Clinical outcomes of PRESERFLO^® Microshunt procedure with mitomycin C alone versus with mitomycin C and bevacizumab

Abstract

Background:

The PRESERFLO^® Microshunt procedure offers a safe and effective alternative to trabeculectomy procedure. The adjuvant use of anti-vascular endothelial growth factor (VEGF) with trabeculectomy has been well studied, yet little is known on the effect and safety of anti-VEGF agents used as an adjunctive therapy in PRESERFLO^® procedures.

Objectives:

To compare the outcomes of PRESERFLO Microshunt procedure using either mitomycin C (MMC) alone versus MMC augmented with bevacizumab in patients with open angle glaucoma to a follow-up time of 6 months.

Design:

Retrospective observational single-center single-surgeon study.

Methods:

Patients’ records of patients who consecutively underwent PRESERFLO Microshunt at Queen Victoria Hospital, United Kingdom, were examined. From December 2018 to January 2020, eligible patients underwent PRESERFLO Microshunt implantation with MMC alone (0.2–0.4 mg/ml), whereas from February 2020 to January 2022, patients underwent PRESERFLO Microshunt implantation with MMC (0.2–0.4 mg/ml) and adjuvant intracameral 0.1 ml of bevacizumab (1.25 mg/0.05 ml). Efficacy outcomes were analyzed, including changes in intraocular pressure (IOP) and changes in medication use. Postoperative complications and postoperative interventions were also reviewed.

Results:

A total of 75 eyes were included in the analysis, 38 eyes received MMC alone, whereas 37 eyes received MMC combined with bevacizumab. There were no statistically significant differences in the demographic or clinical profile of patients between treatment groups.

Both strategies were effective in terms of IOP lowering (baseline vs 6 months postoperatively: 20.0 (6.8) mmHg vs 12.8 (3.7) mmHg in the MMC group; 23.6 (6.9) mmHg vs 11.9 (4.2) mmHg in the MMC+ bevacizumab group; p < 0.001 in both comparisons). Mean anti-glaucoma medication use also reduced significantly from baseline with no difference between both groups. A higher proportion of patients required postoperative interventions (5-FU injection, needling, and bleb revision) in the MMC alone group (N = 16; 42.1%) compared to the MMC+ bevacizumab group (N = 4; 9.9%) which was statistically significant (Pearson’s χ² test; p = 0.002).

Conclusions:

Adjuvant use of bevacizumab in MMC-augmented PRESERFLO Microshunt procedure is at least as effective as MMC alone but helps reduce rates of postoperative interventions.

Precis

Glaucoma patients undergoing PRESERFLO^® Microshunt procedure with mitomycin C combined with adjunct intracameral bevacizumab had at least similar safety and efficacy as mitomycin C alone but with statistically significant lower rates of postoperative interventions including revisions and needling at 6-months postoperatively.

Introduction

Trabeculectomy remains the mainstay of surgical treatment for medically uncontrolled glaucoma or when glaucoma progression warrants surgery. ¹ Despite its efficacy at lowering intraocular pressure (IOP), trabeculectomy requires significant postoperative management, which delays recovery and could be associated with potentially sight-threatening complications. ²

The PRESERFLO^® Microshunt (Santen, Miami, FL, USA) offers an alternative to trabeculectomy, if the single digit eye pressure is not needed. It is a device measuring 8.5 mm long with a 350-µm outer diameter and 70-µm lumen and is made from a biocompatible material: poly(styrene-block-isobutylene-block-styrene), alternatively known as “SIBS.”^3,4 The MicroShunt is inserted via an ab externo procedure. After implantation, the proximal tip of the device sits in the anterior chamber, parallel to the iris, and the distal tip sits underneath the conjunctiva and Tenon’s capsule, approximately 6 mm behind the limbus, allowing flow of aqueous humor through the lumen to form a posterior bleb.^5,6 The device drains aqueous humor from the anterior chamber to a bleb in the subconjunctival space similar to trabeculectomy. However, the procedure is considered to be less invasive and is associated with a lower risk of complications and faster recovery. The risk of hypotony is minimized in the PRESERFLO Microshunt by a valve-less intrinsic flow limiting design. Moreover, the biocompatible material was designed to decrease postoperative inflammation and the risk of fibrosis. As with trabeculectomy, Mitomycin C (MMC) is used intraoperatively to improve bleb morphology and achieve higher surgical success rates by inhibiting subconjunctival fibrosis, which is the main cause of bleb failure. ⁷

Antimetabolites such as MMC and 5-fluorouracil (5-FU) are commonly used to inhibit fibroblast activity following glaucoma filtration surgery.^8,9 However, due to their nonspecific mechanisms of action, both MMC and 5-FU are associated with widespread cell death, resulting in vision-threatening complications such as ocular hypotony, shallow anterior chamber, choroidal detachment, hypotonous maculopathy, bleb leakage, blebitis, and bleb-related endophthalmitis.^{10 –13}

Several studies demonstrated reasonable safety and effectiveness of the PRESERFLO Microshunt in patients with primary open-angle glaucoma (POAG), with greater success reported when higher doses of MMC (0.4 mg/ml) were used.^{14 –18} However, the standard use of MMC at high doses (0.4 mg/ml) regardless of each patient’s risk factors for scarring could be associated with higher rates of MMC-related complications.^{10 –13}

To improve surgical success rates of the PRESERFLO Microshunt without compromising its safety, adjuvant therapeutic agents could be used with MMC (0.2–0.4 mg/ml) to modulate wound healing and maintain bleb functioning. Anti-vascular endothelial growth factor (VEGF) agents have less cytotoxic but more specific physiological actions while modulating wound healing. ²⁴ VEGF plays a crucial role in wound healing and scarring as it stimulates angiogenesis and increases vascular permeability, in addition to enhancing fibroblast proliferation and activity. One study showed that VEGF is expressed in aqueous humor from glaucoma patients and rabbits that have undergone surgery. VEGF was also found to stimulate the proliferation of Tenon’s fibroblasts in cell cultures. ¹⁹ Bevacizumab (Avastin^®) is a recombinant full-length humanized antibody that binds to all types of VEGF, which has been shown to inhibit fibroblast proliferation and improve scarring in an animal model by reducing angiogenesis and collagen deposition. ¹⁹

Some studies show that the adjuvant use of anti-VEGF agents with MMC in trabeculectomy offers higher success rates, whereas other studies showed no added benefit.^{20 –23} Given the growing appeal of PRESERFLO Microshunt as an alternative to trabeculectomy, the purpose of this study was to compare the outcomes of PRESERFLO Microshunt using MMC alone versus MMC augmented with intracameral anti-VEGF agent; bevacizumab (Avastin), in patients with open angle glaucoma to a follow-up time of 6 months.

To our knowledge, this is the first study that investigates the use of intracameral bevacizumab as an adjunctive treatment to MMC during PRESERFLO Microshunt procedure.

Methods

Study design

Retrospective analysis of patients who underwent PRESERFLO Microshunt procedure between December 2018 and January 2022 at Queen Victoria Hospital, NHS Foundation Trust, United Kingdom. An electronic patient record system was used to review safety and efficacy outcomes over 6 months postoperatively in both groups. We adopted similar approach that was used by Tanner et al. for analysis. ²⁴

Inclusion criteria

Clinical criteria to list patients for PRESERFLO Microshunt procedure were: glaucomatous patients with uncontrolled IOP or progressive glaucomatous visual field loss despite maximum tolerated medical therapy. All glaucoma patients who were listed for the procedure during this period and completed 6 months of follow-up were included in the study.

Exclusion criteria

Patients who did not complete 6 months of follow-up after surgery were excluded from the study. In this study, only one patient was lost to follow up in the bevacizumab group due to relocation.

A total of 38 patients underwent PRESERFLO Microshunt procedure with MMC alone, between December 2018 and January 2020, whereas a total of 37 patients underwent PRESERFLO Microshunt procedure with adjuvant use of intracameral bevacizumab together with MMC, between February 2020 and January 2022.

The initial surgical approach involved the use of MMC alone with the microshunt. The incorporation of intracameral bevacizumab later on followed reflections from clinical experts on the potential clinical utility of bevacizumab in augmenting the effect of MMC.

Surgical procedure

All procedures were performed by a single consultant surgeon. Most of the procedures were performed under sub-Tenon’s anesthesia in both groups. A fornix-based conjunctival peritomy and Tenon’s dissection was made, either in the supero-nasal or supero-temporal quadrant. A deep sub-Tenon’s pocket was formed, over an area of 6–8 mm to ensure adequate placement of the 8.5-mm long microshunt. MMC (0.2–0.4 mg/ml) was applied under Tenon’s for 3 min using 3 LASIK shield sponges, then irrigated thoroughly with a balanced salt solution.

The concentration of MMC was titrated according to each patient’s risk factor for hypotony and excessive scarring. In eyes predisposed to hypotony such as those with high myopia, chronic uveitis, and thin conjunctiva, a concentration of 0.2 mg/ml MMC was used. In contrast, for eyes that were more prone to scarring such as Afro-Caribbean patients, eyes with previous failed trabeculectomy, thick conjunctiva, and Tenon’s, a concentration of 0.4 mg/ml MMC was used.

A scleral tunnel, about 3–3.5 mm from the limbus, was marked and then created using the supplied 1-mm blade, and then anterior chamber entry was completed using a 25-gauge needle provided with the Microshunt kit. Using forceps, the Microshunt was threaded, beveled up and fins flat, through the scleral tunnel into the anterior chamber. The fins were then wedged into the scleral pocket. The flow was confirmed visually at the end of the tip. Additionally, the anterior chamber was inflated with balanced salt solution. The distal end of the Microshunt was tucked underneath Tenon’s capsule and the conjunctiva, ensuring that it was straight and free of tissue. Conjunctiva and Tenon’s were then closed in two separate layers using 9/0 monofilament Vicryl sutures.

At the end of the procedure, 0.5 ml of subconjunctival dexamethasone and 0.1 ml of intracameral cefuroxime were given in both groups. In the bevacizumab group, 0.1 ml of bevacizumab (1.25 mg/0.05 ml) was injected intracamerally.

Perioperative management

According to the Queen Victoria Hospital protocol, all patients received preservative-free topical steroids for 2 weeks three times daily preoperatively to optimize their ocular surface. When conjunctival inflammation was detected, patients were switched to preservative free glaucoma eye drops at the time of listing. In case of an IOP increase, oral acetazolamide was considered if discontinuation of glaucoma drops was needed while awaiting surgery. To optimize the ocular surface preoperatively, blepharitis was carefully looked for and treated promptly with topical lubricants and oral antibiotics, if necessary.

Postoperatively, patients were reviewed on the same day after surgery, then at week 1, week 2, week 4, week 8, 3 months, and at 6 months postoperatively. The postoperative topical regimen was the same for both groups which is preservative-free topical antibiotic prophylaxis for 4 weeks and preservative-free topical steroid eye drops six times daily, tapering off over 6–8 weeks. Subconjunctival 5-FU injections were given in the clinic at the discretion of the treating surgeon depending on postoperative IOP and bleb morphology.

Data collection

Demographic data, including age, gender, ethnicity, and any previous ocular surgeries and laser procedures were collected from the electronic medical records. Relevant clinical data, including type of glaucoma and visual field mean deviation (dB) were also collected. Clinical outcome parameters including best-corrected visual acuity (BCVA), IOP, and anti-glaucoma medications (topical and oral) were recorded preoperatively at each postoperative visit (day 1, week 1, month 1, month 3, month 6). BCVA was measured using Snellen visual acuity chart and was converted later to LogMAR visual acuity for statistical analysis. IOP was measured using Goldmann applanation tonometry during all visits. Each class of topical anti-glaucoma medication (prostaglandin inhibitors, beta blockers, carbonic anhydrase inhibitors, alpha 2 agonists) was recorded to determine total number of agents used at each visit. In addition, the use of oral carbonic anhydrase inhibitor was recorded separately at each visit.

Bleb morphology was reported in both groups at each postoperative visit, with special attention to bleb heights, area, and vascularity scores. Systematic bleb assessment at the slit lamp was recorded using Moorfield’s Bleb Grading System. ²⁵ Postoperative complications and interventions were also reviewed up to 6 months of follow-up. The findings were reported following the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) checklist for cohort studies.

Outcome measures

The primary outcome at 6 months was defined as a complete success (CS), qualified success (QS), or failure (F). CS was defined as: an IOP of 6–18 mmHg (inclusive) without anti-glaucoma medications. QS was defined as the same parameters as CS but with anti-glaucoma medications. F was defined as IOP >18 mmHg, IOP ⩽5 mmHg with any decreased vision on two consecutive visits; reoperation for glaucoma; or loss of light perception vision.

Secondary outcomes at 6 months were IOP, BCVA, anti-glaucoma medications, intraoperative and postoperative complications, postoperative interventions, revisions, and reoperations.

Statistical analysis

Statistical analysis was performed using SPSS version 27.0. Between-subjects ANOVA and independent t-test were used to examine differences in continuous variables between the bevacizumab and control group. Within-subject ANOVA was used to examine differences in continuous variables over the study period. Categorical variables were analyzed using Pearson’s χ² or Fisher’s exact test.

Results

A total of 75 eyes of 75 patients were included in the analysis. Thirty-seven eyes received bevacizumab, and 38 eyes were the control. Baseline characteristics are shown in Table 1. There were no statistically significant differences in the demographic or clinical profile of patients between treatment groups. In both groups, most patients were Caucasian, with only a few Asian and Afro-Caribbean patients. Most patients were pseudophakic in both groups. All the phakic patients in this study had standalone PRESERFLO Microshunt surgery without combining it with phacoemulsification at the time of surgery. Most patients had 0.4 mg/ml MMC at the time of surgery; 30 eyes in bevacizumab group and 31 eyes in control group. Most patients had primary open angle glaucoma with moderate-to-advanced glaucomatous field damage, as determined by the extent of visual field loss (Hodapp–Parrish–Anderson classification). ²⁶ Visual field was assessed using Humphrey Field Analyzer.

Table 1.

Baseline characteristics, % (n).

Baseline characteristics, % (n)	Bevacizumab (N = 37)	Control (N = 38)	p
Age ± SD (years)	72.0 (13.4)	74.8 (13.0)	0.18
Gender, n (%)
Female	19 (48.6)	19 (50)
Male	18 (51.4)	19 (50)	0.90
Ethnicity, n (%)
Caucasian	33 (89.2)	35 (92.1)
Asian	1 (2.7)	0
Afro-Caribbean	1 (2.7)	1
Indian	1 (2.7)	1
Middle Eastern	1 (2.7)	1
Type of glaucoma, n (%)
POAG	27 (75.0)	28 (73.7)
Secondary OAG (PXF, PDG)	6 (16.2)	7 (18.4)
PACG	1 (2.7)	1 (2.6)
NTG	1 (2.7)	2 (5.3)
JOAG	2 (5.4)	—
Visual field severity (dB)
Mean deviation ± SD	−12.8 (7.2)	−14.8 (6.8)	0.11
Lens status, n (%)
Phakic	6 (16.2)	7 (18.4)
Pseudophakic	31 (83.8)	30 (78.9)
Aphakic	—	1 (2.6)
History of glaucoma surgery, n (%)
Yes	10 (27.0)	10 (10.1)
No	27 (73.0)	28 (73.3)	0.94
History of glaucoma laser, n (%)
Yes	6 (16.2)	3 (7.9)
No	31 (83.8)	35 (92.1)	0.27

Baseline characteristics and comparisons between treatment groups using independent samples t-test or Pearson’s χ² test.

PACG, primary angle closure glaucoma; POAG, primary open angle glaucoma; secondary OAG, secondary open angle glaucoma;, NTG, normal tension glaucoma; JOAG, Juvenile open angle glaucoma; Glaucoma surgery, trabeculectomy, phacoemulsification + minimally invasive glaucoma surgery (MIGS) ± endoscopic cyclophotocoagulation (ECP); glaucoma laser, selective laser trabeculoplasty.

Clinical outcomes

Intraocular pressure

Within subjects, repeated measures ANOVA showed a statistically significant change in IOP for both the bevacizumab and control group from baseline across all time points over the 6-month follow-up (all p ⩽ 0.001; Figure 1). Between subjects, post-hoc comparisons showed there was a statistically significant difference in IOP at month 3 between the bevacizumab and control group (9.9 ± 2.6 mmHg vs 11.3 ± 3.8 mmHg; p = 0.03).

Figure 1.

Average IOP measurements (mmHg) and average medication use between bevacizumab and the control group over the 6-month study period.

IOP, intraocular pressure.

Both surgical approaches were effective in terms of IOP lowering (baseline vs 6 months postoperatively: 20.0 (6.8) mmHg vs 12.8 (3.7) mmHg in the MMC group; 23.6 (6.9) mmHg vs 11.9 (4.2) mmHg in the MMC+ bevacizumab group; p < 0.001 in both comparisons).

Anti-glaucoma medication usage

There were no statistically significant differences in anti-glaucoma medication usage between groups across all study time points (Table 2). A similar pattern of medication use was observed where an average number of drops was significantly reduced in the immediate postoperative period before slowly increasing by month 6 where, at this point, patients treated with bevacizumab were, on average, using 0.5 ± 0.9 medications, whereas patients in the control group were, on average, using 0.7 ± 1.0 medications (Figure 1).

Table 2.

Mean and standard deviation measurements for IOP, VA, and number (no.) of medications at each of the study time points.

	Bevacizumab (N = 37)	Control (N = 38)	p
Baseline
IOP	23.6 (6.9)	20.0 (6.8)
VA	0.38 (0.4)	0.26 (0.2)
No. of medications	2.8 (0.8)	3.0 (0.8)	0.15
Day 7
IOP	8.5 (3.0)	10.2 (5.7)
VA	0.51 (0.3)	0.35 (0.2)
No. medications	0.1 (0.3)	0.0 (0.2)	0.32
Month 1
IOP	10.1 (2.7)	11.3 (4.3)
VA	0.46 (0.3)	0.30 (0.2)
No. medications	0.2 (0.5)	0.3 (0.9)	0.16
Month 3
IOP	9.9 (2.6)	11.3 (3.8)
VA	0.43 (0.3)	0.31 (0.3)
No. medications	0.3 (0.6)	0.4 (0.9)	0.22
Month 6
IOP	11.9 (4.2)	12.8 (3.7)
VA	0.40 (0.6)	0.30 (0.3)
No. medications	0.5 (0.9)	0.7 (1.0)	0.20

IOP, intraocular pressure; VA, Visual acuity.

Best corrected visual acuity

As shown in Table 2, there was no statistically significant difference between the two groups in VA over the study period.

Surgical success rates

Kaplan–Meier survival curves for surgical success are shown (Figure 2). The probability of CS (Graph A) was 89% and 86% at 1 month, and 80% and 78% at 3 months of follow-up for the bevacizumab group and the control group, respectively. At 6 months, the probability of CS was 67% for the bevacizumab group and 58% for the control group. The probability of QS (Graph B) was 97% and 89% at 1 month for the bevacizumab group and the control group, respectively, which remained the same for both groups at 3 months (97% vs 89%). At 6 months, the probability of QS in the bevacizumab group was 90% versus 86% in the control group.

Figure 2.

Kaplan–Meier survival curve showing the probability of success after PRESERFLO^® Microshunt implantation using MMC alone versus MMC augmented with bevacizumab. (a) Complete success is defined as an IOP of 18 mmHg or less, without anti-glaucoma medications. (b) Qualified success is defined as the same IOP parameters as complete success but with anti-glaucoma medications.

IOP, intraocular pressure; MMC, mitomycin C.

Postoperative complications

Postoperative complications are shown in Table 3.

Table 3.

Summary of complications during follow-up.

Complication	Bevacizumab (N = 37), n (%)	Control (N = 38), n (%)
Bleb leak	—	—
Shallow anterior chamber	—	—
Choroidal detachment	1 (2.7)	1 (2.6)
	—	—
Bleb encapsulation/Encystment	1 (2.7)	3 (7.9)
Flat/scarred bleb	1 (2.7)	3(7.9)
Exposure	—	—
Hyphema	1 (2.7)	1 (2.6)
Glaucoma drainage device	1 (2.7)	2 (5.3)

Hypotony

Numerical hypotony (IOP below 6 mmHg without any clinical sequel) was quite common in both groups within the first postoperative month. Clinical hypotony leading to choroidal detachment was detected in one eye (2.6%) in the control group, and in one eye (2.7%) in the bevacizumab group. All cases of choroidal detachment in both groups were early and self-limited as they resolved without visual implications by week 4–6 postoperatively. No cases of shallow anterior chamber or hypotony maculopathy were detected in either group.

Bleb morphology

With regard to bleb morphology, the bevacizumab group generally had more diffuse and less vascular blebs. Three (7.9%) eyes showed bleb encapsulation and three (7.9%) eyes had flat blebs in the control group. In contrast, only one (2.7%) eye had a flat bleb in the bevacizumab group and no bleb encapsulation was detected.

Hyphema

Transient hyphema was detected in one eye in both groups. It resolved by 2–4 weeks postoperatively with conservative management in both cases.

Glaucoma drainage device

The need for a glaucoma drainage device, either Baerveldt 350-mm² (BVT 350-mm²) or Paul glaucoma implant (PGI), to control IOP was considered as a postoperative complication. Two (5.3%) eyes in the control group required a drainage device (one BVT 350-mm² and one PGI) within the 6-month postoperative period. In the bevacizumab group, one eye (2.7%) needed glaucoma drainage device (BVT 350-mm²) over the study period. It is worth mentioning the patient requiring a further glaucoma drainage device in the bevacizumab group was Afro-Caribbean as opposed to Caucasian in the control group. Afro-Caribbean patients are known for having a strong tendency for scarring and fibrosis in comparison to other ethnicities. ²⁷

Postoperative interventions

Postoperative interventions (Table 4).

Table 4.

Summary of postoperative interventions between treatment groups.

Postoperative intervention	Bevacizumab (N = 37), n (%)	Control (N = 38), n (%)
Bleb needling	1 (2.7)	3 (7.9)
Bleb revision for high IOP	1 (2.7)	3 (7.9)
5-fluorouracil injection	3 (8.1)	14 (36.8)

Fourteen (36.8%) eyes in the control group underwent 5-fluorouracil injections within the 6-month postoperative period. In addition, two (5.3%) eyes required bleb needling and three (7.9%) eyes underwent bleb revision. In contrast, only three (8.1%) eyes underwent 5-FU injection in the bevacizumab group. In addition, one eye (2.7%) required bleb needling and one eye (2.7%) underwent bleb revision. In total, a higher proportion of patients underwent postoperative interventions in the control group (N = 19; 50%) compared to the bevacizumab group (N = 5; 13.5%) which was statistically significant (Pearson’s χ² test; p = 0.002).

Postoperative 5-FU injection was performed in the clinic when IOP was increasing above target pressure for the level of disease, and/or when bleb morphology showed either high vascularity scores or signs of scarring (flat or encapsulated bleb). Bleb needling was performed less commonly and was only considered when localized bleb encapsulation was detected. If the previously mentioned measures failed in restoring bleb function, bleb revision in theaters was then performed. Bleb revision was successful to restore bleb function in only one eye in the control group, as it managed to relieve obstruction of the distal microshunt tip by subconjuctival Tenon’s.

Kaplan–Meier survival curves for postoperative intervention rates are shown in Figure 3. The probability of not requiring an intervention was 91% versus 81% at 1-month, and 89% versus 78% at 3-month follow-up for the bevacizumab group and the control group, respectively. At 6 months, the probability of not requiring an intervention in the bevacizumab group was 86% versus 50% in the control group. Interventions included needling, 5-FU, and bleb revision.

Figure 3.

Kaplan–Meier survival curve showing the probability of requiring an intervention after PRESERFLO^® Microshunt implantation using MMC alone versus MMC augmented with bevacizumab.

MMC, mitomycin C.

Discussion

The study demonstrates at least similar safety and efficacy of intracameral bevacizumab as an adjunct to MMC during PRESERFLO Microshunt procedure compared to MMC alone. In both study groups, IOP was statistically significantly improved from baseline across all time points over the 6-month follow-up. At 3 months, there was a statistically significant difference between the two groups in IOP, showing that the bevacizumab group had lower IOP. At 6 months, this difference was no longer statistically significant between both groups. This could be explained by postoperative interventions performed in the control group, in an attempt to modulate wound healing and restore bleb function. In other words, higher rates of postoperative interventions in the control group may have helped reduce the difference in IOP between both groups at 6 months of follow-up. Number of anti-glaucoma medications was slightly lower in the bevacizumab group at 6 months; however, this difference was not statistically significant.

Complete and qualified surgical success was defined as an IOP of 18 mmHg, without and with anti-glaucoma medications, respectively. Measurement of success according to percentage reduction of IOP may not be considered the optimal approach to evaluating the clinical success of a filtering procedure; arguably, assessment of the absolute IOP level attained may be more indicative of what constitutes success in clinical practice. At 6 months, the probability of CS was 67% for the bevacizumab group and 58% for the control group, whereas the probability of QS in the bevacizumab group was 90% versus 86% in the control group. At 6 months, the mean IOP was 11.9 (4.2) mmHg, on 0.5 ± 0.9 medications, in the bevacizumab group versus 12.8 (3.7) mmHg, on 0.7 ± 1.0 medications, in the MMC alone group; p < 0.001 in both comparisons.

The efficacy outcomes reported here are consistent with findings from published clinical studies investigating Microshunt implantation with adjunctive use of MMC (0.2–0.4 mg/ml, 2–3 min). However, the average IOP at 6 months in our sample was lower in both groups compared to other studies.^6,28

Complication rates were similar for both groups except for rates of bleb scarring and bleb encapsulation, which were higher in the control group. In comparison, the bevacizumab group generally had more diffuse and less vascular blebs.

The major difference between both groups was detected in the rates of postoperative interventions performed to restore bleb function and modulate wound healing. The control group required higher rates of post-operative interventions (bleb needling, 5-FU injections, bleb revisions) to restore bleb function compared to the bevacizumab group; and this difference was statistically significant. The decision to perform postoperative interventions was based on both the target IOP level and bleb morphology.

Fewer postoperative interventions in the bevacizumab group are likely to reflect the number of clinic visits and return to theater in the postoperative period, which subsequently reflects the total cost needed to manage patients post-operatively. Therefore, the use of intracameral bevacizumab could be more cost-effective. However, formal analysis is required to confirm this assumption.

Bevacizumab was used over other anti-VEGF agents because of lower cost and larger molecular weight, allowing binding to the sclera for a longer period, possibly resulting in a longer-lasting effect. ²⁹ Bevacizumab was injected intracamerally, as opposed to the subconjunctivally, as the direct injection into the anterior chamber could possibly lead to diffuse and more homogeneous distribution of the drug to the bleb. Moreover, VEGF levels in the aqueous humor of glaucoma patients rise after glaucoma surgery. ³⁰ Thus, intracameral administration of the drug might promote direct anti-VEGF action in the aqueous humor, resulting in lower VEGF levels reaching the bleb. Finally, bevacizumab has been well documented to be safe for corneal endothelial cells.^31,32

Many studies investigated the use of bevacizumab as an adjunctive treatment to trabeculectomy. Anti-VEGF agents including bevacizumab were used as an effective medication for rescuing failing filtering blebs that exhibit neovascularization. ³³

Previous studies reported higher rates of bleb leak when bevacizumab was used versus placebo during trabeculectomy. Some studies have reported higher rates of bleb encapsulation with bevacizumab alone versus MMC alone during trabeculectomy procedure. However, in our study, there was no bleb leak detected in any of the study groups. Moreover, bevacizumab was used as an adjuvant to MMC rather than as an alternative to it, which could explain lower rates of bleb encapsulation detected in the bevacizumab group. It is also worth mentioning that bevacizumab was administered subconjunctivally rather than intracamerally in most of these studies. ²²

Our study yielded similar results to those demonstrated by Landers et al., who investigated the use of a single-dose intravitreal bevacizumab during MMC augmented trabeculectomy procedure. They reported that bevacizumab was associated with a significant reduction in the need for additional medication or further surgery to achieve target IOP. In their study, bevacizumab was also associated with larger blebs that were less inflamed and required fewer subsequent interventions. ³⁴

This study has some limitations. First, it is a non-randomized retrospective comparative study. However, both groups were well-matched regarding ethnic origins, clinical profile, and surgical history. The relatively small sample size in both groups and the short follow-up period were also limiting factors. Further assessment of our cohort of patients at 1 year and beyond is necessary.

In both groups, most patients were Caucasian; this population is known to have reduced risk of bleb fibrosis and subsequent failure relative to other ethnicities; therefore, this study could not be generalized to non-white ethnicities.

In both groups, the MMC dose was variable (0.2–0.4 mg/ml); it was titrated according to each patient’s clinical profile. The lack of standardization of MMC dose is a limitation; however, this reflects real-world clinical practice. In this study, most patients had 0.4 mg/ml MMC at the time of surgery; 30 eyes in bevacizumab group and 31 eyes in control group. This helped minimize confounding effect of variable MMC dose between both groups.

In conclusion, the adjunctive use of bevacizumab with MMC in PRESERFLO Microshunt procedure appears to be superior to MMC alone. In other words, the adjunctive use of bevacizumab with MMC could help optimize the outcomes of PRESERFLO Microshunt procedure without compromising its safety.

This study can inform further investigations through prospective randomized trials with longer follow-up and inclusion of a larger cohort of high-risk group patients, for example, non-White ethnicities.