Survey on Glaucoma Drainage Device Usage: A United Kingdom Perspective


Survey on Glaucoma Drainage Device Usage: A United Kingdom Perspective

Corresponding author: Dr. Andrew Feyi-Waboso, Royal Gwent Hospital, Aneurin Bevan University Health Board, Cardiff Road, Newport, NP20 2UB, United Kingdom, Tel: +44 1633 234 234; Email:  Tel: +234-8061302925; Email:

To determine current practice preferences for glaucoma drainage device (GDD) usage among members of the United Kingdom and Eire Glaucoma Society (UKEGS).
An anonymised web survey was emailed to UKEGS members on the 30th of June 2014 and open for three months. The responders’ demographics, experience and complication rate were determined. Fifteen poor prognosis and high risk clinical scenarios were presented.

A total of 83 (42%) of UKEGS members responded to the survey. Twenty four (32%) members were glaucoma fellowship trained but did not undertake GDD insertion. Fifteen (27%) members carried out between 5 and 9 cases annually, which was the most common group. Two (4%) members performed up to 45-50 GDD’s per annum. The most popular implant was Baerveldt 71%), followed by Ahmed (27%) and Molteno (2%). Preference for inserting a GDD was highest for neovascular glaucoma (81%), failed trabeculectomy with MMC (88%), previous penetrating keratoplasty (63%) and previous scleral buckle surgery (56%). The most common GDD complication encountered over the last year was hypotony (43%), followed by tube erosion (40%), tube extrusion (15%), visual loss (25%), corneal decompensation (25%), suprachoroidal haemorrhage (8%) and endophthalmitis (3%).
This is the first survey describing the practice preference in among the UKEGS members in the UK for GDD surgery. The management choice and scenario varies significantly between surgeons.
Keywords: Survey; Glaucoma Drainage Device; United Kingdom 
Systemic literature review showed that glaucoma drainage devices controls the intraocular pressure (IOP) effectively in glaucoma refractory to trabeculectomy [1]. The use of such devices has been demonstrated to be on an upward trajectory a few years ago [2,3]. Landmark studies such as the Tube Versus Trabe culectomy (TVT) study found Baerveldt implant had a higher success rate compared to trabeculectomy with mitomycin-C (MMC) when carried out in eyes with previous trabeculectomy, and/or cataract extraction with intraocular  lens implantation over 5 years [4]. The Ahmed Baerveldt Comparison (ABC) and Ahmed Versus Baerveldt (AVB) studies have shown that the BGI rendered up to 1.3mmHg lower IOP with 0.7 fewer medications compared to Ahmed implant up to 3 years follow-up [5,6]. However, both studies found that the Baerveldt implant was related with more serious postoperative complications.
The use of, and preference for GDDs in the UK has not been established. It is unknown how many glaucoma specialists are trained to utilise this modality. The aims of this study were to establish the use and practice preferences for GDD among the United Kingdom and Eire Glaucoma Society (UKEGS) members.
Materials and Methods
Guidance was sought from the Local Research and Development team, the study was agreed to be a service evaluation based on NISCHR guidance and approved [7]. An anonymised web survey (Survey Monkey) was emailed to UKEGS members on the 30/06/2014 and remained opened until 15/09/2014. Participation in the survey was voluntary. Demographic data was collected including age and location.
We established whether respondents viewed glaucoma as their primary sub-specialist interest and had fellowship training. We enquired whether they included GDD in their practice. If they did not, we enquired the reason and whether they would be keen to have the skill as an additional management option for uncontrolled IOP.
The members’ experience of using GDD, the total number inserted in a year and the types of device primarily used were ascertained. We asked whether antimetabolites were used in more than half of the GDD cases. Respondents were prompted whether they encountered specified complications in the previous year. Fifteen poor prognosis/high risk clinical scenarios were presented and members were advised to choose between trabeculectomy with MMC, or a GDD in the following: neovascular glaucoma; previous failed trabeculectomy  with and without MMC, previous extracapsular cataractextraction(ECCE)/intracapsular cataract extraction (ICCE), aphakic/pseudophakic glaucoma, previous penetrating keratoplasty, previous scleral buckling surgery, uveitic glaucoma, combined phacoemulsification and drainage device, previous pars plana vitrectomy, increased episcleral venous pressure, conjunctival scarring, patient younger than 50 years old and whether respondents would chose GDD as a primary surgery. Respondents were also encouraged to state other scenarios not included in the survey they deemed necessary for a GDD. They were also given the opportunity to choose other surgical modalities in the comments section. We finally established whether respondents audited their outcomes and complications.

Eighty-three (42%) of the 200 UKEGS members responded to our survey. The demographics of the respondents are presented in Table 1. Seventy-six (95%) out of 80 respondents viewed glaucoma as a primary subspecialist interest. Twenty-four (32%) members were glaucoma fellowship trained but did not undertake GDD insertion. Of these, nine (38%) respondents felt a lack of specialist training limited their ability to use them, seven (29%) stated they did not see a place in their current practice, 3 (13%) stated lack of experience, 1 lack of funds and 4 did not give a reason. The number of UKEGS members who perform GDD operation by region is given in Table 1.
Fifty-six (67%) members provided the estimated number of GDD surgeries performed in the previous year (Table 2). Fifteen (27%) members carried out between 5 and 9 cases, which was the most common group. Two (4%) members performed up to 45-50 GDD’s per annum. The most popular implant was Baerveldt (71%), followed by Ahmed (27%) and Molteno (2%).
Fifty-one (91%) out of 56 respondents claimed that they were fellowship trained. Of the 53 members who replied to the questions of antimetabolites, 27 (51%) used it in over half of the cases and 26 (49%) did not. Members would use antimetabolites in the scenarios where there is high risk for scarring e.g. uveitis and Afro-Caribbean, thicken Tenon’s capsule, young patients, previous multiple procedures, revisions and adults undergoing Ahmed valve surgery.
Forty members (49%) responded to the question on complications experienced over a year ago (Table 1). Seventeen (43%) encountered hypotony, 16 (40%) tube erosion, 6 (15%) tube extrusion, 10 (25%) visual loss, 10 (25%) corneal decompensation, 3 (8%) suprachoroidal haemorrhage and 1  (3%) endophthalmitis. The Spearman correlation between the numbers of GDD’s performed in a year and total complications was 0.4 (p=0.02).
Eight (14%) members would consider GDD as the primary surgical intervention for poor prognosis/high risk scenarios rather than trabeculectomy or any other glaucoma surgical procedure. The most common reasons for respondents to choose GDD were previous failed trabeculectomy with MMC
Table 1. Demographics of respondents, total GDD performed yearly and complications.
Table 2. Clinical scenario where respondents favour GDD over trabeculectomy and MMC.
(88%) and neovascular glaucoma (81%). Table 2 reveals other scenarios likely to result in respondents choosing a GDD. Ninety three per cents preferred a trabeculectomy and MMC for younger patients (age<50).
Fifty three (64%) members responded to the question of surgical outcome audit and 44 (83%) audited their outcomes regularly.
This is the first survey to establish the practice preferences for GDD surgery amongst UKEGS members in the UK. The American Glaucoma Society has undertaken a similar survey in 1996 and 2002. The latter showed an increased usage of GDDs for poor prognosis and high risk scenarios e.g. neovascular glaucoma, previous failed trabeculectomy, uveitic glaucoma, glaucoma after penetrating keratoplasty, glaucoma after scleral buckle for retinal detachment, and previous extracapsular or intracapsular cataract surgery [8]. Medicare claims data in the USA from 1995 to 2004 show a 43% decrease in the number of trabeculectomy procedures and a concurrent 184% increase  in tube shunt surgery [2]. We hope the result of this survey will provide a baseline for future comparison in the UK.
Surprisingly, 29% of the respondents who did not perform GDD procedure mentioned it did not have a place in their practice. The reason for this was uncertain. It is possible that other modalities such as diode laser contact cyclophotocoagulation may be used as it has been reported to have similar IOP reduction and complication rate as Ahmed implant in neovascular glaucoma patients [9]. Members in the age groups above 40 years old were the only ones to respond that did not carry out GDDs due to a lack of specialist training. We expect GDD usage to increase in the future for the UK; as the new generation of glaucoma specialists are more likely fellowship trained and would have gained GDD surgical exposure.
Baerveldt implant was preferred by more than 70% of the respondents. It is uncertain whether this is due to their previous training on one type of implant or by choice. The ABC and AVB studies showed Baerveldt implant to have lower IOP and required fewer glaucoma medications at 3-year followups [5,6]. The ABC study group has produced the 5-year data recently which revealed that the mean±standard deviation IOP for Baerveldt implant was 14.7±4.4 mmHg and Ahmed 12.7±4.5 mmHg (p=0.015) [10]. The number of eyes failed because of inadequately controlled IOP or reoperation was 80% in the Ahmed implant group and 53% in the Baerveldt implant group (p=0.003) [10]. However, the same group found that Baerveldt implant had more than twice failure (47% versus 20%) due visual threatening complications. It is not clear how these results will influence the current practice. Could Baerveldt implant be more popular because as the surgeon experience increases they may expect to have a lower rate of complication?
Just over half of the respondents admitted to using antimetabolites in over 50% of GDD operations. Two members mentioned they would use antimetabolites routinely in patients with very thick Tenon’s and adults undergoing Ahmed valve surgery. This is an interesting finding as two retrospective studies reported no benefit of intraoperative use of MMC with Baerveldt implants [11,12]. The use of MMC was also studied in 2 prospective randomized trials with Molteno and Ahmed implantats. Neither showed higher success rates with MMC in terms of final IOP and number of anti-glaucoma medications required postoperatively [13,14].
For the ‘poor prognosis/high risk’ scenarios our results were similar to the 2008 survey of American glaucoma society members survey which showed members preferred GDD compared to trabeculectomy in the neovascular glaucoma, penetrating  keratoplasty, prior scleral buckling procedure, prior pars plana vitrectomy and uveitic glaucoma [3].
Our study has some limitations. Only 42% of the ophthalmologists sent the survey responded. The survey was sent out to UKEGS members only and may not reflect the practice of all glaucoma specialists in the UK as some may not be a member. However this group represents the largest body of UK glaucoma specialists. This also does not mean only glaucoma specialist carry out glaucoma surgery; we recognise not all UKEGS members are glaucoma specialists. Our survey is unlikely to have captured the entire number of surgeons who undertake GDD surgery. We believe the survey has captured the largest group of glaucoma specialists in the UK and has avoided repetition.
The larger UK deaneries had a larger number of glaucoma specialists carrying out GDDs. We do hope these data will provide an idea of the spread of GDD usage in the UK.
We acknowledge that glaucoma surgery is undergoing a transition involving other surgical modalities and therefore encouraged respondents to include comments for most questions, hoping this would capture practice variance. Interestingly no surgeons who responded to the survey mentioned their use of other glaucoma surgeries e.g. Microinvasive Glaucoma Surgery.
This is the first survey in the UK attempting to capture the practice of GDD usage. The survey has determined the management decisions taken by the UK surgeons when considering GDD surgery vary greatly. This is probably due to the lack of definite guidance as to when the devices are best implanted. It would be interesting to see how the trend in the UK changes with time in light of some landmark studies and increasing fellowship training. We hope the result of this survey will provide a baseline for future comparison.

No author has a financial or proprietary interest in the material or method mentioned in the article. This study was not funded.

We are very grateful to UKEGS for facilitating this survey. We thank the all UKEGS members who responded.


1. Hong CH, Arosemena A, Zurakowski D, Ayyala RS. Glaucoma Drainage Device: A systematic literature Review and Current Controversies. Surv Ophthal. 2005, 50(1): 48-60.

2.Ramulu PY, Corcoran KJ, Corcoran SL, Robin AL. Utilization of various glaucoma surgeries and procedures in Medicare beneficiaries from 1995 to 2004. Ophthalmology. 2007, 114(12): 2265–2270.

3. Desai M, Gedde SJ, Feuer WJ, Shi W, Chen P et al. Practice Preference for Glaucoma Surgery: A Survey of the American Glaucoma Society in 2008. Ophthalmic Surg Lasers Imaging. 2011, 42(3): 202-208.

4. Gedde SJ, Schiffman JC, Feuer WJ, Herndon LW, Brandt JD et al. Treatment outcomes in the Tube Versus Trabeculectomy (TVT) study after five years of follow-up. Am J Ophthalmol. 2012, 153(5):789-803.

5. Barton K, Feuer WJ, Budenz DL, Schiffman J, Ahmed Baerveldt Comparison Study Group et al. Three-year treatment outcomes in the Ahmed Baerveldt Comparison Study. Ophthalmology. 2014,121(8): 1547–1557.

6. Christakis PG, Tsai JC, Kalenak JW, Cantor LB et al. The Ahmed Versus Baerveldt Study: three-year treatment outcomes. Ophthalmology. 2013,120(11): 2232–2240.

7. Research taking Place within the NHS.

8. Joshi AB, Parrish RK, Feuer WF. 2002 Survey of the American Glaucoma Society. Practice preferences for glaucoma surgery and antifibrotic use. J Glaucoma. 2005, 14(2): 172–174.

9. Yildirim N, Yalvac IS, Sahin A, Ozer A, Bozca T. A comparative study between diode laser cyclophotocoagulation and the Ahmed glaucoma valve implant in neovascular glaucoma: a long-term follow-up. J Glaucoma. 2009, 18(3): 192–196.

10. Budenz DL, Barton K, Gedde SJ, Feuer WJ, Schiffman J et al. Five-year treatment outcomes in the Ahmed Baerveldt comparison study. Ophthalmology. 2015, 122(2): 308-316.

11. Trible JR, Brown DB. Occlusive ligature and standardized fenestrations of a Baerveldt tube with and without antimetabolites for early postoperative intraocular pressure control. Ophthalmology. 1998, 105(12): 2243-2250.

12.Irak I, Moster MR, Fontanarosa J. Intermediate-term results of Baerveldt tube shunt surgery with mitomycin-C use. Ophthalmic Surg Laser and Imaging. 2004, 35(3):189-196.

13. Cantor L, Burgoyne J, Sanders S, Bhavnani V, Hoop J et al. The effect of mitomycin C on Molteno implant surgery: A 1-year randomized, masked, prospective study. J Glaucoma. 1998, 7(4): 240-246.

14. Costa VP, Azuara-Blanco A, Netland PA, Lesk MR, Arcieri ES et al. Efficacy and safety of adjunctive mitomycin C during Ahmed glaucoma valve implantation: A prospective randomized clinical trial. Ophthalmology. 2004,111(6): 1071- 1076.

Be the first to comment on "Survey on Glaucoma Drainage Device Usage: A United Kingdom Perspective"

Leave a comment

Your email address will not be published.