1. Trabeculectomy
Trabeculectomy has been the preferred surgical procedure for UG for many years
[1][2]. Although studies evaluating the success rate of trabeculectomy in UG are retrospective with a small number of patients, most of them agree that the success rate of trabeculectomy with MMC is reduced in UG in comparison to POAG
[3]. There are at least two reasons for this: Inflammatory activity is likely to be more pronounced in uveitic eyes following intraocular surgery, leading either to hypotony due to ciliary body impairment or to bleb failure due to subconjunctival scarring. Almarobac et al. reported that the cumulative probabilities of success were 60% and 35.7% at 36 and 60 months postoperation, respectively, whereas in the largest up-to-date series, Iwao et al. reported probabilities of success 1, 3 and 5 years after trabeculectomy in the UG group of 89.5%, 71.3% and 61.7%, respectively
[3][4]. In a recent study, Kanaya et al. reported that the success rates in UG and POAG were 91.7% and 88.0% at 12 months, 82.2% and 75.6% at 36 months, and 66.5% and 61.8% at 120 months, respectively
[5]. The authors attributed the increased success rate of trabeculectomy in UG (which was similar to that in POAG) to the successful control of the inflammation. Different studies evaluated risk factors for the failure of trabeculectomy in cases with UG. Iwao et al. considered cataract surgery and granulomatous uveitis as a risk factor for failure
[3]. Almobarac et al. reported that in UG eyes that underwent phacoemulsification following MMC-enhanced trabeculectomy, the bleb survived but the eyes required more medication to control the IOP after the procedure
[6]. In contrast to Iwao et al.
[3], Kanaya et al. reported that granulomatous uveitis was significantly associated with favorable prognosis
[5]. There is controversy regarding whether preoperative inflammation affects the results of surgery. On the contrary, most studies agree that postoperative inflammation is a risk factor of worsening failure rate for trabeculectomy surgery
[7][8].
Kwon et al. looked specifically at the effect of the activity of inflammation on the success rate of trabeculectomies in UG and concluded that the initial activity of inflammation did not affect the success rate, but relapses of the inflammation were risk factors for failure
[9]. In contrast, recently, Magliya et al. concluded that proper perioperative uveitis control in patients attending UG surgeries results in lower IOP levels and less inflammation over 2 years postoperatively
[10]. Finally, Gregory et al. evaluated the effect of race on the course of UG and concluded that trabeculectomy has a higher risk of failure in black patients
[11].
2. Minimally Invasive Glaucoma Surgical (MIGS) Devices in Uveitic Glaucoma
Minimally invasive glaucoma surgical (MIGS) devices have been developed as a surgical option for glaucoma, to improve surgical safety, conserve conjunctiva and maintain efficacy in terms of lowering IOP. Procedures that disrupt, ablate, or bypass the trabecular meshwork (TM) constitute MIGS. These procedures include ab interno trabeculectomy using the Trabectome® (NeoMedix, Tustin, CA, USA), goniotomy with the Kahook Dual Blade® (New World Medical, Cucamonga, CA, USA) and gonioscopy-assisted transluminal trabeculotomy (GATT). These techniques are blebless and target the TM, the primary anatomic structure responsible for aqueous outflow resistance.
Goniotomy has traditionally been used to treat pediatric UG. It has been shown that it results in a significant decrease in IOP and number of IOP-lowering medications, although multiple interventions are often needed
[12][13]. Recently, Meerwijk et al. reported success rates of 100%, 93% and 80% at 1, 2 and 5 years, respectively, after performing goniotomy in children with a mean age of 7 years and non-infectious UG
[14]. There were no significant changes in visual acuity and uveitis activity or its treatment, and there were no major complications.
Trabectome (Neomedix, Tustin, CA, USA) is a MIGS device that uses electrocautery, irrigation and aspiration to selectively ablate the trabecular meshwork and the inner wall of Schlemm’s canal and allow aqueous free access to the canal and its collector channels. Anton et al. used Trabectome to treat 24 patients with UG and reported that there was no patient who achieved absolute success but 87.5% of the patients achieved qualified success after 1 year. Three (12.5%) patients needed further glaucoma surgery
[15]. According to Swamy et al., who reported the results of the operation from 45 eyes with UG from the Trabectome Study Group database, the qualified success rate at 12 months was 91%. Six (13.4%) cases required secondary glaucoma surgery and no other serious complication were noticed
[16].
The Kahook dual blade (KDB) is a disposable handpiece that employs two parallel blades to remove a strip of trabecular meshwork to improve outflow. It may be combined with phacoemulsification. Murata et al. performed ab interno trabeculotomy with KDB in 24 eyes with UG and reported that the success rate was 33% in 1 year
[17]. In contrast, Chen et al. performed KDB in 24 eyes of 22 patients and reported an 86% success rate after 2 years
[18]. TrabEx+ (MST, Redmond, WA, USA) consists of a handpiece with a dual blade that is also connected with an irrigation and aspiration system that adapts to each machine for phacoemulsification. Tanev et Kirkova reported 100% qualified success 18 months after performing TrabEx in 12 patients with UG
[19].
Gonioscopy-assisted transluminal trabeculotomy (GATT) is a minimally invasive ab interno procedure that has evolved from traditional trabeculotomy techniques and is performed with a prolene suture or with the guidance of an illuminating micro-catheter device. The surgical procedure involves cutting through the trabecular meshwork, cannulating the Schlemm’s canal 360° and unroofing the Schlemm’s canal. GATT is believed to reduce IOP by fracturing the trabecular meshwork and removing the resistance to aqueous outflow. Initially, Sachdev et al. successfully used GATT in three young patients with JRA uveitis
[20]. Very recently, Gunay et al. reported favorable results in two other patients
[21]. Parkish et al., in a small study of 16 eyes with uncontrolled UG, reported a cumulative success rate of 81% at 12 months. Transient hyphema was seen in 44% of eyes
[22]. In the largest series to date, Belkin et al. used GATT in 33 eyes of 32 patients with UG who underwent GATT with or without concomitant cataract extraction. Surgical success was achieved in 71.8% of cases in 1 year. No sight-threatening complications occurred during surgery or follow-up
[23]. Sotani et al. performed microhook trabeculotomy with a straight Tanito microhook (M-2215 s, Inami & Co., Ltd., Tokyo, Japan) in 36 eyes of 30 patients and reported that after 1 year, surgical success was achieved in 67% of eyes
[24]. Other MIGS used are the i-stent and Hydrous, which might have a role as primary or secondary conjunctival sparing procedures in UG
[25].
Bleb-Forming Devices
Similar to traditional filtering surgery, another MIGS approach to reducing IOP is to shunt aqueous from the anterior chamber to the subconjunctival space. The Xen® gel stent (Allergan INC, Dublin, Ireland) and PreserFlo® (Santen, Osaka, Japan) microShunt utilize this approach. Because this approach results in the formation of a filtering bleb, there is debate as to whether they may truly be classified in the MIGS category.
The Xen implant stent is a hydrophilic tube that is 6 mm long with a lumen of 45 μm, and it is composed of porcine gelatine crosslinked with glutaraldehyde to prevent degradation when implanted
[26]. In 2018, Sng et al. published the first results with Xen-45 in UG. They implanted Xen-45 in 24 consecutive UG patients, in the majority of whom conventional glaucoma surgery was considered inevitable. The 12-month cumulative Kaplan–Meier survival probability was 79.2%
[27].
Qureshi et al. performed Xen-45 implantation in urgent basis in 37 eyes with uncontrolled glaucoma. At the end of the follow-up period (12 to 23 months; mean: 16.7 months) five eyes (13.5%) failed, needing further glaucoma surgery. The cumulative probability of absolute success was 89.2% 1 year after surgery
[28].
Recently, Evers et al. reported results for Xen-45 implantation in 25 eyes with uncontrolled UG. Six eyes (24%) underwent surgical revision and were considered failures. At the final follow-up (mean: 17.7 months), 72% of eyes achieved complete success and 4% of eyes qualified success. Notably, the Xen implant did not prevent IOP spikes during uveitis activity
[29].
Serar et al. reported the successful use of Xen-63 with a larger lumen in the case of a refractory neovascular glaucoma due to Fuchs heterochromic iridocyclitis and retinal vein occlusion after the failure of an Ahmed tube. After one-year, intraocular pressure was 16 mmHg without any medication and the bleb was well-formed
[30].
The PreserFlo
® microShunt is an 8.5-mm-long glaucoma filtration surgical device with a 350 μm outer diameter and a 70 μm lumen that is implanted through an ab externo technique. The device’s proximal tip rests in the anterior chamber while the distal tip sits under the conjunctiva and Tenon’s capsule, about 6 mm beyond the limbus, enabling aqueous humor to pass through the lumen to produce a posterior bleb after implantation
[31]. Triolo et al. reported 36-month results of PreserFlo implants in a consecutive series of 21 patients with UG. The mean rates of success were 68%, 47% and 47% at 12, 24 and 36 months postoperation, respectively
[32].
3. No Penetrating Glaucoma Procedures in Uveitic Glaucoma
In uveitic patients, non-penetrating surgery offers the advantage of minimal post-operative anterior chamber inflammation and a reduced risk of delayed complications such as hypotony and bleb-related infections, which are more common with trabeculectomy. The absence of an iridectomy and anterior chamber penetration is supposed to reduce the inflammatory response while the presence of a trabecular meshwork may act as a barrier to infectious organisms entering the eye. Satisfactory long-term results have been reported for non-penetrating glaucoma procedures (deep sclerectomy (DS) and viscocanalostomy) in the management of UG. Obeidan et al. performed DS in 33 consecutive eyes of 21 patients and after a mean follow-up of 33.2 months reported that complete success was obtained in 72.7% of eyes, whereas qualified success was obtained in 21.2% of eyes, yielding an overall success rate of 93.9%
[33].
Mercieca et al. reported that after performing DS with 0.2–04 mgr/l MMC in 43 eyes of 43 patients, the probabilities of IOP < 22 mmHg and <19 mmHg were 69% and 62% at 3 years and 60% and 51% at 5 years, respectively. Most eyes (60%) had a Nd:Yag laser goniopuncture (LGP) by the fifth year. Recurrence of uveitis was observed in 16 eyes. Seven eyes (16.3%) had subsequent glaucoma procedures
[34].
The limitation of deep sclerectomy is that it is technically difficult to perform manually, which has limited its popularity. CO
2 laser-assisted sclerectomy surgery (CLASS) is an improved version of DS that uses a CO
2 laser, which is precise and easily strips the deep sclera, unroofs the Schlemm’s canal (SC) and leaves the trabecular meshwork thin enough for aqueous humor percolation. Xiao et al. performed CLASS in 22 eyes with UG and in 25 eyes with POAG and compared the results. After 1 year, the qualified surgical success was comparable between the UG (86.9%) and POAG (96.0%) groups, and the complete success rates were 60.9% and 64.0% in the UG group and POAG group, respectively
[35].
Recently, Salloukh et al. presented long-term results after performing vicocanalostomy in 16 patients with UG. Complete and qualified success was seen in 75% and 94% of patients at year 1, 50% and 86% of patients at year 3 and 19% and 75% of patients at year 5
[36].
4. Tube Shunt Surgery in Uveitic Glaucoma
Tube shunt (aqueous shunt) surgery has traditionally been reserved for refractory glaucoma. Therefore, shunts are commonly performed as primary surgery in UG.
Ahmed Valve and UG: The Ahmed Glaucoma Valve® (AGV, New World Medical Inc., Rancho Cucamonga, CA, USA) is a glaucoma drainage device commonly used for the treatment of glaucoma. It can be used as a primary surgical procedure or after failure of a previous filtration procedure. As the AGV has an internal valve mechanism consisting of thin silicone elastomer membranes, it does not require additional restrictive mechanisms to limit aqueous humour flow from the anterior chamber to the subconjunctival space. The above valve mechanism prevents early hypotony, which is considered advantageous, especially in UG. The body plate is usually placed 8–10 mm from the limbus while the tube is inserted 2–3 mm into the anterior chamber (AC), sulcus or even the vitreous cavity, depending on AGV type. More recently, a new type of AGV was introduced (Ahmed ClearPath GDD) which lacks the internal valve mechanism. It is available in 250 mm2 and 350 mm2 sizes and a pre-threaded 4–0 rip cord is provided by the manufacturer to prevent hypotony during the early postoperative period.
Data for more than 20 years can give us a general idea of what to expect when using these shunts as far as outcomes and complications are concerned. Earlier studies indicated that a significant IOP reduction (at least 25% from preoperative values) was achieved in more than 70% and 50% of patients at 1 and 4 years postoperation
[37][38][39]. A significant reduction of glaucoma medications was also detected in all cases, but up to 17% of eyes experienced complications during the follow-up period
[37]. The most important complications were tube occlusion, valve exposure and corneal decompensation
[37][39]. It has been proposed that sulcus placement of the tube is associated with a moderate decrease in endothelial cell count and is strongly recommended in eyes at high risk of corneal failure. Macular edema as well as ocular hypotony is still a concern even with the use of AGV. Ramdas et al. reported that 13.2% and 15.8% of UG patients developed macular edema and hypotony, respectively, after AGV and Baerveldt-350 implantation. These percentages were higher compared to non-uveitic patients, but the difference was not statistically significant. IOP reduction was comparable to that of non-uveitic glaucoma patients (44.9% vs. 42.8% decrease)
[40]. When AGV performance was evaluated as the mean IOP decrease postoperatively, this was ranged from 11 mmHg to 25.2 mmHg
[41][42][43][44][45]. A mean decrease in the number of antiglaucoma medications was also achieved (1.88)
[41][42][43][44][45]. Combining AGV with fluocinolone implant resulted in even less need for glaucoma medications
[46]. The success rate was relatively higher in eyes with pars planitis and lower in eyes with ankylosing spondylitis, suggesting that there might be differences in valve performance depending on the uveitis cause
[42]. Another study indicated that aqueous suppression early after surgery, when IOP is 10–15 mmHg, was associated with lower IOP later
[47].
Baerveldt Valve and UG: The Baervedt Implant
® (BGI-250/350, Johnson & Johnson Vision, Irvine, CA, USA) has been used for more than three decades in glaucoma practice worldwide. It consists of a non-valved silicone tube attached to a silicon plate of 250 mm
2 or 350 mm
2 total surface. The implant is placed under two recti muscles (usually superior and lateral) and the absence of any internal valve mechanism requires additional surgical steps to restrict aqueous flow during the early postoperative period. BGI was extensively used and evaluated in a tube vs. trabeculectomy study (TVT) and primary TVT study where surgery was performed in either glaucoma patients with previous glaucoma and/or cataract surgery (TVT) or in patients with no prior incisional surgery (PTVT). BGI surgery and trabeculectomy with MMC produced similar IOPs at 5 years postoperatively in both studies while tube shunt surgery had a higher success rate (TVT), suggesting BGI’s good performance in a wide range of glaucoma patients
[48][49]. Tan et al. reported results after using BGI in 47 eyes with UG. With an upper limit of 18 mmHg, the qualified success was 87% and 74% at 1 and 5 years, respectively
[50]. The presence of a tube did not prevent IOP spikes during inflammation. Tan et al. reported a high rate of corneal decompensation (9%) and hypotony maculopathy (11%) as well. Chambra et al. reported a 76% qualified success rate at 5 years
[51]. Casana compared the results of BGI implantation in 24 eyes with UG and 38 eyes with other forms of glaucoma. The median follow-up period was 592 days for UG and 764 days for other forms of glaucoma. At the end of follow-up time, 52.5% of UG and 32.5% of other glaucoma cases showed qualified success
[52]. Manako et al. have recently reported significant IOP reductions from around 30 mmHg to 15 mmHg 1 year postoperation following BGI surgery in UG patients. This corresponded to a 1-year success rate of 88%
[53]. The authors reported that the use of immunosuppressive treatment (that indicated a strong inflammatory response) was a risk factor for failure.
It is difficult to compare efficacy between different types of aqueous shunts in UG. Data from the Ahmed–Baerveldt comparison study group showed a higher vision-threatening complication rate in the BGI group at a 5-year follow up, but authors included all types of refractory glaucomas
[54]. Several studies compared AGV and BGI in the management of UG. Shisha et al. compared results after performing AGV and BGI implantation in 122 eyes and concluded that after a mean follow-up of 29.6 ± 3.6 months in the AGV group and 33.1 ± 3.8 months in the BGI group, the BGI group had a greater IOP reduction (60.3% vs. 44.5%) and complete success rate (30% vs. 9%) with a higher complication rate (51.4% vs. 20.9%). The glaucoma reoperation rate was significantly higher in the AGV group (19% in the AGV group and 4% in the BGI group). Hypotony resulted in failure in 7 eyes (10%) in the BGI group and none in the AGV group
[55]. The same group reported a greater incidence of corneal complications in BGI compared to AGV. Previous trabeculectomy was considered a risk factor for corneal decompensation
[56].
Molteno Valve and UG: The Molteno
® Implant (IOP, Inc., Costa Mesa, CA, USA/Molteno Ophthalmic Limited, Dunedin, New Zealand) is a non-valved device which consists of a silicon tube attached to a single or a double plate. The double-plate model provides more surface for aqueous humour drainage but implantation is considered surgically demanding. Molteno implants have also been used for UG patients even if this is not the most frequently inserted valve. Vuori et al. reported an 85% success rate after 4 years
[57].
Recently, Garagani et al. reported the results of tube implantation (mostly Molteno) in 50 eyes of 36 children with UG. Success rates were 98% at 1 year, 87% at 5 years, and 59% at 15 years; postoperative complications occurred in 36% of patients and included hypotony (22%), tube exposure (6%), tube obstruction (4%), corneal decompensation (2%) and cystoid macular edema (2%)
[58].
The question of whether UG patients lose visual acuity and/or the visual field deteriorates postoperatively needs to be addressed. Vision loss in UG patients is multifactorial and can be associated not only with IOP control but also with the level of inflammation, cataract/macula status and ciliary body function. Tan et al. reported that approximately 1/3 of BGI patients suffered significant vision loss (mean follow-up: 63.6 months)
[50]. Earlier reports estimated the rate to be no more than 26% after AGV implantation but the follow-up period was shorter
[37][39]. A tendency of visual field loss over the first 2 years postoperation with further stabilization has been observed in BGI patients, but contemporary literature provides insufficient data regarding visual field deterioration.
5. Comparison of Tubes to Trabeculectomy
The choice of a surgical procedure for UG is not an easy task. The benefits of tube surgery over trabeculectomy remains a matter of debate. Various studies have compared the two procedures. The most reported complications have been hypotony, corneal edema and hyphema for tube implantation and aqueous leakage, macular edema and cataract progression for trabeculectomy
[59]. Initially, Bettis et al., in a retrospective study of 41 eyes, reported that AGV had a higher success rate compared to trabeculectomy (100% vs. 66.7% after 1 year). Most trabeculectomies failed because of relapse of the inflammation
[41]. Similar results were reported by Iverson et al., who compared BGI to trabeculectomy
[60]. Later studies failed to detect a difference in the success rate
[9][45]. Chow et al.
[45] reported significantly worse IOP control and a higher number of antiglaucoma medications in the AGV group, compared with the trabeculectomy and BGI group. Lee et al. found that trabeculectomy had a significant benefit over AGV implantation; namely, its lower postoperative IOP values, as achieved with significantly fewer antiglaucoma medications
[59].
Recently, El-Saied HMA et al. prospectively compared three surgical modalities for treatment of UG in a total of 105 patients: trabeculectomy AGV implantation and trans-scleral diode laser cyclophotocoagulation. They concluded that the three modalities had the same efficacy in reducing IOP and no significant difference in complications. After 2 years, complete success was achieved in 60% via trabeculectomy, 68.6% via AGV and 62.9% via TD-CPC
[61].
Nevertheless, according to most experts, certain situations most clearly call for a tube as the first surgical intervention. These include patients with active inflammation at the time of surgery as well as those with other known risk factors for trabeculectomy failure: young age, black race, aphakia or pseudophakia and prior failed glaucoma surgery.
No matter which surgical approach is elected, up to 1/3 of UG will need a second or even a third operation. The activity of postoperative inflammation may be a critical factor for the longevity of the procedure
[62].