPEN IN VIEWER

Figure 1.

Percent of participants by group that had improved or worsened dizziness measured with the Dizziness Handicap Inventory (DHI) ≥18 point change criterion. Pre-intervention total DHI score was compared to post-intervention score obtained after ~371 days.

We also considered post-intervention change in DHI severity category, displayed in Figure 2. Across participants, 10 improved by at least 1 category post-intervention, and 2 worsened by at least 1 category. Within the PLM group and the RLM group, 3 improved and 1 worsened. For the LMO group, 4 improved and none worsened.

OPEN IN VIEWER

Figure 2.

Percent of participants by group that had improved or worsened dizziness determined by a change in Dizziness Handicap Inventory (DHI) severity category. Pre-intervention DHI severity category was compared to post-intervention severity category obtained after ~371 days.

Discussion

The purpose of this investigation was to determine the long-term (~372 days) effects of interventions including medications and lifestyle modification on dizziness handicap for participants with VM. We considered both change in DHI total score with a criterion of ≥18 points as well as change in severity category to evaluate intervention effects. Using this approach, 30.4% to 43% of participants reported improvement in dizziness long-term. The RLM intervention group performed the best with 50% of participants reporting improvement and only—approximately 17% reporting worsened dizziness. The LMO group performed next best with 25% to 50% of participants reporting improvement and no worsening in dizziness. The PLM group still had participants with improvement, but this ranged from 22% to 33% with 11% reporting worsened dizziness.

Follow-up periods of about 90 to 120 days after intervention are common.^4,5 While Celik et al ¹⁰ did follow patients for 180 to 960 days post-intervention, the follow-up period varied for each participant. Reploeg and Goebel ⁸ had a follow-up period more like the current report with a mean ~382 days. However, their range was large and included times as short as ~28 days. Comparison with other long-term data is challenging given these issues.

This is a preliminary investigation and there is an obvious limitation of number of participants per group. Still, the results are encouraging and support moving forward with an investigation of larger scale. In addition to disease-specific measures like DHI, future work should also include a generic measure of health state like the EQ-5D. ¹³ Generic measures not only allow for comparison among different diseases but a tool like the EQ-5D provides additional information on multiple dimensions including mobility, self-care, activities, pain/discomfort, and anxiety/depression. This information could prove valuable in helping to identify which individuals would be expected to perform optimally with varied interventions.

Conclusions

Positive treatment outcomes for VM-related dizziness can be measured at periods exceeding 1 year using validated tools (DHI), suggesting long-term effectiveness of the interventions. In the current report, rescue medications plus lifestyle modification produced the best results, followed by lifestyle modifications only. Preventive medication plus lifestyle modification had the third best results but it is possible these participants were more complex, given the need to trial various medications during intervention. The positive outcomes observed at such a long follow-up period coupled with the low risk of worsening symptoms is encouraging. Based on these initial data, we hypothesize that lifestyle medications are an effective solution for long-term reductions in DHI for about half of people with VM, especially when combined with rescue medication. Additional work is needed to better understand which intervention method should be used for specific patients.