recovery of peripheral versus central nerves identified by ......abducens nerve palsy lowers saccade...

417

Ann. N.Y. Acad. Sci. 1039: 417–429 (2005). © 2005 New York Academy of Sciences.doi: 10.1196/annals.1325.039

Recovery of Peripheral Versus CentralNerves Identified by Saccadic VelocityAfter Abducens Neuropathy

JAMES A. SHARPE, KYLEN MCREELIS, AND AGNES M.F. WONG

Division of Neurology and Department of Ophthalmology and Vision Sciences,University Health Network, University of Toronto, Toronto, Ontario, Canada

ABSTRACT: The abducens is the motor nerve with the most substantial course,both within and outside the brain and it innervates only one muscle. Sixthnerve palsy affords an opportunity to compare recovery after central versusperipheral nerve damage by assessing the dynamics of abduction. Horizontalsaccade peak velocities and durations in 14 patients with unilateral peripheralsixth nerve palsies (5 acute, 9 chronic) are compared with those in 5 patientswith central sixth nerve palsies (2 acute, 3 chronic) and with those in 10 normalsubjects. Acutely, abducting saccades in the paretic eye were slow in both cen-tral and peripheral palsies, as anticipated from weakness of the lateral rectusmuscle. In chronic central palsies, abducting saccadic velocities remained re-duced, but in chronic peripheral palsies, they increased to normal within thelimited range of excursion. The chronically damaged peripheral nerve behaveslike a high-pass filter in transmitting phasic velocity commands, whereas tonicposition commands remain defective, accounting for limited abduction butnormal velocities within the range of duction. In chronic central (fascicular)palsies, saccade velocities remain reduced. Impaired conduction from damageto central myelin or axons is more persistent in central palsies, consistent withlimited regeneration within the brain. Recording of saccade velocities may aidthe distinction of fascicular from peripheral palsies. Saccade speed is repairedin peripheral palsies, probably by remyelination, and perhaps also by centralmonocular adaptation of innervation selectively to the paretic eye in order todrive both eyes rapidly and simultaneously to a target in the paretic field ofmotion.

KEYWORDS: saccades; sixth nerve palsy; abducens nerve; repair; regenera-tion; adaptation; saccade velocity; saccade duration; Hering’s law

INTRODUCTION

The abducens is the motor nerve with the most substantial course both within andoutside the brain, where it is vulnerable to damage. Comparison of recovery aftercentral versus peripheral nerve palsy provides an opportunity to investigate differ-

Address for correspondence: Dr. James A. Sharpe, University Health Network—WW5-440TWH, 399 Bathurst Street,, Toronto, Ontario, Canada M5T 2S8. Voice: 416-603-5950; fax:416-603-5596.

[email protected]

418 ANNALS NEW YORK ACADEMY OF SCIENCES

ences in repair properties between the central and peripheral nervous systems. Theabducens nerve innervates only one muscle, the lateral rectus, so the amount of re-covery can be quantified readily by assessing the dynamic properties of that musclealone.

Abducens nerve palsy lowers saccade amplitude and velocity in the direction ofaction of the paretic lateral rectus muscle.1–7 We investigated patients with unilateralsixth nerve palsy and assessed the saccades of both the paretic and nonparetic eyesduring monocular viewing with each eye. We studied idiopathic peripheral palsyversus central (fascicular) palsy of structural cause and compared the effects of thetwo sites of damage and the duration of the palsy on the speed of saccades.

METHODS AND SUBJECTS

Among the 19 patients with unilateral sixth nerve palsy, 14 patients had idio-pathic, presumed ischemic, peripheral palsy (mean age 63 ± 10 years, range 46 to77, median 64; nine were men). All 14 had normal MR or CT imaging. Four patientshad identifiable ischemic risk factors (hypertension or diabetes mellitus). Symptomdurations ranged from three weeks to 96 months (mean duration 21 months). Pa-tients with diplopia of less than one month duration were classified as having acutepalsy; all others were designated as chronic. Five patients with peripheral palsy weretested acutely (within one month of symptom onset). Serial eye-movement record-ings were performed on these five patients with acute peripheral palsy, first at pre-sentation and then at two months after symptom onset.

Five patients had central (fascicular) lesions identified on MR imaging. Two hadcavernomas and three had demyelinating lesions from multiple sclerosis, involvingthe abducens nerve fascicle in the pons, but sparing the abducens nucleus. They hadno other ocular motor abnormalities. Their mean age was 52 ± 18 years (range 30 to75 years, median 56; two were men). The duration of symptoms ranged from oneweek to 132 months (mean duration 38 months). Two patients were tested acutely(within one month of symptom onset). Serial recordings were performed on thesetwo patients at presentation and at two months after symptom onset.

All patients had an incomitant esotropia with a clinically obvious unilateral ab-duction deficit. Mean abduction deficit was the same in both lesion-site groups: 80%of normal (range 60 to 95%) in peripheral, and 75% of normal (range 70 to 90%) incentral fascicular palsy. Ductions in the fellow eye were normal.

Ten normal subjects served as controls (five men; mean age 49 ± 12 years; medianage 55; age range 19 to 69). Investigations followed the tenets of the Declaration ofHelsinki and the Ethics Review Board of the University Health Network, and in-formed consent was obtained from each participant.

Oculography

Participants fixated a red laser spot of 0.25-degree diameter 1 m away from thenasion, which stepped at 3-s intervals 10° right, then back to center, then 10° left,cycling through this pattern 20 times. The study design of 10° target steps assuredthat testing was within the limited range of abduction, as patients were prospectivelyentered. Head position was stabilized with an occipital support and monitored by a

419SHARPE et al.: SACCADES IN SIXTH NERVE PALSY

search coil taped to the forehead. With the normal eye covered, patients were in-structed to follow the laser spot as it stepped among positions. The target sequencewas then repeated viewing with the normal eye while the paretic eye was covered.

Eye positions were measured by a three-dimensional magnetic search coil tech-nique, using 6 ft (183 cm) diameter coils field arranged in a cube (CNC Engineering,Seattle, WA). In each eye, the patient wore a dual-lead scleral coil annulus designedto detect horizontal, vertical, and torsional gaze positions (Skalar Instrumentation,Delft, The Netherlands). Eye position data were filtered with a bandwidth of 0 to90 Hz and digitized at 500 Hz.

Saccade amplitudes, peak velocities, and durations were analyzed for the pareticand nonparetic eyes. In all patients, the peak velocity–amplitude and duration–amplitude relationships in both the paretic and nonparetic eyes did not differ be-tween viewing with the paretic or the nonparetic eye. We report the peak velocity–amplitude and duration–amplitude relationships during paretic eye viewing.

RESULTS

Saccades in Peripheral Abducens Nerve Palsy

Peak velocity–amplitude and duration–amplitude relationships were correlatedwith bins of amplitude in the 2-degree ranges (FIGS. 1 and 2) because the palsies pre-cluded measuring large-amplitude saccades to determine asymptotic features of thepeak velocity–amplitude main sequence. In the acute state, centrifugal abductingsaccades in the paretic eye had reduced peak velocities (FIG. 1) and longer durations

FIGURE 1. Group mean peak velocity versus amplitude (in bins) of initial saccades inthe paretic eye of 14 patients with idiopathic (presumed ischemic) sixth nerve palsies. Cen-trifugal abducting saccades from center to 10° abducted position in peripheral palsy weresignificantly slowed acutely (<1 month of diplopia) but normal in the chronic state.


FIGURE 2. Mean duration versus amplitude (in bins) of initial saccades in the pareticeye. (A) Centrifugal abducting saccades from center to 10° abducted position in peripheralpalsy are prolonged acutely but normal in the chronic state (>1 month). (B) Centripetal ab-ducting saccades from 10° adducted position to center in peripheral palsy are normal bothacutely and chronically.


FIGURE 2 — continued. (C) In central (fascicular) palsy, centrifugal abducting sac-cades from center to 10° abducted position are slowed both acutely and chronically. (D)Centripetal abducting saccades from 10° adducted position to center in central (fascicular)palsy are normal.


for any given amplitude (FIG. 2A) when compared with normal controls (P < .05;Wilcoxon rank sum test), as anticipated from weakness of the lateral rectus muscle.However, in the chronic state, centrifugal abducting saccadic peak velocities (FIG. 1)and durations (FIG. 2A) in the paretic eye were within the normal range, despite lim-ited abduction in the nine chronic palsies. In the five initially acute palsies tested se-rially, saccade peak velocities and durations became normal within two months ofsymptom onset despite persisting limited abduction. Centripetal abducting saccadesfrom 10° adducted position to center in the paretic eye had normal peak velocities(FIG. 3A) and durations (FIG. 2B), both in acute and chronic states. Saccade peak ve-locities and durations in the nonparetic eye were normal in both acute and chronicstates.

FIGURE 3. Centripetal abducting saccades from 10º adducted position to center in pe-ripheral (A) and in central (fascicular) palsy (B) had normal peak velocities in both acuteand chronic palsies.


Saccades in Central Abducens Nerve Palsy

Centrifugal abducting saccades in the paretic eye had reduced peak velocities(FIG. 4) and had longer durations (FIG. 2C) for any given amplitude, both in the acuteand chronic states, when compared with normal controls (P < .05; Wilcoxon ranksum test). Persistent slowing of abducting saccades was recorded in all five chronicpalsies (including the two cases of acute central palsies that were recorded withinone month of onset of diplopia and later at two months after onset). Centripetal ab-ducting saccades from the 10° adducted position to center in the paretic eye had nor-mal peak velocities (FIG. 3B) and durations (FIG. 2D), both in acute and chronicstates. Saccade peak velocities and durations in the nonparetic eye were normal inboth acute and chronic states.

DISCUSSION

Information about saccadic dynamics in paralytic strabismus is sparse. The ef-fects of duration and lesion site of abducens nerve palsy on saccade dynamics hadnot been systematically investigated and the potential for recovery or adaptation hadnot been assessed. We measured the effects of idiopathic (presumed ischemic) pe-ripheral palsy, and central fascicular palsy on the peak velocity and duration of sac-cades, and assessed changes over time. Abducting saccades are hypometric and slowin the paretic eye in lateral rectus palsy.1–4 In the present investigation, abductingsaccades in the paretic eye had amplitudes appropriate to the ±10° target steps, be-cause our patients had mild to moderate palsies with full ductions within the testedrange. Acutely, centrifugal abducting saccades in the paretic orbital hemirange ofduction (from center to an abducted position) in the paretic eye were slow in both

FIGURE 4. Centrifugal abducting saccades from center to 10° abducted position incentral (fascicular) palsy had reduced peak velocities in both acute and chronic stages.


central and peripheral palsy, as anticipated from weakness of the lateral rectus mus-cle. In the chronic state, centrifugal abducting saccadic peak velocities remained re-duced and durations remain prolonged in the paretic eye in central palsy, but theybecame normal in peripheral palsy.

Centripetal abducting saccades (from an adducted position to center) in the paret-ic eye were normal in amplitude, peak velocity, and duration in both central and pe-ripheral palsy, whether acute or chronic. The two horizontal rectus muscles in eacheye contribute reciprocally to horizontal eye movements. Abducting movements re-quire an increase in force of the lateral rectus and a decrease in force of the medialrectus. The amount by which the force of the agonist increases and the antagonistdecreases depends on the initial position of the eye in the orbit.8 Most of the forcefor centrifugal abducting movements (starting from the orbital midposition to the ab-ducted hemirange) comes from contraction of the agonist lateral rectus muscle.However, when abducting movements are centripetal (starting in the adductedhemirange toward orbital midposition), most of the force for abduction comes fromrelaxation of the antagonist medial rectus muscle.8 Thus, in sixth nerve palsy, abduc-tion is little affected when the eye starts from an adducted (nasal) position, as mostof the change in force is contributed by relaxation of the normal antagonist, that is,the medial rectus. When the movement begins from orbital midposition, however,the change in force comes mainly from contraction of the agonist, that is, the pareticlateral rectus, and hence saccades in the paretic orbital hemirange of duction are ab-normal. Our finding that abducting saccades in the paretic orbital hemirange of duc-tion (from center to an abducted position) in the paretic eye are slow, whileabducting saccades from an adducted position to center in the paretic eye are normalcan be explained by the relative contribution of the agonist and antagonist musclesin different orbital positions.8

Paralytic strabismus may be accompanied by contracture (shortening and in-creased stiffness) in the nonparetic antagonist muscle,9–12 while the paretic musclelengthens in response to a change in orbital position of the globe. Contracture of thenonparetic antagonist is associated with a decrease in the number of sarcomeres,whereas lengthening of the paretic muscle is accompanied by an increase in sarcom-eres.13 The persistent reduction in centrifugal abducting saccadic velocities found inour patients with chronic central palsy is unlikely to be the result of contracture ofthe medial rectus (antagonist) muscle because the function of the medial rectuswould not be expected to differ between peripheral and central palsy of the same du-ration, but it did. Also, contracture of the medial rectus might be expected to causeslowing of centripetal abducting saccades (from adducted hemirange to orbital mid-position) and adducting saccades in the paretic eye; however, those velocities werenormal.

Recovery in Peripheral Abducens Nerve Palsy

The study of abducens nerve palsy provides a unique opportunity to investigatethe different repair properties of the central and peripheral nervous systems. Centralmyelin, consisting of membrane layers of oligodendrocytes and maintained by theircell bodies, extends for a few millimeters distal to the emergence of the nerve fromthe brain.14 The rest of the peripheral abducens nerve is ensheathed in peripheralmyelin, which consists of layers of Schwann cell membranes and is maintained by


their somata. The central (fascicular) portion of the nerve is surrounded by a glialenvironment that is generally inhospitable to neural recovery, whereas the peripheralnerve should have the capacity for recovery. In addition, because the abducens nerveonly innervates one muscle, the lateral rectus, the degree of recovery can be quanti-fied readily by assessing the dynamic properties of that muscle alone by the speedof the paretic eye during abduction.

By comparing saccade peak velocities and durations in central fascicular versusperipheral idiopathic abducens nerve palsy, we found that in peripheral palsy, ab-ducting saccadic peak velocity and duration for the tested range of excursion (10°)becomes normal, despite persistent esotropia and limitation of abduction. In con-trast, there was no improvement in saccadic peak velocity or duration in chronic cen-tral fascicular palsy. Our findings are concordant with the capacity for recovery ofperipheral nerves, as exemplified here by function of the abducens nerve after it exitsfrom the brain stem, and also concordant with the limited capacity for recovery ofcentral nerves, as measured by function of the abducens nerve after damage to itsfascicle.

The time course for recovery of peripheral palsy in our patients is consistent withdemyelinating damage, or neurapraxia consisting of local conduction block with nodistal Wallerian degeneration.15,16 If the myelin sheath is disrupted, recovery de-pends on remyelination, which has a time course of many days or weeks. In axonot-mesis or second-degree injury, the axons are disrupted. Wallerian degenerationoccurs, in which the distal axon and myelin undergo degradation and the proximalneuron soma undergoes chromatolysis. Regeneration occurs by the outgrowth ofmultiple axonal sprouts proximal to the site of lesion, and by growth of axons at arate of 1 to 2 mm a day.17

Histopathological correlation of peripheral ocular motor nerve palsy in sparse.18–21

In diabetic third nerve palsy18–20 the lesion is ischemic and primarily demyelinating,with a paucity of Wallerian degeneration distally, and an absence of chromatolysisin neuronal cell bodies in the oculomotor nucleus. These findings are compatiblewith the rapid and complete recovery of ocular motor nerve palsy commonly seen indiabetes mellitus.

No histopathological study of diabetic, hypertensive, or idiopathic sixth nervepalsy is available. In cats, axotomy of the abducens nerve is followed by structuraland functional recovery in about three months.21 We found that abducting saccadicpeak velocity and duration became normal within two months in patients with idio-pathic (presumed ischemic) peripheral palsy. Although the range of abduction re-mained limited in the peripheral sixth nerve palsies we studied, in contrast to thetypical complete recovery of abduction after presumed ischemic palsies, the timecourse of recovery of saccade speed and duration to normal in these patients suggeststhat their palsies resulted from a primarily demyelinating event, rather than fromaxonal disruption.

Recovery of Saccadic Velocity in Peripheral Abducens Nerve Palsy

In our patients with central palsy, both acute and chronic abducting saccades wereslow and of longer duration. In contrast, in our patients with peripheral palsy, ab-ducting saccadic peak velocity and duration became normal, in the face of persistentesotropia.


The partially recovered peripheral nerve appears to act as a high-pass filter, al-lowing the transmission of high firing rates during the pulse of innervation, such thatthe resulting saccades are of normal peak velocity and duration. In contrast, the par-tially recovered nerve may be unable to transmit low-frequency tonic signals (thestep of saccadic innervation), resulting in an abnormal tonic position (i.e., esotropia)and the limited abduction of the paretic eye, evident in each of our patients in theacute (<1 month) and chronic states.

A cellular correlate of disparity between the repair of saccade speed and the per-sistence of defective ranges of abduction may be the selective nature of the nerve fi-bers damaged or the tempo of their regeneration and remyelination. Although allocular motoneurons participate in saccadic and slow eye movements, and in the tonicdischarges that maintain eye positions,22,23 some may be specialized in function.Large motoneurons within the abducens nucleus innervate singly innervated twitch mus-cle fibers, whereas smaller motoneurons around its periphery innervate multiply inner-vated nontwitch muscle fibers.24,25 The high-pass filter response of the damagedperipheral abducens nerve to saccadic velocity commands, as described earlier,might be explained either by damage predominantly to nerve fibers from nontwitchmotoneurons just outside the abducens nucleus, or by better repair of nerve fibersfrom twitch motoneurons within the nucleus.

Central (Fascicular) Abducens Nerve Palsy

Damage to the fascicle of the nerve in the pons was determined by MR imaging.Sparing of the abducens nucleus was assured both by imaging and by normal speedsof adduction in the nonparetic eye (field of action of the contralateral yoked medialrectus muscle). In addition to lateral rectus motoneurons, the abducens nucleus con-tains internuclear neurons that excite medial rectus motoneurons of the opposite eye.

The central nervous system is a hostile molecular and cellular milieu for neuralregeneration.26 We found that abducting saccadic velocity remained reduced in pa-tients with central palsy from cavernomas or multiple sclerosis. Although multiplesclerosis is generally considered to be an inflammatory demyelinating disease withrelative axonal sparing, axonal damage, including loss, transection, or disturbed ax-onal transport, occurs early in the disease and during lesion development.27–31 Theabsence of recovery of saccade speed and range in our patients with central fascicu-lar palsy indicates limited regeneration of damaged axons and myelin in the glialenvironment of the brain.

Possible Role of Monocular Adaptation

We found that in chronic peripheral palsy, velocities of abducting saccades in theparetic eye became normal, without a conjugate increase in adducting saccadic ve-locity in the nonparetic eye (i.e., field of action of the contralateral yoked medial rec-tus muscle). These findings may be due to partial recovery of the peripheral nerve,as discussed earlier. Alternatively, they may represent a monocular readjustment ofinnervation selectively to the paretic eye. Hering32 suggested that the brain controlsgaze by two systems, one for conjugate movements and the other for vergence. Ifcommon and conjugate premotor signals were sent to motoneurons to both eyes, ad-ducting saccadic velocity in the yoked antagonist of the nonparetic eye would be in-


creased as well. For example, in the case of a left lateral rectus weakness from a leftsixth nerve palsy, any adaptive increase in innervation to the left lateral rectus wouldbe accompanied by increased innervation to the right medial rectus, in accord withHering’s “law.” However, this was not the case; yoked adduction of the fellow eyehad a normal velocity–amplitude relationship. Monocular adaptation of saccades tosixth nerve palsy would be consistent with evidence of monocular adaptation of thevestibulo-ocular reflex in sixth, fourth, or third nerve palsy.33–36

An anatomical and physiological prenuclear substrate for possible monocular ad-justments of saccade speed in response to peripheral nerve palsy is found in presac-cadic burst neurons of the caudal paramedian pontine reticular formation (PPRF).The great majority (about 80%) of short-lead burst neurons in the caudal PPRF of mon-keys, which was once thought to encode conjugate velocity commands for horizontal sac-cades, actually encodes monocular movements of either abduction or adduction.37 Thisbehavior is contrary to Hering’s law. Whether the present results indicate partial recov-ery, monocular adaptation, or both, the recovery of saccadic speed in chronic periph-eral palsy allows both eyes to reach targets in the paretic hemifield of motion rapidlyand simultaneously.

Clinical Implications

Idiopathic abducens nerve palsy in patients over 50 years of age is usually pre-sumed to be caused by microvascular ischemia, occurring with greater frequency inpatients with diabetes mellitus or hypertension.38,39 Because most such patients recov-er within three months,40 they may not require imaging of the brain or skull base at thetime of initial presentation, provided that the sixth nerve palsy is an isolated neurolog-ical finding.39 However, if there is no improvement within three months, imaging stud-ies are indicated. Normal MR imaging in our peripheral cases indicated an idiopathic(presumed ischemic) origin of palsy. Assessment of abducting saccadic velocity maybe an adjunct to imaging in distinguishing between a central and peripheral cause ofabducens nerve palsy. Study of palsies having comparable severity after partial struc-tural damage to the peripheral nerve will be needed to determine whether their saccad-ic dynamics differ from those in idiopathic (presumed ischemic) palsies or fromcentral palsies, as investigated here. Normal saccadic velocity over a month after onsetof palsy may indicate a peripheral demyelinating or ischemic cause. Conversely, re-duced saccadic velocity after a month may signify a central lesion.

ACKNOWLEDGMENTS

This work was supported by Canadian Institutes of Health Research (CIHR)grants MT 5404, ME 5509 (J.A.S.), MSH 55058 (A.W.), and MOP 57853 (A.W. andJ.A.S.), and the University Health Network Ophthalmology (A.W.) and Neurology(J.A.S.) Practice Plans

REFERENCES

1. METZ, H.S., A.B. SCOTT, D. O’MEARA & H.L. STEWART. 1970. Ocular saccades in lat-eral rectus palsy. Arch. Ophthalmol. 84: 453–460.


2. KOMMERELL, G., D. OLIVIER & H. THEOPOLD. 1976. Adaptive programming of phasicand tonic components in saccadic eye movements. Investigations in patients withabducens palsy. Invest. Ophthalmol. 15: 657–660.

3. ZEE, D.S. & R.D. YEE. 1977. Abnormal saccades in paralytic strabismus. Am. J. Oph-thalmol. 83: 112–114.

4. OPTICAN, L.M., D. ZEE & F. CHU. 1985. Adaptive response to ocular muscle weaknessin human pursuit and saccadic eye movements. J. Neurophysiol. 54: 110–122.

5. BRONSTEIN, A.M., J. MORRIS, G. DU BOULAY, et al. 1990. Abnormalities of horizontalgaze. Clinical, oculographic and magnetic resonance imaging findings. I. Abducenspalsy. J. Neurol. Neurosurg. Psychiatry 53: 194–199.

6. BARTON, J.J., A.G. HUAMAN & J.A. SHARPE. 1994. Effects of edrophonium on saccadicvelocity in normal subjects and myasthenic and nonmyasthenic ocular palsies. Ann.Neurol. 36: 585–594.

7. ABEL, L.A., D. SCHMIDT, L.F. DELL’OSSO & R.B. DAROFF. 1978. Saccadic system plas-ticity in humans. Ann. Neurol. 4: 313–318.

8. COLLINS, C.C. The human oculomotor control system. 1975. In Basic Mechanisms ofOcular Motility and Their Clinical Implications. G. Lennerstrand & P. Bach-y-Rita,Eds.: 145–180. Pergamon Press. New York.

9. JAMPOLSKY, A. 1971. A simplified approach to strabismus diagnosis. In Symposium onStrabismus. Trans. New Orleans Acad. Ophthalmol.: 34–92. Mosby. St. Louis, MO.

10. SCOTT, A.B. 1971. Active force tests in lateral rectus paralysis. Arch. Ophthalmol. 85:397–404.

11. SCOTT, A.B., C.C. COLLINS & D.M. O’MEARA. 1972. A forceps to measure strabismusforces. Arch. Ophthalmol. 88: 330–333.

12. SCOTT, A.B. & S.P. KRAFT. 1985. Botulinum toxin injection in the management of lat-eral rectus paresis. Ophthalmology 92: 676–683.

13. SCOTT, A.B. 1994. Change of eye muscle sarcomeres according to eye position. J.Pediatr. Ophthalmol. Strabismus 31: 85–88.

14. FRAHER, J.P., P.F. SMIDDY & V.R. O’SULLIVAN. 1988. The central-peripheral transitionregions of cranial nerves. Trochlear and abducent nerves. J. Anat. 161: 115–123.

15. SEDDON, H.J. 1943. Three types of nerve injury. Brain 66: 237.16. SUNDERLAND, S. 1977. Some anatomical and pathophysiological data relevant to the

facial nerve injury and repair. In Facial Nerve Surgery. U. Fisch, Ed.: 47–61. Kugler.Amstelveen.

17. SCHAUMBERG, H.H., A.R. BERGER & P.K. THOMAS. 1992. Disorders of peripheralnerves. Davis. Philadelphia.

18. DREYFUS, P.M., S. HAKIM & R.D. ADAMS. 1957. Diabetic ophthalmoplegia: report ofcase, with postmortem study and comments on vascular supply of human oculomotornerve. Arch. Neurol. Psychiatry 77: 337–349.

19. WEBER, R.B., R.B. DAROFF & E.A. MACKEY. 1970. Pathology of oculomotor nervepalsy in diabetics. Neurology 20: 835–838.

20. ASBURY, A.K., H. ALDREDGE, R. HERSHBERG & C.M. FISHER. 1970. Oculomotor palsyin diabetes mellitus: a clinico-pathological study. Brain 93: 555–566.

21. DELGADO-GARCIA, J.M., F. DEL POZO, R.F. SPENCER & R. BAKER. 1988. Behavior ofneurons in the abducens nucleus of the alert cat—III. Axotomized motoneurons.Neuroscience 24: 143–160.

22. SCOTT, A.B. & C.C. COLLINS. 1973. Division of labor in human extraocular muscles.Arch. Ophthalmol. 90: 319–322.

23. ROBINSON, D.A. 1978. The functional behavior of the peripheral ocular motor appara-tus: a review. In Disorders of Ocular Motility. G. Kommerell, Ed.: 43–61. BergmanVerlag. Munich.

24. BUTTNER-ENNEVER, J.A., A.K. HORN, H. SCHERBERGER & P. D’ASCANIO. 2001. Moto-neurons of twitch and nontwitch extraocular muscle fibers in the abducens, trochlear,and oculomotor nuclei of monkeys. J. Comp. Neurol. 438: 318–335.

25. BUTTNER-ENNEVER, J.A., A. EBERHORN & A.K. HORN. 2003. Motor and sensory inner-vation of extraocular eye muscles. Ann. N.Y. Acad. Sci. 1004: 40–49.

26. FENRICH, K. & T. GORDON. 2004. Axonal regeneration in the peripheral and centralnervous system—current issues and advances. Can. J. Neurol. Sci. 31: 142–156.


27. FERGUSON, B., M.K. MATYSZAK, M.M. ESIRI & V.H. PERRY. 1997. Axonal damage inacute multiple sclerosis lesions. Brain 120: 393–399.

28. MEWS, I., M. BERGMANN, S. BUNKOWSKI, et al. 1998. Oligodendrocyte and axonpathology in clinically silent multiple sclerosis lesions. Mult. Scler. 4: 55–62.

29. TRAPP, B.D., J. PETERSON, R.M. RANSOHOFF, et al. 1998. Axonal transection in thelesions of multiple sclerosis. N. Engl. J. Med. 338: 278–285.

30. LOVAS, G., N. SZILÁGYI, S. KOMOLY, et al. 2000. Axonal changes in chronic demyeli-nated cervical spinal cord plaques. Brain 123: 308–317.

31. KORNEK, B., M.K. STORCH, J. BAUER, et al. 2001. Distribution of a calcium channelsubunit in dystrophic axons in multiple sclerosis and experimental autoimmuneencephalomyelitis. Brain 124: 1114–1124.

32. HERING, E. 1868. Die Lehre vom Binokularen Sehen. Wilhelm Englemann. Leipzig.33. SHARPE, J.A., D. TWEED & A.M.F. WONG. 2003. Adaptations and deficits in the vestib-

ulo-ocular reflex after peripheral ocular motor palsies. Ann. N.Y. Acad. Sci. 1004:111–121.

34. WONG, A.M.F., D. TWEED & J.A. SHARPE. 2002. Adaptations and deficits in the vestib-ulo-ocular reflex after sixth nerve palsy. Invest. Ophthalmol. Vision Sci. 43: 99–111.

35. WONG, A.M., D. TWEED & J.A. SHARPE. 2002. The vestibulo-ocular reflex in fourthnerve palsy: deficits and adaptation. Vision Res. 42: 2205–2218.

36. WONG, A.M.F. & J.A. SHARPE. 2002. Adaptations and deficits in the vestibulo-ocularreflex after third nerve palsy. Arch. Ophthalmol. 120: 360–368.

37. ZHOU, W. & W.M. KING. 1998. Premotor commands encode monocular eye move-ments. Nature 393: 692–695.

38. RICHARDS, B.W., F.R. JONES & B.R. YOUNGE. 1992. Causes and prognosis in 4,278cases of paralysis of the oculomotor, trochlear, and abducens cranial nerves. Am. J.Ophthalmol. 113: 489–496.

39. CHOU, K.L., S.L. GALETTA, G.T. LIU, et al. 2004. Acute ocular motor mononeuropa-thies: prospective study of the roles of neuroimaging and clinical assessment. J. Neu-rol. Sci. 219: 35–39.

40. KING, A.J., E. STACEY, G. STEPHENSON & R.B. TRIMBLE. 1995. Spontaneous recoveryrates for unilateral sixth nerve palsies. Eye 9: 476–478.

recovery of peripheral versus central nerves identified by ......abducens nerve palsy lowers saccade...

Documents