The 5 Pillars in Tourette Syndrome Deep Brain Stimulation Patient Selection

Abstract

The selection of patients with Tourette syndrome (TS) for deep brain stimulation (DBS) surgery rests on 5 fundamental pillars. However, the operationalization of the multidisciplinary screening process to evaluate these pillars remains highly diverse, especially across sites. High tic severity and tic-related impact on quality of life (first 2 pillars) require confirmation from objective, validated measures, but malignant features of TS should per se suffice to fulfill this pillar. Failure of behavioral and pharmacologic therapies (third pillar) should be assessed taking into account refractoriness through objective and subjective measures supporting lack of efficacy of all interventions of proven efficacy, as well as true lack of tolerability, adherence, or access. Educational interventions and use of remote delivery formats (for behavioral therapies) play a role in preventing misjudgment of treatment failure. Stability of comorbid psychiatric disorders for 6 months (fourth pillar) is needed to confirm the predominant impact of tics on quality of life, to prevent pseudo-refractoriness, and to maximize the future DBS response. The 18-year age limit (fifth pillar) is currently under reappraisal, considering the potential impact of severe tics in adolescence and the predictive effect of tic severity in childhood on tic severity when transitioning into adulthood. Future advances should aim at a consensus-based definition of failure of specific, noninvasive treatment strategies for tics and of the minimum clinical observation period before considering DBS treatment, the stability of behavioral comorbidities, and the use of a prospective international registry data to identify predictors of positive response to DBS, especially in younger patients.

Deep brain stimulation (DBS) for the treatment of tics in Tourette syndrome (TS) has been rapidly expanding worldwide, with several hundred cases deposited into the publicly available Tourette Association of America (TAA) DBS International Database and Registry (IDR; tourettedeepbrainstimulationregistry.ese.ufhealth.org/). Despite methodological limitations and lack of consensus on the optimal target, randomized controlled trials reported benefits of DBS of ventral globus pallidus internus and centromedian thalamic regions for the treatment of tics.^1–4 Open-label data from the IDR indicate a 45% decrease of tic severity at 1 year following DBS implantation, with similar effect from thalamic and pallidal targets.⁵

A set of selection criteria for TS DBS was previously proposed in 2015 (figure 1),⁶ most of which supported by American Academy of Neurology 2019 treatment recommendations.¹ This selection criteria relied on multidisciplinary screening and the development of clinical judgment, which is based on an individual risk/benefit ratio. Current clinical experience suggests that this selection is grounded on 5 pillars, evaluated through multidisciplinary assessment: tic severity, quality of life, failure of noninvasive treatments, behavioral comorbidities, and age.

Figure 1. Summary of the 2015 Patient Selection Criteria for Deep Brain Stimulation Treatment for Tics in Tourette Syndrome.

This article critically reviews the criteria currently applied to each of the 5 pillars of candidate selection for TS DBS. We examined content and applicability of these criteria and knowledge gaps that should be addressed to harmonize and reach consensus on the screening process across multiple centers. Finally, we propose a practical decision-making process for TS DBS based on multidisciplinary screening and risk-benefit analysis and a process that is heavily weighted on the 5 pillars.

Critical Appraisal of the 5 Pillars for Candidate Selection

The collaboration on this critical review and proposed screening process was initiated during an expert panel meeting organized by TAA in New York City in January 2020 to discuss important advances in the field, based on the TAA IDR for TS DBS. The operationalization of a selection process of suitable candidates for DBS in TS was identified as a priority objective. Interested meeting panelists formed the author group. All authors who were also clinicians conducted, for at least one of the agreed-upon selection pillars, a review of the literature to formulate the defining existing criteria for their pillar and to highlight existing knowledge gaps. To achieve consensus, the content for each pillar was reviewed by the whole team in a teleconference meeting and through iterative interventions to the text using a shared document system. Any disagreement on a pillar's content or proposed criteria was resolved through discussion and consensus.

Tic Severity

When considering the escalation of therapy from behavioral and pharmacologic interventions to surgery, it is crucial to establish that the potential benefits on tic severity would outweigh the risk of undergoing an invasive surgical procedure. Selecting a strict threshold for tic severity as a prerequisite for DBS has notable challenges. It has been suggested that an expert examine a prospective DBS candidate, consider a video-recording of tics, and use standardized scales to follow severity pre- and post-DBS.⁶ The Yale Global Tic Severity Scale (YGTSS) is the standard measure most experts use for capturing tic severity and impact,⁷ and this scale has shown validity, including responsiveness to change and widespread familiarity among practitioners. The scale assesses motor and phonic tics separately, assigning a score of 0–5 for each of 5 domains of tic severity—number, frequency, intensity, complexity, and interference—with a maximal total tic severity score of 50; a separate item rates tic-related impairment, that is, disruption of well-being, also within a 0–50 score range. A total tic severity score of 35/50 or more has been proposed as a general minimum criterion for DBS consideration. This recommendation has been consistent across multiple expert consensus groups (figure 2).⁶ A score of ≥35 is also commonly accepted as marked severity, generally affecting everyday life.

Figure 2. Algorithm for the Assessment of Pillars 1 and 2 of Candidate Selection for Deep Brain Stimulation (DBS) Surgery for the Treatment of Tics in Patients With a Confirmed Diagnosis of Tourette Syndrome (TS).

The members of the multidisciplinary team involved in the assessment are indicated in the figure.

However, beyond a minimum qualifying score, there may be additional modifiers of severity to be considered. In particular, self-injurious tics such as whiplash or head-snapping tics may impart an exceptional burden and excessive risk for injury, including, amongst others, cervical spinal cord injury, retinal detachment, bone fractures, and burns.^8,9 In such cases in which tics consistently produce a risk of bodily harm, it may be more dangerous to withhold DBS than to offer it, and therefore, offering DBS based on such malignant features “may be ethically and medically appropriate.”¹⁰ A provisional definition for these cases of malignant TS would suggest that the symptom complex should lead to 2 or more emergency department visits or at least 1 hospitalization associated with self-injury.⁸ A single-center observational study reported that approximately 5% of patients with TS in a tertiary care setting would exhibit a severe and debilitating form of TS.⁸ Importantly, such cases highlight aspects of severity that may not be captured by standard rating instruments like the YGTSS.

In addition, tic severity is best appreciated across multiple time points. The YGTSS typically characterizes the last week of experience; however, tics commonly wax and wane over time and across environments and situations. Different guidelines have proposed that severity criteria be met for 6–12 months before screening for potential DBS therapy.¹⁰ Establishing this persistence in marked severity may require sequential longitudinal examinations, record review, and careful history. This approach may also allow assessing for symptom consistency and for the presence of coexisting functional tic-like behaviors.¹¹

Finally, in assessing the burden of tic severity holistically, it is essential to supplement scale data with an assessment of personal impact from tics. Interference with academic or professional affairs, personal relationships, and declining self-esteem are variables that should be explored to appreciate the gravity of tic burden fully. However, existing guidelines do not provide sufficient indications for the use of measurement instruments to quantify functional impairment.

Quality of Life

The assessment of the health-related quality of life (QoL) in individuals with TS is complex, as most patients have 1 or more comorbid psychiatric disorders (see pillar 4), which can greatly impair function and enjoyment of life. Before contemplating an invasive procedure for tics, the clinician must determine to what extent tics vs other psychiatric symptoms impair QoL. To truly understand the impact of tics on QoL, the use of disease-specific QoL tools, such as the YGTSS impairment score⁷ and the Gilles de la Tourette Syndrome Quality of Life (GTS-QOL) Scale,¹² may yield a more accurate understanding of how tics affect physical and psychosocial aspects of health (figure 2).

QoL studies in TS using generic QoL measures have demonstrated that tic severity is not always the primary determinant of QoL in children or adults.¹³ Studies in children have revealed that symptom severity of attention-deficit hyperactivity disorder (ADHD), obsessive-compulsive disorder (OCD), and anxiety worsen substantially psychosocial QoL when measured with generic instruments and that TS or comorbidities minimally affect the physical QoL domains.^14,15 In adults, depression severity has the greatest impact on QoL as measured on the EQ-Visual Analogue Scale, followed by anxiety and obsessive-compulsive symptom severity.¹⁶ It may be useful to complete assessments of QoL with disease-specific QoL instruments such as the GTS-QOL to understand the specific impact of tics and generic QoL instruments to understand the impact of psychiatric comorbidities. If a patient's score falls in the normal range on the GTS-QOL, but there is evidence of impairment on generic QoL instruments, it is likely that the treatment of comorbidity, rather than the treatment of tics, should be the focus for QoL enhancement.

Controlled trials of DBS for TS have assessed the impact of treatment on QoL as a secondary outcome, generally yielding results that support the use of disease-specific instruments. In a DBS trial of the centromedian thalamic region in 5 adults, Okun et al. demonstrated a small but significant improvement on the YGTSS Impairment score from baseline to 6 months of scheduled stimulation (mean ± SD change −11.3 ± 5.0; p = 0.007).¹⁷ However, no significant difference from baseline was demonstrated on 2 generic QoL instruments, the 36-item Short-Form Health Survey Quality of Life Assessment and the Quality of Life Assessment Schedule. In a DBS trial of the globus pallidus internus in 15 adults, significant improvements in QoL were demonstrated with the GTS-QOL and YGTSS Impairment score when comparing baseline to open-label stimulation.³ Overall, it is important for practitioners to appreciate that, in some patients, the decrease of tic severity produced by DBS may not translate into a large improvement of QoL scores.

Failure of Noninvasive Treatments

The 2015 proposed criteria from the TAA DBS Registry and Database suggested that consideration for DBS should be limited to patients with TS who are resistant to at least 3 drug classes (an alpha-adrenergic agonist, a typical and an atypical dopamine antagonist, and 1 from another class, e.g., clonazepam, topiramate, and tetrabenazine).⁶ Potential candidates should be offered comprehensive behavioral intervention for tics by a trained therapist, and there should be documentation of treatment adherence (figure 3). However, what constitutes a failure to each pharmacologic or behavioral therapy remains undefined. Clinical judgment is frequently used and can be highly variable across TS specialists.¹⁸ Practitioners should therefore consider multiple aspects that, individually or combined, may lead to asserting treatment failure, that is, poor tolerability, insufficient efficacy (refractoriness), and insufficient adherence.

Figure 3. Algorithm for the Assessment of Pillar 3 of Candidate Selection for Deep Brain Stimulation (DBS) Surgery for the Treatment of Tics in Patients With a Confirmed Diagnosis of Tourette Syndrome (TS).

The members of the multidisciplinary team involved in the assessment are indicated in the figure. The steps required to verify whether this pillar is fulfilled should also take into account potential limitations of the access to behavioral and pharmacologic treatments due to lack of financial coverage. The inclusion of a behavioral therapist in the multidisciplinary team for the screening, treatment, and postoperative monitoring of patients with TS would optimize the access to comprehensive behavioral interventions for tics. Failure to respond to any given medication includes the possibility that the medication is contraindicated in that patient.

Medications represent the least tolerated noninvasive treatment category for tics. Discontinuation secondary to objective, intolerable adverse effects and contraindication to a specific agent represent insurmountable reasons for specific medications' failure. Moreover, the metabolic side effect profile of the antipsychotic medications most frequently used to treat tics and the recent population-based report of higher risk for obesity, type 2 diabetes, and circulatory system diseases in individuals with TS or chronic tic disorders¹⁹ suggest that DBS might represent a safer long-term option for the subgroup of patients with TS with comorbid metabolic and cardiovascular disorders who do not respond to other medical therapies. Regular, close safety monitoring of antipsychotic treatment allows the early identification of patients with TS for whom cardiometabolic risks outweigh therapeutic benefits and should lead to drug discontinuation. Moreover, a trial of medications with FDA pregnancy category D designation (e.g., benzodiazepines or topiramate) in women of childbearing potential does not seem justified to meet the noninvasive treatment failure criterion of eligibility for DBS. At the same time, there may be instances in which medications are discontinued in patients with TS due to misinterpretation of milder side effects that could be minimized by small dose changes or lifestyle adjustments (e.g., dietary changes in the presence of mild-moderate weight gain) or to a nocebo effect that is potentially preventable through more effective patient education. We strongly believe that poor tolerability to drugs should be confirmed only if appropriate education on adverse effects, adequate agent selection based on the patient's medical profile, and active strategies to counteract specific adverse effects have been provided during the course of pharmacologic treatment. A proportion of patients may also have limited access to coverage for medications for tics, as there are only 3 currently Food and Drug Administration–approved medications for the treatment of tics (haloperidol, pimozide, and aripiprazole), with several medications frequently prescribed off label, for example, clonidine and guanfacine.

While typically not generating adverse effects, behavioral therapies for tics have been hindered over the past 10–15 years by limitation of access due to the absence of qualified therapists in the patient's geographical area or by limited financial coverage. However, in recent years, important progress has been made on the development and validation of remote delivery formats supported by telemedicine, promising in terms of cost-effectiveness, as well as integrated by online therapeutic modules.^20,21 Hence, the judgment of failure of behavioral interventions for tics could consider whether these alternative formats have been explored in cases in which access is the main limiting issue. Poor adherence and compliance to a full treatment course may also be responsible for failure of behavioral therapies for tics. Behavioral therapists and physicians should ideally cooperate within a multidisciplinary team to address compliance difficulties and to provide support for families and patients. This process can identify and possibly integrate individualized care solutions. In addition, behavioral interventions like the Comprehensive Behavioral Intervention for Tics (CBIT) may not be appropriate for patients with a severe form of TS who manifest with comorbid learning disabilities, autistic features, or stable but severe ADHD. In such cases, DBS could still be considered an option if a qualified behavioral therapist on the multidisciplinary team confirms a lack of patient's ability to engage in the behavioral treatment protocol.

A clinically meaningful improvement has been proposed for YGTSS as a 25% decrease on the YGTSS Tic Severity score, which has been considered an adequate reference to define insufficient efficacy.²² It would be sensible to consider ineffective also any change that does not bring the YGTSS impairment score into the moderate degree or below, given that all degrees above moderate indicate substantial difficulties in multiple domains of daily living. This insufficient change of impairment should include the persistence of disabling or malignant tics. Additional elements to consider are whether the highest tolerated and permitted medication dose had been used and whether the treatment duration was adequate to demonstrate change. Considering the known fluctuations of tic severity, it would make sense for the duration to fall within a range of 4–12 weeks. Medications should be tailored to the individual patient's baseline fluctuation pattern. This critical component, however, should lead to discussion and consensus agreement among expert clinicians.

It remains unclear whether, in order to declare resistance to pharmacologic treatment, it is necessary to try all drug classes that are not contraindicated in a specific patient and include agents for which placebo-controlled or active-comparator trials support efficacy to treat tics. Some agents may be helpful but lack adequate supporting data. For example, the use of fluphenazine is supported by open-label observations and clinical experience, which largely suggest good efficacy and superior tolerability when compared with other first-generation antipsychotics.²³ Similarly, open-label evidence is available for tetrabenazine,²⁴ although recent randomized controlled trials of newer selective vesicle membrane–associated transporter 2 inhibitors^24,25—deutetrabenazine (Coffey et al., personal communication) and valbenazine²⁶—did not demonstrate superiority over placebo in decreasing tic severity. Cannabis-based medications, for example, δ-9-tetrahydrocannabinol, have shown to have possible efficacy in small studies and might be considered as an alternative class before declaring failure of pharmacotherapy,^27,28 especially in adults who already use cannabis efficiently as self-medication (and when use is in agreement with regional legislation; see also tourette.org/research-medical/medical-marijuana-research/for more details). Moreover, there is emerging evidence for new pharmacologic treatments, including an endocannabinoid modulator (ABX-1431),²⁹ taurine,³⁰ and vitamin D,³¹ all of which will need to be tested in more rigorous and larger studies before including them in the eligibility screening for DBS. Finally, although there is no direct evidence supporting a greater therapeutic value in combining behavioral and pharmacologic treatments compared with use in isolation, it would be sensible and feasible to apply this strategy before concluding that the threshold for treatment refractoriness has been reached.

Behavioral Comorbidities

Comorbid psychiatric disorders occur in approximately 90% of all patients with TS.¹ ADHD, OCD, anxiety, and mood disorders are the most common. These comorbidities are associated with a significant impact on QoL and can also be more impairing than motor and vocal tics. It was recommended that psychiatric comorbidities be stabilized for 6 months before DBS surgery (figure 4).⁶ Untreated or poorly controlled psychiatric comorbidities have been associated with lower quality of life scores regardless of DBS status.³² Indeed, OCD is strongly associated with potentially harmful, hospitalization-requiring, or life-threatening tics and has been identified as a common behavioral comorbidity in the context of malignant TS.⁸ Moreover, poorly controlled psychiatric comorbidities can exert a detrimental effect on the ability to control tics, thereby generating an illusion of refractoriness of tics to the medical or surgical treatment. ADHD is associated with behavioral disturbances and may contribute to decreased efficacy of TS therapies, such as comprehensive behavioral intervention for tics.³³ DBS in the setting of undertreated comorbidities might, therefore, not result in the expected reduction of tics. A recent cohort study of the Swedish National Patient Register has highlighted the increased risk of dying by, or attempting, suicide in patients with TS and chronic tic disorders.³⁴ Whereas this risk may be linked to psychiatric comorbidities, the significance of the increased odds of these suicidal behaviors persisted after adjusting for preexisting psychiatric comorbidities, suggesting the possibility that malignant TS might per se contribute to greater suicidality.³⁴ In those cases in which multidisciplinary assessment concludes for a potentially relevant contribution of tics to suicidal behaviors, the recommendation to observe stable psychiatric symptoms for 6 months before DBS might not be justified and even lead to increased risk of self-harm.³⁵ Overall, these considerations stress the importance of a thorough, multidisciplinary assessment of comorbid psychiatric problems in screening patients with TS for DBS. However, the optimal level of control of these comorbid disorders and the ideal treatment trials remain unclear and will require further data and consensus.

Figure 4. Algorithm for the Assessment of Pillars 4 and 5 of Candidate Selection for Deep Brain Stimulation (DBS) Surgery for the Treatment of Tics in Patients With a Confirmed Diagnosis of Tourette Syndrome (TS).

The members of the multidisciplinary team involved in the assessment are indicated in the figure. *A potential exception could be represented by unique cases in which there is suicidal risk that is highly linked to the severity and impact of tics.

In addition to the potential relationship between tic severity and the severity of behavioral comorbidities, the effect of DBS on comorbid conditions remains controversial.^32,36,37 A systematic review and meta-analysis published in 2016 evaluated the outcomes of 156 cases of TS DBS. The authors calculated a median reduction of 31.2% on the Yale-Brown Obsessive-Compulsive Scale and a median reduction of 38.9% on the Beck Depression Inventory.³⁸ This significant reduction in OCD and depression scores did not vary by DBS target, and the majority of the cases were implanted in the thalamus (n = 78) or the globus pallidus internus (n = 64). Earlier reports favored the use of the nucleus accumbens–anterior limb of the internal capsule (NA-ALIC) for the dual management of OCD and tics, given the role of this target in DBS treatment for OCD.³⁹ More recently, the NA-ALIC target has become less common in the treatment of TS, presumably given its lower performance in tic control compared with other targets. There are, however, not enough data available to properly judge the NA-ALIC target in TS DBS. Finally, some psychiatric problems might hinder post-DBS programming and threaten the integrity of the hardware. For example, infection risk is higher in patients with self-injurious behavior and in those who may inappropriately (and repeatedly) touch, poke, or rotate the implanted devices.⁴⁰

In summary, comorbid psychiatric disorders can affect and be affected by DBS. Assessment for the presence of comorbidity and appropriate treatment is recommended before DBS to improve patient selection (i.e., to avoid pseudo-refractoriness), to ensure that the tics are the primary source of impairment, and to create an ideal setting to maximize DBS treatment response. There is a lack of consensus on what outcomes are considered acceptable in advance of pursuing DBS, and there is a lack of a reliable method to determine whether residual symptoms will influence tic manifestations following DBS. The effect of the different brain target regions and programming parameters on the comorbid TS symptoms is an area that will require critical reappraisal.

Age

The 2015 selection criteria for TS DBS state that severe cases presenting below age 18 years can be considered for DBS when there is involvement of a local ethics committee and a multidisciplinary screening team (figure 4).⁷ Although all 5 randomized and blinded trials of TS DBS enrolled only adults,^2,3,36,41,42 real-world data from the TAA DBS Registry and Database have revealed that DBS in younger individuals may have a positive impact. The mean age at implantation in the registry was 28 years (SD 10.47, range 11–73 years), although 24 individuals were younger than 18 years and collectively had a mean baseline YGTSS Global score of 81.5 (SD 17.5, range 39–100), indicating the presence of severe tics.⁵

Pediatric DBS is an accepted treatment for idiopathic or monogenic dystonia and can also be considered in secondary forms of dystonia. In these patients, shorter disease duration at the time of surgery has been correlated with better long-term outcomes.⁴³ However, an age-dependent response and a tolerability profile have not been identified for TS. Coulombe et al.⁴⁴ conducted a meta-analysis of DBS safety and efficacy data in 58 patients with TS who were aged 21 years or younger. Across brain targets, the mean improvement in the YGTSS Global score was 57.5% ± 24.6% over 34.2 ± 23.4 months (range 3–95 months) of treatment, with 27.6% of patients (16/58) experiencing stimulation-related side effects and 13.8% (8/58) with surgical complications including infections (3/8, 5.2%), abdominal wall hematoma, need for wound revision (2/58, 3.4%), and hardware malfunction. These findings were consistent with the known side effect profile of DBS in other indications. A key difference was the rate of infection. Devices were removed in 6 of the 58 patients (10.3%) due to marked improvement/resolution or stability of symptoms or due to infection. Consideration of safety should include assessing the risk of infections and the potential for symptom improvement and future discontinuation of DBS.

Although severe tics during childhood and adolescence can compromise social, psychological, and educational attainment,^45–47 this must be weighed against the potential harm from multiple surgeries, including battery replacements over time, the chance of limited improvement, adverse effects, or complications from DBS.^47–49 The potential for tic resolution after the teenage years in 50% of patients with TS has led to reluctance in offering DBS before adulthood.⁵⁰ However, to date, there is no reliable method to predict whether tics will persist or worsen, nor is it clear what impact earlier resolution of disability from tics may have on subsequent development and well-being. A cohort of 227 patients with TS evaluated first at a mean age of 12.4 years (range 5.3–19.8 years) and 4–8 years later at a mean age of 18.5 years (range 11.1–25.9 years) showed an overall age-related decline in tic severity, although 22.7% of patients at follow-up still experienced a YGTSS Total Tic score ≥20 despite pharmacotherapy. In the same cohort, tic severity scores in childhood predicted tic severity in early adulthood.⁵¹

Approximately 56% of adults presenting to a tertiary clinic for their first TS evaluation reported persistent/worsening of tics since childhood without remission.⁵² In other series of adult patients, 24% had moderate/severe tics,⁵³ and 32% continued to report significant social, occupational, or educational dysfunction.⁵⁴ Finally, a population study reported the risk of suicidality among individuals with TS, demonstrating that persistence of tics beyond young adulthood and previous suicide attempts were the strongest predictors of death by suicide.³⁴ It is not yet established how or whether tic reduction through DBS intervention will mitigate these long-term risks and/or modify the disease course.

Toward an Algorithmic Decision-Making Process for Patient Selection

The prevalence of patients with TS who would be deemed eligible for DBS on the basis of the 5 pillars is probably higher than what is typically believed. In the United States, a national survey conducted in 2012 and based on parents' report estimated a prevalence for lifetime diagnosed TS of 1 in 360 youth between 6 and 17 years of age (∼138,000 children),⁵⁵ which is consistent with the results of the National Survey of Children's Health published in 2009 that reported the prevalence of a lifetime diagnosis of TS to 3 per 1,000 youth population.⁵⁶ Data from the Center for Disease Control estimate that about 37% of patients with TS exhibited moderate to severe tics across pediatric age groups.⁵⁷ Given that greater tic severity in childhood is associated with a higher risk of persistence of tics in adulthood (approximately 40–50%), this would translate into 15–20% of individuals with moderate-severe tics persisting in adulthood in the United States (20,500–25,600). Integrating the estimated number of adults with moderate-severe TS (indicated above) with a 30% conservative estimate of refractoriness to noninvasive treatments,^58,59 a highly conservative prevalence estimate of patients with TS potentially eligible for DBS in the United States would range between 6,150 and 7,680.

The estimated number of patients with TS in the United States to be considered for DBS supports the clinical importance of establishing a robust decision-making process for patient selection. Further work is needed to refine the existing selection criteria to minimize intersite procedural heterogeneity because the practical application of the proposed 5 pillars remains incompletely defined in some areas. Improving the efficiency and reproducibility of patients' evaluation during the selection process will in turn serve to optimize the success rate of DBS in TS. Moreover, standardizing the selection process will accelerate our understanding of the predictors of success of DBS therapy, which will improve further the quality of patient selection and potentially prevent unnecessary surgical and stimulation-related risks for subgroups of patients. Figure 5 presents a simple workflow providing the basic framework for patient selection, to which local, site-specific adjustments should be made depending on differences in the composition of the multidisciplinary team and geographical differences in the access to noninvasive therapies. Consideration should be given to the less tangible domains, such as patient and family expectations and to the burden of technology (especially in the case of rechargeable implantable pulse generators).

Figure 5. Summary of the Multiple Steps That Should Be Undertaken by a Multidisciplinary Team (MDT) Responsible for the Screening, Treatment, and Postoperative Monitoring of Patients With Tourette Syndrome (TS) Undergoing Deep Brain Stimulation (DBS) Surgery.

In addition to a neurologist and/or a psychiatrist and a neurosurgeon, the MDT should ideally include a nurse, a social worker, a behavioral therapist, and a neuropsychologist. The assessment of the five pillars for DBS candidacy is described in more depth as an algorithmic procedure in Figure 1. Within the workplan related to this assessment, allied health professionals would ideally complement the role of physicians: a social worker would be primarily involved in the evaluation of the impact of tics on functioning and quality of life and facilitate a clear communication of the patient's and family's expectations around surgery; a behavioral therapist would aid in defining whether an adequate trial of comprehensive behavioral interventions for tics has been offered and, if financially viable, performed; if available, a specialist DBS nurse would aid in the education phase and could follow up postoperative programming. We propose that all 5 pillars for candidacy be fulfilled before proceeding with the second phase of selection. The latter would encompass a thorough discussion of the surgical risk and expected benefits related to the preferred stimulation target and rule out any existing contraindications to surgery.

In our opinion, the proposed screening procedure should be considered a work in progress, as further refinement can address emerging limitations and knowledge gaps. The table separates the established concepts on which these pillars rest from the persistent knowledge gaps that will need to be addressed by future clinical research.

Table. Summary of Established Concepts and Questions for Future Research Related to Each of the Five Pillars of Tourette Syndrome Patient Selection for Deep Brain Stimulation.

With respect to the first pillar, a critical research objective will be the development of a comprehensive instrument to rate the functional impairment specifically related to tics, across different age groups. Existing instruments merge different functional domains (e.g., social interactions, academic or work-related performance, and family life) in cumulative indices that may lead to approximate judgments of both the presurgical state and the response to DBS. Moreover, a phenomenology-driven, consensus-based definition of malignant TS would help screening for this specific presentation of the disorder: for example, it is unclear whether certain manifestations that are not physically harmful but socially disruptive, for example, severe coprophenomena, should be included in the definition of malignant TS.

Ascertaining failure of non- (or less) invasive treatments for tics remains, to date, a challenging pillar of the DBS selection process. Figure 3 proposes an algorithm to screen for treatment failure, but this should be refined once a formal, consensus-based definition of this concept is achieved. A crucial step toward this refinement is an efficient, multidisciplinary education of patients and families on the objectives, potential side effects, and costs of noninvasive therapies. This educational intervention should be performed long before DBS becomes a possible treatment option. Indeed, the concept that a safe and accurate screening for invasive therapies requires a previous, adequate administration of the less invasive ones is especially pertinent to TS. Appropriate support from the family setting, fostered by productive interactions between physicians and families with integration of digital applications (e.g., medication reminder apps and applications for remote monitoring of behavioral treatments), could optimize adherence to pharmacologic and behavioral therapies. Tolerability of pharmacologic treatment should also be enhanced through a tailored choice of agents suitable to the patient's medical profile, alongside an accurate monitoring and active intervention to minimize side effects. Moreover, the assessment of efficacy of noninvasive therapies should be planned when treatment is initiated, selecting with the patient the most adequate treatment duration and dose titration, as these should be individualized in the management of tics. Given the potential mismatch between tic severity per se and overall impact of tics on functioning and quality of life, the expected goals of noninvasive therapies should be discussed and agreed upon together with patients and families before treatment initiation, especially with respect to the type and degree of functional improvement needed by the individual patient.

A similar challenge for clinicians is presented by the fulfillment of the behavioral comorbidity pillar. The relevance of the multidisciplinary nature of the team working on screening patients with TS for DBS eligibility is stressed by the need to minimize the risk of worsening coexisting psychiatric conditions and underestimating the functional impact of these comorbidities when compared with tics. Whether ADHD, OCD, anxiety, and mood disorders require a different management approach when they are associated with severe tics remains uncertain. This is particularly relevant to behavioral conditions like OCD, which may exhibit phenomenological differences depending on whether they are isolated or associated with tics.⁶⁰ New observational studies should inform clinicians on the level and duration of clinical stability of comorbid disorders that can be considered safe in confirming eligibility for DBS and offer new insight on the response of comorbidities to single and multiple-target DBS approaches. Until then, we suggest clinicians should cautiously apply management strategies valid for these comorbidities regardless of their association with tics and use the 6-month period of stability as a reference criteria that can be individually adjusted on the patient's history following a careful multidisciplinary assessment.

Finally, we propose an in-depth evaluation of potential risks and benefits of DBS in those patients younger than 18 years who clearly fulfill the first 4 pillars. As registry cohorts increase in size, predicting DBS response in these younger patients will likely improve. In the meantime, achieving a consensus-based operational definition of malignant TS is reasonable.

To summarize, future research to advance the quality of patient selection for DBS in TS should rest on 3 crucial objectives: (1) to achieve a consensus-based definition of refractoriness of tics to specific treatment strategies and the minimum clinical observation period before considering DBS treatment; (2) to achieve consensus on the definition of stability of behavioral comorbidities and on treatment algorithms to reach this stability before considering DBS for tics; and (3) to use prospective international registry data to identify factors that predict a positive response to DBS, especially in younger patients.

As clinical research progresses, we conclude that multidisciplinary teams responsible for screening, treating, and following up patients with TS undergoing DBS would benefit from structuring their approach using the 5 selection pillars. Adopting this systematic approach while advancing consensus on uncertain clinical aspects relevant to decision making will allow clinicians and researchers from different regions to pool cases together, refine our understanding of effectiveness and safety of DBS in TS, inform new clinical trials, consolidate an operational framework that will diminish heterogeneity of clinical practice, and potentially facilitate the upgrade of this therapeutic modality from rarely used and experimental to established in routine clinical practice with appropriate guidelines.

Glossary

ADHD

attention-deficit hyperactivity disorder

DBS

deep brain stimulation

GTS

Gilles de la Tourette Syndrome

IDR

International Database and Registry

NA-ALIC

nucleus accumbens–anterior limb of the internal capsule

OCD

obsessive-compulsive disorder

QoL

quality of life

TAA

Tourette Association of America

TS

Tourette syndrome

YGTSS

Yale Global Tic Severity Scale

Appendix. Authors

Study Funding

No targeted funding reported.

Disclosure

Dr. Davide Martino has received honoraria for lecturing from the Movement Disorders Society, Tourette Syndrome Association of America, and Dystonia Medical Research Foundation Canada; honoraria for advisory board attendance from Sunovion Pharmaceuticals Canada Inc.; research and education funding support from Dystonia Medical Research Foundation Canada, the University of Calgary, the Michael P Smith Family, the Owerko Foundation, Ipsen Corporate, the Parkinson Association of Alberta, and the Canadian Institutes for Health Research; and royalties from Springer-Verlag. Dr. Irene Malaty has participated in research funded by the Parkinson Foundation, Tourette Association, Dystonia Coalition, AbbVie, Biogen, Boston Scientific, Eli Lilly, Impax, Neuroderm, Prilenia, Revance, and Teva but has no owner interest in any pharmaceutical company. She has received travel compensation or honoraria from the Tourette Association of America, Parkinson Foundation, International Association of Parkinsonism and Related Disorders, Medscape, and Cleveland Clinic and royalties for writing a book with Robert rose publishers. Dr. Tamara M. Pringsheim has received research funding support from the Maternal Newborn Child and Youth Strategic Clinical Network of the Alberta Health Services and the Canadian Institutes for Health Research and salary support from the American Academy of Neurology. Dr. Alfonso Fasano has received honoraria from AbbVie, Abbott, Boston Scientific, Ceregate, Ipsen, Medtronic, and UCB; research support from AbbVie, Boston Scientific, and Medtronic; and research funding support from the Chair in Neuromodulation and Multidisciplinary Care at University of Toronto and University Health Network. Dr. Michael Okun serves as a consultant for the Parkinson's Foundation and has received research grants from the NIH, Parkinson's Foundation, the Michael J. Fox Foundation, the Parkinson Alliance, Smallwood Foundation, the Bachmann-Strauss Foundation, the Tourette Syndrome Association, and the UF Foundation. Dr. Okun's DBS research is supported by NIH R01 NR014852 and R01NS096008. Dr. Okun is PI of the NIH R25NS108939 Training Grant. Dr. Okun has received royalties for publications with Demos, Manson, Amazon, Smashwords, Books4Patients, Perseus, Robert Rose, Oxford, and Cambridge (movement disorders books). Dr. Okun is an associate editor for New England Journal of Medicine Journal Watch Neurology. Dr. Okun has participated in CME and educational activities on movement disorders sponsored by the Academy for Healthcare Learning, PeerView, Prime, QuantiaMD, WebMD/Medscape, Medicus, MedNet, Einstein, MedNet, Henry Stewart, American Academy of Neurology, Movement Disorders Society, and by Vanderbilt University. The institution and not Dr. Okun receives grants from Medtronic, AbbVie, Boston Scientific, Abbott, and Allergan, and the PI has no financial interest in these grants. Dr. Okun has participated as a site PI and/or co-I for several NIH, foundation, and industry-sponsored trials over the years but has not received honoraria. Research projects at the University of Florida receive device and drug donations. Drs. Deeb, Jimenez-Shahed, Ganos, and Wu report no disclosures relevant to this manuscript. Go to Neurology.org/Nhttps://n.neurology.org/lookup/doi/10.1212/WNL.0000000000011704 for full disclosures.