Treatment of medication overuse headache: Effect and predictors after 1 year-A randomized controlled trial.

BACKGROUND Combined withdrawal and early preventive medication was the most effective treatment for medication overuse headache (MOH) within the first 6 months in a previous study, but results from a longer follow-up period are lacking. OBJECTIVE (1) To measure the efficacy at 1 year of three different treatment approaches to MOH; (2) to compare withdrawal and early preventives (W+P), preventives with potential withdrawal therapy after 6 months (P+pW), and withdrawal with delayed potential preventives (W+pP); and (3) to identify predictors of chronic headache after 1 year. METHODS Patients with MOH and migraine and/or tension-type headache were randomly assigned to one of the three outpatient treatments. Headache calendar and questionnaires were filled out. Primary outcome was a reduction in headache days/month after 1 year. RESULTS Of 120 patients, 96 completed 1-year follow-up, and all were included in our analyses. Overall headache days/month were reduced from 24.6 (23.4-25.8) to 15.0 (13.0-17.0) (p < 0.0001), and only 11/96 patients (11%) relapsed. Reduction in monthly headache days was 10.3 days (95% CI: 6.7-13.9) in the W+P group, 10.8 days (95% CI: 7.6-14) in the P+pW group, and 7.9 days (95% CI: 5.1-10.7) in the W+pP group. No significant differences in treatment effect were seen between the three groups (p = 0.377). After 1 year, 39/96 (41%) had chronic headache. Predictors of chronic headache after 1 year were higher headache frequency (aOR 1.19; 1.09-1.31), more days with acute medication (aOR 1.11; 1.03-1.19), higher pain intensity (aOR 1.04; 1.01-1.08), and depression (aOR 4.7; 1.38-18.95), whereas higher self-rated health (aOR 0.61; 0.36-0.97) and high caffeine consumption (aOR 0.40; 0.16-0.96) were predictors of a lower risk of chronic headache. No adverse events were reported. CONCLUSIONS All treatment strategies proved effective in treating MOH with a low relapse rate. The W+P strategy leads to the fastest effect, confirming earlier treatment recommendations. Identification of predictors for chronic headache may help identify more complex patients.