A Retrospective Cohort Study Analyzing Postoperative Safety and Efficacy of a Baerveldt Glaucoma Implant Modification for Earlier Intraocular Pressure Lowering, Compared to Nonmodified Baerveldt Implants at 3 Months

Abstract

Aim and background

The Baerveldt glaucoma implant (BGI) is a tube shunt that has proven long-term safety and efficacy in lowering intraocular pressure (IOP). However, it requires temporary ligation for 6–8 weeks before it can take effect. This study compares the efficacy and safety of the BGI modified with a reversible ventilating mattress suture (vsBGI) to allow early flow with the unmodified BGI.

Materials and methods

A retrospective cohort study was performed at the University of Iowa, including patients who underwent either an unmodified Baerveldt tube shunt (BGI) or a vsBGI between 2016 and 2023. Primary outcomes were IOP, number of glaucoma medications, and best-corrected visual acuity. Adverse events within 3 months postoperatively were recorded.

Results

A total of 124 eyes from 114 patients were included in the study. The vsBGI group had 62 eyes from 58 patients, and the BGI group had 62 eyes from 56 patients. At baseline, median IOP was significantly higher in the vsBGI group compared to those receiving the standard BGI (p < 0.001), but on day 1 and week 1, the median IOP was significantly lower in the vsBGI group than BGI (p < 0.001). By month 3, there was no significant difference in IOP between the two groups. From week 1 through month 3, patients in the vsBGI group required significantly fewer topical medications compared to those with the standard BGI (all p ≤ 0.017), even though no baseline differences were observed. Visual acuity was not significantly different between the groups at any time point. The number of adverse events or rate of reoperations in the 3-month postoperative period were not statistically different between vsBGI and BGI.

Conclusion

The reversible ventilating suture modification to the BGI tube shunt is a safe and effective method of allowing early aqueous humor flow and reducing IOP in the early postoperative period.

Clinical significance

The ventilating suture modification to the BGI provides a practical approach to overcoming the delay in IOP lowering that accompanies traditional tube ligation. By permitting controlled early aqueous flow, the vsBGI achieves meaningful pressure reduction in the immediate postoperative period without increasing adverse events or reoperation rates.

Introduction

Glaucoma is a leading cause of irreversible blindness worldwide, and lowering intraocular pressure (IOP) is currently the only proven treatment strategy.^1,2 Tube shunts play a critical role in achieving target IOP in cases of refractory glaucoma or when other treatment options, such as medication and trabeculectomy, do not provide adequate reduction.^2,3

The Baerveldt glaucoma implant (BGI) is a tube shunt that has proven safety and efficacy in the management of glaucoma.^4–9 It has reliable long-term IOP reduction and a well-established safety profile.^5,7,9 However, valveless tube shunts like the BGI must be temporarily occluded with a ligation suture, which will remain in place for approximately 6–8 weeks until the suture dissolves to prevent immediate postoperative hypotony. During this time of ligation, a fibrin capsule forms around the drainage plate, creating adequate resistance to flow to maintain proper IOP when the tube opens. This temporary occlusion prevents immediate reduction of IOP, making the BGI unfit to treat eyes with very high IOPs that require prompt IOP reduction postoperatively. Valved glaucoma shunts such as the Ahmed glaucoma valve (AGV) can be effective immediately after implantation; however, at 5 years, the BGI has a lower failure rate, rate of de novo glaucoma surgery, and provides lower IOP on fewer glaucoma medications, but has a higher incidence of hypotony compared to the AGV.⁷ Previous attempts at mitigating this delay in BGI IOP reduction have included making fenestrations in the drainage tube. However, this has shown limited efficacy, as they commonly seal shut, preventing the shunting of aqueous humor and leading to an inadequate IOP-lowering effect.^10,11

A ventilating suture modification to the BGI (vsBGI) has been described that aims to hold venting slits patent in the early postoperative period to allow for improved outflow and earlier IOP reduction.¹² Akil et al. retrospectively compared this ventilating suture method against the traditional ventilating slit technique; their results suggested the ventilating suture modification to be a reliable, safe, and reversible method to promote early shunting of aqueous humor and achieve postoperative IOP control in the immediate postoperative period after BGI implantation comparable to a valved tube shunt.¹² The modification typically loosens and loses efficacy in time for the ligation suture's dissolution at week 6. Additionally, this modification can easily be reversed at the slit lamp if the patient experiences hypotony in the initial postoperative period. The previous investigation by Akil et al. assessed follow-up out to the immediate weeks after tube opening (roughly 8 weeks postoperative).¹² In this study, we sought to build upon their results with a larger sample size between both vsBGI and BGI groups, and with a longer follow-up period. Our goal was to compare the efficacy and safety of the vsBGI to the unmodified BGI with follow-up out to the third postoperative month.

Materials and Methods

Study Design and Patient Selection

A retrospective cohort study was performed at the University of Iowa and included patients who underwent either an unmodified BGI or a vsBGI between 2016 and 2023. The study was approved by the University of Iowa Institutional Review Board (Protocol Number: 202409674), and the requirement for informed consent was waived due to the retrospective nature of the study. Manual chart review of electronic medical records was conducted. Patients younger than 18 years were excluded. The University of Iowa's Institutional Review Board approved this study. Informed consent was waived due to the retrospective study design, consistent with institutional policy. Surgical technique selection was at the discretion of the treating surgeon and reflected clinical disease severity at presentation.

Surgical Procedure

The ventilating suture modification is performed after inserting the tube into the eye through a sclerostomy. A ventilating slit is made in the proximal tube by passing a 10-0 nylon suture through the tube lumen. The same 10-0 nylon is then used to make a partial-thickness pass to the peripheral cornea in a horizontal mattress technique. The suture tension is adjusted to allow for the desired amount of flow through the ventilating slit and tied off. The resulting horizontal mattress suture will have a visible loop in the peripheral cornea, which is accessible postoperatively in case too much flow is noted and it needs to be removed in clinic. The remainder of the BGI surgery technique is unchanged.

Study Variables and Outcomes

Baseline demographic and clinical variables included age, sex, race, central corneal thickness (pachymetry), cup-to-disk ratio (CDR), glaucoma type, lens status (phakic or pseudophakic), history of systemic steroid use, diabetes, and prior glaucoma surgeries.

Primary outcomes were IOP, number of glaucoma medications, and best-corrected visual acuity. Visual acuity was reported as logMAR values at day 1, week 1, month 1, and month 3 using a previously described Snellen fraction to logMAR calculator.¹³ CDR was also assessed longitudinally.

Adverse events recorded included hyphema, corneal blood staining, corneal edema or keratoplasty graft compromise, shallow anterior chamber (AC), choroidal hemorrhage, clinically significant hypotony, cystoid macular edema, retinal detachment, tube erosion, endophthalmitis, diplopia, and need for additional glaucoma surgery.¹⁴ The complications we chose to assess for were based on the Tube vs Trabeculectomy study designs.^15,16 We defined clinically significant hypotony as IOP ≤5 mm Hg accompanied by choroidal effusion, choroidal hemorrhage, and/or hypotony maculopathy at any postoperative time point. This definition was made to distinguish transient, clinically benign hypotony from hypotony with vision-threatening sequelae.^17,18 Events with 0 incidence in both groups were documented but not included in statistical comparisons.

Data Analysis and Statistics

Baseline continuous variables were assessed for normality with the Shapiro–Wilk test. Due to frequent violations of normality, all baseline continuous variables were compared using Wilcoxon rank-sum tests. For categorical variables, expected counts were calculated. Fisher's exact test was applied when all expected counts were <5, and the Chi-squared test was used otherwise. All analyses were performed in R version 4.5.0 (R Foundation for Statistical Computing, Vienna, Austria).

Longitudinal outcomes were analyzed with mixed-effects models. Continuous outcomes, including IOP, best-corrected visual acuity (logMAR), and CDR, were modeled using linear mixed-effects models, with surgery type, time point, and their interaction as fixed effects and random intercepts for subjects. Surgical selection was nonrandom, and preoperative IOP differed significantly between groups; therefore, baseline IOP was included as a covariate in all longitudinal IOP models to account for confounding by disease severity (channeling bias). This modeling approach was chosen to support comparative effectiveness analyses while accounting for baseline disease severity in this retrospective cohort. Glaucoma medication counts were analyzed using a Poisson generalized linear mixed-effects model with surgery type, time point, baseline medication, and their interaction as fixed effects, with random intercepts for each subject. Model dispersion was calculated by dividing the Pearson residual deviance by the model's residual degrees of freedom.

Estimated marginal means with 95% confidence intervals were obtained for each model using the emmeans package, and pairwise comparisons were performed across groups and time points. Adverse events occurring up to month 3 postoperative were compared between groups using Fisher's exact test. Events with 0 incidence in both groups were reported descriptively but excluded from hypothesis testing.

Results

A total of 124 eyes from 114 patients were included in the study. The vsBGI group had 62 eyes from 58 patients, and the BGI group had 62 eyes from 56 patients (Table 1). At baseline, the two groups were similar in age, sex, race, lens status, diabetes, steroid use, and visual acuity. Other variables, including CDR, medication count, and pachymetry, were not significantly different. Implant quadrant location was not significantly different between groups; however, there was a trend toward increased superotemporal placement in the vsBGI group (89 vs 78%).

Table 1.

Baseline characteristics by surgery type

Characteristic	N	BGI N = 62	Modified BGI N = 62	p-value
Age	124	72 (64, 76)	70 (59, 77)	0.168
Sex	124			0.150
Male		34 (55%)	25 (40%)
Female		28 (45%)	37 (60%)
Race	124			0.181
Caucasian		57 (92%)	52 (84%)
Black		4 (6%)	8 (13%)
Hispanic		0 (0%)	2 (3%)
Other/unknown		1 (2%)	0 (0%)
Lens status	124			0.455
Phakic		20 (32%)	25 (40%)
Pseudophakic		42 (68%)	37 (60%)
Diabetes	124			0.654
No		48 (77%)	51 (82%)
Yes		14 (23%)	11 (18%)
Steroid use	124			0.761
No		55 (89%)	57 (92%)
Yes		7 (11%)	5 (8%)
Glaucoma type	124
POAG		29 (47%)	36 (58%)
PXG		3 (5%)	4 (6%)
Neovascular		0 (0%)	0 (0%)
Uveitic		3 (5%)	3 (5%)
Congenital		0 (0%)	0 (0%)
Juvenile		3 (5%)	1 (2%)
Steroid-induced		4 (6%)	3 (5%)
Trauma		0 (0%)	0 (0%)
Angle-closure		1 (2%)	2 (3%)
Other		11 (18%)	8 (13%)
Preop IOP	124	19.5 (16.0, 23.0)	25.0 (19.0, 32.0)	<0.001
Preoperative medications	124	4 (3, 4)	4 (3, 4)	0.188
Pachymetry	113	558 (522, 593)	555 (534, 586)	0.728
Baseline CDR	118	0.85 (0.75, 0.90)	0.85 (0.80, 0.90)	0.429
Preoperative LogMAR	124	0.18 (0.00, 0.40)	0.18 (0.00, 0.54)	0.561
Implant quadrant location	123			0.153
Superotemporal		47 (77%)	55 (89%)
Superonasal		1 (2%)	0 (0%)
Inferonasal		12 (20%)	7 (11%)
Inferotemporal		1 (2%)	0 (0%)

There was a significant difference in baseline IOP, with the vsBGI group having a significantly higher preoperative IOP compared to those receiving standard BGI [median: 25.0 (19.0, 32.0) mm Hg vs 19.5 (16.0, 23.0) mm Hg, p < 0.001]. In the postoperative period, the vsBGI group achieved significantly lower adjusted IOP than BGI at day 1 [mean difference 9.5 mm Hg, 95% CI (6.7, 12.2), p < 0.001] and week 1 [mean difference 5.8 mm Hg, 95% CI (3.0, 8.5), p < 0.001] during the period when the ventilating suture is in effect but before the BGI tube opens (Fig. 1). However, at postoperative month 1 [mean difference 2.4 mm Hg, 95% CI (−0.32, 5.2), p = 0.083] and month 3 [mean difference −1.7 mm Hg, 95% CI (−4.4, 1.1), p = 0.241], there was no significant difference between the two groups.

Fig. 1.

Estimated IOP (mm Hg) for vsBGI and BGI groups up to month 3 postoperative. There was a statistically significant decrease in IOP at day 1 and week 1 in the vsBGI group compared to BGI (p < 0.001). Preoperative IOP is not displayed, as baseline IOP was included as a covariate in the model due to significant preoperative differences between groups

The number of glaucoma medications decreased in both groups over the 3-month follow-up period. At baseline, median medication use was similar between groups [BGI 4, 95% CI (3, 4) vs vsBGI 4, 95% CI (3, 4); p = 0.372]. By week 1, patients in the vsBGI group required significantly fewer medications compared to BGI [mean difference 0.97, 95% CI (0.65, 1.29), p < 0.001], and this difference persisted at month 1 [mean difference 0.56, 95% CI (0.30, 0.82), p < 0.001] and month 3 [mean difference 0.31, 95% CI (0.06, 0.57), p = 0.017]. No significant between-group difference in medication use was observed at day 1 [mean difference 1.94 medications, 95% CI (−0.15, 4.03), p = 0.069] (Fig. 2). Visual acuity (logMAR) did not differ significantly between groups at any time point, with all Holm-adjusted pairwise comparisons nonsignificant and between-group differences small in magnitude (absolute logMAR differences ≤0.20), indicating no clinically meaningful visual acuity differences (all p ≥ 0.081) (Fig. 3). CDR was also not significantly different between groups at baseline [mean difference 0.025, 95% CI (−0.076, 0.026), p = 0.342] or month 3 [mean difference 0.017, 95% CI (−0.070, 0.035), p = 0.514] (Fig. 4).

Fig. 2.

Number of glaucoma medications from baseline to month 3 postoperative for BGI and vsBGI. There was a significant difference in medication burden between BGI and vsBGI at month 3 (p < 0.05)

Fig. 3.

Estimated visual acuity in logMAR, showing no significant differences between the BGI and vsBGI groups out to month 3 postoperative

Fig. 4.

CDRs of BGI and vsBGI at baseline and month 3

In addition to model-based estimates, descriptive analyses of median values are shown in Figure 5 and include baseline measurements. These plots highlight the sharp early reduction in IOP in the vsBGI group compared to the BGI group. Median visual acuity and CDR remained largely stable and comparable between groups across follow-up. In contrast, median glaucoma medication use decreased in both groups but remained consistently lower in the vsBGI group from week 1 through month 3.

Fig. 5.

Median IOP, best-corrected visual acuity (logMAR), number of glaucoma medications, and CDR over time for the BGI and modified BGI groups. This figure presents unadjusted median values, including preoperative measurements, to illustrate overall clinical trajectories

Within 3 months postoperatively, there were 19 eyes (31%) in each group that experienced at least 1 adverse event (Fig. 6). There were no statistically significant differences between groups in the distribution of specific adverse events. Clinically significant hypotony occurred in 3 eyes (4.8%) in each of the two groups by month 3. However, the timing of when these events occurred differed between groups. In the vsBGI group, there was 1 case of clinically significant hypotony at week 1 and 2 cases at month 1, whereas in the BGI group, 1 case occurred at month 1 and 2 cases at month 3.

Fig. 6.

Types of adverse events and surgical failures in the BGI and vsBGI groups up to month 3 postoperative. There was no significant difference between the two groups

Hyphema was more common in the vsBGI group (9 eyes, 14.5%) compared to the BGI group (2 eyes, 3.2%), although this did not reach statistical significance (p = 0.054). Of the 9 hyphemas in the vsBGI group, 7 occurred on day 1, 1 at week 1, and 1 at month 1. In the BGI group, 1 case occurred on day 1, 2 at week 1, and 1 at month 1. Other complications were infrequent and not statistically significant between groups, including cystoid macular edema, corneal edema or keratoplasty worsening, diplopia, shallow anterior chamber, and choroidal hemorrhage. No instances of corneal blood staining, retinal detachment, tube erosion, or endophthalmitis occurred in either group (Table 2).

Table 2.

Adverse events within 3 months by surgery type

Complication	BGI	vs BGI	Test	p-value
Hyphema	2 (3.2%)	9 (14.5%)	Fisher	0.054
Cystoid macular edema	4 (6.5%)	4 (6.5%)	Fisher	1.000
Clinically significant hypotony	3 (4.8%)	3 (4.8%)	Fisher	1.000
Corneal edema or keratoplasty worsening	2 (3.2%)	4 (6.5%)	Fisher	0.680
Diplopia	1 (1.6%)	2 (3.2%)	Fisher	1.000
Reoperation	1 (1.6%)	2 (3.2%)	Fisher	1.000
Shallow anterior chamber	1 (1.6%)	1 (1.6%)	Fisher	1.000
Choroidal hemorrhage	0 (0.0%)	1 (1.6%)	Fisher	1.000
Corneal blood staining	0 (0.0%)	0 (0.0%)	Fisher	1.000
Endophthalmitis	0 (0.0%)	0 (0.0%)	Fisher	1.000
Retinal detachment	0 (0.0%)	0 (0.0%)	Fisher	1.000
Tube erosion	0 (0.0%)	0 (0.0%)	Fisher	1.000

The vsBGI group resulted in 2 instances of surgical reoperation compared to 1 in the BGI group. In the vsBGI group, 1 eye underwent reoperation due to occlusion of the tube by the iris, resulting in increased IOP. Another case required anterior segment revision due to a conjunctival leak leading to hypotony. This eye subsequently developed hemorrhagic choroidals and severe, debilitating pain, and the patient ultimately elected enucleation. The eye had a history of trauma-induced open-angle glaucoma and was only light-perception vision preoperatively.

There were 5 instances of ventilating suture removal in the vsBGI group. One eye required removal due to very low IOP (1 mm Hg), causing corneal striae and blurry vision. At week 3 follow-up, IOP increased to 12 mm Hg and was within the target range. In another case, the suture was preemptively removed at 6 weeks to accommodate tube opening and avoid hypotony. One week later, the IOP had increased from 9 to 34 mm Hg, necessitating restart of latanoprost and dorzolamide. A third case of removal was performed for hypotony (IOP of 1 mm Hg). The eye was Seidel positive prior to the suture being removed and negative afterward. IOP increased to 17 mm Hg 1 week later. Another case of suture removal was performed for shallowing of the anterior chamber and choroidal effusions despite an IOP of 12 mm Hg. Shallowing and effusions improved with the removal of the suture, but IOP increased to 28 mm Hg. Glaucoma medications and prophylactic atropine were started along with prednisolone. The final ventilating suture was removed at week 1 due to clinically significant hypotony (3 mm Hg) with shallow choroidal effusions. IOP increased to 31 mm Hg, and choroidals improved 2 weeks after removal of the suture.

Discussion

The vsBGI produced significant IOP reduction in the early postoperative period compared to the BGI. While overall adverse events, visual acuity, and CDR were comparable between groups, the vsBGI group demonstrated trends toward fewer medications. This suggests potential advantages in early pressure control compared to the BGI.

Perhaps the most consequential finding from this study is the ability of the ventilating mattress suture technique to provide early IOP reduction, with a significant difference seen at both day 1 and week 1 postoperative, despite similar preoperative medication burden (both median 3–4 medications). This significant decrease in early postoperative IOP is more impressive given that the vsBGI group had a significantly higher mean baseline IOP in its cohort. There were, however, no significant differences in IOP lowering at month 1 or month 3 postoperative. The ventilating suture modification explored in this study allows for early IOP lowering but tends to gradually loosen spontaneously before the ligation suture dissolves at 6 weeks postoperative and aqueous begins draining through the endplate. This may explain why we do not see a sustained pressure-lowering effect in comparison to the BGI at month 1 or month 3. Based on the significantly lower IOP in the early (day 1 and week 1) postoperative period, the ventilating suture modification for the BGI could be a potential option to provide immediate IOP reduction with the benefit of the long-term efficacy of the BGI. This suggests that the vsBGI may be particularly advantageous in eyes with markedly elevated IOP or advanced glaucoma, where prompt postoperative IOP reduction may be critical to limiting further visual deterioration. This approach may also be advantageous when continuation of medical therapy during the several weeks required for suture dissolution and full BGI function is not feasible, or when the glaucoma is refractory to medical treatment. However, the long-term effects (greater than 3 months) of the vsBGI remain to be determined, and a future study on this topic would be warranted.

Additionally, pooled analyses from the Ahmed Baerveldt Comparison Study and the Ahmed versus Baerveldt Study have suggested that the BGI achieves better IOP reduction and longer survival compared with the AGV.⁷ A comparative study of vsBGI and AGV with both short- and long-term follow-up would be useful to determine whether eyes could achieve prompt IOP lowering without sacrificing the advantages of the BGI.

In addition to decreased IOP, the vsBGI group also significantly lowered medication burden at month 3 postoperative compared to the BGI group. This finding suggests that the ventilating suture modification may provide earlier and more reliable IOP control, thereby reducing the need for adjunctive topical therapy in the early postoperative period. Reducing medication dependence is clinically meaningful, as it not only alleviates the cost and adherence challenges faced by patients with glaucoma but also minimizes the ocular surface toxicity and risk of medication tolerance associated with chronic drop use.¹⁹ It should be noted that medication reduction was clinician-dependent. Medications were held in the postoperative period and added back in a stepwise manner if IOP exceeded the intended goals. These results further support the superior efficacy of the vsBGI to the BGI in the short term.

Visual outcomes remained stable between groups, with no statistically significant differences in best-corrected visual acuity at any time point through month 3 postoperative. This finding indicates that the addition of a ventilating suture to the BGI does not adversely affect short-term visual recovery compared to the standard device. Similarly, there were no significant differences in CDR between groups at baseline or at month 3, suggesting that optic nerve cupping was comparable and did not change differentially with either surgical approach at month 3. Together, these results reinforce that the vsBGI modification provides early IOP-lowering benefits without compromising vision or optic nerve appearance in the early postoperative period.

In this study, the frequency of adverse events was similar between the vsBGI and the traditional BGI groups, with each reporting complications in approximately one-third of eyes. This finding suggests that the vsBGI does not substantially alter the overall risk profile of the procedure. Importantly, while the spectrum of adverse events was comparable across groups, certain differences in the timing and distribution of specific complications were observed.

Both groups demonstrated an equal incidence of clinically significant hypotony (4.8%), though the onset varied. In the vsBGI group, hypotony tended to occur earlier, with events clustered within month 1 postoperative, whereas in the BGI group, most cases were delayed, presenting between months 1 and 3. This difference in temporal distribution suggests that device modifications influence early postoperative fluid dynamics, even if the overall incidence of hypotony remains unchanged. Additionally, hyphema was more frequently observed in the vsBGI group compared to the BGI group (14.5 vs 6.5%). While this did not reach statistical significance, the trend may indicate subtle differences in surgical technique or device design that warrant attention.

Reoperation rates were infrequent in both groups, with 2 cases in the vsBGI cohort and 1 in the BGI cohort. Given the equal sample sizes of 62 eyes per group, this marginal difference is not of clinical or statistical significance (p = 1.000), thereby reinforcing the conclusion that the overall safety profiles of the 2 devices are largely comparable.

In this study, we found several instances of ventilating mattress suture removal, which led to the resolution of the hypotony caused by the modification in the immediate postoperative period. While nonvalved tube shunts do typically have lower IOP results than valved tube shunts, hypotony is still a possible complication for both types of tube shunts, and it is of great utility to have a modification that can be reversed at the slit lamp to allow for relatively less invasive and safe IOP elevation in situations where postoperative hypotony requires intervention.^20–22

It should be noted that the efficacy of the vsBGI may have led to surgeon bias in procedure selection. In cases with severely elevated IOP, the patients needing immediate IOP reduction may have preferentially received a vsBGI (as opposed to a traditional, nonmodified BGI). This may have led to the discrepancy in baseline IOP between the two groups (Fig. 2). Similarly, a 2024 study by Zhang and Qiu noted half of respondents stating a preference for valved tubes over nonvalved tubes in circumstances of higher baseline or preoperative IOP in order to lower postoperative IOP earlier.²³ Given that the vsBGI has shown clinical significance in lowering IOP in the postoperative period, our surgeons likely began to preferentially use the vsBGI on patients with very high IOPs as time went on, leading to issues of selection bias in our retrospective dataset. Future studies should utilize multivariable Cox regression to determine if the bias affects adverse events and surgical failure rates. A randomized prospective study design would also be helpful for eliminating potential bias.

The authors acknowledge the limitations of this study. The retrospective design limits the ability to control for confounding variables; there was a wide variety of comorbidities, glaucoma medication regimens, and glaucoma types, which could potentially influence the results of the study. Future studies should be conducted with multivariable Cox regression analysis to determine the extent of comorbidity effects. Additionally, this study followed patients out to month 3 only, due to the recent adoption of this surgical method. Future studies should investigate the long-term effects of this procedure and demonstrate whether the long-term efficacy of the vsBGI is maintained. As mentioned earlier, the vsBGI group had a significantly higher baseline IOP, likely attributable to surgeons increasingly selecting this device for patients with very high preoperative IOPs. Future studies employing a randomized prospective design will be necessary to clarify its impact on adverse events and surgical failure rates, in addition to better analyzing the safety and efficacy of the vsBGI in a controlled manner.

Conclusion

The reversible ventilating suture modification to the BGI tube shunt is a safe and effective method of allowing early aqueous humor flow and reducing IOP in the early postoperative period in comparison to an unmodified BGI.

Clinical Significance

The ventilating mattress suture modification to the BGI provides a means of achieving earlier postoperative IOP reduction while maintaining the long-term efficacy and safety profile of the traditional device. This feature is particularly important for patients presenting with severely elevated preoperative IOP and those with advanced stages of glaucoma, in whom immediate pressure control can be vision-saving. In addition to lowering IOP more rapidly, the modification reduced medication burden in the early postoperative period, which may translate into improved adherence, reduced cost, and decreased ocular surface toxicity for patients. Importantly, these benefits were achieved without compromising visual acuity, optic nerve appearance, or overall complication rates. The modification adds minimal procedural complexity and can be performed with only a few additional minutes added to the standard BGI operation. The ability to remove the ventilating suture in cases of hypotony further enhances its safety by offering a minimally invasive and reversible management option. Taken together, these findings suggest that the vsBGI is a clinically meaningful refinement of the BGI, offering earlier and more reliable pressure control in patients with advanced or refractory glaucoma while reducing the medication burden in the early postoperative period.

Author Contributions

Conception and design: A.E.P., Q.H.C., and K.L.D.

Data collection: Q.H.C., K.L.D., J.A.V., D.R.P., R.S.S., P.J.D., M.D.M., and H.M.A.

Analysis and interpretation: A.E.P., Q.H.C., K.L.D., D.R.P., and H.M.A.

Writing - original draft preparation: Q.H.C., K.L.D., and D.R.P.

Writing - review and editing: All authors.

Obtained funding: Not applicable.

Overall responsibility: A.E.P., Q.H.C., and K.L.D.

All authors have read and agreed to the published version of the manuscript.

Acknowledgment

The authors thank the Department of Ophthalmology and Visual Sciences at University of Iowa Health Care for their support.

AI Use Declaration

Use of Artificial Intelligence (ChatGPT, OpenAI) was limited to proofreading grammar and editing tone and clarity.

Orcid

Quinton H Christensen https://orcid.org/0009-0005-7435-9424

Kristin L Davis https://orcid.org/0009-0002-8501-4040

Joel A VandeLune https://orcid.org/0009-0009-1666-7380

Danielle R Pellack https://orcid.org/0000-0002-3322-3820

Ronak S Singh https://orcid.org/0009-0004-5016-9610

Patrick J Donegan https://orcid.org/0000-0002-8368-6743

Matthew D Meyer https://orcid.org/0009-0006-5765-6387

Hend M Al-Kaylani https://orcid.org/0000-0001-5188-5521

Erin A Boese https://orcid.org/0000-0003-1124-3991

Andrew E Pouw https://orcid.org/0000-0001-8594-7612

Data Availability

The datasets generated during the current study are available from the corresponding author on reasonable request.