Extraocular Muscle Insertions Relative to the Fovea and Optic Nerve: Humans and Rhesus Macaque

Abstract

purpose. To identify extraocular anatomic relationships of muscle insertions relative to the fovea and the optic nerve.

methods. Thirty-eight human eye bank eyes and 10 rhesus macaque (Macaca mulatta) eyes were measured. Ten human volunteers were used to determine the horizontal rectus muscle-to-globe apposition in primary, left, and right gaze.

results. External globe measurements (human/rhesus; mm ± SD) from the temporal border of the optic nerve (ON) to the center of the fovea (F) were 3.7 ± 0.6 and 2.6 ± 0.2; F to the posterior border of the inferior oblique (IO) insertion, 2.5 ± 0.8 and 0.5 ± 0.4; ON to the posterior border of the IO, 5.6 ± 0.9 and 2.8 ± 0.3; horizontal axial plane (H) of the eye, defined by the long posterior ciliary artery, to the IO, 2.0 ± 0.8 and 0.5 ± 0.4; and H to F, 1.0 ± 0.6 and 0.4 ± 0.3, respectively. The IO insertion formed an arc, inferior to H, with an anterior-to-posterior cord insertion width of 9.2 ± 0.7 and 7.7 ± 0.3. The IO angle of insertion (θ) was 30° in 84% (32/38) and 0° in 16% (6/38) of human eyes and 25 to 30° in all rhesus. In 20 human volunteers, from the ON to the apex of lateral rectus globe apposition was 13.9 ± 1.1 in primary, 17.2 ± 1.9 in lateral, and 9.3 ± 1.7 in medial gaze.

conclusions. The fovea is located mostly superior and slightly posterior to the posterior border of the IO insertion. Topographic relationships of the extraocular muscles relative to the fovea are essential for the design of extraocular drug delivery systems.

Posterior segment eye disease represents a major cause of irreversible blindness worldwide. ^{1 2 3 4 5 6 7} New pharmacotherapies are being developed to address diseases of the retina, choroid, and optic nerve. Systemic delivery exposes the entire body to potential adverse effects from a given drug or therapeutic agent to a greater extent than local, targeted delivery. Several factors limit the diffusion of topically applied medications to posterior segment tissues. These include the distance required for simple diffusion, aqueous flow kinetics, conjunctival and episcleral blood flow, and permeability barriers such as the sclera, corneal epithelium, stroma, and endothelium. To specifically target the macular region for retinal disorders such as age-related macular degeneration and diabetic maculopathy, the transscleral drug delivery (TSDD) route offers an excellent safety profile and eliminates many complications associated with intravitreal delivery. 

Human sclera has been shown to be permeable to many compounds, depending on the size of the molecule delivered and the thickness of the sclera. ^{8 9} Extraocular drug delivery systems that use the transscleral route offer a safety advantage over intraocular delivery systems. Intraocular surgery or intravitreal injections may be associated with endophthalmitis, retinal detachment, cataract progression, device migration, and recurrent risk with each subsequent intervention. An extraocular TSDD system would decrease these risks and could be more easily replaced or “recharged” for maintenance therapy. This capability is particularly important in treating chronic diseases, such as age-related macular degeneration (AMD), diabetic eye disease, and glaucoma. Early studies with TSDD systems tested in rabbits (Yaacobi Y, et al. IOVS 2003;44:ARVO E-abstract 4210) and rhesus monkeys (Olsen TW, et al. IOVS 2003;44:ARVO E-Abstract 4213) have shown promising results. Herein, we document a detailed anatomy of specific extraocular relationships in humans and rhesus monkey to optimize design parameters of extraocular devices to direct TSDD to the macula and to establish a relationship from an animal model to humans. Although the monkey has thinner sclera, ¹⁰ its extraocular muscle anatomy is similar to that of humans. Moreover, all primates have a fovea that allows direct comparison of anatomic relationships to humans. Our study investigates the important anatomic relationships of the extraocular muscle insertions to the fovea and optic nerve in rhesus and human eyes. 

Methods

Eye Preparation

Human Eye Bank Eyes.

Thirty-eight enucleated human cadaver whole globes were obtained from the Minnesota Lions Eye Bank. Age, time of death, preexisting medical conditions, and cause of death were recorded for all donors. Eyes were obtained according to standard tissue procurement techniques used by the eye bank and with the consent of the donor or donor family for the purposes of medical research. Consent was obtained in accordance with the principles outlined in the Declaration of Helsinki. Tenon capsule and adnexa were dissected from each eye to expose the optic nerve and the four rectus and oblique muscles. Globes were then preserved in 10% formalin. 

Rhesus Macaque Eyes.

Ten rhesus macaque eyes preserved in 10% formalin were obtained from the Wisconsin Regional Primate Research Center (Madison, WI). Date of birth and sex of each primate were documented. No animal was killed specifically for this study. Tissue was obtained as it became available from other research projects unrelated to the present study. Extraocular structures were isolated according to the same procedures performed in human eyes. 

Human Subjects.

Subjects volunteered and signed informed consent; approval was granted by the Institutional Review Board at the University of Minnesota. Ultrasonography was performed with the subjects in the supine position and with their eyelids closed. 

Measurements

Long posterior ciliary arteries were used to identify the horizontal meridian. A precision dial caliper (Pierre Roch Ltd., Renens, Switzerland) was used to measure the greatest globe diameter at the horizontal equator (accurate to 0.01 mm). A standard Vernier caliper was used to measure the cord length between anatomic landmarks (accurate to 0.5 mm) (Table 1) . A flexible silicon ruler (accurate to 0.5 mm) was used to measure the arc length between anatomic landmarks (Fig. 1) . A 360° sclerotomy was made at the limbus using a razor blade and scissors. Cornea, iris, and lens were removed to allow visualization of the macula through a surgical microscope (OPMI 6-SFR; Zeiss, Oberkochen, Germany). A 30-gauge needle was inserted perpendicularly through the center of the fovea, exiting the sclera posteriorly (Fig. 2) . Specific dimensions and relationships of the extraocular muscles relative to the fovea were determined. All measurements were taken from the scleral surfaces of the globe. Angles were measured with a standard protractor. The measurement was taken from a line along the inferior border of the inferior oblique (IO) insertion. 

Ten volunteers with no history of ocular disease or motility disturbances were examined using B-scan ultrasonography (I³; Innovative Imaging Inc., Sacramento, CA) to identify the effect of horizontal eye movements on the triangle formed by the orbital apex, the apposition of the lateral rectus (LR) to the globe, and the lateral border of the optic nerve. 

Results

Human Eyes

Of the 25 human donors, 12 were women and 13 were men. Age at the time of death ranged from 49 to 80 years (mean, 67 ± 8 years). Subjects had no history of ocular diseases or strabismus surgery. The fovea (F) lies temporal to the optic nerve and superior-nasal to the posterior IO insertion. The F is located at 1.0 ± 0.6 mm inferior to the horizontal meridian (H). Distances from the fovea to the posterior-inferior border (Y) and the anterior-inferior (Z) borders of the IO were 2.5 ± 0.8 mm and 11.0 ± 1.0 mm. The distance from F to the ON lateral edge was 3.7 ± 0.6 mm. 

The cord and arc width of the IO muscle insertion was 9.2 ± 0.7 mm and 9.5 ± 0.6 mm. The borders of the IO at points Y and Z were located 2.0 ± 0.8 mm and 3.9 ± 0.8 mm inferior to H. Distances from the ON lateral edge to Y and Z were 5.6 ± 0.9 mm and 14.2 ± 0.9 mm. The angle of the IO insertion (θ) relative to the horizontal meridian was 30° in 84% (32/38) of donors. The insertion was parallel (0°-5°) to the horizontal meridian in 16% (6/38) of donors. In eyes with a smaller θ, point Z was closer to H, thereby decreasing the angle of insertion θ. 

The arc width of the LR insertion was 10.5 ± 1.0 mm. Distances from the posterior-superior (L₂) and the posterior-inferior (L₄) borders of the LR to H were 4.6 ± 0.7 mm and 5.6 ± 0.9 mm. The arc length from the L₂ to Y or Z was 19.6 ± 1.5 mm and 12.5 ± 1.3 mm. The arc length between L₄ and Z was 10.0 ± 1.5 mm. The arc length from the 9 o’clock limbus (J) to the ON lateral edge was 33.0 ± 1.0 mm. The transverse diameter ( ) of the entire globe at H was 23.7 ± 0.6 mm. 

In 20 human volunteers, the ultrasonic cord length measurement from the ON lateral edge to the apex formed by lateral rectus apposition to the globe was 13.9 ± 1.1 in primary gaze, 17.2 ± 1.9 in lateral gaze, and 9.3 ± 1.7 in medial gaze (Fig. 3) . 

Rhesus Macaque Eyes

Of the 5 primates (10 eyes), one was male and four were female. Ages ranged from 5 to 30 years (mean, 15 ± 13 years). Primate globes did not show evidence of prior muscle surgery. F was temporal to the ON and superior-nasal to Y and was located 0.4 ± 0.3 mm inferior to H. Distances from F to Y and Z were 0.5 ± 0.4 mm and 7.7 ± 0.4 mm. The distance from F to the ON lateral edge was 2.6 ± 0.2 mm. 

Cord and arc widths of the IO muscle insertion were equivalent at 7.7 ± 0.3. The IO borders at points Y and Z were located 0.5 ± 0.4 mm and 2.6 ± 0.3 mm inferior to H. Distances from the ON lateral edge to Y and Z were 2.8 ± 0.3 mm and 10.1 ± 0.3 mm. The angle θ relative to H ranged from 25° to 30°. 

Arc width of the LR insertion was 6.5 ± 0.5. Distances from L₂ and L₄ to H were 2.5 ± 0.4 mm and 4.0 ± 0.7 mm. Arc lengths from L₂ to Y or Z were 16.9 ± 0.9 mm and 10.6 ± 0.5 mm. The arc length between L₄ and Z was 9.1 ± 0.7 mm, and the arc length from J to the ON lateral edge was 25.8 ± 0.9 mm. Transverse diameter ( ) of the entire globe at H was 20.0 ± 0.2 mm. 

Discussion

Transscleral drug delivery is a viable option in the management of posterior segment disorders. Anatomic configurations and variations determine a direct transscleral route to the macula. The primary anatomic obstacle to TSDD directed at the fovea is the IO insertion and the ON. Distances from the ON lateral edge to the posterior insertion of the IO are 5.6 ± 0.9 mm and 2.8 ± 0.3 mm in humans and in rhesus monkeys, respectively. Distances from the fovea to the posterior border of the IO were 2.5 ± 0.8 mm and 0.5 ± 0.4 mm, respectively. The posterior inferior border of the IO is nearly at the fovea in the rhesus, whereas in humans, there is a separation measuring nearly 2 to 3 mm. Therefore, the IO is less of a barrier to TSDD in the human eye than in the rhesus monkey eye. Extraocular muscles have intrinsic vasculature that could diminish the effective delivery of a drug core located on the scleral surface when the muscle lies between the device and the sclera. In addition, muscle movement may contribute to instability or device migration. 

Most of these anatomic measurements were performed in the first half of the twentieth century. Walter Fink ¹¹ analyzed the anatomic variation in the attachment of the oblique muscles of the eyeball by looking at 100 human globes and analyzing the oblique muscles. He found that there was variation in the insertion of the oblique muscle, but it occurred to a lesser degree than in the superior oblique. The distance from the lateral rectus tendon to the anterior end of the insertion of the inferior oblique was 9.6 mm, similar to our measurement of 10 ± 1.5 mm. The distance he observed from the posterior border of the IO to the ON was slightly less. The average was 4.26 mm, whereas our measurement was 5.6 ± 0.9 mm. Our measurement, at 5.2 mm, is closer to that documented by Whitnal ¹² in 1921 (it is higher in more myopic globes). This disparity may be attributed to the optic nerve sheath. Fink’s cautious dissection might have left more of the loose fibrous sheath around the optic nerve, hence the smaller measurement. One could consider this tissue surrounding the optic nerve a cushion to protect the optic nerve or a barrier to the placement of a drug or a device in the immediate perioptic nerve region. Whitnal ¹² reports that the distance from the IO to a “spot corresponding to the fovea” measures 2.2 mm, similar to our measurement of 2.5 ± 0.8 mm. According to Fink, ¹³ the angle of IO insertion in the horizontal plane is 15° to 20°, corresponding closely to our measurement of 23.7° ± 0.58°. Our measurement was taken from the inferior border of the IO to simulate a device that may be supported at the inferior border. Fink’s ¹³ measure is schematically demonstrated at the superior line of insertion, thereby resulting in a smaller angle. 

Another important relationship is the distance from the anterior-inferior border of the inferior oblique muscle to the fovea (Fig. 1 , Z). There is 11 mm of separation in the human eye and nearly 8 mm in the rhesus monkey eye. The IO insertion is a firm anatomic and surgical landmark that could serve to stabilize a TSDD system. Localizing this structure surgically is relatively simple and is commonly performed during strabismus operations. Direct visualization of this region is difficult because of the posterior location of the insertion and the surrounding orbital adnexa. 

The orbital tissue conforms and accomodates an extraocular device and helps to stabilize it against the globe. Eye movement should not be obstructed or altered by placement of a TSDD system in the episcleral foveal region. We found that in human volunteers who underwent ultrasonography, the apposition of the LR muscle to the globe would not create a significant barrier to a posteriorly located, small TSDD device. Medial gaze is the most restrictive gaze for a device placed in the sub-Tenon space over the macula. The apposition of the LR to the ON in a human volunteer decreases from 14 mm in primary gaze to approximately 9 mm in medial gaze. Considering the subjective nature of each person’s gaze, these measurements are relatively reproducible with low standard deviations. Appositional distance increases in lateral gaze to approximately 17 mm. The orbital apex insertion of the LR creates a space to accommodate a drug delivery system that should not interfere with the function of this muscle. Because of the angle of insertion of the LR muscle, medial gaze would not likely create an obstruction for a device placed near the fovea because the belly of the LR would not approach the sclera in this region. Orbital fat is accommodating and could serve to cushion a device posteriorly and to provide forward pressure to maintain device apposition of the sclera. As long as the TSDD system is under the Tenon capsule, orbital fat should not limit drug diffusion. 

The distance from the insertion of the inferior oblique to the horizontal aspect of the globe is 2 mm in humans and 0.5 mm in rhesus monkeys, demonstrating that the IO insertion is closer to the foveal region of the rhesus monkey eye than in the human eye. Once again, the human eye is less prone to interference by the IO muscle for drug diffusion to the fovea. 

The closest anatomic approach for placement of a TSDD system is over the superior border of the LR or through a superior-temporal approach. The LR is decentered slightly inferiorly to the horizontal aspect of the eye. Alternatively, the inferior approach is limited by the IO muscle insertion and by the greater distance to the foveal region. A device placed through the inferior approach would have to rest under the IO insertion or travel around the insertion to approximate the foveal region. 

In summary, these data identify important anthropomorphic measurements of human and rhesus extraocular musculature in the episcleral region that are relevant to important intraocular structures, such as the fovea and the optic nerve. New advances in pharmacotherapy of posterior segment disorders are evolving. TSDD systems offer several advantages over intraocular or systemic delivery. ¹⁴ Most posterior segment disorders are chronic and, therefore, would be best treated with a continuous drug delivery system. Local therapy is advantageous to minimize systemic toxicities. An extraocular device that makes use of simple diffusion through the sclera could be easily replaced or recharged. A major limitation of drug delivery directly through TSDD to the macula includes the unknown effects of diffusion through the highly vascularized choroid. Our data help to document the important anatomic structures for developing TSDD systems to treat many common posterior segment disorders. Detailed comparison of the human eye and the rhesus model will provide important relationships to translate primate study data to the future design and development of devices for humans. 

 Supported by Minnesota Lions Macular Degeneration Center, Minneapolis, MN; University of Minnesota Vision Foundation, Minneapolis, MN; Alcon Laboratories Ltd., Fort Worth, TX; and by an unrestricted research grant from Prevent Blindness Foundation.

 Submitted for publication March 5, 2005; revised May 28, 2005; accepted August 19, 2005.

 Disclosure: X. Feng, Alcon Laboratories, Inc. (F); K. Pilon, None; Y. Yaacobi, Alcon Laboratories, Inc (E); T.W. Olsen, Alcon Laboratories, Inc. (F)

 The publication costs of this article were defrayed in part by page charge payment. This article must therefore be marked “advertisement” in accordance with 18 U.S.C. §1734 solely to indicate this fact.

 

Table 1.

 

Relationship of the IO to the LR and Macular

	F-Y	Y-Z	F-Z	θ	F-ON	Y-ON	Z-ON	Y-H	Z-H	F-H	HX-HY	L4-Z2 *	J-N*	J-L2 *	L2-L4 *	L2-Y*	L2-Z*	Y-Z*	L2-H	L4-H	@H, †
Human
Mean	2.5	9.2	11.0	30°, ‡	3.7	5.6	14.2	2.0	3.9	1.0	1.8	10.0	33.0	8.8	10.5	19.6	12.5	9.5	4.6	5.6	23.7
±SD	0.8	0.7	1.0		0.6	0.9	0.9	0.8	0.8	0.6	0.7	1.5	1.0	0.7	1.0	1.5	1.3	0.6	0.7	0.9	0.6
Monkey
Mean	0.5	7.7	7.7	28°, §	2.6	2.8	10.1	0.5	2.6	0.4	0.3	9.1	25.8	6.3	6.5	16.0	10.6	7.7	2.5	4.0	20.0
±SD	0.4	0.3	0.4		0.2	0.3	0.3	0.4	0.3	0.3	0.3	0.7	0.9	0.4	0.5	0.9	0.5	0.4	0.4	0.7	0.2

 All values are in millimeters except for θ, which is measured in degrees. , the transverse diameter of the entire globe. See Figure 1for expansions of abbreviations.

*  Arc length.

†  Transverse diameter at the horizontal meridian.

‡  84% θ = 30°; 16% θ = 0°.

§  θ ranged from 25° to 30°.

Figure 1.

 

Anatomic landmarks found in Table 1 . L₁, superior-anterior border of the LR; L₂, superior-posterior border of the LR; L₃, inferior-anterior border of the LR; L₄, inferior-posterior border of the LR; J, 9 o’clock limbus at horizontal meridian; θ, angle of the IO insertion relative to the horizontal meridian.

Figure 2.

 

A 30-gauge needle was inserted perpendicularly through the center (arrow) of the fovea (1.5-mm circle around the needle). The optic nerve (ON) and the inferior oblique (IO) are identified.

Figure 3.

 

Human ultrasonographic images of the right eye showing the angle formed by the orbital apex, apposition of the lateral rectus to the globe (upper arrow), and optic nerve (lower arrow) in primary position (upper left), lateral gaze (upper right), and medial gaze (lower left).