Journal of Pediatric Ophthalmology and Strabismus

Extraocular Muscle Surgery in Early Infancy - Anatomical Factors

Kenneth C Swan, MD; John H Wilkins, MD

Abstract

Recognition that good alignment of the eyes in infancy is essential for development of normal binocular vision has led to increasingly early surgery to correct strabismus. In a recent study1 of congenital esotropia, 5 of 23 patients had had surgery between four and six months of age. In another report,2 20 of 106 patients with congenital esotropia had had surgery by six months of age, and an additional 46 had surgery before one year of age. About one third of the patients had vertical as well as horizontal muscle surgery. It is well known that diameters of the human eye are only about 17 mm at birth, and that its rate of growth is relatively rapid in infancy; however, there have been few references to the anatomy of the extraocular muscles of the neonatal eye and the effect of growth upon their insertional sites during the first year of Ufe. It is our purpose to present additional information about these anatomical factors which might affect the predictability and safety of extraocular muscle surgery on the infant eye.

Materials and Methods

The topographical anatomy and dimensions of 26 eyes from 14 infants, obtained post-mortem, were studied and photographed with special attention to the insertions of the extraocular muscles. The specimens, obtained from the Eye Bank of the Oregon Health Sciences University and the Oregon Lions Eye Bank, excluded abnormal eyes and those of prematurely born infants. Five pairs of eyes were obtained from full-term neonatal infants who died within two days after birth. Other specimens were obtained at 2, 3, 5Vz, 6, 9 and 20 months. The specimens were fixed in 10% formalin within a few hours after death. Linear dimensions of the globe, the widths of the insertions of the rectus muscle and their distances from clear cornea and the equator were measured with calipers. Measurements of the oblique insertions also included their distances from the optic nerve, clear cornea and from each other.

The volume of each globe was determined by measuring fluid displacement in a calibrated vial. The sclera of one eye of each pair was dissected free of other tissues and the total scleral weight determined. Eight-mm wide buttons then were trephined from all parts of the shell and weighed. From the weight of the sclera, the weight and surface area of the buttons, it was possible to calculate the total surface of the sclera in representative eyes from various age groups. Weights of the scleral buttons also made it possible to indirectly compare the relative thickness of the sclera in corresponding areas of specimens of varying ages. The second eye of each pair was prepared for histologic study. Twelve additional histologic preparations of normal infant eyes in our John E. Weeks Eye Pathology Laboratory also were studied and measured.

Results

The overall dimensions of our five pairs of neonatal eyes were similar to those reported by other authors.35 The sagittal, vertical and transverse diameters averaged slightly less than 17 mm as compared to 24 mm for the average emmetropic adult eye. Measurements of diameters, although traditionally used for comparative studies of eye size, did not reflect the small size of the neonatal eye as well as other parameters and photographs (Figure 1). The average volume (2.8 cc) of our neonatal eyes was only half that of adult eyes (6.8 - 7.5 cc). The average scleral surface (812 mm2) was only one-third that (2450 mm2) of the adult specimens. The average neonatal corneal surface (102 mm2 ) was three fourths that (138 mm2) of the adult specimens.

The most rapid growth of the human eye is known…

Recognition that good alignment of the eyes in infancy is essential for development of normal binocular vision has led to increasingly early surgery to correct strabismus. In a recent study1 of congenital esotropia, 5 of 23 patients had had surgery between four and six months of age. In another report,2 20 of 106 patients with congenital esotropia had had surgery by six months of age, and an additional 46 had surgery before one year of age. About one third of the patients had vertical as well as horizontal muscle surgery. It is well known that diameters of the human eye are only about 17 mm at birth, and that its rate of growth is relatively rapid in infancy; however, there have been few references to the anatomy of the extraocular muscles of the neonatal eye and the effect of growth upon their insertional sites during the first year of Ufe. It is our purpose to present additional information about these anatomical factors which might affect the predictability and safety of extraocular muscle surgery on the infant eye.

Materials and Methods

The topographical anatomy and dimensions of 26 eyes from 14 infants, obtained post-mortem, were studied and photographed with special attention to the insertions of the extraocular muscles. The specimens, obtained from the Eye Bank of the Oregon Health Sciences University and the Oregon Lions Eye Bank, excluded abnormal eyes and those of prematurely born infants. Five pairs of eyes were obtained from full-term neonatal infants who died within two days after birth. Other specimens were obtained at 2, 3, 5Vz, 6, 9 and 20 months. The specimens were fixed in 10% formalin within a few hours after death. Linear dimensions of the globe, the widths of the insertions of the rectus muscle and their distances from clear cornea and the equator were measured with calipers. Measurements of the oblique insertions also included their distances from the optic nerve, clear cornea and from each other.

The volume of each globe was determined by measuring fluid displacement in a calibrated vial. The sclera of one eye of each pair was dissected free of other tissues and the total scleral weight determined. Eight-mm wide buttons then were trephined from all parts of the shell and weighed. From the weight of the sclera, the weight and surface area of the buttons, it was possible to calculate the total surface of the sclera in representative eyes from various age groups. Weights of the scleral buttons also made it possible to indirectly compare the relative thickness of the sclera in corresponding areas of specimens of varying ages. The second eye of each pair was prepared for histologic study. Twelve additional histologic preparations of normal infant eyes in our John E. Weeks Eye Pathology Laboratory also were studied and measured.

Results

The overall dimensions of our five pairs of neonatal eyes were similar to those reported by other authors.35 The sagittal, vertical and transverse diameters averaged slightly less than 17 mm as compared to 24 mm for the average emmetropic adult eye. Measurements of diameters, although traditionally used for comparative studies of eye size, did not reflect the small size of the neonatal eye as well as other parameters and photographs (Figure 1). The average volume (2.8 cc) of our neonatal eyes was only half that of adult eyes (6.8 - 7.5 cc). The average scleral surface (812 mm2) was only one-third that (2450 mm2) of the adult specimens. The average neonatal corneal surface (102 mm2 ) was three fourths that (138 mm2) of the adult specimens.

The most rapid growth of the human eye is known to occur during the first two years of life;3 however, measurements of this series indicate that the most dramatic changes actually take place in the first six months. As shown in Table 1, diameters of the six-month-old specimens had increased to 20.3 mm; that is, about half of the expected total in diameter, volume and surface area lifetime increase had occurred by this age! The diameters of 20- and 30month specimens had grown to 22 to 23 mm; that is, within 2 mm of the diameters of the adult emmetropic eye.

Comparison of sagittal histologic sections (Figure 2) documents that the dramatic overall enlargement of the coats of the human eye in the first two and a half years of life is in the sclera and vitreous chamber In keeping with its rapid growth, the sclera of the infant was much more cellular than that of adults. The growth was circumferential; that is, there was little variance in the weight of comparable scleral buttons in the neonatal versus the 6-, 9- and 20-month specimens. The median thickness of the sclera at the equator, measured by calipers and on microscopic sections was about 0.45 mm, similar to that found in adult eyes and in agreement with measurements by Scammon and Armstrong.4

FIGURE 1: The cornea and iris, the visible parts of the eye of a neonatal infant (left), appear almost adult size, but the volume and area of scleral surface are less than half those of the eye of an emmetropic adult (right).

FIGURE 1: The cornea and iris, the visible parts of the eye of a neonatal infant (left), appear almost adult size, but the volume and area of scleral surface are less than half those of the eye of an emmetropic adult (right).

FIGURE 2: A sagittal section of an eye from a 12-day -old infant superimposed on a section of the eye of a 30-month-old infant documents the dramatic growth of the sclera and enlargement of the vitreous chamber, as compared to the minimal growth of the anterior segment. (Hematoxylin andeosin - original magnification one time.)

FIGURE 2: A sagittal section of an eye from a 12-day -old infant superimposed on a section of the eye of a 30-month-old infant documents the dramatic growth of the sclera and enlargement of the vitreous chamber, as compared to the minimal growth of the anterior segment. (Hematoxylin andeosin - original magnification one time.)

Table

TABLE 1GROWTH OF THE INFANT EYE

TABLE 1

GROWTH OF THE INFANT EYE

There were individual variations in the breadth of the rectus insertions, but as shown in Table 2, the widths of the insertions in the neonatal eyes were about 3 mm narrower than in the adult eyes of 24 mm diameters. The thickness of the insertions was not measured, but they appear thinner and more easily stripped from their scleral attachments than in the adult eye. Histologic studies revealed relatively superficial insertions into the sclera.

Table

TABLE 2BREADTH OF RECTUS MUSCLE INSERTIONS IN MILLIMETERS

TABLE 2

BREADTH OF RECTUS MUSCLE INSERTIONS IN MILLIMETERS

Table

TABLE 3MILLIMETERS FROM CORNEA TO RECTUS INSERTIONS

TABLE 3

MILLIMETERS FROM CORNEA TO RECTUS INSERTIONS

Table

TABLE 4DISTANCE IN MILLIMETERS OF OBLIQUE INSERTIONS FROM CORNEA AND OPTIC NERVE

TABLE 4

DISTANCE IN MILLIMETERS OF OBLIQUE INSERTIONS FROM CORNEA AND OPTIC NERVE

In several specimens the borders of the limbus and the corneal scleral sulcus could not be sharply defined; therefore, the margins of clear cornea had to be used in all eyes as the primary reference for measuring insertional positions of the rectus muscles. Our measurements, therefore, were slightly greater than those traditionally made from the posterior border of the limbus. This technique also introduced two other variants; in the infant eyes the horizontal diameter of clear cornea was about 1 mm wider than the vertical and the corneal diameters varied with individual specimens.

There also were variations in the insertional positions of the rectus muscles, but as shown in Table 3, the median distance of the insertions from the edge of clear cornea in the ten neonatal eyes was about 1 mm less than those of the 6- to 9-month specimens and about 2 mm less than the adult distance.

FIGURE 3A: In the neonatal eye (left) the medial rectus insertion may be less than 2 mm from the equator, but in its relationship to the cornea it is similar to its emmetropic adult eye (right).

FIGURE 3A: In the neonatal eye (left) the medial rectus insertion may be less than 2 mm from the equator, but in its relationship to the cornea it is similar to its emmetropic adult eye (right).

FIGURE 3B: The insertion of the lateral rectus tendon (folded toward the cornea) in a neonatal eye (left) may be at the equator. Adult eye (right).

FIGURE 3B: The insertion of the lateral rectus tendon (folded toward the cornea) in a neonatal eye (left) may be at the equator. Adult eye (right).

FIGURE 4A: At birth, the superior and inferior oblique insertions arem uch closer to each other and to the horizontal meridian than in the larger adult eye (right).

FIGURE 4A: At birth, the superior and inferior oblique insertions arem uch closer to each other and to the horizontal meridian than in the larger adult eye (right).

FIGURE 4B: In the neonatal eye (left) the superior oblique insertion is closer to the superior rectus and optic nerve than in the larger adult eye (right).

FIGURE 4B: In the neonatal eye (left) the superior oblique insertion is closer to the superior rectus and optic nerve than in the larger adult eye (right).

Of particular significance in relation to extraocular muscle surgery on infants was the distance from the rectus insertions to the equator. The lateral rectus insertions of neonatal eyes were at or close to the equator, and the medial rectus insertions were only 1 to 2 mm in front of it (Figure 3). Furthermore, the equatorial zone in these eyes, with an average circumference of only 54 mm, was relatively narrow as compared to that of an emmetropic adult eye with a circumference of 85 mm.

Fink6 reported that the insertions of the oblique muscles of the infant eye were similar to those in the adult, but the measurements (Table 4) were quite different in our specimens. As shown in Figure 4A, the oblique insertions were much closer together and closer to the optic nerve, but in these small eyes the anterior border of the superior oblique insertions (Figure 4B) was a median distance of only 9 mm from the cornea as compared to 15 for the adult eyes. The anterior edge of the inferior oblique was only 10 mm from the cornea of the neonatal eye as compared to 16 mm for the adult eyes. In some specimens the posterior edge of the inferior oblique insertions were within 1 mm of the optic nerve and in direct apposition with some of the short posterior ciliary arteries and veins (Figure 5). In several of the neonatal and specimens of early infancy the vortex vein was under the inferior oblique near its insertion.

FIGURE 5: In some neonatal eyes the medial end of the inferior oblique insertion extends to the optic nerve and posterior ciliary vessels and nerves. Near its insertion, the muscle may cover a vortex

FIGURE 5: In some neonatal eyes the medial end of the inferior oblique insertion extends to the optic nerve and posterior ciliary vessels and nerves. Near its insertion, the muscle may cover a vortex

FIGURE 6A: In contrast to the small size of the neonatal eye shown in Figures 3 and 4, the eye has obtained about half of its lifetime growth by 6 months of age. The superior oblique and superior rectus insertions of a 6-month-old eye (left) and an adult eye (right).

FIGURE 6A: In contrast to the small size of the neonatal eye shown in Figures 3 and 4, the eye has obtained about half of its lifetime growth by 6 months of age. The superior oblique and superior rectus insertions of a 6-month-old eye (left) and an adult eye (right).

FIGURE 6B: Insertions of the lateral rectus, superior oblique and inferior oblique of a 6-month -old eye (left) andan adult eye (right).

FIGURE 6B: Insertions of the lateral rectus, superior oblique and inferior oblique of a 6-month -old eye (left) andan adult eye (right).

With rapid enlargement of the posterior segment in the first six months of life the separation between the oblique insertions had increased by 4 to 5 mm. The insertion of the inferior oblique had moved temporally from its neonatal position at the posterior pole, but still was only 3.2 mm from the nerve sheath and adjacent vessels.

Discussion

Several aspects of this study have relevance to extraocular muscle surgery on the infant eye. One is that simple linear measurements of diameters, traditionally used to compare the size of the eye, mislead the surgeon into believing that the neonatal eye is about three fourths of the adult size, whereas its volume is only half and surface area even less. The surgeon is further misled by the fact that the most visible parts of the eye in infants, the cornea and iris, have more than 80% of their adult dimensions at birth.

Several anatomical factors make it more difficult to predict the outcome of extraocular muscle surgery on the neonatal eyes as compared to surgery on older children. Although the distance of the rectus insertions from the limbus is about 80% of the adult eye, the insertions may be bo close to the equator, especially those of the lateral rectus, that even a 3- to 4-mm recession would place them posterior to it. As the positions of the oblique insertions in very young infants differ considerably from those of older children, the ultimate outcome of early surgery on these muscles also will be difficult to predict until there is more information about the effects of growth of the globe on the ultimate positions of insertions shifted surgically early in infancy.

Our studies confirm earlier anatomical reports that the sclera of the neonatal eye, although thin (.4 to .5 mm), is comparable in thickness to that of older infants and children (Figure 2), so that the risk of accidental perforation during placement of recession sutures would appear to be the same. In some neonatal eyes the close proximity of the inferior oblique insertion to the optic nerve places the posterior ciliary vessels and nerves, as well as the vortex vein, at a risk if tenotomy is performed at the insertion.

The rapid enlargement of the human eye in the first five to six months dramatically changes its topographical anatomy, thereby placing insertional positions of the extraocular muscles much more like those of older children and adults (Figure 6). The wider and more posterior equatorial zone permits room for rectus recessions of 3 mm or more without placing them behind the equator. The dramatic expansion of the posterior sclera has shifted the inferior oblique insertions away from the posterior ciliary vessels and nerves. The results of surgical procedures, therefore, should be comparable to those performed on older children.

It is planned to continue these studies with a larger number and wider scope of infant eyes. More specimens are especially needed in the two - five-month-old age group. In addition, there are several other aspects which merit investigation in relation to the safety and predictability of early extraocular muscle surgery. These include the relationships of the eye to the orbit and lids, as well as histologic studies to better establish the pattern and mechanisms of scleral growth. Long-term clinical studies are needed to determine the ultimate positions of muscle insertions shifted by surgery on small immature eyes.

Summary and Conclusion

The dimensions and topographical anatomy of 26 eyes from 14 neonates and infants were measured and photographed with special attention to the insertional position of the extraocular muscles. Additional measurements were made in histologic preparations of 12 normal infant eyes.

Although the diameters of the neonatal eyes were found to be about 70% of the adult eyes, the volumes of the globes were only about half, and the surface areas even less. The insertions of the rectus muscles were about 2 mm nearer to the cornea than in emmetropic adult eyes, but some were close to or at the equator. The posterior segments of the neonatal eyes were much less developed than the anterior, so that the oblique insertions, as compared to the adult eye, were closer to each other, to the horizontal meridian and to the posterior pole. In some neonatal specimens the inferior oblique insertion was so close to the optic nerve that tenotomy at that point would have jeopardized some of the posterior ciliary vessels and nerves.

The dramatic postnatal growth of the eye occurs in the scleral segment; there is minimal corneal growth. About half of the total lifetime increase in the diameters, volume and total surface area of the human eye occurs in the first six months of life. In the six-month-old specimens the volume of the globe and the surface area of the sclera had almost doubled as compared to the neonatal eyes. The inferior oblique insertion had moved temporally away from the optic nerve. The distance between the insertion of the rectus muscles and the equator had increased significantly. The predictability and safety of extraocular muscle surgery, therefore, would appear to be greater if the surgical correction of congenital deviations could be deferred until the eyes reached this stage of development at about six months of age; however, additional anatomical and clinical studies are needed, especially on eyes of three- to fivemonth-old infants.

References

1. Freeley DA, Nelson LB, Calhoun JH: Recurrent esotropia following early successful surgical correction of congenital esotropia. J Pediatr Ophthalmol Strabismus 1983; 20:68-71.

2. Ing MR: Early surgical alignment for congenital esotropia. J Pediatr Ophthalmol Strabismus 1983; 20:11-18.

3. Weiss L: Über das Wachstum des menschlichen auges und über die Veränderung der muskelinsertionen am wachsenden auge. Anatomische Hefte 1897; 8:193-248.

4. Scammon RE, Armstrong EL: On the growth of the human eyeball and optic nerve. J Comp Neurol 1925; 38:165-219.

5. Sorsby A, Sheridan M: The eye at birth: Measurement of the principle diameters in forty-eight cadavers. J Anat I960; 94:192-197.

6. Fink WH: The development of the extrinsic muscles of the eye. Am J Ophthalmol 1953; 46:10-23.

TABLE 1

GROWTH OF THE INFANT EYE

TABLE 2

BREADTH OF RECTUS MUSCLE INSERTIONS IN MILLIMETERS

TABLE 3

MILLIMETERS FROM CORNEA TO RECTUS INSERTIONS

TABLE 4

DISTANCE IN MILLIMETERS OF OBLIQUE INSERTIONS FROM CORNEA AND OPTIC NERVE

10.3928/0191-3913-19840301-03

Sign up to receive

Journal E-contents