You are here

Article Text

Article menu
Movement disorders
Original research
Differential response to pallidal deep brain stimulation among monogenic dystonias: systematic review and meta-analysis
  1. Carlo Alberto Artusi1,
  2. Alok Dwivedi2,
  3. Alberto Romagnolo1,
  4. Sara Bortolani1,
  5. Luca Marsili3,
  6. Gabriele Imbalzano1,
  7. Andrea Sturchio3,
  8. Elizabeth G Keeling3,
  9. Maurizio Zibetti1,
  10. Maria Fiorella Contarino4,5,
  11. Alfonso Fasano6,7,
  12. Michele Tagliati8,
  13. M S Okun9,
  14. Alberto J Espay3,
  15. Leonardo Lopiano1,
  16. Aristide Merola10
  1. 1 Department of Neuroscience "Rita Levi Montalcini", University of Turin, Turin, Italy
  2. 2 Texas Tech University Health Sciences Center El Paso, El Paso, Texas, USA
  3. 3 Department of Neurology, University of Cincinnati, Cincinnati, Ohio, USA
  4. 4 Department of Neurology, Haga Teaching Hospital, The Hague, The Netherlands
  5. 5 Leids Universitair Medisch Centrum, Leiden, The Netherlands
  6. 6 Morton and Gloria Shulman Movement Disorders Clinic. Division of Neurology, Toronto Western Hospital, Toronto, Ontario, Canada
  7. 7 Krembil Research Institute, Toronto, Ontario, Canada
  8. 8 Department of Neurology, Cedars Sinai Medical Center, Los Angeles, California, USA
  9. 9 Department of Neurology, University of Florida, Gainesville, Florida, USA
  10. 10 Department of Neurology, Ohio State University Wexner Medical Center, Columbus, Ohio, USA
  1. Correspondence to Dr Carlo Alberto Artusi, Department of Neuroscience "Rita Levi Montalcini", University of Turin, Turin 10126, Italy; caartusi{at}gmail.com

Abstract

Objective Genetic subtypes of dystonia may respond differentially to deep brain stimulation of the globus pallidus pars interna (GPi DBS). We sought to compare GPi DBS outcomes among the most common monogenic dystonias.

Methods This systematic review and meta-analysis followed the Preferred Reporting Items for Systematic Reviews and Meta-analyses and Meta-analysis of Observational Studies in Epidemiology guidelines. We searched PubMed for studies on genetically confirmed monogenic dystonia treated with GPi DBS documenting pre-surgical and post-surgical assessments using the Burke–Fahn–Marsden Dystonia Rating Scale Motor Score (BFMMS) and Burke–Fahn–Marsden Disability Score (BFMDS). We performed (i) meta-analysis for each gene mutation; (ii) weighted ordinary linear regression analyses to compare BFMMS and BFMDS outcomes between DYT-TOR1A and other monogenic dystonias, adjusting for age and disease duration and (iii) weighted linear regression analysis to estimate the effect of age, sex and disease duration on GPi DBS outcomes. Results were summarised with mean change and 95% CI.

Results DYT-TOR1A (68%, 38.4 points; p<0.001), DYT-THAP1 (37% 14.5 points; p<0.001) and NBIA/DYT-PANK2 (27%, 21.4 points; p<0.001) improved in BFMMS; only DYT-TOR1A improved in BFMDS (69%, 9.7 points; p<0.001). Improvement in DYT-TOR1A was significantly greater than in DYT-THAP1 (BFMMS −31%), NBIA/DYT-PANK2 (BFMMS −35%; BFMDS −53%) and CHOR/DYT-ADCY5 (BFMMS −36%; BFMDS −42%). Worse motor outcomes were associated with longer dystonia duration and older age at dystonia onset in DYT-TOR1A, longer dystonia duration in DYT/PARK-TAF1 and younger age at dystonia onset in DYT-SGCE.

Conclusions GPi DBS outcomes vary across monogenic dystonias. These data serve to inform patient selection and prognostic counselling.

  • dystonia
  • deep brain stimulation
  • globus pallidus
  • genetics
  • monogenic

https://doi.org/10.1136/jnnp-2019-322169

Statistics from Altmetric.com

    Request Permissions

    If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

    Introduction

    Dystonia, defined as a condition characterised by sustained or intermittent muscles contractions causing abnormal, often repetitive, movements, postures or both, is a common movement disorder leading to generalised, segmental or focal impairment of motor activities and physiological postures.1 2 It has been linked to over 200 genetic mutations leading to autosomal dominant, recessive and X-linked disorders.3 4 Genetic dystonias may present with variable age at onset, body distribution, temporal pattern and associated features.1 Clinically, they are classified into ‘isolated dystonias’, such as DYT-TOR1A, childhood-onset foot dystonia followed by generalisation, and DYT-THAP1, adolescent-onset cranio-cervical dystonia; ‘combined dystonias’, such as DYT-SGCE, in which dystonia is ‘combined’ with myoclonus; and ‘complex dystonias’ such as NBIA/DYT-PANK2, a childhood-onset or adolescent-onset dystonia due to pantothenate kinase-associated neurodegeneration with brain iron deposition (PKAN).3

    Deep brain stimulation (DBS) of the globus pallidus pars interna (GPi) is one of the most effective treatments for chronic, medically intractable, dystonia. Multiple trials have demonstrated that GPi DBS is safe and effective for patients with isolated generalised and segmental dystonia.5 6 Outcomes after GPi DBS, however, can have variable success, which highlights the importance of patient selection. Narrative reviews have summarised data with divergent conclusions on the variables affecting GPi DBS outcomes.4 Established predictors include electrodes location,7 age at dystonia onset,8 disease duration prior to GPi DBS9 and whether fixed postures have complicated the phenotype.10

    DYT-TOR1A patients with segmental or generalised dystonia are considered optimal candidates for surgery,4 11 but no trial data are available to assess whether the outcomes on DYT-TOR1A-positive patients are comparable to those of patients with alternative genetic mutations. We sought to evaluate the differential extent to which monogenic dystonias respond to GPi DBS using a systematic review and meta-analysis of outcomes with data from published clinical trials, observational studies, case series and case reports.

    Methods

    Search methods

    We conducted a systematic review and meta-analysis following the Preferred Reporting Items for Systematic Reviews and Meta-analyses and the Meta-analysis of Observational Studies in Epidemiology guidelines.12 13 We searched PubMed for interventional and non-interventional studies prior to January 1, 2019 reporting data on GPi DBS-treated patients screened for monogenic forms of dystonia using the following searching terms: ‘deep brain stimulation’, ‘mutation’, ‘gene’, ‘genetics’, ‘inherited’, ‘familial’, ‘dystonic’ and ‘dystonia’ (search string: ‘((Deep brain stimulation AND (dystonia OR dystonic) AND (mutation OR gene OR genetics OR inherited OR familial)))’).

    Abstracts and full-text articles were independently reviewed for eligibility criteria by four authors (CAA, AR, LM and SB). Duplicated studies were identified and excluded. Only studies referring to human subjects and published in English were considered. No restrictions were applied to sex, age, ethnicity, follow-up duration, disease duration, disease severity or type of dystonia. The reference list of each article was searched to screen for additional pertinent studies not captured by the original search strategy.

    Inclusion and exclusion criteria

    We included studies with cases of genetically confirmed dystonias treated with GPi DBS, with a minimum post-surgical follow-up of 3 months, and with Burke–Fahn–Marsden Dystonia Rating Scale (BFMDRS) scores before and ≥3 months after GPi DBS.14 Studies of patients with pooled rather than segregated genetic subtyping were excluded, as well as studies with assumed but not confirmed genetic data or incomplete follow-up.

    Data extraction

    We used a standardised data collection form to extract, from the BFMDRS, the Motor (BFMMS) and Disability Scores (BFMDS) before (baseline), at 3–24 months (short-term post-GPi DBS follow-up), and, when available, at >24 months (long-term follow-up). Additional data extracted included sample size, genetic mutations evaluated, year of publication, study design, age at disease onset, age at GPi DBS surgery and follow-up duration in months. Data were expressed as mean with SD. If two or more studies reported data from the same population, we included the most recent publication with the longest follow-up. In addition, if a study provided data on multiple genetic populations we allocated each to their corresponding dataset.

    Assessment of risk of bias

    Two investigators (CAA and AR) independently performed the quality appraisal of qualifying studies, which were rated as ‘good’, ‘fair’ or ‘poor’ as per the National Heart, Lung, and Blood Institute tools (Research Triangle Institute International. National Heart, Lung, and Blood Institute Quality Appraisal Tools) according to Cochrane handbook recommendations (see online supplementary figure 1).15 Only studies with a ‘good’ or ‘fair’ rating were included in the analyses. Visual inspection of funnel plots was also conducted to assess for publication bias.16

    Supplemental material

    Study aims and statistical analysis

    We sought to analyse motor (BFMMS) and disability (BFMDS) outcomes of patients with confirmed monogenic dystonias treated with GPi DBS and to examine the impact of disease duration prior to GPi DBS, age, and sex. The following sets of analysis were conducted:

    1. To analyse the effect of genetic mutations on GPi DBS outcomes, we performed a meta-analysis of BFMMS and BFMDS short-term and long-term results for each gene separately. This analysis was performed for at least two studies on a defined monogenic dystonia cohort with a sample size ≥5 patients each containing pre-GPi DBS and post-GPi DBS BFMDRS scores. The mean change between pre-intervention and post-intervention along with pooled SD (PSD) for each study was computed. Pearson’s correlation coefficient between pre-value and post-value per gene mutation was estimated and used for computing PSD. Given the sample size, the inclusion of observational studies, and the heterogeneity across the studies, the pooled effect size was computed using a random effects model with DerSimonian and Laird method.17 The heterogeneity across the studies was measured using the I2 statistic. An I2 statistic >50% was considered as representing substantial heterogeneity.15 The results of the meta-analysis were summarised using I2 statistic, per cent relative improvement, pooled mean difference along with a 95% CI and p value for each specific gene mutation separately for each outcome. Meta-analysis was conducted separately for short-term and long-term outcomes and results were summarised using forest plots.

    2. To analyse the outcomes of genetic mutations compared with DYT-TOR1A, we performed unadjusted and adjusted (age at dystonia onset and dystonia duration at surgery) comparisons of BFMMS and BFMDS between DYT-TOR1A and other distinct forms of monogenic dystonia. This analysis was performed when at least two studies, one of which with a sample size >1 patient, were available. We performed unadjusted and adjusted weighted ordinary linear regression analyses with robust variance estimation using Huber and Sandwich approach. The weight was assigned according to the sample size of each study. Larger sample size studies received relatively large weights compared with small sample size studies. After regression analysis, unadjusted paired-wise comparisons were performed across all gene mutations. The results of regression analyses were displayed using regression coefficient, 95% CI and p value. In the adjusted analysis, age and disease duration were adjusted irrespective of their significance levels.

    3. To analyse the effect of age at dystonia onset, disease duration at surgery and sex on GPi DBS outcomes, we used weighted linear regression analyses with robust variance estimation after adjusting for gene differences and separately for each gene. This analysis was performed for each monogenic dystonia when at least two studies were identified from the systematic review of the literature.

    Continuous data were summarised using mean and SD; categorical data with frequencies and proportions. All the statistical analyses were carried out using STATA V.15.1. P values less than or equal to 5% were considered statistically significant results.

    Results

    Of 165 studies, 87 met full criteria and underwent data extraction (see online supplementary table 1), individual quality assessment and evaluation of the risk of bias (figure 1). In all, 34 articles met criteria for meta-analysis yielding 36 datasets and a sample size of 311 patients distributed in DYT-TOR1A (n=269), DYT-THAP1 (n=16) and NBIA/DYT-PANK2 (n=26) mutations. In total, 81 articles met criteria for the analysis of outcomes vs DYT-TOR1A, providing 87 datasets and a sample size of 432 patients in DYT-TOR1A (n=306), DYT-THAP1 (n=24), NBIA/DYT-PANK2 (n=40), DYT/PARK-TAF1 (n=23), ACTB (n=3), CHOR/DYT-ADCY5 (n=3), GNAO1 (n=8) and DYT-SGCE (n=25) mutations. In addition, individual datasets of single case reports (n=13) were extracted from seven studies reporting data on DYT/PARK-GCH1, CHOR-VPS13A, DYT-SGCE + DYT-TOR1A, DYT/PARK-ATP1A3, DYT-PRKRA, DYT/PARK-GLB1, NBIA/DYT-DCAF17, trisomy X, SCA-ATXN3, SCA-ATXN2, PxMD-SLC2A1 and ATM (figure 1, online supplementary table 2). Demographic data are reported in table 1.

    Supplemental material

    Supplemental material

    View this table:
    Table 1

    Demographic and clinical data of included patients

    Figure 1
    Figure 1

    Study flow chart phases of the systematic review according to the Preferred Reporting Items for Systematic Reviews and Meta-analyses flow diagram recommendations. RCT, randomised clinical trial.

    Meta-analysis

    Motor endpoint (BFMMS)

    In the short term (11.8±3.6 months; range: 6–12), BFMMS improved by 68% in DYT-TOR1A (−38.4, p<0.001), 37% in DYT-THAP1 (−14.5, p<0.001) and 27% in NBIA/DYT-PANK2 (−21.4, p<0.001). Long-term (46.6±16.4 months; range: 24–72) BFMMS data were available for DYT-TOR1A, showing sustained improvement (68%, −37.1, p<0.001) (figure 2).

    Figure 2
    Figure 2

    Forest plot-motor endpoints. Forest plot of the motor endpoints obtained by the meta-analysis of the following genetic dystonias: DYT-TOR1A, DYT-THAP1 and NBIA/DYT-PANK2. Data presented refer to absolute change of the BFMMS. n=number of datasets included. BFMMS, Burke–Fahn–Marsden Dystonia Rating Scale Motor Score.

    Disability endpoint (BFMDS)

    In the short term (12.5±2.9 months; range: 10–24), BFMDS improved by 69% in DYT-TOR1A (−9.7, p<0.001) but not in DYT-THAP1 (−0.4; p=0.895) or NBIA/DYT-PANK2 (0; p=0.993). Long- term (45.9±18.5 months; range 25–72) BFMDS data were available for DYT-TOR1A, showing sustained improvement (68%, −10; p<0.001) (figure 3).

    Figure 3
    Figure 3

    Forest plot-disability endpoints. Forest plot of the disability endpoints obtained by the meta-analysis of the following genetic dystonias: DYT-TOR1A, DYT-THAP1 and NBIA/DYT-PANK2. Data presented refer to absolute change of the BFMDS. n=number of datasets included. BFMDS, Burke–Fahn–Marsden Disability Score.

    Comparison of GPi DBS effect on monogenic dystonias

    Motor endpoint (BFMMS)

    DYT-TOR1A was associated with greater motor improvement than DYT-THAP1 (−31%; p<0.001), NBIA/DYT-PANK2 (−35%; p=0.016) and CHOR/DYT-ADCY5 (−36%; p<0.001). We found no significant differences between DYT-TOR1A and DYT/PARK-TAF1, DYT-SGCE, GNAO1 and ACTB (figure 4).

    Figure 4
    Figure 4

    Comparison of GPi DBS outcomes between DYT-TOR1A and other monogenic forms of dystonia Motor outcomes were assessed by the score of the Burke–Fahn–Marsden Dystonia Rating Scale; disability outcomes by the disability Burke–Fahn–Marsden Dystonia Rating Scale. GPi DBS: deep brain stimulation of the globus pallidus pars interna. RC: regression coefficient after adjusting for age and disease duration. *Indicates significant difference compared with DYT-TOR1A.

    Disability endpoint (BFMDS)

    DYT-TOR1A was associated with greater reduction in disability than NBIA/DYT-PANK2 (−53%; p=0.004) and CHOR/DYT-ADCY5 (−42%; p=0.003). We found no significant differences between DYT-TOR1A and DYT/PARK-TAF1, DYT-THAP1, DYT-SGCE, GNAO1 and ACTB (figure 4).

    Effect of disease duration, age at onset and sex on GPI DBS outcome

    Longer dystonia duration prior to GPi DBS was associated with worse motor outcome in DYT-TOR1A (p=0.040) and DYT/PARK-TAF1 (p=0.001), but not in DYT-THAP1, DYT-SGCE and NBIA/DYT-PANK2 (Table 2, Supplementary Figure 2). Older age at dystonia onset was associated with worse motor outcome in DYT-TOR1A (p<0.001) but not in DYT/PARK-TAF1, DYT-THAP1 and NBIA/DYT-PANK2, and better outcome in DYT-SGCE (p=0.010) (table 2).

    View this table:
    Table 2

    Effect of disease duration, age and sex on GPI DBS outcome

    Sex was not associated with GPi DBS outcomes in DYT-TOR1A, DYT-SGCE and NBIA/DYT-PANK2. In DYT-THAP1, there was a trend towards better GPi DBS motor outcomes in women compared with men (p=0.051) (table 2).

    Analysis of single cases or rarer dystonia-associated mutations

    There was improvement in motor and disability outcomes reported in single cases of DYT/PARK-GCH1 (−44.5% BFMMS; −85.7% BFMDS), CHOR-VPS13A (−75% BFMMS; −50% BFMDS) and DYT-SGCE + DYT-TOR1A mutations (−83.3% BFMMS; −75% BFMDS); in motor outcomes in DYT/PARK-ATP1A3 (−26.1% and −7% BFMMS), DYT-PRKRA (−74.4% BFMMS), DYT/PARK-GLB1 (−20% BFMMS), NBIA/DYT-DCAF17 (−42.0% BFMMS; −7.7% BFMDS), trisomy X (−44.2% BFMMS) and SCA-ATXN3 (−10% BFMMS;+33% BFMDS); and worsening of motor and disability outcomes in SCA-ATXN2 (+35.5% BFMMS;+112.5% BFMDS), PxMD-SLC2A1 (+2.6% BFMMS) and ATM (+13.0% BFMMS;+12.5% BFMDS) (table 3).

    View this table:
    Table 3

    Single case analysis

    Discussion

    We confirmed the beneficial effect of GPi DBS on motor and disability outcomes in DYT-TOR1A, DYT-THAP1 and NBIA/DYT-PANK2 cases, with a greater improvement in DYT-TOR1A compared with other monogenic dystonias. Differential improvement was observed in ACTB, GNAO1, DYT-SGCE, DYT/PARK-TAF1 and CHOR/DYT-ADCY5, affected negatively by longer disease duration prior to GPi DBS (DYT-TOR1A and DYT/PARK-TAF1) and older age at dystonia onset (DYT-TOR1A; the opposite for DYT-SGCE).

    Previous studies alternatively suggested that dystonia duration before surgery may or may not predict outcome after GPi DBS.5 8 18–22 In agreement with a recent meta-regression analysis reporting an association between shorter dystonia duration and greater GPi DBS motor improvement in patients with isolated dystonia,9 we found that monogenic dystonias DYT-TOR1A and DYT/PARK-TAF1 may benefit from an earlier surgical treatment. This result is in accordance with previous findings indicating that an early modulation of dystonia-associated cortical plasticity might slow or prevent progression.19 Our findings suggest that different genetic mutations may account for a different impact of disease duration on GPi DBS outcome, and similar results emerged from the analysis of age at dystonia onset, with younger DYT-TOR1A and older DYT-SGCE patients benefiting most from GPi DBS. Also, there was a trend towards better motor outcomes in women compared with men in DYT-THAP1.

    GPi DBS demonstrated short-term and long-term efficacy in DYT-TOR1A,3 4 and a lower but also significant improvement in DYT-THAP1 and NBIA/DYT-PANK2.23 24 The differential extent of motor benefit observed in DYT-TOR1A and DYT-THAP1 might reflect the differential pattern of metabolic abnormalities reported by connectivity studies in these two monogenic forms of dystonias,25 26 as well as the older age at onset of DYT-THAP1 and its predominant axial and cranio-cervical distribution,27 which have been reported as factors negatively influencing GPi DBS outcome.28 The lower extent of motor benefits observed in NBIA/DYT-PANK2 might result from the high variability in associated features (eg, spasticity) of the PKAN phenotype.29

    Available data were insufficient for a meta-analysis in cases of DYT/PARK-TAF1, associated with X-linked dystonia-parkinsonism (XDP or Lubag disease)30; DYT-SGCE, responsible for an early-onset form of myoclonus-dystonia predominantly involving the neck and arms3; CHOR/DYT-ADCY5, which may cause a variable phenotype characterised by chorea, dystonia or myoclonus31 32; GNAO1, associated with a form of childhood dystonia combined with other hyperkinetic movement disorders, cognitive impairment and seizures33; and ACTB, responsible for a clinical phenotype characterised by dystonia and deafness.34 35 However, outcomes may not be significantly different from those observed in DYT-TOR1A using age- and disease-duration-adjusted motor and disability outcomes, with the only exception of CHOR/DYT-ADCY5 that showed significantly lower benefits from GPi DBS than DYT-TOR1A.

    The cases of DYT/PARK-TAF1 treated with GPi DBS (n=23) showed good motor and disability improvement at 12 months. While these outcomes are encouraging, it is important to note that the clinical picture associated with DYT/PARK-TAF1 (ie, XDP) may change over time. Focal dystonia tends to generalise within the first 5 years; after 10 years, dystonia becomes relatively less severe in the face of increasingly prominent parkinsonian signs.30 A recent observational study from 16 patients showed that GPi DBS may also improve the parkinsonian features associated with XDP.36 Favouring earlier timing at surgery, the authors observed an association between less atrophy in the caudate nucleus and better GPi DBS clinical outcomes.36

    In DYT-SGCE, GPi DBS has been reported to benefit both the myoclonus and the dystonia.37 38 Our analysis of 26 patients with DYT-SGCE showed a significant improvement on motor and disability outcomes similar to those observed in DYT-TOR1A, thus confirming that GPi DBS may be an optimal therapeutic option for carriers of DYT-SGCE-associated dystonia. GNAO1 patients treated with GPi DBS (n=8) showed mixed results likely related to the heterogeneous clinical presentation of this particular genotype, which is frequently associated with a progressive encephalopathy. GPi DBS seems a viable (palliative) option for cases with extreme dystonia severity with difficulties in feeding and/or life-threatening clinical conditions.39 ACTB patients (n=3) showed good preliminary results, which suggest the potential for GPi DBS as a therapeutic option for this particular form of dystonia but also warrant further evaluation due to the limited number of cases reported to date. Finally, the analysis of the few CHOR/DYT-ADCY5 cases (n=3) treated with GPi DBS suggested lower clinical efficacy than DYT-TOR1A. However, it should be considered that combined (DYT-SGCE, DYT/PARK-TAF1, CHOR/DYT-ADCY5) and complex (GNAO1, ACTB) dystonias present with a plethora of accompanying neurological features, which may or may not respond to GPi-DBS. The greater clinical benefit observed in DYT-SGCE could be explained by the effect of GPi-DBS on both dystonia and myoclonus.40 The limited efficacy observed in GNAO1 may be explained by the concomitance of features characteristically not responsive to GPi-DBS, such as intellectual disability and drug-resistant seizures.39

    GPi DBS in monogenic dystonias confined to single case reports yielded results ranging from excellent (DYT/PARK-GCH1, CHOR-VPS13A and DYT-SGCE +DYT-TOR1A), mild/moderate (DYT/PARK-ATP1A3, DYT-PRKRA, DYT/PARK-GLB1, NBIA/DYT-DCAF17 and trisomy X), to futile (SCA-ATXN3, SCA-ATXN2, PxMD-SLC2A1 and ATM). The significant variability observed in these rare forms of dystonia highlights the critical need for prospective clinical registries reporting all cases of genetically defined dystonia treated with GPi DBS. For DYT/PARK-ATP1A3, there are only few reports with results ranging from moderate to no improvement after GPi DBS.41 42 Also, we found 13 cases of DYT-KMT2B, an emerging form of childhood onset, generalised dystonia with prominent cervical, cranial and laryngeal involvement.43 44 While these patients had a variable extent of dystonia improvement after GPi DBS, sometimes with ‘dramatic’ amelioration in walking, scoliosis and dysphonia,44 none of them received pre-surgical versus post-surgical evaluations with validated clinical scales. Therefore, we could not include them in this analysis.

    Some limitations should be considered in the interpretation of our results. First, the limited number of cases limited the possibility of extending the meta-analysis beyond DYT-TOR1A, DYT-THAP1 and NBIA/DYT-PANK2. Second, the small number of cases available and heterogeneous clinical presentation of GNAO1, ACTB and CHOR/DYT-ADCY5 precludes the generalisability of our findings to the entire spectrum of clinical phenotype associated with these rare subtypes of monogenic dystonias. Third, we lacked patient-centred outcome measures, such as quality of life, which are critical to provide a comprehensive analysis of GPi DBS outcomes. Fourth, only few studies reported data on the accuracy of DBS lead placement. Thus, results could not be adjusted for this important clinical variable. Fifth, the lack of randomised clinical trials or blind prospective observational studies inevitably limited the strength of the data available for analyses. Finally, the effects of DBS on targets other than GPi, such as the subthalamic nucleus, could not be investigated due to paucity of data.

    These limitations notwithstanding, this study provides robust support for both short-term and long-term efficacy of GPi DBS in DYT-TOR1A, a modest but still significant improvement in DYT-THAP1 and NBIA/DYT-PANK2 patients, and promising but preliminary results for dystonia associated with DYT-SGCE, DYT/PARK-TAF1, ACTB and GNAO1 mutations. We also confirmed that dystonia duration prior to surgery and age at dystonia onset may differentially affect GPi DBS motor outcome in patients with DYT-TOR1A, DYT/PARK-TAF1 and DYT-SGCE mutations. We suggest that genetic testing should be employed in dystonia cases being considered for GPi DBS. Prospective clinical registries will be required to confirm these findings and further clarify the role of GPi DBS in patients with rare genetic forms of dystonia.

    Supplemental material

    References

    1. ↵
      2. Albanese A ,
      3. Bhatia K ,
      4. Bressman SB , et al
      . Phenomenology and classification of dystonia: a consensus update. Mov Disord 2013;28:863–73.doi:10.1002/mds.25475
      OpenUrlCrossRefPubMed
    2. ↵
      2. Girach A ,
      3. Vinagre Aragon A ,
      4. Zis P
      . Quality of life in idiopathic dystonia: a systematic review. J Neurol 2019;266:2897–906.doi:10.1007/s00415-018-9119-x
      OpenUrl
    3. ↵
      2. Lohmann K ,
      3. Klein C
      . Update on the genetics of dystonia. Curr Neurol Neurosci Rep 2017;17:26.doi:10.1007/s11910-017-0735-0
    4. ↵
      2. Jinnah HA ,
      3. Alterman R ,
      4. Klein C , et al
      . Deep brain stimulation for dystonia: a novel perspective on the value of genetic testing. J Neural Transm 2017;124:417–30.doi:10.1007/s00702-016-1656-9
      OpenUrl
    5. ↵
      2. Vidailhet M ,
      3. Vercueil L ,
      4. Houeto J-L , et al
      . Bilateral deep-brain stimulation of the globus pallidus in primary generalized dystonia. N Engl J Med 2005;352:459–67.doi:10.1056/NEJMoa042187
      OpenUrlCrossRefPubMedWeb of Science
    6. ↵
      2. Kupsch A ,
      3. Benecke R ,
      4. Müller J , et al
      . Pallidal deep-brain stimulation in primary generalized or segmental dystonia. N Engl J Med 2006;355:1978–90.doi:10.1056/NEJMoa063618
      OpenUrlCrossRefPubMedWeb of Science
    7. ↵
      2. Cheung T ,
      3. Noecker AM ,
      4. Alterman RL , et al
      . Defining a therapeutic target for pallidal deep brain stimulation for dystonia. Ann Neurol 2014;76:22–30.doi:10.1002/ana.24187
      OpenUrlCrossRefPubMed
    8. ↵
      2. Moro E ,
      3. LeReun C ,
      4. Krauss JK , et al
      . Efficacy of pallidal stimulation in isolated dystonia: a systematic review and meta-analysis. Eur J Neurol 2017;24:552–60.doi:10.1111/ene.13255
      OpenUrlCrossRefPubMed
    9. ↵
      2. YS W ,
      3. LH N ,
      4. Fan RM , et al
      . Meta-regression analysis of the long-term effects of pallidal and subthalamic deep brain stimulation for the treatment of isolated dystonia. World Neurosurg 2019;8750:31442–1.
      OpenUrl
    10. ↵
      2. Fox MD ,
      3. Alterman RL
      . Brain stimulation for torsion dystonia. JAMA Neurol 2015;72:713–9.doi:10.1001/jamaneurol.2015.51
      OpenUrl
    11. ↵
      2. Bronte-Stewart H ,
      3. Taira T ,
      4. Valldeoriola F , et al
      . Inclusion and exclusion criteria for DBS in dystonia. Mov Disord 2011;26:S5–16.doi:10.1002/mds.23482
      OpenUrlCrossRefPubMed
    12. ↵
      2. Moher D ,
      3. Liberati A ,
      4. Tetzlaff J , et al
      . Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. PLoS Med 2009;6:e1000097.doi:10.1371/journal.pmed.1000097
    13. ↵
      2. Stroup DF ,
      3. Berlin JA ,
      4. Morton SC , et al
      . Meta-analysis of observational studies in epidemiology (moose) group. JAMA 2000;283:2008–12.
      OpenUrlCrossRefPubMedWeb of Science
    14. ↵
      2. Burke RE ,
      3. Fahn S ,
      4. Marsden CD , et al
      . Validity and reliability of a rating scale for the primary torsion dystonias. Neurology 1985;35:73–7.doi:10.1212/WNL.35.1.73
      OpenUrlCrossRefPubMed
    15. ↵
      2. Higgins JPT ,
      3. Altman DG ,
      4. Gøtzsche PC , et al
      . The Cochrane collaboration's tool for assessing risk of bias in randomised trials. BMJ 2011;343:d5928.doi:10.1136/bmj.d5928
    16. ↵
      2. Sterne JAC ,
      3. Sutton AJ ,
      4. Ioannidis JPA , et al
      . Recommendations for examining and interpreting funnel plot asymmetry in meta-analyses of randomised controlled trials. BMJ 2011;343:d4002.doi:10.1136/bmj.d4002
    17. ↵
      2. DerSimonian R ,
      3. Laird N
      . Meta-analysis in clinical trials revisited. Contemp Clin Trials 2015;45:139–45.doi:10.1016/j.cct.2015.09.002
      OpenUrlCrossRefPubMed
    18. ↵
      2. Holloway KL ,
      3. Baron MS ,
      4. Brown R , et al
      . Deep brain stimulation for dystonia: a meta-analysis. Neuromodulation 2006;9:253–61.doi:10.1111/j.1525-1403.2006.00067.x
      OpenUrlPubMed
    19. ↵
      2. Isaias IU ,
      3. Alterman R ,
      4. Tagliati M
      . Outcome predictors for DBS in primary dystonia: the role of disease duration. Brain 2008;131:1895–902.
      OpenUrlCrossRefPubMedWeb of Science
    20. ↵
      2. Andrews C ,
      3. Aviles-Olmos I ,
      4. Hariz M , et al
      . Which patients with dystonia benefit from deep brain stimulation? A metaregression of individual patient outcomes. J Neurol Neurosurg Psychiatry 2010;81:1383–9.doi:10.1136/jnnp.2010.207993
      OpenUrlAbstract/FREE Full Text
    21. ↵
      2. Isaias IU ,
      3. Volkmann J ,
      4. Kupsch A , et al
      . Factors predicting protracted improvement after pallidal DBS for primary dystonia: the role of age and disease duration. J Neurol 2011;258:1469–76.doi:10.1007/s00415-011-5961-9
      OpenUrlCrossRefPubMedWeb of Science
    22. ↵
      2. Markun LC ,
      3. Starr PA ,
      4. Air EL , et al
      . Shorter disease duration correlates with improved long-term deep brain stimulation outcomes in young-onset DYT1 dystonia. Neurosurgery 2012;71:325–30.doi:10.1227/NEU.0b013e318258e21b
      OpenUrlCrossRefPubMedWeb of Science
    23. ↵
      2. Vloo P ,
      3. Lee DJ ,
      4. Dallapiazza RF , et al
      . Deep brain stimulation for pantothenate kinase‐associated neurodegeneration: a meta‐analysis. Mov Disord 2019;34:264–73.doi:10.1002/mds.27563
      OpenUrl
    24. ↵
      2. Bruggemann N ,
      3. Kuhn A ,
      4. Schneider SA , et al
      . Short- and long-term outcome of chronic pallidal neurostimulation in monogenic isolated dystonia. Neurology 2015;84:895–903.doi:10.1212/WNL.0000000000001312
      OpenUrlCrossRefPubMed
    25. ↵
      2. Carbon M ,
      3. Eidelberg D
      . Abnormal structure-function relationships in hereditary dystonia. Neuroscience 2009;164:220–9.doi:10.1016/j.neuroscience.2008.12.041
      OpenUrlCrossRefPubMedWeb of Science
    26. ↵
      2. Sadnicka A ,
      3. Teo J ,
      4. Kojovic M , et al
      . Genotype specific cerebellar involvement in DYT1 and DYT6 dystonia? J Neurol Neurosurg Psychiatry 2013;84:e2.67–e2.doi:10.1136/jnnp-2013-306573.159
      OpenUrl
    27. ↵
      2. LeDoux MS ,
      3. Xiao J ,
      4. Rudzińska M , et al
      . Genotype–phenotype correlations in THAP1 dystonia: molecular foundations and description of new cases. Parkinsonism Relat Disord 2012;18:414–25.doi:10.1016/j.parkreldis.2012.02.001
      OpenUrlCrossRefPubMedWeb of Science
    28. ↵
      2. Mills KA ,
      3. Starr PA ,
      4. Ostrem JL
      . Neuromodulation for dystonia: target and patient selection. Neurosurg Clin N Am 2014;25:59–75.doi:10.1016/j.nec.2013.08.014
      OpenUrl
    29. ↵
      2. Hayflick SJ ,
      3. Kurian MA ,
      4. Hogarth P
      . Neurodegeneration with brain iron accumulation. Handb Clin Neurol 2018;147:293–305.doi:10.1016/B978-0-444-63233-3.00019-1
      OpenUrl
    30. ↵
      2. Hanssen H ,
      3. Heldmann M ,
      4. Prasuhn J , et al
      . Basal ganglia and cerebellar pathology in X-linked dystonia-parkinsonism. Brain 2018;141:2995–3008.doi:10.1093/brain/awy222
      OpenUrl
    31. ↵
      2. Carecchio M ,
      3. Mencacci NE ,
      4. Iodice A , et al
      . ADCY5-related movement disorders: frequency, disease course and phenotypic variability in a cohort of paediatric patients. Parkinsonism Relat Disord 2017;41:37–43.doi:10.1016/j.parkreldis.2017.05.004
      OpenUrl
    32. ↵
      2. Vijiaratnam N ,
      3. Bhatia KP ,
      4. Lang AE , et al
      . ADCY5‐related dyskinesia: improving clinical detection of an evolving disorder. Mov Disord Clin Pract 2019. Accepted Author Manuscript.
    33. ↵
      2. Danti FR ,
      3. Galosi S ,
      4. Romani M , et al
      . GNAO1 encephalopathy: Broadening the phenotype and evaluating treatment and outcome. Neurol Genet 2017;3:e143.doi:10.1212/NXG.0000000000000143
    34. ↵
      2. Conboy E ,
      3. Vairo F ,
      4. Waggoner D , et al
      . Pathogenic Variant in ACTB, p.Arg183Trp, Causes Juvenile-Onset Dystonia, Hearing Loss, and Developmental Delay without Midline Malformation. Case Rep Genet 2017;2017:9184265.doi:10.1155/2017/9184265
    35. ↵
      2. Verloes A ,
      3. Di Donato N ,
      4. Masliah-Planchon J , et al
      . Baraitser–winter cerebrofrontofacial syndrome: delineation of the spectrum in 42 cases. Eur J Hum Genet 2015;23:292–301.doi:10.1038/ejhg.2014.95
      OpenUrlCrossRef
    36. ↵
      2. Brüggemann N ,
      3. Domingo A ,
      4. Rasche D , et al
      . Association of pallidal neurostimulation and outcome predictors with X-linked dystonia parkinsonism. JAMA Neurol 2019;76:211–6.doi:10.1001/jamaneurol.2018.3777
      OpenUrl
    37. ↵
      2. Rughani AI ,
      3. Lozano AM
      . Surgical treatment of myoclonus dystonia syndrome. Mov Disord 2013;28:282–7.doi:10.1002/mds.25326
      OpenUrlCrossRefPubMed
    38. ↵
      2. Roze E ,
      3. Lang AE ,
      4. Vidailhet M
      . Myoclonus-dystonia: classification, phenomenology, pathogenesis, and treatment. Curr Opin Neurol 2018;31:484–90.doi:10.1097/WCO.0000000000000577
      OpenUrl
    39. ↵
      2. Honey CM ,
      3. Malhotra AK ,
      4. Tarailo-Graovac M , et al
      . GNAO1 Mutation–Induced Pediatric Dystonic Storm Rescue With Pallidal Deep Brain Stimulation. J Child Neurol 2018;33:413–6.doi:10.1177/0883073818756134
      OpenUrlCrossRefPubMed
    40. ↵
      2. Menozzi E ,
      3. Balint B ,
      4. Latorre A , et al
      . Twenty years on: Myoclonus‐dystonia and ε‐sarcoglycan — neurodevelopment, channel, and signaling dysfunction. Mov Disord 2019;34:1588–601.doi:10.1002/mds.27822
      OpenUrl
    41. ↵
      2. Albanese A ,
      3. Di Giovanni M ,
      4. Amami P , et al
      . Failure of pallidal deep brain stimulation in DYT12-ATP1A3 dystonia. Parkinsonism Relat Disord 2017;45:99–100.doi:10.1016/j.parkreldis.2017.09.008
      OpenUrl
    42. ↵
      2. Fearon C ,
      3. McKinley J ,
      4. McCarthy A , et al
      . Failure of sequential pallidal and motor thalamus DBS for rapid-onset dystonia-parkinsonism (DYT12). Mov Disord Clin Pract 2018;5:444–5.doi:10.1002/mdc3.12559
      OpenUrl
    43. ↵
      2. Meyer E ,
      3. Carss KJ ,
      4. Rankin J , et al
      . Mutations in the histone methyltransferase gene KMT2B cause complex early-onset dystonia. Nat Genet 2017;49:223–37.doi:10.1038/ng.3740
      OpenUrl
    44. ↵
      2. Balint B ,
      3. Mencacci NE ,
      4. Valente EM , et al
      . Dystonia. Nat Rev Dis Primers 2018;4:25.doi:10.1038/s41572-018-0023-6
      OpenUrl

    Supplementary materials

    Footnotes

    • Contributors CAA: Design and conceptualised study; acquisition of data; drafted the manuscript for intellectual content. AKD: Analysed the data; interpreted the data; revised the manuscript for intellectual content. AR: Acquisition of data; interpreted the data; revised the manuscript for intellectual content. SB: Acquisition of data; interpreted the data; revised the manuscript for intellectual content. LM: Acquisition of data; interpreted the data; revised the manuscript for intellectual content. GI: Acquisition of data; interpreted the data; revised the manuscript for intellectual content. AS: Acquisition of data; interpreted the data; revised the manuscript for intellectual content. EK: Acquisition of data; interpreted the data; revised the manuscript for intellectual content. MZ: Revised the manuscript for intellectual content; interpreted the data. MFC: Revised the manuscript for intellectual content; interpreted the data. AF: Revised the manuscript for intellectual content; interpreted the data. MT: Revised the manuscript for intellectual content; interpreted the data. MO: Revised the manuscript for intellectual content; interpreted the data. AJE: Revised the manuscript for intellectual content; interpreted the data. LL: Revised the manuscript for intellectual content; interpreted the data. AM: Design and conceptualised study; interpreted the data; drafted the manuscript for intellectual content.

    • Funding CAA received travel grants from Zambon and Abbvie, and educational grants from Ralpharma. AD is supported as a co-investigator by the NIH (1R01HL125016-01), (1 R21 HL143030-01) and (1R21 AI133207) grants and as a collaborator in NIH R21 AI118228 grant. He has been also serving as a statistician in CPRIT grants (PP180003, PP170068, PP170004, PP140164, 140211, PP110156, PP150031 and PP130083), CCTST K12 (consultant) award, Coldwell (co-investigator) and TMF (co-investigator). AD is a director of Biostatistics & Epidemiology Consulting Lab at the TTUHSC EP. AR received grant support and speaker honoraria from AbbVie, speaker honoraria from Chiesi Farmaceutici and travel grants from Medtronic, Lusofarmaco and UCB Pharma. MZ received speaker's honoraria from Medtronic, Lundbeck, UCB Pharma and AbbVie. LL has received honoraria for lecturing and travel grants from, UCB Pharma, AbbVie, DOC, Zambon and Bial. MFC received travel support from Boston Scientific. Advisory board: Medtronic, Boston Scientific. Independent consultant for Medtronic for research and educational issues. Received a grant from the Stichting Parkinson Fonds. The DBS center of the Haga Teaching Hospital/LUMC received compensation for DBS training activities from Medtronic and an unrestricted educational grant from Medtronic. AF reports grants, personal fees and non-financial support from Abbvie, grants, personal fees and non-financial support from Boston Scientific, grants, personal fees and non-financial support from Medtronic, personal fees from Chiesi, personal fees and non-financial support from Ipsen, personal fees from UCB, grants and personal fees from Sunovion, outside the submitted work. MT received grant support from the Cure Parkinson Trust and Drown Foundation; personal compensation as a consultant/scientific advisory board member for Abbott, Boston Scientific, Medtronic and Revance; honoraria from Boston Scientific, Medtronic, the American Academy of Neurology. MSO serves as a consultant for the Parkinson’s Foundation, and has received research grants from 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. MSO DBS research is supported by: R01 NR014852 and R01NS096008. MSO has received royalties for publications with Demos, Manson, Amazon, Smashwords, Books4Patients, Perseus, Robert Rose, Oxford and Cambridge (movement disorders books). MSO is an associate editor for New England Journal of Medicine Journal Watch Neurology. MSO 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 MSO receives grants from Medtronic, Abbvie, Abbott and Allergan and the PI has no financial interest in these grants. MSO 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. AJE received grant support from the NIH, Great Lakes Neurotechnologies and the Michael J Fox Foundation; personal compensation as a consultant/scientific advisory board member for Abbvie, TEVA, Impax, Acadia, Acorda, InTrance, Cynapsus/Sunovion, Lundbeck, and USWorldMeds; publishing royalties from Lippincott Williams & Wilkins, Cambridge University Press, and Springer; and honoraria from Abbvie, UCB, USWorldMeds, Lundbeck, Acadia, the American Academy of Neurology, and the Movement Disorders Society. LL received honoraria for lecturing and travel grants from Medtronic, UCB Pharma, and AbbVie. AM is supported by NIH (KL2 TR001426) and received speaker honoraria from CSL Behring, Cynapsus Therapeutics, Theravance, Abbott, and AbbVie. He received grant support from Lundbeck.

    • Competing interests None declared.

    • Patient consent for publication Not required.

    • Provenance and peer review Not commissioned; externally peer reviewed.

    • Data availability statement All data relevant to the study are included in the article or uploaded as supplementary information.