Early Corneal Nerve Damage and Recovery Following Small Incision Lenticule Extraction (SMILE) and Laser In Situ Keratomileusis (LASIK)

Abstract

Purpose.: We compared early corneal nerve changes after small incision lenticule extraction (SMILE) and laser in situ keratomileusis (LASIK).

Methods.: A total of 12 rabbits underwent LASIK in one eye and SMILE in the fellow eye. Baseline and follow-up evaluations at 1, 2, and 4 weeks postoperatively were performed with in vivo confocal microscopy to evaluate 5 different areas within the treated zone: center, superior, inferior, nasal, and temporal. Cryosections of the corneas and whole mount of the extracted SMILE lenticules were analyzed with immunostaining of βIII-tubulin.

Results.: One week after SMILE and LASIK, a decrease in nerve length and density was observed in all evaluated areas. A trend toward greater subbasal nerve length and density (SLD), more eyes with subbasal nerves (ESN), more eyes with subbasal nerves longer than 200 μm (SNL), and higher mean number of subbasal nerves by frame (NSN) in SMILE than in LASIK groups was observed at subsequent follow-up time points. Only the SMILE group showed a recovery of SLD, ESN, and NSN by week 4 (P > 0.05). A trend toward more eyes with sprouting subbasal nerves and greater mean number of sprouting nerves was observed in LASIK than in SMILE, indicating that more subbasal nerves were disrupted and undergoing regeneration after LASIK. Immunostaining at postoperative week 4 revealed a faster stromal nerve recovery in post-SMILE eyes compared to post-LASIK eyes.

Conclusions.: Our findings suggest that SMILE results in less nerve damage and faster nerve recovery than LASIK.

Introduction

Laser in situ keratomileusis (LASIK) involves the creation of a corneal flap with either a femtosecond laser or a mechanical microkeratome, followed by ablation of the stroma with an excimer laser. ¹ Refractive lenticule extraction (ReLEx) is a new corneal refractive procedure that does not require stromal ablation with the excimer laser. ² In ReLEx, the refractive error is corrected by means of creating an intrastromal refractive lenticule using only a femtosecond laser. ² The original ReLEx procedure, femtosecond lenticule extraction (FLEx), mimics a LASIK procedure with the formation of an anterior hinged flap. The lenticule is peeled away after the flap is lifted. In the refined version of ReLEx, small incision lenticule extraction (SMILE), the creation of a flap is obviated. Instead, the lenticule is dissected and subsequently extracted from a small arcuate incision (2.5–3 mm), usually placed superiorly. ^2,3  

The cornea is one of the most richly innervated tissues in the body, composed of sensory and autonomic nerve fibers. ⁴ These nerves have specific physiologic roles, including those of regulating epithelial integrity, proliferation, and wound healing. ^5,6 Corneal nerves are derived from the nasociliary branch of the ophthalmic division of the trigeminal nerve. Nerve fiber bundles enter the cornea at the limbus in a radial manner at the level of middle and anterior stroma, and then travel parallel to the cornea surface to the center of the cornea. As these nerve bundles travel, they undergo division into smaller branches to innervate the anterior and mid stroma, and eventually turn 90° toward the surface, perforating Bowman's layer. ⁷ They then turn again 90° to travel between Bowman's layer and the basal epithelial cells forming the subbasal nerve plexus. From there the individual subbasal nerves give branches to innervate the more superficial epithelial cells. ^8–11  

One of the most common complications observed in LASIK is dry eye, which is secondary to the surgical-induced neurotrophic cornea. ¹² The corneal nerves are cut during the flap creation and stromal ablation. The magnitude of corneal nerve damage is related directly to the flap/ablation diameters, and the depth and degree of laser correction. ¹³ Hence, SMILE would be expected to improve postoperative corneal sensitivity and induce less of a neurotrophic cornea. If the majority of the stromal nerves are severed, the subbasal and epithelial nerves will undergo degeneration. Therefore, evaluation of damage in the subbasal nerve plexus can be an ideal way to assess the general nerve damage of the cornea. Previous studies have shown that corneal nerves are degenerated during LASIK and regrowth of the subbasal nerves is not complete even several years after surgery, although nerves in the hinge area are preserved. ^14,15  

This study was performed in New Zealand white rabbits to analyze the subbasal nerve changes after SMILE and LASIK surgeries. We used in vivo confocal microscopy (IVCM) to visualize the subbasal nerves in 4 quadrants of the cornea following laser correction (superior, inferior, nasal, and temporal) and also at the center of the treatment area. Finally, immunofluorescent staining of βIII-tubulin was performed to analyze the postoperative nerve damage and regrowth. 

Methods

Experimental Design

A total of 12 New Zealand white rabbits was used in this study. One eye of each rabbit, selected at random, had SMILE (3 right eye and 9 left eyes) and the fellow eye had LASIK (9 right eyes and 3 left eyes) surgery. Both groups (SMILE and LASIK) underwent a myopic treatment correction of −6.00 diopters (D). In vivo evaluations at baseline, and at 1, 2, and 4 weeks after surgery were performed with IVCM. Three rabbits were sacrificed at postoperative weeks 2 and 4. Excised corneoscleral rims and extracted SMILE lenticules were analyzed with immunofluorescent staining (anti βIII-tubulin antibody) for corneal nerves. 

Animals

The New Zealand White rabbits were obtained from National University of Singapore. They were 12 to 15 weeks old and with a body weight of 3 to 4 kg. All surgeries and in vivo evaluations were performed under anesthesia with xylazine hydrochloride (5 mg/kg intramuscularly; Troy Laboratories, Smithfield, Australia) and ketamine hydrochloride (50 mg/kg intramuscularly; Parnell Laboratories, Alexandria, Australia). Rabbits were euthanized under anesthesia at different time points after surgery (2 and 4 weeks) by overdose intracardiac injection of sodium pentobarbital. All animals were treated according to the guidelines of the Association for Research in Vision and Ophthalmology (ARVO) Statement for the Use of Animals in Ophthalmic and Vision Research. The protocol was approved by the Institutional Animal Care and Use Committee of SingHealth. 

Surgical Procedures

Femtosecond Laser-Assisted LASIK Procedure.

Our rabbit experimental model for LASIK was used as described previously. ¹⁶ All procedures were performed by one surgeon (JSM). LASIK flaps were created by using a 500-kHz femtosecond laser (VisuMax; Carl Zeiss Meditec, Jena, Germany). The laser parameters were as follows: 110 μm flap thickness, 7.9 mm flap diameter, 170 nJ power, spot distance, and tracking spacing of 4.8 μm for lamellar flap and 2 μm for flap side cuts, respectively, flap side cut at 90°, hinge position at 90°, hinge angle of 50°, and spiral in (centripetal) scanning pattern direction. After the flap was lifted, the underlying stroma underwent a 6.5 mm optical zone myopic ablation of −6.00 D using an excimer laser (Technolas; Bausch & Lomb, Rochester, NY) with the following laser parameters: spot size 2.0 mm diameter, fluence 120 mJ/cm2, and repetition rate 50 Hz. The flap then was repositioned and a bandage contact lens (Bausch & Lomb) applied, and the eyelid was closed with a temporary tarsorrhaphy using a 6-0 Vicryl suture. 

SMILE Procedure.

Our rabbit experimental model for SMILE was performed as described previously. ¹⁷ All the procedures were performed by one surgeon (JSM). A myopic correction of −6.00 D was performed with a 500-kHz femtosecond laser (Visumax; Carl Zeiss Meditec). The laser was centered visually on the pupil. The eye was docked under the small (S) curved interface cone and application of suction was applied. The femtosecond incisions were performed in a spiral in/out scanning pattern direction. ¹⁸ That is, first the posterior surface of the lenticule was cut (spiral in, centripetal direction), second the anterior surface of the lenticule and a cap extension were cut (spiral out, centrifugal), followed by the lenticule side cut and a small arcuate incision. The femtosecond laser parameters were: 200 nJ power, 110 μm cap thickness, 7.5 mm cap diameter, 6.5 mm lenticule diameter, and minimum and maximum lenticule thicknesses of 15 and 105 μm, respectively. The spot distance and tracking spacing were set at 3 μm for the cap and lenticule, and at 2 μm for the side cuts. Side cut angles were at 90°, incision position 120°, and incision width 3 mm. After completion of the laser firing, the cornea incision was opened with a Sinskey hook (Rhein Medical, Inc., Petersburg, FL). The adhesions of anterior and posterior surfaces of the lenticule were released with a femto-lamellar dissector (Asico, Westmont, IL). The refractive lenticule then was grasped and extracted from the cornea with a Tan DSAEK forceps (Asico). Finally, a 24-gauge cannula was used to flush the pocket insertion with balanced salt solution. 

In Vivo Confocal Microscopy

The IVCM was performed with a confocal scanning microscope, the Heidelberg HRT3 Rostock Cornea Module (Heidelberg Engineering GmbH, Heidelberg, Germany). Each examination was performed as follow: Rabbits were anesthetized and placed over a platform for evaluation. A drop of topical anesthetic 0.5% proparacaine hydrochloride ophthalmic solution (Alcaine; Alcon Laboratories, Inc., Fort Worth, TX) was instilled on the eye. A carbomer gel (Vidisic; Mann Pharma, Berlin, Germany) was applied onto and over the objective tip as an immersion fluid. Image acquisition time was 30 frames/s (0.024 s/frame), images consisted of 384 × 384 pixels covering a field of view of 400 × 400 μm. 

All corneas were examined in 5 different areas within the treated zone (central, superior, inferior, nasal, and temporal). At least 3 z-axis full thickness scans, from epithelium to endothelium, were taken in each area. To ensure the correct position of the periphery scans, the objective lens first was moved manually in the x/y axis to localize the peripheral edge of the laser cutting interface and then moved centrally to be within the treated zone. The treated zone of LASIK refers to the excimer laser ablated optical zone, which was 6.5 mm in diameter. The treated zone of SMILE corresponds to the size of the lenticule, also 6.5 mm in diameter. Every scan sequence was verified further to be within the treated zone by confirming the presence of interface in each sequence. The interface could be identified as a highly reflective layer in the anterior stroma. ¹⁹  

The best-focused and most representative images of subbasal nerves were selected. Up to 3 frames (images) were selected from each area of the optical zone for every eye. Each nerve was counted only once; branched nerves were counted as one nerve. The selected frames from each eye and area were analyzed, averaged, and accepted as the value for that examination session. Quantitative and qualitative analysis of the images were performed using ImageJ software (Developed by Wayne Rasband, National Institutes of Health, Bethesda, MD; available in the public domain at http://rsb.info.nih.gov/ij/index.html). Nerve length was calculated using NeuronJ, a semi-automated nerve tracing image analysis software that is a plug-in program to ImageJ. ^10,20 All images were evaluated by one observer (KMN). 

Subbasal Nerve Plexus Analysis

Analysis of subbasal nerves was performed in the 5 areas of the optical zone (central, superior, inferior, nasal, and temporal) before and after surgery in SMILE and LASIK groups as described previously. ^7,10 The presence, number, density by frame, density by cornea area, length, diameter, tortuosity grade, presence and number of beadings, and presence and number of branchings, and the presence and number of sprouting nerves in the subbasal plexus area also were recorded. Finally, each cornea area was classified according to the observed nerve morphology as follow: number (%) of corneas with (0) no nerves, (1) sprouting nerves, (2) only subbasal nerves shorter than 200 μm, and (3) subbasal nerves longer than 200 μm (adapted from the studies of Linna et al. ²¹ and Bragheeth et al.²²). All measurements were performed at a zoom of 100%, except nerve diameter, which used a 200% zoom. ⁵  

To analyze the number and presence of subbasal and sprouting nerves, total subbasal nerve density and nerve morphology, all eyes were included, and 3 frames per cornea area were analyzed and averaged. Cornea areas that showed no nerves were included with a 0 value. For subbasal nerve density by frame, cornea areas with no nerves were excluded, 3 frames per cornea area were analyzed and averaged. For total nerve length, diameter, tortuosity, branching, and beading, cornea areas with no nerves were excluded and only frames with nerves were included (up to 3 frames with nerves were selected, analyzed and averaged per cornea area). 

Definition of each nerve analysis: 

1.  
   Presence of eyes with subbasal nerves: Percentage of eyes that show subbasal nerves.
2.  
   Number of subbasal nerves: The sum of all the subbasal nerves observed within a frame. ^5,9
3.  
   Total nerve length: The total length of all visible subbasal nerves in a frame.
4.  
   Subbasal nerve density by frame (mm/mm²): The total length of all visible subbasal nerves in a frame divided by the frame area (400 × 400 μm). ^{5,6,9,10,14,23,24} Corneal areas with no visible nerves were excluded. Three frames from each cornea area were selected, analyzed, and averaged per cornea area. ¹⁴ If less than 3 frames for cornea area with different nerves were available, the remaining empty frames were considered as 0 value.
5.  
   Total subbasal nerve density (mm/mm²): The total length of all visible subbasal nerves in a frame divided by the frame area (400 × 400 μm). ^{5,6,9,10,14,23,24} In cornea areas with no visible nerves, the density was counted as 0. Three frames from each cornea area were selected, analyzed, and averaged per cornea area. ¹⁴ If less than 3 frames per cornea area with different nerves were available, the remaining empty frames were considered as 0 value.
6.  
   Nerve diameter (μm): The average of three random measurements on every nerve. ^5,9,25
7.  
   Nerve tortuosity was calculated using the Oliveira-Soto and Efron score. ^5,10 Grade 0: The nerve fibers appear almost straight. Grade 1: The nerve fibers are slightly tortuous. Grade 2: The nerve fibers appear moderately tortuous; there are frequent changes in the direction of the fiber, although these are of small amplitude. Grade 3: The nerve fibers are quite tortuous; the amplitude of the changes in the fiber direction was quite severe. Grade 4: The nerve fibers appear very tortuous, showing abrupt and frequent changes in the nerve fiber direction. ⁵ Tortuosity was expressed as the median tortuosity grade, except when only one eye showed nerves; therefore, the only value was used.
8.  
   Presence of eyes with nerve branching: Percentage of eyes that showed at least one nerve branching in the subbasal plexus area.
9.  
   Branching grade was calculated using, the Midena et al. score. ¹⁰ Grade 0: No branching. Grade 1: One or more fibers presenting 1 direct branch from the principal trunk. Grade 2: One or more fibers with 1 branch originating from a grade 1 fiber. Grade 3: One or more fibers with 1 branch originating from a grade 2 fiber. ¹⁰ It was expressed as the median branching grade.
10.  
    Presence of eyes with nerves with beading formations: Percentage of eyes that have at least one nerve with a beading formation in the subbasal plexus area.
11.  
    Number of beading: It was defined as the number of beading formations present in a random 100 μm of nerve fiber. ^5,9,10 Beadings are identified as hyperreflective and well defined areas that protrude slightly from both sides of the nerve fibers. ^5,10
12.  
    Presence of eyes with sprouting nerves: Percentage of eyes that show sprouting nerves in the subbasal plexus area.
13.  
    Number of sprouting nerves: The number of sprouting nerves by frame. ²⁶

Tissue Fixation and Sectioning

The corneas of 3 rabbits were excised 2 and 4 weeks after surgery, and embedded in an optimal cutting temperature (OCT) cryo-compound media (Leica Microsystems, Nussloch, Germany) for immunofluorescent staining. Frozen tissue blocks were stored at −80°C until sectioning. Serial sagittal 30 μm cryosections were cut in a superior-inferior orientation using a Microm HM550 cryostat (Microm, Walldorf, Germany). Sections also were made in a modified technique that oriented the corneal tissue obliquely, increasing the area and allowing for better visualization of corneal nerves. Only sections within ±2 mm of the center of the corneas were collected for immunofluorescent staining, allowing better visualization, and inclusion of longer and continuous nerves. All sections were placed on polylysine-coated glass slides and air dried for 15 minutes. 

Immunofluorescent Staining of Cryosections

Cryosections were fixed with 4% paraformaldehyde (Sigma-Aldrich Corp., St. Louis, MO) for 15 minutes, washed three times for 5 minutes each with ×1 PBS, and incubated with mouse monoclonal antibody against βIII-tubulin (Covance, Princeton, NJ) at 4°C for 24 hours to stain the corneal nerves. The antibody was diluted in 4% bovine serum albumin (Sigma), ×1 PBS, and 0.15% Triton X-100 (Sigma-Aldrich Corp.). After three thorough washings for 5 minutes each with ×1 PBS, the sections were incubated with goat anti-mouse Alexa Fluor 488-conjugated secondary antibody (Invitrogen, Carlsbad, CA) at room temperature for 2 hours. Slides then were mounted with UltraCruz Mounting Medium containing 4′,6-diamidino-2-pheylindole (DAPI; Santa Cruz Biotechnology, Santa Cruz, CA). Sections were observed and imaged with a Zeiss AxioImager Z1 fluorescence microscope (Carl Zeiss, Oberkochen, Germany). To reconstruct the sagittal image of the whole cornea, consecutive images from the superior to the inferior of the cornea were taken at a magnification of ×50. The Panorama module in the Axiovision software (Carl Zeiss) was used to stitch the consecutive images. Sagittal images obtained from two consecutive sections then were aligned and merged using Adobe Photoshop CS3 software (Adobe Systems, San Jose, CA). 

Whole Mount Immunofluorescent Staining of Lenticules

The extracted lenticules from the eyes that underwent SMILE were collected and then fixed with 4% paraformaldehyde (Sigma-Aldrich Corp.) at 4°C overnight. The following day, the corneas were washed with ×1 PBS for 3 times 10 minutes each, blocked with 4% bovine serum albumin (Sigma-Aldrich Corp.) in ×1 PBS, 0.15% Triton X-100 (Sigma-Aldrich Corp.) for 1 hour, and incubated with mouse monoclonal antibody against βIII-tubulin (Covance) at 4°C for 24 hours. After washing with ×1 PBS for 3 times 10 minutes each, the corneas were incubated with goat anti-mouse Alexa Fluor 488-conjugated secondary antibody (Invitrogen) at room temperature for 2 hours. Following washing with ×1 PBS, the whole mount was viewed using a Zeiss AxioImager Z1 fluorescence microscope (Carl Zeiss). To acquire the whole image of the stromal nerve structure within the lenticules, consecutive images from the center to the periphery of the lenticules were taken at a magnification of ×50. The Panorama module in the Axiovision software (Carl Zeiss) was used to stitch the consecutive images. 

Statistical Analysis

Independent nonnormally distributed samples (number of subbasal nerves, number of sprouting nerves, nerve length, nerve density, nerve diameter, nerve tortuosity grade, nerve branching grade, number of nerve beadings) were analyzed with a nonparametric test, the Mann Whitney U 2-tailed test. Dependent nonnormally distributed categorical data: The morphologic analysis of the subbasal nerve plexus based on nerve size and type (number of eyes with no nerves, with sprouting nerves, with subbasal nerves, with only subbasal nerves shorter than 200 μm, and with subbasal nerves longer than 200 μm), number of eyes with branching, and number of eyes with beading were analyzed with a nonparametric test, the Fisher's exact 2-tailed test. Statistics were performed in Prism 6 for Mac OS X, v.6.0 2012 (GraphPad Software, Inc., La Jolla, CA) software program. Differences were considered statistically significant at P < 0.05. 

Results

Length and Density of Corneal Subbasal Nerves

A detailed description in the 5 analyzed areas, with comparison and P values of subbasal nerve length and diameter before and after surgery in the SMILE and LASIK groups, is shown in Table 1. Both groups showed similar nerve length and density (P > 0.05) in preoperative evaluation. The SMILE group showed greater subbasal nerve length and density than the LASIK group in postoperative evaluations at weeks 1, 2, and 4 in all 5 areas; however, the differences were not significant (P > 0.05), except the total nerve density in the inferior (P = 0.009) and central (P = 0.011) areas at postoperative week 2 (Table 1). 

Table 1

 

Nerve Length and Density of Corneal Subbasal Nerves

	Baseline		P Value	Week 1		P Value	Week 2		P Value	Week 4		P Value	P Value Baseline vs. Week 4		P Value Week 1 vs. Week 4
	SMILE, n = 6	LASIK, n = 6		SMILE, n = 8	LASIK, n = 8		SMILE, n = 9	LASIK, n = 9		SMILE, n = 6	LASIK, n = 6		SMILE	LASIK	SMILE	LASIK
Central area
Total nerve length by frame,  μm	658 (365)	393 (146)	0.238	176 (86)	106 (–)	–	252 (216)	84 (48)	0.143	487 (350)	277 (–)	–	0.476	–	0.400	–
Total density, mm/mm2	2.5 (1.3)	2.0 (1.5)	0.571	0.2 (0.4)	0.03 (0.1)	0.323	1.1 (1.0)	0.1 (0.1)	0.011*	1.3 (1.8)	0.2 (0.5)	0.303	0.307	0.006*	0.3520	0.736
Density by frame, mm/mm2	2.5 (1.3)	2.0 (1.5)	0.571	0.6 (0.5)	0.2 (–)	–	1.2 (1.0)	0.2 (0.1)	0.105	2.6 (1.7)	1.3 (–)	–	0.714	–	0.200	–
Eyes with subbasal nerves,  n (%)	6 (100%)	6 (100%)	1.000	3 (38%)	1 (13%)	0.569	8 (89%)	4 (44%)	0.131	3 (50%)	1 (17%)	0.545	0.181	0.015*	1.000	1.000
Superior quadrant
Total nerve length by frame,  μm	305 (84)	313 (82)	0.513	237 (243)	286 (209)	0.800	234 (134)	225 (179)	1.000	630 (554)	189 (–)	–	0.352	–	0.200	–
Total density, mm/mm2	1.78 (0.79)	1.74 (0.71)	0.898	0.3 (0.5)	0.2 (0.3)	0.478	0.4 (0.6)	0.5 (0.7)	0.927	1.1 (1.3)	0.1 (0.2)	0.106	0.290	0.002*	0.200	0.846
Density by frame, mm/mm2	1.78 (0.79)	1.74 (0.71)	0.898	0.5 (0.5)	0.6 (0.5)	0.800	0.8 (0.6)	1.0 (0.8)	0.556	1.6 (1.2)	0.4 (–)	–	0.826	–	0.114	–
Eyes with subbasal nerves,  n (%)	6 (100%)	6 (100%)	1.000	4 (50%)	2 (25%)	0.608	5 (56%)	4 (44%)	1.000	4 (67%)	1 (17%)	0.242	0.454	0.015*	0.627	1.000
Inferior quadrant
Total nerve length by frame,  μm	316 (93)	313 (88)	0.898	166 (69)	103 (–)	–	135 (100)	–	–	324 (239)	–	–	0.695	–	0.228	–
Total density, mm/mm2	1.72 (0.81)	1.69 (0.66)	0.787	0.3 (0.6)	0.03 (0.08)	0.200	0.4 (0.7)	0.0 (0.0)	0.009*	0.6 (0.7)	0.0 (0.0)	0.061	0.062	0.002*	0.249	>0.999
Density by frame, mm/mm2	1.72 (0.81)	1.69 (0.66)	0.787	0.7 (0.8)	0.2 (–)	–	0.6 (0.7)	–	–	0.9 (0.7)	–	–	0.247	–	0.571	–
Eyes with subbasal nerves,  n (%)	6 (100%)	6 (100%)	1.000	3 (38%)	1 (13%)	0.569	6 (67%)	0 (0%)	0.009*	4 (67%)	0 (0%)	0.060	0.454	0.002*	0.592	1.000
Nasal quadrant
Total nerve length by frame,  μm	331 (202)	318 (188)	0.675	82 (33)	125 (–)	–	203 (46)	136 (12)	0.095	845 (422)	105 (–)	–	0.095	–	0.200	–
Total density, mm/mm2	1.54 (0.71)	1.50 (0.68)	0.788	0.05 (0.1)	0.04 (0.1)	1.000	0.6 (0.7)	0.1 (0.2)	0.076	1.6 (1.8)	0.1 (0.2)	0.181	0.939	0.002*	0.202	0.736
Density by frame, mm/mm2	1.54 (0.71)	1.50 (0.68)	0.788	0.2 (0.1)	0.3 (–)	–	1.0 (0.5)	0.5 (0.3)	0.191	3.3 (0.7)	0.5 (–)	–	0.050	–	0.200	–
Eyes with subbasal nerves,  n (%)	6 (100%)	6 (100%)	1.000	2 (25%)	1 (13%)	1.000	5 (56%)	2 (22%)	0.335	3 (50%)	1 (17%)	0.545	0.181	0.015*	0.580	1.000
Temporal quadrant
Total nerve length by frame,  μm	260 (102)	248 (97)	0.788	156 (52)	–	–	213 (122)	173 (–)	–	413 (282)	–	–	0.476	–	0.400	–
Total density, mm/mm2	1.35 (0.78)	1.26 (0.38)	0.576	0.1 (0.2)	0.00 (0.00)	0.200	0.5 (0.7)	0.04 (0.13)	0.135	1.1 (1.4)	0.00 (0.00)	0.061	0.455	0.002*	0.195	>0.999
Density by frame, mm/mm2	1.35 (0.78)	1.26 (0.38)	0.675	0.4 (0.1)	–	–	1.1 (0.7)	0.4 (–)	–	1.7 (1.4)	–	–	0.895	–	0.400	–
Eyes with subbasal nerves,  n (%)	6 (100%)	6 (100%)	1.000	3 (38%)	0 (0%)	0.200	4 (44%)	1 (11%)	0.294	4 (67%)	0 (0%)	0.060	0.454	0.002*	0.592	1.000

 Total nerve length, density by frame, and total density are expressed as mean (SD).

*  Indicate statistically significant (P < 0.05).

The SMILE and LASIK groups showed an initial decrease in nerve length and density after surgery. At postoperative week 4, only the SMILE group showed signs of nerve recovery identified by an increase in nerve length and nerve density with no significant differences (P > 0.05) between preoperative values and postoperative week 4 in all 5 areas of the treatment zone. In contrast, the LASIK group at postoperative week 4 maintained a significantly reduced nerve density (P < 0.05) as compared to preoperative values in all 5 areas of the treated zone. We were not able to perform meaningful statistical analysis of the nerve length between preoperative and post-LASIK week 4 values due to the absence of nerves in most viewing frames. We observed a trend in improvement in nerve length and density from postoperative week 1 to postoperative week 4 that was greater in the SMILE than in the LASIK group (Table 1, representative confocal images were taken from the central area of the treated zone and are shown in Fig. 1). 

Figure 1

 

Subbasal nerves before and after SMILE and LASIK. Representative in vivo confocal microscopy images of observed subbasal nerves in the subbasal nerve plexus before and after surgery in SMILE and LASIK groups in central area of the laser treated zone. (A–D) SMILE group. (E–H) LASIK group. (A, E) Preoperative image of subbasal nerves before surgery. (B, F) Postoperative images 1 week after SMILE and LASIK, respectively. (C, G) Postoperative images 2 weeks after SMILE and LASIK, respectively. (D, H) Postoperative images 4 weeks after SMILE and LASIK, respectively.

Subbasal and Sprouting Nerve Morphology

A detailed analysis of subbasal nerve morphology is displayed in Table 2. Preoperative evaluation revealed that both groups presented similar subbasal nerve morphology (P > 0.05). More eyes with no subbasal nerves or with sprouting nerves were observed in the LASIK than in the SMILE group in the designated 5 analyzed areas at postoperative weeks 1, 2, and 4. At post-SMILE and LASIK week 4, we were able to observe nerve recovery, evident by the tendency toward a decrease in the number of eyes observed with no nerves. However, this change was not significant between preoperative and postoperative week 4 values, and between postoperative weeks 1 and 4 values in the SMILE and LASIK groups (P > 0.05), except in the treated area inferiorly of post-LASIK eyes (P = 0.025). 

Table 2

 

Presence and Number of Corneal Subbasal and Sprouting Nerves

	Baseline		P Value	Week 1		P Value	Week 2		P Value	Week 4		P Value	P Value Baseline vs. Week 4		P Value Week 1 vs. Week 4
	SMILE, n = 6	LASIK, n = 6		SMILE, n = 8	LASIK, n = 8		SMILE, n = 9	LASIK, n = 9		SMILE, n = 6	LASIK, n = 6		SMILE	LASIK	SMILE	LASIK
Central area
Eyes with no nerves, n (%)	0 (0%)	0 (0%)	1.000	5 (63%)	6 (75%)	1.000	1 (11%)	3 (33%)	0.576	3 (50%)	2 (33%)	1.000	0.181	0.454	1.000	2.774
Eyes with sprouting nerves, n (%)	0 (0%)	0 (0%)	1.000	0 (0%)	1 (13%)	1.000	0 (0%)	3 (33%)	0.206	1 (17%)	3 (50%)	0.545	1.000	0.181	0.428	0.224
Eyes with subbasal nerves, n (%)	6 (100%)	6 (100%)	1.000	3 (38%)	1 (13%)	0.569	8 (89%)	4 (44%)	0.131	3 (50%)	1 (17%)	0.545	0.181	0.015	1.000	1.000
Eyes with only subbasal nerves shorter than 200 μm, n (%)	0 (0%)	0 (0%)	1.000	2 (25%)	1 (13%)	1.000	3 (33%)	4 (44%)	1.000	1 (17%)	1 (17%)	1.000	1.000	1.000	1.000	1.000
Eyes with subbasal nerves longer than 200 μm, n (%)	6 (100%)	6 (100%)	1.000	1 (13%)	0 (0%)	1.000	5 (56%)	0 (0%)	0.029*	2 (34%)	1 (17%)	1.000	0.060	0.015*	0.538	0.428
Mean number of Sprouting nerves (SD)	0 (–)	0 (–)	>0.999	0 (–)	0.04 (0.12)	>0.999	0 (–)	1.44 (3.62)	0.206	0.11 (0.27)	0.44 (0.66)	0.424	>0.999	0.181	0.428	0.164
Mean number of subbasal nerves (SD)	1.17 (0.51)	1.50 (0.98)	0.787	0.17 (0.25)	0.04 (0.12)	0.446	0.93 (0.55)	0.19 (0.24)	0.007*	0.67 (0.82)	0.17 (0.41)	0.303	0.277	0.008*	0.328	0.736
Superior quadrant
Eyes with no nerves, n (%)	0 (0%)	0 (0%)	1.000	3 (38%)	6 (75%)	0.314	4 (44%)	3 (33%)	1.000	1 (17%)	2 (33%)	1.000	1.000	0.454	0.580	0.277
Eyes with sprouting nerves, n (%)	0 (0%)	0 (0%)	1.000	2 (25%)	0 (0%)	0.466	1 (11%)	2 (22%)	1.000	1 (17%)	3 (50%)	0.545	1.000	0.181	1.000	0.055
Eyes with subbasal nerves, n (%)	6 (100%)	6 (100%)	1.000	4 (50%)	2 (25%)	0.608	5 (56%)	4 (44%)	1.000	4 (67%)	1 (17%)	0.242	0.454	0.015*	0.627	1.000
Eyes with only subbasal nerves shorter than 200 μm (n (%)	0 (0%)	0 (0%)	1.000	3 (38%)	1 (13%)	0.569	3 (33%)	3 (33%)	1.000	1 (16%)	1 (16%)	1.000	1.000	1.000	0.580	1.000
Eyes with subbasal nerves longer than 200 μm (n (%)	6 (100%)	6 (100%)	1.000	1 (13%)	1 (13%)	1.000	3 (33%)	2 (22%)	1.000	3 (50%)	0 (0%)	0.181	0.181	0.002*	0.245	1.000
Mean number of sprouting nerves (SD)	0 (–)	0 (–)	>0.999	0.13 (0.25)	0.00 (0.0)	0.467	0.04 (0.11)	0.19 (0.38)	0.471	0.11 (0.27)	0.28 (0.33)	0.545	>0.999	0.181	>0.999	0.055
Mean number of subbasal nerves (SD)	1.22 (0.46)	1.33 (0.30)	0.766	0.21 (0.25)	0.08 (0.15)	0.445	0.30 (0.42)	0.37 (0.45)	0.917	0.50 (0.51)	0.06 (0.14)	0.106	0.045*	0.002*	0.307	>0.999
Inferior quadrant
Eyes with no nerves, n (%)	0 (0%)	0 (0%)	1.000	5 (63%)	7 (88%)	0.569	3 (33%)	8 (89%)	0.050	2 (33%)	1 (17%)	1.000	0.454	1.000	0.592	0.025*
Eyes with sprouting nerves, n (%)	0 (0%)	0 (0%)	1.000	0 (0%)	1 (13%)	1.000	1 (11%)	1 (11%)	1.000	1 (17%)	5 (83%)	0.080	1.000	0.015*	0.428	0.025*
Eyes with subbasal nerves, n (%)	6 (100%)	6 (100%)	1.000	3 (38%)	1 (13%)	0.569	6 (67%)	0 (0%)	0.009*	4 (67%)	0 (0%)	0.060	0.454	0.002*	0.592	1.000
Eyes with only subbasal nerves shorter than 200 μm, n (%)	0 (0%)	0 (0%)	1.000	3 (38%)	1 (13%)	0.569	6 (67%)	0 (0%)	0.009*	2 (33%)	0 (0%)	0.454	0.454	1.000	1.000	1.000
Eyes with subbasal nerves longer than 200 μm, n (%)	6 (100%)	6 (100%)	1.000	1 (13%)	0 (0%)	1.000	2 (22%)	0 (0%)	0.470	2 (33%)	0 (0%)	0.454	0.060	0.002*	0.538	1.000
Mean number of Sprouting nerves (SD)	0 (–)	0 (–)	>0.999	0.00 (0.00)	0.04 (0.12)	>0.999	0.04 (0.11)	0.07 (0.22)	>0.999	0.06 (0.14)	0.78 (0.89)	0.054	>0.999	0.015*	0.428	0.021*
Mean number of subbasal nerves (SD)	1.22 (0.40)	1.17 (0.46)	0.652	0.25 (0.46)	0.04 (0.12)	0.446	0.41 (0.46)	0.00 (0.00)	0.009*	0.50 (0.55)	0.00 (0.00)	0.061	0.048*	0.002*	0.373	>0.999
Nasal quadrant
Eyes with no nerves, n (%)	0 (0%)	0 (0%)	1.000	5 (63%)	6 (75%)	1.000	4 (44%)	6 (67%)	0.637	1 (17%)	1 (17%)	1.000	1.000	1.000	0.137	0.102
Eyes with sprouting nerves, n (%)	0 (0%)	0 (0%)	1.000	1 (13%)	1 (13%)	1.000	1 (11%)	2 (22%)	1.000	2 (33%)	5 (83%)	0.242	0.454	0.015*	0.538	0.025*
Eyes with subbasal nerves, n (%)	6 (100%)	6 (100%)	1.000	2 (25%)	1 (13%)	1.000	5 (56%)	2 (22%)	0.335	3 (50%)	1 (17%)	0.545	0.181	0.015*	0.580	1.000
Eyes with only subbasal nerves shorter than 200 μm, n (%)	0 (0%)	0 (0%)	1.000	2 (25%)	1 (13%)	1.000	4 (44%)	2 (22%)	0.619	0 (0%)	1 (17%)	1.000	1.000	1.000	0.472	1.000
Eyes with subbasal nerves longer than 200 μm, n (%)	6 (100%)	6 (100%)	1.000	0 (0%)	0 (0%)	1.000	3 (33%)	0 (0%)	0.206	3 (50%)	0 0%)	0.181	0.181	0.002*	0.055	1.000
Mean number of sprouting nerves (SD)	0 (–)	0 (–)	>0.999	0.04 (0.12)	0.04 (0.12)	>0.999	0.04 (0.11)	0.37 (0.99)	0.735	0.33 (0.56)	0.61 (0.57)	0.264	0.454	0.015*	0.318	0.008*
Mean number of subbasal nerves (SD)	1.11 (0.50)	1.22 (0.40)	0.678	0.08 (0.15)	0.04 (0.12)	>0.999	0.56 (0.71)	0.11 (0.24)	0.145	0.44 (0.54)	0.11 (0.27)	0.424	0.067	0.004*	0.202	0.736
Temporal quadrant
Eyes with no nerves, n (%)	0 (0%)	0 (0%)	1.000	4 (50%)	7 (88%)	0.282	5 (56%)	5 (56%)	1.000	2 (33%)	3 (50%)	1.000	0.454	0.181	0.627	0.244
Eyes with sprouting nerves, n (%)	0 (0%)	0 (0%)	1.000	1 (13%)	1 (13%)	1.000	0 (0%)	3 (33%)	0.206	0 (0%)	3 (50%)	0.181	1.000	0.181	1.000	0.244
Eyes with subbasal nerves, n (%)	6 (100%)	6 (100%)	1.000	3 (38%)	0 (0%)	0.200	4 (44%)	1 (11%)	0.294	4 (67%)	0 (0%)	0.060	0.454	0.002*	0.592	1.000
Eyes with only subbasal nerves shorter than 200 μm, n (%)	1 (16.7%)	0 (0%)	1.000	3 (38%)	0 (0%)	0.200	3 (33%)	1 (11%)	0.576	1 (17%)	0 (0%)	1.000	1.000	1.000	0.580	1.000
Eyes with subbasal nerves longer than 200 μm, n (%)	5 (83.3%)	6 (100%)	1.000	0 (0%)	0 (0%)	1.000	2 (22%)	0 (0%)	0.471	3 (50%)	0 (0%)	0.181	0.545	0.002*	0.055	1.000
Mean number of sprouting nerves (SD)	0 (–)	0 (–)	>0.999	0.08 (0.24)	0.04 (0.12)	>0.999	0.00 (0.00)	0.67 (1.76)	0.206	0.00 (0.00)	0.56 (0.81)	0.182	>0.999	0.181	>0.999	0.164
Mean number of subbasal nerves (SD)	1.06 (0.39)	1.11 (0.27)	0.844	0.12 (0.17)	0.00 (0.00)	0.200	0.44 (0.60)	0.04 (0.11)	0.135	0.61 (0.61)	0.00 (0.00)	0.060	0.182	0.002*	0.123	>0.999

*  Indicate statistically significant (P < 0.05).

The LASIK group showed a tendency toward more eyes with sprouting nerves and more sprouting nerves by frame than the SMILE group in all corneal areas and at all time points after surgery (representative confocal images were taken from the central area of the treated zone and are shown in Fig. 2); however, no significant differences were observed between the 2 groups at any time point (P > 0.05). No significant differences in number of eyes with sprouting nerves and the mean number of sprouting nerves by frame were observed in the SMILE and LASIK groups between preoperative and postoperative week 4 values, and between postoperative weeks 1 and 4 values, except in the treated area inferiorly and nasally of post-LASIK eyes (P < 0.05). In addition, at postoperative week 4, the LASIK group presented significantly reduced number of subbasal nerves, reduced number of eyes with subbasal nerves, and less subbasal nerves longer than 200 μm as compared to baseline values in all areas (P < 0.05, Table 2). The SMILE group presented more eyes with presence of subbasal nerves, more eyes with subbasal nerves longer than 200 μm, and more subbasal nerves by frame than the LASIK group in all areas and at all time points after surgery (Table 2); however, in most cases, no significant differences were observed between the 2 groups (P > 0.05), except centrally at postoperative week 2, where the SMILE group showed significantly more eyes with subbasal nerves longer than 200 μm (P = 0.029) and more subbasal nerves by frame (P = 0.007) than the LASIK group. Inferiorly, the SMILE group also presented with more eyes with subbasal nerves (P = 0.009), more eyes with subbasal nerves longer than 200 μm (P = 0.009), and more subbasal nerves by frame (P = 0.009) than the LASIK group at postoperative week 2 (Table 2). The SMILE group revealed no significant differences between postoperative week 4 and baseline values in number of subbasal nerves by frame, number of eyes with subbasal nerves, and number of eyes with subbasal nerves longer than 200 μm in all corneal areas (P > 0.05), except a significantly reduced mean number of subbasal nerves by frame in superior (P = 0.045) and inferior (P = 0.048) of the treated zone. No significant differences were observed between postoperative weeks 1 and 4 in number of subbasal nerves by frame, number of eyes with presence of subbasal nerves, and number of eyes with subbasal nerves longer than 200 μm in all areas in the SMILE and LASIK groups (P > 0.05) (Table 2). 

Figure 2

 

Sprouting nerves before and after SMILE and LASIK. Representative in vivo confocal microscopy images of observed sprouting nerves in the subbasal nerve plexus before and after surgery in SMILE and LASIK groups in central area of the laser treated zone. (A–D) SMILE group. (E–H) LASIK group. (A, E) Preoperative images; no spouting nerves were observed in the subbasal nerve plexus in either group. (B, F) Postoperative images 1 week after SMILE and LASIK, respectively. (C, G) Postoperative images 2 weeks after SMILE and LASIK, respectively. (D, H) Postoperative images 4 weeks after SMILE and LASIK, respectively.

Immunofluorescent Staining

The whole mount staining of the extracted lenticule revealed abundant presence of stromal nerves (Fig. 3), indicating that stromal nerves lying within the lenticule were removed in the SMILE procedure. The absence of stromal nerves in the central cornea still was evident 2 weeks post-SMILE (Fig. 4B) and similar observation could be made on corneas that underwent LASIK (Fig. 4A). 

Figure 3

 

Whole mount view of corneal stromal nerves in the extracted lenticule. The flat mount βIII-tubulin staining of the lenticule showed that extraction of lenticule in a SMILE procedure is associated with a significant removal of stromal nerves. Images were obtained with a Carl Zeiss AxioImager Z1 microscope at a magnification of ×50. To acquire the whole image of the stromal nerve structure within the lenticule, the consecutive images were merged utilizing the Panorama module in the Carl Zeiss AxioVision software. Dashed circle indicates the edge of the extracted lenticule.

Figure 4

 

Sagittal view of the corneal nerves 2 weeks after SMILE and LASIK procedure. Cross-sectional view of the whole cornea after LASIK (A) and SMILE (B) in superior-inferior orientation was constructed by merging consecutive images taken at ×50 magnification. (C) Subbasal and stromal nerve networks were severed by the LASIK flap side cut, which was placed inferiorly. (D) An abrupt discontinuation of the subbasal nerve network (arrow) was seen at the central cornea. Stromal nerves were not detected at this time point. (E) Subbasal nerves were abundant at the superior cornea, where the LASIK flap hinge was placed. (F) In post-SMILE eyes, subbasal nerves were abundant at the inferior quadrant. (G) Similar to LASIK, an abrupt discontinuation of the subbasal nerve network (arrow) and no stromal nerves were seen at the central cornea. (H) Subbasal nerves were severed by the small incision, which was placed superiorly. Images (C–H) were taken at ×100 magnification. Asterisks (*) indicate the flap side cut and small incision created in LASIK and SMILE procedure, respectively. Scale bar: 50 μm.

Two weeks postoperatively, the sagittal section showed the presence of subbasal nerves in the superior-central cornea, where the LASIK flap hinge was placed (Figs. 4A, 4E). In contrast, the creation of LASIK side cut (asterisk) severed the subbasal nerves in the inferior-central cornea (Figs. 4A, 4C). The absence and presence of the subbasal nerves in the inferior-central and superior-central cornea, respectively, created an abrupt discontinuation of the nerve network at the central cornea (arrow, Fig. 4D). In the SMILE-treated corneas, no subbasal nerves in the superior-central quadrant were seen, where the 3-mm incision was located (asterisk; Figs. 4B, 4H). The subbasal nerves were intact in the inferior-central quadrant, where no incision or side cut was created (Figs. 4B, 4F). Similar to the LASIK-treated corneas, the subbasal nerve network appeared to be abruptly disconnected at the central cornea (arrow, Fig. 4G). 

Four weeks post-LASIK, we could see discontinuous and fine stromal nerves, suggesting formation of new nerves, at the periphery of the superior and inferior laser-ablated zones (Fig. 5A). In a higher magnification image at the inferior cornea, where the side cut was created (asterisk), this formation of new stromal nerves (arrowhead) could be seen clearly (Fig. 5C). However, the subbasal nerves in the inferior-central cornea still were not detected at this time point (Figs. 5A, 5C). In contrast, the subbasal nerve network was abundantly present in the superior-central cornea (Figs. 5A, 5E). At the central cornea, no stromal nerve regeneration was observed and the subbasal nerve from the center to the inferior portion of the cornea was absent (arrow, Fig. 5D). 

Figure 5

 

Sagittal view of the corneal nerves 4 weeks after SMILE and LASIK procedure. Cross-sectional view of the whole cornea after LASIK (A) and SMILE (B) in superior-inferior orientation was constructed by merging consecutive images taken at ×50 magnification. (C) Discontinuous and fine stromal nerves (arrowhead), suggesting formation of new nerves, were seen at the inferior cornea. Subbasal nerves still were hardly present at this time point. (D) An abrupt discontinuation of the subbasal nerve network (arrow) and no stromal nerves were seen at the central cornea. (E) Subbasal nerves were abundant at the superior cornea, where the LASIK flap hinge was placed. (F) In post-SMILE eyes, subbasal nerves were abundant at the inferior quadrant. (G) An abrupt discontinuation of the subbasal nerve network (arrow) was seen at the central cornea. In contrast to LASIK, central stromal nerve sprouting (arrowhead) could be seen clearly 4 weeks post-SMILE. (H) Subbasal nerves still were absent superiorly. Formation of new stromal nerves was observed below the small incision site (arrowhead). Images (C–H) were taken at ×100 magnification. Asterisks (*) indicate the flap side cut and small incision created in LASIK and SMILE procedure, respectively. Scale bar: 50 μm.

Four weeks post-SMILE, we could find discontinuous and fine new stromal nerves at the periphery of the laser disrupted zone (Fig. 5B). In a higher magnification image at the superior cornea, where the small incision was created (asterisk), the stromal nerve sprouting (arrowhead) could be seen clearly (Fig. 5F). The stromal nerve regeneration appeared to be more prominent in post-SMILE corneas than in post-LASIK corneas (arrowhead, Fig. 5G). The subbasal nerves in the superior-central cornea still were not detected on week 4 after SMILE (Figs. 5B, 5F). In contrast, the subbasal nerve network was abundantly present in the inferior-central cornea (Figs. 5B, 5H). 

Discussion

The subbasal nerve plexus is located in the area between basal epithelial cells and below Bowman's layer. The subbasal nerve plexus allows the analysis of corneal nerve density in a reproducible and reliable way in contrast to the analysis of the stromal nerves. ^13,19 Signs of nerve damage can be observed with IVCM as decrease in nerve length, decrease in nerve density, presence of eyes with sprouting nerves, and decrease in the number of eyes with subbasal nerves. Signs of nerve recovery or regrowth can be observed as increase in nerve length, increase in nerve density, decrease in the mean number of sprouting nerves, and increase in the number of eyes with subbasal nerves. We observed signs of nerve damage and nerve recovery in the SMILE and LASIK groups. However, there was a tendency toward greater nerve damage after LASIK, which contrasted with the less nerve disruption and faster nerve recovery after SMILE. 

After surgery, the SMILE group showed a tendency toward greater subbasal nerve length and density, more subbasal nerves by frame, more eyes with subbasal nerves and more eyes with subbasal nerves longer than 200 μm than the LASIK group in all quadrants and central area of the treatment zone, and at all postoperative evaluation time points. The LASIK group showed a tendency toward more eyes with sprouting nerves and more sprouting nerves by frame than the SMILE group in all areas and time points after surgery. These trends suggested a better preservation of corneal subbasal nerves and earlier nerve recovery after SMILE relative to LASIK. Similarly, a recent study in human patients found that FLEx surgery, which requires a flap to be fashioned, induced a significant greater decrease in subbasal nerve density in central cornea as compared to SMILE 6 months after surgery. ²⁷  

In this study, we observed significantly more eyes with no nerves and significantly less eyes with subbasal nerves longer than 200 μm superiorly in SMILE relative to LASIK at postoperative week 4 compared to the values obtained preoperatively and also to postoperative week 1 values. The finding was confirmed further by our immunostaining results 2 weeks postoperatively, that revealed preservation of corneal subbasal nerves in the area of the hinge and absence of nerves at flap edge after LASIK. The opposite was true for SMILE, where our immunostaining showed the preservation of subbasal nerves in the area opposite to the small incision, but not the nerves located at the superior quadrant, the area of the small incision. 

The SMILE group presented earlier and faster recovery of subbasal nerves than the LASIK group in all cornea areas. After surgery, both groups presented an expected decrease in nerve length, density number of subbasal nerves by frame, and number of eyes with subbasal nerves. However, 4 weeks after surgery, only the SMILE group showed significant recovery of subbasal nerve length, density, number of subbasal nerves by frame, and number of eyes with subbasal nerves, showing no significant differences with baseline values, with an exception of the superior corneal area that continued to show significantly reduced number of subbasal nerves. At postoperative week 4, the LASIK group continued to show significant decrease in nerve length, density, number of subbasal nerves by frame, and number of eyes with subbasal nerves compared to baseline values. The LASIK group continued to show significantly more sprouting nerves 4 weeks after surgery relative to baseline values in inferior and nasal cornea areas, and a tendency toward more sprouting nerves in central, superior, and temporal areas. In addition, our immunostaining results 4 weeks postoperatively revealed a more prominent stromal nerve regeneration after SMILE than after LASIK. 

The reason for the differential nerve damage and subsequent recovery observed between SMILE and LASIK may be 2-fold. The first reason lies on the innervations area within the cornea that is compromised by the laser incision. In SMILE, this affected area includes the lenticule cut (6.5 mm in diameter) and the small pocket incision (2.5–3 mm in arcuate diameter). In LASIK, the affected area is substantially larger (80% larger side-cut area and 30% larger cap incision area than SMILE). ²⁸ The second reason may lie in the higher level of collateral damage and decellularization of keratocytes following the excimer laser-based LASIK procedure compared to the femtosecond-only SMILE procedure. ^16,29 Whether keratocytes are required to sustain function of corneal nerves or vice versa still is ambiguous, but keratocyte acellularity has been noted after photorefractive keratectomy (PRK) and LASIK, in which there is an extended period of denervation postoperatively. ^19,30 No specific physiologic relationships between nerves and keratocytes have been described, but, anatomically, human corneal nerve fibers have been found to invaginate the cytoplasm of some keratocytes, ³¹ raising the possibility of a cross-talk between the keratocytes and neuronal cells. Since keratocytes remain viable and have been able to repopulate along the femtosecond laser incision site after ReLEx, ^32,33 they may be able to provide nutrients and biological cues for proliferation, leading to improvement of the disrupted nerves regeneration rate. 

The sculpting and extraction of the lenticule in SMILE removes stromal nerves that lie within the lenticule, evident by the presence of βIII-tubulin–positive nerves on our flat mount staining of the extracted lenticule. The absence of βIII-tubulin staining at the central area of the corneal cross sections still could be observed 2 weeks post-SMILE. Although the nerve disruption is significantly less than LASIK, post-SMILE patients still experience immediate dry eye symptoms and reduced corneal sensation. ²⁷ Since corneal nerves are located predominantly in the anterior third of the human cornea, ⁸ it may be more advantageous to perform SMILE at a deeper depth in the stroma. 

In conclusion, to our knowledge this is the first study to compare early nerve damage and nerve recovery after SMILE and LASIK surgery. Although we were not able to show statistical significance in all of the analyzed parameters, our findings suggested that SMILE has a tendency to produce less nerve damage with faster and earlier nerve recovery than LASIK. 

Acknowledgments

Supported by National Research Foundation of Singapore-Funded Translational and Clinical Research Programme Grant NMRC/TCR/002-SERI/2008 and Centre Grant NMRC/CG/SERI/2010. 

Disclosure: K. Mohamed-Noriega, None; A.K. Riau, None; N.C. Lwin, None; S.S. Chaurasia, None; D.T. Tan, None; J.S. Mehta, None