ORIGINAL STUDY
Silicone Ahmed Glaucoma Valve With and Without
Intravitreal Triamcinolone Acetonide for Neovascular
Glaucoma: Randomized Clinical Trial
Sergio Henrique Teixeira, MD,* Larissa Morimoto Doi, MD,* Andre´ Luiz de Freitas Silva, MD,*
ˆ
Karine Duarte Silva, MD,* A ngela Tavares Paes, PhD,w Fabiana Shinzato Higa, MD,* Marcelo
Mendonc¸a, MD,z Joa˜o Antoˆnio Prata, Jr, MD,y and Augusto Paranhos, Jr, MD*w
Purpose: To compare the e ect on intraocular pressure (IOP) of
the silicone Ahmed glaucoma valve with and without an
intravitreal injection of triamcinolone acetonide.
Patients and Methods: Forty-nine patients with clinically uncon-
trolled neovascular glaucoma were included in the study; 22 were
randomly assigned to the study group (silicone Ahmed glaucoma
valve implant with intravitreal triamcinolone acetonide) and 27 to the
control group (silicone Ahmed glaucoma valve). IOP was the
primary outcome measure in this study. The secondary outcome
measure was success, defined by IOP lower than 22 mm Hg and
higher than 5 mm Hg, and no serious complications. Success rates
in both the groups were compared using Kaplan-Meier survival
curves and the log-rank test. IOP levels were compared using mixed
linear model analysis to correct for repeated measures correlation.
Results: Forty-three patients, 18 in the study group and 25 in
the control group, completed the study (follow-up of 12 mo).
The mean IOP was significantly lower after 1 year in both the
groups (P<0.001). The mean IOP in the first month of follow-up
was lower in the study group (control; 20.4±9.7, study;
13.6±6.5, P<0.01). The success rate at 1 year was 78% for the
study group and 76% for the control group (P=0.82).
Complication rates were not di erent between the groups.
Conclusions: Intravitreal injection of triamcinolone acetonide in
neovascular glaucoma did not a ect the intermediate-term
success of the silicone Ahmed valve nor reduce the incidence
of com-plications. The mean IOP spike in the first month was
lower in the triamcinolone group.
Key Words: neovascular glaucoma, glaucoma drainage
implants, triamcinolone acetonide
(J Glaucoma 2012;21:342–348)
N eovascular glaucoma (NVG) is an aggressive vision-
threatening disease, mainly caused by diabetic retino-pathy
(DR), ischemic central retinal vein occlusion
Received for publication April 19, 2010; accepted December 28,
2010. From the *Department of Ophthalmology, Federal University
of Sao Paulo, R Botucatu Vila Clementino; wHospital Israelita Albert
Einstein, Av. Albert Einstein, Morumbi, Sa˜o Paulo; zHORP, Av
Jose´Munia, Sa˜o Jose´do Rio Preto, SP; and yFederal University of
Triangulo Mineiro. Av. Frei Paulino, Abadia, Uberaba. MG, Brazil.
Supported by the National Counsel of Technological and Scientific
Development (CNPQ), Brazil.
The authors declare no conflict of interest.
Valves were donated by New World Medical.
Reprints: Sergio Henrique Teixeira, MD, Av Indianopolis, 1797,
Planauto Paulista, Sao Paulo, SP, Brazil. ZIP: 04063-003 (e-
mail: sergiohteixeira@gmail.com).
r
Copyright 2012 by Lippincott Williams & Wilkins
DOI:10.1097/IJG.0b013e31820d7e4e
342 | www.glaucomajournal.com
(CRVO), and ocular ischemic syndrome. As it often has
a poor response to conventional glaucoma surgery,
posterior aqueous drainage devices have been used as
a surgical option, with variable success rates.1–5
As the introduction of the Molteno implant,6 many
types of posterior aqueous drainage devices have been
evaluated in an e ort to treat all forms of glaucoma. A
systematic literature review on various glaucoma
drainage devices concluded that the Molteno implant,
Baerveldt implant, Ahmed glaucoma valve, and Krupin
valve did not significantly di er in either the percentage
change in intraocular pressure (IOP) or the overall
surgical success rate at the last follow-up.5 The Ahmed
glaucoma valve is a drainage device with a 185 mm 2
end plate made of polypropylene (model S2) or silicone
(model FP-7). The critical di erence between the Ahmed
glaucoma valve and other glaucoma implants is the
double sheet silicone valve that allows aqueous
drainage to begin immediately after placement. The FP-
7 model of the Silicone Ahmed glaucoma valve (SAGV)
became available in January 2003 and uses silicone
instead of polypropylene for the external plate. In
addition, the posterior edge of the new model is tapered
to facilitate insertion of the device under the conjunctiva
and Tenon’s capsule, and it has 3 fenestration holes.
The valve principle is the same in both the models.
Surgery with adjuvant treatments using mitomicin, 7–9
bevacizumab,10 laser,7,11 and corticosteroids12,13 have
been proposed to manage NVG. Corticosteroids inhibit
vascular permeability and fibroblast proliferation, reduce
and regulate wound healing, and suppress inflammation.
Intravitreal injections of corticosteroids such as dexametha-
sone alcohol and, especially, the more lipophilic, triamci-
nolone acetonide, have been used to treat macular edema
associated with DR and CRVO.14–17 Intravitreal delivery of
corticosteroids has allowed many posterior segment dis-
eases to be locally treated without adverse systemic side e
ects. Intravitreal delivery also allows the steroid to bypass
the blood-retinal barrier, leading to a more concentrated
dose of steroid for a prolonged period of time. It has been
shown that a 4 mg dose of intravitreal triamcinolone
acetonide (IVTA), in a nonvitrectomized human eye,
maintains measurable concentrations for approximately 3
months.18
In addition to its e ect on lowering inflammation and
improving visual acuity in patients with DR and earlier
CRVO, IVTA has been proposed as a potential adjuvant
treatment for NVG. In this setting, the regression of iris
neovascularization provides the IOP lowering e ects. 19–
21
J Glaucoma Volume 21, Number 5, June/July 2012
J Glaucoma Volume 21, Number 5, June/July 2012
The objective of this study was to compare the e ect
of the SAGV on IOP with and without IVTA in eyes with
NVG.
METHODS
This was a randomized controlled clinical trial. The
research ethics committee of the Federal University of
Sa˜o Paulo approved the protocol. Informed consent
was obtained from all participants, in accordance with
the tenets of the Declaration of Helsinki 1989.
Patient Eligibility
All patients older than 17 years, with uncontrolled NVG
from any etiology except intraocular tumors or uveitis, in the
Glaucoma Section of Federal University of Sa˜o Paulo,
Federal University of Triangulo Mineiro and Rio Preto’s Eye
Hospital were candidates for inclusion in the study. Clinically
uncontrolled NVG was defined as an IOP above 22 mm Hg
using maximum tolerated glaucoma medication. Patients
were excluded from the study if any of the following criteria
were present: (1) no light perception;
(2) NVG secondary to intraocular tumors or uveitis; (3)
unwilling or unable to return for follow-up; (4) pregnancy;
or (5) earlier cyclodestructive procedure, scleral buckle
procedure, or silicone oil surgery.
As a part of the protocol, all patients who had not had
panretinal photocoagulation by the time of enrollment were
referred for the procedure before surgical intervention.
Interventions and follow-up took place at each
center, and all of the analysis was carried out at Federal
University of Sao Paulo.
Treatment Assignment
Patients with NVG were randomized into the follow-ing
groups using a computer-generated randomization table: (1)
study group; silicone Ahmed valve implant with IVTA and (2)
control group; silicone Ahmed valve implant.
Masking
Technical sta and statisticians participated in a
masked manner. Surgeons were told to inject or withhold
the TA at the end of the surgical procedure. The surgeon
who performed the tube implantation was responsible for
the follow-up.
Surgical Procedure
The surgical procedures in each group were standard-
ized, but did allow the surgeon to adapt the technique to
patients on an individual basis. In both the groups, SAGV
implantations were performed in the following way. An FP-7
SAGVwas placed preferably in the superotemporal quad-rant.
The implant was primed with a profuse infusion of balanced
saline solution until the valve was opened. The implant was
positioned between 2 rectus muscles and sutured to sclera with
8-0 silk sutures at a measured distance of 10 mm posterior to
the limbus, using the 2 fixation holes in the Ahmed plate. The
implant tube was trimmed bevel up to extend 1 to 2 mm into the
anterior chamber. A 23-gauge needle was used to create a tight
entry incision into the anterior chamber at the posterior limbus.
The tube was inserted through this entry incision and positioned
away from the corneal endothelium, just above the iris. A patch
graft of sclera was used to cover the limbal portion of the tube.
The sclera was sutured with a minimum of 2, but
r |
2012 Lippincott Williams & Wilkins
Ahmed Glaucoma Valve and Triamcinolone Acetonide
usually 4 10-0 nylon sutures, and the conjunctiva was
sutured with 7-0 Vicryl continuous sutures.
In the study group, after the conjunctiva was
sutured, 0.1 mL of TA (40 mg/mL) was injected into the
vitreous cavity through the pars plana, 3.0 to 3.5 mm
posterior to the limbus, with a 27-gauge needle.
Gentamicin and dexametasone were injected into
the subconjunctival space at the end of the surgery in
both the groups, and after the surgical procedure,
patients received atropine drops and an occlusive patch
to be removed by the attendant at the first follow-up visit.
In all surgeries, anesthesia included 5 to 10 mL
peribulbar injection of 0.75% ropivacaine or a 1:1
mixture of 2% lidocaine and 0.75% bupivacaine.
Surgeons who performed the surgery were at least
second year glaucoma fellows with earlier glaucoma
implant surgery experience.
Baseline and Follow-up Evaluation
A list of study measurements for scheduled follow-
up visits is presented in Table 1 and detailed here:
Best-Corrected Visual Acuity: Snellen visual acuity
was measured at each follow-up visit with padronized
charts in the same room.
Tonometry: IOP was measured by Goldmann appla-
nation tonometry.
Gonioscopy was performed with Sussman lens
during the baseline examination.
Outcomes
IOP was the primary outcome measure in this study.
The secondary outcome measure was success defined
by the following 2 criteria:
Success was defined as the absence of:
IOP above 21 mm Hg on 2 consecutive
measurements, or
IOP lower than 6 mm Hg, or
No light perception, glaucoma surgery, serious
complication.
Complete success was defined as the absence of:
IOP above 21 mm Hg on 2 consecutive
measurements, or
IOP lower than 6 mm Hg, or
No light perception, glaucoma surgery, serious
complication, or
Use of more than 2 medications to achieve target IOP.
The time frame of at least 1 month between 2
consecutive IOP measurements was used to
characterize a failure. Surgeons could see the patients
as many times as they deemed necessary to conduct
any particular case; however, if the IOP was still above
21 mm Hg after 30 days, it was considered a failure.
TABLE 1. Follow-up Chronology Scheme
1 1 1 3 6 9 1
Baseline d wk mo mo mo mo y
Snellen visual acuity
Slit-lamp biomicroscopy
Tonometry
Gonioscopy
Fundoscopy
www.glaucomajournal.com 343
Teixeira et al J Glaucoma Volume 21, Number 5, June/July 2012

Serious complications were defined as

TABLE 2. Baseline Patient Characteristics
complications that might result in a significant decrease
of vision or those that might require surgical correction. Study Group Control Group
Serious complica-tions included, but were not limited to, (n=22)* (n=27)w P
tube extrusion, endophthalmitis, corneal Age (y) 0.134z
decompensation, suprachoroidal hemorrhage, retinal Mean±SD 62.91±7.26 57.48±15.32
detachment, phthisis bulbi, and loss of light perception. Sex, n (%) 0.254y
Male 16 (72.7%) 15 (55.6%)
Statistics Sample Size and Power Calculation Female 6 (27.3%) 12 (44.4%)
Statistical analysis was carried out using R software Race, n (%) 0.407y
version 2.8.1 (R Foundation for Statistical Computing, White 14 (63.6%) 14 (51.9%)
Vienna, Austria). Baseline demographic data were analyzed Other 8 13
Black 3 (13.6%) 4 (14.8%)
using the Student t test for continuous variables and the w2
Hispanic 4 (18.2%) 9 (33.3%)
or Fisher exact tests for categorical variables. Other 1 (4.5%) 0 (0.0%)
Comparisons of mean IOP between groups at each Etiology, n (%) 0.215y
visit and over time were analyzed using linear mixed-e DR 9 (40.9%) 15 (55.6%)
ects (LME) models,22 and the number of preoperative CRVO 13 (59.1%) 12 (44.4%)
and postoperative antiglaucoma medications was Comorbities,
analyzed using generalized estimating equations 23 for n (%)
count data (Poisson distribution). SAH 18 (81.8%) 21 (77.8%) 0.999J
Kaplan-Meier analysis and the log-rank test were DM 11 (50.0%) 18 (66.7%) 0.238y
used to compare the survival distribution in both the *Silicone Ahmed valve+intravitreal triamcinolone acetonide.
groups. Cox proportional hazards regression was used wSilicone Ahmed valve.
to evaluate baseline demographic and clinical features zStudent t
as possible predictors of failure. P values of <0.05 were test. yw2 test.
JFisher exact test.
considered statistically significant. The minimum sample CRVO indicates central retinal vein occlusion; DM, diabetes mellitus;
size, setting a at 0.05, with a 0.8 power to detect a di DR, diabetic retinopathy; SAH, systemic arterial hypertension.
erence of 20% between the groups, and an estimated
standard deviation (SD) of the IOP measurement of
25%, was 41 patients (1-tailed test).
blood pressure, diabetes, and renal failure. This patient
RESULTS died 2 months after the surgical procedure as a result of
Forty-nine patients were included in the study. complications of his systemic condition. By the time the
Randomization assigned 27 patients to the control group study was done, no patient had received any
(SAGV) and 22 to the study group (SAGV+IVTA). The
antiprolifera-tive medication or any other ocular injection.
progress of patients in the study is shown in Figure 1. Two
patients died during the first year of follow-up, 1 in each Demographics
group. The patient who died in the control group completed
Patients’ baseline characteristics are shown in
the first 6 months of follow-up. In the study group, 1 patient
Table 2. No significant di erences were found between
had a supracoroidal hemorrhage during the surgery, the groups in clinical and demographic features.
probably as a result of uncontrolled systemic
IOP and Glaucoma Medications
The baseline and follow-up IOPs for both the groups
are reported in Table 3 and Figure 2. Baseline IOPs and the
number of medications did not di er between the groups
(Table 3). Patients who underwent additional glaucoma
surgery and patients with phthisis bulbi were censored from
analysis after the time of reoperation or phthisis onset. Both
the surgical procedures produced a significant reduction in
IOP. In the study group, mean (±SD) IOP decreased from
42.1±9.3 mm Hg at baseline to 13.9±3.7 mm Hg at the 1-
year follow-up visit (P<0.001, LME); in the control group,
mean IOP decreased from 40.4±10.8 mm Hg at baseline to
15.5±4.4 mm Hg at the 1-year follow-up visit (P<0.001,
LME). There was no significant di erence in mean IOP
between groups at 1 year (P=0.316, LME). However, the
study group had a significantly lower mean IOP than the
control group at the 1-month postoperative follow-up visit
(P=0.010, LME). In addition, the need for medical therapy
decreased signifi-cantly after the surgery in both groups
[study group (mean±SD); 2.4±1.1 to 0.8±0.8, P<0.001,
FIGURE 1. Flowchart of patient progress. Control group: control group (mean±SD); 2.3±0.9 to 1.3±1.2, P<0.001, gen-
silicone Ahmed glaucoma valve. Study group: silicone Ahmed eralized estimating equation considering preoperative and
glaucoma valve with intravitreal triamcinolone acetonide.

344 | www.glaucomajournal.com r
2012 Lippincott Williams & Wilkins
J Glaucoma Volume 21, Number 5, June/July 2012 Ahmed Glaucoma Valve and Triamcinolone Acetonide

control groups, respectively P=0.31; success, 77.8% and

TABLE 3. IOP and Medical Therapy at Baseline and Follow-up
76.0%, study and control groups, respectively P=0.82; log-
Study Group* Control Groupw P rank test). In Table 4, we detailed the reasons for failure in
Baseline both the groups. Baseline demographic and clinical features
IOP (mm Hg) 42.1±9.3 40.4±10.8 0.476z were evaluated as possible predictors of failure using Cox
Glaucoma medications 2.4±1.1 2.3±0.9 0.709y proportional-hazards regression for survival data, and no
1d statistically significant predictors were found.
IOP (mm Hg) 9.8±11.1 11.8±13.6 0.221z
1 wk Complications
IOP (mm Hg) 8.7±6.1 10.8±8.8 0.237z Complication rates are presented in Table 5. Compli-
Glaucoma medications 0.1±0.5 0.1±0.6 0.924y cations that were considered failure criteria were labeled as
1 mo serious complications. In the control group, 1 patient had
IOP (mm Hg) 13.6±6.5 20.4±9.7 0.010z phthisis bulbi. In the study group, 1 patient had bullous
Glaucoma medications 0.3±0.7 0.3±0.6 0.879y
3 (2-4) mo keratopathy, and 1 had phthisis bulbi secondary to a
IOP (mm Hg) 15.8±7.3 16.6±9.6 0.418z subcoroidal hemorrhage during the surgery.
Glaucoma medications 0.8±0.8 0.7±0.8 0.797y Other complications occurred, particularly hyphema
6 (5-7) mo and early hypotony. One patient had a misdirection
IOP (mm Hg) 15.0±5.7 16.7±7.7 0.318z glaucoma treated medically. After 2 weeks, this patient
Glaucoma medications 0.8±0.7 1.0±1.1 0.257y underwent a pars plana vitrectomy.
9 (8-10) mo Tube obstructions by the iris (3 patients) were
IOP (mm Hg) 15.9±5.5 15.3±4.9 0.623z successfully treated with slit lamp needling. Early
Glaucoma medications 0.6±0.6 1.2±1.2 0.071y
12 (11-13) mo
shallow anterior chamber obstructions (6 patients) were
IOP (mm Hg) 13.9±3.7 15.5±4.4 0.316z treated with air injections (3 patients), 2%
Glaucoma medications 0.8±0.8 1.3±1.2 0.284y methylcellulose viscoe-lastic injections (1 patient), and
ligature of the tube (2 patients).
Data presented as mean±standard deviation. There were no statistically significant di erences in
*Silicone Ahmed valve+intravitreal triamcinolone acetonide. the rates of complications between the groups (Table 5).
wSilicone Ahmed valve.
zLinear mixed-e ects models.
yGeneralized estimating equations for count DISCUSSION
data. IOP indicates intraocular pressure.
The management of NVG remains a challenge, as it
has a rapid and devastating course. Most of the time, it
presents with very high IOPs that are often di cult to control
1-year visits]. There were no di erences between the and frequently cause irreversible visual loss. In this context,
groups at any follow-up visit regarding need for medical posterior drainage implants are valuable tools because, in
therapy (Table 3).
comparison to other treatment modalities, their e cacy is
less influenced by inflammation and complica-tions such as
Success Rates hyphema or peripheral anterior synechiae. In this study, we
Figure 3 presents the success and complete success assessed a silicone valved drainage implant with and
rates for the 2 treatment groups. There were no significant without the adjunct use of IVTA to manage NVG. The
di erences in the rates at 1 year after surgery between the success rate at 1 year did not di er significantly between the
groups (complete success, 77.8% and 64.0%, study and groups, and the overall success rate was 76.7%.

FIGURE 2. Intraocular pressure (IOP) at the baseline and follow-up. Data are presented as mean±standard error of the mean.
Control group: silicone Ahmed glaucoma valve. Study group: silicone Ahmed glaucoma valve with intravitreal triamcinolone
acetonide. Follow-up time is not in a linear scale.

r |
2012 Lippincott Williams & Wilkins www.glaucomajournal.com 345
Teixeira et al
FIGURE 3. Kaplan-Meier curves for the probability of complete success (intraocular pressure between 6 and 21 mm Hg on 2 consecutive
measurements, no serious complications, no more than 2 glaucoma medications) and success (intraocular pressure between 6 and 21 mm
Hg on 2 consecutive measurements and no serious complications). Control group: silicone Ahmed glaucoma valve. Study group: silicone
Ahmed glaucoma valve with intravitreal triamcinolone acetonide. P values were calculated using the log-rank test.
Some studies have assessed glaucoma drainage implants
in a population restricted to patients with NVG (mainly
retrospective series). One-year success rates vary 37% to
72% for Molteno implants, 61% to 72% for Ahmed S2
implants, 79% for Baerveldt implants, and 89% for Krupin-
Denver valves.1,4,9,24–28 All of these studies defined
success rates as an absence of an IOP more than 21 mm
Hg. Most of them did not include the absence of glaucoma
medication as a part of the success criteria. Our success
criteria also used 21 mm Hg as an IOP cuto point, but we
also included an additional requirement that this IOP could
not rise above 21 mm Hg more than 2 consecutive visits.
This seemed logical because when medications are used to
treat an IOP spike, the case should not be immediately
labeled as a failure. For instance, in patients with NVG,
transient IOP spikes are not uncommon during follow-up
[(causes include hyphema, inflammation, and hypertensive
phases (HPs)]. Recent clinical trials studying glaucoma
surgery also used similar constraints in defining failure. 13,29
We found a statistically significant di erence in the
mean IOP between the 2 groups at the 1-month follow-up
visit (study group; 13.6±6.5 mm Hg, control group; 20.4±9.7
mm Hg, P=0.010). In the control group, this was the highest
mean IOP after the surgery, consistent with the HP of the
implant. HP is defined as an IOP greater than 21 mm Hg
from the third week to the sixth month after sur-gery. Ayyala
et al30 evaluated 85 patients who underwent insertion of an
AGV for control of refractory glaucoma and reported an HP
in 70 patients (82%). The IOP peaked at 1 month and
stabilized at 6 months. Apparently, the HP is caused by a
fibrotic response in the subconjunctival space with the bleb
becoming visibly inflamed and encapsulated. Some authors
have shown experimentally31 and clinically8 the positive e
ect of an antifibrotic in the HP. One could suspect that the
antinflammatory e ects of triamcinolone could also have
some beneficial e ects on the HP, which would be
consistent with our findings. As the study was not originally
designed to study HP, further research must be done to
address this hypothesis.
Hypotony was our most common complication in the
early postoperative period (26% considering both the
346 | www.glaucomajournal.com
J Glaucoma Volume 21, Number 5, June/July 2012
groups). The incidence of postoperative hypotony has been
reported in 5.3% and 8% with AGV.32,4 Law et al33 found,
during a 12-month follow-up period, an IOP r5 mm Hg
postoperatively in 38.0% of patients after implantation of a
S2 AGV and 50.0% after implantation of a FP-7 AGV. Our
patients had very high IOPs in the immediate preoperative
period and usually had inflamed eyes, which could partially
explain the low pressures during this period. Our hypotony
cases resulted in shallow anterior chambers in only 6
patients and could be managed with air injections (3
patients), viscoelastic injections (1 patient), and ligature of
the tube (2 patients). In both the cases, silicone tube suture
was performed with 7-0 Vicryl: 1 patient in the study group
(78-year-old male, has CRVO, on the seventh postoperative
day) and 1 patient in the control group (70-year-old female,
has CRVO, on the sixth postoperative day). At that time,
surgeons assumed that the silicone valve was not working
properly. Transient hyphema was observed in 10 of the 49
patients (20.4%) at some point during the follow-up period.
Other studies have reported this rate between 8% and 28%
with tube implantation in NVG.25–27 In our study, all the
hyphemas resorbed without surgical intervention in both the
groups.
TABLE 4. Reasons for Treatment Failure
Study Group Control Group
(n=4)* (n=9)w
Reasons for failure
IOP >21 mm Hg 2 (50) 5 (56)
More than 2 0 (0) 3 (33)
glaucoma
medications
IOP <6 mm Hg 1 (25) 0 (0)
Serious 1 (25) 1 (11)
Complications
Data presented as number (percentage) and refers to the earlier
failure criteria occurrence.
*Silicone Ahmed valve+intravitreal triamcinolone acetonide.
wSilicone Ahmed valve.
IOP indicates intraocular pressure.
r
2012 Lippincott Williams & Wilkins
J Glaucoma Volume 21, Number 5, June/July 2012 Ahmed Glaucoma Valve and Triamcinolone Acetonide

valve nor reduce the incidence of complications.

TABLE 5. Complications
However, the IOP spike in the first month after surgery
Study Control was lower with IVTA.
Group* Groupw Pz
Serious complications REFERENCES
Loss of light perception 1 (5) 1 (4) >0.999 1. Every SG, Molteno AC, Bevin TH, et al. Long-term results of
Phthisis bulbi 1 (5) 1 (4) >0.999 Molteno implant insertion in cases of neovascular
Corneal descompensation 2 (9) 1 (4) 0.581 glaucoma. Arch Ophthalmol. 2006;124:355–360.
Hemorrhagic coroidal 1 (5) 0 (0) 0.460 2. Minckler D, Vedula SS, Li T, et al. Aqueous shunts for
detachment glaucoma. Cochrane Database Syst Rev. 2006;2:CD004918.
Other complications 3. Krishna R, Godfrey DG, Budenz DL, et al. Intermediate-
Hyphema 4 (18) 6 (22) >0.999 term outcomes of 350-mm(2) Baerveldt glaucoma implants.
(First week) 1 (5) 5 (19) Ophthal-mology. 2001;108:621–626.
Late 3 (14) 1 (4) 4. Yalvac IS, Eksioglu U, Satana B, et al. Long-term results of
Hypotony (First week) 7 (32) 6 (22) 0.752 Ahmed glaucoma valve and Molteno implant in neovascular
Serous coroidal 2 (9) 3 (11) >0.999 glaucoma. Eye. 2007;21:65–70.
detachment 5. Hong CH, Arosemena A, Zurakowski D, et al. Glaucoma
Atalamy (First week) 3 (14) 3 (11) >0.999 drainage devices: a systematic literature review and current
Vitreous hemorrhage 0 (0) 3 (11) 0.242 controversies. Surv Ophthalmol. 2005;50:48–60.
Tube obstruction 1 (5) 2 (7) >0.999 6. Molteno AC. New implant for drainage in glaucoma: clinical
Misdirection glaucoma 1 (5) 0 (0) 0.460 trial. Br J Ophthalmol. 1969;53:606–615.
7. Elgin U, Berker N, Batman A, et al. Trabeculectomy with
Data presented as number (percentage).
mitomycin C combined with direct cauterization of
*Silicone Ahmed valve+intravitreal triamcinolone acetonide.
wSilicone Ahmed valve.
peripheral iris in the management of neovascular glaucoma.
zFisher exact test.
J Glaucoma. 2006;15:466–470.
IOP indicates intraocular pressure. 8. Alvarado JA, Hollander DA, Juster RP, et al. Ahmed valve
implantation with adjunctive mitomycin C and 5-fluorouracil:
long-term outcomes. Am J Ophthalmol. 2008;146:276–284.
We measured the frequency of diabetes mellitus and 9. Susanna R Jr., Latin American Glaucoma Society Investiga-
systemic hypertension in our patients, and as expected, we tors. Partial Tenon’s capsule resection with adjunctive mito-
found a fairly large number of both the conditions (59% and mycin C in Ahmed glaucoma valve implant surgery. Br J
80%, respectively). As for the etiology of NVG, we found Ophthalmol. 2003;87:994–998.
that a large proportion of patients had signs of CRVO 10. Ichhpujani P, Ramasubramanian A, Kaushik S, et al. Bevacizu-mab
(51%), and in those cases, we assumed CRVO to be the in glaucoma: a review. Can J Ophthalmol. 2007;42:812–815.
cause of NVG. Five of these patients had diabetes mellitus 11. Kiuchi Y, Nakae K, Saito Y, et al. Pars plana vitrectomy and
concomitant with signs of CRVO, none of them with panretinal photocoagulation combined with trabeculectomy
for successful treatment of neovascular glaucoma. Graefes
proliferative DR. DR and CRVO are the most common
Arch Clin Exp Ophthalmol. 2006;244:1627–1632.
causes of NVG,34 and as all of our patients had either 1 of 12. Tham CC, Li FC, Leung DY, et al. Intrableb triamcinolone
these conditions, we did not perform any additional acetonide injection after bleb-forming filtration surgery (tra-
investigation to look for other NVG etiologies. Two patients beculectomy, phacotrabeculectomy, and trabeculectomy revi-
died during the first year of follow-up, 1 in each group. sion by needling): a pilot study. Eye. 2006;20:1484–1486.
Patients with NVG usually have systemic and serious 13. Yuki K, Shiba D, Kimura I, et al. Trabeculectomy with or
comorbidities. One study found a mean length of remaining without intraoperative sub-tenon injection of triamcinolone
life of 6.5 years in patients with NVG and drainage implant acetonide in treating secondary glaucoma. Am J
surgery compared with 14 years in an age matched New Ophthalmol. 2009;147:1055–1060.
Zealand general population (52% reduction in the expected 14. Grover D, Li TJ, Chong CC. Intravitreal steroids for macular
edema in diabetes. Cochrane Database Syst Rev. 2008;23:
remaining life span).35 In our study, 4 patients (8%) were CD005656.
lost to follow-up before the 1-year visit. Of these, 1 had a 15. Jonas JB, Kamppeter BA, Harder B, et al. Intravitreal
foot amputation caused by uncontrolled diabetes and was triamcinolone acetonide for diabetic macular edema: a pro-
unable to maintain follow-up visits. spective, randomized study. J Ocul Pharmacol Ther. 2006;
Our study has limitations, 1 of which is that our 22:200–207.
16. Kwong YY, Lai WW, Lam DS. Intravitreal triamcinolone acetonide in
sample size only allowed us to detect fairly large di
eyes with cystoid macular edema associated with central retinal vein
erences between the groups. Another limitation was the occlusion. Am J Ophthalmol. 2004;137:593–594.
inability to mask the IVTA injection and the follow-up 17. Gewaily D, Greenberg PB. Intravitreal steroids versus obser-
evaluation performed by the investigators. In addition, vation for macular edema secondary to central retinal vein
we had mild asymmetry between the groups, but this occlusion. Cochrane Database Syst Rev. 2009;21:CD007324.
occurred by chance. Strengths of the study are its 18. Beer PM, Bakri SJ, Singh RJ, et al. Intraocular concentration
randomized controlled design and the uniformity of the and pharmacokinetics of triamcinolone acetonide after a single
population studied (only NVG patients). intravitreal injection. Ophthalmology. 2003;110:681–686.
We are now experiencing an important improvement in 19. Jonas JB, So¨fker A. Intravitreal triamcinolone acetonide for
cataract surgery with iris neovascularization. J Cataract
the treatment of retinal diseases like DR and CRVO. The
Refract Surg. 2002;28:2040–2041.
development of new delivery systems and new drugs should 20. Jonas JB, Hayler JK, So¨fker A, et al. Regression of neovascular
provide possibilities for the management of NVG, with and iris vessels by intravitreal injection of crystalline cortisone. J
without glaucoma surgery. Future research should address Glaucoma. 2001;10:284–287.
these new possibilities. 21. Bonanomi MT, Susanna R Jr. Intravitreal triamcinolone
In conclusion, IVTA in patients with NVG did not a acetonide as adjunctive treatment for neovascular glaucoma.
ect the intermediate-term success of the silicone Ahmed Clinics (Sao Paulo). 2005;60:347–350.

r |
2012 Lippincott Williams & Wilkins www.glaucomajournal.com 347
Teixeira et al
22. Pinheiro JC, Bates D. Mixed-e ects models in S and S-
PLUS. New York: Springer; 2000.
23. Liang KY, Zeger SL. Longitudinal data analysis using
generalized linear models. Biometrika. 1986;73:13–22.
24. Yildirim N, Yalvac IS, Sahin A, et al. A comparative study
between diode laser cyclophotocoagulation and the Ahmed
glaucoma valve implant in neovascular glaucoma: a long-
term follow-up. J Glaucoma. 2009;18:192–196.
25. Faghihi H, Hajizadeh F, Mohammadi SF, et al. Pars plana
Ahmed valve implant and vitrectomy in the management of
neovascular glaucoma. Ophthalmic Surg Lasers Imaging.
2007; 38:292–300.
26. Mermoud A, Salmon JF, Alexander P, et al. Molteno tube
implantation for neovascular glaucoma: long-term results
and factors influencing the outcome. Ophthalmology.
1993;100: 897–902.
27. Sidoti PA, Dunphy TR, Baerveldt G, et al. Experience with
the Baerveldt glaucoma implant in treating neovascular
glaucoma. Ophthalmology. 1995;102:1107–1118.
28. Mastropasqua L, Carpineto P, Ciancaglini M, et al. Long-term
results of Krupin-Denver valve implants in filtering surgery for
neovascular glaucoma. Ophthalmologica. 1996;210:203–206.
348 | www.glaucomajournal.com
J Glaucoma Volume 21, Number 5, June/July 2012
29. Gedde SJ, Schi man JC, Feuer WJ, et al. Treatment
outcomes in the tube versus trabeculectomy study after one
year of follow-up. Am J Ophthalmol. 2007;143:9–22.
30. Ayyala RS, Zurakowski D, Monshizadeh R, et al.
Comparison of double-plate Molteno and Ahmed glaucoma
valve in patients with advanced uncontrolled glaucoma.
Ophthalmic Surg Lasers. 2002;33:94–101.
31. Prata JA, Minckler DS, Mermoud A, et al. E ects of
intraoperative mitomycin-C on the function of Baerveldt
glaucoma drainage implants in rabbits. J Glaucoma.
1996;5:29–38.
32. Huang MC, Netland PA, Coleman AL, et al. Intermediate-
term clinical experience with the Ahmed Glaucoma Valve
implant. Am J Ophthalmol. 1999;127:27–33.
33. Law SK, Nguyen A, Coleman AL, et al. Comparison of
safety and e cacy between silicone and polypropylene
Ahmed glaucoma valves in refractory glaucoma.
Ophthalmology. 2005;112:1514–1520.
34. Hayreh SS. Neovascular glaucoma. Prog Retin Eye Res.
2007;26:470–485.
35. Blanc JP, Molteno AC, Fuller JR, et al. Life expectancy of
patients with neovascular glaucoma drained by Molteno
implants. Clin Experiment Ophthalmol. 2004;32:360–363.
r
2012 Lippincott Williams & Wilkins