Phototherapeutic Keratectomy in Diffuse Stromal Haze in Granular Corneal Dystrophy Type 2 Se Hwan Jung, MD,* Kyung Eun Han, MD,* R. Doyle Stulting,… [605310]

CLINICAL SCIENCE
Phototherapeutic Keratectomy in Diffuse Stromal
Haze in Granular Corneal Dystrophy Type 2
Se Hwan Jung, MD,* Kyung Eun Han, MD,* R. Doyle Stulting, MD, PhD, †
Bradford Sgrignoli, DO,* Tae-im Kim, MD, PhD,* and Eung Kweon Kim, MD, PhD* ‡
Purpose: To determine the minimum depth of phototherapeutic
keratectomy (PTK) required for diffuse haze removal in granular
corneal dystrophy type 2 and to determine whether Fourier domain
optical coherence tomography (FD-OCT) can be an effective
technique for predicting the exact required depth of ablation.
Methods: The depth of ablation used for diffuse stromal haze
removal was evaluated with the slit lamp and serially taken
photographs during the PTK procedure. The depth of diffuse haze
was measured preoperatively using FD-OCT.
Results: Forty-three eyes of 30 patients were included in this study.
The mean age of the patients was 62.0 ± 8.4 years. The mean depth of
PTK required was 43.7 ± 6.2 mm (range, 31 –59mm). The mean
follow-up period for 29 eyes of 22 patients, who had follow-up peri-
ods of more than 6 months, was 21.0 ± 12.0 months. The mean best
spectacle-corrected visual acuity of these 29 eyes was 0.43 ± 0.15
preoperatively and 0.71 ± 0.16 ( P= 0.022) 1 month postoperatively.
Of the 43 eyes of 30 patients, FD-OCT was evaluated in 29 eyes of 22
patients. The mean preoperative depth of diffuse haze using FD-OCT
was 44.3 ± 6.4 mm. The mean depth of ablation required to remove
diffuse stromal haze was 44.5 ± 5.9 mm. The actual ablated depths
correlated well with the depth of haze detected by FD-OCT preoper-
atively (intraclass correlation coef ficient = 0.719).
Conclusions: FD-OCT is an accurate method of predicting the
depth of PTK required to remove visually signi ficant diffuse haze in
patients with granular corneal dystrophy type 2. We advocate the use
of slit-lamp biomicroscopy after the initial 30- mm ablation to
determine the necessity for any further ablation.
Key Words: granular corneal dystrophy type 2, Avellino corneal
dystrophy, diffuse haze, phototherapeutic keratectomy, Fourier
domain optical coherence tomography
(Cornea 2013;32:296 –300)Granular corneal dystrophy type 2 (GCD2; also known as
Avellino corneal dystrophy) is an autosomal dominant
disorder showing 3 kinds of deposits in sequence: granular
deposits at an early stage, lattice deposits with progression,
andfinally, diffuse stromal haze.1Each characteristic deposit
occurs at a different depth: diffuse stromal haze is found at
the sub-Bowman layer, granular deposits are below the dif-
fuse stromal haze, and lattice deposits are in the deeper
stroma.1–3In the early stage of expression in the GCD2
heterozygote, the intervening cornea between granular or
lattice deposits remains clear without compromising vision.4
As diffuse stromal haze is deposited centrally, visual acuity is
reduced and treatment is indicated.
Phototherapeutic keratectomy (PTK) has been
frequently used for the treatment of anterior corneal dystro-
phies.5Although exacerbation and recurrence of GCD2 have
been reported after refractive or therapeutic excimer laser
ablation,6–9PTK may delay the need for keratoplasty in
patients with severe diffuse stromal haze.10If the exact depth
of deposits could be evaluated preoperatively, the surgeon
could determine the minimal amount of tissue required to
be removed.
Fourier domain optical coherence tomography (FD-OCT)
is a noninvasive method that allows clear identi fication of
corneal layers in various pathological conditions.11−13The mor-
phology and depth of the deposits in GCD2 using FD-OCT
have been previously reported.2However, the relationship
between the measured depth of diffuse stromal haze and the
ablation depth required for haze removal during PTK has not
been determined.
In this study, the actual depth of PTK ablation required
for diffuse stromal haze removal was determined by retro-
spective chart review and compared with the depth of stromal
haze measured preoperatively by FD-OCT.
MATERIALS AND METHODS
Patient Selection and Analysis
Medical records of patients who underwent PTK for
heterozygous GCD2 from January 2007 to December 2010 at
Severance Hospital (Seoul, Korea) were reviewed. In each
case, the diagnosis was con firmed by DNA analysis. Patients
were excluded from this study if they had signi ficant corneal
thinning or edema, previous excimer laser ablation, uveitis,
glaucoma, and other signi ficant ocular surface diseases.
Patients lacking diffuse stromal haze, but having either
granular or lattice deposits, were also excluded. PTK wasReceived for publication October 18, 2011; revision received December 7,
2011; accepted January 8, 2012.
From the *Department of Ophthalmology, Corneal Dystrophy Research
Institute, Yonsei University College of Medicine, Seoul, Korea; †Stulting
Research Center, Woolfson Eye Institute, Atlanta, GA; and ‡Brain Korea
21 Project for Medical Science, Severance Biomedical Science Institute,
Yonsei University College of Medicine, Seoul, Korea.
Supported by a National Research Foundation of Korea grant funded by the
Korea government (No. 2011-0028699).
The authors state that they have no con flicts of interest to disclose.
Reprints: Eung Kweon Kim, Department of Ophthalmology, Yonsei
University College of Medicine, 50 Yonsei-ro, Seodaemun-gu, Seoul
120-752, Korea (e-mail: eungkkim@yuhs.ac).
Copyright © 2012 by Lippincott Williams & Wilkins
296 |www.corneajrnl.com Cornea /C15Volume 32, Number 3, March 2013

performed until diffuse haze disappeared on 43 eyes of 30
patients. Slit-lamp photographs were also taken before and
during the PTK procedure for the analysis of deposits in the
remaining stroma.
Three studies were carried out among the 43 eyes of 30
patients. Part I: total stromal thickness for the removal of
diffuse haze was collected and evaluated. Part II: in 29 eyes of
22 patients who had follow-up periods of more than 6 months
from part I, the best spectacle-corrected visual acuity
(BSCVA) was examined before PTK and 1 month post-
operatively. The incidence of recurrences was also evaluated
during the follow-up periods. In Part III, 29 eyes of 22
patients from part I, consisting of a different group of eyes
than those used in part II, were evaluated for actual ablation
depth. These results were then compared with the preopera-
tive diffuse haze depth values from FD-OCT analysis. This
study was performed in accordance with the Declaration of
Helsinki and was approved by the Severance Hospital
Institutional Review Board (IRB 4-2011-0155).
Ablation Depth Evaluation During PTK and
Clinical Evaluation
All PTKs were performed with the VISX S4 IR (VISX
Inc, Santa Clara, CA) by the same surgeon (E.K.K.). The
radiant exposure was 160 mJ/cm2. Slit-lamp photographs
were taken before PTK for reference. After insertion of
a lid speculum, the epithelium was removed mechanically
with a spatula before the excimer laser ablation was
performed. An ablation zone of 6.0 mm with no transition
zone, at a pulse rate of 10 Hz, was used for all eyes. The
density and depth of the remaining diffuse stromal haze were
checked and photographed after each 5- to 10- mm excimerlaser ablation to prevent overablation. Preservative-free
hydroxypropyl methylcellulose eye drops (Tears Naturale
Free; Alcon Laboratories, Fort Worth, TX) were instilled dur-
ing ablation every 10 mm to ensure smoothness of the abla-
tion. PTK was performed until the diffuse stromal haze was
no longer visible (Fig. 1) and the depth of ablation recorded.
Postoperatively, a bandage contact lens was applied and 0.3%
ofloxacin (Ocu flox; Samil Pharmaceutical Co, Seoul, Korea)
and 0.02% fluorometholone (Ocumetholone; Samil Pharma-
ceutical Co) eye drops were instilled 4 times a day for 2
months. The contact lens was removed 4 days after PTK.
Measurement of Haze Depth by FD-OCT
FD-OCT with a resolution of 5.0 mm( R T V u e – 1 0 0 ;
Optovue Inc, Fremont, CA) was performed preoperatively
for the measurement of the depth of diffuse stromal haze in
29 corneas of 22 patients with GCD2 after August 2008. In the
central 6.0 mm of the cornea, an ophthalmologist unrelated to
this study measured the perpendicular distance from the surface
of the Bowman layer to the margin of the deepest haze.
Statistical Analysis
Visual acuity results were converted to decimal equiv-
alents for analysis. All data were analyzed using SAS software
(version 9.2; SAS Institute Inc, Cary, NC). The paired ttest
was used to compare BSCVA before and after PTK. To ana-
lyze the difference and correlation between the actual ablated
depth during PTK and the depth of diffuse stromal haze mea-
sured by FD-OCT, the paired ttest, intraclass correlation coef-
ficient (ICC), and Bland –Altman plots were used. A Pvalue
,0.05 was considered statistically signi ficant.
FIGURE 1. Serial photographs taken before and during PTK for diffuse stromal haze in GCD2. The preoperative slit-lamp pho-
tograph of the right eye of a 69-year-old woman shows a few discrete large and small opacities with diffuse stromal haze between
opacities (A). B–E, Slit-lamp photographs taken after PTK ablation depths of 21, 32, 39, and 42 mm, respectively. F, Slit-lamp
photograph taken 2 months after PTK.Cornea /C15Volume 32, Number 3, March 2013 Phototherapeutic Keratectomy in GCD2
/C2112012 Lippincott Williams & Wilkins www.corneajrnl.com |297

RESULTS
Forty-three eyes of 30 patients (11 men and 19 women)
were included in this study. PTK for the removal of diffuse
stromal haze was performed in all patients without compli-
cations. The mean age of the patients was 62.0 ± 8.4 (SD)
years (range, 42 –76 years). There was a mean hyperopic shift
of 0.83 ± 0.96 diopters on manifest refraction after PTK in 23
eyes of 18 patients. Refractive error, however, could not be
determined before PTK in 20 eyes of 14 patients secondary to
severe corneal opacities and/or cataract.
In part I of this study, 43.7 ± 6.2 mm (range, 31 –59mm)
was needed to remove diffuse haze (Table 1). In part II of this
study, the mean age of the 22 patients (29 eyes, 9 men and 13
women) was 61.4 ± 8.3 years (range, 47 –76 years). The mean
PTK ablation depth was 43.4 ± 6.6 mm (range, 32 –59mm).
The mean follow-up period of the patients after PTK was
21.0 ± 12.0 months (range, 6 –40 months). There were no
patients requiring repeat PTK for severe exacerbation or
recurrence of GCD2 during the follow-up period. The mean
BSCVA of the patients was 0.43 ± 0.15 (range, 0.2 –0.7)
preoperatively and 0.71 ± 0.16 (range, 0.3 –1.0) 1 month post-
operatively. There was a statistically signi ficant improvement
in BSCVA from the preoperative examination to the postop-
erative examination ( P= 0.022; Table 1).
In part III of this study, FD-OCT was evaluated in 29
eyes of 22 patients (11 men and 11 women; mean age, 64.3 ±
7.8 years; range, 42 –76 years). The preoperative depth of
diffuse stromal haze, measured by FD-OCT, was 44.3 ±
6.4mm (range, 34 –56mm). The mean ablation depth required
to remove diffuse stromal haze was 44.5 ± 5.9 mm (range,
32–55mm). There was no statistical difference between the
actual ablated depth during PTK and the preoperative depth
of haze measured by FD-OCT ( P= 0.78; Tables 1 and 2;
Fig. 2). The Bland –Altman plot and ICC both showed that the
actual ablated depths were very consistent with the haze
depths detected by FD-OCT preoperatively (ICC = 0.719;
Table 2; Fig. 3).
DISCUSSION
This study shows that diffuse haze can be removed
successfully with PTK to a depth of 31 to 59 mm (mean,
43.7 ± 6.2 mm) and that the necessary ablation depth canbe predicted accurately by FD-OCT. Nevertheless, we peri-
odically checked for the presence of residual haze with the slit
lamp during ablation because the resolution of the laser opti-
cal system did not provide suf ficient resolution for our
needs.10As the ablation depth was always more than 31
mm, the surgeon can reduce the frequency of slit-lamp exami-
nations by beginning them after 30 mm of ablation when
FD-OCT is not available. With FD-OCT, we recommend
beginning slit-lamp examination at 10 mm less than the
measured depth of diffuse haze.
We postulated that the diffuse stromal haze, always
located centrally and widely so as to block the visual axis,
would be responsible for decreasing vision in GCD2. In
previous studies, diffuse stromal haze and granular deposits
were always adjacent to the Bowman layer, whereas lattice
deposits were located 65.4 ± 48.0 mm from the Bowman
layer. The thickness of diffuse haze, granular, and lattice
deposits were 47.7 ± 10.2, 91.3 ± 39.5, and 313 ± 71.4 mm,
respectively,2showing that diffuse stromal haze had the clin-
ical features of being both the most super ficial and the thinnest
among the 3 kinds of deposits. Therefore, the removal of
diffuse stromal haze could be easier and with less tissue loss
compared with granular or lattice deposits with PTK.
Every refractive procedure involving ablation of the
visual axis with excimer laser is known to be contra-
indicated in patients with GCD2.6–9,14,15Jun et al7reported
a recurrence of corneal opacities 12 months or later after
laser in situ keratomileusis in 13 eyes of 7 relatively young
GCD2 patients (mean age, 26.4 years). Lee et al,8in 15 eyes
of 8 relatively young GCD2 patients (mean age,
27.6 years), reported that recurrence of disease was first
o b s e r v e d1t o3y e a r s( m e a n ,2 3 . 1±9 . 1m o n t h s )a f t e rl a s e r
epithelial keratomileusis. However, Inoue et al6reported
a recurrence-free interval of 38.4 ± 6.2 months (range,
30–49 months) in 7 eyes of 4 heterozygous GCD2 patients
(mean age, 66 years) after PTK. In our series of elderly
patients (mean age, 61.4 ± 8.3 years), the BSCVA improve-
ment after PTK was statistically signi ficant, and no patients
needed repeat PTK for severe recurrence or exacerbation
of disease through the latest follow-up visit (mean, 21.0 ±
12.0 months; range, 6 –40 months). Even though the postop-
erative mean follow-up period was relatively short, these
results suggest that PTK can be performed in GCD2 corneas
TABLE 1. Demographics and Clinical Results of Patients With GCD2 in This Study
Part of study I II III
No. eyes/patients43/30 29/22 29/22
Sex (M:F)11:19 9:13 11:11
Mean ± SD Range Mean ± SD Range Mean ± SD Range
Age (y) 62.0 ± 8.4 42 –76 61.4 ± 8.3 47 –76 64.3 ± 7.8 42 –76
Depth of PTK ablation ( mm) 43.7 ± 6.2 31 –59 43.4 ± 6.6 32 –59 44.5 ± 5.9 32 –55
Follow-up period (mo) —— 21.0 ± 12.0 6 –40 ——
Preoperative BSCVA —— 0.43 ± 0.15 0.2 –0.7 ——
Postoperative BSCVA —— 0.71 ± 0.16 0.3 –1.0 ——
F, female; M, male.Jung et al Cornea /C15Volume 32, Number 3, March 2013
298 |www.corneajrnl.com /C2112012 Lippincott Williams & Wilkins

of elderly patients with caution to improve visual acuity and
postpone keratoplasty.
We found that the depth of diffuse stromal haze
deposits measured by FD-OCT (44.3 ± 6.4 mm) corre-
sponded well to the actual minimal required depth of PTK
ablation (44.5 ± 5.9 mm) performed with serial ablations
using frequent slit-lamp observation. Ef ficiency of excimer
l a s e ra b l a t i o ni sk n o w nt ob ei n fluenced by deposit charac-
teristics such as depth and density.16Swelling of the stroma
secondary to adding methylcellulose eye drops during abla-
tion could also be a contributing factor in fluencing any
difference seen between the actual ablation depth and the
calculated depth. Swelling of corneal stroma, however, was
reported to occur mainly at the posterior stroma because of
the difference in the glycosaminoglycans between the ante-
rior and posterior stroma.17Because diffuse haze in GCD2
is typically shallow and located in the anterior stroma, one
would anticipate a good correlation between the actual
required ablation depth and the depth of haze measured by
FD-OCT. Fine adjustment with additional slit-lamp examina-
tion, however, might be needed to minimize the ablationdepth. We successfully performed PTK to a depth as deep
as 59 mm to remove diffuse haze in 1 case. The maximum
ablation depth, however, needs more research.
There are some limitations to this study. Applying our
results to GCD2 patients without diffuse stromal haze could
be problematic. We hope, however, that our results will
validate the use of FD-OCT to estimate the required ablation
depth needed to remove diffuse haze in eyes with advanced
GCD2 deposits.
FIGURE 2. Representative photo-
graphs showing the correlation
between the PTK ablation depth and
the FD-OCT depth of the diffuse
stromal haze. The preoperative slit-
lamp photograph of the left eye of
a 64-year-old woman shows multi-
ple granular and lattice opacities
with diffuse stromal haze (A). B, After
a 41- mm excimer laser ablation,
some granular and lattice opacities
remained; however, diffuse stromal
haze was removed. C, Preoperative
FD-OCT image along the black
a r r o wo f( A ) .T h ed e p t ho fd i f f u s e
stromal haze was measured at 38 mm
from Bowman layer.TABLE 2. Clinical Results of the 29 Eyes of 22 Patients (Part III)
That Had FD-OCT Performed Preoperatively
Depth of PTK
Ablation ( mm)FD-OCT Depth
of Diffuse Stromal
Haze ( mm) p* ICC
Mean ± SD 44.5 ± 5.9 44.3 ± 6.4 0.78 0.719
Range 32 –55 34 –56 ——
*Paired ttest.
FIGURE 3. Bland–Altman plot showing the difference
between the depth of PTK ablation and the haze depth
measured by FD-OCT. The line of mean difference (solid line)
and limits of ±1.96 SD (dotted line) are shown.Cornea /C15Volume 32, Number 3, March 2013 Phototherapeutic Keratectomy in GCD2
/C2112012 Lippincott Williams & Wilkins www.corneajrnl.com |299

REFERENCES
1. Holland EJ, Daya SM, Stone EM, et al. Avellino corneal dystrophy.
Clinical manifestations and natural history. Ophthalmology . 1992;99:
1564–1568.
2. Hong JP, Kim TI, Chung JL, et al. Analysis of deposit depth and
morphology in granular corneal dystrophy type 2 using fourier domain
optical coherence tomography. Cornea . 2011;30:729 –738.
3. Seitz B, Behrens A, Fischer M, et al. Morphometric analysis of deposits
in granular and lattice corneal dystrophy: histopathologic implications for
phototherapeutic keratectomy. Cornea . 2004;23:380 –385.
4. Jee DH, Lee YD, Kim MS. Epidemiology of corneal dystrophy in Korea.
J Korean Ophthalmol Soc . 2003;44:581 –587.
5. Rapuano CJ. Excimer laser phototherapeutic keratectomy. Int Ophthal-
mol Clin . 1996;36:127 –136.
6. Inoue T, Watanabe H, Yamamoto S, et al. Recurrence of corneal dystro-
phy resulting from an R124H Big-h3 mutation after phototherapeutic
keratectomy. Cornea . 2002;21:570 –573.
7. Jun RM, Tchah H, Kim TI, et al. Avellino corneal dystrophy after LA-
SIK. Ophthalmology . 2004;111:463 –468.
8. Lee JH, Stulting RD, Lee DH, et al. Exacerbation of granular corneal
dystrophy type II (Avellino corneal dystrophy) after LASEK. J Refract
Surg. 2008;24:39 –45.9. Park SH, Mok J, Joo CK, et al. Heterozygous Avellino corneal dystrophy
9 years after photorefractive keratectomy: natural or laser-induced accel-
erated course? Cornea . 2009;28:465 –467.
10. Kim EK. PTK in corneal dystrophy. Cornea . 2004;23:323 –324.
11. Wojtkowski M, Leitgeb R, Kowalczyk A, et al. In vivo human retinal
imaging by Fourier domain optical coherence tomography. J Biomed
Opt. 2002;7:457 –463.
12. Kim TI, Hong JP, Ha BJ, et al. Determination of treatment strategies for
granular corneal dystrophy type 2 using Fourier-domain optical coher-
ence tomography. Br J Ophthalmol . 2010;94:341 –345.
13. Gaucher D, Saleh M, Sauer A, et al. Spectral OCT analysis in Bietti
crystalline dystrophy. Eur J Ophthalmol . 2010;20:612 –614.
14. Lee WB, Himmel KS, Hamilton SM, et al. Excimer laser exacerbation of
Avellino corneal dystrophy. J Cataract Refract Surg . 2007;33:133 –138.
15. Han KE, Kim TI, Chung WS, et al. Clinical findings and treatments of
granular corneal dystrophy type 2 (Avellino corneal dystrophy): a review
of the literature. Eye Contact Lens . 2010;36:296 –299.
16. Kornmehl EW, Steinert RF, Pulia fito CA. A comparative study of masking
fluids for excimer laser phototherapeutic keratectomy. Arch Ophthalmol .
1991;109:860 –863.
17. Meek KM, Leonard DW, Connon CJ, et al. Transparency, swelling and
scarring in the corneal stroma. Eye (Lond) . 2003;17:927 –936.Jung et al Cornea /C15Volume 32, Number 3, March 2013
300 |www.corneajrnl.com /C2112012 Lippincott Williams & Wilkins