Toward electron-beam sterilization of a pre-assembled Boston keratoprosthesis

Abstract

Purpose:

To evaluate the effects of electron-beam (E-beam) irradiation on the human cornea and the potential for E-beam sterilization of Boston keratoprosthesis (BK) devices when pre-assembled with a donor cornea prior to sterilization.

Methods:

Human donor corneas and corneas pre-assembled in BK devices were immersed in recombinant human serum albumin (rHSA) media and E-beam irradiated at 25 kGy. Mechanical (tensile strength and modulus, and compression modulus), chemical, optical, structural, and degradation properties of the corneal tissue after irradiation and after 6 months of preservation were evaluated.

Results:

The mechanical evaluation showed that E-beam irradiation enhanced the tensile and compression moduli of human donor corneas, with no impact on their tensile strength. By chemical and mechanical analysis, E-beam irradiation caused a minor degree of crosslinking between collagen fibrils. No ultrastructural changes due to E-beam irradiation were observed. E-beam irradiation slightly increased the stability of the cornea against collagenase-induced degradation and had no impact on glucose diffusion. The optical evaluation showed transparency of the cornea was maintained. E-beam irradiated corneal tissues and BK-cornea pre-assembled devices were stable for 6 months after room-temperature preservation.

Conclusions:

E-beam irradiation generated no detrimental effects on the corneal tissues or BK-cornea pre-assembled devices and improved native properties of the corneal tissue, enabling prolonged preservation at room temperature. The pre-assembly of BK in a donor cornea, followed by E-beam irradiation, offers the potential for an off-the-shelf, ready to implant keratoprosthesis device.

Introduction

Corneal diseases are one of the primary causes of blindness affecting people of all ages¹. Transplantation of a corneal allograft currently is the primary treatment for corneal blindness, including penetrating keratoplasty, deep anterior lamellar keratoplasty, Descemet’s stripping automated endothelial keratoplasty, and Descemet’s membrane endothelial keratoplasty. Nearly 50,000 corneal transplants were performed in the US in 2015, a 53% increase since 2005². While donor corneas are readily available in economically advantaged countries, there is a severe scarcity of donor corneas globally³. Tissue suitability compounds the issue, as regulations for donor screening, serological and microbiological testing are stringent^4–6. In addition, the necessity for short donor-to-recipient time frames and the potential for microbial contamination contributes further to the paucity of donor cornea availability^7–8. Substantial recent efforts have focused on developing new modalities for corneal replacement, ranging from tissue engineering strategies⁹ to decellularization of xenografts;^10–11 these technologies are still in their infancy. At present, tissue-engineered scaffolds are incapable of mimicking the native cornea’s biomechanical properties and molecular microarchitecture^12–13. Xenografts, although anatomically and biomechanically similar to the human cornea, are antigenic and susceptible to rejection^{11, 14–17}. In addition to these obstacles, severe ocular surface disorders in which the eyelids and conjunctiva are preserved but there is deficiency of limbal stem cells, such as after chemical burn and in aniridic keratopathy^18–21, represents an unmet need among corneal disorders, and in most cases is not amenable to corneal transplantation. At present, keratoprosthesis implantation is the only viable alternative to corneal allograft surgery in such patients²², but the most commonly used device, the Boston keratoprosthesis^23–25, still requires a donor cornea for implantation.

Prolonging the viability of corneal tissue in storage could mitigate some of the issues associated with the lack of donor corneas, improving overall accessibility. This is especially significant to those living in underprivileged countries⁷. Cryopreservation of corneal tissue in different media^26–29, glycerol-preservation³⁰, freeze-drying³¹, and ionizing irradiation (gamma and electron beam [E-beam]) represent different approaches aimed at extending the shelf life of corneas for transplantation. Unlike frozen and glycerin-preserved corneas, irradiated grafts are sterile and can be immediately used without the need for reconstitution. It has been shown that irradiation may reduce the allogenicity of donor corneas⁷. Tissue banks frequently use gamma irradiation for sterilization of biological tissues against various contaminations (e.g. bacterial, viral, fungal, prional contamination)³². Moreover, the transplanted gamma-irradiated human cornea has demonstrated satisfactory properties without graft rejection, neovascularization, or loss of transparency³³.

E-beam irradiation is capable of much higher dosing rates than gamma irradiation, and this consequently reduces the necessary exposure time and potential degradation of the irradiated tissue^34–36. In studies of long-term storage, E-beam irradiation has been shown to effectively sterilize corneal tissue, while only slightly increasing light scattering, and permitted tissue storage for up to 2 years³⁷. However, to our knowledge, E-beam irradiation’s effect on the biomechanical, structural, and degradation properties of the cornea has not been comprehensively studied.

Herein, corneal tissues immersed in rHSA were irradiated with 25 kGy and their chemical, mechanical, optical, structural, and enzymatic degradation properties were evaluated. We have found E-beam irradiation does not induce detrimental effects on the cornea’s properties. E-beam irradiation potentially enables a simplified pre-assembly of the BK to a donor cornea in a central facility by a skilled technician, storage at ambient temperature for a prolonged period of time, allowing accessibility to keratoprosthesis treatment for corneal diseases world-wide (Fig. 1).

Figure 1.

Illustration of E-beam irradiation and preservation of corneas for full-thickness, lamellar, and keratoprosthesis applications.

Materials and Methods:

Materials.

Recombinant Human Albumin (rHSA) was acquired from InVitria (Junction City, KS) and dissolved in phosphate buffer saline (PBS) solution to a 20% w/w solution. Donor human corneas were kindly provided by Lions VisionGift (Portland, OR) for research purposes only and sent without donor identifying data, not requiring ethical approval.

E-beam irradiation.

Corneal tissues were placed in 5 mL glass vials, filled with rHSA solution, and irradiated using a Van de Graaff (Model K) electron accelerator (Electron Technology Company; South Windsor, CT) at 2.6 MeV. A dose rate of 5 kGy per pass was used to reach 25 kGy irradiation.

Fourier transform infrared spectroscopy (FTIR).

6-mm-diameter discs were dissected from native corneas (NC), E-beam irradiated corneas (E-beam), and E-beam irradiated corneas that were preserved for 6 months (E-beam-6), washed 3 times with distilled water, and freeze-dried overnight under 0.02 mbar. FTIR spectra of tissues were collected in the Mid-IR range using a Nicolet iS50 FT-IR Spectrometer (Thermo Scientific; Waltham, MA) equipped with a diamond Attenuated Total Reflectance (ATR) accessory. Spectra were obtained from 500 to 4000 cm⁻¹ averaging 64 scans at 0.5 cm⁻¹ resolution.

Swelling ratio.

Corneal Tissues (NC, E-Beam, and E-beam-6) were dissected into 6 mm discs, blot dried, and weighed (W_i). The tissues were then immersed in PBS solution and incubated at 37 °C for a recorded time (2–24 h). Finally, the swollen tissues were blot dried and weights were measured again (W_s). The swelling ratio (S) for the corneas [n = 5] were obtained by the following equation:

$$S(\%)=\frac{\left(W_S-W_i\right)}{W_i}\times 100$$

Evaluation of mechanical properties.

The tensile modulus and strength of the corneal tissues were measured using a Mark-10 ESM 303 motorized test stand (Mark-10 Corporation, NY, USA) using a 50N load cell. Sample corneal tissues were dissected into dumbbell-shaped specimens. The crosshead speed was 1 mm/min; the initial grip separation was 5 mm. The compression modulus was measured on the same instrument on 4-mm-diameter discs dissected from those groups [n = 5]. Both tensile and compression modulus was calculated from the linear derivative of the stress-strain curve in the high stiffness range (i.e. 20–40% strain).

Assessment of optical properties.

The optical properties of trephined corneas (6-mm-diameter central) from each group were evaluated in PBS using a UV–Vis spectrometer (Molecular Devices SpectraMax 384 Plus Microplate Reader, CA, USA). Transmittance spectra (%T) were acquired from 300–900 nm at 1 nm wavelength increments [n = 4] and corrected with PBS background. The mean %T for each group was plotted as a function of wavelength.

Transmission electron microscopy (TEM).

Specimens were fixed with Karnovsky’s fixative (50% strength, pH 7.4) (Electron Microscopy Sciences, Hatfield, PA) overnight at room temperature. The samples were washed with 0.1 M Cacodylate Buffer (Electron Microscopy Sciences) for 5 min, then washed three times with PBS before post-fixing with immersion in 2% osmium tetroxide (Electron Microscopy Sciences) for 1.5 h at RT. Following this, the specimens were en bloc stained with 2% aqueous uranyl acetate for 30 min, then dehydrated in ethanol. The samples were then embedded in epoxy resin (Tousimis, Rockville, Maryland, USA) and 80nm thin sections were obtained with a Leica EM UC7 ultramicrotome (Leica Microsystems, Buffalo Grove, IL, USA, diamond blade). TEM analysis was obtained at 120 kV using an FEI Tecnai G2 Spirit transmission electron microscope, (FEI, Hillsboro, Oregon, USA). Collagen fibril diameter and collagen interfibrillar Brag spacing were quantified using ImageJ software (NIH, Bethesda, Maryland) from multiple images taken from each sample (n = 3).

In Vitro Biodegradation.

Enzymatic degradation of the corneal tissues was evaluated using collagenase from Clostridium histolyticum, as reported elsewhere¹⁰. Briefly, dissected specimens were placed in 0.1M Tris-HCL buffer (ph 7.4) with collagenase (5U/mL) and CaCl₂ (5 mM) and incubated at 37 °C, with solution replacement at 8-hour intervals. Remaining residues were removed from the solution, immersed in DI water, and lyophilized. The dried masses at different time points (W_f) were weighed, and retention was calculated [n = 4] using the following equation:

$$\mathit{\text{Retention}}(\%)=\frac{W_f}{W_i}\times 100$$

Glucose diffusion.

A Static Franz cell system with a diameter of 9 mm, equipped with a small stirrer bar, (PermeGear, PA, USA) was used to evaluate diffusion of glucose through non-irradiated vs. irradiated corneas. Briefly, corneas were inserted between the two compartments of the Franz cell. The upper chamber was filled with 1 mL PBS and the bottom one was filled with a glucose solution with a concentration of 2000 mg/dL. The unit was placed inside an incubator at 37 °C, and the solutions were stirred with the magnetic stirrer. The diffusion of glucose through each cornea into the upper chamber was measured over time using a Counter Next EZ blood glucometer (Bayer, Parsippany, NJ, USA). Diffusion coefficients were determined as previously described³⁸.

BK-Cornea fitting properties.

Corneal tissues were trephined with a 3 mm biopsy punch, assembled with BK, immersed in rHSA media, E-beam irradiated, and preserved at room temperature for 6 months.

For the adhesion test, the backplate was carefully removed, and the assemblies were placed on the designed corneal holder (Fig. 6d) and set up with a mechanical tester as previously described³⁹. A pushing method was carried out with a crosshead speed of 1 mm/min. The maximum force measured was averaged (n = 5) and compared to a control group (Control: after removing the irradiated BK stem from the corneal tissue, the stem was inserted back in the trephined cornea and test performed similarly).

Figure 6.

a) Burst pressure set up. The pressure of an artificial chamber set up with BK-carrier assembly for corneal tissues (NC, E-beam, E-beam-6) as a function of injected PBS volume (b) and consequent pressure/PBS volume slope (c). d) Schematic illustration of pushing test apparatus to measure the adhesion force between BK stem and human corneal tissue (NC, E-beam, E-beam-6) compared to control (irradiated BK stem was removed from the corneal tissue, the stem-carrier was assembled again and adhesion test was repeated) (e). f) Trephination diameter of the corneal tissues assessed and averaged from 4 orthogonal measurements shown in the inset [n = 4].

In the analysis of burst pressure, BK-cornea assemblies were placed in an artificial corneal chamber (Barron Precision Instruments; Grand Blanc, MI) equipped with a PBS delivery syringe pump and a digital pressure transducer (PS-2017, PASCO; Roseville, CA) for data collection. The syringe pump (NE-300, ArrEssPro Scientific; Farmingdale NY) delivered a consistent flow (0.2 ml/min) into the chamber until the pressure reached ca. 200 mm Hg.

The specimen’s trephination size was photographically assessed averaging 4 orthogonal measurements acquired from each cornea specimen (n = 5) using a high-resolution camera (Dino-Lite Edge, AM73915MZTL 5MP; Torrance, CA).

Optical Coherence Tomography (AS-OCT).

To assess how each step of treatment impacts the specimens and their thickness, Anterior Segment Optical Coherence Tomography (AS-OCT) was performed (OCT Spectralis; Heidelberg Engineering Inc.; Franklin, MA). After acquiring several images for each cornea (n = 4), the thicknesses of both the central and peripheral sections of the cornea were measured and compared to those of NC control specimens. AS-OCT was also used in diagnostic analysis of donor tissue interface (between the donor tissue and the BK assembly) after irradiation and preservation for 6 months.

Results

Mechanical properties

Uniaxial extension of the corneal tissue pre and post-treatment showed that E-beam irradiation increased the tensile modulus of corneal tissue from 13.9 ± 0.6 MPa (NC) to 16.2 ± 1.6 MPa (E-beam) and 16.1 ± 1.3 MPa (E-beam-6) with p values of 0.0359, for both comparisons (Fig. 2a–b). A statistically insignificant decrease in tensile strength was observed from 7.81 ± 0.72 MPa (NC) to 7.27 ± 0.47 MPa (E-beam) and 7.22 ± 0.45 MPa (E-beam-6) with p values of 0.314 and 0.261, respectively (Fig. 2c). Extension at break showed similar behavior (NC: 110 ± 10 %, E-beam: 96 ± 12 % and 95 ± 11 %) (Fig. 2d). Uniaxial compression revealed significantly higher compression moduli in irradiated corneas (E-beam and E-beam-6) (4.50 ± 0.50 MPa and 4.32 ± 0.37 MPa, respectively) compared to those of NC (3.54 ± 0.30 MPa) with the p values of 0.0068 and 0.0231, respectively (Fig. 2e–f). In comparing 6-month preserved specimens versus ‘new’ corneas, we did not observe any significant differences in mechanical properties.

Figure 2.

Mechanical characterization of the human cornea after E-beam irradiation. a) Tensile measurement set up. b) Tensile modulus of the E-beam irradiated (E-beam: immediately after E-beam irradiation, and E-beam-6: 6 months after E-beam irradiation) corneas is significantly higher than that of the native cornea (NC). c) Tensile strength and extension break of E-beam irradiated cornea is slightly less than those of NC. d) Compression measurement set up. e) compression modulus of the E-beam irradiated corneas is significantly higher than that of NC.

Optical evaluation

The optical evaluation of corneal tissues noted high optical transmission (> 79%) in the visible spectrum for all three groups (Fig. 3). There is a transmittance loss in the UV range (300–400 nm) observed from 36.9 ± 3.2 % (NC) to 31.8 ± 2.6% (Ebeam) and 26.6 ± 4.9 % (E-beam6) with the p values 0.14 and 0.002, respectively (Fig. 3b). We also did not observe any noticeable color change in the corneal tissue upon irradiation and after preservation for 6 months, suggesting irradiation and preservation have negligible visible-range optical effects on corneal tissue.

Figure 3.

a) Transmission of NC, E-beam, and E-beam-6 as a function of wavelength (nm), and b) its quantifications in both UV (300–400 nm) and Visible (Vis: 400–900 nm) range. E-beam irradiation and preservation does not impact the optical transmission in the Vis range, yet they decrease transmission in the UV range.

Chemical and structural characterization

FTIR spectra of all three groups are nearly superimposable in the lower wavenumber range (450–2700 cm⁻¹) (Fig. 4a). Significant increases in the intensities of the vibrational bands were observed in the upper acquired range from 2850 and 2917 cm⁻¹, containing C-H -sp2 and C-H -sp3 stretching bands. These changes indicate E-beam exposure may have induced free radical formation in the irradiated specimens, recombining to form C-C and C=C bonds as previously described in polymeric materials^40–41. FTIR spectra of E-beam and E-beam-6 are nearly superimposable across the acquired range, indicating perseveration of the irradiated cornea at room temperature for 6 months did not significantly impact the chemical properties of the cornea.

Figure 4.

a) Chemical characterization of corneas before and after irradiation (NC, E-beam, and E-beam-6) using attenuated total reflection-Fourier transform infrared (ATR-FT-IR) spectroscopy. Swelling ratio (b), retention of the corneal tissues in the presence of collagenase (c) before and after irradiation as a function of incubation time (T = 37 °C and 5% CO₂). Glucose diffusion (d) through corneal tissues before and after irradiation.

Swelling studies up to 24 h show irradiated corneas exhibit significantly lower swelling ratios with a maximum swelling ratio of 25 ± 7 % compared to the NC (80 ± 8 %) (Fig. 4b). The preserved corneas showed similar swelling behavior to those of freshly irradiated corneas with a maximum swelling ratio of 26 ± 6 %.

In studies of biodegradation in collagenase, the retention of corneal tissue is a function of incubation time. E-beam irradiated corneas had statistically insignificantly higher stability compared to NC (Fig. 4c). While NC was completely digested after 35 h of incubation, nearly 5% of the residual mass was preserved in E-beam and E-beam-6 specimens. In contrast, glucose diffusion studies showed no significant difference between the permeability of the corneal tissue before (NC) and after irradiation (E-beam and E-beam-6) with p values > 0.9 (Fig. 4d).

Transmission electron microscopy (TEM)

The structural organization of the anterior, mid, and posterior sections of the corneal stroma, pre, and post-irradiation, are shown in Fig. 5a–b. The arrangements of collagen fibrils at both lower and higher magnifications seem similar among all groups for all three sections. Quantitative image analysis showed that the collagen fibril diameter and collagen interfibrillar Bragg spacing of all analyzed sections of the cornea were similar, indicating that irradiation and preservation did not alter the ultrastructure.

Figure 5.

Transmission electron microscopy (TEM) micrographs of corneal tissues (NC, E-beam, and E-beam-6) at different layers of the cornea (anterior, mid, and posterior) and low (a) and high (b) magnifications. Quantifications of the collagen fibril diameter (c) and collagen interfibrillar Brag spacing (d). Scale bars: (a), 10μm; (b), 1μm.

BK-Cornea fitting properties

To examine if E-beam irradiation impacts the fitting properties of the assembled device and the adhesion between the BK stem and the trephined cornea in the preassembled device after irradiation or after preservation for 6 months, we performed burst pressure and adhesion tests.

Burst pressure measurements did not show leakage from the interface of the BK stem with donor tissue up to 200 mmHg (Fig. 6a–b) in all three specimen groups (NC, E-beam, and E-beam-6). This suggests irradiation and preservation did not induce expansion of the trephination opening or degradation of the corneal tissue adjacent to the stem; the latter could create a physical gap between the BK stem and donor tissue. The burst pressure analysis shows that irradiation significantly increased the slope of pressure/volume graphs. This suggests irradiated corneas possess higher stiffness, a finding supported by prior mechanical testing (e.g. higher tensile and compression moduli) (Fig. 2).

The adhesion test showed that the force necessary to remove the BK stem from the irradiated BK-carrier assembly immediately after irradiation (0.096 ± 0.035 N), and after preservation for 6 months (0.098 ± 0.034 N), are significantly higher than that of NC (0.020 ± 0.013) with p < 0.001 for both comparisons. We also re-assembled the BK stem with the irradiated corneal tissue and performed adhesion studies to determine if adhesion originates from mechanical (e.g. cornea contraction) or chemical factors (e.g. interfacial reactions at the donor cornea/BK stem). Our studies indicated that the adhesion force between the irradiated cornea and BK stem is 0.018 ± 0.014 N, similar to NC-BK stem specimens. We also recorded the trephination diameter before and after irradiation, and after irradiation and 6-month preservation. No significant differences were observed between groups (NC, E-beam, E-beam-6) (p > 0.7). These results suggest trephination size was not affected by irradiation, indicating the observed higher adhesion properties did not originate from a shrinking of the trephination opening induced by E-beam irradiation.

Optical Coherence Tomography (AS-OCT)

To evaluate the impact of E-beam irradiation on the physical properties of the cornea, AS-OCT analysis was performed (Fig. 7). This study indicates irradiated corneas are thicker than NC controls in both the corneal periphery and center (Fig. 7b). Cornea thickening or swelling effects did not appear to originate from irradiation but rather from the addition of rHSA. In our study, the rHSA incubated cornea control group exhibited higher thickness than post-irradiation specimens (Fig. 7a–b). After E-beam irradiation, corneal thickness measurements were stable for up to 6 months. In addition, AS-OCT analyses performed immediately after irradiation and after 6 months of preservation, the BK-carrier assembly was stable without deterioration of the BK-corneal interface, generation of unintended gaps, or any retraction or abnormality for the corneal tissue (Fig. 7a).

Figure 7.

a) AS-OCT of human corneal tissue, or/and Boston Keratoprosthesis (BK)-cornea pre-assembled devices (native cornea: NC) and after each step of treatment: i) addition of rHSA media, ii) E-beam irradiation, and iii) and after 6 months of preservation at room temperature along with b) the quantifications of their thickness from central and peripheral sections of the cornea. After 6 months of irradiation, the AS-OCT showed that there is no unintended space at the interface of the corneal tissue (1) and BK stem (2), suggesting that there is no deterioration at the interface or dilation of the trephination opening.

Discussion

A full characterization of biological implants is paramount to determining their suitability for clinical applications. Despite US FDA approval, and wide availability of E-beam-irradiated human corneas for human implantation, with few exceptions³⁷, the biological effects of E-beam irradiation on corneal tissue have not been published. We evaluated the impact of E-beam irradiation on the chemical, optical, structural, mechanical, and biological properties of corneal tissue. We demonstrated that E-beam irradiation induces no apparent detrimental effects on the tissue and that the irradiated cornea’s properties are retained for up to 6 months after irradiation.

E-beam irradiation increased the tensile modulus of the cornea yet decreased its tensile strength. The former indicates the interfibrillar crosslinking of the collagen fibrils and the latter irradiation-induced chain/fibril scission. Such crosslinking between collagen fibrils also accounts for the increased compression modulus of the corneal tissue (Fig. 2d–e). A crosslinked structure retaining mechanical properties over time suggests the stability of irradiated cornea and indicates the suitability of E-beam treatment to preserve native biomechanical characteristics of cornea.

Sterilization of bone, bone-tendon-bone, and anterior cruciate ligament using E-beam have been well studied. Most studies suggest irradiation conditions such as media compositions affect the biomechanical properties of the tissues^42–43. While some studies pointed out that the higher doses have detrimental effects on biomechanical properties of studied tissues, most reports indicate minimal changes at 25 kGy^{36, 42–46}, consistent with our study’s findings.

Corneas irradiated at 25 kGy showed slightly higher visible optical transmittance. The higher transmission stems from inter-fibril crosslinking induced during irradiation, impeding tissue swelling. The latter has previously been shown to cause light scattering, reducing the transparency of the cornea^47–49. Alternatively, gamma irradiation has been shown to lead to the yellowing of corneal tissue⁵⁰. A reduced UV transmittance is likely due to the formation of absorbing functional moieties such as C=C bonds. After preservation, however, the transmission of light in the entire spectrum slightly declines, which is believed to originate from minor swelling of tissue over time as previously described³⁷.

The swelling ratio of the irradiated corneas was significantly lower than in the NC control group (Fig. 4c). This observation complements the chemical and mechanical studies performed, denoting the formation of crosslinking between collagen fibrils. Such crosslinking is also responsible for resistance to biodegradation of irradiated specimens versus the control group (Fig. 4c). The irradiation and the resultant collagen fibril crosslinking did not significantly alter the ultrastructural appearance (Fig. 5). Such an arrangement of collagen in the cornea is responsible for the optical and mechanical integrity of the cornea and also regulates corneal cellular functions^51–53.

Corneal allograft surgery is the primary solution when the cornea has lost its optical transparency. However, some conditions, including repeated corneal allograft rejection, are associated with a poor prognosis. Implantation of a BK, the most widely used prosthetic cornea, is a valued alternative for such patients^54–57. Although the most common indication for the implantation is multiple allograft failure⁵⁴, patients with chemical ocular injury^58–59, atopic keratoconjunctivitis⁶⁰ and other causes of limbal stem cell deficiency⁶¹ may also benefit from BK surgery, and indications for its use are expanding^62–67. However, complications of surgery are significant, and include glaucoma, keratolysis (corneal melt), retro-prosthetic membrane, sterile vitritis, infectious keratitis, and endophthalmitis^68–71, any of which can lead to loss of recovered vision.

Retroprosthetic membrane is the most common complication of BK implantation, occurring in up to 65% of BK recipients⁷⁰, and is most common when the recipient has congenital aniridia⁵⁹. Retroprosthetic membranes not only obstruct the visual axis to reduce vision, but are also associated with sterile keratolysis⁷², presumably due to reduced access of the cornea to nutrients from the aqueous humor⁷⁰. Keratolysis can lead to corneal perforation, hypotony, infection, suprachoroidal hemorrhage, and device extrusion⁶⁹. Our evaluation of glucose permeability of corneal samples before and after irradiation showed no impact on the diffusion of glucose through irradiated corneal tissue, indicating that E-beam irradiation should not contribute to keratolysis (Fig. 4d). In addition, chemical and biodegradation analyses showed no impact on tissue stability and less degradation by collagenase after irradiation. Based on our results, an E-beam irradiated cornea within a pre-assembled BK should be no less resistant to melt than native corneal tissue.

Currently, the BK device is ethylene oxide sterilized, shipped to a hospital, then assembled to a fresh corneal donor graft in the operating room at the time of surgery. Pre-assembly of BK followed by E-beam sterilization could permit room temperature storage of the assembled device, bypassing many current obstacles for cornea-blind patients. Gamma irradiation of an assembled BK has been previously proposed⁷³. Our data indicate that E-beam irradiation presents distinct advantages for the preservation of properties of the PMMA component and the corneal carrier for the BK device^{34, 74}.

Remarkably, our burst pressure studies showed no interfacial leakage between BK stem and donor cornea to at least 200 mmHg pressure. Our measurements of trephination size also imply a negligible change as a consequence of irradiation. The adhesion study suggests irradiation may induce chemical bonding between BK stem and corneal tissue, possibly improving seal quality. Along with the AS-OCT images, preservation for 6 months post-irradiation did not introduce any loss of adhesion between the BK stem and corneal tissue. The findings here demonstrate that 6 months after pre-assembly and irradiation, the corneal tissues remained stable in the assembled devices. This is analogous to what is required in BK implanted patients^75–77.

In summary, E-beam irradiation of corneal tissue, and its pre-assembly with the BK, does not have a significant negative impact on the structural, chemical, mechanical, or optical properties of the corneal tissue. E-beam irradiation sterilizes the tissue and enables its prolonged storage at room temperature. This may address obstacles associated with the lack of donor corneas globally, especially in underprivileged countries lacking organ banks and robust, rapid transportation systems. Preassembled, E-beam sterilized BK devices would improve access in locales where donor tissues are scarce or unavailable, potentially providing an accessible, expert-assembled, room temperature device for patients in regions where corneal blindness from severe ocular surface disease is most common. This would also reduce operating time and avoid potential errors in assembly by surgeons. Further advancements in the field may arrive in the form of more affordable carrier donors, including adapted xenografts^{11, 50} or artificial constructs^{9, 78–79}. Until then, we suggest E-beam sterilization of a preassembled BK, which this study demonstrates would allow long-term ambient temperature storage, offers a potentially more accessible and safer process to treat corneal blindness.