Trabeculectomy with concurrent intravitreal bevacizumab in neovascular glaucoma

Abstract

Purpose:

To evaluate the clinical efficacy of concurrent intravitreal bevacizumab (IVB) injection with trabeculectomy with mitomycin-C (MMC) in neovascular glaucoma (NVG).

Methods:

Patients with NVG who underwent trabeculectomy with concurrent IVB (group 1) and those who underwent IVB sequentially, followed by trabeculectomy with MMC (group 2) in 1–2 weeks between January 2021 and August 2022, were included in this retrospective hospital-based study. The need for medications for intraocular pressure (IOP) control at 6 months in the two groups was the primary outcome measured and compared between the groups. The association of the need for medications postoperatively with clinical variables was assessed using stepwise multivariate regression statistics.

Results:

We finally included 40 patients (n = 12 in group 1, n = 28 in group 2) with no significant differences in presenting age between groups. The IOP at 1 day and 1 week were not significantly different between groups though the IOP at 1, 3, and 6 months. IOP was lower in group 1 eyes with the 6-month IOP, being significantly lower in group 1, P = 0.05. Three eyes in group 1 and 11 eyes in group 2 required anti-glaucoma medications in the postoperative period. Multivariate regression identified preoperative IVB >3 (β =0.7, P < 0.001) and recurrent vitreous hemorrhage (β = 0.7, P = 0.004) as prognostic factors (R² = 40.6%) determining the need for anti-glaucoma medication (AGM) postoperatively in both groups.

Conclusion:

Concurrent IVB with trabeculectomy with mitomycin-C is a feasible alternative in patients with NVG with refractory high-presenting IOP. This may serve to address raised IOP as well as retinal ischemia, thereby improving surgical success rates in the most challenging NVG cases.

Neovascular glaucoma (NVG) is a form of secondary refractory glaucoma, caused by the formation of new vessels on the iris and the anterior chamber angle.[1,2,3,4] The pathogenesis of NVG is mostly due to retinal ischemia (97%) and some caused due to inflammation.[1,2] Retinal ischemia causes the release of pro-angiogenic factors such as vascular endothelial growth factor (VEGF) that tip the balance toward angiogenesis causing neovascularization.[1] Management of NVG is difficult as it requires multiple factors to be addressed concurrently or sequentially, including raised intraocular pressure (IOP), retinal ischemia, and control of systemic risk factors.[3,4,5,6,7,8] Formation of new vessels in the angle causes refractoriness to conventional trabeculectomy, poor prognosis, and high risk of failure.[6,7,8]

Anti-VEGF molecules (bevacizumab, ranibizumab, and aflibercept) have helped improve outcomes in such vision-threatening clinical situations and are a boon for patients with refractory NVG.[5,6,8,9] They help reduce neovascularization at the iris and angles, and help indirectly in the reduction of IOP, thereby helping prevent blindness in such patients.[5,9] It is common practice to administer anti-VEGF 5–7 days before glaucoma-filtering surgeries to allow a reduction in the level of ischemia, which can be a poor prognostic factor for surgical success. Although modulated trabeculectomy with mitomycin-C (MMC) is recommended for NVG, high failure rates with the former have caused glaucoma implants to be favored as the primary procedure for refractory NVG.[5,7,8,9,10,11] A few randomized controlled trials have used concurrent anti-VEGF with concurrent glaucoma implants and pan-retinal photocoagulation (PRP), with the paucity of consensus and clarity on the surgical paradigms for treating NVG.[4,7,11] In developing countries, it is common to see patients with very high IOP, closed angles, and very minimal useful vision at presentation.[8] Lack of awareness, poor socioeconomic status, and inability to afford glaucoma medications, and loss of follow-up after anti-VEGF injections are additive issues responsible for poor surgical outcomes and high rates of failure in NVG.[7,8] Such clinical scenarios mandate exploring measures to tackle retinal and glaucoma procedures concurrently while improving surgical outcomes. This study evaluates the outcomes of adjuvant intravitreal bevacizumab with modulated trabeculectomy in NVG compared to sequential procedures as performed traditionally.

Methods

A retrospective hospital database patient chart review was performed from January 2021 to August 2022, to review the details of patients with neovascular glaucoma who underwent trabeculectomy with MMC at a tertiary center. Patients unlikely to come for review or afford medications after injections were offered concurrent glaucoma surgery with intravitreal bevacizumab (IVB) injections. Patients who underwent trabeculectomy with concurrent intravitreal bevacizumab (group 1) and those who underwent IVB sequentially followed by trabeculectomy with MMC (group 2) in 1–2 weeks were identified. The study was approved by the institutional review board of LV Prasad Eye Institute, MTC campus, Bhubaneswar (IEC-16-IM-3), and patients were included after receiving their written informed consent. The preoperative and postoperative best-corrected visual acuity (recorded in equivalent logMAR units), slit-lamp examination, IOP measured by Goldmann Applanation Tonometry at each follow-up visit, including presenting IOP before surgery, gonioscopy to identify open or closed angles, and for the presence of neovascularization of the angle (NVA), dilated fundus examination, the number of medications, and need for additional surgeries/anti-VEGF injections/retinal lasers at 6 months were retrieved from the hospital database. The diagnosis of neovascular glaucoma was performed based on the presence of neovascularization of the iris or/and angle (NVI and NVA), with raised IOP. Both open and closed angles with NVG were included. Patients with loss of vision (perception of light only), concurrent vitreous hemorrhage at presentation, uncontrolled diabetes, lack of follow-up of at least 6 months after surgery, non-consenting patients, and patients with a history of previous retinal laser or intravitreal injections 1–4 months before presentation, were excluded.

Surgical procedure

The trabeculectomy surgery with MMC was performed with a fornix- or limbal-based conjunctival flap in the superior quadrant. Subconjunctival application of 0.04% MMC was performed for 3 min in all cases, followed by a thorough wash with a balanced salt solution (20 mL). The scleral flap was closed with 10-nylon sutures, whereas the conjunctiva was closed with 8–0 vicryl interrupted wing sutures. For group 1, IVB 0.05 mL was injected in the inferotemporal quadrant at the end of the procedure at 3.5 mm from the limbus, followed by topical 5% betadine drops instillation at the end of the procedure. The patients were started on topical steroids (1% prednisolone acetate) 6 to 8 times, for 1 week and tapered thereafter over 4–6 weeks, and topical antibiotics (ofloxacin 0.3%) 4 times daily for 1 week. The patients were followed at 1 day, 1 week, 1 month, 3 months, and 6 months. For patients who had not received PRP before presentation, PRP in multiple sittings was performed in the immediate 1-week postoperative period. Add-on PRP was performed within 1 month of surgery in both groups for patients who had received previous PRP before surgery or presentation. The need for medications for IOP control at 6 months in the two groups was the primary outcome compared between groups, whereas the need for additional surgeries, complications, and additional vitreoretinal or other procedures, was analyzed between groups.

Descriptive statistics are presented as mean ± standard deviation or as proportions. The clinical variables between groups were compared using the independent student’s t-test student and the Chi-square test. The association of final IOP and the need for medications postoperatively with clinical variables such as presenting IOP at the time of surgery, age, sex, cause of NVG, and need for IVB and PRP preoperatively were assessed using stepwise multivariate regression statistics. All analyses were performed using Stata (Stata Corp., USA) with P < 0.05 considered statistically significant.

Results

Of a total of 54 patients who underwent trabeculectomy for NVG during the study period, 12 were excluded for lack of follow-up beyond 1 month, whereas 2 patients who had received IVB 0.5–2 months before trabeculectomy, were excluded. We finally included 40 patients (n = 12 in group 1, n = 28 in group 2) with no significant differences in the presenting age between groups [Table 1]. The baseline IOP was higher in group 1 eyes with most of the eyes at presentation in both groups on >2 medications at the time of surgery. Patients had received preoperative IVB in 5 of 22 and 22/28 eyes in each group respectively, with 3 eyes in group 1 and 5 in group 2 requiring >3 injections with or without pars plana vitrectomy for recurrent vitreous hemorrhage 6 months before surgery. Five eyes in group 1 and 13 eyes in group 2 had received complete PRP before surgery. There were no statistical differences in the cause for NVG (though group 1 had more eyes with OIS) or the proportion of eyes with NVA at presentation between the two groups [Table 1]. The visual acuity in both eyes at presentation did not significantly differ between the two groups.

Table 1.

Clinical profile of patients with neovascular glaucoma who underwent modulated trabeculectomy with or without concurrent IVB

	Group 1 n=12	Group 2 n=28	P
Age (years)	58±10.1	57±14.3	0.8
Presenting IOP at the time of surgery (mm Hg)	41±5.08	33±10.7	0.01
Presenting visual acuity (LogMAR)	1±0.7	1±0.8	0.3
Male:female (n)	7:5	24:4	0.057
Cause of NVG
OIS	3	1	0.03
PDR	3	13	0.2
RVO	4	12	0.6
Others	2	2	0.3
Gonio status (n)
Closed angle	6	16	0.6
NVA	9	18	0.5
Preop PRP (n)	5	13	0.8
Preop IVB (n)	5	22	0.02

IVB-Intravitreal bevacizumab; NVA-neovascularization of the angle; PRP-pan-retinal photocoagulation; NVG-neovascular glaucoma, OIS-ocular ischemic syndrome; PDR-proliferative diabetic retinopathy; RVO-retinal vein occlusion; IOP-intraocular pressure

All eyes underwent uneventful trabeculectomy with MMC with no intraoperative complications observed in any eye in either group. Two eyes in group 2 developed decompression retinopathy with one eye having concurrent shallow choroidals that resolved with conservative treatment over 4 weeks, 1 eye in group 1 and 2 in group 2 had transient hyphema that resolved with routine medications over 2 weeks.

The IOP at 1 day and 1 week were not significantly different between groups though the IOP at 1, 3, and 6 months IOP were lower in group 1 eyes, with the latter being significantly lower, P = 0.05 (Table 1 and Fig. 1). Three eyes in group 1 and 11 eyes in group 2 required anti-glaucoma medications in the postoperative period[Table 2]. All 3 eyes in group 1 and 10 of 11 eyes in group 2 that required medications for IOP control postoperatively had received >3 IVB preoperatively with pars plana vitrectomy for recurrent vitreous hemorrhage. Of these, 1 patient in group 1 underwent cataract surgery 8 months after surgery, whereas 9 eyes in group 2 underwent vitreoretinal (VR) surgery (n = 8) and cataract surgery (n = 1). Four in group 1 and 6 eyes in group 2 also received added PRP at 1–2 months of surgery after review with retina services. None of the eyes underwent needling in either group or any repeat surgery for bleb-related procedures. Two eyes in group 2 had decreased vision due to repeat surgeries and optic disc pallor at 1 year, whereas none of the eyes in group 1 experienced visual loss at the final follow-up. The preop-(1.4 ± 0.8, group 1 and 1.2 ± 0.8 group 2, P = 0.2) and postoperative visual acuity (1.2 ± 0.7 group 1 and 1.3 ± 0.9 group 2, P = 0.6) was not significantly different between the groups.

Figure 1.

IOP profile of patients with neovascular glaucoma who underwent modulated trabeculectomy with or without concurrent IVB

Table 2.

Postoperative outcomes in patients with neovascular glaucoma who underwent modulated trabeculectomy with or without concurrent IVB

	Group 1 n=12	Group 2 n=28	P
IOP 1 day (mm Hg)	12±9.4	14±9.8	0.6
IOP 1 week (mm Hg)	10±9.7	14±12.8	0.056
IOP 1 month (mm Hg)	14±8.6	20±11.8	0.04
IOP 3 months (mm Hg)	17±3.04	19±9.4	0.6
IOP 6 months (mm Hg)	13±1.02	17±11.7	0.05
Eyes needing AGM after surgery (n)	3	11	0.3
Additional retinal surgeries	1	9	0.1
Postop IVB (n)	2	8	0.4
Postop PRP (n)	4	6	0.4
Final VA (LogMAR)	1±0.7	1±0.8	0.8

IOP-intraocular pressure, VA-visual acuity, AGM-anti-glaucoma medication; IVB-Intravitreal bevacizumab; PRP-pan-retinal photocoagulation; NVG-neovascular glaucoma

Multivariate regression identified preoperative IVB >3 (β = 0.7, P < 0.001) and recurrent vitreous hemorrhage (β = 0.7, P = 0.004) as prognostic factors (R² = 40.6%) determining the need for anti-glaucoma medication (AGM) postoperatively in both groups, whereas age, presenting IOP, or cause of NVG did not have any significant influence on the final IOP or the need for AGM.

Discussion

This study found significant IOP reduction with reduced need for additional medications after concurrent IVB with modulated trabeculectomy with MMC compared to the two procedures performed sequentially as routinely practiced. The final IOP at 6 months and the number of eyes that needed medications after surgery were lower in group 1 than in group 2, suggesting that concurrent administration of IVB with filtering surgery may help relieve the ischemic load while reducing IOP at the same time. This allows for faster control of the underlying diseases such as diabetic retinopathy with additional procedures such as PRP that can be completed 1–2 weeks after surgery, allowing salvage of useful vision in most eyes with NVG.

NVG has two important factors that need to be controlled, namely ischemia and raised IOP.[1,2,3] Retinal ischemia in NVG adds to the VEGF load in the eye, which, in turn, causes angiogenesis and new vessels causing raised IOP by obstruction of the outflow pathways. The raised OP in turn causes more ischemic in an eye with underlying retinal ischemia causing a vicious loop.[1,3] The management of NVG, therefore, not only requires control of IOP but also requires correction and reduction in the ischemic load by anti-VEGF, retinal ablation, or other additional surgeries along with control of the systemic risk factors such as diabetes or hypertension. Conventionally, IVB is administered 5–7 days before any intraocular surgery aiming to reduce the VEGF load inside the eye.[3,5,8] While this standard approach works for early cases and in developed countries, this, however, poses a challenge in developing countries with most patients presenting late with very high presenting IOP, severe disease/closed angles at presentation, and frequent loss of follow-up after IVB injections.[8] Concurrent IVB with trabeculectomy with MMC may offer a feasible solution in such challenging clinical scenarios, enabling both IOP control (and the IOP-related ischemia thereby) and retinal ischemia in the same sitting. This also allows faster administration of PRP in the immediate 12 weeks postoperative period, which otherwise would get delayed because of high IOP, poor uptake of the laser with high IOP, coexisting corneal edema precluding fundus view by raised IOP by IVB injection alone until filtering surgery is performed.

The significant reduction in IOP and the need for medications sustained at 6 months suggests that concurrent IVB administration can be a feasible alternative to conventional sequential surgery for most NVG scenarios in developing countries for selected patients. This can be offered to patients with closed-angle NVG at presentation, status post multiple surgeries/interventions, delayed presentation, and very high presenting IOP coming from remote areas in developing countries. Studies have reported a cumulative success of 83% at 1 and 3 years of follow-up with the use of bevacizumab before trabeculectomy.[6,8,9,12,13,14,15,16,17] Saito et al.[9] reported complete and qualified success of 95% versus 50% trabeculectomy with bevacizumab compared to 95% versus 75% in patients who underwent trabeculectomy alone. A study by Takihara et al.[13] comparing two groups with or without bevacizumab pre-operatively before trabeculectomy with MMC showed similar success rates of 65.2% in the bevacizumab group and 65.3% in the control group at 5 years with significant IOP reduction immediately after surgery in the bevacizumab group. Similar results have been reported with good success rates of bevacizumab adjuvant use with glaucoma implants in NVG.[11,18,19,20] However, there is still a lack of consensus on standard treatment regimens in difficult challenging scenarios in developing countries. This study suggests that adjuvant IVB may be useful with filtering surgeries to achieve good IOP control while also tacking retinal ischemia in the same sitting.

Heightened angiogenesis and wound epithelialization lead to aggressive wound healing and failure with traditional trabeculectomy in NVG eyes.[8,10] The mechanism of bevacizumab lies in the reduction of VEGF concentration in the anterior and posterior chambers, which, in turn, leads to a reduction in neovascularization at the angle/iris/retina.,[3,5] The use of anti-VEGF results in a decrease in the pro-inflammatory response in NVG eyes, leading to improved success rates with trabeculectomy or implants.[3,5,7,20,21] There are many studies showing benefits with the use of bevacizumab before trabeculectomy showing improvement in results in the management of NVG.[9,12,13,14,15,16,17] To our knowledge, there is no study reporting the usage of concurrent IVB along with trabeculectomy in NVG. A study by Kaushik et al.[21] compared the use of concurrent bevacizumab along with AGV or AGV alone and reported fewer complications along with higher NVI regression in eyes receiving concurrent bevacizumab. Surgical success rates were comparable in both groups with an IOP reduction of 14.25+ −2.05 mm Hg in the bevacizumab group and 15.25 + 2.95 mm Hg in the group without bevacizumab. In trabeculectomy, the regression of new vessels on the iris and angle by IVB caused decreased chances of hyphema, whereas reduced inflammatory drive results in decreased wound fibrosis at the surgical site. Saito et al.[9] reported a decrease in postoperative hyphema in trabeculectomy performed after bevacizumab. This study did not find any increased rate of hyphema with concurrent IVB with trabeculectomy, suggesting that careful surgery, with slow decompression during filtering surgery, may help reduce hyphema rates rather than IVB per se. Further, reduced angiogenesis by initial IVB administration may be offset by ischemia, resulting from raised uncontrolled IOP in a sequential surgery regimen as conventionally practiced. This study, therefore, suggests that concurrent control of IOP along with treating the primary cause of retinal ischemia by IVB, followed by PRP in the immediate postoperative period can be completed within shorter intervals. This may help reduce the number of office visits for patients in developing countries while improving success rates and minimizing long-term blindness by NVG.

This study identified that repeated IVB before presentation with NVG or recurrent VH may be a bad prognostic factor determining the need for AGM in the postoperative period in NVG. Our earlier study had identified closed angles, suggesting severe disease as prognosticating worse clinical outcomes in NVG.[8] Both groups in this study had a similar number of eyes with closed angles. However, the need for medications was lower in eyes receiving concurrent IVB, suggesting that this was not a factor predicting the need for medications after trabeculectomy. The causes for recurrent VH in diabetic retinopathy can be varied, one of them being rubeotic/pre-rubeotic glaucoma.[1,2] The need for repeated IVB > 3, as observed in this study, may also indicate heightened angiogenic factors in that eye signaling the development of NVG or a pre-rubeotic stage of NVG. Eyes that require >3 IVB injections for any non-resolving disease or recurrent VH should alert the surgeon to the possible development of NVG in that eye. This mandates gonioscopy and closer follow-up and early diagnosis and treatment in such eyes to prevent the onset of NVG or for preventing blindness in NVG.

The limitations of this study include a retrospective design and a short follow-up of 6 months. We did not evaluate the bleb status of the eyes in the two groups, nor did we evaluate the quantitative VEGF levels in the eyes in the two groups.

Conclusion

Nevertheless, the advantage of IVB along with modulated trabeculectomy with MMC includes simultaneous control of IOP and retinal ischemia while allowing reduced office visits for subsequent PRP and reduced need for medications. This is a feasible alternative to treat NVG eyes with very high presenting IOP and challenging clinical scenarios that can help reduce blindness rates in NVG in developing countries.

Financial support and sponsorship:

Nil.

Conflicts of interest:

There are no conflicts of interest.