Introduction

Introduction of deep brain stimulation (DBS) has been a remarkable milestone in the treatment of several neurological disorders resistant to pharmacotherapy1,2,3. The technical evolution of available systems has resulted in continuously improving efficacy and safety of DBS. Among the released innovations, the NeuroSphere™ platform by Abbott allowing remote programming of the implanted DBS device seems to resolve several difficulties of postoperative care4,5,6,7,8,9,10,11. A recent study by our group has shown that among others, teleprogramming can save time and money for both patients and their caregivers, and in line with some other pilot trials, can help the maintenance of DBS care even in special situations such as a pandemic4,5,6,7,12. In addition, we also found that patients who live further from our center, are of older age or represent higher level of disease severity and disability have less frequent in-person visits at our department both during the first year and five years of stimulation treatment4,5,6,7,12. However, regular reprogramming of the DBS device that was possible only during in-person visits prior to the introduction of teleprogramming may be necessary especially after the initiation of the stimulation until the optimal settings are found and as the disease progresses to maintain the efficacy of stimulation treatment. Some recent studies suggest that remote programming, even in itself without in-person sessions, may be an effective and safe tool for finding the initial settings, the initial optimalization of the stimulation and maintaining the achieved improvement13,14. Patients who cannot have the desired number of in-person visits because of the long-distance, time-consuming, costly, and challenging travels may receive a less efficient treatment that could be achieved which can have serious clinical and economic consequences. To in-depthly investigate this thus far unstudied potential issue of postoperative DBS care, we conducted a single-center, registry-based analysis using data of patients who are being treated with DBS for Parkinson’s disease (PD) in our unit. We hypothesized that far-living patients who had more frequent in-person visits in the first year and during five years of DBS treatment showed a better outcome.

Results

Demographics and disease specific data

A total of 161 patients who had at least one-year follow-up and fulfilled all eligibility criteria were identified in our MD registry, of them, 86 subjects had also been followed for at least five years. Of the 75 patients who could not be included in the five-year analyses, 56 patients (74.7%) had a follow-up period shorter than 5 years, 4 patients (5.3%) died before the 5-year visit, and 15 patients (20.0%) were lost to follow up. Of the 15 patients who were lost to follow up, 4 patients lived close (≤ 50 km), while 11 subjects lived far (> 50 km) from our unit.

Most of the included patients (n = 128, 79.5%) were living further than 50 km from our MD center meaning their homes were located outside of Baranya county in which Pécs lies. All demographic and disease-specific data of the whole study population and different patient subgroups can be seen in Table 1. One-year and five-year follow-up visits took place between August 9, 2013 and August 12, 2022.

Table 1 Characteristics of the study population at deep brain stimulation surgery.
Full size table

Travel distance and number of in-person visits

The mean travel distance between the homes of the patients and our center was 27.5 ± 14.9 km in the group of patients who were living close to our unit, while it was 220.2 ± 106.7 km in the far-living group. There were negative correlations between travel distance from home to our center and the numbers of in-person visits both during the first-year follow-up period (Spearman’s rho = −0.364, p < 0.001) and over five years (Spearman’s rho = −0.368, p < 0.001). The portion of patients with five-year follow-up who had the desired number of in-person visits during the first year (at least 3 in-person visits) but visited our unit less frequently than considered as optimal (less than 2 in-person visits per year) from the second year to the five-year follow-up was higher in the far-living group (11/46 vs. 3/19, p = 0.041). In the far-living group, patients who visited our unit at least as frequent as considered as optimal had 3.8 ± 1.1 in-person visits in the first year, while 14.5 ± 3.2 visits in the remaining 4 years meaning that these subjects had in general 3.6 in-person visits per year throughout the whole five-year follow-up period. In the group of far-living patients who had less frequent in-person visits than desired, mean visit numbers were 1.8 ± 0.6 and 5.2 ± 0.9 for the first year and the remaining 4 years, respectively, which shows that these patients had a decrease in the number of in-person visits after the first year (generally, 1.8 visits in the first year and 1.3 visits per year in the remaining period of the follow-up). Considering the first-year and five-year data, respectively, 42% (54 subjects out of 128) and 41% (25 subjects out of 61) of the far-living patients had less in-person visits than considered optimal. Travel- and visit-related data can be found in Table 2.

Table 2 Travel- and visit-related data.
Full size table

Primary endpoint

In the primary outcome (39-item Parkinson’s Disease Questionnaire - PDQ-39 - summary index), there was no significant difference between patients living far from the center and having the desired in-person visit number and living far from the center and having less in-person visits than desired comparing scores assessed at the first-year follow-up visit (considering first-year data of all included patients: 21.8 ± 13.2 vs. 25.3 ± 15.0, p = 0.211; analyzing only data of patients who had both first-year and five-year follow-up visits: 19.5 ± 12.8 vs. 20.2 ± 11.4, p = 0.813). Of the domains of the PDQ-39, only bodily discomfort showed a significantly larger decrease in patients of subgroup 3 (−17.2 ± 16.7 vs. −7.8 ± 19.2, p < 0.001). However, the five-year global global health-related quality of life (HRQoL) was better in patients who had the desired number of in-person visits despite the long distance between their homes and our center (PDQ-39 summary index: 23.3 ± 18.5 vs. 32.2 ± 17.9, p = 0.030) and improvement in the PDQ-39 summary index was also higher in this group from baseline (preoperative status) to the five-year follow-up (−9.8 ± 18.7 vs. −1.1 ± 19.8). Significantly lower scores in the mobility, activities of daily living, stigma, and bodily discomfort domains of the PDQ-39 assessed at five years (27.9 ± 25.4 vs. 47.8 ± 29.6, p = 0.041; 23.2 ± 20.6 vs. 32.7 ± 17.2, p = 0.030; 17.8 ± 19.8 vs. 36.2 ± 20.8, p = 0.001; and 23.8 ± 16.7 vs. 34.1 ± 18.5, p = 0.036, respectively) might have contributed to the better outcome in the group of patients who visited our unit as frequently as it considered as optimal by our protocol. Improvement in the mobility, activities of daily living, stigma, communication, and bodily discomfort domains of the PDQ-39 from baseline to the five-year follow-up was larger in subgroup 3 (Table 3).

Table 3 The primary outcomes at the first-year and five-year follow-up visits in patient groups 3 and 4.
Full size table

The findings on HRQoL were also confirmed by the five-year summary index of the EuroQoL 5-Dimension 5-Level (EQ-5D-5 L, 0.802 ± 0.122 vs. 0.705 ± 0.181, p = 0.033). Improvement in the EQ-5D-5 L summary index from baseline to the five-year follow-up was significantly larger in far-living patients with regular in-person visits both after one year of stimulation and at five years (−0.118 ± 0.098 vs. −0.071 ± 0.103; p = 0.009 and − 0.107 ± 0.101 vs. −0.022 ± 0.132; p < 0.001, respectively). Of the dimensions of the EQ-5D-5 L, numbers of patients who reported better outcome on mobility and usual activities were significantly higher in group of patients having the desired number of in-person visits over five years of stimulation (p = 0.009 and p = 0.023, respectively). The five-year Shwab-England Activities of Daily Living Scale (SE-ADL) outcome was also better in patients of subgroup 3 (p = 0.021, Table 3).

Comparing data of subgroups 1 and 3, we found no significant difference in HRQoL measures (Table 4).

Table 4 The primary outcomes at the first-year and five-year follow-up visits in patient groups 1 and 3.
Full size table

Secondary endpoint

On the first-year follow-up visit, the motor and non-motor symptomatic control did not differ significantly between subgroups 3 and 4. However, at five years, far-living patients with the desired number of in-person visits reported a better outcome on the motor experiences of daily living (Movement Disorder Society-Sponsored Unified Parkinson’s Disease Rating Scale - MDS-UPDRS - Part II: 14.7 ± 8.7 points vs. 18.7 ± 8.3 points, p = 0.035). In this group of patients, compared to baseline, the improvement in MDS-UPDRS Parts II, III and IV was significantly higher (−1.7 ± 7.9 points vs. 1.0 ± 8.3 points, p < 0.001; −8.5 ± 13.5 points vs. −3.5 ± 14.8 points, p = 0.012; and − 2.5 ± 3.6 points vs. −0.8 ± 3.9 points, p = 0.049, respectively, Table 5).

Table 5 The secondary outcomes at the first-year and five-year follow-up visits in patient groups 3 and 4.
Full size table

At five years, the portions of patients on levodopa and monoamine oxidase B inhibitor treatment were higher in the far-living group of patients who visited our unit less frequently as considered as optimal, while portion of patients on dopamine agonist treatment was higher in the far-living patient group with the desired number of in-person visits over five years. There was a higher decrease in the number of patients on monotherapy from the one-year visit to the five-year follow up. The number of different antiparkinsonian medication types could be decreased in a tendentiously higher manner in far-living patients with the desired number of in-person visits and manner of decrease in the number of tablets daily taken from baseline to five years was also significantly higher in this group (−1.9 ± 2.5 vs. −0.6 ± 2.9, p = 0.039; Table 5).

Comparing data of subgroups 1 and 3, we found no significant differences in motor and non-motor symptomatic control and per oral antiparkinsonian treatment (Table 6).

Table 6 The secondary outcomes at the first-year and five-year follow-up visits in patient groups 1 and 3.
Full size table

Discussion

Remote DBS programming is an emerging innovation that seems to have several clinical and economic benefits4,5,6,7,8,9,10,11,12,13,14. In this study, we aimed to explore potential future beneficial effects of teleprogramming on the efficacy of stimulation treatment by analyzing the effects of travel distance and number of in-person visits on the one-year and five-year outcomes of DBS in PD.

According to our previous and present results, travel distance has a negative impact on the number of in-person follow-up visits both on short- and particularly long-term12, while lower number of in-person visits seems to be associated with a less favorable outcome of DBS especially on long-term. We found that PD patients who are living further than 50 km from our center and had less in-person follow-up visits than considered as optimal by our protocol before the introduction of remote DBS programming showed worse HRQoL and symptomatic improvement at the five-year follow-up compared to those far-living subjects who had regular in-person visits, while after one year of stimulation the outcome of these two groups of patients was almost comparable. In the analyzed groups, there was a difference in the numbers of in-person visits already in the first year, however, this difference became more pronounced over the remaining 4 years of the analyzed follow-up period. Long-distance travels seem to be one of the underlying causes for the decreasing number of follow-up visits among a not negligible part of DBS patients who live far from the movement disorder center (42% in the present study) over the follow-up period after the implantation. However, before the availability of remote DBS programming, travel for in-person visits was a prerequisite for optimalization of stimulation treatment which put high financial burdens on and required time from patients and their caregivers12. Patients who live closer to the center may be more likely to visit their doctors in-person to discuss any problems and before remote programming, this provided the opportunity to clinicians for checking and optimizing the stimulation. Patients who live far from the center may be more likely to contact the center on other routes (e.g., via phone and e-mail), however, without remote programming, this does not allow clinicians to optimize DBS settings. If the patient cannot attend the necessary follow-up visits for optimizing the parameters of the stimulation among others because of mobility disturbances, worsening disability resulting from both suboptimal stimulation and disease progression can become a barrier to travel12. The more severe the mobility problems, the harder it is for the patient to travel long distances. Without remote programming, if the patient cannot travel to the center, the stimulation cannot be optimized. Based on this, an assumed bidirectional relationship between travels for in-person visits and regular optimalization of DBS settings may be a factor to be considered in far-living PD patients treated with DBS.

At the one-year follow-up visit, compared to the preoperative status, DBS led to clinically important improvement in HRQoL in both analyzed groups according to the PDQ-39 SI, however, at five years, this improvement was significantly larger and remained clinically relevant only in those far-living patients who had the desired number of in-person visits over five years15. In this group of patients, the improvement in the EQ-5D-5 L summary index was also significantly larger both at the first-year and five-year follow-up visits, and considering the lowest reported minimal clinically important difference (MCID) thresholds for improvement in the summary index of the EQ-5D (from 0.03 to 0.08)16,17,18 and the MCID threshold for the Hungarian population (0.0705)19, only the improvement measured in far-living patients with regular follow-up visits exceeded these values at five years. Larger improvement in mobility and ability to perform activities of daily living (ADL) were the main contributors to the better HRQoL in the group of patients who visited our unit at least as frequently as considered as optimal. A study by Gorecka-Mazur et al. found significant correlation between ADL performance and HRQoL after subthalamic DBS, and this study also showed that DBS has the largest improving effect on the HRQoL in the first 3–6 months of stimulation which improvement remains stable during the remaining follow-up period20. These results, in line with some other findings21,22,23, highlight the importance of regular visits and on-demand optimalization of stimulation settings in the early follow-up period. In our study, the difference in number of in-person visits was smaller and there was no significant difference in global HRQoL and ADL performance between the compared far-living patient groups during the first year, however, over five years, difference in number of in-person visits increased between the far-living groups and beneficial effects of DBS on HRQoL and ADL could be maintained only in patients with regular follow-up visits. In addition, our results did not show difference in HRQoL measures between patients who live close to our center and had the desired number of in-person visits and those who live far from our unit and visited our center regularly either during short- or long-term follow-up. Based on this, regular follow-up visits and on-demand optimalization of stimulation parameters seems to be important also in maintaining the effectiveness of DBS21,22,23. Continuously maintaining an effective treatment is of high importance in several aspects of everyday living of a PD patient because among others losing ability to work seems to have irreversible negative impact on beneficial effects of DBS on the HRQoL24.

In our study, patients who visited our center regularly despite long-distance travel showed similar motor and non-motor symptomatic control than patients who had to travel shorter distances to visit our unit both during short- and long-term follow up. Of the motor symptoms, disabling dyskinesia including OFF periods with painful dystonia and choreiform hyperkinesia is one of the main indications for DBS in PD. Depending on the target of stimulation, DBS can directly or by enabling dose reduction in oral antiparkinsonian treatment, indirectly decrease the severity of dyskinesia and the impact of dyskinesia on ADL25,26. In our study, dyskinesia could be reduced in a clinically relevant manner27 in both far-living groups of patients at the first-year follow-up visit, however, this beneficial effect of DBS remained clinically meaningful only in those far-living patients who visited our center regularly. Higher number of patients on levodopa treatment which is a main contributor to the development of dyskinesia in PD and presumably suboptimal DBS settings in the group of far-living patients who did not have the desired number of in-person visits can partially explain the previous finding. In contrast, dopamine agonists could be used in a higher portion of far-living patients with regular in-person follow-up visits. Based on these results, it also seems that number of in-person visits and possibility of maintaining the optimal stimulation treatment may influence the choice of adjuvant antiparkinsonian medications which can have both clinical and economic consequences28,29. In addition to possible beneficial effect of remote programming on patient compliance to regular check-ups regarding stimulation treatment, possibility of greater simplification to per oral antiparkinsonian pharmacotherapy on long-term with regular DBS adjustments may have not just beneficial economic effects but increase the patient compliance also with pharmacological treatment. However, regarding this, it should be also noted that subtype and leading symptoms of PD may influence both adjustments to per oral antiparkinsonian medications postoperatively and DBS programming. As an example, tremor-dominant patients may need higher frequency stimulation to relive tremor, however, it has been described that high frequency stimulation may worsen gait difficulties and postural instability which is not optimal therefore for patients having such problems30. Tremor may poorly be responsive to levodopa which may necessitate the use of other antiparkinsonian medications such as dopamine agonists and anticholinergics31, however, the latter may again worsen gait difficulties and postural instability32. In addition to subtype of PD and clinical symptoms, stimulation parameters may also have an impact on per oral antiparkinsonian treatment. In our study population, distribution of PD subtypes was balanced across the subgroups.

Although this analysis is based on data collected by only our center, because our movement disorder unit is a tertiary center, we could include a relatively large number of patients from the whole area of the country. However, during the interpretation of our results, some limitations should also be considered. This study was a retrospective analysis of data from our prospective movement disorder registry and from healthcare information systems, and neither our registry nor the other used systems contain data on the exact cause of not attending our unit. Therefore, the present analysis could not exactly reveal all possible causes of and the level of their contributions to not achieving the optimal number of in-person visits. We found a negative correlation between travel distance and the number of in-person visits, and assumed that mobility disturbances and worsening disability resulting from both suboptimal stimulation in the lack of the desired number of in-person visits providing the opportunity for optimizing the stimulation and disease progression can become a barrier to travel. It can be suspected that long-distance, time-consuming and costly travel can be one of the main contributors to not attending the center among patients living far from the center. However, some other factors such as unsatisfactory outcome after DBS implantation may also contribute to not achieving the desired number of visits but one can assume that factors less related or unrelated to travel are more important for patients living closer to the center. Remote programming may be a useful tool also for patients in whom it takes longer time to find the optimal settings but lose motivation for travelling long distances because not having the desired improvement after the first few programming sessions or experiencing decreasing efficacy as the disease progresses. During the interpretation of our results, it should also be considered that clinical practice on selecting antiparkinsonian medications after initiating the stimulation treatment and protocol for DBS care may vary among centers.

According to the results of a search in the PubMed database conducted on February 12, 2025 using the terms ‘deep brain stimulation’, ‘patient selection’, ‘decision making process’, ‘effectiveness’ and ‘cost-effectiveness’, in available articles from the era before the wider introduction of DBS teleprogramming, travel distance between the residence of the DBS candidate and the clinic has not been taken into account during the selection of patients for whom DBS may be effective and cost-effective. However, according to our results, travel distance might have been an important but overlooked factor influencing long-term DBS outcomes primarily through its impact on access to in-person follow-up and stimulation optimization. Travel distance would have been therefore a not negligible factor during selecting the best PD candidates for DBS treatment. By highly decreasing the impact of travel distance on stimulation treatment, remote programming may be a feasible tool to make close- and far-living PD patients equal candidates for DBS, however, analyses of real-life data on DBS teleprogramming should confirm this hypothesis.

Materials and methods

Ethical approval

All study-related procedures were approved by the National Ethical Board of Hungary and the National Institute of Pharmacy and Nutrition (031037/2015/OTIG and OGYÉI/6916/2021). The study was performed in accordance with the Declaration of Helsinki. Our registry partly serving as a source of data analyzed in this study fulfills all security and data protection requirements and was approved by the Hungarian Medical Research Council.

Study population

From our MD registry, all subjects who fulfilled the following criteria were included: the patient (1) must have signed an informed consent form approved by the respective Ethical Committees before the inclusion in our MD registry; (2) underwent DBS implantation for medication-refractory PD; (3) had the DBS system activated at our clinic; and (4) had at least a one-year follow-up in our unit after the activation. Patients who had the one-year follow up after August 12, 2022 were excluded. Five-year outcome data of subjects who had the five-year follow up visit after August 12, 2022 was also not used. Our center started using the NeuroSphere™ platform by Abbott on August 12, 2022, and teleprogramming may have had an impact on the number of in-person visits and DBS outcome.

Sources of data

This study was based on the prospective MD registry of our department which is a tertiary MD center in Hungary. It served as the basis for identifying patients eligible for inclusion in the analyses, and it also included all data on the one-year and five-year outcomes of the stimulation treatment.

Additional data (e.g., address of the patient, number of in-person visits during the first year and five years of stimulation) were collected from the healthcare information systems used at our unit. During the analyses, we considered only those in-person visits where changes to stimulation settings (e.g., changing the lower and/or upper limits of patient amplitude control, the contact used for stimulation, the frequency or pulse width of the stimulation or initiating a completely new program) were performed.

Travel distances and times between the homes of subjects and our department were determined using the route planer called ViaMichelin (https://www.viamichelin.com). In every case, the fastest route was considered. Hatchback and E5 unleaded petrol were selected for type of car and type of fuel, respectively.

We used data of patients who signed an informed consent form prior to all study related procedures.

Primary endpoint

Our main aim was to explore the effects of travel distance and number of in-person visits during the first year and five years of stimulation treatment on HRQoL measured by different dimensions and the summary index of the PDQ-39 and the generic quality of life scale called EQ-5D-5 L at the one-year and five-year follow-up visits. To measure the level of independence in daily chores, an important contributor to HRQoL, the SE-ADL was also assessed both at the one-year and five-year follow-up.

Secondary endpoint

We also investigated the effects of travel distance and number of in-person visits during the first year and five years of DBS treatment on the motor and non-motor outcome of DBS measured by different parts of the MDS-UPDRS at the one-year and five-year follow-up visits.

Patient categorization

Patient subgroups were developed based on travel distance between the home of the patient and our center and the number of in-person-visits during the first year and five years of stimulation treatment. Because about 50 km can be generally travelled over 60 min in Hungary and our unit can also be reached within 60 min from most parts of Baranya county in which Pécs lies, one of the grouping aspects was whether the patient lives close (≤ 50 km) to or far (> 50 km) from our center. Included subjects were further categorized based on the number of in person visits during the first year and five years. Our protocol considered as optimal if the patient had at least 3 visits (3 months, 6 months, and 12 months after activation) during the first year of DBS treatment and at least 2 in-person visits per year from the second year which means that a patient who was being treated with DBS for PD should have had at least 11 visits over a 5-year follow-up period. Based on these aspects, the following groups of patients were developed for both the first-year and the five-year analyses: (1) living close to the center and having the desired in-person visit number, (2) living close to the center and having less in-person visits than desired, (3) living far from the center and having the desired in-person visit number, and (4) living far from the center and having less in-person visits than desired.

Statistical analysis

Because most of our PD patients treated with DBS live further than 50 km from our center, the large difference in the numbers of subjects living close and far did not allow us to reliably compare data of these two groups of patients. Therefore, we decided to compare DBS outcomes of far-living patients who had and missed the desired numbers of in-person visits for the first year and five years of stimulation treatment. To test normality, the Kolmogorov-Smirnov test or the Shapiro-Wilk test was used depending on sample size. Because data did not follow the normal distribution, to analyze correlation between travel distance and numbers of in-person visits during the first year and five years, Spearman’s rank correlations were used. To compare data of patient subgroups, Mann-Whitney U and Chi-squared tests were performed. The level of statistical significance was set at 0.05. All statistical analyses were performed using jamovi version 2.2.5.