A 2-Year, Single-Center Study to Assess the Safety and Effectiveness of the MicroShunt in Primary Open-Angle Glaucoma

The primary goal of glaucoma management is long-term reduction in intraocular pressure (IOP) [1, 2]. When medical treatment fails to achieve adequate IOP reduction, laser or incisional surgeries are indicated [3]. Trabeculectomy and tube-shunt surgery remain the most commonly performed glaucoma surgeries for the treatment of moderate-to-severe and refractory glaucoma [4]; however, despite being effective at lowering IOP, these incisional surgical techniques are often associated with the need for substantial postoperative management [4, 5]. In recent years, several minimally invasive glaucoma surgery (MIGS) procedures have been introduced with the aim of providing a less invasive means of reducing IOP than traditional surgery, with a reduced requirement for postoperative management [6]. To date, the available MIGS procedures offer modest effectiveness compared with traditional glaucoma surgery with the benefit of fewer adverse events (AEs) and complications [6].

The MicroShunt (Santen Inc., Miami, FL, USA), formerly known as the MIDI-Arrow (when this study began in 2011) or the InnFocus MicroShunt® (Fig. 1), is an 8.5-mm-long (350-μm outer diameter and 70-μm lumen) controlled ab externo glaucoma filtration surgery device made from a biocompatible, bioinert material called poly(styrene-block-isobutylene-block-styrene) or SIBS [7]. In Europe, the MicroShunt received Conformité Européenne (CE) mark approval in 2012 for the reduction of IOP in eyes of patients with primary open-angle glaucoma (POAG) in which IOP remains uncontrollable while on maximum tolerated medical therapy and/or in which glaucoma progression warrants surgery [8]. The MicroShunt is implanted in glaucomatous eyes via an ab externo approach [9] and has been designed to achieve the desired pressure range by draining aqueous humor from the anterior chamber to a bleb formed under the conjunctiva and Tenon’s capsule [7].

Data from a previously published study of MicroShunt implantation as a standalone procedure, or in combination with cataract surgery, in 23 eyes with POAG showed a decrease in mean medicated IOP values ± standard deviation (SD) from 23.8 ± 5.3 mm Hg (N = 23) at baseline to 10.7 ± 2.8 (n = 23), 11.9 ± 3.7 (n = 22), and 10.7 ± 3.5 mm Hg (n = 22) at years 1, 2, and 3, respectively (corresponding to a mean percentage reduction in IOP of 55%, 50%, and 55%, respectively) [9]. Furthermore, the mean number of glaucoma medications per patient was reduced from 2.4 ± 0.9 (N = 23) at baseline to 0.3 ± 0.8 (n = 23), 0.4 ± 1.0 (n = 22), and 0.7 ± 1.1 (n = 22) at years 1, 2, and 3, respectively [9]. A total of 21 device-related intra- and postoperative AEs were reported in 7 patients, including tube in contact with iris (3/23 [13%]), transient hypotony (3/23 [13%]), shallow or flat anterior chamber (3/23 [13%]; all of which resolved without intervention), and choroidal effusion (2/23 [9%]) [9]. There were no cases of persistent corneal edema, late tube-corneal touch, bleb leaks, bleb infection, chronic hypotony, endophthalmitis, migration, or erosion [9]. This 2-year adaptive feasibility study sought to further investigate the long-term safety and effectiveness of the MicroShunt in patients with POAG.

This was a single-center, single-surgeon, nonrandomized, open-label, adaptive feasibility clinical study (www.clinicaltrials.gov identifier: NCT01563237) conducted in France from June 2011 to November 2016. Patients were enrolled over a 2-year period, and each patient was followed until at least study completion; follow-up visits were at day 1, day 7, week 3, week 6, month 3, month 6, month 9, year 1, and year 2. The study was conducted in accordance with the Declaration of Helsinki, the requirements for medical device investigations as presented in EN/ISO 14155 (2011), and applicable local regulatory requirements. The study protocol was reviewed and approved by the Committee for the Protection of Persons of the Groupe Hospitalier Pellegrin, Bordeaux, France, approval number 2011-A00076-35. Patients provided written informed consent prior to their enrollment in the study.

Eligible patients were 18–85 years of age and had POAG inadequately controlled on maximum tolerated medical therapy with IOP ≥18 and ≤40 mm Hg. Key exclusion criteria included no light perception; unwillingness to discontinue contact lens use after surgery; <0.1 (20/200) visual acuity (VA) in the nonstudy eye; infectious conjunctivitis, a narrow-angle glaucoma, endophthalmitis, infection, severe dry eye, or severe myopia; prior cataract surgery involving a conjunctival incision; need for glaucoma surgery combined with other ocular procedures (except for cataract surgery) or anticipated need for additional ocular surgery during the investigational period; and previous ophthalmic surgery, excluding cataract surgery, endocyclophotocoagulation, cyclophotocoagulation, or corneal refractive surgery.

For patients who required cataract surgery, phacoemulsification was performed prior to MicroShunt implantation. The MicroShunt was provided by InnFocus Inc. (Miami, FL, USA) in a sterile packaged kit containing a 3-mm scleral marker, a 1-mm triangular-bladed knife, three LASIK Shields (EYETEC^TM, Antwerp, Belgium), a marker pen, and a 25-gauge needle. The procedure for MicroShunt implantation is described in the following text. Topical anesthesia (eye drops followed by application of gel) was applied to the cornea and conjunctiva; in the case of known allergies to local anesthesia, general anesthesia was used. A peritomy was made between two rectus muscles of the surgeon’s choice, and a posterior sub-Tenon’s delamination was performed that extended 90–120° in width and approximately 8 mm retrograde to the equator of the eye. The subconjunctival space was treated with topical mitomycin C (MMC; 0.2 or 0.4 mg/mL) via three LASIK Shields for 2–3 min. MMC concentration and exposure times were decided on a case-by-case basis where younger patients and patients with higher inflammation, severity of glaucoma, thickness of Tenon’s capsule, and number of glaucoma medications were considered suitable to receive a higher concentration of MMC for a longer period of exposure. As the study advanced, learnings about the most effective concentration of MMC and the positioning of the MMC-soaked sponges also guided the intervention.

Following treatment with MMC, the location for the scleral pocket was marked 3 mm from the limbus with the pre-inked 3-mm scleral marker. A 1-mm-wide, 1–2-mm-long, shallow scleral pocket was made 3 mm posterior to and toward the limbus with a triangular-bladed knife. A 25-gauge needle was then passed through the scleral pocket into the anterior chamber, remaining parallel to the iris plane to decrease the risk of corneal endothelial cell loss, and then retracted, thereby creating a dissected tunnel. The MicroShunt was threaded through the needle tunnel with forceps, and the 1.1-mm wingspan planar fins of the device were secured into the 1-mm scleral pocket. The flow of aqueous humor through the lumen of the MicroShunt was visually confirmed by the formation of a slowly growing droplet on the distal tip of the device. The distal end of the device was then tucked beneath the conjunctiva and Tenon’s capsule, followed by separate closures of the conjunctiva and Tenon’s layer with 10–0 Nylon sutures. The site was checked for bleb leaks.

IOP was measured using Goldmann applanation tonometry before pupillary dilation and at the same time of day as the qualifying assessment to minimize the effect of diurnal fluctuation of IOP, where possible. Safety of the MicroShunt was evaluated by slit lamp biomicroscopy for anterior chamber examination and a 90D lens for peripheral fundus examination. Patients were assessed preoperatively and postoperatively and at defined follow-up intervals for choroidal effusion, hypotony, leakage, inflammation, hyphema, infection, and migration of the shunt, as well as several other potential complications. VA (measured using a Monoyer chart and converted to logMAR) was also evaluated at follow-up intervals.

The prespecified primary effectiveness endpoints were the reduction in IOP relative to the preoperative value assessed at each postoperative visit and the measurement of success at years 1 and 2. The prespecified definition of success for patients with baseline IOP >18 to <21 mm Hg was a ≥20% decrease in IOP from baseline with no reoperation for glaucoma or loss of light perception vision; for patients with baseline IOP >21 mm Hg, success was defined as achieving an IOP of <21 mm Hg and ≥20% decrease in IOP from baseline with no reoperation for glaucoma or loss of light perception vision. The definition of success was modified post hoc to standardize the ≥20% decrease in IOP for all patients irrespective of baseline IOP. Further, a lower limit was included as per the World Glaucoma Association guidelines for glaucoma clinical trials [10]. As such, overall success was defined as patients who were not pressure or surgical failures, with or without glaucoma medication. A target pressure zone was defined as IOP ≥6 mm Hg and <21, <18, or <14 mm Hg. Pressure failure was defined as patients with an IOP outside of the target pressure zone or without a decrease in IOP ≥20% from baseline on two consecutive scheduled follow-up visits after 3 months and did not achieve a decrease in IOP ≥20% from baseline at the last visit in which the success rate was reported. Surgical failure was defined as patients requiring reoperation in the operating room; bleb needlings were not considered a reoperation. Complete success and qualified success were defined as above but without glaucoma medications and with glaucoma medications, respectively.

The prespecified secondary effectiveness endpoint was the mean number of glaucoma medications per patient. The prespecified primary safety endpoint was the incidence of all procedure- and/or device-related AEs during the study, and the prespecified secondary safety endpoints were change in VA from baseline and the incidence of glaucoma reoperations.

The analysis population for effectiveness and safety endpoints comprised the per-protocol (PP) population, which was defined as patients who had at least one IOP measurement collected at month 6 or later and had no major deviations from the protocol. Safety data for the intention-to-treat (ITT) population, defined as patients who enrolled in the study and received MicroShunt implantation, are also reported. Quantitative endpoints were reported in terms of mean and SD, or median and interquartile range; qualitative endpoints were reported in terms of number and percentage of each modality. No formal sample size calculations were conducted. Statistical analyses were performed using SAS System® version 9.1 or higher, and all statistical analyses were conducted on locked databases following a careful review of the data to identify any protocol deviations and their potential effect on endpoint analysis. Descriptive summaries were based on observed cases, with the exception of the calculation of success rates, where missing IOP scores at years 1 and 2 were imputed using the last observed IOP score. For patients who were using acetazolamide 1 day prior to or on the surgery date, baseline IOP was recorded as the IOP value prior to acetazolamide use. Data collected after reoperation were excluded from efficacy and safety analyses [5]. For all descriptive summaries, any records after reoperation were set to missing.

This was an adaptive feasibility study that aimed to establish the surgical technique and further understand the optimal placement of MMC for this novel device. In the first group of patients implanted with the MicroShunt, MMC placement was posterior and localized under the conjunctiva and Tenon’s capsule, away from the limbus. As the study advanced and the understanding of optimal placement increased [9], the technique was adapted so that the remaining patients received wide placement of MMC, including close to the limbus. Subanalyses were conducted retrospectively to further explore the effect of MMC placement on surgical outcomes. Findings were not validated as a function of MMC concentration.

In addition, statistical methods for generating p values, AE summaries, pressure failure rates and time to first failure, box-whisker plots, and scatter plots, as well as censoring of patients who underwent bilateral MicroShunt surgery from all analyses, were performed post hoc. The p values for differences between baseline and postbaseline timepoints were based on a paired t test and were not adjusted for multiplicity; for between-group differences, p values were based on a two-sample t test and were not adjusted for baseline IOP/number of glaucoma medications or multiplicity. Summaries of AEs were based on AEs that started on or before year 2 (or day 730 if year 2 data were missing). All AEs related to the procedure and device are presented together to control for double counting. The definition for the PP population was amended post hoc to incorporate the requirement for patients to have at least one IOP score at month 6 or later.

A total of 69 patients were enrolled in the study, of whom 67 underwent MicroShunt implantation (ITT population). Of these patients, 61 underwent unilateral MicroShunt implantation, met all eligibility criteria, and had at least one IOP score at month 6 or later; these patients were included in the PP population (Fig. 2). Fifteen patients from the PP population underwent MicroShunt implantation in combination with cataract surgery; the remaining patients received the MicroShunt as a standalone procedure. A total of 6 patients from the overall PP population discontinued the study before year 2; 1 patient died during study participation for reasons not related to the procedure or device (as assessed by the study investigator), and 5 patients were lost to follow-up. Patient demographics and baseline characteristics for the PP and ITT populations are shown in Table 1 and online supplementary Table 1 (for all online suppl. material, see www.karger.com/doi/10.1159/000526960), respectively.

In the overall population, mean IOP ± SD was reduced from 25.7 ± 6.1 mm Hg at baseline (N = 61) to 15.8 ± 4.7 mm Hg at year 1 (n = 58) and 16.5 ± 6.0 mm Hg at year 2 (n = 52), representing a mean decrease from baseline of 34.7 ± 24.0% and 34.1 ± 25.3% at years 1 and 2, respectively (Fig. 3). IOP was significantly reduced from baseline at all postbaseline timepoints to year 2 (p < 0.001). In the cohort of patients who underwent MicroShunt implantation in combination with cataract surgery, mean IOP was reduced from 24.7 ± 6.3 mm Hg at baseline (n = 15) to 14.3 ± 2.6 mm Hg at year 1 (n = 15) and 14.7 ± 3.7 mm Hg at year 2 (n = 13).

Overall success rates for patients achieving IOP ≥6 mm Hg and <21 mm Hg were 80.3% at year 1 and 75.4% at year 2. Overall, complete, and qualified success rates stratified by IOP target range at year 1 and year 2 are shown in Figure 4 and online supplementary Table 2. In the overall population, pressure failure rates were 18.0% and 19.7%, and surgical failure rates were 3.3% and 6.6% at years 1 and 2, respectively. In patients undergoing MicroShunt implantation in combination with cataract surgery, overall success rates for those achieving IOP ≥6 mm Hg and <21 mm Hg were 86.7% at year 1 and 93.3% at year 2.

In the overall population, the mean ± SD number of glaucoma medications per patient was reduced from 2.9 ± 1.1 at baseline (N = 61) to 0.6 ± 1.0 at year 1 (n = 58) and 1.0 ± 1.3 at year 2 (n = 52) (Fig. 5a). The percentage of medication-free patients at year 1 and year 2 was 67.2% and 53.8%, respectively (Fig. 5b). The number of glaucoma medications per patient decreased significantly from baseline at all postbaseline timepoints to year 2 (p < 0.001). In the cohort of patients who underwent MicroShunt implantation in combination with cataract surgery, the mean number of glaucoma medications per patient was reduced from 2.9 ± 1.2 at baseline (n = 15) to 0.4 ± 0.8 at year 1 (n = 15) and 0.6 ± 0.9 at year 2 (n = 13), with 80.0% and 61.5% of patients medication free at years 1 and 2, respectively.

In the ITT population, nonserious AEs and serious AEs were reported in 48 (71.6%) and 3 (4.5%) patients, respectively, through year 2 (online suppl. Table 3). In the overall PP population, nonserious AEs were reported in 42 (68.9%) patients through year 2. The most commonly reported procedure- and/or device-related AEs in the study eye were investigator-reported increased IOP (31.1%), hyphema (24.6%), transient hypotony (IOP <6 mm Hg at any time; 16.4%), and keratitis (defined as small defects in the corneal epithelium; 11.5%) (Table 2). There were three reports (4.9%) of serious procedure- and/or device-related AEs in the study eyes (bleb leak [1.6%; resolved by flap resuture], implant migration [1.6%; resolved by reimplantation of the MicroShunt], investigator-reported increased IOP [1.6%; resolved by trabeculectomy with MMC]). All serious procedure- and/or device-related AEs were resolved in ≤14 days. No cases of persistent corneal edema, infections, or erosions were reported up to 2 years post-surgery. In the cohort of patients who underwent MicroShunt implantation in combination with cataract surgery, nonserious AEs were reported in 12 (80.0%) patients through year 2; no serious procedure- and/or device-related AEs were reported in this subgroup.

In the overall PP population, mean ± SD VA logMAR scores changed from 0.28 ± 0.57 at baseline (N = 61) to 0.22 ± 0.55 at year 1 (n = 60) and 0.25 ± 0.59 at year 2 (n = 55). A worsening of VA from baseline was reported in 2 (3.3%) patients by year 2.

Four patients (6.6%) required reoperation during the study, three of whom had serious procedure- and/or device-related AEs as described above. One patient required reimplantation of the MicroShunt because of disinsertion during the first procedure, 1 patient required endocyclophotocoagulation for glaucoma, 1 patient required flap resuture to prevent a leak, and 1 patient required trabeculectomy with MMC. No reoperations were required in the cohort of patients who underwent MicroShunt implantation in combination with cataract surgery. Bleb needlings were not considered reoperation. Thirteen (21.3%) patients had needling by year 1 and 19 (31.1%) by year 2.

In the first 36 consecutive PP patients implanted with the MicroShunt (group 1), MMC placement was posterior and localized under the conjunctiva and Tenon’s capsule, away from the limbus; in the remaining 25 PP patients (group 2), MMC placement was wide, including close to the limbus. The MMC concentration and exposure also varied between these groups, with the majority (72.2%) of group 1 receiving 0.4 mg/mL MMC for 3 min and the majority (92.0%) of group 2 receiving 0.2 mg/mL for 2 min. Patient demographics and baseline characteristics for each MMC subgroup are shown in online supplementary Table 4, and further details of MMC concentration and exposure are shown in online supplementary Table 5.

Online supplementary Figure 1 presents IOP and medication use over time by MMC subanalysis group. Mean IOP ± SD was reduced from 26.1 ± 6.5 mm Hg at baseline (N = 36) to 17.0 ± 6.4 mm Hg at year 2 (n = 30; 32.1 ± 26.5% reduction; group 1) and from 25.0 ± 5.5 mm Hg (N = 25) to 15.7 ± 5.5 mm Hg (n = 22; 36.7 ± 24.0% reduction; group 2). The mean number of glaucoma medications per patient ± SD was reduced from 3.1 ± 1.0 at baseline to 1.1 ± 1.4 at year 2 (53.3% medication-free patients at year 2; group 1) and from 2.6 ± 1.3 to 0.8 ± 1.1 (54.5% medication-free patients at year 2; group 2). Between-group differences in IOP and medications at year 2 were not significant (p = 0.445 and p = 0.375, respectively). At year 2, overall success rates for patients achieving IOP ≥6 mm Hg and <21 mm Hg were 63.9% and 92.0% in groups 1 and 2, respectively. Overall, complete, and qualified success rates stratified by IOP target range at years 1 and 2 are shown in online supplementary Table 6.

Overall, 21 (58.3%) patients in group 1 and 21 (84.0%) patients in group 2 had experienced any nonserious AE by year 2 (online suppl. Table 7). Mean ± SD VA logMAR scores changed from 0.35 ± 0.70 at baseline (N = 36) to 0.27 ± 0.57 at year 2 (n = 32; group 1) and from 0.18 ± 0.24 (N = 25) to 0.22 ± 0.62 (n = 23; group 2). Reoperation rates were 8.3% and 4.0% in groups 1 and 2, respectively. In group 1, 1 patient required reimplantation of the Micro­Shunt because of disinsertion during the first procedure, 1 patient required endocyclophotocoagulation for glaucoma, and 1 patient required flap resuture to prevent a leak. In group 2, 1 patient required trabeculectomy with MMC.

Results from this 2-year adaptive feasibility study are similar to the 3-year efficacy profile previously observed by Batlle et al. [9] (2016), with both studies demonstrating sustained reductions in IOP and medications at 2 years of follow-up. In comparison with this study, Batlle et al. [9] (2016) reported notably greater reductions in IOP; the differences recorded may be a result of the variations in MMC placement and concentration used in this study (as opposed to the administration of 0.4 mg/mL MMC close to the limbus by Batlle et al. [2016]) [9]. In accordance with the data presented in Batlle et al. [9] (2016), no cases of persistent corneal edema, infections, or erosions were reported in this study. In the present study, a subgroup analysis of patients who underwent MicroShunt implantation in combination with cataract surgery showed possible outcome improvements in this cohort with regard to overall success and glaucoma medication reduction, as well as a higher incidence of nonserious AEs, relative to the overall study population.

A number of bleb-based procedures have been performed with MMC, including trabeculectomy. Results from the Primary Tube Versus Trabeculectomy (PTVT) study reported a reduction in mean IOP from 23.9 mm Hg at baseline to 12.4 mm Hg at year 1 in the trabeculectomy group and from 23.3 mm Hg to 13.8 mm Hg in the tube group, representing a mean decrease from baseline of 46.0% and 37.5%, respectively [4]. The mean number of glaucoma medications was also reduced from 3.2 at baseline to 0.9 at year 1 in the trabeculectomy group and from 3.1 to 2.1 in the tube group [4]. Despite the greater reductions in IOP reported for trabeculectomy and tube implantation compared with the MicroShunt results reported here, MicroShunt implantation has been shown to be associated with fewer postoperative complications and interventions [4]. In the PTVT study, the most common early postoperative complications associated with trabeculectomy and tube implantation were choroidal effusion (10% and 7%, respectively), wound leak (12% and 1%, respectively), and shallow or flat anterior chamber (9% and 10%, respectively) [4]; these complications were not observed following implantation of the MicroShunt in this study. Rates of postoperative intervention were high in both the trabeculectomy (63%) and tube (60%) groups [4].

In recent years, bleb-based MIGS devices have also been clinically evaluated for the treatment of patients with mild-to-moderate glaucoma [6]. The effectiveness and safety of XEN® ab interno implantation (Allergan, Dublin, Ireland; device placed under the conjunctiva) were evaluated in a multicenter European study of 171 patients with open-angle glaucoma, of whom 115 underwent standalone XEN® implantation [11]. In this study, a reduction in mean IOP from 25.0 mm Hg (N = 115) at baseline to 15.4 mm Hg (n = 89) was reported at year 1 in the standalone group [11]. Reductions in IOP were similar to those reported following MicroShunt implantation, yet lower than the reductions observed following trabeculectomy, reinforcing the modest effectiveness typically observed with micro-incisional devices versus trabeculectomy and tube-shunt surgery. A needling rate of 45.2% was required postoperatively in the standalone group in response to an IOP increase over target, flat or fibrotic bleb, or high risk of bleb failure [11]; this is notably lower than the reported needling rates following MicroShunt implantation. Prospective, comparative studies of the MicroShunt versus traditional filtration surgeries and MIGS devices have not yet been published. Direct comparisons with data from published studies cannot be made, given the differences in study populations and statistical approaches.

Throughout this study, several key observations were noted. Following learnings from the Batlle et al. [9] (2016) study, MMC placement was adapted throughout this study in an attempt to improve outcomes for the patient. Indeed, wide placement of MMC close to the limbus, as opposed to being posterior and localized under the conjunctiva and Tenon’s capsule, away from the limbus, resulted in improved success rates, regardless of variations in MMC concentration. Similar results have been reported in a recent trabeculectomy study where wide placement of MMC (application to the scleral flap area and upper temporal and nasal quadrant), compared with application to the scleral flap only, was shown to increase long-term success without increasing the complication rates [12]. This may be explained by examining the effect of MMC placement on bleb survival, morphology, and histology. A preclinical study investigating the effect of varying the treatment area of subconjunctival MMC during glaucoma filtration surgery in New Zealand White rabbit eyes indicated that the size of the area of MMC application significantly affected surgical outcome [13]. Wide placement of MMC resulted in improved bleb survival with diffuse and elevated blebs; in contrast, small areas of treatment resulted in reduced bleb survival and thin-walled, localized blebs with scarring at 21 days [13]. Another key observation concerned healing time following surgery. Unlike trabeculectomy, where MMC penetrates deep into the tissue because of the incision in the sclera [14], the smaller scleral incision performed for MicroShunt implantation results in reduced penetration of MMC into the tissue. This, together with the less invasive approach of MicroShunt implantation compared with trabeculectomy, may explain why wound healing following MicroShunt implantation was relatively quick in this study. Notably, cystic bleb formation was not observed; this challenges the perception arising from trabeculectomy surgery that placing MMC close to the limbus is associated with cystic bleb formation [15]. A 2-year study by Wells et al. [15] (2003) comparing fornix- and limbus-based conjunctival flaps with respect to cystic bleb-related complications of trabeculectomy with high-dose MMC suggested an increased risk of serious bleb-related complications (including cystic bleb formation, late hypotony, and bleb-related ocular infection) following limbus-based versus fornix-based MMC application [15]. The fact that no cystic blebs were reported in this study, despite the application of MMC close to the limbus in some eyes, may also be correlated with the length of the MicroShunt and the posterior subconjunctival/Tenon’s positioning of the device. Another observation concerned the morphology of the bleb. The bleb generated through the MicroShunt implantation procedure extends both anteriorly and posteriorly (online suppl. Fig. 2); observations during this study suggest that blebs that only extend posteriorly may lead to an increase in IOP over time. This observation is also supported by findings from a study that retrospectively analyzed varying MMC concentration (0.2 and 0.4 mg/mL) and placement in a pooled dataset of 87 patients from NCT01563237 and NCT00772330 [16]. In this study, patients who received MMC away from the limbus obtained a more posterior filtration bleb; this placement resulted in a lesser reduction in IOP than in the group of patients who received MMC near to the limbus [16].

This study had limitations: it was a single-center, procedure-optimizing study and comprised a small study population. In addition, the approach to device implantation, including placement, dose, and exposure time of MMC, was adapted throughout the study in order to optimize surgical outcomes. Data were not validated as a function of these changing parameters, thus introducing several confounding factors into the study. The study was not powered to compare the surgical approaches used; therefore, a comparative analysis of the varied placement of MMC was conducted retrospectively. Furthermore, it should be noted that a few analyses reported here were conducted retrospectively.

In this 2-year, single-center, single-surgeon study, MicroShunt implantation in patients with POAG was associated with significant reductions in IOP and number of glaucoma medications, which were maintained below baseline levels in most patients. No long-term sight-threatening AEs were reported following MicroShunt implantation. Further data from studies conducted across multiple centers are needed to fully characterize the safety and effectiveness of the MicroShunt.

Medical writing support, including preparation of manuscript drafts for critical revision and approval by the author in accordance with GPP3, was provided by Lucy Cartwright, MChem, Helios Medical Communications, Alderley Park, Cheshire, UK, which was funded by Santen Inc., Emeryville, CA, USA. The study was sponsored by InnFocus Inc., a Santen Pharmaceutical Co. Ltd. Company, Osaka, Japan. The author would like to thank Santen for their support in the development of this manuscript, in particular Zhengyang Shi for statistical support and data analysis, and Dr. Leonard Pinchuk and Dr. Omar Sadruddin for their critical review of the data presented.

The study was conducted in accordance with the requirements for medical device investigations as presented in EN/ISO 14155 (2011) and applicable local regulatory requirements. Institutional Review Board approval was obtained for each site. All research adhered to the tenets of the Declaration of Helsinki. The study protocol was reviewed and approved by the Committee for the Protection of Persons of the Groupe Hospitalier Pellegrin, Bordeaux, France, approval number 2011-A00076-3. All participants provided written informed consent prior to their enrollment in the study.

Isabelle Riss reports grants and personal fees from Santen Inc. during the conduct of the study.

The sponsor (InnFocus Inc., a Santen Pharmaceutical Co. Ltd. Company, Osaka, Japan) participated in the design and conduct of the study, data collection, and management. This analysis was also sponsored by Santen Inc., who participated in the data analysis, interpretation of the data, preparation, review, and approval of the manuscript.

Isabelle Riss obtained funding, designed and directed the study, collected, analyzed, and interpreted the data, reviewed the manuscript, and provided final approval.

Due to the varying rights of individuals and contractual rights of parties involved, the sponsor does not make a practice of sharing datasets. Further inquiries can be directed to the corresponding author.