SURVEY OF OPHTHALMOLOGY VOLUME 49 • NUMBER 6 • NOVEMBER–DECEMBER 2004

DIAGNOSTIC AND SURGICAL

TECHNIQUES
MARCO ZARBIN AND DAVID CHU, EDITORS

Evolution, Techniques, Clinical Outcomes, and

Pathophysiology of LASEK: Review of the Literature
Suphi Taneri, MD,1,2,3 James D. Zieske, PhD,3 and Dimitri T. Azar, MD2,3

1
Zentrum für Refraktive Chirurgie, Münster, Germany; 2Cornea and Refractive Surgery Service, Massachusetts Eye and
Ear Infirmary, Boston, Massachusetts; and 3Schepens Eye Research Institute, Harvard Medical School, Boston,
Massachusetts, USA

Abstract. Laser subepithelial keratomileusis (LASEK) is a relatively new laser surgical procedure that
combines certain elements of both laser in situ keratomileusis (LASIK) and photorefractive keratectomy
(PRK) to improve the risk/benefit ratio. Diluted alcohol is used to loosen the epithelial adhesion to
the corneal stroma. The loosened epithelium is moved aside from the treatment zone as a hinged
sheet. Laser ablation of the subepithelial stroma is performed before the epithelial sheet is returned to
its original position. We reviewed the literature regarding modifications of the technique, indications,
outcomes, and complications, as well as wound healing after LASEK surgery. This literature review of
1,421 LASEK-treated eyes provided many findings: 1) The long-term stable results in the absence
of serious complications, such as infections, recurrent erosions, scars, or late-onset corneal haze
formation in patients re-examined up to 5 years after LASEK; 2) Epithelial closure with recovery
of functional vision was completed at days 4 to 7 in most cases; 3) A tendency toward overcorrection
with PRK nomograms; 4) We hypothesize that this tendency may be due to the decreased wound
healing response, which may lead to myopic regression in PRK; and 5) Postoperative discomfort
and prolonged visual recovery until the epithelium closes remain the biggest disadvantages of LASEK
compared to LASIK. LASEK surgery is especially valuable in patients with thin corneas who would not
qualify for LASIK surgery. However, a potential superiority of LASEK to LASIK in wavefront guided
ablations still remains speculative. (Surv Ophthalmol 49:576–602, 2004. 쑖 2004 Elsevier Inc. All
rights reserved.)

Key words. laser subepithelial keratomileusis (LASEK) • laser in situ keratomileusis

(LASIK) • photorefractive keratectomy (PRK)

Introduction correction of refractive error, combines certain el-

ements of both laser in situ keratomileusis (LASIK)
DEFINITION AND TERMINOLOGY and photorefractive keratectomy (PRK).
Laser subepithelial keratomileusis (LASEK),2,3,6,8,9, Instead of completely removing the epithelium as
13,14,22,24,30,49,50,68,71,75,76,78,82,84,115,117–119,123,126,130,151
a with PRK, dilute alcohol is used to loosen the epithe-
relatively new laser surgical procedure for the lial adhesion to the corneal stroma. The loosened

576
쑖 2004 by Elsevier Inc. 0039-6257/04/$–see front matter
All rights reserved. doi:10.1016/j.survophthal.2004.08.003
EVOLUTION AND PATHOPHYSIOLOGY OF LASEK 577

epithelium is then moved aside from the treatment concepts of keratectomy, keratotomy, and thermo-
zone as a hinged sheet. Laser ablation of the subepi- keratoplasty.73
thelial stroma is performed before the epithelial The ideal refractive procedure would be simple,
sheet is returned to its original position, as with the effective, minimally invasive, safe, and applicable in
LASIK flap. all patients desiring vision correction. It would not
The first LASEK procedure was performed at the regress but be adjustable as visual needs of the pa-
Massachusetts Eye and Ear Infirmary in 1996 by one of tients change over time and yet be reversible, and also
us (DTA).9 Nevertheless, Camellin popularized the allow for instant visual recovery without discomfort.
procedure and coined the term LASEK for laser epi- Refractive power at the central cornea is about ⫹43
thelial keratomileusis (Cimberle M, Camellin M: D, and is the sum of refractive power at the air–tear
LASEK technique promising after 1 year of experi- (⫹44 D), tear–cornea (⫹5 D), and cornea–aqueous
ence. Ocul Surg News 14:14–7, 2000; Condon P, humor (–6 D) interfaces. Thus the cornea provides
Camellin M: LASEK may offer the advantages of both about two-thirds of the total refractive power of the
LASIK and PRK. Ocular Surgery News International eye.96 In addition, the cornea is easily accessible with-
Edition, 1999; Cimberle M, Condon M: LASEK per- out the inherent risks of intraocular surgery and
forms better than LASIK in selected cases. Ocul Surg without compromising accommodation. Therefore it
News, 2002). Alternative expressions include laser seems reasonable to look for a vision correction pro-
subepithelial keratomileusis,9 and subepithelial pho- cedure for changing the anterior corneal curvature.
torefractive keratectomy,71,117 epithelial flap photore- Contemporary corneal refractive surgery modifies
fractive keratectomy,118 laser-assisted subepithelial corneal curvature, including ablative (PRK, LASIK,
keratectomy,82,119 excimer laser subepithelial abla- LASEK), additive (intracorneal ring segments,104 cor-
tion,84 and Epi-LASEK.6 neal inlays), incisional (astigmatic keratotomy, radial
keratotomy), and thermal (laser thermokeratoplasty,
THEORETICAL ADVANTAGES OF LASEK
conductive keratoplasty) methods. In our review of
LASEK we will focus on PRK and LASIK for compari-
The main rationale for combining elements of son as the other methods are either limited in their
LASIK and PRK to LASEK is to avoid the flap-related indications, yield no long-term stable results, or are
LASIK complications and the slow visual recovery still experimental and therefore not widely used
and haze risk of PRK. (Table 1).
LASEK may avoid several of the inherent complica- Photorefractive keratectomy treats ametropia by
tions, including free caps, incomplete pass of the employing a 193 nm argon fluoride excimer laser to
microkeratome, flap wrinkles, epithelial ingrowth, reshape the anterior corneal stroma by photoabla-
flap melt, interface debris, corneal ectasia, and dif- tion after mechanically removing the epithelium.2
fuse lamellar keratitis, after LASIK4,10,16,26,28,40,46, Postoperative pain is explained by this absence of the
52,55,69,91,101,121,129,135–137,140,147
and postoperative pain, epithelial layer. The concept of ablative surgery is
subepithelial haze, and slow visual rehabilitation after that by removing small amounts of tissue from the
PRK.1,25,26,29,67,113,114,124,128,144,145 anterior surface of the cornea, a significant change of
No refractive surgery has been established as ideal refraction can be attained. The process of excimer
for all patients. The choice among the three major laser photoablation at the corneal surface is highly
procedures—LASIK, PRK, and LASEK—is at the dis- predictable, accurate, and reproducible. However,
cretion of the surgeon. the outcome is, among other factors, influenced by
Current ophthalmic literature does not provide the comparatively strong wound-healing response
the specific indications, visual outcomes, complica- as the central Bowman’s layer is ablated. At times,
tions, and limitations of LASEK. This review gath- this wound-healing response leads to a regression
ers the data to evaluate the potential benefits and of the desired effect over time or to subepithelial
risks of LASEK and the visual outcome in a semi- haze formation. PRK has undergone extensive inves-
quantitative fashion. tigation and has become a widely performed proce-
dure. From the huge database of treated patients
currently available, there is evidence that over 90%
Background: Corneal Refractive Surgery
of myopic corrections of up to ⫺6.00 D have a final
Ametropic people have sought ways to improve outcome of within ⫾ 1.00 D of the targeted refrac-
their vision and rid themselves of eyeglasses for centu- tion. The higher the intended corrections, the less
ries, and surgical attempts at reducing ametropia predictable the result becomes, accompanied by an
have challenged some of the great minds of history. increase in complications such as postoperative
In 1898, Lans, of Leiden, the Netherlands, published pain, subepithelial haze,17 and slow visual rehabilita-
the results of experiments with rabbits in treating tion. Symptoms of dry eye and recurrent erosion syn-
astigmatism and he described many of the current drome are more frequent and severe in PRK than after
578 Surv Ophthalmol 49 (6) November–December 2004 TANERI ET AL

TABLE 1
Widely Accepted Relative Differences between PRK, LASEK, and LASIK*
Factor PRK LASEK LASIK
Range of correction Low to moderately high Low to moderately high Low to moderately high

Postoperative pain Moderate 24–48 hrs Mild to moderate Minimal 12 hrs

24–48 hrs in ∼50%
Postoperative 3 weeks to several months 3 weeks to several months 1–2 weeks
medications

Functional 3–7 days 3–7 days ⬍24 hrs

vision recovery
Refractive 3 weeks to several months 3 weeks to several months 1–6 weeks
stability achieved

Specific Haze formation, scarring Haze formation, scarring Free caps, incomplete pass
complications of the microkeratome, flap
wrinkles, epithelial ingrowth,
flap melt, interface debris,
corneal ectasia, and diffuse
lamellar keratitis

Risk of scarring 1–2% Possibly less than PRK ⬍1%

Dry eye sensitive 1–4 weeks or longer 1 to 4 weeks or longer Could last up to 12 months
or more

Thin corneas Often not contraindicated Often not contraindicated May be contraindicated
or wide pupils depending on amount
of intended correction
Special (relative) — Thin corneal pachymetry Concern about postoperative
indications Wide scotopic pupil pain
LASIK complications in Requirement of rapid
fellow eye visual recovery
Predisposition to trauma Retreatment after
Keratoconus suspects incisional surgery
(irregular astigmatism) or PRK/LASEK
Glaucoma suspects
Recurrent erosion syndrome
Dry eye syndrome
Basement membrane
disease
Special (relative) Concern about Concern about postoperative pain Thin corneas
contraindications postoperative pain Requirement of rapid Wide pupils
Predisposition visual recovery Recurrent erosion syndrome
to haze formation Glaucoma Scleral buckle
Deep-set eyes
Small palpebral fissure
*
Additional factors such as surgeon experience, type of laser, age of patient, amount of correction, and administrative
regulations of various countries may influence these comparisons

LASIK, according to Hovanesian et al,58 and there- and PRK.7 However, Lee et al found the decrease in
fore they recommend LASIK in order to prevent tear production and severity of dry eye was greater
recurrent erosions. Ang et al found the decreased in LASIK than PRK.77
corneal sensation to be the main cause inducing or In LASIK, a hinged corneal flap of approximately
exacerbating dry eye after laser refractive surgery, 80 to 200 µm thickness is created using a microkera-
lasting from a few weeks up to 1 year, but they did tome. To facilitate cutting a uniform and regular flap
not describe significant differences between LASIK and a stromal bed of the appropriate diameter, the
EVOLUTION AND PATHOPHYSIOLOGY OF LASEK 579

IOP must exceed 65 mm Hg, which may be inadvis- leave variable amounts of epithelium.19,51,122 Chemi-
able in patients with glaucoma (Cimberle M, Condon cal agents such as 0.5% proparacaine, iodine, co-
M: LASEK performs better than LASIK in selected caine, alkali n-heptanol, and ethanol have been used
cases. Ocul Surg News, 2002),83 or it may be impossi- to remove the corneal epithelium in experimental
ble after scleral buckling surgery. The creation of studies.20,57 Today, 18–20% ethanol is commonly uti-
the flap remains the most unpredictable part of the lized in LASEK, but there are only a few experimental
procedure as the actual flap thickness may vary con- evaluations of its use. The use of 100% ethanol for
siderably from the intended one.39 It is currently 2 minutes on rabbit corneas leads to a significant
accepted that the residual thickness in the stromal decrease in stromal keratocytes after 24 hours.20 Simi-
bed after surgery should be at least 250 µm to avoid larly, when using 70% isopropyl alcohol for 2 minutes
iatrogenic keratoectasia.98,116 In addition, it has been for epithelium removal in rabbit eyes, Agrawal et
suggested to leave a certain percentage (perhaps 50– al found an increased inflammatory response and
55%) of intact posterior stroma in high myopes86 or damaging effect on keratocytes.5 Helena et al ob-
take into account patient age and attempted correc- served increased keratocyte loss but decreased in-
tion.98 The limitation of the residual stromal bed with flammation after using 50% ethanol for 1 minute
LASIK may not allow creation of optical zones wide compared to mechanical debridement.54
enough to correspond to the patient’s scotopic pupil Abad et al showed that alcohol-assisted epithelial
diameter due to the flap thickness, which is not removal was a simple and safe alternative to mechani-
available for ablation. For the same reason, at times cal epithelial removal before PRK. 2 Applying 25%
undercorrection instead of full correction must be ethanol for 3 minutes, Stein et al were able to grasp,
planned in LASIK. In patients with corneal pa- lift, pull apart, and split the corneal epithelium using
chymetry of 500 µm or lower and in patients with two McPherson forceps.123 Similarly, Shah et al ex-
asymmetric corneal curvature, the use of the Azar- posed the epithelium using a dry sponge.117 These
Lu MEEI keratoconus classification91 is helpful in early reports revealed that epithelial removal using
identifying patients with keratoconus or forme fruste 18–25% alcohol for 20–25 seconds was fast, easy, and
keratoconus in whom corneal ablation surgery safe compared to mechanical debridement. They also
should be avoided. Even in keratoconus suspects, showed that this concentration can produce sharp
we do not recommend performing LASIK. Although wound edges and clean, smooth Bowman’s layer, and
technically more challenging than PRK, LASIK is the that the central epithelium can be translocated in
most popular refractive procedure as functional part or completely.3 Carones et al found significantly
vision usually returns within 1 day, there is mini- better results in terms of haze and corneal regularity in
mal pain, and the risk of visually affecting corneal epithelial debridement using a 20% alcohol solu-
haze formation is lower than in PRK.21,118,119 How- tion compared to mechanical debridement.21
ever, dissecting corneal nerves by creating the flap The dilution of alcohol in balanced salt solution
may also increase a dry eye syndrome58 until the (BSS), physiologic solution, or sterile water, thus ob-
nerves regenerate after up to 12 months. In addition, taining different osmolarities, is an area of active
flap-related complications,134,140 such as free caps, debate but none of the LASEK studies has shown a
incomplete pass of the microkeratome, flap wrin- definite advantage of a specific formulation. Cam-
kles, epithelial ingrowth,133 flap melt, interface ellin strongly points out the importance of a hypo-
debris, and diffuse lamellar keratitis,18,64,80,81,95,148,150 tonic solution obtained by diluting alcohol in
may be seriously threatening to sight. LASIK flap distilled water for facilitating epithelial detachment
complications tend to occur more frequently when (Cimberle M, Camellin M: LASEK technique promis-
the preoperative anterior corneal curvature is less ing after 1 year of experience. Ocul Surg News 14:
than 41 D or greater than 46 D.140 Melki et al reported 14–7, 2000), whereas Vinciguerra uses BSS for dilu-
four cases of late traumatic LASIK flap dislocation tion.130 We found that both ways work equally well
that could be treated by refloating and scraping of the (unpublished data).
stromal bed without loss of best corrected visual acuity
within 12 hours of the injury.92 This long-term risk
of flap dislocation should be considered in patients EFFECT OF ALCOHOL ON EPITHELIAL CELL
predisposed to ocular trauma, like military personnel SURVIVAL IN VITRO
or contact sports athletes. Cultured epithelial cells of animal34,97 and
human32,33,35,66 corneal and conjunctival32,33,35 ori-
gins have been used to reduce the number of animals
Alcohol-assisted Epithelial Removal needed in the preclinical evaluation of ocular toxicity
Manual epithelial debridement produced scratch- of various substrates. Monolayered cultures of cor-
es and nicking in Bowman’s layer and was found to neal epithelial cells have been shown to be equally
580 Surv Ophthalmol 49 (6) November–December 2004 TANERI ET AL

sensitive as three-dimensional corneal constructs for basement membrane or between the basement mem-
evaluation of acute toxicity.100 brane and Bowman’s layer.12,123 Additionally, the ad-
Fluorescent viability staining provides information herence of the basement membrane to the basal layer
regarding the functional alterations of cell mem- of the epithelium is significant because it is believed
brane integrity and cell metabolism. It has previously that the basement membrane provides the stability
been used to evaluate the viability of corneal donor and support that keeps the epithelium intact even
tissue,90 the in situ organization of keratocytes,102,103 after manipulation.12 The preservation of the hem-
apoptosis of normal surface corneal epithelial idesmosomes in the basal epithelial layer provides a
cells,106,138 Fas/Fas ligand-stimulated keratocytes,141 structure that may promote the adhesion of a viable
toxicity of natural tear substitutes,48 and surfactant- epithelium to the ablated stroma.12
induced cell death.61,62 Kim et al evaluated the toxicity of 20% ethanol
We detected TdT-mediated dUTP nick-end label- exposure for 30 seconds, 1 minute, and 3 minutes
ing (TUNEL)-positive cells after 8 and 24 hours of on rabbit corneal epithelium with scanning and
incubation after 20 and 40 seconds exposure to 20% transmission electron microscopy.68 They found
alcohol.22 Most of the nuclei in TUNEL-positive cells widespread partial or total damage of microvilli, focal
were round without cellular condensation. The breaks of intercellular junction, and cellular edema.
TUNEL-positive cells with condensed nuclei showed The damage increased with exposure time. After 1
the characteristic morphology of apoptosis. The con- minute of ethanol exposure, slough of superficial
densed nuclei were referred to as apoptotic bodies epithelium was observed, which progressed with
by light or electron microscopy.22 time.
Our in vitro studies also suggested a dose- and We examined several corneal epithelium speci-
time-dependent effect of alcohol on epithelial cells. mens from patients who underwent alcohol-assisted
The 25% concentration of alcohol was the inflection epithelial removal prior to PRK.8,9,22 Electron micros-
point of epithelial survival. Significant increase in copy showed variability in the histological appear-
cellular death occurred after 35 seconds of alcohol ance of the basement membrane zone. Most
specimens showed intact epithelial cell layers (Fig. 1).
exposure; 40 seconds of exposure further induced
Edematous cells and abnormal vacuoles were ob-
apoptosis after 8 hours of incubation. These findings
served in other specimens. Higher magnification re-
are consistent with the clinical observations of vari-
vealed variable configurations of the epithelial
able epithelial attachment to the stromal bed on the
basement membranes, including normal areas and
first postoperative days after LASEK surgery, and also
areas of discontinuities and irregularities in the base-
with the recent report of 50% ethanol in the treat-
ment membrane. Basement membrane fragments
ment of progressive or recurrent epithelial ingrowth
were still attached to the epithelial basal cells in most
after LASIK.68
specimens. Bowman’s layer and corneal stroma were
The in vitro mono-layered results may apply to in absent indicating that the epithelial sheets separate
vivo multi-layered epithelium. The critical alcohol from Bowman’s layer with variable amounts of basal
concentration and its exposure duration are thus laminae attached to the basal epithelial cell layer. The
frequently exceeded during surgery. Increased dura- ultrastructure of desmosomes and hemidesmosomes
tion of alcohol application can be used intentionally were normal in most specimens.8,22
to weaken the epithelial adhesions, which contri-
butes to the variability in alcohol-induced toxicity
that is observed in vivo.
Surgical Techniques
The four major surgical techniques are variations
of PRK (Fig. 2).
TRANSMISSION ELECTRON MICROSCOPY OF
CORNEAL EPITHELIUM SPECIMEN AZAR FLAP TECHNIQUE
To study the effect of alcohol exposure and me- Our LASEK technique (Fig. 2A), evolving from PRK
chanical manipulation on corneal epithelium, we car- after alcohol-assisted epithelial removal, is described
ried out electron microscopy studies on specimens elsewhere.8,9,13 Briefly, all our LASEK patients receive
obtained after conventional alcohol-assisted PRK. a preoperative evaluation consisting of uncorrected
The images revealed that the epithelial cell layer is visual acuity (UCVA), best spectacle-corrected visual
intact and the epithelial cells are still viable immedi- acuity (BSCVA), manifest and cycloplegic refraction,
ately after exposure to alcohol and surgical peeling. ocular dominance, keratometry, tonometry, pa-
The presence of the basement membrane attached chymetry, slit-lamp examination, dilated fundus ex-
to the basal epithelial cell layer indicates that the amination, and computerized videokeratography.
point of separation was likely to be within the Infrared pupillometry has also been performed after
EVOLUTION AND PATHOPHYSIOLOGY OF LASEK 581

then peeled back as a single hinged sheet using a

dry Merocel sponge (Fig. 3). More recently, we have
used one arm of a modified Vannas scissors (ASICO,
Westmont, IL) to delineate the epithelial margin
and fashion a hinged epithelial flap (Fig. 4). The
modified Vannas scissors also allow for creative varia-
tions of the LASEK incision to be customized for
different corneal types (Fig. 5).
After pushing aside the epithelial flap, the underly-
ing stromal bed is ablated with either a VISX Star S4
laser (VISX Inc., Santa Clara, CA), an Alcon LADAR-
Vision 4000 laser (Alcon Surgical, Orlando, FL), or a
Technolas 217A laser (Bausch & Lomb, Rochester,
NY). A Summit Apex SVS excimer laser (Summit,
Inc., Waltham, MA) or a VISX Star S2 laser were used
in the early treatments. After ablation, an anterior
chamber cannula is used to hydrate the stroma and
float the epithelial flap over a layer of balanced salt
solution. The epithelial flap is then replaced under
intermittent irrigation and careful attention to re-
alignment of the epithelial flap margins using the
previous marks. The epithelial flap is then allowed
to dry for 2–5 minutes. Topical steroids and antibiotic
medications are applied, and a bandage contact lens
(Soflens 66; Bausch & Lomb, Rochester, NY) is placed.
Fig. 1. (A) Electron microscopy showing intact epithelial The bandage contact lens is removed after complete
cells with intact hemidesmosomes (arrowheads) and base-
ment membrane (arrows). (B) Higher magnification of re-epithelialization (generally postoperative day 3 or
another specimen reveals discontinuities and irregularities 4); early removal or manipulation of the contact lens
in the basement membrane. Basement membrane frag- prior to postoperative day 3 risks peeling the epithe-
ments were still attached to the epithelial basal cells lial flap with the contact lens. Oral analgesics are
(arrows). The ultrastructure of hemidesmosomes (arrow-
heads) is preserved. Bar ⫽ 1 µm. prescribed to be taken every 4 hours as needed.
From our experience, we have learned that our
technique may not require specialized instruments
1999. Patients are counseled about potential risks, but does necessitate several key steps for consistent
benefits, and alternatives of the procedures. After epithelial flap creation and replacement. Pre-
application of topical 0.05% proparacaine (Oph- treatment with several drops of 4% tetracaine prior to
thetic; Allergan, Inc., Irvine, CA) and 4% tetracaine alcohol exposure helps to loosen the epithelium and
(formulated in the Massachusetts Eye and Ear Infir- lessen intraoperative discomfort. Placement of over-
mary pharmacy), a lid speculum is applied. The lapping corneal marks is crucial in ensuring correct
cornea is then marked with overlapping 3 mm circles
epithelial alignment and avoiding irregular epithelial
around the corneal periphery, simulating a floral pat-
placement and mismatch. We use an alcohol dis-
tern. An alcohol dispenser consisting of a customized
penser, but any optical zone marker with a barrel
7- or 9-mm semi-sharp marker (ASICO, Westmont,
could be used to expose the epithelium to alcohol and
IL) attached to a hollow metal handle serves as the
reservoir for 18% alcohol. After 25–30 seconds, the avoid spillage. A modified Vannas scissors or a jewel-
ethanol is absorbed using an aspiration hole followed er’s forceps to delineate the wound edge and locate
by dry sponges (Weck-cel or Merocel). If necessary, the dissection plane and a dry, non-fragmenting, cel-
the ethanol application may be repeated for an addi- lulose sponge to peel the epithelial sheet are easily
tional 10–15 seconds. available instruments for creating the flap. The flap
Over time the step of epithelial removal has been can be repositioned with an irrigating cannula under
modified. Initially one arm of a jeweler’s forceps was intermittent hydration using the preplaced corneal
inserted under the epithelium and traced around to marks as a guide. No overlap of the flap and wound
delineate the epithelial margin, leaving a hinge of edge has been observed that would have been attrib-
2–3 clock hours of intact margin, preferably at the utable to stretching of the flap during peeling or
12 o’clock position. The loosened epithelium was overexpansion due to generous hydration.
582 Surv Ophthalmol 49 (6) November–December 2004 TANERI ET AL

Fig. 2. Schematic illustration of different surgical LASEK techniques. (A) Azar Flap technique. (B) Camellin technique.
A sharp partial-thickness trephination of the epithelium is carried on prior to alcohol application. (C) Vinciguerra
butterfly technique. A thin paracentral epithelial line from 8 to 11 o’clock is abraded with a spatula, and 20% alcohol
in BSS is placed in contact with the cornea for 5–30 seconds and removed with a sponge. The epithelium is separated
from Bowman’s layer, proceeding from center to periphery on both sides. The two sheets of loose epithelium are
moved sideways toward the limbus. After drying the surface, excimer laser ablation is performed. (D) McDonald alcohol-
free technique. A rounded cataract blade is used to make a small linear abrasion through which a LASEK spatula is
slipped. Using that hole as a fulcrum, a spatulating motion is made and the epithelium stripped off. A dedicated curved
cannula is slipped under the epithelium and GenTeal Gel is blown out to dome up the epithelium. Finally, the
raised epithelium is bisected with Vannas scissors and parted sideways.

CAMELLIN TECHNIQUE three times while maintaining a constant pressure.

In the Camellin technique a sharp partial-thickness Then 20% alcohol solution (96⬚ pure alcohol in in-
trephination of the epithelium is carried on prior to jectable distilled water) is instilled into a small hold-
alcohol application: a pre-incision of the corneal epi- ing well (Janach J2905) on the corneal surface for
thelium is done to circumscribe the flap area and to 30 seconds. The well serves two functions: holding
allow the alcohol solution to penetrate under the the eye still and avoiding discharge of fluid. The
flap using a dedicated trephine (J2900; Janach, surface is dried and rinsed thoroughly with BSS and
Como, Italy), which has a blunt section of 90⬚ for the a final irrigation with an antihistamine to reduce
formation of a hinge (Fig. 2B). A rotation of about the amount of histamine induced by the alcohol.
10⬚ is performed, repeating the maneuver two or Subsequently the epithelium is detached with the
EVOLUTION AND PATHOPHYSIOLOGY OF LASEK 583

Fig. 3. Former variant of our technique using a Merocel sponge to fashion the epithelial flap.

short side of a dedicated epithelial detaching spatula In the Vinciguerra butterfly technique,130 a thin
(Janach J2910A). By making tiny movements almost paracentral epithelial line from 8 to 11 o’clock is
perpendicular to the margin, the epithelial sheet is abraded with a specially designed spatula, and 20%
folded at the 12-o’clock position to keep it moist alcohol in BSS is placed in contact with the cornea
during the treatment. Before laser ablation, the for 5–30 seconds (Fig. 2C). With the same spatula, the
longer side of this spatula is passed over the stromal epithelium is separated from Bowman’s layer, pro-
surface to remove any debris. If necessary, the ex- ceeding from center to periphery on both sides. A
posed stromal area may be enlarged by slightly strip- special retractor is used to move the two sheets of
ping the epithelium in the periphery. Camellin has loose epithelium sideways toward the limbus and
adjusted his PRK nomogram by reducing the preset hold them in place. After drying the surface, excimer
values by 10% when treating myopia up to 10 D and laser ablation is performed. Smoothing with a hya-
by 20% for myopias of 10–20 D, thus avoiding over- luronic acid masking solution (Laservis; Cheme-
correction. He advises protection of the flap with a dica, München, Germany) is then carried out,
masking fluid if smoothing is performed. The epithe- followed by repositioning of the stretched epithelial
lial flap is returned after laser ablation with another flaps with the margins overlapping.
spatula (Janach J2920A) and a soft contact lens is
applied. Postoperatively, antibiotic and cortisone eye
McDONALD TECHNIQUE
drops are administered for a few days, and a mild
cortisone treatment is continued for up to a month. The alcohol-free McDonald technique105 uses mi-
If complete re-epithelialization has not taken place crokeratome suction and a methylcellulose gel to
at postoperative day 3 or 4, a new lens is fitted for 3 create the epithelial sheet (Fig. 2D). A new curved
more days. Camellin strongly points out the impor- cannula (Mastel Precision, Rapid City, SD) is used; it
tance of a hypotonic solution, obtained by diluting has fine holes along the side through which GenTeal
alcohol in distilled water, for facilitating epithelial Gel (hydroxypropyl methylcellulose 0.3%; Novartis
detachment (Cimberle M, Camellin M: LASEK tech- Ophthalmics, Duluth, GA) can simultaneously ema-
nique promising after 1 year of experience. Ocul nate. Because methylcellulose gel, unlike alcohol,
Surg News 14:14–7, 2000). does not stiffen the epithelial cells, metallic instru-
ments should never touch the epithelium. Instead,
the cells are stripped with the assistance of suction
VINCIGUERRA BUTTERFLY TECHNIQUE and manipulated on a cloud of gel.
Decreased epithelial viability after alcohol expo- Generous amounts of GenTeal Gel are applied to
sure is a postoperative complication of standard the corneal surface to keep the epithelium in good
LASEK that may prolong visual recovery and cause condition. A rounded cataract blade is used to make a
temporary reduced visual acuity as well as discomfort. small linear abrasion in the far periphery of the
Vinciguerra developed a modification of the stan- cornea. Ten drops of NaCl 5% are added for 10
dard approach of creating the epithelial flap that, by seconds to slightly stiffen the epithelium, which is
preserving the limbal connection of epithelial stem followed by placement of the suction ring. While the
cells and limbal vascular connections, aims at increas- suction is on, a LASEK spatula is slipped through
ing epithelial viability, thus reducing the occurrence the 1- or 2-mm linear abrasion. Using that hole as a
of these complications. fulcrum, a spatulating motion is made and the
584 Surv Ophthalmol 49 (6) November–December 2004 TANERI ET AL

Fig. 4. Our current LASEK technique. (A) Multiple marks are applied around the corneal periphery, simulating a floral
pattern. (B) An alcohol dispenser consisting of a customized 7- or 9-mm semi-sharp marker attached to a hollow
metal handle serves as a reservoir for 18% alcohol. Firm pressure is exerted on the cornea and alcohol is released into
the well of the marker. (C) After 25–30 seconds, the ethanol is absorbed using a dry cellulose sponge. (D) One arm of
a modified Vannas scissors (note the knob at the tip of the lower arm) is then inserted under the epithelium and
traced around the delineated margin of the epithelium, leaving a hinge of 2–3 clock hours of intact margin, preferably
at the 12 o’clock position. (E) The loosened epithelium is peeled as a single sheet using a Merocel sponge or the edge
of a jeweler’s forceps, leaving it attached at its hinge. (F) After laser ablation is performed, an anterior chamber cannula
is used to hydrate the stroma and epithelial flap with balanced salt solution. (G) The epithelial flap is replaced on the
stroma using the cannula under intermittent irrigation. (H) Care is taken to realign the epithelial flap using the previous
marks and to avoid epithelial defects. The flap is allowed to dry for 2–5 minutes. Topical steroids and antibiotic
medications are applied. (I) A bandage contact lens is placed.

epithelium stripped off. After a maximum of 30 sec- After parting the two sides, a wet Weck-cel sponge
onds’ suction time, the dedicated curved cannula is used to remove the gel from Bowman’s layer. Then
is slipped under the epithelium and GenTeal Gel is ablation is performed, after which GenTeal Gel is
blown out to dome up the epithelium. Finally, the applied again, the epithelial sheet is herded back
raised epithelium is bisected with Vannas scissors. into position, and a bandage contact lens is placed.105
EVOLUTION AND PATHOPHYSIOLOGY OF LASEK 585

Clinical Outcomes STABILITY

In our analysis we have tried to adhere to the sys- Stability was defined as the number and percentage
tematic reporting format proposed by Koch et al.70 A of eyes with a change in spherical equivalent of mani-
summary is provided in Table 2. fest refraction of ⬎1.00 D within a specified interval;
Snellen visual acuity does not take into account the recommended minimal interval is 6 months. Only
many aspects of vision, such as contrast sensitivity Rouweyha et al112 report such a regression of about
and visual field, that are important to the overall 2 D in four eyes of two patients (8% of their eyes)
assessment of refractive procedures. However, it does with visually significant haze at 6 months, whereas
provide a reasonable and widely accepted indicator several other authors point out the absence of re-
of visual function that may serve as an outcome gression (Cimberle M, Camellin M: LASEK tech-
for comparison purposes. nique promising after 1 year of experience. Ocul
Surg News 14:14–7, 2000).24
Fig. 6 shows spherical equivalent of manifest refrac-
SAFETY tion plotted against time for the same cohort of pa-
Safety was defined as the number and percentage of tients followed throughout the postoperative period.
eyes losing two or more lines of BSCVA. In the reports
QUALITY OF VISION
that met our inclusion criteria and specifically men-
tion loss of BSCVA,1,3,24,41,76,82,84,112,118,119,130 only one Despite the lack of standardization in this area,
of 1,421 eyes(⬍0,1%) lost two lines of BSCVA. This we note some results of studies indicating quality
occurred after a macular cyst developed.6 In this cal- of vision. Scerrati compared LASIK to LASEK in 60
culation, we excluded a patient that we reported with myopic eyes. After 6-month follow-up he found slight-
loss of BSCVA of 2 lines at his final visit (1 month),41 ly better refractive results in the LASEK-treated group
because he regained BSCVA after ending this study regarding corneal topography, BSCVA, and con-
at his 3-month visit. trast sensitivity.115

EPITHELIAL CLOSURE AND PAIN

EFFICACY Epithelial closure time may serve as an indirect
Efficacy was defined as the percentage of eyes with sign of definite return of functional vision and end
UCVA of 20/20 and 20/40 or better, respectively. of pain perception after LASEK. However, this data
Only three studies24,76,112 report this outcome in a is not uniformly reported. Kornilovsky gives 4 days,
total of 352 eyes aimed at emmetropia at the 6-month Camellin 4–5 days, Lee 3.68 ⫾ 0.69 days,75–78 and
follow-up. UCVA of 20/20 or better at that time was Claringbold up to 2 weeks.24 We observed an epithe-
achieved in 76% and of 20/40 or better in 99%. lial defect in five eyes on day 3 that had no defect
Efficacy index is the ratio of the mean postopera- on day 1, and a closure rate of 78% on day 3 and
tive UCVA to the mean preoperative BSCVA. This 98.8% at 1 week. None of our patients had an epithe-
measure is particularly useful in describing out- lial defect after 1 week.
comes of patients with high myopia when the preop-
erative BSCVA is worse than 20/20. We computed SECONDARY SURGICAL MODIFICATION
an overall efficacy index of 0.9465 based on 421 eyes Having treated 222 eyes, Claringbold24 reported no
treated in the three studies that provide data enabling re-treatment, whereas we re-treated 11 eyes (6.7%) in
determination of the efficacy within their cohort at the time interval of 6 months to 4 years, all with
the 6-month follow-up (0.76,118 0.89,41 1.0224). satisfactory results. The earliest retreatment was done
We contrasted the reported LASEK results to those because of overcorrection of ⫹1.0 D; the other pa-
of a study by Walker and Wilson that compared early tients required enhancement.41 Rouweyha reports
postoperative visual recovery with PRK and LASIK in that three eyes required secondary surgery because
low myopes (Fig. 6).132 of a mean overcorrection of ⫹1.0 D.112

COMPLICATIONS
PREDICTABILITY (SPHERICAL EQUIVALENT) Possible complications of LASEK may be classified
The mean postoperative spherical equivalent of as sight-threatening, non-sight-threatening, intraop-
152 eyes at the 6-month follow-up was ⫺0.32 erative, and early and late postoperative (Table 3).
D.112,118,130 At the 6-month follow-up 83% of Possible complications during and immediately after
eyes24,76,112 were within ±0.50 D and 98.35% of eyes LASEK include free epithelial flap, dissolution, frag-
were within ±1.00 D of the desired postoperative re- ments, fold, and slip. Complications in the days fol-
fractive error.24,76 lowing surgery include persistent epithelial defects
586 Surv Ophthalmol 49 (6) November–December 2004 TANERI ET AL

Fig. 5. (A) Ying-yang or S-shaped cut of epithelium as a variant of our technique. (B) Variant of Vinciguerra butterfly
technique (without epithelial abrasion prior to alcohol application). (C) Star-shaped incision of epithelium as another
variant of our technique. (D) Z-shaped incision of epithelium as another variant of our technique.
EVOLUTION AND PATHOPHYSIOLOGY OF LASEK 587

Fig. 5. (Continued )
588 Surv Ophthalmol 49 (6) November–December 2004 TANERI ET AL

TABLE 2
Definitions and Occurrence of LASEK Indices of Safety, Efficacy, Predictability, and Stability
Safety at 52 weeks Loss of two lines of BSCVA 1 of 1421 eyes (⬍0.1%)
Safety index [mean postop 1.0
BSCVA]/[mean preop BSCVA]
Efficacy at 6 months UCVA of 20/20 or better 76%
UCVA of 20/40 or better 99%
Efficacy index [mean postop 0.9465
UCVA]/[mean preop BSCVA]
Predictability (spherical Mean spherical equivalent ⫺0.32 D
equivalent at 6 months) Within ⫾0.50 D 83%
Within ⫾1.00 D 98.35%
Stability at 6 months Regression ⬎ 1 D 4 eyes of 2 patients
(of ∼ 2 D with haze)112

and subepithelial foreign body. Minor complica- SUMMARY OF CLINICAL REPORTS

tions are probably not reported, but serious compli- After treating 249 patients, Camellin observed that
cations such as infections, stromal melting, recurrent intraoperative flap management was easy in 60% of
erosions, keratoectasia, scars, and severe haze forma-
cases, average in 28%, and difficult in 12% (Cimberle
tion certainly would have been.
M, Camellin M: LASEK technique promising after 1
Clinical outcomes of studies involving LASEK treat-
year of experience. Ocul Surg News 14:14–7, 2000).
ments are summarized in Tables 4 and 5. There has
No pain was experienced by 44% of his cases in the first
been no large, prospective, randomized clinical trial
24 hours after surgery, and 80% of preoperative BCVA
examining this procedure. Several patients seem to
be reported in various studies conducted by the same was achieved by 90% of his patients 10 days postopera-
examiners, as they report their results at different tively. In re-treatments after LASEK, he found the
time-points with increasing experience or to high- flap to be almost as easily detachable within 3 months
light diverse aspects of LASEK. For our calculations as in a primary operation.
we have neglected these redundant reports and took Claringbold24 reported a retrospective case series
only the largest study of each group into account. of 222 consecutive eyes with myopia ranging from
Outcomes noted by Azar8,9,13 are included in Feit’s ⫺1.25 D to ⫺11.25 D and astigmatism up to ⫹2.25 D
report,41 and those of Lee et al75,78 are included in treated with LASEK. At 12 months after surgery, there
Lee et al76 (personal communication), and we was no loss of BSCVA and no eye required retreatment
assume that the outcomes noted by Vinciguerra131 (of the 84 eyes followed up). Claringbold pointed
are included in Vinciguerra.130 out that young men, postmenopausal women, and

Fig. 6. Comparison of UCVA over time based on Walker and Wilson’s study.132
EVOLUTION AND PATHOPHYSIOLOGY OF LASEK 589

TABLE 3
LASEK Complications
Complication n % Reference Comments
Sight-threatening
Intraoperative
Perforation 0
Decentration of ablation 0
Early postoperative
Infection 0
Diffuse lamellar keratitis 0
Stromal melting 0
Persistent epithelial defect 0
Late postoperative
Scarring 0
Keratoectasia 0
Visually significant haze 4 8 112 Myopic regression of ∼ 2 D with visually
significant haze at 6 mo
Recurrent erosions 0
Severe dry eye syndrome 0

Non-sight-threatening
Intraoperative
Alcohol leakage 3 3.5 15
Incomplete flap 3 3.5 15
Fragmented flap 2 0.9 8 Tearing of flap during lifting
Allergic or toxic reactions 1 * Keratitis secondary to GenTeal
Overcorrection 8 114 ⬎ 0.5 D
Early postoperative
Pain 55.6 On day 1
13.5 44 On day 3
Superficial punctate keratopathy 31 44 At 1 week (peak time)
Loss of epithelial sheet 2.53 ** In association with loss of BCL
Contact lens intolerance 5 5.9 76
Steroid-induced rise in IOP 1 1.2 76
Late postoperative
Regression 4 8 114 Myopic regression of ∼ 2 D with
visually significant haze at 6 mo
Trace to mild haze 31 41 At 3 mo (peak time)
3.8 130 At 1 yr (trace)
2 0.8 *** Trace haze for 2 mos
*
Piechocki M, McDonald M: Alcohol-free LASEK procedure proves too effective in pilot study. Ocular Surg News.
Waikoloa, Hawaii, 2002.
**
Cimberle M, Camellin M: LASEK technique promising after 1 year of experience. Ocul Surg News 14:14–7, 2000
***
Cimberle M, Condon M: LASEK performs better than LASIK in selected cases. Ocul Surg News, 2002.

long-time contact lens users required prolonged alco- of 25% of the preoperative sphere. In the compound-
hol exposure as their epithelium was difficult to lift. treated eyes he found central islands that induced sig-
The LASEK patients achieved visual recovery sooner nificant optical aberrations.
than previous PRK patients and nearly at the same Lee et al76 found relatively good postoperative
time as those having LASIK. He thought that the ab- visual acuity and early visual recovery after LASEK,
sence of early stromal haze was due to improved enabling them to perform simultaneous surgery in
ablation with modern lasers and the presence of the both eyes. As complications they report alcohol leak-
epithelial flap. He found the lack of late-onset cor- age during surgery, which led to increased postopera-
neal haze, even in the higher refractive errors, a tive pain but not to conjunctival or limbal damage.
promising aspect of LASEK. They also noted contact lens intolerance associated
Vinciguerra131 used LASEK in his experimental with folds of Descemet’s membrane that required
model to disprove the hypothesis that compound pressure patching. As this intolerance occurred
ablation for myopia reduces postoperative spherical mainly in patients over 40 years old, they do not
aberration. Compound ablation consists of treating recommend LASEK in this age group.
myopic ablation by increasing the preoperative Examining the role of the epithelial flap in a pro-
sphere 25% and then applying a hyperopic ablation spective, comparative paired-eye trial in 36 patients
 TABLE 4
LASEK Results
Preoperative Epithelial
SE/Sph/Cyl Preop Preop Postoperative Postoperative Postop Duration of Loss of Complica- Closure
Study n (D) UCVA BSCVA SE (D) UCVA BCVA Follow-up BCVA tion/Remarks Time Comments
Piechocki *
42 ⫺5.53 [⫺1.25 n.a. n.a. n.a. At 1 mo. 6% n.a. 3 mo n.a. Allergic toxic n.a. GenTeal
to 9.125] achieved 20/12.5 keratitis Gel used
visual acuity, instead
which improved of alcohol
to 13% at 3 mo
and 75% were
20/25 or better
and 100% were
20/32 or better
at 3 mo
Rouweyha 58 (46 SE ⫺7.78 ⫾ n.a. n.a. 71% within 45%, 83%, 85%, n.a. Up to 6 mo No ⬎ 1 line 8 eyes overcor- n.a. 12 mono-
et al 112 targeted 2.9 Sph ⫾0.50 at and 89% ⱖ20/ rection vision
for emme- [⫺1.5 to 3 mo 40 at 1day, 1 wk, ⬎0.5D eyes
tropia) ⫺14.75] 68% within 2 wk, 1 mo 8% myopic re- excluded
⫾0.50, respectively gression in results
mean ⫺0.51 73% 20/20, with visu-
⫾ 0.72 97% ⱖ 20/40, ally signifi-
at 6 mo mean ⫺0.51 at cant haze at
6 mo 6 mo
Shah et al118 36 ⫺3.49 ⫾ 20/200 1.20 (SF) ⫺0.30 at 6 mo 0.91 at 6 mo 1.20 at 1 yr 62.6 weeks 5 eyes 1 line No flap n.a. Prospective,
1.36[⫺1.0 (graph) ⫺0.24 ⫾ 0.97 ⫾ 0.30 [52–70] No eye ⱖ 2 related paired
to ⫺6.10] 0.43 at 1 at 1 yr lines eye trial
year 100%
follow-up
Azar et al9 20 ⫺2.75 ⫾ 5% ⱖ 20/40 n.a. 58% within 92% ⱖ 20/25 n.a. Up to 4 yr n.a. No RES. no Defect in 63%
1.36 ⫾ 0.50 at at 1 mo pain after 1 at day 1
1 mo week Defect in 9%
0.15 ⫾ 0.53 at day 3
at 1 mo None
⫺0.06 ⫾ thereafter
0.08 at 1 yr
Azar et al8 57 ⫺2.55 ⫾ 20/40 to CF n.a. 75.6% within 25.4% 20/40 n.a. Up to 5 yr n.a. 58.1% pain at Up to 1 At 5 year
1.31 ⫾0.50 or better at (1 patient) day 1 week 5 followup 1
day 1 No pain after had patient
All 20/40 or better 1 week defect at reported
at 1 wk and 1 15.6% SPK at day no RES.
mo day 3 3 that had no
1 eye SPK at 6 had no haze
mo defect
at day 1
(Continued)
 TABLE 4
Continued
Preoperative Epithelial
SE/Sph/Cyl Preop Preop Postoperative Postoperative Postop Duration of Loss of Complica- Closure
Study n (D) UCVA BSCVA SE (D) UCVA BCVA Follow-up BCVA tion/Remarks Time Comments
9 eyes trace to
slight haze ‡
at 1 mo
5 new cases of
haze at 3
mo, 4 of
which re-
solved after
6 mo
Feit et al41 163 Mean ⫺2.92 1.05 logmar 3 20/30 74.7% within 0.95 (SF). No signif 1 mo to 2 yr 13 eyes 1 line 13.5% mild 78% at day 3 Including 12
Range ⫹0.25 (0.09 SF 5 20/25 ⫾ 0.50 0.02 logMAR at change (1 patient 5 1 eye 2 lines pain at 98.8% at 1 wk monovision
to ⫺.75 20/200) 115 20/20 93.2% within 6 mo (n ⫽ 73) yr) at final visit day 3 eyes
40 20/15 ⫾ 1.00 at 0.03 (20/20 at 31% SPK at 1
Mean final visit 12 mo week (peak
1,07SF time)
18% up to
mild haze
at 3 mo
(peak time)
Cimberle M** 249 total, 29 Myopic n.a. n.a. n.a. Almost 90% achieve 80% of Longest 14 n.a. 55.6% 4–5 days In re-treatments
re-treated group: their preop BCVA a day 10 months discomfort after LASEK,
after PRK, Mean 6.1 D (unclear if UCVA or BCVA) or pain at the flap
RK, PK ⫾ 4.2 SD day 1 was easily
or LASEK Hyperopic or 2.53% loss of detachable
hyperopic epithelial within 3
astigmatic sheet with mo, almost
group loss of BCL as in a
(n ⫽ 41): 20% fine primary
Mean 2.2 D subepithel- operation.
⫾ 2 SD ial dotting
at 1 month
Vinciguerra, 70 eyes ⫺5.30 ⫾ n.a. n.a. ⫺0.21 at 6 mo n.a. n.a. 12 mo No 3.8% trace n.a. “Butterfly”
Camesasca130 3.70 D (graph) haze at 12 technique
⫺0.10 ⫾ months
0.40 at 1 y No epithelial
(conflicting flap comp.
with graph)
Vinciguerra Group I: ⫺4.75 ⫾ 071 n.a. 0.99 ⫾ 0.02 0 ⫾ 0.74 0.81 ⫾ 0.23 0.95 ⫾ 0.10 3 mo n.a. Haze ⱕ 1 n.a. Group I:
et al131 4 eyes compound
ablation
Group II: ⫺4.44 ⫾ n.a. 0.96 ⫾ 0.07 ⫺0.19 ⫾ 38 0.94 ⫾ 0.06 0.98 ⫾ 0.03 40.5 ⫾ 30.2 Haze ⱕ 1‡ Group II:
4 eyes 1.25 days standard
ablation
Cimberle M*** 250 All ⬍ ⫺6 D n.a. n.a. n.a. 67% 20/20 n.a. Up to 1 yr No 2 haze for 2 n.a.
at 3 mo mo
(Continued)
 TABLE 4
Continued
Preoperative Epithelial
SE/Sph/Cyl Preop Preop Postoperative Postoperative Postop Duration of Loss of Complica- Closure
Study n (D) UCVA BSCVA SE (D) UCVA BCVA Follow-up BCVA tion/Remarks Time Comments

Claringbold 24
222 Sph:mean n.a. 219 20/20 97.7% within 83.8% ⱖ 20/40 at n.a. Up to 12 mo No 2 eyes devel- Up to 2 weeks 16.7% of
⫺4.89 2 20/25 ⫾ 0.50 at day 4 oped tears eyes
[⫺1.25 to 1 20/30 3 mo 86.5% ⱖ 20/20 at during lift- required
⫺1.25] Mean 1.0 98.7% within 3 mo ing of flap more than
Cyl mean (calculated) ⫾ 0.50 at 100% ⱖ 20/20 at no time 35-45 s of
⫹0.95 6 mo 90% ⱖ 20/20 more than ethanol
[⫹0.25 96.4% within Mean 1.02 at 6 mo trace haze exposure
to ⫹2.25] ⫾ 0.50 at (n ⫽ 156) No late-onset
12 mo 82% ⱖ 20/20 at haze
12 mo No recurrent
crosions
No severe
dry-eye
problems
No
retreatment
More in graphs
Lohmann et al34 24 SE mean n.a. n.a. 84% within n.a. n.a. 3 mo 5 (21%) lost 1 No epithelial n.a.
⫺4.54 ⫾ 0.50 line instability
[⫺2.0 to 100% within 6 eyes gained 1 Haze peaked
⫺6.0] ⫾1.00 at line at month
3 mo
Cyl up to ⫺1.0
mean ⫺0.14 0.3 ⫾ 0.37
⫾ 0.43 at 3 mo
0.22 ⫾ 0.25
Foreign body
sensation
Lee et al78 27 ⫺4.69 ⫾ 0.96 n.a. ⱖ20/20 in all ⫺0.34 ⫾ 0.42 All eyes 20/30 or n.a. 3 mo No early 3.64 ⫾ 0.63 Significantly
[⫺3.25 to eyes better at 3 mo postop [3-5] less pain
⫺6.50] than in
PRK
treated
fellow eye
(Continued)
 TABLE 4
Continued
Preoperative Epithelial
SE/Sph/Cyl Preop Preop Postoperative Postoperative Postop Duration of Loss of Complica- Closure
Study n (D) UCVA BSCVA SE (D) UCVA BCVA Follow-up BCVA tion/Remarks Time Comments
Lee et al 76
84 ⫺4.72 ⫾ 1.08 n.a. ⱖ20/20 in all 50% within 78.8% 20/30 or n.a. 6 mo 24% lost 1 line Mild to mod- 3.68 ⫾ 0.69
[⫺3.25 to eyes ⫾ 0.50 better at 1 week No loss of 2 erate pain days [3 to 6]
⫺7.0] 94% within ⫾ 97.6% at 1 mo lines 0.56 ⫾ 0.36
1.00 at 96.4% at 6 mo haze at 1 mo
6 mo 39.3% ⱖ 20/20 at 0.16 ⫾ 0.25
6 mo haze at 6 mo
3 alcohol
leakage
3 incomplete
flap
5 contact lens
intolerance
1 steroid-
induced
rise in
IOP

Lee et al75 15 ⫺4.67 ⫾ 0.80 n.a. n.a. n.a. n.a. n.a. 6 mo n.a. n.a. Significantly
[⫺3.25 to less tear
⫺6.0] fluid TGF-
β1 than
in PRK
at postop
day 1 and 2
Significantly
less corneal
haze at 1
mo than
PRK but
not at
3 mo and
6 mo
Scerrati115 30 ⫺6.00 to n.a. n.a. n.a. n.a. n.a. 6 mo n.a. n.a. n.a. Better
⫺12.00 contrast
sensitivity
than LASIK
Komilovsky71 12 n.a. n.a. 20/30 and n.a. n.a. n.a. n.a. n.a. Small to 4 days
below significant
pain in 6
eyes
Flap loss with
CL loss
Gierek-Ciaciura, 115 3 mo No postopera- Abstract
et al50 tive only

n ⫽ number of eyes; SE ⫽ spherical equivalent; SF ⫽ Snellen fraction; Sph ⫽ sphere; Cyl ⫽ cylinder; SD ⫽ standard deviation; n.a. ⫽ not available; KP ⫽ keratoplasty; BCL ⫽ bandage contact lens; SPK ⫽ Super-
ficial punctate keratopathy; RES ⫽ recurrent erosion syndrome; CF ⫽ counting fingers; TGF ⫽ transforming growth factor.
*
Piechocki M, McDonald M: Alcohol-free LASEK procedure proves too effective in pilot study. Ocular Surg News. Waikoloa, Hawaii, 2002.
**
Cimberle M, Camellin M: LASEK technique promising after 1 year of experience. Ocul Surg News 14:14–7, 2000.
***
Cimberle M, Condon M: LASEK performs better than LASIK in selected cases. Ocul Surg News, 2002.
‡
Granting according to Braunsetin et al.17
 594
TABLE 5

Surv Ophthalmol 49 (6) November–December 2004

LASEK Results in Prospective Studies

Preopera- Epithelial
tive SE/ Closure
Sph/Cyl Preop Postoperative Postoperative Postop Duration of Loss of Complication/ Time
Study N (D) UCVA Preop BCVA SE (D) UCVA BCVA Follow-up BCVA Remarks [days] Comments
Anderson 343 Sph ⫺5.43 n.a. 20/40 or At 1 mo 73% within At 1 mo, 70% n.a. Up to 6 1 eye lost 2 87% no postop 4.76 ⫾ 1.13 PRK nomogram
et al6 ⫾ 2.62 D better in all ⫾0.5 D, 92% achieved 20/20 mo lines (to pain decreased by
[⫺1.00 to eyes within ⫾ 1.0 D and 97% 20/40 20/40) 1.6% clinically 10%
⫺14.00] At 3 mo 78% within at 3 mo 80% 20/20 after significant haze 88% bilateral
Cyl 0.87 ⫾ ⫾0.5 D, 92% and 95% 20/40 developing on or before 3 simultaneous
0.75 D [0 within ⫾1.0 D at 6 mo 84% 20/20 a macular mo follow-up surgery
to 4.75] At 6 mo 85% within and 98% 20/40 cyst 20% ethanol in
⫾0.5 D, 94% dist. water for
within ⫾1.0 D 30 s
Chilled BSS
irrigation
Litwak 25 SE ⫺3.1 ⫾ n.a. 20/30 or n.a. At 1 wk 20/25 or n.a Up to 1 48% at 1 wk No corneal haze 3.6 ⫾ 0.5 Intraindividual
et al82 2.0 [⫺0.75 better in all better in 48% mo No line at 1 Patient report comparison to
to ⫺7.75] eyes At 1 mo 20/25 or mo follow- more PRK in fellow
better in 80% up discomfort and eye
worse Hispanic patients
VA in early 18% ethanol in
postop period dist.
compared to Water for 40 s
PRK-treated Chilled BSS
fellow eye irrigation
Shahinian119 146 SE ⫺5.32 n.a. n.a. 1 wk ⫹0.11 D At 6 mo 20/20 in n.a. 1 to 12 No ⬎ 1 line at No visually 88% by day 4 20% ethanol in
[⫺1.25 to 1 mo 0.0 D 57% and 20/40 mo 6 and 12 affecting haze 100% by day refrigerated
⫺14.38] 3 mo ⫺0.13 D or better in 96% mo No serious 6 dist. water
Cyl 0 to 6 mo ⫺0.14 D At 12 mo 20/20 in complication 3 for 35 s
⫹4.5 12 mo ⫺0.02 D 56% and 20/40 eyes of 2 PRK nomogram
At 6 mo 81% within or better in 96% patients late decreased by 10
⫾0.5 D and 94% overcorrection to 20%
within ⫾1.0D Minor corneal
tear film
changes,
similar to those
in epithelial
membrane
dystrophy in
33%

TANERI ET AL
Shah et al118 36 ⫺3.49 ⫾ 20/200 1.20 (Snellen ⫺0.30 at 6 mo 0.91 at 6 mo 1.20 at 52 62.6 wk 5 eyes 1 line No flap related n.a. Prospective, paired
1.36[⫺1.0 (graph) fraction) ⫺0.24 ⫾ 0.43 0.97 ⫾ 0.30 wk [52–70] No eye ⱖ 2 eye trial
to ⫺6.10] at 1 year at 1 y lines 100% follow-up
N ⫽ Number of eyes; SE ⫽ spherical equivalent; sph ⫽ sphere; cyl ⫽ cylinder.
EVOLUTION AND PATHOPHYSIOLOGY OF LASEK 595

who underwent PRK and LASEK with a Nidek EC-5000 achieve uneventful wound healing and optimal visual
laser, Shah et al118 found no statistically significant recovery.
postoperative UCVA in both groups. In the 36 Lee et al speculate that the epithelial flap in LASEK
LASEK-treated eyes they report a mean BCVA of acts as a barrier to tear fluid similar to applying a
1.20 ⫾ 0.14 and no loss of BCVA of more than 1 human amniotic membrane after PRK,78 as described
line after 52 weeks. This was statistically significantly by Choi et al in rabbit eyes.23
better than the PRK group. Corneal haze was signifi- The role of tear fluid transforming growth factor-β1
cantly lower than in the PRK group at all follow-up (TGF-β1) in wound healing was investigated by Lee
examinations (6, 12, 26, and 52 weeks). et al in 15 patients who underwent PRK in one eye and
Rouweyha et al112 report overcorrection as their LASEK in the other eye.75 They found less TGF-β1
most frequent LASEK complication in myopia from released in the early postoperative days following
⫺1.5 D to ⫺14.75 D. On the other hand, they ob- LASEK than following PRK. They also observed a
served myopic regression of approximately ⫺2.0 D significant lower grade of haze in LASEK after 1
due to unknown reasons in four high myopic eyes month but no significant difference at 3 and 6
(2 patients) within 6 months. All four eyes had haze months. Their conclusion was that decreased TGF-
formation causing decreased vision and requiring β1 in LASEK may have reduced wound healing and
treatment. early haze formation.
Efficacy and safety of LASEK was also seen in a Comparing the corneal wound-healing process
small series of 24 myopic eyes by Lohmann et al.84 after PRK and LASIK in 12 rabbit eyes, each, Park
Kornilovsky found comparable results in 12 high et al99 found that in the LASIK group, there was no
myopic patients whom he treated with LASEK in one observed regenerated collagen between the corneal
eye and with either LASIK, ordinary PRK, or transepi- flap and the ablated stromal bed except in the wound
thelial PRK in the other eye.71 However, he did not margin. Lamellated, parallel collagen fibers in the
report the length of follow-up. Given the potential corneal stroma were not disturbed. However, in
risks of late-onset scarring after PRK in high myopia, the wound margin, corneal epithelial ingrowth be-
most surgeons try to avoid the use of LASEK in tween the flap and the stromal bed was observed, as
higher myopes. This practice may be modified when was some regenerated stromal tissue. The amount
longer-term results in this subgroup become available. of regenerated stromal tissue and the number of
Shahinian reports minor corneal tear film changes, keratocytes in the wound area were statistically
similar to those seen in epithelial basement mem- smaller than those in the PRK group (p ⬍ 0.05).
brane dystrophy, in 48 of 146 eyes (33%).119 He also Stromal wound healing occurs in phases after kera-
reports that five female patients with these subtle torefractive laser surgery.14,142,143 In the first phase,
changes had occasional pain if they opened their the keratocytes adjacent to the area of epithelial
eyes too quickly in the morning, eye sticking in the wounding (including the area affected by the microk-
morning, occasional scratchy sensation, slight pulling eratome cut in LASIK) undergo apoptosis,93,139 leav-
sensation, and slight foreign-body sensation, at 3–12 ing a zone devoid of cells. This cell death has been
months. From this information we cannot exclude suggested to initiate the healing response,14,152 which
a mild form of iatrogenic recurrent erosion syn- in turn appears to be regulated by growth factors
drome after LASEK, which is otherwise not reported such as TGF-β.14,63 Epithelial injury is an important
in the literature.119 factor modulating keratocyte apoptosis.53,94
In the second phase of stromal wound healing, the
keratocytes immediately adjacent to the area of cell
death proliferate to repopulate the wound area. In
Wound Healing rat corneas proliferation occurs 24–48 hours after
Visual recovery is delayed until epithelial healing wounding.152,153 The keratocytes transform into fi-
is complete, and the healing process is likely to play broblasts and migrate into the wound area.138,141 This
a significant role in haze production.79 migration may take up to a week. Transformation
The first stage of wound healing of the cornea of keratocytes to fibroblasts can be visualized at the
after PRK and LASEK is epithelial migration. This is molecular level as reorganization of the actin cy-
followed by epithelial hyperplasia and subsequent toskeleton (with development of stress fibers and
stromal regeneration. By cytokine induction, the focal adhesion structures) and activation of new
epithelium can activate the process of keratocyte genes encoding extracellular matrix (ECM) compo-
apoptosis and myofibroblast transformation, leading nents such as fibronectin, ECM adhesion molecule,
to keratocyte replenishment of the anterior stroma. 5 integrin, ECM-degrading MMPs, and cytokines.36,43
Therefore, it is important to preserve epithelial viabil- This same transition occurs when keratocytes are
ity and integrity during the refractive surgery to isolated from the corneal stroma and cultured in
596 Surv Ophthalmol 49 (6) November–December 2004 TANERI ET AL

serum-containing medium; by the time these cells transformation may take up to a month to become
are subcultured, they have acquired the fibroblast apparent.
phenotype. The migratory repair fibroblasts (fila- TGF-ß and fibroblast growth factor (FGF)-2 have
mentous-actin positive) are elongated and spindle- opposing effects on the stromal cell phenotype
shaped; they are present in the wound edge and during wound healing and in culture conditions.
within the wound. Repair fibroblasts turn on the TGF-ß1 (and activin A) stimulates myofibroblast dif-
synthesis of the 5 integrin chain which results in ferentiation.31,45,65,149 Cytokines such as FGF-2 and
formation of the 5β1 integrin heterodimer, the clas- platelet-derived growth factor (PDGF) inhibit the
sic fibronectin receptor.44,125 This occurs concomi- transformation.47,56,60,111 In addition, FGF-2 induces
tant with deposition of fibronectin in the wound area.
conversion of myofibroblasts to the fibroblast pheno-
Dermatan sulfate proteoglycan synthesis increases,
type.85,88 The opposing effects of FGF-2 and TGF-ß
and lumican synthesis decreases.37
are consistent with the findings that these two growth
Freshly isolated keratocytes differ from subcul-
tured cells or from wound fibroblasts in their incom- factors participate in different signaling pathways and
petence to synthesize collagenase in response to that these pathways converge on the regulation of
treatment with agents that stimulate remodeling of Smad proteins downstream of TßRI and –II.72,74,88
the actin cytoskeleton, such as phorbol myristate ace- In culture, if the fibroblasts are confluent, addition
tate and cytochalasin B.42 Incompetence is due to of TGF-ß does not induce myofibroblast differentia-
failure to activate an autocrine interleukin (IL) 1 tion.89 This density-dependent differentiation corre-
feedback loop required to mediate cell response. To lates with the finding that high-density cells express
activate the positive feedback loop, cells must be able fewer receptors (TGF-ß receptor [TßR]II and TßRI)
to respond to two different stimulators: phorbol my- than do low-density cells.154
ristate acetate or cytochalasin B acts as the initiating The final phase of stromal healing involves stromal
stimulus, but cells must subsequently be able to re- remodeling and is greatly dependent on the origi-
spond to the IL-1 they synthesized if feedback amplifi- nal wound. Completely healed wounds contain few
cation is to occur. Failure to activate the transcription if any myofibroblasts, presumably because they revert
factor NF-κB explains incompetence for expression of to the fibroblast phenotype or undergo apoptosis
IL-1 in corneal stromal cells. Because NF-κB regulates during wound healing.60,61 This may take a year or
many cell functions with potential to disturb corneal more.
structure, including expression of inflammatory,
To date, only topical corticosteroids are widely
stress, and degradative proteinase genes, protection
utilized for modulation of wound healing after re-
against apoptosis, and cell replication, this seems
fractive surgery. They act by inhibiting activated kera-
likely to be an important mechanism protecting cor-
neal stasis and preserving function.15,27,120 In the tocytes, probably by interfering with DNA synthesis,
third phase of stromal wound healing, fibroblasts which decreases cellular activity and reduces collagen
may be transformed into myofibroblasts (evidenced synthesis.14 The use of mitomycin C to modify the
by smooth muscle actin staining). Myofibroblasts wound-healing process was proposed many years ago
appear as stellate cells; they are highly reflective but but is still controversial.127 Mitomycins are a group
are limited to within the wound area. Myofi- of antitumor antibiotics that covalently bind to DNA
broblasts are thought to be responsible for corneal after reductive activation. Mitomycin C inhibits fibro-
haze. The haze appears to be the result of ECM depo- blast function by a dose-dependent inhibition of fi-
sition by the myofibroblasts and also to the highly broblast proliferation.146 We believe the use of
reflective property of the cells themselves. Myofi- mitomycin C for treating patients with visually signifi-
broblasts are usually present directly beneath the cant preexisting corneal scarring may be justified by
healed epithelium. The extent of transformation into the excellent data reported by Majmudar et al87 and
myofibroblasts seems to depend on the type of wound Raviv et al.105 However, we think that extrapolation of
and on the extent of the stromal tissue removed.93 this therapeutic approach to the routinely preventive
In general, gaping wounds and wounds that remove
intraoperative or early postoperative application of
the epithelial basement membrane as well as Bow-
mitomycin C may be premature until more long-term
man’s layer (as in PRK) result in greater myofi-
broblast generation than wounds that do not results are available,11,59 given the potential adverse
penetrate the basement membrane or Bowman’s long-term effects. Our results suggest that the benefi-
layer (as in LASIK). LASEK, which removes Bow- cial effect of mitomycin C may result from inhibition
man’s but leaves the basement membrane relatively of keratocytes underlying the application zone.59
intact, may result in an intermediate level of myofi- The effect of laser ablation surgery on biomechani-
broblast formation in agreement with clinical obser- cal stability and corneal curvature changes has been
vations of intermediate levels of haze. Myofibroblast investigated by Dupps et al38 and Roberts et al.107–110
EVOLUTION AND PATHOPHYSIOLOGY OF LASEK 597

LASEK vs. PRK frailer than a LASIK flap. Among the complication-
Lee et al compared LASEK and PRK in a random- related factors were improper concentration of alco-
ized intra-individually controlled trial in 27 patients hol solution, dry center of epithelial flap, corneal
for the treatment of manifest refraction of –3.00 D epithelial injury by irrigation, and too early removal of
to –6.50 D. At the end of 3 months they found no the bandage contact lens.151
significant differences in both methods regarding Scerrati compared LASIK to LASEK in 60 myopic
epithelial healing time, UCVA, or refractive error. eyes. At postoperative 6 months the refractive results
However, LASEK-treated eyes had less postoperative were slightly better in the LASEK-treated group re-
pain and lower corneal haze scores 1 month after garding corneal topography, BSCVA, and contrast
treatment; 63% of their patients preferred the sensitivity.115
LASEK procedure.78 In the rare event of intraoperative infections, the
The main disadvantages of LASEK remain the un- process in LASEK starts at the epithelial level, as
predictable postoperative pain and epithelial heal- opposed to the intrastromal level in LASIK, and will
ing. Even after ensuring that no epithelial defects therefore theoretically be easier to manage.112
were present at the end of each procedure, de-epithe- Looking at ocular dryness symptoms after PRK and
lialized areas were observed in more than half the LASIK, Hovanesian et al found symptoms suggestive
cases 1 day after surgery. There were a similar number of mild recurrent erosions, including sharp pains,
of reports of postoperative pain. Because pain is the the sensation of the eyelid sticking to the eyeball,
most compelling drawback of PRK, rapid reepitheli- and soreness of the eyelid to touch a previously unrec-
alization is paramount to ensuring patient comfort ognized symptom of this condition. These symptoms
in the immediate postoperative period. LASEK in its occurred commonly after excimer laser procedures
numerous variations cannot guarantee it can achieve but were significantly more common, more severe,
this consistently. It can be argued, however, that half and more prolonged after PRK. These symptoms had
of the LASEK patients may not have epithelial defects a significant effect on patient satisfaction.58
and postoperative pain on postoperative day 1 and In patients who experience recurrent corneal
this is an improvement over conventional PRK. Per- erosions, LASIK is generally considered a poor op-
haps a better understanding of how the epithelium tion. LASEK may not only correct the refractive error
adheres to the ablated stroma would enable us to but may also cure the recurrent erosions in this
further improve the technique and the outcomes. population.8,24
Managing the corneal epithelium as a hinged flap From our study LASEK appears to be a safe and
with 20% ethanol is a safe technique with faster visual effective option when patients request refractive sur-
rehabilitation and reduced haze compared with de- gery, especially when “saving” approximately 80–120
bridement of the epithelium with alcohol. Further µm of corneal stroma in terms of untouched tissue
studies must be performed to compare pain levels depth is a decisive factor for the risk of an elective pro-
postoperatively with the epithelial flap and epithe- cedure, such as in the presence of thin corneas or
lial debridement.118 wide pupils and comparatively high corrections or
Examining the role of the epithelial flap in a pro- forme fruste keratoconus.
spective, comparative paired-eye trial in 36 patients There were insufficient data in prospective, com-
who underwent PRK and LASEK with a Nidek EC- parative trials to describe the relative advantages and
5000 laser, Shah et al found no statistically significant disadvantages of different lasers or nomograms.
postoperative UCVA in both groups.118 In the 36 Some authors using a PRK nomogram report under-
LASEK-treated eyes they report a mean BCVA of correction,118 whereas others indicate overcorrection
1.20 ⫾ 0.14 and no loss of BCVA of more than 1 and nomogram adjustments (Cimberle M, Camellin
line after 52 weeks. This was statistically significantly M: LASEK technique promising after 1 year of experi-
better than the PRK group. Corneal haze was signifi- ence. Ocul Surg News 14:14–7, 2000).8,9,13,24,41,112
cantly lower than in the PRK group at all follow-up
examinations (6, 12, 26, and 52 weeks).118
In the worst-case scenario—complete devitaliza- Future Applications/Wavefront
tion of the epithelial flap—the patient is assumed to Technology
be no worse than having had PRK with alcohol-
assisted epithelial removal in the first place.1,22,118 Achieving the optimal treatment dose can be ham-
pered by patient subjectivity in establishing an accu-
rate refraction. New wavefront technology will be
LASEK vs. LASIK able to obtain objective refractive data and may
In 14 eyes with abnormal epithelial flaps of 309 decrease the need for re-treatment in all types of
eyes treated, Zhou et al found the LASEK flap to be laser corrective surgery.126
598 Surv Ophthalmol 49 (6) November–December 2004 TANERI ET AL

Many investigators think that LASEK may become Researchsoft, Berkeley, CA) in September 2002.
the procedure of choice in wavefront-guided custo- Search terms employed were LASEK, laser subepithelial
mized ablations, as the benefit of these complex abla- keratomileusis, laser epithelial keratomileusis, laser subepi-
tions may not be negated by variable iatrogenic thelial keratectomy, laser epithelial keratectomy, subepithelial
aberrations due to a microkeratome-created stromal photorefractive keratectomy, epithelial flap photorefractive
flap.8,24,112,130 However, the greater wound healing keratectomy, laser-assisted subepithelial keratectomy, and
response in LASEK patients compared to LASIK pa- excimer laser subepithelial ablation in the search fields
tients may also mask the fine contours provided by title or abstract. All entries considered to be of signifi-
wavefront guided ablations and cause significant ab- cance were utilized, including those written in En-
errations itself. glish and German and also from the non-English
Additional study of the biochemical and histopath- literature if an English abstract was available. The
ological causes of the healing response may lead to reference section of each article was reviewed for
the development of a flap-making solution superior articles not captured by the Medline search. If these
to the ethyl alcohol now mainly used. A separation were felt to be of significance by adding additional
below Bowman’s layer would be desirable to further data or refuting existing information, they were in-
minimize haze formation and quicken visual recov- cluded. Articles failing to add significant new data
ery. Perhaps a better understanding of how the epi- were excluded. Whenever necessary, data were com-
thelium adheres to the ablated stroma would enable
puted from graphs and tables.
us to promote the technique and the outcomes.

Conclusions References
The remarkable aspects of this literature review 1. Abad JC: Posterior corneal protrusion after PRK. Cornea
17:456–7, 1998
are the following:
2. Abad JC, An B, Power WJ, et al: A prospective evaluation of
1) LASEK provides long-term stable results in alcohol-assisted versus mechanical epithelial removal before
photorefractive keratectomy. Ophthalmology 104:1566–74;
complete absence of serious complications, discussion 1574–5, 1997
such as infections, recurrent erosions, scars, 3. Abad JC, Talamo JH, Vidaurri-Leal J, et al: Dilute ethanol
and late-onset corneal haze formation. versus mechanical debridement before photorefractive ker-
atectomy. J Cataract Refract Surg 22:1427–33, 1996
2) Epithelial closure with recovery of functional 4. Agarwal A, Agarwal A, Agarwal T, et al: Laser in situ keratomi-
vision was completed at day 4 to day 7 in leusis for residual myopia after primary LASIK. J Cataract
most cases. Refract Surg 27:1013–7, 2001
3) There was a tendency toward overcorrection 5. Agrawal VB, Hanuch OE, Bassage S, et al: Alcohol versus
mechanical epithelial debridement: effect on underlying
with PRK nomograms. cornea before excimer laser surgery. J Cataract Refract Surg
4) Postoperative pain and prolonged visual recov- 23:1153–9, 1997
ery until the epithelium closes remain the big- 6. Anderson NJ, Beran RF, Schneider TL: Epi-LASEK for the
correction of myopia and myopic astigmatism. J Cataract
gest disadvantages of LASEK compared to Refract Surg 28:1343–7, 2002
LASIK. 7. Ang RT, Dartt DA, Tsubota K: Dry eye after refractive
surgery. Curr Opin Ophthalmol 12:318–22, 2001
The tendency toward overcorrection may be due 8. Azar DT, Ang RT: Laser subepithelial keratomileusis: evolu-
to the decreased wound healing response, which tion of alcohol assisted flap surface ablation. Int Ophthal-
mol Clin 42:89–97, 2002
may lead to myopic regression in PRK. A potential 9. Azar DT, Ang RT, Lee JB, et al: Laser subepithelial keratomi-
superiority of LASEK to LASIK in wavefront-guided leusis: electron microscopy and visual outcomes of flap pho-
ablations still remains speculative. LASEK surgery is torefractive keratectomy. Curr Opin Ophthalmol 12:323–
8, 2001
especially valuable in patients with thin corneas who 10. Azar DT, Farah SG: Laser in situ keratomileusis versus photo-
would not qualify for LASIK surgery. Additionally, refractive keratectomy: an update on indications and safety.
LASEK has become a viable option in patients with Ophthalmology 105:1357–8, 1998
professions or lifestyles that predispose to flap trauma 11. Azar DT, Jain S: Topical MMC for subepithelial fibrosis after
refractive corneal surgery [letter]. Ophthalmology 108:239–
(contact sports athletes and military personnel) and 40, 2001
in patients with low myopia who are at a lower risk 12. Azar DT, Spurr-Michaud SJ, Tisdale AS, et al: Altered epithe-
for subepithelial haze. lial-basement membrane interactions in diabetic corneas.
Arch Ophthalmol 110:537–40, 1992
13. Azar DT, Taneri S, Chen CC: Laser sub-epithelial keratomi-
leusis (LASEK) review and clinicopathological correlations.
Method of Literature Search Middle East J Ophthalmol 10:54–9, 2002
14. Baldwin HC, Marshall J: Growth factors in corneal wound
We conducted a search of the MEDLINE database healing following refractive surgery: a review. Acta Ophthal-
using an online search tool (Endnote 5.0, ISI mol Scand 80:238–47, 2002
EVOLUTION AND PATHOPHYSIOLOGY OF LASEK 599

15. Beales MP, Funderburgh JL, Jester JV, et al: Proteoglycan 38. Dupps WJ, Roberts C: Effect of acute biomechanical changes
synthesis by bovine keratocytes and corneal fibroblasts: on corneal curvature after photokeratectomy. J Refract Surg
maintenance of the keratocyte phenotype in culture. Invest 17:658–69, 2001
Ophthalmol Vis Sci 40:1658–63, 1999 39. Durairaj VD, Balentine J, Kouyoumdjian G, et al: The pre-
16. Bianchi C: LASIK and corneal ectasia. Ophthalmology dictability of corneal flap thickness and tissue laser ablation
109:619–21; author reply 621–2, 2002 in laser in situ keratomileusis. Ophthalmology 107:2140–
17. Braunstein RE, Jain S, McCally RL, et al: Objective measure- 3, 2000
ment of corneal light scattering after excimer laser keratec- 40. Farah SG, Azar DT, Gurdal C, et al: Laser in situ keratomi-
tomy. Ophthalmology 103:439–43, 1996 leusis: literature review of a developing technique. J Cataract
18. Bühren J, Cichocki M, Baumeister M, et al: [Diffuse lamellar Refract Surg 24:989–1006, 1998
keratitis after laser in situ keratomileusis. Clinical and confo- 41. Feit R, Taneri S, Ang RT, et al: Lasek results. Ophthalmol Clin
cal microscopy findings]. Ophthalmologe 99:176–80, 2002 North Am 16:127–35, 2003
19. Campos M, Hertzog L, Wang XW, et al: Corneal surface 42. Fini ME: Keratocyte and fibroblast phenotypes in the re-
after deepithelialization using a sharp and a dull instrument. pairing cornea. Prog Retin Eye Res 18:529–51, 1999
Ophthalmic Surg 23:618–21, 1992 43. Fitch JM, Linsenmayer CM, Linsenmayer TF: Collagen fibril
20. Campos M, Raman S, Lee M, et al: Keratocyte loss after assembly in the developing avian primary corneal stroma.
different methods of de-epithelialization. Ophthalmology Invest Ophthalmol Vis Sci 35:862–9, 1994
101:890–4, 1994 44. Friedlander M, Theesfeld CL, Sugita M, et al: Involvement of
21. Carones F, Fiore T, Brancato R: Mechanical vs. alcohol integrins alpha v beta 3 and alpha v beta 5 in ocular neovas-
epithelial removal during photorefractive keratectomy. cular diseases. Proc Natl Acad Sci USA 93:9764–9, 1996
J Refract Surg 15:556–62, 1999 45. Friedman SL: Seminars in medicine of the Beth Israel Hospi-
22. Chen CC, Chang JH, Lee JB, et al: Human corneal epithelial tal, Boston. The cellular basis of hepatic fibrosis. Mecha-
cell viability and morphology after dilute alcohol exposure. nisms and treatment strategies. N Engl J Med 328:1828–
Invest Ophthalmol Vis Sci 43:2593–602, 2002 35, 1993
23. Choi YS, Kim JY, Wee WR, et al: Effect of the application 46. Frisch L, Dick HB: [Bilateral simultaneous LASIK. Pro and
of human amniotic membrane on rabbit corneal wound contra]. Ophthalmologe 97:881–4, 2000
healing after excimer laser photorefractive keratectomy. 47. Funderburgh JL, Funderburgh ML, Mann MM, et al: Proteo-
Cornea 17:389–95, 1998 glycan expression during transforming growth factor beta
24. Claringbold TV: Laser-assisted subepithelial keratectomy for
-induced keratocyte-myofibroblast transdifferentiation.
the correction of myopia. J Cataract Refract Surg 28:18–
J Biol Chem 276:44173–8, 2001
22, 2002
48. Geerling G, Daniels JT, Dart JK, et al: Toxicity of natural
25. Cochener B, Le Floch-Savary G, Colin J: [Excimer photore-
tear substitutes in a fully defined culture model of human
fractive keratectomy (PRK) versus intrastromal corneal ring
corneal epithelial cells. Invest Ophthalmol Vis Sci 42:948–
segments (ICRS) for correction of low myopia]. J Fr Ophtal-
56, 2001
mol 23:663–78, 2000
49. Gierek-Ciaciura S: [History of excimer laser refractive sur-
26. Cochener B, Savary-Le Floch G, Colin J: [Excimer surface
gery in Poland]. Klin Oczna 104:67–9, 2002
photoablation versus Lasik for correction of mild myopia].
50. Gierek-Ciaciura S, Obidzinski M, Rokita-Wala I, et al:
J Fr Ophtalmol 24:349–59, 2001
[LASEK—new laser refractive surgical technique. Personal
27. Cook JR, Mody MK, Fini ME: Failure to activate transcription
experience]. Klin Oczna 104:7–11, 2002
factor NF-kappaB in corneal stromal cells (keratocytes).
Invest Ophthalmol Vis Sci 40:3122–31, 1999 51. Griffith M, Jackson WB, Lafontaine MD, et al: Evaluation of
28. Dada T, Sharma N, Vajpayee RB, et al: Sterile central disci- current techniques of corneal epithelial removal in hyper-
form keratopathy after LASIK. Cornea 19:851–2, 2000 opic photorefractive keratectomy. J Cataract Refract Surg
29. Damji KF, Munger R, Herndon LW, Allingham RR: Reduc- 24:1070–8, 1998
tion of IOP after PRK. Ophthalmology 104:1525–6, 1997 52. Güell JL, Gris O, de Muller A, et al: LASIK for the correction
30. Dastjerdi MH, Soong HK: LASEK (laser subepithelial kera- of residual refractive errors from previous surgical proce-
tomileusis). Curr Opin Ophthalmol 13:261–3, 2002 dures. Ophthalmic Surg Lasers 30:341–9, 1999
31. Desmoulière A, Geinoz A, Gabbiani F, et al: Trans- 53. Helena MC, Baerveldt F, Kim WJ, et al: Keratocyte apoptosis
forming growth factor-beta 1 induces alpha-smooth muscle after corneal surgery. Invest Ophthalmol Vis Sci 39:276–
actin expression in granulation tissue myofibroblasts and 83, 1998
in quiescent and growing cultured fibroblasts. J Cell Biol 54. Helena MC, Filatov VV, Johnston WT, et al: Effects of 50%
122:103–11, 1993 ethanol and mechanical epithelial debridement on corneal
32. Diebold Y, Calonge M, Carretero V, et al: Expression of structure before and after excimer photorefractive keratec-
ICAM-1 and HLA-DR by human conjunctival epithelial cul- tomy. Cornea 16:571–9, 1997
tured cells and modulation by nedocromil sodium. J 55. Helena MC, Meisler D, Wilson SE: Epithelial growth within
Ocul Pharmacol Ther 14:517–31, 1998 the lamellar interface after laser in situ keratomileusis
33. Diebold Y, Calonge M, Fernández N, et al: Characteriza- (LASIK). Cornea 16:300–5, 1997
tion of epithelial primary cultures from human conjunc- 56. Hirschi KK, Rohovsky SA, D’Amore PA: PDGF, TGF-beta,
tiva. Graefes Arch Clin Exp Ophthalmol 235:268–76, 1997 and heterotypic cell-cell interactions mediate endothelial
34. Diebold Y, Herreras JM, Callejo S, et al: Carbomer- versus cell-induced recruitment of 10T1/2 cells and their differen-
cellulose-based artificial-tear formulations: morphologic tiation to a smooth muscle fate. J Cell Biol 141:805–14, 1998
and toxicologic effects on a corneal cell line. Cornea 57. Hirst LW, Kenyon KR, Fogle JA, et al: Comparative studies
17:433–40, 1998 of corneal surface injury in the monkey and rabbit. Arch
35. Diebold YC, Calonge MC, Callejo SC, et al: Ultrastruc- Ophthalmol 99:1066–73, 1981
tural evidence of mucus in human conjunctival epithelial 58. Hovanesian JA, Shah SS, Maloney RK: Symptoms of dry eye
cultures. Curr Eye Res 19:95–105, 1999 and recurrent erosion syndrome after refractive surgery.
36. Doane KJ, Babiarz JP, Fitch JM, et al: Collagen fibril assembly J Cataract Refract Surg 27:577–84, 2001
by corneal fibroblasts in three-dimensional collagen gel cul- 59. Jain S, McCally RL, Connolly PJ, et al: Mitomycin C reduces
tures: small-diameter heterotypic fibrils are deposited in corneal light scattering after excimer keratectomy. Cornea
the absence of keratan sulfate proteoglycan. Exp Cell Res 20:45–9, 2001
202:113–24, 1992 60. Jester JV, Barry-Lane PA, Cavanagh HD, et al: Induction of
37. Doane KJ, Yang G, Birk DE: Corneal cell-matrix interactions: alpha-smooth muscle actin expression and myofibroblast
type VI collagen promotes adhesion and spreading of cor- transformation in cultured corneal keratocytes. Cornea
neal fibroblasts. Exp Cell Res 200:490–9, 1992 15:505–16, 1996
600 Surv Ophthalmol 49 (6) November–December 2004 TANERI ET AL

61. Jester JV, Li HF, Petroll WM, et al: Area and depth of surfac- 84. Lohmann CP, Winkler Von Mohrenfels C, Gabler B, et al:
tant-induced corneal injury correlates with cell death. Invest [Excimer laser subepithelial ablation (ELSA) or laser epi-
Ophthalmol Vis Sci 39:922–36, 1998 thelial keratomileusis (LASEK)—a new kerato-refractive
62. Jester JV, Maurer JK, Petroll WM, et al: Application of in procedure for myopia. Surgical technique and first clinical
vivo confocal microscopy to the understanding of surfactant- results on 24 eyes and 3 months follow-up]. Klin Mo-
induced ocular irritation. Toxicol Pathol 24:412–28, 1996 natsbl Augenheilkd 219:26–32, 2002
63. Jester JV, Petroll WM, Cavanagh HD: Corneal stromal 85. Long CJ, Roth MR, Tasheva ES, et al: Fibroblast growth
wound healing in refractive surgery: the role of myofi- factor-2 promotes keratan sulfate proteoglycan expression
broblasts. Prog Retin Eye Res 18:311–56, 1999 by keratocytes in vitro. J Biol Chem 275:13918–23, 2000
64. Johnson JD, Harissi-Dagher M, Pineda R, et al: Diffuse lamel- 86. Magallanes R, Shah S, Zadok D, et al: Stability after laser
lar keratitis: incidence, associations, outcomes, and a new in situ keratomileusis in moderately and extremely myopic
classification system. J Cataract Refract Surg 27:1560–6, 2001 eyes. J Cataract Refract Surg 27:1007–12, 2001
65. Joyce NC, Zieske JD: Transforming growth factor-beta recep- 87. Majmudar PA, Forstot SL, Dennis RF, et al: Topical mitomy-
tor expression in human cornea. Invest Ophthalmol Vis Sci cin-C for subepithelial fibrosis after refractive corneal sur-
38:1922–8, 1997 gery. Ophthalmology 107:89–94, 2000
66. Kahn CR, Young E, Lee IH, et al: Human corneal epithelial 88. Maltseva O, Folger P, Zekaria D, et al: Fibroblast growth
primary cultures and cell lines with extended life span: in factor reversal of the corneal myofibroblast phenotype.
vitro model for ocular studies. Invest Ophthalmol Vis Sci Invest Ophthalmol Vis Sci 42:2490–5, 2001
34:3429–41, 1993 89. Masur SK, Dewal HS, Dinh TT, et al: Myofibroblasts differen-
67. Katlun T, Wiegand W: [Change in twilight vision and glare tiate from fibroblasts when plated at low density. Proc Natl
sensitivity after PRK]. Ophthalmologe 95:420–6, 1998 Acad Sci USA 93:4219–23, 1996
68. Kim SY, Sah WJ, Lim YW, et al: Twenty percent alcohol 90. Means TL, Geroski DH, L’Hernault N, et al: The corneal
toxicity on rabbit corneal epithelial cells: electron micro- epithelium after optisol-GS storage. Cornea 15:599–605,
scopic study. Cornea 21:388–92, 2002 1996
69. Knorz MC, Jendritza B, Liermann A, et al: [LASIK for 91. Melki SA, Azar DT: LASIK complications: etiology, manage-
myopia correction. 2-year follow-up]. Ophthalmologe 95: ment, and prevention. Surv Ophthalmol 46:95–116, 2001
494–8, 1998 92. Melki SA, Talamo JH, Demetriades AM, et al: Late traumatic
70. Koch DD, Kohnen T, Obstbaum SA, et al: Format for re- dislocation of laser in situ keratomileusis corneal flaps. Oph-
porting refractive surgical data. J Cataract Refract Surg
thalmology 107:2136–9, 2000
24:285–7, 1998
93. Mohan RR, Hutcheon AEK, Choi R, et al: Apoptosis, necro-
71. Kornilovsky IM: Clinical results after subepithelial photore-
sis, proliferation, and myofibroblast generation in the stroma
fractive keratectomy (LASEK). J Refract Surg 17:S222–3,
following LASIK and PRK. Exp Eye Res 76:71–87, 2003
2001
94. Mohan RR, Mohan RR, Kim WJ, et al: Defective keratocyte
72. Kretzschmar M, Doody J, Massagué J: Opposing BMP and
apoptosis in response to epithelial injury in stat 1 null mice.
EGF signalling pathways converge on the TGF-beta family
Exp Eye Res 70:485–91, 2000
mediator Smad1. Nature 389:618–22, 1997
95. Nakano EM, Nakano K, Oliveira MC, et al: Cleaning solu-
73. Lans L: Experimentelle Untersuchungen über Entstehung
tions as a cause of diffuse lamellar keratitis. J Refract Surg
von Astigmatismus durch nichtperforirende Corneawun-
18:S361–3, 2002
den. Graefes Arch Ophthalmol 45:117–52, 1898
96. Nishida T: Basic Science: Cornea, sclera, and ocular adnexa
74. Lawler S, Feng XH, Chen RH, et al: The type II transforming
growth factor-beta receptor autophosphorylates not only on anatomy, biochemistry, physiology, and biomechanics. in
serine and threonine but also on tyrosine residues. J Biol Krachmer JH, Mannis MJ, Holland EJ (eds): Cornea: Funda-
Chem 272:14850–9, 1997 mentals of Cornea and External Disease. St Louis, Mosby,
75. Lee JB, Choe CM, Kim HS, et al: Comparison of TGF-beta1 1997, p 5
in tears following laser subepithelial keratomileusis and 97. North-Root H, Yackovich F, Demetrulias J, et al: Evaluation of
photorefractive keratectomy. J Refract Surg 18:130–4, 2002 an in vitro cell toxicity test using rabbit corneal cells to
76. Lee JB, Choe CM, Seong GJ, et al: Laser subepithelial kera- predict the eye irritation potential of surfactants. Toxicol
tomileusis for low to moderate myopia: 6-month follow-up. Lett 14:207–12, 1982
Jpn J Ophthalmol 46:299–304, 2002 98. Pallikaris IG, Kymionis GD, Astyrakakis NI: Corneal ectasia
77. Lee JB, Ryu CH, Kim J, et al: Comparison of tear secretion induced by laser in situ keratomileusis. J Cataract Refract
and tear film instability after photorefractive keratectomy Surg 27:1796–802, 2001
and laser in situ keratomileusis. J Cataract Refract Surg 99. Park CK, Kim JH: Comparison of wound healing after photo-
26:1326–31, 2000 refractive keratectomy and laser in situ keratomileusis in
78. Lee JB, Seong GJ, Lee JH, et al: Comparison of laser epithe- rabbits. J Cataract Refract Surg 25:842–50, 1999
lial keratomileusis and photorefractive keratectomy for low 100. Parnigotto PP, Bassani V, Montesi F, et al: Bovine corneal
to moderate myopia. J Cataract Refract Surg 27:565–70, stroma and epithelium reconstructed in vitro: characterisa-
2001 tion and response to surfactants. Eye 12(Pt 2):304–10, 1998
79. Li DQ, Tseng SC: Three patterns of cytokine expression 101. Petersen H, Seiler T: [Laser in situ keratomileusis (LASIK).
potentially involved in epithelial-fibroblast interactions of Intraoperative and postoperative complications]. Ophthal-
human ocular surface. J Cell Physiol 163:61–79, 1995 mologe 96:240–7, 1999
80. Linebarger EJ, Hardten DR, Lindstrom RL: Diffuse lamellar 102. Poole CA, Brookes NH: Confocal imaging of the keratocyte
keratitis: diagnosis and management. J Cataract Refract network in porcine cornea using the fixable vital dye 5-
Surg 26:1072–7, 2000 chloromethylfluorescein diacetate. Curr Eye Res 15:165–
81. Linebarger EJ, Hardten DR, Lindstrom RL: Diffuse lamellar 74, 1996
keratitis: identification and management. Int Ophthalmol 103. Poole CA, Brookes NH, Clover GM: Keratocyte networks
Clin 40:77–86, 2000 visualised in the living cornea using vital dyes. J Cell Sci
82. Litwak S, Zadok D, Garcia-de Quevedo V, et al: Laser-assisted 106(Pt 2):685–91, 1993
subepithelial keratectomy versus photorefractive keratec- 104. Rapuano CJ, Sugar A, Koch DD, et al: Intrastromal corneal
tomy for the correction of myopia. A prospective compara- ring segments for low myopia: a report by the American
tive study. J Cataract Refract Surg 28:1330–3, 2002 Academy of Ophthalmology. Ophthalmology 108:1922–8,
83. Loewenstein A, Goldstein M, Lazar M: Retinal pathology 2001
occurring after excimer laser surgery or phakic intraocular 105. Raviv T, Majmudar PA, Dennis RF, et al: Mytomycin-C for
lens implantation: evaluation of possible relationship. Surv post-PRK corneal haze. J Cataract Refract Surg 26:1105–
Ophthalmol 47:125–35, 2002 6, 2000
EVOLUTION AND PATHOPHYSIOLOGY OF LASEK 601

106. Ren H, Wilson G: Apoptosis in the corneal epithelium. 130. Vinciguerra P, Camesasca FI: Butterfly laser epithelial kera-
Invest Ophthalmol Vis Sci 37:1017–25, 1996 tomileusis for myopia. J Refract Surg 18:S371–3, 2002
107. Roberts C: The cornea is not a piece of plastic. J Refract 131. Vinciguerra P, Munoz MI, Camesasca FI: Reduction of
Surg 16:407–13, 2000 spherical aberration: experimental model of photoablation.
108. Roberts C: Future challenges to aberration-free ablative pro- J Refract Surg 18:S366–70, 2002
cedures. J Refract Surg 16:S623–9, 2000 132. Walker MB, Wilson SE: Recovery of uncorrected visual acuity
109. Roberts C: The accuracy of ‘power’ maps to display curva- after laser in situ keratomileusis or photorefractive keratec-
ture data in corneal topography systems. Invest Ophthalmol tomy for low myopia. Cornea 20:153–5, 2001
Vis Sci 35:3525–32, 1994 133. Walker MB, Wilson SE: Incidence and prevention of epithe-
110. Roberts CW, Koester CJ: Optical zone diameters for photo- lial growth within the interface after laser in situ keratomi-
refractive corneal surgery. Invest Ophthalmol Vis Sci leusis. Cornea 19:170–3, 2000
34:2275–81, 1993 134. Walker MB, Wilson SE: Lower intraoperative flap complica-
111. Rønnov-Jessen L, Petersen OW: Induction of alpha-smooth tion rate with the Hansatome microkeratome compared
muscle actin by transforming growth factor-beta 1 in quies- to the Automated Corneal Shaper. J Refract Surg 16:79–
cent human breast gland fibroblasts. Implications for myo- 82, 2000
fibroblast generation in breast neoplasia. Lab Invest 68: 135. Waring GO, Carr JD, Stulting RD, et al: Prospective, random-
696–707, 1993 ized comparison of simultaneous and sequential bilateral
112. Rouweyha RM, Chuang AZ, Mitra S, et al: Laser epithelial LASIK for the correction of myopia. Trans Am Oph-
keratomileusis for myopia with the autonomous laser. J Re- thalmol Soc 95:271–84, 1997
fract Surg 18:217–24, 2002 136. Webber SK, Lawless MA, Sutton GL, et al: LASIK for post
113. Rozakis GW: Halos after PRK. J Refract Surg 13:340, 1997 penetrating keratoplasty astigmatism and myopia. Br J Oph-
114. Sakarya Y, Ozatep V, Ermip SS: Drift index to explain patient thalmol 83:1013–8, 1999
complaints after PRK. J Cataract Refract Surg 26:161, 2000 137. Webber SK, Lawless MA, Sutton GL, et al: Staphylococcal
115. Scerrati E: Laser in situ keratomileusis vs. laser epithelial infection under a LASIK flap. Cornea 18:361–5, 1999
keratomileusis (LASIK vs. LASEK). J Refract Surg 17:S219– 138. Wilson G, Ren H, Laurent J: Corneal epithelial fluorescein
21, 2001 staining. J Am Optom Assoc 66:435–41, 1995
116. Seiler T, Koufala K, Richter G: Iatrogenic keratectasia after 139. Wilson SE: Role of apoptosis in wound healing in the
laser in situ keratomileusis. J Refract Surg 14:312–7, 1998 cornea. Cornea 19:S7–12, 2000
117. Shah S, Doyle SJ, Chatterjee A, et al: Comparison of 18% 140. Wilson SE: LASIK: management of common complications.
ethanol and mechanical debridement for epithelial removal Laser in situ keratomileusis. Cornea 17:459–67, 1998
before photorefractive keratectomy. J Refract Surg 14:S212– 141. Wilson SE, He YG, Weng J, et al: Epithelial injury induces
4, 1998 keratocyte apoptosis: hypothesized role for the interleukin-
118. Shah S, Sebai Sarhan AR, Doyle SJ, et al: The epithelial flap 1 system in the modulation of corneal tissue organization and
for photorefractive keratectomy. Br J Ophthalmol 85:393– wound healing. Exp Eye Res 62:325–7, 1996
6, 2001 142. Wilson SE, Mohan RR, Hong JW, et al: The wound healing
119. Shahinian L: Laser-assisted subepithelial keratectomy for response after laser in situ keratomileusis and photorefrac-
low to high myopia and astigmatism. J Cataract Refract Surg tive keratectomy: elusive control of biological variability and
28:1334–42, 2002 effect on custom laser vision correction. Arch Ophthalmol
120. Sivak JM, Fini ME: MMPs in the eye: emerging roles for 119:889–96, 2001
matrix metalloproteinases in ocular physiology. Prog Retin 143. Wilson SE, Mohan RR, Mohan RR, et al: The corneal wound
Eye Res 21:1–14, 2002 healing response: cytokine-mediated interaction of the epi-
thelium, stroma, and inflammatory cells. Prog Retin Eye
121. Spigelman AV: Complications of LASIK. J Refract Surg
Res 20:625–37, 2001
17:475, 2001
144. Winkler von Mohrenfels C, Hermann W, Gabler B, et al:
122. Spurr SJ, Gipson IK: Isolation of corneal epithelium with
[Topical Mitomycin C for the prophylaxis of recurrent haze
Dispase II or EDTA. Effects on the basement membrane
after excimer laser photorefractive keratectomy (PRK)—a
zone. Invest Ophthalmol Vis Sci 26:818–27, 1985
pilotstudy of 5 patients]. Klin Monatsbl Augenheilkd
123. Stein HA, Stein RM, Price C, et al: Alcohol removal of the
218:763–7, 2001
epithelium for excimer laser ablation: outcomes analysis.
145. Wu G, Xie L, Yao Z: Post-PRK muscular asthenopia and
J Cataract Refract Surg 23:1160–3, 1997 eccentric ablation. Chin Med J (Engl) 114:167–9, 2001
124. Steinert RF, Hersh PS: Spherical and aspherical photorefrac- 146. Yamamoto T, Varani J, Soong HK, et al: Effects of 5-fluoro-
tive keratectomy and laser in-situ keratomileusis for moder- uracil and mitomycin C on cultured rabbit subconjunctival
ate to high myopia: two prospective, randomized clinical fibroblasts. Ophthalmology 97:1204–10, 1990
trials. Summit technology PRK-LASIK study group. Trans 147. Yavitz EQ: Diffuse lamellar keratitis caused by mechanical
Am Ophthalmol Soc 96:197–221; discussion 221–7, 1998 disruption of epithelium 60 days after LASIK. J Refract Surg
125. Strömblad S, Fotedar A, Brickner H, et al: Loss of p53 17:621, 2001
compensates for alpha v-integrin function in retinal neovas- 148. Yeoh J, Moshegov CN: Delayed diffuse lamellar keratitis after
cularization. J Biol Chem 277:13371–4, 2002 laser in situ keratomileusis. Clin Experiment Ophthalmol
126. Sugar A, Rapuano CJ, Culbertson WW, et al: Laser in situ 29:435–7, 2001
keratomileusis for myopia and astigmatism: safety and effi- 149. You L, Kruse FE: Differential effect of activin A and BMP-
cacy: a report by the American Academy of Ophthalmology. 7 on myofibroblast differentiation and the role of the Smad
Ophthalmology 109:175–87, 2002 signaling pathway. Invest Ophthalmol Vis Sci 43:72–81, 2002
127. Talamo JH, Gollamudi S, Green WR, et al: Modulation of 150. Yuhan KR, Nguyen L, Wachler BS: Role of instrument clean-
corneal wound healing after excimer laser keratomileusis ing and maintenance in the development of diffuse lamellar
using topical mitomycin C and steroids. Arch Ophthalmol keratitis. Ophthalmology 109:400–3; discussion 403–4, 2002
109:1141–6, 1991 151. Zhou X, Wu L, Dai J, et al: [The epithelial-flap abnormality
128. Van Gelder RN, Steger-May K, Yang SH, et al: Comparison of of laser epithelial keratomileusis]. Chung Hua Yen Ko Tsa
photorefractive keratectomy, astigmatic PRK, laser in situ Chih 38:69–71, 2002
keratomileusis, and astigmatic LASIK in the treatment of 152. Zieske JD: Extracellular matrix and wound healing. Curr
myopia. J Cataract Refract Surg 28:462–76, 2002 Opin Ophthalmol 12:237–41, 2001
129. Velou SM, Colin J: Photo essay: disastrous complications 153. Zieske JD, Guimarães SR, Hutcheon AE: Kinetics of kerato-
following a bilateral, same-day laser in situ keratomileusis cyte proliferation in response to epithelial debridement.
(LASIK) procedure. Arch Ophthalmol 120:226–7, 2002 Exp Eye Res 72:33–9, 2001
602 Surv Ophthalmol 49 (6) November–December 2004 TANERI ET AL

154. Zimmerman CM, Padgett RW: Transforming growth factor the New England Corneal Transplant Research Fund, Massachu-
beta signaling mediators and modulators. Gene 249:17– setts Lions Eye Research Fund, and Research to Prevent Blindness,
30, 2000 Lew R. Wasserman Merit Award, Northborough, Massachusetts
(Dr. Azar).
The authors reported no proprietary or commercial interest Reprint address: Dimitri T. Azar, MD, Director, Corneal and
in any product mentioned or concept discussed in this article. Refractive Surgery Services, Massachusetts Eye and Ear Infirmary,
Supported by the National Institutes of Health (NEI 10101), 243 Charles Street, Boston, MA 02114 USA.