figure a

Introduction

Glaucoma is one of the leading causes for irreversible visual impairment worldwide. It is a progressive optic neuropathy associated with characteristic functional visual field loss and structural optic disk damage [1]. An estimated 76 million people worldwide are living with the disease in 2020 and the number is anticipated to increase to 111.8 million by 2040 [2]. Generally, glaucoma is classified based on the appearance of the draining apparatus (anterior chamber angle) into open-angle and angle closure glaucoma with further subclassification of primary and secondary causes. The most common type is primary open-angle glaucoma (POAG), which accounts for 74% of all cases reported [3]. However, the true number of people with POAG still remains uncertain due to geographical variations in underlying POAG prevalence and geographical representation of surveys to ascertain prevalence as well as differences in study methods of case ascertainment [3].

Glaucoma is multifactorial in origin; however, the only modifiable risk factor is the elevated intraocular pressure (IOP) [4]. It has been shown that lowering IOP by 20–40% can reduce the rate of progressive visual field loss by 50% [5, 6]. Thus, treatments in glaucoma are mainly focused on lowering IOP. The clinical options for reducing IOP are topical/systemic drugs—aiming to increase aqueous humor outflow or to decrease its production—laser therapy that targets the trabecular meshwork and ciliary body, and incisional surgery[7]. The mode and level of treatment are determined by several factors, including glaucoma subtype, estimated life expectancy, prognosis of visual acuity/field, and ocular and systemic comorbidities [8]. In case of open-angle glaucoma (OAG), trabeculectomy is considered to be the standard procedure for lowering IOP [9,10,11,12]. This technique creates an alternative route for the drainage of aqueous humor from the anterior chamber to the subconjunctival space. The long-term success rate has varied from 35 to 97%, depending on the follow-up time, the criteria used to define successful outcome, and the population studied [13,14,15,16,17,18,19]. Despite its effectiveness, trabeculectomy is frequently accompanied by short- and long-term complications such as hypotony [20, 21], bleb leaks [22,23,24,25], accelerated cataract progression [26], choroidal effusion and hemorrhaging [27], and prolonged or permanent visual impairment due to hypotonic maculopathy [28]. Considering this fact, new novel devices and surgical techniques, described as minimally invasive glaucoma surgery (MIGS), have been developed. In this regard, our department of ophthalmology has already gained valuable experience with iStent inject®, Trabectome®, and canaloplasty ab interno (ABiC) [29,30,31]. XEN® gel (Allergan Inc., CA, USA) implantation is a further, recently developed alternative filtrating glaucoma surgical procedure. The device is a 6-mm hydrophilic tube [32] composed of collagen-derived gelatin cross-linked with glutaraldehyde, which is implanted ab interno through the anterior chamber into the irido-corneal angle to optimize aqueous drainage to the subconjunctival space [33]. Compared to traditional filtering surgeries, XEN® has the potential benefits of increased safety and faster visual recovery due to its minimally invasive approach [34]. XEN® gel implant received a CE mark in 2013 and was approved by the FDA in 2016. The surgery can be performed alone or in combination with a cataract surgery. Previous studies have highlighted the efficacy and safety of the XEN® gel implantation in decreasing the IOP and medication burden [35,36,37]. The mean IOP reduction at the 1-year follow-up ranges from 22.7 to 60.4% [38].

In this study, we aim to evaluate the outcome after XEN® gel implantation with respect to the IOP as well as the number of glaucoma medication—and if this outcome can be preserved over a longer period of time. Subsequently, the influence of the type of anesthesia, prior MIGS and postoperative interventions on the outcome, is analyzed.

Material and methods

This retrospective study is based on the data of 96 eyes of 67 patients (mean age 69.9 ± 12.4 years, range 37–87 years, 47 men, 49 women) with the diagnosis of POAG receiving a XEN® gel implantation in the time period from November 2017 to January 2019. The data were collected in the MIGS database of the department of ophthalmology of the Charité—University Medicine Berlin. The study complied with the ethical standards of the Declaration of Helsinki and followed the regulations of the Good Clinical Practice (GCP) guidelines. The ethics committee of the Charité—University Medicine Berlin has given its approval (EA4/047/20).

Subjects

Only patients suffering from POAG were included in this study. Consequently, all patients had a sufficiently visible trabecular meshwork, detected by gonioscopy (Shaffer III–IV). Further inclusion criteria were a best-corrected visual acuity of at least 0.1 LogMAR and informed patient consent for surgery. The indication for glaucoma surgery complied with the guidelines of the European glaucoma society (EGS, 5th edition, 2021). The main indication was an IOP above the target pressure, following already exhausted local therapy or it could not be intensified due to allergies. The severity of the glaucoma was determined using the cup to disk ratio (CDR), visual field examination (static perimetry; Aulhorn–Karmeyer staging), and measurement of the retinal nerve fiber layer using optical coherence tomography (average thickness in μm; Spectralis OCT, Heidelberg Engineering, Germany). In accordance with the EGS guidelines, the target pressure was defined individually, depending on the severity of the POAG. In most cases, a target pressure below 16 mmHg was ideal.

The exclusion criteria entailed active inflammation in the anterior as well as posterior chamber or corneal infections. Additional criteria for exclusion were also considered in patients who underwent intraocular surgery less than three months before XEN® gel implantation.

Basic data, such as age, gender, glaucoma therapy, best-corrected visual acuity (BCVA), cup to disk ratio (CDR), and lens status, was collected and analyzed. Furthermore, previous laser and glaucoma surgery like selective laser trabeculoplasty (SLT), Trabectome®, or iStent inject® were recorded.

With regard to surgery itself, it was recorded whether the intervention was performed under local anesthesia (peribulbar or local anesthesia) or general anesthesia (intubation anesthesia). The number of glaucoma therapy and the IOP level, recorded by using the Goldmann applanation tonometry, was measured before the surgery and compared with the results after 1 day, 6 weeks (range = 35–49 days), 3 months (range = 77–106 days), 6 months (range = 161–204 days), 12 months (range = 334–387 days), 24 months (range = 670–788 days), and 36 months (range = 1042–1130 days).

As a matter of course, peri- and postoperative complications (postoperative hypotony, IOP spikes, bleeding complications, choroidal effusion or hemorrhage, conjunctival perforation, stent dislocation, infection), postoperative interventions (bleb needling or revisions), and secondary glaucoma surgery (trabeculectomy, Ahmed valve, cyclophotocoagulation) were documented.

Surgical technique

After informed consent was provided, the surgery was performed by one of two experienced surgeons using the same surgical protocol under local or general anesthesia.

The XEN® gel implantation was performed as a stand-alone procedure. First Suprarenin® (synthetic epinephrine, Sanofi Germany GmbH) was used at the side of XEN® gel implantation to reduce bleeding. Then, a 3-mm limbus distance was marked in this area. After this, 0.1 ml mitomycin C was injected in a concentration of 0.1 mg/ml with a 30-gauge needle and massaged peripheral. Following a 1.8-mm micro-incision, an ophthalmic viscosurgical device was injected into the anterior chamber for stabilization. The XEN® gel microstent is a 6-mm hydrophilic tube [32], which is implanted through the anterior chamber under gonioscopic view into the irido-corneal angle to optimize aqueous drainage to the subconjunctival space. Optimal positioning of the XEN® gel microstent is defined as 1 mm in the anterior chamber, 2 mm in the sclera, and 3 mm below the tenon. After implantation, the position of the XEN® gel microstent was controlled and a primary needling was conducted if the part of the XEN® gel microstent below the tenon was not freely movable. Subsequently, the ophthalmic viscosurgical device was washed out with a balanced salt solution. During this procedure, the filling of the bleb was observed.

Postoperative treatment included a topical combination of steroids and antibiotics that were reduced over 4 to 6 weeks following the surgery.

Data analysis

The Shapiro–Wilk test and the Levene test were used to evaluate the normal distribution of the variables and the equality of variances. The paired sample t-test was performed to compare normally distributed data. In case of not normally distributed variables, non-parametric tests were used: the Wilcoxon test for two dependent variables, the Mann–Whitney-U test for two independent variables, the Friedmann test for multiple dependent variables, and the Kruskall–Wallis test for multiple independent variables. As post hoc test, the Bonferroni test was performed. P-values lower than 0.05 were determined to be significant. In case of missing data, a pairwise deletion was made. Furthermore, the success rate of the procedure was assessed with the help of the Kaplan–Meier analysis. According to the World Glaucoma Association (WGA) criteria, qualified success was defined as IOP ≤ 18 mmHg and ≥ 20% reduction in IOP with or without medication compared with baseline values. If the target criteria were not reached on two consecutive visits or additional glaucoma surgery (e.g., trabeculectomy, aqueous shunt, cyclophotocoagulation) was necessary, then the surgery is defined as failure [39]. The data analysis was performed with the program SPSS Statistics (Version 23.0 for Mac).

Results

A total of 96 eyes from 67 patients (mean age 69.9 ± 12.4 years with a range of 37–87 years; 47 men, 49 women) were included in this study. The mean BCVA was 0.25 ± 0.27 LogMAR (range: 0.1–1.3) and the mean CDR was 0.83 ± 0.16 (range: 0.3–0.95). On the Aulhorn–Karmeyer scale, an average value of 2.83 ± 1.4 (range: 1–4) was achieved, whereas the average thickness of the retinal nerve fiber layer was 62.25 ± 13.9 μm (healthy RNFL: > 90 μm).

XEN® gel implantation was performed on phakic eyes in 23 cases of the cohort. A total of 50 eyes had already received minimally invasive glaucoma surgery in the past (SLT: n = 18, Trabectome®: n = 14, iStent inject®: n = 18). General anesthesia was performed on 86 eyes (89.6%) and local anesthesia on 10 eyes (10.4%). After 36 months, 36 eyes were still available for analysis (n = 79 after 12 months, n = 52 after 24 months). The baseline data of the study cohort is summarized in Table 1.

Table 1 Baseline descriptive data of the study cohort
Full size table

Development of intraocular pressure and glaucoma therapy

IOP was reduced from preoperative values of 20.7 ± 5.1 to 12.6 ± 2.4 mmHg at the 12 months postoperative follow-up, 12.5 ± 2.2 mmHg at 24 months, and 12.8 ± 2.5 mmHg at 36 months. In Fig. 1, the mean pre- and postoperative IOP level in total is illustrated. It shows a statistically significant decrease of the IOP level during the whole follow-up period (1 day p < 0.001; 6 months p < 0.001; 12 months p < 0.001; 24 months p < 0.001; 36 months p < 0.001). The scatterplot in Fig. 2 demonstrates the relation between the preoperative IOP and the IOP 36 months after surgery.

Fig. 1
figure 1

Change in intraocular pressure (IOP) as a timeline graph after XEN® gel implantation, *p < 0.001

Full size image
Fig. 2
figure 2

Scatterplot: each point represents one eye showing the preoperative IOP value on the abscissa, and the postoperative IOP values on the ordinate, respectively. The oblique line indicates no change, weather the diagonal lines illustrate the percentage reductions (20%, 30%, and 50%). Different IOP targets (21, 18, and 16 mmHg) are represented by the horizontal lines

Full size image

Using the definition of success as stated previously, the Kaplan–Meier graph shows an estimated survival rate of 79.2% after 12 months and 74% in total after 36 months (Figs. 3 and 4). The type of anesthesia had a significant influence on the qualified success (p = 0.044; after 12 months: p = 0.0043). Intubation anesthesia resulted in a significant improvement of the survival rate (77%; after 12 months: 81%) compared to local anesthesia (50%; after 12 months: 60%).

Fig. 3
figure 3

Kaplan–Meier graph for the probability of qualified success. a All cases in total. b intubation anesthesia: solid, black line; local anesthesia: dotted, black line. The endpoints were defined according to the following success criteria: IOP ≤ 18 mmHg and ≥ 20% reduction in IOP compared with baseline values. Failure was assumed if the target criteria were not reached on 2 consecutive visits. General anesthesia has a significant effect on the qualified success (p = 0.044, Breslow (generalized Wilcoxon))

Full size image
Fig. 4
figure 4

Kaplan–Meier graph for probability of qualified success. a All cases in total 12 months following surgery. b All cases with complete follow-up until 36 months postoperatively. The endpoints were defined according to the following success criteria: IOP ≤ 18 mmHg and ≥ 20% reduction in IOP compared with baseline values. Failure was assumed if the target criteria were not reached on 2 consecutive visits

Full size image

Lens status and preoperative laser such as SLT (p ≥ 0.05) or glaucoma surgical interventions such iStent® inject and Trabectome® had no significant influence on the qualified success 12 and 36 months after surgery (respectively p ≥ 0.05). However, iStent® inject tended to have an unfavorable influence with a survival rate of 67% (Trabectome®: 79%, SLT: 72%) at the 36-month follow-up.

In order to support the validity of the entire data despite the high loss to follow-up after 36 months, the data of the patient cohort who participated in all follow-up visits over the course of all 3 years (complete follow-up cohort) was also analyzed in addition to the evaluation of the 1-year data. The corresponding data showed that the results of the entire study cohort did not differ significantly from the results of the complete follow-up cohort after 36 months. The overall survival rate was 74.4%. Intubation anesthesia had a significant positive influence on the survival rate (p = 0.048), whereas preceded iStent® inject tended to have an unfavorable influence with a survival rate of 62.5% compared to Trabectome® with a survival rate of 80% at the 36-month follow-up.

The number of glaucoma medication changed over time and is illustrated in Fig. 5. Compared to the preoperative value of 3.3 ± 0.8, it was significantly reduced at each examination during the follow-up period. After 12 months, the number of glaucoma medication was 0.6 ± 1.1, after 24 months 0.8 ± 1.3 and after 36 months 1.2 ± 1.6 (1 day p < 0.001; 6 months p < 0.001; 12 months p < 0.001; 24 months p < 0.001; 36 months p < 0.001).

Fig. 5
figure 5

Change in glaucoma medication as a time line graph after XEN® gel implantation, *p < 0.001

Full size image

Complications

The most common complication was a transient postoperative hypotony with an IOP < 6 mmHg, which occurred in 26 eyes (27.1%). In case of two subjects (2.1%), Healon® was placed in the anterior chamber for stabilization. Six weeks postoperatively, only two eyes had an IOP of 5 mmHg and at the 3-month follow-up, all eyes had an IOP of 6 mmHg or higher. IOP spikes or bleeding complications such as a significant intraoperative bleeding or postoperative hyphema were not observed. The same applied to choroidal effusion or hemorrhage. The conjunctiva was perforated by the XEN® gel implant in three cases (3.1%), two of whom needed an oral mucosal transplantation. In the further course, one subject developed an endophthalmitis, which was successfully treated with intravitreal injections of antibiotics. Three more subjects (3.1%) experienced a dislocation of the XEN® gel implant into the anterior chamber (Tables 2 and 3).

Table 2 Drop out data of the study cohort
Full size table
Table 3 Data of peri- and postoperative complications and subsequent glaucoma interventions
Full size table

Postoperative interventions

In order to support the development of an effective filter bleb or to maintain its functionality, needlings or bleb revisions were carried out in 58 eyes (60.4%) in the further postoperative period. In the case of a needling, adhesions are loosened with a 30-gauge needle or a bleb knive. As part of a filter bleb revision, the conjunctiva was opened to such an extent that a suture was necessary afterwards. Needling was performed in 53 (55.2%) eyes (once: n = 24, twice: n = 20, three times: n = 6, four times: n = 3), a bleb revision was performed in 18 (18.8%) eyes. Table 3 provides an overview of the complications and postoperative interventions.

With the aid of a Kaplan–Maier analysis, we analyzed whether the postoperative IOP, a filter bleb revision and the number of needlings had an influence on the qualified success of an XEN® gel implantation. With regard to the postoperative IOP, it seemed to be that a lower IOP tends to be more favorable regarding the survival rate (81% vs. 71%; after 12 months: 81% vs. 79%; complete follow-up cohort: 83% vs 70%). However, this effect was not significant (respectively p > 0.05). A filter bleb revision had no effect (72% vs. 74%; after 12 months: 72% vs 81%; complete follow-up cohort: 71% vs 75%; respectively p > 0.05) in this study cohort. The number of postoperative needlings, however, showed a significantly negative influence on the qualified success (p = 0.012). The more often a needling had to be performed, the lower the survival rate was (no needling: 88%, one needling: 71%, two needlings: 60%, three needlings: 50%, four needlings: 33%). Comparable results were obtained at the 12-month follow-up (no needling: 93%, one needling: 71%, two needlings: 70%, three needlings: 67%, four needlings: 33%; p = 0.018) and with the complete follow-up cohort at the 36-month follow-up (no needling: 86.7%, one needling: 78%, two needlings: 67%, three needlings: 67%, four needlings: 33%; p = 0.044). With the aim of maintaining clarity and comprehensibility, only the results of the entire study cohort are listed below or in the discussion.

Drop out

To date, a subsequent glaucoma surgery was necessary for 8 subjects because the target pressure was not reached or could not be met in the follow-up period (failure rate: 8.3%, mean duration: 10.9 months; range: 3–24 months). In three cases, a trabeculectomy was performed, whereas in one case both a cyclophotocoagulation and an implantation of an Ahmed Valve was carried out. A second XEN® gel implantation was attempted in a further 3 cases. Table 2 displays the characteristics of the dropout cases cohort.

Discussion

The XEN® gel microstent is an ab interno approach without conjunctival incision, aiming to provide a safer and less invasive IOP-lowering surgery in glaucoma patients compared to traditional filtrating surgery.

In the present study, we observed a mean IOP reduction of 40% 36 months after surgery. The number of medications at the 36-month follow-up was reduced by 62% compared to the initial respectively baseline value. Our results 3 years following XEN® gel implantation meet the corresponding data published so far. Furthermore, the results emphasize the conclusion that XEN® gel implantation is an effective procedure for the long-term in lowering the IOP and reducing glaucoma medication. Previous studies on the XEN® gel stent have shown promising results in reducing IOP and medication primarily for patients with primary open-angle glaucoma in a period of up to 4 years. In a follow-up of 8.5 months, Widder et al. reported a 30% reduction of IOP and a decrease of 92% of glaucoma medication [37]. Several studies published resulted with a follow- up of 12 months, with IOP and medication reduction from 22 to 54% and from 37 to 96% versus the baseline, respectively [35, 40,41,42,43,44,45]. In addition, S. Grover et al. [46] reported of 75% of patients with ≥ 20% IOP lowering in combination with 51% reduction of glaucoma medication. Kalina et al. [47] reported of 68% of patients with ≥ 20% IOP lowering from baseline without medication or any secondary glaucoma intervention. Moreover, the gel implant was effective over 2 years, reducing both IOP and medication needs [48,49,50], while during the first 3 years, IOP changes between 22 and 19.5% were reported, respectively [51]. After a 4-year follow-up, Lenzhofer et al. showed a postoperative drop in IOP of about 40% [52].

Analyzing to what extent the XEN® gel implantation can be used as part in a gradual therapeutically scheme of glaucoma surgery (e.g., following an iStent® inject implantation), we investigated the influence of previous minimally invasive glaucoma surgical procedures, typically performed at an early stage in glaucoma disease, on the effectiveness of XEN® gel implantation—which is usually carried out in more advanced glaucoma cases. We could not observe any significant influence of preoperative, minimally invasive interventions on the outcome of the XEN® gel implantation. Preoperative glaucoma interventions such as selective laser trabeculoplasty (SLT) laser and MIGS (iStent® inject and Trabectome®) had no significant influence on the qualified success. However, iStent® inject tended to have an unfavorable influence on the qualified success.

In their publication, Lewczuk et al. retrospectively examined the influence of various glaucoma surgical interventions on the outcome two years after XEN® gel implantation [53]. In contrast to our study, however, eyes with different primary and secondary glaucoma diagnoses were included and prior procedures involved mainly trabeculectomy (69%), sclerectomy (55%), and cyclodestructive procedures (41%). The Kaplan–Meier cumulative incidence rate of qualified success (defined as 20% decrease in IOP or IOP 18 mmHg with up to two glaucoma medications) after 24 months was 76.5% for the naïve group and 68.7% for the group with prior glaucoma surgeries. However, there was no significant difference in qualified success between both groups. Although it is not truly comparable to ours, the study of Lewczuk et al. yet indicates that the surgical success rate after XEN® gel implantation as an initial procedure did not differ from that in patients with previous glaucoma surgeries [53]. Consequently, XEN® gel implantation can be considered as an alternative approach in cases of failed previous laser or glaucoma surgical interventions if the conjunctiva scarring is minimal.

With regard to previous MIGS, a publication about the impact of prior Trabectome® surgery on the outcome of XEN® by Kiessling et al. provides a first more differentiated analysis [54]. Even though they included eyes with exfoliative glaucoma in addition to POAG, this study is more comparable to ours. The mean follow-up was 22 ± 12 months and the mean IOP reduction 38% with a postoperative IOP of 14.5 ± 5.0 mmHg and a postoperative medication score of 0.4 ± 0.8. In total, 42% of their patients underwent an open conjunctival revision surgery and 12% needed a repeated surgery. As conclusion, Kiessling et al. stated that a failed Trabectome® surgery does not affect the postoperative IOP, medication score, revision rate, repeated surgery rate, and success rate of a subsequent XEN® gel implantation [54].

In case of this study, the XEN® gel implantation was carried out as a stand-alone procedure, either in pseudophakic or phakic patients. The lens status did not have a significant influence on the survival rate. In other literature, however, pseudophakic eyes showed a better primary success rate compared to phakic eyes in the XEN® alone group and both in XEN® alone group compared to the combined surgery [37, 49, 55].

Among the documented complications of this study, postoperative ocular hypotension was the most common, with a necessary treatment only being needed in few cases. In general, however, postoperative ocular hypotension seemed to have a favorable, but not significant, effect on the success rate.

Furthermore, documented complications after XEN® implant included XEN dislocation into the anterior chamber and perforation of the conjunctiva. As also found in other literature, a low rate of intra- or postoperative complications or unexpected postoperative ocular adverse events were reported using XEN® gel stent minimal surgery [37, 46, 56], not considering needling of a scarring filtering bleb as a complication. Galal et al. [40] reported one case of a dislocated stent; the situation was managed by applying conjunctival sutures and relocation of the implant in the subconjunctival space. Widder et al. estimated that 34% of patients that could not achieve an adequate IOP without postoperative medication required revisional surgery defined as removal of scar tissue [37].

In case of scarring, depending on the extent, needling and open revision of the filtering bleb have been established as an effective part of the follow-up treatment. In our study cohort, 60.4% of the eyes required at least one further intervention in the area of the filtering bleb in order to maintain the target pressure. A higher needling rate, resulted in a lower success rate, whereas filter pad revisions had no negative effect. Our results are in line with the findings in literature. Steiner et al. published an interesting analysis regarding needling and open filtering bleb revision after XEN® gel implantation [57]. Comparable to our study, they reported as well transient hypotension in 30.2% of the eyes after primary surgery, which lasted longer than 30 days in 4 patients. In addition, a total of 49.6% of the eyes required at least one postoperative secondary intervention. In summary, their data appear to indicate that open filtering bleb revision might have beneficial outcomes in terms of Kaplan–Meier success rates, and the reintervention rate compared to the needling procedure [57].

The effect of local anesthetic regimens on the increase of IOP—being that local anesthesia is the commonly chosen procedure for glaucoma surgeries—is still part of recent research [58, 59]. Studies showed IOP values did not differ between peribulbar and retrobulbar anesthesia [59] or between topical and retrobulbar [60], and no significant differences were found in terms of pain or supplemental anesthesia requirement under topical or retrobulbar anesthesia [60, 61]. In case of this study, we compared the effect of general and local anesthesia on the outcome. As result, there was a significant and positive influence of general anesthesia on the survival rate (77% vs. 50%, p = 0.44), which corresponds to the available literature. Gedar et al. found that general anesthesia may have advantages for glaucoma surgery in terms of (i) relaxation of the patient; (ii) more control of the IOP; and (iii) fixation of the operated eye [62].

Our study has a number of limitations which have to be addressed. The main study limitation is the retrospective design including real-life clinical data with the data variability across the different follow-ups. This may increase a possible bias of the study. The effects of the Covid-19 pandemic are also not negligible. Some patients failed to attend the follow-up visits, due to national health system rules to reduce the risk of infection. Furthermore, due to emergency surgeries, non-urgent appointments were canceled and the further postoperative control was delegated to the outpatient departments. In line with the above mentioned, one more limitation is the small number of patients obtained who received local anesthesia.

To conclude, the XEN® gel stent appears to be effective for reducing IOP and the number of glaucoma medications in POAG despite prior MIGS. Even so, controlled follow-up and evaluation of the indication for subsequent interventions such as needling and open filtering bleb revision are advised after surgery [55]. It offers an effective and safe approach as a sole procedure in POAG in the context of a gradual therapeutically scheme of glaucoma surgery.