OCULAR TORSION AND T H E FUNCTION O F T H E VERTICAL EXTRAOCULAR MUSCLES ROBERT S. JAMPEL,
M.D.
Detroit, Michigan
When the heads of human and nonhuman primates are tilted to the shoulder there are five possible eye responses: (1) The eyes ro tate in the opposite direction around the pupillary axes (lines perpendicular to the plane of the iris that go through the center of the pupil used as a reference line for eye movement) so that the vertical corneal meridia (connect the 12 and 6 o'clock points on the corneoscleral limbus when the head is erect) remain perpendicular to the horizon, i.e., so-called compensatory torsion or coun ter-rolling. (2) The eyes undergo partial counter-rolling around the pupillary axes, i.e., the eyes rotate in the opposite direction a small fraction of the amplitude of head or body tilt. (3) The eyes counter-roll either completely or partially around some axes other than the pupillary axes—so-called "eccentric" rotation—or the eyes are trans lated slightly in their orbits, i.e., the whole eyeball moves horizontally rather than the eyes rotating around a fixed center of rota tion. (4) The vertical corneal meridia remain parallel to the sagittal plane of the head and there is no ocular movement. (5) The verti cal corneal meridia counter-roll or move slowly a few degrees in the opposite direc tion to the head and body inclination owing to inertia or gravity. When the head or body movement stops, the eyes reverse di rection and move rapidly (saccadically) in the direction of the previous head movement until the vertical corneal meridia reestab lish parallelism with the sagittal planes of the head. The first idea of compensatory torsion is widely accepted in clinical ophthalmology From the Kresge Eye Institute, Wayne State University, Detroit, Michigan. Reprint requests to Robert S. Jampel, M.D., Kresge Eye Institute, 3994 John R. St., Detroit, MI 48201.
and it is taught that the eyes conjugately ro tate around their pupillary axes in a direc tion opposite to a head or body tilt so that the vertical corneal meridia maintain a posi tion perpendicular to the horizon.1"5 This important idea was basic in establishing concepts of the function of the vertically acting extraocular muscles, in explaining the physiology of the vestibular apparatus, and in developing systems of clinical analysis for ocular motor defects.6 However, a sur vey of the literature beginning with John Hunter 7 in 1786 does not reveal any ex perimental evidence or documented clinical observations to prove that compensatory ocular rotation is a phenomenon occurring in man.6-9 Ocular torsion was measured in various positions of head and body tilt in man showing that torsional eye movements cannot fulfill a compensatory function.10"15 The maximum ocular movement or tor sion measured in man was about 11° for large amplitudes of head, neck, and body inclination (more than 90°).10"14 For 20° of head torsion the amount of counterrolling or displacement is less than 3°. The eyes of a person whose head is tilted toward the shoulder apparently roll and lag behind the head about 10% of the tilt.15 Similar measurements of counter-rolling were made in the rhesus monkey.16 Those investigators who measured some ocular displacement or torsion did not demonstrate experimentally that the move ment occurred around the pupillary axis. They did not appear concerned about how this movement took place ; possibly it was due to an ocular rotation around some axis other than the pupillary axis or to displace ment of the whole eye (translation). 17 When the head is tilted the eyes are also displaced inward and downward or upward (Figs. 1 and 2 ) . According to my unpublished ob-
292
VOL. 79, NO. 2
OCULAR TORSION
293
servations, no compensation for these hori to 20° of head torsion. I also attempted to measure counter-rolling of the eyes with zontal and vertical movements takes place. Some investigators including Helmholtz neck and body inclinations over a wide and Müller18 have denied the existence of amplitude of up to 100° (Fig. 3). My tech any ocular torsion or eye niovement with nique is described briefly in the legends head or body tilt.19-20 They point to the com of Figures 3 and 4. A thin strip of egg plexity of the techniques and consider the shell membrane was placed horizontally on the cornea and a plumb line was hung from measurement as artifactuous. When I first strove to measure counter- a trial frame. Motion pictures were taken as rolling of the eyes I permitted the person the head and body inclined. While the head to rotate his head only around an antero- was inclining I noted a slight counter-move posterior axis that emerged through the ment of the eye but when the movement of root of his nose and halfway between his the head stopped the ocular counter-move eyes (Figs. 1 and 2). The maximum head ment was not sustained. No ocular countertorsion I measured was about 20° and I rolling could be measured with the head or prefer to call this "true" head torsion. Head body in a sustained inclined position (Fig. tilting beyond 20° requires movement of 3). Experiments with lightly anesthetized the whole cervical column and the shoulders rhesus monkeys failed to reveal any sus and ought to be termed "head and body in tained compensatory counter-rolling to head clination" rather than head torsion.21 I tilt (Fig. 4). thought that if compensatory counter-rolling Recently, a more sophisticated technique occurred it would most likely be a response than I employed detailed dynamically the re-
Fig. 1 (Jampel). Head torsion has an amplitude of about 20° and takes place in the atlanto-occipital articulation, i.e., around an anteroposterior axis that emerges through the root of the nose and halfway between the eyes. Movements of greater in clination are executed by the whole cervical column. If counter-rolling takes place at all, it is 3° or less for a head torsion of 20°.
294
AMERICAN JOURNAL OF OPHTHALMOLOGY
FEBRUARY, 1975
Fig. 2 (Jampel). Eye displacement with head torsion. When the head is rotated 20° around O (a point on the horizontal plane that bisects the line connecting the centers of rotation of the eyes), the pupillary axes are displaced downward or upward (P to P') about 11 mm and inward about 2 mm. If AB (the vertical corneal meridian) is to re main perpendicular to the horizon then point A would be expected to move to point C as the head twisted 20°. Angle POP' would equal angle A'P'C. In our experiments AB moved to A'B' and minimal or no counter-rolling took place. There is no com pensation for the vertical or inward movements. sponse of the eyes to head and body tilt. 22 The authors found that smooth tilts of the head to the shoulder caused ja slow counterrotation of the eyes followed by a saccadic rotation in the direction of the head tilt. T h e net effect was maintenance of the verti cal corneal meridian more or less parallel to the sagittal plane of the head. Residual torsional displacement to sustained head tilt was considered artifactuous. However, for the purpose of this paper, the existence of some slight ocular torsion or displacement with head torsion and in clination may be granted (artifact or n o t ) and the current question put to rest for the moment because such slight movements a r e of no proven value or importance to the clinician though interesting to the physiolo gist. Also, if slight sustained ocular dis placement or torsion does take place with
head tilt, the question of around which axis it takes place has not yet been answered. However, there is no compensatory ocular rotation or torsion in man in response to head inclination, i.e., the vertical meridian of the cornea is not maintained perpendicu lar to the horizon when the head is inclined. In the light of this fact, I will examine the function of the vertical extraocular muscles: how can the oblique muscles and vertical rectus muscles be significant tortors when no significant ocular torsion takes place in the normal human or nonhuman primate? The
oblique
muscles—PREVALENT
CON
CEPT OF OBLIQUE MUSCLE FUNCTION—Based
on an analysis of the position of its origin and insertion, we have been taught that the superior oblique muscle ( S O ) is mainly an intortor, secondly a depressor, and thirdly an abductor. 3 ' 4 T h e functions of intortion
VOL. 79, NO. 2
295
OCULAR TORSION
and abduction increase with abduction of the eye and the function of depression in creases with adduction of the eye. The in ferior oblique muscle (IO) is mainly an extortor, secondly an elevator, and thirdly an abductor. The functions of extorsion and
abduction increase with abduction of the eye and the function of elevation increases with adduction of the eye. The IO is an antagonist of the SO for torsion and de pression but a synergist for abduction. As the eye abducts owing to the contraction
D
Fig. 3 (Jampel). T o demonstrate that compensatory ocular counter-rolling is not clinically detectable in a normal subject, a white marker (eggshell membrane) is placed on the subject's eye. The vertical white line is a plumb line hanging from a t r h l frame. The eye marker mnintrins its relative position in regard to the lower eyelid in all positions of head inclination. A, head erect ; B, head inclined about 20° ; C, head inclined about 25° ; D, head inclined 50° ; E, head inclined 60°.
296
AMERICAN JOURNAL
OF OPHTHALMOLOGY
FEBRUARY, 1975
pensatory rolling of the eye and is assisted by the inferior rectus muscle. E X P E R I M E N T A L EVALUATION OF OBLIQUE
F U N C T I O N I N T H E MONKEY—Different con
Fig. 4 (Jampel). Simple experiment with a lightly anesthetized monkey to demonstrate the absence of compensatory counter-rolling. The eyes were held open with eyelid retractors. The white markers on the corneas are eggshell membrane. The black marker between the eyes is a strip of black tape. The relative positions of the various markers are held constant in spite of head tilt. Top, head erect; center, head inclined left about 27°; bottom, head inclined right about 17°. of the lateral rectus muscle, simultaneous contraction of the oblicjue muscles causes ad ditional abduction that completes the lateral movement. T h e S O is the main intortor in compensatory rolling of the eye and is as sisted in this function by the superior rectus muscle. The I O is the main extortor in com-
cepts of the function of the S O and the I O were formulated from experimental work in the monkey. 2 3 - 2 5 W e found that the axis of rotation of the oblique muscles is fixed in the orbit and is not displaced when the eye moves in the horizontal plane. T h e S O and the I O share the same rotational axis and are definite antagonists in all horizontal gaze positions. T h e amount of rotation around this axis is always the same for a given con traction of an oblique muscle. A s the eye adducts, the amount of both depression or elevation and lateral displacement increases (Fig. 5 ) . W h e n the S O contracts with the eye located above the horizontal plane, the pupillary axis undergoes an increasing ad duction until it reaches the horizontal plane where it undergoes an increasing abduction. W h e n the I O contracts with the eye located below the horizontal plane, the pupillary axis undergoes an increasing adduction until it reaches the horizontal plane where it under goes an increasing abduction. Experimental co-contraction of the S O and I O does not produce abduction or propulsion of the globe ; the eye remains immobile. Contraction of the S O alone or in combination with the superior rectus muscle does not produce wheel-rotation (intortion) around the pupil lary axis. Contraction of the I O alone or in combination with the inferior rectus muscle does not produce wheel-rotation (extorsion) around the pupillary axis. Function of the superior oblique muscle in man—The principles of oblique muscle action derived from experimentation in monkeys (Macaco mulatta) were confirmed in man by clinical observations and experi ments. Oblique muscle action differs between monkey and man in that the amplitude of horizontal eye movement in man is twice that of the monkey, i.e., about 90° in m a n and 45° in the monkey. PATIENTS
WITH
OCULOMOTOR
NERVE
297
OCULAR TORSION
VOL. 79, NO. 2
an angle of about 60° with the Y-axis of Fick (Fick's axes comprise a rectangular coordinate system that is absolutely fixed in the orbit. The center point of this system is the center of rotation of the eye. The hori zontal axis is denoted X, the anteroposterior axis Y, and the vertical axis Z) (Fig. 6, B). This so-called eccentric rotation is charac teristic of the SO and is clearly seen when the cornea is marked with a vertical strip of eggshell membrane (Fig. 6, C). The rare patient with paralysis of both the oculomotor and abducens nerves and
PARALYSIS—When total paralysis of the oculomotor nerve with intact trochlear and abducens nerves occurs, the levator palpebrae, superior rectus, medial rectus, inferior rectus, and IO muscles are paralyzed. The SO and lateral rectus muscles are function ing. The eye is deviated outward, not down ward and outward as is usually taught (even though the SO is unopposed by the para lyzed IO and superior rectus muscles) (Fig. 6, A ) . When commanded to look down or to pursue a moving object, the eye will rotate downward around a fixed axis that forms
AD
AB A\B
P
P -
-
B A' P' B' Fig. S (Jampel). Demonstrating superior oblique (SO) muscle action in the monkey. The anterior axis pole (C) of the rotational axis of the SO remains fixed in the orbit when the eye moves in the horizontal plane (CP) owing to the activity of the horizontal rectuses. A given contraction of the SO causes the pupil lary axis (P) to transcribe larger and larger parallel arcs of latitude (P to P') as the eye moves from a posi tion of abduction {AB) to a position of adduction {AD). Both the outward displacement {PF) of the pupillary axis ( P ) and the backward displacement {PV) increase as the eye adducts. The angle of ocular rotation {PCP1) is equal to the angle of displacement of the vertical corneal meridian {A'P'F) in all horizontal gaze positions. The lower diagram shows the eye movement projected onto a frontal plane.
298
AMERICAN JOURNAL OF OPHTHALMOLOGY
--A
sparing of the SO has an eye located in the midposition. When commanded to look down, the eye rotates downward around the same axis it rotated around when the eye was in the abducted position (Fig. 6, D ) . The same type of eye movement is common ly seen after the retrobulbar injection of anesthesia into the muscle cone in prepara tion for cataract surgery. In this situation all the extraocular muscles except the SO may be paralyzed since the nerve to the SO is outside of the muscle cone. When commanded to look down, the eye under goes the downward eccentric rotation char acteristic of the SO. In patients with oculomotor nerve paralysis and sparing of the trochlear nerve, a silk suture was carefully sewn through the lateral corneoscleral limbus after topical
FEBRUARY, 1975
Fig. 6 (Jampel). Action of the superior oblique ( S O ) muscle in man is demonstrated in a patient with oculomotor nerve paralysis. ( A ) In oculo motor nerve paralysis the eye deviates outward and not outward and - down ward as we have been taught. ( B ) On command to look down, the eye ro tates around point C from P to P' transcribing an arc of latitude. (C) The cornea is marked with a strip of eggshell mem brane to make the type of movement evident. ( D ) The eye from the straight-ahead position also rotates around point C transcribing an arc of latitude. ( E ) The eye can be passively moved into different positions in the horizontal plane by means of a corneoscleral limbal suture. The eye move ments produced by the SO have the same char acteristics from all hori zontal gaze positions and an estimate of the magni tude of these movements is possible.
anesthesia and the eye was passively ro tated medially in small increments. On com mand to look down, the eye rotated down ward from any position in the horizontal plane around the same fixed eccentric axis (Fig. 6, E ) . No rotation of the eye around the pupillary axis produced by isolated SO contraction has even been observed. Thus, in man and in monkey, a given contraction of the SO causes the eye to transcribe larger and larger arcs of latitude as the eye moves medially away from the fixed SO axis. These downward arcs in crease in magnitude until a point 90° (30° of adduction) from the SO axis is reached at which time the arcs decrease in magni tude for the same SO contraction (Fig. 7). Patients with isolated SO paralysis also provide insight into the function of that
VOL. 79, NO. 2
OCULAR TORSION
muscle. Interpretation of the Lancaster redgreen tests performed on such patients is revealing (Fig. 8 ) . In right SO paralysis with the head erect and the left eye fixing, the right eye is dis placed upwards and outwards when the pa tient is looking in the horizontal plane or downward from all positions of horizontal gaze. When he looks upward there is no displacement. The axis around which the SO rotates the eye may be interpolated by connecting the midpoints of the projections of the fixation lights in adduction, primary position, and abduction. This angular rota
LATERAL
299
tion of the projection of the vertical fixation light to the horizontal plane is the same in adduction, primary position, and abduction, and is about 17°. Function of inferior oblique muscle in man—We have measured the action of an isolated IO in man 23 in a patient with an acquired paralysis of the superior rectus muscle (Figs. 9 and 10). The axis, of rota tion of the human IO was located in the horizontal plane about 60° lateral to the Yaxis of Fick and behaved as if it were fixed in the orbit. The IO rotated the eye about 17° upward from the horizontal plane
MEDIAL
Fig. 7 (Jampel). Diagram of the action of the superior oblique (SO) muscle in man : Arc AMKDLC is the horizontal plane. Arc L to M is the approximate range of hori zontal ocular movement in man. The rotational axis of the SO is line ABC with B the center of rotation of the eye. This axis is located about 60° from line BD which rep resents the primary position and corresponds to the Y-axis of Fick. When the SO con tracts with the eye in any position in the horizontal plane the center of the pupil (P) transcribes an arc of latitude (P to P'). The arc of latitude has three components : down. lateral (PF), and back into the orbit. These components increase in magnitude until the eye rotates 90° (to point K) from the rotational axis (line ABC) and then they decrease. Angle ACP' is equal to angle A'P'F; thus the angle of ocular rotation is equal to the angular displacement of the corneal meridian (lines AB and A'B' in the lower diagram).
300
AMERICAN JOURNAL OF OPHTHALMOLOGY
x :
FEBRUARY, 1975
M i l l i π Τ Ρι' I1 1 1 1 1 Tu
AO_ i
1
i
AB 1 i 1
»
1
1
*
!1 ! 1 ■■
j
.AD
1 î
\
4
i
|
|
*
W
lï M
1 1 l i II M
i-4-»-i-»-n-M-i»-ii-»
; 1 ! ''
'
1
1' I Ì1 1
' '
»
f/
f~~\ ]'}*& ^ / ' 1
irL
"
/ !
I>
AB
> 1 /
/
i
1H i ! — l - D i — I D ■
io-
~j
-Γ -Q* * ! ~V X- \ -
■ 1I D — l - D — I t
1
Y *^ì
1
t
9
:ΐJU^ π ^^^ > ;: H ¥ ±f « : 2Δ i Kv> i7 :
îi
1 !Î i i ii ' J
itJ_
Head tilted to left
Head erect
Head tilted to right
A
B
c
Fig. 8 (Jampel). The Lancaster red-green projection test in isolated superior oblique muscle paralysis. (A) The projections are normal when the head is tilted to the contralateral shoulder. (B) With the head erect, the projections are abnormal in horizontal gaze (not shown) and in downward gaze. In downward gaze, the vertical projections undergo the greatest vertical and lateral displacement (abduction) in the position of adduction (AD). The angular displacements of the vertical projections to the horizontal plane are the same in all gaze positions. (C) The projections are abnormal in all gaze positions when the eye is tilted to the homolateral shoulder (open bar, R.E. ; solid, L.E.).
M
Fig. 9 (Jampel). The eye movements produced in man by the inferior oblique (IO) and superior oblique (SO) muscles projected onto a frontal plane are shown. When the eye (P, the center of the pupil) rotates around the axis pole (A) from the hori zontal plane, an arc of latitude (P to P' or P to P") is produced. The arcs increase in magnitude until the eye moves horizontally to E, which is 90° from A or 30° from the straight-ahead position. The arcs decreased in magnitude with further medial rotation from E. The horizontal displacement (PF) also increases until point E is reached and then decreases ; L, lateral ; M, medial.
VOL. 79, NO. 2
OCULAR TORSION
301
3mm
2 mm
Z
mm
o I< >
UJ
_i UJ
mm
2 mm 3 mm
4 mm
ADduction
Fig. 10 (Jampel). The approximate amount in millimeters of depression produced by the superior oblique (SO) and elevation produced by the inferior oblique (IO) muscles in man projected onto a frontal plane with the head in the normal erect position and the eye in different positions in the horizontal plane. The maximum rotation of the eye (0) is about 17° around the axis of the oblique muscles. Note that the maximum elevation or depression occurs when the eye is adducted about 30°. (The SO measure ments were obtained from the experiment in Fig. 6, E.)
around its axis (Figs. 9 and 10). As the eye moved across the horizontal plane from lateral to medial, because of the contrac tion of the medial rectus muscle and the inhibition of the lateral rectus muscle, con traction of the IO produced the following: the eye moving medially away from its axis transcribed larger and larger arcs of latitude (each arc represented 17° of ocular rotation around the IO axis) until the eye was 90° from the IO axis and in a position of 30° of adduction, and the arcs then became smaller. These arcs were composed of an elevating and lateral component. There was increasing elevation and lateral displacement of the eye until 30° of adduction was reached where the elevation and lateral dis placement began to decrease.
DISCUSSION
The eyes in man and monkey in the physiologically integrated state either do not undergo conjugate counter-rolling at all or do to a minimal degree that is wholly in sufficient to compensate for head tilt or body displacement or both. Even investi gators who have demonstrated some slight ocular torsion have not shown that this movement represents a wheel-like rotation around the pupillary axis. This movement could take place around an axis other than the pupillary axis or be the result of translatory movement of the eye. The possibility that the movements are artifacts has not been completely eliminated. With our pres ent limited sophistication in the clinical analysis of ocular motor defects, we con-
302
AMERICAN JOURNAL OF OPHTHALMOLOGY
sider that no compensatory rotation or dis placement of the eyes takes place. During extensive experiments with mon keys, I observed the actions of individual and various combinations of extraocular muscles and produced no wheel-like rotation of the eyes around the pupillary axis.23"25 Stimulation of the ocular motor nuclei in the brainstem, the ampullary nerves of the labyrinths, and the frontal eye fields never resulted in wheel-like ocular rotation.24 For example, simultaneous stimulation of the SO and superior rectus muscles never produced intortion of the eyes and simultaneous stim ulation of the IO and inferior rectus muscles never produced extorsion of the eyes. The mechanic and neurologic mecha nisms for wheel-rotation of the eyes around the pupillary axis apparently do not exist. When an oblique muscle acts alone ex perimentally—a situation which never occurs normally—it produces a rotary movement of the eye around a fixed eccentric axis that forms an angle in man of about 60° with the Y-axis of Fick. This axis remains fixed in the orbit regardless of horizontal eye movements around the Z-axis of Fick. As a result of the location of this axis contraction of an oblique muscle produces an angular rotation of the eye equal to the angular displacement of the vertical corneal merid ian when projected onto the frontal plane (Figs. 5 and 7). The oblique muscles in ex periments with monkeys never produce a wheel-rotation of the eye around the pupil lary axis when acting alone or in combina tion with the vertical rectuses. There is no compensatory conjugate counter-rolling of the eyes to head tilt in order to maintain the retinal horizon parallel to the terrestrial horizon. Also, the oblique muscles appear incapable of producing wheel-like rotation of the eyes, or any sig nificant ocular torsion for that matter, in the normal state. Thus, the neurologic mechanisms that govern the vertical extraocular muscles and the structure of these muscles appear de
FEBRUARY, 1975
signed to inhibit or prevent ocular rolling or torsion and to maintain the eyes in a stable position relative to that of the head, i.e., to keep the vertical corneal meridian as parallel as possible to the sagittal plane of the head. When the head or body or both are inclined, the vertical muscles acting to gether hold the eyes in a stable position. Any residual torsional movements that may exist might be vestigial. Having maintained stable eyes, the vertical extraocular muscles, which in the normal state are neither sig nificant tortors nor horizontal ductors, work in concert to produce elevation or depression of the eyes with the head or body in any position or posture. When the head or body is tilted, the verti cal corneal meridia are held parallel to the sagittal plane of the head. Therefore, the in sertions of the extraocular muscles are dis placed equal to the angle of head tilt from their original position. The insertions of the horizontal rectus muscles, for example, are no longer parallel to the horizon and yet the eye with the head tilted is capable of a smooth pursuit or saccadic movement in the terrestrial horizontal plane. Such a movement is difficult to explain employing what is al ready known about ocular mechanics. Pos sibly an innervational shift or redistribution among the vertical and horizontal extraocular muscles produced by head tilt or body dis placement enables such an eye movement to take place. The head-erect position sends the innervations for elevation and depres sion of the eyes to both the vertical rectus and oblique muscles although not necessarily equally. When the head is tilted, innervation may be shifted to the vertical rectus muscles in the contralateral eye and to the obliques in the homolateral eye for the production of ele vation and depression. This is suggested by some cases of superior oblique muscle paralysis in man (Fig. 8). The otolith organs of the vestibular apparatus may be responsi ble for this innervational shift. Electromyographic studies have shown increasing electrical activity in the SO and
VOL. 79, NO. 2
303
OCULAR TORSION
superior rectus muscles and decreasing elec trical activity in the IO and inferior rectus muscles in the homolateral eye during head tilt in man and the reverse in the contralateral eye.26 These findings might be em ployed as evidence for ocular torsion. How ever, these electrical activities probably represent isometric muscle contractions that stabilize the eyes in relation to the head and prevent both torsion and vertical ocular dis placement. Measuring the innervation in the vertical extraocular muscles when the subject looks up and down from all gaze positions with the head tilted would be an important experiment. Torsional displacement of the vertical corneal meridian takes place only in patho logic conditions (not to be confused with so-called pseudotorsion seen in normal oblique muscle gazes) such as extraocular muscle paralysis or nystagmus and in certain abnormal phorias and tropias, but this dis placement has never been wheel-like, i.e., it does not take place around the pupillary axis. Spatial orientation and the recognition of an "objective" vertical when the head tilts or when the body is displaced is maintained in spite of the absence of compensatory eye movements. The mechanisms responsible for this phenomenon reside in the brain or in an interplay between the retina and the brain and not in the peripheral ocular motor apparatus. Recent experiments indicate that cyclofusion can take place without ocular movement.27 Wheel-like rotational eye movements (cycloduction) around the pu pillary axis do not take place and torsional displacement of the vertical corneal merid ian is always associated with a vertical ocular displacement. Tilting of a perceived vertical straight edge of 40° was produced in cer tain psychologic experiments that involved rotating displays around the line of sight, without significant ocular rotation or dis placement.23 Thus, visual sensory adapta tions to body and head inclinations are not dependent on compensatory eye movements
and certainly not an ocular torsion around the pupillary axis. SUMMARY
The vertical corneal meridia are not kept perpendicular to the horizon in human and nonhuman primates when the head or body is tilted, i.e., compensatory counter-rolling of the eyes does not occur. The slight tor sional displacement of the vertical corneal meridia noted by many observers may be the result of rotation around an axis other than the pupillary axis or to translation of the globe. The neurologic and structural systems that control the actions of the vertical muscles in human and nonhuman primates do not appear to provide a mechanism for wheel-rotation of the eyes around the pu pillary axis. Ocular torsion is not a normal function of the vertical extraocular muscles. Their function is probably the reverse, i.e., the inhibition or prevention of ocular torsion and the stabilization of the eyes when the head or body inclines. Torsional displace ment of a vertical corneal meridian occurs only when there is an abnormal muscle im balance. Wheel-like movements (cycloduction) around the pupillary axis or visual line do not occur. Torsional displacement of a verti cal corneal meridian occurs only with a simultaneous vertical movement. The vertical rectus and the oblique muscles in man work together to produce vertical ocular movements regardless of head position or body posture while main taining the vertical corneal meridia parallel to the sagittal plane of the head. The vestibular apparatus may be responsible for dis tributing innervations among these muscles, enabling them to function in this manner. REFERENCES
1. Walls, G. L. : The evolutionary history of eye movements. Vision Res. 2:69, 1962. 2. Adler, F. H. : Physiologic factors in the dif ferential diagnosis of paralysis of the superior
304
AMERICAN JOURNAL OF OPHTHALMOLOGY
rectus and superior oblique muscles. Arch. Ophthalmol. 36:661, 1946. 3. Moses, R. A. : Adler's Physiology of the Eye. St. Louis, C. V. Mosby, 1970, pp. 81, 187, 200, and 236. 4. Lyle, T., and Bridgeman, "G. : Worth and Chavasse's Squint. London, Baillere, Turdall and Cox, 1959, p. 236. 5. Marquez, M. : Supposed torsion of the eye around the visual axis in oblique directions of gaze. Arch. Ophthalmol. 41:704, 1949. 6. Parks, M. M. : Isolated cyclovertical muscle palsy. Arch. Ophthalmol. 60:1027, 1958. 7. Hunter, cited by Palmer, J. F. (ed.) : The Works of John Hunter, F.R.S., with Notes. Lon don, Longman, Rees, Orme, Brown, Green, and Longman, 1837. 8. Lancaster, W. B. : Terminology in ocular motility and allied subjects. Am. T. Ophthalmol. 26: 122, 1943. 9. Stevens, G. T. : Motor Apparatus of the Eyes. Philadelphia, F. A. Davis Company, 1906, p. 123. 10. Duke-Elder, S. : Textbook of Ophthalmology. St. Louis, C. V. Mosby, vol. 1, 1939, pp. 623628. 11. Woellner, R. C, and Graybiel, A.: Counterrolling of the eyes and its dependence on the mag nitude of gravitational or inertial forces acting laterally on the body. J. Appi. Physiol. 14:632, 1959. 12. Belcher, S. J. : Ocular torsion. Br. J. Physiol. Opt. 2:1, 1964. 13. Miller, E. F. : Counterrolling of the human eyes produced by head tilt with respect to gravity. Acta Otolaryngol. 54:479, 1962. 14. Linwong, M., and Herman, S. : Cycloduction of the eyes with head tilt. Arch. Ophthalmol. 85 : 570, 1971.
FEBRUARY, 1975
15. Davson, H. : The Physiology of the Eye. London, Churchill, 1963. 16. Krejcova, H., Highstein, S., and Cohen, B. : Labyrinthine and extralabyrinthine effects on ocu lar counter-rolling. Acta Otolaryngol. 72:165, 1971. 17. Levine, M. : Pendulum-like eye movement. Am. J. Ophthalmol. 75:979, 1973. 18. Southall, J. : Helmholtz's Treatise on Physio logical Optics, vol. 3. New York, Dover Publica tions, Inc., 1962, pp. 120 and 139. 19. Hewitt, R. S. : Torsional eye movements. Am. J. Ophthalmol. 34:253, 1951. 20. Levine, M. : Evaluation of the Bielschowsky head-tilt test. Arch. Ophthalmol. 82:433, 1969. 21. Rouviere, H. : Anatomie Humaine. Descrip tive et Topographique. Paris, Masson et Cie, 1959, p. 132. 22. Petrov, A., and Zenkin, G. M. : Torsional eye movements and constancy of the visual field. Vision Res. 13 :2465, 1973. 23. Jampel, R. S. : The fundamental principle of the action of the oblique ocular muscles. Am. J. Ophthalmol. 69:623, 1970. 24. : The action of the superior oblique muscle. Arch. Ophthalmol. 75 :535, 1966. 25. : Experimental evaluation of the ocu lar rolling (torsion) in the Macaque. Proceedings of the Annual Conference on Engineering in Medi cine and Biology, 9:7.6, 1967. 26. Scott, A. B. : Extraocular muscles and head tilting. Arch. Ophthalmol. 78:397, 1967. 27. Kertesz, A. E. : The effect of stimulus com plexity on human cyclofusional response. Vision Res. 12:699, 1972. 28. Dichgans, J., and Held, R. : Moving visual scenes influence the apparent direction of gravity. Science 178:1217, 1972.
M.D.
Detroit, Michigan
When the heads of human and nonhuman primates are tilted to the shoulder there are five possible eye responses: (1) The eyes ro tate in the opposite direction around the pupillary axes (lines perpendicular to the plane of the iris that go through the center of the pupil used as a reference line for eye movement) so that the vertical corneal meridia (connect the 12 and 6 o'clock points on the corneoscleral limbus when the head is erect) remain perpendicular to the horizon, i.e., so-called compensatory torsion or coun ter-rolling. (2) The eyes undergo partial counter-rolling around the pupillary axes, i.e., the eyes rotate in the opposite direction a small fraction of the amplitude of head or body tilt. (3) The eyes counter-roll either completely or partially around some axes other than the pupillary axes—so-called "eccentric" rotation—or the eyes are trans lated slightly in their orbits, i.e., the whole eyeball moves horizontally rather than the eyes rotating around a fixed center of rota tion. (4) The vertical corneal meridia remain parallel to the sagittal plane of the head and there is no ocular movement. (5) The verti cal corneal meridia counter-roll or move slowly a few degrees in the opposite direc tion to the head and body inclination owing to inertia or gravity. When the head or body movement stops, the eyes reverse di rection and move rapidly (saccadically) in the direction of the previous head movement until the vertical corneal meridia reestab lish parallelism with the sagittal planes of the head. The first idea of compensatory torsion is widely accepted in clinical ophthalmology From the Kresge Eye Institute, Wayne State University, Detroit, Michigan. Reprint requests to Robert S. Jampel, M.D., Kresge Eye Institute, 3994 John R. St., Detroit, MI 48201.
and it is taught that the eyes conjugately ro tate around their pupillary axes in a direc tion opposite to a head or body tilt so that the vertical corneal meridia maintain a posi tion perpendicular to the horizon.1"5 This important idea was basic in establishing concepts of the function of the vertically acting extraocular muscles, in explaining the physiology of the vestibular apparatus, and in developing systems of clinical analysis for ocular motor defects.6 However, a sur vey of the literature beginning with John Hunter 7 in 1786 does not reveal any ex perimental evidence or documented clinical observations to prove that compensatory ocular rotation is a phenomenon occurring in man.6-9 Ocular torsion was measured in various positions of head and body tilt in man showing that torsional eye movements cannot fulfill a compensatory function.10"15 The maximum ocular movement or tor sion measured in man was about 11° for large amplitudes of head, neck, and body inclination (more than 90°).10"14 For 20° of head torsion the amount of counterrolling or displacement is less than 3°. The eyes of a person whose head is tilted toward the shoulder apparently roll and lag behind the head about 10% of the tilt.15 Similar measurements of counter-rolling were made in the rhesus monkey.16 Those investigators who measured some ocular displacement or torsion did not demonstrate experimentally that the move ment occurred around the pupillary axis. They did not appear concerned about how this movement took place ; possibly it was due to an ocular rotation around some axis other than the pupillary axis or to displace ment of the whole eye (translation). 17 When the head is tilted the eyes are also displaced inward and downward or upward (Figs. 1 and 2 ) . According to my unpublished ob-
292
VOL. 79, NO. 2
OCULAR TORSION
293
servations, no compensation for these hori to 20° of head torsion. I also attempted to measure counter-rolling of the eyes with zontal and vertical movements takes place. Some investigators including Helmholtz neck and body inclinations over a wide and Müller18 have denied the existence of amplitude of up to 100° (Fig. 3). My tech any ocular torsion or eye niovement with nique is described briefly in the legends head or body tilt.19-20 They point to the com of Figures 3 and 4. A thin strip of egg plexity of the techniques and consider the shell membrane was placed horizontally on the cornea and a plumb line was hung from measurement as artifactuous. When I first strove to measure counter- a trial frame. Motion pictures were taken as rolling of the eyes I permitted the person the head and body inclined. While the head to rotate his head only around an antero- was inclining I noted a slight counter-move posterior axis that emerged through the ment of the eye but when the movement of root of his nose and halfway between his the head stopped the ocular counter-move eyes (Figs. 1 and 2). The maximum head ment was not sustained. No ocular countertorsion I measured was about 20° and I rolling could be measured with the head or prefer to call this "true" head torsion. Head body in a sustained inclined position (Fig. tilting beyond 20° requires movement of 3). Experiments with lightly anesthetized the whole cervical column and the shoulders rhesus monkeys failed to reveal any sus and ought to be termed "head and body in tained compensatory counter-rolling to head clination" rather than head torsion.21 I tilt (Fig. 4). thought that if compensatory counter-rolling Recently, a more sophisticated technique occurred it would most likely be a response than I employed detailed dynamically the re-
Fig. 1 (Jampel). Head torsion has an amplitude of about 20° and takes place in the atlanto-occipital articulation, i.e., around an anteroposterior axis that emerges through the root of the nose and halfway between the eyes. Movements of greater in clination are executed by the whole cervical column. If counter-rolling takes place at all, it is 3° or less for a head torsion of 20°.
294
AMERICAN JOURNAL OF OPHTHALMOLOGY
FEBRUARY, 1975
Fig. 2 (Jampel). Eye displacement with head torsion. When the head is rotated 20° around O (a point on the horizontal plane that bisects the line connecting the centers of rotation of the eyes), the pupillary axes are displaced downward or upward (P to P') about 11 mm and inward about 2 mm. If AB (the vertical corneal meridian) is to re main perpendicular to the horizon then point A would be expected to move to point C as the head twisted 20°. Angle POP' would equal angle A'P'C. In our experiments AB moved to A'B' and minimal or no counter-rolling took place. There is no com pensation for the vertical or inward movements. sponse of the eyes to head and body tilt. 22 The authors found that smooth tilts of the head to the shoulder caused ja slow counterrotation of the eyes followed by a saccadic rotation in the direction of the head tilt. T h e net effect was maintenance of the verti cal corneal meridian more or less parallel to the sagittal plane of the head. Residual torsional displacement to sustained head tilt was considered artifactuous. However, for the purpose of this paper, the existence of some slight ocular torsion or displacement with head torsion and in clination may be granted (artifact or n o t ) and the current question put to rest for the moment because such slight movements a r e of no proven value or importance to the clinician though interesting to the physiolo gist. Also, if slight sustained ocular dis placement or torsion does take place with
head tilt, the question of around which axis it takes place has not yet been answered. However, there is no compensatory ocular rotation or torsion in man in response to head inclination, i.e., the vertical meridian of the cornea is not maintained perpendicu lar to the horizon when the head is inclined. In the light of this fact, I will examine the function of the vertical extraocular muscles: how can the oblique muscles and vertical rectus muscles be significant tortors when no significant ocular torsion takes place in the normal human or nonhuman primate? The
oblique
muscles—PREVALENT
CON
CEPT OF OBLIQUE MUSCLE FUNCTION—Based
on an analysis of the position of its origin and insertion, we have been taught that the superior oblique muscle ( S O ) is mainly an intortor, secondly a depressor, and thirdly an abductor. 3 ' 4 T h e functions of intortion
VOL. 79, NO. 2
295
OCULAR TORSION
and abduction increase with abduction of the eye and the function of depression in creases with adduction of the eye. The in ferior oblique muscle (IO) is mainly an extortor, secondly an elevator, and thirdly an abductor. The functions of extorsion and
abduction increase with abduction of the eye and the function of elevation increases with adduction of the eye. The IO is an antagonist of the SO for torsion and de pression but a synergist for abduction. As the eye abducts owing to the contraction
D
Fig. 3 (Jampel). T o demonstrate that compensatory ocular counter-rolling is not clinically detectable in a normal subject, a white marker (eggshell membrane) is placed on the subject's eye. The vertical white line is a plumb line hanging from a t r h l frame. The eye marker mnintrins its relative position in regard to the lower eyelid in all positions of head inclination. A, head erect ; B, head inclined about 20° ; C, head inclined about 25° ; D, head inclined 50° ; E, head inclined 60°.
296
AMERICAN JOURNAL
OF OPHTHALMOLOGY
FEBRUARY, 1975
pensatory rolling of the eye and is assisted by the inferior rectus muscle. E X P E R I M E N T A L EVALUATION OF OBLIQUE
F U N C T I O N I N T H E MONKEY—Different con
Fig. 4 (Jampel). Simple experiment with a lightly anesthetized monkey to demonstrate the absence of compensatory counter-rolling. The eyes were held open with eyelid retractors. The white markers on the corneas are eggshell membrane. The black marker between the eyes is a strip of black tape. The relative positions of the various markers are held constant in spite of head tilt. Top, head erect; center, head inclined left about 27°; bottom, head inclined right about 17°. of the lateral rectus muscle, simultaneous contraction of the oblicjue muscles causes ad ditional abduction that completes the lateral movement. T h e S O is the main intortor in compensatory rolling of the eye and is as sisted in this function by the superior rectus muscle. The I O is the main extortor in com-
cepts of the function of the S O and the I O were formulated from experimental work in the monkey. 2 3 - 2 5 W e found that the axis of rotation of the oblique muscles is fixed in the orbit and is not displaced when the eye moves in the horizontal plane. T h e S O and the I O share the same rotational axis and are definite antagonists in all horizontal gaze positions. T h e amount of rotation around this axis is always the same for a given con traction of an oblique muscle. A s the eye adducts, the amount of both depression or elevation and lateral displacement increases (Fig. 5 ) . W h e n the S O contracts with the eye located above the horizontal plane, the pupillary axis undergoes an increasing ad duction until it reaches the horizontal plane where it undergoes an increasing abduction. W h e n the I O contracts with the eye located below the horizontal plane, the pupillary axis undergoes an increasing adduction until it reaches the horizontal plane where it under goes an increasing abduction. Experimental co-contraction of the S O and I O does not produce abduction or propulsion of the globe ; the eye remains immobile. Contraction of the S O alone or in combination with the superior rectus muscle does not produce wheel-rotation (intortion) around the pupil lary axis. Contraction of the I O alone or in combination with the inferior rectus muscle does not produce wheel-rotation (extorsion) around the pupillary axis. Function of the superior oblique muscle in man—The principles of oblique muscle action derived from experimentation in monkeys (Macaco mulatta) were confirmed in man by clinical observations and experi ments. Oblique muscle action differs between monkey and man in that the amplitude of horizontal eye movement in man is twice that of the monkey, i.e., about 90° in m a n and 45° in the monkey. PATIENTS
WITH
OCULOMOTOR
NERVE
297
OCULAR TORSION
VOL. 79, NO. 2
an angle of about 60° with the Y-axis of Fick (Fick's axes comprise a rectangular coordinate system that is absolutely fixed in the orbit. The center point of this system is the center of rotation of the eye. The hori zontal axis is denoted X, the anteroposterior axis Y, and the vertical axis Z) (Fig. 6, B). This so-called eccentric rotation is charac teristic of the SO and is clearly seen when the cornea is marked with a vertical strip of eggshell membrane (Fig. 6, C). The rare patient with paralysis of both the oculomotor and abducens nerves and
PARALYSIS—When total paralysis of the oculomotor nerve with intact trochlear and abducens nerves occurs, the levator palpebrae, superior rectus, medial rectus, inferior rectus, and IO muscles are paralyzed. The SO and lateral rectus muscles are function ing. The eye is deviated outward, not down ward and outward as is usually taught (even though the SO is unopposed by the para lyzed IO and superior rectus muscles) (Fig. 6, A ) . When commanded to look down or to pursue a moving object, the eye will rotate downward around a fixed axis that forms
AD
AB A\B
P
P -
-
B A' P' B' Fig. S (Jampel). Demonstrating superior oblique (SO) muscle action in the monkey. The anterior axis pole (C) of the rotational axis of the SO remains fixed in the orbit when the eye moves in the horizontal plane (CP) owing to the activity of the horizontal rectuses. A given contraction of the SO causes the pupil lary axis (P) to transcribe larger and larger parallel arcs of latitude (P to P') as the eye moves from a posi tion of abduction {AB) to a position of adduction {AD). Both the outward displacement {PF) of the pupillary axis ( P ) and the backward displacement {PV) increase as the eye adducts. The angle of ocular rotation {PCP1) is equal to the angle of displacement of the vertical corneal meridian {A'P'F) in all horizontal gaze positions. The lower diagram shows the eye movement projected onto a frontal plane.
298
AMERICAN JOURNAL OF OPHTHALMOLOGY
--A
sparing of the SO has an eye located in the midposition. When commanded to look down, the eye rotates downward around the same axis it rotated around when the eye was in the abducted position (Fig. 6, D ) . The same type of eye movement is common ly seen after the retrobulbar injection of anesthesia into the muscle cone in prepara tion for cataract surgery. In this situation all the extraocular muscles except the SO may be paralyzed since the nerve to the SO is outside of the muscle cone. When commanded to look down, the eye under goes the downward eccentric rotation char acteristic of the SO. In patients with oculomotor nerve paralysis and sparing of the trochlear nerve, a silk suture was carefully sewn through the lateral corneoscleral limbus after topical
FEBRUARY, 1975
Fig. 6 (Jampel). Action of the superior oblique ( S O ) muscle in man is demonstrated in a patient with oculomotor nerve paralysis. ( A ) In oculo motor nerve paralysis the eye deviates outward and not outward and - down ward as we have been taught. ( B ) On command to look down, the eye ro tates around point C from P to P' transcribing an arc of latitude. (C) The cornea is marked with a strip of eggshell mem brane to make the type of movement evident. ( D ) The eye from the straight-ahead position also rotates around point C transcribing an arc of latitude. ( E ) The eye can be passively moved into different positions in the horizontal plane by means of a corneoscleral limbal suture. The eye move ments produced by the SO have the same char acteristics from all hori zontal gaze positions and an estimate of the magni tude of these movements is possible.
anesthesia and the eye was passively ro tated medially in small increments. On com mand to look down, the eye rotated down ward from any position in the horizontal plane around the same fixed eccentric axis (Fig. 6, E ) . No rotation of the eye around the pupillary axis produced by isolated SO contraction has even been observed. Thus, in man and in monkey, a given contraction of the SO causes the eye to transcribe larger and larger arcs of latitude as the eye moves medially away from the fixed SO axis. These downward arcs in crease in magnitude until a point 90° (30° of adduction) from the SO axis is reached at which time the arcs decrease in magni tude for the same SO contraction (Fig. 7). Patients with isolated SO paralysis also provide insight into the function of that
VOL. 79, NO. 2
OCULAR TORSION
muscle. Interpretation of the Lancaster redgreen tests performed on such patients is revealing (Fig. 8 ) . In right SO paralysis with the head erect and the left eye fixing, the right eye is dis placed upwards and outwards when the pa tient is looking in the horizontal plane or downward from all positions of horizontal gaze. When he looks upward there is no displacement. The axis around which the SO rotates the eye may be interpolated by connecting the midpoints of the projections of the fixation lights in adduction, primary position, and abduction. This angular rota
LATERAL
299
tion of the projection of the vertical fixation light to the horizontal plane is the same in adduction, primary position, and abduction, and is about 17°. Function of inferior oblique muscle in man—We have measured the action of an isolated IO in man 23 in a patient with an acquired paralysis of the superior rectus muscle (Figs. 9 and 10). The axis, of rota tion of the human IO was located in the horizontal plane about 60° lateral to the Yaxis of Fick and behaved as if it were fixed in the orbit. The IO rotated the eye about 17° upward from the horizontal plane
MEDIAL
Fig. 7 (Jampel). Diagram of the action of the superior oblique (SO) muscle in man : Arc AMKDLC is the horizontal plane. Arc L to M is the approximate range of hori zontal ocular movement in man. The rotational axis of the SO is line ABC with B the center of rotation of the eye. This axis is located about 60° from line BD which rep resents the primary position and corresponds to the Y-axis of Fick. When the SO con tracts with the eye in any position in the horizontal plane the center of the pupil (P) transcribes an arc of latitude (P to P'). The arc of latitude has three components : down. lateral (PF), and back into the orbit. These components increase in magnitude until the eye rotates 90° (to point K) from the rotational axis (line ABC) and then they decrease. Angle ACP' is equal to angle A'P'F; thus the angle of ocular rotation is equal to the angular displacement of the corneal meridian (lines AB and A'B' in the lower diagram).
300
AMERICAN JOURNAL OF OPHTHALMOLOGY
x :
FEBRUARY, 1975
M i l l i π Τ Ρι' I1 1 1 1 1 Tu
AO_ i
1
i
AB 1 i 1
»
1
1
*
!1 ! 1 ■■
j
.AD
1 î
\
4
i
|
|
*
W
lï M
1 1 l i II M
i-4-»-i-»-n-M-i»-ii-»
; 1 ! ''
'
1
1' I Ì1 1
' '
»
f/
f~~\ ]'}*& ^ / ' 1
irL
"
/ !
I>
AB
> 1 /
/
i
1H i ! — l - D i — I D ■
io-
~j
-Γ -Q* * ! ~V X- \ -
■ 1I D — l - D — I t
1
Y *^ì
1
t
9
:ΐJU^ π ^^^ > ;: H ¥ ±f « : 2Δ i Kv> i7 :
îi
1 !Î i i ii ' J
itJ_
Head tilted to left
Head erect
Head tilted to right
A
B
c
Fig. 8 (Jampel). The Lancaster red-green projection test in isolated superior oblique muscle paralysis. (A) The projections are normal when the head is tilted to the contralateral shoulder. (B) With the head erect, the projections are abnormal in horizontal gaze (not shown) and in downward gaze. In downward gaze, the vertical projections undergo the greatest vertical and lateral displacement (abduction) in the position of adduction (AD). The angular displacements of the vertical projections to the horizontal plane are the same in all gaze positions. (C) The projections are abnormal in all gaze positions when the eye is tilted to the homolateral shoulder (open bar, R.E. ; solid, L.E.).
M
Fig. 9 (Jampel). The eye movements produced in man by the inferior oblique (IO) and superior oblique (SO) muscles projected onto a frontal plane are shown. When the eye (P, the center of the pupil) rotates around the axis pole (A) from the hori zontal plane, an arc of latitude (P to P' or P to P") is produced. The arcs increase in magnitude until the eye moves horizontally to E, which is 90° from A or 30° from the straight-ahead position. The arcs decreased in magnitude with further medial rotation from E. The horizontal displacement (PF) also increases until point E is reached and then decreases ; L, lateral ; M, medial.
VOL. 79, NO. 2
OCULAR TORSION
301
3mm
2 mm
Z
mm
o I< >
UJ
_i UJ
mm
2 mm 3 mm
4 mm
ADduction
Fig. 10 (Jampel). The approximate amount in millimeters of depression produced by the superior oblique (SO) and elevation produced by the inferior oblique (IO) muscles in man projected onto a frontal plane with the head in the normal erect position and the eye in different positions in the horizontal plane. The maximum rotation of the eye (0) is about 17° around the axis of the oblique muscles. Note that the maximum elevation or depression occurs when the eye is adducted about 30°. (The SO measure ments were obtained from the experiment in Fig. 6, E.)
around its axis (Figs. 9 and 10). As the eye moved across the horizontal plane from lateral to medial, because of the contrac tion of the medial rectus muscle and the inhibition of the lateral rectus muscle, con traction of the IO produced the following: the eye moving medially away from its axis transcribed larger and larger arcs of latitude (each arc represented 17° of ocular rotation around the IO axis) until the eye was 90° from the IO axis and in a position of 30° of adduction, and the arcs then became smaller. These arcs were composed of an elevating and lateral component. There was increasing elevation and lateral displacement of the eye until 30° of adduction was reached where the elevation and lateral dis placement began to decrease.
DISCUSSION
The eyes in man and monkey in the physiologically integrated state either do not undergo conjugate counter-rolling at all or do to a minimal degree that is wholly in sufficient to compensate for head tilt or body displacement or both. Even investi gators who have demonstrated some slight ocular torsion have not shown that this movement represents a wheel-like rotation around the pupillary axis. This movement could take place around an axis other than the pupillary axis or be the result of translatory movement of the eye. The possibility that the movements are artifacts has not been completely eliminated. With our pres ent limited sophistication in the clinical analysis of ocular motor defects, we con-
302
AMERICAN JOURNAL OF OPHTHALMOLOGY
sider that no compensatory rotation or dis placement of the eyes takes place. During extensive experiments with mon keys, I observed the actions of individual and various combinations of extraocular muscles and produced no wheel-like rotation of the eyes around the pupillary axis.23"25 Stimulation of the ocular motor nuclei in the brainstem, the ampullary nerves of the labyrinths, and the frontal eye fields never resulted in wheel-like ocular rotation.24 For example, simultaneous stimulation of the SO and superior rectus muscles never produced intortion of the eyes and simultaneous stim ulation of the IO and inferior rectus muscles never produced extorsion of the eyes. The mechanic and neurologic mecha nisms for wheel-rotation of the eyes around the pupillary axis apparently do not exist. When an oblique muscle acts alone ex perimentally—a situation which never occurs normally—it produces a rotary movement of the eye around a fixed eccentric axis that forms an angle in man of about 60° with the Y-axis of Fick. This axis remains fixed in the orbit regardless of horizontal eye movements around the Z-axis of Fick. As a result of the location of this axis contraction of an oblique muscle produces an angular rotation of the eye equal to the angular displacement of the vertical corneal merid ian when projected onto the frontal plane (Figs. 5 and 7). The oblique muscles in ex periments with monkeys never produce a wheel-rotation of the eye around the pupil lary axis when acting alone or in combina tion with the vertical rectuses. There is no compensatory conjugate counter-rolling of the eyes to head tilt in order to maintain the retinal horizon parallel to the terrestrial horizon. Also, the oblique muscles appear incapable of producing wheel-like rotation of the eyes, or any sig nificant ocular torsion for that matter, in the normal state. Thus, the neurologic mechanisms that govern the vertical extraocular muscles and the structure of these muscles appear de
FEBRUARY, 1975
signed to inhibit or prevent ocular rolling or torsion and to maintain the eyes in a stable position relative to that of the head, i.e., to keep the vertical corneal meridian as parallel as possible to the sagittal plane of the head. When the head or body or both are inclined, the vertical muscles acting to gether hold the eyes in a stable position. Any residual torsional movements that may exist might be vestigial. Having maintained stable eyes, the vertical extraocular muscles, which in the normal state are neither sig nificant tortors nor horizontal ductors, work in concert to produce elevation or depression of the eyes with the head or body in any position or posture. When the head or body is tilted, the verti cal corneal meridia are held parallel to the sagittal plane of the head. Therefore, the in sertions of the extraocular muscles are dis placed equal to the angle of head tilt from their original position. The insertions of the horizontal rectus muscles, for example, are no longer parallel to the horizon and yet the eye with the head tilted is capable of a smooth pursuit or saccadic movement in the terrestrial horizontal plane. Such a movement is difficult to explain employing what is al ready known about ocular mechanics. Pos sibly an innervational shift or redistribution among the vertical and horizontal extraocular muscles produced by head tilt or body dis placement enables such an eye movement to take place. The head-erect position sends the innervations for elevation and depres sion of the eyes to both the vertical rectus and oblique muscles although not necessarily equally. When the head is tilted, innervation may be shifted to the vertical rectus muscles in the contralateral eye and to the obliques in the homolateral eye for the production of ele vation and depression. This is suggested by some cases of superior oblique muscle paralysis in man (Fig. 8). The otolith organs of the vestibular apparatus may be responsi ble for this innervational shift. Electromyographic studies have shown increasing electrical activity in the SO and
VOL. 79, NO. 2
303
OCULAR TORSION
superior rectus muscles and decreasing elec trical activity in the IO and inferior rectus muscles in the homolateral eye during head tilt in man and the reverse in the contralateral eye.26 These findings might be em ployed as evidence for ocular torsion. How ever, these electrical activities probably represent isometric muscle contractions that stabilize the eyes in relation to the head and prevent both torsion and vertical ocular dis placement. Measuring the innervation in the vertical extraocular muscles when the subject looks up and down from all gaze positions with the head tilted would be an important experiment. Torsional displacement of the vertical corneal meridian takes place only in patho logic conditions (not to be confused with so-called pseudotorsion seen in normal oblique muscle gazes) such as extraocular muscle paralysis or nystagmus and in certain abnormal phorias and tropias, but this dis placement has never been wheel-like, i.e., it does not take place around the pupillary axis. Spatial orientation and the recognition of an "objective" vertical when the head tilts or when the body is displaced is maintained in spite of the absence of compensatory eye movements. The mechanisms responsible for this phenomenon reside in the brain or in an interplay between the retina and the brain and not in the peripheral ocular motor apparatus. Recent experiments indicate that cyclofusion can take place without ocular movement.27 Wheel-like rotational eye movements (cycloduction) around the pu pillary axis do not take place and torsional displacement of the vertical corneal merid ian is always associated with a vertical ocular displacement. Tilting of a perceived vertical straight edge of 40° was produced in cer tain psychologic experiments that involved rotating displays around the line of sight, without significant ocular rotation or dis placement.23 Thus, visual sensory adapta tions to body and head inclinations are not dependent on compensatory eye movements
and certainly not an ocular torsion around the pupillary axis. SUMMARY
The vertical corneal meridia are not kept perpendicular to the horizon in human and nonhuman primates when the head or body is tilted, i.e., compensatory counter-rolling of the eyes does not occur. The slight tor sional displacement of the vertical corneal meridia noted by many observers may be the result of rotation around an axis other than the pupillary axis or to translation of the globe. The neurologic and structural systems that control the actions of the vertical muscles in human and nonhuman primates do not appear to provide a mechanism for wheel-rotation of the eyes around the pu pillary axis. Ocular torsion is not a normal function of the vertical extraocular muscles. Their function is probably the reverse, i.e., the inhibition or prevention of ocular torsion and the stabilization of the eyes when the head or body inclines. Torsional displace ment of a vertical corneal meridian occurs only when there is an abnormal muscle im balance. Wheel-like movements (cycloduction) around the pupillary axis or visual line do not occur. Torsional displacement of a verti cal corneal meridian occurs only with a simultaneous vertical movement. The vertical rectus and the oblique muscles in man work together to produce vertical ocular movements regardless of head position or body posture while main taining the vertical corneal meridia parallel to the sagittal plane of the head. The vestibular apparatus may be responsible for dis tributing innervations among these muscles, enabling them to function in this manner. REFERENCES
1. Walls, G. L. : The evolutionary history of eye movements. Vision Res. 2:69, 1962. 2. Adler, F. H. : Physiologic factors in the dif ferential diagnosis of paralysis of the superior
304
AMERICAN JOURNAL OF OPHTHALMOLOGY
rectus and superior oblique muscles. Arch. Ophthalmol. 36:661, 1946. 3. Moses, R. A. : Adler's Physiology of the Eye. St. Louis, C. V. Mosby, 1970, pp. 81, 187, 200, and 236. 4. Lyle, T., and Bridgeman, "G. : Worth and Chavasse's Squint. London, Baillere, Turdall and Cox, 1959, p. 236. 5. Marquez, M. : Supposed torsion of the eye around the visual axis in oblique directions of gaze. Arch. Ophthalmol. 41:704, 1949. 6. Parks, M. M. : Isolated cyclovertical muscle palsy. Arch. Ophthalmol. 60:1027, 1958. 7. Hunter, cited by Palmer, J. F. (ed.) : The Works of John Hunter, F.R.S., with Notes. Lon don, Longman, Rees, Orme, Brown, Green, and Longman, 1837. 8. Lancaster, W. B. : Terminology in ocular motility and allied subjects. Am. T. Ophthalmol. 26: 122, 1943. 9. Stevens, G. T. : Motor Apparatus of the Eyes. Philadelphia, F. A. Davis Company, 1906, p. 123. 10. Duke-Elder, S. : Textbook of Ophthalmology. St. Louis, C. V. Mosby, vol. 1, 1939, pp. 623628. 11. Woellner, R. C, and Graybiel, A.: Counterrolling of the eyes and its dependence on the mag nitude of gravitational or inertial forces acting laterally on the body. J. Appi. Physiol. 14:632, 1959. 12. Belcher, S. J. : Ocular torsion. Br. J. Physiol. Opt. 2:1, 1964. 13. Miller, E. F. : Counterrolling of the human eyes produced by head tilt with respect to gravity. Acta Otolaryngol. 54:479, 1962. 14. Linwong, M., and Herman, S. : Cycloduction of the eyes with head tilt. Arch. Ophthalmol. 85 : 570, 1971.
FEBRUARY, 1975
15. Davson, H. : The Physiology of the Eye. London, Churchill, 1963. 16. Krejcova, H., Highstein, S., and Cohen, B. : Labyrinthine and extralabyrinthine effects on ocu lar counter-rolling. Acta Otolaryngol. 72:165, 1971. 17. Levine, M. : Pendulum-like eye movement. Am. J. Ophthalmol. 75:979, 1973. 18. Southall, J. : Helmholtz's Treatise on Physio logical Optics, vol. 3. New York, Dover Publica tions, Inc., 1962, pp. 120 and 139. 19. Hewitt, R. S. : Torsional eye movements. Am. J. Ophthalmol. 34:253, 1951. 20. Levine, M. : Evaluation of the Bielschowsky head-tilt test. Arch. Ophthalmol. 82:433, 1969. 21. Rouviere, H. : Anatomie Humaine. Descrip tive et Topographique. Paris, Masson et Cie, 1959, p. 132. 22. Petrov, A., and Zenkin, G. M. : Torsional eye movements and constancy of the visual field. Vision Res. 13 :2465, 1973. 23. Jampel, R. S. : The fundamental principle of the action of the oblique ocular muscles. Am. J. Ophthalmol. 69:623, 1970. 24. : The action of the superior oblique muscle. Arch. Ophthalmol. 75 :535, 1966. 25. : Experimental evaluation of the ocu lar rolling (torsion) in the Macaque. Proceedings of the Annual Conference on Engineering in Medi cine and Biology, 9:7.6, 1967. 26. Scott, A. B. : Extraocular muscles and head tilting. Arch. Ophthalmol. 78:397, 1967. 27. Kertesz, A. E. : The effect of stimulus com plexity on human cyclofusional response. Vision Res. 12:699, 1972. 28. Dichgans, J., and Held, R. : Moving visual scenes influence the apparent direction of gravity. Science 178:1217, 1972.
