Impact of NSAID and Triptan use on developing chronic migraine: results from the American Migraine Prevalence and Prevention (AMPP) study.

Objectives: To assess the influence of triptan or nonsteroidal anti-inflammatory drug (NSAID) use on the likelihood of developing chronic migraine (CM) among persons with episodic migraine (EM).

Background: CM is common in tertiary headache care, and relative to EM, CM is associated with a number of deleterious outcomes, including higher headache-related disability, reduced health-related quality of life, and increased direct and indirect costs. Symptomatic medication use has emerged as an important risk factor for the development of CM. Limited evidence based on a single year of follow up suggests that the association between NSAID and triptan use with the onset of CM varies in a dose-dependent manner that interacts with headache frequency. However, this interaction has never been explicitly studied. Herein, we evaluate results from a large-scale, 5-year, population-based observational study to characterize these relationships and test the hypothesis that NSAID use may modify the effect of triptan use on CM onset.

Methods: In the American Migraine Prevalence and Prevention (AMPP) study, 11,249 participants had EM in 2005 and provided up to 5 years of annual follow-up data. We analyzed the characteristics of persons with EM 1 year that predicted new onset CM in the subsequent year, focusing on treatment with NSAIDs and triptans as exposures. These adjacent years of data provide the basis for analysis and are termed "couplets." Repeated measures logistic regression with a subject-specific random intercept was used to model the likelihood of transition from EM to CM as a function of NSAID or triptan dose while controlling for a number of covariates including headache features, use of other medications, and the number of couplets per individual.

Results: The analysis included 9031 individuals with EM contributing up to 5 years of data and up to 4 couplets each. Results indicated that on average, 55% of the participants used NSAIDs in any given year and 2% transitioned to CM over subsequent years. Among the 20% using triptans, 3% per year transitioned to CM. Among persons with less than 10 headache days per month, frequency of NSAID use was associated with dose-dependent reductions in risk of CM onset. Among those with 10-14 headache days per month, increasing days per month of NSAID use was associated with increasing risk of CM onset. Increasing days per month of triptan use was associated with increased risk of transitioning to CM. Combination use of NSAIDs and triptans was not protective against transition to CM, but was also not statistically significantly associated with increased risk of CM onset.

Conclusion: Triptan use in EM is associated with an increased risk of CM onset that increases with days of medication use. For NSAIDs, effects depend on headache days per month. NSAIDs are protective in individuals with less than 10 headache days per month but associated with increased risk with 10 or more headache days per month. Combining a triptan and NSAID was not associated with a statistically significant increased risk of CM onset, whereas increased risk of CM onset was significantly associated with triptan monotherapy.