Long-term effect and reasons for switching and combining device-aided therapies in Parkinson's Disease.

Introduction: In the advanced stages of Parkinson's disease (PD), when standard drug adjustments fail to sufficiently improve patients' quality of life, device-aided therapies (DATs) such as deep brain stimulation (DBS), continuous subcutaneous apomorphine infusion (CSAI), levodopa-carbidopa intestinal gel infusion (LCIG), levodopa-carbidopa-entacapone intestinal gel infusion, or continuous subcutaneous foslevodoa-foscarbidopa infusion are beneficial in the long run. However, sometimes patients need to switch or combine DATs due to either adverse events or loss of efficacy.

Aim of study: The aim of this article was to summarise the existing data on the long-term efficacy and adverse events of DATs, and to review the data on the rationale and efficacy for switching or combining DATs in advanced PD.

State of the art: A total of 50 studies on the long-term efficacy of DBS (N = 28), LCIG (N = 12), CSAI (N = 10) and 13 studies on switching and combining DATs were included in this review. Although the DATs show a favourable long-term effect on the main motor and non-motor symptoms of PD they all feature specific adverse events that need to be considered when deciding which DAT to offer to a particular patient. Occasionally, switching or combining DATs is recommended, e.g. if the first DAT shows inadequate symptom control, or due to adverse events. The choice of the second DAT depends above all on the main problems of the first DAT being correctly recognised.

Clinical implications: DATs are a safe and long-term effective option for the treatment of advanced PD. Switching and/or combining DATs is recommended for patients in whom the first treatment option is not optimal.

Future directions: Future studies are warranted to address the unresolved issues related to long-term efficacy, side effect profile and switching and combination of DATs in multicentric studies and using advanced analytical approaches such as machine learning.