ACTA OPHTHALMOLOGICA SCANDINAVICA 2003

Needling revision of failed

filtering blebs after
trabeculectomy: a retrospective
study
Marit Fagerli, Kjell T. Løfors and Tor Elsås
Department of Ophthalmology, St. Olav University Hospital, Trondheim, Norway

ABSTRACT.
Purpose: To determine the efficacy and safety of needling revision of failed Mitomycin C (MMC) leads to a lower
filtering blebs. intraocular pressure (IOP) than surgery
Methods: We reviewed retrospectively 26 eyes that had undergone needling alone (Kitazawa et al. 1991; Skuta et al.
revision for a failed trabeculectomy. The needling revisions were performed either 1992; Katz et al. 1995; Azuara-Blanco
with adjunctive use of Mitomycin C, 5-Fluorouracil or without antimetabolites. et al. 1997). Postoperative subconjunc-
The procedure was usually performed as a clinic procedure, using a 27-gauge tival injections of 5-Fluorouracil (5-FU)
also improve the success rate of glau-
needle.
coma filtering surgery (Skuta et al. 1987;
Results: The mean follow-up time was 14.5
11.3 months (range 6.0–
Kitazawa et al. 1991; Fluorouracil
48.0 months). Intraocular pressure (IOP) decreased from 28.8
6.8 mmHg
Filtering Surgery Study Group 1989).
(range 19.0–40.0 mmHg) to 15.3
5.2 mmHg (range 7.0–35.0 mmHg). Twelve Several reports advocate needle revi-
eyes (46.2%) achieved success, defined as IOP 18 mmHg without medication; 11 sion with or without adjunctive MMC
eyes (42.3%) achieved qualified success, defined as IOP 18 mmHg with anti- or 5-FU as an effective, simple and low
glaucomatous medication, and three of 26 eyes (11.5%) were classified as fail- cost method of re-establishing filtration
ures. The success rate after the initial needling was 64% at 6 months and the in eyes with failed filtering blebs
same after 1 year and 2 years. The success rate after one or more needlings was (Pederson & Smith 1985; Ewing &
96% at 6 months and 77% at 1 year and 2 years. Complications developed in six Stamper 1990; Shin et al. 1993; Mardelli
of the 26 eyes (23.1%). These involved transient corneal epithelial defects in three et al. 1996; Chen & Palmberg 1997;
eyes (11.5%), temporary conjunctival wound leak in two eyes (7.7%), and Kapetansky & Kapetansky 1999).
development of bullous keratopathy in one high risk eye (3.8%). We present a retrospective study of
Conclusion: Our results are comparable to the results of other studies. Needling transconjunctival needle revision per-
revision appears to be a useful tool in the management of glaucoma. formed on eyes with a failed filtering
bleb after trabeculectomy.
Key words: glaucoma – needling revision – filtering bleb – trabeculectomy – Mitomycin C –
5-Fluorouracil

Acta Ophthalmol. Scand. 2003: 81: 577– 582
Copyright # Acta Ophthalmol Scand 2003.
doi: 10.1046/j.1600-0420.2003.00126.x
Introduction
Trabeculectomy is the standard treat-
ment for patients who have failed max-
imal tolerated medical therapy. Failure
of the filtering bleb occurs at various
times after a trabeculectomy. Scarring
of the filtering bleb is caused mainly by
proliferation of subconjunctival fibro-
blasts and the biosynthesis of collagen
and other extracellular materials (Skuta
& Parrish 1987).
Adjunctive antifibrotic agents are
being used increasingly in glaucoma
filtering surgery. The combination of
trabeculectomy and intraoperative
Patients and Methods
Patient material
We reviewed the records of patients
who underwent needling revision for a
failed filtering operation with at least
6 months of follow-up during the per-
iod from June 1997 to November 2001.
Patients with one or more failed tra-
beculectomies were considered candi-
dates for needling revision if their IOP
was regarded as too high to preserve
the optic nerve. One or more needling
revisions were made if there was an

577
ACTA OPHTHALMOLOGICA SCANDINAVICA 2003
open corneoscleral window and a bleb
that appeared present to some extent.
With increasing experience, in terms of
both technique and knowledge about
how the bleb reacts post needling, we
aimed to attack a diminishing bleb
quite early. This explains why we per-
formed needling on fast-shrinking blebs
with rising pressure in nine eyes less
than 4 weeks after trabeculectomy. In
all eyes, we first tried a period of digital
bulbus massage to enhance the con-
junctival filtration. If the pressure rise
was considered to be caused by a tightly
sutured scleral flap, we performed laser
suture lysis.
Multiple needling revisions were some-
times required because of inadequate
pressure control after the first procedure.
The needling technique is described below.
Bleb status was divided into dome-
shaped cystic blebs (later called cystic
blebs), diffuse low-prominent blebs
(diffuse blebs), and small, almost flat-
tened blebs (flattened blebs).
Needling procedure
After careful slit-lamp examination, top-
ical anaesthesia (Oxibuprocaine; Chauvin,
Essex, England) was applied to the eye
three times, with 1 min between each
drop. Chloramphenicol eye drops were
then applied to the conjunctival surface
twice and some patients were also given
one drop of povidone-iodine 5% solution
(Betadine; Alcon, TX, USA). An aseptic
technique was used.
The procedure was performed in the
clinic with the use of magnification
with loupes or at the slit-lamp. A few
patients were needled under an operat-
ing microscope.
The first cases of needling revision
were performed using MMC (Mardelli
et al. 1996; Kapetansky et al. 1999).
After gaining more experience, we
increasingly used 5-FU in our practice
(Ewing et al. 1990; Shin et al. 1993).
We used MMC 0.02 ml (0.2 mg/ml)
and 0.02 ml of xylocaine with epineph-
rine 2%, making a total volume of
0.04 ml (taken from a larger volume of
1 : 1 mixture of the two agents), drawn
through a 30-gauge needle, attached to
a 1.0-ml syringe.
The injections of 5-FU consisted of
0.1 ml of 5-FU (50 mg/ml) drawn
through a 30-gauge needle attached to
a 1.0-ml syringe.
For both MMC and 5-FU injections,
the subconjunctival space was entered
578
about 10 mm distal to the desired site of
injection. The mixture was then injected
near the intended site of revision, rais-
ing a subconjunctival bleb. The eye was
irrigated with saline solution after the
injection until no backflow of MMC or
5-FU could be seen from the needle
injection hole when examined with the
Seidel test.
After approximately 20–30 min, the
needling procedure was performed. A
sterile and bent 27-gauge needle
attached to a syringe was used, pene-
trating the conjunctiva about 10 mm
from the assumed site of aqueous resist-
ance. The needle was advanced sub-
conjunctivally toward the failed
filtering bleb site, and then over the
scleral flap of the trabeculectomy. This
was repeated 3–10 times using a to-and-
fro movement. If local elevation of the
conjunctiva did not occur, the needle
tip was introduced underneath the
scleral flap to the level of the sclerot-
omy if the scleral flap could be seen
through the conjunctiva. Special care
was taken throughout the procedure
not to buttonhole the conjunctiva or
to penetrate the sclera.
After the procedure, the eye was
examined for aqueous leak through
the needle puncture point or elsewhere,
using the Seidel test. If necessary, the
conjunctival puncture wound was
closed with a brief tamponade with a
cotton applicator (Mardelli et al. 1996).
All patients were placed on a combin-
ation of steroid and antibiotic eye drops,
to be used four times daily for several
weeks (usually 4–12 weeks).
Follow-up
Patients were examined at the slit-lamp
immediately after the needling revision
was performed, again after 1 day and
1 week, and then as needed. For the
purpose of this study, follow-up is
registered after 3 weeks, 2 months and
6 months, and longer for some patients.
The procedure was regarded as suc-
cessful if IOP was 18 mmHg without
any antiglaucomatous medication.
Qualified success was defined as
IOP 18 mmHg with antiglaucomatous
medication. Failure was defined as
IOP 19 mmHg with medication.
Results
The mean age of the subjects was
62.7
21.4 years (range 5–84 years). They
included one 5-year-old female child, 10
women and 15 men. Mean IOP before
revision was 28.8
6.8 mmHg (range
19–40 mmHg). The average time between
the last failed filtration surgery and needle
revision was 1.91
2.48 months (range
0.25–12 months).
There were 18 patients with primary
open-angle glaucoma (OAG), one with
congenital glaucoma and seven with
secondary glaucoma. Among the
patients with secondary glaucoma, one
had pigmentary glaucoma, one had
haemorrhagic glaucoma, one had
Cogan–Reese iris nevus syndrome,
two had chronic uveitis, one had glau-
coma that developed after penetrating
keratoplasty for keratoconus, and one
had glaucoma following serious trauma
to the eye.
Seventeen eyes had undergone one
trabeculectomy, three had undergone
two trabeculectomies, two had under-
gone trabeculectomy three times, one
four times, one five times and two
seven times. Nine eyes had a cystic
bleb, three eyes had a diffuse bleb and
14 eyes had a flattened bleb (Table 1).
Eight eyes had previously undergone
surgery other than filtration surgery.
Six patients had undergone cataract
extraction only. One eye had received
surgical treatment of a retinal detach-
ment and had undergone cataract
extraction associated with prior trauma
to the eye. One patient had undergone a
penetrating keratoplasty, with subse-
quent development of secondary glau-
coma.
Thirteen eyes underwent needle revi-
sion with MMC, 10 eyes with 5-FU and
three eyes underwent needle revision
without the use of antimetabolites.
Sixteen eyes underwent one single
needle revision procedure, four eyes
underwent two procedures. Two eyes
underwent three procedures, two eyes
underwent four, one eye underwent 10
and one eye underwent 13 needle revi-
sions (Table 2).
Three patients who underwent
several needling procedures had their
eye needled with the use of MMC
the first time and later with 5-FU.
This will be mentioned in the Dis-
cussion. For the convenience of this
study, and because we regard MMC
as the most potent antimetabolite,
these patients have been placed in
the group with eyes needled with
MMC. Nineteen of the patients
received subconjunctival injections of
 ACTA OPHTHALMOLOGICA SCANDINAVICA 2003

Table 1. Preoperative patient characteristics.

Eye Age Type of No. of Peroperative use Pre-needling Bleb

no. glaucoma trabeculectomies of MMC during IOP status
performed last trabeculectomy

1 46 pigmentary 7 yes 34 flattened

2 42 CoganReese sec 7 yes 26 flattened
3 5 congenital 3 yes 25 cystic
4 75 pseudophakic 3 yes 23 cystic
5 72 OAG 1 no 32 diffuse
6 72 OAG 1 no 22 cystic
7 79 OAG 1 no 40 cystic
8 62 exfoliative 1 yes 38 cystic
9 54 uveitic 4 yes 24 flattened
10 84 pseudophakic 2 yes 40 flattened
11 75 OAG 1 no 20 cystic
12 75 exfoliative 1 no 25 diffuse
13 84 exfoliative 1 no 28 flattened
14 67 OAG 1 no 30 cystic
15 28 uveitic 5 yes 40 flattened
16 81 haemorrhagic 1 no 20 diffuse
17 80 pseudophakic 2 yes 35 flattened
18 79 exfoliative 1 no 38 flattened
19 76 exfoliative 1 no 29 flattened
20 75 exfoliative 1 no 20 cystic
21 74 pseudophakic 1 no 24 flattened
22 66 exfoliative 1 no 26 flattened
23 62 OAG 1 no 19 flattened
24 27 post-traumatic 2 yes 35 flattened
25 22 secondary 1 yes 30 cystic
26 67 OAG 1 no 26 flattened
Mean
SD 62.7
21.4 2.0
1.8 28.8
6.8

5-FU (5 mg in 0.1 ml) shortly after the
needling procedure and in total
received 2–10 injections.
Mean follow-up from last needling
procedure was 14.5
11.3 months (range
6.0–48.0 months). Intraocular pressure
decreased from 28.8
6.8 mmHg (range
19.0–40.0 mmHg) to 15.3
5.2 mmHg
(range 7.0–35.0 mmHg). Overall, 12 of
26 eyes (46.2%) achieved a successful
result, 11 of 26 eyes (42.3%) achieved
qualified success and three of 26 eyes
(11.5%) did not achieve adequate control
(Table 3, Fig. 1).
The survival rate after the initial
needling was 64% at 6 months, and
the same after 1 year and 2 years. All
the failures after the initial needling
were early failures (rising IOP within
less than 1 month). The survival rate
after one or more needling revisions
was 96% at 6 months and 77% at
1 year and 2 years (Fig. 2).
Table 4 shows the results in the
subgroups receiving either MMC,
5-FU or no antimetabolites. All 10
eyes in the 5-FU group and all three
eyes needled without antimetabolites
were successes or qualified successes.
The MMC group of 13 had three
(23.1%) failures.
Table 5 compares the results of the
needling revision according to the type
of bleb. There was no difference in the
overall success between eyes with a flat-
tened filtering bleb prior to needling
and those with a cystic bleb. There
were only three eyes with a diffuse
bleb; they all achieved a qualified suc-
cess but this group is too small in num-
ber to compare with the other groups.
Among the three eyes that failed to
achieve IOP of 18 mmHg, one eye did,
however, decrease IOP by 50% and one
eye decreased IOP by 18.5%. The third
failure eye had increasing IOP after
6 months of follow-up and underwent
another needling. This patient had sec-
ondary glaucoma after penetrating ker-
atoplasty.
Only one patient showed a greater
than two-line drop in visual acuity
after needling, due to developing bul-
lous keratopathy.
Complications developed in six of
the 26 eyes (23.1%). There were corneal
epithelial defects in three eyes (11.5%)
and conjunctival wound leaks in two
eyes (7.7%). These complications
resolved spontaneously within a few
days. However, one eye (3.8%) of a
patient with CoganReese syndrome
developed bullous keratopathy. This
patient had earlier undergone seven
filtering operations, followed by 10
needling procedures with MMC and
several subconjunctival injections of
5-FU. He later underwent a penetrating
keratoplasty.
Discussion
Failed filtering procedures are a pro-
blem after glaucoma surgery. We
found needling procedures to be a
very useful tool in this respect. The
technique is a relatively simple way of
restoring a filtering bleb, with a fairly
small degree of surgical trauma to the
eye. In our material, 23 of 26 eyes
achieved success or qualified success,
yielding a total success rate of 88.5%
over a mean follow-up of 14.5 months.
We found a highly significant drop in
IOP at the last follow-up after the need-
ling procedure.
Our overall rate of complete or qua-
lified success is comparable to the
results of other authors. Ewing & Stam-
per (1990) presented a retrospective
study including 12 eyes that underwent
one to five needling revisions for a

579
ACTA OPHTHALMOLOGICA SCANDINAVICA 2003

Table 2. Postoperative patient characteristics.

Eye Months No. of Doses of Doses of 5-FU 5-FU Post- Final Follow-up Final
between needle MMC 5-FU doses doses needling medication from last result
surgery revisions during during before after IOP at needle
and the needle needle Final final final revision
first revision revision needle needle follow-up (months)
needle revision revision
revision

1 12.00 1 1 0 0 2 8 0 48 S
2 6.00 10 10 0 8 0 22 3 36 F
3 1.50 13 7 6 3 0 17 0 12 S
4 5.00 1 1 0 1 2 12 0 24 S
5 1.00 1 1 0 3 4 15 2 30 QS
6 3.00 1 1 0 0 1 15 2 24 QS
7 1.25 2 1 1 0 0 14 0 24 S
8 1.50 1 0 0 0 1 10 3 24 QS
9 1.50 3 2 1 9 2 14 1 6 QS
10 0.50 1 1 0 5 3 16 1 6 QS
11 1.00 1 0 0 0 9 18 0 12 S
12 1.50 1 0 0 0 0 14 3 13 QS
13 0.25 1 0 1 0 6 14 0 9 S
14 0.75 3 0 2 2 3 10 0 6 S
15 0.25 4 4 0 6 4 20 3 24 F
16 2.75 1 0 1 7 0 16 2 6 QS
17 0.50 1 0 1 3 3 16 2 7 QS
18 0.50 1 0 1 0 6 13 0 6 S
19 1.00 1 0 1 1 5 18 1 10 QS
20 2.50 2 0 2 3 0 13 0 6 S
21 0.50 1 0 1 0 4 15 0 6 S
22 0.75 4 1 3 8 2 16 1 6 QS
23 1.00 1 0 1 2 1 7 0 6 S
24 1.00 2 2 0 2 11 13 1 7 QS
25 1.50 2 2 0 2 1 35 3 12 F
26 0.50 1 0 1 0 3 16 0 6 S
Mean
SD 1.9
2.48 2.3
2.9 1.3
2.4 0.9
1.3 2.5
2.9 2.8
2.8 15.3
5.2 1.1
1.2 14.5
11.3

failed or failing filtering operation,
without the use of antimetabolites dur-
ing the needling. Seven of the patients
(the most complicated) received post-
operative injections of 5-FU. All the
12 eyes had at least some evidence of a
bleb present before the needling, and the
last needling was performed from
6 weeks to 35 years after prior glaucoma
surgery. The average follow-up period
in this study was 9 months. The needling
revision was successful in 11 of 12 eyes
(91.7%) and all seven eyes receiving
postoperative 5-FU succeeded.
Table 3. Number of earlier performed trabeculectomies and classification of results after one or
more needling revisions.
No. of earlier No. of patients
trabeculectomies
1 17
2 3
3 6
Success ¼ IOP 18 mmHg without medication; qualified success ¼ IOP 18 mmHg with medication;
failure ¼ IOP > 18 mmHg on medication.
Shin et al. (1993) reported the results
of needling revision in a retrospective
study including 30 eyes that underwent
from one to seven revisions with
adjunctive 5-FU. None of the eyes had
an encapsulated bleb before the need-
ling. The time between prior glaucoma
surgery and the first needling ranged
from 0.4 weeks to 216.1 weeks. The
average follow-up period from the last
revision was 49.9 weeks. Success was
achieved in 24 of 30 eyes (80%).
Mardelli et al. (1996) reviewed the
results achieved in 62 eyes that under-
No. of patients No. of patients
with success with qualified
success
9 7
0 3
3 1
went one to seven needling revisions
with adjunctive MMC, with a mean
follow-up of 8.3 months from the last
needling. The time between the last fil-
tration surgery and needle revision ran-
ged from 3 months to 46 years. They
reported an overall success rate of
91.9%.
Our needling procedures are per-
formed with adjunctive use of MMC
or 5-FU or without antimetabolites.
The choice between the three alterna-
tives partly depended upon the experi-
ence of the surgeon and partly on the
nature of the glaucoma. Our study
included only three patients needled
without the use of antimetabolites. All
three had undergone their first trabecu-
No. of lectomies 4–6 weeks before the need-
failures ling, and had open-angle glaucoma
with diffuse or cystic blebs and pressure
rises up to 2038 mmHg before need-
1 ling. This is too small a sample to allow
0
for any firm conclusions even if there
2
were no failures among them, but need-
ling without antimetabolites is usually
not considered very successful. Costa

580
 ACTA OPHTHALMOLOGICA SCANDINAVICA 2003

et al. (1997) reported a prospective,

randomized study comparing needling
revisions without antimetabolites and
medical treatment in eyes with encap-
sulated blebs after trabeculectomy with
MMC. Needling was performed in 14
eyes, with success in only 7.1% and
qualified success in 57.1%. Ewing et al.
(1990) could not demonstrate a statisti-
cally significant benefit from the use of
5-FU, but their clinical impression was
that it was advantageous.
Three patients who underwent sev-
eral needling procedures had their eyes
needled with MMC the first time and
later with 5-FU. We did this because we
Fig. 1. Comparison of pre-needling and post-needling IOP at last follow-up. started with great respect for and no
personal experience of subconjunctival
injections of MMC, while we knew 5-
FU from other types of treatment, such
as recurrent pterygea. As our results
were good, we sometimes used this
combination according to our clinical
estimation.
We tended to use MMC for the more
complicated glaucomas and after sev-
eral previous failed filtration surgery
procedures (Kitazawa et al. 1991;
Skuta et al. 1992; Katz et al. 1995).
In the groups receiving 5-FU or no
antimetabolites there were no failures.
The group receiving MMC included
23.1% failures. This result is probably
explained by the selection of the
patients to the different groups, as the
more complicated glaucoma eyes were
Fig. 2. Kaplan Meier survival plots of 26 eyes following a single needling revision versus 26 eyes needled with MMC, as is illustrated by
after one or more needling revisions. the three failure eyes, which all had
secondary glaucoma. Two of these
Table 4. Results of one or more needling revisions in the subgroups receiving MMC, 5-FU or no had undergone several filtration sur-
antimetabolites. gery procedures.
Our clinical experience is that need-
Antimetabolites No. of No. of No. of No. of Pre-needling Post-needling
ling with MMC is a valuable tool, espe-
used during eyes patients patients patients IOP (mmHg) IOP (mmHg)
needle with with with
cially in the management of the more
revision success qualified failure complicated glaucomas.
success As mentioned above, multiple need-
lings have been reported by several
MMC 13 4 6 3 30.5
6.8 16.7
6.5 authors. Two of the eyes in our study
5-FU 10 7 3 0 26.9
6.4 13.8
3.3 underwent needling more than four
None 3 1 2 0 27.7
9.3 14.0
4.0
times. One was a congenital glaucoma
eye in a child who had three trabeculec-
tomies before needling. She underwent
Table 5. Results of needling revision according to the type of filtering bleb. 13 needling procedures and finally
Bleb No. of No. of No. of
achieved success. The other eye
appearance patients patients failures belonged to a patient with Cogan
with with Reese syndrome who had undergone
success qualified seven earlier trabeculectomies. He
success underwent 10 needling procedures. He
was considered a bad candidate for
Flattened 6 6 2
Diffuse 0 3 0
other filtering surgery because of exten-
Cystic 6 2 1 sive conjunctival fibrosis. His eye was
classified as a failure. Implantation of a

581
ACTA OPHTHALMOLOGICA SCANDINAVICA 2003
glaucoma drainage device might have
been an alternative treatment for these
patients.
The complications after the needling
procedure were mainly transient and
tolerable and healed within less than a
week without sequelae. They were the
same as reported by other authors.
Ewing & Stamper (1990) reported
few and manageable complications
including corneal epithelial defects
(41.7%), hyphema and shallowing of
the anterior chamber (8.3%), conjunc-
tival leaks from the needle entry
(16.7%) and an enlargement of a non-
functioning overhanging bleb after tra-
beculectomy (8.3%).
Shin et al. (1993) reported comp-
lications in six of 30 eyes (20%),
including small hyphemas (6.7%),
corneal epithelial defects (10%), and
conjunctival wound leaks and
shallow anterior chamber (3.3%).
All complications resolved sponta-
neously.
Mardelli et al. (1996) reported serous
choroidal detachment (16.1%), bleb
leak (8.1%), hyphema (3.2%), iris
blocking sclerostomy (3.2%), corneal
abrasion (3.2%) and hypotony (1.6%).
However, one of our patients devel-
oped bullous keratopathy. We believe
that this patient, who had undergone
seven earlier filtration operations, was
at high risk of developing the same
complication if another surgical inter-
vention had been chosen (McDermott
et al. 1994; Katz et al. 1995; Burgoyne
et al. 2000). Compared to filtering
surgery, the rate of complications
after bleb needling seems to be low
(McDermott et al. 1994; Katz et al.
1995; De Bry et al. 2002).
Compared to other authors, we
tended to perform needling revision
quite early after filtering operations.
Quite often, we chose needling instead
of resumption of topical medication,
which makes patient compliance a
smaller problem. Because of this, we
have not been able to compare the
need for antiglaucoma medication
before and after needling procedures.
Our good results so far with early need-
ling have led us to consider early
needling as something to look at separ-
582
ately on a larger number of patients.
Kapetansky & Kapetansky (1999)
showed that bleb needling is more
effective when there still is a bleb
present. Their article evaluated the
efficacy of MMC needling in 30 eyes
that were classified into three different
groups according to bleb appearance.
They suggest that early bleb needling
affords the best opportunity to success-
fully restore the function of a failing
bleb (Kapetansky & Kapetansky 1999).
For patients with side-effects from
topical medication, successful needling
may even contribute to a better quality
of life.
Needling revision is most often per-
formed as a clinic procedure. This
makes it very time efficient for both
doctor and patient. The cost is lower
than for repeat filtering procedures.
Compared to supplemental medication,
it will lead to more frequent clinic visits,
and it requires a fully co-operative
patient.
Our study supports needling revision
as a useful tool in the management of
glaucomas.
Acknowledgement
The authors would like to thank Harald
Johnsen MD, PhD for statistical analysis.
References
Azuara-Blanco A, Bond JB, Wilson RP, Mos-
ter MR & Schmidt CM (1997): Encapsulated
filtering blebs after trabeculectomy with
Mitomycin C. Ophthalmic Surg Lasers 28
(10): 805–809.
Burgoyne JK, Dunn WD, Lakhani W & Can-
tor LB (2000): Outcomes of sequential tube
shunts in complicated glaucoma. Ophthal-
mology 107 (2): 309–314.
Chen P & Palmberg P (1997): Needling revi-
sion of glaucoma drainage device filtering
blebs. Ophthalmology 104 (6): 1004–1010.
Costa V, Correra M & Kara-Jose N (1997):
Needling versus medical treatment in encap-
sulated blebs. A randomized, prospective
study. Ophthalmology 104 (8): 1215–1122.
De Bry PW, Perkins TW, Heatley G,
Kaufmann P & Brumback LC (2002):
Incidence of late-onset bleb-related compli-
cations following trabeculectomy with
Mitomycin. Arch Ophthalmol 120: 297–300.
Ewing R & Stamper R (1990): Needle revision
with and without 5-Fluorouracil for the
treatment of failed filtering blebs. Am J
Ophthalmol 110: 254–259.
Fluorouracil Filtering Surgery Study Group
(1989): Fluorouracil filtering surgery study
one-year follow-up. Am J Ophthalmol 108:
625–635.
Kapetansky FM & Kapetansky SD (1999):
Antimetabolite use in revising failing filter-
ing blebs. Semin Ophthalmol 14: 144–151.
Katz GJ, Higginbotham EJ, Lichter PR,
Skuta GL, Musch DC, Bergstrom TJ &
Johnson AT (1995): Mitomycin C versus
5-Fluorouracil in high risk glaucoma filtering
surgery. Extended follow-up. Ophthalmology
102: 1263–1269.
Kitazawa Y, Kawase K, Matsushita H & Min-
obe N (1991): Trabeculectomy with Mitomy-
cin. A comparative study with Fluorouracil.
Arch Ophthalmol 109: 1693–1698.
Mardelli P, Lederer CM, Murray PL, Pastor
SA & Hassanein KM (1996): Slit-lamp nee-
dle revision of failed filtering blebs using
Mitomyin C. Ophthalmology 103 (11):
1946–1955.
McDermott ML, Wang J & Shin DH (1994):
Mitomycin and the human corneal endothe-
lium. Arch Ophthalmol 112: 533–537.
Pederson JE & Smith G (1985): Surgical man-
agement of encapsulated filtering blebs.
Ophthalmology 92: 955–958.
Shin DH, Juzych MS, Khatana AK, Swendris
RP & Parrow KA (1993): Needling revision
of failed filtering blebs with adjunctive 5-
Fluorouracil. Ophthalmic Surg 24: 242–248.
Skuta GL, Beeson CC & Higginbotham EJ
(1992): Intraoperative Mitomycin C versus
postoperative 5-Fluorouracil in high risk
glaucoma filtering surgery. Ophthalmology
99: 440–444.
Skuta GL & Parrish RK II (1987): Wound
healing in glaucoma filtering surgery. Surv
Ophthalmol 32: 149–170.
Received on February 5th, 2003.
Accepted on June 3rd, 2003.
Correspondence:
Marit Fagerli MD
Department of Ophthalmology
St. Olav University Hospital
N-7006 Trondheim
Norway
Tel: þ 47 73 86 87 30
Fax: þ 47 73 86 87 65
Email: marit.fagerli@stolav.no