JOURNALOF NEUROPHYSIOLOGY Vol. 62, No. 1, July 1989. Printed
in U.S.A.
Lidocaine-Induced Unilateral Internuclear Ophthalmoplegia: Effects on Convergence and Conjugate Eye Movements PAUL D. R. GAMLIN, JAMES W. GNADT, AND LAWRENCE E. MAYS Department of Physiological Optics, School of Optometry, and The Neurobiology Research Center, University of Alabama at Birmingham, Birmingham, AL 35294 SUMMARY
AND
CONCLUSIONS
INTRODUCTION
As described in the accompanying paper, most abducens 1. To characterize the vergence signalcarried by the medial longitudinal fasciculus (MLF), it was subjected to reversible internuclear neurons decrease their firing rate during symblockadeby smallinjections of 10%lidocainehydrochloride. The metrical convergence (Gamlin et al. 1989). This should effects of theseblockadeson both conjugate and vergence eye result in an inappropriate vergence signal (i.e., decreased movementswere studied. activity) being sent to medial rectus motoneurons. We 2. With this procedure, experimentally induced internuclear have proposed that the inappropriate medial longitudinal ophthalmoplegia(INO) and its effects on conjugate eye move- fasciculus (MLF) vergence signal is compensated for by an mentscould be studied acutely, without possiblecontamination from long-term oculomotor adaptation. In the eye contralateral appropriate signal of higher gain that arises from cells close to the MLF blockade,saccadicand horizontal smooth-pursuiteye to the oculomotor nucleus (Gamlin et al. 1989). Thus the movementswerenormal. Horizontal abductingnystagmus,often results of the accompanying paper predict that a unilateral lesion of the MLF will affect horizontal eye movements in seenin patientswith INO, wasnot observedin this eye. 3. As previously reported for INO, profound oculomotor defi- two ways. First, there will be a decreased conjugate drive to cits were seenin the eye ipsilateralto the MLF blockade.During ipsilateral medial rectus motoneurons. This effect is a maximal blockade, adducting saccadesand horizontal smooth- characteristic of the clinical syndrome of internuclear pursuit movementsin this eyedid not crossthe midline. Adduct- ophthalmoplegia seen in patients with MLF lesions (Cogan ing saccadeswere reduced in amplitude and peak velocity and 1970). Second, as a result of reducing the inappropriate showed significantly increaseddurations. Abducting saccades, MLF vergence signal, there should be an increased verwhich were slightly hypometric, displayeda marked postsaccadic gence signal reaching ipsilateral medial rectus motoneucentripetal drift. 4. The eyeipsilateralto the blockadedisplayeda pronounced, rons. The present study was specifically designed to test the upward, slow drift, whereasthe eye contralateral to the blockade showedvirtually no drift. Furthermore, althoughvertical saccades second prediction and thus assess the behavioral signifito visual targets remained essentiallyconjugate, the size of the cance of the vergence signal carried by the MLF. Reversible resettingquick phasesin eacheye wasrelatedto the amplitude of unilateral blockades of the MLF were produced by small the slowphasemovementin that eye.Thus the eyeon the affected injections of lidocaine hydrochloride, and vergence gain side displayedlarge quick phases,whereasthe eye on the unaf- was assessed employing the monocular accommodative fected sideshowedonly slight movements.On occasion,unilat- vergence paradigm (Miiller 1843). This approach, while eral downbeatingnystagmuswasseen.This strongly suggests that necessary for open-loop measurements of vergence gain, the vertical saccadegeneratorsfor the two eyescan act indepenrequires some explanation for the results to be interpreted dently. correctly. 5. The effect of MLF blockadeon the vergencegain of the eye Accommodative convergence, determined during monon the affected sidewasinvestigated.As a measureof open-loop ocular viewing, is the adduction of the covered eye obvergencegain, the relationshipof accommodativeconvergenceto accommodation(AC/A) wasmeasuredbefore, during, and after served when the viewing eye accommodates to a near tarreversible lidocaine block of the MLF. After taking conjugate get. This is a consequence of the coupling of the vergence deficits into account, the net vergencesignalto the eyeipsilateral and accommodation systems (Schor 1985). During acto the injection was found to increasesignificantly during the commodative convergence, the adduction of the covered reversibleblockade. eye consists of two components: a symmetrical vergence 6. The most parsimoniousexplanation for this increasedver- component and a conjugate component that maintains the gencesignalis suggested by the accompanyingsingle-unitstudy. viewing eye on target (Keller 1973; Kenyon et al. 1978). This study showedthat abducensinternuclear neurons, whose axons coursein the MLF, provide medial rectus motoneurons This is shown in Fig. IA, which schematically represents a vergence eye movement with the with an appropriate horizontal conjugateeyeposition signalbut normal accommodative an inappropriate vergencesignal. Ordinarily, this incorrect ver- right eye viewing. As the eyes accommodate to the near gencesignalisovercomeby another, more potent, vergencesignal target, convergence is elicited in both eyes. However, the impinging on the medial rectus motoneurons.This strongersig- right eye undergoes only a transient movement and renal is revealedwhen the inappropriate MLF vergencesignal is mains aligned on the target. This is due to conjugate eye reducedby lidocaineblockade. movements to the right, either saccadic or smooth pursuit, that counteract the tendency for the right eye to move to
82
0022-3077/89 $1 SO Copyright 0 1989 The American Physiological Society
INTERNUCLEAR RIGHT
EYE
OPHTHALMOPLEGIA
AND
VIEWING
LEFT c
LEFT EYE 5O e
RIGHT EYE 0
2.
5O --=i
IO0
“:GV”ET
LEEV’E’
COMPONENT
CONJUGATE COMPONENT
2.5O +
OBSEEy.VED
7.5O
O0
MOVEMENT
-
4
0
2
EYE
VIEWING LIDOCAINE
D BLOCKADE
NORMAL
LEEyFET 5O 4
VERGENCE
83
VERGENCE
RIGHT EYE
5O
5O &-
VERGENCE COMPONENT
5O --i
’
CONJUGATE COMPONENT
,5
0
2-
2 IO0 -
RIGHT EYE
LEFT EYE
0
OBSE&VED
0’
MOVEMENT
4
&
5O
!o”
I’
FIG. 1. Predicted effects on accommodative vergence of lidocaine blockade of the left MLF if this pathway carries no net vergence signal. For ease of explanation, this figure presents a situation in which the normal accommodative eye movement results in an observed convergence of 10”. A: a normal accommodative vergence eye movement with the right eye viewing. This movement consists of a vergence and conjugate component (Keller 1973; Kenyon et al. 1978). First, the eyes begin to accommodate. This elicits symmetrical vergence in both eyes. At a variable latency, the eyes make conjugate movements to the right, either saccadic or smooth pursuit, to maintain the right eye on target. At the completion of the movement, the right eye is on target, and the left eye has moved to the right by an amount that is equal to the total of the vergence movement and the conjugate movement in that eye. In this case, it corresponds to a 10” movement. B: a hypothetical situation in which MLF blockade has reduced the conjugate gain of the left eye with respect to the right eye by a factor of 0.5 (i.e., G = 0.5). As in A, accommodation of the right eye causes both eyes to converge symmetrically, and overall, if the right eye is to stay on target, it must make a conjugate movement of 5” to the right. However, because of the MLF blockade, when the right eye moves 5” to the right, the left eye will move only 2.5” to the right. Overall, at the completion of the movement the left eye will have moved only 7.5” to the right, a 5” vergence component and a 2.5” conjugate component. This would result in a change in the AC/A to 0.75 of its initial value. C and D: comparable to A and B except that now the situation is considered with the left eye viewing. In D, accommodation causes both eyes to converge symmetrically. If the left eye is to stay on target, it must make a conjugate movement of 5” to the left. However, because of the MLF blockade when the left eye moves 5” to the left, the right eye will move 10” to the left. Overall, at the completion of the movement the right eye will have moved 15 O to the left, a 5” vergence component and a 10” conjugate component. This would result in the AC/A increasing by a factor of 1.5.
the left as a result of convergence (Keller 1973; Kenyon et al. 1978). The net result is that the new position of the eyes is a combination of conjugate and vergence movements. These cancel in the right eye, but in the left eye the rightward vergence and conjugate movements are summed. Over a certain range, accommodative convergence is linearly related to accommodation, and it is customary to plot accommodative convergence against accommodation to yield an AC/A value [the ratio of accommodative convergence (AC) to accommodation (A)]. When accommodative response is used, a response AC/A is produced. Alternatively, when accommodative demand is used, a stimulus AC/A is produced. Figure 1B summarizes the result of lidocaine blockade that would be expected if the MLF carries no vergence signal; this represents the null hypothesis being tested. In this example, there is a blockade of the left MLF, and the right eye is viewing a target moving from far to near. As in Fig. lA, the right eye shows no net movement, and the rightward movement of the left eye is a combination of both conjugate and vergence movements. However, because of the MLF blockade, the rightward-acting conjugate drive to the left eye will be reduced by some measurable amount (in this example, it is reduced 50%). If the vergence input to the left eye is unaffected by the blockade, then the left eye will move only 75% as far as a result of the MLF blockade. Figure 1, C and D, considers the outcome of viewing with the other eye. As before, it is assumed that the MLF blockade has reduced the conjugate drive to the left eye by
50%. The null hypothesis again states that the vergence input to each eye is unaffected by the blockade. Accommodative vergence causes both eyes to converge, and the conjugate eye movement system attempts to counteract the movement of the left eye and keep it on target. However, because of the reduced conjugate gain of the left eye, the conjugate signal to it must be increased to twice the usual 1evel. Because the horizontal conjugate signals are coupled, this will cause the right eye to move twice as far to the left. Th us, as a result of the blockade, the right eye will move 150% of normal to the left. In the general case, with G being the ratio of the conjugate gain of the affected eye to that of the normal eye, the change in AC/A predicted by the null hypothesis is l/2( G + 1) with the normal eye viewing. With the affected-eye viewing, the predicted change in AC/A is 1/2((G + 1)/G). Now consider the outcome if the MLF does carry an inappropriate vergence signal. Lidocaine blockade of the left MLF would significantly reduce this inappropriate signal and thus result in an increased vergence signal to the left eye. With either eye viewing, the resulting movements of the nonviewing eye would be larger than predicted by the null hypothesis. For the experiments, accommodative vergence and conjugate eye movements were measured before, during, and after lidocaine blockade with each eye viewing separately. During unilateral blockade, the observed eye movements were larger than those predicted by the null hypothesis and were consistent with the hypothesis that the MLF carries an
84
P. D. R. GAMLIN,
J. W. GNADT,
A I I
M7
I/
I
I L-
--I 3.
I I
fiBSORBING FILTER
I
I Lll
I
> RECORDING INFRARED OPTOMETER
I I I I
AND
VR
L. E. MAYS
*
L-&L vL-7
I
2 deg
I I I
I -
BEAM
I CHOPPER
I I .
T 1
PEN PL(TTTER
I
0 w
I
TARGET
1
I
1 TIME in
2
seconds
MM1 1/ M4 \
FIG. 2. Schematic of the apparatus used in this study. BPD, bipartite photodetector; HS, horizontal slit; L 1- 13, 10-D lenses; L14, 20-D lens; M l-4, beam splitters; M&7, front surface mirrors; M6, pellicle; MM l-3, front surface mirror galvanometers.
inappropriate vergence signal to medial rectus motoneuTons. A preliminary report of some of these results has appeared elsewhere (Gamlin et al. 1988). METHODS
Some of the methods used were described in detail in the accompanying paper (Gamlin et al. 1989). These methods are described only briefly. Other methods, not previously used by us, are described more fully.
Animal preparation Two juvenile rhesus monkeys (Macaca mulatta) were used in this study. Their preparation was the same as in the accompanying paper (Gamlin et al. 1989).
Behavioral
training
Animals were trained to look at targets in an apparatus that was a combined visual stimulator and recording infrared optometer.
0
1 TIME
in
2
seconds
FIG. 3. A: a representative downward vertical eye movement during MLF blockade. It also shows the pronounced downbeating nystagmus in the left eye. Despite this nystagmus, when the animal looks with the right eye to a target 4” below, the saccade in the left eye is comparatively normal in amplitude. Although quick phases in the left eye are generally seen as smaller amplitude movements in the right eye, the arrows indicate 2 occasions when this was not the case, and unilateral vertical eye movements can be seen. A movement in which the left eye went down while the right eye went up is also shown (*). B: normal upward saccades of 4” in both eyes despite the pronounced slow drift and nystagmus. VR, vertical right eye position; VL, vertical left eye position. The unlabeled, dashed line in this and subsequent eye movement traces indicates target position (from monkey A, session 1).
INTERNUCLEAR
OPHTHALMOPLEGIA
AND
85
VERGENCE
A
4 deg
I -------------------------------------. II I II
EYE
I I
HL
!
ADDUCTS “*7
TIME in seconds
TIME in seconds
4 deg
I LEFT EYE ABDUCTS
1
0 TIME in seconds
0
1 TIME in seconds
FIG. 4. Effects of MLF blockade on saccadic eye movements. A and B: saccades in which the left eye views. A: a series of undershooting saccades that occur when the left eye attempts to view a target 8” to the right. B: the animal is required to look at a target 12” to the left. The initial saccade slightly undershoots the target and shows clear postsaccadic centripetal drift. Additional saccades are required to compensate for postsaccadic drift. C and D: saccades in which the right eye views. C: the right eye looks to a target 10” to the right of primary position. The left eye makes a saccade -4O in amplitude. Because the starting position of the left eye was 6” to the left of primary position, it did not cross the midline. D: the right eye looks to a target 12” to the left. Although the saccade in the left eye is only slightly hypometric, it displays a pronounced centripetal postsaccadic drift. In this and subsequent figures, HR, horizontal right eye position; HL, horizontal left eye position; VA, vergence angle (from monkey B, sessions 1 and 2).
It allowed visual stimuli to be presentedto either, or both, eyes while accommodation was continuously monitored. Vergence and accommodativedemandscould be changedindependently and saccadicand smooth-pursuiteye movementscould alsobe elicited. Figure 2 is a schematicof the apparatus.The infrared recording optometer and visual display can be consideredseparately. The recording optometer continuously measuredaccommodation in the right eye. This optometer is basedon a design publishedby Kruger (1982)and can be thought of asa motorized streakretinoscope.A horizontal slit of infrared light is sweptvertically acrossthe retina by the beam chopper. Lamp 2 and an infrared transmitting filter form the infrared source. Light reflected from the retina goesthrough L2, M2, L6, MM3, L7, a horizontal slit, and L8, to a bipartite photodetector (BPD). The horizontal slit and the BPD function like the observer’seye in streakretinoscopy.In streakretinoscopy, the observer’stask is to determineif the retinal reflection is moving “with” or “against” the stimulusmotion. The BPD is attachedto a timing circuit that determines whether the retinal reflection is moving with or againstthe stimulusand also measuresthe speedof this move-
ment. This speedgivesa measureof the accommodativestateof the eye(Kruger 1982).Dependingon pupil diameter,the output of the device wasunaffected by horizontal eye rotations of up to 8”. The optometerhasa sensitivity < 0.1 D. An infrared-sensitive TV camera(Hitachi HV-120) and its associatedoptics for aligning the right eyeare not shown. The other component of the apparatuswasthe visual display. Briefly, it was composedof Lamp 1 and an infrared-absorbing filter that illuminated a target on a flatbed plotter through lens L 13.The light beamwassplit by pellicle M6, and the target image wasrelayedto the left eye(LE) by way of MMl, M5, L3, M3, Ml and Ll, and to the right eye (RE) by way of MM2, L4, M4, M2 and L2. The target consistedof a back-lit white cross,subtending 0.5O,on a black background. This optical arrangementformed a Badal optometer, suchthat 1 cm of target motion produced a 1 diopter (D) changein its accommodative demand. Horizontal target motion was produced by movable mirrors for the left (MM 1) and right eyes(MM2) that were placedat a point conjugate with the centersof rotation of the eyes.Thesemirrors were rotated by precisehigh-speedgalvanometers(General Scanning
86
P. D. R. GAMLIN,
J. W. GNADT,
A
AND L. E. MAYS
C 0 n
NORMAL LIDOCAINE
250
0 0
5 AMPLITUDE
10 (degrees)
15
B
D
5 AMPLITUDE
10 (degrees)
5 AMPLITUDE
10 (degrees)
50 40 30 20 10
0 ! 0
. 5 AMPLITUDE
. 10 (degrees)
1
15
0 0
15
FIG. 5. A and B: peak velocity (deg/s) - amplitude and duration (ms) - amplitude relationships, respectively, of the left eye for abducting saccades. C and D: peak velocity (deg/s) - amplitude and duration (ms) - amplitude, respectively, relationships of the left eye for adducting saccades. Results for normal horizontal saccades (0) and for horizontal saccades during MLF blockade (m) are shown (from monkey B, sessions I and 2).
300PD). Vertical motion wasprovided by the vertical movement of the plotter. The lenseswere all 75 mm in diameter, allowing nearly 20” of horizontal and vertical target motion. The galvanometerscould be rotated to act as independentshuttersfor the two eyes. To elicit accommodativevergenceeyemovements,the animal wasrequired to view a target monocularly at a certain accommodative demand. After a variable delay, the target would make a positive accommodativestep ranging from 1 to 7 D. The animal wasrewardedfor viewing the new target appropriately.
Recording and injection procedures With the useof a Kopf microdrive, an assemblyconsistingof a parylene-insulatedtungsten microelectrodewithin teflon tubing insidea 26-gacannulawasadvancedthrough a 2 1-gahypodermic needleusedto puncture the dura. Unit activity wassharply filtered above 5 kHz, and the occurrence of a spikewas detected with a window discriminator and recordedto the nearest0.1 ms. Initially, the oculomotor, trochlear, and abducensnuclei were located basedon the characteristicsof their cellular activity and on their responseto microstimulation. The location of the MLF was then identified basedon its relationship to the oculomotor nuclei,on the presenceof horizontal burst-tonic fibers,and on the
presenceof fibers that were modulated for vertical eye movementsbut pausedfor saccades. To minimize lidocaine spreadto important oculomotor regions[e.g., paramedianpontine reticular formation (PPRF), the motor nuclei], the electrodeassemblywas directed to the dorsal MLF midway betweenthe abducensand trochlear nuclei. On a given day, the dorsoventral extent of the MLF wasmappedout by measuringthe threshold cathodal current (20-mstrain at 500 Hz, 0.2-mspulse)required to produce a detectableadductive twitch in the ipsilateral eye. This proved a sensitive method, asthe thresholdcurrent required to produce an adductive twitch wasdoubled (from 5 to 10 PA) when the electrode was moved back 200 pm from the presumedlocation of the MLF. Once the location of the MLF wasdetermined,the microelectrode and teflon tubing were removed from the 26.ga cannula, which remainedin situ. The microelectrodewasreplacedwith a premeasuredlength of 32-gahypodermic tubing. The 32-gacannula wasattached to a 0.54 Hamilton syringeby polyethylene tubing (2 mm OD, 0.18 mm ID) and wasfilled with 10%lidocaine hydrochloride. In all instances,it was found that effective MLF blockade could be induced by injections of 100 nl of 10%lidoCainehydrochloride. It was estimatedthat this amount of lidoCaineblockeda region - 1mm in diameter(seeRESULTS). Saline injections of equal volume producedno significant effects.
INTERNUCLEAR
OPHTHALMOPLEGIA
AND
VERGENCE
HR HL
TIME in seconds
B
TIME in seconds
20 15 -
k
;
10.
F
5.
f
0.
8 f
-5.
T !i
-10. -15
10
:
T
5
f
0
LID0
-
-2o!
,
-20
-15
,
-10
,
-5
RIGHT
,
,
.
0
5
10
1
15
1
20
-20
I
,20
I
-15
I
-10
1
-5
I
0
I
1
5
10
I
15
1
20
LEFT EYE POSITION
EYE POSITION
FIG. 6. A: horizontal smooth pursuit with the right eye viewing. B: position of the left eye is plotted against the position of the right eye for this movement and, for reference, a normal movement. C horizontal smooth pursuit with the left eye viewing. D: position of the right eye is plotted against the position of the left eye for this movement and for a normal movement (from monkey B, session 1).
The MLF was approachedfrom the left chamber and the left MLF was blocked with lidocaine. Becausethe optometer measuredaccommodationof the right eye, the ratio of accommodative convergenceto accommodation (responseAC/A) could be obtainedwith the right eyeviewing. For someinitial experiments, becauselidocaine blockadereducedthe oculomotor rangeof the left eyeand produceddownbeatnystagmus,only the right eyewas allowedto view the target. However, as the anesthesiawore off, the downbeat nystagmusbecamelesspronounced, and the animal would reliably view with the left eye. In later experiments, therefore, accommodativeconvergenceand conjugateeyemovementswere measuredwith the left eyeviewing during this period of submaximalblockade. Becausethe accommodativeresponse of the left eyecould not be measured,the ratio of accommodative convergenceto accommodative demand (stimulus AC/A) was usedasa measureof accommodativevergencegain. To facilitate comparisons,the accommodative stimuli and the vergenceresponseare expressedin equivalent units. Thus accommodationis expressedin diopters (D) and vergence is expressedin meter angles(MA). One MA is the angle betweenthe two eyes when viewing a target at 1 m (1-D accommodative demand). These values were equivalent to 1.9’ and 2.0° for animals A and B, respectively.
Eye movement
recording
The horizontal and vertical gainsof each eye were calibrated independently at the beginning of each recording session.Animals showedlittle variability in fixation from trial to trial, and saccadesof 0.2’ could be detected reliably. The right pupil was dilated slightly with 3 drops of 2.5% phenylephrine HCl before experimental sessions. The optometer was calibrated by placing trial lensesin the detector portion of the device at a point conjugatewith the principal plane of the eye.The positionsof the right and left eyes,accommodation of the right eye, and target positions were sampledat 1,000Hz and storedon computer tape for analysis.
Data analysis The stored data were analyzed using a PDP- 1l/73 computer equippedwith an interactive graphicsdisplay. Eye position and accommodationwere displayed, and periods of steady fixation weremanually delineatedby a cursor. ResponseAC/A plots were producedby measuringsuccessivelOO-mssamplesof accommodation and vergenceduring periodsof steadyfixation for trials in
P. D. R. GAMLIN,
88
which
the animal
J. W. GNADT,
viewed a target with various accommodative
demands. Stimulus AC/A plots were produced from a number of
separate trials by measuring the vergence angle before a movement and subtracting this value from successive lOO-ms samples of vergence angle during periods of steady fixation at the completion of the movement. These values were then plotted against the accommodative demand. Eye velocity was calculated from eye position by a two-point derivative (Usui and Amidror 1982), and the beginning and end of saccadic eye movements were automatically marked by computer based on a 4O”/s velocity criterion. The peak velocity of these movements were recorded, and their duration and amplitudes were measured.
Histology
AND L. E. MAYS
(Bender and Weinstein 1944; Evinger et al. 1977). Also, in one study (Burde et al. 1977), vestibular commissural fibers were probably damaged. In addition, long-term oculomotor adaptation was not controlled in previous studies. This is important, because a number of clinical observations, such as abducting nystagmus and saccadic trajectory changes, may be explained by adaptive mechanisms (Baloh et al. 1978a; Zee et al. 1987). The effects of unilateral blockade on conjugate eye movements are considered before the effects on accommodative convergence. Conjugate eye movement e&cts
NYSTAGMUS. Within minutes of lidocaine injection, the left eye developed a slow-phase, upward drift that was essentially confined to this eye (Fig. 3). This drift was generally present at primary position and increased on downward gaze. During maximal blockade, the drift in the affected eye could not be visually suppressed. The size of the resetting, vertical quick phases in each eye was related to the amplitude of the slow phase movement in that eye. Thus large quick phases were prominent in the left eye, while synchronized quick phases of much smaller amplitude were generally evident in the right eye. However, on a number of occasions, the left eye displayed a quick phase movement, whereas the right eye remained virtually stationary. Two examples of this are indicated by the arrows in Fig. 3A. In addition, a movement (*) can be seen in Fig. RESULTS 3A in which the left eye shows a downward quick phase The results described below were essentially the same while the right eye shows an upward quick phase. Importantly, vertical saccadic eye movements to visual targets, between the two animals and for the two experimental which were examined qualitatively for movements oft 10’ sessions run on each animal. Therefore, the data presented or less in amplitude, were essentially conjugate. Examples are representative of all sessions. Although the conjugate data were necessary for the interpretation of the vergence of saccades to targets at -4 and +4” are shown in Fig. 3, A effects, it became clear that they also provided the first and B, respectively. In no case was horizontal nystagmus observed in either eye. description of unilateral internuclear ophthalmoplegia (INO) in which the observed eye movements were unaf- HORIZONTALSACCADICEYEMOVEMENTS. Duringmonfected by long-term oculomotor adaptation. Previous ex- ocular viewing with the left eye, adducting saccades in this perimental studies were potentially confounded for a numeye were hypometric, and a series of additi onal saccades ber of reasons. Generally, bilateral MLF lesions were made were needed to acquire the target (Fig. 4A). During these DOWNBEAT
In the first animal (A), at the end of the experiment, a small injection of wheat germ agglutinin-horseradish peroxidase (Sigma) was placed at the same X- Y coordinates and depth, as confirmed by stimulation, as the reversible blockade had been placed on the previous day. Subsequently, standard HRP histochemistry revealed the MICA-HRP marking spot clearly. In Nissl-stained sections, slight gliosis associated with the lidocaine injection sites was also seen, which was sufficient to allow reconstruction of these sites. The second animal (B) was, therefore, not injected with WGA-HRP. In both cases, animals were deeply anesthetized with pentobarbital sodium and then perfused through the aorta with saline followed by a suitable fixative. The brain was sectioned at 40 pm, and a Nissl-stained series was prepared.
TABLE
1.
Monkey
Summary of data obtainedfrom the experimentalsessions Session
During
Before
Predicted
After
G
Response AC/A with right eye viewing A A B B
I 2
1
2
1.66 1.68 1.26 1.18
t I!I I!I t
0.03 0.04 0.03 0.04
(43) (24) (17) (19)
1.61 1.73 1.20 1.21
AI k t *
0.04* 0.04* 0.05* 0.03*
(31) (26) (12) (19)
1.29 1.28 0.89 0.89
1.82 k 0.06-t 1.28 t 0.06 1.19 t 0.04
(18) (10) (15)
0.56 0.52 0.41 0.51
1.32 1.14 1.45
1.01 t 0.04 0.89 z!z 0.06 1.01 to.12
(10) (15) (13)
0.66 0.77 0.66
Stimulus AC/A with left eye viewing A B B
2
1
2
1.06 AI 0.04 (8) 0.99 IL 0.06 (8) 1.16 + 0.05 (8)
1.59 + 0.07* 1.36 + 0.08* 1.89 I!I 0.12*
(16) (11) (12)
Values for the AC/A t 95% confidence limits are expressed in meter angles over diopters of accommodative response or demand (MA/D); number of separate trials in parentheses. The predicted AC/A assumes the null hypothesis that the MLF carries no significant vergence signal and is calculated from the AC/A value before lidocaine blockade and G (see INTRODUCTION). G, conjugate gain of the affected eye with respect to the normal eye during lidocaine blockade; AC/A, relationship of accommodative convergence to accommodation. *During AC/A is si gnificantly different from the predicted AC/A (Student’s t test for differences of slope; P < 0.001). tAC/A is significantly different from before and during (P < 0.01).
INTERNUCLEAR
OPHTHALMOPLEGIA
AND VERGENCE
89
HR
TIME in seconds
C
D n
6 I
NORMAL LIDOCAINE
20 1 15 -
ii
10 -
T"
5-
YE E Fi s T ::
-6
! 0
I 1
1 2
1 I 3 4 ACCOMMODATIVE
I
2
TIME in seconds
a 0
I
1
I 1 5 6 RESPONSE
1 7
1
0
0, -5-10
-
-20
! -20
I -15
1 -10
1 I 1 -5 0 5 RIGHT EYE POSITION
I
I
1
10
15
20
FIG. 7. With the right eye viewing, A shows an accommodative vergence eye movement before lidocaine blockade. ACC, accommodation. B: an accommodative vergence eye movement during MLF lidocaine blockade. Note the reduced conjugate gain of the left eye in B compared with A as evidenced by the reduction in saccadic gain during the movement. The amount of eye movement predicted by the null hypothesis that the MLF carries no vergence signal is indicated (dotted line; Null). C compares the normal response AC/A (slope = 1.18 MA/D) with the response AC/A during MLF blockade (slope = 1.2 1 MA/D). The observed AC/A is significantly higher than the AC/A (slope = 0.89 MA/D) predicted by the null hypothesis (Null). D: horizontal conjugate position of the left eye as the right eye viewed targets from - 12” to + 12”. The slope of this relationship when the right eye viewed from -6” to +6” of primary position was -0.5 (i.e., G = 0.5). Note that for conjugate eye movements, the left eye did not adduct past the midline (from monkey B, session 2).
movements, the right eye displayed larger amplitude saccades (Fig. 4A). Abducting saccades in the left eye were slightly hypometric compared with normal saccades and showed pronounced, centripetal postsaccadic drift, which was not seen in the right eye (Fig. 4B). Because of this postsaccadic drift, additional saccades were required to bring the eye onto target (Fig. 4B). During monocular viewing with the right eye, the saccades in this eye were normal in amplitude and duration (Fig. 4, C and D). However, the concurrent adducting saccades in the left eye were hypometric (Fig. 4C). The concurrent, abducting saccades in the left eye were somewhat hypometric and showed pronounced postsaccadic centripeta1 drift (Fig. 40). The relationships between peak velocity and amplitude, and duration and amplitude, were examined for both ab-
ducting and adducting saccades. The duration and velocity relationships were normal in the right eye for both adducting and abducting saccades. The relationship between amplitude and duration was also normal in the left eye for abducting saccades (Fig. 5B). However, for these same saccades, the relationship between peak velocity and amplitude was significantly different from normal (Student’s t test for differences of slope; P < O.OOl), with a somewhat increased peak velocity for a given amplitude (Fig. 5A). Furthermore, as expected, adducting saccades in the left eye were slower than normal (Fig. 5C) and of longer duration (Fig. 5D). HORIZONTALSMOOTH
suit eye movements by MLF blockade.
PURSUIT. Horizontalsmooth-purin the left eye were seriously impaired To illustrate these deficits, a single
90
P. D. R. GAMLIN,
J. W. GNADT,
smooth-pursuit trial with the right eye viewing is shown in Fig. 6A as the animal looks from 10’ left to 10’ right. As the left eye adducts, the amount that it diverges from the right eye increases. This can be seen in the vergence trace (VA). This movement, which is plotted in Fig. 64 contrasts with the normal situation in which there is a 1: 1 coupling between right eye position and left eye position. With the left eye viewing, the horizontal smooth-pursuit deficits are even more pronounced. Pursuit gain is reduced and “catch-up” saccades are required to maintain the eye on target (Fig. 6C). The increased conjugate signal to the left eye required to overcome the MLF blockade is reflected in the larger-amplitude movements in the right eye. This is clearly seen in Fig. 6D, in which left eye position is plotted against right eye position for normal smooth pursuit and smooth pursuit with the left eye viewing during MLF blockade.
AND
L. E. MAYS
Vergence eye movement e&cts Accommodative vergence eye movements, with the left or right eye viewing, were recorded before, during, and after reversible block with lidocaine. Table 1 provides a summary of the values obtained in these experimental sessions. EYE VIEWING. Figure 7A shows an accommodative convergence eye movement before lidocaine blockade in which the animal accommodated 5.6 D. The right eye shows slight movements, but there is no net change in its position. The entire accommodative convergence movement is made by the nonviewing left eye. For the period before the lidocaine block, a response AC/A plot was constructed for movements to targets of various accommodative demands (Fig. 7C, normal).
RIGHT
r
I
2MA
1 0
1
TIME in seconds c
2
TIME in seconds
D
12
20
1
15 -
v ii
R T;
10 -
F
5.
F E
0.
Fl $
-5.
8
G
i E A
6
6 k
4
T::
-10
-
-15 -2o! 0
1
2
3 ACCOMMODATIVE
4 DEMAND
5
6
1 -20
, -15
, -10
, -5 LEFT
, 0
, 5
, 10
EYE POSITION
FIG. 8. With the left eye viewing before lidocaine blockade, A shows an accommodative vergence eye movement in response to a step increase in accommodative demand of 5 D. B shows an accommodative vergence eye movement to the same accommodative demand during MLF blockade with lidocaine. Note that the reduced conjugate gain in the left eye is evident from its decreased ability to keep on target during the vergence movement. The amount of eye movement predicted by the null hypothesis that the MLF carries no vergence signal is indicated (dotted line; Null). C compares the normal stimulus AC/A (slope = 1.16 MA/D) with the stimulus AC/A (slope = 1.89 MA/D) during lidocaine blockade. It also shows that the AC/A measured during blockade is significantly higher than the AC/A (slope = 1.45 MA/D) predicted by the null hypothesis (Null). D: horizontal conjugate position of the right eye as the left eye viewed targets from -8” to +4”. The slope of this relationship is 1.5 (i.e., G = 0.66) (from monkey B, session 2).
, 15
) 20
INTERNUCLEAR
OPHTHALMOPLEGIA
AND
VERGENCE
91
sponded to 8 MA (Fig. 8B). This is greater than that predicted by the null hypothesis (Fig. 8B; Null). For this session, the slope of the relationship between accommodative demand and vergence response increased by a factor of 1.6, as is shown in Fig. 8C. Based on the null hypothesis, the stimulus AC/A should have increased by a factor of 1.25 (Fig. SC, Null). Thus the effect of the blockade was to significantly increase the stimulus AC/A beyond the prediction of the null hypothesis. After the lidocaine blockade had ended, the stimulus AC/A returned to its preblockade value (Table 1, monkey B, session 2). 165 TIME
(minutes)
9. Effect of injections of 100 nl 10% lidocaine HCl at different locations relative to the MLF. The phoria was determined while the animal was looking with the right eye at a target 12” to the right of primary position. See text for more details (from monkey B, session 2). FIG.
For comparison with the preblockade data, Fig. 7B shows an accommodative vergence eye movement during blockade of the MLF as the right eye accommodated 5 D. For this movement, accommodation and the resultant vergence eye movement are in the same ratio as before the reversible blockade. Thus the movement is larger than that predicted by the null hypothesis (Fig. 7B; Null). An AC/A plot constructed for the period of the blockade (Fig. 7C, lidocaine) shows that, apart from a lidocaine-induced exophoria that shifts the intercept down -3 MA, the AC/A values are similar for the two conditions. As is shown in Table 1, with the right eye viewing, there were no significant differences between the AC/A values before and during the blockade for any of the sessions. In contrast, as has been described above, the lidocaine blockade had a profound effect on horizontal conjugate eye movements over this period. With the right eye viewing either side of primary position, Fig. 70 shows the reduction in conjugate gain of the left eye with respect to the right eye. The slope of the relationship (G) is reduced to -0.5. That is, a conjugate eye movement in the right eye results in a movement of only one-half the amplitude in the left eye. Thus according to the null hypothesis, the AC/A with the right eye viewing during lidocaine blockade should have been reduced to 75% of its preblockade value (Fig. 7C; Null). LEFT EYE VIEWING. To more fully investigate the effects of MLF blockade on the accommodative vergence gain, the left eye viewed the target during submaximal MLF blockade (see METHODS). During this blockade, the conjugate gain with the left eye viewing was measured. It is presented in Fig. 80 and shows that any conjugate movement in the left eye resulted in a conjugate movement in the right eye of 1.5 times that amplitude (i.e., G = 0.66). The effects of the blockade on accommodative convergence can be demonstrated if a normal accommodative vergence movement is compared with that seen during lidocaine blockade (Fig. 8, A and B). Both movements were in response to a 5 D stimulus. In the normal condition, the increase in vergence angle corresponded to 5 MA (Fig. 84. After lidoCaine blockade, the increase in vergence angle corre-
Extent of lidocaine anesthesia To assessthe amount of the lidocaine blockade, the right eye viewed a target at 12” to the right of primary position.
A
FIG. 10. Chartings of representative sections showing the location of the cannulae in monkeys A and B. The arrow in A indicates the location of a small WGA-HRP injection, while that in B indicates an area of gliosis. Both arrows indicate the angle of approach. NIV, trochlear nerve; SC, superior colliculus; SCP, superior cerebellar peduncle; A, aqueduct.
92
P. D. R. GAMLIN,
J. W. GNADT,
Because of the yoking of conjugate movements, under normal conditions, the nonviewing left eye would also be directed - 12” to the right. The difference between the left and right eye is the phoria for this viewing distance. After lidocaine injection into the left MLF, the greater the effect of the blockade, the less the left eye will adduct. For example, while the right eye is viewing a target at 12’ to the right, the left eye may adduct only to primary position. This would result in a measured phoria of - 12’. Thus the greater the exophoria, the more effective the MLF blockade. Using this measure, the spread of lidocaine was assessed by the following procedure (Fig. 9). Based on stimulation mapping, an injection of 100 nl of lidocaine was placed -300 pm into the MLF. As shown by the large exophoria in Fig. 9, this injection (1) substantially blocked the MLF for a period of -20 min, with the effect wearing off after a total of 50 min. At this point, the cannula was raised 1 mm, allowed to stabilize, and another injection of 100 nl lidocaine was made. This injection (2) produced no detectable phoria change. After 15 min, the cannula was lowered 0.5 mm, allowed to stabilize, and another 100 nl of lidocaine injected. This injection (3) caused an increased exophoria. However, the effect was smaller and of a shorter duration than that seen with the cannula tip 500 pm below. Location oflidocaine
blockades
Based on histological sections, the locations of the cannulae were reconstructed for each animal. They were found to be located above or slightly penetrating the MLF between the trochlear and abducens nuclei (Fig. 10). In addition, Fig. 1OB shows one cannula site that was - 1 mm below the left trochlear nerve before its crossing. If the lidocaine anesthesia had spread to affect this nerve, it would have been apparent in the movements of the right eye. No such trochlear nerve palsies were seen. This suggests, as did the experiment above, that the effective spread of 100 nl of 10% lidocaine HCl was no greater than 1 mm. These anatomic reconstructions and limited lidocaine spread are also consistent with the absence of eye movement deficits that would have indicated lidocaine spread to the ipsilateral paramedian pontine reticular formation (PPRF) and abducens nuclei (see DISCUSSION). DISCUSSION
As described above, lidocaine blockade of the MLF profoundly affected both conjugate and vergence eye movements. Before considering these effects, the extent and location of the blockades will be considered. Extent and speczfzcity of lidocaine
blockades
The spread of injections of 100 nl of 10% lidocaine HCl was assessed by moving the tip of the injection cannula small distances. This procedure suggested that the extent of the spread was - 1 mm. A similar degree of spread was also suggested by the anatomic reconstructions of the cannulae locations and their proximity to the trochlear nerve. Clearlv. lidocaine iniections did not spread much bevond
AND
L. E. MAYS
the left MLF. They did not involve the PPRF, because this would have affected eye movements bilaterally. They did not involve the left abducens nucleus, because this would have affected the abducens internuclear pathway to the right eye in addition to the lateral rectus motoneurons for the left eye. Finally, they did not involve the left trochlear nucleus or nerve as this would have affected movements of the right eye. EfSect of MLF blockade on conjugate eye movements Unilateral, gaze-evoked vertical nystagmus was seen during MLF lidocaine blockade. This nystagmus was usually present at primary position, becoming more pronounced on downward gaze. Presumably, this nystagmus reflects inactivation of the fibers carrying vertical eye position signals that course in the MLF (King et al. 1976; Pola and Robinson 1978). After bilateral MLF lesions, bilateral gaze-evoked vertical nystagmus with centripetal drift has been reported (Evinger et al. 1977). Vertical nystagmus was also reported by Bender and Weinstein (1944), although its precise nature was not documented. Clinically, vertical gaze-evoked nystagmus has been reported as a symptom of IN0 (Cogan 1970; Kirkham and Katsarkas 1977). Furthermore, ipsilateral downbeat nystagmus combined with contralateral incyclorotary nystagmus has been reported for unilateral IN0 (Nozaki et al. 1983). However, we do not know if a contralateral incyclorotary nystagmus was also present in our experiments, because cyclorotation was not measured. The most interesting aspect of the observed vertical nystagmus was not the disconjugate nature of the slow phase, which can be explained by selective blockade of tonic vertical vestibular fibers, but the disconjugate nature of the quick phase movements. It is well known that MLF lesions have differential effects on horizontal and vertical eye movements [see Evinger et al. (1977) for a discussion of this]. Briefly, because abducens internuclear neurons carry both the eye position and velocity signals for horizontal saccades and quick phases, damage to their axons affects both aspects of these eye movements. In contrast, the vertically active fibers in the MLF only carry a tonic eye position signal for saccades and quick phases (King et al. 1976; Pola and Robinson 1978). The eye velocity signal during these eye movements is provided by vertical short-lead burst neurons in the rostra1 mesencephalic reticular formation (Biittner et al. 1977; King and Fuchs 1979). Thus damage to the MLF compromises the tonic, but not the phasic, component of vertical saccades and quick phases. Consequently, during MLF blockade vertical saccades appear normal: movements in the two eyes are yoked and of approximately equal amplitude. In contrast, quick phases, which are produced by the same saccade generators (Biittner et al. 1977; King and Fuchs 1979), are often not yoked or are of distinctly different amplitudes. Thus, during these quick phases, the saccade generator for one eye is acting independently of the saccade generator for the other eye. Although this is an unexpected finding, it may not be too surprising that, under the unusual circumstances produced bv unilateral lidocaine blockade. the vertical saccade
NYSTAGMUS.
INTERNUCLEAR
OPHTHALMOPLEGIA
generators can act independently. Although clinical studies have reported dissociated monocular downbeat nystagmus in IN0 (Aschoff et al. 1974; Bogousslavsky and Regli 1985; Jacome 1986; Nozaki et al. 1983), they did not discuss the relevance of these observations to the issue of independent quick-phase generators. Although vertical nystagmus is not always seen clinically, horizontal abducting nystagmus in the eye contralatera1 to a unilateral MLF lesion is often seen (Baloh et al. 1978a; Feldon et al. 1980; Zee et al. 1987). However, it has not been reported consistently in studies of experimental IN0 in the primate (Bender 1980; Carpenter and McMasters 1963; Carpenter and Strominger 1965; Evinger et al. 1977) This was also the case in the present study, which failed to find horizontal abducting nystagmus during monocular viewing, even with extensive MLF blockade. Clearly, horizontal abducting nystagmus is not due to the RIGHT EYE VIEWING LEFT EYE 7.5O
LEFT EYE VIEWING
R:Gv”ET
FirfiT
5O 6-
VERGENCE COMPONENT
7.3O -
5O -+
CON JUGATE COMPONENT
7.3O -
O0
oesEEyREVED MOVEMENT
O0 +
+
06
Rkt;T x
AND
VERGENCE
93
acute effects of blocking MLF signals. Although many explanations for the origins of horizontal abducting nystagmus have been presented (e.g., Baloh et al. 1978a; Pola and Robinson 1976; Stroud et al. 1973), this finding supports the previously offered explanation that it results from an adaptive strategy of the oculomotor system (Baloh et al. 1978a; Zee et al. 1987). SACCADES. Clinically, IN0 is often diagnosed from abnormalities in the amplitude or dynamics of adducting saccades. The slowing of adducting saccades that results from IN0 or experimental lesions has been reported widely (Baloh et al. 1978a; Bender and Weinstein 1944; Bird and Leech 1976; Carpenter and Strominger 1965; Crane et al. 1983; Evinger et al. 1977; Feldon et al. 1980; Kirkham and Katsarkas 1977; Metz 1976) and was clearly seen in the present study. Some studies have also reported abnormalities in abducting saccades (Bird and Leech 1976; Feldon et al. 1980). Consistent with these latter reports, the present study showed that abducting saccades were somewhat hypometric during lidocaine blockade. Also, these abducting saccades were characterized by a small increase in peak velocity for a given amplitude.
HORIZONTAL
SMOOTH PURSUIT. Evinger and colleagues (1977) studied qualitatively the effects of bilateral MLF lesions on horizontal smooth-pursuit eye movements. These authors could see no obvious deficits in either nasal or temporal pursuit. Clinically, pursuit eye movements with normal characteristics have been reported in a quantitative study of IN0 patients (Baloh et al. 1978b). However, the results of the present study show that during acute blockade adducting pursuit is impaired, especially as the movement approaches the midline. The reason that chronic IN0 patients show normal smooth-pursuit eye movements, but that animals with acute MLF lesions do not, may be related to oculomotor adaptation, which can increase the gain of the pursuit system (Carl and Gellman 1986). However, this would require unilateral changes in the gain of smooth pursuit and, to our knowledge, this issue has not been specifically examined.
HORIZONTAL
16O
6’0
FIG. 1 1. Explanation of the observed effects on accommodative vergence of MLF lidocaine blockade based on the examples in Figs. 7 and 8. For ease of explanation, as in Fig. 1, this figure presents a situation in which the normal accommodative eye movement results in an observed convergence of 10”. Left: shows the situation in which blockade of the left MLF has reduced the conjugate gain of the left eye with respect to the right eye by a factor of 0.5 (i.e., G = 0.5). Therefore, during the movement the right eye will make a conjugate movement of 5” to the right to stay on target, but because of the MLF blockade the left eye will move only 2.5” to the right. Despite this, the AC/A ratio in Fig. 7 is unchanged, and at the completion of the movement, the left eye will have moved 10” to the right. This suggests that the vergence component in the left eye increased from 5 Oto 7.5 O. Comparable reasoning can be applied to the example in Fig. 8 with the left eye viewing. Right: situation in which blockade of the left MLF has increased the conjugate gain of the right eye with respect to the left eye by a factor of 1.5 (i.e., G = 0.66). Accommodation of the left eye causes both eyesto converge. However, the vergence gain to the left eye is now greater than unity, and it converges more than the right eye. If the size of this vergence component is considered to be X0, then for the left eye to stay on target it must make a conjugate movement of X0 to the left. In addition, because of the MLF blockade, as the left eye moves X0 to the left, the right eye will move 1.5 times X0 to the left. Overall, at the completion of the movement, the right eye will have moved 5” + 1.5 *X0 to the left. The AC/A in Fig. 8 increased by 1.6 times (from 1.16 to 1.89 MA/D), which would correspond in this example to an eye movement of 16”. Thus 5 O+ 1.5 *X0 = 16”. This yields a value for X of 7.3 O,which suggests that the vergence component in the left eye increased from 5 to 7.3”.
MLF carries an inappropriate
vergence signal
The results indicate that the effect of MLF lidocaine injection on horizontal eye movements is due primarily to blockade of the ascending axons of abducens internuclear neurons. It is unlikely that the lidocaine anesthesia spread to the ascending tract of Deiter’s, because this courses laterally to the MLF at the level that the injections were made (McCrea et al. 1987). Also, it is unlikely that the observed effects were produced by blockade of the descending axons of oculomotor internuclear neurons. The vast majority of cells recorded in and around the oculomotor nucleus that carry a horizontal conjugate eye position signal increase their firing rate for convergence (Judge and Cumming 1986; Keller 1973; Mays et al. 1984a, b). However, only one fiber with this characteristic was encountered in the MLF in the primate (Gamlin et al. 1989). Consistent with this, the fibers of oculomotor internuclear
P. D. R. GAMLIN,
94
J. W. GNADT,
neurons in the cat have been reported to course in the “wings” of the MLF (Maciewicz et al. 1975, 1977). Therefore, the discussion will concentrate on the observed affects on vergence eye movements as they relate to lidocaine blockade of abducens internuclear axons in the MLF. An explanation for the AC/A values seen in the present study is presented in Fig. 11. For illustrative purposes, the left panel (right eye viewing) considers the example shown in Fig. 7, while the right panel (left eye viewing) considers the example shown in Fig. 8. In both these examples, during MLF lidocaine blockade, the observed AC/A values were significantly higher than would have been predicted by the null hypothesis. As can be seenin Table 1, this was also the casefor all other sessions.These increasesin AC/A are consistent with an increased net vergence signal to the left eye during lidocaine blockade of the MLF. This, in turn, indicates that an inappropriate vergence signal is normally present in the abducens internuclear pathway. However, it is difficult to accurately determine the magnitude of this signal. Our results indicate that during MLF lidocaine blockade, the vergence signal to medial rectus motoneurons is increased by - 50%. This result implies that the inappropriate MLF vergence signal also has a value of - 50%. However, the absolute value of the inappropriate MLF vergence signal could only be definitively determined by measuring the changes in AC/A at a number of orbital positions other than primary position. Nevertheless, the value of the inappropriate MLF vergence signal that we suggestfrom our present results is consistent with the accompanying single-unit study (Gamlin et al. 1989) which reported that the magnitude of the inappropriate MLF vergence signal was -50%. Our hypothesis that the MLF carries an inappropriate vergence signal is also supported by some other findings. A previous study (Evinger et al. 1977) reported that lesions of the MLF result in enhanced convergence movements. However, the authors suggestedthat these larger-than-normal convergence movements resulted from the need to overcome a large exophoria. Additionally, two electromyographic studies of the medial rectus muscle in cases in which conjugate signals were absent reported vergence activity (presumably) that was greater than expected (Loeffler et al. 1966; Orlowski et al. 1965). This prompted Kommere11(198 1) to speculate that the medial rectus motoneurons were exhibiting a denervation supersensitivity to the unimpaired vergence input. However, these results could be interpreted more parsimoniously as reflecting the activity of the higher gain midbrain vergence signal in the absenceof the inappropriate abducens internuclear signal. We thank S. Hayley for computer programming and L. Millican for technical assistance. We also thank Dr. David A. Corliss for designing the combined visual display and infrared optometer. This research was supported by National Eye Institute Grant RO 1-EY03463 to L. E. Mays and core Grant P30-EY-03039. P. D. R. Gamlin and J. W. Gnadt were supported by NE1 training Grant T32-EY-07033. Address for reprint requests: P. D. R. Gamlin, Dept. of Physiological Optics, School of Optometry, University of Alabama at Birmingham, Birmingham, AL 35294. Received
5 July
1988; accepted
in final form
3 March
1989.
AND
L. E. MAYS
REFERENCES
ASCHOFF, J.C., CONRAD, B., ANDKORNHUBER, H.H.Acquiredpendular nystagmus with oscillopsia in multiple sclerosis: a sign of cerebellar nuclei disease. J. Neurol. Neurosurg. Psychiatry 38: 570-577, 1974. BALOH, R. W., YEE, R. D., AND HONRUBIA, V. Internuclear ophthalmoplegia. I. Saccades and dissociated nystagmus. Arch. Neurol. 35: 484-489, 1978a. BALOH, R. W., YEE, R. D., AND HONRUBIA, V. Internuclear ophthalmoplegia. II. Pursuit, optokinetic nystagmus, and vestibulo-ocular reflex. Arch. Neurol. 35: 490-493, 1978b. BENDER, M. B. AND WEINSTEIN, E. A. Effects of stimulation and lesion of the median longitudinal fasciculus in the monkey. Arch. Neurol. Psychiatry 52: 106- 113, 1944. BENDER, M. B. Brain control of conjugate horizontal and vertical eye movements: a survey of the structural and functional correlates. Brain 103: 23-69, 1980. BIRD, A. C. AND LEECH, J. Internuclear ophthalmoplegia. An electrooculographic study of peak angular saccadic velocities. Br. J. Ophthalmol. 60: 645-65 1, 1976. BOGOUSSLAVSKY, J. AND REGLI, F. Monocular downbeat nystagmus. J. Neurol. 232: 99-101, 1985. BURDE, R. M., LEHMAN, R. A., ROPER-HALL, G., BROOKS, J., AND KELTNER, J. L. Experimental internuclear ophthalmoplegia. Br. J. Ophthalmol. 6 1: 233-239, 1977. BUTTNER, U., BUTTNER-ENNEVER, J. A., AND HENN, V. Vertical eye movement related unit activity in the rostra1 mesencephalic reticular formation of the alert monkey. Brain Res. 130: 239-252, 1977. CARL, J. R. AND GELLMAN, R. S. Adaptive responses in human smooth pursuit. In: Adaptive Processes in Visual and Oculomotor Systems in Advances in the Biosciences, edited by E. L. Keller and D. S. Zee. New York: Pergamon, 1986, vol. 57, p. 335-339. CARPENTER, M. B. AND MCMASTERS, R. E. Disturbances of conjugate horizontal eye movements in the monkey. II. Physiological effects and anatomical degeneration resulting from lesions of the medial longitudinal fasciculus. Arch. Neurol. 8: 347-368, 1963. CARPENTER, M. AND STROMINGER, N. L. The medial longitudinal fasciculus and disturbances of conjugate horizontal eye movements in the monkey. J. Comp. Neurol. 125: 41-66, 1965. COGAN, D. G. Internuclear ophthalmoplegia, typical and atypical. Arch. Ophthalmol. 84: 583-589, 1970. CRANE, T. B., YEE, R. D., BAI,OH, R. W., ANDHEPLER, R.S. Analysis of characteristic eye movement abnormalities in internuclear ophthalmoplegia. Arch. Ophthalmol. 10 1: 206-2 10, 1983. EVINGER, L. C., FUCHS, A. F., AND BAKER, R. Bilateral lesions of the medial longitudinal fasciculus in monkeys: effects on the horizontal and vertical components of voluntary and vestibular induced eye movements. Exp. Brain Res. 28: l-20, 1977. FELDON, S. E., HOYT, W. F., AND STARK, L. Disordered inhibition in internuclear ophthalmoplegia: analysis of eye movement recordings with computer simulations. Brain 103: 113- 137, 1980. GAMLIN, P. D. R., GNADT, J. W., AND MAYS, L. E. Lidocaine-induced unilateral internuclear ophthalmoplegia: Effects on accommodative vergence. Sot. Neurosci. Abstr. 14: 6 12, 1988.
GAMLIN,P. D. R., GNAIX, J. W., AND MAY& L. E. Abducens internuclear neurons
carry
an inappropriate
Neurophysiol. 62: 70-8 1, 1989. JACOM.E, D. E. Monocular downbeat
signal
for ocular
convergence.
J.
nystagmus. Ann. Ophthalmol. 18: 293-296, 1986. JUDGE, S. J. AND CUMMING, B. G. Neurons in the monkey midbrain with activity related to vergence eye movement and accommodation. J. Neurophysiol. 55: 9 15-930, 1986. KELLER, E. L. Accommodative vergence in the alert monkey. Motor unit analysis. Vision Rex 13: 1565- 1575, 1973. KENYON, R. V., CIUFFREDA, K.J., AND STARK, L.Binoculareye movements during accommodative vergence. Vision Res. 18: 545-555, 1978. KING, W. M. AND FUCHS, A. F. Reticular control of vertical saccadic eye movements by mesencephalic burst neurons. J. Neurophysiol. 42: 861-876, 1979. KING, W. M., LISBERGER, S. G., AND FUCHS, A. F. Responses of fibers in medial longitudinal fasciculus (MLF) of alert monkeys during horizontal and vertical conjugate eye movements evoked by vestibular or visual stimuli. J. Neurophysiol. 39: 1135- 1149, 1976.
INTERNUCLEAR
OPHTHALMOPLEGIA
T. H. AND KATSARKAS, A. An electrooculographic study of internuclear ophthalmoplegia. Ann. Neural. 2: 385-392, 1977. KOMMERELL, G. Unilateral internuclear ophthalmoplegia. The lack of inhibitory involvement in medial rectus muscle activity. Invest. Ophthalmol. Visual Sci. 2 1: 592-599, 198 1. KRUGER, P. Infrared recording retinoscope for monitoring accommodation. Am. J. Optom. Physiol. Opt. 59: 726-734, 1982. LOEFFLER, J. D., HOYT, W. F., AND SLATT, B. Motor excitation and inhibition in internuclear palsy. An electromyographic study. Arch. Neural. 15: 664-67 1, 1966. MACIEWICZ, R. J., KANEKO, C. R., HIGHSTEIN, S. M., AND BAKER, R. Morphophysiological identification of interneurons in the oculomotor nucleus that project to the abducens nucleus in the cat. Brain Res. 96: 60-65, 1975. MACIEWICZ, R. J. AND SPENCER, R. F. Oculomotor and abducens internuclear pathways in the cat. In: Control ofGaze by Brain Stem Neurons, edited by R. Baker and A. Berthoz. Amsterdam: Elsevier/North Holland, 1977, p. 99-108. MAYS, L. E. Neural control of vergence eye movements: convergence and divergence neurons in the midbrain. J. Neurophysiol. 5 1: 109 l- 1108, 1984a. MAYS, L. E. AND PORTER, J. D. Neural control of vergence eye movements: activity of abducens and oculomotor neurons. J. Neurophysiol. 52: 743-76 1, 1984b. MCCREA, R. A., STRASSMAN, A., DAY, E., AND HIGHSTEIN, S. M. Anatomical and physiological characteristics of vestibular neurons mediatKIRKHAM,
AND VERGENCE
95
ing the horizontal vestibulo-ocular reflex of the squirrel monkey. J. Comp. Neurol. 264: 547-570, 1987. METZ, H. S. Saccadic velocity measurements in internuclear ophthalmoplegia. Am. J. Ophthalmol. 8 1: 296-299, 1976. MUELLER, J. Elements ofphysiology, translated by W. Baly. London: Taylor and Walton, 1843. NOZAKI, S., MUKUNO, K., AND ISHIKAWA, S. Internuclear ophthalmoplegia associated with ipsilateral downbeat nystagmus and contralateral incyclorotatory nystagmus. OphthaImoZogica 187: 2 10-2 16, 1983. ORLOWSKI, W. J., SLOMSKI, P., AND WOJTOWICZ, S. Bielschowsky-LutzCogan Syndrome. Am. J. Ophthalmol. 59: 416-430, 1965. POLA, J. AND ROBINSON, D. A. An explanation of eye movements seen in internuclear ophthalmoplegia. Arch. Neural. 33: 447-452, 1976. POLA, J. AND ROBINSON, D. A. Oculomotor signals in medial longitudinal fasciculus of the monkey. J. Neurophysiol. 41: 245-259, 1978. SCHOR, C. M. Models of mutual interactions between accommodation and convergence. Am. J. Optom. Physiol. Opt. 62: 369-374, 1985. STROUD, M. H., NEWMAN, N. M., KELTNER, J. L., AND GAY, A. J. Abducting nystagmus in the medial longitudinal fasciculus (MLF) syndrome (internuclear ophthalmoplegia (INO)). Adv. Oto-rhino-Zaryngol. 19: 367-376, 1973. USUI, S. AND AMIDROR, I. Digital low-pass differentiation for biological signal processing. IEEE (Inst. Electr. Electron. Eng.) Trans. Biomed. Eng. 29: 686-693, 1982. ZEE, D. S., HAIN, T. C., AND CARL, J. R. Abduction nystagmus in internuclear ophthalmoplegia. Ann. Neural. 21: 383-388, 1987.
