THE FIRST EXPERIENCES

The origin of keratoprostheses is as old as that of keratoplasty. The same year in which the French Revolution began (1789), Pellier de Quengsy – a then obscure ophthalmologist from Montpellier – published for the first time the idea of ​​replacing the opacified cornea with a glass lens (Figure 1)¹. But it would not be until more than half a century later when in 1853 Nussbaum would try it on rabbits. He first implanted a glass disc into the corneal stroma and later a penetrating prosthesis in the form of a cylinder of 3 mm in diameter and 1 mm in height, which widened in front and behind (Figure 2). This required trepanation of the same diameter and a radial incision that was then sutured. Although his prostheses were expelled after a few weeks, he demonstrated the low reaction to glass, even experimenting with himself with cutaneous implants of different materials².

The design of Nussbaum would be applied clinically in the following years, by Weber (1855), Heusser (1860) and von Hippel (1877), in general with retentions limited to a few months, except in one of the 7 patients of the last in whom the keratoprosthesis remained for one year⁴. These first prostheses were glass or quartz glass – in the case of Heusser –. Later, Salzer (1895) achieved 9- and 33-month retentions through a quartz disc supported by a ring and small platinum hooks. The first to use a thermoplastic was Dimmer in 1891. He implanted in human eyes celluloid discs that he molded, although after a few weeks, retroprosthetic membranes were formed⁴.

THE 20^th CENTURY

With the arrival of the new century, interest in artificial corneas declined due to the success of corneal transplants, starting with the famous case of Zirm and the work of other pioneers (see chapter 1.3). In 1935 Filatov designed a keratoprosthesis that would not need fixation because it had a double trunk shape with an opposite cone ("diabolo"), although it was also expelled after 6 weeks⁴. It was not until the middle of the century that some corneal surgeons were again attracted to this field when reconsidering the limitations of keratoplasty⁵. No doubt technological advances such as the introduction of polymethyl methacrylate (PMMA) and other plastics, as well as the appreciation of its easy handling, good intraocular tolerance and refractive properties – which would lead Ridley to perform in 1949 the first intraocular lens implant of the modern era.

Single-component PMMA keratoprosthesis

The first PMMA keratoprostheses were one-piece, based on the prototype "shirt button" with different variations in the anterior and posterior disc amplitudes, the type of fixation – superficial or intrastromal (interlaminar) – of the same, or as they were of partial penetration like that of Franceschetti (1949) or full-thickness such as those of Dorzee (1948), Baron and Legrand (1954) – these already with holes in the haptic disc to pass sutures –, or that of Barraquer-Cardona (1958)^4,5.

Another group of designs presents a single interlaminar haptic disc, generally with perforations to improve the transit of metabolites – in an attempt to avoid corneal necrosis – and to favor the healing that would fix the prosthesis, such as those of Stone and Herbert (1953) (Figure 3), that of Binder and Binder (1956) – with 8 mm polyethylene haptics (Figure 4) –, that of McPherson and Anderson (1953) with the haptic reduced to a series of radii terminated in a mace (Figure 5), or that of Cardona and Castroviejo (1962), with a 2 mm cylinder and an interlaminar plate of 4 mm^3,4.

Keratoprostheses composed of two pieces

In 1951, Györffy and Vanisek introduced one of the most relevant modifications in the design of keratoprosthesis: its composition in two pieces that were assembled – in their case by means of a screw and nut-like thread – in front and behind the altered cornea, with a piston rod or optical cylinder that passes through it (Figure 6)³.

This idea – which will later lead to Boston's keratoprosthesis – is picked up by Cardona with his nut & bolt design (1969), in which the 8.5 mm anterior disc has coloration for cosmetic purposes and to avoid glare, and it is attached to a longer optical cylinder (5.5 to 9.5 mm) on which the rear plate is screwed (Figure 7). The greater length of the cylinder possibly helps to avoid the retroprosthetic membranes. Stone (1965) and Choyce (1969) apply this concept to make the interchangeable optics, as well as Fyodorov, Moroz and Zuev (1972), with threaded models of PMMA and titanium haptics (Figure 8). Tantalum metal supports are also used by the Odessa group (see 8.5) and Bedilo (1969), or platinum by Torres and Ruiz (1963)^3,4.

TOWARDS INTEGRATED HAPTICS AND BIOLOGICAL SUPPORT

Despite the multiple designs of keratoprostheses at this time, the main problem remained that of extrusion. It became evident that the plastic or metal never really joined the living tissues, which led to look for other materials that would achieve the integration or at least improve the adhesion of the prosthesis. Prostheses were designed with a skirt formed by a mesh of synthetic fibers, which would be colonized by the recipient tissue. Among these, those of Girard with nylon meshes (1967) and later Teflon, proplast or Dacron (1969)⁶, material used by Cardona and in particular Pintucci (1979) (Figure 9)⁷ stand out. Hofmann (1978) and shortly after Polack and Heimke (1980) propose using a very hard and very porous ceramic, which would facilitate its adhesion to the tissue⁸.

On the other hand, some techniques began to combine coatings in several layers with biological materials. In the Cardona through-through, in addition to the optical cylinder with thread and a Teflon anterior disk of 0.2 mm thick that is sutured to the cornea by 8 holes, a patch of tibial periosteum anchored to the sclera was applied – material also used by Polack – and on it a conjunctival flap⁹. Girard completed his techniques with a scleral homograft and a conjunctival flap⁶.

Strampelli’s osteo-odonto-keratoprosthesis

The search for a material strong enough to support the optical element and at the same time capable of bonding with the eye tissues would culminate in Benedetto Strampelli's idea of ​​using as a haptic a sheet taken from the root of a tooth, which he presented in 1963 as osteo-odonto-keratoprosthesis (OOKP)¹⁰. Strampelli starts from the concept that a keratoprosthesis is a mesoprosthesis, that is, it is on the border between the outside and the inside of the organism, unlike the exoprostheses as an artificial limb or the endoprostheses as a heart valve. This particular situation causes the organism to tend to encompass it from the front or to surround it from behind and extrude it, a problem that can only be solved when the prosthesis joins the surrounding tissues in a way that is as close to the physiological and the superficial epithelial tissue stabilizes without triggering the lysis processes associated with an incomplete repair. The osteo-dentary piece presents the advantages of a great hardness and resistance to the degradation of the dentin, together with the ease of the bone to scar with other connective tissues such as the cornea, the sclera or the mucous membranes. It is also a heterotopic autograft not susceptible to immune rejection.

The first Strampelli’s technique used a disc cut transversely to the root axis of a canine and implanted interlamellar, with an optical cylinder of 2 mm in diameter and 3 mm in length cemented to a hole in the dentine. However, this technique also presented frequent extrusions due to necrosis of the superficial corneal tissue, even applying a cornea homograft on it⁴. He went through it in 1966 to what is today the classic technique of OOKP, in which the lamella is cut lengthwise to the root of the tooth – which provides more bone (Figure 10A) – and is fixed in an epicorneal plane under a full thickness mucosal covering previously performed. It requires a longer optical cylinder, 9 mm, with diameters between 2.5 and 4.0 mm. This technique is still used today with some modifications such as those proposed by the Falcinelli group¹¹ (see chapter 8.3.1).

Despite the success of the OOKP, one of its problems is the lack of adequate teeth in a part of the patients. In these, Strampelli (1967) and Casey proposed the use of ear cartilage, and Falcinelli that of relatives' teeth¹¹. However, our experience with them has not been good, possibly due to a problem of immunological rejection. Thus, José Temprano proposed, in 1984, the use of a hard bone disc of the patient's own tibia (Figure 10B), the so-called tibial osteo-keratoprosthesis⁴ (see chapter 8.3.2).

KERATOPROSTHESES IN SPAIN

The implantation of keratoprostheses in Spain began in 1955 in Barcelona by Joaquín Barraquer, who treated a young woman who had suffered a severe chemical injury of both eyes with a Dorzee-type PMMA model (Figure 11). A good functional result was obtained for 5 years, at which time an extrusion of the prosthesis with retinal detachment occurred. In 1958 he implanted a second prosthesis – in this case a model modified by him and Cardona from Dorzee’s – in another chemical injury patient (Figure 12). The prosthesis remained well until the death of the patient 12 years later, although the functional result was limited by presenting a terminal glaucoma¹².

In 1960, Barraquer implanted another model of Cardona in a patient after two failed keratoplasties, with good functional results until he developed an endophthalmitis 8 years later. Later, he tried Girard's keratoprosthesis with Teflon mesh, which was extruded after a few months (Figure 13). The first OOKP was also carried out in Spain in 1965, with the initial (interlamellar) technique of Strampelli and which remained well for 10 years (Figure 14).

THE NEW KERATOPROSTHESES

In spite of the good results of the OOKPs, the complexity of the technique and other inconveniences limited its diffusion to a few centers in the world. Other groups continued with the investigation of new models.

Boston keratoprosthesis

In the 1960s, Claes H. Dohlman and M.G. Doane began to develop, at the Massachusetts Eye and Ear Infirmary in conjunction with the Schepens Eye Institute in Boston, a keratoprosthesis based on the two-piece design: an optical PMMA cylinder with a relatively short anterior flap, and a larger diameter posterior disc that initially was also of PMMA and was screwed to the stem as in Cardona’s nut & bolt; later the disc slides directly and is fixed with a titanium locking ring that fits into a notch in the stem (Figure 15).

The Boston keratoprosthesis (B-Kpro), as it was called, was implanted for the first time by Dohlman in 1968, although it was not approved as a medical device in the USA until 1992 and in Europe until 2013. However, due to its simplicity and good results in the most common indications, it has become the most used worldwide^12,13. Perhaps its most original contribution is the fact that it is always mounted on a cornea homograft instead of the patient's own, with which the present pathology is avoided, and the graft is made very tight between the two parts of the prosthesis. In addition to the initial model or B-Kpro type I, Dohlman devised a type II for application through the eyelid in patients with very severe alterations of the ocular surface (see chapters 8.4). The transpalpebral fixation had already been applied, however, with several other keratoprostheses at least since 1978⁴.

Flexible keratoprostheses

Other recent developments based on more advanced alloplastic materials include AlphaCor keratoprosthesis, KeraKlear and Miro Cornea, among others. The first was developed in Australia, implemented since 1992, and approved in the USA in 2003. It uses a hydrophilic acrylic material and is therefore flexible. The second one is also made of a flexible and injectable acrylic material for its intrastromal (non-penetrating) lamellar implantation. The third uses a hydrophobic acrylic material combined with a hydrophilic optical surface and a coating of the plate that favors adhesion to the tissue. These new types are described in chapter 8.2.