Primary implantation of glaucoma drainage device in secondary glaucoma: Comparison of Aurolab aqueous drainage implant versus Ahmed glaucoma valve

Abstract

Glaucoma drainage devices are invaluable in the management of secondary/ refractory glaucomas. This study aimed to compare the efficacy and safety of Aurolab Aqueous Drainage Implant (AADI) and the Ahmed Glaucoma Valve (AGV) in filtration-surgery-naïve secondary glaucoma eyes. For this purpose, a retrospective, comparative review was conducted on patients with secondary glaucoma (open and closed) who underwent primary tube procedures, either AADI or AGV. The primary outcome measure was intraocular pressure (IOP), and secondary measures included best-corrected visual acuity (BCVA), number of antiglaucoma medications (AGMs), and complications. This study included 59 eyes in the AADI group with a mean follow-up of 20.3 ± 12.9 months and 61 eyes in the AGV group with a mean follow-up of 19.8 ± 11.8 months. Preoperative IOP, AGM use, and BCVA did not significantly differ between the groups. However, at the last visit, both IOP and AGM use were significantly lower in the AADI group (12.9 ± 3.7 mmHg and 0.6 ± 0.9 vs. 15.7 ± 2.7 and 1.8 ± 1.0 respectively, all p < .001). Moreover, the AADI group exhibited a significantly higher rate of complete success (57.6%) compared to the AGV group (14.7%, p < .001); corresponding qualified success was 91.5% and 80.3%. Serious complication rates were comparable between the two groups. In conclusion, toth AADI and AGV procedures effectively reduced IOP and the need for AGMs. However, the reductions were significantly greater in the AADI group, which also showed a higher rate of complete success. Considering its affordability, AADI could have a substantial positive impact, particularly in resource-constrained settings.

Secondary glaucomas constitute a significant proportion of all glaucomas, with the majority proving highly refractory to treatment. These eyes often exhibit heightened inflammation, develop membranes over the angle leading to its closure, or present with scarred conjunctiva due to prior surgeries, or a combination thereof.[1] Despite efforts with medical management and even 'oral therapy', they frequently exhibit treatment failure, necessitating multiple trabeculectomies before considering a glaucoma drainage device (GDD).

Several authors have also noted that when GDDs are implanted in eyes with previous trabeculectomy failures, they too tend to fail.[2,3,4,5] Consequently, a therapeutic strategy yielding improved outcomes in refractory secondary glaucoma cases involves primary GDD implantation.

The reported incidence of secondary glaucomas in India varies widely, ranging from 6% to 21.8%.[6,7] Despite this discrepancy, consensus exists among most authors regarding the refractory nature of such glaucomas with GDD implantation being necessitated in a substantial proportion.

In India, three types of GDDs are available. Ahmed glaucoma valve (AGV, New World Medical Inc., California, USA) has been available for >20 years. The cost-effective, locally manufactured non-valved Aurolab Aqueous Drainage Implant (AADI, Aurolabs, India) with a surface area of 350 mm², modeled after the Baerveldt glaucoma implant (BGI, Advanced Medical Optics, California, USA) has been available since 2013. In addition, a newer non-valved device known as Ahmed Clear Path (ACP, New World Medical Inc., California, USA) is available since 2023 in India; it is available in two sizes: 250 mm² and 350 mm².

While some studies have compared AGV and AADI in refractory glaucomas in adults,[8,9,10] none have exclusively focused on the outcomes of primary GDD implantation in secondary glaucomas. Therefore, this study aimed to investigate the comparative efficacy and safety of AADI and AGV when implanted in eyes with secondary glaucoma naive to filtration surgery.

Methods

This comparative study conducted a retrospective review of consecutive adult patients who underwent primary GDD surgery between January 2017 and December 2021 under the care of a single senior glaucoma surgeon. Ethical clearance was obtained from an independent ethics committee, and the study adhered to the principles outlined in the Declaration of Helsinki. Informed written consent for surgery was obtained from all eligible participants.

Inclusion criteria encompassed consecutive surgeries using either the AGV or the AADI in eyes with secondary open and closed angle glaucoma that had not undergone filtration surgery, with a minimum follow-up period of 6 months or more.

Exclusion criteria comprised eyes that had previously undergone trabeculectomy and eyes in which Goldmann applanation tonometry was either not feasible or compliance was poor (e.g. in cases involving keratoprosthesis and pediatric eyes).

Surgical methods

The surgical procedure for AADI has been previously described in detail.[9,11] To summarize, all AADI plates were positioned under adjacent recti without trimming, occlusion of tube was done with 6-0 vicryl, and 3–4 fenestrations were made anterior to the occlusive ligature to assist control of IOP prior to suture autolysis, anticipated at 5–6 weeks postoperatively. Technique of ripcord was not used.

For AGV implantation, a standard procedure between muscles was followed under peribulbar block and sterile conditions. Notably, after priming the implant, it was anchored to the sclera, 10 mm posterior to the limbus, by using a preplaced 6-0 vicryl suture (polyglactin 910 violet: Ethicon, Johnson and Johnson, HP, India). Long pass of 6-0 Vicryl was used 10 mm posterior to the limbus, and this was then threaded through the eyelets of the plate and secured in a mattress configuration.

For both types of tubes, almost the entire length of the tube was covered with either a corneal or scleral patch graft, prepared in advance.[11] The patch graft was secured over the tube by using fibrin glue or 10-0 nylon sutures. The conjunctiva and tenon were then repositioned with wing and continuous sutures. Following the procedure, a subconjunctival injection of dexamethasone (4 mg) was administered, and the eye was patched; it was then unpatched after 24 hours.

Postoperatively, moxifloxacin topical antibiotic (0.5%) was used four times daily for 1 week, and cycloplegic eye drops (homatropine 2%) were used as needed for 1–3 weeks. Topical steroid drops (difluoro-prednisolone butyrate acetate, or DFBA 0.05%) were started every 2 hours and gradually tapered over 8–12 weeks.

In the AGV group, antiglaucoma medications (AGMs) were reintroduced early postoperatively (usually 2–3 weeks) if IOP exceeded 14 mmHg. Topical aqueous suppressants such as beta-blockers, carbonic anhydrase inhibitors, or alpha-adrenergic agents, either alone or in fixed combinations, were used. AGM was continued postoperatively in the AADI group as needed until suture autolysis occurred.

Follow-up visits were scheduled based on clinical indications, with data documented at day 1, 1 week, 6 weeks, 3 months, 6 months, 1 year, and at the last follow-up postoperatively.

Outcome criteria

The primary outcome measure was intraocular pressure (IOP), while secondary outcome measures included the number of AGMs, LogMAR best-corrected visual acuity (BCVA), and complications.

Complete success was defined as an IOP between 5 mmHg and 21 mmHg. Meeting these IOP criteria with AGMs was considered a qualified success. Failure was defined as an inability to meet IOP criteria, loss of light perception, device explantation, or additional glaucoma surgery to reduce IOP.

Any eye undergoing a re-procedure or experiencing a reduction in BCVA of 2 lines or more was considered a serious complication.

The hypertensive phase was defined by a tense cystic bleb around the plate with significantly increased height, accompanied by an IOP exceeding 21 mmHg with or without AGM, following the reduction of IOP to less than 22 mmHg during the early postoperative period. This was observed from the third week onwards for AGV and post-suture autolysis after the sixth week for AADI.

Statistics

Descriptive statistics were employed to compare the baseline demographic and ocular characteristics of the treatment groups. Descriptive data are presented as mean ± standard deviation. Normality was assessed using the Shapiro-Wilk test. Accordingly, univariate comparisons were conducted using the paired t-test or Wilcoxon signed rank test within groups and the independent t-test or Mann-Whitney U test for between-group comparisons. Categorical variables were analyzed using the Chi-squared test or Fisher’s exact test. Snellen visual acuity was converted to logarithm of minimal angle of resolution (logMAR) for analysis. Data on IOP and AGMs were censored if explanted or if a second glaucoma surgery was needed, whereas visual acuity data was not censored. Survival analysis was performed using the Kaplan-Meier method, and risk factors for treatment failure were assessed for statistical significance using the Cox proportional hazard model.

All statistical tests were two-sided, and statistical significance was defined as P < 0.05. Statistical analyses were conducted using Stata 12.1 (StataCorp, College Station, TX).

Results

A total of 126 eyes from 119 subjects underwent primary GDD surgery for secondary glaucoma. All surgeries were performed by a single senior fellowship-trained glaucoma specialist. Fifty-nine eyes underwent AADI, while 67 eyes underwent AGV surgery. All eyes in the AADI group had a follow-up of more than 6 months and were included in the analysis; however, six out of 67 eyes in the AGV group had less than 6 months of follow-up and were excluded. The mean follow-up was 20.3 ± 12.9 months in the AADI group and 19.8 ± 11.8 months in the AGV group.

Superotemporal (ST) was the most common quadrant for GDD placement in both groups, with 98.4% (n = 60) in the AGV group and 91.5% (n = 54) in the AADI group. The remaining placements in both groups were in the inferotemporal (IT) quadrant. Furthermore, 50.8% (n = 31 eyes) of AGV tubes and 40.7% (n = 24 eyes) of AADI tubes were placed in the ciliary sulcus (CS).

A small proportion of eyes underwent simultaneous GDD and cataract surgery in both groups (AADI: n = 7, 11.8%; AGV: n = 4, 6.5%). In addition, 4.9% of eyes (n = 3) in the AGV group underwent simultaneous sclera-fixated intraocular lens implantation using the flanged technique.

Baseline characteristics between groups are summarized in Table 1.

Table 1.

Baseline characteristics between the two groups – Aurolab aqueous drainage implant (AADI) and Ahmed glaucoma valve (AGV)

	AADI n=59	AGV n=61	P
Age in years	47.2±19.7	45.9±17.0	0.689
Intraocular pressure in mmHg Mean±SD	35.8±10.6	33.1±11.7	0.139
Antiglaucoma medications (number) Mean±SD	3.8±0.9	4.1±0.7	0.542
BCVA (LogMAR) Mean±SD	1.1±0.7	1.2±0.7	0.363
Etiology:
Neovascular glaucoma	12	25	0.055
Post retinal surgery	13	19	0.177
Aphakic/pseudophakic	6	11	0.423
Post corneal transplant (PK/DSEK)	7	1	0.024
Uveitis	10	2	0.012
Others	11	3	0.019

IOP and AGM:

In the AADI group, mean IOP decreased significantly from 35.8 ± 10.6 mmHg at baseline to 12.9 ± 3.7 mmHg at the last follow-up (P < 0.001), a 64% reduction. Similarly, in the AGV group, the mean IOP decreased from 34.8 ± 13.3 mmHg at baseline to 15.7 ± 2.7 mmHg at the last follow-up (P < 0.001), a 54.8% reduction. The AGV group had a significantly lower mean IOP than the AADI group only on day 1 and week 1 (both P < 0.001), whereas the AADI group showed a significantly lower mean IOP at every subsequent postoperative visit [Table 2].

Table 2.

Baseline and follow-up intraocular pressures (IOP) and antiglaucoma medications (AGM) for the two groups, Aurolab aqueous drainage implant (AADI) and Ahmed glaucoma valve (AGV)

	AADI IOP mmHg Mean±SD	AGV IOP mmHg Mean±SD	P	AADI number of antiglaucoma medications Mean±SD	AGV number of antiglaucoma medications Mean±SD	P
Pre-op	35.8±10.6	34.8±13.3	0.139	3.8±0.9	4.1±0.7	0.542
†POD1	22.3±11.7	12.1±5.2	<0.001	-	-
‡POW1	20.6±9.6	14.0±7.0	<0.001	1.7±1.2	0.4±0.7	<0.001
POW6	9.3±7.6	16.9±6.5	<0.001	0.8±1.4	1.5±1.0	0.002
§POM3	13.1±4.2	16.0±5.0	0.008	0.6±0.9	1.8±1.1	<0.001
POM6	11.9±3.7	14.8±3.6	<0.001	0.7±1.1	1.7±1.0	<0.001
Last follow-up	12.9±3.7	15.7±2.7	<0.001	0.6±0.9	1.8±1.0	<0.001
†POD	‡POW	§POM
Post-op day	Post-op week	Post-op month

There was also a significant reduction in the need for medical therapy in both treatment groups (P < 0.001), with a significantly lower need for AGM in the AADI group at every postoperative visit after week 1 [Table 2]. At the final follow-up, reduction in AGM was 84% in the AADI group and 56% in the AGV group.

Fig. 1 plots the IOP and AGM between groups at various timepoints.

Figure 1.

Intraocular pressure (IOP) (top) and number of antiglaucoma medication/s (AGM) (bottom) in Aurolab aqueous drainage implant (AADI) and Ahmed glaucoma valve (AGV) groups

Complications

The early complications and total number of complications were significantly higher in the AGV group compared to the AADI group. However, late and serious complications did not differ significantly between the groups [Table 3].

Table 3.

Early, late, and serious complications for the two groups and the interventions for these in Aurolab aqueous drainage implant (AADI) and Ahmed glaucoma valve (AGV)

	AADI n, %	AGV n, %	Interventions for all complications	AADI n, %	AGV n, %
Early complications ≤3 months Between groups
Hyphaema	2, 3.3%	8, 13.1%	AC wash/reformation	1, 1.7%	2, 3.2%
Hypotony/Choroidals	4, 6.7%	5, 8.2%	CD drainage	2, 3.3%	2, 3.2%
Shallow AC/aqueous misdirection	-	2, 3.2%	Irido-zonulo-hyaloido-vitrectomy	-	1, 1.6%
Fibrin/Hypopyon	3, 5.1%	5, 8.2%	-	-	-
Endophthalmitis	-	1, 1.6%	Vitrectomy (pars plana)	-	1, 1.6%
Conj gape/retraction/tube/plate exposure	8, 13.5%	8, 13.1%	Conj suturing/CLAG/re-do patch graft	2, 3.3%	5, 8.2%
Tube block	1, 1.7%	2, 3.2%	Anterior vitrectomy	1, 1.7%	1, 1.6%
Vitreous Haem	1, 1.7%	2, 3.2%	Vitrectomy pars plana	1, 1.7%	1, 1.6%
Total early complications ≤3 months P=0.015	19, 32.2%	33, 54%
Late complications >3 months
Corneal decompensation	2, 3.3%	2, 3.2%
Tube block	1, 1.7%	-	Anterior vitrectomy	1, 1.7%	-
Patch graft melt	-	1, 1.6%	Re-do patch graft	-	1, 1.6%
Vitreous Haem	-	2, 3.2%
Endophthalmitis	-	1, 1.6%	Vitrectomy pars plana	-	1, 1.6%
Spontaneous extrusion	-	1, 1.6%
Tube-endo touch	1, 1.7%	-	Tube Repositioning	1, 1.7%	-
Late complications >3 months	4, 6.7%	7, 11.4%	Total interventions P=0.201	9, 15.2%	15, 24.5%
Total early and late complications P=0.373	23, 38.9%	40, 65.5%	Serious Complications (Interventions + VA worse ≥2 lines or NLP) P=0.780	13, 22%	19, 31.1%

Visual acuity

There was no statistically significant difference in visual acuity outcomes between the AADI and AGV groups. A similar proportion of eyes in both groups either had no change or showed improvement in BCVA (n = 47, 79.7% in the AADI group and n = 49, 80.3% in the AGV group; P = 0.927). One eye in the AGV group lost perception of light, but none in the AADI group did so.

Hypertensive phase

The hypertensive phase (HTP) was observed less frequently in the AADI group (23.7%) compared to the AGV group (50.8%) (P = 0.006). This phase was most commonly observed 3–6 weeks postoperatively in the AGV group, despite the use of early aqueous suppressants.

Outcomes

The AADI group showed a statistically significant higher rate of complete success (P < 0.001) compared to the AGV group. Complete success was seen in 34 eyes (57.6%) in the AADI group, and another 20 eyes (33.9%) achieved IOP control with medication (qualified success). Therefore, total success (complete + qualified) in the AADI group was 91.5% (n = 54). Complete success was seen in 14.7% (n = 9) in the AGV group and an additional 65.6% (n = 40) as qualified success (total success 49 eyes, 80.3%).

Kaplan-Meier analysis demonstrated a higher cumulative probability of complete success in the AADI group compared to the AGV group over time [Fig. 2]. A Kaplan-Meier plotting of cumulative probability of complete success was 66%, 60%, and 50% in the AADI group at year 1, 2, and 3, respectively, whereas it was 22%, 18%, and 15% in the AGV group (log-rank < 0.001) at year 1, 2, and 3, respectively [Fig. 2 left]. The total success was 92%, 90%, and 82% at year 1, 2, and 3, respectively, in the AADI group and that in the AGV group was 88%, 72%, and 62% [Fig. 2 right].

Figure 2.

Kaplan-Meier survival by group Aurolab aqueous drainage implant (AADI) and Ahmed glaucoma valve (AGV) – complete success (left) and total success (right)

Five eyes failed in the AADI group (8.4%), and 11 eyes (18%) did so in the AGV group (P = 0.123). Device explantation was commoner in AADI (n = 4), whereas failure on IOP criterion was higher in the AGV group (n = 8).

The Cox proportional hazards model identified the hypertensive phase and AGV as significant risk factors for treatment failure.

Discussion

This study compares the primary implantation outcomes of AADI and AGV in secondary glaucomas and highlights that AADI achieves significantly lower IOP with fewer AGM compared to AGV. Although the overall success rate did not differ between the groups, the complete success rate was notably higher in the AADI group.

Most studies, including prominent randomized controlled trials such as the Ahmed versus Baerveldt study (AVB study) or the Ahmed Baerveldt Comparison study (ABC study),[12,13] have typically evaluated mixed cohorts of primary and secondary glaucomas, with a substantial proportion having undergone previous trabeculectomy. However, subgroup analyses from these studies have shown varying outcomes; for instance, primary glaucomas previously operated showed better results than neovascular glaucomas. Pooled data from these studies indicate that non-valved devices such as the Baerveldt glaucoma implant (BGI) offer lower failure rates compared to valved devices such as the AGV, with better achievement of target IOP.[14]

AADI closely resembles BGI 350, and previous studies have reported varying failure rates over time, potentially influenced by factors such as prior filtration surgeries.[15] Our results align with these trends at year 1, showing comparable success rates to previous reports.[15] The failure rate reported by Puthuran et al.[15] at year 2 is much higher (27.8%) than that reported previously by Ray VP et al.,[16] as well as in this study (10%). Puthuran et al. reported increased failure rate in primary (vs. secondary) glaucomas perhaps due to the higher rate of previous filtration surgery, lending credence to the hypothesis that this may be implicated in lower success rates in GDD surgery, as noted by several other authors too.[2,3,4,5]

Results in this study are similar to the pooled data from the ABC and AVB studies at the end of 5 years,[13] which showed that non-valved surgery group produced a lower mean IOP on fewer medications and had lower failure rates than the AGV group. However, it differed from these studies in as much that the BGI group carried a risk of hypotony, which was not seen in this study.

While Indian studies comparing AADI with AGV are scarce, existing research corroborates our findings, showing higher success rates with AADI, particularly in longer-term follow-ups. Pathak Ray previously reported the outcomes of the two in a mixed cohort of primary and secondary glaucomas at 1 year – 92.3% overall success in the AADI group and 80.5% in the AGV group (P < 0.001).[9] Pandav et al.,[8] also reported in a mixed cohort a total success of 73.08% and 58.18% in the AADI and AGV groups, respectively, at 3 years. The corresponding total success rates in the current study are 92% and 88% at 1 year and 82% and 62% at 3 years in the AADI and AGV groups, respectively.

Regarding tube positioning, our study adheres to the ciliary sulcus (CS) in pseudophakic and aphakic eyes where possible,[17] with notable benefits observed in minimizing complications related to tube-endothelial touch and corneal decompensation. However, in eyes that undergo simultaneous sclera fixated IOL, it may be better to place the tube in the AC rather than risk IOL tilt and/or subluxation.[18] The rate of corneal decompensation was low in this study (<5%), and this is comparable to the rate reported in a recent large retrospective cohort;[19] however, it has also been reported to be as high as 17% and 19%.[20,21] Beatson et al.[19] found age, postoperative hypotony, tube-cornea touch, Fuchs and Irido-corneal endothelial (ICE) syndrome to be significant risk factors. The preferred option of tube positioning in ICE syndrome cases, whenever GDD is indicated, should be the CS as far as feasible, as reported in a previous case series,[22] delaying or avoiding the need for keratoplasty.

The hypertensive phase, a concern with valved devices, occurs less frequently and later with non-valved devices such as BGI and AADI, consistent with the literature. It tends to occur early and with greater severity in the valved devices as it is hypothesized that the early access of the aqueous to the nascent bleb around the endplate immediately post surgery allows pro-inflammatory cytokines and other ligands to set up a fibrotic response.[23] It is postulated that this thickens and forms encystment of the bleb lining, leading to increased resistance and thus an increase in IOP occurs. The incidence of HTP in AGV has been reported variably in the literature – Ishida et al. reported 26% and 35%, respectively, in a White and African-American cohort,[24] 33% was reported by Abe et al.[25] in a mixed racial cohort, 31.3% was reported in a Korean population,[26] 41.1% in an Egyptian population,[27] and 58.3%, 41.8%, and 63.93% in three different studies in the Indian population.[9,16,28] Strategies such as early aqueous suppression in AGV aim to mitigate this, albeit with marginal improvement.[29] On the contrary, HTP occurs rather less frequently and presents much later in AADI, only after the occlusive suture autolyses, and has been reported to occur in approximately 20% of eyes.[10,15]

Complication rates, though higher in AGV, were manageable and comparable to previous studies.

Even though complications occurred more frequently in the AGV group (65.5% vs. 38.9%), most of these in both groups were transient and were managed conservatively. These rates are comparable to those reported by Pandav et al. in the AGV group (62.9%) but not in the AADI group (75.2%).[8] Hypotony only occurred as an early complication (<3 months postoperatively) in both groups and was not statistically significant. Early hypotony in the AGV group was likely due to peritubal leakage; severe choroidal exudation occurred in two such eyes that required drainage. Most notably, none of the eyes in this study had late hypotony related complications in the AADI group, whereas Pandav et al. reported a rate of 3.5% in the AADI group and 0.5% in the AGV group.[8] This difference in the AADI group between studies may be explicable by the fact that all the surgeries in this series were performed by a single experienced surgeon. Though there is not much variability in the surgical technique for the performance of an AGV, considerable technical skill is required in the implantation of a non-valved implant and may differ from surgeon-to-surgeon based on their experience.

However, the interventions required for some of these complications are rather more comparable between the groups in both the studies – 24.5% in AGV vs. 17% reported by Pandav et al. and 15.2% in the AADI group vs. 16% reported by the same authors.[8] In a previous report by Pathak-Ray et al.,[9] rate of interventions was 30.7% (vs. 15.2% in this study), whereas it remained comparable in AGV implantation in both the studies. This may be indicative of a reduced rate of interventions in primary implantation of non-valved surgery in secondary glaucomas. Nonetheless, this hypothesis can be proven only when a study is conducted inspecting primary and secondary implantation of AADI in secondary glaucomas.

Visual acuity outcomes were similar between groups, with a few cases of vision decline. Four eyes had vision worse than 2 lines in the AADI group and two eyes in the AGV group. Only one eye developed no light perception (following endophthalmitis) in the entire cohort in the AGV group.

Limitations of our study include its retrospective nature and inherent selection bias. This limitation was partially overcome by the inclusion of consecutive eligible eyes with the surgery being performed by a single surgeon in both groups, eliminating variability of surgical technique as a confounding factor. Nonetheless, the study design ensured consistency in surgical technique, reducing confounding factors.

In conclusion, both AADI and AGV are effective and safe primary implants in filtration-surgery-naïve eyes with secondary glaucoma. AGV, while controlling IOP, requires more AGM and is prone to a higher incidence of hypertensive phase. Conversely, AADI exhibits higher complete success rates, lower IOP, fewer AGM, and fewer complications, establishing its efficacy and safety in this context. Notably, persistent hypotony-related complications were absent in the AADI group, possibly attributed to consistent technique.

Conflicts of interest:

Nil relevant (BVI, Santen, Alcon/Novartis, Allergan, Glaukos, Viatris).

Funding Statement

Nil.