Efficacy and safety of robot-assisted deep brain stimulation for Parkinson’s disease: a meta-analysis

Objective This meta-analysis aims to assess the effectiveness and safety of robot-assisted deep brain stimulation (DBS) surgery for Parkinson’s disease(PD). Methods Four databases (Medline, Embase, Web of Science and CENTRAL) were searched from establishment of database to 23 March 2024, for articles studying robot-assisted DBS in patients diagnosed with PD. Meta-analyses of vector error, complication rate, levodopa-equivalent daily dose (LEDD), Unified Parkinson’s Disease Rating Scale (UPDRS), UPDRS II, UPDRS III, and UPDRS IV were performed. Results A total of 15 studies were included in this meta-analysis, comprising 732 patients with PD who received robot-assisted DBS. The pooled results revealed that the vector error was measured at 1.09 mm (95% CI: 0.87 to 1.30) in patients with Parkinson’s disease who received robot-assisted DBS. The complication rate was 0.12 (95% CI, 0.03 to 0.24). The reduction in LEDD was 422.31 mg (95% CI: 68.69 to 775.94). The improvement in UPDRS, UPDRS III, and UPDRS IV was 27.36 (95% CI: 8.57 to 46.15), 14.09 (95% CI: 4.67 to 23.52), and 3.54 (95% CI: −2.35 to 9.43), respectively. Conclusion Robot-assisted DBS is a reliable and safe approach for treating PD. Robot-assisted DBS provides enhanced accuracy in contrast to conventional frame-based stereotactic techniques. Nevertheless, further investigation is necessary to validate the advantages of robot-assisted DBS in terms of enhancing motor function and decreasing the need for antiparkinsonian medications, in comparison to traditional frame-based stereotactic techniques. Clinical trial registration: PROSPERO(CRD42024529976).

Studies have demonstrated that it was more effective than medicinal intervention in individuals with Parkinson's disease and primary motor difficulties (Burchiel et al., 2013;Sato et al., 2019;Paff et al., 2020).DBS has relied on arc-radius frame-based systems since its creation in 1949 by Leksell.These systems are considered the benchmark for achieving precise and accurate results (Rahman et al., 2009;Khan and Henderson, 2013).Enhancing operational efficiency and precision is crucial consideration in enhancing DBS procedure for movement disorders (Bari et al., 2015).By prioritizing the optimal utilization of operating room and anesthetic time, it is anticipated that patients' surgical experiences, comfort, and safety will be enhanced (Fenoy and Simpson Jr., 2014;Tolleson et al., 2014).Stereotactic precision is known to be crucial to results in movement disorders, but it can frequently be tedious, error-prone, and timeconsuming to achieve acceptable levels (Lanotte et al., 2002;Starr et al., 2004).
With the advent of cutting-edge robotic guiding systems, stereotaxy has undergone a sea change for numerous procedures, such as stereoelectroencephalography (Joseph et al., 2017;Brandmeir et al., 2018).The primary goal of surgical robots is to guarantee and improve the accuracy of a specific operation.With their precise, repeatable, and predefined pathways, robots can safely navigate around obstacles and avoid harming neighboring structures (Davies, 2000).Recent years have seen a rise in the use of robot-assisted procedures for DBS, first in Europe (Lefranc and Le Gars, 2012;Moran et al., 2020), then in Asia (Lefranc and Le Gars, 2012;Liu et al., 2020), and most recently in the US (Vadera et al., 2017;Faraji et al., 2020).Numerous facilities have begun incorporating robotic systems like ROSA (Zimmer Biomet Inc.) and neuromate (Renishaw plc) into their workflow due to the high levels of accuracy and reproducibility that these systems provide (Lefranc and Le Gars, 2012;Liu et al., 2020;Moran et al., 2020).
This study involved a meta-analysis to thoroughly evaluate the available evidence in studies regarding the effectiveness and safety of robot-assisted DBS for PD.The primary outcome was vector error, while the secondary outcomes included complication rate, LEDD, UPDRS, UPDRS II, UPDRS III, and UPDRS IV.

Search strategy
The present meta-analysis followed the 2020 guidelines established by the Preferred Reporting Project for Systematic Review and Meta-Analysis (PRISMA; Page et al., 2021).The study has been registered at PROSPERO with the registration number CRD42024529976.A comprehensive search was performed in four databases, including PubMed, Embase, Web of Science, and the Cochrane Library, to retrieve literature published up until March 23, 2024.The search technique adhered to the PICOS principle and utilized a blend of MeSH terms and unrestricted text phrases.The search strategy employed was to combine the terms "Parkinson's Disease, " "Deep Brain Stimulation" and "robot." Supplement material 1 offered a thorough summary of the search record.

Selection of studies
The procedure of selecting literature, which included eliminating duplicate entries, was carried out using EndNote (Version 20; Clarivate Analytics).Two independent reviewers conducted the first search.They removed any duplicate records, evaluated the titles and abstracts to determine their relevance, and classified each study as either included or excluded.We reached a resolution by achieving consensus.In the absence of consensus among the parties, a third reviewer assumed the position of a mediator.

Data extraction
The data was extracted by two reviewers independently.The extracted data included: (1) Basic information of the study, including the first author, publication year, country, study design, sample size, and main outcomes; (2) Baseline characteristics of study subjects, including number of patients, age, disease; (3) The data analyzed included Vector error, complication, LEDD, UPDRS, UPDRS II, UPDRS III, and UPDRS IV.In the absence of consensus between the two independent reviewers, a third reviewer assumed the position of a mediator.

Quality assessment
Two independent reviewers assessed the quality evaluation in the trials that were included.The Newcastle-Ottawa Scale (NOS; Stang, 2010) was utilized to assess the quality of retrospective cohort studies included, while the methodological index for non-randomized studies (MINORS; Slim et al., 2003) for single-arm studies.If there were any discrepancies, the disputed conclusions were resolved through collaborative discussion.

Statistical analysis
The analyses were performed using Stata 12.0.The comparison of continuous variables was performed using the weighted mean difference (WMD) and a 95% confidence interval (CI).The relative ratio (RR) was used to compare binary variables, along with a 95% CI.The medians and interquartile ranges of continuous data were converted to the mean and standard deviation.The Cochrane 'Sq test and the I 2 index were used to evaluate the statistical heterogeneity among the studies included (Cumpston et al., 2022).Considering that the papers included in the analysis are sourced from the public literature, it is generally more rational to select the random effect model as the initial choice.A p-value below 0.05 was considered to have statistical significance.

Search results
Figure 1 depicted the process of selecting and incorporating articles.We initially identified a total of 198 studies.After removing redundant articles, there were only 138 articles left.Upon evaluating the titles and abstracts, a total of 116 publications were determined to be irrelevant and thus excluded.After a comprehensive inspection of the entire text, a total of 16 articles (Delavallee et al., 2016;Lefranc et al., 2017;Ho et al., 2018Ho et al., , 2019;;VanSickle et al., 2019;Faraji et al., 2020;Jin et al., 2020;Moran et al., 2020;Paff et al., 2020;Eross and Halasz, 2021;Ribault et al., 2021;Liang et al., 2022;Mei et al., 2022;Soler-Rico et al., 2022;Hegde et al., 2023;Wu et al., 2023) were chosen for inclusion in this meta-analysis.

Patient characteristics and quality assessment
This meta-analysis comprised 15 publications, consisting of six retrospective cohort studies and nine retrospective single-arm studies.The analysis was limited to the data of individuals who underwent robot-assisted DBS for PD, and totally 732 patients were included.The Newcastle-Ottawa Scale (NOS) was utilized to assess the quality of retrospective cohort studies included, while the methodological index for non-randomized studies (MINORS) for single-arm studies.
Table 1 presents detailed data on patient characteristics and quality assessment.

Clinical outcomes
The meta-analysis results for clinical outcomes were consolidated and shown in Table 2.

Vector error
Totally 10 studies reported vector error.The pooled results revealed that the vector error was measured at 1.09 mm (95% CI, 0.87 to 1.30; Figure 2) in patients with Parkinson's disease who received robot-assisted DBS.

Complication rate
Totally 12 studies reported adverse events.The pooled results revealed that the complication rate in patients with Parkinson's disease who received robot-assisted DBS was 0.12 (95% CI, 0.03 to 0.24; Figure 3).Common complications included hemorrhage, infection and transient confusion.

LEDD
A total of eight articles recorded the reduction in LEDD.The pooled results showed that the reduction in LEDD was 422.31 mg (95% CI: 68.69 to 775.94; Figure 4) after patients with PD received robot-assisted DBS.

Discussion
Since its initial clinical description in 1995, DBS has been shown to be an effective treatment in multiple randomized controlled trials for patients with Parkinson's disease who experienced motor fluctuations.This therapy improved severe periods of reduced movement (akinesia) during off-periods and reduced troublesome involuntary movements (dyskinesia) during on-periods (Limousin et al., 1995;Krack et al., 2003).Randomized controlled trial results, with quality of life as the key outcome, have conclusively demonstrated the advantages of neurostimulation over a medically treated control group that did not get stimulation (Deuschl et al., 2006).A recent meta-analysis of five randomized controlled studies comparing STN-DBS to the best available medical treatment has proven that Forest plot of the meta-analysis for vector error.
neurostimulation is more effective than the best medical treatment (Krack et al., 2019).The study found a significant improvement of 35.4% in motor symptoms, as judged by the Unified Parkinson's Disease Rating Scale Part 3 in the absence of medication.Additionally, there was a reduction of 50.8% in off-time and a 49.1% reduction in levodopa-induced dyskinesias.Maximizing operational efficiency and Forest plot of the meta-analysis for complication rate.Forest plot of the meta-analysis for reduction in LEDD. 10.3389/fnagi.2024.1419152 Frontiers in Aging Neuroscience 07 frontiersin.orgprecision continue to be crucial elements in enhancing DBS surgery for movement disorders (Bari et al., 2015).The advancement of contemporary robotic-guidance systems has brought about a significant transformation in stereotaxy, particularly in procedures such as stereoelectroencephalography (Joseph et al., 2017;Brandmeir et al., 2018).Currently, there is significant interest in the use of robotassisted brain pacemaker surgery for the treatment of Parkinson's disease.The present meta-analysis aimed to thoroughly evaluate the available evidence in studies regarding the effectiveness and safety of robot-assisted DBS for PD.Since most of studies included were retrospective single-arm studies (Table 1), a single-arm meta-analysis limited to robot-assisted DBS was performed.

FIGURE 6
Forest plot of the meta-analysis for improvement in UPDRS III. 10.3389/fnagi.2024.1419152 Frontiers in Aging Neuroscience 08 frontiersin.orgMei et al., 2022;Chuang et al., 2023;Hegde et al., 2023;Li, 2023;Schulder et al., 2023) when DBS was performed by conventional frame-based stereotactic methods.The vector error was quantified at 1.09 mm in individuals with PD who underwent robot-assisted DBS, according to our findings.Our findings suggest that the precision of robot-assisted DBS appears to surpass that of the traditional framebased approach.Over time, the precision and dependability of metallic devices used for lead implantation might be affected by wear and deformation.Robot systems are easier to maintain than traditional frame-based systems.The frame-based approach necessitates the repeated autoclaving of the following items following each surgery: the head frame, screws, phantom base, microdriver, and targeting bow.Though the robot does not come into direct touch with the patient during surgery, only a few of components like screws, microdrivers, and instrument holders require autoclaving in the robot system.The frame-based group may have had inferior precision due to deformation, wear, and dullness caused by the mechanical components' prolonged exposure to heat and accidental bumps.The robot system may be able to keep a better level of precision and accuracy for lead implantations than the frame-based system, but only with long-term, thoughtful and dependable maintenance (Mei et al., 2022).DBS robot-assisted surgery achieves high precision by the implementation of a redesigned registration process, intraoperative registration, and simulated target verification (Xu et al., 2018;Gong et al., 2020;Jin et al., 2020).
The aggregated findings demonstrated that the improvement in UPDRS, UPDRS III, and UPDRS IV was 27.36, 14.09, and 3.54, respectively.Prior research has demonstrated that traditional framebased stereotactic procedures resulted in improvements of 12.3 to 50 in UPDRS, 3.2 to 37.9 in UPDRS III, and 3.4 to 44.4 in UPDRS IV (Weaver et al., 2009;Fasano et al., 2010;Keller, 2013;Ostrem et al., 2013;Cheng et al., 2014;Bezchlibnyk et al., 2020;Paff et al., 2020;Zhang, 2020;Holewijn et al., 2021;Ribault et al., 2021;Li, 2023).Our study indicates that the increased accuracy of robot-assisted DBS does not provide substantial advantages over traditional frame-based stereotactic procedures in terms of motor improvement.However, the conclusion drawn may not be entirely accurate due to the limited number of studies that reported findings on motor improvement (Table 2).
While the existing evidence does not disprove the benefits of robot-assisted DBS in terms of motor improvement and reduction of antiparkinsonian drugs, it still provides several advantages (Mei et al., 2022).These include increased patient comfort without the need for a heavy frame, shorter operation time, consistent and reliable positioning of the arm along a specific path, facilitating easier adjustments, and eliminating the manual setup of coordinates.Furthermore, the use of robotic approach is constrained by several restrictions, such as its exorbitant expense, sluggish acceptance and popularization, the necessity of a substantial team, and a protracted learning curve.Nevertheless, the financial strain on both doctors and patients could be alleviated by reducing the duration of surgeries and streamlining the surgical procedure.On the other hand, the framebased technology poses challenges for the patient due to its extensive processes and extended operation periods, which can result in increased costs.The advantages of our research are evident.This study is the inaugural meta-analysis to assess the effectiveness and safety of robotassisted DBS surgery for Parkinson's disease.Our findings can offer empirical medical support for the utilization of robot-assisted DBS in the management of individuals diagnosed with Parkinson's disease.However, it is indisputable that our study has certain limitations.At first, because most research were single-arm studies, a single-arm meta-analysis was undertaken particularly on robot-assisted DBS.Consequently, there was a deficiency in directly comparing robot-assisted DBS with normal frame-based stereotactic treatments.Furthermore, all of the studies included in the analysis were retrospective in nature, which increases their susceptibility to bias.In addition, several factors may cause notable differences among the included research' results.The robotic devices used to perform DBS surgery were different among included studies (Table 1).The study designs varied, with some studies being single-arm studies while others were controlled cohort studies.There were disparities in the basic characteristics of the patients, including their age, gender, LEDD, and UPDRS scores prior to the surgical procedure.The surgical techniques employed by surgeons varied across various medical centers.
In conclusion, our research findings indicate that robot-assisted DBS is a viable and secure method for treating PD.Robot-assisted DBS offers improved precision compared to traditional frame-based stereotactic procedures, which deserves further promotion in clinical application.However, the evidence gathered indicates that there are no significant benefits of robot-assisted DBS in terms of motor improvement and reduction of antiparkinsonian drugs, compared to standard frame-based stereotactic procedures.Therefore, to ensure further verification of the safety and efficacy of robot-assisted DBS, it is crucial to carry out additional multicenter, randomized controlled trials that compare robot-assisted DBS with standard frame-based stereotactic procedures.

FIGURE 1
FIGURE 1Flow chart of literature search strategies.

FIGURE 7
FIGURE 7Forest plot of the meta-analysis for improvement in UPDRS IV.

TABLE 1
Characteristics of included studies and patients.

TABLE 2
The results of the meta-analysis.