Abstract
purpose. To analyze the path of extraocular muscles (EOMs) quantitatively in
highly myopic subjects with and without restricted eye motility versus
control. To elucidate the cause of the acquired motility disorder in
patients with high myopia.
methods. Thirty-three orbits were imaged using a Magnetom or Siemens Vision
(Siemens, Erlangen, Germany; both 1.5 Tesla) MRI (magnetic resonance
imaging) scanner. Coronal T1-weighted, spin-echo images
were obtained with repetition time of 550 msec and echo time of 15
msec. Subjects had to fixate in different positions of gaze. Orbits of
three patient groups were analyzed: group 1 (n = 14),
patients with high axial myopia and restricted eye motility (average
axial length, 31.4 mm; refractive error more than −15 D); group 2
(n = 8), subjects with high axial myopia and normal eye
motility (average axial length, 29.2 mm); control group (n= 11), emmetropic subjects with normal eye motility (average
axial length, 23.6 mm).
results. Highly myopic patients showed significant displacements of recti EOMs
in comparison with control subjects. Mean displacements, measured in
the plane 3 mm anterior to the globe–optic nerve junction in primary
gaze, were in group 1, lateral rectus (LR) 2.9 mm (2.5 downward, 1.4
medial), medial rectus (MR) 1.3 mm downward and in group 2, LR 1.4 mm
(1.3 downward, 0.6 medial) and MR 1.2 mm downward. In groups 1 and 2
the inferior rectus (IR) was displaced 1.3 mm medially and upward. In
both groups of myopic patients the superior rectus (SR) was displaced
1.5 mm medially and downward.
conclusions. In patients with high axial myopia, displacements of all recti EOMs can
be detected by MRI. Displacements of SR, MR, and IR were very similar
in groups 1 and 2 versus control. LR displacement into the lateral and
inferior quadrant of the orbit was greatest in patients with restricted
eye motility. Thus, LR displacement is probably the major
pathophysiological factor for the restrictive motility disorder in high
myopia. EOM dislocations can be explained by myopia-associated
alterations in the orbital connective tissues confining EOM positions
in relation to the orbital wall.
Adult patients with unilateral or bilateral high (axial) myopia
may acquire a typical restrictive motility disorder, resulting in
esotropia and often hypotropia. Many different theories on the
underlying cause of this restrictive disease can be found in the
literature. Duke–Elder and Wybar
1 suggested structural
changes in the ocular muscles—that is, a reduced number of muscular
fibers included in fibrous tissue in the lateral rectus (LR). In some
textbooks
2 3 the deviation, especially the hypotropia, is
characterized by the name heavy eye syndrome. Bagolini et
al.
4 inferred from echography, myopathic paralysis of the
LR due to pressure from the lateral orbital wall. Demer and von
Noorden
5 found that rotation was limited due to contact
between the posterior aspects of the elongated globe and the bones of
the orbital apices. Herzau and Ioannakis
6 observed during
surgery an abnormal LR path but could not confirm this finding before
surgery by computed tomography or magnetic resonance imaging (MRI).
More recent theories concerning restrictive strabismus take into
account new aspects of the functional anatomy of the
orbit.
7 Thus, it has been shown by Demer et
al.
7 that the EOMs pass through connective tissue sleeves
in the posterior Tenon’s fascia that constrain muscle paths during
gaze shifts and act as functional origins of the muscles. These
tissues, termed pulleys, structurally and functionally include many of
the orbital connective tissues known as check ligaments, Lockwood’s
ligament, intermuscular membranes, or orbital septa.
7 8 9 10 Pulleys are composed of collagen, elastin, and smooth muscle and are
connected to the orbital bones and to one another by connective tissue
bands.
7 In normal orbits, location of the pulleys is
highly uniform. However, displacements of normal pulley positions of
some millimeters have been found in cases with incomitant (A- or
V-pattern) strabismus.
11
Recently, Krzizok et al.
12 described a significant
inferior dislocation of the lateral rectus (LR) based on MRI scans. Our
purpose was to analyze the path of all recti extraocular muscles (EOMs)
quantitatively in highly myopic subjects with and without strabismus
versus control by means of fixation-controlled MRI. We were looking for
a specific pattern of myopia-associated EOM displacement related to
restricted motility. Results of our study should not only contribute to
a better understanding of the motility disorder associated with high
myopia but should also to provide further anatomic data for improvement
of strabismus surgery in these patients.
We examined prospectively 22 orbits of 15 patients with
unilateral or bilateral high axial myopia and 11 orbits of 8 emmetropic
volunteers without strabismus. All subjects had given their informed
consent according to a protocol approved by the institutional review
board for the protection of human subjects (Declaration of Helsinki).
Routine ophthalmologic and orthoptic examinations and A- and B-scan
echography were performed in all subjects to determine axial length of
the globes (Ocuscan; Alcon, Fort Worth, TX,). A simultaneous and
alternate prism cover test at distances of 5 m and 0.33 m, in
25° up-, down-, and side gaze was performed in all patients with
central (foveolar) fixation, before surgery and after surgery at 1 week
and at least 3 months. The range of ocular rotation was measured with a
synoptometer. Forced duction tests were performed during surgery to
determine horizontal, vertical, and torsional motility.
Highly myopic patients unable to fixate in primary gaze because
of very limited abduction or elevation had to be excluded from this
analysis. Subjects were divided into three groups according to axial
length of the globes and degree of motility restriction (clinical
characteristics are described in
Table 1 ): control group (11 orbits): emmetropic subjects (spherical
shaped globes with axial length of 21–24 mm) and normal eye motility
(abduction ≥40° and elevation ≥30°); group 1 (14 orbits): highly
myopic subjects (axial length >27 mm) with limited abduction (<40°)
and elevation (<30°); and group 2 (8 orbits): highly myopic subjects
(axial length >27 mm) without restricted motility (abduction ≥40°
and elevation ≥30°).
The position of the centroid of the globe–optic nerve
junction in coronal MRI scans (plane 0 in
Fig. 1 ) was used to determine
the exact eye position (e.g., primary gaze, upgaze, downgaze,
abduction, and adduction). Tertiary positions were not necessary for
this study. The geometric analysis of contiguous MRI scans was
performed according to the description of Miller,
13 except
that coronal planes were not orthogonal (perpendicular) to the orbital
axis—that is, 26° tilted to the coronal plane—but were
perpendicular to the visual axis of each eye in primary gaze.
The original MRI films were scanned (Scanmaker E-6; Mikrotek,
Hsinchu, Taiwan) in tagged-image file format and analyzed
quantitatively using NIH Image with personal computers (Macintosh;
Apple, Cupertino, CA) or the same program, named PC-image, with
IBM-compatible personal computers and Windows 95 (Microsoft, Redmond,
WA). Both software programs were written by Wayne Rasband (National
Institutes of Health) and are available as public domain programs on
the Internet (anonymous - ftp//:zippy.nimh.nih.gov). Images of left
orbits were reflected digitally (flipping horizontally) to the
orientation of a right orbit. Each image was calibrated according to
the reference scale on the MRI scans: 10 mm equaled 38 to 55 pixels.
Positional errors during the MRI examination were accounted for.
Rotation in the coronal plane (head tilt) was checked and corrected by
rotating the image to align the interhemispheric fissure of the brain
with the scanner-defined vertical meridian.
A Cartesian coordinate system was used in the orbit. The
z-axis consisted of a line, connecting the geometric centers
of all coronal planes. After outlining the bony orbit of each scan with
a trackball, the geometric middle, that is, the centroid, was
determined (function XY Center of NIH Image software). This centroid is
the reference or zero point for all measured coordinates of EOMs
(Fig. 2) . The plane containing the globe–optic nerve junction was designated
plane 0. More anteriorly located planes have a negative prefix
(Fig. 1) . If the MRI scans did not fit exactly with these planes, data for
plane 0, −1, and so on, were found by interpolation. The position of
recti EOMs was determined by their centroid in this coordinate system.
A simulation with the computer model Orbit 1.6 of Miller and
Shamaeva
14 was used to examine the functional results of
the different morphologic findings.
Recti EOMs had a regular, nearly symmetrical location in relation
to the geometric center of the bony orbit. The medial rectus (MR) and
LR were insignificantly shifted below the horizontal meridian, and the
inferior rectus (IR) and superior rectus (SR) were insignificantly
located medially to the vertical meridian. The entrance of the optic
nerve into the globe in primary gaze was 2.9 mm medial and 1.3 mm
superior to the orbital center (
Table 2 and
Fig. 2 ).
Figure 3 describes the muscle path of the significantly displaced LR of a
subject of group 1 in contiguous MRI scans from the posterior to the
middle right orbit. The measured inferior displacement of LR relative
to the MR (difference of
y-value of MR to LR) in different
anteroposterior positions of the orbit constantly decreased from the
anterior MRI scan in
Figure 3 to the posterior MRI scan. This was
primarily because of the displaced curved path of the LR, which shifted
from the normal insertion site into the lateral and inferior quadrant
of the orbit and proceeded to the origin and secondly according to the
conical path of recti EOMs toward the apex of the orbit.
Recti EOMs, the optic nerve, the entrance of the optic nerve into the
globe, the superior oblique muscle, the levator palpebrae superioris
and the lateral levator aponeuris can be reliably imaged in the orbit
by means of MRI. Even with the highest quality orbital MRI, it is not
yet consistently possible to image the pulleys of recti EOMs directly.
This applies to the medium-to-lower–resolution images used in our
study. Unfortunately, the most anterior portions of the recti EOMs
consist of tendons that are difficult to distinguish in MRI scans from
intermuscular connective tissue and sclera. Moreover, the localization
of the LR in the anterior orbit is made difficult by the path and
insertion of the inferior oblique muscle.
In a previous study of 13 highly myopic patients with
restriction-induced esodeviations or hypodeviations, Krzizok et
al.
12 found an inferior dislocation of the LR relative to
the MR in the midorbital region, with a median value of 3.4 mm. The
study did not include calculation of the centroids of the recti EOMs in
relation to a reference point. In the present study, relative vertical
positions of the horizontal recti EOMs were taken into account. We
found a significant relative inferior dislocation of the LR in relation
to the MR of 1.9 mm only in group 1 (
Table 2 ; mean value of
y −1.5 [MR] versus −3.4 [LR]). No significant inferior
dislocation of LR relative to MR was present in group 2 (0.8 mm [mean
value of
y −1.4 [MR] versus −2.2 [LR]), and in the
control group (0.7 mm [mean value of
y −0.2 [MR] versus−
0.9 [LR]). This is consistent with an earlier finding (Krzizok et
al.
12 ) of inferior dislocation of LR of 3.4 mm in myopes
with very restricted abduction and elevation, equivalent to group 1 in
the current study.
Fixation-controlled MRI scans showed significant midorbital
displacement of all recti EOMs in high myopia (groups 1 and 2).
Surprisingly, the displacement of SR and IR medially and the
displacement of MR inferiorly was detectable in the same amount in high
myopes with (group 1) and without (group 2) restricted motility. The
displacement of the superior and the inferior rectus muscles in high
myopes without restricted motility is likely the result of the larger
globe diameter in myopic patients. Because this displacement is
symmetrical and small, no clinically detectable strabismus is
associated.
However, inferior displacement of LR was significantly greater in high
myopes with restricted abduction and elevation (group 1) compared with
group 2. Thus, LR displacement can be considered to be the major
pathophysiological factor for the restrictive motility disorder in high
myopia. Computer simulation of group 1 patients by using the Orbit 1.6
Gaze Mechanics Simulation program
14 is consistent with the
main clinical deviations: esotropia, hypotropia, and excyclorotation.
LR displacement explains the acquired strabismus: the abducting
function of the LR is reduced, and an unphysiological depressing and
excyclorotating force is created.
The pathophysiology can be explained by stretching in the suspensory
tissues that regulate the position of the LR pulley,
7 permitting the pulley to slip inferomedially, and the globe to herniate
superotemporally through the suspensory tissues that ordinarily confine
the globe to the center of the orbit. Between SR and LR, this
superotemporal tissue forms the lateral levator aponeurosis, consisting
of dense collagen, elastin, and smooth muscle.
7 However,
the histopathologic changes and the trigger for them occurring in this
particular anatomic site are not yet known. Demer et al.
7 found in histologic studies that the MR pulley contains the most
fibroelastic and smooth muscle tissue, resulting in the highest
rigidity compared with the other recti EOMs.
If the inferior displacement of the LR is due to a weakening of the
suspensory tissues of the LR pulley, then it would be expected that LR
dislocation might be greatest when the LR contracts and increases its
tension—-that is, in abduction of the myopic globe. This is what we
found in the simulation using Orbit 1.6. Clark et al.
15 found in normal subjects a displacement of the LR superiorly 1 mm
during abduction and explained this with the abducting action of the
SR, which under increased tension in abduction may pull superiorly on
the LR pulley through the lateral levator aponeurosis. Unfortunately,
it is not possible to implement abnormal globe shape or a dehiscence of
the lateral levator aponeurosis into computer simulations of highly
myopic patients with the Orbit 1.6 Gaze Mechanics Simulation
program.
14 Thus, more fixation-controlled MRI scans in
adduction and abduction must be analyzed.
Dislocation of recti EOMs in high myopia without restricted
motility (group 2) has not yet been described. Because horizontal and
vertical deviations in group 2 were small, the displacements of SR, MR,
and IR were probably not a secondary adaptation to the misalignment in
adduction and hypotropia. On the contrary, they were probably the
result of abnormalities of orbital connective tissues (intermuscular
membranes, lateral levator aponeurosis, pulleys, and Lockwood’s
ligament
8 9 10 ) in high myopia, resulting in pathologic EOM
paths of different extents. Other than Duane’s syndrome, high myopia
is the only nontraumatic disease with evident sideslip of
EOMs.
16
To have reproducible and comparable MRI scans, it is important to
perform measurements in a defined (coronal) plane. A reliable reference
point is the center of the bony orbit, but not the mobile optic nerve,
which may lead to erroneous measurements. In high myopes, the optic
nerve sometimes assumes a variable S shape near the orbital apex. Thus,
the location of the optic nerve, even in constant gaze positions,
depends on the scan plane. To determine and control the gaze position
in the MRI scan procedure by means of the optic nerve position, the
same coronal scan plane must be chosen. Most suitable for this purpose
is scan plane 0, the globe–optic nerve junction. Although the mobile
optic nerve is not a reliable reference point for measurements, the
optic nerve in scan plane 0 may be helpful to determine the actual gaze
position. It is important to position the head properly when
quantitatively measuring positions of the EOMs.
Data for the path of recti EOMs of emmetropic subjects without
strabismus have been determined by means of CT and MRI by
Miller,
13 Clark et al.,
15 and Miller and
Robinson.
17 In those publications coronal planes were
orthogonal to the orbital axis, 26° tilted to a classic radiologic
coronal plane
(Fig. 1) . We preferred exact coronal scan planes, 116°
off the orbital axis. This procedure has two advantages: Both orbits
can be scanned simultaneously; if coronal scan planes are perpendicular
to the orbital axis, this means 26° abduction for the gaze position
defined as primary gaze in the study of Clark et al.,
15 necessitating that the patient abduct 52° to reach the gaze position
abduction. The disadvantage of our method is that the cross section of
the LR is oblique to the imaging plane.
The choice of scan plane produces different data for normal subjects
(Table 3) , because of the projection. Thus, in our study, in plane −1
the LR was measured 0.5 mm more laterally, MR 1.0 mm more medially, IR
2.1 mm and SR 3.4 mm more medially in comparison with the measurements
of Clark et al.
15 Their study and ours have the consistent
finding that recti EOMS have a stable, invariable path in the middle
and posterior orbit in normal subjects.
Unfortunately, the most anterior portions of recti EOMs consist of
tendons that are difficult to distinguish in MRI scans from
intermuscular connective tissue and sclera. Thus, recognition of recti
EOMs more anterior than plane −3
(Fig. 1) is not reliable.
Nevertheless,
Figure 3 shows a realistic impression of the LR path.
Additional information about the LR path was available from former
observations made during surgery.
12 Thus, the LR has a
displaced curved path, taking its course from the origin, shifting into
the lateral and inferior quadrant of the orbit and proceeding to the
normal insertion site.
Miller,
13 Clark et al.,
15 Miller and
Robinson,
17 and Simonsz et al.
18 demonstrated
in vivo the stability of position and path of recti EOMs in the middle
and posterior orbit during eye movements, allowing displacements only
in the tendinous anterior part of recti EOMs. The stability of the
recti EOMs’ path is maintained by the pulleys, which encompass the
anatomic structures previously known as check ligaments, and probably
the so-called intermuscular membranes as well.
8 9 10 19 Demer et al.
20 21 described in a patient with Marfan
syndrome deficient fibrillin in recti EOMs’ pulleys, resulting in
instability of the MR pulley in one patient in different gaze
positions.
Thus, in the future it seems it would be very useful to examine those
motility disturbances by means of standardized, high-resolution MRI, in
which the orbital connective tissue is possibly altered. Another
therapeutic step is to find a surgical treatment that can normalize a
dislocated EOM or the pulley of an EOM.
Presented in part at the annual meeting of the Association for
Research in Vision and Ophthalmology, Fort Lauderdale, Florida, May
1998.
Supported in part by Deutsche Forschungsgemeinschaft (KR 853/5-1).
Submitted for publication May 29, 1998; revised December 17, 1998, and May, 25, 1999; accepted July 6, 1999.
Commercial relationships policy: N.
Corresponding author: Thomas H. Krzizok, Department of Strabismus and
Neuroophthalmology, 18 Friedrichstrasse, 35392 Giessen,
Germany. E-mail:
[email protected]
| Number of Orbits | Age (y) | Gender (F/M) | Refraction (D) | Axial Length of Globe (mm) |
Group 1 | 14 | 55 ± 13 | 6/3 | −19.2 ± 5.9 | 31.4 ± 2.0 |
Group 2 | 8 | 46 ± 22 | 3/3 | −17.1 ± 6.7 | 29.2 ± 2.2 |
Control | 11 | 39 ± 18 | 4/4 | +0.25 ± 0.7 | 23.6 ± 0.8 |
Table 2. Position of Recti EOMs in Primary Gaze
Table 2. Position of Recti EOMs in Primary Gaze
| MR | | LR | | SR | | IR | |
| x | y | x | y | x | y | x | y |
Plane 0 | | | | | | | | |
Group 1 | 12.4 ± 0.9 | −1.7 ± 1.0 | −9.5 ± 1.2 | −3.0 ± 1.4 | 3.6 ± 1.1 | 10.7 ± 1.0 | 4.6 ± 1.0 | −9.8 ± 0.8 |
| (0.377)* | (0.049) | (<0.001) | (0.004) | (<0.001) | (0.001) | (0.010) | (<0.001) |
Group 2 | 12.6 ± 0.6 | −1.6 ± 1.8 | −10.3 ± 0.5 | −2.5 ± 1.9 | 3.8 ± 0.8 | 10.6 ± 1.4 | 4.9 ± 0.8 | −10.1 ± 1.3 |
| (0.362)* | (0.055) | (0.018) | (0.076) | (0.001) | (0.015) | (0.006) | (0.003) |
Control | 12.5 ± 0.6 | −0.4 ± 1.3 | −10.9 ± 0.6 | −1.4 ± 1.3 | 1.9 ± 1.3 | 11.8 ± 0.8 | 3.5 ± 1.2 | −11.6 ± 0.8 |
Plane−1 | | | | | | | | |
Group 1 | 13.1 ± 0.9 | −1.5 ± 0.8 | −10.8 ± 1.6 | −3.4 ± 1.3 | 3.3 ± 1.3 | 11.9 ± 0.9 | 5.0 ± 1.1 | −11.0 ± 0.8 |
| (0.432) | (0.014) | (0.011) | (<0.001) | (0.018) | (0.019) | (0.060) | (0.038) |
Group 2 | 13.2 ± 0.5 | −1.4 ± 1.7 | −11.6 ± 0.7 | −2.2 ± 1.5 | 3.3 ± 1.1 | 12.3 ± 0.9 | 5.0 ± 0.6 | −11.2 ± 1.0 |
| (0.440) | (0.067) | (0.097) | (0.071) | (0.033) | (0.176) | (0.055) | (0.173) |
Control | 13.1 ± 0.8 | −0.2 ± 1.6 | −12.2 ± 0.8 | −0.9 ± 1.4 | 2.0 ± 1.6 | 12.7 ± 0.9 | 3.8 ± 2.2 | −11.6 ± 0.8 |
Table 3. Comparison of Recti EOMs’ Position in Primary Gaze and Plane −1 in
Normal Subjects in the Study of Clark et al. and in the Current Study
Table 3. Comparison of Recti EOMs’ Position in Primary Gaze and Plane −1 in
Normal Subjects in the Study of Clark et al. and in the Current Study
| MR | | LR | | SR | | IR | |
| x | y | x | y | x | y | x | y |
Clark et al., 15 primary gaze* | 12.1 ± 0.4 | 0.1 ± 0.7 | −11.7 ± 0.3 | −0.8 ± 0.4 | −1.4 ± 0.3 | 12.3 ± 0.5 | 1.7 ± 0.6 | −12.3 ± 0.5 |
Clark et al., adduction | 11.8 ± 0.4 | 0.1 ± 0.2 | −12.2 ± 0.3 | −1.3 ± 0.5 | −1.0 ± 0.5 | 12.4 ± 0.4 | 1.5 ± 0.5 | −12.4 ± 0.3 |
Present study, primary gaze | 13.1 ± 0.8 | −0.2 ± 1.6 | −12.2 ± 0.8 | −0.9 ± 1.4 | 2.0 ± 1.6 | 12.7 ± 0.9 | 3.8 ± 2.2 | −11.6 ± 0.8 |
The authors thank Earl A. Palmer, Portland, Oregon, for advice and
useful discussions.
Duke–Elder S, Wybar K. System of Ophthalmology. Ocular Motility and Strabismus. Vol. 6. 1973;607–608. Henry Kimpton London.
Taylor R, Whale K, Raines M. The heavy eye phenomenon: orthoptic and ophthalmic characteristics. Ger J Ophthalmol
. 1995;4:252–255.
[PubMed]Basghaw J. The “heavy eye” phenomenon: a preliminary report. Br J Ophthalmol. 1966;23:73–78.
Bagolini B, Tamburrelli C, Dickmann A, Colosimo C. Convergent strabismus fixus in high myopic patients. Doc Ophthalmol
. 1990;74:309–320.
[CrossRef] [PubMed]Demer JL, von Noorden JK. High myopia as an unusual cause of restrictive motility disturbance. Surv Ophthalmol
. 1989;33:281–284.
[CrossRef] [PubMed]Herzau V, Ioannakis K. Zur Pathogenese der Konvergenz und Hypotropie bei hoher Myopie. Klin Monatsbl Augenheilkd
. 1996;208:33–36.
[CrossRef] [PubMed]Demer JL, Miller JM, Poukens V, Vinters HV, Glasgow BJ. Evidence for fibromuscular pulleys of the recti extraocular muscles. Invest Ophthalmol Vis Sci
. 1995;36:1125–1136.
[PubMed]Bron AJ, Tripathi RC, Tripathi BJ. The extraocular muscles and ocular movements. Bron AJ Tripathi RC Tripathi BJ eds. Wolff’s Anatomy of the Eye and Orbit. 1997; 8th ed. 147–152. Chapman and Hall London.
Lockwood CB. The anatomy of the muscles, ligaments and fasciae of the orbit. J Anat Physiol. 1886;20:1–25.
Lang Joh. Praktische Anatomie. Begr. von T. von Lanz W. Wachsmuth Band 1 Kopf Teil B Gehirn- und Augenschädel von Lang Joh in Zsarb mit Bushe KA Buschmann W & Linnert D eds. Lehr- und Hilfsbuch der anatomischen Grundlagen ärztlichen Handelns. 1980; Springer–Verlag Berlin.
Clark RA, Miller JM, Rosenbaum AL, Demer JL. Heterotopic muscle pulleys or oblique muscle dysfunction ?. J Am Acad Pediatr Ophthalmol Strabismus
. 1998;2:17–25.
[CrossRef] Krzizok Th, Kaufmann H, Traupe H. Elucidation of the restrictive motility disorder in high myopia by MRI. Arch Ophthalmol
. 1997;115:1019–1027.
[CrossRef] [PubMed]Miller JM. Functional anatomy of normal human rectus muscles. Vision Res
. 1989;29:223–240.
[CrossRef] [PubMed]Miller JM, Shamaeva I. Orbit 1.6 Gaze Mechanics Simulation. 1995; Eidactics San Francisco.
Clark RA, Miller JM, Demer JL. Location and stability of rectus muscle pulleys. Invest Ophthalmol Vis. Sci.
. 1997;38:227–240.
[PubMed]Miller JM, Demer JL, Rosenbaum AL. Two mechanisms of up-shoots and down-shoots in Duane’s syndrome revealed by a new magnetic resonance imaging (MRI) technique. Campos EC eds. Strabismus and Ocular Motility Disorders. 1990;229–234. Macmillan Press London.
Miller JM, Robinson DA. Extraocular muscle sideslip and orbital geometry in monkeys. Vision Res
. 1987;27:381–392.
[CrossRef] [PubMed]Simonsz HJ, Harting F, de Waal BJ, Verbeeten BWJM. Sideways displacement and curved path of recti eye muscles. Arch Ophthalmol
. 1985;103:124–128.
[CrossRef] [PubMed]Koorneef L. New insights in the human orbital connective tissue. Arch Ophthalmol. 1977;96:1269.
Demer JL, Poukens V, Clark RA, Miller JM, Porter JD. Molecular mechanism of strabismus in Marfan syndrome: deficient fibrillin destabilizes extraocular muscle pulleys (abstract). Proceedings of the 24th annual meeting of the American Academy for Pediatric Ophthalmology and Strabismus. April, 1998; Palm Springs, California.
Demer JL, Miller JM, Poukens V. Surgical implications of the recti extraocular muscle pulleys. J Pediatr Ophthalmol Strabismus
. 1996;33:208–218.
[PubMed]