Download Long-term Outcomes of Keratolimbal Allograft for Total Limbal Stem

CLINICAL SCIENCES
Long-term Outcomes of Keratolimbal Allograft
for Total Limbal Stem Cell Deficiency
Using Combined Immunosuppressive Agents
and Correction of Ocular Surface Deficits
Lingyi Liang, MD, PhD; Hosam Sheha, MD, PhD; Scheffer C. G. Tseng, MD, PhD
Objective: To determine the long-term outcomes of kera-
tolimbal allograft (KLAL).
Methods: Scores of such risks as infrequent blinking,
blink-related microtrauma, conjunctival inflammation,
elevated intraocular pressure, dry eye, symblepharon, lagophthalmos, and previous KLAL or penetrating keratoplasty (PKP) failure were calculated and recorded before,
during, and after KLAL. Prolonged oral mycophenolate
mofetil and tacrolimus and short-term prednisone and
acyclovir were administered in 12 eyes (10 consecutive
patients) with total limbal stem cell deficiency after KLAL.
Ten eyes underwent subsequent PKP.
Results: More corrective measures were required in eyes
with higher risk scores. During a follow-up of 61.2 months
(standard deviation [SD], 18.2; range, 36-91 months) after KLAL, postoperative epithelial breakdown due to exposure occurred late in the period after PKP and re-
T
Author Affiliations: State Key
Laboratory of Ophthalmology,
Zhongshan Ophthalmic Center,
Ocular Surface Disease Center,
Sun Yat-sen University,
Guangzhou, China (Dr Liang);
Ocular Surface Center, Ocular
Surface Research and Education
Foundation, Miami, Florida
(Drs Liang, Sheha, and Tseng);
and Research Institute of
Ophthalmology, Cairo, Egypt
(Dr Sheha).
mained a primary risk. Mean daily doses of 1.4 g of
mycophenolate mofetil and 1.6 mg of tacrolimus were
administered for 52.7 months (SD, 22.5; range, 23-91
months) with few adverse effects and reached trough levels of 1.6 µg/mL (SD, 0.6 µg/mL) and 4.5 ng/mL (SD, 2
ng/mL), respectively. Keratolimbal allograft and PKP rejection was noted in 2 and 3 eyes, respectively, though
there was a reversal in 1 eye in each group, yielding final KLAL and PKP survivals in 10 and 8 eyes, respectively, and ambulatory visual acuity of up to 20/20 in 10
eyes for 67.2% of the entire follow-up period.
Conclusion: Correction of ocular surface deficits combined with an immunosuppressive regimen further improves the long-term outcome of KLAL in eyes with total
limbal stem cell deficiency.
Arch Ophthalmol. 2009;127(11):1428-1434
HE MAINTENANCE OF A
healthy corneal epithelium
relies on stem cells located at
the limbus.1,2 When limbal
stem cells or their supporting stromal environment is destroyed, corneal blindness occurs.3 Histologically, corneas with limbal stem cell deficiency (LSCD)
carry the hallmark of conjunctivalization associated with chronic inflammation, superficial neovascularization, scarring, and poor
epithelial integrity.4,5 Consequently, patients with LSCD experience severe photophobia and decreased vision.
Conventional penetrating keratoplasty
(PKP) is contraindicated. Visual rehabilitation for LSCD resorts to transplantation
of autologous or allogeneic limbal stem cells,
depending on the laterality, the extent of involvement, and the patient’s compliance.6
For bilateral total LSCD, or in unilateral total
LSCD when the fellow eye is not chosen as
a donor, transplantation of allogeneic limbal stem cells is achieved by either limbal
(REPRINTED) ARCH OPHTHALMOL / VOL 127 (NO. 11), NOV 2009
1428
conjunctival allograft from living, related donors7,8 or keratolimbal allograft (KLAL) from
cadaveric donors,7,9,10 or both.11 Compared with limbal conjunctival allograft from
living, related donors,9,10,12,13 KLAL provides more stem cells without compromising the healthy limbal tissue in a normal eye.
Transplantation of an allograft poses
a high risk of rejection even in HLAmatched recipients. Therefore, KLAL survival depends on systemic immunosuppression for a prolonged, if not indefinite,
period.6 Immunosuppression using topical and systemic steroids and cyclosporine
A cannot halt allograft rejection in highrisk PKP14 or KLAL.13,15-17 Combined systemic immunosuppression based on mycophenolate mofetil and tacrolimus has been
shown to be more effective and safer than
cyclosporine A in promoting the survival in
solid organ transplantations by lessening
acute rejection18 and can reduce rejection
in high-risk PKP.19 However, it remains unclear whether this regimen is also effective
WWW.ARCHOPHTHALMOL.COM
Downloaded from www.archophthalmol.com by VadrevuRaju, on March 12, 2010
©2009 American Medical Association. All rights reserved.
Table 1. Preoperative Clinical Characteristics and Risk Factors
Preoperative Risk Factors and Scores, No.
Case No./
Sex/
Age, y
1/M/41
2/M/46
3/F/59
4/M/70
5/M/65
6/M/28
7/F/64
8/M/50
9/M/27
10/M/52
Eye
Cause
of LSCD
Previous
Operations
Left
Idiopathic
Right
CB
Left
CB
Right
SJS
LKP
Right
CB
AMT; CE⫹IOL
Left
CB
Right
CB
Left
CB
2 PKPs, trabeculectomy
Left
CB
CE⫹IOL, KLAL, AMT, T
Right
CB
T
Left
CB
KLAL, 2 PKPs, CE⫹IOL
Left
CB
AMT, CE⫹IOL, 2 KLALs
Previous
Failure
Infrequent
Conjunctival Increased Dry
Pathogenic
Total
Blinking BRMT Inflammation
IOP
Eye Lagophthalmos Symblepharon PKP KLAL Score
0
1
1
1
1
1
1
1
1
1
1
1
0
0
0
1
1
1
1
1
0
1
0
1
0
1
1
0
1
1
1
0
0
0
1
1
0
0
0
0
1
0
1
1
1
1
1
1
0
0
0
1
0
1
1
0
0
1
1
1
0
0
0
0
0
0
0
0
1
1
0
1
0
0
0
0
0
0
0
0
1
0
0
1
0
0
0
0
0
0
0
2
0
0
2
0
0
0
0
0
0
0
0
0
1
0
1
2
0
2
2
3
4
4
5
5
5
5
7
9
Abbreviations: AMT, amniotic membrane transplantation; BRMT, blink-related microtrauma, CB, chemical burn; CE, cataract extraction; FR, fornix
reconstruction; IOL, intraocular lens implantation; IOP, intraocular pressure; KLAL, keratolimbal allograft; LKP, lamellar keratoplasty; LSCD, limbal stem cell
deficiency; PKP, penetrating keratoplasty; SJS, Stevens-Johnson syndrome; T, tarsorrhaphy.
in prolonging the survival of KLAL and, if so, what the optimal long-term dosage is.
Although KLAL rejection is considered the major cause
of failure,6 it only accounts for a part of reported cases.16,20
In fact, such risk factors as keratinization,21 symblepharon,16 inflammation,22 and dry eye23 have been implicated for KLAL failure. We wonder whether there are asyet-unknown clinical variables that might affect the KLAL
outcome and whether measures taken to correct these
risk factors may augment aforementioned immunosuppression to improve the KLAL outcome.
METHODS
PATIENTS
This study was approved by the Baptist Health South Florida
institutional review board (Miami, Florida) to retrospectively
review the medical records of 10 patients (12 eyes) with LSCD
who were consecutively operated on by 1 surgeon (S.C.G.T.)
at the Ocular Surface Center, Miami, from 2001 through 2005.
All patients completed at least 36 months of follow-up after
KLAL. Their clinical data concerning preoperative diagnosis,
previous operations, measures taken to correct ocular surface
deficits, outcomes, and complications were retrieved. Diagnosis of LSCD was confirmed in 2 eyes by impression cytology
that showed goblet cells on the corneal surface24,25 and in 10
eyes by late fluorescein staining, loss of limbal palisade of Vogt,
and superficial vascularization.
OCULAR SURFACE DEFICIT SCORES
To gauge the overall risk that may lead to graft failure or rejection, we arbitrarily assigned a total score by assigning 1 point for
each of the following 7 ocular surface deficits: infrequent blinking; blink-related microtrauma due to eyelid margin abnormalities, such as malposition, scar, keratinization, or misdirected lashes
as previously defined26; conjunctival inflammation; increased intraocular pressure (IOP); aqueous tear–deficient dry eye detected by the fluorescein clearance test27; lagophthalmos; and
pathogenic symblepharon as recently defined28; as well as 1 point
each for prior KLAL or PKP failure (Table 1).
SURGICAL TECHNIQUES
OF KLAL AND AMNIOTIC
MEMBRANE TRANSPLANTATION
All operations were performed under general anesthesia and
included KLAL and amniotic membrane transplantation,
similar to what has previously been reported with some
modifications. 1 7 Briefly, pannus and scar tissue were
removed from entire limbal and corneal surfaces. The
denuded corneal and limbal and some of the bulbar surface
was covered by a sheet of cryopreserved amniotic membrane
(Bio-Tissue, Miami) with sutures in 10 eyes and with fibrin
glue (Baxter Inc, AG, Vienna, Austria) in 2 eyes. Following a
7.5- to 8.0-mm trephination, the remaining donor ring without HLA matching was prepared for KLAL by trimming off
the excess sclera, removing the posterior half of the stroma,
and tapering both corneal and scleral edges. After KLAL was
secured to the recipient’s conjunctival edge with interrupted
10-0 nylon sutures, another layer of amniotic membrane was
sutured by a purse-string running 10-0 nylon sutures in 11
eyes, or ProKera (Bio-Tissue) was inserted in 1 eye. Temporary sutured tarsorrhaphy was performed in 11 eyes and permanent tarsorrhaphy in 1 eye.
IMMUNOSUPPRESSIVE REGIMENS
All patients received both 1 g of oral mycophenolate mofetil
and 1 mg of tacrolimus twice a day starting 1 week preoperatively and 0.5-mg/kg prednisone daily and 200 mg of acyclovir 5 times daily starting 3 days prior to surgery. Both
prednisone and acyclovir were tapered off 2 to 3 months
postoperatively when the ocular surface was not inflamed.
The trough levels of mycophenolate mofetil and tacrolimus
were checked every 1 to 3 months, and their doses were
adjusted according to ocular surface inflammation and systemic adverse effects that mycophenolate mofetil and tacrolimus are known for based on blood pressure and levels of
creatinine, blood urea nitrogen, complete blood count, platelets, and liver profile evaluated at baseline and every 1 to 3
months. When the ocular surface was not inflamed, dosages
of mycophenolate mofetil and tacrolimus were tapered to 0.5
g and 0.25 mg/kg per day, respectively, and discontinued if
irreversible graft failure or significant adverse effects were
noted.
(REPRINTED) ARCH OPHTHALMOL / VOL 127 (NO. 11), NOV 2009
1429
WWW.ARCHOPHTHALMOL.COM
Downloaded from www.archophthalmol.com by VadrevuRaju, on March 12, 2010
©2009 American Medical Association. All rights reserved.
OTHER POSTOPERATIVE MANAGEMENT
Postoperatively, topical ofloxacin (Allergan, Irvine, California) and
prednisolone acetate, 1% (Allergan), or unpreserved dexamethasone, 0.1%, eye drops were applied 4 times a day. Treatment with
ofloxacin was discontinued when epithelialization was completed
but resumed if there was a corneal epithelial defect. Dexamethasone dosage was tapered off in 2 to 3 months after the ocular surface became uninflamed. The suture for temporary tarsorrhaphy
was removed usually 2 weeks after surgery or when the amniotic
membrane dissolved, and a bandage contact lens, such as Night
and Day (CIBA Vision, Duluth, Georgia) and a scleral lens from
Kontur Kontact Lens Co (Hercules, California) were inserted. During follow-up, increased limbal inflammation or vascular engorgement with or without corneal epithelial defects9 was treated with
subconjunctival injection of 20 mg of triamcinolone (Bristol-Myers
Squibb, Princeton, New Jersey) and hourly topical steroid drops
and an increase in mycophenolate mofetil and tacrolimus dosages
to the preoperative level. Any epithelial breakdown, such as irregular epithelium, superficial punctate keratopathy, or epithelial defect, was promptly treated according to the underlying cause. Infrequent blinking and/or increased scleral show was managed by
encouraging blinking and by punctal occlusion with or without
autologous serum 4 times a day, followed by permanent tarsorrhaphy. A bandage contact lens was used continuously and an autologous serum was used until an intact epithelium remained stable.
For eyes with total punctal occlusion and extended wear of
bandage contact lens, irrigation of preservative-free saline solution
twice daily was instructed. Recurrent trichiasis was corrected by
periodic epilation. Eyelid malposition or deformity, resulting in
blink-related microtrauma and/or lagophthalmos, was corrected
by eyelid reconstruction. Pathogenic symblepharon was treated
by symblepharon lysis, fornix reconstruction, and amniotic membrane transplantation as recently reported.28
OUTCOME MEASURES
The first outcome measure was the change in best-corrected visual acuity from before to after surgery and in the postoperative
period during which ambulatory vision (defined as a best-corrected
visual acuityⱖ20/200) was maintained similar to visual acuities
reported by Dohlman and Terada29 in patients with keratoprosthesis. The second outcome measure was KLAL survival, evidenced
by the maintenance of an intact and stable corneal epithelium postoperatively. In contrast, KLAL failure was indicated by recurrent
late fluorescein staining, conjunctivalization of the cornea, or persistent epithelial breakdown after the aforementioned corrective
measures.17 The third outcome measure was conventionally defined PKP survival, while PKP failure was diagnosed by classic signs
of endothelial or epithelial rejection lines and subepithelial infiltrates, keratic precipitates, stromal edema, and corneal opacity.
RESULTS
These 10 patients included 8 males and 2 females with a
mean age of 50.2 years (standard deviation [SD], 15 years).
Limbal stem cell deficiency was bilateral in 8 and unilateral in 2 patients (cases 9 and 10) and caused by chemical burns (10 eyes [83%]) and Stevens-Johnson syndrome (case 3); case 1 was idiopathic. Table 1 lists cases/
eyes ranked by preoperative risk score. The more severe
cases (cases 6 to 10) had received more prior operations.
Except for case 1, all had infrequent blinking. Blinkrelated microtrauma, conjunctival inflammation, increased IOP, and aqueous tear–deficient dry eye were found
in more than 50% of eyes. Lagophthalmos or pathogenic
symblepharon was found in cases 7 to 10, and prior PKP
or KLAL failure was found in cases 6 to 10.
Preoperatively, measures were taken to correct all ocular surface deficits, except for infrequent blinking (Table 2).
A nonpreservative steroid was applied in 7 eyes with conjunctival inflammation. Intraocular pressure was controlled
medically in 4 eyes and by implantation of a glaucoma drainage device in another 3 eyes. Punctal occlusion, artificial
tears, and/or autologous serum was applied in all 6 eyes with
dry eye. Eyelid reconstruction was performed in 2 eyes with
blink-related microtrauma caused by eyelid malposition.
Fornix reconstruction and amniotic membrane transplantation were performed in 2 eyes with pathogenic symblepharon as recently reported.28 Bandage contact lenses were inserted in 2 eyes with large epithelial defects. Epilation was
also performed in those with misdirected lashes.
Keratolimbal allograft/amniotic membrane transplantation was uneventfully performed in all 12 eyes. Intraoperatively, mitomycin C was applied in 5 eyes with conjunctivalinflammation,andfornixreconstructionwasperformed
in 1 eye with remaining pathogenic symblepharon (Table 2).
Postoperatively, the remaining infrequent blinking in all
patients (except for case 2) was treated by a bandage contact lens or scleral lens immediately after the suture for temporary tarsorrhaphy was removed or when the amniotic
membrane bandage dissolved. For case 2, a bandage contact lens was inserted at the first postoperative visit, ie, 2
months after KLAL, when an epithelial defect was noted.
Punctal occlusion and autologous serum drops were administered in 3 and 5 eyes, respectively (Table 2).
For a mean period of 9.6 months (SD,9.2; range, 1-31
months) after KLAL when the eye was not inflamed, 10 eyes
in 8 cases (except for cases 1 and 3) underwent a total of
13 PKPs. Despite all aforementioned measures, 9 of 12 eyes
still experienced a total of 14 episodes of corneal epithelial
breakdown, including epithelial defects (n=10), superficial punctate keratopathy (n=3), and irregular epithelium
(n=1) during a mean follow-up period of 61.2 months (SD,
18.2; range, 36-91 months) after KLAL. Notably, 13 of 14
episodes happened after PKP; among them 11 developed
4 months to 4 years later. Except for 2 eyes with the highest risk scores (cases 9 and 10), epithelial breakdown was
resolved by additional measures (n=7), including additional
punctal occlusion and autologous serum drops—with or
without permanent tarsorrhaphy in 5 eyes in which epithelial breakdown (cases 2-6) was caused by infrequent blinking with or without increased scleral show. One such
example is shown in Figure 1A and B. The corneal epithelial defect (Figure 1C and D) in 1 eye (case 7) caused
by lagophthalmos and pathogenic symblepharon healed after fornix reconstruction (Figure 1E and F). The epithelial
defect (Figure 1G) in another eye (case 8) caused by lagophthalmos due to skin burn led to the first PKP failure
(Figure 1H), though it was resolved after skin and oral mucosal grafts and permanent tarsorrhaphy. These measures
resulted in a stable corneal epithelium 27 months after a
repeat PKP (Figure 1I and J). A total of 7 (SD, 3; range, 2-12)
corrective measures were required, with more used in eyes
with high risk scores (Table 2). Specifically, 9 eyes with a
risk score of more than 2 needed 6 or more corrective measures, while the remaining 3 eyes with risk scores of less
than 3 needed 5 or fewer corrective measures. The 3 eyes
(REPRINTED) ARCH OPHTHALMOL / VOL 127 (NO. 11), NOV 2009
1430
WWW.ARCHOPHTHALMOL.COM
Downloaded from www.archophthalmol.com by VadrevuRaju, on March 12, 2010
©2009 American Medical Association. All rights reserved.
Table 2. Corrective Measures, Operations, Complications, and Final Outcome
Corrective Measures
Case
No.
Preoperative
1
2
3
4
5
6
7
8
9
10
Intraoperative
T
T
PO, AS,
epilation, LR
AG, epilation,
PO
PO, AS,
epilation
PO, AS, AG,
epilation
Epilation, GDD
GDD, FR, MMC,
BCL
PO, AG,
epilation
PO, AG, BCL
PO, GDD, FR,
MMC,
epilation, LR
Postoperative,
Early/Late a
Total
No.
PKP,
mo
PostKLAL
2
5
12
T
BCL
BCL, b PO,
AS/#T
BCL/#T
3
13
T
BCL, AS
6
MMC, T
BCL, AS/PO
7
5
MMC, T
BCL, AS, PO
7
4
MMC, T
9
4
SPK (1.5 y PKP)
T
BCL, AS/PO,
#T
BCL/PO, AS/#T
7
3
#T
BCL, PO/FR
8
18, 44
BCL/#T/#T/
LR ⫻ 2/LR
10
1,8
BCL/PO
BCL, AS/#T
7
12
31
5, 39
SPK (6 mo PKP),
ED (4 y PKP)
ED (18 mo first
PKP)
ED (14 d first
PKP), ED (1 mo
second PKP),
SPK (8 mo 2nd
PKP), ED (2 y
second PKP)
ED (4 mo PKP)
ED (11 mo second
PKP)
T, LR
MMC, T
MMC, T, FR
Outcome Measures
Complications
Corneal Epithelial
Breakdown (Time
After KLAL/PKP)
ED (2 mo KLAL),
ED (13 mo PKP)
Irregular
epithelium (15
mo PKP)
ED (17 mo PKP)
Rejection
KLAL
PKP
Follow-up,
mo
VA Before;
After
KLAL
PKP
0
0
NA
0
36
78
20/300; 20/80
CF; 20/200
S
S
NA
S
0
0
70
20/300; 20/20
S
S
0
NA
84
20/200; 20/25
S
NA
0
0
60
20/400; 20/40
S
S
0
0
67
HM; 20/20
S
S
1, R
0
67
HM; 20/20
S
S
0
2, R
79
HM; 20/20
S
S
0
0
50
CF; 20/60
S
S
0
0
36
LP; 20/80
S
S
0
2, R, IR
1, IR
1, IR
37
56
CF; HM
LP; 20/400
F
F
F
F
Abbreviations: AG, anti-glaucoma medications; AS, autologous serum; BCL, bandage contact lens or scleral lens; CF, counting fingers; ED, epithelial defect;
F, failure; FR, fornix reconstruction; GDD, glaucoma drainage device implantation; HM, hand motion; IR, irreversible; KLAL, keratolimbal allograft; LP, light
perception; LR, eyelid reconstruction; MMC, mitomycin C; NA, not applicable; PO, punctual occlusion; PKP, penetration keratoplasty; R, reversible; S, successful;
SPK, superficial punctuate keratopathy; T, temporary tarsorrhaphy; #T, permanent tarsorrhaphy; VA, visual acuity.
a Measures taken immediately after KLAL (early) and later when corneal epithelial breakdown ensued (late).
b Starting when ED occurs but not immediately after KLAL.
without postoperative epithelial breakdown had a low risk
score of less than 5; among them 2 did not undergo PKP.
Systemic immunosuppression was continued in all 8
patients in whom it was successful without any known
significant adverse effects, though it was discontinued in
2 patients in whom it had failed because of irreversible PKP
rejection (case 9) and persistent hypertension (case 10). As
a result, it had been used for 52.7 months (SD, 22.5 months)
in all patients, ie, 86% of the entire follow-up period. Persistent hypertension and hyperbilirubinemia in case 10 were
reversed after discontinuation of tacrolimus; transient gastric upset and loss of appetite in case 8 was spontaneously
resolved. During the period without any rejection, the average daily medication doses were 1.5 g (range, 0.8-2 g) and
1.6 g (range, 1.2-2 g) of mycophenolate mofetil and tacrolimus, respectively, resulting in mean trough levels of
1.6 µg/mL (SD, 0.6 µg/mL; range, 0.8-2.7 µg/mL) and
4.5 ng/mL (SD, 2 ng/mL; range, 1.7-9.8 ng/mL), respectively.
Among all 13 PKPs performed in 10 eyes, 3 were repeated
because of a retrocorneal membrane (case 7), persistent corneal epithelial defect (case 8), and irreversible corneal graft
rejection (case 10). As shown in Table 2, 2 eyes (17%) encountered a total of 3 episodes of KLAL graft rejection, which
occurred 11, 36, and 48 months after KLAL; 3 eyes (30%)
experiencedatotalof4episodesofPKPgraftrejection,which
occurred 6, 11, 46, and 70 months after PKP. Rejections occurred in cases 5, 6, 9, and 10, which had risk scores of 5
or higher; 3 of them had more than 1 prior KLAL/PKP failure (Table 1 and Table 2). Three KLAL and PKP graft rejections occurred after discontinuation of mycophenolate
mofetil and tacrolimus owing to noncompliance to or intolerance of the drugs (case 9 and 10). For the remaining
4 rejections, daily doses of 1.1 g (SD, 0.7 g) of mycophenolate mofetil (P=.3) and 1.8 mg (SD, 1.0 mg) of tacrolimus (P=.9) were taken, which is no different from the doses
taken when there was no rejection. The trough level at the
time of KLAL or PKP rejection was not available. Temporally, 5 of 7 episodes occurred shortly after at least 1 of the
following ocular surface deficits: trichiasis, loose sutures,
dry eye, and exposure because of lagophthalmos and exotropia. After the anti-rejection treatment and corrective measures, 2 of 3 KLAL rejections (Figure 2A and B) and 2 of
4 PKP rejections (Figure 2C) were successfully reversed.
All irreversible rejections were noted in 2 eyes of cases 9
and10.Keratolimbalallograftgraftmeltatthe12-to3-o’clock
position limbus was noted in 1 eye (case 10) with severe
chemical burn after KLAL and coincided with the location
of prior scleral ischemia (Figure 2D). This area subsequently
developed partial LSCD, which was confirmed by late fluorescein staining (Figure 2E). However, this eye ended up
with irreversible KLAL and PKP graft rejections (Figure 2F).
Postoperatively, IOP increased in 6 eyes with prior ocular
hypertension, but was controlled by anti-glaucoma medications. There was no postoperative infection.
(REPRINTED) ARCH OPHTHALMOL / VOL 127 (NO. 11), NOV 2009
1431
WWW.ARCHOPHTHALMOL.COM
Downloaded from www.archophthalmol.com by VadrevuRaju, on March 12, 2010
©2009 American Medical Association. All rights reserved.
A
C
E
B
A
B
C
D
E
F
D
F
G
H
I
J
Figure 1. Postoperative management of corneal epithelial breakdown. Poor
contact lens fit and epithelial defect caused by infrequent blinking and
increased scleral show in the right eye of case 2 (A) was corrected by
punctal occlusion, autologous serum drops, and permanent tarsorrhaphy
and remained stable for 63 months thereafter (B). Poor contact lens fit
caused by recurrent symblepharon (C) and lagophthalmos (C, inset) in case
7 resulted in a corneal epithelial defect (D). After symblepharon lysis and
amniotic membrane transplantation, a deep fornix (E), complete eyelid
closure (E, inset), and a stable epithelium were maintained 14 months
thereafter (F). Progressive cicatricial eyelid deformity (G) caused by
lagophthalmos (G, inset) and persistent epithelial defect led to penetrating
keratoplasty graft failure in case 8 (H). Eyelid margin reconstruction (I)
allowed the eye to close completely (I, inset) and the cornea remained clear
27 months after the subsequent penetrating keratoplasty ( J).
For visual rehabilitation, cataract extraction and intraocular lens implantation were performed in 1 eye after KLAL (case 3) and in 6 eyes during PKP. Bestcorrected visual acuity improved from hand movement
to 20/200 (before KLAL) to ambulatory 20/200 to 20/20
in 10 eyes (cases 1 to 8) at the last visit. Furthermore,
the ambulatory vision was maintained for 41 months (SD,
26.9; range, 3-90 months), accounting for 67.2% of the
entire post-KLAL follow-up period. Ten of 12 eyes (83%)
maintained uninflamed ocular surfaces, which is no dif-
Figure 2. Postoperative complications. Perilimbal injection signifying early
keratolimbal allograft (KLAL) rejection (A) was accompanied by several small
epithelial defect spots (B), dry eye, and trichiasis (not shown) in the right eye
of case 5. One week after subconjunctival injection of 20 mg of
triamcinolone, an increase in the dose of mycophenolate mofetil from 500
mg to 1000 mg daily, a change to a soft bandage contact lens, and epilation,
the eye became noninflamed without epithelial defect for 32 months (C).
Keratolimbal allograft failure was associated with diffuse neovascularization
and focal melt at the 12- to 3-o’clock position, where there was scleral
ischemia (D, arrows) 1 year after KLAL in case 10. Vivid late fluorescein
staining suggestive of limbal stem cell deficiency was noted 12 months later
(E). This eye had irreversible KLAL and penetrating keratoplasty graft failure
3 months later (F).
ferent from eyes receiving conjunctival limbal autograft
(Figure 3). The second outcome measure showed that
these eyes also maintained KLAL survival (Figure 3). The
third outcome measure showed that 8 of 10 eyes (80%)
maintained PKP survival, resulting in a stable and smooth
corneal epithelium (Figure 3). The failure in cases 9 and
10, with the highest risk scores, was associated with irreversible KLAL and PKP rejection because of exposure
refractory to all measures taken.
COMMENT
Although KLAL with or without PKP has been advised for
restoring vision in eyes with total LSCD, its outcome is not
simply determined by successful transplantation of limbal
stem cells, but can also be adversely affected by several risk
factors threatening ocular surface health.6 We know that
corneal surface health depends on a stable preocular tear
film that helps supply oxygen to an avascular clear cornea
when the eye is open. To achieve this goal, ocular surface
epithelia, all external adnexal glands, and eyelids work in
concert via neuroanatomic integration by 2 neural reflexes controlling secretion of different tear components
and eyelid blinking/closure.30 The present study illustrates what ocular surface deficits are prevalent and threatening and how corrective measures may augment the longterm success of KLAL for eyes with total LSCD.
Among ocular surface deficits, we have identified infrequent blinking, blink-related microtrauma, conjunctival inflammation, increased IOP, aqueous tear–deficient dry eye,
lagophthalmos, pathogenic symblepharon, and prior KLAL
or PKP failure as potential risk factors. As summarized in
Tables 1 and 2, 11 of 12 eyes initially had 1 or more risk
factors before KLAL. Eyes with higher scores had more prior
failed operations, required more corrective measures, developed more postoperative corneal epithelial breakdown,
showed more KLAL/PKP rejections, resulted in irreversible
rejection, and had overall poor outcomes. Keratinization21,
symblepharon16, inflammation22, and dry eye23 have been
(REPRINTED) ARCH OPHTHALMOL / VOL 127 (NO. 11), NOV 2009
1432
WWW.ARCHOPHTHALMOL.COM
Downloaded from www.archophthalmol.com by VadrevuRaju, on March 12, 2010
©2009 American Medical Association. All rights reserved.
recognized as potential risk factors in eyes receiving KLAL.
In this study, only case 3 had mild eyelid margin keratinization, which caused blink-related microtrauma. Glaucoma
is a known risk factor for PKP.31 Schwartz et al32 proposed
a prognostic staging system according to underlying diseases with or without inflammation. Besides these, we also
identified 3 other risk factors, namely, infrequent blinking,
blink-related microtrauma, and lagophthalmos; the former
2 were present in the overwhelming majority of our cases,
while the last 1 was presented in 3 severe cases (Table 1).
Together, these 3 factors could destabilize the preocular tear
film, hence contributing to postoperative epithelial breakdown and PKP failure as illustrated in case 8 (Figure 1G
and H). Blink-related microtrauma could further damage
the ocular surface epithelium and cause inflammation.26 Future studies are needed to determine whether a risk score
as proposed herein can help predict the outcome in studies with a larger sample size.
Identification of these risk factors helped us formulate
a corrective strategy before, during, and after KLAL. Nine
of 12 eyes required at least 6 types of corrective measures
(Table 2). Similar to what has been reported,15-17,23 punctalocclusion,artificialtears,autologousserum,bandagecontact lenses, and tarsorrhaphy were needed in all patients (except for case 1) (Table 2). Previously, epithelial defect was
considered 1 major postoperative complication in 36% to
73% of eyes receiving KLAL with or without PKP,15-17,23 while
37%16 to 83%23 of these defects lead to PKP failure. We believed that such postoperative complications should also
include superficial punctate keratopathy and any irregular epithelium, as they might precede an epithelial defect
(cases 2, 5, 6, and 8). Previous studies, however, did not
provide information linking epithelial breakdown to PKP.
Herein, we noted that such epithelial breakdown still developed late after PKP in 9 of 12 eyes despite all aforementioned measures. A continuous follow-up to protect against
any insult to the ocular surface cannot be overstated. Such
conventional measures as punctal occlusion, artificial tears,
and autologous serum drops were not sufficient. Extended
bandage contact lens wear not only protected the eye from
blink-related microtrauma during blinking, but also held
a stable precorneal tear film in eyes with infrequent blink.
We did not note any bandage contact lens–associated infectious keratitis during the entire follow-up, when eyes also
received topical antibiotic eye drops only during epithelialization and daily irrigation with preservative-free saline.
The remaining exposure problem due to increased scleral
show was solved by permanent tarsorrhaphy, while problems due to lagophthalmos were solved by other reconstructive operations. As a result, 7 of 9 eyes with epithelial breakdown healed and maintained a stable corneal epithelium
and helped improve the surgical outcome.
Combined immunosuppressive agents also play a role
in our long-term maintenance of a clear and functional graft.
Before KLAL, patients had their routine check-ups with an
internist to rule out underlying diseases, such as significant
hypertension, liver or kidney dysfunction, and hematocytopenia, all of which may affect the use and dosage of systemic immunosuppressants. During immunosuppressant
treatment, internists and transplant specialists were consulted if there was any issue related to toxicity. The goal of
immunosuppression is to keep the eye uninflamed, which
A
B
C
D
E
F
G
H
Figure 3. Preoperative and postoperative appearances of eyes with
successful treatment. A vascularized and scarred cornea of case 3 following
prior lamellar keratoplasty (A) became stable and clear with a best-corrected
visual acuity (BCVA) that improved from 20/200 to 20/25 for 84 months after
keratolimbal allograft (KLAL) (B). A vascularized and scarred cornea in case
4 (C) had BCVA improve from 20/400 to 20/40 and a clear and stable cornea
60 months after KLAL (D). A vascularized and scarred cornea in the left eye
of case 5 (E) had BCVA recover from hand motions to 20/20 with a stable
and clear cornea 67 months after KLAL (F). A vascularized and scarred
cornea of case 6 (G) became clear and safe, with BCVA improving from hand
motions to 20/20 seventy-nine months after KLAL (H).
the ophthalmologist monitors to avoid systemic adverse effects, which are monitored by an internist. Our daily dosages were less than 2 g of mycophenolate mofetil and 2 to
12 mg of tacrolimus, which are routinely used in liver or
renal transplantation, and reached an average trough level
that was also less than the target level of 8 to 13 ng/mL for
tacrolimus.33-35 This might explain partly why there were
fewer and milder adverse effects in these patients than in
those receiving liver or renal transplantation.33-35 Alloway
et al36 also noted that combined mycophenolate mofetil and
tacrolimus resulted in significantly fewer systemic adverse
effects in KLAL patients than renal transplant patients. The
PKP rejection rate was 30%, not significantly different from
46% (28 eyes),15 57% (7 eyes),23 and 64% (14 eyes)25 (all
P⬎.05)whensystemiccyclosporineAwasusedalone.However, the KLAL rejection rate was 17%, which is significantly
lower than the 80% rate in 5 eyes receiving a daily dose of
4.4 mg of tacrolimus alone37 and 87.5% in 15 eyes receiving cyclosporine A alone (both P⬍.01).16 Because 5 of 7
rejections occurred shortly after trichiasis, loose sutures,
(REPRINTED) ARCH OPHTHALMOL / VOL 127 (NO. 11), NOV 2009
1433
WWW.ARCHOPHTHALMOL.COM
Downloaded from www.archophthalmol.com by VadrevuRaju, on March 12, 2010
©2009 American Medical Association. All rights reserved.
dry eye, or remaining exposure, we believe that the primary
cause of rejection was recurring or persistent ocular surface deficits. Indeed, 4 such episodes were reversed after
adjusting immunosuppressant doses and correcting the remaining ocular surface deficits. It is likely that previous failed
KLALs in eyes with the highest risk scores decreased the
success rate of subsequent KLALs despite adnexal surgery,
corrective measures, and strong immunosuppression. Future studies are needed to address this likelihood and to determine whether these eyes require a different immunosuppressive regimen.
Judged by all 3 outcome measures, aforementioned corrective measures and combined immunosuppressive treatment resulted in a long-term (more than 3 years) success
rate of 80% to 83%, significantly higher than the long-term
(over 2 years) success rate of less than 50% reported by others (a total of 121 eyes) (P⬍.05) with various ocular surface diseases13,15-17,23 but comparable with the long-term (an
average of 3 years) success rate of 74.2% in 31 eyes (P⬎.05)
with aniridia11 when oral cyclosporin A alone is administered. Future studies with a larger sample size may help determine the relative importance between measures taken to
correct different ocular surface deficits in various ocular surface diseases and the combined immunosuppressive regimen in attaining the long-term success of KLAL with or without PKP in eyes with total LSCD caused by different diseases.
Submitted for Publication: January 10, 2009; final revision received March 24, 2009; accepted March 30, 2009.
Correspondence: Scheffer C. G. Tseng, MD, PhD, Ocular Surface Center, 7000 SW 97 Ave, Ste 213, Miami, FL
33173 ([email protected]).
Financial Disclosure: Dr Tseng and his family are more
than 5% shareholders of TissueTech Inc, which owns US
patents 6,152,142 and 6,326,019 on the method of preparation and clinical uses of cryopreserved human amniotic membrane and ProKera distributed by Bio-Tissue Inc.
No other author has any proprietary interest in any material mentioned in this study.
Funding/Support: Supported in part by grants
2006B36006002 from the Technological Project Foundation of Guangdong Province and B2006061 from the
Medical Science Fund of Guangdong Province (Dr Liang).
REFERENCES
1. Schermer A, Galvin S, Sun T-T. Differentiation-related expression of a major 64K
corneal keratin in vivo and in culture suggests limbal location of corneal epithelial stem cells. J Cell Biol. 1986;103(1):49-62.
2. Cotsarelis G, Cheng SZ, Dong G, Sun T-T, Lavker RM. Existence of slow-cycling
limbal epithelial basal cells that can be preferentially stimulated to proliferate:
implications on epithelial stem cells. Cell. 1989;57(2):201-209.
3. Lavker RM, Tseng SC, Sun TT. Corneal epithelial stem cells at the limbus: looking
at some old problems from a new angle. Exp Eye Res. 2004;78(3):433-446.
4. Puangsricharern V, Tseng SCG. Cytologic evidence of corneal diseases with limbal stem cell deficiency. Ophthalmology. 1995;102(10):1476-1485.
5. Dua HS, Saini JS, Azuara-Blanco A, Gupta P. Limbal stem cell deficiency: concept, aetiology, clinical presentation, diagnosis and management. Indian J
Ophthalmol. 2000;48(2):83-92.
6. Espana EM, Di Pascuale M, Grueterich M, Solomon A, Tseng SC. Keratolimbal
allograft in corneal reconstruction. Eye. 2004;18(4):406-417.
7. Holland EJ, Schwartz GS. Epithelial stem-cell transplantation for severe ocular
surface disease. N Engl J Med. 1999;340(22):1752-1753.
8. Daya SM, Ilari L. Living related conjunctival limbal allograft for the treatment of
stem cell deficiency. Ophthalmology. 2001;108(1):126-133.
9. Tsai RJF, Tseng SCG. Human allograft limbal transplantation for corneal surface
reconstruction. Cornea. 1994;13(5):389-400.
10. Tsubota K, Toda I, Saito H, Shinozaki N, Shimazaki J. Reconstruction of the corneal epithelium by limbal allograft transplantation for severe ocular surface
disorders. Ophthalmology. 1995;102(10):1486-1496.
11. Holland EJ, Djalilian AR, Schwartz GS. Management of aniridic keratopathy with
keratolimbal allograft: a limbal stem cell transplantation technique. Ophthalmology.
2003;110(1):125-130.
12. Tan DTH, Ficker LA, Buckley RJ. Limbal transplantation. Ophthalmology. 1996;
103(1):29-36.
13. Holland EJ. Epithelial transplantation for the management of severe ocular surface disease. Trans Am Ophthalmol Soc. 1996;94:677-743.
14. Hill JC. Systemic cyclosporine in high-risk keratoplasty: short- versus longterm therapy. Ophthalmology. 1994;101(1):128-133.
15. Tsubota K, Satake Y, Kaido M, et al. Treatment of severe ocular surface disorders with corneal epithelial stem-cell transplantation. N Engl J Med. 1999;340
(22):1697-1703.
16. Ilari L, Daya SM. Long-term outcomes of keratolimbal allograft for the treatment of
severe ocular surface disorders. Ophthalmology. 2002;109(7):1278-1284.
17. Solomon A, Ellies P, Anderson DF, et al. Long-term outcome of keratolimbal allograft with or without penetrating keratoplasty for total limbal stem cell deficiency.
Ophthalmology. 2002;109(6):1159-1166.
18. Boratyńska M, Banasik M, Patrzalek D, Klinger M. Conversion from cyclosporinebased immunosuppression to tacrolimus/mycophenolate mofetil in patients with
refractory and ongoing acute renal allograft rejection. Ann Transplant. 2006;
11(2):51-56.
19. Tabbara KF. Pharmacologic strategies in the prevention and treatment of corneal transplant rejection. Int Ophthalmol. 2008;28(3):223-232.
20. Maruyama-Hosoi F, Shimazaki J, Shimmura S, Tsubota K. Changes observed in
keratolimbal allograft. Cornea. 2006;25(4):377-382.
21. Holland EJ, Schwartz GS. The evolution of epithelial transplantation for severe
ocular surface disease and a proposed classification system. Cornea. 1996;
15(6):549-556.
22. Samson CM, Nduaguba C, Baltatzis S, Foster CS. Limbal stem cell transplantation
in chronic inflammatory eye disease. Ophthalmology. 2002;109(5):862-868.
23. Shimazaki J, Shimmura S, Fujishima H, Tsubota K. Association of preoperative
tear function with surgical outcome in severe Stevens-Johnson syndrome.
Ophthalmology. 2000;107(8):1518-1523.
24. Prabhasawat P, Tseng SCG. Impression cytology study of epithelial phenotype
of ocular surface reconstructed by preserved human amniotic membrane. Arch
Ophthalmol. 1997;115(11):1360-1367.
25. Tseng SCG, Prabhasawat P, Barton K, Gray T, Meller D. Amniotic membrane transplantation with or without limbal allografts for corneal surface reconstruction in patients with limbal stem cell deficiency. Arch Ophthalmol. 1998;116(4):431-441.
26. Cher I. Blink-related microtrauma: when the ocular surface harms itself. Clin Experiment Ophthalmol. 2003;31(3):183-190.
27. Pflugfelder SC, Tseng SCG, Sanabria O, et al. Evaluation of subjective assessments and objective diagnostic tests for diagnosing tear-film disorders known
to cause ocular irritation. Cornea. 1998;17(1):38-56.
28. Kheirkhah A, Blanco G, Casas V, Hayashida Y, Raju VK, Tseng SC. Surgical strategies for fornix reconstruction based on symblepharon severity. Am J Ophthalmol.
2008;146(2):266-275.
29. Dohlman CH, Terada H. Keratoprosthesis in pemphigoid and Stevens-Johnson
syndrome. Adv Exp Med Biol. 1998;438:1021-1025.
30. Tseng SC, Tsubota K. Important concepts for treating ocular surface and tear
disorders. Am J Ophthalmol. 1997;124(6):825-835.
31. Ayyala RS. Penetrating keratoplasty and glaucoma. Surv Ophthalmol. 2000;45(2):
91-105.
32. Schwartz GS, Gomez JAP, Holland EJ. Pre-operative staging of disease severity.
In: Holland EJ, Mannis MJ, eds. Ocular Surface Disease Medical and Surgical
Management. New York, NY: Springer; 2002:158-167.
33. Jorge S, Guerra J, Santana A, Mil-Homens C, Prata MM. Mycophenolate mofetil: ten
years’ experience of a renal transplant unit. Transplant Proc. 2008;40(3):700-704.
34. Brum S, Nolasco F, Sousa J, et al. Leukopenia in kidney transplant patients with
the association of valganciclovir and mycophenolate mofetil. Transplant Proc.
2008;40(3):752-754.
35. Müller H, Solari S, Zuniga C, et al. Immunosuppression with generic tacrolimus
and mycophenolate mofetil in renal transplant recipients: preliminary report in
Chile. Transplant Proc. 2008;40(3):705-707.
36. Alloway R, Mogilishetty G, Cole L, et al. Ocular surface transplant recipients experience minimal immunosuppression complications: implications for composite tissue transplants [abstract]. Am J Transplant. 2007;1058:419.
37. Sloper CM, Powell RJ, Dua HS. Tacrolimus (FK506) in the management of highrisk corneal and limbal grafts. Ophthalmology. 2001;108(10):1838-1844.
(REPRINTED) ARCH OPHTHALMOL / VOL 127 (NO. 11), NOV 2009
1434
WWW.ARCHOPHTHALMOL.COM
Downloaded from www.archophthalmol.com by VadrevuRaju, on March 12, 2010
©2009 American Medical Association. All rights reserved.