Hydrostatic pressure difference between the anterior chamber and the suprachoroidal space is the basis for surgical use of the unconventional outflow pathway.1 Implants enhancing the unconventional outflow pathway have already undergone specific evolution through GMS (SOLX Inc., Occulogix, Waltham, MA, USA) used via an ab externo approach to CyPassMicro-Stent (Transcend Medical) the implantation of which is performed from the anterior chamber to the suprachoroidal space.2, 3 However, it seems that the physiology of suprachoroidal space has not been well recognized yet as regards the possibility of IOP regulation and the reaction to the presence of a foreign body in the long-term follow-up.

Case reports

There were four patients aged 70.2±15.5 years with POAG. Mean IOP before GMS implantation was 22.5±3.5 mm Hg, and the patients used 2.5±0.6 (median: Me−2.5) medications. The mean IOP before GMS removal was 28.8±4.3 mm Hg (P=0.002), and the patients used 2.3±0.5 (Me−2.0) medications (P=0.361). The mean time for the implant removal was 29.8±8.3 months.

Case 1

A 90-year-old man with GMS removed from the right eye 29 months after implantation. IOP was 25.0 mm Hg (Ganfort), and progression in the visual field was observed (MD=−17.91). The central corneal thickness (CCT) increased from 480 to 568 μm and the corneal decompensation was detected despite the lack of GMS contact with the endothelium (OCT Visante). In the middle portion of GMS the presence of collagen-rich connective capsule-like reaction about 100-μm thick was found (Figure 1). The presence of cell-rich collagen-poor connective tissue was also found between the plates. There were also observed chronic inflammatory infiltration cells, polynuclear giant cells, and small blood vessels (Figure 2a).

Figure 1
figure 1

Macroscopic image of the removed microimplant (case 1). Colonization of the connective tissue around the microimplant. (a) Proximal portion; fragments of the connective tissue at the side of the sclera, (b–d) distal portion: connective tissue (at the choroidal side) blocking the channels of the microimplant (Cannon EF 100 mm f/2.8 L Macro IS USM).

Figure 2
figure 2

Microphotograph of the removed microimplant. (a) Case 1: in the middle portion the implant is surrounded by collagen-rich connective tissue capsule. A cluster of polynuclear giant cells present between the implant plates (asterisk). On the left side the giant cells combine with collagen-poor granulation tissue with numerous small blood vessels filling the remaining part of the microimplant channel. Below, within the sclera, is shown a visible focus of dispersed inflammatory infiltration (arrow) (H&E × 200). (b) Case 3: amorphous thickened protein masses between microimplant plates (H&E × 200). (c) Case 4: inflammatory infiltrations from lymphocytic cells between the GMS plates (H&E × 400). (d) Case 4: the membrane expression of LCA (CD45) in the infiltration from mononuclear cells (immunohistochemical staining with a anti-LCA antibody × 400).

Case 2

In the case of a 69-year-old woman GMS+ was removed from the left eye 21 months after surgery. Before removal, IOP was 24.0 mm Hg (Xalacom). Progression in the visual field (MD=−8.55) and corneal decompensation were observed. CCT increased from 534 to 676 μm despite the lack of GMS contact with the endothelium. In the middle portion of GMS the presence of collagen-rich connective capsule-like reaction was detected, but the capsule was thin (about 50 μm). In case no. 2 the presence of cell-rich collagen-poor connective tissue was found between the plates.

Case 3

The cause of GMS removal from a right eye of a 70-year-old man was the decrease of endothelium from 2579 to 1685 cells mm−2 despite the lack of contact with GMS, and progression in the visual field was observed (MD=−3.49). IOP was 22.0 mm Hg (Ganfort) 28 months after the surgery. In the middle and distal part the GMS was surrounded by collagen-rich connective tissue capsule. The proximal, middle and distal portion channels were partially filled with dense amorphous protein masses (Figure 2b).

Case 4

In a 52-year-old man, 41 months after primary surgery, GMS+ was removed from the right eye because of the lack of IOP control despite of the maximal antiglaucoma therapy (Duotrav, Azopt). There were observed an increase of IOP to 41.0 mm Hg and progression in the visual field (MD=−4.23). No contact of the implant with the endothelium was observed. Collagen-rich connective tissue in the implant channels and dispersed chronic inflammatory infiltrations both around and inside the implant were detected. Lymphocytic infiltration (LCA+) was accompanied with clusters of numerous polynuclear giant cells (Figure 2c and d).


GMS is one of the most modern implants improving the unconventional pathway outflow. Despite encouraging initial clinical reports, in the medium-term follow-up the IOP reduction decreased, whereas the number of applied anti-glaucoma medications increased.4 It was suggested that the formation of a fibrotic capsule around the device was responsible for the failure of the GMS surgery.4, 5, 6 We found fibroblasts both around the microimplant and within GMS channels. Apart from connective tissue, amorphous masses were also found in the proximal portion of the microimplant. It cannot be excluded that due to nutritional disorders there came to the degeneration of collagenized connective tissue.

The presence of giant cells both around and inside the microimplant may point to the role of activated monocytes and macrophages in the pathomechanism of the development of GMS obstruction. Distinct inflammatory infiltrations from mononuclear cells were also observed in the specimens. Their occurrence proves specific immune response to the antigen presented by macrophages.

Summing up, it should be concluded that microimplant obstruction caused by the connective tissue proliferation is the cause of the failure of surgery with GMS. This process can be non-inflammatory, but it can also result from chronic inflammation in the suprachoroidal space.