The uveoscleral outflow pathway was first described in animal experiments by Anders Bill in 1965.  Aqueous exits the anterior chamber (AC) across the ciliary body and iris root, driven by the negative pressure gradient from the AC to the suprachoroidal space. From there, fluid is resorbed into scleral vessels, the choriocapillaris, and through scleral pores into the episclera. Prostaglandin analogues, commonly prescribed as first-line medical therapy for glaucoma, increase aqueous outflow through this uveoscleral pathway, thereby lowering intraocular pressure.
Surgical access to the suprachoroidal space to lower intraocular pressure in patients with glaucoma is not a novel concept. Surgical creation of a cyclodialysis cleft to lower intraocular pressure was described by Heine in 1905 for intractable glaucoma; however, this procedure was largely abandoned due to frequent postoperative hypotony and a rapid increase in intraocular pressure after spontaneous closure of the cleft. Various techniques, implants, and space-maintaining substances have been described in an attempt to create a safer and more stable suprachoroidal drainage pathway, with limited success.    
With advances in device design and biocompatibility, the uveoscleral outflow pathway is again a target of active research and development. Several new microsurgical devices seek to lower intraocular pressure by exploiting the outflow facility of the suprachoroidal space and providing a controlled and sustainable fluid egress from the anterior chamber. Both ab externo and ab interno approaches to the suprachoroidal space are possible, each with unique advantages and drawbacks.
Ab Externo Approach
The suprachoroidal space can be accessed ab externo via a trans-scleral dissection. This approach has the disadvantage of requiring both conjunctival and scleral manipulation, which may make additional glaucoma filtering surgery difficult in the future due to scarring. However, an ab externo approach may allow for more extensive tissue dissection and potentially greater IOP lowering compared to an ab interno approach. Three notable ab externo devices include The Gold Micro-Shunt (GMS, SOLX Corp, Waltham, MA), the STARflo (iStar Medical, Isnes, Belgium), and the Aquashunt (OPKO Health, Miami, FL).
The GMS is made of medical-grade 24-karat gold. The concept for this device was based on the observation that a gold foreign body was well tolerated in the eye. The original GMS is 3.2 mm wide, 5.2 mm long, 44 um thick and weighs 6.2 mg. The shunt is composed of two rectangular fused leaflets with a rounded proximal end, and a distal end with fins to anchor the device in the suprachoroidal space. The proximal end of the shunt has sixty 100-um holes and one 300-um hole to allow aqueous to enter. The distal end contains a grid of 117 110-um holes on each side of the implant to allow free flow of fluid into the suprachoroidal space. Nineteen channels are present within the shunt of which 9 are open, each with a width of 24 um and a height of 50 um. The GMS+, a revised model, is 3.2 mm wide and 5.5 mm long and weighs 9.2 mg. The design has been modified such that the fins are located at the proximal end of the device in the anterior chamber. The external gold plates of the GMS+ are separated by posts rather than channels, creating a larger area of aqueous flow (Figure 1). The GMS was approved for use in Canada, as well as in some European countries. In the U.S., the GMS had a high failure rate in investigational studies, and FDA approval was never granted.
Following creation of a fornix-based conjunctival flap, a full-thickness scleral incision is made 2-3 mm posterior to the limbus, exposing the suprachoroidal space. The anterior chamber is deepened using either viscoelastic or an anterior chamber maintainer. A crescent blade is then used to tunnel anteriorly into the AC at approximately 90% scleral thickness. Blunt dissection is also directed posteriorly into the suprachoroidal space. The proximal end of the GMS is fed into the anterior chamber using the inserter which comes with the device, such that 1-1.5 mm of the device is visible in the AC. The distal portion of the shunt is then tucked underneath the posterior lip of the scleral flap. The scleral and conjunctival flaps are the then closed tightly to avoid leakage of aqueous and bleb formation.
Skaat et al compared two different models of the GMS (24-um internal channels or 48-um internal channels) with the Ahmed Glaucoma Valve in a three-armed randomized, prospective, interventional trial for patients with refractory glaucoma. All patients included in the study had either primary open angle glaucoma, pseudoexfoliation glaucoma or pigment dispersion glaucoma and had undergone at least one failed trabeculectomy. Twenty-nine patients were randomized for implantation with either an Ahmed glaucoma valve, 24-um GMS or 48-um GMS. Success was defined as IOP greater than 5 mmHg and less than 22 mmHg with reduction of IOP at least 20% below preoperative IOP. IOP was significantly lowered in all 3 groups after 5 years of follow up. In the Ahmed group, IOP was lowered from 33.5±6.7 to 17.3±2.6 mm Hg. In the 24-um GMS group, IOP was lowered from 25.7±0.7 to 17.8±2.4 mm Hg. In the 48-um GMS group, IOP was lowered from 35.6±2.2 to 19.6±5.2 mm Hg. The number of glaucoma medications required was not significantly different between groups. Cumulative success rates at 5 years were 77.8%, 77.8% and 72.7% in the Ahmed, 24-um GMS and 48-um GMS, respectively.
Hueber et al reported very different results in a retrospective review of 31 eyes which were implanted with the second-generation GMS+. The patients in this study were diverse; 48% of eyes had undergone previous glaucoma surgery, 55% had primary open angle glaucoma, 16% had pseudoexfoliation glaucoma, 13% had secondary glaucoma, 10% had pseudophakic angle closure glaucoma, and 6% had pigmentary glaucoma. Failure was defined as IOP greater than 21 or less than 5 at any visit at least 6 months after GMS+ implantation, any serious complications (retinal detachment, endophthalmitis, suprachoroidal hemorrhage, low grade inflammation or new onset rubeosis iridis) or need for additional glaucoma surgery. At the end of 4 years, 97% of eyes met criteria for failure, with 71% of eyes failing within the first year. 77% of eyes required a secondary glaucoma surgery for uncontrolled IOP, with mean time to second surgery 7.3 months (range 0.5-37.1). Two GMS+ were explanted because of newly developed rubeosis iridis, and two were explanted for persistent low-grade inflammation. At the conclusion of the study, mean IOP for all patients was 27.9±10.4 mmHg, slightly increased from the preoperative mean IOP of 26.58 ± 10.14 mmHg. No definitive rationale for the poor performance of the GMS in this study was provided by the authors.
Figus et al reported on 55 eyes of 55 patients with refractory glaucoma. All patients had uncontrolled IOP on maximum tolerated medical therapy and had undergone at least 1 previous glaucoma surgery (range 1-5). Patients underwent GMS implantation and were followed for two years. Success was defined by IOP less than 21 mmHg and >33% IOP reduction. At 2 years, 67.3% of patients were considered a qualified success, while only 5.5% were considered a complete success. The most common complication was transient hyphema (21.8%). Choroidal detachment was noted in 10.9% of eyes. Explantation was required in three eyes (5.45%): two secondary to corneal edema due to endothelial-shunt contact and one secondary to over-filtration causing exudative retinal detachment. Most failures occurred within the first year. The most common cause of failure was an inflammatory membrane forming on the anterior end of the shunt, obstructing inflow from the AC (66.6%).
The STARflo is made of a proprietary flexible, micro-porous medical grade silicone called STAR Biomaterial, which is designed to optimize tissue integration and reduce fibrosis. It is a single piece porous implant with 3 parts: a head, neck and body. The STARflo device received CE marking approval in Europe in 2012.
A conjunctival flap is created followed by a partial-thickness scleral flap. The head is implanted in the anterior chamber at the anterior aspect of the scleral flap, and a full-thickness scleral incision is made at the posterior edge of the flap to allow the body to be inserted into the suprachoroidal space. The scleral and conjunctival flaps are sutured tightly to avoid leakage of aqueous and bleb formation.
In an early clinical trial  , 4 patients had the STARflo device implanted. Three patients completed 12-month follow up and 1 was removed from the study after undergoing cyclophotocoagulation 4.5 months postoperatively for uncontrolled IOP. In those completing the study, mean IOP was 14.5 mmHg after 12 months compared to 37.0 mmHg preoperatively. IOP was reduced to less than 21 mmHg and by more than 30% from baseline in all patients completing follow up. Early postoperative complications included bleb formation, hypotony, choroidal hemorrhage and abnormal macula.
A prospective clinical trial was performed at a single center in Europe. In this study, the STARflo glaucoma implant was placed in 40 eyes in 32 patients with treatment-refractory, moderate to severe open-angle glaucoma. After 24 months, 45% of treated eyes in the study required an additional glaucoma procedure, while 55% of patients saw a decrease in the average number of topical ocular hypotensive medications from 2.7 pre-operatively to 1.2 at 24 months. Unfortunately, the study failed to demonstrate effective IOP reduction at 24 months. Additionally, many patients exhibited decreased endothelial cell counts, with corneal decompensation occurring in 5 eyes (12.5%).
The Aquashunt was originally designed by Vidus Ocular, a start up led by Dr. Bruce Shields at Yale University. After initial pre-clinical trials, Vidus was purchased by OPKO Health (Miami, FL). The Aquashunt is made of polypropylene and is 4 mm wide, 10 mm long, and 0.75 mm thick. It has a single 250-micron lumen.
After creation of a conjunctival flap, a full-thickness scleral incision is made to expose the suprachoroidal space. The shunt is advanced through the suprachoroidal space using a custom inserter, breaking attachments between the ciliary body and the scleral spur to enter the anterior chamber. The distal end of the device is tucked underneath the sclera, posterior to the scleral incision and the device is sutured to the sclera in order to secure its position. The scleral and conjunctival flaps are then closed tightly to prevent leakage of aqueous and bleb formation.
In 2009, OPKO Health funded a clinical trial of the Aquashunt in the Dominican Republic. The results of this trial were not published. Reportedly, there were mixed results after implantation in the first 15 patients with refractory open-angle glaucoma. After 12 months, 1 device was explanted, 3 patients did not achieve adequate IOP lowering, 3 patients developed hypotony, and 8 patients achieved an average IOP reduction of 31% (range 13-46%). OPKO Health decided not to pursue clinical trials in the United States, and the intellectual property license was transferred back to Yale University. Currently there are no ongoing clinical trials investigating the Aquashunt.
Ab Interno Approach
The suprachoroidal space can also be accessed through an ab interno approach using a clear corneal incision and intraoperative gonioscopy. This approach seeks to provide increased outflow via the suprachoroidal space without manipulation of the conjunctiva and sclera. The main advantages of this approach are the relatively simple, straightforward insertion of devices and avoidance of conjunctival manipulation to allow for traditional bleb-based filtering surgery, if needed later in the future. Two ab interno devices currently in development are the Cypass Micro-Stent (Transcend Medical Inc, Menlo Park, CA) and the iStent Suprachoroidal Bypass System (iStent Supra, Glaukos Corporation, Laguna Hills, CA).
The Cypass Micro-Stent is a 6.35 mm long polyimide tube with an outer diameter of 0.5 mm and an inner diameter of 300 um. It has fenestrations to facilitate outflow along the entire length of the device (Figure 2).
The Cypass Micro-Stent can be inserted through a clear corneal incision combined with cataract extraction or as a stand-alone procedure. The anterior chamber is deepened using viscoelastic, and intraoperative gonioscopy is used to visualize angle structures. The stent is positioned on a guide-wire which is passed across the anterior chamber and used to bluntly dissect the ciliary body from the scleral spur, effectively creating a miniature cyclodialysis. The device is then passed along the guide-wire into the suprachoroidal space, with the proximal collar of the device remaining in the anterior chamber. The guide-wire is withdrawn once appropriate position is verified. Anterior segment OCT or ultrasound biomicroscopy can be used to verify the position of the stent postoperatively.
Feijoo et al followed 65 eyes of patients with open angle glaucoma and IOP >21 on medical therapy which were treated with Cypass micro-shunt implantation through a clear corneal incision. Cypass implantation was not combined with cataract extraction in this study. IOP was reduced from 24.5±2.8 to 16.4 ±5.5 mmHg at 12 months (n=55). Mean medication usage was decreased from 2.2±1.1 at baseline to 1.4±1.3 at 12 months. The most common adverse events were IOP>30 beyond 1 month (11%), transient hyphema (6%) and progression of cataracts (12%). No serious complications such as retinal detachment, choroidal effusion, persistent uveitis, persistent hyphema or hypotony maculopathy were observed. Secondary glaucoma surgery was required in 17% of eyes during the 1-year follow up.
Hoeh et al prospectively studied 167 eyes of 142 patients with open angle glaucoma and cataract. Patients underwent combined phacoemulsification with posterior chamber intraocular lens (IOL) placement and Cypass micro-shunt placement. Analysis was divided into two cohorts. Cohort 1 (n=65) had baseline IOP ≥21 mmHg, and Cohort 2 (n=102) had baseline IOP <21. Mean follow up was 294±121 days. Cohort 1 demonstrated a 35% reduction in mean IOP and 49% reduction in mean medication usage. Cohort 2 demonstrated 75% reduction in mean medication use, and maintained IOP <21 in all eyes. Overall, mean IOP at 12 months was 15.9±3.1 mmHg compared to 20.2±6.0 mmHg preoperatively. No serious complications or adverse events were observed. Early postoperative IOP elevation occurred in 1.2% of eyes, and late IOP elevation in 1.8% of eyes. Two patients developed transient hyphema.
In another study by Hoh et al, patients with cataract and open-angle glaucoma were included in a prospective consecutive case series. All patients underwent combined phacoemulsification with posterior chamber IOL placement and Cypass Micro-Stent implantation. Analysis was divided into two cohorts. Cohort 1 (n=23) had baseline IOP ≥21, and Cohort 2 (n=59) had baseline IOP <21. After 24 months, Cohort 1 had a mean IOP of 15.8±3.8 mmHg, a 37±19% reduction from baseline. IOP reduction in Cohort 1 was statistically significant at 6, 12, and 24 months. For cohort 2, mean IOP at 24 months was 16.1±3.2 mmHg, a 0±28% change from baseline. Mean glaucoma medications were reduced from 2.2 at baseline to 1.0 at last follow-up in Cohort 1 and from 2.0 to 1.1 in Cohort 2 at 24 months. Mean decrease in glaucoma medications from baseline was statistically significant at 6, 12 and 24 months in both cohorts. With regard to complications, 15.4% of patients experienced transient hypotony and 8.8% of patients developed micro-stent obstruction due to iris overgrowth. Secondary incisional glaucoma surgery was required in 11% of patients during the 2-year follow up.
The COMPASS trial is a large FDA-approved randomized controlled trial in which 505 patients were been enrolled at 27 sites to receive either Cypass Micro-Stent implantation combined with cataract surgery or cataract surgery alone. The COMPASS trial found higher IOP-lowering efficacy in patients randomized to the Cypass Micro-Stent compared to cataract surgery alone. However, a post-approval study with a follow-up of 5 years found a statistically significant difference in corneal endothelial cell density in patients randomized to the Cypass Micro-Stent compared to cataract surgery alone, with one patient receiving the Cypass Micro-Stent experiencing corneal edema. The number of Cypass retention rings extending into the anterior chamber was independently and negatively associated with corneal endothelial cell density. As a result of findings from this post-approval study, the Cypass Micro-Stent was voluntarily withdrawn from the market in August 2018.
iStent Suprachoroidal Bypass System (iStent Supra)
The iStent Supra (Model G3) is the newest IOP lowering stent from Glaukos. It is made of a biocompatible polymer with a titanium sleeve. It is a 4-mm long curved stent with a lumen of 0.165 mm (Figure 3). The iStent Supra has received CE Mark approval in Europe and is undergoing clinical trials in the United States.
The iStent Supra can be inserted into the suprachoroidal space at the time of cataract surgery or as a stand-alone procedure through a 1-mm clear corneal incision. It is inserted using a similar ab interno approach to the Cypass device under gonioscopic visualization, as described above. The device is implanted into the suprachoroidal space using a disposable injector which comes pre-loaded with the device.
Junemann described a series of 42 eyes with advanced open angle glaucoma which were followed for 12 months after implantation of the iStent Supra. Mean preoperative IOP was 20.4±2.4mmHg on two medications. All patients were treated postoperatively with topical travoprost. 98% of eyes met the primary endpoint of greater than 20% reduction in IOP with reduction of 1 medication. 90% of eyes met the secondary endpoint of IOP<15mmHg and reduction of 1 medication. No complications were observed at 1-year follow up.
In a similar study, Myers et al described a series of 25 eyes with moderate to advanced open-angle glaucoma which were followed for 12 months after medication washout and implantation of the iStent Supra. All patients were treated postoperatively with travoprost unless IOP was less than 6 mmHg. Mean IOP was 19.7±1.4 mmHg on two medications preoperatively and 24.9±1.5 mmHg after medication washout. Postoperatively, mean IOP was 13.8±1.4 mmHg through 1 year of follow up on travoprost alone; 24 of 25 patients experienced >20% reduction in IOP compared to preoperative un-medicated IOP, and 21 of 25 eyes experienced >20% reduction in IOP vs. preoperative medicated IOP. Hypotony (IOP <5 mmHg) was observed in two eyes at postoperative week one, but resolved by postoperative month one. Choroidal detachment was observed in one eye, but resolved by postoperative month three. BCVA improved or was unchanged from preoperative BCVA in all subjects.
Further clinical trials are ongoing in the United States to evaluate the safety and efficacy of the iStent Supra.
Augmenting the uveoscleral outflow pathway with implantation of suprachoroidal devices represents an alternative approach for treating glaucoma without creating a filtering bleb. Initial results with the Cypass Micro-Stent were promising. However, longer-term negative effects on corneal endothelial cell density led to voluntary recall of the device.
- Bill A, Hellsing K. Production and Drainage of Aqueous Humor in the Cynomolgus Monkey (Macaca irus). Invest Ophthalmol, 1965. Vol. 4, pp. 920-6.
- Bill A. The Aqueous Humor Drainage Mechanism in the Cynomolgus Monkey (Macaca irus) with Evidence for Unconventional Routes. Invest Ophthalmol, 1965. Vol. 4, pp. 911-9.
- Emi K, Pederson JE, Toris CB. Hydrostatic pressure of the suprachoroidal space. Invest Ophthalmol Vis Sci, 1989. Vol. 30, pp. 233-8.
- Alm A, Nilsson SF. Uveoscleral outflow – A review. Exp Eye Res, 2009. Vol. 4, pp. 760-8.
- Weinreb RN, Toris CB, Gabelt BT, Lindsey JD, Kaufman PL. Effects of Prostaglandins on the Aqueous Humor Outflow Pathways. Survey of Ophthalmology, 2002. Vol. 47, pp. S53–S64.
- Heine L. Cyclodialyse, eine neue glaukomoperation. Dtsch Med Wochenschr, 1905. Vol. 31, pp. 824–6.
- Klemm M, Balazs A, Draeger J, Wiezorrek R. Experimental use of space-retaining substances with extended duration: functional and morphological results. Graefes Arch Clin Exp Ophthalmol, 1995. Vol. 233(9), pp. 592-7.
- Portney GL. Silicone elastomer implantation cyclodialysis. A negative report. Arch Ophthalmol. 1973. Vol. 89(1), pp. 10-2.
- Miller RD, Nisbet RM. Cyclodialysis with air injection in black patients. Ophthalmic Surg, 1981. Vol. 12(2), pp. 92-4.
- Alpar JJ. Sodium hyaluronate (Healon) in cyclodialysis. CLAO J, 1985. Vol. 11(3), pp. 201-4.
- Pinnas G, Boniuk M. Cyclodialysis with teflon tube implants. Am J Ophthalmol, 1969. Vol. 68(5), pp. 879-83.
- Sen SC, Ghosh A. Gold as an intraocular foreign body. Br J Ophthalmol, 1983. Vol. 67(6), pp. 398-9.
- Hueber A, Roters S, Jordan JF, Konen W. Retrospective analysis of the success and safety of Gold Micro Shunt Implantation in glaucoma. BMC Ophthalmol, 2013. Vol. 13, p. 35.
- Skaat A, Sagiv O, Kinori M, Simon GJ, Goldenfeld M, Melamed S. Gold Micro-Shunt Implants Versus Ahmed Glaucoma Valve: Long-term Outcomes of a Prospective Randomized Clinical Trial. J Glaucoma, 2016. Vol. 25(2), pp. 155-61.
- Hueber A, Roters S, Jordan JF, Konen W. Retrospective analysis of the success and safety of Gold Micro Shunt Implantation in glaucoma. BMC Ophthalmol, 2013. Vol. 13, p. 35.
- Figus M, Lazzeri S, Fogagnolo P, Iester M, Martinelli P, Nardi M. Supraciliary shunt in refractory glaucoma. Br J Ophthalmol, 2011. Vol. 95(11), pp. 1537-41.
- Pourjavan S, Collignon N, De Groot V. STARflo™ Glaucoma Implant: 12 month clinical results. Acta Ophthalmologica: Special Issue: Abstracts from the 2013 European Association for Vision and Eye Research Conference, 2013. Vol. 91(S252).
- Pourjavan S, Collignon N, De Groot V. STARflo glaucoma implant: 12 month clinical results. Acta Ophthalmol, 2013. Vol. 91(Suppl), p. 252.
- Fili S, Janoud L, Vastardis I, Wölfelschneider P, Kohlhaas M. The STARflo™ glaucoma implant: a single-centre experience at 24 months. Graefes Arch Clin Exp Ophthalmol, 2019. Vol. 257(12), pp. 2699-2706.
- Gigon A, Shaarawy T. The Suprachoroidal Route in Glaucoma Surgery. Journal of Current Glaucoma Practice, 2016. Vol. 10(1), pp. 13-20.
- Saheb H, Ianchulev T, Ahmed I. Optical coherence tomography of the suprachoroid after CyPass Micro-Stent implantation for the treatment of open-angle glaucoma. Br J Ophthalmol, 2014. Vol. 98(1), pp. 19-23.
- García-Feijoo J, Rau M, Grisanti S, Grisanti S, Höh H, Erb C, Guguchkova P, Ahmed I, Grabner G, Reitsamer H, Shaarawy T, Ianchulev T. Supraciliary Micro-stent Implantation for Open-Angle Glaucoma Failing Topical Therapy: 1-Year Results of a Multicenter Study. Am J Ophthalmol, 2015. Vol. 159(6), pp. 1075-1081.
- Hoeh H, Vold SD, Ahmed IK, Anton A, Rau M, Singh K, Chang DF, Shingleton BJ, Ianchulev T. Initial Clinical Experience With the CyPass Micro-Stent: Safety and Surgical Outcomes of a Novel Supraciliary Microstent. J Glaucoma, 2016. Vol. 25(1), pp. 106-12.
- Höh H, Grisanti S, Grisanti S, Rau M, Ianchulev S. Two-year clinical experience with the CyPass micro-stent: safety and surgical outcomes of a novel supraciliary micro-stent. Klin Monbl Augenheilkd, 2014. Vol. 231(4), pp. 377-81.
- Junemann A. Twelve-month Outcomes Following Ab Interno Implantation of Suprachoroidal Stent and Postoperative Administration of Travoprost to Treat Open Angle Glaucoma. 31st Congress of the European Society of Cataract and Refractive Surgeons, Oct 2013. Amsterdam, Netherlands.
- Myers J, Katz J. Open Angle Glaucoma Treated with a Suprachoroidal Stent and Topical Travoprost. 23rd Annual American Glaucoma Society Meeting. March 2013, San Francisco, CA.