Dry Eye Disease after Refractive Surgery

Comparative Outcomes of Small Incision Lenticule

Extraction versus LASIK
Alexandre Denoyer, MD, PhD,1,2,3 Elise Landman, MD,1 Liem Trinh, MD,1,4 Jean-Franois Faure, MD,1,4
Franois Auclin, MD,1 Christophe Baudouin, MD, PhD1,2,3,5

Purpose: To compare small incision lenticule extraction (SMILE) versus LASIK for post-refractive dry eye
disease.
Design: Prospective, comparative, nonrandomized clinical study.
Participants: Thirty patients scheduled for bilateral myopic SMILE and 30 age-, sex-, and refraction-
matched patients scheduled for bilateral myopic LASIK were enrolled and followed for 6 months after the surgery.
Methods: Complete evaluation of dry eye disease was performed at 1 and 6 months postoperatively, which
included vision-related quality of life (Ocular Surface Disease Index [OSDI]), clinical examinations (tear lm
breakup time [TBUT], Schirmer I test, corneal staining), and tear osmolarity measurements, together with an
overall severity score. Function and morphology of the corneal innervation were evaluated by corneal esthesi-
ometry and subbasal nerve imaging using in vivo confocal microscopy (IVCM).
Main Outcome Measures: Overall analysis of dry eye disease and corneal innervation.
Results: High incidence of mild to moderate dry eye disease was observed in both groups 1 month post-
operatively, which remained signicantly higher in the LASIK group than in the SMILE group 6 months after
surgery (overall severity score [0e4]: 1.2
1.1 vs. 0.2
0.4, respectively, P < 0.01), leading to more frequent use of
tear substitutes over the long term. Corneal sensitivity was better in SMILE than in LASIK eyes 1 month post-
operatively (3.5
1.79 vs. 2.45
2.48, respectively, P < 0.01) and then recovered to statistically similar values at 6
months. Corneal nerve density, number of long bers, and branchings as assessed by IVCM were signicantly
higher in the SMILE group compared with the LASIK group 1 and 6 months after surgery. Corneal sensitivity was
negatively correlated with dry eye-related corneal damage (R2 0.48, P < 0.01), and the long ber nerve density
was independently correlated with the OSDI score (R2 0.50, P < 0.01) and the Schirmer test (R2 0.21, P <
0.01) 6 months postoperatively.
Conclusions: The SMILE procedure has a less pronounced impact on the ocular surface and corneal innervation
compared with LASIK, further reducing the incidence of dry eye disease and subsequent degradation in quality of life
after refractive surgery. Ophthalmology 2015;122:669-676 2015 by the American Academy of Ophthalmology.

For the past 2 decades, LASIK has become the most popular
corneal refractive surgery with approximately 1 million
procedures per year in the United States.1 Although a high
satisfaction rate is reported, dry eye is still the most common
adverse effect of LASIK. Many patients experienced mild-
severity dry eye symptoms for a few months after LASIK,
which are sufciently eased using conventional tear sub-
stitutes, but patients still reported dry eye over the long term,
with an occurrence of chronic dry eye disease ranging from
20% to 40% at least 6 months after surgery.2 Dry eye causes
damage to the ocular surface and symptoms of ocular
discomfort associated with visual disturbance, which de-
grades not only the visual outcomes but also the quality of
everyday life.3e5 Thus, it could be assumed that hundreds of
thousands of patients are likely to develop chronic dry eye
disease after LASIK every year, further affecting the health
status of this young and active population.
Total disruption of corneal nerves due to ap making
combined with excimer photoablation is a likely cause of
post-LASIK dry eye. A 90% decrease in central nerve ber
density has been reported in the rst few months after
LASIK, lasting for years until it recovers preoperative
values or does not.6 As a result, LASIK induces a decrease
in tear lm quality, tear secretion, blinking rates, and
epithelial wound healing, all factors known to be involved in
the pathogenesis of dry eye disease. Femtosecond laser was
developed for LASIK to improve ap making and allows
customization. However, the control and optimization of
ap features brought few improvements in post-LASIK dry
eye,7 thus the need to develop new procedures to better
protect the ocular surface after refractive surgery.
Small incision lenticule extraction (SMILE) is a recent
procedure using femtosecond laser to create an intrastromal
lenticule that is then removed through a small corneal

 2015 by the American Academy of Ophthalmology http://dx.doi.org/10.1016/j.ophtha.2014.10.004 669

Published by Elsevier Inc. ISSN 0161-6420/14
 Ophthalmology Volume 122, Number 4, April 2015

incision.8,9 Contrary to LASIK, this all-in-one femtosecond Patients

refractive surgery no longer needs excimer laser photo-
ablation or a full ap cut. As a result, SMILE could
constitute a minimally invasive approach to corneal refrac-
tive surgery because it only requires a small tunnel, which
may have less impact on corneal innervation, thus further
protecting patients against iatrogenic dry eye disease.
Although clinical studies have reported refractive outcomes,
corneal sensitivity, and clinical dryness after SMILE,
nothing has been done to evaluate the overall severity of the
disease, which requires a combination of objective tests and
subjective symptom assessment as recommended by
Delphi,10 and its precise relationship with morphologic and
functional changes in corneal innervation. Thus, uncertainty
remains about long-term ocular surface recovery and sub-
sequent patient health status after SMILE, which could
become the new gold standard for corneal refractive surgery,
provided its theoretic benets are demonstrated.
This nonrandomized study was designed to prospectively
compare SMILE with LASIK for surgically induced corneal
changes and dry eye disease in accurately matched pop-
ulations in terms of age, gender, and refraction. Clinical
evaluation, a vision-related quality of life questionnaire, tear
osmolarity assessment, corneal esthesiometry, anterior
segment optical coherence tomography (OCT), and in vivo
confocal microscope (IVCM) imaging were performed 1
and 6 months postoperatively to determine whether the
SMILE procedure preserves the ocular surface and put an
end to the common but sometimes deleterious post-refrac-
tive dry eye disease.
Methods
This prospective, comparative, nonrandomized clinical study was
conducted in the Clinical Center for Investigation of Ocular Sur-
face Pathology (Quinze-Vingts National Ophthalmology Hospital,
National Institute for Health and Medical Research 503, Paris,
France) in accordance with the Declaration of Helsinki, Scotland
amendment, 2000. Previous approval was obtained from the Na-
tional Ethical Research Committee (Comit de Protection des
Personnes Ile de France V, National Agreement Number 10793).
All patients gave informed consent to participate in the study.
Thirty European subjects scheduled for bilateral SMILE (J.F.F.)
and 30 European age-, sex-, and spherical equivalentematched
subjects scheduled for bilateral LASIK (F.A.) were prospectively
included. Inclusion criteria were planned myopic SMILE or
LASIK (spherical correction range, 1 to 8 diopters; cylinder
range, 0 to 1.5 diopters), willingness to participate in the study,
and the ability to give informed consent. Exclusion criteria were
any ocular pathology but myopia, any clinical sign or symptom of
dry eye disease (Schirmer I test >10 mm/5 minutes, tear lm
breakup time [TBUT] >10 seconds, no corneal/conjunctival
staining, no Meibomian gland dysfunction, and Oxford score 0),
previous ocular/eye lid medical or surgical treatment, systemic
disorder, and pregnancy. All the examinations were performed 1
month and 6 months after the surgery, except OCT, which was
performed at 6 months only. Detailed patient data are shown in
Table 1.
Surgical Technique
All procedures were performed with topical anesthesia (0.8% oxy-
buprocaine tetrachloride) after standard sterile draping and insertion of
an eyelid speculum. SMILE was performed bilaterally using femto-
second laser (Visumax, Carl Zeiss Meditec, Jena, Germany) with a
110-mm depth and a 6.5-mm diameter lenticule, and a 10-oclock small
tunnel incision. The lenticule was gently detached and extracted
through the corneal tunnel using a spatula. The maximum lenticule
thickness ranged from 51 to 152 mm. For the LASIK group, the corneal
ap was made by femtosecond laser (IFS500, Abbot Medical Optics,
Abbot Laboratories, Chicago, IL) with a 110-mm depth, 9-mm diam-
eter ap, and 50 superior hinge. Excimer photoablation was per-
formed (Allegretto, Alcon Laboratories, Fort Worth, TX) for a 6.5-mm
optical zone, and then the ap was repositioned. In both groups, 0.3%
tobramycin/0.1% with dexamethasone suspension (Tobradex, Alcon
Laboratories) associated with preservative-free tear substitutes were
used 3 times per day for 1 month, and then preservative-free uid tear
substitutes or gels were administered when needed.
Clinical Examinations and Questionnaire
Slit-lamp evaluations were conducted in a dened sequence11 and
included TBUT measurements (mean of 3 consecutive tests),
ocular surface uorescein staining (grade 0e5, according to the
Oxford score), and the Schirmer I test (mm/5 minutes, without
anesthesia). Before clinical examination, a trained interviewer

Table 1. Patient Features, Visual Outcomes, and Corneal Morphology as Assessed by Anterior Segment Optical Coherence Tomography

SMILE Group (n[30) LASIK Group (n[30) t Test (P)

Preoperative data
Age (yrs) 31.1
4.7 32.2
7.5 NS
Gender (M/F) 0.47 0.47 NS
Mean keratometry (D) 42.9
1.6 43.7
1.5 NS
Spherical equivalent (D) 4.65
2.38 4.42
1.78 NS
6-mo outcomes
Mean keratometry (D) 40.9
1.4 39.9
2.5 NS
Spherical equivalent (D) 0.03
0.41 0.13
0.4 NS
Distant best-uncorrected visual acuity (logMAR) 0.07
0.10 0.08
0.10 NS
Epithelial thickness (mm) 39.9
14.1 39.6
13.6 NS
Total corneal thickness (mm) 496.3
39.3 506.7
39.5 NS
Depth of the interface (mm) 117.2
4.6 117
4.5 NS

Data are reported as mean
standard deviation. D diopters; logMAR logarithm of the minimum angle of resolution; NS not signicant;
SMILE small incision lenticule extraction.

670
 Denoyer et al 
Dry Eye and Small Incision Lenticule Extraction
(E.L.), who was masked to patient group, administered the French
version of the Ocular Surface Disease Index (OSDI) questionnaire,
which was developed to quantify the specic impact of dry eye
disease on vision-targeted health-related quality of life.12 Tear
osmolarity was also measured using the TearLab osmolarity sys-
tem (TearLab Corp, San Diego, CA). An overall classication of
the severity of dry eye disease was determined according to the
modied scheme of the Delphi Panel Report.3
Corneal Esthesiometry
Corneal sensitivity was measured using the contact nylon thread
Luneau 12/100 mm CocheteBonnet esthesiometer (Luneau,
Prunay-Le-Gillon, France). A 6.0-cm adjustable nylon lament is
applied perpendicularly to the cornea, which creates a pressure
gradient ranging from approximately 11 to 200 mg/mm2. Starting
from 6.0 cm, the lament length was progressively reduced in 5-
mm steps until the rst response occurred. Measurements were
taken in a masked manner, that is, by a clinician (E.L.) who was
not aware of the type of surgery, and data are reported as the mean
of 3 measurements at the center of the cornea and 3 measurements
in each of 4 peripheral quadrants.
Anterior Segment Optical Coherence Tomography
A spectral-domain OCT tted with an anterior segment module
(Optovue Corp, Fremont, CA) was used. The OCT axial and lateral
optical resolutions were 18 and 60 mm, respectively. All acquisi-
tions were taken using the high-resolution mode with an acquisi-
tion time of 0.125 seconds/cross-section for overall examinations.
For each eye, 2 orthogonal images of the cornea were acquired, and
then epithelial thickness, total thickness, and depth of the interface
were measured at the apex.
In Vivo Confocal Microscopy
In vivo confocal microscopy of the cornea was performed using the
Rostock Cornea Module of the Heidelberg Retina Tomograph
(Heidelberg Engineering GmbH, Heidelberg, Germany). Briey, the
microscope lens is an immersion lens (Olympus, Hamburg, Ger-
many) with 60 magnication. The images comprised 384384
pixels covering an area of 400400 mm with 2-mm transversal res-
olution and 4-mm axial resolution, and an acquisition time of 0.024
seconds. Before IVCM evaluation, 1 drop of topical anesthetic
(oxybuprocaine 0.4%; MSD-Chibret, Paris, France) and 1 drop of gel
tear substitute (Lacrigel, carbomer 0.2%; Europhta, Monte-Carlo,
Monaco) were instilled in the lower conjunctival fornix.
Images of the subbasal nerves of the central cornea were acquired
using the same illumination intensity (manual mode) and by focusing
the microscope beneath the basal epithelium. Approximately 20 im-
ages were acquired in the central cornea for each eye, and 5 images
were randomly selected for quantitative measurements. Images were
anonymized and then analyzed using NeuronJ software by a single
researcher (A.D.), who was masked to patient group and features.
NeuronJ is a free ImageJ plug-in (National Institutes of Health,
Bethesda, MD) allowing semiautomated tracing and quantication of
nerves. Corneal innervation was evaluated by quantitating (i) the
density of corneal nerves, dened as the total length of the nerves
visible within a frame (mm/mm2); (ii) the number of main nerves,
dened as the sum of the long nerve ber bundles observed within a
frame (number/mm2); and (iii) the number of branchings, dened as
the mean number of secondary bers connecting to long nerve bers
within a frame (number/mm2), as previously detailed.13 Data on the
corneal innervation in healthy patients (n 14, sex ratio 0.75 [M/
F], mean age 45.4
9.2 years, no ocular sign or symptom of dry
eye), which were previously reported by our team following the same
procedure,13 were also included for discussion purpose. In parallel,
the density of dendritic cells (number/mm2) in the epithelial layer and
the density of activated keratocytes (number/mm2) in the anterior
stroma were also calculated.
Statistical Analysis
All data are given as the mean
standard deviation. The primary
outcome was the severity of dry eye 6 months after SMILE versus
LASIK. Pre-study analysis was conducted to determine the sample
size, which was based on the proportion of patients with moderate
to severe dry eye (overall dry eye score >1) at 6 months post-
surgery. Given the reportedly known incidence of post-refractive
dry eye and an assumed difference of 10% or more between
SMILE and LASIK, a sample size of 29 patients per group was
calculated for a 2-sided signicance level of a 0.05 and type II
error of b 0.1. Considering an expected dropout rate less than
5%, we included 60 patients. For ocular clinical examinations and
imaging, 1 eye only per patient was selected using a random
number table to not bias the statistical relevance of the results. Data
were controlled for normality, homogeneity of variances, and
sphericity to perform the adequate tests. The SMILE and LASIK
groups were compared using the t test or the ManneWhitney test
depending on the assumptions. Scatter plots, the R2 correlation
coefcient, and Spearmans rank test were used to assess the as-
sociation between pairs of variables. A correlation matrix followed
by a stepwise regression procedure was performed to determine
accurate multiple regression models. The probability level of sig-
nicance was adjusted according to the post hoc Bonferroni pro-
cedure to maintain an overall type I error equal to 0.05.
Results
No adverse effects occurred in any of the 120 procedures. Six-
month postoperative best-uncorrected visual acuity and residual
spherical equivalent were not statistically different between SMILE
and LASIK eyes, as detailed in Table 1. Morphologic parameters
including the keratometry, epithelial thickness, and total thickness
of the cornea, and the mean depth of the interface, as assessed by
OCT, did not differ between the 2 groups.
Clinical Dry Eye Disease after Refractive Surgery
One month after the surgery, a high rate of signs and symptoms of
dryness was reported in both groups (Table 2). The data collected
did not show any signicant difference between the groups, except
for tear osmolarity, which was higher in the LASIK group than in
the SMILE group. The distribution of the severity of dry eye dis-
ease 1 month after SMILE versus LASIK is detailed in Figure 1A.
Six months after the surgery, patient-reported vision-related
quality of life (OSDI) together with tear lm quality (TBUT) and
osmolarity were signicantly impaired in the LASIK group
compared with the SMILE group, leading to a worse severity score
of dry eye disease in the LASIK group (Table 2, Fig 1B). At this
point, 80% of patients in the SMILE group did not use any eye
drops at 6 months postoperatively versus 57% in the LASIK group,
with 20% of the LASIK group needing daily and frequent use of
tear substitutes or even gels versus none of the patients in the
SMILE group (Fig 2).
Corneal Sensitivity and Innervation
Corneal sensitivity, as assessed by the CocheteBonnet esthesi-
ometer, was hampered in both groups 1 month after the surgery,
but was slightly higher in the SMILE group than in the LASIK
group (Fig 3A). Both groups recovered to normal and did not have
statistically different values at 6 months post-surgery.
671
 Ophthalmology Volume 122, Number 4, April 2015

Table 2. Dry Eye Disease 1 and 6 Months after Refractive Surgery

1 Month 6 Months
SMILE LASIK P SMILE LASIK P
OSDI (0e100) 19.7
12.7 23.9
14.8 0.09 7.54.5 20.620.8 <0.01
TBUT (s) 5.9
1.7 5.1
1.9 0.16 71.8 5.21.8 0.01
Schirmer I test (mm/50 ) 13.2
6.1 19.9
10.5 0.08 17.3
8.2 16.9
7.8 0.85
Oxford score (0e5) 0.08
0.28 0.27
0.55 0.16 0 0.4
0.6 0.06
Tear osmolarity (mOsm) 305.112.5 316.311.6 <0.01 300.311.4 315.011.9 <0.01
Dry eye severity score (0e4) 1.0
0.8 1.5
0.9 0.06 0.20.4 1.21.1 <0.01

OSDI Ocular Surface Disease Index; SMILE small incision lenticule extraction; TBUT tear lm breakup time.
The dry eye severity score was calculated according to the modied Delphi approach.3,10 Data are reported as mean
standard deviation. Boldface indicates
statistical signicance (P < 0.01).

The subbasal nerve plexus of the cornea was evaluated using
IVCM. The nerve density, number of long nerve bers, and nerve
branchings in the SMILE group were statistically superior to those in
the LASIK group 1 month and 6 months after the surgery (Fig 3B). In
addition, the density of dendritic cells at the corneal surface was
greater in the LASIK group compared with the SMILE group at 1
month and 6 months postoperatively (156.4
63.2 cells/mm2 vs.
75.9
26 cells/mm2 at 1 month and 107
34.5 cells/mm2 vs. 52
18.6
cell/mm2 at 6 months, P < 0.01 for both), whereas the density of
activated keratocytes was lower in the LASIK group 1 month post-
operatively (252.9
62.9 vs. 331
71.5, P 0.01) and not statistically
different from that of the SMILE eyes 6 months after the surgery.
Correlations of Dry Eye Disease Severity with
Surgically Induced Changes in Corneal Innervation.
The correlation matrix revealed a statistical association of corneal
sensitivity with corneal staining and the overall severity score of dry
eye disease (Table 3). In regard to IVCM innervation data, nerve
density was negatively correlated with the OSDI score and corneal
staining; the number of long bers was negatively correlated with the
OSDI score, corneal staining, overall severity score, and corneal
sensitivity (Table 3). Because clinical data were highly interlinked, a
stepwise regression routine was performed to determine independent
correlations. Corneal sensitivity appeared to correlate with corneal
staining only (R2 increment 0.48, P < 0.01). In addition, nerve
density was found to independently correlate with the OSDI score and
the Schirmer test (R2 increment 0.35, P < 0.01; and R2 increment
0.16, P 0.02, respectively), as was the number of long nerve bers
(R2 increment 0.50, P < 0.01; and R2 increment 0.21, P < 0.01,
respectively).
Discussion
Dry eye is still the most frequent complication of refractive
surgery.14 Post-refractive dry eye disease is an immediate
issue because the patients visual comfort and quality of life
dictate their overall satisfaction.15 On a more global scale,
ocular dryness after refractive surgery may lead to perma-
nent pain, visual uctuations, and the need for daily treat-
ment, which affects health status, quality of life, and vision
when it lasts for months or even years. New corneal pro-
cedures, such as SMILE, could both reduce the risk for
iatrogenic dry eye disease and improve the overall outcome,
and make indications for refractive surgery evolve by
reducing limitations related to preoperative ocular dryness.
According to the denition of the Dry Eye WorkShop,
dry eye disease is a multifactorial pathology at the ocular
surface, which includes tear lm changes with or without

Figure 1. Distribution of post-refractive dry eye disease. The severity of dry eye disease was scored according to an overall index (range, 0e4) including signs
and symptoms3,8 in small incision lenticule extraction (SMILE) and LASIK cases 1 month (A) and 6 months (B) after the surgery. The mean score was
signicantly higher in the LASIK than in the SMILE group at 6 months postoperatively (P < 0.01) (Table 2).

672
 Denoyer et al 
Dry Eye and Small Incision Lenticule Extraction

innervation and corneal changes. Although post-LASIK dry

Figure 2. The use of tear substitute eye drops 6 months after small incision
lenticule extraction (SMILE) versus LASIK.
corneal damage, ocular symptoms, visual degradation, and
increased tear osmolarity, together leading to degradation of
quality of life.3 As a result, a full appropriate evaluation is
required to accurately diagnose and evaluate the disease,
especially because no specic marker reecting its overall
severity has been identied. The 2 main strengths of the
present study are (1) the matching of SMILE with LASIK
cases in terms of age, gender, and refraction, the main
recognized risk factors for dry eye, to optimize the relevance
of the results; and (2) the complete evaluation of the disease
combined with a morphologic and functional analysis of
eye is a well-documented complication, a few studies have
analyzed both objective and subjective symptoms after
LASIK, one including the OSDI-validated questionnaire for
dry eye disease14e21 as recommended by Delphi, and only
one has analyzed them for SMILE.22 The present study was
designed within a complete, original, and comprehensive
approach to dry eye disease after refractive surgery, and it
demonstrated an increase in symptoms (OSDI score), signs
(TBUT), and tear osmolarity 6 months after LASIK
compared with SMILE. Accordingly, some authors have
reported decreased values on these scales in patients
receiving LASIK up to years after surgery,17,21,23 although
clinical signs are often found to be highly variable, as
pointed out by Feng et al.7 As for SMILE, Shah et al24
found that approximately 40% of the patients thought that
their eyes were dryer than before the surgery using a self-
reported symptom questionnaire. Li et al22 recently
compared LASIK and SMILE for dry eye disease and re-
ported a better OSDI and TBUT in SMILE eyes compared
with LASIK eyes than reported in this article. We also report
that the higher incidence of dry eye disease after LASIK was
associated with more frequent use of eye drops over the long
term, which may also degrade the quality of life and raises
the question of the cost of treatments in this young
population.
LASIK-induced dry eye relies mainly, but not exclu-
sively, on the dramatic breakdown of subbasal nerves as a
result of making the ap. In contrast, the SMILE procedure
protects corneal innervation because it creates a 40 - to 60 -
wide penetrating tunnel only, compared with the

Figure 3. Morphologic and functional assessment of corneal innervation 1 month and 6 months after small incision lenticule extraction (SMILE) versus
LASIK. A, Corneal sensitivity as measured by the CocheteBonnet esthesiometer (Luneau, Prunay-Le-Gillon, France). LASIK eyes showed lower sensitivity
than SMILE eyes at 1 month postoperatively (*P<0.05). B, Corneal innervation as assessed using in vivo confocal microscopy. Nerve density, number of
long bers, and secondary branchings were higher in SMILE than in LASIK eyes (*P < 0.05; **P < 0.01). Control group is presented for illustrative
purposes, which is taken from a previous study conducted by our team (n 15).13

673
 Ophthalmology Volume 122, Number 4, April 2015
Table 3. Correlations of Signs and Symptoms of Dry Eye with Corneal Sensitivity and Innervation
Esthesiometry
Corneal Sensitivity
Age 0.28
Preoperative spherical equivalent 0.31
OSDI 0.65y
TBUT 0.01
Schirmer 0.15
Corneal staining 0.69y
Overall dry eye severity score 0.60y
Tear osmolarity 0.01
Corneal sensitivity
OSDI Ocular Surface Disease Index; TBUT tear lm breakup time.
R correlation coefcient is reported; Spearman rank test: appears in boldface.
*P < 0.05
y
P < 0.01.
approximately 300 -wide penetrating tunnel for LASIK.
From the early beginnings of refractive surgery, studies have
reported a signicant decrease in corneal sensitivity after
LASIK, which likely lasts for months or even years despite
the use of a femtosecond laser for ap making.25e29 As for
SMILE, 3 recent studies clinically reported that SMILE
maintained corneal sensitivity compared with LASIK.22,30,31
The present study has conrmed a signicant decrease in
corneal sensitivity at 1 month postoperatively after LASIK,
with similar values recorded between groups at 6 months,
according to the ndings by Demirok et al.31 In both groups,
the corneal sensitivity at 6 months appeared not to be
different than that obtained in healthy controls in a previous
study conducted by our team,13 suggesting a progressive
recovery of the normal values in both procedures. More
interestingly, we originally combined clinical assessment of
corneal innervation (i.e., sensitivity measurement) with
IVCM imaging of the nerves to determine the relationship
between morphology and function. Basically, LASIK was
objectively found to reduce the corneal nerve density over the
long term,29,32,33 whereas SMILE preserved nerve density as
previously reported.34,35 Objective features of the corneal
innervation 6 months after SMILE were found to be normal
and not differ from those previously obtained by our team
after the same image analysis in healthy patients.13 More
precisely, we investigated specic parameters of corneal
innervation (e.g., total nerve density, number of long bers,
and secondary branchings) and analyzed their relationship
with the overall clinical data. We demonstrate (1) a positive
correlation between corneal sensitivity and the number of
long nerve bers and (2) signicant associations of corneal
innervation parameters with clinical signs of dryness and
patient-reported symptoms, further improving our clinical
understanding of what occurs at the ocular surface after
SMILE compared with LASIK surgery. Vestergaard et al35
recently conducted a prospective controlled study to compare
SMILE with femtosecond lenticule extraction (FLEX) using
the same analysis methods as in the present study.35
Accordingly, TBUT, corneal sensitivity, and corneal nerve
morphology at 6 months were found to be signicantly better
after SMILE than after FLEX. No statistical association
674
In Vivo Confocal Microscopy
Nerve Density No. of Long Fibers Branchings
0.05 0.02 0.22
0.1 0.32 0.20
0.46* 0.55y 0.34
0.12 0.06 0.33
0.11 0.29 0.17
0.41* 0.40* 0.36
0.32 0.44* 0.20
0.02 0.13 0.02
0.36 0.45* 0.13
between the nerve density and the other variables was re-
ported except a moderate correlation for the Schirmer test
(P 0.05) in the FLEX group, consistent with what we
found. Even if FLEX is not the same procedure as LASIK, it
could be assumed that it also causes a dramatic breakdown of
corneal subbasal nerves, further conrming the benets of
SMILE.
The pathophysiology of post-refractive dry eye is a wide
open eld of basic and clinical research. Surgically induced
changes in corneal innervation play a pivotal role in the
pathogenesis of tear dysfunction and dryness, considered as
neurogenic/neuropathic dry eye disease, as recently reported
by Chao et al.2 We have reported signicant hypoesthesia in
the LASIK group versus the SMILE group 1 month post-
operatively but not at 6 months, which correlated with tear
secretion, but we could have improved the outcomes of the
study by optimizing the assessment of corneal sensitivity,
for example, using a noncontact esthesiometer,36 and by
measuring the blink rate and tear clearance,23 which was not
possible because of the lack of appropriate tools and stan-
dardized procedures. Surgical factors for corneal denerva-
tion (e.g., preoperative refraction, ap diameter and depth,
and hinge position) and related dry eye have been studied,
but their true impact has not been fully elucidated.7,18 The
aim of the present study was not to investigate risk factors
for dry eye disease, but it would have been valuable to make
a lateral hinge ap to specify this point. It should also be
noted that high variability of post-surgery corneal reinner-
vation, whose duration ranges from 3 months to 5 years
depending on the studies, could be another limitation in this
study because the end point was 6 months after surgery.37
This study reports a pathologic increase in tear osmolarity 6
months after LASIK, thus including this mechanism within
the pathogenesis of post-LASIK dry eye, contrary to
SMILE. In fact, the neurogenic origin of post-refractive dry
eye is undoubtedly associated with inammatory mecha-
nisms.38,39 This study originally reported a high density of
dendritic cells at the ocular surface 6 months after LASIK,
further conrming the role of inammatory processes in the
disease, which could be better analyzed by measuring other
inammatory mediators, such as cytokine in tears.39 Central
 Denoyer et al 
Dry Eye and Small Incision Lenticule Extraction
epithelial thickness as assessed by OCT did not differ be-
tween SMILE and LASIK at 6 months after surgery.
Repeated mapping of the epithelial thickness in the entire
cornea could better detail the impact of post-SMILE dry eye
and inammation on epithelial changes. Last, we revealed
transient changes in keratocyte activation, for which a long-
term follow-up could be valuable.40,41
In this study, we decided not to conduct a randomized
paired-eye study, but performed bilateral surgery using the
same technique, thus performing interindividual com-
parisons. As a result, even if the patients undergoing SMILE
had been paired with age-, gender-, and refraction-matched
patients undergoing LASIK, the study design could consti-
tute a limitation compared with a paired-eye approach.
Because we aimed at comparing SMILE with LASIK for 6-
month outcomes, data on preoperative morphology and
function of the corneal innervation were not collected. On
the basis of previous studies conducted by our team, how-
ever, we discussed the results regardless of those in the
healthy control group, which were previously obtained us-
ing exactly the same corneal analyses as in the present
study. A long-term follow-up of all preoperative and post-
operative variables could strengthen the present conclusions.
Today, a better understanding of the mechanisms
involved in the pathogenesis of post-refractive dry eye
disease is a crucial issue for 2 main reasons. First, this
should better dene the severity and impact of this common
surgically induced disease on quality of life, further
contributing to the development and evaluation of innova-
tive refractive procedures such as SMILE. Second, this
should determine the risk factors for surgically induced dry
eye, further making indications evolve according to the
procedure and preoperative condition of the ocular surface.
The present study demonstrates that SMILE greatly de-
creases the incidence of post-refractive dry eye disease
compared with LASIK in a comparable population with a
perfect preoperative ocular surface; further studies will
determine whether this procedure could be recommended to
those with low- to mild-severity dry eye, who are especially
represented within the population of contact lens wearers
and should not undergo refractive surgery using current
conventional procedures.
References
1. Eye health statistics at a glance. Compiled by American
Academy of Ophthalmology. Available at: http://www.aao.
org/newsroom/upload/Eye-Health-Statistics-April-2011.pdf.
Updated April 2011. Accessed November 4, 2014.
2. Chao C, Golebiowski B, Stapleton F. The role of corneal
innervation in LASIK-induced neuropathic dry eye. Ocul Surf
2014;12:3245.
3. The denition and classication of dry eye disease: report of
the denition and classication subcommittee of the Interna-
tional Dry Eye Workshop. Ocul Surf 2007;5:7592.
4. Denoyer A, Rabut G, Baudouin C. Tear lm aberration dy-
namics and vision-related quality of life in patients with dry
eye disease. Ophthalmology 2012;119:18118.
5. Deschamps N, Ricaud X, Rabut G, et al. The impact of dry eye
disease on visual performance while driving. Am J Oph-
thalmol 2013;156:1849.
6. Calvillo MP, McLaren JW, Hodge DO, Bourne WM. Corneal
reinnervation after LASIK: prospective 3-year longitudinal
study. Invest Ophthalmol Vis Sci 2004;45:39916.
7. Feng Y, Yu J, Wang D, et al. The effect of hinge location on corneal
sensation and dry eye after LASIK: a systematic review and meta-
analysis. Graefes Arch Clin Exp Ophthalmol 2013;251:35766.
8. Sekundo W, Kunert KS, Blum M. Small incision corneal
refractive surgery using the small incision lenticule extraction
(SMILE) procedure for the correction of myopia and myopic
astigmatism: results of a 6 month prospective study. Br J
Ophthalmol 2011;95:3359.
9. Ivarsen A, Asp S, Hjortdal J. Safety and complications of more
than 1500 small-incision lenticule extraction procedures.
Ophthalmology 2014;121:8228.
10. Behrens A, Doyle JJ, Stern L, et al. Dysfunctional tear syn-
drome. A Delphi approach to treatment recommendations.
Cornea 2006;25:9007.
11. Smith J, Nichols KK, Baldwin EK. Current patterns in the use of
diagnostic tests in dry eye evaluation. Cornea 2008;27:65662.
12. Schiffman RM, Christianson MD, Jacobsen G, et al. Reli-
ability and validity of the Ocular Surface Disease Index. Arch
Ophthalmol 2000;118:61521.
13. Labb A, Liang Q, Wang Z, et al. Corneal nerve structure and
function in patients with non-Sjogren dry eye: clinical corre-
lations. Invest Ophthalmol Vis Sci 2013;54:514450.
14. Toda I. LASIK and the ocular surface. Cornea 2008;27:S706.
15. Ambrsio R Jr, Tervo T, Wilson SE. LASIK-associated dry eye
and neurotrophic epitheliopathy: pathophysiology and strategies
for prevention and treatment. J Refract Surg 2008;24:396407.
16. Yu EY, Leung A, Rao S, Lam DS. Effect of laser in situ ker-
atomileusis on tear stability. Ophthalmology 2000;107:21315.
17. Toda I, Asano-Kato N, Komai-Hori Y, Tsubota K. Dry eye after
laser in situ keratomileusis. Am J Ophthalmol 2001;132:17.
18. Donnenfeld ED, Solomon K, Perry HD, et al. The effect of
hinge position on corneal sensation and dry eye after LASIK.
Ophthalmology 2003;110:10239.
19. De Paiva CS, Chen Z, Koch DD, et al. The incidence and risk
factors for developing dry eye after myopic LASIK. Am J
Ophthalmol 2006;141:43845.
20. Hassan Z, Szalai E, Berta A, et al. Assessment of tear osmolarity
and other dry eye parameters in post-LASIK eyes. Cornea
2013;32:e1425.
21. Lee JB, Ryu CH, Kim J, et al. Comparison of tear secretion and
tear lm instability after photorefractive keratectomy and laser in
situ keratomileusis. J Cataract Refract Surg 2000;26:132631.
22. Li M, Zhao J, Shen Y, et al. Comparison of dry eye and corneal
sensitivity between small incision lenticule extraction and
femtosecond LASIK for myopia. PloS One 2013;8:e77797.
23. Battat L, Macri A, Dursun D, Pugfelder SC. Effects of laser
in situ keratomileusis on tear production, clearance, and the
ocular surface. Ophthalmology 2001;108:12305.
24. Shah R, Shah S, Sengupta S. Results of small incision lenticule
extraction: all-in-one femtosecond laser refractive surgery.
J Cataract Refract Surg 2011;37:12737.
25. Kanellopoulos AJ, Pallikaris IG, Donnenfeld ED, et al.
Comparison of corneal sensation following photorefractive
keratectomy and laser in situ keratomileusis. J Cataract Refract
Surg 1997;23:348.
26. Petznick A, Chew A, Hall RC, et al. Comparison of corneal
sensitivity, tear function and corneal staining following in situ
keratomileusis with two femtosecond laser platforms. Clin
Ophthalmol 2013;7:5918.
675
 Ophthalmology Volume 122, Number 4, April 2015
27. Patel SV, McLaren JW, Jittkeson KM, et al. Subbasal nerve
density and corneal sensitivity after laser in situ keratomileu-
sis: femtosecond laser vs. mechanical microkeratome. Arch
Ophthalmol 2010;128:14139.
28. Benitez-del-Castillo JM, del Rio T, Iradier T, et al. Decrease in
tear secretion and corneal sensitivity after laser in situ kerat-
omileusis. Cornea 2001;20:302.
29. Linna TU, Prez-Santonja JJ, Tervo KM, et al. Recovery of
corneal nerve morphology following laser in situ keratomil-
eusis. Exp Eye Res 1998;66:75563.
30. Wei S, Wang Y. Comparison of corneal sensitivity between FS-
LASIK and femtosecond lenticule extraction (ReLEx ex) or
small-incision lenticule extraction (ReLEx smile) for myopic
eyes. Graefes Arch Clin Exp Ophthalmol 2013;251:164554.
31. Demirok A, Ozgurhan EB, Agca A, et al. Corneal sensation
after corneal refractive surgery with small incision lenticule
extraction. Optom Vis Sci 2013;90:10407.
32. Lee BH, McLaren JW, Erie JC, et al. Reinnervation in the cornea
after LASIK. Invest Ophthalmol Vis Sci 2002;43:36604.
33. Zhang F, Deng S, Guo N, et al. Confocal comparison of
corneal nerve regeneration and keratocyte reaction between
FS-LASIK, OUP-SBK, and conventional LASIK. Invest
Ophthalmol Vis Sci 2012;53:553644.
34. Li M, Niu L, Qin B, et al. Confocal comparison of corneal
reinnervation after small incision lenticule extraction (SMILE)
and femtosecond laser in situ keratomileusis (FS-LASIK).
PloS One 2013;8:e81435.
Footnotes and Financial Disclosures
Originally received: August 26, 2014.
Final revision: October 6, 2014.
Accepted: October 7, 2014.
Available online: November 22, 2014. Manuscript no. 2014-1378.
1
Quinze-Vingts National Ophthalmology Hospital, Paris, France.
2
French National Institute of Health and Medical Research, UPMC
University Paris, Institut de la Vision, CNRS, Paris, France.
3
Clinical Center for Investigations, Paris, France.
4
Espace Nouvelle Vision, Paris, France.
5
Ambroise Par Hospital, APHP, University of Versailles St-Quentin en
Yvelines, Versailles, France.
Financial Disclosure(s):
The author(s) have made the following disclosure(s):
A.D.: Consultant e Alcon, Thea, Abbott Medical Optics, Bausch & Lomb;
Grants/grants pending e INSERM and UPMC University Paris 6; Inventor e 2
patents belonging to INSERM.
676
35. Vestergaard AH, Grnbech KT, Grauslund J, et al. Sub-
basal nerve morphology, corneal sensation, and tear lm
evaluation after refractive femtosecond laser lenticule
extraction. Graefes Arch Clin Exp Ophthalmol 2013;251:
2591600.
36. Gallar J, Acosta MC, Moilanen JAO, et al. Recovery of
corneal sensitivity to mechanical and chemical stimulation
after laser in situ keratomileusis. J Refract Surg 2004;20:
22935.
37. Murakami Y, Manche EE. Prospective, randomized compari-
son of self-reported postoperative dry eye and visual uctua-
tion in LASIK and photorefractive keratectomy.
Ophthalmology 2012;119:22204.
38. Rodriguez AE, Rodriguez-Prats JL, Hamdi IM, et al. Com-
parison of goblet cell density after femtosecond laser and
mechanical microkeratome in LASIK. Invest Ophthalmol Vis
Sci 2007;48:25705.
39. Leonardi A, Tavolato M, Curnow SJ, et al. Cytokine and
chemokine levels in tears and in corneal broblast cultures
before and after excimer laser treatment. J Cataract Refract
Surg 2009;35:2407.
40. Ivarsen A, Laurberg T, Mller-Pedersen T. Role of keratocyte
loss on corneal wound repair after LASIK. Invest Ophthalmol
Vis Sci 2004;45:3499506.
41. Erie JC, Nau CB, McLaren JW, et al. Long-term keratocyte
decits in the corneal stroma after LASIK. Ophthalmology
2004;111:135661.
C.B.: Board member and consultant e Alcon, Allergan, MSD, Thea,
Santen; grants/grants pending e Allergan, MSD, Thea, Santen; Co-inventor
e 2 patents belong to INSERM.
Public grant from the French National Institute of Health and Medical
Research. The funding organization had no role in the design or conduct of
this research.
Abbreviations and Acronyms:
FLEX femtosecond lenticule extraction; IVCM in vivo confocal
microscope; OCT optical coherence tomography; OSDI Ocular
Surface Disease Index; SMILE small incision lenticule extraction;
TBUT tear lm breakup time.
Correspondence:
Alexandre Denoyer, MD, PhD, Centre Hospitalier National dOphtalmolo-
gie des Quinze-Vingts, 28 rue de Charenton, F-75012 Paris, France. E-mail:
alexandre.denoyer@gmail.com.