A Transatlantic Viewpoint on the Role of Pallidal Stimulation for Parkinson's Disease

Abstract

For many neurologists, crossing borders during training is part of the trajectory to becoming a movement disorders specialist and offers novel insights. In this viewpoint, we discuss how crossing the Atlantic Ocean, moving from Europe to North America, changed our approach to choosing a deep brain stimulation (DBS) target for the treatment of Parkinson's disease (PD). Although the debate on whether the subthalamic nucleus (STN) or the globus pallidus internus (GPi) is the preferred target for DBS is no longer a hot topic and is possibly considered a solved controversy, a new debate has recently emerged, after the publication of the European guidelines on advanced therapies for PD.1-3 Moreover, with the recent implementation of magnetic resonance imaging–guided focused ultrasound (MRgFUS),4 which can also be used to target the GPi, pallidotomy (performed using either radiofrequency or MRgFUS) may gain traction again. Therefore, we believe that sharing our view would be of interest to readers on both sides of the Atlantic. STN DBS just celebrated its 30th anniversary in Grenoble, France, where it was introduced in the early 1990s.5 Pallidotomies were the preferred stereotactic intervention for advanced PD before the introduction of DBS.6 Despite the fact that shortly after the first report of STN DBS was published,5 a series of patients treated with GPi DBS were reported,7 STN quickly became the target of choice for PD, and GPi DBS was performed primarily for dystonia.8 This was true especially in Europe, where DBS entered the clinical space prior to U.S. Food and Drug Administration approval. In the past decade, several important clinical trials comparing STN to GPi were published.9-12 Even though evidence has confirmed that both STN and GPi are effective targets in PD,13 in Europe STN is still perceived as the DBS target of choice for PD, substantially outnumbering DBS procedures targeting the GPi. Despite our European training with STN being the preferred target, since arriving and practicing in Canada, each GPi case we have cared for has made us recognize why GPi as a target has been embraced in North America. Although the majority of patients respond well to STN DBS,14 every DBS expert would concur that the potential outcome ranges between the miraculous effect often portrayed by the media, usually young tremor-dominant patients, and the disappointed patients in whom surgery and/or stimulation had a profound detrimental effect on speech,15 cognition,16 gait, and balance.17 Despite definite improvement in tremor and motor complications,18 which is generally observed, the impact of these problems on function and quality of life makes these gains less significant. In contrast, the effect of GPi DBS is in our experience somewhat more modest, especially in the longer term,10 and it rarely comes with the nearly overnight transformation seen in STN DBS. On the contrary, we believe that the postoperative phase of a GPi patient is less complicated. We noticed that programming is faster as there is no risk of dyskinesia, apathy after medication reduction, or stimulation-induced mania, just to name a few complications that occur during STN DBS programming.13 Nevertheless, stimulation-induced side effects (eg, speech disturbances) are also seen in GPi DBS patients19 although possibly less often than in STN DBS.11, 20, 21 In the trials referenced earlier,9-12 age-matched patients were randomized to STN or GPi DBS. In real-world practice, in keeping with an individualized patient-centered approach, GPi DBS is proposed for frailer elderly patients, with poorer cognition,13 to reduce the risk of adverse effects. In our practice, we have noted a vast demographic difference between STN and GPi patients with more advanced age, poorer cognition, and higher comorbidities in the latter cohort.22 By considering only STN DBS in our toolbox, we would be excluding some patients with cognitive impairment or motor frailty23 from the opportunity to benefit from DBS. In our center, we recently completed a DBS trial targeting the GPi and the nucleus basalis of Meynert in PD patients with severe cognitive impairment, all of which tolerated the procedure well.24 Similarly, there were no significant side effects when targeting the GPi in another recently published trial enrolling patients with multiple system atrophy.25 In addition, although difficult to measure, our experience indicates that fulfilling patient expectations in GPi is easier than in the STN patients. In fact, being usually older, people to whom GPi is offered may have less functional expectations and less PD-related psychosocial issues. In fact, we may be consciously or unconsciously underpromising outcomes to patients who are frailer. Further, because there is no or only minimal medication reduction after GPi DBS,13 one may argue that the target might be perceived as less effective by the patients, thus creating lower expectations and overall reducing the risk of postoperative disappointment. The decrease in dopaminergic medication in the STN cohort can in turn lead to apathy and depression, which can also contribute to patients' negative perceptions of the procedure.26 Noteworthy, the high volumes of DBS procedures in North America can be partly related to the financial incentives of its “fee per service” system. In this context, GPi can potentially be favored not only because candidate selection is less strict (ie, offered to a large proportion of patients) but also due to its easier management and more frequent generator replacement surgeries caused by the higher energy demand compared to STN DBS. Other cultural differences may influence the choice between STN and GPi as North American physicians may choose the less risky option (ie, GPi) to avoid litigation, for example, in case of adverse events. Moreover, in keeping with a more risk-averse practice, dopamine agonists may be less frequently used in North America due to perceived risk of impulse control disorder.27 The lack of a levodopa-sparing approach, and therefore the higher dose of levodopa, may be associated with an earlier onset of dyskinesias,28 once again contributing to greater adoption of DBS, and particularly of the GPi, given its efficacy against this motor complication. Due to greater profit, insurance-based healthcare policies sometimes favor unilateral (or staged) DBS procedures, which are more feasible for the GPi as opposed to the STN.12 Furthermore, unilateral GPi DBS can complement pallidotomies.4 Finally, our transatlantic view on this interesting topic also highlights the importance of the relevance of how we were trained. We both trained in Europe and were rarely exposed to GPi DBS for PD. It is well known that doing something often is the best way to guarantee good results. It is possible that the negative reputation of GPi DBS in Europe comes from the fact that physicians have limited training, exposure to GPi DBS, and therefore the opportunity to develop expertise. This may result in suboptimal management of the patients, including postsurgical programming. To conclude, in this viewpoint we would like to reconsider GPi as a target in PD, especially for patients who are prevented from having the opportunity to be considered for STN DBS a priori due to their age, frailty, or comorbid conditions. The question is not STN or GPi for everyone, as it is clear that whenever possible, STN should be preferred in younger, healthier patients as it provides an opportunity to lower the dose of dopaminergic medications. However, in those patients who are frailer, the question to be posed is whether to choose GPi DBS, levodopa-carbidopa infusion (intestinal or subcutaneous), or even radiofrequency ablation.29 Beyond the traditional comparison of motor and nonmotor outcomes in DBS, future DBS trials should investigate whether a personalized approach of available advanced treatments is effective for individual patients and—even more importantly—what the patient's (and family's) journey is after implantation. Our experience teaches us that this journey may be more challenging after STN DBS. The complexity of STN DBS also involves more visits and number of personnel. This results in increased healthcare utilization in addition to societal costs. Considering these indirect consequences30 should be the goal of future research, which will need to be based on real-life data also taking into account the implicit selection bias of these two patient populations (Table 1). We are grateful to Galit Kleiner MD, FRCPC (Baycrest Health Sciences, Katz Interprofessional Research Program in Geriatric and Dementia Care Rotman Research Institute, Baycrest Academy for Research and Education, Toronto, ON—Canada), for the insightful comments on our manuscript. Conception of the work: all authors. Acquisition of data: all authors. Data interpretation: all authors. Statistical analysis: all authors. Writing of the first draft: all authors. Critical revision and approval of the manuscript: all authors. (1) Research project: A. Conception, B. Organization, C. Execution; (2) Statistical Analysis: A. Design, B. Execution, C. Review and Critique; (3) Manuscript: A. Writing of the First Draft, B. Review and Critique. AB: 1C, 3A AF: 1A, 3B Not applicable.

Topics & Concepts