OBJECTIVE: To estimate the real-world, long-term cardiovascular safety of lisdexamfetamine dimesylate (LDX) in new adult LDX users vs remote users of other ADHD treatments.
METHODS: This noninterventional, population-based cohort study used Danish (2013–2017) and Swedish (2013–2018) national electronic health registers. The LDX cohort included adults initiating LDX with data available within 12 months preceding the first dispensing of LDX (index date). Remote users included adults with ≥1 non-LDX ADHD medication dispensing 6–24 months before the index date and no prescriptions within 6 months before index date. Cohorts were matched for age, sex, region, and calendar year. The primary endpoint, composite of major adverse cardiovascular events (MACE), included the first hospitalization for acute myocardial infarction or stroke and out of-hospital coronary heart disease or cerebrovascular disease death. Incidence rates (IRs) and IR ratios (IRRs) per 1000 person-years with 95% confidence intervals (CIs) of MACE were estimated using Poisson regression. Confounding by matching factors, obesity, socioeconomic factors, comorbidities, and comedications was adjusted using propensity score estimation, stratification, and trimming.
RESULTS: After matching and trimming, there were 5516/40,163 LDX users and 27,494/200,389 remote-user index dates from Denmark/Sweden. Approximately 50% were aged ≤29 years and 50% were female. In Denmark/Sweden, 52.1%/58.1% of LDX users received ≥6 prescriptions and 78.8%/71.4% had 1–2 episodes of LDX use. In Denmark, IRs of MACE per 1000 person-years (95% CI) were similar for LDX (1.63 [0.85–3.14]) and remote users (1.61 [1.28–2.01]). In Sweden, the IR (95% CI) among LDX users was 1.40 (1.09–1.79) vs 1.17 (1.00–1.38) in remote users. Adjusted MACE IRRs (95% CI) for LDX vs remote use were 1.01 (0.48–2.13) in Denmark and 1.13 (0.75–1.71) in Sweden.
CONCLUSIONS: LDX was not associated with an increased risk of MACE compared with non-LDX ADHD medication in either Denmark or Sweden.
