patients with parkinson's disease dementia stay in the hospital twice as long as those without...

Botulinum Toxin Type A forLevodopa-Induced Cervical

Dyskinesias in Parkinson’s Disease:Unfavorable Risk-Benefit Ratio

Peak-dose levodopa (L-dopa)-induced dyskinesias (LIDs)are disabling side effects of long-term L-dopa therapy in Par-kinson’s disease (PD), affecting 45–85% of patients within afew years of treatment initiation.1 LIDs are a therapeuticchallenge for neurologists as limited treatment options exist.Functional chemodenervation with botulinum toxin type A(BTX-A, Botox, Allergan), first-line treatment for cervicaldystonia,2 may be considered a treatment strategy againstthis motor complication when predominantly affectingthe cervical region. Although the pathophysiology of LID isdistinct from that of CD, abnormal muscle contractionsof antagonistic sets of muscles suggest a phenomenologicaloverlap.3 We sought to determine the extent to whichcervical intramuscular BTX-A injections can reduce cervical-predominant LID in PD.Twelve consecutive L-dopa-treated PD patients with fre-

quent and bothersome cervical-predominant LID, regardlessof previous or concurrent antidyskinetic treatment, wereinvited to participate in this six-month double-blind, cross-over clinical trial. Patients were randomized to receive stand-ardized EMG-guided injections of BTX-A or equal-volumenormal saline (placebo) into the sternocleidomastoid (25 Ubilaterally), splenius capitis (50 U bilaterally, divided), andtrapezius (25 U bilaterally) muscles from a blinded injector.The total BTX-A dose of 200 U (10:1 dilution) was selectedas the average optimal dose used in cervical dystonia.4 Sub-jects were assessed always in their best ‘‘on’’ state, whenmaximal benefits and peak-dose LID were expected to occur.The timing for the greatest severity of LID was chosen atenrollment with patient and caregiver input, and all study-related assessments were carried out at the same time.Assessments occurred at 0, 1, 3 (crossover visit), 4, and 6months after enrollment, with blinded injections adminis-tered at the 0- and 3-month visits. Primary outcome measurewas change in the Goetz dyskinesia rating scale (GDRS, 0–4, higher is worse),5 modified for the cervical region, 1month after each injection (1- and 4-month study visits).

The GDRS was obtained off-line by three blinded ratersfrom randomized videotape sequences. Secondary outcomesincluded changes in the patient-reported clinical globalimpression of change (CGIC) and in the duration, severity,and pain of LID using items 32–34 of the UPDRS-IV. Non-parametric analyses to compare BTX-A with baseline andplacebo groups were carried out with the Friedman or Wil-coxon ranked test, as appropriate. A two-point difference inthe CGIC was prespecified as clinically significant. Institu-tional Review Board-approved informed consent was admin-istered to all participating subjects. The study was registeredin ClinicalTrials.gov as NCT00477802.

From 12 eligible PD patients, 8 patients (4 men; 70.36 5.6 years; disease duration, 11.1 6 2.7 years; ‘‘on’’UPDRS-III, 20.2; L-dopa-equivalent dose, 1375 6 601 mg)were randomized. Only 4 patients completed the 6-monthstudy before it was voluntarily stopped due to safety con-cerns, namely excessive neck weakness (Supporting Infor-mation Fig. 1, Consort flow diagram). BTX-A wasassociated with improved GDRS scores for the resting butnot action-induced dyskinesias (1.8 6 0.7 vs. 2.1 6 0.8,P ¼ 0.018; Table 1). There was a trend for reduction in‘‘on’’ time with LID in the BTX-A group compared withbaseline (UPDRS-IV, item 32, 1.7 6 1.0 vs. 2.5 6 1.0, P¼ 0.16). CGIC in the BTX-A group ranged from mildworsening to marked improvement. Dyskinesias self-ratedas moderately and severely disabling (UPDRS-IV, item 33,n ¼ 4/6) became either mildly (n ¼ 3/4) or not disablingwith BTX-A. This intervention also converted moderatelyand severely painful dyskinesias (UPDRS-IV, item 34, n ¼4/6) into mildly or not painful dyskinesias (n ¼ 3/4). Onaverage, the BTX-A group showed a one-point differenceabove placebo in the CGIC (mild improvement vs.unchanged, respectively), falling below the prespecified cut-off for clinical significance.

Transient neck weakness was the most frequently reportedadverse reaction (2 patients experiencing head drop andmoderate dysphagia for 2 weeks). Only 1 subject requestedBTX-A injections for ongoing poststudy management of hisLID, which, uniquely, were exclusively restricted to the cer-vical region.

Although BTX-A reduced cervical dyskinesia severity andpain, these benefits were outweighed by (1) the developmentof excessive neck weakness, leading to a larger-than-antici-pated dropout rate and forcing a premature termination ofthe trial; and (2) the lack of a favorable effect on the action-induced component of LID, which may be the most dis-abling component. This dichotomy of improvement on rest-ing but not action-induced dyskinesias is similar to theantidyskinetic response reported for clozapine.6 Overall, theadverse risk-benefit ratio uncovered by the study’s paradigmprecludes recommending the use of BTX-A for treatment ofcervical LID.

We recognize that in considering any future role of BTX-A for individual PD patients with focal LID, clinicians willhave to balance the discouraging outcome of this truncatedrandomized clinical trial with the observation that a

------------------------------------------------------------Additional Supporting Information may be found in the online versionof this article.Allergan, Inc. had no role in study design, data collection, dataanalysis, data interpretation, or writing of the report.

*Correspondence to: Alberto J. Espay, The Neuroscience Institute,Department of Neurology, Movement Disorders Center, University ofCincinnati, Cincinnati, Ohio, USA; [email protected]

Relevant conflicts of interest/financial disclosures: Nothing to report.Full financial disclosures and author roles may be found in the onlineversion of this article.

Published online 22 March 2011 in Wiley Online Library(wileyonlinelibrary.com). DOI: 10.1002/mds.23522

L E T T E R S T O T H E E D I T O R : N E W T O P I C S

Movement Disorders, Vol. 26, No. 5, 2011 913

satisfactory response may exist for this indication if LID arepredominantly cervical in distribution and dosages are tai-lored to prevent excessive weakness.

Acknowledgments: The authors thank Noel Jones for assistance on data

collection.

Alberto J. Espay, MD, MSc,1* Jennifer E. Vaughan, BPhil,1

Rakesh Shukla, PhD,2 Maureen Gartner, RN, MEd,1

Alok Sahay, MD,1 Fredy J. Revilla, MD,1

Fredy J. Revilla, MD,1 and Andrew P. Duker, MD1

1The Neuroscience Institute, Department of Neurology,Movement Disorders Center, University of Cincinnati,

Cincinnati, Ohio, USA; and 2Center for Biostatistical Services,Department of Environmental Health, University of Cincinnati,

Cincinnati, Ohio, USAE-mail: [email protected]

References1. Wagner ML, Fedak MN, Sage JI, Mark MH. Complications of dis-

ease and therapy: A comparison of younger and older patientswith Parkinson’s disease. Ann Clin Lab Sci 1996;26:389–395.

2. Comella CL, Jankovic J, Brin MF. Use of botulinum toxin type Ain the treatment of cervical dystonia. Neurology 2000;55:S15–S21.

3. Marconi R, Lefebvre-Caparros D, Bonnet AM, Vidailhet M, DuboisB, Agid Y. Levodopa-induced dyskinesias in Parkinson’s disease phe-nomenology and pathophysiology. Mov Disord 1994;9:2–12.

4. Jankovic J. Treatment of cervical dystonia with botulinum toxin.Mov Disord 2004;19 (Suppl 8):S109–S115.

5. Goetz CG, Stebbins GT, Shale HM, et al. Utility of an objectivedyskinesia rating scale for Parkinson’s disease: Inter- and intraraterreliability assessment. Mov Disord 1994;9:390–394.

6. Durif F, Debilly B, Galitzky M, et al. Clozapine improves dyskine-sias in Parkinson disease: A double-blind, placebo-controlled study.Neurology 2004;62:381–388.

Retinal Nerve Fiber Layer Loss inMultiple System Atrophy

The retina can be regarded as a ‘‘window to the centralnervous system’’ that might provide new insights into themultisystemic character of neurodegenerative diseases.1 Reti-nal nerve fibers (RNFs) can in particular be studied with op-tical coherence tomography (OCT), which allows fast,noninvasive imaging of retinal tissue with resolution reach-ing that of low-power histological sections.2 Latest genera-tion OCT devices are able to substantially reduce test–retestvariability by real time eye tracking and to automaticallysegment and quantify the RNF layer, providing immediatedata on the distribution and structural integrity of RNF tis-sue. We investigated whether multiple system atrophy(MSA) is associated with decreased RNF layer thickness(RNFLT) and if so, whether such an effect correlates withindices of disease severity or disease duration.

Ten MSA patients of cerebellar (MSA-C; n ¼ 6) and par-kinsonian subtype (MSA-P; n ¼ 4) meeting the consensuscriteria for MSA were prospectively recruited.3 All patientswere assessed for disease severity by the United MSA ratingscale (UMSARS).4 Ten age-matched healthy subjects servedas controls (7 men per group; for subject characteristics, seeSupporting Information Table 1).

Spectral domain OCT (SD-OCT) imaging was performedusing the Spectralis HRA þ OCT platform (Heidelberg Engi-neering, Heidelberg, Germany) with real time compensationfor eye movements (TruTrack).2 For quantification of centralretinal thickness (CRT) and RNFLT, we used the

Table 1. Clinical measurements

Variable Baseline (n ¼ 6) Placebo (n ¼ 5) Percent change (%) BTX-A (n ¼ 6) Percent change (%)

GDRS, at rest 1.5 6 1.0 1.4 6 1.1 �6.7 1.0 6 0.6 �33.3GDRS, on action 2.2 6 0.6 2.3 6 0.7 þ4.6 1.8 6 0.8 �17.9UPDRS-II (off) 18.3 6 3.8 17.0 6 3.2 �7.0 15.7 6 2.9 �14.3UPDRS-II (on) 7.5 6 4.0 7.6 6 4.8 þ1.3 6.3 6 2.8 �15.6UPDRS-III (on) 19.5 6 5.5 19.1 6 7.0 �2.1 16.7 6 5.2 �14.5UPDRS-IV 11.0 6 2.1 8.2 6 3.4 �25.5 7.5 6 1.4 �31.8Item 32 2.5 6 1.0 1.8 6 1.0 �30.0 1.7 6 1.0 �33.3Item 33 1.7 6 1.0 1.0 6 0.8 �40.0 0.7 6 0.5 �60.0Item 34 1.5 6 1.2 1.0 6 1.2 �33.3% 0.5 6 0.5 �66.7

PDQ-8 37.9 6 8.7 30.6 6 11.4 �19.3 25.0 6 11.2 �34.1CGIC-Investigator N/A 0.2 6 0.4 N/A 1.0 6 1.4 N/ACGIC-Patient N/A 0.0 6 0.7 N/A 1.0 6 1.4 N/A

Placebo and BTX-A columns reflect the 1-month blinded scores. GDRS: Goetz Dyskinesia Rating Scale; UPDRS: Unified Parkinson’s disease Rating Scale;PDQ-8: 8-item Parkinson’s disease Quality of Life Questionnaire; N/A: not applicable.

------------------------------------------------------------Additional Supporting Information may be found in the online versionof this article.

*Correspondence to: Daniela Berg, Department ofNeurodegeneration, Hertie Institute of Clinical Brain Research,Tubingen, Germany; [email protected]



914 Movement Disorders, Vol. 26, No. 5, 2011


proprietary Eye Explorer software (version 5.1). Briefly, wequantified the distance between the inner limiting membrane(ILM) and the outer border of the retinal pigment epitheliumof two linear B-Scans at the foveola (¼CRT) and betweenILM and ganglion cell layer along a circular B-Scan centeredon the optic disc (¼RNFLT).We used JMP V8.02 (SAS Institute, Cary, NC) for the

analysis of variance and to calculate Spearman correlationcoefficients between retinal parameters and UMSARS scoresof clinical severity. Data are shown as mean 6 SD, and a P-value of <0.05 was regarded as statistical significant.Data of the right eyes were chosen for analyses. Central

visual function as assessed by best-corrected visual acuity(logMAR) was unaffected in MSA patients (0.03 6 0.08,range ¼ �0.10 to 0.18) compared to healthy controls (0.006 0.08, range ¼ �0.10 to 0.10). Patients demonstrated areduced overall RNFLT compared with the control group(Fig. 1; globalMSA ¼ 91.30 6 1.45 lm vs. globalCTRL ¼97.20 6 1.45 lm; P ¼ 0.01). Nasal sectors were mostseverely affected (nasal-superiorMSA ¼ 87.60 6 2.78 lm vs.nasal-superiorCTRL ¼ 104.30 6 2.78 lm; nasalMSA ¼ 61.706 1.63 lm vs. nasalCTRL ¼ 71.10 6 1.63 lm; nasal-inferiorMSA ¼ 84.20 6 4.05 lm vs. nasal-inferiorCTRL ¼107.20 6 4.05 lm; all with P < 0.01, see also Fig. 1). Incontrast, differences in the temporal sectors did not reachstatistical significance (temporal-superiorMSA ¼ 134.50 63.29 lm vs. temporal-superiorCTRL ¼ 131.30 6 3.29 lm;

temporalMSA ¼ 83.40 6 3.25lm vs. temporalCTRL ¼ 75.306 3.25 lm; temporal-inferiorMSA ¼ 144.10 6 2.81 lm vs.temporal-inferiorCTRL ¼ 140.20 6 2.81 lm; all with P >0.05, see also Fig. 1). Likewise, CRT did not differ betweenMSA patients versus controls (CRTMSA ¼ 234.20 6 5.14 vs.CRTCTRL ¼ 232.30 6 5.14; P ¼ 0.80). Subgroup analysisfor MSA-C versus MSA-P did not demonstrate significantdifferences in RNFLT (P > 0.05, data not shown). RNFLTdid not correlate with disease severity (R2 ¼ 0.01) or diseaseduration (R2 ¼ 0.04; for single-subject data see SupportingInformation Table 2).

Our findings reveal that RNFL is significantly reduced inMSA, thus providing first evidence for retinal pathology inMSA and adding further support for the multisystemic na-ture of this neurodegenerative disorder. Reduction ofRNFLT most severely affects the nasal RNFs. This patternof RNFL thinning contrasts findings observed, for example,in Alzheimer’s disease (reduced RNFLT mainly in the supe-rior quadrant5) but partially resembles findings in Parkin-son’s disease (PD).6 Future OCT studies directly comparingMSA and PD patients have to confirm our preliminary find-ings and, in particular, identify potential overlaps as well asdiscriminative specificities in the respective RNFL pattern ofMSA and PD patients.

MSA patients featured normal visual acuity, which isconsistent with a nasal RNFLT reduction, as the nasalaxons are responsible for relaying the visual input

FIG. 1. Analysis of variance of retinal nerve fiber layer thickness (RNFLT) in age matched control subjects (n 5 10) and multiple system atrophypatients (n 5 10). Center of diamonds represent mean, tops and bottoms the 95% confidence interval.



originating from the peripheral rather than central retina.Correspondingly, VEP was also reported to be normal inMSA patients as it measures activity mainly originatingfrom the central visual field.7 Future studies will have todetermine whether MSA patients show predominantly pe-ripheral visual field defects.

M. Dominik Fischer, MD,1 Matthis Synofzik, MD,2,3 RobertHeidlauf, MD1, Julia Schicks, MD,2,3 Karin Srulijes, MD,2,3

Christoph Kernstock, MD1, Daniela Berg, MD,2,3

Ludger Schols, MD,2,3 and Ulrich Schiefer, MD1

1Centre for Ophthalmology, Institute for OphthalmicResearch, 2Department of Neurodegeneration, Hertie Institute

of Clinical Brain Research and 3German Center ofNeurodegenerative Diseases (DZNE), University of Tubingen,

Tubingen, Germany*E-mail: [email protected]

References1. Miller N, Drachman DA. The optic nerve: A window into diseases

of the brain? Neurology 2006;67:1742–1743.

2. Baleanu D, Tornow RP, Horn FK, et al. Retinal nerve fiber layerthickness in normals measured by spectral domain OCT. J Glaucoma2009.

3. Gilman S, Low PA, Quinn N, et al. Consensus statement on thediagnosis of multiple system atrophy. J Neurol Sci 1999;163:94–98.

4. Geser F, Seppi K, Stampfer-Kountchev M, et al. The clinical pre-sentation of multiple system atrophy (MSA) in Europe: An interimanalysis of the EMSA-SG (European MSA-Study Group) Registry.Mov Disord 2004;19:347.

5. Berisha F, Feke GT, Trempe CL, McMeel JW, Schepens CL. Retinalabnormalities in early Alzheimer’s disease. Invest Ophthalmol VisSci 2007;48:2285–2289.

6. Hajee ME, March WF, Lazzaro DR, et al. Inner retinal layer thin-ning in Parkinson disease. Arch Ophthalmol 2009;127:737–741.

7. Delalande I, Hache J, Forzy G, et al. Do visual-evoked potentials andspatiotemporal contrast sensitivity help to distinguish idiopathic Parkin-son’s disease and multiple system atrophy? Mov Disord 1998;13:446–452.

Interaction Between HemeOxygenase-1 Genotypes

and Exposure to Pesticides inParkinson’s Disease

Parkinson’s disease (PD) results from a combination ofgenetic and environmental factors,1 and there is some evi-

------------------------------------------------------------*Correspondence to: Jon Infante, University Hospital ‘‘Marques deValdecilla’’ University of Cantabria (UC), CIBERNED Santander, Spain;[email protected]



dence that pesticide exposure might particularly affect genet-ically susceptible individuals.2–4 Oxidative stress induced bythe pesticide Paraquat has been shown to upregulate theantioxidant protein heme oxygenase-1 (HO-1), the overex-pression of HO-1 protecting dopaminergic neurons againstMPPþ neurotoxicity.5 We assessed whether the interactionbetween exposure to pesticides and HO-1 (�413,rs2071746) genotypes might be a risk factor for PD.

The study included 237 patients with PD (45.5%women; mean age, 65 years; range, 35–83 years). Patientsfulfilling UK PD Society Brain Bank criteria for PD wereconsecutively recruited from outpatients at the UniversityHospital ‘‘Marques de Valdecilla’’ in Santander, NorthernSpain. Carriers of mutations in either LRRK2 or Parkingenes were excluded. Control group consisted of 203 unre-lated individuals (59.9% women; mean age, 73 years;range, 35–90), comprising 105 healthy spouses of neurolog-ical patients admitted to the same institution and 98healthy individuals randomly selected from a nursing home.Both patients and controls were Caucasians, originatingfrom a limited geographic area in northern Spain. Thestudy was approved by the local ethical committee. Geno-typing of HO-1 (�413, rs2071746) was performed by aTaq-Man single-nucleotide-polymorphism assay (AppliedBiosystems, Warrington, Cheshire, UK) and an ABI PRISM7000 sequence detection system (Applied Biosystems). Pesti-cide exposure was assessed through personal interview. Par-ticipants who reported applying any pesticides (insecticides,herbicides, or fungicides) for work or gardening prior to thereference age were considered ever exposed. Participantswho reported never applying any pesticides were considerednever exposed. Association between dichotomous variableswas analyzed with odds ratio, and 95% confidence intervalswere estimated by the Cornfield method or the exactmethod; P values for association were estimated by Waldtest. Hardy-Weinberg equilibrium was tested by v2 test.Interaction analysis was assessed by logistic regression usingSPSS 15.0 for windows software 8, SPSS, Inc., Chicago, IL.

Control group for HO-1 (�413, rs2071746) (P ¼ 0.20)was in Hardy-Weinberg equilibrium. Neither exposure topesticides nor being HO-1 (�413) TT genotype carrierincreased the risk for PD in the univariant analysis (P¼ 0.10and P ¼ 0.16, respectively). However, when consideringsimultaneously both factors, exposure to pesticides in thegroup of HO-1 (�413) TT carriers significantly increased infivefold the risk for PD (OR 5.49; P ¼ 0.03; Table 1).

Gene–environment interactions might explain, at least inpart, the conflicting results reported in some epidemiologicalstudies exploring an association between pesticide use andPD1; indeed, we did not observe an association of pesticideuse with PD risk until we considered HO-1 genotypes. Westudied a polymorphism in the HO-1 promoter region(�413, rs2071746) with a functional effect, the HO-1(�413, rs2071746) T allele having significantly lower tran-scriptional activity and therefore a less effective antioxidantresponse, than the HO-1 (�413, rs2071746) A allele pro-moter.6 We report for the first time that exposure to pesti-cides increased PD risk only in those individualshomozygous for the HO-1 (�413, rs2071746) T allele thatconfers a less effective antioxidant response. Our studyexpands the number of reports demonstrating gene–pesticideinteractions in PD risk2–4; as many of these reported genes



are involved in the oxidative stress pathway, the interactionbetween environmental and genetic factors that modify oxi-dative stress balance might account for individual PDsusceptibility.

Jon Infante, MD,1* Marıa Sierra, MD1

Pascual Sanchez-Juan, MD,1 Ines Garcıa-Gorostiaga, MD,2

Isabel Gonzalez-Aramburu, MD,1

Carlos Fernandez-Viadero, MD,3 Jose Berciano, MD,1

and Onofre Combarros, MD1

1Service of Neurology, University Hospital‘‘Marques de Valdecilla’’, University of Cantabria

(UC), CIBERNED, Santander, Spain

2Service of Neurology, Galdakao Hospital,Galdakao, Spain

3Parayas Psychiatric Hospital, Cantabria, Spain*E-mail: [email protected]

References1. Benmoyal-Segal L, Soreq H. Gene-environment interactions in spo-

radic Parkinson’s disease. J Neurochem 2006;97:1740–1755.

2. Hancock DB, Martin ER, Vance JM, Scott WK. Nitric oxide syn-thase genes and their interactions with environmental factors inParkinson’s disease. Neurogenetics 2008;9:249–262.

3. Dutheil F, Beaune P, Tzourio C, Loriot MA, Elbaz A.Interaction between ABCB1 and professional exposure to organo-chlorine insecticides in Parkinson disease. Arch Neurol 2010;67:739–745.

4. Kiyohara C, Miyake Y, Koyanagi M, et al.; for the FukuokaKinki Parkinson’s Disease Study Group. GST polymorphisms, interac-tion with smoking and pesticide use, and risk for Parkinson’s diseasein a Japanese population. Parkinsonism Relat Disord, 2010;16:447–452

5. Olesen BT, Clausen J, Vang O. Characterization of the transcrip-tional profile in primary astrocytes after oxidative stress inducedby Paraquat. Neurotoxicology 2008;29:13–21.

6. Ono K, Goto Y, Takagi S, et al. A promoter variant of the hemeoxygenase-1 gene may reduce the incidence of ischemic heart dis-ease in Japanese. Atherosclerosis 2004;173:315–319.

Immediate Response toBotulinum Toxin Injections inPatients with Fixed Dystonia

Fixed dystonia is characterized by a fixed abnormal pos-ture of the affected limb,1 typically precipitated by a minorperipheral trauma.1 Some patients may fulfill criteria forchronic regional pain syndrome (CRPS).1 Disability can besevere, and standard treatments are generally unsuccessful.1,2

Intrathecal baclofen has been reported to be effective insome such cases.3,4 There is continued debate whether thesecases have an organic or psychogenic basis. Here we reportand demonstrate on video the dramatic and immediateresponse to botulinum toxin in 3 patients with typical fixeddystonia.

CasesCase 1

This 43-year-old woman had a 3-year history of a fixedabnormal posture and severe pain in the left foot follow-ing traction for low back pain. She had no voluntarymovement at the left ankle and minimal movement of thetoes. No passive movement was possible at the left ankle.Distal sensory loss, allodynia, and a dusky color of thefoot were noted. She received an injection of 100 U ofbotulinum toxin type A (Dysport, Ipsen, UK) into the leftgastrocnemius and 30 U into the right teres anterior.There was immediate (within minutes) relaxation of thefoot posture, and she was able to walk unsupported withthe left foot almost flat on the floor (see Video). Thiseffect persisted for approximately 2 weeks before gradu-ally returning to baseline. She is currently awaiting admis-sion to an inpatient multidisciplinary program involvingphysical and psychological therapy.

Table 1. Odds ratios for PD risk according to interaction between pesticide exposure and HO-1 genotypes

Pesticide exposure HO-1 (rs2071746) TT genotype Patients with PD Controls OR 95% CI* P* OR 95% CI** P**

� � 134 133 1 (Reference) 1 (Reference)� þ 54 38 1.41 (0.87–2.27) 0.16 1.35 (0.83–2.19) 0.22þ � 35 21 1.65 (0.91–2.99) 0.10 1.71 (0.94–3.13) 0.07þ þ 11 2 5.45 (1.18–25.1) 0.03 5.49 (1.17–25.6) 0.03

HO-1 (rs2071746) TT (�) ¼ AA þ AT.*Crude odds ratios.**Odds ratios adjusting by age and sex (using multiple logistic regression).

------------------------------------------------------------*Correspondence to: Kailash P. Bhatia, Sobell Department of MotorNeuroscience and Movement Disorders, Institute of Neurology, UCL,Queen Square, London, United Kingdom; [email protected]





Case 2

This 44-year-old woman had a 9-month history of persis-tent in-turning of the right foot following lumbar disc sur-gery. Imaging revealed a good surgical result from thediscectomy. On examination, there was a fixed in-turning ofthe right foot, and no active or passive movement was possi-ble. There was patchy alteration in sensation and allodynia,and the foot was cold and dusky with flaking of toenails.One hundred units of botulinum toxin type A (Dysport,Ipsen) was injected into the right gastrocnemius, with imme-diate resolution of the abnormal posture (see Video). Shewas able to walk with the foot almost flat on the floor. Thiseffect persisted at follow-up 5 months later, although shestill needed a stick to help, and there was persistent flexionof toes on the left.

Case 3

This 48-year-old woman had developed a fixed abnormalposture of the left hand with severe pain 3 years previouslyafter sleeping awkwardly on her arm. On examination, shehad a fixed flexed posture of the fingers of the left handand neither passive nor active extension of the fingers.There was allodynia and hyperesthesia. Twenty units ofbotulinum toxin type A (Dysport, Ipsen) was injected intothe left hand divided between the second, third, and fourthlumbricals. Immediate relaxation of the abnormal postureand easing of her pain occurred. The posture recurred after2 weeks. The patient declined further follow-up or repeatinjections.

DiscussionThese cases illustrate that some patients with fixed dysto-

nia can respond immediately and dramatically to injectionsof botulinum toxin. The neuromuscular blocking actions ofbotulinum toxin take at least 72 hours to have a clinicaleffect. We therefore surmise that the dramatic responsewithin minutes of injection was due to the expectation ofthe patient of benefit: a placebo effect.The mere existence of a placebo effect does not auto-

matically suggest that the disorder is a psychogenic one.However, compared with the degree of improvement seenin the placebo arm of studies using botulinum toxin in typi-cal primary dystonia,5,6 the improvement seen in our casesis clearly dramatic. This finding calls into question theresults of previous trials of treatment in fixed dystonia.Although trials of intrathecal baclofen in patients withfixed dystonia and CRPS were said to be placebo controlled(by giving test doses of either placebo or baclofen3), it ispossible that there was systematic unblinding of patientsbecause of the systemic effects typically produced by an in-trathecal injection of baclofen. We urge caution in theinterpretation of the usefulness of open-label or inad-equately controlled treatment trials that are said to be ben-eficial in this group of patients.We do not believe that these patients are malingering,

and we appreciate the real and severe disability that theysuffer. However, these data, particularly if confirmed in alarger cohort of patients, need to be incorporated intoany model of the pathophysiology of this enigmaticcondition.

Video 1. Case 1—patient is in wheelchair with a fixedabnormal posture of the left foot that does not move volun-tarily or passively.

Video 2. Case 1—immediately following botulinum toxininjections, the patient can walk with a near-normal postureof the left foot.

Video 3. Case 2—fixed abnormal posture of the right footat the ankle with splaying of the toes. Foot is fixed to activeand passive movement.

Video 4. Case 2—immediately following botulinum toxininjections, the patient can walk with a significantly improvedposture of the right foot.

Video 5. Case 3—fixed abnormal posture of the left handwith no active or passive movement at the fingers or wrist.

Video 6. Case 3—immediately following botulinum toxininjections, the patient can actively move the fingers and wristwith a full range of movement.

Mark J. Edwards, PhD,1 Kailash P. Bhatia, MD,1*and Carla Cordivari, MD2

1Sobell Department of Motor Neuroscience andMovement Disorders, Institute of Neurology

UCL, Queen Square, London, United Kingdom; and2Department of Clinical Neurophysiology

National Hospital for Neurology and NeurosurgeryQueen Square, London, United Kingdom

*E-mail: [email protected]

References1. Schrag A, Trimble M, Quinn N, Bhatia K. The syndrome of fixed

dystonia: an evaluation of 103 patients. Brain. 20042007;127:2360–2372.

2. Ibrahim NM, Martino D, van de Warrenburg BP, et al. The prog-nosis of fixed dystonia: a follow-up study. Parkinsonism Relat Dis-ord. 2009;15:592–597.

3. van Hilten BJ, van de Beek WJ, Hoff JI, Voormolen JH, DelhaasEM. Intrathecal baclofen for the treatment of dystonia in patientswith reflex sympathetic dystrophy. N Engl J Med. 2000;343:625–630.

4. van Rijn MA, Munts AG, Marinus J, et al. Intrathecal baclofen fordystonia of complex regional pain syndrome. Pain. 2009;143:41–47.

5. Truong D, Duane DD, Jankovic J, et al. Efficacy and safety of bot-ulinum type A toxin (Dysport) in cervical dystonia: results of thefirst US randomized, double-blind, placebo-controlled study. MovDisord. 2005;20:783–791.

6. Greene P, Kang U, Fahn S, Brin M, Moskowitz C, Flaster E. Dou-ble-blind, placebo-controlled trial of botulinum toxin injections forthe treatment of spasmodic torticollis. Neurology. 1990;40:1213–1218.



Patients with Parkinson’s DiseaseDementia Stay in the Hospital Twice as

Long as Those without Dementia

It is well recognized that patients with dementia have on av-erage longer hospital admissions than those without. Dementiacare is expensive; the Alzheimer’s Society estimates that at least£80 million a year could be saved within the UK health system ifhospital length of stay (LOS) of patients with dementia wasreduced by a week,1 and there is currently an intense drive bylocal health authorities to reduce hospital length of stay indemented patients toward that of nondemented patients.2

Dementia in Parkinson’s disease (PD) is an advanced-stagecomplication, and it is therefore likely that PD patients withdementia (PD-D) may spend longer in the hospital thanthose without. Although several studies have shownincreased length of stay of PD patients relative to controls,none so far has compared the length of stay of PD patientswith that of PD-D patients.We therefore conducted a retrospective study comparing

the LOS of the 2 groups of PD patients admitted to a largeLondon teaching hospital between April and December 2009.The electronic discharge summary of every patient with a cod-ing of ‘‘parkinsonism’’ was reviewed, and any patient with idi-opathic PD was included. Admission date and reason, lengthof stay, and any mention of dementia (on any recorded admis-sion) within the summary were recorded. In addition, dis-charge prescriptions for every admission were checked forcholinesterase inhibitors, as a further marker of dementia.We recorded a total of 201 admissions for 147 patients, of

which 53 were of PD-D patients and 148 were of PD patients.The PD-D group stayed significantly longer than the PD group(18.6 and 9.1 days, respectively; P ¼ .00005). The averagelength of stay for all patients was 11.6 days, which is similarto the result (11.3 days) of a recent audit of the health trustconducted externally by an external management consultancy(unpublished data). This is lower than the national average of19.2 days.3 Therefore, the difference in length of stay betweenthe PD-D and the PD groups is likely to reflect the nature ofthe disease rather than any institutional management issues.The average LOS was greatest for patients admitted with

a PD-related problem (20.3 and 11.6 days for PD-D and PDpatients, respectively) and was the only statistically signifi-cant difference between the 2 groups (P ¼ .004). Within thePD-related admissions, the longest admissions occurred forpatients admitted with dysphagia followed by infection forboth PD and PD-D groups (Table 1).We also examined the number of admissions by each admit-

ting specialty. Sixty percent of admissions were directly relatedto a PD-related complication, with falls the most common rea-

son in both groups (see Table 1), which is the same as that foundin a previous study from the Royal Adelade Hospital in Aus-tralia.4 Cardiovascular-related problems were the next maincause, which is also similar to previous studies4,5 and likelyreflects the general incidence of heart disease in the community.

The main finding of this study is that patients with PD-Dstay in the hospital twice as long as PD patients. We hopethis study will help to increase awareness of health author-ities and physicians and aid in the development of new strat-egies to reduce the admissions and length of stay PD-Dpatients, both to improve the quality of life of these patientsand to reduce the financial pressure on society.

Phillip Fletcher,* Alison Leake, and Marie-Helene Marion

Atkinson Morley Neuroscience Centre, St.George’s Hospital, London, United Kingdom*E-mail: [email protected]

References1. Lakey L. Counting the cost—caring for people with dementia on

hospital wards. Alzheimer’s Society, 2009. Available at: http://www.alzheimers.org.uk/site/scripts/download_info.php?fileID¼787.

2. Available at: http://www.healthcareforlondon.nhs.uk/assets/Mental-health/08-Dementia-Outcomes-v1.0.pdf.

3. Hospital Episode statistics. Available at: http://www.hesonline.ns.uk.

4. Temlett JA, Thompson PD. Reasons for admission to hospital forParkinson’s disease. Intern Med J. 2006;36:524–526.

5. Woodford H,Walker R.Emergency hospital admissions in idio-pathic Parkinson’s disease. Mov Disord. 2005;20:1104–1108.

------------------------------------------------------------*Correspondence to: Phillip Fletcher, Atkinson Morley NeuroscienceCentre, St. George’s Hospital, London, United Kingdom;[email protected]



Table 1. Length of stay (LOS) in days and number ofadmissions (n) by main admission reason for patientswith Parkinson’s disease (PD) and Parkinson’s disease

dementia (PD-D)

Admission reason

PD PD-D

P valueLOS n LOS n

Gastroenterology 7.1 13 17 3 .08Neurology 15.3 7 14 1 —Other infection 11.5 2 4 1 .12Urology 5.5 14 7.5 3 .29Cardiovascular 4.5 37 2 2 .25PD related 11.6 76 20.3 44 .00005a

Dysphagia 24.6 5 27.8 4 .3Urine infection 11.6 16 26 8 .04a

Pulmonary infection 15.2 11 11.3 3 .38Falls 10.7 26 24.5 11 .02a

Dyskinesia 5.5 2 — — —Worsening mobility 10.7 9 23.6 7 .1Dementia — — 10.3 9 —Postural hypotension 5.3 5 8 1 —Back pain 0.5 2 8 1 —

Admissions directly related to PD are broken down by main problem. P valuesare shown for the difference in LOS between groups (astatistically significant).



Acute Effects of Whole-BodyVibration at 3, 6, and 9 Hz on Balanceand Gait in Patients with Parkinson’s

Disease

Vibration as a stimulus to treat Parkinson’s disease (PD)patients was first recommended by Charcot in 1892,1 andalthough this approach was subsequently abandoned,recently, whole-body vibration (WBV) protocols have beensuggested as a modern substitute.2–5

A small number of studies have assessed the effects of WBVon motor symptoms in PD, with apparently positive results,obtained in open-trial designs. In each case, however, only afixed frequency of stimulation was used, and no placebo groupwas included.2,3 It is therefore of fundamental interest toexamine a range of ‘‘doses’’ (in this case, frequencies) in orderto optimize potential therapeutic effects. Using an appropriateplacebo-controlled experimental design, we have explored theuse of different vibration frequencies, some of which had al-ready been reported to have an effect after 1 session.2,3

Forty-eight patients with PD diagnosed as idiopathic par-ticipated in this study. Possible participants were excluded ifany other disease or impairment potentially affected thevalidity of the results, and selected PD were naive to WBVprotocols. Patients were randomly allocated to each of 4groups: placebo, vibration at 3 Hz, vibration at 6 Hz, andvibration at 9 Hz (n ¼ 12 each).The protocol followed previous work reporting excellent

results of single-session WBV3 by means of a vibrating plat-form comprising 5 vibration sets of 10 each (interset restperiod, 10). The stepped platform (amplitude of 13 mm)thrusts the right and left legs upward alternately.4,5 Duringstimulation, patients stood on the platform with their feetseparated at a stable and comfortable position and with theknees slightly flexed. In the placebo group, patients adoptedthe same posture without vibration. Instead, they wererequired to stay still, trying to minimize hip oscillation; thiscontrolled for a placebo effect.5 Stimulation and evaluationwere done during ON periods. The protocol conformed tothe Declaration of Helsinki.The effect of vibration on gait and balance was evaluated

by the timed-up-and-go (TUG) and functional reach (FR)tests. Patients were evaluated just before (PRE), after(POST), and 48 hours after (POST-2) stimulation. Exam-iners were blind to protocol and group assignment.

—FR: Subjects had to displace, as far as they could,the movable stick of a Harpenden Anthropometer,pushing with their fists, without flexing the knees or

------------------------------------------------------------*Correspondence to: Javier Cudeiro, Neuroscience and Motor ControlGroup (NEUROcom), Deparment of Medicine, University of A Coruna,A Coruna, Spain; [email protected]



Table 1. Differences between groups in the pretest(shaded) and effect of the protocol (unshaded)

Functional reach (mm) Timed up and go (s)

Placebo group 237.90 (665.48) 16.39 (67.28)3-Hz group 260.99 (657.78) 14.62 (62.60)6-Hz group 237.51 (670.95) 13.38 (63.25)9-Hz group 264.79 (648.20) 15.61 (62.57)

Pretest Posttest 1 Posttest 2

Functionalreach (mm)

250.30 (660.59) 264.70 (661.20) 255.42 (665.97)

Timed upand go (s)

15.00 (64.39) 14.15 (63.79) 13.75 (64.13)

The shaded part of the table shows the values of each group in thepretest. No significant differences were seen between groups in FR or TUG(see Results section). The unshaded part of the table shows the effect ofthe protocol on the analyzed variables. Results were pooled for the 4groups, given the lack of significant interaction of evaluation � group. Thisdemonstrates that the effect of the protocol equally affected all 4 groups,therefore suggesting no vibration frequency was superior to the placebostimulation.

lifting the heels. The FR distance was obtained fromthe difference between the starting and end posi-tions of the stick. Each subject did so 3 times witheach arm, with 30-second rests between trials. Thevariable analyzed was FR distance.—TUG: Patients had to stand up from a chair, walk3 m using their preferred pattern, turn, come back,and sit down again. The time taken was recorded,and this was the analyzed variable. Patients per-formed this task 3 times.

Differences between groups at PRE were assessed by a1-way-ANOVA. A possible WBV effect was evaluated by a3 � 4 ANOVA with repeated measures. The within-subjectsfactor was evaluation (PRE, POST, POST-2); the between-subjects factor was group (3 Hz, 6 Hz, 9 Hz, placebo). Nor-mality was checked by the 1-sample Kolmogorov–Smirnovtest. Significance was set at P < .05.

Before stimulation, groups were comparable for FR dis-tance (F3,44 ¼ 0.685, P ¼ .566) and TUG (F3,44 ¼ 1.052, P¼ .379). Stimulation led FR to increase (F1.590,69.968 ¼4.255, P ¼ .026) and to reduce the time at TUG (F2,88 ¼14.128, P � .001). However, the lack of significant interac-tion in evaluation � group showed that none of the vibra-tion frequencies had an effect different from the placebo:F4.771,69.968 ¼ 0.717, P ¼ .606 for FR; F6,88 ¼ 1.332, P ¼.251 for TUG (Table 1).

In summary, previous work has shown that a single sessionof WBV can appear to induce significant short-term improve-ment on postural stability in PD,2,3 an effect subsequentlyconfirmed by others using different protocols.4 However, therole of placebo was either not investigated2,3 or unsatisfacto-rily controlled.4 We have shown here that a single session ofWBV with 3 frequencies does not have acute effects on gaitand balance of PD patients that are different than the effectsof a placebo. However, the lack of significant improvementsshould not mean complete rejection of the use of WBV in the



search for parkinsonian therapies: some reports suggest fur-ther analysis of the potential roles of the different vibrationsprotocols.6 Future designs involving WBV should be devisedin order to explore other parameters, such as application du-ration, number of sets, frequency, and the nature of vibration;however, in all cases, it is essential that the effect of the pla-cebo be properly controlled for.

Acknowledgments: We are indebted to Dr. K. L. Grieve for his valua-ble comments and corrections on the manuscript.

Marcelo Chouza, PhD,1 Pablo Arias, PhD,2

Susana Vinas, PhD,1 and Javier Cudeiro, MD, PhD3*

1Neuroscience and Motor Control Group(NEUROcom), Department of Physical Therapy,

University of A Coruna, A Coruna, Spain;2Neuroscience and Motor Control Group

(NEUROcom), Department of Physical Educationand Sport Science, University of A Coruna, A

Coruna, Spain; 3Neuroscience and Motor ControlGroup (NEUROcom), Department of Medicine,

University of A Coruna, A Coruna, Spain*E-mail: [email protected]

References1. Charcot JM. La medicine vibratoire: application des vibrations

rapides et continues au traitement de quelques maladies du systemenerveux. Prog Med. 1892;16:149–151.

2. Turbanski S, Haas CT, Schmidtbleicher D, et al. Effects of randomwhole-body vibration on postural control in Parkinson’s disease.Res Sports Med. 2005;13:243–256.

3. Haas CT, Turbanski S, Kessler K, Schmidtbleicher D. The effectsof random whole-body-vibration on motor symptoms in Parkin-son’s disease. NeuroRehabilitation. 2006;21:29–36.

4. Ebersbach G, Edler D, Kaufhold O, Wissel J. Whole body vibrationversus conventional physiotherapy to improve balance and gait inParkinson’s disease. Arch Phys Med Rehabil. 2008;89:399–403.

5. Arias P, Chouza M, Vivas J, Cudeiro J. Effect of whole bodyvibration in Parkinson’s disease: a controlled study. Mov Disord.2009;24:891–898.

6. Goetz CG. Jean-Martin Charcot and his vibratory chair for Parkin-son disease. Neurology. 2009;73:475–478.

Treatment of a Dystonic Storm WithPallidal Stimulation in a Patient With

PANK2 Mutation

Pantothenate kinase–associated neurodegeneration (PKAN)is a progressive autosomal recessive neurological disorderlinked to PANK2 gene mutations. PKAN accounts for

------------------------------------------------------------*Correspondence to: Francisco Grandas, Movement DisordersResearch Unit, Hospital General Universitario Gregorio Maranon,Madrid; [email protected]



approximately 50% of cases of neurodegeneration with brainiron accumulation and is associated with changes in the basalganglia on magnetic resonance imaging (MRI), particularlythe eye-of-the-tiger sign—a globus pallidus central hyperinten-sity with surrounding hypointensity on T2-weighted images—regarded as highly specific of PANK2 mutations.1 Althoughthere is wide phenotypic heterogeneity in PKAN cases, severegeneralized dystonia is frequently a prominent clinical sign,which can be very disabling, impair gait, and sometimes leadto life-threatening dystonic storms.2

At present, there is no effective pharmacological treatmentfor this disorder. Several case reports2–5 and 2 small series6,7

of PKAN patients with dystonia treated with deep brainstimulation (DBS) of the internal globus pallidus (GPi) havebeen published. Clinical outcome was mostly favorable, withresults ranging from marked sustained improvement6 tomodest improvement lasting only a few months.7 Althoughmany of the reported PKAN patients experienced severedystonia, data on the effect of DBS of the GPi in PKANdystonic storms are lacking.

We report on the effectiveness of DBS of the GPi in treat-ing dystonic storm in a patient with PKAN who initially didnot show the eye-of-the-tiger sign on MRI.

A 19-year-old man, the son of consanguineous parents

and the third of 5 children, had normal neurological devel-

opment until the age of 13 years, when he began to notice

tremor in both hands. Four years later, left foot dystonia

appeared, followed by dystonic movements of the fingers of

the left hand. Dystonia became generalized within 6 months

with prominent extensor trunk spasms. The left foot adopted

a permanent varus position, and the patient was unable to

walk and became wheelchair-ridden.

Examination revealed generalized dystonia involving the

face, neck, trunk, and left extremities, as well as postural

tremor of both upper limbs. The remainder of the neurologi-

cal and general examination was unremarkable. An ophthal-

mologic evaluation was normal, and neuropsychological

tests showed only mild executive dysfunction. A bilateral

hypointense globus pallidus signal with no central hyperin-

tensity was obseved on T2-weighted MRI images (Fig. 1A),

suggesting brain iron accumulation. Plasma concentrations

of ceruloplasmin, ferritin, and lipoproteins were normal. No

acanthocytes were found in a peripheral blood smear, and

no axonal spheroids were observed in a sural nerve biopsy.

A homozygous 1021C!T mutation (p.T237M) was found

in exon 6 of the PANK2 gene.Treatment with clonazepam, baclofen, levodopa, anticholi-

nergics, tetrabenazine, and botulinum toxin was ineffective.Dystonia worsened, leading, with no apparent precipitatingfactors, to status dystonicus with continuous repetitive violentopisthotonic axial spasms (Burke-Fahn-Marsden DystoniaRating Scale scores of 96/120 and 29/30 for movement anddisability, respectively) that did not improve with intravenousinfusions of midazolam and propofol and required inductionof barbiturate coma. Three days later, the patient underwenta bilateral implant of DBS electrodes into the GPi under gen-eral anesthesia with intraoperative microrecording using bis-pectral analysis monitoring, as described elsewhere.8 Animpulse generator (Kinetra, Medtronic, Minneapolis, MN)was inserted. A subtle central pallidal hyperintensity sur-rounded by hypointensity was observed only on proton



density sequences of the preoperative MRI, suggesting theeye-of-the-tiger sign (Figure 1B).Within days, DBS of the GPi caused an important and

progressive improvement in dystonia that persists 9 monthsafter surgery (Burke-Fahn-Marsden Dystonia Rating Scalescores of 10/120 and 4/30 for movement and disability,respectively). Currently, the DBS parameters are: left GPicontacts 0 (�), 1 (�), case (þ), 2.4 V, 90 ls, 130 Hz; rightGPi contacts 5 (�), case (þ), 2.4 V, 90 ls, 130 Hz. A fortui-tous disconnection of the DBS system produced a similardystonic storm that resolved progressively within 6 hoursafter the DBS was reconnected.Arthrodesis was performed to correct the fixed dystonic

posture of the left foot, and the patient has regained inde-pendent ambulation.This case illustrates the usefulness of DBS of the GPi to treat

dystonic storms in patients with PKAN and confirms that mod-ulation of GPi neuronal firing can improve severe generalizeddystonia, even in neurodegenerative disorders with brain ironaccumulation in the basal ganglia. In addition, it is noteworthythat the characteristic eye-of-the-tiger sign on MRI appearedlate in the disease course and only on proton density sequences.This finding suggests that genetic testing for mutations in thePANK2 gene should be considered in patients with suspectedneurodegeneration with brain iron accumulation, even in theabsence of this neuroradiological sign.

Francisco Grandas, MD, PhD1*,Carlos Fernandez-Carballal, MD1,Juan Guzman-de-Villoria, MD1 and

Israel Ampuero, PhD2

1Movement Disorders Research Unit, HospitalGeneral Universitario Gregorio Maranon

Madrid; 2Banco de Tejidos para InvestigacionesNeurologicas, Universidad Complutense, Madrid

*E-mail: [email protected]

References1. McNeill A, Bichall D, Hayflick SI, et al. T2* and FSE MRI distin-

guishes four subtypes of neurodegeneration with brain iron accu-mulation. Neurology 2008;70:1614–1619.

2. Mariotti P, Fasano A, Contarino F, et al. Management of statusdystonicus: our experience and review of the literature. Mov Dis-ord 2007;22:963–968.

3. Krause M, Fogel W, Tronnier V, et al. Long-term benefit to pallidaldeep brain stimulation in a case of dystonia secondary to pantothe-nate kinase-associated neurodegeneration. Mov Disord 2006;21:2255–2257.

4. Isaac C, Wright I, Bhattacharyya D, Baxter P, Rowe J. Pallidalstimulation for pantothenate kinase-associated neurodegenerationdystonia. Arch Dis Child 2008;93:239–240.

5. Mitaki M, Yehya A, Darwish H, Karam P, Comair Y. Deep brainstimulation as a mode of treatment of early onset pantothenatekinase-associated neurodegeneration. Eur J Paediatr Neurol 2009;13:61–64.

6. Castelnau P, Cif L, Valente EM, et al. Pallidal stimulationimproves pantothenate kinase-associated neurodegeneration. AnnNeurol 2005;57:738–741.

7. Timmermann L, Pauls KAM, Wieland K, et al. Dystonia in neu-rodegeneration with brain iron accumulation: outcome of bilateralpallidal stimulation. Brain 2010;133:701–712.

8. Duque P, Mateo O, Ruiz F, Guzman de Villoria J, Contreras A,Grandas F. Intraoperative microrecording under general anesthesiawith bispectral analysis monitoring in a case of deep brain stimula-tion surgery for Parkinson’s disease. Eur J Neurol 2008;17:76–77.

FIG. 1. A: Brain MRI showing bilateral globus pallidus hypointensity on T-2-weighted axial images, when the patient was 17 years old. B: BrainMRI, performed 2 years later, showing the eye-of-the-tiger sign on proton density sequences.



Bromocriptine Use and MyocardialFunction

Many studies have established that ergotamine-deriveddopamine agonists are associated with valvular heart dis-ease.1 The underlying mechanism is thought to be related tothe stimulation of serotonergic (5-HT2B) receptors that exerta trophic effect on fibroblasts with resultant cardiac valvularfibrosis. 5-HT2B receptors are also found in the myocardium.In rats, serotonin injections not only affected the heartvalves but also caused development of subendocardial pla-ques of myofibroblasts and enhanced extracellular matrixformation in the atria and ventricles.2 Moreover, in micelacking the membrane-bound transporter, responsible for theclearance of serotonin, fibrotic lesions developed not only inthe valves but also focally in the myocardium.3 Hence,ergotamine-derived dopamine agonist treatment might beassociated with increased myocardial fibrosis and resultingreduction in myocardial function.In addition to pergolide and cabergline, bromocriptine

also has 5-HT2B agonist properties. We had previouslyshown that bromocriptine was associated with an increasedrisk of developing valvular heart disease.4 As an extension ofthis study, we conducted a prospective echocardiographicstudy to investigate the risk of myocardial dysfunctionamong PD patients on bromocriptine compared with a con-trol group of PD patients who had not been exposed to do-pamine agonists before.The method of recruitment has been described previously.4

Seventy-one PD patients treated on bromocriptine and 47control PD patients were recruited. Complete transthoracicechocardiographic examinations were carried out with aVivid 7 ultrasounds system (G.E., Wingmed Ultrasound,Horten, Norway) and left ventricular chamber measurementswere made according to American Society of Echocardiogra-phy guidelines.5 Diastolic function was assessed by means ofthe mitral inflow profile, namely, E-wave velocity (E),A-wave velocity (A), and deceleration time (DT). Diastolicfunction was then divided into different grades based on theE/A ratio and DT.6 Left and right ventricular longitudinalfunctions were assessed using tissue Doppler imaging (TDI).TDI measuring myocardial contraction and relaxationvelocities at the septal mitral annulus was carried out, andearly-diastolic apically directed myocardial velocity (E0),late-diastolic atrially directed myocardial velocity (A0) andventricularly directed myocardial velocity (S0) were mea-sured. This was repeated at the lateral right ventricularannulus to obtain the corresponding velocities.All echocardiograms were read by a qualified cardiologist

who is an experienced echocardiographer (K.N.) and who

was blinded to participants’ prior drug history. Analyseswas performed using SPSS version 11. Our study had 96%power to detect a difference of 5% or more in EF and93% power to detect a difference of 0.2 or more in theE/A ratio. We tested differences between bromocriptinecases and controls using the chi-square test for categoricalvariables and the t test for continuous variables, after con-firming that normality and homogeneity assumptions weresatisfied.

The clinical characteristics of patients were similar tothose previously published.4 Patients on bromocriptine hadbeen treated an average of 43.0 6 32.5 months. The ventric-ular and atrial dimensions, wall thickness, and left ventricu-lar ejection fraction did not differ between bromocriptinecases and controls. The transmitral flow features E, E/Aratio, DT, and TDI measurements of septal mitral annulusvelocities—S0, E0, and A0—were also similar in both groups.There was also no significant difference in diastolic dysfunc-tion between the groups (Table 1). There was no associationbetween valvular stenosis and LV dysfunction.

In this study, we used conventional and recently intro-duced echocardiographic techniques to analyze LV systolicand diastolic function in PD patients but were unable todemonstrate any adverse impact of bromocriptine treatmenton LV function. LVEF, longitudinal displacement, and dia-stolic function did not seem to be impaired by treatment

------------------------------------------------------------*Correspondence to: Louis C.S.Tan, Department of Neurology,National Neuroscience Institute, Singapore; [email protected]

This project was supported by the National Medical Research Council,Singapore project no. NMRC/0995/2005.Relevant conflicts of interest/financial disclosures: Nothing to report.Full financial disclosures and author roles may be found in the onlineversion of this article.


Table 1. Echocardiographic analysis

Bromocriptine

group (n ¼ 71)

Control group

(n ¼ 46) P value

LVEDV (mL) 79.7 6 19.3 74.9 618.1 0.18LVESV (mL) 27.9 6 10.9 25.6 6 8.3 0.24LAA (cm2)a 17.2 6 3.8 16.5 6 3.4 0.31LVEF (%) 65.5 6 7.9 66.0 6 7.2 0.73E wave (E, cm/s)a 0.74 6 0.17 0.75 6 0.14 0.68A wave (A, cm/s)a 0.70 6 0.21 0.69 6 0.16 0.82E/A ratioa 1.13 6 0.37 1.12 6 0.28 0.96DT (ms)a 199.3 6 45.4 190.6 6 40.7 0.3Septal annulus E0 (cm/s) 0.075 6 0.024 0.074 6 0.020 0.75Septal annulus A0 (cm/s) 0.102 6 0.024 0.102 6 0.021 0.85Septal annulus S0 (cm/s) 0.070 6 0.015 0.071 6 0.013 0.67RV annulus E0 (cm/s) 0.121 6 0.036 0.111 6 0.027 0.12RV annulus A0 (cm/s) 0.164 6 0.100 0.151 6 0.036 0.41RV annulus S0 (cm/s) 0.125 6 0.024 0.120 6 0.024 0.18E0/A0 ratio 0.78 6 0.33 0.75 6 0.26 0.56E/E0 ratio, n (%)a

E/E0 > 15 8 (11.8) 7 (15.6)E/E0 8–15 44 (64.7) 28 (62.2) 0.84E/E0 < 8 16 (23.5) 10 (22.2)

Diastolic dysfunction,n (%)a

None 53 (77.9) 32 (71.1)Stage I 2 (2.9) 13 (28.9) 0.27Stage II 13 (19.1) 0 (0)Stage III 0 (0) 0 (0)

LVEDV, left ventricular end-diastolic volume; LVESV, left ventricular end-systolic volume; LAA, left atrial area; LVEF, left ventricular ejection fraction;DT, mitral deceleration time.All values are expressed as mean 6 SD unless otherwise stated.aThese values were based on n ¼ 68 for bromocriptine group and n ¼ 45for controls, as these measurements could not be performed on somepatients.



with bromocriptine. Our results are consistent with a previ-ous study that also did not detect myocardial dysfunctionamong patients treated with pergolide and cabergoline.7 Toour best knowledge, no other similar study has beenperformed.Although assessment of LV longitudinal systolic and dia-

stolic function has been widely used to diagnose subclinicaldiastolic LV dysfunction in various diseases associated withmyocardial fibrosis, it is recognized that a definite diagnosisof myocardial fibrosis can only be made by histopathologicalexamination of the myocardium.We conclude that although bromocriptine use is associated

with an increased risk of developing valvular heart disease,4

bromocriptine does not have a detectable adverse impact onmyocardial systolic and diastolic function.

Acknowledgments: We thank Ms. Irene Seah Soo Hoon for assisting inthis study.

Louis C.S. Tan, FRCP,1,2* Kenneth K.C. Ng, MRCP,3

Wing-Lok Au, FRCP,1,2 Raymond K.K. Lee, FRCP,3 andNigel C.K. Tan, FRCP2

1Parkinson’s Disease and Movement DisordersCentre, National Neuroscience Institute

Singapore, USA National Parkinson FoundationCentre of Excellence; 2Department of Neurology,

National Neuroscience Institute, Singapore*E-mail: [email protected];

3Department of Cardiology, Tan Tock SengHospital, Singapore

References1. Antonini A, Poewe W. Fibrotic heart-valve reactions to dopamine

agonist treatment in Parkinson’s disease. Lancet Neurol. 2007;6:826–829.

2. Gustafsson BI, Tommeras K, Nodrum I, et al. Long-term serotoninadministration induces heart valve disease in rats. Circulation2005;111:1517–1522.

3. Mekontso-Dessap A, Brouri F, Pascal O, et al. Deficiency of 5-hy-droxytryptamine transporter gene leads to cardiac fibrosis andvalulopathy in mice. Circulation. 2006;113:81–89.

4. Tan LCS, Ng KKC, Au WL, Lee RKK, Chan YH, Tan NCK. Bro-mocriptine use and the risk of valvular heart disease. Mov Disord.2008;24:344–349.

5. Lang RM, Bierig M, Devereux RB, et al. Recommendations forchamber quantification:a report from the American Society ofEchocardiography’s guidelines and standards committee and thechamber quantification writing group, developed in conjunctionwith the European association of echocardiography, a branch ofthe European Society of Cardiology. J Am Soc Echocardiogr.2005;18:1440–1463.

6. Nagueh SF, Appleton CP, Gillebert TC, et al. Recommendationsfor the evaluation of left ventricular diastolic function by echocar-diography. J Am Soc Echocardiogr. 2009;22:107–133.

7. Rasmussen VG, Poulsen SH, Dupont E, Ostergaard K, SafikhanyG, Egeblad H. Ergotamine-derived dopamine agonists and left ven-tricular function in Parkinson patients: systolic and diastolic func-tion studied by conventional echocardiography, tissue Dopplerimaging, and two-dimensional speckle tracking. Eur J Echocar-diogr. 2008;9:803–808.

DYT 6—A Novel THAP1 Mutation WithExcellent Effect on Pallidal DBS

DYT6 is an early-onset dystonia caused by variablemutations of the gene encoding the thanatos-associated pro-tein (THAP1).1 It usually starts in the cranial region or on theupper extremities with subsequent generalization. It has beenreported that deep brain stimulation of the internal globuspallidus (GPi-DBS) has moderate or unsatisfactory effects inthese patients.2,3 We describe a novel mutation of the THAP1gene in 2 siblings (a boy and a girl) with a rapid generaliza-tion into a life-threatening status dystonicus (SD) in the boy.Unlike in the 7 previously reported patients,2,3 in his case, weobserved excellent long-term effects from bilateral GPi-DBS.

We report on a Czech family with 2 affected memberswith dystonia and 5 carriers in 2 preceding generations (Fig.1). Asymptomatic mutation was noted in the mother, 2aunts, grandfather, and granduncle of the affected siblings.A sequence analysis of genomic DNA from peripheral bloodleukocytes revealed a new mutation (NM_018105.2)c.89C>G in exon 2 of the THAP1 gene in a heterozygotestate, leading to amino acid substitution (p.Pro30Arg).

The girl, now 20 years old, started with writer’s cramp inher right hand when she was 15. Two years later, she devel-oped a slight dysarthria, and since the age of 18 she has suf-fered from disabling dystonia in the left hand. Her actualclinical symptoms correspond to multifocal dystonia involv-ing fingers, toes, and oromandibular muscles (Burke FahnMarsden Dystonia motor score [BFMDS] ¼ 8).

Her brother, now 14.5 years old, developed writer’scramp at the age of 8, which gradually deteriorated intosevere generalized dystonia that lapsed into status dystoni-cus (SD) when he was 12 (BFMDS ¼ 41). We have previ-ously published details of the case (including video) as anexample of the excellent effect of GPi-DBS in SD occur-ring in non-DYT1 dystonia (BFMDS ¼ 3).4 Subsequently,30 months after the implantation, with the DBS parame-ters set on pulse duration 450 ls, frequency 130 Hz, am-plitude left 1.6 V, and amplitude right 1.8 V, cervicaldystonia suddenly reappeared because of repeated unpre-dictable switching off of the left neurostimulator, probablyrelated to battery exhaustion. Soon after DBS wasswitched on again, cervical dystonia disappeared.Recently, 2 Soletra stimulators were replaced by a rechar-gable RC stimulator (Medtronic, Minneapolis, MN).Awakening from propofol anesthesia was accompanied bysevere generalized dystonic jerks of the trunk and extrem-ities resembling the original signs of SD. These symptoms

------------------------------------------------------------*Correspondence to: Robert Jech, Department of Neurology andCenter of Clinical Neuroscience, First Faculty of Medicine, CharlesUniversity in Prague and General University Hospital in Prague, CzechRepublic; [email protected]

This work was supported by the Czech Science Foundation (grantproject 309/09/1145) and by the Czech Ministry of Education (researchproject MSM 0021620849).Relevant conflicts of interest/financial disclosures: Nothing to report.Full financial disclosures and author roles may be found in the onlineversion of this article.




were suppressed within minutes after restarting the DBS.Now 32 months after implantation, the patient is stable,with only mild dystonia in his right hand and slight dys-tonic dysarthria (BFMDS ¼ 6).Since the discovery of the THAP1 mutation c.89C>G,

this is the first documented case of excellent DBS effects ina patient with DYT6 and probably the first case of SD inDYT6. There may be a number of reasons why our patientresponded fairly better to GPi-DBS (98% improvement in1 month, 85% 32 months after surgery) than the 5patients of Groen et al3 (33% mean improvement 6months after surgery) or the 2 patients treated by Zittelet al2 (mild to moderate improvement). Apart from differ-ences in the clinical picture and in DBS parameters, thismay be because of the specific THAP1 mutation. THAP1mutations are marked by considerable variability. Theirnumber is steadily growing; in 2009 the first 15 mutationswere reported, and by August 2010, another 25 new muta-tions had been described (list of references limited).2,3,5–7

Of 8 DBS-treated DYT6 patients, ours4 is the only onewho has the mutation in the first half of exon 2. Fourpatients3 have the mutation in the second half of exon 2, 2patients2,3 in exon 3, and 1 patient2 in exon 1. We canonly speculate that each exon or its part may influence dif-ferent properties of the translation product, clinical mani-festations, and perhaps also the patient’s response to DBS.Therefore, patients with other THAP1 mutations mayimprove after GPi-DBS to a similar extent as our patientwith SD. We thus believe that in advanced cases of DYT6,deep brain stimulation of GPi should continue beingregarded as a method of treatment.

Robert Jech, MD, PhD,1,2* Martin Bares, MD, PhD,3

Anna Krepelova, MD, PhD,4 Dusan Urgosık, MD, PhD,1,2,5

Petra Havrankova, MD,1,2 and Evzen Ruzicka, MD, DSc1,2

1Department of Neurology and Center of ClinicalNeuroscience, First Faculty of Medicine, CharlesUniversity in Prague, Czech Republic; 2GeneralUniversity Hospital in Prague, Czech Republic;

3First Department of Neurology, St. Anne’sHospital, Faculty of Medicine, Masaryk UniversityBrno, Czech Republic; 4Institute of Biology andMedical Genetics, Second Faculty of Medicine,Charles University in Prague, Prague, Czech

Republic; and 5Department of Stereotactic and

Radiation Neurosurgery, Na Homolce Hospital,Prague, Czech Republic*E-mail: [email protected]

References1. Fuchs T, Gavarini S, Saunders-Pullman R, et al. Mutations in the

THAP1 gene are responsible for DYT6 primary torsion dystonia.Nat Genet 2009;41:286–288.

2. Zittel S, Moll CK, Brggemann N, et al. Clinical neuroimaging andelectrophysiological assessment of three DYT6 dystonia families.Mov Disord 2010;25:2405–2412.

3. Groen JL, Ritz K, Contarino MF, et al. DYT6 dystonia: Mutationscreening, phenotype, and response to deep brain stimulation. MovDisord 2010;25:2420–2427.

4. Jech R, Bares M, Urgosik D, et al. Deep brain stimulation in acutemanagement of status dystonicus. Mov Disord 2009;24:2291–2292.

5. Van Gerpen JA, Ledoux MS, Wszolek ZK. Adult-onset leg dystoniadue to a missense mutation in THAP1. Mov Disord 2010;25:1306–1307.

6. Houlden H, Schneider SA, Paudel R, et al. THAP1 mutations(DYT6) are an additional cause of early-onset dystonia. Neurology2010;74:846–850.

7. Bressman SB, Raymond D, Fuchs T, Heiman GA, Ozelius LJ,Saunders-Pullman R. Mutations in THAP1 (DYT6) in early-onsetdystonia: a genetic screening study. Lancet Neurol 2009;8:441–446.

Caudate Glucose Hypometabolism ina Subject Carrying an Unstable Alleleof Intermediate CAG33 Repeat Lengthin the Huntington’s Disease Gene

Huntington’s disease (HD) is caused by an abnormalCAG expansion in the huntingtin (htt) gene. Current guide-lines indicate that individuals with �36 CAG repeats willinevitably develop HD, whereas those with 27–35 CAG

FIG. 1. Pedigree of the DYT6 affected individuals and relatives in the Czech family with the c.89C>G mutation (? denotes nontested, clinicallyasymptomatic subjects potentially at risk of having this mutation).

------------------------------------------------------------*Correspondence to: Ferdinando Squitieri, Neurogenetics and RareDiseases Centre, IRCCS Neuromed, Pozzilli, Italy; [email protected]





repeats exhibit reduced penetrance and generally do notmanifest the disease, although as a result of unstable trans-mission of paternal alleles, subsequent generations mayexpand into the HD range.1 Several reports in recent years,however, have provided evidence that an intermediate allele(IA) length may be associated with clinical (and pathologi-cal) features of HD.2–5 Experimental reports on lympho-blastoid cell lines from normal and affected subjects haveshown progressively diminishing ATP production levels asa function of the CAG repeat length, with IAs showingsome degree of energy and metabolic impairment.6 Otherreports have documented a striking CAG repeat somaticmosaicism in postmitotic neurons, particularly in the cortex

and striatum, thus contributing to the variability of age atonset of manifest disease.7

Case PresentationWe describe the case of a 68-year-old man carrying 33

CAG repeats who transmitted a full ‘‘novel’’ expansionmutation of 48 CAG repeats to his offspring, who thenmanifested early-onset HD (new mutation). Automatedsequencing determined the exact repeat number and pater-nity confirmation. The IA subject manifested only subtle,soft neurological signs suggesting initial HD, including jerkyocular pursuit, increased latency saccade initiation, slight arm

FIG. 1. Cerebral glucose metabolic rates (MRGlu) in the subject with the intermediate-length allele and in 10 age-matched controls. The MRGluwas 2.6 (right hemisphere) and 1.8 (left hemisphere) times lower in the HD patient (red triangle) than the lower bound of CI calculated in the controls(blue triangles). Below, FDG-PET scan of the IA subject versus controls; arrows, caudate. [Color figure can be viewed in the online issue, which isavailable at wileyonlinelibrary.com.]



incoordination, and a Mini Mental Status examination scoreof 28, missing 1 point for delayed recall and 1 point for cal-culation. Despite the scarce clinical evidence, a fluorodeoxy-glucose (FDG) positron emission tomography (PET) scandisclosed severe caudate glucose hypometabolism (Fig. 1).Dynamic PET images were analyzed using integrated soft-ware (http://www.brainnest.org). The regional difference inthe brain metabolic rate of glucose (MRGlu) between theHD subject and healthy age-matched controls was assessedfor each region of interest (ROI) by plotting the estimateddifferences between the mean MRGlu values in a given ROIobtained from all healthy subjects and the absolute ‘‘truevalue’’ obtained from a given subject’s ROI (see online mate-rial). This procedure yielded a repeatability coefficient (RC),twice the standard deviation of the differences (see onlinematerial).

DiscussionThis case illustrates an intergenerational expansion of

CAG repeats from a father (n ¼ 33) to his offspring (n ¼48) associated with clear worsening of the clinical pheno-type, from subtle motor and cognitive findings in the fatherto the full spectrum of HD in the offspring. Although thefather had only subtle clinical features suspicious of mildHD, he demonstrated markedly impaired glucose uptake inthe caudate nucleus. Several studies have documented de-fective glycolysis and progressively impaired cerebral glu-cose metabolism, especially in the caudate nucleus, detectedyears before clinical manifestations of HD.8 Recent reportssuggest that, in addition to the mutation length, somaticinstability in the CAG repeat stretch influences HD age atonset.7 A partial toxic effect of the intermediate-size alleledetected in the patient’s blood cells might therefore be associ-ated with increased striatal toxicity due to mosaic alleles pos-sibly expanded beyond the 33 CAG repeats. The increase insize and the IA length itself may have contributed to a clini-cally incomplete or mild phenotype, similar to observationsin cases of reduced penetrance alleles. On the other hand, itis also possible that IAs by themselves could contribute tothe HD phenotype in vivo because of subclinical huntingtintoxicity. Our neuroimaging finding (ie, the impaired glucosemetabolism in the caudate nucleus) therefore highlights theinteresting hypothesis that in certain conditions, subjects car-rying premutations may show an incomplete phenotype(endophenotype) associated with reduced brain glucoseuptake. It is possible that the abnormal caudate FDG uptakecould be evidence for subclinical (ie, mild) huntingtin toxicityas a consequence of IA effect. Thus, although we cannot pro-vide conclusive proof without neuropathological findings, ourreport supports the hypothesis that, in addition to mutationlength, CAG repeat instability contributes to the clinical phe-notype in HD and that some IA subjects show incompletephenotype.5

In conclusion, documented CAG repeat instability inHD families (ie, siblings with different mutation length,large intergenerational CAG repeat changes) independentlyof blood triplet size (either expanded or of intermediatelength) may contribute to the HD phenotype because ofpotentially repeat-related mosaicism in the brain. In suchcases, the unstable IA may increase the risk to expandinto a full mutation not only in offspring, but also in thetransmitting subject’s nervous system. This possibility,

together with the potential intrinsic IA toxicity, shouldbe considered during genetic counseling of subjects withIAs.

Ferdinando Squitieri, MD, PhD,1*

Mouna Esmaeilzadeh, MD,2 Andrea Ciarmiello, MD,3

and Joseph Jankovic, MD, PhD4

1Neurogenetics and Rare Diseases Centre, IRCCSNeuromed, Pozzilli, Italy; 2Stockholm Brain

Institute, Department of Clinical Neuroscience,Karolinska Institute, PET Centre,Karolinska University Hospital,

Stockholm, Sweden; 3Unit of Nuclear Medicine,Sant Andrea Hospital,

La Spezia, Italy; 4Department of Neurology,Baylor College of Medicine,

Houston, Texas

References1. ACMG/ASHG statement. Laboratory guidelines for Huntington

disease genetic testing. The American College of Medical Genetics/American Society of Human Genetics Huntington Disease GeneticTesting Working Group. Am J Hum Genet 1998;62:1243–1247.

2. Kenney C, Powell S, Jankovic J. Autopsy-proven Huntington’s dis-ease with 29 trinucleotide repeats. Mov Disord 2007;22:127–130.

3. Andrich J, Arning L, Wieczorek S, Kraus PH, Gold R, Saft C.Huntington’s disease as caused by 34 CAG repeats. Mov Disord2008;23:879–881.

4. Herishanu YO, Parvari R, Pollack Y, et al. Huntington disease insubjects from an Israeli Karaite community carrying alleles of in-termediate and expanded CAG repeats in the HTT gene: Hunting-ton disease or phenocopy? J Neurol Sci 2009;277:143–146.

5. Groen JL, de Bie RM, Foncke EM, et al. Late-onset Huntingtondisease with intermediate CAG repeats: true or false? J NeurolNeurosurg Psychiatry 2010;81:228–230.

6. Seong IS, Ivanova E, Lee JM, et al. HD CAG repeat implicates adominant property of huntingtin in mitochondrial energy metabo-lism. Hum Mol Genet 2005;14:2871–2880.

7. Swami M, Hendricks AE, Gillis T, et al. Somatic expansion of theHuntington’s disease CAG repeat in the brain is associated with anearlier age of disease onset. Hum Mol Genet 2009;18:3039–3047.

8. Squitieri F, Ciarmiello A. Key role of nuclear medicine in seekingbiomarkers of Huntington’s disease. Eur J Nucl Med Mol Imaging2010;37:1124–1127.

Motor Tic of the Neck: A ProbableCause of Stroke in a Child with Gilles

de La Tourette Syndrome

The male patient had several motor and vocal tics fromage 7 on. There was no family history for tics and no

------------------------------------------------------------*Correspondence to: Andreas Hartmann; Departement de Neurologie,Pole des Maladies du Systeme Nerveux, Groupe HospitalierPitie-Salpetriere, Universite Pierre et Marie Curie Paris VI, Paris, France;andreas.hartmann.aphp.fr

Relevant conflicts of interest/financial disclosures: Nothing to report.




psychiatric comorbidities such as OCD or ADHD. Violentand frequent hyperextension and lateral flexion of the neckwas the main motor tic at onset and remained prominentthroughout the disease course. At age 12, the patient sud-denly developed a left transient hemianopsia lasting for 24hours. Ten days later, a left hemiplegia set in abruptly,which partially resolved within 30 min. Neurological exami-nation showed a left facial paresis, a subtle left hemiparesia,and brisk reflexes on the left side. Brain MRI showed anacute right striatal stroke with a secondary hemorrhage (Fig.1). Cerebral angiography supported the diagnosis of a rightvertebral artery dissection (Fig. 1). A comprehensive cardio-vascular work-up was normal, including transthoracic andtransesophageal echocardiography, Holter ECG, and coagu-lation exams. The striatal stroke was probably caused by anembolism from the vertebral artery dissection through theposterior communicating artery as the carotid artery was notobstructed. A few days later, a new cerebellar strokeoccurred, and anticoagulation was started despite the hemor-rhagic transformation. The outcome was good, with only amild left motor deficit 12 months after the initial stroke. Thevertebral artery dissection was considered to be the conse-quence of the cervical motor tic, and treatment with pimo-zide up to 3 mg/day was started, with little efficacy.Risperidone was then tried up to 4 mg/day but discontinueddue to important sedation. Aripiprazole was discussed butrejected because it is contraindicated in stroke patients. Be-havioral psychotherapy (Habit Reversal Training1) led to adecrease in neck tic frequency but not severity. According to

the movement observed, he received local injection of botuli-num toxin in the right trapezius (40 U dysport), right levatorscapula (140 U dysport), and in the right splenius (120 Udysport), which totally abolished the neck tic after the sec-ond injection with no reccurence since.

Our patient had a right vertebral artery dissection, likelyfrom traumatic origin, secondary to a severe motor tic withneck hyperextension. Despite different neuroleptic medica-tions at sufficient dosages and durations, violent tic of theneck persisted. As illustrated in this case report, tics can leadto incidental self-injurious behavior. Severe complicationssuch as bone fractures, joint sprains, skin or eye injuries, cer-vical myelopathies, or nerve lesions have been reported inGTS patients.2–6 However, stroke due to motor tics has notbeen reported previously. The only report of stroke in anadult GTS patient was related to a concomitant antiphos-pholipid syndrome.7

A survey by the Canadia Stroke Consortium showed that in74 patients studied (age range 16–87 years, mean age 44 years)vertebrobasilar artery dissections (72%) clearly predominateover carotid artery dissections (28%).8 This can be explainedby the vertebral artery’s extreme vulnerability to torsioninjury, because it winds around the atlas to enter the skull.Most dissection (81% in this series8) were associated with sud-den and vigorous neck movements but it should be noted thattrivial traumas such as coughing, vomiting, and prolongedhead tilting may cause dissections in predisposed persons.

Efficient and aggressive treatment is crucial in this patient.As neuroleptic medications and habit reversal training1 wereineffective, we resorted to botulinum toxin injections,8,9 previ-ously reported to be effective in reversing compressive myelop-athy associated with intense neck extension (whiplash) typetics.10 Interestingly, and as described previously,9 tics subsidedcompletely after the botulinum toxin injections, suggestingthat interrupting a feed-forward loop between tic executionand tic generation may represent an effective therapy.

In case of treatment failure, deep brain stimulation (DBS)could have been a worthwhile therapeutic option.11 To date,DBS is restricted to adult patients given the substantial remis-sion rates for tics once patients reach adulthood. Accordingly,recent guidelines suggest not to operate in GTS patientsyounger than 25 years.12 Nonetheless, one 16-year-old patientwith very severe GTS has recently been operated with goodresults.13 As DBS can be switched off to evaluate if stimula-tion is still required and devices can be explanted in case oftic remission, we suggest that DBS could be proposed to rig-orously select young patients to avoid tragic and irreversibleneurological sequelae of severe tics. We believe that the bene-fits clearly outweigh the surgical risk in this age group, as pre-viously shown in juvenile dystonia patients, where earlyintervention may even have a positive influence on diseasecourse.14 However, in GTS patients, this remains speculativeand requires demonstration in a controlled clinical trial.

Priscilla van Meerbeeck, MD,1 Cecile Behar, BS,1,2

Virginie Czernecki, PhD,1,2 Emmanuel Roze, MD, PhD,1

Emmanuelle Deniau, MD,2,3 andAndreas Hartmann, MD,1,2*

1Departement de Neurologie, Pole des Maladiesdu Systeme Nerveux, Groupe Hospitalier

FIG. 1. A/B: Brain MRI: (A) Acute stroke of the right caudatus nucleusand right putamen and (B) subacute multiple right cerebellar stroke inthe PICA and the superior cerebellar artery territories. C/D: Cranial CTangiography: Occlusion of the right vertebral artery (VA) at the C4 level(V2-V3 segment).



Pitie-Salpetriere, Universite Pierre et Marie CurieParis VI, Paris, France;

*E-mail: [email protected] de reference ‘Syndrome Gilles de la

Tourette’, Pole des Maladies du Systeme Nerveux,Groupe Hospitalier Pitie-Salpetriere, Universite

Pierre et Marie Curie Paris VI, Paris, France; and3Service de Psychiatrie de l’Enfant et de

l’Adolescent, Pole des Maladies du SystemeNerveux, Groupe Hospitalier Pitie-Salpetriere,

Universite Pierre et Marie Curie Paris VI,Paris, France

References1. Piacentini J, Woods DW, Scahill L, et al. Behavior therapy for chil-

dren with Tourette disorder: A randomized controlled trial. JAMA2010;303:1929–1937.

2. Lin JJ, Wang HS, Wong MC, Wu CT, Lin KL. Tourette’s syn-drome with cervical disc herniation. Brain Dev 2007;29:61–63.

3. Dobbs M, Berger JR. Cervical myelopathy secondary to violent ticsof Tourette’s syndrome. Neurology 2003;60:1862–1863.

4. Krauss JK, Jankovic J. Severe motor tics causing cervical myelopa-thy in Tourette’s syndrome. Mov Disord 1996;11:563–566.

5. Cheung MY, Shahed J, Jankovic J. Malignant Tourette syndrome.Mov Disord 2007;22:1743–1745.

6. Isaacs JD, Adams M, Lees AJ. Noncompressive myelopathy associ-ated with violent axial tics of Tourette syndrome. Neurology 2010;74:697–698.

7. Seijo-Martinez M, Mosquera-Martınez JA, Romero-Yuste S, Cruz-Martinez J. Ischemic stroke and epilepsy in a patient with Tour-ette’s syndrome: association with the antiphospholipid syndromeand good response to leviracetam. Open Neurol J 2008;2:32–34.

8. Norris JW, Beletsky V, Nadareishvili ZG. Sudden neck movementand cervical artery dissection. The Canadian Stroke Consortium.CMAJ 2000;163:38–40.

9. Aguirregomozcorta M, Pagonabarraga J, Diaz-Manera J, Pascual-Sedano B, Gironell A, Kulisevsky J. Efficacy of botulinum toxin insevere Tourette syndrome with dystonic tics involving the neck.Parkinsonism Relat Disord 2008;14:443–445.

10. Marras C, Andrews D, Sime E, Lang AE. Botulinum toxin for sim-ple motor tics: a randomized, double-blind, controlled clinical trial.Neurology 2001;56:605–610.

11. Welter ML, Mallet L, Houeto JL, et al. Internal pallidal and tha-lamic stimulation in patients with Tourette syndrome. Arch Neurol2008;65:952–957.

12. Mink JW, Walkup J, Frey KA, et al.; Tourette Syndrome Associa-tion, Inc. Patient selection and assessment recommendations fordeep brain stimulation in Tourette syndrome. Mov Disord 2006;21:1831–1838.

13. Shahed J, Poysky J, Kenney C, Simpson R, Jankovic J. GPi deepbrain stimulation for Tourette syndrome improves tics and psychi-atric comorbidities. Neurology 2007;68:159–160.

14. Isaias IU, Alterman RL, Tagliati M. Outcome predictors of pallidalstimulation in patients with primary dystonia: the role of diseaseduration. Brain 2008;131:1895–1902.



patients with parkinson's disease dementia stay in the hospital twice as long as those without...

Documents