MOH Treatment Strategies
(2020)Objective
To compare the effectiveness of three treatment strategies for treating medication overuse headache (MOH).
Study Summary
• All 3 treatment strategies were effective in reducing headache days per month with no significant difference among groups, but withdrawal therapy combined with preventive medication from the start of withdrawal achieved the best outcomes for reverting to episodic headache and curing MOH, and is recommended as the preferred management strategy
Intervention
Three arms: (1) Withdrawal plus preventive treatment from the start, (2) Preventive treatment alone, (3) Withdrawal alone with optional preventive treatment after 2 months.
Inclusion Criteria
Adults ≥18 years with MOH from underlying tension-type headache and/or migraine, eligible for outpatient management, and able to complete a headache calendar.
Study Design
Arms: Withdrawal + Preventive vs. Preventive alone vs. Withdrawal alone
Patients per Arm: 120 total randomized; 102 completed follow-up
Outcome
Reduction in monthly headache days: withdrawal+preventive (-12.3), preventive alone (-9.9), withdrawal alone (-8.5), p=0.20. MOH cure rates: 96.8% (withdrawal+preventive) vs. 74.3% (preventive) vs. 88.9% (withdrawal). Reversion to episodic headache: 74.2% vs. 60.0% vs. 41.7%, p=0.03.
Bottom Line
All 3 treatment strategies were effective in reducing headache days per month with no significant difference among groups, but withdrawal therapy combined with preventive medication from the start of withdrawal achieved the best outcomes for reverting to episodic headache and curing MOH, and is recommended as the preferred management strategy
Major Points
- First randomized controlled trial directly comparing the 3 debated treatment strategies for MOH in a pragmatic clinical setting
- Primary outcome showed no significant difference among groups: headache days reduced by 12.3 (withdrawal+preventive) vs 9.9 (preventive) vs 8.5 (withdrawal) days/month at 6 months (P=0.20)
- Significantly more patients reverted to episodic headache with withdrawal+preventive (74.2%) compared to preventive alone (60.0%) and withdrawal alone (41.7%), P=0.03
- Significantly higher MOH cure rates with withdrawal+preventive (96.8%) compared to preventive alone (74.3%) and withdrawal alone (88.9%), P=0.03
- Withdrawal+preventive showed 30% higher chance of MOH cure compared to preventive alone (RR 1.3, 95% CI 1.1-1.6, P=0.03)
- Withdrawal+preventive showed 80% higher chance of reverting to episodic headache compared to withdrawal alone (RR 1.8, 95% CI 1.1-2.8, P=0.03)
- Complete discontinuation of analgesics achieved by 58.1% in withdrawal+preventive group and 55.6% in withdrawal group
- Preventive group reduced medication use by ~50% by 2 months despite no formal withdrawal requirement
- Most common preventive medications used: candesartan (51.6% withdrawal+preventive, 54.3% preventive), metoprolol, amitriptyline
- 95.1% completion rate (18 dropouts from 120 enrolled) with equal distribution across groups
Study Design
- Study Type
- Prospective, longitudinal, open-label, randomized controlled trial
- Randomization
- Yes
- Blinding
- Open-label (withdrawal therapy impossible to blind); randomization conducted by independent project nurse using Sealed Envelope applications
- Sample Size
- 120
- Follow-up
- 6 months with visits at baseline, 2 months, and 6 months; telephone follow-up at 1 and 4 months
- Centers
- 1
- Countries
- Denmark
Primary Outcome
Definition: Change in headache days per month from baseline to 6 months
| Control | Intervention | HR/OR | P-value |
|---|---|---|---|
| -9.9 (95% CI 7.2-12.6) in preventive group | -12.3 (95% CI 9.3-15.3) in withdrawal+preventive group; -8.5 (95% CI 5.6-11.5) in withdrawal group | - | 0.20 (no significant difference among 3 groups) |
Limitations & Criticisms
- Open-label design with withdrawal impossible to blind, though this was the most feasible and practicable approach to address the clinical problem
- Potential unblinding and nocebo/placebo effects, though elaborate efforts made to minimize bias
- Patients in preventive group may have reduced medication use inspired by information about withdrawal or national awareness campaign about MOH in Denmark (autumn 2016)
- Study population limited to outpatient-eligible patients; excludes more complex MOH cases (significant opioid overuse, severe comorbidities, daily/almost daily barbiturate use)
- Single-center study at tertiary Danish Headache Center limits generalizability
- More than 75% had single medication overuse type (simple analgesics); less applicable to polyoveruse patients
- Only 41 of 483 screened patients (8.5%) excluded for opioid overuse, reflecting Danish prescribing patterns; less applicable to regions with high opioid prescribing
- Withdrawal group had lower preventive treatment uptake at 4 months (61.1%) compared to other groups (93.5% and 85.7%), possibly explaining lower outcomes
- No placebo group to judge nocebo and placebo effects, though both preventive and withdrawal established as superior to placebo in prior studies
- Headache days increased from 4-6 months in withdrawal group, suggesting need for better education and earlier preventive treatment initiation
- CGRP monoclonal antibodies not available at time of study
- Relatively small sample size (n=102 completers) may have limited power to detect differences in primary outcome
- 15% dropout rate, though equally distributed among groups and characteristics similar to completers
- Study funded by TrygFonden with potential industry conflicts disclosed
Citation
JAMA Neurol. 2020;77(9):1069-1078. doi:10.1001/jamaneurol.2020.1179