Botulinum toxin use in patients with post-stroke spasticity: a nationwide retrospective study from France

Background: Current guidelines recommend intramuscular botulinum toxin type A (BoNT-A) injection as first-line treatment for spasticity, a frequent and impairing feature of various central nervous system (CNS) lesions such as stroke. Patients with spasticity commonly require BoNT-A injections once every 3 to 4 months. We conducted a nationwide, population-based, retrospective cohort study, using the French National Hospital Discharge Database (PMSI), to describe BoNT-A use for spasticity in clinical practice in France between 2014 and 2020. The PMSI database covers the whole French population, corresponding to over 66 million persons.

Methods: We first searched the PMSI database for healthcare facility discharge of patients who received BoNT-A injections between 2014 and 2020, corresponding to the first set. For each BoNT-A-treated patient, we identified the medical condition for which BoNT-A may have been indicated. Another search of the PMSI database focused on patients admitted for acute stroke between 2014 and 2016 and their spasticity-related care pathway (second set). Overall, two subpopulations were analysed: 138,481 patients who received BoNT-A injections between 2014 and 2020, and 318,025 patients who survived a stroke event between 2014 and 2016 and were followed up until 2020.

Results: Among the 138,481 BoNT-A-treated patients, 53.5% received only one or two BoNT-A injections. Most of these patients (N = 85,900; 62.0%) received BoNT-A because they had CNS lesions. The number of patients with CNS lesions who received ≥1 BoNT-A injection increased by a mean of 7.5% per year from 2014 to 2019, but decreased by 0.2% between 2019 and 2020, corresponding to the COVID-19 outbreak. In stroke survivors (N = 318,025), 10.7% were coded with post-stroke spasticity, 2.3% received ≥1 BoNT-A injection between 2014 and 2020, and only 0.8% received ≥3 injections within the 12 months following BoNT-A treatment initiation, i.e., once every 3 to 4 months.

Conclusion: Our analysis of the exhaustive PMSI database showed a suboptimal implementation of BoNT-A treatment recommendations in France. BoNT-A treatment initiation and re-administration are low, particularly in patients with post-stroke spasticity. Further investigations may help explain this observation, and may target specific actions to improve spasticity-related care pathway.