Abstract
Purpose
This study presents an evolving decision-making approach for deep brain stimulation (DBS) in children with movement disorders at an Australian tertiary centre.
Methods
A retrospective review of paediatric patients referred for DBS was conducted. We present decision-making determinants and clinical details for DBS referrals. We also describe assessment frameworks before and after establishment of an interdisciplinary team (IDT) model in 2017. Motor, non-motor, and functional outcomes are presented, including patient-reported outcome measures (PROMs), patient-reported experience measures (PREMs), and a qualitative sub-study.
Results
Before the IDT was established, 4 patients underwent DBS (2 idiopathic, 1 inherited, 1 acquired dystonia). The IDT considered 44 referrals, determining 25/42 unsuitable for DBS given MRI abnormalities, unsuitable phenomenology, or parental apprehension. Twelve patients underwent DBS at our centre after discussion in the IDT (11 inherited, 1 idiopathic dystonia). Treatment was most effective for inherited dystonia and severe acute motor exacerbations (SAME) (n = 4). Younger age was not a limiting factor for considering DBS. Outcome measures showed improved motor symptoms, quality of life, and satisfaction.
Conclusion
We propose an interdisciplinary framework for collaborative decision making for DBS in children including family expectations. Outcome evaluation should be multifaceted, including PROMs and PREMs that reflect patient priorities.
Deep brain stimulation (DBS) is a potentially life-changing intervention for individuals with disabling movement disorders. It has been described to be potentially lifesaving in individuals with episodes of severe motor exacerbations 1 with accompanying changes to haemodynamic stability and vital functions—a phenomenon identified as status dystonicus or more recently, as severe acute motor exacerbations (SAME). 2 In 1996, describing the dramatic improvement with DBS in the first paediatric patient—an 8-year old girl with idiopathic generalised dystonia—the authors stated, “Although her future remains uncertain, we believe that chronic bilateral pallidal stimulation may prove to be the treatment of choice for early onset generalized dystonia especially in children.” 3 Nearly 3 decades later, many clinicians remain unsure which children with a movement disorder to refer for DBS, when to refer, and what the associated risks are. 4 In this retrospective study we summarise our experience from a tertiary centre in Australia to explore the evolution of our service in parallel to advances in the understanding of indications suitable for DBS, patient selection, and frameworks for quantitative and qualitative evaluation. In addition to describing cases with rare disorders not previously reported, we also highlight the implementation of interdisciplinary adult-paediatric, neurology-rehabilitation, and medical-allied health collaborative frameworks that have enabled us to set up a comprehensive paediatric DBS service.
Methods
The study was approved by the institutional ethics committee (LNR/18/SCHN/414), and informed consent was obtained from all participants. Individual patient information was obtained from review of electronic medical records. All patients who were referred for or underwent DBS at our centre from January 2013 to January 2025 were reviewed, including the decision-making processes for the treated cohort and those who did not proceed to DBS. In addition to outlining decision-making processes, we reported relevant clinical and investigative findings, including cases of SAME. We set up an interdisciplinary team (IDT) model in 2017 for collaborative decision making. This included paediatric neurologists, paediatric rehabilitation medicine physicians, an adult DBS specialist neurologist, a paediatric neuropsychologist, a paediatric occupational therapist, and paediatric physiotherapist as well as referring clinicians. All patients underwent pre-operative magnetic resonance imaging (MRI) with sequences suitable for stereotactic planning. Where available, any previous neuroimaging and clinical videos were requested to be presented by referring clinicians in the IDT meetings or, if urgent, to be shared with the IDT. After the IDT setup, patients who had acquired brain injury were offered pre-operative neurophysiology including central motor times (CMCT) and somatosensory evoked potentials (SSEP). Formal neuropsychological evaluation was offered to suitable patients after setting up the interdisciplinary framework and data from previous evaluations were included instead, if available, and performed as part of clinical care in the preceding 2 years. In parallel, a framework for standardised baseline and follow-up measures was set up, outlined in Supplementary Table 1. Additional detail on how these measures were applied is provided in Supplementary Box 1. Details on preoperative neuroimaging and DBS surgical procedures are summarised in Supplementary Box 2. Prior to 2017, patients progressed to DBS based on individual referrals to adult neurologists from paediatric neurology and rehabilitation clinics and underwent varying baseline and follow-up measures. The new model included questionnaires to assess patient and family expectations, perceived change, and satisfaction with DBS (Supplementary Table 2), adapted from previous adult patient studies (Supplementary Table 3). From 2019 to 2021, five patients and their accompanying parents also participated in video-recorded semi-structured interviews to assess expectations for DBS before surgery. These interviews were conducted by an occupational therapist and physiotherapist who specialise in paediatric rehabilitation and who also conducted the other outcome assessment measures. Interviews ranged in duration between 3 and 13 minutes depending on the length of participant responses. Topic guides can be found in Supplementary Box 3. Responses were transcribed and thematically analysed (Supplementary Box 4). Thematic saturation was considered achieved when core concepts were repeatedly identified and no new themes emerged in later interviews.
Results
During the study period, 16 patients (11 male, 5 female) underwent DBS at our institution. After implementing the IDT decision-making model in 2017, 44 patients were referred for consideration (Figure 1). Referrers included neurology (same hospital [n = 15]; other hospital/clinic [n = 14]), rehabilitation (same hospital [n = 11]; other hospital/clinic [n = 3]), and a general paediatrician (n = 1). The IDT considered 19 patients suitable for DBS, with 14 proceeding to surgery, 12 of which were performed at our centre and are described here. An additional 3 were awaiting DBS at the time this manuscript was compiled.

Patient suitability for DBS following referral to the IDT. This figure outlines the sources of patient referrals to the IDT for consideration of DBS and breaks down patient suitability for DBS by (A) aetiology, (B) presentation with SAME, (C) phenomenology, and (D) further quantifies reasons patients were evaluated as unsuitable for DBS by the IDT, including magnetic resonance imaging abnormalities, significant parental apprehension, or unsuitable phenomenology. Nineteen patients were considered suitable for DBS (14 proceeded, 3 are planned, 2 did not proceed), whereas 25 were unsuitable. DBS, deep brain stimulation; IDT, interdisciplinary team; ITB, intrathecal baclofen; SAME, severe acute motor exacerbations; SDR, selective dorsal rhizotomy.
IDT Decision Making and Alternative Interventions
By consensus, the IDT determined 25 patients as unlikely to benefit from DBS due to factors including MRI abnormalities (severe brain injury/atrophy), unsuitable movement disorder phenomenology, mild clinical severity, or unsuitable due to significant parental apprehension (Figure 1). Specific neuroimaging findings on MRI that suggested an unpredictable DBS response (leading to consideration as possibly unsuitable for DBS) included diffuse brain injury, severe basal ganglia or thalamic damage, or white matter injury affecting motor pathways. Phenomenology most likely to respond to DBS included isolated dystonia, dystonia with associated hyperkinetic movements (myoclonus; chorea), or parkinsonism, and phenomenology that was not considered suitable for DBS included dominant spasticity with or without dystonia.
Three of these patients with dominant spasticity proceeded to have an intrathecal baclofen (ITB) pump insertion, and another received an ITB test dose but did not progress to pump insertion. One patient proceeded to ITB for severe dystonia secondary to progressive basal ganglia encephalopathy. A patient with UBA5 related movement disorder and epilepsy was described in a previous cohort before having any interventions. 5 She underwent ITB for dominant spasticity after initial IDT discussion but then underwent emergent DBS 1.5 years later because of life-threatening SAME. 2 Another patient with pantothenase kinase–associated neurodegeneration (PKAN) underwent ITB 2 years after DBS because of increasing spasticity. One patient with kernicterus responded to emergent bilateral pallidotomy (further details not described here).
Clinical Details
The DBS cohort is summarised in Table 1 in chronological order with further details in Supplementary Tables 4.1 and 4.2. The cohort included 12 patients with inherited dystonia, 3 with idiopathic dystonia, and 1 with acquired dystonia. We report 2 patients with monogenic disorders (TNPO2 and SOX2) not previously reported to have undergone DBS. Clinical and diagnostic details for the patient with TNPO2-associated movement disorder prior to DBS were previously published as part of the first cohort of this disorder. 6 Three patients—one each with GNB1-, 7 GNAO1-, 8 and GABRB2 9 -associated movement disorders—have been previously described. These prior reports mentioned DBS but did not include details of the assessment process, patient selection, motor and non-motor measures, and longer-term outcomes.
Paediatric Deep Brain Stimulation Cohort: Clinical Summary.
Abbreviations: DBS-Deep brain stimulation; SAME, severe acute motor exacerbations.
At last follow-up.
Stimulation discontinued.
All patients in this cohort had hyperkinetic movement disorders. The 3 patients with DYT-SGCE and 1 with GNB1 variant had myoclonus-dominant myoclonus dystonia whereas patients with DYT-TOR1A and DYT-KMT2B had isolated dystonia. Cervical dystonia that progressed to become generalised was the dominant phenomenology in patient 6 with PKAN and in patients 3 and 13 with idiopathic isolated dystonia. Associated spasticity was noted in the patient with PKAN as well as in patient 16 with UBA5, both of whom also underwent ITB pump implantation as outlined above. Most children (14/16) had associated psychiatric or neurodevelopmental comorbidities, and 4 patients had epilepsy. Other accompanying clinical features are summarised in Supplementary Table 4.1.
All patients underwent DBS following unsuccessful attempts at managing their refractory movement disorders, having tried a median of 3 medications (range: 1-8) (Supplementary Table 4.2). SAME led to referrals in 9 patients. Of these, 2 patients with acute brain injury—one with pyruvate dehydrogenase deficiency and another with strangulation related basal ganglia injury—underwent urgent ITB insertion as they had rigid tone in wakefulness and sleep along with pallidal injury. Along with the significant radiologic injury, this phenomenology was considered to make the efficacy of DBS less predictable. 10 One patient with dyskinetic cerebral palsy secondary to kernicterus was referred for DBS during prolonged admission with SAME but was discharged after stabilising on medication. Another patient with kernicterus and SAME responded to bilateral pallidotomy. Emergent DBS for SAME was performed in 4 patients with UBA5-, GNAO1-, GABRB2-, and SOX2-related hyperkinetic movement disorders. An additional patient referred with dystonic cerebral palsy following extreme prematurity and possible kernicterus underwent DBS for SAME at another centre.
Since the introduction of our IDT framework, all but one of the children who underwent DBS (who had idiopathic dystonia) had monogenic diagnoses (patients 5-16). The 4 patients implanted prior to implementation of the IDT framework included 2 with idiopathic dystonia, 1 with GNAO1-associated movement disorder with SAME, and 1 with post-encephalitic dystonic cerebral palsy. Currently, 3 patients are awaiting DBS, including 1 with a monogenic hyperkinetic movement disorder due to ATP8A2 variants and 2 patients with brain injury–related dystonia—1 with kernicterus and 1 with thalamic atrophy after successful treatment with chemoradiotherapy for bilateral thalamic germinoma. In both patients with brain injury, the IDT considered the phenomenology as potentially amenable to DBS. CMCTs and SSEPs were abnormal in the patient with kernicterus and normal in the patient with thalamic atrophy, who were the only 2 patients who had these neurophysiology assessments. Both teenage patients, in conjunction with their families, have decided to progress to DBS because of the severe hyperkinetic movement disorder and prominent side effects, or lack of benefit with sequential, appropriate medication trials.
Implantation Details
The median age at implantation was 11.5 years (range: 3.5-17.1). On parental request, patients 7 and 13 were fitted with non-rechargeable implantable pulse generators (IPGs) (one with Medtronic Activa PC and the other with Medtronic Percept PC) because of concerns about intellectual ability, behavioural challenges, and fragile social support in the family limiting regular charging. Rechargeable IPGs were implanted in all other patients: Medtronic Activa RC in 13 patients and the Medtronic Percept RC with neurosensing ability in 2 patients. All patients were implanted with bilateral globus pallidus interna (GPi) leads, and directional leads have been used in the last 3 patients, all of whom remain on circumferential stimulation. IPGs were placed in the pectoral region in all patients except patient 4, where this was placed in the subcostal region because of lack of sufficient muscle and soft tissue cover in the pectoral region. Four-lead systems were implanted for 2 patients. Patients 2 had a second IPG with a separate pair of leads implanted into the STN whereas patient 6 had a second IPG with leads implanted into the subthalamic nucleus (STN) extending through the ventral occipito-posterior (Vop) nucleus of the thalamus. For patient 2, STN stimulation was switched off at 9-month follow-up because of perceived lack of benefit while GPi stimulation was continued. Patient 6 trialled GPi stimulation for 2 months but continued to experience worsening painful dystonia, subsequently benefitting from activation of STN stimulation and switching off GPi leads. Specific surgical and implantation variations were made for the youngest patients implanted (patients 11, 14, and 15): use of a double set of screws for the stereotactic frame and intentionally implanting the leads 2 mm below target to accommodate for potential lead retraction due to anticipated head growth with age. Amongst those patients who have had more than 1-year follow up and still have stimulators on (12/16), stimulation parameters that were set in the first 3 months after surgery did not require subsequent alteration in 4 of 12 patients.
Post-operative Outcomes
The median duration of follow up for the cohort was 2 years (0.5-10 years), excluding patient 4 who is currently awaiting reimplantation after device removal because of wound breakdown and antibiotic-resistant Staphylococcus aureus hardware infection 2 months after implantation. Motor exacerbations due to SAME (UBA5-, GNAO1-, GABRB2-, and SOX2-associated disorders) abated after DBS, with none requiring hospital admission for movement disorder postsurgery. Broadly, we grouped movement disorder outcomes into 3 categories at the time of their last follow up: (1) those with a complete response without any noticeable movement disorder on follow-up (patient 9 with DYT-TOR1A and patient 12 with DYT-SGCE), (2) those with a marked and persisting clinical response in movement disorder but with either some reduction in efficacy over time (patients 2 and 3) or with some residual motor symptoms (patients 4, 5, 7, 8, 10, 11, 13, 14, 15, 16), and (3) those with complete loss of efficacy after initial response (patient 6) or, worsening (patient 1), both of whom have stopped stimulation. Diagnostic details of all patients are found in Table 1.
Patient 3 with idiopathic isolated dystonia had marked initial benefit with suppression of dystonic neck and arm movements and normalisation of gait in the first 18 months after DBS. She then lost efficacy with gait improvement after loss of battery charge to zero. With reprogramming, she regained best benefit with interleaving and adjusting stimulation frequency to 180 Hz. Patient 15, with SOX2
Outcomes for Patients With Severe Acute Motor Exacerbations
DBS was undertaken after prolonged inpatient stays in 3 of 4 patients who received DBS with SAME at our centre (patient 4 with GNAO1-related dyskinetic crises, patient 11 with GABRB2, and patient 16 with UBA5). DBS enabled ICU or hospital discharge within a month in all cases. Prior to undergoing DBS, patient 11 had been hospitalised for 7 months because of a severe disabling movement disorder that required multiple medications and infusions, as documented in Supplementary Table 4.1 and 4.2. Remarkably, this patient achieved significant remission of dystonic episodes after DBS, which facilitated discharge home 1 month post-procedure. However, he passed away 15 months post-DBS as a result of refractory seizures, despite stable control of the movement disorder. Patient 16, with UBA5-related movement disorder, had a 3-month admission with SAME that continued to worsen despite medication and ITB titration. She required intravenous midazolam and ketamine sedation and intubation in intensive care. After emergent DBS, she was able to be discharged from intensive care within a week, and infusions were stopped.
Motor Scores
Structured motor scales were used to assess all patients pre- and post-implantation. Dystonia scales (Burke-Fahn-Marsden Dystonia Rating Scale Movement Subscale [BFMDRS-M] and Barry-Albright Dystonia Scale) were applied to 13 of 16 patients. Follow-up data were available for 12 of 16 children: 2 patients at 6 months (patient 11 passed away before the 12-month follow-up, and patient 16 has not yet reached 12 months post-operative), 6 patients at 12 months, and 4 patients at 2-3 years. Patient 15 was assigned pre-operative motor scores, but the stimulator was explanted before the 6-month follow-up) and only the Unified Myoclonus Rating Scale was used for 3 patients with SGCE associated myoclonus-dystonia. Individual scores are outlined in Supplementary Table 5.1–5.9. On last observation carried forward, 9 of 12 improved on BFMDRS-M (6 improved beyond the smallest detectable difference [SDD] of 27.39% 11 ) and 7 of 12 improved on the Barry-Albright Dystonia Scale (5 of 12 met the SDD of 17.72% 11 ). Patient 11 showed 100% improvement on both scales with near complete remission of motor symptoms. Patients 1 and 14 had slightly worsened scores on both.
The Unified Myoclonus Rating Scale was used for 4 patients with myoclonus-dominant movement disorder (patient 5 with GNB1 and patients 7, 10, and 12 with SGCE). All showed improved scores across Unified Myoclonus Rating Scale domains by 12 months, except patient 5, who showed improved scores only at 2-year follow-up.
A structured assessment of motor proficiency was applied in 4 of 16 patients using the Bruininks-Oseretsky Test of Motor Proficiency, Second Edition (BOT-2), Short Form, following introduction of the IDT framework. The BOT-2 Short Form comprises 14 items sampled across fine manual control, manual coordination, body coordination, and strength and agility domains, yielding an age-normed standard score. As the BOT-2 is designed for children with normal to mildly impaired motor control who can follow test instructions, it could not be administered to the remaining patients because of severe motor disability or intellectual impairment. At the 12-month follow-up, 3 of 4 patients demonstrated improvement in their standard scores compared with baseline, and 2 of 4 exceeded the published minimum clinically important difference (MCID) of 6.53 points, 12 indicating clinically meaningful gains in overall motor proficiency over the study period.
Disability and Function
The Disability Subscale of the BFMDRS (BFMRDS-D) was applied to 12 suitable patients (in 3 of 16 patients, the BFMDRS-D was not used because of myoclonus-dominant phenomenology, and 1 patient was explanted before the 6-month follow-up). On last observation carried forward, 5 of 12 improved on BFMDRS-D (all met the MCID of 0.5 points 13 ) and 4 of 12 had no change. The Caregiver Priorities and Child Health Index of Life with Disabilities (CPCHILD), a measure of quality of life, was applied to 9 of 12 patients post-IDT, but only to 2 of 4 patients who underwent DBS before formation of the IDT in 2017. At last observation, 9 of 11 patients had improved on the CPCHILD (6 met the MCID of 10 points 14 ).
The Dyskinetic Cerebral Palsy Functional Impact Scale (DFIS) was only applied after commencement of the IDT in 2017. Five participants were assessed with the DFIS and all demonstrated improvement.
Fatigue
Fatigue scores on the Paediatric Quality of Life Multidimensional Fatigue Scale (PedsQL-MFS) improved for 3 of 7 patients where this was applied (it was not implemented before the implementation of the IDT). Patient 13 with idiopathic dystonia reported initial improvement in fatigue at 6 months post-DBS but returned to baseline at the 12-month follow-up.
Pain
Pain was not formally assessed on an evaluation scale until the implementation of the IDT in 2017. Seven patients were assessed on the Paediatric Pain Profile (PPP), with 7 of 7 showing improved ongoing pain scores.
Expectations and Satisfaction
The Canadian Occupational Performance Measure (COPM) was applied to all patients to assess their performance and satisfaction on personally set treatment goals. One was explanted before follow-up. Performance on treatment goals set with COPM improved for 13/15 patients (9 met the commonly cited MCID of 2 points 15 —Supplementary Table 1 and footnote D). Satisfaction with goal performance improved for all patients beyond (13 of 15 beyond 2 points).
Locally developed questionnaires on expectations, perceived change, and satisfaction were completed by 6 families (Supplementary Figure 1) who were treated after application of the IDT framework. Pre-DBS semistructured interviews were conducted with 5 of 6 families, conducted after completion of the DBS workup process including at least 3 consultations with DBS clinicians (paediatric neurologist, paediatric rehabilitation medicine physician, adult neurologist, neurosurgeon), including education regarding the potential benefits and risks of surgery. On the structured questionnaires, families expected moderate to extreme motor and non-motor improvement, in contrast with prior consultations with DBS clinicians that included discussion of potentially attainable goals. Perceived motor gains were reported consistently at 6 and 12 months—except for patient 14, who reported improvement only at 12 months—whereas non-motor symptoms were mostly perceived to be unchanged at 6 months but improved by 12, especially pain. Four patients had low functional expectations; most showed no change at 6 months despite COPM and Dyskinetic Cerebral Palsy Functional Impact Scale (D-FIS) gains, although improvement emerged by 12 months. Only patient 12 (with highest expectations) and patient 14 (whose family did not complete pre-DBS questionnaires) reported no perceived functional improvement. All reported high overall satisfaction (Supplementary Figure 2).
On thematic analysis of transcribed patient and carer interviews (Supplementary Table 6), 5 themes emerged. “Addressing physical symptoms” was a salient theme for all patients, as was the “impact on function.” They valued regaining control of movements to enable easier and independent participation in activities of daily living. The focus of interviewees was on returning to normal functioning or normal self when the movement disorder was less severe and less impactful on daily function. The third theme was “discordance between hopes and expectations” for DBS outcomes among patients and carers. Although some identified specific preferred areas of improvement, others expressed a hope of broad, non-specific improvement following DBS. “Exploration and exhaustion of treatment options” was the fourth theme and a major motivator for pursuing DBS, and information provided to carers was often interpreted as a last resort. Most carers were overall satisfied with the information they were provided with, but some suggested that information could have been provided earlier in the process. The final theme was “psychosocial wellbeing.” Many patients and carers indicated an adverse impact on psychosocial health because of the movement disorder. Prior to DBS, 3 of 5 patients had a label of a mood disorder, 3 of 5 had a label of attention deficit hyperactivity disorder, and patient 4 had autism spectrum disorder along with a history of delusional symptoms. The impact of the disorder on patients’ ability to learn and socialise was explored. Formal neuropsychological evaluations prior to DBS revealed or confirmed pre-existing mental health issues in some patients, including anxiety, difficulty with mood-regulating behaviour and emotion, and difficulties with attention and other aspects of “executive functioning” (Supplementary Table 4.1).
Discussion
We present the evolving decision-making approach for children with movement disorders from a single centre, and our experience with the use of structured outcome measures and patient-reported outcome and experience measures. Overall, as described above and discussed here, a one-size-fits-all approach is difficult to apply to this group of children and their families, because of nuances in differences not only in their age and underlying aetiology but also clinical and perceived severity as well as varying goals, expectations, and satisfaction.
The cause of the movement disorder is a crucial factor for estimation of the potential benefits of DBS. As a general rule, genetic or idiopathic isolated movement disorders with normal brain structure on imaging are more likely to respond to DBS. However, DBS teams need to be aware of potential exceptions to the general rule, such as good responses with tardive dystonia and poor responses with ATP1A3-related disorder, and that this is a rapidly evolving field, particularly in terms of response in monogenic disorders.16–18 Furthermore, close scrutiny of clinical signs (such as evidence of upper motor neuron dysfunction or ataxia) and imaging may reveal important but easily overlooked changes—such as atrophy and increased T2 signal in the pallidum and STN in kernicterus, mild reduction in white matter volume—particularly if because of slice thickness or specific imaging data sets are not optimum. Other investigations may yield further information, including neurophysiological measures. To aid clinicians and families considering DBS for genetic movement disorders, our centre has collaborated with international centres to set up a live resource called the DBS matchmaker (https://www.dbsmatchmaker.com, last accessed on February 25, 2025). 19
Although the magnitude of measured motor change in individuals with cerebral palsy is generally lower than for those with some genetic disorders, some individuals will respond better than others. The exact determinants behind this need further exploration but there is a suggestion that intact CMCTs (measured by motor evoked potentials following transcranial magnetic stimulation) and SSEPs suggest that intact central sensory and motor pathways indicate a higher likelihood of benefit. 20 This group of individuals requires careful consideration and framing of expectations of likely motor or quality of life change with DBS. 21 The treating team also needs to be aware of other therapies that may be useful, particular in situations where the benefit of DBS may be limited, such as ITB, selective dorsal rhizotomy, and invasive or non-invasive lesioning. Overall, if the movement disorder severity dictates the consideration of DBS, a well-informed family may decide to proceed with DBS despite a relatively low chance of benefit, and collaborative decision making with an IDT framework is crucial.
Several reports suggest that SAME, often referred to as “status dystonicus,” are likely to respond to DBS irrespective of aetiology, including in individuals with CP. 1 This is concordant with our experience. Vogt et al 22 provide a comprehensive approach to SAME, emphasising the need to consider DBS including performing suitable imaging to enable quick decision making and implantation.
In our centre, we have adopted a model where phenomenology and aetiology are the primary considerations rather than the patient's age as to whether or not to offer DBS, although age remains an important consideration in terms of the natural history of the disease as well as for surgical technique adjustments. The youngest patient in this cohort was 3.5 years at the time of implantation. Mechanical issues such as skull pliability, pin penetration, or frame dislocation are considerations in younger children with a thinner and more deformable skull. 23 In our cohort, we mitigated this by using extra screws to distribute the load of the frame. Both families and referring clinicians also frequently ask about the risk of lead migration as the skull grows with age. In our cohort, only patient 3, who was implanted at the age of 6.3 years, had a follow-up MRI scan at the age of 15 years to check for lead migration in view of ongoing partial benefit. This did not detect a change in the lead position within the GPi. More extensive studies have shown that the projected distance from skull entry to the DBS target is likely to change by 5 to 10 mm from 4 to 18 years of age, with the maximum change occurring before the age of 7 years. 24 We implant children under the age of 7 years by placing the lead deeper than the target to allow for any potential lead retraction with skull growth, if safe to do so in relation to cerebral vasculature. It has also been suggested that DBS should be considered earlier in the life course of dystonia because the proportion of life lived with dystonia may correlate with the degree of motor improvement. 25 In terms of considering early DBS in severe or progressive disorders, our experience with patient 6 with PKAN and patient 11 with GABRB2-related movement disorder, who ultimately passed away, was that DBS provided a period of significant comfort and respite to carers, resulting in better quality of life and relief from distress.
The invasive nature of DBS, with its attendant risks needs to be weighed against risks related to complications of their movement disorder. In our small cohort, severe adverse effect of system infection leading to explantation occurred in 1 case. Overall, wound or hardware-related infections have been reported in a varying proportion of children from 8% to 12.5%,26,27 slightly higher than in adult cohorts.
Stimulation-related adverse effects noted in our cohort were manageable through programming adjustments, but these can sometimes necessitate balancing less stimulation to reduce adverse effects. Freezing of gait is a recognised complication of pallidal lesions and overstimulation of the GPi (particularly the ventral part of the nucleus), which occurred in 2 patients, one with DYT-SGCE and another with DYT-KMT2B. This complication was recently reported in a few individuals with DYT-KMT2B, 28 raising the possibility of selective vulnerability for some complications in specific conditions.
Measures that are administered and recorded primarily by health professionals can sometimes have an inherent bias. Therefore, it is suggested that patient-reported outcome measures (PROMs) and patient-reported experience measures (PREMs) are critical to the development of a comprehensive assessment framework. 29 Recent Delphi-based consensus work in DBS similarly recommends core sets of outcome measures that combine motor scales with quality-of-life and patient-reported domains, underscoring the need for multidimensional outcome assessment in DBS trials and registries. 30
Many of the measures we used were based on patient reports representing PROMs (Supplementary Table 1) whereas the expectations and satisfaction questions represented PREMs. Proxy-report bias is also inherent to these outcome measures when a proxy is used, either because it is required by the instrument (eg, the CPCHILD) or due to a child's inability to provide direct self-report because of disability or young age. This is particularly relevant in the emotionally and cognitively complex context of parental decision making in paediatric brain surgery. Some studies suggest that proxies may report worse outcomes than patients themselves.31,32 However, a recent large-scale systematic review 33 found that proxies are reliable reporters for PROMs related to physical outcomes and quality of life, although greater variability is observed in reports of psychological outcomes. We did not employ structured measures for some non-motor symptoms such as mental health symptoms, but these were addressed by clinicians and by neuropsychology assessments in some patients and should also form part of assessments. 34
Although motor symptoms and their improvement remain a focus for patients and carers, the results presented here reconfirm that other aspects such as non-motor symptoms as well as functional abilities and mental well-being are important. Even in this small cohort, it was obvious that standard and commonly used motor rating scales do not adequately measure outcomes that are meaningful to children with movement disorders, consistent with other studies. 35 Therefore, individualised occupational performance goal setting and activity performance measures (such as the COPM and goal attainment scale) may play an increasingly important role in the documentation of outcome with interventions like DBS. In addition to core measures, co-design of approaches with families can ensure that specific and relevant evidence is ascertained for each child, grounding outcomes towards the people living with the movement disorder.
This emphasis on individualised, function-focused outcomes is consistent with the World Health Organisation's International Classification of Functioning, Disability and Health (ICF), which highlights the importance of activities and participation, in addition to body structures and function, to develop a comprehensive understanding of an individual's health and disability status. 36 In line with this, Gimeno and Lin 37 have outlined why DBS outcome measurement in dystonia should explicitly incorporate ICF-based domains beyond impairment, including activity, participation, and contextual factors, rather than relying solely on motor severity scales. Goal-oriented DBS programmes that assess change beyond motor symptoms are increasingly recommended in contemporary clinical guidance, 38 and research increasingly shows that goal attainment and participation-focused measures capture meaningful changes that are not reflected in motor severity scales alone. 39 The present study describes an IDT-based service model designed to address these broader outcomes, building on previously reported multidisciplinary DBS service frameworks. 37
Although most outcomes showed improvement, fatigue scores were mixed. Patient 5, 7, and 8 worsened. This reflects the multifactorial nature of fatigue including the influence of comorbidities and possible increased activity demands after motor improvement. Patient 5 experienced significant psychological comorbidity including anxiety and OCD whereas patient 7 had epilepsy, both of which are associated with an effect on fatigue. Patient 8 had intellectual disability including cognitive fatigue. Fatigue is influenced by cognitive load, emotional regulation, and adaptive functioning, and may increase even when motor function improves.
It is our practice to ask patients and families to restate their understanding of their expectations during pre-DBS consultations. We also highlight the differences between hopes and expectations of patients and their families. This suggests that there may be a discrepancy between what people understand at a verbal and intellectual level, vs what they feel, hope, and wish for, aligning with previous reports. 40 It is possible that patient and carer expectations can also influence perceived change. 41 The potential benefit from DBS ranges from no response to near complete response. Many disorders include non-motor symptoms such as speech delay that cannot be expected to improve with DBS; however, patients and families may remain hopeful that most or all symptoms will improve with an intervention. Hopes and expectations are therefore important to moderate and should form an important part of pre-DBS counselling. Some carers are keen to know about interventions earlier rather than later and this may also hold true for referring clinicians who may not be aware of invasive interventions like DBS for patients with neurodisability. In response, we have co-developed a suite of psychoeducational resources for and with families promoting information provision, health care engagement, and clinical decision making 42 (https://www.schn.health.nsw.gov.au/advanced-therapies-handbook, last accessed June 2025). We are also in the process of finalizing a co-designed suite of neuromodulation specific resources tailored to young people, families, and referring clinicians.
Strengths of our study include the detailed assessments of outcome, even if the same measures could not be applied to all participants. The diversity of the participants (including age, aetiology, severity, phenomenology) is also a strength, suggesting that the lessons learned herein have broad applicability in common as well as very rare paediatric movement disorders. Inclusion bias is unlikely in this consecutive cohort of all treated patients. This study has some limitations: the cohort size was relatively small and diverse rather than large and homogeneous that might allow statistical interrogation, and the questionnaires used were derived from previous adult studies lacking validation in this population. Another limitation of this study is the inherent bias involved in qualitative thematic analysis; however, we attempted to mitigate this through multiple independent coders and consensus meetings. In conjunction with surveys, patient experience with respect to the treatment (improving symptoms improvement, side effects, and delivery of the intervention) can provide important specific evidence of impact and contextual information important for determining meaningful change. We acknowledge that our service evolved and therefore the tools used were variable between patients, reflecting improvements, we hope, in our understanding of meaningful and useful outcome measures in this population.
Interventions like DBS are likely to be offered to individuals with rare neurologic disorders, where the lack of evidence precludes accurate prediction of outcomes—something that we emphasise herein as well as in our pre-DBS consultations. Based on our experience, we propose a decision-making algorithm (Figure 2), to guide clinicians, patients, and families on the suitability of DBS for each individual. We also propose that a comprehensive assessment program should include measures beyond standard motor scales, including assessment of function and a breadth of symptoms including non-motor domains. In addition, we recommend the use of PROMs and PREMs, particularly for goal setting and achievement, and for assessing patient and carer expectations and satisfaction to better evaluate outcomes according to the priorities of patients and carers. This also enables families to be part of the shared decision-making process. Future research into the discrepancy between patient-reported outcomes and structured outcome measures may reveal further insights into how patient priorities can be measured and addressed most effectively.

Decision-making framework for assessment of DBS suitability by an IDT. DBS, deep brain stimulation; IDT, interdisciplinary team; SAME, severe acute motor exacerbations.
Supplemental Material
sj-docx-1-jcn-10.1177_08830738261435155 - Supplemental material for Implementation of Paediatric Deep Brain Stimulation: Experience from an Australian Tertiary Centre
Supplemental material, sj-docx-1-jcn-10.1177_08830738261435155 for Implementation of Paediatric Deep Brain Stimulation: Experience from an Australian Tertiary Centre by Jamie M. Barnacoat, Wafa Bani Uraba, Dianah Hadi, Kirsty Stewart, Jennifer Lewis, Sara Coombes, Mary-Clare Waugh, Michelle A. Farrar, Brian Owler, Neil Mahant, Russell C. Dale, Simon Paget and Shekeeb S. Mohammad in Journal of Child Neurology
Footnotes
Acknowledgements
We would like to thank the patients and their carers for their participation and Cerebral Palsy Alliance Australia for supporting this study. We also thank many other clinicians including allied health professionals and support workers who contributed to the individual care of each patient and their family.
ORCID iDs
Ethics Statement
The study was approved by the institutional ethics committee (LNR/18/SCHN/414).
Informed Consent Statement
Informed consent was obtained from all participants.
Author Contributions
Conceptualization, S.S.M., R.C.D.; Methodology, S.S.M., R.C.D., S.P., K.S., J.L.; Data Curation, J.M.B., W.B.U., D.H., K.S., J.L., S.C., S.S.M.; Formal Analysis, J.M.B., K.S, J.L, N.M, S.S.M.; Investigation, J.M.B., W.B.U., D.H., K.S., J.L., N.M., B.O.; Writing – Original Draft, J.M.B., S.S.M.; Writing – Review & Editing: All authors; Visualization, J.M.B., S.S.M.; Supervision: S.S.M., R.C.D., M.A.F.; Project Administration, S.S.M.
Funding
The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the Cerebral Palsy Alliance (grant number PG01217).
Declaration of Conflicting Interests
The authors declared the following potential conflicts of interest with respect to the research, authorship, and/or publication of this article: Jamie Barnacoat's position was partially funded by the Cerebral Palsy Alliance, Australia, grant PG01217 (CIA Shekeeb Mohammad).
Data Availability Statement
The authors confirm that the data supporting the findings of this study are available within the article and its supplementary materials.
Supplemental Material
Supplemental material for this article is available online.
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
