VIDEO Eye movementabnormalities in stiff

person syndrome

AbstractThe authors describe a 38-year-old woman with stiff person syn-drome (SPS) and gaze-holding nystagmus, limited abduction, vertical and hor-izontal ocular misalignment, deficient smooth pursuit, and impaired saccadeinitiation. There was no evidence of ocular myasthenia, indicating that abnor-malities of ocular motor function can occur as a primary manifestation of SPS,perhaps from depletion of GABA.

NEUROLOGY 2005;65:14621464

John R. Economides, PhD; and Jonathan C. Horton, MD, PhD

Stiff person syndrome (SPS) is an autoimmune dis-ease characterized by muscle rigidity, spasm, andcirculating antibodies against glutamic acid decar-boxylase (GAD), the synthetic enzyme for GABA.1Surprisingly, ocular symptoms have not been de-scribed as a prominent feature of SPS, althoughGABAergic neurons are critical for brainstem controlof eye movements.2 In a recent report, a patient wasdescribed with alternating esodeviation, bilateral ab-duction weakness, hypometric saccades, nystagmusand antiacetylcholine receptor antibodies.3 The au-thors identified four similar reports in the literatureand suggested that gaze disorders in SPS arise frommyasthenia gravis. Here we describe a patient witheye movement abnormalities and SPS, without evi-dence of myasthenia gravis.

Case report. A 38-year-old woman reported that in 1993 shedeveloped diplopia on lateral gaze to either side and nystagmus.Within a few years, she became disabled by progressive musclestiffness, cramps, and ataxia. Spinal fluid analysis was unreveal-ing, except for an elevated IgG synthesis rate. Tests for anti-Hu,Yo, and Ri antibodies as well as thyroid autoantibodies were neg-ative. In 1998, electromyography revealed continuous motor activ-ity in the right gastrocnemius, quadriceps, and lumbarparaspinous muscles after brief, low-intensity stimulation of theright ankle. The diagnosis of SPS was confirmed by a serumanti-GAD65 antibody level of 187 nmol/L (normal 0.02), docu-mented in 1998 by radioimmunoassay.

In 2004, our examination (see video on the Neurology Web siteat www.neurology.org) showed mild deficiency of abduction ineach eye. In primary gaze, there was an alternating esotropia ofabout 8 degrees and a right hypertropia of 1 to 2 degrees, with aleft-eye fixation preference. Horizontal smooth pursuit had a low

gain and was asymmetric. Horizontal saccades were often initi-ated with a blink. When the patient was instructed to avoid blink-ing, the saccades became hypometric. Gaze-holding nystagmuswas present on eccentric gaze in all directions. Vertical saccadesand pursuit were normal. There was no ptosis. Serological testingrevealed no evidence for antiacetylcholine receptor antibodies(0.1 nmol/L). A chest CT scan was normal. A repeat determina-tion of the anti-GAD antibody level yielded a value 30 nmol/L.An MRI was normal, except for subtle midline atrophy of thecerebellar cortex (figure 1).

Methods. The patients eye movements were recorded binocu-larly while she was seated with her head in a chin rest facing atangent screen. Computer-controlled spots (size 0.5 degrees;Cambridge Research Systems, England) were rear-projected ontothe screen to control the patients fixation behavior. Eye move-ments were monitored at 60 Hz with a spatial resolution of 1degree using two infrared video eye-tracking cameras (SensoMo-toric Instruments, Germany). Analog voltages representing theposition of each eye and the target spot were recorded digitally foroffline analysis (Power1401 and Spike2, Cambridge ElectronicsDesign, England). The gain and offset of each eye was calibratedindependently (while the fellow eye was covered) by asking thepatient to fixate a nine-point grid of known visual field locations.The first task involved smooth pursuit of a target moving sinusoi-dally in different directions at different speeds and amplitudes.The second task involved making a saccade to a target appearingintermittently in different locations. Eye movement traces weresmoothed with a Gaussian kernel and records of the patientsblinks were removed with a running median filter. Eye and targetvelocities were computed offline using central point digitaldifferentiation.

Results. Figure 2A demonstrates the change in themagnitude of the patients horizontal misalignment withrespect to orbital location of the preferred left eye. She wasasked to fixate a target spot appearing randomly at 25degrees left to 25 degrees right in 5-degree steps along thehorizontal meridian. Each data point represents the posi-tion of the fixating left eye plotted against the difference inposition between the two eyes (angle of esodeviation). Dur-ing rightward gaze, the angle of esotropia was stable, outto an eccentricity of 25 degrees. The mean value of theright esodeviation was 8.5 degrees. On leftward gaze, theeyes became progressively more crossed at eccentricitiesgreater than 10 degrees, indicating an incomitant ocularmisalignment. Her vertical deviation was comitant.

Figure 2B shows the effect of gaze position on nystag-mus. At central fixation the eyes were relatively stable. At30 degrees of eccentric gaze in any direction, a jerk nystag-mus emerged beating in the direction of the eccentric gaze.The velocity of the slow drifts during fixation at 30 degreesto the left and to the right was approximately equal.

During smooth pursuit of sinusoidally oscillating targetspots (20 degrees, 0.1 to 1 Hz), the patient demonstrateddeficient gain, compensated for by catch-up saccades (fig-

Additional material related to this article can be found on the NeurologyWeb site. Go to www.neurology.org and scroll down the Table of Con-tents for the November 8 issue to find the title link for this article.

From the Beckman Vision Center, University of California, San Francisco,CA.

Supported by the National Eye Institute (EY015343, EY10217, andEY02162), the Larry L. Hillblom Foundation, and Research to PreventBlindness.

Disclosure: The authors report no conflicts of interest.

Received April 1, 2005. Accepted in final form July 12, 2005.

Address correspondence and reprint requests to Dr. Jonathan C. Horton,Beckman Vision Center, University of California San Francisco, 10 KoretWay, San Francisco, CA, 94143-0730; e-mail: horton@itsa.ucsf.edu

1462 Copyright 2005 by AAN Enterprises, Inc.

ure 3A). Her deficiency in smooth pursuit gain was moreobvious on leftward pursuit. There were large rightwardnystagmic drifts that could be overcome only by makingnumerous catch-up saccades to the left. Rightward pursuitshowed a higher gain with fewer catch-up saccades. How-ever, it was also deficient compared with normal subjects.

Perhaps to compensate for deficient smooth pursuit, thepatient often switched ocular fixation while binocularlytracking a target moving in a predictable sinusoidal trajec-tory. As the target moved to the left, the angle of rightesotropia increased, prompting the patient to switch fixa-tion to her right eye (figure 3B). When the target movedback to the right, the patient switched back to her left eye.

To investigate the possibility of myasthenia gravis, 10mg of edrophonium chloride was injected IV during sinu-soidal smooth pursuit 20 degrees to the left and right. Nochange was seen in the magnitude of the right esodevia-tion or pursuit velocity (data not shown).

Discussion. The only previous study to include eyemovement recordings of a patient with SPS con-cluded that oculomotor deficits occur from coexistingocular myasthenia.3 The risk of developing a secondautoimmune disease, such as myasthenia gravis, ap-parently is elevated in patients with SPS. However,our patient showed no clinical evidence of myasthe-nia gravis over 12 years. She had no detectable levelof antiacetylcholine receptor antibodies, no thy-moma, no ptosis, and a negative edrophonium test.Therefore, we conclude that her right esodeviation,deficient pursuit, impaired saccadic initiation, andnystagmus represent primary manifestations of SPS.

In normal subjects, the folia of the vermis oftenappear more prominent than those of the cerebellarhemispheres. Nonetheless, their prominence in ourpatient was excessive. The Purkinje cells of the ver-mis encode gaze velocity during smooth pursuit.2Their depletion, documented by neuropathologicalstudies,4 may explain our radiologic finding of ver-mal atrophy and could contribute to eye movementabnormalities in SPS.

The gaze-evoked nystagmus that we documented

Figure 1. T1-weighted MRIs showing mild atrophy of thecerebellar vermis. (A) Normal right cerebellar hemi-spheres, 2.1 cm from the sagittal plane. (B) Through themidline, subtle vermal atrophy is revealed by the relativeprominence of the cortical folia. Scale bar 2 cm.

Figure 2. (A) Increasing esodeviation onleft gaze. The fixating left eyes positionis graphed on the abscissa; the ordinateshows the amplitude of the right esotro-pia, determined by subtracting right eyeposition from left eye position for datasampled at 60 Hz during episodes ofattempted stable fixation (rather thanduring purposeful dynamic eye move-ments). The exponential fit line for thedata shows an increasing eso-deviationfor leftward, but not rightward gaze.(B) Horizontal eye position recordingsfor the left eye during attempted fixa-tion show gaze-holding nystagmus. The

three sample traces show fixation of a spot directly in front of the patient and 30 degrees to either side. Upward deflec-tions of the eye position trace represent rightward horizontal movements of the eye. In primary position there was a slightrightward drift (0.3 0.7 degree/second); left eccentric fixation resulted in a mean drift velocity back to the right of 6.1 1.0 degree/second, right eccentric fixation produced a mean drift velocity to the left of 8.4 1.3 degree/second.

November (1 of 2) 2005 NEUROLOGY 65 1463

may be due to an inappropriate eye position signaloriginating from the brainstem. The nucleus preposi-tus hypoglossi (NPH) and the medial vestibular nu-cleus (MVN) form the horizontal neural integrator,responsible for maintaining the eyes at eccentric or-bital positions. Injections of GABAA antagonists intothe NPH and the MVN disrupt normal gaze-holdingat different eccentricities.5,6

Axons forming the cortico-ponto-cerebellar path-way subserving smooth pursuit make synapses inthe dorsolateral pontine nucleus (DLPN). The ponssignals the vestibulo-cerebellum via mossy fibers. In-activation of the DLPN with muscimol, a GABAAantagonist, results in ipsilateral deficits in pursuitgain.7 The climbing fiber input to the flocculus origi-nates from the dorsal cap of the inferior olive. It isdriven by input from the nucleus of the optic tract(NOT). The NOT, which is thought to underlie theslow build up of eye velocity in optokinetic nystag-mus, has also been shown to mediate gaze holdingand smooth pursuit. Pharmacological manipulationsof the GABAergic neurons in the NOT leads to bothnystagmus and ipsilateral smooth pursuit asymme-try.8 In addition, other areas (including the deep cer-ebellar nuclei and the superior colliculus) haveGABA-mediated effects on gaze.9,10

We propose that the oculomotor deficits exhibitedby our patient were caused by dysfunction of

GABAergic pathways. This report expands the spec-trum of clinical findings in SPS to include horizontalgaze limitation, ocular misalignment, nystagmus,impaired smooth pursuit, and poor saccadeinitiation.

References1. Murinson BB. Stiff-person syndrome. Neurologist 2004;10:131137.2. Leigh RJ, Zee DS. The Neurology of Eye Movements, 3rd ed. New York:

Oxford University Press, 1999.3. Thomas S, Critchley P, Lawden M, et al. Stiff person syndrome with eye

movement abnormality, myasthenia gravis, and thymoma. J NeurolNeurosurg Psychiatry 2005;76:141142.

4. Warich-Kirches M, Von Bossanyi P, Treuheit T, et al. Stiff-man syn-drome: possible autoimmune etiology targeted against GABA-ergiccells. Clin Neuropathol 1997;16:214219.

5. Arnold DB, Robinson DA, Leigh RJ. Nystagmus induced by pharmaco-logical inactivation of the brainstem ocular motor integrator in monkey.Vision Res 1999;39:42864295.

6. Mettens P,Godaux E, Cheron G, Galiana HL. Effect of muscimol micro-injections into the prepositus hypoglossi and the medial vestibular nu-clei on cat eye movements. J Neurophysiol 1994;72:785802.

7. Ono S, Das VE, Mustari MJ. Role of the dorsolateral pontine nucleus inshort-term adaptation of the horizontal vestibuloocular reflex. J Neuro-physiol 2003;89:28792885.

8. Mustari MJ, Tusa RJ, Burrows AF, Fuchs AF, Livingston CA. Gaze-stabilizing deficits and latent nystagmus in monkeys with early-onsetvisual deprivation: role of the pretectal not. J Neurophysiol 2001;86:662675.

9. Basso MA, Krauzlis RJ, Wurtz RH. Activation and inactivation of ros-tral superior colliculus neurons during smooth-pursuit eye movementsin monkeys. J Neurophysiol 2000;84:892908.

10. Robinson FR, Straube A, Fuchs AF. Participation of caudal fastigialnucleus in smooth pursuit eye movements. II. Effects of muscimol inac-tivation J Neurophysiol 1997;78:848859.

Figure 3. (A) Horizontal sinusoidal smooth pursuit shows deficient gain, especially on leftward pursuit. The bottom panelshows the position of the left eye (blue trace) superimposed over the target position. The top panel shows horizontal eyevelocity (green trace) and target velocity (black trace); the purple trace is a fit line plotted through the nonsaccadic por-tions of the eye velocity trace and represents the smooth pursuit velocity of the eye during tracking. The eye velocity traceis artificially clipped at 50 degrees. Note the lower peak eye velocity on leftward tracking, resulting in more catch-upsaccades. (B) Binocular horizontal eye movement recordings during smooth pursuit. The patient initially pursued the tar-get with her favored left eye, but at eccentric leftward gaze she began to pursue with the right eye (highlighted in gray).Near central fixation, she resumed tracking with her left eye.

1464 NEUROLOGY 65 November (1 of 2) 2005