Ocular Implants and Methods for Their Manufacture

1. Field of the Invention

The present invention relates to ocular implants and methods for improving surfaces thereof.

2. Prior Art

Studies have shown that the surgical implantation of ocular implants such as intraocular lenses (IOL), etc., can result in the loss of significant corneal endothelial tissue unless great care is taken to ensure a lack of contact between the device and the endothelium. Most ocular implants are constructed of hydrophobic polymethylmethacrylate (PMMA) polymers because of their superior optical qualities, resistance to biodegradation, etc. It has been found, however, that PMMA surfaces adhere to endothelial cells upon even casual contact and that separation of the surface therefrom results in a tearing away of the endothelial tissue adhered to the polymer surface. Similar adhesive interactions with other ocular tissues, i.e., the iris, can also cause adverse tissue damage. Other hydrophobic polymers which are used or have been proposed for use in ocular implants (i.e., polypropylene, polyvinylidene fluoride, polycarbonate, polysiloxane) also can adhere to ocular tissue and thereby promote tissue damage.

It is well documented in the prior art that a significant disadvantage inherent in PMMA IOLs resides in the fact that any brief, non-traumatic contact between corneal endothelium and PMMA surfaces results in extensive damage to the endothelium. See Bourne et al, Am. J. Ophthalmol., Vol. 81, pp. 482-485 (1976). Forster et al, Trans. Am. Acad. Ophthalmol. Otolaryngol., Vol. 83, OP-195-OP-203 (1977); Katz et al, Trans. AM. Acad. Ophthalmol. Otolaryngol., Vol. 83, OP-204-OP-212 (1977); Kaufman et al, Science, Vol. 198, pp. 525-527 (1977) and Sugar et al, Arch. Ophthalmol., Vol. 96, pp. 449-450 (1978) for a discussion of the problem associated with implant surface/endothelium contact.

Since it is extremely difficult to avoid any contact between implant surfaces and endothelium during surgical procedures, efforts have been undertaken to modify the PMMA ocular implant surfaces to reduce the tendency thereof to adhere to and damage corneal endothelium.

Ocular implant surfaces have been coated with various hydrophilic polymer solutions or temporary soluble coatings such as methylcellulose, polyvinylpyrrolidone [Katz et al, supra, and Knight et al, Chem. Abs., Vol. 92:203547f (1980)] to reduce the degree of adhesion between the implant surfaces and endothelial tissue cells. While offering some temporary protection, these methods have not proven entirely satisfactory since such coatings complicate surgery, do not adhere adaquately to the implant surfaces, become dislodged or deteriorate after implantation, dissolve away rapidly during or soon after surgery or may produce adverse post-operative complications. Moreover, it is difficult to control the thicknesses and uniformity of such coatings.

Yalon et al [Acta: XXIV, International Congress of Ophthalmology, ed. Paul Henkind (1983)] attempted to produce protective coatings on PMMA implant surfaces by gamma-radiation induced polymerization of vinylpyrrolidone thereon [See also Knight et al, supra]. Their efforts were not altogether successful, however, since their methods also presented problems in controlling the optical and tissue protective qualities of the coatings. Process conditions and parameters (i.e., monomer concentration solvent, dose and dose rate) were not specified. The resulting coatings were of poor quality and non-uniform mechanical stability.

Gamma-PVP treatment of PTFE has been reported but under severe process conditions requiring gamma doses above 1 Mrad which are impractical in that both bulk and surface properties of the PTFE are changed [Boffa et al, J. Biomed. Mater. Res., Vol. 11, p. 317 (1977)]. Non-aqueous solutions of high monomer concentrations (50% NVP in pyridine) are required at relatively high doses of gamma radiation (1-5 Mrad) resulting in a high degree of grafting but with extensive changes in the bulk and surface properties of the PTFE since PTFE is readily degraded at gamma doses above 1 Mrad.

In U.S. Pat. No. 4,806,382, issued Feb. 21, 1989, there are described improved methods for producing hydrophilic, gamma irradiation induced polymerized and chemically grafted coatings on ocular implants constructed of a variety of polymeric materials, which methods overcome the above-noted difficulties and disadvantages.

The invention described in that application is predicated on the discovery of certain process conditions and parameters that produce thin hydrophilic gamma irradiation induced polymerized and chemically grafted coatings of N-vinyl-pyrrolidone (NVP) [PVP], copolymerized NVP and 2-hydroxyethylmethacrylate (HEMA) [P(NVP-HEMA)], or HEMA [PHEMA] on the surfaces of ocular implants constructed of materials including polymethylmethacrylate (PMMA) and of other process conditions and parameters which produce thin gamma irradiation induced graft PVP, P(NVP-HEMA), or PHEMA coatings on the surfaces of ocular articles constructed of materials including polypropylene (PP), polyvinylidene fluoride (PVDF), polycarbonate (PC) and silicone (PSi). The coatings increase the hydrophilicity of the implant surface and minimize adhesion between the surface and sensitive ocular tissues such as corneal endothelium or iris thereby minimizing tissue damage and post-operative complications occasioned by contact between the implant surface and ocular tissue. The coatings produced by the improved method of the invention described in U.S. Pat. No. 4,806,382 are thin and reproducibly uniform. Moreover, they are chemically bound to the surface of the ocular implant and, therefore, far more durable and less subject to removal, degradation or deterioration during or following surgery than the coatings produced by prior art methods.

The improved gamma-irradiation induced graft polymerization of NVP, HEMA or mixtures of NVP and HEMA on ocular implant surfaces comprising PMMA to form optimum PVP, P(NVP-HEMA) or PHEMA graft polymer surface modifications thereon comprises carrying out the graft polymerization in an aqueous solution under specific combinations of the following conditions:

a) monomer concentration in the range of from about 0.5 to about 50%, by weight;

b) total gamma dose in the range of from about 0.01 to about 0.50 Mrad;

c) gamma does rate in the range of from about 10 to about 2500 rads/minute; and

d) maintaining the molecular weight of the polymer in solution in the range of from about 250,000 to about 5,000,000.

Optimally, the method may also be carried out under one or more of the following conditions:

e) substantially excluding free oxygen from the aqueous graft polymerization solution;

f) maintaining the thickness of the PVP or P(NVP-HEMA) surface graft in the range of from about 100 .ANG. to about 150 microns;

g) including a free radical scavenger in the aqueous graft polymerization solution; and

h) including in the aqueous graft polymerization solution a swelling solvent for PMMA or other polymer substrate surface.

The improved gamma-irradiation induced graft polymerization of NVP, mixtures of NVP and HEMA or HEMA on ocular implant surfaces comprising PP, PVDF, PC or PSi to form optimum PVP or P(NVP-HEMA) surface grafts thereon may also be carried out under specific conbinations of the process parameters as indicated above for PMMA but also under conditions which involve excluding free oxygen from the polymerization solution for preferred surface modification of these ocular implant polymer substrates.

It is an object of the present invention to provide a still further improved method for producing hydrophilic coatings on the surfaces of ocular implants.

The present invention is predicated on the discovery that, in order to produce the hydrophilic coatings on the surfaces of ocular implants according to the method described in U.S. Pat. No. 4,806,382, the "maintenance of the molecular weight of the polymer in solution in the range of from about 250,000 to about 5,000,000" is not a critical condition.

The present invention is further predicated on the discovery that in order to successfully carry out the method described in U.S. Pat. No. 4,806,382, the total gamma dose range may be extended to a minimum value of 0.001 Mrad.

The success of the improved method of the present invention is, however, predicated on strictly maintaining the remainder of the conditions outlined in U.S. Pat. No. 4,806,382 for achieving the graft polymerization coating on ocular implant surfaces.

The drawings, FIGS. 1-3, depict examples of ocular implants according to the present invention.

FIG. 1 depicts a top view of a one-piece intraocular lens,

FIG. 2 depicts a top view of an intraocular lens with fiber haptics which may be made of a different substrate polymer than the optic, and

FIG. 3 depicts a top view of a kerotoprosthesis.

The maintenance of the molecular weight of the polymer in solution at certain values, identified in U.S. Pat. No. 4,806,382 as a critical condition of the method is not actually a "condition" of the method, but rather, as stated in the specification, a result which is dependent on the reaction conditions employed in carrying out the graft polymerization process. It is, therefore, not appropriate to specify the molecular weight of the polymer in solution as the reaction conditions used in this invention and may be widely varied depending on specific gamma graft monomer-substrate-process conditions. If a certain set of fixed conditions are employed, namely: monomer, monomer concentration, total gamma dose and gamma dose rate, the molecular weight of the polymer formed in solution cannot be independently varied but will be an output of the process which is dependent upon the values of the above-noted monomer concentration, total gamma dose and gamma dose rate conditions. For example, in the presence of certain ionic monomers, solvents or radical inhibitors, solution polymerization may be inhibited significantly without sacrificing efficient surface graft polymerization and the resulting solution polymer molecular weight may thereby be relatively low (i.e., as low as 5,000-10,000).

Since the application which matured into U.S. Pat. No. 4,806,382 was filed, the inventors of the subject matter defined therein conducted additional research and unexpectedly found that although relatively low doses of 0.01 to 0.20 Mrad are generally preferred for the compositions of this invention, the process could be conducted at a total gamma dose as low as 0.001 Mrad.

The state of the art prior to the application which matured into U.S. Pat. No. 4,806,382 taught the use of relatively high gamma doses, generally greater than 0.5 Mrad, for gamma polymerization grafting, and it was therefore surprising to find that surface grafting could be achieved at doses as low as 0.01 Mrad. The achievement of effective grafting at doses as low as 0.001 Mrad is consequently an even more unexpected result of the process of this invention. Furthermore, although grafting with monomer concentrations as low as 0.5 wt % was indicated in prior U.S. Pat. No. 4,806,382, further research has revealed that monomer concentrations as low as 0.1 wt % may be utilized in some embodiments of the graft process of this invention.

Yalon et al (supra) and Knight et al (supra) disclose gamma-irradiation coatings on PMMA using N-vinylpyrrolidone (NVP) and 2-hydroxyethylmethacrylate (HEMA) and indicate poor dynamic (abrasive) protection of endothelium for these coatings. Dissolvable coatings of polyvinyl-alcohol (PVA) were regarded as optimal for intraocular lenses (IOLs) by Knight et al, supra, and commercial development of a PVA-coated IOL was attempted with unsatisfactory clinical results. The gamma polymerization surface modifications reported were carried out under process conditions of monomer concentration, solvent, dose and dose rate which were not specified and which apparently yielded poor quality, readily abraded coatings. Conditions for producing useful permanent PVP or PHEMA coatings on PMMA IOLs or any other plastic surface are not taught in the prior art. Neither Knight et al, Yalon et al or the literature on gamma-graft polymerization of the past 30 years suggest the process conditions required to achieve the complicated requirements for useful coatings on plastics. These requirements include:

a) Thin, permanent, optically clear (in the case of contact lenses) and uniform graft coatings. The literature generally discloses conditions which produce distortion and degradation of the substrate due to the use of high gamma-radiation dose (>1 Mrad) and non-aqueous solvent media, and yield thick, cloudy, non-uniform coatings (e.g., Chapiro, Radiation Chemistry of Polymeric Systems, John Wiley and Sons, Inc., New York (1962); Henglein et al, Angew. Chem., Vol. 15, p. 461 (1958).

b) Long-term biocampatibility in vivo.

c) Low contact angle (high wettability) for water or underwater air bubble (less than about 30.degree.).

d) Non-adherent to tissue (adhesive force to endothelium less than about 150 mg/cm.sup.2).

e) Non-damaging to endothelium (less than ca. 20% damage for in vitro contact tests).

f) Measurable graft coating by ESCA or FTIR analysis.

g) Abrasion resistance by sliding (dynamic) friction testing showing no change in wetting (contact angle) and confirming before and after presence of graft coating.

h) Rapid hydration-change from dry state to wetted lubricous state on immersion in water (within five minutes).

Yalon et al (supra) disclose an in vitro technique for measuring endothelium damage. Results for PMMA were used to illustrate the method. Although it was noted that PVP coatings reduced cell damage with less damage at higher monomer concentrations, the conditions for the experiment (i.e., irradiation dose, dose rate, etc.) were not disclosed nor were any of the critical process-product relationships indicated.

The improved process conditions and parameters of the invention described in U.S. Pat. No. 4,806,382 (the entire disclosure of which is incorporated herein by reference) which are necessary to produce useful polymers having a surface modified by gamma-irradiation induced graft polymerization therein of PVP, P(NVP-HEMA) or PHEMA include: % monomer, gamma dose, dose rate, penetration time or swelling time of monomer into the substrate prior to polymerization and oxygen (air) degassing. Other optimal process conditions include catalysts, free radical scavengers, polymer swelling solvents and temperature. The solution polymer molecular weight and M.W. distribution, the % conversion and residual monomer, the graft polymer thickness and surface properties, etc., are process results which can change markedly as the process variables change. For example, the surface modification achieved for PVP on polymer surfaces will be different when using 10% monomer and 0.1 Mrad if prepared at low dose rates since low dose rates (slower polymerization) favor higher molecular weights. Similarly, degassed oxygen-free reaction media result in improved grafts at much lower doses. The presence of free radical scavengers such as copper or iron salts or organic reducing agents (i.e., ascorbic acid) also greatly influences other process parameters, generally reducing solution polymer molecular weight and preventing solution gelation at high monomer concentrations.

Each of the above-described process conditions and parameters of the method of the invention may be varied within the ranges discussed below to produce certain specific combinations which are particularly advantageous for the surface modification of a particular polymeric surface.

a) Monomer concentration: Increasing monomer concentration increases polymer mol. wt. in the graft solution and reduces contact angle (C.A.), i.e., renders the surface more hydrophilic. For example, in the case of forming PVP coatings on PMMA, in the range of from about 3-15% NVP the PVP viscosity mol. wt. (M.sub.v) increases from 560,000 to 2,700,000 and the PMMA graft C.A. decreases from 29.degree. to 21.degree. at 0.1 Mrad and 309 rads/min. However, this effect is sensitive to dose rate and total dose. For example, at 1-10% NVP, but at a lower dose rate of 64 rads/min., the mol. wt. increases from 400,000 to 4,590,000 and the C.A. decreases from 49.degree. to 18.degree..

In general, monomer concentrations in the range of 0.1-50% are preferred depending on other parameters. Concentrations as low as 0.1-0.5% at low dose rates can yield hydrophilic surface grafts with C.A. below 30.degree.-40.degree. under conditions of this invention. At monomer concentrations greater than 20-30%, effective grafting without solution polymer gelation requires low doses and use of free radical scavengers. Monomer concentrations greater than 50% are feasible but not preferred since high concentrations of radical scavengers must be used and polymer mol. wts. and monomer conversion are lowered significantly by their use. For producing PHEMA coatins, HEMA concentrations of between 0.5% and 10, by weight, are sufficent.

b) Dose: In general, increasing total gamma dose increases mol. wt. of the polymer and reduces C.A. However, an important practical limit exists in that at higher doses, lower dose rates and higher monomer concentrations, reaction media beomes extremely viscous or form gels which are very difficult to wash and to remove (e.g., about 0.25 Mrad and 10% NVP at 309 rads/min).

It will be understood by those skilled in the art that electron beam radiation will also induce graft polymerization. Therefore, electron beam radiation of energies equivalent to that described herein for gamma radiation may be substituted for gamma radiation in the practice of the method of the invention. Electron beam voltages in the range of from about 50 KeV to about 10 MeV may be employed at currents of from about 4 mA to about 100 mA. For electron beam initiated polymerization grafting, conditions which produce dose rates substantially higher than for gamma graft polymerization, i.e., in the range of from about 10 to about 10.sup.8 rads/min or more may be employed.

c) Dose rate: Decreasing gamma radiation dose rate generally increases solution PVP M.W., e.g., from 1,150,000 to 5,090,000 at 10% NVP and 0.1 Mrad as dose rate decreases from 1235 to 49 rads/min. The C.A. also goes down at lower rates of up to 10.sup.8 rads/min or more are practical when employing electron beam irradiation.

d) Solution Polymer Mo. Wt.: The mol. wt. may vary widely depending upon process conditions, monomers and radical inhibitors used. Effective grafting with low C.A. may therefore be achieved with even low mol. wt. solution polymer (V.sub.v as low as 5000-10,000 or less). However, solution polymer M.sub.v greater than 5,000,000 or gels which form during grafting are generally impractical because of washing problems.

e) Degassing: Removal of oxygen from the graft solutions by vacuum and/or inert gas (e.g., argon purging) has an important effect: lower total doses are required (practical grafting at less than 0.1 Mrad). Oxygen degassing also has a large effect on PVP M.sub.w and % conversion of monomer. For example, with degassing, good grafting or PVP on polypropylene (PP) is achieved at 0.05 Mrad and 10% NVP (C.A. 15.degree.). Without degassing, little grafting occurs under these conditions. Oxygen degassing is critical to hydrophilic surface modification grafting where the substrate polymer is PP, PVDF or PSi. It has been found that graft polymerization is inefficient when using these materials as substrates in the presence of oxygen. Oxygen degassing is also beneficial fo PMMA and PC substrates in that much lower radiation doses (0.01-0.15 Mrad) become effective compared with grafting these polymers in the presence of oxygen.

f) Grafting thickness: Surfage grafts less than 100-200 angstroms, although non-adhesive and hydrophilic, are useful but may exhibit somewhat less mechanical "softness" or compliiant gel-like surfaces than thicker coatings for reduced tissue-contact trauma. Graft coatings greater than ca. 300-500 .ANG. (or 0.03-0.05 microns) up to 50 microns or more are probably more desirable for many applications as long as they are smooth, uniform, optically clear for optic surfaces, and quickly hydrated.

Using no swelling solvent ansd no prolonged monomer contact with substrates prior to irradiation, surface grafts which exhibit desired implane properties under preferred process conditions have thicknesses of about 0.1 to 5 microns. However, using swelling solvent such as ethyl acetate, polymer grafts on PMMA of 100 microns or more can be prepared. For certain applications it may be preferred to have thicker "spongy" coatings of 20-100 microns.

g) Free-Radical Scavengers: Free radical traps, usually reducing agents such as Cu.sup.+, Fe.sup.+2 asocrobic acid, etc., are known to inhibit radical polymerization in solution and thus be effective (especially at high gamma doses, high dose rates and high monomer concentrations) in slowing the onset of solution gelation during grafting. However, under practical grafting conditions, this may result in lower mol. wts., high concentrations of unreacted monomer and broad mol. wt. distributions. Use of metal salts may also be objectionable where maximum biocompatibility is critical.

Although most preferred graft conditions avoid the use of radical scavengers, useful conditions for graft coatings of PVP, P(NVP-HEMA) ro PHEMA have also been defined using ascorbi acid to limit high viscosity and gelation of the graft polymer solution. These conditions use high monomer concentrations (up to 50%) and thicker grafter are obtained using ethyl acetate as a swelling solvent (0.5-5%).

h) Swelling solvents: The use of substrate polymer solvents in the aqueous monomer grafting solution facilitates swelling and monomer diffusion into the polymer before and during gamma polymerization. Penetration of monomers into the substrate increases graft coating thickness and enhances bonding to the surface. Solvent such as ethyl acetate have been shown to greatly facilitate this process with some substrates such as PMMA.

Although the above-described method represents a significant improvement over prior art methods, optimum results in each case depend upon the selection of a combination of numerous process parameters and conditions.

Where mixtures of NVP and HEMA are employed to form graft copolymerized coatings of P(NVP-HEMA), the mixtures may contain up to about 50% by weight of HEMA, based on the weight of the monomer mixture. However, above 20-30% HEMA, radical scavengers and low monomer concentrations should be used to prevent gelation since HEMA enhances the onset of gelation.

It will be understood by those skilled in the art that the PVP, P(NVP-HEMA) or PHEMA graft coatings of this invention may be modified by copolymerization with various ionic monomers. Mixtures of hydrophilic and ionic monomers may also be copolymerized therewith. For example, graft coplymerization incorporation vinylsulfonic acid, styrene sulfonic acid, sulfoethylmethacrylate, sulfopropylmethacrylate, or other vinyl sulfonic acids or vinyl-carboxylic acids such as acrylic acid, crotonic acid or methacrylic acid can afford surface modifications which are anionic. Similarly, graft copolymerization incorporating basic or amino-functional monomers, e.g., vinylpyridines, aminostyrenes, aminoacrylates, or aminomethacrylates such as dimethylaminomethylmethacrylate or dimethylaminostyrene afford surface modifications which are cationic. It is also useful to use salts of ionic monomers or to convert ionic grafts to the salt form by post-treatment.

Amounts of ionic monomers up to about 50 wt. % of the total monomer weight may be employed, it being understood that the critical process parameters listed above may be maintained.

Based on the foregoing considerations and the many process studies conducted, preferred conditions for various article substrate polymers by way of example are provided in the examples below. Some key points may be summarized as follows:

Several ranges of process conditions appear useful. Choice of the "best" process will depend on such factors as: molecular structure of substrate and coating thickness desired. In general, those conditions which produce extreme solution viscosities and gels or conditions which could produce solvent stress cracking or crazing of the IOL polymers (e.g., high conc. than about 20% for a PMMA swelling solvent such as ethyl acetate) should be avoided. The following four sets of process conditions appear most practical for the preparation of improved surface modified articles.

(1) Aqueous Mononer Concentration:

5-20% (preferred 10%)

Dose: 0.05-0.20 Mrad (preferred 0.10)

Dose Rate: 20-15,000 rads/min.

(preferred 50-2,000)

C.A.: <30.degree.

(2) Same as (1) except that system is oxygen degassed (vacuum or inert gas purge, e.g., argon) with Dose:

0.010.15 Mrad (0.05 preferred) and % NVP: 1-15% (5-10% preferred). This system is generally preferred to (1).

(3) Same as (1) and (2) with swelling solvent (e.g., ethyl acetate for PMMA) gives greater monoer penetration of substrate and thicker grafts.

(4) High monomer concentrations (25-50%) using 21 5.0% ethyl acetate swelling agent and radical inhibitor such as ascorbic acid (0.1-1.0 mM) at 0.100.20 Mrad and 20-5000 rads/min.

All percentages expressed in the examples are by weight unless otherwise stated.

All contact angles (C.A.) and other surface characteristics for gamma polymerization grafts, unless otherwise indicated, are for samples washed with water or water-alcohol at room temperatures or elevated temperatures to remove soluble residual monomer and ungrafted polymer for the improved surface graft processes of this invention. The resulting graft polymers are stable and permanent for long-trem use and are not dissolvable by aqueous media.

It will aslo be understood by those skilled in the art that the ocular implants to be graft coating may be also constructed of materials othe than PMMA, PP, PVDF, PC or PSi to facilitate their use. It will be understood by those skilled in the art that such materials may also be at least partially graft polymer surface modified so as to improve their properties as implant materials.

The hydrophilic graft polymer surface modifications of this invention are especially advantageous for intraocular lenses (anterior chanber, posterior chamber and phakic), but are also of great vaule in affording improved tissue protection and improved biocompatibility for other ocular implants, such as corneal inlays, keratoprosthesis, epikeratophakia devices, glaucoma drains, retinal staplees, sclearl buckles, etc.

This example illustrates the important effects which result from varying the above-discussed process conditions and polymerization parameters for gamma-irradiated polymer graft surface modification of PMMA with PVP.

PMMA slab samples were washed twice each by soap solution and distilled water using a sonicator. After complete drying, the samples were put into NVP solutions in glass vials. The samples were then .tau.-irradiated at various conditions. After .tau.-irradiation, the surface modified PMMA samples were rinsed several times with H.sub.2 O and evaluated.

The polymerized NVP grafting solutions or gels were freeze-dried under a vacuum. The solution PVP samples were evaluated for molecular weight by viscosity measurement (M.sub.v) or gel permeation chromatography (M.sub.w). For M.sub.v, PVP was dissolved in distilled water and intrinsic viscosity [.eta.] was measured at 30.degree. C. in a capillary viscometer.

PVP grafted PMMA samples were evaluated by water drop or underwater air bubble contact angle measurements. The bubble technique is regarded as more reliable for very hydrophilic surfaces. For air bubble C.A., the grafted PMMA was held horizontally in distilled water. An approximately 0.8 .mu.l bubble was formed and positioned underneath the test surface. Angles on opposite sides of the bubble were measured assuring symmetry. Five measurments were usually made for each sample. The results are set forth in the following tables:

The effect of dose rate was evaluated by PVP solution viscosity measurements. These results show that the molecular weight increased as dose rate decreased due to the slower and reduced initiation of radicals and the increased time of polymerization while maintaining the same total absorbed dose. At the lowest dose rate in this experiment, 49 rads/min (at 10" from the Cobalt-60 gamma source), the highest molecular weight PVP polymer, M.sub.v =5.09.times.10.sup.6, was obtained.

Table 2 shows the effect of total .tau.-irradiation dose on molecular weight at 309 rads/min. Increasing the total dose gives a higher molecular weight. A polymer gel was formed at a dose of 0.25 Mrad and higher. These results show that a high irradiation dose can cause gelation or cross-linking of the PVP polymer.

These results show the relation between the concentration of NVP monomer and the molecular weights of PVP at constant dose and dose rate. The results indicate that higher NVP concentrations give higher molecular weight polymers. The importance of dose rate is also indicated by the fact that even at 15% NVP, the PVP molecular weight (M.sub.v) was only 2.7.times.10.sup.6 at 309 rads/min. compared to 5.0.times.10.sup.6 at a lower dose rate of 49 rads/min.

The results in Table 4 show that the contact angles for PVP grafted PMMA decreased due to hydrophilic PVP grafting and that the lower dose rates give lower contact angles.

These results show the effect of total dose on the contact angles of PVP .tau.-grafted PMMA. The contact angle showed little change above 0.05 Mrad at constant dose rate of 309 rads/min.

The effect of different monomer concentrations was evaluated for PVP .tau.-grafts on PMMA by contact angle measurement. Even at 3% NVP and 0.1 Mrad a major increase in hydrophilicity was observed as compared with non-grafted PMMA. The contact angle decreased slightly at monomer concentrations above 3%.

These results show the relationship between the concentration of NVP monomer and molecular weight of PVP at a dose rate of 64 rads/min.

The molecular weight of PVP increases significantly with increasing concentration of NVP monomer.

The contact angle of PMMA was evaluated after .tau.-grafting with NVP at different solution concentrations at a dose rate of 64 rads/min. These results show that the contact angles of PVP-grafted PMMA decreased with increasing concentration of NVP monomer. This result, at 64 rads/min dose rate is qualitatively similar to results at 309 rads/min (Table 6). Hydrophilicity at 10% monomer appears to be favored somewhat by the lower dose rate (C.A. 18.degree. vs. 25.degree.).

Polar organic solvents or aqueous-polar organic solvent mixtures may be useful for hydrophilic monomer graft polymerization. Typical of such organic solvents are alcohols or ethers such as methanol, ethylene glycol, polyethylene glycols, dioxane, etc. However, when such organic solvents acts as radical traps or radical chain transfer agents, they must be used at concentrations lower than 50% or with high hydrophilic monomer concentrations (i.e., >25%). For example, methanol has some radical scavenger properties but may be used for PVP gamma grafts on PMMA in water-methanol mixtures up to 50-60% methanol for PVP grafts on PMMA using 0.1 Mrad and 10% monomer (Table 9). Hydrophilic grafts result although radical chain transfer by methanol appears to require low dose rates at 10% monomer. In general these systems yield low viscosity solutions indicative of low molecular weight solution polymer which forms in the presence of radical inhibitors.

This example illustrates the effect of swelling solvents on the surface modification process.

For hydrophilic gamma grafts on PMMA as the substrate, for example, addition of the swelling solvent, ethyl acetate (EtOAc), to aqueous monomer solutions is advantageous to achieve more efficient diffusion of monomer into the PMMA surface. Although EtOAc is not very soluble in water, a homogeneous reaction medium can be achieved in the presence of a monomer such as NVP.

The thickness of the graft polymer surface modification can be increased by higher ethyl acetate concentrations and by longer diffusion times prior to irradiation, i.e., the time of pre-swelling. In general, without oxygen degassing, gamma radiation doses of 0.10-0.15 Mrad are suitable to achieve significant amounts of grafting.

The NVP-ethyl acetate-water solvent system is also a solvent for PVP and keeps the solution polymer phase homogenous.

"Embedded grafting" of PVP into the PMMA surface is made possible by irradiating the PMMA after exposure for various times to the monomer-swelling solvent-water mixture.

In experiments using this process techniques, samples were cleaned by sonication in a 10% soap solution followed by washing with distilled water. Prior to surface modification, PMMA samples were dried for 18 hours in a vacuum desiccator and weighed. NVP monomer was purified by vacuum distillation and stored at 4.degree. C.

For gamma radiation grafting, the PMMA substrate was immersed in aqueous monomer-solvent solutions and exposed to gamma radiation. Typically, cleaned substrates were immersed in NVP-ethyl acetate-H.sub.2 mixtures and irradiated in a 600 Curie Co-60 source. The samples were exposed to the monomer solution for various lengths of time. Gamma doses ranging from 0.01-0.15 Mrad as measured by Fricke dosimetry were used in this experiment. Dose rates were also varied. After irradiation, samples were removed from the gamma polymer solution and washed several times with distilled water and in deionized water with agitation. Some samples were weighed hydrated after blotting with filter paper to remove surface water and then dried for 24 hours in a vacuum desiccator. The polymerization solutions ranged from clear viscous solutions to gels. The following parameters were measured.

One measure of the degree of grafting was obtained from the weight increase of the substrate according to the following equation: ##EQU1## where W.sub.o is the initial weight of PMMA and W.sub.1 is the weight of grafted PMMA. Likewise, percent hydration was calculated according to the following equation: ##EQU2## where W.sub.w is the weight of PMMA after equilibration in water (after blotting it dry) and W.sub.d is the weight of dry sample (after desiccation). In most cases, the maximum water uptake was reached after 12 hours.

Captive air bubble and n-octane contact angles were measured for the radiation grafted PMMA surfaces to estimate the hydrophilicity of modified surfaces. Static contact angles were measured on a Rame-Hart contact angle goniometer. At least five measurements on different surface regions of each sample were made.

IR/ATR surface analysis of the grafted and ungrafted surfaces was made by using a Perkin-Elmer Model 283B IR Spectrometer using attenuated total reflectance.

Samples of 1 cm.sup.2 grafted and ungrafted PMMA were analyzed using a Kratos ES 300 ESCA spectrometer employing a magnesium K.alpha. x-ray source. Graft analysis consisted of N/C ratio determination.

The molecular weights of PVP solution polymers were determined by solution intrinsic viscosity measurements at 30.degree. C. in a Ubbelhode viscometer.

Radiation doses ranged from 0.01 to 0.15 Mrad and monomer concentrations ranged from 5 to 15%.

Data for PVP grafting onto PMMA using EtOAc as a swelling solvent are shown in Table 10. Since no preradiation swelling time is used here, diffusion penetration of the surface by EtOAc and monomer occurs during gamma radiation. Some pre-radiation swell-time is considered preferable. This system exhibits behavior typical of a reaction which involves monomer diffusion control. Partitioning of NVP monomer into the hydrophobic surface of PMMA is favored initially because of the presence of the ethyl acetate, which is a swelling solvent for PMMA.

By the use of a swelling solvent for the graft substrate (i.e., EtOAc), the NVP-EtOAc-H.sub.2 system swells the surface layrs of PMMA and polymerization grafting of monomer molecules in the vicinity of radiation induced radical species near the surface is immediate. Under such conditions, more efficient grafting is achieved at lower doses and with deeper penetration of the graft polymer into the solvent swollen surface.

Measurement of percent swelling of PMMA samples in NVP-ethyl actate-H.sub.2 O (1:1:8) vs. time shows that swelling of about 6% is attained after 12 hours. In this system, the thickness of the grafted layer could be controlled by changing the time allowed for diffusion prior to irradiation, thus controlling the thickness of the grafted zone. Table 11 shows the graft behavior after 24 hours of pre-swelling of PMMA in 1:9 ethyl acetate: water containing 15% of NVP. Comparing this data with Table 10 (no swelling time), it is clear that the % graft is significantly higher for pre-swelling PMMA. At a given ethyl acetate concentration, this difference is generally more pronounced at lower monomer concentrations, e.g., 5% monomer compared to 15% monomer.

In this system, NVP is the monomer but also acts as a mutual solvent to maintain a homogeneous phase of otherwise poorly miscible solvents, i.e., ethyl acetate and water. At a given monomer concentration (e.g., 10%), it is necessry to keep the concentration of ethyl acetate below 10% to avoid phase separation to a microemulsion. Variation of the ethyl acetate concentration, being a swelling agent, affects graft yield. Table 12 summarizes the observations made by varying the concentration of ethyl acetate while keeping other factors constant showing that the percent grafting does increase with higher ethyl acetate concentrations. Greater grafting efficiency is also indicated by the significant % grafting and reduction of C.A. in the solvent swelling monomer system at low doses. For example, up to 0.05 Mrad, little grafting occurs in a simple aqueous monomer system. In contrast, at only 0.01 Mrad C.A. is reduced to 35.degree. (Table 11, 24 hr. pre-swell) and to 23.degree. at 0.03 Mrad.

Techniques used for the chemical analysis of bulk polymers are usually not very satisfactory for analysis of the surfaces of polymers. The surface region, which is significantly different in structure and/or chemistry from the bulk, is present only as a fraction of the mass of the polymer. Thus, the traditional techniques of chemical analysis are inadequate. Special surface analysis techniques are required for graft copolymers since the surface region is a complex mixture of graft, substrate, cross-linking groups and chain transfer products. Two spectroscopic methods, ATR-IR and ESCA are the most useful methods now available for this purpose and were used to help characterize grafted surfaces.

The results for ATR-IR (attenuated total reflection infrared) shown in Table 13 indicate that the ration of C.dbd.O (ester) and C.dbd.O (amide) groups in the surface changes from 7.67 to 1.68 as the gamma dose increases from 0.01 to 0.10 Mrad and then levels off which is consistent with PVP grafting on PMMA.

ESCA analyses are shown in Table 14 and indicate increasing nitrogen composition with increasing dose (and grafting) as expected for a PVP graft.

Scanning electron microscopic examinations of the grafted samples were performed in order to observe their surface morphologies. All of the coated surfaces appeared smooth even at 10,000.times.. The graft polymer surface modifications appear to provide uniform coverage across the surface of PMMA substrate. This is important to insure excellent retention of optical properties for an optical implant such as an intraocular lens.

Major conclusions to be drawn from the results of this example are:

The NVP-ethyl acetate-water system produces uniform hydrophilic graft polymer surfaces with controllable graft penetration using PMMA as the substrate.

The monomer-ethyl acetate-water grafting front gradually penetrates into the substrate and may be controlled by varying the concentration of swelling agent and the time of pre-swelling.

The presence of the PVP surface graft was confirmed by gravimetric, contact angle, ATR-IR and ESCA measurements.

Unusually low radiation doese are required to achieve significant grafting. Hence, any possible radiation damage to the surface or substrate is minimized.

The following experiment demonstrates the very significant influence of oxygen on gamma polymerization and gamma grafting and the important beneficial effects of carrying out graft polymerizations in the substantial absence of oxygen.

Gamma radiation induced polymerization of NVP was carried out in 10 % NVP aqueous solution as follows:

(a) polymerization in presence of oxygen (air);

(b) polymerization in absence of oxygen using argon degassing; and

(c) polymerization in absence of oxygen. For Group (a), aqueous 10% NVP solutions were irradiated to total doses of 0.01, 0.05, 0.10, 0.20 and 0.25 Mrad in each case at 213 rads/min in the presence of air. An argon purge for 10 minutes was used in the case of Group (b). A vacuum freeze-thaw (FT) method was employed for degassing in the case of Group (c). In the freeze-thaw experiments, the monomer solution was frozen in liquid nitrogen and then vacuum (0.3 mm) was applied to eliminate oxygen. The frozen solution was thawed and brought to room temperature before irradiation. Some samples were subjected to three freeze-thaw cycles (3 FT). Experiments were run in duplicate to establish reproducibility.

To determine the oxygen degassing effects on gamma radiation grafting and polymerization, monomer conversions and molecular weights were determined for the different NVP solutions irradiated at 0.01 Mrad to 0.25 Mrad at 213 rads/min.

A method used for determining unreacted NVP after irradiation was as follows: 5 ml of the gamma irradiated NVP solution was extracted using 50 1 acetonitrile. NVP is soluble in acetonitrile, but PVP precipitate was centrifuged and the supernatant solution was analyzed for NVP. The NVP monomer solution (10% NVP/aqueous) was used as a control. NVP analysis was as follows: The 10% by weight aqueous solution was diluted with acetonitrile to appropriate concentrations (0.5 g/ml to 5.0 .mu.g/ml). The U.V. absorbance was measured for each solution at 323 nm to develop a standard curve of NVP concentration vs. U.V. absorbance. The regression coefficient was 0.99 for this curve. GPC was used for molecular weight measurements and gives M.sub.w as well as molecular weight distribution.

The % NVP conversion (amount of monomer reacted) is significantly affected by Ar purge deoxygenation and by FT oxygen degassing. At the very low dose of 0.01 Mrad, virtually no polymerization occurs in the non-degassed oxygen (air) containing solutions. However, 46%, 61%, and 63% conversion to PVP occurred for the AR-purged, 1 FT and 3 FT samples, respectively. Even at 0.10 Mrad, samples irradiated in air showed only 90% conversion (10% unreacted NVP monomer) compared to virtually complete conversion (99%) for oxygen degassed systems. This is important for biological implants where unreacted monomers can cause serious adverse toxicological behavior.

To demonstrate more efficient grafting of PVP on PMMA at low gamma doses in the oxygen degassed system, 10% aqueous NVP was argon purged to remove oxygen and irradiated with PMMA samples at 157 rads/min to 0.05 Mrad. The resulting hydrophilic surface modification had C.A. 20.degree. and was stable (no change in C.A.) to mechanical abrasion. As indicated above, this mechanically stable and very hydrophilic graft of PVP on PMMA graft is achieved with high monomer conversion (98%) and a high degree of polymerization for the solution polymer (1.65.times.10.sup.6 mol. wt.). In the presence of air (oxygen), higher radiation doses (>0.1 Mrad) and/or higher monomer concentration (15% or more) are required to achieve low C.A. with high conversion and high molecular weight. For hydrophilic monomer gamma polymerization grafts on other substrate polymers, i.e., polypropylene, fluorocarbons (e.g., PTFE or PVDF), or silicones, the beneficial effect of oxygen degassing can be even greater. Oxygen removal may also be used for improved gamma grafting in combination with the use of substrate swelling solvents and free radical inhibiting agents such as oxidizable metal salts or organic compounds (e.g., ascorbic acid). In the presence of radical inhibitors effective grafting may be achieved but solution polymer may be of low mol. wt.

PVP molecular weight is also greatly affected by oxygen degassing. The Ar-purged and FT samples yield PVP polymers with molecular weights of about 1.6.times.10.sup.6 at only 0.01 Mrad. In sharp contrast, the non-degassed samples do not form high mol. wt. polymer. At 0.05 Mrad, oxygen degassed samples yield PVP with molecular weights of 1.65-1.8.times.10.sup.6 compared with only about 0.35.times.10.sup.6 in air. At 0.10 Mrad, all samples have molecular weights of about 1.8 to 2.0.times.10.sup.6.

The following experiments were carried out to demonstrate the advantageous effects of free radical scavengers in inhibiting solution polymerization and gelation during the graft polymerization process, especially at high monomer concentration.

PMMA samples were surfaces grafted with PVP using gamma irradiation as in Example 1. Ascorbic acid (AscA) was used as a radical inhibitor in these experiments. The irradiation conditions are set forth in Table 15.

a) 30% NVP/0.5 mM AscA/2.5% EtoAc/0.2 Mrad*

b) 30% NVP/0.5 mM AscA/2.5% EtoAc/0.15 Mrad

c) 40% NVP/1.0 mM AscA/0.1 Mrad

d) 50% NVP/1.0 mM AscA/0.1 Mrad

e) 50% NVP/1.0 mM AscA/0.2 Mrad*

C. A. for all PMMA samples in Table 15 were 18.degree.-24.degree. indicating very high hydrophilic grafts. Dose rates used were 33 rads/min. A dose rate of 667 rads/min for (b) was also used. Solution polymer gelation can occur under these conditions at these monomer concentrations (30-50%) if a radical inhibitor such as AscA is not used. The AscA significantly inhibits solution polymerization without interfering with grafting yielding low mol. wt. solution polymer. In addition to C. A., PVP grafting was verified by ESCA and FTIR-ATR analysis showing the presence of surface nitrogen and the PVP imide carbonyl group. Good mechanical properties were demonstrated by an abrasion test showing little change in C. A. or surface nitrogen after abrasion.

This example demonstrates the large favorable effects of hydrophilic gamma graft surface modification on reducing tissue adhesion by measuring corneal endothelium adhesion and cell adhesion using fibroblast cells. These are important factors in demonstrating the improved biocompatibility and minimal tissue irritation or damage afforded by the hydrophilic graft surface modifications of this invention.

An apparatus which measures the force of adhesion (mg/cm.sup.2) between contacting polymer and tissue surfaces was used to determine adhesion between rabbit corneal endothelium and polymer surfaces. Adhesion force values of about 250-400 mg/cm.sup.2 where measured for PMMA and other hydrophobic polymers evaluated for implants, i.e., silicone, polypropylene, etc. The improved hydrophilic gamma graft surfaces, prepared under preferred process conditions, exhibit much lower adhesion; below 150 mg/cm.sup.2 and often less than 100 mg/cm.sup.2. This is accompanied by a major reduction in endothelium cell damage as measured by SEM; from about 50-80% damage for PMMA or silicone to 20% or less for surfaces gamma grafted under preferred process conditions of this invention.

The gamma graft surface modifications of this invention also shown a major reduction in cell adhesion as demonstrated by exposure to live cel cultures of chick embryo fibroblast cells (CEF) or rabbit lens epithelial cells (LE). Experiments indicate that 2-4 times more CEF or LE cells adhere to PMMA as compared to PVP graft modified PMMA. Grafts prepared at 0.1 Mrad and using 15% NVP, for example, showed adherence of only 35% of the number of CEF cells which adhere to PMMA. Similarly, PHEMA grafts on PMMA exhibited only 38% cell adhesion and 15:1 NVP: HEMA (at 16% total monomer) exibited only 20% CEF cell adhesion compared to PMMA. Under optimal conditions of the method of the invention for PVP surface modified PMMA, PC or PSi, less than 1-2 LE cells per sq. mm. adhere as compared to about 10 LE cells or more to unmodified PMMA, PC or PSi.

This example demonstrates the graft polymerization of HEMA and mixtures of NVP and HEMA on PMMA.

The method of Example 1 was repeated utilizing a 16% NVP/HEMA (15:1) aqueous solution at about 1300 rads/min and 0.10 Mrad dose. The PVP-PHEMA surface modified PMMA had a C. A. of 17.degree.. Under similar conditions, a 7, NVP/HEMA solution (5:2) gave a surface with C. A. 23.degree., and a 2.5% HEMA solution gave a surface with C. A. 18.degree..

This example demonstrates the graft copolymerization of anionic or cationic monomers with the hydrophilic monomers of this invention using ionic monomers with NVP.

a. The method of Example 1 was used with PMMA substrate and 15% NVP plus 1-5 wt % of acrylic acid (AA) or crotonic acid (CA) as comonomers at 0.1 Mrad and 1235 rads/min. Contact angles were 18.degree.-22.degree. and endothelium adhesion was about one half or less that of unmodified PMMA indicating formation of a good hydrophilic graft coating. Similar results can be obtained using dimethylaminoethylacrylate to produce cationic graft coatings. Styrene sulfonic acid (SSA) was also used to produce anionic grafts with NVP on PMMA according to the method of Example 1. Using an SSA:NVP ratio of 1:2 (33% SSA) and total monomer concentration of 30% at 0.15 Mrad and about 700 rads/min. dose rate, hydrophilic grafts with 30.degree.-40.degree. C. A. were prepared.

b. Styrene sulfonic acid sodium salt (NaSSA) was used to prepare highly hydrophilic anionic copolymer grafts with NVP on silicones (PDMS). PDMS samples were cleaned by sonication in ethanol and vacuum dried prior to irradiation in aqueous monomer solutions. Table 16 lists grafting conditions, monomer concentrations and contact angles for graft surfaces prepared at a dose rate of about 700 rads/min.

As shown in Table 16, under conditions of even a relatively low total dose of 0.05 Mrad, using 40% total monomer and 50% anionic NaSSA comonomer with NVP, very hydrophilic (C. A. 17.degree.) anionic grafts were achieved.

This example demonstrates the hydrophilic monomer surface grafting of polypropylene (PP) and the importance of oxygen degassing for effective surface modification.

Hydrophilic surface grafts on polypropylene are not readily prepared by gamma irradiation of aqueous NVP in the presence of oxygen. Under conditions of Example 1, even at gamma doses<0.1 Mrad and monomer concentrations<10%, little surface hydrophilicity and little reduction in C. A. occurs. However, in oxygen degassed media, at 157 rad/min, and doses as low as 0.01-0.05 Mrad with 10% NVP, contact angles were about 15.degree.. Very hydrophilic PP grafts which are also mechanically stable by a mechanical abrasion test are thereby readily prepared using oxygen degassed process conditions. This is especially important for gamma graft surface modification of IOLs with PMMA optics and PP haptics.

Polycarbonate is a useful engineering plastic for ocular implants. Surface modification of polycarbonate is most readily accomplished using gamma radiation of oxygen degassed aqueous monomer NVP solutions, e.g., grafting conditions of oxygen degassed 10% NVP at 93 rad/min and 0.05 Mrad dose yield C. A. 19.degree..

Although silicone (Psi) does not gamma graft with NVP as readily as PMMA, PSi surfaces were modified using oxygen degassed 10% NVP solutions. Irradiation to 0.05 Mrad at 93 rad/min yields C. A. of about 45.degree. indicating significant surface hydrophilicity. Higher doses, swelling solvents, higher monomer concentrations and different hydrophilic monomers can produce improved hydrophilicity. For example, gamma grafting of NVP/HEMA (10:1) at 0.10 Mrad and 157 rad/min even without oxygen degassing yields grafts with 30.degree. C. A.

Polyvinylidene fluoride (PVDF) is an example of a fluorocarbon polymer which can be surface modified by gamma irradiation of aqueous NVP, NVP/ water-methanol solutions or EtOAc-water systems. Hydrophilic grafts, with C. A. about 30.degree., are prepared at 326 rad/min and 0.20 Mrad. However, PVDF is preferably grafted using oxygen degassed process conditions. Conditions of 157 rad/min, 0.05 Mrad, and 10% aqueous NV produce PVP grafts with C. A. 17.degree.. Since NVP monomer is also an effective swelling solvent for PVDF, allowing pre-radiation swelling time is favorable for producing improved grafts. For example, C. A. as low as 14.degree. is obtained using 5 hrs. swelling time with 7% NVP, 0.10 Mrad and 94 rads/min.

Lenses with Haptics of Different Polymers

One of the important aspects of this invention is the discovery that certain specific grafting process conditions make it feasible to surface modify combinations of materials to be used as lens/haptic pairs in ocular implants. Surface grafting of an assembled IOL can then take place in a one-step simultaneous grafting procedure yielding improved more biocompatible surfaces. Lens materials such as PMMA, PC and PSi can thereby be grafted under specific conditions of this invention which also achieve good grafting of haptic fiber materials such as PVDF or PP. Table 16 summarizes some lens/haptic combinations with preferred mutual grafting conditions for obtaining improved PVP grafts.

PMMA/PP and PMMA/PVDF

It has been demonstrated that PMMA and PP gamma graft under degassed conditions at 157 rad/min, 0.05 Mrad, 10% NVP. These conditions yield contact angles of 20.degree. and 15.degree. for PMMA and PP, respectively, and are mechanically stable. Non-degassed PP does not graft efficiently under conditions similar to PMMA because of the adverse effect oxygen has on PP surface grafting.

PVDF surface graft studies also indicate the importance of oxygen degassing. A 10% degassed aqueous NVP solution, irradiated at 157 rad/min to 0.05 Mrad, gives good hydrophilic grafts on both PMMA and PVDF. See Table 16.

PC/PP and PC/PVDF

PC and PP graft under similar gamma irradiation conditions when NVP solutions are degassed. Using 157 rad/min, 0.05 Mrad, and 10% aqueous NVP solutions, efficient hydrophilic grafting occurs on both polymers yielding contact angles of 19.degree. and 15.degree., respectively.

PVDF and PC are both grafted under the same conditions which graft PC/PP and PMMA/PP combinations; e.g., 157 rad/min, 0.05 Mrad, 10% degassed NVP. Since PVDF swells in NVP, gamma grafting with prior swelling time can result jin improved binding of PVP to the PVDF. Conditions are thereby afforded for simultaneous hydrophilic polymer grafting to IOLs, or other ocular implants which are made of two or more polymers as indicated above. See Table 16.

Intraocular lenses (IOLs) were surface modified using several conditions described in the above examples and implanted in rabbit eyes for periods of up to one year to demonstrate the good bioacceptance of hydrophilic gamma polymerization surface modified IOL ocular implants prepared by the process conditions of this invention. For example, Sinskey-style-037 J-loop lenses (PMMA optic/PP haptics) were surface modified with PVP, ethylene oxide sterilized, and implanted in the anterior chambers, and one-piece flexible haptic PMMA IOLs were implanted in the posterior chambers of New Zealand white rabbits. Process conditions for IOL surface modifications include:

(a) 15% NVP, 0.10 Mrad, 30 and 12 rads/min, C. A. 20.degree.-25.degree.;

(b) Conditions of Example 4, Table 15, a, b, d.

Periodic slit lamp examinations of eyes, histopathology after one year and microscopic examination of explanted lenses (compared to ungrafted PMMA control IOLs), indicated good biocompatibility and normal behavior for the hydrophilic polymer surface graft modifications of this invention.

This example illustrates the efficient grafting which can be achieved by the process of this invention at extremely low gamma doses (0.005 Mrad or less) even at very low aqueous monomer concentrations (0.5 wt % or less).

PVDF surfaces were surface modified using conditions described in the above examples at the extremely low gamma-radiation doses (0.01 and 0.005 Mrad) and low HEMA monomer concentrations (0.5-2.0%) summarized in Table 18. PVDF samples were cleaned, gamma irradiated in aqueous HEMA solutions and washed according to the general method of Example 1. Highly hydrophilic surface graft modifications are achieved as indicated by the low contact angles listed in Table 18. Good graft efficiency for PHEMA on PVDF under these extremely low dose and monomer concentration conditions is further confirmed by the XPS analyses given in Table 19 which shows little surface fluorine and a corresponding increase in carbon for the PHEMA-g-PVDF; a surface analysis which closely approximates the composition of PHEMA.

Even at doses as low as 0.005 Mrad or less and monomer concentrations as low as 0.5 wt % or less, extremely hydrophilic PHEMA grafts are obtained. For comparison PVDF itself is very hydrophobic and has a contact angle greater than 58.degree..

The XPS surface analysis clearly shows that efficient surface grafting of PHEMA occurred at 0.005 Mrad. The surface carbon concentration for the graft was about that expected for a PHEMA surface and very little surface fluorine for PVDF was detected.