Comparison of the effectiveness of the XEN microstent in the treatment of different types of glaucoma

Purpose The aim of this study was to compare the e�cacy and safety pro�le of the XEN-microstent in different types of primary and secondary open angle glaucoma. Methods The XEN microstent was implanted in patients with primary open-angle glaucoma (POAG), normal-tension glaucoma (NTG), pseudoexfoliation glaucoma (PEX) and secondary glaucoma (Sec.Gl). The intraocular pressure (IOP), the active substances of the applied IOP-lowering drugs, the best corrected visual acuity (BCVA) and the mean deviation (MD) of the perimetry were measured at baseline and at regular follow-ups, scheduled at 2 days and 1, 3, 6 and 12 months after surgery


Background
Glaucoma is one of the leading causes of blindness worldwide, with an estimated prevalence of 2-4% in the 40 years and older age group [1,2].Glaucoma is a very heterogeneous group of diseases characterized by progressive atrophy of the optic disc caused by apoptotic retinal ganglion cell death [3,4].Known risk factors for the development of glaucoma are an elevated intraocular pressure (IOP), familial history of glaucoma, myopia and thin central corneal thickness [5][6][7].Reduction of IOP by means of medication and/or surgery is the only known way to effectively slow down further disease progression [8,9].Usually, IOP reduction by means of medication is the rst-line treatment in glaucoma therapy.Surgical intervention is indicated in cases of insu cient medical IOP reduction, problems with side effects, insu cient adherence and persistence to therapy and further progression of optic nerve damage.[10].Trabeculectomy (TE) with or without the use of antimetabolites was rst introduced in the mid-1960s and has since remained the "gold standard" for longterm surgical IOP reduction [11,12].However, TE has numerous risks and unintended side effects that limit its use to advanced disease stages [13][14][15][16].In recent years, various techniques have been developed to minimize surgical trauma and the risk of serious side effects.These newly developed techniques make use of different mechanisms to reduce the resistance of aqueous humor out ow, thereby reducing IOP.These new techniques are summarized under the term minimally invasive glaucoma surgery (MIGS) [17].
A promising MIGS procedure is the XEN microstent.The XEN microstent is approved for the treatment of primary open-angle glaucoma (POAG).The XEN microstent facilitates subconjunctival drainage of aqueous humor from the anterior chamber, which is comparable to the mode of action utilized in TE [18].The XEN microstent does not comprise a valve mechanism, but maintains a minimum IOP of approximately 8 mmHg through its length and internal lumen [19].
However, subconjunctival drainage of aqueous humor is suitable for more than one type of glaucoma (i.e.POAG).Therefore, the concept of this study was to analyze the IOP-lowering e cacy of the XEN microstent in the treatment of other types of glaucoma.Other glaucoma entities suitable for XEN implantation were pseudoexfoliation glaucoma (PEX), secondary open-angle glaucoma of various causes (Sec.Gl) and normal tension glaucoma (NTG).
The main objective of the investigations was to analyze whether or not the XEN implant is also suitable for the treatment of these types of glaucoma.

Methods
For this retrospective analysis, we reviewed the medical records of all patients/eyes scheduled for XEN microstent implantation from January 1, 2017 to December 31, 2018 at the University Eye Hospital in Leipzig.
The study was approved by the local ethics committee of the University Leipzig, Leipzig, Germany.Written informed consent was obtained from all patients.All procedures performed met the ethical standards of the institutional research commission as well as the Helsinki Declaration of 1964 and its later amendments.The trial was registered with the DRKS (registration number: DRKS00020800, Registered 25.February 2020 -Retrospectively registered,https://www.drks.de/drks_web/navigate.do?navigationId=trial.HTML&TRIAL_ID=DRKS00020800).
Presence of optic disc changes such as an increased cup/disc ratio, optic disc hemorrhages, nerve ber layer defects and/or indentations of the papillary vessels veri ed glaucoma diagnosis.Untreated IOP had to be ³21 mmHg in POAG, PEX and secondary glaucoma cases.In NTG IOP was by de nition lower than 21 mmHg.In the analyzed cases surgical intervention was necessary due to disease progression or medically uncontrollable IOP.This was demonstrated by deterioration of repeated automated standard perimetry results and/or reduction of RNFL thickness on repeated optical coherence tomography (RNFL-OCT) examinations.Topical IOP-lowering therapy was discontinued four weeks prior to surgery in eyes scheduled for XEN implantation and preservative-free steroid eye drops were administered four times daily to achieve an irritation free conjunctiva.Exclusion criteria were a patient age below 40 years and a narrow anterior chamber angle.If both eyes of a single patient required surgical treatment for glaucoma, only the results of the rst-operated eye were included into this analysis.
Four weeks prior to surgery, a complete ophthalmological examination was performed with maximum tolerable anti-glaucomatous local medication, which included measurement of best-corrected visual acuity (BCVA) using Snellen charts (transformed into logMAR for statistical analysis), objective refraction, slit-lamp examination of anterior and posterior eye segments with evaluation of the optic disc by indirect ophthalmoscopy and RNFL-OCT scans; Spectralis, Heidelberg Engineering, Heidelberg, Germany), automated standard perimetry (Twin eld 2, Oculus, Wetzlar, Germany; 24-2 test strategy, 55 targets), IOP measurement with Goldmann applanation tonometry and gonioscopy with the Sussmann four-mirror contact lens.The indication for the planned surgical procedure was re-evaluated, whereby the aim was to lower IOP and/or to reduce the number of necessary IOP-lowering agents.The demographic patient data collected prior to surgery were age, sex and laterality of the operated eye.

Surgical Technique:
Ocular surface was disinfected using povidone iodine.Two side port paracenteses were laid and the eyes anterior chamber was lled with a dispersive viscoelastic agent.In the superior nasal quadrant a mark 3 mm posterior to the limbus was made.0.1 ml of 0.01% Mitomycin C (MMC) was administered 6 mm from the limbus through the conjunctiva and into the Tenon´s capsule to form a bleb.The MMC was then massaged posteriorly away from the corneal limbus.The XEN microstent was then inserted into the eyes anterior chamber mounted in its applicator.The applicator´s tip was aimed at the chamber angle and pierced through the chamber angle anterior to the trabecular meshwork and was directed through the sclera forming a scleral tunnel of approximately 3 mm in length.The outer ori ce was targeted so as to lie 3 mm behind the limbus without piercing the overlying conjunctiva.The XEN stent was then released from the applicator.If performed correctly approximately 1 mm of the microstent lay in the anterior chamber, 3 mm in the formed intrascleral tunnel and 2 mm outside the tunnel underneath the conjunctiva.
The subconjunctival portion of the XEN stent was then examined and freed from Tenon´s capsule if bent (or not moving freely) using a 30G needle.
Postoperative visits were scheduled at day 1 and 2 as well as at month 1, 3 and 6 and 12 after surgery.All local and oral antiglaucoma medications were discontinued.Preservative-free steroid eye drops were given 6 times daily for 4 weeks and then slowly reduced over 3 months.Additionally preservative-free antibiotic eye drops were prescribed four times daily for one week.At each visit BCVA and IOP were measured and the anterior and posterior eye segments were evaluated for adverse events (conjunctival scarring, at bleb, shallow anterior chamber, choroidal detachment, hypotony maculopathy etc.).The number of glaucoma medications taken was queried.In addition, visual eld examinations were performed during the visits 6 and 12 months after the surgery.
For additional necessary post-surgical needling procedures a 27G cannula was used on a syringe with 5-FU (5 mg in 0.1 ml).The needle was inserted under the conjunctiva 3 mm temporal from the XEN stent and needling was performed by sweeping the needle tip posteriorly above and beneath the XEN microstent.Finally 0.1 ml of 5-FU was injected posterior to the end of the XEN microstent.
Clinical success was de ned according to the recommendations of the World Glaucoma Association [20].For complete success (A) an IOP reduction of ≥20% from baseline and resulting IOP of <21 mmHg without the use of IOP-lowering drugs had to be achieved.For quali ed success (A) the same criteria applied as for complete success A with the use of additional IOP-lowering drugs permitted.For complete success (B) an IOP reduction of ≥30% from baseline and resulting IOP of <18 mmHg without the use of IOP-lowering drugs had to be achieved.For quali ed success (B) the same criteria applied as for complete success B with the use of additional IOP-lowering drugs permitted.For complete success (C) an IOP reduction of ≥40% from baseline and resulting IOP of <15 mmHg without the use of IOP-lowering drugs had to be achieved.For quali ed success (C) the same criteria applied as for complete success C with the use of additional IOP-lowering drugs permitted.
Data collection and statistical analysis were performed using Excel (Version 2007, Microsoft; Redmond, USA) and SPSS (IBM Version 22.0; Chicago, Illinois, USA).The tested indices for patient age, BCVA, objective refraction and visual eld are given as mean and standard deviation.The differences between the pre-and postoperative results for IOP, IOP reduction, BCVA, number of active IOP-lowering agents applied and the mean defect of static automated perimetry were analyzed using the Kruskal-Wallis nonparametric test or one-way ANOVA (where applicable) for intergroup comparisons and the Friedman test for intragroup comparisons of results measured in groups of different glaucoma types.A p ≤ 0.05 indicated statistical signi cance.

Results
For this analysis the clinical course of 153 eyes of 153 patients (80 female and 73 male; 77 right and 76 left eyes) were included.Four groups of eyes suffering from different types of glaucoma were treated.113 eyes (74%) were treated for POAG, 5 eyes (3%) for NTG, 13 eyes (9%) for secondary glaucoma and 22 eyes (14%) for pseudoexfoliation glaucoma.Of the 13 secondary glaucoma cases, four were due to previous eye trauma, four were due to neovascularization (2 cases of proliferative diabetic eye disease and 2 cases of central retinal vein occlusion, the underlying disease being considered in a stable phase at the time of XEN implantation), three cases were secondary to uveitis and the last two cases were due to steroid-induced glaucoma.In the complete patient group the mean age of patients at the time of surgery was 70.2±10.8years.The difference in mean patient age was not statistically signi cant between the four treated patient groups (p=0.24).Before surgery the mean IOP of the 153 eyes was 23.9±7.4mmHg in the 153 eyes.Upon comparison of the mean IOP before surgery between the four treated groups the difference was statistically signi cant (p<0.01).The mean number of active substances contained in the applied IOP-lowering eye drops was 2.6±1.2before surgery, and the comparison between the four treated groups showed no statistically signi cant difference (p=0.56).The mean defect tested by standard automated perimetry was 10.1±4.3 dB before surgery, and the comparison between the four treated groups showed no statistically signi cant difference (p=0.96).The exact demographic data are also presented in Table 1.

IOP results
In the complete group of eyes analyzed mean IOP before surgery was 23.9±7.4mmHg.Two days after XEN implantation, mean IOP decreased to 9.2±5.2mmHg and then increased again to 15.5±7.9mmHg one month and to 15.9±6.1 mmHg three months after surgery.Thereafter, mean IOP remained stable with a mean value of 15.4±5.1 mmHg one year after XEN implantation, which corresponds to a mean IOP reduction of 31% compared to baseline before surgery.For the complete group of eyes analyzed the difference between the mean IOP results measured at all follow-up examinations and those measured at baseline was statistically signi cant at all-time points (p<0.01).
The exact values and the course of mean IOP during the rst 12 months after surgery for the complete group as well as for the four treated subgroups are summarized in Figures 1 & 2 and Table 2.

IOP results in different types of glaucoma
Mean IOP at baseline before XEN implantation in the 113 POAG eyes was 22.8±6.5 mmHg.Two days after surgery mean IOP dropped to 8.9±5.1 mmHg.From the follow-up examination one month after surgery the IOP remained stable at about 15 mmHg and one year postoperatively the mean value was 15.1±4.6 mmHg.This corresponds to an average IOP decrease of 31% compared to baseline.Comparison of mean IOP results before and after surgery showed a difference of statistical signi cance (p<0.01) at all follow-up examinations.
In the 5 NTG eyes the mean IOP before XEN implantation was 16.6±3.4mmHg.Mean IOP decreased to 5.6±2.7 mmHg two days after surgery and rose again to 10.0±3.0 mmHg one month after surgery and remained relatively stable thereafter with a mean value of 11.6±2.2mmHg 12 months after surgery.This corresponds to an IOP reduction of 29% from baseline.IOP was well regulated, with all ve eyes achieving an IOP <15 mmHg 12 months after surgery.Comparison of mean IOP results before and after surgery showed a difference of statistical signi cance (p<0.05) at all follow-up examinations.
In the 22 treated eyes suffering from PEX mean IOP before surgery was 28.0±7.9mmHg.Similarly to the other treatment groups mean IOP decreased to 11.9±6.4mmHg two days after surgery and increased to 16.5±8.0mmHg one month after surgery.Thereafter, IOP remained reasonably stable with a mean value of 17.1±6.6mmHg one year after XEN implantation, which corresponds to a decrease of 38% compared to baseline.The comparison between the average IOP results before and after surgery showed a difference of statistical signi cance (p<0.01) at all follow-up examinations.
In the 13 eyes suffering from secondary glaucoma mean IOP before surgery was 28.9±13.9mmHg.Mean IOP dropped to 8.7±4.6 mmHg 2 days after surgery.Thereafter, as in the other treatment groups, mean IOP increased to 16.7±10.6mmHg one month after XEN implantation and remained stable with a mean of 15.5±6.9mmHg at the follow-up examination one year after surgery, which corresponds to a mean IOP decrease of 36% compared to baseline.Comparison of follow-up to baseline results again showed a difference of statistical signi cance at alltime points (p<0.05).
A further statistical analysis before and 12 months after surgery comparing mean IOP results between the four treatment groups revealed differences of statistical signi cance at baseline (p<0.01) as well as 12 months after surgery (p<0.01).However, when comparing results for mean IOP reduction from baseline to 12 months after surgery between the four groups, no difference of statistical signi cance was found (p=0.2).

Complete and quali ed success
Taking into account the WGA guidelines and the criteria summarized above de ning complete and quali ed success levels A to C, in the complete group of 153 eyes 60% achieved a complete success (A) and 74% a quali ed success (A).Complete success levels B and C were achieved by 51% and 45% respectively.Quali ed success B and C was achieved by 61% and 56% in the complete group of 153 eyes operated.The exact results for complete and quali ed success levels A to C achieved in the four subgroups of different glaucoma entities is summarized in Table 3.

IOP-lowering medication
The exact course of the number of IOP-lowering eye drops applied is summarized in Figure 3 and Table 4.The average number of active substances applied in the entire group of 153 eyes fell from 2.6±1.2before to 0.8±1.3 12 months after XEN implantation (p<0.01).In the four subgroups treated the number of applied drugs decreased from 2.5±1.2 to 0.8±1.3 in the POAG group (p<0.01),from 2.6±0.9 to 0.6±0.9 in the NTG group (p<0.05),from 2.9±1.2 to 1.0±1.3 in the PEX group (p<0.01) and from 2.7±0.9 to 0.7±1.4 in the group of eyes suffering from secondary glaucoma (p<0.01)during the rst 12 months after surgery.

Visual Acuity
In the entire group of 153 eyes treated with the XEN microstent, mean BCVA remained stable and showed no difference of statistical signi cance during postoperative follow-up.Mean BCVA was 0.38±0.52logMAR before surgery and 0.36±0.49logMAR 6 months and 0.37±0.49logMAR 12 months after surgery (p=0.97 after 6 months and p=0.88 after 12 months).In the four subgroups of the different glaucoma types, changes in mean BCVA over the postoperative follow-up period also showed no statistical signi cance (POAG: p=0.31;NTG: p=0.22;PEX: p=0.92; secondary glaucoma: p=0.91).(See also Figure 4 for the exact course).However, the comparison of mean BCVA between the four treatment groups before and 12 months after surgery showed a difference of statistical signi cance (before surgery: p<0.01; 12 months: p<0.01).

Visual eld
Simultaneously with the drop of mean IOP, mean visual eld defect remained stable.In the complete group of 153 eyes treated mean defect was 10.1±4.3 dB at baseline, increased slightly to 10.8±4.3 dB at 6 months (p=0.11) and was 10.8±4.2 dB at 12 months after surgery, which was not statistically signi cant (p=0.15).Also in the four subgroups, mean visual eld defect showed no difference of statistical signi cance compared to the baseline (POAG: p=0.8;NTG: p=0.23;PEX: p=0.17; secondary glaucoma: p=0.37) (see Table 5 for details).

Complications and needling rate
To achieve the results described, 64 needling procedures had to be performed in 54 of the 153 eyes (35.3%).44 eyes needed 1 needling and another 10 eyes needed 2 needlings during the rst year after XEN implantation.None of the eyes needed >2 needlings.The needling rate was different between the four subgroups analyzed (see Table 6).The lowest needling rate was found in NTG eyes (20%) and the highest rate in PEX glaucoma eyes (55%).In the total group of 153 eyes treated 20 eyes showed benign choroidal detachment during the rst postoperative month.
All cases responded well to conservative treatment (contact lens and cycloplegia).A large and prominent subconjunctival hemorrhage (to the extent that it was classi ed as clinically signi cant) was seen in 12 cases immediately after surgery, but was resorbed after 1 to 4 weeks.Blood in the anterior chamber immediately after surgery was seen in 10 eyes and dissolved in all cases without further consequences after 1 to 4 weeks.
Postoperative macular edema was seen in 3 eyes, which all responded well to treatment with carbonic anhydrase inhibitors.Apart from the above described incidents, no serious complications were observed in the treated eyes.

Discussion
Classic TE with or without the use of antimetabolites is the "gold standard" for surgical treatment of a wide range of glaucoma entities.The XEN microstent is designed to bypass the transtrabecular out ow pathway whereby reducing IOP.Since its initial description, a number of reports have been published on the e cacy and safety pro le of the XEN microstent for the treatment of POAG [21,18,22].In summary, implantation of the XEN microstent can signi cantly reduce IOP and the amount of IOP-lowering eye drops applied, resulting in acceptable success rates in POAG after 12 and 24 months.Reitsamer and colleagues already showed in their one and two-year multicenter study (APEX study) that IOP and required medication can be reduced effectively by implantation of the XEN microstent either alone or in combination with phacoemulsi cation and posterior chamber lens implantation [21] Some groups even reported on achieving lower mean postoperative IOP values of around 12 mmHg when starting from a lower initial mean level at baseline before surgery [23].
Grover et al reported on their 12-month XEN results in a group of 65 eyes with uncontrolled and already pre-operated glaucoma cases (85% of eyes had already been unsuccessfully treated with ltering surgery and/or cilioablative techniques) [24] The results were similar to the data hereby reported with a quali ed success rate of 75%, a reduction of mean IOP from 25.1±3.7 mmHg at baseline to 15.9±5.2mmHg 12 months after surgery and a needling rate of about 32%.The patient cohort reported on also included a large percentage of POAG cases and some PEX cases, but no case of secondary glaucoma.Grover et al. also did not report on the e cacy of the XEN microstent in subgroups of different glaucoma entities.Lenzhofer et al. already demonstrated the long-term effectiveness of the XEN microstent in a group of 64 eyes of 64 POAG patients.It was shown that low mean IOP values and reduced necessity for application of IOP-lowering medication can even be found as long as four years after implantation of the XEN microstent [25].However, these data were collected in eyes in which the XEN 63 was implanted (with a larger internal lumen of 63 µm).
Given the mode of action utilized, the XEN microstent should in theory also be applicable in glaucoma entities other than POAG.Although some groups reported on the treatment of different types of glaucoma with the XEN microstent, they so far failed analyzing the e cacy in the treated subgroups [26,22].We analyzed the 12-month postoperative follow-up results in NTG, PEX and secondary glaucoma cases and compared the results with those obtained in a medium-sized comparison group of POAG eyes.With the reported postoperative results we could clearly and for the rst time demonstrate that IOP and necessary IOP-lowering medication can be lowered with a comparable e cacy in NTG, PEX and secondary glaucoma cases as in POAG.Apart from this visual acuity and visual eld indices remained stable without evidence for further postoperative deterioration, although only the rst 12 months after surgery were monitored.There was a notable difference in mean BCVA, which was worse but also stable during the postoperative course in the subgroup of treated secondary glaucoma cases.This must be interpreted within the context of the underlying disease process leading to secondary glaucoma, which in most cases has devastating consequences on visual acuity.
Postoperative results after implantation of the XEN microstent other types of glaucoma than POAG have also been described (but not been analyzed in more detail) by other authors.Ibanez-Munoz and colleagues implanted the XEN microstent in 36 eyes with POAG and 37 eyes with secondary glaucoma, 34 of which had PEX glaucoma [27].The difference in the postoperative results regarding the reduction of IOP and the required medication was not statistically signi cant in both groups.Apart from that, no difference was found when the XEN microstent was implanted as a solo procedure or in combination with cataract extraction and IOL implantation.

Karimi et al. (like our group
) have implanted the XEN microstent in different glaucoma types, including traumatic and neovascular glaucoma, but have not reported on postoperative results in these treated subgroups [28].They also found no statistically signi cant differences in their results regarding IOP reduction and medication in cases where the XEN microstent was implanted alone or in combination with cataract extraction.In the 259 eyes treated, 41% required needling during the rst 18 months after implantation.However, it should be noted that results originating from only 34% and 12% of eyes were included for analysis of the 12-and 18-month postoperative follow-up.
The higher rate of needling after XEN microstent implantation compared to TE, the "gold standard" in glaucoma surgery, has been the subject of criticism in the past.Schlenker et al. and others reported on further necessary secondary interventions and safety problems when comparing the results following both XEN microstent implantation and TE±MMC, which showed a slightly higher needling rate after XEN, but which did not reach statistical signi cance during further analysis [29].However, the different rates of necessary postoperative interventions and the rates of postoperative complications were not statistically signi cant between the groups treated with XEN microstent or TE.In summary, XEN microstent and TE are comparable in terms of postoperative safety.However TE led to more visits, especially in the rst 3 months after surgery.
The main strengths of this analysis are the results for glaucoma types other than POAG.In this analysis only one eye per patient was included (even if both eyes were treated).And for all included eyes a full 12-month postoperative follow-up was available.One limiting factor, concerning the hereby presented data, is that the sizes of the four treatment groups were different.However, the largest group, which comprises POAG patients, should only be regarded as a comparison group for the other, less frequent glaucoma entities, treated under the same conditions.In the entire study group, XEN microstent implantation not only resulted in a reduction of IOP and necessary IOP-lowering drugs, but also had no signi cant effect on BCVA, neither in the total group of 153 eyes nor in the four subgroups analyzed.Lenzhofer et al. showed similar results in their treated POAG cases [30].It has also been shown earlier that BCVA in eyes after TE decreases, even when taking into account (and ruling out) the known side effect of cataract formation.[31].Therefore, our hereby presented data also complement the growing evidence in favor for the XEN microstent in POAG and other types of glaucoma with an unblocked anterior chamber angle.

Conclusion
Implantation of the XEN microstent was effective in reducing intraocular pressure and the number of drugs applied to reduce intraocular pressure, while BCVA and perimetry results remained unchanged.In conclusion, the data presented here on the e cacy and safety of the XEN microstent in glaucoma types other than POAG are encouraging for the use of this (comparatively new) technique also in the treatment of PEX and other secondary glaucoma cases.Longer follow-up time, a larger number of eyes treated and a comparison group comparing the e cacy of the XEN microstent to the "gold standard" of glaucoma surgery, which remains the TE, are necessary to further demonstrate the usefulness of this technique.Needling rate for the whole group of patients (as well as for the four subgroups of the different types of glaucoma treated) and the rate of other complications observed during the rst 12 months of follow-up after XEN microstent implantation in uncontrolled glaucoma in all groups together.

Figures Figure 1
Figures
Secondary Glaucoma:10.1±6.8Patientdemographic data for the 153 eyes treated with XEN microstent implantation in uncontrolled glaucoma.Also shown are the glaucoma types and the percentage of each glaucoma type as a proportion of the total treated eye group.

Table 3 Complete
ned as complete (without the use of IOP-lowering drugs) and quali ed (with the use of additional IOP-lowering drugs) success for the complete treated group of eyes and the four subgroups of the different glaucoma types (A) IOP reduction of ≥20% from baseline and resulting IOP of <21 mmHg; complete success (B) IOP reduction of ≥30% from baseline and resulting IOP of <18 mmHg; complete success (C) IOP reduction of ≥40% from baseline and resulting IOP of <15 mmHg)