*5.3. Recommendations*

Future studies should be designed to capture predictors of QoL and ADL improvements after DBS, taking into account the heterogeneity of the disease, the contribution of non-motor symptoms, and the impact of axial symptoms. We recommend evaluating both the ADL and QoL of candidates for DBS by means of validated scales (e.g., UPDRS part-II and PDQ-39 or PDQ-8) and carefully discuss with patients the disease burden and their determinants. After surgery, these scales can inform more than the in-clinic motor assessment about the clinical status of the patient and the impact of stimulation on their functioning in daily life, guiding possible changes in stimulation parameters, medical therapy or non-medical interventions, such as psychological support, physiotherapy, and emotional and social stimulation.

### **6. The Complexity of PD Spectrum Integrated into the Selection Process: Relevance of Non-Motor Symptoms**

### *6.1. The Standard Rule*

Despite the fact that DBS was developed to treat motor symptoms, the growing relevance given to non-motor symptoms (NMS) in the last 15 years fostered investigations into the effect of DBS for these features as well [65,66]. There are no indications nor clues so far on how to consider the presence and burden of non-motor symptoms in PD candidates for DBS.

### *6.2. Pros and Cons*

A few studies have demonstrated the improvement of di fferent NMS (cardiovascular, sleep/fatigue, perceptual problems/hallucinations, gastrointestinal, urinary, and miscellaneous domains) six months after surgery [65], which were maintained at 24 months for the sleep/fatigue, urinary and miscellaneous domains [66], and at 36 months for the sleep domain [67]. These findings were confirmed in a small cohort of young onset PD patients, for whom STN-DBS provided sustained improvement of the sleep domain of the Non-Motor Symptoms Scale and Parkinson's disease sleep scale-2 up to 24 months and correlated to the decrease in dopamine-agonist medication [68].

Remarkably, change in NMS frequency and severity after STN-DBS is strongly correlated to the improvement in QoL both in uncontrolled [66] and controlled studies [67,69] performed in the same STN-treated cohort at di fferent follow-up.

In the attempt to define profiles for the best DBS candidates which may encompass the complexity of PD clinical spectrum and its heterogeneity, a new data-driven approach to PD, supported by biomarkers and neuropathology, disclosed three di fferent PD subtypes: mild-motor predominant, intermediate, and di ffuse malignant [70,71]. These three groups, defined based on the progression of disability and mortality, di ffer in the presentation of motor and non-motor symptoms at onset, in particular for the contribution of three types of NMS: cognitive impairment, rapid eye movement sleep behavior disorder, and dysautonomia. When the same subtyping criteria were applied to a cohort of STN-DBS patients at the time of the surgical selection, the mild phenotype seem to perform better on ADL independence at the short and long-term follow-up compared to the malignant phenotype, despite similar e fficacy of stimulation on motor symptoms, fluctuations, and ambulatory capacity [72].
