Auditory Startle Reaction inPrimary Blepharospasm

Jorg Muller, MD,1 Martina Rinnerthaler, MD,1,2

Werner Poewe, MD,1 and Markus Kofler, MD1,2*

1Department of Neurology, Medical University Innsbruck,Austria; 2Department of Neurology, Hospital Hochzirl,

Austria

Abstract: Primary dystonia is associated with abnormalbrainstem function, as shown by abnormalities of the blinkreflex in blepharospasm (BSP) and of the auditory startlereaction in cervical dystonia. We examined the auditorystartle reaction—a brainstem reflex elicited by an unex-pected loud stimulus—in patients with primary BSP toexpand knowledge on brainstem pathophysiology in pri-mary focal dystonia. Thirteen patients with primary BSPwere included and 13 age- and sex-matched healthy volun-teers served as controls. Auditory startle responses (ASRs)were elicited by binaural high-intensity auditory stimuli,and reflex electromyographic activity was recorded simul-taneously with surface electrodes bilaterally from masseter,orbicularis oculi, sternocleidomastoid, and biceps brachiimuscles. Patients with BSP showed higher ASR probabili-ties (masseter, sternocleidomastoid, biceps brachii), shorterASR onset latencies (masseter, orbicularis oculi, sternoclei-domastoid), and larger ASR area-under-the-curve (masse-ter, sternocleidomastoid) as compared with normal con-trols. Habituation of ASRs did not differ significantlybetween patients and controls. These results corroborateprevious findings of increased brainstem excitability in pri-mary BSP but point to a different pattern of brainstemdysfunction compared to cervical dystonia, indicating thatdifferent pathophysiological mechanisms are involved inthe two types of focal dystonia. © 2006 Movement DisorderSociety

Key words: auditory startle reaction; auditory startle re-sponse; botulinum toxin; brainstem reflex; essential blepharo-spasm; pedunculopontine nucleus (PPN)

Primary blepharospasm (BSP) is a focal dystonia char-acterized by excessive involuntary closure of the eyelids.Primary BSP is not associated with any known etiology,whereas secondary BSP may occur following lesions inlower pons, upper brainstem, thalamus, or basal ganglia.1

Previous studies on blink reflex recovery curves indicateincreased interneuronal excitability in brainstem circuits

in patients with cranial and cervical dystonia.2–4 Loss ofinhibition of brainstem reflexes in dystonia might reflectalterations in descending projections to inhibitory neu-rons, which are indirectly modulated by the pallido-thalamo-cortical motor circuit.5

The auditory startle reaction is a brainstem reflexevoked by an unexpected loud stimulus. The initial re-action involves eye closure, facial grimacing, neck flex-ion, and abduction or flexion of the arms. ASRs aremediated and integrated by the nucleus reticularis pontiscaudalis6–8; basal ganglia as well as higher-order struc-tures are implicated in the modulation of this reflex.9–11

Disorders associated with dysfunction in pallido-thalamo-cortical circuits could therefore result in ASRabnormalities. Indeed, a recent study in patients withcervical dystonia demonstrated ASR alterations.12 ASRprobability and size were significantly smaller in cervicaldystonia patients than in controls, while normal latenciesand recruitment patterns indicated a preserved organiza-tion of intrinsic neural pathways mediating ASRs. Thepresent study investigated ASR characteristics in patientswith essential BSP.

PATIENTS AND METHODS

Patients

Thirteen consecutive patients with primary BSP and13 age- and sex-matched controls were recruited (Table1). Of the 13 patients, 11 had isolated BSP, and 2 hadadditional mild oromandibular involvement. Secondarycauses of dystonia such as basal ganglia lesions andprevious drug exposure were ruled out in all patients. Allpatients had received regular botulinum toxin (BTX)injections for at least 1 year, but ASRs were investigatednot earlier than 12 weeks after the last BTX injection.BTX injections were restricted to the orbicularis oculimuscle in all patients. None of the patients was taking

*Correspondence to: Dr. Markus Kofler, Department of Neurology,Medical University Innsbruck, A-6020 Innsbruck, Austria. E-mail:markus.kofler@uibk.ac.at

Received 13 July 2006; Revised 28 August 2006; Accepted 6 Sep-tember 2006

Published online 5 December 2006 in Wiley InterScience (www.interscience.wiley.com). DOI: 10.1002/mds.21270

TABLE 1. Demographic data of BSP patients and healthycontrol subjects

Primaryblepharospasm

(N � 13)Controls(N � 13)

Age (yr) 63.8 (7.6) 63.1 (7.8)Gender 6 female, 7 male 6 female, 7 maleBSP severity 2 (0.7)Disease duration 10.8 years (6.1)Duration of BTX

treatment 6.9 years (4.2)Previous BTX injection 15.2 weeks (5.5)Co-medication None None

Numbers are given as mean values, with standard deviations inparentheses. BSP, blepharospasm; BTX, Botulinum toxin.

268 J. MULLER ET AL.

Movement Disorders, Vol. 22, No. 2, 2007

anticholinergic drugs, tetrabenazine, benzodiazepines, orantidepressants. The mean severity of BSP assessed by a0 to 3 BSP total score (0 � no BSP, 1 � mild, 2 �moderate, 3 � severe BSP) was 2.0 (SD 0.7). All pa-tients and controls consented to participate in the study,which was approved by the institutional ethics board.

Methods

All participants were examined in supine position in aquiet semidarkened room, and were asked to remainawake and relaxed. Care was taken to avoid potentialvisual or auditory prepulse stimuli.13 After normal bilat-eral hearing thresholds were ascertained, ASRs wereelicited by eight binaurally presented tone bursts differ-ing in tonal frequency and intensity (250 Hz, 90 dB; 500Hz, 105 dB; 750 Hz, 105 dB; 1,000 Hz, 110 dB nHL,each stimulus presented twice at random) to enhance thenovelty of the stimulus.11,14,15 Each participant receivedthe same amount and type of stimuli. Consecutive stimuliwere given at intervals of 2 to 3 minutes. Nonrectifiedsurface electromyographic (EMG) recordings were ob-tained simultaneously after each stimulus from masseter,orbicularis oculi, sternocleidomastoid, and biceps brachiimuscles bilaterally. Silver–silver chloride cup electrodeswere attached over the muscle belly and tendon whereapplicable. Single sweeps of 500 msec, including a 20-msec prestimulus delay, were recorded with filters set at10 and 10,000 Hz. No trace contained background ac-tivity with a mean amplitude exceeding 50 �V, whichwould have precluded inclusion in further statisticalanalysis, but eight traces had to be discarded (one mas-seter, three sternocleidomastoid, four orbicularis oculimuscles) in the patient group because of continuousbackground EMG activity that prevented exact measure-ment of probable ASRs.

Statistical Analysis

ASR probability was calculated by dividing the num-ber of all accepted reflex responses by the total numberof recorded traces and multiplying by 100. ASR latencieswere measured from stimulus onset to ASR onset. ASRarea-under-the-curve was calculated during the first 100msec after response onset to avoid possible contamina-tion with voluntary muscle activation. ASR habituationwas assessed by comparing the number of muscles show-ing startle responses after stimuli 1�2 with the numberof responses to stimuli 7�8. Kolmogorow–Smirnov test-ing confirmed non-normal distribution of latency andarea data. Comparison between BSP patients and con-trols was performed with the Fisher’s exact test for ASRprobability, with the Mann–Whitney U test for ASR

latency and area-under-the-curve, and with a �2 test forhabituation. The significance level was set at P � 0.05.

RESULTS

ASRs were elicited more frequently in patients withBSP (71%) than in normal controls (58%; P � 0.001).Patients with BSP had higher ASR probabilities in mas-seter (78% vs. 56%; P � 0.001), sternocleidomastoid(70% vs. 56%; P � 0.006), and biceps brachii muscles(45% vs. 23%; P � 0.001) as compared with controls.Interestingly, BSP patients had lower ASR probabilitiesin orbicularis oculi muscles (93% vs. 98%; P � 0.017).

Patients with BSP had significantly shorter medianASR onset latencies in masseter (58 vs. 63 msec; P �0.026), orbicularis oculi (31 vs. 36 msec; P � 0.001),and sternocleidomastoid muscles (58 vs. 71 msec; P �0.001) as compared with controls, while ASR onsetlatencies were significantly longer in biceps brachii mus-cles (105 vs. 91 msec; P � 0.029).

Median ASR area-under-the-curve was significantlylarger in masseter (1743 vs. 1,195 �Vms; P � 0.001)and sternocleidomastoid (3,203 vs. 2,359 �Vms, P �0.035) muscles, while being significantly smaller in or-bicularis oculi muscle (2,778 vs. 4,015 �Vms, P �0.001) in patients with BSP as compared with controls.Figure 1 shows a representative example of ASRs in apatient with BSP and in a normal control subject.

Habituation of ASRs did not differ significantly be-tween patients with BSP and controls in all musclescombined (P � 0.66), nor in masseter (P � 0.49),orbicularis oculi (P � 0.67), sternocleidomastoid (P �0.85), or biceps brachii (P � 0.12) muscles.

FIG. 1. Representative example of auditory startle responses in apatient with primary blepharospasm and a normal control subject. Thearrows indicate stimulus onset. MSS, masseter; OOc, orbicularis oculi;SCM, sternocleidomastoid; BB, biceps brachii; R, right; L, left.

AUDITORY STARTLE REACTION IN BLEPHAROSPASM 269

Movement Disorders, Vol. 22, No. 2, 2007

DISCUSSION

The present study demonstrates alterations of ASRs inpatients with primary BSP as compared with healthycontrol subjects matched for age and gender.16,17 A nor-mal muscle recruitment pattern suggests no major orga-nizational disturbance of intrinsic neural pathways me-diating and integrating ASRs. However, several featuressuggest disinhibition of ASR circuits in BSP patientsmediated by higher-order control pathways: ASRs weresignificantly more frequent in masseter, sternocleidomas-toid, and biceps brachii muscles, were recruited signifi-cantly faster in masseter, orbicularis oculi, and sterno-cleidomastoid muscles, and were significantly larger inmasseter and sternocleidomastoid muscles. The reducedASR probability and area-under-the-curve in orbicularisoculi in BSP patients are most likely due to chemoden-ervation following repeated BTX therapy. Although pa-tients were investigated at least 3 months after their lastBTX injection, the toxin may exert its direct muscle-relaxing effect for up to 6 months.18 Therefore, theresults of the present study indicate disinhibition of cra-nial muscles in patients with essential BSP. Loss ofinhibition of various brainstem reflexes has previouslybeen reported in patients with BSP: despite normal la-tency and duration of the blink reflex itself, recovery ofits R2 component following double-pulse stimulationwas reported to be enhanced.19 This increased reflexexcitability was not affected by BTX,4 but was partiallynormalized by apomorphine,20 and a “geste antagonis-tique.”21 BSP patients without a geste also displayeddefective somatosensory prepulse inhibition.22 The ex-teroceptive EMG suppression in sternocleidomastoid fol-lowing supraorbital nerve stimulation was reduced,23

although the excitatory short-latency trigemino-sterno-cleidomastoid response was normal in BSP patients.24

These findings suggest increased brainstem interneuronalexcitability, in agreement with apparent ASR disinhibi-tion in our BSP patients.

ASR onset latencies were not only shorter as com-pared to normal controls (in masseter, orbicularis oculi,and sternocleidomastoid), but also as compared to pa-tients with cervical dystonia12 who were previously stud-ied in the same laboratory applying identical methods (inmasseter, P � 0.031, and orbicularis oculi, P � 0.001;Mann–Whitney U). BSP patients in this study were olderthan the previously reported patients with cervical dys-tonia,12 further corroborating ASR disinhibition in BSP,as physiologically ASR latencies increase with advanc-ing age.11,16 Larger ASR area-under-the-curve (in mas-seter and sternocleidomastoid) as compared to normalcontrols and as compared to previously published cervi-

cal dystonia patients (all muscles combined, P � 0.004;masseter, P � 0.001; sternocleidomastoid, P � 0.021;biceps brachii, P � 0.001; Mann–Whitney U) clearlyindicate different pathophysiological mechanisms inBSP as compared to cervical dystonia.12

Similar to patients with cervical dystonia, those withBSP exhibited significant habituation of ASR probabilitywith repeated stimulation. Habituation is a characteristicfeature of polysynaptic reflexes. ASR habituation isthought to result from synaptic depression by cerebellarinput of brainstem interneurons localized in the pontinereticular formation, which seems to be intact in bothdystonic entities.25,26

At this point, we can only speculate about possibleneural candidate pathways underlying the observed ASRdisinhibition: The ASR is mediated by a pathway com-prising auditory afferents, the cochlear nucleus and thenucleus reticularis pontis caudalis (NRPC), whichproject to cranial and spinal motor neurons by means ofreticulobulbar and reticulospinal pathways.6,27 TheNRPC is the central key structure of this ASR-mediatingpathway. It receives cholinergic projections from thepedunculopontine tegmental nucleus (PPN), which ex-erts predominantly inhibitory effects by modulating pre-pulse inhibition and by reducing baseline ASRs.27 ThePPN receives dense and predominantly inhibitory basalganglia input, mainly from the globus pallidus and sub-stantia nigra pars reticulata.28,29 Intraoperative micro-electrode recordings in patients with primary cervicaland generalized dystonia demonstrate reduced activity ofthe internal pallidal segment (GPi) at rest and duringmovement. Therefore, reduced pallidal inhibition of thePPN would be expected to reduce startle probability andmagnitude, as was shown in a previous study by ourgroup of patients with cervical dystonia.

However, Garcia-Rill and colleagues30 were able todemonstrate a frequency-dependent differential effect ofdescending projections from the PPN in the cat. Stimu-lation at approximately 60 Hz induced prolonged tonicfiring in NRPC cells, with subsequent recruitment ofreticulospinal neurons while stimulation at lower (10 Hz)or higher (100 Hz) frequencies induced shorter or noresponses in some NRPC cells, which may suppressmuscle tone by means of active inhibition.30 It thusappears that stimulus frequency serves to determine adifferential effect within the same target cell popula-tion.31 In that context, reduced pallidal inhibition of thePPN could result in overactivity of the NRPC withsubsequent ASR facilitation. Indeed, patients with BSPshowed higher ASR probabilities (masseter, sternoclei-domastoid, biceps brachii), shorter ASR onset latencies(masseter, orbicularis oculi, sternocleidomastoid), and

270 J. MULLER ET AL.

Movement Disorders, Vol. 22, No. 2, 2007

larger ASR area-under-the-curve (masseter, sternoclei-domastoid) as compared with controls.

An alternative explanation might be primary overac-tivity of the GPi in BSP, which was shown recently bydetailed microelectrode recordings in a patient with se-vere primary blepharospasm and mild oromandibulardystonia,32 a finding that is contradictory to microelec-trode recording results in generalized and cervical dys-tonia. Therefore, GPi overactivity in BSP would increasepallidal inhibition of the PPN, and in turn result indisinhibition of the primary startle circuit.

The differential pattern of changes in orbicularis oculimuscle as compared with other recorded muscles in thisand other previously published studies on ASRs warrantsfurther discussion: there is ongoing debate as to whetherASRs in orbicularis oculi truly belong to the startlereaction, or whether they additionally reflect an auditoryblink reflex.8,11,25 The basal ganglia circuitry controllingthe excitability of the startle reaction by means of PPN isdifferent from the one controlling trigemino-facial inter-neuronal excitability. The latter involves the substantianigra pars reticulata, which inhibits the superior collicu-lus and modulates the blink reflex by means of tecto-bulbar projections.33 Thus a pattern of abnormalities inorbicularis oculi being different from that in other mus-cles seems possible.

In summary, the present study demonstrates: (1) pre-served organization of intrinsic neural pathways mediat-ing the startle response in primary BSP, and (2) disinhi-bition of ASR circuits in cranial muscles of BSP patientswith the exception of the orbicularis oculi muscle show-ing smaller and fewer ASRs, probably resulting fromBTX-induced chemodenervation in this muscle. Thesefindings are consistent with previous findings of blinkreflex abnormalities in patients with primary blepharo-spasm but point to a different pattern of brainstem dys-function in patients with BSP as compared to cervicaldystonia.

Acknowledgments: The authors thank Katharina Jeller,Sandra Voppichler, and Maria Hoch for their expert technicalassistance, and Ellen Quirbach for her help with editing of themanuscript.

REFERENCES

1. Schmidt KE, Linden DE, Goebel R, Zanella FE, Lanfermann H,Zubcov AA. Striatal activation during blepharospasm revealed byfMRI. Neurology 2003;60:1738–1743.

2. Tolosa E, Montserrat L, Bayes A. Blink reflex studies in focaldystonias: enhanced excitability of brainstem interneurons in cra-nial dystonia and spasmodic torticollis. Mov Disord 1988;3:61–69.

3. Pauletti G, Berardelli A, Cruccu G, Agostino R, Manfredi M. Blinkreflex and the masseter inhibitory reflex in patients with dystonia.Mov Disord 1993;8:495–500.

4. Valls-Sole J, Tolosa ES, Ribera G. Neurophysiological observa-tions on the effects of botulinum toxin treatment in patients withdystonic blepharospasm. J Neurol Neurosurg Psychiatry 1991;54:310–313.

5. Vitek JL. Pathophysiology of dystonia: a neuronal model. MovDisord 2002;17:S49–S62.

6. Yeomans JS, Frankland PW. The acoustic startle reflex: neuronsand connections. Brain Res Rev 1996;21:301–314.

7. Davis M, Gendelman DS, Tischler MD, Gendelman PM. A pri-mary acoustic startle circuit: lesion and stimulation studies. J Neu-rosci 1982;2:791–805.

8. Brown P, Rothwell JC, Thompson PD, Britton TC, Day BL,Marsden CD. New observations on the normal auditory startlereflex in man. Brain 1991;114:1891–1902.

9. Fariello RG, Schwartzman RJ, Beall SS. Hyperekplexia exacer-bated by occlusion of posterior thalamic arteries. Arch Neurol1983;40:244–246.

10. Swerdlow NR, Braff DL, Geyer MA. GABAergic projection fromnucleus accumbens to ventral pallidum mediates dopamine-in-duced sensorimotor gating deficits of acoustic startle in rats. BrainRes 1990;532:146–150.

11. Kofler M, Muller J, Valls-Sole J. Auditory startle responses as aprobe of brainstem function in healthy subjects and patients withmovement disorders. In: Cruccu G, Hallett M, editors. BrainstemFunction and Dysfunction. Amsterdam: Elsevier Science BV;2006. p 232–248.

12. Muller J, Kofler M, Wenning GK, Seppi K, Valls-Sole J, PoeweW. Auditory startle response in cervical dystonia. Mov Disord2003;18:1522–1526.

13. Valls-Sole J, Valldeoriola F, Tolosa E, Nobbe F. Habituation of theauditory startle reaction is reduced during preparation for execu-tion of a motor task in normal human subjects. Brain Res 1997;751:155–159.

14. Kofler M, Muller J, Wenning GK, et al. The auditory startlereaction in parkinsonian disorders. Mov Disord 2001;16:62–71.

15. Gogan P. The startle and orienting reactions in man. A study oftheir characteristics and habituation. Brain Res 1970;18:117–135.

16. Kofler M, Muller J, Reggiani L, Valls-Sole J. Influence of age onauditory startle responses in humans. Neurosci Lett 2001;307:65–68.

17. Kofler M, Muller J, Reggiani L, Valls-Sole J. Influence of genderon auditory startle responses. Brain Res 2001;921:206–210.

18. Billante CR, Zealear DL, Billante M, et al. Comparison of neuro-muscular blockade and recovery with botulinum toxins A and F.Muscle Nerve 2002;26:395–403.

19. Berardelli A, Rothwell JC, Day BL, Marsden CD. Pathophysiol-ogy of blepharospasm and oromandibular dystonia. Brain 1985;108:593–608.

20. Napolitano A, Bonuccelli U, Rossi B. Different effects of levodopaand apomorphine on blink reflex recovery cycle in essential bleph-arospasm. Eur Neurol 1997;38:119–122.

21. Gomez-Wong E, Marti MJ, Cossu G, Fabregat N, Tolosa ES,Valls-Sole J. The “geste antagonistique” induces transient modu-lation of the blink reflex in human patients with blepharospasm.Neurosci Lett 1998;251:125–128.

22. Gomez-Wong E, Marti MJ, Tolosa ES, Valls-Sole J. Sensorymodulation of the blink reflex in patients with blepharospasm.Arch Neurol 1998;55:1233–1237.

23. Carella F, Ciano C, Musicco M, Scaioli V. Exteroceptive reflexesin dystonia: a study of the recovery cycle of the R2 component ofthe blink reflex and of the exteroceptive suppression of the con-tracting sternocleidomastoid muscle in blepharospasm and torti-collis. Mov Disord 1994;9:183–187.

24. Quartarone A, Girlanda P, Di Lazzaro V, Majorana G, Battaglia F,Messina C. Short latency trigemino-sternocleidomastoid responsein muscles in patients with spasmodic torticollis and blepharo-spasm. Clin Neurophysiol 2000;111:1672–1677.

AUDITORY STARTLE REACTION IN BLEPHAROSPASM 271

Movement Disorders, Vol. 22, No. 2, 2007

25. Chokroverty S, Walczak T, Hening W. Human startle reflex:technique and criteria for abnormal response. ElectroencephalogrClin Neurophysiol 1992;85:236–242.

26. Kofler M, Muller J, Seppi K, Wenning GK. Exaggerated auditory startleresponses in multiple system atrophy: a comparative study of Parkinsonand cerebellar subtypes. Clin Neurophysiol 2003;114:541–547.

27. Koch M. The neurobiology of startle. Prog Neurobiol 1999;59:107–128.

28. Fendt M, Koch M. Cholinergic modulation of the acoustic startleresponse in the caudal pontine reticular nucleus of the rat. EurJ Pharmacol 1999;370:101–107.

29. Nandi D, Aziz TZ, Liu XG, Stein JF. Brainstem motor loops in thecontrol of movement. Mov Disord 2002;17:S22–S27.

30. Garcia-Rill E, Skinner RD, Miyazato H, Homma Y. Pedunculo-pontine stimulation induces prolonged activation of pontine retic-ular neurons. Neuroscience 2001;104:455–465.

31. Takakusaki K, Habaguchi T, Ohtinata-Sugimoto J, Saitoh K, Saka-moto T. Basal ganglia efferents to the brainstem centers controllingpostural muscle tone and locomotion: a new concept for under-standing motor disorders in basal ganglia dysfunction. Neuro-science 2003;119:293–308.

32. Foote KD, Sanchez JC, Okun MS. Staged deep brain stimulationfor refractory craniofacial dystonia with blepharospasm: case re-port and physiology. Neurosurgery 2005;56:E415.

33. Basso MA, Powers AS, Evinger C. An explanation for reflex blinkhyperexcitability in Parkinson’s disease. 1. Superior colliculus.J Neurosci 1996;16:7308–7317.

Holmes-Like Tremor of theLower Extremity Following

Brainstem Hemorrhage

Melanie Walker, MD, Hojoong Kim, MD, andAli Samii, MD*

Department of Neurology, University of Washington Schoolof Medicine, Seattle, Washington, USA

Abstract: Holmes tremor is an arrhythmic, 2- to 5-Hz rest-ing, postural, and kinetic upper extremity movement dis-order that occurs weeks to months after acute mesence-phalic pathology. We present a patient who developedtremor in three body parts postbrainstem hemorrhage withsubsequent hypertrophic olivary degeneration and discussthe relevant clinical evolution. Our case is unique because

in addition to expected upper extremity and cervical dys-tonic head tremors, the patient also developed a severelower extremity movement disorder, which we believe tobe a form of Holmes tremor. Tremor involving the lowerextremity in this setting has not been previouslyreported. © 2006 Movement Disorder Society

Key words: hypertrophic olivary degeneration; leg tremor;Holmes tremor; hemorrhage; brainstem

In 1904, Gordon Holmes described a unique motorabnormality characterized by low frequency tremor atrest, which was exacerbated by posture and accentuatedwith intention.1 Brainstem pathology, such as intracra-nial hemorrhage, can result in upper extremity Holmestremor, which occurs late in the course of recovery, andmay correlate anatomically with hypertrophic olivarydegeneration (HOD).2,3 Hemidystonia and torticollismay rarely result from brainstem lesions that extendrostrally into the thalamic or subthalamic regions.4 Cer-vical dystonia with superimposed dystonic head tremor,upper limb tremor, and severe lower limb tremor havenot previously been described in the same patient fol-lowing a brainstem hemorrhage with unilateral HOD.

CASE REPORT

A healthy 43-year-old man experienced sudden onsetleft hemiparesis, left hypesthesia, right lower motor neu-ron facial paralysis, severe dysarthria, and right upperand lower limb dysmetria. The first brain magnetic res-onance imaging (MRI) performed 10 days after symptomonset (Fig. 1A) corroborated initial head computerizedtomography (CT), which showed hemorrhage extendingfrom the brachium pontis through the dorsal midbrain onthe right. Routine serum and urine analysis were normal.Coagulation studies were within the standard range. Thepatient had no history of hypertension. After an extendedIntensive Care Unit stay, the patient was eventuallytransferred to a Rehabilitation Center at 1 month pos-themorrhage where he continued to have left hemipare-sis, right peripheral facial weakness, dysphagia requiringa feeding tube, and profound disequilibrium. MRI/MRangiography 2 months posthemorrhage did not reveal anunderlying arteriovenous malformation, but T2-weightedsequences showed a central area of hyperintensity sur-rounded by a hypointense ring suggestive of a cavernousmalformation as the cause of the hemorrhage.

Five months after his initial hemorrhage, the patientdeveloped a coarse, large amplitude tremor of the leftarm present at rest and posture, but much more severewith action. The patient also developed a torticollis to theleft and a resulting irregular dystonic head tremor. Initialpharmacotherapies to treat the tremors were unsuccessful

This article includes Supplementary Video, available online at http://www.interscience.wiley.com/jpages/0885-3185/suppmat

*Correspondence to: Dr. Ali Samii, Department of Neurology, VAPuget Sound Health Care System, 1660 S. Columbian Way, MS-127,Seattle, WA 98108. E-mail: asamii@u.washington.edu

Received 5 July 2006; Revised 22 August 2006; Accepted 7 Sep-tember 2006

Published online 5 December 2006 in Wiley InterScience (www.interscience.wiley.com). DOI: 10.1002/mds.21271

272 M. WALKER ET AL.

Movement Disorders, Vol. 22, No. 2, 2007