Poster Number-35 - Association of Oral Triptans and Other Common Oral Migraine Treatments With Cardiovascular Outcomes in Patients With a History of Cardiovascular Conditions

Association of Oral Triptans and Other Common Oral Migraine Treatments With Cardiovascular Outcomes in Patients With a History of Cardiovascular Conditions

Purpose: Although US prescribing information for triptans notes history of cardiovascular (CV) disease as a contraindication for treatment, evidence is lacking on whether triptans pose greater risks of CV events in high-risk CV patients compared with other acute migraine therapies. This large, retrospective longitudinal study is the first to assess the risk of major adverse CV events (MACE) associated with oral triptans versus opioids/butalbital (O/B) and non-aspirin nonsteroidal anti-inflammatory drugs (NSAIDs) in migraine patients with preexisting CV conditions. A subgroup analysis assessing the risk of MACE associated with oral sumatriptan only versus O/B and non-aspirin NSAIDs was also conducted.

Methods: Adults treated with acute migraine treatment (ie, oral triptans, O/B, or non-aspirin NSAIDs) between 1/1/2006 and 12/31/2020 were identified from the Mass General Brigham Research Patient Data Registry database. The index date was defined as the date of each eligible prescription, and multiple prescriptions per patient were included. Baseline was defined as the 12-month period before index to assess patient characteristics and probability of baseline CV risk; ≥1 diagnosis of a baseline CV condition was required. Prescriptions with a cluster headache diagnosis at baseline or on index date were excluded; a washout period was set to exclude any prescriptions for acute migraine treatment 60 days before index. Nonfatal myocardial infarction (MI), nonfatal stroke, and in-facility all-cause mortality were assessed separately and together as a composite MACE-proxy (MACE) endpoint from index to earliest of treatment switch, 60 days post-index, end of clinical activity/data availability, or death. Inverse probability of treatment-weighted Cox proportional hazards models were used to estimate the adjusted risk of MACE comparing oral triptans with O/B and non-aspirin NSAIDs; subgroup analyses were conducted to compare only oral sumatriptan with O/B and non-aspirin NSAIDs.

Results: A total of 12,121 prescriptions were identified, including 4,016 (33%) oral triptans, 6,084 (50%) O/B, and 2,021 (17%) non-aspirin NSAIDs. The most frequent index treatments were sumatriptan (59%) in the oral triptan cohort, butalbital (39%) and oxycodone (27%) in the O/B cohort, and ibuprofen (62%) in the non-aspirin NSAIDs cohort. The mean ages at index were 49, 54, and 47 years for the oral triptan, O/B, and non-aspirin NSAIDs cohorts, respectively; 86%–87% of patients in each cohort were female. Compared with the oral triptan cohort (13%), the O/B and non-aspirin NSAIDs cohorts had higher baseline usage of low dose aspirin (24% and 20%, respectively) and the O/B cohort had higher baseline usage of antihyperlipidemics, including statins (38%) compared with non-aspirin NSAIDs and oral triptan cohorts (22% and 24%, respectively). MACE occurred with 1.0%, 4.5% and 3.8% of oral triptan, O/B, and nonaspirin NSAIDs prescriptions, respectively. The adjusted hazard ratios of MACE were 0.38 (95% CI: 0.22, 0.67) comparing oral triptans with O/B and 0.46 (95% CI: 0.29, 0.71) comparing oral triptans with non-aspirin NSAIDs. Adjusted comparisons for individual endpoints of nonfatal MI, nonfatal stroke, and in-facility all-cause mortality were underpowered to detect statistically significant differences between cohorts, although results from these analyses were consistent with those for MACE. Of the 4,016 oral triptan prescriptions, 2,368 sumatriptan prescriptions (59%) were identified. The mean age at index was 50 years in the sumatriptan cohort and 85% of patients were female. Similar to the oral triptan cohort, the sumatriptan cohort had lower baseline usage of low-dose aspirin (15%) compared with the O/B and non-aspirin NSAIDs cohorts (24% and 20%, respectively) and lower baseline usage of antihyperlipidemics, including statins (25%), compared with the O/B cohort (38%; non-aspirin NSAIDs cohort: 22%). MACE occurred with fewer oral sumatriptan prescriptions (1.1%) compared with O/B (4.5%) and non-aspirin NSAIDs (3.8%) prescriptions. The adjusted hazard ratios of MACE were 0.34 (95% CI: 0.21, 0.56) comparing oral sumatriptan with O/B and 0.47 (95% CI: 0.28, 0.79) comparing oral sumatriptan with non-aspirin NSAIDs.

Conclusion: Analysis of real-world data from clinical practice shows that oral triptans are associated with a significantly lower risk of MACE in migraine patients with a history of CV conditions compared with O/B and non-aspirin NSAIDs. A subgroup analysis including only oral sumatriptan showed similar results to the oral triptan cohort compared with the O/B and non-aspirin NSAIDs cohorts. The results suggest more deliberation may need to be given when considering migraine treatments for adults with CV conditions. These data should be considered in the context of all available treatment options for acute migraine relief. This study is subject to the typical limitations of observational research, including residual confounding, selection bias, and misclassification. Decreasing barriers to effective acute migraine therapy will help reduce the burden of migraine, which remains an undertreated and debilitating disease.

References: