In the original keratomileusis procedure for myopia, an approximately 300 mm

thick disc was dissected from the anterior cornea in a freehand fashion and reshaped
using a cryolathe. In the late 1980s Ruiz developed an automated microkeratome that
controlled speed as it passed across the cornea, leading to more consistent results.
This procedure has become known as automated lamellar keratoplasty (ALK). In the
1990s the combination of a microkeratome and an excimer laser (for the refractive
cut) was developed by Pallikaris, Burato and others, further improving the
predictability of the refractive procedure. This procedure, known as laser assisted in
situ keratomileusis (LASIK) has gained wide acceptance wordwide1. The limitations
of this procedure have been already proven in long term 6 and 10 year follow-up
studies related to the induction of aberrations and regression2,3. Recently,
femtosecond lasers have become available for the cutting of the intrastromal lenticule
and subsequent lenticule extraction4-6. In the femtosecond-only technique for
correction of refractive errors (femtosecond lenticule extraction), the intrastromal
lenticule is removed using flap-like access4,7. In small-incision lenticule extraction
(SMILE), the incision is minimized and the procedure does not use a flap4,5. Both
types of procedures seem safe and promising for corneal refractive correction of
myopia, although small-incision lenticule extraction is not yet widely performed

Surgical Procedure

Patients were administered a mild oral anxiolytic and sedative (diazepam,

Valium® 5 mg; Roche Farma S.A. Barcelona, Spain) 30 minutes before surgery.
Topical anaesthesia was given (oxibuprocaine hydrochloride 0.4%, Prescain®;
Laboratorios Llorens S.A., Barcelona, Spain) and after a povidone–iodine
(Betadine®) scrub of the skin and eyelids, the patients were draped with a sterile head
towel, their eyelashes were taped with sterile tape and they were positioned on the
ergonomic pivoting patient table of the laser unit.
 The SMILE femtosecond procedure was initiated with the application of an
eyelid speculum to keep the eye open; the patient’s eye was positioned under a curved
contact glass interface. A completely clean and diffusely wet cornea was ensured
using a wet microsponge. This contact glass is similar to a gonioscopic lens in that it
possesses a curved surface designed to couple with the cornea with only minimal
applanation force. Before coupling, the VisuMax® femtosecond laser system (Carl
Zeiss Meditec AG, Jena, Germany) calibrates the contact glass. The eye’s keratometry
data is entered into the VisuMax® to account for the difference between the relaxed
cornea and the contact glass curvature.

The first femtosecond laser used was version 2.7.3 with maximum spot spacing
of 3.0 µm.

A fully device-integrated suction system ensures that low suction is only applied
to the cornea during the actual laser treatment. Centration is assisted using several
features including an internal fixation target for the patient which is designed to
attract their attention, self-adjustment of the eye during docking to the contact glass,
an easy to find line of sight for treatment positioning, and automated adaptation to
patient eye refraction. Flap parameters that can be adjusted include flap thickness, flap
diameter, hinge width, side-cut angle and hinge location.

Using this procedure, once contact is made between cornea and the contact glass,
the patient is able to see the flashing fixation target in clear focus. The patient’s
cooperation is fundamental for properly aligned treatment. When full contact glass
application is achieved, suction is applied, the eye immobilized, and the laser
activated by the surgeon by pressing on a foot pedal. A very accurately focused laser
beam is guided through to the cornea, whereby the laser beam moves across and
through the cornea in a spiral manner, creating a layer of very tiny bubbles under its
path. These bubbles quickly disappear, and the tissue above the bubbles becomes a
corneal lenticule that can be easily lifted by the surgeon.
 In the SMILE procedure, instead of making a complete flap side cut, only a small
incision is created, and the flap is never lifted. Instead, the lenticule is extracted from
within the cornea through a small incision.

In these cases, the duration of the entire procedure was relatively consistent, from
50 to 55 seconds, regardless of the refractive error to be corrected. This is an
advantage in terms of postoperative assessment of results among different correction
groups. The shape of the lenticule generated was designed to correct for refractive
errors. In all cases in this particular preliminary study group, the anterior surface of
the lenticule was 100 μm deep and the maximum diameter of the first cleavage plane,
6.0 mm. For technical reasons, astigmatism corrections result in an oval posterior
surface of the lenticule.

For all myopic corrections, the optical zone size was 6.0 mm. The spot-and-track
spacing for the cleavage plane, which defines the posterior surface of the lenticule,
was slightly higher than the spot-and-track spacing for the cleavage plane, which
defines the anterior surface of the lenticule (i.e. the cap). The posterior part of the
lenticule was created by laser scanning in spirals from the centre of the pupil to the
periphery of the optical zone. The anterior part of the lenticule was created by laser
scanning in spirals from the periphery to the centre of the pupil. All passes and
spot-and-track distance changes were automatically performed by the laser software
and hardware with no user intervention (Figure 1).
 Our standard postoperative treatment consisted of antibiotic and steroid drops
twice a day during one week which was the slowly tapered over a four week period,
with preservative-free artificial tears five times a day for almost a month.

Complications

Intra-operative complications

Early evidence suggests that the SMILE procedure is safe. Intra-operative

complications may occur; however, these are not likely to be sight threatening. Suction
loss can occur albeit the literature suggests an incidence of between 0.9 to 4.4 per cent
of patients. If suction loss occurs prior to the laser ablation, the patient can be
re-engaged and the procedure can continue immediately. If suction is lost once the laser
has be-gun, the post-operative course will be determined by the amount of laser
completed.

Early suction loss may represent an opportunity to re-engage and complete

with-out complication, whereas suction loss at the middle of the surgery may require
the procedure to be aborted in favour of later surface ablation. Epithelial disturbances
may occur due to the docking procedure or by the man-ual removal of the lenticule
through the laser-created channel. A bandage contact lens is rarely required in our
experience; however, it may aid the post-operative comfort and healing process for
some patients.

The difficulty of extracting the lenticule may vary depending on the laser energy
settings. This is usually resolved quickly following installation. It is possible that the
lenticule may be partially torn during the procedure. Further at-tempts to remove the
remaining lenticule should be undertaken. Dong and Zhou43 re-ported a case of
irregular astigmatism following the SMILE procedure, possibly related to incomplete
dissection and subsequent residual lenticule left within the pocket.

Post-operative complications

The absence of a corneal flap, as in the LA-SIK procedure, removes the risk of
related concerns such as flap displacement or striae. Otherwise, the SMILE procedure
maintains a similar risk profile following surgery with al-ternate laser refractive
procedures. Reported post-operative complications include; epithe-lial ingrowth,
diffuse lamellar keratitis, cor-neal haze, irregular astigmatism, increased

aberrations, ectasia, punctate corneal staining and mild dry eye.

EPITHELIAL INGROWTH
 Ingrowth may be caused by epithelial cells displaced at the time of the procedure or
through cells invading the SMILE incision site. As per LASIK surgery, no treatment is
of-ten required for mild cases. The risk for the progression of irregular astigmatism or
cor-neal melt increases significantly, if the in-growth is considered severe. Surgical
intervention is warranted in these cases.

The risk of epithelial ingrowth in LASIK procedures has been estimated to range
from 0.01 to 6.1 per cent. The relative potential risk for ingrowth in SMILE cases
should be less as the entry point for epithelial cells is negligible compared to the full
LASIK flap. Further-more, minimal washout is typically required, thereby diminishing
the risk of cells or debris being introduced. A recent review of existing literature on the
safety and efficacy of SMILE revealed an accumulative risk of epithelial in-growth of
0.5 per cent.37 Ivarsen, Asp and Hjortdal2 describe 10 eyes (out of 1,500) with islands of
epithelial cells seen near the inci-sion site. No progression was seen over three months
and in 5 cases the cells spontaneously disappeared by one year. Our current experi-ence
reflects this with only one case requiring secondary treatment. The patient was returned
to surgery and the cells removed. Vi-sual acuity was 6/6 at day 1 with no pain or
photophobic symptoms.

DIFFUSE LAMELLAR KERATITIS

Diffuse lamellar keratitis (DLK) was first de-scribed in 1998 in conjunction with
the LASIK procedure. Representing an acute inflamma-tory response, the patient
usually presents with increasing discomfort, photophobia and a reduction in visual
acuity. Biomicroscopic signs have been described,50 briefly small white, granular cells
may be seen under slitlamp investigation increasing with the se-verity of the condition.
Severe cases may lead to stromal necrosis without further treatment. Typically, cases
present during the immediate post-operative period. A definitive cause of diffuse
lamellar keratitis remains unclear; however, the condition has variously been
at-tributed to the introduction of bacteria, surgi-cal debris, meibomian gland secretions
and immune infiltrates.50 The existence of epithe-lial defects has been shown to
significantly in-crease the risk of diffuse lamellar keratitis.

The incidence of diffuse lamellar keratitis following SMILE has been reported as
be-tween 0.04 and 1.6 per cent.37,50 Of particular interest, Zhao et al50 found that diffuse
lamel-lar keratitis was associated with larger lenti-cule diameter and thinner lenticules.
The authors proposed that gas bubble accumula-tion in cases of thin lenticules may
induce a greater inflammatory response given the an-terior and posterior poles may
accordingly be under more significant compressive force.

Alternatively the thinner lenticules may repre-sent a greater technical challenge

increasing the intra-operative manipulation and thereby possible inflammatory
response. Zhao et al50 suggest that the proximity to the limbus and limbal vasculature
of the larger diameter len-ticules may further give rise to inflammatory cells. Further
studies are required to eluci-date the causes. It should be noted that all cases resolved
without sequellae following treatment with topical corticosteroids.

CORNEAL HAZE AND OEDEMA

Up to eight per cent of patients in a study by Ivarsen, Asp and Hjortdal2 were
found to have trace haze at the initial post-operative visit. This did not impact upon
visual acuity and subsequently subsided. Yao et al51 de-scribe micro-distortions in
Bowman’s layer of up to 88.5 per cent of SMILE cases. This was in comparison to
only 42.1 per cent per cent of LASIK comparable cases. These micro-dis-tortions
were diagnosed through optical co-herence tomography (OCT). There was no
additional evidence of corneal striae under biomicroscopy and the micro-distortions
ap-peared to stabilise at 1 week suggesting mini-mal impact upon visual acuity or
quality. Similarly Moshifar et al37 reported only 0.2 per cent of patients in the
combined review with mild corneal oedema. In summary, cor-neal distortion does not
appear to represent a significant clinical issue in SMILE patients.
 Miao et al52 describe the optical quality and intraocular scattering after the
SMILE proce-dure. The authors found that SMILE had lit-tle impact upon the retinal
image and further that the objective scatter index quickly returned to normal
following a tem-porary increase at three weeks. As corneal haze and anterior
keratocyte loss have been indicated as the main cause of a decrease in both parameters
following excimer laser pro-cedures, these results would appear to con-firm the lack
of corneal distortion in SMILE patients. Riau et al53 describe a statistically
sig-nificant increase in inflammatory cells in pa-tients undergoing LASIK as
compared to lenticule extraction. All patients observed had moderate to high myopia
(-6.00 to -9.00 D) suggesting a possible further benefit for the SMILE procedure in
reducing the inci-dence of corneal inflammation post-surgery.

IRREGULAR ASTIGMATISM AND INCREASED HIGHER-ORDER

ABERRATIONS

Irregular astigmatism may result in ghost im-ages or loss of visual acuity. Ivarsen,
Asp and Hjortdal2 describe six eyes of five patients with monocular ghosting
following surgery. In all cases, corneal topography showed irreg-ular astigmatism.
Interestingly, only one eye could be directly attributed to a peri-opera-tive
complication (suction loss and attempted re-treatment). All cases showed irregular
Bowman’s layer with compensatory epithelial hyperplasia on OCT.48 Due to the
epithelial hyperplasia, topography-guided surface abla-tion was undertaken in these
patients with a reduction in visual complaints and subse-quent improvement in visual
acuity in most cases. Two eyes developed grade 2 corneal haze following surface
treatment resulting in worsening of symptoms. Both cases did not use mitomycin C
during surgery suggesting this may be beneficial in these cases.

Li et al54 describe decentration following the SMILE procedure in a cohort of

100 pa-tients. The mean decentred displacement was 0.17 ± 0.09 mm. Although 99
per cent of patients had post-operative uncorrected vi-sion of 6/6 or better, fifteen
eyes lost one line of visual acuity and two eyes lost two lines at the final
post-operative visit. Vertical coma showed the greatest increase in magnitude possibly
contributing to the decrease in visual acuity.54

ECTASIA

Ectasia remains a significant concern for re-fractive surgeons.55 As discussed

previously, one of the proposed benefits of SMILE over LASIK in particular is the
minimisation of biomechanical changes following surgery. This would suggest a
reduction in the risk for ectatic changes following the surgical pro-cedure. Through
mid-2015, only one case of ectasia related to a SMILE procedure has been reported in
the peer-reviewed literature. In this case report, the patient appeared to have
pre-existing keratoconus.36 Although this is likely to correspond with the proposed
biome-chanical benefits of SMILE, it is likely also a consideration of our greater
understanding of potential risk factors for ectasia. Longer-term data, with
corresponding data sets, will provide confirmation of the potential benefits of SMILE
over LASIK surgery in particular.

INFECTIOUS KERATITIS

Infectious keratitis remains a potentially dev-astating complication of corneal

refractive surgery. Fortunately, the incidence is rare with published rates as low as
0.011 per cent in LASIK cases and 0.066 per cent following surface ablation.56
Patients may complain of significant loss of vision accompanied by pain, photophobia
and epiphora. The condi-tion usually presents in the early post-opera-tive period and
may appear similar to diffuse lamellar keratitis or epithelial in-growth.
Biomicroscopic signs should distin-guish keratitis from these conditions. Initial
treatment is with broad-spectrum antibiotics and amended depending on the resolution
of signs.
 Ivarsen, Asp and Hjortdal2 noted five eyes with interface infiltrates and ocular
irritation following SMILE (0.33 per cent). Samples were taken for microbiological
investigation however no specific antigens could be identi-fied. Patients were treated
topically in four cases with the further case requiring the in-terface to be flushed with
cefuroxime. All eyes cleared without sequellae by three months. There are no further
reports in the current literature.

DRY EYE

The majority of literature relates the pres-ence of significant dry eye symptoms
with the transection of corneal nerves during sur-gery.57 This may lead to the
suppression of tear secretion from the lacrimal gland, mucin expression on the corneal
epithelium and blink rate changes, which impact upon the corneal surface.57 Further
risk factors for post-surgery dry eye, include pre-existing dry eye disease, deeper laser
ablations, LASIK flap size, thickness and position albeit evi-dence remains
variable.58–60

As described earlier, results suggest that SMILE patients recover more quickly
and have less significant symptoms than patients undergoing the LASIK procedure.
These findings have clinical significance. Li et al30 describe significantly less corneal
fluorescein staining on slit lamp investigation; however, no significant differences in
Schirmer test re-sults were found between corresponding SMILE and LASIK groups.
Subjective patient satisfaction remains increased in SMILE pa-tients in comparison to
LASIK groups.30,31 Anecdotally, this is what we have experienced in our clinical
practice, namely, despite the longer visual recovery of SMILE patients com-pared to
LASIK, the patients themselves re-port being more comfortable and happier during
the post-operative recovery period.