intravitreal bevacizumab and aqueous shunting surgery for neovascular glaucoma: safety and efficacy

CAN J OPHTHALMOL—VOL. 44, NO. 4, 2009 451

Intravitreal bevacizumab and aqueous shunting surgery for neovascular glaucoma: safety and efficacyTarek M. Eid,*† MD; Ashraf Radwan,‡ MD; Wael el-Manawy,† FRCS; Ihab el-Hawary,† FRCS

Objective: To study the safety and efficacy of intravitreal injection of bevacizumab followed by aqueous shunting tube surgery for the management of neovascular glaucoma (NVG).

Study Design: A prospective, non-randomized study with a historical control group.Participants: Twenty eyes of 20 patients with intractable NVG were treated with intravitreal injection of bevacizumab

followed by aqueous shunting surgery (IVB group). A historical group of 10 NVG eyes treated with panretinal photocoagulation followed by aqueous shunting surgery without bevacizumab injection was used for comparison (PRP group).

Injection of bevacizumab (1.25 mg/0.05 mL) was performed under topical anesthesia. An Ahmed valve was implanted in all cases after 1–2 weeks. In the IVB group, 10 eyes received postoperative panretinal photoco-agulation (subgroup 1A), and 10 eyes were followed without further photocoagulation (subgroup 1B). Minimum follow-up was 1 year or when failure was diagnosed.

Results: Mean preoperative intraocular pressure (IOP) was 46.5 mm Hg in the IVB group and 49.2 mm Hg in the PRP group (p = 0.5). After bevacizumab injection, iris neovessels regressed markedly. The final IOP after aqueous shunting tube surgery was 18.8 mm Hg in the IVB group and 15.9 mm Hg in the PRP group (p = 0.2). Postsur-gical complications were comparable between the groups. The success rate was 85% and 70% in the 2 groups, respectively. Two eyes were considered failures, and 3 required repeated bevacizumab injections in subgroup 1B as compared with 1 in subgroup 1A.

Intravitreal bevacizumab is a useful preparatory step to safely and effectively implant an aqueous shunt-ing tube in NVG. Panretinal photocoagulation after bevacizumab injection promotes the success rate of aqueous shunt surgery by permanent ablation of the ischemic retina.

Objet : Étude de la sécurité et de l’efficacité de l’injection intravitréenne de bévacizumab suivie d’une chirurgie du tube de drainage de l’humeur aqueuse pour le traitement du glaucome néovasculaire (GNV).

Nature : Étude prospective et non randomisée avec groupe témoin historique.Participants : Vingt yeux de 20 patients avec un GNV résistant au traitement, qui ont été traités par injection intra-

vitréenne de bévacizumab suivie d’un implant de drainage de l’humeur aqueuse (groupe BIV), ainsi qu’un groupe témoin historique de 10 yeux atteints de GNV et traités par photocoagulation panrétinienne suivie d’un implant de drainage de l’humeur aqueuse sans injection de bévacizumab, qui a servi de comparaison (groupe PPR).

: L’injection de bévacizumab (1,25 mg / 0,05 mL) a été effectuée sous anesthésie topique. Une valve d’Ahmed a été implantée dans tous les cas après 1 à 2 semaines. Dans le groupe BIV, 10 yeux ont reçu une photocoagulation panrétinienne postopératoire (sous-groupe 1A) et 10 yeux ont été suivis sans autre photoco-agulation (sous-groupe 1B). Le suivi a duré 1 an, ou jusqu’à un diagnostic d’échec.

: La moyenne de pression intraoculaire (PIO) postopératoire était de 46,5 mm Hg dans le groupe BIV et de 49,2 mm Hg dans le groupe PPR (p = 0,5). Après l’injection de bévacizumab, les néovaisseaux de l’iris ont régressé remarquablement. La PIO finale après le drainage chirurgical de l’humeur aqueuse était de 18,8 mm Hg dans le groupe BIV et de 15,9 mm Hg dans le groupe PPR (p = 0,2). Les complications postchirurgicales se com-paraient dans les deux groupes. Le taux de réussite y était de 85 % et 70 % respectivement. Il y a eu échec dans 2 yeux et 3 yeux ont requis une reprise des injections de bévacizumab dans le sous-groupe 1B comparativement à 1 dans le groupe 1A.

: Le bévacizumab intravitréen est une étape préparatoire utile pour assurer la sécurité et l’efficacité de l’implant d’un tube de drainage de l’humeur aqueuse pour le GNV. La photocoagulation panrétinienne après injection de bévacizumab favorise le taux de réussite de l’implant de drainage de l’humeur aqueuse par l’ablation permanente de la rétine ischémique

From *the Ophthalmology Department, Tanta University, Egypt, and the †Cataract and Glaucoma Unit and ‡Vitreoretina Unit, Magrabi Eye & Ear Center, Jeddah, Saudi Arabia

Presented as a poster at the American Academy of Ophthalmology meeting, Atlanta, 2008, and the European Glaucoma Society meeting, Berlin, 2008.

Originally received Oct. 22, 2008. Revised Feb. 25, 2009Accepted for publication Mar. 10, 2009Published online July 13, 2009

Correspondence to Tarek M. Eid, MD, Magrabi Eye & Ear Center, Madina Rd., P.O. Box 20377, Jeddah 21455, Saudi Arabia; [email protected]

This article has been peer-reviewed. Cet article a été évalué par les pairs.

Can J Ophthalmol 2009;44:451–6doi:10.3129/i09-108

Bevacizumab and tube shunts for NVG—Eid et al.

452 CAN J OPHTHALMOL—VOL. 44, NO. 4, 2009

Neovascular glaucoma (NVG) refractory to laser (panretinal photocoagulation) and medical (anti-

inflammatory and antiglaucoma medications) treatment represents a surgical challenge to glaucoma specialists. Glaucoma filtering surgery with or without antimetabolites carries a high failure rate.1 Cyclodestructive procedures2,3 and aqueous shunting tube surgery4,5 provide comparable success rates for NVG, but the former carries a high risk of phthisis and loss of light perception.6 Cyclodestruction is usually reserved for eyes with poor visual potential, and aqueous shunting tube surgery is considered by many as the primary treatment of choice for refractory NVG.1 Regres-sion of iris and angle rubeosis by good panretinal photoco-agulation to ablate ischemic retina is a key to safe and successful shunt surgery. In patients with markedly elevated intraocular pressure (IOP), florid iris neovascularization, and opaque media (corneal edema, dense cataract, or vit-reous hemorrhage), panretinal photocoagulation is difficult to perform properly. In this condition, glaucoma operation carries a high risk of complications and failure.1

Several mediators are involved in intraocular neovascular-ization. Vascular endothelial growth factor-A (VEGF-A), an endothelial-cell-specific mitogen, is 1 of these mediators.7 Bevacizumab is a recombinant antibody against VEGF-A approved for treatment of colorectal cancer.8 Recently, it has been widely applied for treatment of cystoid macular edema and neovascular age-related macular degeneration because of marked reduction of neovascular activity and vascular permeability in ocular tissue.9,10 Some case reports have shown the efficacy of intravitreal administration of bevacizumab in reducing IOP in patients who have NVG with nonsynechial angle closure due to marked regression of neovascularization in the iris and angle and significant reduction of intraocular inflammatory reactions.11–16 After injection of bevacizumab the eye receives proper panretinal photocoagulation to ablate the ischemic retina. In neovascu-lar eyes with total synechial angle closure, IOP may not be expected to decrease significantly after these procedures, and such eyes will need glaucoma surgical intervention.15

In this study, we prospectively evaluated the operative and postoperative complications and IOP reduction in patients with intractable NVG treated with an aqueous drainage implant (Ahmed valve) after bevacizumab injec-tion compared with NVG patients treated with panretinal photocoagulation and aqueous shunting tube surgery with-out bevacizumab injection.

METHODS

PatientsAll consecutive patients with intractable NVG scheduled

for intravitreal injection of bevacizumab and implanta-tion of Ahmed valve (2-stage procedure) seen in 2007 in the Glaucoma Unit at Magrabi Eye & Ear Center, Jeddah, Saudi Arabia, were included in a prospective study (group 1, IVB group). A historical group of NVG eyes treated with

panretinal photocoagulation followed by aqueous shunting tube surgery without bevacizumab injection in the pre-ceding 6 months was used for comparison (group 2, PRP group). The study was approved by the ethics review board of the Magrabi eye institution, which follows the principles outlined in the Declaration of Helsinki.

Inclusion criteriaAll consecutive patients with NVG and uncontrolled

IOP on maximal antiglaucoma medications, evident iris neovascularization, and active retinal pathology (e.g., pro-liferative diabetic retinopathy [PDR], central retinal vein occlusion [CRVO]) with or without previous panretinal photocoagulation were referred to the Retina Department for bevacizumab injection. Patients were referred back to the Glaucoma Department, where the aqueous shunting tube surgery was performed within 1–2 weeks of the injection. After the IOP had been controlled, patients were followed up for IOP control (Glaucoma Department) and the stabil-ity of the retinal condition (Retina Department). Accord-ing to the status of the retina and clinical judgment of the retina specialist, patients might or might not have received subsequent photocoagulation to guard against a recurrence of neovascularization. Patients in the IVB group, accord-ingly, were divided into 2 subgroups: subgroup 1A con-sisted of eyes that received postoperative photocoagulation (IVB-ASTS-PRP), and subgroup 1B consisted of eyes that did not receive further photocoagulation after bevacizumab injection and Ahmed valve implantation (IVB-ASTS only).

Surgical techniqueFor the intravitreal bevacizumab injection, all patients

signed a consent form after being told the nature of the pro-cedure and the fact that the injection was a step in a multi-stage treatment plan for glaucoma that would be followed by panretinal photocoagulation and aqueous shunting tube surgery, whichever came first according to IOP level. Intra-vitreal injection through the pars plana was administered by the retina specialist under topical anesthesia. The dose given was the recommended dose of 0.05 mL (1.25 mg) of the sterile, undiluted, commercially available bevacizumab (Avastin 100 mg/4 mL intravenously; Roche, Switzerland).

With regression of the neovessels and decreased anterior segment inflammation, patients were scheduled for aque-ous shunting tube surgery. They signed another consent form for the glaucoma procedure. The operation was per-formed under peribulbar or general anesthesia. A fornix-based flap was fashioned in the superotemporal quadrant, and this was followed by blunt dissection of the episcleral space. An Ahmed valve (New World Medical Inc, Calif.) was implanted after the patency of the valve system had been checked by saline injection. The implant was secured to the scleral bed 10 mm from the limbus by nonabsorb-able sutures. The tube was cut, bevel up, and inserted into the anterior chamber through a track made by a 23-G needle 2 mm from the limbus. The tube was directed par-



allel to the iris plane and extended for 2–3 mm into the anterior chamber. The tube was secured in place by 10/0 nylon sutures and covered by a dehydrated human fascia lata graft (Tutoplast, Tutogen Medical GmbH, Germany). The conjunctival flap was closed and the anterior chamber reformed. A combination of topical steroids and antibiotics was prescribed, as well as cycloplegic drops, and the patient was followed closely.

OutcomesMain outcome measures included intraoperative and

postoperative complications and IOP control with or with-out glaucoma medications after aqueous shunting tube surgery in each group. The potential promoting effect of combining bevacizumab and retinal photocogulation on the success of the glaucoma tube shunt surgery in group 1 was further studied by comparing outcome measures be-tween subgroups 1A and 1B.

The criteria of failure of shunt surgery in the study were lack of control of IOP to a safe level with or without glau-coma medications, operative or postoperative devastating complications, loss of light perception, or the need for addi-tional glaucoma surgical intervention. An arbitrary number of 25 mm Hg was used as a target pressure for NVG eyes treated with aqueous shunting tube surgery.6 This pressure was deemed suitable for NVG eyes with a relatively rapid increase of IOP to extremely high values and an anticipated IOP after surgery usually ranging from the high teens to low twenties. Neither repeated panretinal photocoagulation or bevacizumab injection to control iris neovascularization, nor postoperative surgical intervention to treat surgical complications of shunt surgery, nor laser treatment of tube blockage were criteria of failure. Patients were followed up for a minimum of 1 year or when failure of the glaucoma procedure was diagnosed.

The 2 groups were compared using a paired t test for continuous variables and 2 testing for categorical vari-ables. The probability of failure was compared between the 2 main groups and the 2 IVB subgroups using Kaplan-Meier survival analysis. Nominal significance level (p value) was equal to or less than 0.05.

RESULTS

The study involved 20 eyes in the IVB group (group 1) and 10 eyes in the PRP group (group 2). No significant difference was found between the groups in terms of age, sex, medical history, lens status, or previous glaucoma treatment. Five eyes in group 1 had had previous panret-inal photocoagulation compared with 3 in group 2. PDR (11 eyes in group 1 and 7 in group 2) and CRVO (8 eyes in group 1 and 2 in group 2) were the main causes of ocular ischemia. Synechial angle closure was seen in 65% of eyes in group 1 and 80% in group 2.

In group 1, marked regression of iris neovascularization was noticed from the first day after bevacizumab injec-tion, with improvement of corneal edema and inflamma-tory reactions (in 11 eyes there was complete regression of neovascularization, in 6 [30%] it was confined to 1 quad-rant, and in 3 [15%] it was confined to 2 quadrants at the time of shunt surgery). No intraoperative or postoperative complications were associated with bevacizumab injection. IOP decreased in some eyes compared with the preinjection level while patients were continuing on maximum glau-coma treatment (preinjection pressure was 46.5, SD 10.0, mm Hg and postinjection was 42.1, SD 11.6, mm Hg). After shunt surgery, IOP was significantly reduced in both groups (Table 1) as compared with preoperative values (p < 0.001) with significant reduction in the number of glau-coma medications. No statistically significant difference in final IOP (p = 0.2) or number of glaucoma medications (p = 0.6) was found between the groups. Table 2 lists a com-parable number of postoperative events and surgical inter-ventions in both groups. Intraoperative and postoperative anterior chamber bleeding was less pronounced in group 1 than group 2 (Figs. 1 and 2). One patient in group 1 had suffered pseudo-Brown syndrome with persistent diplo-pia due to superonasal insertion of the implant. The same implant was removed and fixed superotemporally with re-covery of normal ocular motility, disappearance of diplo-pia, and adequate control of IOP. Additional surgeries not related to the management of postoperative complications involved 2 eyes in group 1 (2 phacoemulsifications) and

Table 1— Comparison of continuous variables between the two study groups

Mean (SD)

VairableGroup 1 (IVB-ASTS)

n = 20Group 2 (PRP-ASTS)

n = 10 p value

Age, years 56.0 (9.7) 53.7 (17.6) 0.7Glaucoma duration, years 3.3 (2.0) 6.0 (5.6) 0.001Duration of ASTS after IVB or PRP, days 19.5 (20.6) 90.0 (71.6) 0.001Preoperative visual acuity 0.1 (0.2) 0.1 (0.2) 0.8Postoperative visual acuity 0.1 (0.2) 0.1 (0.1) 0.3Preoperative IOP, mm Hg 46.5 (10.0) 49.2 (12.3) 0.5Postoperative IOP, mm Hg 18.8 (4.6) 15.9 (6.7) 0.2Preoperative glaucoma medications, no. (%) 3.7 (1.6) 3.2 (1.6) 0.5Postoperative glaucoma medications, no. (%) 1.1 (1.4) 0.8 (1.3) 0.6Follow-up duration, months 12.5 (2.2) 16.4 (6.9) 0.03Note: IVB, intravitreal bevacizumab; ASTS, aqueous shunting tube surgery; PRP, panretinal photocoagulation; IOP, intraocular pressure.



4 eyes in group 2 (1 phacoemulsification, 1 trabeculectomy, 1 penetrating keratoplasty, and 1 pars plana vitrectomy).

Subgroup analysis of eyes that received bevacizumab re-vealed that 3 eyes (30%) in subgroup 1A and 7 (70%) in subgroup 1B had NVG with open anterior chamber angles. On the other hand, subgroup 1A had a predominance of PDR (7 eyes), whereas subgroup 1B had a predominance of CRVO (6 eyes). One eye in subgroup 1A and 3 eyes in subgroup 1B had repeated bevacizumab injection for ret-inal indications. In subgroup 1B, IOP was reduced by ap-proximately 8 mm Hg after injection (from 45.7, SD 12.1, to 37.6, SD 8.0, mm Hg, p = 0.1), whereas in subgroup 1A, post-bevacizumab pressure (46.7, SD 13.1, mm HG) was about the same (p = 0.9) as preoperative measures (47.3, SD 8.8, mm Hg). Final IOP measures after shunt surgery in both subgroups was similar (19.1, SD 2.5, and 18.5, SD 6.2, mm Hg, respectively). Kaplan-Meier sur-vival analysis (Fig. 3) revealed a 90% success rate of aque-ous shunting tube surgery when bevacizumab injection was followed by panretinal photocoagulation compared with

80% when bevacizumab was not followed by photocoagu-lation and 70% when photocoagulation was done without bevacizumab injection before the shunt surgery.

DISCUSSION

Elevated levels of VEGF-A have been identified in the aqueous humour of patients with rubeosis and NVG.17 Sone and coworkers18 found higher levels of VEGF in aque-

Table 2—Postoperative course and clinical outcome of aqueous shunting tube surgery

Count (%)

Variable OutcomeGroup 1 (IVB-ASTS)

n = 20Group 2 (PRP-ASTS)

n = 10

Operative complications Hyphema 0 1 (10)Postoperative complications Hyphema

Vitreous hemorrhageLost anterior chamber

Choroidal effusionTube blockage

Motility disturbanceCorneal graft failure

1 (5)00

2 (10)1 (5)1 (5)

0

2 (20)1 (10)1 (10)2 (20)

00

1 (10)Postoperative surgical intervention AC reformation

AC wash of hyphemaTube lavage

Choroidal drainageImplant repositioning

00

1 (5)1 (5)1 (5)

1 (10)1 (10)

01 (10)

0Postoperative PRP – 10 (50) 2 (20)Postoperative glaucoma medications 1 drop

2 drops3 drops

4 (20)2 (10)4 (20)

1 (10)1 (10)3 (30)

Success of ASTS Complete successQualified success

Failure

10 (50)7 (35)3 (15)

5 (50)2 (20)3 (30)

Note: IVB, intravitreal bevacizumab; ASTS, aqueous shunting tube surgery; PRP, panretinal photocoagulation; AC, anterior chamber.

Fig. 1—Postoperative hyphema and clotted blood around the tube tip after Ahmed valve im-plantation in neovascular glaucoma patients with florid iris neovascularization without intravitreal bevacizumab.

Fig. 2—Epithelial edema and iris neovasculariza-tion with dilated pupil and high intraocular pres-sure before intravitreal bevacizumab (A). After bevacizumab injection and Ahmed valve implanta-tion with quietness of the eye and disappearance of iris neovessels (B).



ous and vitreous humour in patients with diabetic NVG compared with patients with only PDR. These findings substantiate the importance of VEGF in the development of iris and angle neovessels.1 The IOP-reducing effect of bevacizumab injection in eyes with NVG has been dem-onstrated in some case reports.11–16 The mechanism of this IOP reduction is not clearly understood, especially in eyes with complete synechial angle closure. In NVG eyes with nonsynechial angle closure, the regression of neovascular tissue in the iris and angle decreases inflammatory reaction in the anterior chamber and may improve filtration via the remaining functioning trabecular tissue. This may not be the case when the angle is totally closed by contracting fi-brovascular membrane. After bevacizumab injection for refractory glaucoma in the first group in our study, iris neovascularization and inflammatory reactions regressed markedly with only mild reduction of IOP, which was more pronounced in subgroup 1B (7 of 10 eyes had nonsynech-ial angle closure). The decrease in ocular inflammation and regression of angle rubeosis may improve ocular responses to glaucoma medications and at the same time improve filtration through residual parts of the trabecular mesh-work in eyes with incomplete angle closure. Yazdani and associates13 postulated a “reversible anatomic closure” in angles not yet completely and permanently compromised by the abnormal tissue; with regression of the neovascular membrane, the angle may be partially relieved from the pretrabecular obstruction. Another explanation by the same authors is the presence of functional trabecular tissue at a microscopic level in eyes considered gonioscopically to be totally closed. Regression of angle neovessels may improve filtration through this microscopic tissue.13

The hypotensive effect of bevacizumab in NVG is tem-porary (signs of reactivation of the neovascular process appear within 6–8 weeks) and may require another ap-plication, especially if the ischemic process in the retina is not properly treated.13,15,16 However, repeated intravitreal injections expose the patient to the hazards of ocular infec-tion and bleeding beside the stress of being in the operating room several times. Accordingly, bevacizumab injection by itself cannot be considered a definitive treatment for NVG, even in eyes with nonsynechial angle closure, unless com-bined with panretinal photocoagulation with complete ab-lation of retinal ischemic tissue and permanent regression of angle and iris rubeosis.

Our study supports the adjuvant rule of bevacizumab injection to more definitive treatment strategies for re-fractory NVG, such as aqueous shunting tube surgery. The dramatic effect of bevacizumab on regression of iris neovascularization and intraocular inflammation made the shunt surgery less eventful compared with the antici-pated high rate of operative complications if the drainage tube is implanted in the presence of florid neovascular-ization with very high IOP and markedly inflamed eye. The introduction of bevacizumab injection seems to re-duce surgical complications in the anterior and posterior segments if glaucoma surgery is performed. We found a comparable rate of operative and postoperative compli-cations and postoperative surgical interventions in the group that had prior bevacizumab and the group with previous photocoagulation. In the latter, the operative and postoperative adverse events were more promin-ent in eyes with persistent iris neovascularization after photocoagulation.

The success of aqueous shunting tube surgery in patients who received bevacizumab injection (85%) in our study was better than that seen in eyes treated with photoco-agulation only (70%). Complete success in IOP control (50% each) as well as the number of glaucoma drops at final analysis was similar in both groups. In group 1, further treatment of the retina with photocoagulation after shunt surgery may have promoted the success of the glaucoma surgery (1 failure was reported after additional photoco-agulation compared with 2 failures in eyes not treated with photocoagulation after bevacizumab injection). Pro-longed follow-up is necessary to confirm the promoting ef-fect of bevacizumab on the long-term success of aqueous shunting tube surgery. The dramatic short-term effect of bevacizumab on regression of neovessels and inflammatory reaction may be followed by recurrence of the neovessels and pressure elevation. However, combining bevacizumab with good panretinal photocoagulation assures permanent ablation of ischemic retina and the disappearance of iris neovascularization with low recurrence rate. In the sub-group without photocoagulation treatment, 3 eyes required repeated bevacizumab injection compared with 1 eye in the subgroup that had photocoagulation after bevacizumab in-jection and shunt surgery.

Fig. 3—Kaplan-Meier survival analysis of probability of failure of aqueous shunting tube surgery with intravitreal bevacizumab injec-tion only (IVB-ASTS, subgroup 1A) or followed by panretinal pho-tocoagulation (IVB-ASTS-PRP, subgroup 1B) compared with tube surgery with previous panretinal photocoagulation without bevaci-zumab injection (PRP-ASTS, group 2).



In summary, intravitreal bevacizumab is a useful prepar-atory step to safely and effectively implanting an aqueous shunting tube in eyes with severe NVG and intractable IOP. Postoperative panretinal photocoagulation reduces the recurrence of neovascular activity and improves the success rate of aqueous shunting surgery without the need for re-peated bevacizumab injection.

The authors have no proprietary interest in any aspect of this article.

REFERENCES

1. Sivack-Callcott JA, O’Day DM, Gass DM, Tsai JC. Evidence-based recommendations for the diagnosis and treatment of neo-vascular glaucoma. Ophthalmology 2001;108:1767–76.

2. Oguri A, Takashashi E, Tomita G, Yamamoto T, Jikihara S, Kitazawa Y. Transscleral cyclophotocoagulation with the di-ode laser for neovascular glaucoma. Ophthalmic Surg Lasers 1998;29:722–7.

3. Bloom PA, Tsai JC, Sharma K, et al. “Cyclodiode”. Trans-scleral diode laser cyclophotocoagulation in the treatment of advanced refractory glaucoma. Ophthalmology 1997;104:1508–19.

4. Mermoud A, Salmon JE, Alexander P, Straker C, Murray AD. Molteno tube implantation for neovascular glaucoma. Long-term results and factors influencing the outcome. Ophthalmol-ogy 1993;100:897–902.

5. Sidoti PA, Dunphy TR, Baerveldt G, et al. Experience with the Baerveldt glaucoma implant in treating neovascular glaucoma. Ophthalmology 1995;102:1107–18.

6. Eid TM, Katz LJ, Spaeth GL, Augsburger JJ. Tube-shunt surgery versus neodymium:YAG cyclophotocoagulation in the management of neovascular glaucoma. Ophthalmology 1997;104:1692–1700.

7. Funatsu H, Yamashita H, Sakata K, et al. Vitreous levels of vas-cular endothelial growth factors and intracellular adhesion mol-ecule 1 are related to diabetic macular edema. Ophthalmology 2005;112:908–16.

8. Hurwitz H, Fehrenbacher L, Novotny W, et al. Bevacizumab plus irinotecan, fluorouracil, and leucovorin for metastatic colo-rectal cancer. N Engl J Med 2004;350:2335–42.

9. Rosenfeld PJ, Fung AE, Puliafito CA. Optical coherence tom-ography findings after intravitreal injection of bevacizumab (Avastin) for macular edema from central retinal vein occlusion. Ophthalmic Surg Lasers Imaging 2005;36:336–9.

10. Avery RL, Pieramici DJ, Rabena MD, Castellarin AA, Nasir MA, Giust MJ. Intravitreal bevacizumab (Avastin) for neo-vascular age-related macular degeneration. Ophthalmology 2006;113:363–72.

11. Gheith ME, Siam GA, de Barros DS, Garg SJ, Moster MR. Role of intravitreal bevacizumab in neovascular glaucoma. J Ocul Pharmacol Ther 2007;23:487–91.

12. Batioglu F, Astam N, Ozmert E. Rapid improvement of retinal and iris neovascularization after a single intravitreal bevacizumab injection in a patient with central retinal vein occlusion and ne-ovascular glaucoma. Int Ophthalmol 2008;28:59–61.

13. Yazdani S, Hendi K, Parkravan M. Intravitreal bevacizumab (Avastin) injection for neovascular glaucoma. J Glaucoma 2007;16:437–9.

14. Vatavuk Z, Bencic G, Mandic Z. Intravitreal bevacizumab for neovascular glaucoma following central retinal artery occlusion. Eur J Ophthalmol 2007;17:269–71.

15. Iliev ME, Domig D, Wolf-Schnurrbursch U, Wol S, Sarra GM. Intravitreal bevacizumab (Avastin) in the treatment of neovascu-lar glaucoma. Am J Ophthalmol 2006;142:1054–6.

16. Kahook MY, Schuman JS, Noecker RJ. Intravitreal bevacizumab in a patient with neovascular glaucoma. Ophthalmic Surg Lasers Imaging 2006;37:144–6.

17. Tripathi RC, Li J, Tripathi BJ, Chalam KV, Adamis AP. In-creased levels of vascular endothelial growth factor in aqueous humor of patients with neovascular glaucoma. Ophthalmology 1998;105:232–7.

18. Sone H, Okuda Y, Kawakami Y, et al. Vascular endothelial growth factor level in aqueous humor of diabetic patients with rubeotic glaucoma is markedly elevated (Letter). Diabetes Care 1996;19:1306–7.

Keywords: bevacizumab, neovascular glaucoma, aqueous shunting tube surgery

intravitreal bevacizumab and aqueous shunting surgery for neovascular glaucoma: safety and efficacy

Documents