Idiopathic Unilateral Corneal Endothelial Failure in Phakic Eyes (845)
Download
Report
Transcript Idiopathic Unilateral Corneal Endothelial Failure in Phakic Eyes (845)
Idiopathic Unilateral Corneal
Endothelial Failure in Phakic Eyes
Adrian Tey MBChB, MRCOphth1;
Weng Lee FRCSEd(Ophth)2;
Hardeep Mudhar BSc,PhD,MBBChir,FRCPath3;
Francisco Figueiredo,MD,PhD,FRCOphth1
1. Department of Ophthalmology, Royal Victoria Infirmary, Newcastle, UK ;
2. Hermitage Medical Clinic, Dublin, Ireland ;
3. National Specialist Ophthalmic Pathology Service, Royal Hallamshire
Hospital, Sheffield, UK
The authors have no financial interest in the subject matter of this
poster
Introduction
Bullous keratopathy occurs as a result of corneal endothelial cell
failure and can occur unilaterally or bilaterally.
Unilateral bullous keratopathy can be Iatrogenic e.g. following
intraocular surgery, mitomycin C toxicity and laser iridotomies or as
a result of pathological diseases e.g. Fuch’s endothelial dystrophy
(FED), posterior polymorphous corneal dystrophy (PPD),
iridoschisis,
iridocorneal
endothelial
syndrome
(ICE),
pseudoexfoliation syndrome (PxF) and herpes viruses.
Purpose
To describe a cohort of consecutive phakic patients with
progressive, irreversible, unilateral bullous keratopathy (BK) of
undetermined aetiology.
Methods
Patients presenting to the Royal Victoria Infirmary (United Kingdom
(U.K)) corneal clinic were recruited consecutively from January 1999 to
July 2009.
Informed consent was obtained from all patients.
Data was collected retrospectively from patient records for
demographics, thorough medical history, slit lamp ocular examination,
treatment, pachymetry (Mentor) and Specular microscopy (Non contact
Robo Konan, Inc) in 5 zones: central, superior, nasal, inferior and
temporal zones.
All penetrating keratoplasties (where required) were performed by a
single surgeon (FF) utilising standardised technique.
Corneal buttons following penetrating keratoplasty (PK) were
investigated for histology, electron microscopy, immunohistochemistry
(IHC) for herpes viral surface antigen, polymerase chain reaction (PCR)
and in situ polymerase chain reaction (iPCR) for herpes viral DNA using
established protocols1,2.
Aqueous samples were also obtained at the time of surgery for herpes
viral DNA analysis.
1. Cohrs RJ, Barbour M, Gilden DH. Varicella-zoster virus (VZV) transcription during latency in human ganglia: detection of transcript
mapping to genes 21,29,62, and 63 in a cDNA library enriched for VZV RNA. J. Virol 1996;70(5): 2789-2796
2. Lewis F. An approach to In Situ PCR manual. PE Biosystems, Foster City, CA, USA.
Age
Results
A total of 11 patients presented
with unilateral BK.
The group were non contact
lens wearers with no past
ocular, medical, surgical, social
or drug history that may affect
the cornea.
Both eyes were phakic with
normal corneal sensations.
There
were
no
clinical
evidence of PxF, herpetic eye
disease,
FED
and
ICE
syndrome in either eye.
Table 1 summarises the
demographics.
Table 2 summarises corneal
measurement findings.
Mean:
71.7 years
Range: 59 to 80 years
Gender
Male:
2
Female: 9
Affected eye
Ethnic group
Right:
8
Left:
3
Asian:
1
Caucasian: 10
Referral source
Emergency room: 2
Optometrist: 3
Ophthalmologist: 4
General Practitioner: 2
Duration of
symptoms
Mean:
11.8 months
Range: 1.5 months to 36
months
Mean follow up
period
Mean:
82.2 months
Range: 34 to 163 months
Table 1: Patient demographics and presentation
Results
Affected eyes
Non affected eyes
BCVA
6/9 or worse
6/5 to 6/18
(2 patients had ARMD)
Mean central corneal
pachymetry (µm)
684m
536m
(range 595 to 800m)
(range 503 to 590m)
Endothelial cell
density (cells/mm2)
Not recordable
1980cells/mm2
(range 1584 to 2325)
Table 2: Vision and corneal measurements at presentation
Treatment of affected eyes on presentation:
Topical steroid (guttae prednisolone acetate 1%) for a mean period
of 5.6 months (range 1 to 12 months). Therapeutic trial of topical
acyclovir 3% ointment 5 times per day, oral acyclovir 400mg twice a
day or a combination of both for up to 3 months (None showed any
significant improvements).
Results
8 affected eyes eventually underwent PK. The interval from disease
onset to PK ranged from 6 to 67 months (mean 22.5 months). There
were no peri-operative complications.
The PK group had a post-operative follow up of 16 to 120 months
(mean 61.5 months). Table 3 summarises the post-operative course
of the 8 patients.
Mean BCVA of grafted eye: 6/18 (range 6/6 to hand movements)
Two patients with age related macular degeneration accounted for
poor BCVAs of hand movements.
Complications: Controlled secondary glaucoma - 3,
Endothelial rejection - 1,
Grafts failed - 4 ( 3 primary failure, 1 due to
fungal infection)
Only one patient required multiple PKs (2 regrafts).
Two patients have since deceased and 1 was discharged back to
the care of the district general hospital.
Results
Table 3: Penetrating Keratoplasty (PK) group
Patient
Graft
number
Post-op
follow
up (months)
Duration onset
to PK
(months)
Graft Failure / rejection
episodes
Donor
endothelium
(cells/mm2)
A
1
2
3
120
21
16
6
126
147
Graft failure
Fungal infection
B
1
108
30
No
2500
C
1
72
28
No
2400
D
1
63
19
Graft failure
2600
E
1
91
10
Graft failure
2600
F
1
74
13
No
2400
G
1
21
67
Endothelial rejection x 1
2800
H
1
29
7
No
Not recorded
2600
2500
In the non PK group (n = 3), mean central pachymetry of affected
eyes increased by 9.8% to 793m (range 715m to 857m). One
patient opted for amniotic membrane transplant for symptomatic BK
instead of PK in view of her advanced age related macular
degeneration.
Results
Histology & Electron microscopy showed features of chronic oedema in
all cornea samples
Light microscopy
Transmission electron microscopy
Epithelium
Intracytoplasmic swelling, irregularity, bullous
lifting with subepithelial excess basemant
membrane with or without fibrosis
Intracytoplasmic swelling, irregularity,
bullous lifting with subepithelial excess
basemant membrane with or without fibrosis
Bowman’s membrane
Intact
Intact
Stroma
N/A
Focally degenerate keratocytes
Descemet’s
membrane
Thickening and multilayering of Descemet’s
(thick collagen layer lining the posterior nonbanded zone)
Thickening and multilayering of Descemet’s
(thick collagen layer lining the posterior nonbanded zone)
Endothelium
N/A
Microvilli, multilayered endothelial cells were
seen without epithelial differentiation
Figure 1: Light microscopy (LM)
Figure 2: Transmission electron microscopy (TEM)
LM image (haematoxylin and eosin), showing
a. Bullous lifting of the oedematous corneal epithelium (arrow),
b Excess production of epithelial basement membrane (arrow),
C Thickened Descemet’s (green arrow) and endothelial cell loss (black arrow).
Composite TEM figure, showing the range of abnormalities seen in
some of the cases:
a: This shows a thickened, multilayered Descemet’s (commonest
observation) with the elaboration of loose collagen posterior to the
posterior non-banded zone of Descemet’s (black arrow). The
green arrow points to bi-layering of the endothelial cell processes
in this case, between which are scattered desmosomes,
b. Shows an isolated cell, with expanded lysosomes filled with
loose reticular material, The appearance are those of a
macrophage (yellow circle),
c. Shows dense curly collagen in the bottom half of the image(
green arrow), posterior to the non-banded Descemet’s above.
Results
No virus particles, exfoliation material or guttae were identified.
Post-PK (regrafts) corneal buttons showed features of chronic
bullous keratopathy, secondary to chronic endothelial cell loss, with
a thickened, multilayered Descemet’s. There were no specific
features within the failed grafted tissue that exposed the aetiology of
the original corneal disease.
Herpes viral studies on post-PK corneal buttons were negative for all
samples. In addition, aqueous humour PCR confirmed absence of
any herpes viruses (HSV, VZV and CMV)
Conclusions
Despite extensive investigations and long term follow up, the
aetiology of BK for this group of patients remained elusive and
remains a diagnostic challenge.
Medical treatment was not beneficial. PK appears to offer a limited
prognosis as 2 patients developed early graft endothelial failure and
1 patient required 2 further transplants due to fungal keratitis after
the initial PK.
Features of chronic inflammatory component was seen in 2 cases,
however, all other cases showed no inflammation and therefore the
interpretation of inflammation playing a role is speculative.
Whilst some cases harboured some pathological changes that were
reminiscent of CHED and PPD, non of the cases fitted neatly into a
distinct diagnostic category.
We speculate that the clinical entity we observed may yet represent
previously unreported and currently unknown pathology.