Reshaping and Customization of SMILE-Derived Biological Lenticules for Intrastromal Implantation

Abstract

Purpose: To evaluate the feasibility of excimer laser reshaping of biological lenticules available after small incision lenticule extraction (SMILE).

Methods: Fresh and cryopreserved SMILE-derived human lenticules underwent excimer laser ablation for stromal reshaping. The treatment effects in the lasered group were compared with the nonlasered group with respect to changes in surface functional groups (by Fourier transform infrared spectroscopy [FTIR]) and surface morphology (by scanning electron microscopy [SEM] and atomic force microscopy [AFM]). Ten SMILE-derived porcine lenticules, five nonlasered (107-μm thick, −6 diopter [D] spherical power) and five excimer lasered (50% thickness reduction), were implanted into a 120-μm stromal pocket of 10 porcine eyes. Corneal thickness and topography were assessed before and after implantation.

Results: FTIR illustrated prominent changes in the lipid profile. The collagen structure was also affected by the laser treatment but to a lesser extent. SEM exhibited a more regular surface for the lasered lenticules, confirmed by the lower mean R_z value (290.1 ± 96.1 nm vs. 380.9 ± 92.6 nm, P = 0.045) on AFM. The lasered porcine lenticules were thinner than the nonlasered controls during overhydration (132 ± 26 μm vs. 233 ± 23 μm, P < 0.001) and after 5 hours in a moist chamber (46 ± 3 μm vs. 57 ± 3 μm, P < 0.001). After implantation, the nonlasered group showed a tendency toward a greater increase in axial keratometry (6.63 ± 2.17 D vs. 5.60 ± 3.79 D, P = 0.613) and elevation (18.6 ± 15.4 vs. 15.2 ± 5.5, P = 0.656) than the lasered group.

Conclusions: Excimer laser ablation may be feasible for thinning and reshaping of SMILE-derived lenticules before reimplantation or allogenic transplantation. However, controlled lenticule dehydration before ablation is necessary in order to allow stromal thinning.

Small incision lenticule extraction (SMILE) has become increasingly popular for the treatment of refractive errors.^1,2 Following the creation of the stromal lenticule with a femtosecond laser, the intrastromal lenticule is dissected and subsequently removed through a small incision.³ The achieved myopic correction is dependent on both the thickness and diameter of the lenticule, with greater myopic correction creating centrally thicker lenticules. After SMILE, the extracted lenticule is a by-product of the procedure and is typically discarded after surgery. However, it may potentially be used for several purposes, for example, for the treatment of hyperopia,^4–6 presbyopia,⁷ aphakia,⁸ keratoconus,⁹ and corneal dystrophies.¹⁰ 

Barraquer¹¹ was the first to perform corneal intrastromal transplantation of biological tissue in rabbits, also known as keratophakia. Since then, both rabbit^4,12–14 and nonprimate studies^15,16 have shown the feasibility of stromal implantation using SMILE-derived lenticules. After creating an intrastromal pocket or flap, using the femtosecond laser, the fresh or cryopreserved biological lenticule may be implanted in a recipient cornea to alter the corneal curvature, depending on the shape and size of the lenticule. The lenticule may also be reimplanted into the pocket of previously SMILE-treated eyes to restore the preoperative refractive state.^12,14,15 

Following the initial animal studies, clinical trials in lenticule implantation for patients with hyperopia have shown promising results with clear lenticules after surgery, although an undercorrection has been reported.^4,5,8,17 In one case study of a previously complicated myopic laser-assisted in situ keratomileusis (LASIK) procedure, a SMILE-derived lenticule was successfully implanted under the corneal flap to reduce the amount of postoperative astigmatism.¹⁷ In patients with keratoconus and centrally positioned cones, implantation of centrally trephined lenticules caused a midperipheral steepening and consequently a central corneal flattening.¹⁸ One study of lenticule implantation for presbyopia has found increased uncorrected near vision acuity after implantation of a 1-mm lenticule button in the nondominant eye, trephined from a 2.50- to a 2.75–diopter (D) lenticule.⁷ Lenticule implantation may also be useful to restore myopia in the nondominant eye of SMILE patients, to achieve monovision when reaching the presbyopic state.^12,14,15 

One major limitation of using any SMILE-derived lenticules for transplantation remains the lenticule thickness, since the thickness is predetermined by the refractive status of the donor patients undergoing SMILE. Thus, there may be limited access to low-powered lenticules, especially in countries where the mean myopic correction is high.¹⁹ Furthermore, lenticule reimplantation in SMILE-treated patients for presbyopic correction requires some level of lenticule thinning if the patient is highly myopic before surgery. In a recent article from our center, only 15 lenticules from a series of 424 cases (3.54%) could be considered useful for reimplantation for presbyopic treatment.²⁰ Hence, if lenticule implantation and reimplantation is considered as a future treatment for refractive and corneal disorders, it would be beneficial to develop a technique to modify the lenticule thickness and reshape it to clinical requirements. 

The aim of the current study was to evaluate the logistics of excimer laser reshaping of biological lenticules available after SMILE. The first part aimed at examining the surface morphologic and compositional changes after phototherapeutic keratectomy (PTK) treatment of human stromal lenticules, which also served as indications that excimer laser ablation had been carried out on these lenticules. The second part aimed at examining and comparing the topographic changes after implantation of lasered and untreated stromal lenticules in porcine eyes. 

Methods

Excimer Laser Ablation of Human Stromal Lenticules

A total of 14 stromal lenticules (six pairs and two single) were obtained after SMILE in myopic patients by using the 500-kHz VisuMax femtosecond laser (Carl Zeiss Meditec, Jena, Germany) as previously described.²⁰ The average age of donors was 28.6 ± 6.9 years (53% men) and average spherical equivalent was −5.69 ± 3.00 D. The six pairs were cryopreserved (−80°C storage for more than 2 weeks) as described previously.²¹ The two single lenticules were treated with PTK without prior cryopreservation. All subjects were treated in accordance with the tenets of the Declaration of Helsinki. 

One cryopreserved lenticule from each pair was treated with PTK and the other served as the nonlasered control. The lenticules were thawed by placing the containers in a 37°C water bath for 10 minutes. The lenticules were rinsed twice in phosphate-buffered saline (PBS, 0.01 M; Life Technologies, Carlsbad, CA, USA) to remove the cryoprotectant agent (10% fetal bovine serum and 20% dimethylsulfoxide; Sigma-Aldrich Corp., St. Louis, MO, USA). Three of the six pairs were coated with (3-aminopropyl)triethoxysilane (3-APTES; Sigma-Aldrich) before excimer ablation to establish the ablative effect of the ArF laser analyzed with energy dispersive X-ray spectroscopy (EDX; Supplementary Fig. S1). 

The lenticules were grasped at the edge with a pair of forceps, placed on a firm surface, and flattened with a sponge spear (HighSOAK; Madhu Instruments Pvt. Ltd., New Delhi, India), semiwet with 0.9% sodium chloride solution (Sigma-Aldrich). PTK was performed by one surgeon (JSM) with the Technolas 217z (Bausch & Lomb, Inc., Rochester, NY, USA), using an emission wavelength of 193 nm, a repetition rate of 50 Hz, and a laser fluence of 120 mJ/cm². The ablation depth was set to 50 μm with an optical zone of 6.0 mm. 

The fresh single lenticules (n = 2) underwent cell apoptosis detection immediately after PTK. The cryopreserved noncoated paired lenticules (n = 6) underwent surface functional group analysis, atomic force microscopy (AFM), scanning electron microscopy (SEM), and EDX scanning the subsequent day. The 3-APTES–coated paired lenticules (n = 6) were only examined with EDX. All lenticules were stored in Optisol (Bausch & Lomb) at 4°C before and after PTK (Supplementary Fig. S2A).²² The storage time in Optisol before PTK was 2 hours for both fresh and cryopreserved lenticules. The storage time in Optisol after PTK was <12 hours for both the fresh and cryopreserved lenticules. 

Cell Apoptosis (TdT-dUTP Terminal Nick-End Labeling [TUNEL])

Immediately after excimer laser ablation, the lenticules were fixed in 3% neutral buffered paraformaldehyde (Sigma-Aldrich) for 20 minutes and washed with PBS. Fluorescence-based TUNEL assay was performed according to the manufacturer's instructions (Click-iT TUNEL Alexa-Fluor Imaging Assay; Thermo Fisher Scientific, Waltham, MA, USA). Subsequently, the lenticules were mounted with Fluoroshield containing 4′,6-diamidino-2-phenylindole (DAPI; Santa Cruz Biotechnology, Santa Cruz, CA, USA). Using confocal laser scanning microscopy (SP8; Leica, Wetzlar, Germany), serial z-stack images (1-μm depth) of the central segment of lenticule were acquired. Three-dimensional (3D) imaging from 2D z-stack images was reconstructed and viewed by using LAS X software (Leica). The control section was set to 30-μm depth from the untreated surface, while the lasered section was set to 30-μm depth from the lasered surface. Densities of TUNEL-positive and DAPI-labelled cells, and total cell density (TUNEL+DAPI) were assessed. 

Functional Group Analysis (Attenuated Total Reflection [ATR]–Fourier Transform Infrared Spectroscopy [FTIR])

The lenticules were fixed in neutral buffered 2% glutaraldehyde and 2% paraformaldehyde overnight, rinsed with copious amount of distilled water, and dried in a desiccator for 3 days. Same samples were used for AFM, SEM, and EDX analyses. The infrared (IR) spectra of lenticules were generated with a PerkinElmer Frontier FTIR spectrometer (PerkinElmer, Inc., Waltham, MA, USA), equipped with an ATR sampling and supplied with a top plate for ZnSe crystal. Spectra were obtained with 64 scans and a 4 cm⁻¹ resolution. Analysis of surface IR spectra was performed as previously described.^23,24 A group average spectrum for the lasered and control lenticules was calculated. Average second derivative spectra were generated from the group average spectra by using the Savitsky-Golay algorithm with nine smoothing points to elucidate the overlapping bands in the spectra. 

Lenticule Surface Morphology (AFM and SEM)

AFM (Digital Instruments, Santa Barbara, CA, USA) was performed with a monolithic Silicon NCH-50 Point Probe (NanoWorld AG, Neuchatel, Switzerland), using tapping mode and a scan size area of 10 × 10 μm². Quantitative surface roughness was described by the root mean square (RMS) value and the average vertical height (R_z) of consecutive highest peaks and lowest valleys within a sampling region (Gwyddion software v2.45; Czech Metrology Institute, Brno, Czech Republic). Following AFM, the lenticules were sputter-coated with a 10-nm-thick layer of gold, and examined with SEM with an accelerating rate of 5 kV (JSM-7600F; JOEL, Tokyo, Japan). 

Ex Vivo Lenticule Implantation Analysis

Fresh porcine eyes (within 6 hours from death) from the local abattoir were used to evaluate the topographic change after lenticule implantation. SMILE was performed by using the Visumax femtosecond laser to harvest 10 stromal lenticules of −6 D refractive power, with the following settings: energy index of 32, with a spot-per-spacing distance of 4.5 μm (cap) and 2.0 μm (rim), cap diameter of 120 μm, lenticule diameter of 6.5 mm, and minimum lenticule thickness of 15 μm.²⁵ The lenticules were stored in PBS overnight. 

The following day, five lenticules were PTK treated before implantation (Wavelight EX500 laser, 4.00-mm optical zone, 5.40-mm ablation zone, and 40-μm ablation depth; Alcon Laboratories, Inc., Fort Worth, TX, USA), while the other five lenticules served as control. The ablation depth was set to 50% of the average lenticule thickness, estimated with optical coherence tomography (OCT) (RTVue OCT; Optovue, Inc., Fremont, CA, USA) just before PTK. 

Lenticule implantation was performed the subsequent day into 10 fresh porcine eyes (Supplementary Fig. S2B). A stromal pocket was created by using the Visumax laser, by cutting a 120-μm corneal flap with a 300° hinge, thereby creating a 60° (3.9 mm) incision for lenticule implantation.²⁶ Energy index and spot-per-spacing distance were similar to those used for lenticule harvest. The center of cornea was marked and the lenticule inserted into the stromal pocket and aligned by gentle sweeping movements on the surface. All procedures were performed by an experienced surgeon (JSM) at the same laboratory. 

Lenticule Hydration Level (OCT) and Corneal Topography

Before PTK treatment of five lenticules, all lenticules were immersed in PBS overnight and placed on filter paper (Whatman; GE Healthcare, Buckinghamshire, UK) in a moist chamber for 5 hours to achieve an acceptable hydration level. The process was repeated for all lenticules before implantation. The lenticule thickness was assessed with OCT before and after PTK, by placing the lenticule in a petri dish. The maximum lenticule thickness of 107 μm, calculated by the VisuMax software on the previously mentioned laser settings, was used as a reference value for the hydration level. 

Topographic imaging was performed with the iTrace (Tracey Technologies, Houston, TX, USA) and the ATLAS (Carl Zeiss Meditec). The porcine eyes were positioned in an open holder and marked on the limbus with a surgical pen to ensure the same orientation before and after implantation. The average of three OCT measurements, one central and two at ±0.5 mm from either side of the central measurement, was used for analysis (ImageJ, http://imagej.nih.gov/ij/; provided in the public domain by the National Institutes of Health, Bethesda, MD, USA). 

Paired and unpaired Student's t-test, and Wilcoxon matched-pairs signed rank test were used for statistical comparison (GraphPad Prism 6; GraphPad, La Jolla, CA, USA and STATA version 13; STATACorp, College Station, TX, USA). A P value <0.05 was considered statistically significant. 

Results

Excimer Laser Ablation of Human Stromal Lenticules

TUNEL Assay

We observed TUNEL-positive cells in the superficial layers of the nonlasered lenticules, which was consistent with our previously published study.²¹ The density of TUNEL-positive cells was reduced in the lasered lenticules (Table 1; Supplementary Fig. S3). Total cell densities (DAPI+TUNEL) were lower after excimer laser ablation, which was most likely due to the vaporization of cells from the surface of the lenticules (Supplementary Fig. S3). 

Table 1

 

TUNEL and DAPI Quantification of Lasered Lenticules

 

Functional Group Analysis (ATR-FTIR)

Prominent changes could be seen for the lasered lenticules when compared with the nonlasered control lenticules (Fig. 1). The most prominent changes occurred in the lipid profile of the tissue, evidenced by changes in spectral band intensity and peak shift in the 3100 to 2800 cm⁻¹ region (Fig. 1; Table 2). The changes, as in spectral bands at 2960 cm⁻¹, 2976 cm⁻¹, and 2853 cm⁻¹, could be assigned to the asymmetric CH₃, symmetric CH₃, and asymmetric CH₂ stretching modes of lipids, respectively.²⁷ Changes could also be found in the spectral region of 1800 to 1400 cm⁻¹, but to a lesser extent, which corresponded to absorptions attributable to proteins and collagen. The primary marker of proteins was the amide I absorption bands at 1686 cm⁻¹, 1650 cm⁻¹, and 1630 cm⁻¹.²⁸ Bands corresponding to collagen, such as those at 1336 cm⁻¹, 1283 cm⁻¹, and 1204 cm⁻¹,²⁹ appeared with relatively lower intensity in laser-treated samples. 

Figure 1

 

ATR-FTIR showing functional groups of PTK-treated (n = 3) and control lenticules (n = 3). Second derivative spectra were generated from the group average spectra from the three examined lenticules by using the Savitsky-Golay algorithm with nine smoothing points to elucidate the overlapping bands in the spectra.

Table 2

 

Band Assignments Seen in Figure 2 for the Untreated Lenticules and Excimer-Lasered Lenticules

 

Surface Morphology Analysis (SEM and AFM)

The lasered lenticules appeared to have a smoother surface with reduced irregularities on the SEM images (Fig. 2). Incomplete tissue bridges, commonly observed on the corneal stromal bed after femtosecond laser-assisted extraction,^30,31 were seen on the surface of the untreated lenticules. The surface morphology evaluated with SEM was consistent with the AFM of the lasered and control lenticules, with a significantly higher R_z value in the control group than the lasered group (Table 3), indicating a more significant variation in surface irregularities. 

Figure 2

 

Surface morphology of the lasered and nontreated (control) lenticules. SEM (left column) showed more smoothening of the lenticule surface after PTK than of the untreated lenticules. Tissue bridges could be seen on the surface of nontreated lenticule. AFM results supported the SEM findings.

Table 3

 

Quantification of Surface Roughness Obtained From AFM Analysis (Mean ± SD)

 

Ex Vivo Lenticule Implantation Analysis

Hydration Level (OCT)

The average lenticule thicknesses are shown in Table 4. The average lenticule thickness just before PTK was 82 ± 11 μm, suggesting an underhydration. The laser-treated lenticules were thinner than the control lenticules, both after storage in PBS overnight (132 ± 26 μm vs. 233 ± 23 μm, P < 0.001) and after 5 hours of dehydration on filter paper (46 ± 3 μm vs. 57 ± 3 μm, P < 0.001; Supplementary Fig. S4). Postimplantation lenticule thickness was 67 ± 18 μm for the control lenticules, suggesting dehydration of the lenticules after implantation. When accounting for the difference in thickness of the lasered and control lenticules, there was gradual hydration of the porcine eyes during the study (Table 4). 

Table 4

 

CCT and Lenticule Thickness Before and After Implantation, Porcine Eyes (Mean ± SD)

 

Corneal Topography

For both lasered and control group, the postoperative ATLAS axial keratometry (P < 0.031) and elevation (P < 0.037) were significantly increased after implantation (Fig. 3; Table 5). However, no differences were seen in iTrace variables after implantation, with regard to the effective refractive power, Sim K, and Q values (P > 0.099; Table 5). In Table 6, the implantation of lasered lenticules seemed to cause less steepening than implantation of the untreated control lenticules, seen by the average change in axial keratometry (5.60 ± 3.79 D vs. 6.63 ± 2.17 D, P = 0.613), SimK (2.12 ± 4.16 D vs. 3.35 ± 5.12 D, P = 0.688), and Q values (−0.12 ± 0.31 vs. −1.05 ± 1.10, P = 0.108). 

Figure 3

 

ATLAS topography before (row A) and after (row B) implantation of a lasered and nonlasered control lenticule. Optical coherence tomography (row C) showed a significant difference in thickness of the lasered and nonlasered lenticules after implantation (P = 0.042).

Table 5

 

Keratometry Values Measured With ATLAS and iTrace Before Lenticule Implantation, After Pocket Creation, and After Lenticule Implantation (Mean ± SD)

 

Table 6

 

Average Change in ATLAS and iTrace Keratometry Readings (Mean ± SD)

 

Discussion

This study demonstrated the feasibility of excimer laser reshaping of SMILE-derived lenticules, using excimer ablation. We established the ablative effects of the excimer laser by showing compositional changes in the lipid and collagen structure with ATR-FTIR (Fig. 1; Table 2), lower cell densities with DAPI and TUNEL staining (Table 1; Supplementary Fig. S3), and changes in the surface topography with SEM and AFM (Fig. 2; Table 3). Physically, the excimer-lasered porcine lenticules were significantly thinner than their corresponding nonlasered controls (Table 4), and seemed to cause less curvature steepening after implantation (Table 6). 

Intrastromal implantation of SMILE-derived lenticules resembles the keratophakia procedure first described by Barraquer,¹¹ in which a biological lenticule is lathed from a frozen donor cornea and implanted into a manually dissected stromal pocket. The freezing process caused not only keratocyte necrosis but also stromal swelling that made it difficult to lathe a lenticule of a predetermined thickness. Femtosecond laser technology has now made it possible to create both the lenticule³² and the intrastromal pocket³³ with high accuracy. Furthermore, lenticule cryopreservation (with cryoprotectants)²¹ and lenticule excimer ablation caused only limited cell death and inflammation and may therefore reduce the risk of corneal scarring after implantation. 

Several techniques may be feasible to thin and reshape SMILE-derived lenticules. Cryosectioning of the lenticules is an option but may be impeded by uneven stromal thinning and limited possibilities for astigmatic correction. The use of femtosecond lasers is another alternative to create customized lenticules from donor corneas but is limited by the amount of donor tissue available worldwide. The use of discarded SMILE-derived lenticules may be more advantageous owing to increasing popularity of this procedure. Excimer reshaping could increase the use of locally cryopreserved lenticules in cases where there are only a limited amount of thin (low-powered) lenticules available. PTK has also previously been used in clinical studies, for the removal of corneal scarring following epikeratophakia³⁴ and to reduce residual error after epithelial basement membrane degeneration after LASIK.³⁵ 

Lenticule reimplantation may be performed just after extraction of the SMILE-derived lenticule. However, screening of transferable diseases is necessary if the lenticule is to be used as an allogenic implantation, and both the donor and recipient would need to be scheduled for surgery on the same day. A more optimized workflow would be the use of cryopreserved lenticules that could be stored until implantation. We recommend that nucleic acid amplification testing of the lenticule donor is performed by the lenticule cryopreserving eye bank as soon as the patient is scheduled for surgery, as the response time is typically 2 to 3 days. 

Intrastromal transplantation may have a lower risk of rejection than other commonly used transplantation techniques such as deep anterior lamellar keratoplasty and penetrating keratoplasty, as the implanted stromal tissue is unexposed to the tear film components and aqueous humor.³⁶ The intrastromal lenticule may be removed in cases with severe rejection, but it would be more efficient to reduce the risk of rejection by decellularization of the lenticule. Several protocols for decellularization have been suggested over time that include chemical, biological, and physical tissue processing to remove the cellular components.³⁷ In a previous study, we have made a comprehensive comparison of decellularization protocols and found that 0.1% sodium dodecylsulfate is adequate for stromal decellularization without compromising tissue transparency and extracellular matrix architecture.³⁸ 

Animal and patient case studies have shown that cryopreserved lenticules remain clear and viable and with minimal inflammation up to 6 months after implantation or reimplantation.^5,12,14,15 A study of reimplantation of cryopreserved lenticules in rabbits has found restoration of the central corneal thickness with a keratometry difference of −0.6 ± 0.8 D compared with the preoperative values.¹² A similar study using nonhuman primates has found a 0.6-D error in the keratometry readings 16 weeks after reimplantation.¹⁵ The keratometry change after reimplantation may be explained by the creation of a corneal flap in both studies,³⁹ although we also observed a minor change in the keratometry readings after pocket creation in this study. 

Equal hydration level of the porcine lenticules was important to assess the actual effect of the excimer ablation, owing to a strong dependency of hydration level on the corneal thickness.⁴⁰ The 107-μm reference lenticule thickness was calculated by the VisuMax software, which uses an algorithm based on the Munnerlyn formula.^41,42 Comparison of the lenticule thickness immediately after PTK may not be representative when the porcine lenticules have been under different humidity and storage conditions for a short time span.⁴³ Therefore, we stored both the lasered and nonlasered porcine lenticules in PBS overnight to ensure equal hydration before comparison. 

The lenticule storage in a moist chamber for 5 hours was based on previous pilot studies, examining the conditions and time span needed to ensure adequate hydration before implantation (unpublished data). Numerous storage conditions were tried, including 100% glycerol, which did not dehydrate the lenticules to a sufficient level and caused a hyperreflective surface on the OCT images. Lenticule storage directly on microscopy glass slides seemed to cause uneven dehydration with a crisp dry lenticule edge. The average lenticule thickness in the nonlasered group was 67 ± 18 μm, suggesting an underhydration after implantation (Table 4). However, we may expect that the dehydration level was similar in the two groups, which allows us to compare the lenticule thickness and postoperative corneal topography. We also found that the hydration level was important during laser thinning of the lenticules. Thus, we did not achieve any change in the lenticule thickness when PTK was performed on overhydrated lenticules (unpublished data). This may be explained by a large percentage of energy spent on water vaporization rather than ablation of the stromal components.⁴⁴ It may be worth considering that the amount of stromal thinning with PTK may differ for cryopreserved and fresh lenticules, as prior cryopreservation may affect the hydration level of lenticules. In this study, the fresh human stromal lenticules were stored in Optisol for no longer than 1 hour before TUNEL and DAPI assay. However, lenticule storage for longer time may cause morphologic changes, seen by keratocyte necrosis and shorter intercollagen distance.²² 

FTIR has recently been used to analyze surface biochemical components of biological tissues.^27,28 In this study, the most apparent shift was observed in spectral peaks in the 3100 to 2800 cm⁻¹ region, which could be attributed to changes in the lipid profile of cell membranes (Fig. 2; Table 2).²⁷ These peaks and peaks in the 1300 to 900 cm⁻¹ range (attributable to the presence of nucleic acids in stromal cells) exhibited lower intensity than the nonlasered lenticule, which we hypothesized was due to the reduction of cellular components in the ablation zone.⁴⁵ Our TUNEL assay confirmed the FTIR observation whereby there was a lower number of cells after PTK treatment than in the control group (Table 1; Supplementary Fig. S3). The cells lying in the periphery of the lenticule, which we have previously shown to undergo apoptosis following a femtosecond laser incision,²¹ had been “vaporized” by the excimer laser. We also found changes in terms of peak shift and peak intensity to the collagen and protein content, indicated by peaks in the 1800 to 1200 cm⁻¹ region. These changes were expected because excimer laser treatment has been known to induce alteration of stromal collagen structure and secretion of unique proteins, particularly corneal wound healing–associated proteins, for example, fibronectin, tenascin, and CD90.^30,46,47 

In this study, the SEM and AFM showed a smoother surface of the lenticules after PTK treatment than for control lenticules. The results are consistent with a previous study of PTK in rabbits, where a 37.5-μm ablation depth is used.⁴⁸ The SEM reveals a surface structure of undulating collagen fiber bundles, similar to what was seen after PTK of the human lenticules in this study (Fig. 2). Another study of porcine eyes has also revealed a smoother and regular surface after PTK, using a masking agent.⁴⁹ Following PTK, there were fewer tissue bridges located on the corneal surface, as remnants from the cavitation bubble separation during the lenticule extraction. These are readily seen on the stromal bed and the lenticule surface after SMILE, even after a smooth dissection and extraction.^30,50 Moreover, thicker corneas and thinner lenticules have been associated with a higher risk of opaque bubble layer development during SMILE.⁵¹ The lenticule pairs in this study were matched by their spherical equivalent power, with approximately the same calculated maximum lenticule thickness, whereas the central corneal thickness of the SMILE-recipient was not taken into consideration. We performed PTK to reduce the lenticule thickness without changing the plano-convex shape of the myopic lenticule, as a myopic ablation profile would have caused an uneven thinning of the lenticules. However, the lenticule shape may be altered by using a myopic or hyperopic ablation profile (photorefractive keratectomy) or topography-guided shaping with or without a masking agent.⁴⁹ 

Implantation of PTK-treated lenticules into porcine eyes seemed to cause less curvature steepening than implantation of nonlasered lenticules (Fig. 3; Table 6). Our chosen model had some limitations, as we were not able to evaluate changes in the hydration level of the lenticule over time. We have previously shown in nonhuman primates that implanted lenticules gradually thin out after implantation,¹⁴ and it is therefore necessary to further evaluate the long-term stability after implantation in order to predict the refractive outcome.⁵² However, the lenticule hydration level just before implantation may not be an important factor when lenticule implantation is performed in patients or animal models, as the hydration will stabilize owing to normal endothelial regulation.⁵³ Another limitation was the assessment of the hydration level by the lenticule thickness. We used the maximum lenticule thickness calculated by the VisuMax software as a reference value, although this value is dependent on the femtosecond laser precision, including the achieved minimum lenticule thickness and diameter. 

In conclusion, this study demonstrated the possibility to reshape SMILE-derived lenticules for use in volume restoration or tissue additive surgery. However, following a controlled hydration protocol to achieve an adequate level of lenticule hydration is necessary before stromal thinning with excimer laser. Our study suggests that overhydration of the lenticule in PBS followed by storage in a moist chamber for approximately 5 hours is sufficient before laser ablation of the lenticule. Future animal studies may consider optimization of the implantation depth as well as topography-guided shaping of the lenticules before implantation. 

Acknowledgments

Supported by the Singapore National Research Foundation under its Translational and Clinical Research (TCR) Programme (NMRC/TCR/1021-SERI/2013) and administered by the Singapore Ministry of Health's National Medical Research Council. Supported by The Synoptik Foundation and Fight for Sight Denmark. 

Disclosure: I.B. Damgaard, None; A.K. Riau, None; Y.-C. Liu, None; M.L. Tey, None; G.H.-F. Yam, None; J.S. Mehta, Ziemer (C), Carl Zeiss Meditec, Inc. (C)