SURGICAL TREATMENT FOR MACULAR DEGENERATION
MAHESH
SHANMUGAM, NITANT SHAH
Medical Research Foundation 18,
College Road, Chennai600 006.
The
sequence of events leading to disciform scaning and permanent loss of central
vision is well recognised. Surgeons are now attempting to intervene to disrupt
this sequence, reduce scar formation and photoreceptor damage and improve visual
outcome. Submacular surgery has shown encouraging results in presumed ocular
histoplasmosis (POHS), inflammatory and idiopathic CNVM. This review article
discusses the current indications and outlines the surgical techniques.
INTRODUCTION
Choroidal neovascular membrane is
an important cause of loss of central visual function in adults. The vascular
buds from the choroid grow through the Bruchs membrane, disrupt the macular
anatomy including the critical photoreceptor-retinal pigment epithelium interface,
leak serum and formed blood elements and lead to irreversible loss of overlying
photoreceptors.
One of the principal limiting factors in the laser treatment of CNVM is the
concomitant damage to the overlying sensory retina, which is particularly harmful
when the membrane is under the centre of the fovea.
The MPS group published data on the treatment of subfoveal CNVM in AMD. Both
perifoveal laser treatment and subfoveal obliteration were believed to be beneficial
when compared to no treatment.[1] Despite its documented efficacy
Laser photocoagulation has several limitations viz:
1. Only a minority of patients with CNV due to AMD meet the strict MPS criteria
for treatment; associated haemorrhage especially is problematic.
2. Though treatment benefit has been shown, laser for subfoveal CNVM not only
ablates the neovascular tissue but also the overlying neurosensory retina. Thus
an absolute scotoma and a decreased visual acuity are inevitable consequences
of this treatment.
3. Complications like sudden decline in vision, persistence and recurrence may
occur.
To circumvent these problems some investigators pursued subretinal surgery as
an alternative therapy. Surgical removal is a definitive alternative means of
removing the CNVM with potentially less damage to the neurosensory retina.
Historical perspective
In 1988, de Juan and Machemer[2]
described 4 cases of subretinal scar removal using a large circumferential retinotomy
temporal to the macula. However they had a high rate of complications including
poor vision, retinal detachment and PVR.
In 1991 Binder et al[3] removed subretinal tissue in 2 patients
with advanced AMD by combining preoperative laser and a large retinotomy to
perform transplantation of homologous RPE grafts and transposition of autologous
pedicle grafts. The authors reported some improvement in post-operative vision.
In 1991, Thomas and Kaplan[4] emphasized the removal of CNVM
through a very small retinotomy and reported good postoperative results in 2
patients with POHS.
The SST (Submacular surgery trials) was begun as a randomised, multicentre clinical
trial to evaluevaluate the feasibility of a larger study to determine the efficacy
of surgical removal of subfoveal CNVM.
The submacular surgery trials (SST)[4]
The pilot study was expanded to a full phase trial in 1997 in which 250 patients
are to be enrolled and followed up for 4 years. The main aim was to evaluate
submacular surgery systematically and compare the results to observation in
eyes with subfoveal CNV for which laser treatment is judged inappropriate. The
SST enrolled patients in 3 subsets viz:
1. New subfoveal CNV in AMD trial: These included well-demarcated subfoveal
CNV larger than 3.5 MPS disc areas but not larger than 9 disc areas and poorly
demarcated subfoveal CNV upto 9 MPS areas.
2. Haemorrhagic subfoveal CNV in AMD trial: These included subfoveal lesions
greater than 3.5 MPS disc areas in which at least 50% of the visible lesion
is haemorrhage.
3. Histoplasmosis/idiopathic subfoveal CNV trial: These included cases of subfoveal
CNV upto 9 MPS areas secondary to POHS or idiopathic causes.
Current indications for surgery
With the advent of PDT however an
important question that arose was whether submacular surgery should be done
when non-invasive techniques were available worldwide. The TAP[5]
and VIP[6] study groups have already shown the benefit of PDT
with verteporfin in subfoveal CNVM due to AMD and pathological myopia respectively.
To address this issue the SST data and safety monitoring committee (DSMC) made
the following recommendations for enrollment of patients in the SST.[7]
1. SST Group H : As PDT with verteporfin has not been shown to be beneficial
in CNVM due to POHS or of idiopathic cause, the DSMC recommended the enrollment
of such patients for surgical trials without any change. Though the 2 current
arms of this group are surgery and observation, the investigators feel that
trials to compare surgery and PDT directly may be considered in future.
2. SST Group B : The DSMC has recommended continued enrollment of eligible
patients with AMD and subfoveal lesions that are predominantly blood i.e. the
area of blood occupies at least 50% of the entire subfoveal lesion. Such lesions
have not been included in PDT trials as it is unlikely that the laser light
will penetrate enough to activate the dye. The rationale for inclusion in the
SST is that the visual outcome with surgery may be more favourable than natural
history because clearing subretinal haemorrhage from the fovea may limit the
destruction of photoreceptors secondary to the exposure to blood.
3. SST Group N : The DSMC recommended continued enrollment of patients
with AMD who have subfoveal lesions that have some classic CNV but are not predominantly
classic. Such patients experienced no visual acuity benefit with PDT compared
to observation. For patients with predominantly classic lesions PDT has been
shown to reduce the risk of moderate and severe visual loss for at least 2 years.
Hence such patients are informed about the 2 treatment options and each patient
then decides whether to enrol in the SST or to undergo PDT.
Case selection for surgery in AMD with subfoveal membranes
Intraoperative preservation of foveal RPE and choriocapillaries is the single
most important predictor of good visual outcome. This is possible if neovascular
complexes are selected which lie anterior to the RPE and involve focal defects
in Bruchs membrane. Gass has classified such membranes as type 2 while those
posterior to the RPE are classiclassified as type 1.[8]
Aetiology
Type 2 membranes are found in CNVM
due to:
1. POHS : These are considered as the best candidates for surgical intervention
and good visual outcome.
2. Multifocal choroiditis.
3. Punctate inner choroidopathy.
4. Idiopathic membranes.
Type 1 membranes are found in
1. ARMD-widespread nature of the disease involving Bruchs membrane results in
extensive fibrovascular membrane growth posterior to the RPE.
2. Angioid streaks also involve widespread areas of the RPE and Bruchs.
3. Myopia.
These are not ideal for submacular surgery.
Certain clinical and angiographic
characteristics suggest that the neovascular membrane has an anterior location
viz :
Clinical features
1. Well defined edge of membrane
2. CNVM seen anterior to the plane
of RPE on stereoscopic viewing.
3. A thin rim of blood or a dark pigmented border are indicative of a cleavage
plane between the CNVM and the RPE.
Angiographic features
1. A distinct boundary between the hyperfluorescence
of the membrane and the background choroidal fluorescence.
2. A rim of blocked fluorescence surrounding the hyperfluorescence.
3. A compact membrane with homogeneous fluorescence and regular distinct borders.
4. Absence of late staining in the surrounding tissue beyond the membrane.
5. Stereoscopic viewing of the angiogram demonstrates an anterior location.
Role of surgery for juxtafoveal membranes
For juxtafoveal lesions with CNVM located within 200 mm of the FAZ centre, the
recommended treatment extent is difficult to achieve. These lesions are prone
to either overtreatment or undertreatment and hence are at risk for severe loss
of central vision. Surgical removal of CNVM is an option in patients with such
lesions.[9]
Current surgical technique
The surgical procedure involves a standard three-port pars plana vitrectomy.
After complete vitrectomy the posterior hyaloid is engaged by active suction,
detached and excised. Alternatively it can be removed with a 33 gauge pick (hyaloid
lifter).
|
|
|
|
Fig. 1
|
Fig. 2
|
|
|
|
|
Fig. 3
|
Fig. 4
|
The retinotomy site is selected to
gain the best possible access to the membrane, avoiding damage to major blood
vessels and made with the help of a 130 degrees angled sharply pointed pick.
Diathermy is not necessary prior to retinotomy. The site of retinotomy is chosen
according to the following factors :
1. Exact location of the membrane.
2. Dimensions of the subretinal instruments - especially tip length of angled
instruments is important as it determines how close the retinotomy should be
to the CNVM so that the tips will reach the membrane.
3. Presence of presumed adhesions between the retina and the underlying tissue
e.g. laser scars. (Fig. 1)
The assistant very slowly injects BSS into the sub-retinal space while the surgeon
directs the 33 G bent infusion cannula at the small retinotomy. This creates
a shallow retinal detachment at the posterior pole. Excessive infusion pressure
can cause a break. (Fig. 2)
A 130° bent sub-retinal pick is used to push against the leading edge of
the neovascular complex. It is passed over the anterior surface of the membrane
to break any adhesions to the retina. In some cases a cleavage plane can be
established and the neovascular complex gets separated from the underlying RPE,
while in others the CNVM is inseparably connected with the underlying RPE and
choroid. The horizontal subretinal scissors (130°) can be used to cut firm
adhesions. (Fig. 3)
Then a subretinal forceps is introduced through the retinotomy and the neovascular
complex grasped and extracted. Gentle traction exerted by closing blades of
the forceps is sufficient to break the stalk or adhesion. If this manoeuvre
does not work, subretinal scissors are used to cut the stalk. (Fig. 4)
During any manipulation there is a risk of severing the choroidal vessels, which
can cause a subretinal bleed. The IOP is raised to 80-100 mm Hg for 1 minute
by increasing the bottle height temporarily to stop bleeding. The infusion bottle
is lowered gradually.
Once haemostasis is achieved, the membrane is extracted through the sclerotomy
or cut and aspirated with the vitrectomy probe.
Role of subretinal endophotocoagulation
A 31 G subretinal endolaser probe can be used to obliterate subretinal CNVM.
It is beneficial in
1. Previously lasered CNVM and those associated with inflammatory scars.
2. Small pre-RPE CNVM with large sub-RPE component.
3. For bleeding stalk of the CNVM.
A complete fluid-gas exchange is performed. Air tamponade and prone positioning
favours retinal apposition and minimizes cataract formation.
Postoperative follow-up
The patient is examined on the 1st and 7th day postop to assess the IOP and
rule out any infection.
An angiogram is done 2 weeks after surgery as a baseline for future comparison.
At this stage significant hyperfluorescence (transmitted choroidal hyperfluorescence
through the damaged RPE) is seen. Subsequent angiograms should show a decrease
in hyperfluorescence, if not then recurrence of CNVM is suspected.
Instruments for submacular surgery (Fig. 5)
Visual Outcome
Melberg et al have reported that eyes with lesions with an eccentric ingrowth
(eccentric from the fovea) site have the best visual outcome.[10]
They classified the in growth site. As the ingrowth site is the point of greatest
alteration of ultrastructure causing loss to RPE, choriocapillaries and photoreceptors
it stands to reason that subfoveal membranes with eccentric ingrowth site should
have the best visual prognosis. Eyes with unidentifiable ingrowth sites may
have a guarded prognosis because of more extensive involvement of Bruchs membrane
and adjacent cells.
Complications
1. Recurrence may occur in upto half the
cases in the first 6 months. Eyes with nonsubfoveal recurrences can be treated
with laser, while subfoveal recurrences are surgically treated.
2. Other complications include cataract,
macular pucker, retinal holes, retinal detachment, endophthalmitis, etc.
CONCLUSION
Though submacular surgery shows encouraging results in CNVM caused by POHS,[11]
inflammatory and idiopathic causes; its role in ARMD is limited due to widespread
nature of the disease and as most vascular complexes are posterior to the RPE
surgery entails removal of large amount of RPE cells. In most cases visual benefit
is limited to resolution of the neurosensory detachment and limitation of scotoma
size. However these difficulties may be overcome by RPE transplantation and
foveal translocation, 2 new surgical techniques being tried out at different
centres.
RPE Transplantation
The aim is to facilitate repair of RPE defects that necessarily occur after
removal of the CNVM and prevent choriocapillary atrophy. After surgical excision
of the CNVM following can be used.
- Use of iris pigment epithelium
As it is necessary to transplant RPE cells
in a single layer with proper polarity researchers have concentrated on developing
intact patches of embryonic RPE cells which are grown in culture or dissecting
sheets of RPE derived from cadavers.
The few attempts made so far at RPE transplantation in humans has met with limited
success.
Algvere et al have used RPE suspension transplants and foetal patch grafts in
patients with both forms of ARMD.[12] The transplanted tissue
was tolerated but no significant improvement was noted in any patient.
Subretinal surgery with simultaneous autologous retinal pigment epithelial cells
transplantation consisting of RPE harvested from the nasal subretinal area of
the same eye has recently been reported by Binder et al in 14 eyes of 13 patients.
The BCVA improved 2 or more lines in eight eyes (57.1%) over a median period
of 17 months.
Macular Translocation
The aim is to relocate the fovea i.e. the central neurosensory retina to a new
healthy location so that it may regain some of its visual function. Relocation
of the fovea away from the subfoveal CNVM also enables treatment of the CNVM
by surgical removal or laser destruction (Fig. 6).
Types of macular translocation surgeries
1. Macular translocation with punctate or
no retinotomies with or without chorioscleral shortening.
2. Macular translocation with incomplete
circumferential retinotomy (less than 360 degrees)
3. Macular translocation with 360 degrees retinotomy.
The first technique involving small punctate self-sealing retinotomies is called
limited macular translocation and is currently the preferred technique.
Surgical Principles
Physical separation of the neurosensory retina from the underlying layers.Planned
retinal detachment.Subsequent reattachment.Retinopexy
in case of large retinotomies.Internal
tamponade.
Mechanisms allowing foveal displacement
Disinsertion of the retina from
its anterior attachment: Large retinotomy, 360 degree retinotomy.
Machemer and steinhorst technique[13]
Lensectomy and vitrectomy followed by infusion of fluid into the subretinal
space to achieve retinal detachment.
A peripheral 360° retinectomy is made and the retina is rotated around the
optic nerve and reattached followed by silicone oil tamponade.
Risk of PVR 30%.
Significant tilt in image
Muscle surgery combined to rectify the tilt
Technique by Ninomaya and Tano[14]
To reduce the incidence of PVR a temporal 180 degree retinal flap is created
and elevated to remove the underlying neovascular membrane. The retina is repositioned.
Technique of de Juan[15]
Vitrectomy and induction of retinal
detachment using 40 g needles.
External scleral resection/infolding
in ITQ or INQ to shift the fovea in opposite direction. (Fig. 7)
Fluid air exchange.
Prone positioning given
Advantages
1. Limited macular shift - usually less than
1 disc diameter
2. Surgical risks less than with other techniques
Disadvantages
1. Unpredictable final location of the macula
2. Possibility of folds going across
the macula
Current Status
Macular translocation is a promising treatment method offered to patients with
ARMD and subfoveal CNV of recent onset in order to improve vision to a level,
which makes reading and driving possible. The best candidates have small minimum
desired translocation with a healthy RPE-Bruchs-choriocapillaries complex outside
the lesion.
The surgery may not succeed due to the following reasons:
1. Persistence of the CNVM and inadequate movement of the fovea.
2. Irreversible preoperative retinal damage.
Recurrence of the CNVM.
Intraop and postop complications like
macular hole, retinal breaks, retinal detachment, subretinal-choroidal and vitreous
haemorrhage and macular folds.
The exact role of translocation is not yet clear and multicentred randomized
clinical trials comparing macular translocation with PDT and observation may
probably provide the answers.
REFERENCES
1.Macular photocoagulation study group. Laser photocoagulation
of subfoveal neovascular lesions in age related macular degeneration: results
of a randomized clinical trial. Arch Ophthalmol 1991; 109 : 1220-1.
2.De Juan E, Machemer R. Vitreous surgery for hemorrhagic and
fibrous complications of age related macular degeneration. Am J Ophthalmol 1988;
105 : 25-9.
3.Blinder KJ, Peyman GA, Paris CL, et al. Submacular scar excision
in age related macular degeneration. Int Ophthalmol Clin 1991; 15 : 215-22.
4.SST : Retina Vol. 3 : 2562-2572, third edition, Stephen J
Ryan.
5.Photodynamic therapy of subfoveal choroidal neovascularisation
in age related macular degeneration with verteporfin. TAP report 1. Treatment
of age related macular degeneration with photodynamic therapy (TAP) study group.
Arch Ophthalmol 1999; 117 : 1329-45.
6.Photodynamic therapy of subfoveal choroidal neovascularisation
in pathological myopia with verteporfin: one year results of a randomised clinical
trial - VIP report 1. Verteporfin in photodynamic therapy (VIP) study group.
Ophthalmology.
7.Bressler NM, Hawkins BS, Steinberg P, et al. Are the submacular
surgery trials still relevant in an era of photodynamic therapy? Guest editorial.
Ophthalmology 2001; 108 : 435-6.
8. Gass JDM. Biomicroscopic and histopathological considerations
regarding the feasibility of surgical excision of subfoveal neovascular membranes.
Am J Ophthalmol 1994; 118 : 285-98.
9. Joseph DP, Thomas MA. Surgical treatment of juxtafoveal choroidal
neovascularisation. Invest Ophthalmol Vis Res 1997; 38 : 457.
10. Melberg NS, Thomas MA, Burgess DB. The surgical removal
of subfoveal choroidal neovascularisation: ingrowth site as a predictor of visual
outcome. Retina 1996; 16 : 190-5.
11. Holekamp NM, Thomas MA, Dickinson JD, et al. Surgical removal
of subfoveal choroidal neovascularisation in presumed ocular histoplasmosis:
stability of early visual results. Ophthalmology 1997; 104 : 22-6.
12.Algvere PV, Gouras P, et al. Long term outcome of RPE allografts
in non-immunosuppressed patients with AMD. Eur J Ophthalmol 1999; 9 : 217-30.
13. Machemer R, Steinhorst UH. Retinal separation, retinotomy
and macular relocation. A surgical approach for ARMD? Graefes Arch Clin Exp
Opthalmol 1993; 231 : 635-41.
14. Ninomiya Y, Tano Y. Retinotomy and foveal translocation
for surgical management of subfoveal choroidal neovascular membranes. Am J Ophthalmol
1996; 122 : 613-21.
15. De Juan E Jr, Alexander J. Translocation of the retina for
the management of subfoveal choroidal neovascularisation. Am J Ophthalmol 1998;
125 : 635-46.
