Comparison of surgical outcomes between initial trabeculectomy and Ex-PRESS in terms of achieving an intraocular pressure below 15 and 18 mmHg: a retrospective comparative study

Background To evaluate the postoperative outcomes of initial trabeculectomy (Trab) and Ex-PRESS (EX) in terms of achieving an intraocular pressure (IOP) below 15 and 18 mmHg. Methods This study retrospectively analyzed 64 and 54 cases of Trab and EX, respectively, performed by the same surgeon with uniform management from April 2018 to March 2019. Surgical success was defined as 5 < IOP < 15 mmHg (criterion 1) and 5 < IOP < 18 mmHg (criterion 2) without additional glaucoma medication, needling, and bleb reconstruction 2 months after surgery. Survival analysis with Cox regression was performed to identify factors associated with postoperative outcomes. Results The Trab and EX groups had an IOP of 22.6 ± 6.2 vs. 21.8 ± 6.0 mmHg before surgery (P = 0.507), 12.6 ± 2.6 vs. 14.0 ± 4.4 mmHg (P = 0.06) at 6 months, 12.7 ± 2.3 vs. 12.9 ± 2.8 mmHg (P = 0.678) at 12 months, 13.3 ± 2.6 vs. 12.6 ± 2.8 mmHg (P = 0.260) at 18 months, and 13.2 ± 2.3 vs. 13.6 ± 2.8 mmHg (P = 0.444) at 24 months, respectively. The proportion of those with an IOP < 15 mmHg in the Trab and EX groups was 82% vs. 81% at 6 months, 68% vs. 62% at 12 months, 63% vs. 61% at 18 months, and 57% vs. 53% at 24 months, respectively. The log-rank test showed no significant difference between the groups for Criteria 1 (P = 0.755) and 2 (P = 0.138). The results of the multivariate logistic analysis identified only a high preoperative IOP as a risk factor for surgical failure (odds ratio for Criteria 1: 1.076, P = 0.009 and Criteria 2: 1.068, P = 0.048). Conclusion Postoperative outcomes of Trab and EX suggested similar ability for achieving an IOP below 15 and 18 mmHg without medications and interventions. Supplementary Information The online version contains supplementary material available at 10.1186/s40662-022-00279-1.

of intraocular pressure (IOP) considering that a 1 mmHg increase in IOP promotes a 10-18% increase in the risk of developing glaucoma [2][3][4], whereas an IOP reduction of 1 mmHg decreases the risk of progression by about 10% [5]. Recently, clinicians' preferences have shifted toward minimally invasive glaucoma surgeries, resulting in surgical IOP outcomes of around mid-teens at 1 year [6][7][8][9][10]. Therefore, ophthalmologists perform filtering surgery such as trabeculectomy (Trab) and Ex-PRESS (EX) device implantation are required to produce both high success rates and low IOPs. Xen [11] and Preserflo [12] have recently utilized implant devices like EX, which uses mitomycin C (MMC) and filtering of the aqueous humor into the subconjunctival spaces. Therefore, given the worldwide trend toward minimally invasive surgeries, the surgical outcomes of Trab and EX should be reconsidered to achieve lower IOP. Previous studies comparing Trab and EX have adapted relatively relaxed surgical success criteria for IOP and nonuniform surgical management [13][14][15][16][17][18][19][20]. To determine the actual efficacy of surgery, restarting antiglaucoma medications after surgery should be considered given their influence on success rates.
Here, we evaluated the surgical outcomes of initial Trab and EX performed by a single consultant surgeon. In particular, this study compared the real-world ability of both procedures to achieve an IOP of < 15 and < 18 mmHg without medications and interventions.

Methods
This retrospective comparative study was approved by the Institutional Review Board of Saneikai Tsukazaki Hospital (IRB No. 211027) and was conducted in accordance with the tenets of the Declaration of Helsinki. Information in the electronic database of the Department of Ophthalmology, Saneikai Tsukazaki Hospital, was collected between April 2018 and March 2019. Initially, all patients (N = 198) who underwent Trab or EX (P50PL) (Alcon Laboratories, Fort Worth, TX, the USA) were assessed (by S.N.) during the study period. Thereafter, patients who underwent simultaneous cataract surgery, those with a history of glaucoma surgery within 3 years, and those who received vitrectomy after Trab or EX were excluded. Similarly, those with postoperative follow-up less than 1 year were excluded. However, patients who skipped their examination for 1 year due to Covid-19 but were re-examined consecutively 15 months after surgery were included. A flowchart for study inclusion is presented in Additional file 1: Fig. S1. In the end, 64 and 54 eyes that underwent initial Trab and initial EX were analyzed, respectively. We investigated patient demographics such as the type of glaucoma, severity of visual field defect, pre-and postoperative IOP, number of antiglaucoma eye drops used, and surgical interventions.

Surgical techniques and postsurgical management
Trab and EX were performed as similarly as possible, with the only difference being that the Trab surgery was performed with a second flap excision. Initially, a corneal traction suture was placed using a 9-0 silk suture to fix the eye downward. Thereafter, a 7-mm limbal conjunctiva incision and subtenon anesthesia with 2% xylocaine was performed, including epinephrine administration to stop the bleeding. After making spaces under Tenon's capsule and the conjunctiva for a wider view of the fornix area, coagulation for bleeding was performed, and a 2.5 × 2.5 mm half-thickness quadrangular scleral flap was created. Thereafter, 0.04% MMC was administered under Tenon's capsule and the conjunctival space for 3 min using neurosurgical pad bemsheets, followed by irrigation with approximately 200 mL of balanced salt solution. This procedure was similarly performed during both surgeries.
In eyes receiving Trab, a second 2.0 × 2.0 mm flap was created and cut off to construct a scleral tunnel. After creating a window anterior to the chamber using V-rance and Kelly punch, peripheral iridectomy was performed. The scleral flap was then sutured using 10-0 nylon for approximately 4 or 5 sutures to adjust the IOP to approximately 10-15 mmHg via palpation. In eyes undergoing EX, a 25-G needle was used to enter the anterior chamber parallel to the iris, after which the EX was implanted. The scleral flap was then sutured using 10-0 nylon for approximately 3 or 4 sutures to adjust the IOP to 10-15 mmHg via palpation.
After the surgery, patients were prescribed 1.5% levofloxacin and 0.1% betamethasone four times per day for approximately 1.5 months with gradual tapering. Laser suture lysis was performed to achieve a target IOP of approximately the low teens 1 week after surgery and 8 to 9 mmHg 2 weeks after surgery. However, laser suture lysis and massage were ineffective for creating blebs corresponding to an IOP at the low teens. Thus, revision surgery using a bleb knife [21] was performed after 1 to 2 months. Those who required needling procedure or bleb revision after 2 months and restarting of glaucoma medication were defined as surgical failures. Around 3 months after surgery, needling was the first treatment of choice for raising IOP after Trab or EX [21], followed by limbal-based bleb revision. Patients who refused surgical repair of the bleb were mostly treated with prostaglandin analog monotherapy.
Oral acetazolamide and topical 1% atropine were administered when choroidal detachment occurred following a decrease in IOP, all of which resolved within 2 months. Patients were scheduled to visit the clinic at postoperative day 1; postoperative weeks 1, 2, and 3; and postoperative months 1, 2, and 3. When no intervention for increasing IOP was required after 3 months, patients were scheduled to visit every 3 months. All surgical procedures, postoperative examinations, laser suture lysis, and interventions (e.g., needling, bleb revisions, and medical prescriptions) were performed by S.N. to control for the effects of surgeon factors (surgical skill, bleb management, strategy, etc.) on surgical outcomes and ensure maximum uniformity in the surgical procedure. Best-corrected visual acuity (BCVA) was evaluated before surgery, 6 months after surgery, 1 year after surgery, and during the last visit. Additional glaucoma surgeries, were also determined via the medical records. IOP was measured using the Goldmann applanation tonometer.

Statistical analysis
Statistical analyses were performed using JMP version 10.0.0 (SAS Institute Inc., Cary, NC, USA) and statistical program R software (version 3.6.1, http:// www. rproj ect. org/). For the comparison of patient background and frequency of early postoperative complications, Fisher's exact test, Chi-squared test, and Mann-Whitney U test were utilized. IOP and the number of patients using glaucoma eye drops were presented as mean ± standard deviation (SD) and compared using Welch's t-test and Mann-Whitney U test, respectively. Baseline BCVA (logMAR) was evaluated via the one-way analysis of variance (ANOVA) test and compared using Welch's t-test. Surgical success was defined as 5 < IOP < 15 mmHg (Criterion 1) and 5 < IOP < 18 mmHg (Criterion 2), without any additional glaucoma medication, needling, and bleb reconstruction 2 months after surgery. The surgery was deemed as a failure when the IOP was < 5 mmHg or > 15 mmHg (Criterion 1), and > 18 mmHg (Criterion 2) at two consecutive visits, antiglaucoma medication was restarted, needling revision was performed using MMC [21], or bleb revision and additional glaucoma surgery was conducted at any time after 3 months following surgery.
Survival analysis with Cox regression was performed to identify factors associated with postoperative outcomes. Multivariate logistic regression analysis was conducted using surgery (Trab/EX), age, sex, eye (right/ left), glaucoma types, lens status [phakia/intraocular lens (IOL)], preoperative IOP, and the number of preoperative antiglaucoma medications to identify the risk factors for surgical failure. To determine the number of antiglaucoma medications, a fixed combination was calculated as two medications, whereas oral acetazolamide was counted as one medication. P values of less than 0.05 were considered statistically significant. Table 1 summarizes the demographic data of the 64 eyes from the 64 patients in the Trab group and the 54 eyes from the 54 patients in the EX group. The EX group was significantly older (median age, 75 vs. 61 years; P < 0.001; Mann-Whitney U test) and had significantly a higher IOL implantation rate (P < 0.001; Fisher's exact test) than the Trab group. No significant differences in sex (P = 0.510, Fisher's exact test), treated eye (P = 0.430, Fisher's exact test), and glaucoma types (P = 0.610; Chi-squared test) were observed between the groups.

IOP and number of antiglaucoma medications
IOP and the number of medications received during follow-up are detailed in Table 2 and 3. IOP (Table 2) is presented together with all antiglaucoma medications and additional interventions. Additional glaucoma surgery comprised μ-hook Trab [22] in the Trab (one case) and EX (one case) groups.
After each surgery, IOP decreased significantly at each time point compared with the baseline (P < 0.001; paired t-test).
The number of antiglaucoma eye drops decreased significantly from presurgical to each postoperative time point in each group, respectively (all P < 0.01; Mann-Whitney U test, Table 3). No significant intergroup differences were noted (all P > 0.30).
Presurgical visual field defects measured by the Humphry Field Analyzer (HFA) (24-2 or 30-2 Swedish Interactive Threshold Algorithm Standard test) were only present in 52 and 35 patients in the Trab and EX groups [mean deviation: − 16.03 ± 6.3 vs. − 18.23 ± 6.5 dB for the Trab EX groups, respectively (P = 0.117; Welch's t-test)]. The remaining patients' visual field tests could not be analyzed given that they underwent Goldmann perimetry or the HFA 10-2 program.

Visual acuity
The EX group had significantly worse baseline BCVA than the Trab group (0.29 ± 0.28 vs. 0.17 ± 0.18 logMAR, respectively; P = 0.009). However, there was no significant change in BCVA between each group during followup (before surgery, after 6 and 12 months of surgery, and during the final visit) (P = 0.863 and 0.494 in the Trab and EX groups, respectively; one-way ANOVA).

Complications within 1 month after surgery
No significant difference in the presence of a shallow anterior chamber, choroidal detachment, Seidel treated by a soft contact lens, hyphema, iris inlaying for scleral window or device, and vitreous extrusion in the anterior chamber were observed between the groups (all P > 0.20, Fisher's exact test, Table 4).

Survival analysis results
The median follow-up period (quantile) was 743 days (231-844 days) and 805 days (217-976 days) in the Trab and EX groups, respectively (P = 0.592, Welch's t-test). In our study, 58 events occurred in Criteria 1 and 43 events occurred in Criteria 2. Criterion 1. 5 < IOP < 15 mmHg without any additional glaucoma medication or intervention.

Discussion
The present study showed the real-world surgical outcomes of Trab and EX under uniform management using two strict criteria. Few studies that are available have compared Trab and EX over a follow-up duration of more than 1 year (Table 5) [13][14][15][16][17][18][19][20]. Moreover, studies varied in terms of the primary success criteria utilized, with almost all using 5 < IOP < 18 or < 20 mmHg with medications as their primary success criteria, which were less strict than our second criteria. However, de Jong et al., who used criteria similar to us (i.e., 15 and 18 mmHg for Criteria 1 and 2), showed that EX was superior to Trab under both criteria during all follow-up time points, which was contrary to our findings. Similar to this study, surgical procedures by de Jong et al. were completed by a single surgeon; however, they did not include patients with secondary glaucoma, neovascular glaucoma, and angle closure glaucoma, which may have caused the discrepancy in results.
Most studies showed that Trab and EX had similar IOP lowing efficacy. Notably, three studies utilized primary criteria similar to that of our Criterion 2 [13,15,19]. A randomized control study by Gonzalez-Rodriguez JM et al. [13] reported primary success rates of 42% and 43% at 2 years following Trab and EX, respectively (≤ 18 mmHg, i.e., our Criterion 2, P = 0.78). Meanwhile, Dahan et al. 's study showed no concrete success rate at any year [15]. In contrast, a retrospective study by Liu i.e., our Criterion 2, P > 0.05) [19]. Although the aforementioned study considered Trab and EX to have similar success rates for achieving an IOP of under 18 mmHg without any medication or intervention, their success rates were slightly lower than those presented herein (72% vs. 64% for the Trab and EX groups, respectively). This study identified preoperative IOP as a significant risk factor for surgical failure in both surgical procedures. The Collaborative Bleb-related Infection Incidence and Treatment Study that utilized a large data set also reported preoperative lens status and preoperative higher IOP as significant risk factors for Trab surgical failure [23]. Moreover, the Advanced Glaucoma Intervention Study (AGIS) study found that younger age and higher IOP were significant risk factors for Trab success [24].  Therefore, they concluded that Xen was inferior to EX for complete success [29]. Further studies are needed to determine whether all filtering surgeries using MMC for controlling IOP via laser suture lysis through a scleral flap are superior to single insertion Xen and Preserflo. The present study has several limitations that are worth noting. First, given that our study was retrospective in nature and utilized real-world data, differences in patient backgrounds could not be adjusted for. Accordingly, the EX group were older, with a higher IOL implantation rate, and worse visual acuity. This bias may be attributed to our knowledge that EX promoted rapid visual acuity recovery to baseline after the surgery [14] and that inserting EX is safer in patients with IOL to prevent the device from touching the iris. The second limitation concerns the cost-effectiveness of IOP reduction. Several problems related to costs have remained for EX [30,31], which should be discussed in studies regarding recent micro incision glaucoma devices in the near future. Third, the recent Japanese national survey showed that EX had significantly higher reoperation rates than Trab (adjusted hazard ratio, 1.72; P < 0.001) [31]. In addition, we excised the second flap via Trab. Dada et al. recently reported that deep sclerectomy-an IOP-lowering procedure, like our method-was superior to single-flap Trab based on a randomized control study (12.5 ± 1.67 vs. 13.4 ± 1.83 mmHg at 12 months, P = 0.04) [32]. Therefore, further long-term studies are warranted.

Conclusion
Our study revealed that Trab and EX exhibited similar ability for achieving an IOP below 15 and below 18 mmHg without medications and interventions. Future prospective studies should determine the proper indications for Trab, EX, and minimally invasive glaucoma surgery to control IOP for glaucoma management.