WO1999049814A1

WO1999049814A1 - Method and apparatus for adjusting corneal curvature

Info

Publication number: WO1999049814A1
Application number: PCT/US1999/006933
Authority: WO
Inventors: Joseph Lee
Original assignee: Microoptix Llc
Priority date: 1998-03-30
Filing date: 1999-03-29
Publication date: 1999-10-07
Also published as: AU4068099A

Abstract

A method and apparatus for adjusting corneal curvature of the eye comprising an adjustable split device (15) formed of a flexible hollow shell (16) which is implantable into the cornea in encircling relation to the central optic zone of the cornea. The implant (15) is filled by a predetermined amount with a select quantity of solid biocompatible material in various forms such as rings or strands (30) and composed of flexible polymeric materials of various shape and length. The biocompatible filler material (30) is strategically located within the flexible shell (16) to alter its dimensions in thickness or diameter and thereby adjust the corneal curvature to correct for refractive error. Further adjustment of the implant may be made post-operatively after implantation by select removal of the biocompatible filler material, shifting of the material, or addition of new material to the device.

Description

METHOD AND APPARATUS FOR ADJUSTING CORNEAL CURVATURE

1. RELATED APPLICATIONS

This patent application claims 119(e) benefit of prior Provisional Application No.

60/079,950, filed March 30, 1998.

2. FIELD OF THE INVENTION

This invention relates to a method and apparatus for adjusting corneal curvature and, more particularly, to an implantable device adapted for insertion into the cornea of an eye and which may be modified in the amount of volume it displaces at the time of insertion and at post-operative times to correct refractive error by adjusting or removing solid material from the implanted device or augmenting said device with solid material.

3. BACKGROUND OF THE INVENTION

Ametropia, an undesirable refractive condition of the eye, has three main subdivisions: myopia, hyperopia, and astigmatism. In myopia, by far the most common type of ametropia, the parallel light rays 1 which enter the eye as shown in FIG. 1(c) come to a focus F3 in front of the retina 2. In hyperopia, the rays of light 1 come to a focus F2 behind the retina 2 as shown in FIG. 1(b). When the rays of light converge to not one, but several foci, it is referred to as astigmatism, in which condition the various foci may all lie before the retina; all lie behind the retina; or partly before and partly behind the retina. Ametropia is usually corrected by glasses or contact lenses. However, these refractive disorders may also be corrected by surgery. Refractive eye surgery is defined as that surgery on the eye that acts to change the light-bending qualities of the eye. More 2 common current refractive procedures include radial keratotomy, as described in U.S. Pat. 4,815,463 and 4,688,570 and also laser ablation of corneal stroma, described in U.S. Pat. 4,941,093. Various other surgical methods for the correction of refractive disorders have been tried including thermokeratoplasty for the treatment of hyperopia, epikeratoplasty to correct severe hyperopia, and keratomileusis which can steepen or flatten the central cornea. Keratomileusis was introduced by Barraquer of Colombia in 1961 and essentially involves grinding a corneal button into an appropriate shape to correct the refractive error and replacing the reshaped corneal button. Some of the more common keratorefractive procedures are discussed below, none of which have currently shown itself to have all the characteristics of an ideal keratorefractive procedure. The disadvantages of corneal refractive surgery include limited predictability, lack of reversibility, corneal destabilization, optical zone fibrosis, post-operative discomfort, and visual symptoms such as glare, halos, and starbursts.

In radial keratotomy (RK) multiple peripheral radially directed incisions are made into the cornea at 90-95% depth in an attempt to flatten the central cornea and thus correct myopia. The problem of unpredictability of result was tackled by multiple extensive retrospective analyses of the patients in whom surgery had already been performed. These studies revealed certain factors that seemed to control the outcome of the surgery, such as the size of the optical zone, the initial keratometric readings, corneal diameter, corneal rigidity, number of incisions, incision depth, intra-ocular pressure, thickness of the cornea, and degree of astigmatism. Age and sex are also factors that are taken into consideration in most of the nomograms, which have been devised to predict what effect to expect for a certain surgery.

At one point, many experts in the field considered it nearly impossible to fully and accurately correct patients in one surgery and felt that RK should be considered a two-stage surgery, with the initial surgery to achieve the "ball-park" correction, followed by an enhancement procedure to adjust or titrate the result near the desired outcome for an individual eye. It was felt that because of individual variability which may lead to an under or over-correction in the individual different from that predicted by the nomogram, attempting 3 to fully correct the refractive error in one surgery could lead to over-correction in a not insignificant percent of the surgeries, resulting in hyperopia which is much more difficult to correct. Unfortunately, the second-stage surgery is even less predictable than the initial procedure. No one has yet devised a formula to take into account the profound changes which occur in the cornea after the initial RK, especially when weeks or months have passed. Most studies quote only 50-60% of eyes achieving 20/20 or better visual acuity following RK. Patients who are accustomed to 20/20 or better corrected visual acuity before surgery are not typically satisfied with less than 20/25 or 20/30 uncorrected post-operative visual acuity.

In addition, a gradual hyperopic shift is a major concern after RK. Refractive stability is critical for all refractive procedures but all corneal refractive procedures show significant degrees of instability. To date, there has been no clear explanation of why the cornea is destabilized by RK. A recent report on the long-term results of RK stressed the "natural" hyperopic refractive progression of "normal" eyes as a function of age. It is possible that patients are initially overcorrected and the over-correction masked by the patient's accommodative powers. With time and loss of accommodation, the hyperopia may be gradually unmasked with the hyperopia becoming visually symptomatic. At the time of surgery, a patient may be corrected with resultant slight hyperopia and yet have 20/20 vision because of the ability of the lens to accommodate. There is a range of residual correction within which the patient can have 20/20 uncorrected vision. This range varies depending on the individual but probably spans two to three diopters. Even with this range, the percentage achieving 20/20 is only 50-60%. This reflects poorly on the precision of the technique. It is important to note that this range diminishes with presbyopia, or loss of accommodation which usually begins at about 45 years of age. This results in the percentage achieving 20/20 dropping from the 50-60% described above. It is obvious that RK does not qualify as a simple, safe, predictable procedure to adjust the refractive outcome after the initial RK has been performed. Most ideas to contend with the corneal shape after this event have been purely empirical. Thus an easy method to fine-tune a refractive correction that is minimally invasive and easily performed, would require serious consideration. 4

Laser stromal ablation procedures, such as photorefractive keratectomy (PRK) for correction of refractive disorders are currently popular and have had reasonable success. These procedures are not, however, spared from the problem of unpredictability. Essentially, in the treatment of myopia, laser energy is imparted to the central cornea thereby causing excision of more tissue centrally and a resultant flattening of the cornea. Unfortunately, the final refractive effect is determined not only by the amount of ablation but also by the healing response to the keratectomy. The cornea actively lays down new collagen and the epithelium undergoes a hyperplastic response, among other responses, in an attempt to repair the damage to its surface. This causes regression, or a shift backwards towards myopia, which can gradually occur over a period of months to years. An undesired effect of new collagen deposition is stromal scar formation which manifests as stromal haze and possible decrease in contrast sensitivity by the patient. This corneal stromal opacification is variously referred to as fibrosis, scarring, or haze which is associated with reduced visual acuity and contrast sensitivity, regression of the refractive effect, and poor night vision. Predictability with PRK is an issue, as with RK. Most published results of outcome after PRK treatment for myopia show 80-94% of eyes obtaining uncorrected visual acuity of 20/40 or better while the percentage of patients achieving 20/20 is significantly less. These numbers are in spite of the fact that there is a range of residual refraction at which the patient can still see 20/20 as previously explained. It can be assumed that a significant proportion of those achieving 20/20 after PRK is actually slightly hyperopic. It may very well be that with time, a significant percentage of those patients develop "progressive hyperopia", or an unmasking of the latent hyperopia. So, although the percentage of patients achieving 20/20 after PRK is not acceptable by the definition of an ideal refractive procedure, it may be inflated as was the initial results with RK. Although visual recovery is slow in RK, it is quicker than after PRK. A second laser ablation procedure is usually undertaken with caution since it may cause a greater healing response with even more regression than the initial procedure. Again, as in RK, the laser ablation procedure is not completely predictable, partly because one cannot predict an individual's wound healing response. 5

The criteria for an ideal refractive procedure include adjustability, predictability/efficacy, maintenance of quality of vision, reversibility, simplicity, stability, safety, and low cost. Each of these factors are reviewed:

1. Adjustability. Because of corneal wound healing and the delicate corneal curvature changes required to achieve 20/20 uncorrected vision, it is unlikely that any refractive procedure will achieve 20/20 vision with a single procedure. The refractive literature is abundant with comments and opinions describing the need for an adjustable procedure. PRK attempts adjustments or "enhancements by repeating the same process — removal of epithelium and re-ablation. If there was variability in the initial PRK procedure, it is unlikely that the same procedure will be able to "fine-tune" the refractive outcome.

2. Predictability/Efficacy. The bottom line is that only about 71% to 97% of eyes with baseline refractions less than or equal to 6 D achieve manifest post-op refractions within +/-1 D of the attempted correction one-year after surgery. PRK, RK, and corneal ring literature all report similar efficacy data. 3. Quality of vision. PRK ablates the central cornea and corneal haze is an issue as evidenced by a small percentage of patients with a decrease in best-corrected visual acuity. Visually comprising complications include haze and scarring, halo effect, decreased contrast sensitivity, and decentration of ablation. Any procedure that operates on the central cornea will result in decreasing the best-corrected visual acuity in at least a small percentage of patients.

4. Reversibility. Many prominent refractive surgeons believe that the trend in refractive surgery would e towards reversible procedures. A reversible procedure, such as corneal rings, may likely be the procedure of choice for lower myopia and a reversible procedure, such as implantable intraocular lenses, may be the procedure of choice for higher myopia. Many potential patients have not had vision correction surgery because of its irreversibility. 6

5. Simplicity. A refractive procedure that only a few skilled refractive surgeons can perform will unlikely become a popular procedure. Also, difficult procedures typically have variable outcomes and a steeper learning curve. The complication rate for beginning refractive surgeons for a difficult procedure cannot be dismissed since there will always be new surgeons learning the procedure.

6. Stability. A long-term complication of Radial Keratotomy is progressive hyperopia. PRK undergoes regression of effect.

7. Safety. Refractive surgeries are elective procedures and much of the initial resistance to performing these procedures in the late 80' s was the philosophy among ophthalmologists that surgically manipulating a healthy eye with the potential for disastrous loss of vision was ethically unacceptable. PRK can cause central corneal haze sufficient to decrease best-corrected visual acuity. LASIK invades a central cornea with a microkeratome and is associated with multiple potential complications including lost flaps, button-holed flaps, free caps, thin flaps, and even perforation with extrusion of intraocular contents. In general, a corneal procedure is probably safer than an intraocular procedure and a peripheral corneal procedure safer than a central corneal procedure.

For well over a century, ophthalmologists have searched for a surgical method to permanently correct refractive errors. At least 15 different techniques have been developed and considerable experience has accumulated in both animal and human models. Laser photorefractive keratectomy has come the closest to gaining widespread acceptance in the ophthalmic community, but the difficulty in gaining acceptance for keratorefractive procedures is because of the unsolved problems with poorly predictable and unstable refractive outcomes, adverse effects on the quality of vision, lack of adjustability, and irreversibility. Poor predictability looms as the largest unsolved problem with refractive corneal surgery. The two major factors that contribute to poor predictability are (a) the variability and inaccuracy inherent with manual surgical techniques, and (b) the variable influence of 7 corneal wound healing in determining the refractive outcome. Until these two deficiencies are corrected, it is unlikely that a refractive surgical procedure will predictably correct ametropia to within a half diopter, the margin which can be achieved routinely with glasses or contact lenses. Photorefractive keratectomy offers the possibility of solving one of the major causes of poor predictability by reducing the surgical variability of the procedure. A major unresolved issue is how the second nemesis that causes poor predictability, corneal wound healing, will affect the results of PRK. However, patients typically undergo regression of effect of approximately 1.0 - 1.5 diopters over six months. 9% - 20% of patients continued to show myopic regression of 1 D or more even during the second year after surgery. There are "under-responders" that do not undergo any regression and there are over-responders who undergo up to 3 diopters of regression. So, although it is true that the excimer laser is precise to sub-micron accuracy, the variability in regression removes that advantage.

Typical refractive surgery studies, including RK and PRK, report the post-operative refraction in terms of the percentage of patients achieving +/- 1 diopter of emmetropia.

Approximately 80-90% of patients achieves this range. FIG. 2(a) shows a distribution curve of number of patients and their refractive outcomes. If it is assumed that only patients within the range of -0.30 D and +1.0 D are 20/20 without symptoms, it can be seen that only approximately 50-60% of patients achieve vision of 20/20 without symptoms. The unsolved problem in refractive surgery is that only about half of patients achieve a vision of 20/20 without correction following current refractive procedures.

The goal in refractive surgery is to achieve emmetropia. However, there is a range of residual refractive error at which the patient can see 20/20 without correction. On the myopic side, a patient can be -0.30 D or less and still see 20/20 uncorrected. The issue is slightly more complex on the hyperopic side due to the availability of lens accommodation. The average 30 year-old has a total of 7 diopters of accommodation available to him and can easily supply several diopters from his "storehouse" of reserve. Reading a book or newspaper at arms-length (40 cm) requires 2.5 diopters of accommodation for the emmetrope. This 8 means that a 30 year old patient may be overcorrected by up to 2.5 diopters and still see 20/20 uncorrected when tested, with minimal symptoms for distance vision since he is using only 2.5 diopters of the 7 D available to him. Of course, now to read, he will require 5.0 diopters of accommodation and will most likely be unable to read comfortably. In the ages of 35-40, the 1.50 diopter hyperope who could always get along easily without wearing his correction for distance vision will suddenly find that he cannot.

The modern refractive surgeon has several weapons in his armamentarium to choose from in attacking myopia. The refractive surgeon knows the limitations of his options. It is understood that RK is moderately predictable, adjustable only towards hyperopia, and irreversible. PRK is also moderately predictable, adjustable only towards hyperopia with the caveat that there is some regression towards myopia, but also essentially irreversible. On cardinal rule of refractive surgery is to avoid overcorrection because the options for a patient who is over-corrected to hyperopia are much more limited.

The dilemma results from the surgeon (and patient) wishing to achieve an uncorrected visual acuity of 20/20. The uncorrected visual acuity is poor if the post-op refraction is myopic but 20/20 if the post-op refraction is hyperopic. However, residual myopia can be "enhanced" while residual hyperopia is much more difficult to surgically manage.

This is best illustrated in a study comparing Summit and Visx lasers. The results showed a median refraction of 0.0 D in the Summit group and -0.5 D in the Visx group. The uncorrected visual acuity was 20/40 or better for 100% of the Summit treated eyes, whereas only 85% of the Visx treated eyes achieved 20/40 or better. Overcorrection results in a higher percentage of patients achieving 20/40 or better but results in a higher percentage of patients who are hyperopic. In other words, the higher percentage of patients achieving 20/40 in the Summit group may be explained by the accommodative reserve still present in the younger patients that were overcorrected.

Referring back to FIG. 2(a), it can be seen that there are many more patients with an uncorrected visual acuity of 20/20 on the right side (hyperopic, over-corrected) of the graph. The problem is that only 50% of patients achieve an uncorrected visual acuity of 20/20. Referring to FIG. 2(b), it becomes obvious that the easiest way to increase the percentage of patients achieving 20/20 is to shift the curve to the right. This means a higher percentage of over-corrected patients but also a much higher percentage achieving uncorrected vision of 20/20. Patients on the left side of the graph can be re-treated for their residual myopia.

Because there are fewer options after being over-corrected, it is unethical to shift the curve to the right, even though a much higher percentage achieves 20/20. Also, even the patients who are on the right side of the graph and do see 20/20 now, will become presbyopic and will not have clear vision at any distance without glasses in the future. FIG. 2(b) demonstrates the usefulness of a refractive procedure that can be partially reversed after the initial procedure, such as the Adjustable Corneal Ring (ACR) of the present invention. Assuming that a refractive procedure can easily be partially reversed after the initial procedure, the curve can safely be shifted to the right. All the patients in FIG. 2(b) who significantly overcorrected can undergo a partial reversal. This results in a much narrower distribution of patients and a distribution of patients that surrounds emmetropia.

Another fortunate but coincidental outcome of this particular refractive strategy is that greater than 90% of patients may have an uncorrected visual acuity of 20/20 even prior to the adjustment procedure.

For years it has been thought that refractive surgery with intracorneal implants could be used in the correction of ametropia. Early techniques included lamellar removal or addition of natural corneal stromal tissue, as in keratomileusis and keratophakia. These required the use of a microkeratome to remove a portion of the cornea followed by lathing of either the patient's (keratomileusis) or donor's (keratophakia) removed cornea. The equipment is complex, the surgical techniques difficult, and most disappointingly, the results quite variable. The current trend in keratorefractive surgery has been toward techniques that are less traumatic to the cornea, that minimally stimulate the wound healing response, and behave in a more predictable fashion. The use of alloplastic intracorneal lenses to correct the refractive state of the eye, first proposed in 1949 by Jose Barraquer, have been plagued with 10 problems of biocompatibility, permeability to nutrients and oxygen, corneal and lens hydration status, etc. Other problems with these lenses included surgical manipulation of the central visual axis with the concomitant possibility of interface scarring.

More recent efforts toward the correction of refractive errors have focused on minimizing the effects of the wound healing response by avoiding the central cornea. There have been multiple attempts to alter the central corneal curvature by surgically manipulating the peripheral cornea. These techniques are discussed because of their specific relevance to this invention.

Zhivotovskii, D. S., USSR patent no. 3887846, describes an alloplastic, flat, geometrically regular, annular ring for intracorneal implantation of an inside diameter that does not exceed the diameter of the pupil. Refractive correction is accomplished primarily by making the radius of curvature of the surface of the ring larger than the radius of curvature of the surface of a recipient's cornea in order to achieve flattening of the central area of the cornea. Surgical procedures for inserting the ring are not described. A. B. Reynolds (U.S. Patent No. 4,452,235) describes and claims a keratorefractive technique involving a method and apparatus for changing the shape of the optical zone of the cornea to correct refractive error. His method comprises inserting one end of a split ring shaped dissecting member into the stroma of the cornea, moving the member in an arcuate path around the cornea, releasably attaching one end of a split ring shaped adjusting member to one end of the dissecting member, reversibly moving the dissecting member about the path, and thereby pulling the adjusting member about the circular path, made by the dissecting member, withdrawing the dissecting member, adjusting the ends of the split-ring shaped adjusting member relative to one another to thereby adjust the ring diameter to change the diameter and shape of the cornea and fixedly attaching the ring's ends by gluing to maintain the desired topographical shape of the cornea.

A major advantage of this ring was that a very minimal wound healing effect was expected. A marked corneal wound healing response would decrease the long-term stability 11 of any surgical refractive procedure. However, there are two distinct problem areas affecting the refractive outcome of surgical procedures treating ametropia:

1. The first problem is concerned with the ability to predetermine the shape and size of an implant that will lead to a certain refractive outcome. In RK or PRK, retrospective studies have been performed that led to the development of nomograms which predict that a certain depth cut or a certain ablation amount will result in a predictable amount of correction. In the case of the ring, eventually nomograms will be developed that can be used to predict a given refractive correction for a given thickness or size of the ring. However, these nomograms can never fully account for individual variability in the response to a given keratorefractive procedure.

2. The refractive outcome also depends on the stability of the refractive correction achieved after surgery. To reiterate, the advantage of the ring would be the stability of the refractive outcome achieved because of a presumed minimal wound healing response. This decreases the variability of the long-term refractive outcome but still does not address the problems posed in the first problem area, ~ the inherent individual variability, in that while the outcome may be stable, it may very well be an inadequate refractive outcome that is stable.

Another unaddressed issue is that even with the implant, surgeons will aim for a slight under-correction of myopia because, in general, patients are more unhappy with an over-correction that results in hyperopia. Again, the refractive outcome may be more stable than in RK or PRK but it may be an insufficient refractive result that is stable.

Simon in U.S. Pat. No. 5,090,955 describes a surgical technique that allows for modification of the corneal curvature by inter-lamellar injection of a synthetic gel at the corneal periphery while sparing the optical zone. He does discuss an intra-operative removal of gel to decrease the peripheral corneal volume displaced and thus adjust the final curvature of the central corneal region.

Siepser (U.S. Pat. No. 4,976,719) describes another ring-type device to either flatten or steepen the curvature of the cornea by using a retainer ring composed of a single surgical 12 wire creating a ring of forces which are selectively adjustable to thereby permit selective change of the curvature of the cornea, ~ the adjustable means comprising a turnbuckle attached to the wire.

There are several mechanisms by which peripheral manipulation of the cornea affects anterior corneal curvature. The cornea, like most soft tissues, is nonlinear, viscoelastic, nonhomogeneous, and can exhibit large strains under physiologic conditions. The whole eye is geometrically extremely complex and the biomechanics technique capable of systematically modeling this reality is the finite element method which assumes small strains (a measure of deformity), homogeneity, and linear elastic behavior. Two simple mechanisms will be briefly described.

A simple example is helpful in understanding the first mechanism. Assume a loose rope R between two fixed points PI and P2 as in FIG. 3(a), which forms a curve, the lowest point 10 being in the middle. Referring to FIG. 3(b), a weight W placed on the rope between the middle point P and one fixed point will cause the central portion of the rope to straighten 11. The cornea 13 demonstrated in FIG. 3(c) and FIG. 3(d) behaves similarly, the two fixed points, PI and P2, analogous to the limbus of the eye and the weight W similar to the intrastromal implant 15 which, when inserted in the cornea in surrounding relation to the corneal central optical zone, causes the corneal collagen fibers to deviate upwards above the implant, and downwards below the implant. In essence, this deviation of the cornea around the peripheral implant caused by volume displacement in the peripheral cornea results in other areas of the cornea losing "slack", or relatively straightening 14 as shown.

Mechanical expansion of a peripherally implanted corneal ring also flattens the central corneal curvature whereas constriction of the corneal ring steepens the central corneal curvature. A constricting or expanding corneal ring is likely to cause a less stable refractive outcome because the inward or outward forces of the implant against the corneal stroma may gradually cause further lamellar dissection and dissipation of the forces. A more consistent outcome is likely to be achieved with varying the volume displaced in the peripheral cornea as described by Simon. 13

The second mechanism is aptly described by J. Barraquer in the following quote. Since 1964, "It has been demonstrated that to correct myopia, thickness must be subtracted from the center of the cornea or increased in its periphery, and that to correct hyperopia, thickness must be added to the center of the cornea or subtracted from its periphery." Procedures involving subtraction were called 'keratomileusis' and those involving addition received the name of 'keratophakia'. Intrastromal corneal ring add bulk to the periphery and increasing the thickness of the ring results in a more pronounced effect on flattening of the anterior corneal curvature by "increasing (thickness) in its periphery".

In the February, 1991 issue of Refractive and Corneal Surgery, T. E. Burris states that "the thickening effects of ICR implantation may prove most important for maintenance of corneal flattening" and that "new ICR designs must take into consideration thickness effects on corneal flattening".

The ideal keratorefractive procedure allows all the advantages of eyeglasses or contact lenses, namely, being able to correct a wide range of refractive errors, accuracy or predictability, allowing reversibility in the event that the refractive state of the eye changes and it becomes necessary to adjust the correction again, yielding minimal complications, and associated with technical simplicity, low cost, and being aesthetically acceptable to the patient. The goal of refractive surgeons should be to achieve 20/20 uncorrected visual acuity with long-term stability in greater than 95% of patients. None of the currently available refractive surgery procedures generate this degree of accuracy or stability.

Once again, an easy procedure to post-operatively fine-tune the refractive correction and corneal curvature which is often influenced by changes in corneal hydration status, wound healing responses, and other unknown factors, is not available. Each of the techniques described suffers from a limited degree of precision. In this disclosure of the present invention, an easy method to adjust the refractive outcome after the corneal curvature has stabilized, a method that is minimally invasive, a method causing minimal stimulation of the wound healing processes, allowing repetitive adjustments as deemed necessary, and being almost completely reversible is described. It may make moot the pervasive issue of 14 unpredictability and make obsolete the application of procedures which rely heavily upon nomograms to predict refractive outcome and are thus unable to adequately account for an individual's variable response to the procedure.

4. SUMMARY OF THE INVENTION

The present invention concerns the use of an adjustable intrastromal device adapted for implantation in the cornea and formed of a flexible hollow shell composed of a material such as a silicone or urethane polymer, with an annular chamber that may be augmented with a biocompatible filler material such as polymethylmethacrylate (PMMA). The filler material can be any biocompatible material of a varying shape or length but preferably is ring-shaped and a flexible elongated strand-like filament. The device is filled with a predetermined amount of the biocompatible material described, and implanted in the cornea in surrounding relation to the optical zone of the cornea. The corneal curvature is then adjusted by complete removal of one or more strands thus modifying the volume of the device displacing peripheral corneal tissue in a discrete fashion and resulting in steepening of the corneal curvature and a myopic shift. This relatively simple adjustment for refractive correction can be performed with surgical instruments commonly available and requires minimal post-operative manipulation of the cornea and the implanted device. The apparatus of the invention is an adjustable implantable device including an outer membrane forming an enclosure for receiving a filler material such as multiple strands and adapted to be inserted into the interlamellar space of the corneal stroma for the purpose of correcting refractive error. The volume displaced by the device is easily modified on multiple occasions following the initial surgery of implantation and thus allows for adjustment of the refractive outcome at a later date without necessitating the removal of the implanted device. 15

5. BRIEF DESCRIPTION OF THE DRAWINGS

FIG. 1(a) is a schematic representation of a horizontal section of the human eye; FIG. 1 (b) is a schematic representation showing how the light rays focus in front of the retina of the eye in the condition of hyperopia;

FIG. 1(c) is a schematic representation showing how light rays focus in front of the retina of the eye in the condition of myopia;

FIG. 2(a) is a graph representing the typical distribution of refractive surgery outcomes;

FIG. 2(b) is similar to FIG. 2(a) but demonstrating the expected distribution of refractive surgery outcomes with hyperopic overcorrection;

FIG. 2(c) is similar to FIG. 2(b) but demonstrating the expected distribution of the Adjustable Corneal Ring outcomes following an adjustment procedure, showing a distribution of patients more tightly grouped around emmetropia;

FIG. 3(a) is a schematic illustration for showing a rope suspended at its ends between two fixed points;

FIG. 3(b) is a schematic illustration which shows the rope in FIG. 3(a) with the force of a weight applied to the rope between its midpoint and one of the fixed points; FIG. 3(c) is a schematic illustration showing the cornea of an eye wherein the cornea is fixedly attached at diametrically opposed points on the surrounding limbus;

FIG. 3(d) is an illustration similar to FIG. 3(c) but showing the curvature effects produced on the cornea because of the presence of an intracorneal ring or device in the cornea; FIG. 4(a) is a plan view of the intracorneal device of the invention wherein the device has been severed by a radial cut;

FIG. 4(b) is a plan view of the device of the invention within the cornea of the eye;

FIG. 4(c) is an enlarged diametral cross section view as taken along the section line 4c - 4c in FIG. 4(a); 16

FIG. 5(a) is a radial cross-sectional view of the intracorneal device of the invention demonstrating strands with a circular cross-sectional area;

FIG. 5(b) is a radial cross-sectional view of the intracorneal device of the invention similar to FIG. 5(a) after three strands have been removed from the device, demonstrating a smaller radial cross-sectional area delimited by the outer shell following removal of the strands;

FIG. 5(c)-5(h) are radial cross-sectional views of the intracorneal device of the invention demonstrating strands of varying size and shape within the intracorneal device and also demonstrating a decrease in radial cross-sectional area delimited by the outer shell following removal of the various strands;

FIG. 6(a) is a radial cross-sectional view of the device of the invention and showing typical dimensions thereof;

FIG. 6(b) is a radial cross-sectional view of the device in FIG. 6(a) wherein the interior of the device is filled with several strands; FIG. 6(c) is a radial cross-sectional view similar to FIG. 6(b) but showing the device interior filled with a lesser number of strands which have a greater cross-sectional area;

FIG. 7(a)-7(d) are radial cross sectional views of modified forms of the outer shell of the device of the invention;

FIG. 8 is a plan view of the device showing possible strand connection placement along the strands within the device of the invention;

FIG. 9(a) is an illustration similar to FIG. 3(c) showing the effect produced by constriction of the intracorneal ring of the invention after its implantation in the cornea;

FIG. 9(b) is an illustration similar to FIG. 9(a) showing the central flattening of the cornea after the constriction by the intracorneal ring has been relieved; FIG. 9(c) is an illustration similar to FIG. 3(c) but showing the curvature effects produced on the cornea because of the presence of an intracorneal ring or device in the cornea; 17

FIG. 9(d) is an illustration similar to FIG. 9(c) showing the steepening of corneal curvature following removal of strands from the intracorneal ring;

FIG. 10(a) is a schematic plan view of the orientation and form of a plurality strand material which ma be inserted into the interior of the device; the spacing therebetween exaggerated for purposes of illustration;

FIGS. 10(b) and 10(c) are cross sections of the device of the invention as taken along the section lines 10b - 10b and 10c - 10c in FIG. 10(a), respectively;

FIGS. 10(e), 10(f), and 10(g) show variations in the configuration and orientation of strands which are suitable for insertion in the device; FIG. 11 (a) is a schematic showing a plan view of the device of the invention wherein an arc segment has been inserted in the device; FIG. 11(b) is a view in radial cross section of the device in FIG. 11(a) as taken along the section line 1 lb - l ib;

FIG. 11(c) is a view in radial cross section of the device in FIG. 11 (a) as taken along the section line l ie - l ie; FIG. 12 is a perspective view of the device of the invention with an opening in the anterior shell which is implanted in the cornea; the cornea has an incision made anterior to the opening of the shell and the removal of a strand from the implanted device is demonstrated;

FIG. 13(a) is an enlarged schematic illustration showing the cornea of an eye wherein the cornea is fixedly attached at diametrically opposed points on the surrounding limbus; FIG. 13(b) is an enlarged illustration similar to FIG. 13(a) but showing the curvature flattening effects produced on the cornea because of the presence of an intracorneal device in the cornea;

FIG. 13(c) is an enlarged illustration similar to FIG. 13(b) but showing the partial re- steepening effects produced on the corneal curvature following removal of two strands from the intracorneal device.

6. DETAILED DESCRIPTION OF THE INVENTION 18

Referring more particularly to the drawings, there is shown in FIG. 4(a) the apparatus of the invention which comprises an adjustable device 15. The device 15 forms an enclosure for receiving a filler material which is easily removable, such as polymethylmethacrylate, nylon, polyester, polypropylene, polyimide, or other polymeric materials such as fiuoropolymer resins. The device filler material can be any biocompatible material but preferably is a flexible, filamentous structure that may be constructed from a resilient polymeric substance such as that described above.

The device 15 comprises a tubular shell 16 made of a flexible material, such as a silicone, acrylic or urethane polymer and in FIG. 4(a) is shown as a split donut shape. The major axis of a transverse cross section of the device 15 is such that it corresponds to the slope of the corneal arc of the anterior pole of the cornea, thus forming the conic section 25. This angle is approximately 25 to 35 degrees as shown in FIG. 4(c). The shell material has adequate stiffness such that the device will maintain its generally circular shape in plan view when sufficiently filled and also have adequate flexibility to allow an increase in thickness with filling as shown in the cross section view of FIG. 5(a) and flattening with removal of strands as shown in FIG. 5(b). The shell of the device must have sufficient structural integrity, strength and flexibility to generally maintain its circular shape and be expandable. Its composition material may be similar to that used in producing foldable or deformable intraocular lenses such as a silicone polymer, urethane polymer or acrylic polymer, or that material used in soft contact lenses or materials such as fiuoropolymer resins.

The essence of the invention is an annular device that is implanted intrastromally in the peripheral cornea thus inducing flattening of the central corneal curvature and that is designed such that the amount of peripheral corneal tissue it displaces can be easily modified at a later time in a minimally invasive fashion to thus adjust the refractive effect. The two essential factors which are crucial to the feasibility of the device are 1) biocompatibility without significant biodegradation of the device and 2) collapsibility of the outer annular shell following strand removal. Ideally, the pressure from the surrounding corneal stromal 19 tissue is sufficient to cause collapse of the outer shell following strand removal from the implanted device.

The collapsibility of the outer annular shell following removal of strands from the implanted device is important in effecting the refractive adjustment. FIG. 5 demonstrates what is meant by collapsibility of the shell; a decrease in radial cross-sectional area and thickness (from 31 to 32) following removal of strands from the shell of the device. Factors that determine ease of collapsibility of the shell following removal of strands from the shell include shell wall thickness, wall material composition, flexibility of the material, the memory of the material, and these characteristics of the material at the angle of the inner 26 and outer 27 diameters of the shell. The wall material at the inner 26 and outer 27 diameter angles provide the bulk of the structural integrity which resists shell collapse or flattening. Collapsibility of the shell can be promoted by decreasing wall thickness, thus decreasing the structural mass at the inner and outer diameters which maintain the angle. Collapsibility can be facilitated by forming the shell of a material softer in composition or with less flexural strength.

In another useful embodiment, the outer shell of the implantable device is composed of a biocompatible, porous polymer material such as a microporous polypropylene tube. The characteristics of the porous shell are similar to that already described including sufficient flexibility to allow the thickness of the device to decrease when the biocompatible filler material is removed. Advantages of a porous shell include improved nutrient diffusion to the anterior corneal stroma. Another method to allow improved nutrient diffusion to the anterior corneal stroma is to place openings in the shell of the implant. The openings may be multiple, radially or longitudinally oriented, of variable length and width and situated on the anterior or posterior surface of the device. The composition material of the strands may be any suitable plastic or polymer material such as that used in producing foldable or deformable lenses, silicone polymers, urethane polymers, acrylic polymers, polyesters, fiuoropolymer resins, or materials used in soft contact lenses. It will be understood by those skilled in the art that, among polymers of 20 acrylic esters, those made from acrylate ester monomers tend to have lower glass transition temperatures and to be more flexible than polymers of methacrylate esters. Examples of other medical devices composed of materials which be suitable for the shell of this invention include vascular graft tubing, dialysis tubing or membrane, blood oxygenator tubing or membrane, ultrafiltration membrane, intra aortic balloon, catheter, suture, soft or hard tissue prosthesis, artificial organ, and lenses for the eye such as contact and intraocular lenses. The strands are comprised of a biocompatible material which is preferable a flexible solid material. An example of a suitable biocompatible material is polymethylmethacrylate. There are many other suitable polymeric materials, including but not limited to epoxy resins, polyamides, polyacetals, polycarbonates, polyethers/ether ketones, polyolefins, polyurethanes, polyvinylpyrrolidone, natural or synthetic rubbers, polysulfones, copolymers, and combinations of the above.

Referring to FIG. 5, the cross section of the strand may be of various geometric shapes including circular 30, oval 33, rectangular 34 , square 35, or triangular. The cross- sectional area of the strand can vary in dimension along its length. The device may contain one or more strands, each of which is removable at a later time. The cross section of the device 30 as taken in a radial plane through the center of the implant is elliptically shaped.

The different embodiments shown in FIG. 5 can each be modified to provide a number of sub-embodiments by altering variables such as the composition material of the device wall, manner of device connection, type of ring filler material, and cross-sectional surface parameters of the device, e.g., forming the device from cross sections in the form of a circle, square, rectangle, triangle, oval, etc. The two ends 18, 19 of the device are squared off so that they may juxtapose each other as shown in FIG. 8 and may be fixably joined at the time of surgery by such methods as suturing or gluing.

The device 15 is adapted to be implanted into the peripheral stromal cornea. It is of a thickness and geometry such that when implanted it alters the central corneal curvature 21 without intruding into the central optical zone of the cornea and without decreasing the diffusion of nutrients to the central cornea. It is of a size such that it can be readily inserted into the peripheral human cornea intrastromally and consists of a flexible material which is biocompatible, and more specifically, compatible with ocular tissues. The dimensions as shown in FIG. 6(a) include a device thickness (22) of OJ-1.5 mm, width (21) of 0.4 to 2.0 mm and an outer over-all diameter (23) of 4.00 to 11.0 mm. The thickness of the shell 16 of this device 15 may be varied as shown in FIGS. 7(a)-7(d). The shell wall thickness 24 can vary from approximately 0.001 mm to 0.30 mm. The device may contain only one or multiple strands 30 of varying diameter and composition. The strands may be composed of a biocompatible material commonly used in ophthalmic surgery such as polymethylmethacrylate, nylon, polyester, prolene, or polypropylene and can vary from 0.02 mm in diameter to 1.0 mm in diameter. The strands may be clear or colored. The strand may be marked towards the head and tail end of the device to aid the surgeon in adjusting the tension when connecting the ends of a strand. The strand may have a pre-fabricated loop 66, 67 at one end which would facilitate removal of the strand by using an instrument having a small hook at the operative end with which the loop can be snared. Instead of a loop, the strand end may have some other configuration such as a rounded or thickened end which would also facilitate grasping the strand. The loop also aids in preventing surrounding strands from being pulled out simultaneously by providing resistance at the open end. The two ends of the strand are not necessarily connected.

The device of the invention is designed to be implanted in the cornea of the eye to alter the external curvature of the central optic zone of the cornea without encroachment into the optic zone. It is comprised of a hollow device with a variable internal volume such that the central optic zone may be flattened by disconnecting a strand that has been connected with tension, or steepened in curvature by strand removal to an amount suitable to provide the refractive correction needed and allowing for adjustment of over-correction or under- correction of the refractive error. 22

A typical adjustable device 1 of the invention is shown in FIG. 6. The width of its outer diameter is 0.80 mm, overall thickness is 0.35 mm, and shell wall thickness is 0.005 mm. A device of this size is expected to flatten the central cornea by approximately 3.5 diopters. To calculate the number of strands which will comfortably fit and the diopter change with removal of each ring, the following is assumed. The internal cross-sectional area of the oval-shaped device is approximately 0.20 mm squared. Since this volume cannot be completely filled with strands that have round cross-sections — there are spaces between the round strands, the area that will be occupied by a strand is 78.5% ideally. Referring to FIG. 6(b), approximately six (0J75 mm diameter) strands 43 will fit into this space. The thickness of this device is approximately 0.35 mm (42). Referring to FIG. 6(c), removal of three strands results in a flattening of the device by approximately 0J5 mm with a resultant thickness of 0.20 mm (44). A 0J5 mm decrease in thickness of the implanted device is expected to steepen the central corneal curvature by 1.50 diopter. Since three strands are removed, the average diopter change for each 0.175 mm strand removed from this typical embodiment is approximately 0.5 diopter.

There are parameters which may be altered to effect the desired result. For example, if one strand removal does not effect the refractive adjustment desired, the strand size may be increased or decreased in future device designs. If a thicker ring is required, the size of the strands may be increased, the number of strands increased, or the shell wall thickness increased. There are many variations on the theme but the main concept remains that strand removal minimally disturbs the cornea thus bypassing the wound healing response yet allowing a small discrete change in peripheral corneal tissue volume displaced by the device in a consistent fashion symmetrically or asymmetrically around the peripheral cornea.

Given an initial myopic patient, the outcome can be overshot by 50% of the initial refraction and the hyperopia still reasonably managed by strand removal alone. Over- treatment resulting in hyperopia is a significant disadvantage in most kerato-refractive procedures. In radial keratotomy the wound healing processes occur over a period of years and there is often a progressive hyperopia. Patients who become symptomatically hyperopic 23 after surgery are extremely unhappy. Therefore most surgeons use nomograms that attempt to achieve a slight under-correction. Concerning photorefractive keratectomy, in one study, it was found the main reason patients did not have their second eye corrected with PRK (given that their first eye was corrected with PRK) was because of dissatisfaction with the hyperopia in their operated eye. The technique described herein easily corrects overcorrection hyperopia.

The number of strands within the device and the radial cross-sectional size of the individual strands can be selected pre-operatively to minimize the number of strands that needs to be removed to effect a reasonable diopteric change. A strand may have a radial cross-sectional diameter of 0.02 to 1.0 mm in diameter. Formulas can be developed that predict the diopteric change expected following removal of a single strand of a given radial cross-sectional diameter while taking into consideration such factors as initial refractive error, refractive change effected by the particular implant, number of strands in the initial device, and corneal device diameter. The radial cross-sectional size of strands is chosen such that removal of a single strand will effect a steepening in corneal curvature by approximately 0.1 to 1.00 diopter. Strands of varying radial cross-sectional diameter may be present in a single device and identified by a mark or color to enable the surgeon to determine the strand diameter to make the adjustment procedure more precise. If there is a residual hyperopia of greater than 1.0 diopter after a device is implanted for the correction of myopia, the adjustment procedure would require removal of a larger strand diameter. Alternatively, a greater number of small diameter strands may be removed to achieve the same adjustment.

Depending on the amount of refractive error, an appropriate embodiment varied in shape, size, circumference, strand size, strand composition and number of strands, are chosen. The flexible shell 16 containing the strand material can also be varied as shown by the embodiments of implant illustrated in FIGS. 7(a)-7(d). The choices include:

1. The absence of a supporting polymethylmethacrylate (PMMA) backbone.

2. PMMA or other stiff physiologically acceptable polymer backbone reinforcing the inner circumference of the device wall as shown in FIG. 7(d) . The thickened areas 47 24 shown in FIG. 7(d) may be increased thickness of the flexible material composing the walls or it may be the stiff polymer backbone mentioned above. During surgery, the inner circumference backbone could be appropriately adjusted and fixed with suture or glue, with gross adjustments aided by the use of a keratometer. 3. PMMA or other stiff polymer backbone reinforcing the outer circumference 44 of the device wall as shown in FIG. 7(a). 4. Support of both inner and outer circumferences.

The size of the device chosen should be such that the range of over-correction or under-correction secondary to individual variability of response to surgery may be comfortably corrected (not requiring removal of all of the strands) by the methods described. The maximal thickness, circumference, and type of supporting backbone are chosen prior to insertion of the implant. The ideal embodiment, given the preoperative refractive state and other pertinent data, is chosen prior to operating and then that embodiment further manipulated as necessary to determine the ideal curvature. The device is inserted into the peripheral cornea at an adequate depth and then further adjusted in order to more precisely adjust the shape of the cornea and focus the light entering the eye on the retina. The intra- operative keratoscope or automatic keratometer may be helpful. However, intra-operative curvature measurements in surgeries involving the cornea have not been shown to be predictably reproducible and so later post-operative adjustments of the device will be the most useful in adjusting the refractive outcome.

The device is implanted into a circular lamellar channel formed at 1 to 2/3 corneal depth with a circular dissecting instrument that requires only a small midperipheral corneal incision. A knife is used to make an approximately 2 mm radial incision beginning at 2.5 to 3.5 mm from the corneal center. The surface of the cornea is cut only at this incision. A Suarez spreader or other lamellar dissecting instrument is introduced into the bottom of the incision and a small lamellar pocket created. Application of a suction fixation ring is positioned around the limbus and used to fix the globe while an 8-9 mm outer diameter 25 lamellar channeling tool introduced through the incision into the lamellar channel is rotated to produce a 360 degree channel around the corneal mid-periphery at Vz to 2/3 corneal depth. After the channeling tool is removed, a circular endoscopic-type forceps is inserted into the same channel and rotated 360 degrees such that the forceps tip emerges from the radial incision. One end of the device is inserted into the forceps, the forceps jaws closed thus gripping the device, the circular forceps rotated until the device is progressively pulled into place. Other instruments may be used to insert the device, such as a circular instrument with a hook at the leading end. The head and tail of the device are brought together and may be fixed together with suture or glue. In summary, adjustment or choice of device size, shape, width, shell thickness, circumference, and other factors affecting the corneal curvature and refractive outcome, occurs in three distinct temporal stages:

1. Preoperatively, the above mentioned variables and presence or absence of a supporting backbone are chosen using nomograms developed from retrospective studies as a guide to the selection of each variable.

2. Intra-operatively, the device tightness is adjusted as necessary, aided by the use of the intra-operative keratoscope if necessary. The strand passing completely around the implant may be tightened and connected at various tensions, keeping the following in mind: a. Adjusting the volume of the implant probably results in a more predictable change in corneal curvature than attempting to adjust corneal curvature by either the application of tension or the removal of tension, b. If a hyperopic correction is required, circular radial forces will be necessary to maintain the corneal curvature and either the head and tail of the device connected at tension or one or more strands connected at tension. In other words, strands are connected at tension only for the correction of hyperopia.

3. Post-operative adjustments. Simple, easily performed postoperative adjustments, which avoid the complications of re-operation concomitant with most kerato- 26 refractive procedures, are rendered feasible by this mechanism of adjustment. This postoperative adjustment can compensate for an inadequate preoperative implant choice, for corneal hydration intra-operatively which results in a different corneal curvature after corneal hydration status changes post-operatively, for an unexpected wound healing response in the periphery to the implant, and for later refractive changes caused by unknown factors. This postoperative adjustment is made possible by a flexible corneal device containing several strands which can easily be removed thus modifying the volume of the device and resulting in increased corneal curvature. Strand removal from the device minimally disturbs the corneal stromal-device interface compared to removing the device itself, thus minimizing the effects wound healing and edema will have on the adjustment. This postoperative adjustment appears to be a necessary adjunct to any method that seeks to meet the criteria for the ideal kerato-refractive procedure. If the refractive outcome is not ideal after the initial implant procedure, these are the steps that may be taken: a. As demonstrated in FIGS. 9(a) and 9(b) if the corneal curvature is too steep 50 and the patient has residual myopia, and if there is a strand connected at tension it may be cut thus releasing some of the constricting circle of forces and thus flattening the corneal curvature 52. When comparing the corneal ring diameter, it can be seen that the diameter is smaller 51 prior to releasing the tension and larger 53 afterwards. Ideally, the strand is cut near the initial incision site. The strand may be cut with a sharp needle, knife, or even with a laser. If still inadequate, more than one strand may be cut. The two ends of the device are unlikely to drift even if all the strands are cut. In the case that strand cutting results in excessive flattening, one of the strands may be completely removed from the device and eye, resulting in a relative decrease in volume of the device with a concomitant steepening of the corneal curvature. If, in the unlikely event that a ring is difficult to remove, that strand may be cut at 180 degrees away and then each half removed through the initial incision. Corneal curvature may be flattened by another 27 method. A strand or other solid biocompatible material within the device may be attached to a larger diameter strand such that as the strand within the device is removed, the larger strand is progressively pulled into place thus increasing the volume of the device and flattening the anterior corneal curvature. b. As shown in FIGS. 9(c) and 9(d) , if the corneal curvature is too flat 54 after surgery, strands that have relatively little or no tension may be cut and removed, thus steepening 56 the corneal curvature as shown in FIG. 9(d) with a myopic shift as described above. This is why some of the strands are connected with little or no tension at the initial surgery. If there is over-correction of the adjustment and removal of the strand results in excessive steepening, a strand with tension may be cut and left in place, thus flattening the cornea or a larger new strand may be pulled into place.

In a simple adaptation of this technique, the device may be used to correct astigmatism. Curvature variation of the anterior surface of the cornea is responsible for the majority of cases of astigmatism. The light rays converge upon more than one plane and no one principal focus is formed. Astigmatism ordinarily depends on the presence of toroidal instead of spherical curvatures of the refractory surfaces of the eye. To correct astigmatism certain areas of the cornea must necessarily be corrected to a greater degree than other areas. The device can be varied in thickness along the circumference of the device with the sections of the device having increased volume corresponding to the areas of the cornea having a steeper slope and requiring greater correction. In the illustration of FIG. 10(a), the strand 60 completes almost 360 degrees within the device. Another partial strand 61 is shorter and is absent at approximately 4-6 o'clock in the drawing. The strand 62is the mirror image placement of (33), and is absent at 6-8 O'clock. Strand 63 folds over itself twice in the area of increased thickness. Strand 64 is the mirror image placement of 63. As illustrated by the greater volume 67 of the cross section of the implant in FIG. 10(b) as compared to the size of the cross section 68 in FIG. 10(c), the areas with more strands have augmented volume by up 28 to 50% and thus allow for the differential correction required in astigmatic conditions. If the astigmatism is overcorrected, strand 63 and 64 may be pulled until the loops 65 and 66 are removed and then cut at the point where the strand emerges from the device. The removal of the loops 65 and 66 reduces the ratio of the larger area to smaller area of the implant from 6/3 to 4/3. In the event that the astigmatism is under-corrected, strand 60 may be completely removed, increasing the ratio from 6/3 to 5/2. Many different variations on this theme are possible, with some examples shown in FIGS. 10(e)- 10(g). The variations can occur in the flexible device which may have a supporting backbone of PMMA or other polymeric material. The thickness of the device outer shell or membrane may also be varied. Strand adjustment is based on principles previously discussed. The strands may be manipulated through the initial incision site or at any other point along the circumference of the device. The device may have an opening in the anterior shell through which the strands may be adjusted or removed. The strands are not necessarily 360 degrees in length and they may also be cut at their mid-length so as to facilitate their removal at a later date. Referring to FIG. 11, in another embodiment, the device may have areas of increased thickness formed by the presence of a thicker arc segment 71 that is inserted into the hollow implant shell 16 and that may be composed of the same material as the implant wall or a stiffer substance such as PMMA. This thicker arc segment 71 may have various transverse cross-section shapes, preferably conforming to that of the device cross-section and more than one thick arc segment 71 may be provided. It may be 10 to 360 degrees in chord length. The ends 73 of the arc segment are gradually tapered so that the thickness at the segment ends approximates the thinnest areas of the device. The thickness of the segments can be varied so that the thick section 75 of the device may be several times the thickness of the thinnest sections 74 of the device. 120 to 180 degrees away at the opposite side of the device, there is a similar arc segment 71 that may be similar in length and thickness, but not necessarily so. A strand 72 as demonstrated in FIG. 11(a) connects the two arc segments to each other. The axis of astigmatism may be adjusted at a later date through the initial incision site by pulling the segments in one direction or the other, thus changing the position of the arc segments 29 within the device chamber and with respect to their direction from the central axis of the device. An individual arc segment may have a strand that connects one end to the other such that each arc segment can be adjusted independently. As previously stated, many different variations on this theme are possible. This particular sub-embodiment may be used with any of the previous processes described. An important advantage of this design is the ease of reversibility of the procedure. The procedure may be completely reversed by the surgical removal of the device or the refractive effect may be partially altered as previously described. The adjustments themselves may be reversed.

The strands can be varied in thickness along its circumference with the thicker strand areas displacing a greater amount of peripheral corneal tissue and corresponding to the areas of the cornea having a steeper slope and requiring greater correction. The U.S. Patent, "Adjustable Corneal Arcuate Segments", filed by this author in 1998, is included herein in its entirety by reference. It is anticipated by this author that the various segment shapes and arc lengths described in the "Adjustable Corneal Arcuate Segments" patent can also be applied to the device of the current invention in the treatment of astigmatism, hyperopic astigmatism, myopic astigmatism, and the adjustments thereof.

Referring to FIG. 12, if there is overcorrection after implantation of the device and it is determined that strands need to be removed from the outer shell, the strands are removed as follows. Strand removal from the cornea can be accomplished at the time of surgery with an intraoperative keratoscope to guide the refractive change. However, most of the benefit from adjusting the peripheral corneal volume displaced by the strands will occur at a time after corneal edema from the surgical procedure has resolved and the refractive effect has stabilized. There may or may not be an opening placed in the anterior shell of the device. The opening 17 in the anterior shell 16 of the device may have various geometric shapes such as circular, oval, a concentric slit incision, or a radial slit incision. There may be more than one opening, the placement of the opening along the anterior shell of the device may vary, and various combinations of the different geometric shapes may be present. As seen in FIG. 12, if there is an opening placed, the corneal incision 76 is made anterior to the opening for 30 the removal of strands. Strand removal is accomplished by making an incision, either radial or horizontal, into the anterior cornea at a site near the strands. This incision may be made at the initial incision site or at any other site on the cornea near the strands. Preferably, the incision site is made 180 degrees away from the initial incision site at a location 180 degrees away from the two ends of a strand so that the original incision is left undisturbed. Strands can be removed from the incision site by a forceps type instrument or an instrument with a small hook at the end such as an iris hook. One or more strands are removed 77 as necessary to adjust for the overcorrection. If no opening in the anterior shell is present, when the incision is made anterior to the segment the incision can be made sufficiently deep to cut into the anterior shell of the device. Post-operative adjustments are rendered simple and easy requiring only strand removal and avoid the complications of re-operation concomitant with most kerato-refractive procedures. The decrease in amount of peripheral corneal tissue displaced by the device following removal of strands is

It is therefore to be appreciated that by use of the present invention, the disadvantages of traditional refractive surgery procedures are avoided, such as 1) progressive hyperopia with radial keratotomy. Hyperopia in any refractive procedure is a generally worse outcome because the patient does not have clear vision at any range and because hyperopia is much more difficult to correct. The described procedure is particularly well suited to adjust a hyperopic refractive outcome. 2) The irreversibility of radial keratotomy and laser ablation surgeries. 3) Surgical manipulation of the central visual axis with the potential for scar and stromal haze formation following laser ablation procedures. 4) The need for chronic use of steroid drops with its accompanying complications such as cataract and glaucoma. 5) Regression with laser ablation procedures, especially following re-operation. 6) Reduction of positive sphericity with RK and laser ablation which may result in increased image aberration. 7) The invasiveness of laser in-situ keratomileusis. 8) Lack of precision and predictability with all current procedures. 9) The possible need for repetitive explanting and 31 implanting of ICR' S, which may cause shearing of corneal peripheral channel lamellae with associated decrease in effect and also scar formation.

The last point requires further elaboration. Methods to adjust ring thickness have been described in the prior art. These methods are only discussed in relation to adjusting the ring thickness during implantation, not post-operatively. Attempts to adjust the thickness of the ring are most useful after corneal curvature has essentially stabilized. Adjustments of devices that have been described in the prior art would necessarily require rotation of the ring with resultant shearing of the complete corneal-ring interface. Rotation of the whole ring would be required to allow more or less overlap of the individual ring parts thus increasing or decreasing ring thickness. This shearing of the corneal tissue in the immediate vicinity of the ring may alter the corneal curvature in an unpredictable fashion and probably also cause more scarring with possible unpredictable long-term effects. In the embodiment that is described in this article, the device volume is adjusted with less disturbance of the surrounding tissue. There is some disturbance of the corneal-strand interface but only in the vicinity of the strand that is removed while the other strands and surrounding cornea is relatively undisturbed.

Also, the additional shearing of the cornea caused by re-implanting another complete ring is avoided. In conclusion, a slight decrease in volume of peripheral corneal tissue displaced by the adjustment described is not only easier to perform, but also expected to have a much more predictable effect. Most refractive surgery procedures use nomograms to calculate the correction required and cannot completely account for an individual's variable response to refractive surgery. Oftentimes, an enhancement procedure with all its unpredictability is relied upon to correct the residual refractive error, with its concomitant increase in complication rate and scar formation. This new espoused device allows for the fact that individual tissue response to the calculated correction may not be completely predictable, and permits easy adjustments at the time of surgery and more importantly, at a later date after corneal hydration and would healing responses have stabilized, by simple strand removal from the implanted device. The nature of these adjustments minimally disturb the implant-corneal interface (unlike the 32 explantation of an intracorneal ring) and will thus allow a much more predictable adjustment. If an adjustment is to allow fine-tuning within 0.25 — 0.50 D of emmetropia, it is essential that the adjustment minimally disturb corneal tissue thus precluding a secondary wound healing response. In addition, when correcting myopia, a hyperopic outcome is very difficult to correct with any of the current kerato-refractive procedures and over-correction of myopia does occur. In this invention, a hyperopic outcome is relatively easily reversed by ring removal from the implanted device. Typically, in most kerato-refractive procedures for myopia, the surgeon aims for a slight under-correction because of the wish to avoid a hyperopic outcome. The ease with which a hyperopic outcome is adjusted with the device of the present invention enables the surgeon to aim for full correction, thereby obtaining the full benefit of the nomogram, and resulting in a higher percentage of patients with the desired refractive outcome even without a modification of the device. The surgeon may even choose to slightly overcorrect followed by a modification. Referring to back to FIG. 2(a), it can be seen that there are many more patients with an uncorrected visual acuity of 20/20 on the right side (hyperopic, over-corrected) of the graph. The problem is that only 50% of patients achieve an uncorrected visual acuity of 20/20. Referring to FIG. 2(b), it becomes obvious that the easiest way to increase the percentage of patients achieving 20/20 is to shift the curve to the right. This means a higher percentage of over-corrected patients but also a much higher percentage achieving uncorrected vision of 20/20. Patients on the left side of the graph can be re-treated for their residual myopia. Because there are fewer options after being over-corrected, it is unethical to shift the curve to the right, even though a much higher percentage achieves 20/20. Also, even the patients who are on the right side of the graph and do see 20/20 now, will become presbyopic and will not have clear vision at any distance without glasses in the future.

FIG. 2(c) demonstrates the usefulness of a refractive procedure that can be partially reversed after the initial procedure. Assuming that a refractive procedure can easily be partially reversed after the initial procedure, the curve can safely be shifted to the right. All 33 the patients in FIG. 2(b) who are signficantly overcorrected can undergo a partial reversal by simple strand removal. Even if only slightly hyperopic, if the patients become symptomatic, they may choose to undergo a partial reversal. This results in a much narrower distribution of patients and a distribution of patients that surrounds emmetropia. Another fortunate but coincidental outcome of this particular refractive strategy is that greater than 90% of patients may have an uncorrected visual acuity of 20/20 even prior to the adjustment procedure. This surgical strategy of shifting the nomogram and over-correcting patients' refractive error so that almost all patients are initially 20/20 without correction has not been used in the past with other keratorefractive procedures because of the fewer surgical options to correct the resultant hyperopia.

FIG. 13(a) illustrates a pre-operative cornea in a myopic patient. FIG. 13(b) shows flattening 81 of the central corneal curvature relative to the pre-operative corneal curvature 80, following implantation of the adjustable corneal ring. FIG. 13(c) illustrates the partial re- steepening 83 of the central corneal curvature following removal of strands from the implanted ring. It can also be seen that the corneal ring displaces more 82 peripheral corneal tissue prior to strand removal and that the corneal ring displaces less 84 peripheral corneal tissue following strand removal.

Dr. R. Eiferman in the Journal of Refractive and Corneal Surgery states that "if we can regulate the amount of tissue that is either added to or subtracted from the cornea and control the biological response, we may then be able to optimize refractive surgery". The ideal method to control the biological response is to minimally disturb corneal tissue, thus minimally inciting a wound healing response. Dr. K. Thompson, in the same Journal asks, "will it be possible for a refractive surgery technique to bypass the variable effects of corneal wound healing altogether?" That is unlikely for any initial keratorefractive procedure but the adjustable corneal annular segments of the present invention makes possible an adjustment that avoids the variable effects of corneal wound healing by minimally disturbing corneal tissue. 34

The essence of this invention lies in the assumption that individual responses to any kerato-refractive surgical procedures are variable, that even a "perfect" nomogram will not lead to a reliably predictable result in a particular individual, that a simple, safe, and effective technique for corneal curvature adjustment is necessary and that this modification should minimally disturb surrounding tissue thus allowing for a predictable effect. It should also be easily accomplished at some post-operative date after implantation of the device and after factors affecting corneal curvature changes have stabilized. A key feature of this invention lies in the ability of the device in its various embodiments to allow the amount of peripheral corneal volume displaced to be modified with ease at the time of implantation but more importantly on multiple occasions thereafter by simple removal of strand material from the implanted device, thus allowing fine-tuning of the refractive outcome.

In conclusion, in correcting refractive errors with this technique, the feeling of finality does not set in even with an initial inaccurate correction, with inadequate adjustment, or even when the last strand is removed because the device itself can be easily removed or better yet, left in place while other refractive procedures, such as laser ablation surgery are considered, if that point is ever reached.

It is also to be appreciated that the foregoing description of the invention has been presented for purposes of illustrations and explanation and is not intended to limit the invention to the precise form of apparatus and manner of practice herein. It is to be appreciated therefore, that changes may be made by those skilled in the art without departing from the spirit of the invention.

Claims

35CLAIMSWHAT IS CLAIMED IS:

1. A corneal ring for placement into a human cornea comprising: a hollow, collapsible, outer member to be formed into a ring, the outer member comprising a biocompatible material; and a removable, solid filler material substantially filling the cavity of the outer member, the filler material extending at least partially along the length of the outer member.

2. The corneal ring of claim 1 wherein the solid filler material is comprised of a removable flexible strand within the cavity of the hollow member.

3. The corneal ring of claim 2 further comprising at least one additional strand.

4. The corneal ring of claim 1 comprising one or more physiologically acceptable polymers.

5. The corneal ring of claim 1 wherein the solid filler material comprises a material selected from the group consisting of polymethylmethacrylate, nylon, polyester, polypropylene, polyethylene, polysulfone, TEFLON, or interpolymers.

6. The corneal ring of claim 1 wherein the length of the filler material is about equal to the length of the outer member.

7. The corneal ring of claim 1 wherein the filler material has at least one portion of a larger radial cross-sectional area.

8. The corneal ring of claim 1 wherein the perimeter of a cross-sectional area of the filler material is angular. 36

9. The corneal ring of claim 1 wherein the perimeter of a cross-sectional area of the filler material is angular.

10. The corneal ring of claim 1 wherein the solid filler material is comprised of a pre-formed material.

11. The corneal ring of claim 1 wherein the hollow outer member is collapsible following removal of the solid filler material from within the outer member.

12. The corneal ring of claim 1 wherein the outer member comprises a reinforcing backbone along at least a portion of the length of the outer member.

13. The corneal ring of claim 1 wherein the outer member comprises a plurality of apertures for the exchange of fluids between the cavity and an exterior of the outer member.

14. The corneal ring of claim 1 wherein the outer member comprises an aperture to provide access to the cavity.

15. The corneal ring of claim 1 wherein a radial cross-sectional area defined by the outer member decreases following removal of at least part of the filler material from the corneal ring.

16. The corneal ring of claim 1 wherein the radial cross-sectional area of the ring is largely comprised of the filler material within the outer member.

17. The corneal ring of claim 1 wherein the radial cross-sectional area of the ring is largely comprised of the filler material within the outer member.

18. The corneal ring of claim 1 wherein the radial cross section of the ring has at least one area of comparatively larger dimension than other radial cross sections of the implant and are positioned to substantially correct astigmatism. 37

19. The corneal ring of claim 18 wherein the comparatively larger radial cross sectional dimension of the ring is largely due to the larger dimension of the solid filler material.

20. The corneal ring of claim 18 wherein the comparatively larger radial cross sectional dimension of the ring is largely due to the larger dimension of the walls of the outer member.

21. A process for altering the corneal curvature of an eye, the process comprising the steps of:

Making an incision into the cornea;

Forming an annular channel between lamellae of the corneal tissue, said channel extending about an optical zone of the cornea; and

Implanting into the channel a corneal ring comprising a hollow, collapsible member that is annular in shape and which contains

Solid filler material within the cavity of the hollow member that is removable after implantation of the corneal ring within the cornea.

22. The process of claim 21 wherein the solid filler material comprises at least one strand of material extending along at least a portion of the corneal ring.

23. The process of claim 21 wherein the solid filler material comprises a pre-formed material.

24. The process of claim 22 further comprising the step of adjusting a radial cross- sectional area of the corneal ring.

25. The process of claim 24 wherein the adjusting step comprises removing at least one of the at least one strand. 38

26. The process of claim 24 wherein the adjusting step comprises replacing at least one of the at least one strand with a larger strand.

27. The process of claim 24 wherein the strand has a thicker portion and wherein the adjusting step comprises moving the strand such that the thicker portion is displaced in the corneal ring.

28. The process of claim 24 wherein there are present within the collapsible member at least two strand and wherein the adjusting step comprises moving one strand relative to the other vvithin the collapsible member.

29. The process of claim 21 further comprising the step of adjusting a thickness of at least a portion of the corneal ring.