Comparison of the Intrableb Characteristics of Anterior Segment Optical Coherence Tomography Imaging in Trabeculectomy according to Amniotic Membrane Transplantation

Trabeculectomy, a procedure introduced in the 1960s, remains the gold standard for glaucoma filtering surgery in patients with medically uncontrolled glaucoma [1‒3]. The purpose of trabeculectomy is to allow aqueous humour drainage from the anterior chamber to the subconjunctival or subtenon’s space [3, 4]. The aqueous humour in the filtering bleb passes through the conjunctiva and mixes with the tear film, or is absorbed by vascular or perivascular conjunctival tissue and lymphatic vessels near the surgical area [4].

Failure of trabeculectomy may be due to obstruction of the aqueous outflow at the level of the scleral flap or the ostium, and formation of fibrotic tissue in the subconjunctival and episcleral area [3, 5]. Although the surgical success rate of trabeculectomy increases with the use of antimetabolites, including mitomycin C (MMC) and 5-Fluorouracil, the procedure is associated with an increased risk of complications, such as avascular cystic bleb, bleb leakage, and bleb-related infection [6‒9].

Previous studies reported that trabeculectomy with amniotic membrane transplantation (AMT) had a lower complication rate and higher success rate than trabeculectomy without AMT [10‒12]. In our previous study, trabeculectomy with AMT was found to be a safe and effective method for intraocular pressure (IOP) reduction in patients with primary open-angle glaucoma (POAG) without the development of avascular cystic blebs or bleb-related infection [12].

Intrableb structures which can be analysed using anterior segment optical coherence tomography (AS-OCT) or ultrasound biomicroscopy (UBM) are associated with the surgical success of trabeculectomy [13‒18]. Intrableb morphology, including bleb wall reflectivity, bleb height, bleb wall thickness, and length and height of the fluid-filled cavity, have been associated with bleb function after trabeculectomy [15, 19, 20]. Only one study has evaluated intrableb structures after trabeculectomy with AMT using an anterior segment imaging device [14]. The authors investigated intrableb structures using UBM in patients with glaucoma who underwent trabeculectomy with or without AMT and reported that bleb wall reflectivity in the trabeculectomy alone group and the extent of the subconjunctival fluid-filled space in the AMT-assisted trabeculectomy group were factors associated with long-term IOP control [14]. However, the difference in the type of conjunctival incision between the two groups was a potential limitation of the UBM study as it may influence the IOP-lowering mechanism and intrableb structures associated with IOP control [14]. A fornix-based conjunctival incision was made for trabeculectomy without AMT, while limbal-based conjunctival incision was made for trabeculectomy with AMT [14]. Further studies found that the different types of conjunctival incision in trabeculectomy had distinct features of filtering bleb associated with IOP control [21, 22]. Therefore, the type of conjunctival flaps should be considered in the evaluation of intrableb structures associated with bleb function. Furthermore, bleb wall reflectivity was assessed qualitatively, not quantitatively in the previous study [14]. UBM examination is time-consuming and may be associated with the risk of bleb infection or bleb morphological change as examination requires contact with the eye [22‒24]. On the contrary, bleb imaging with AS-OCT is a non-contact and non-invasive examination that allows imaging of filtering bleb with higher resolution than UBM [22].

To our knowledge, AS-OCT imaging of the filtering bleb after fornix-based trabeculectomy with AMT has not been previously evaluated. Therefore, the purpose of this study was to compare the characteristics of the filtering bleb assessed with high-resolution spectral-domain AS-OCT imaging after trabeculectomy with a fornix-based conjunctival flap according to AMT and to evaluate intrableb parameters associated with successful IOP control in patients with POAG.

This investigation was a retrospective cohort study. The study was conducted according to the guidelines of the World Medical Association Declaration of Helsinki, and approved by the Institutional Review Boards of Pusan National University Hospital (approval no. 2111-004-108). All patients gave written informed consent for the surgical procedures and for their information to be stored in the hospital database and used for research.

The patients with POAG who underwent fornix-based trabeculectomy with or without AMT between November 2014 and March 2020 were enrolled in this study. And all the patients were followed-up for at least 1 year postoperatively. Since August 2017, we have performed the trabeculectomy with AMT for all eyes who were eligible for glaucoma surgery and agreed to receive the AMT. Indications for surgery were as follows: (1) IOP not appropriately controlled with maximal tolerated medical therapy or laser trabeculoplasty or both, and (2) intolerance or allergy to glaucoma medication. Exclusion criteria included secondary glaucoma including pseudoexfoliation syndrome, pigment dispersion syndrome, and any other ocular or systemic disorder known to affect the optic nerve head, macula, or visual field. Patients who had undergone previous ocular surgery, except for uncomplicated phacoemulsification, were excluded.

Prior to the study, all the participants underwent thorough ophthalmologic examination, including best corrected visual acuity (BCVA), slit-lamp examination, IOP measurement with Goldmann applanation tonometry, gonioscopy, dilated fundus examination, stereoscopic optic disc, and red-free retinal nerve fibre layer photography (AFC-210; Nidek, Aichi, Japan), biometry using the IOL Master (Carl Zeiss Meditec, Dublin, CA, USA), and standard automated perimetry. Central corneal thickness (CCT) was measured using ultrasonic pachymetry (Pachmate; DGH Technology, Exton, PA, USA). Keratometry was performed with an Auto Kerato-Refractometer (ARK-510A; NIDEK, Hiroshi, Japan).

POAG was diagnosed by the presence of glaucomatous optic disc changes and corresponding visual field defects as confirmed by two reliable visual field tests and open anterior chamber angle. A glaucomatous optic disc was defined as a disc meeting at least one of the following criteria: (1) a rim notch with a rim width ≤0.1 disc diameter; or (2) a vertical cup-to-disc ratio of >0.7; or (3) a cup-to-disc ratio asymmetry between the 2 eyes ≥0.2; or (3) disc haemorrhage; or (4) a retinal nerve fibre layer defect congruent with visual field defects [25].

Automated perimetry was performed using a Humphrey Visual Field Analyzer 750i instrument (Carl Zeiss Meditec) with the Swedish interactive threshold algorithm 24-2 in all subjects. Glaucomatous visual fields were those that met at least one of Anderson-Patella’s criteria [25]: (1) a cluster of ≥3 points in the pattern deviation plot in a single hemifield (superior/inferior) with p < 0.05, one of which must have been p < 0.01; (2) glaucoma hemifield test result outside normal limits; or (3) abnormal pattern standard deviation with p < 0.05. Reliable visual field tests were defined as false-positive rate <15%, false-negative rate <20%, and fixation loss <20%.

Successful IOP control was defined as an IOP ≤18 mm Hg and an IOP reduction ≥20% without glaucoma medication at AS-OCT examination [26]. IOP was measured at 10:00 AM and 5:00 PM at the AS-OCT test date, and if the difference between the two values was greater than 2 mm Hg, a third IOP was additionally measured, the mean of which was used in the analysis [27]. If eyes did not meet the above criteria, the IOP control was considered unsuccessful.

All procedures were performed by one surgeon (J.W.L). Under local anaesthesia, the limbal conjunctiva was incised 5–6 mm to form the fornix-based conjunctival flap, and the conjunctiva and Tenon’s capsule were separated towards the conjunctival sac. A trapezoidal scleral flap (basal 4 mm, apical 2.5 mm, bilateral 2.75 mm) with 2/3 of the sclera thickness was constructed. Weck-cell sponges soaked with 0.4 mg/mL (0.04%) of MMC were applied between the Tenon’s capsule and sclera for 2–3 min. After the sponges were removed, the area exposed to MMC was irrigated with 20 mL of balanced salt solution. Inner sclerostomy was performed, and then peripheral iridectomy was performed. The scleral flap was closed with two preplaced 9-0 nylon (Ethicon Inc., Johnson & Johnson, Somerville, NJ, USA) releasable sutures. In the AMT group, 15 × 15 mm single layer of cryopreserved amniotic membrane (MS Amnion, MS BIO inc., Seongnam, South Korea) was placed with the stromal side facing up under Tenon’s capsule. The amniotic membrane was secured to the lateral side of scleral flap with two interrupted 10-0 nylon sutures (Ethicon Inc., Johnson & Johnson) (Fig. 1). The anterior chamber was inflated with BSS, and the degree of aqueous outflow through the scleral flap and bleb leakage through the conjunctival sutures were assessed. Postoperatively, administration of topical eye drops, including Levofloxacin (Cravit^Ⓡ, Santen Pharm, Co., Osaka, Japan) four times a day and Prednisolone acetate (Predbell^Ⓡ, CKD Pharm, Co., Seoul, South Korea) six times a day for 1 month, was commenced and tapered over 8–12 weeks according to bleb morphology and IOP.

Postoperative blebs were imaged with the Anterior Segment Module (ASM, volume scan vertical-filtering blebs mode, “VolBleb”) of the Spectralis OCT (Heidelberg Engineering, Heidelberg, Germany) which used shorter wavelength light sources (870 nm) and higher speed (40,000 A-scans/second) than time domain OCT.

In this study, the high-resolution VolBleb mode with enhanced depth imaging mode and automatic real time of 16 frames was used. OCT scan pattern size was 8.3 × 2.8 mm, and the number of B-scans was 21 with a 139 μm-distance between B-scans. Penetration depth was 1.9 mm with lateral resolution scaling of 10.84 μm/pixel and axial resolution scaling of 3.87 μm/pixel. Images with a quality score >25 dB were included for analysis. When the scleral margin was not apparent in OCT image, manual adjustment of the contrast setting enabled identification of the sclera edge. Intrableb structural parameters were measured with the device’s built-in software (Heidelberg Eye Version: 1.10.2.0), while bleb wall reflectivity was measured with ImageJ software (ImageJ 1.50b, http://imagej.nih.gov/ij/; developed by Wayne Rasband, National Institutes of Health, Bethesda, MD, USA).

Horizontal (tangential to the limbus) and vertical scans (radial perpendicular to the limbus) were taken at the maximum elevation of bleb for each eye. The quantitative parameters included maximum bleb height, maximum bleb wall thickness, maximum striping layer thickness, maximum fluid-filled space height and area, fluid-filled space score (FFSS), and bleb wall reflectivity (Fig. 2). The mean of these horizontal and vertical measurements was used for analysis. The symbols used in the figures were defined as follows: The white and white dotted two-way arrows indicate the bleb wall thickness and fluid-filled space height, respectively. The star indicates the fluid-filled space and the asterisk indicates scleral flap. The black arrow indicates the microcyst and the white arrow indicates visible amniotic membrane beneath the bleb wall. The white arrow head indicates the margin of hyporeflective layers with striping phenomenon.

Bleb height was measured as the maximal vertical distance between the first reflective signal from the conjunctiva along a straight line perpendicular to a tangent to sclera (Fig. 2). Bleb wall thickness consisting of conjunctiva, Tenon’s capsule, and/or incorporated amniotic membrane was measured as the maximal vertical distance between the first reflective signal of the conjunctiva to the top of the fluid-filled space (Fig. 2). The striping layer was defined as the multiple parallel and fluid-filled channels in Tenon’s capsule, which resembled a honeycombed structure (Fig. 3d) [16, 28].

Fluid-filled space height was measured as the maximal vertical distance in the signal void or hyporeflective area between the bottom of the inner bleb wall and the top of the sclera along a straight line perpendicular to a tangent to the sclera (Fig. 2). The fluid-filled space area was measured as the maximal area of signal void or hyporeflective area between the bottom of the inner bleb wall and the top of the sclera. The fluid-filled space score was graded from 0 to 2: (1) score 0, not visible fluid-filled space; (2) score 1, limited and demarcated fluid-filled space with a clear posterior margin; and (3) score 2, diffuse fluid-filled space extending posteriorly beyond the field of image view [14].

Bleb wall reflectivity was analysed using ImageJ software. An ellipse mark of the background near the bleb wall and 3 different evenly spaced ellipse marks in the bleb wall (anterior, middle, and posterior) were placed to measure the bleb wall reflectivity. The value of background was subtracted from the 3 bleb wall reflectivity values, and then the average of the 3 bleb wall reflectivity values was calculated [29]. Microcyst formation was a qualitative parameter in the analysis and defined as a hyporeflective or signal void space directly within or beneath the epithelial layer in the bleb wall [16].

Data distribution normality was checked using the Kolmogorov-Smirnov test. Differences between the two groups were analysed with the Mann-Whitney U test or Independent-sample t-test for continuous variables and the χ² or Fisher’s exact test for categorical variables. Multivariate logistic regression analysis with stepwise forward variable selection method was used to determine intrableb AS-OCT-based structural parameter which was significantly associated with successful IOP control in entire eyes and in each AMT and control group, and the odds ratio for these parameters was calculated. The independent variables included were sex, preoperative lens status, performance of AMT (for analysis in the entire group) as categorical variables. Age, the number of preoperative glaucoma medications, and preoperative IOP were included as continuous, independent variables. The intrableb parameters included as variables were bleb height, bleb wall thickness, striping layer thickness, bleb wall reflectivity, fluid-filled space height, fluid-filled space area, fluid-filled space score, and presence of microcyst formation. All statistical analyses were performed using SPSS software version 22 (IBM Corp., Armonk, NY, USA). p values <0.05 were considered statistically significant.

A total of 116 eyes of 103 patients were included in this study. Thirty-one eyes of 27 patients received fornix-based trabeculectomy with MMC alone (control group), and 85 eyes of 76 patients received fornix-based trabeculectomy with MMC and AMT (AMT group). Demographics and clinical characteristics of patients in each group are shown in Table 1. There were no significant differences between the two groups in sex, age at trabeculectomy, MMC soaking time, IOP at AS-OCT test, preoperative IOP, number of preoperative glaucoma medications, preoperative BCVA, preoperative lens status, CCT, axial length, spherical equivalent, and visual field parameters (all ps ≥ 0.053). Overall frequency of successful IOP control (72/85 = 84.7% in the eyes with AMT vs. 23/31 = 74.2% in those without AMT; χ² test, p = 0.274; Table 1) was similar in the two groups.

AS-OCT-based intrableb structural parameters were compared between eyes with successful IOP control and eyes with unsuccessful IOP control in both the AMT and control groups (Table 2). In the AMT group, eyes with successful IOP control had higher bleb (p < 0.001), thicker bleb wall and striping layer (p = 0.004 and p < 0.001, respectively), lower bleb wall reflectivity (p < 0.001), and more frequent microcyst formation (p = 0.002) than eyes with unsuccessful IOP control. In addition, the area and score of the fluid-filled space were greater in eyes with successful IOP control than in those with unsuccessful IOP control in AMT group (p = 0.043 and p < 0.001, respectively).

In the control group, eyes with successful IOP control had higher bleb (p = 0.005), thicker bleb wall and striping layer thickness (p = 0.009 and p = 0.001, respectively), lower bleb wall reflectivity (p = 0.003), and more frequent microcyst formation (p = 0.012) than eyes with unsuccessful IOP control. However, there was no significant difference in parameters associated with the fluid-filled space between the eyes with successful IOP control and those with unsuccessful IOP control (all ps ≥ 0.674).

Multivariate logistic regression analysis was performed to determine the AS-OCT-based intrableb parameters associated with IOP control in entire eyes and in each AMT and control group (Table 3). In the AMT group, lower bleb wall reflectivity (p = 0.003), greater fluid-filled space score (p = 0.027), and the presence of microcyst formation (p = 0.041) were directly associated with successful IOP control. In the control group, lower bleb wall reflectivity alone was directly associated with successful IOP control (p = 0.019). In the entire group, lower bleb wall reflectivity (p < 0.001), greater fluid-filled space score (p = 0.002), and the presence of microcyst formation (p = 0.034) were directly associated with successful IOP control.

The presence of transplanted amniotic membrane was also evaluated on AS-OCT examination in the AMT group. It was found in 52/72 (72.2%) of the eyes with successful IOP control and 6/13 (46.2%) of the eyes with unsuccessful IOP control. However, there was no significant difference between the two groups (p = 0.065, Fisher’s exact test). On univariate logistic regression analysis, the presence of the transplanted amniotic membrane was not associated with IOP control in the AMT group (p = 0.069). Encapsulated bleb was found only in patients with unsuccessful IOP control in both the control group (3/8 eyes, p = 0.012) and the AMT group (6/13 eyes, p < 0.001).

Figure 3 showed distinct characteristics of intrableb structures in the same patient who underwent trabeculectomy with or without AMT. A 34-year-old male patient underwent trabeculectomy with AMT in the right eye and trabeculectomy without AMT in the left eye. Slit-lamp photographic images of both eyes showed that the bleb had similar morphologic findings of diffuse, moderate height, and mild vascularity based on the Indiana Bleb Appearance Grading Scale (H2 E3 V2 S0) (Fig. 3a, c) [30]. However, the AS-OCT-based intrableb structures were remarkably different between the two eyes (Fig. 3b, d).

Figure 4 showed representative cases of AS-OCT images after trabeculectomy with (Fig. 4a, c, e) or without (Fig. 4b, d, f) AMT. A 75-year-old male patient underwent trabeculectomy with AMT in the left eye 2 years before AS-OCT examination (Fig. 4a). The functioning bleb had a posteriorly extended fluid-filled space (FFSS 2) with successful IOP control of 10 mm Hg without medication. Transplanted amniotic membrane was visible at the inner bleb wall. A 53-year-old female patient underwent trabeculectomy alone in the left eye 2.5 years before AS-OCT examination (Fig. 4b). AS-OCT image showed thick bleb wall with multiple parallel hyporeflective layers and fluid-filled channels within the bleb wall, and IOP was 10 mm Hg without medication. A 68-year-old male patient underwent trabeculectomy with AMT in the left eye 2 years before AS-OCT examination (Fig. 4c). The functioning bleb had a posteriorly extended fluid-filled space (FFSS 2), Tenon’s layer with striping phenomenon, and multiple microcysts with successful IOP control of 13 mm Hg without medication. Transplanted amniotic membrane was visible in the fluid-filled space beneath inner bleb wall. A 69-year-old female patient underwent trabeculectomy alone in the right eye 2 years before AS-OCT examination (Fig. 4d). AS-OCT image showed multiple parallel hyporeflective layers and fluid-filled channels inside Tenon’s layer and successful IOP control of 8 mm Hg without medication. A 70-year-old man underwent trabeculectomy with AMT in the left eye 3 years before AS-OCT examination (Fig. 4e). AS-OCT image showed a thin hyperreflective bleb wall and the fluid-filled space was limited with a clear demarcated posterior margin (FFSS 1). IOP was 21 mm Hg with medications. A 60-year-old male patient underwent trabeculectomy alone in the left eye 1 year before AS-OCT examination (Fig. 4f). AS-OCT image showed low height bleb with a thin and hyperreflective bleb wall and limited fluid-filled space (FFSS 1), which suggested a non-functioning bleb. IOP was 20 mm Hg with glaucoma medications.

In the present study, the intrableb structural factors associated with IOP control were investigated using AS-OCT in patients with POAG who underwent fornix-based trabeculectomy with or without AMT. To our knowledge, this is the first study to evaluate intrableb structure after fornix-based trabeculectomy with AMT using AS-OCT.

We found that the functioning bleb in both control and AMT groups had common features of intrableb structure which were different from those of the non-functioning bleb. The functioning bleb was higher and had a thicker bleb wall and striping layer, lower bleb wall reflectivity, and more frequent microcyst formation than the non-functioning bleb in both groups.

However, there were additional distinct characteristics of intrableb parameters in the functioning bleb in AMT group which were not observed in control group. We found that the fluid-filled space area and score in the functioning bleb were greater than those of the non-functioning bleb in the AMT group. Conversely, there were no significant differences in the fluid-filled space area and score between the functioning bleb and non-functioning bleb in the control group. Multivariate logistic regression analysis confirmed that higher fluid-filled space score, as well as lower bleb wall reflectivity, and microcyst formation were significantly associated with surgical success in the AMT group, whereas lower bleb wall reflectivity alone was associated with surgical success in the control group.

These findings are consistent with those of a previous study evaluating the UBM-based intrableb structure associated with IOP control after trabeculectomy with AMT [14]. The authors reported that a wide subconjunctival fluid-filled space was associated with successful IOP control in the AMT group [14]. The results of the present study are also consistent with an earlier study which assessed the filtering bleb after trabeculectomy alone with AS-OCT [20, 24]. Tominaga et al. [24] demonstrated that the IOP levels were significantly lower in eyes with posterior episcleral fluid (PEF) beyond the scleral flap compared to eyes without PEF. Kawana et al. [20] found that IOP was negatively correlated with horizontal and vertical length of the fluid-filled cavity, height of fluid-filled cavity, and volume of the internal fluid-filled cavity.

In this study, there were no significant differences in the fluid-filled space area and score between the functioning bleb and non-functioning bleb in the control group. However, the functioning bleb had greater bleb height, bleb wall thickness, striping layer thickness, and lower bleb wall reflectivity than the non-functioning bleb in the control group. These findings are consistent with the results of a previous study using UBM which reported that eyes without AMT had no or minimal fluid-filled space and that a third of the bleb with good IOP had a hyporeflective bleb wall after trabeculectomy alone [14].

There are two possible explanations for our findings. The interval between trabeculectomy and the time of AS-OCT examinations in this study was 2.48 ± 0.73 years in the control group and 2.23 ± 0.74 years in the AMT group. Since post-trabeculectomy changes over time in the AS-OCT parameter have been reported, the possibility that the subconjunctival fluid-filled space may be diminished due to the wound healing process after trabeculectomy alone cannot be excluded [14, 16].

In patients with POAG, reports found that transforming growth factor-β (TGF-β) was elevated in the aqueous humour and was significantly higher in trabecular meshwork cells, inducing fibrotic changes in the trabecular meshwork leading to increased aqueous outflow resistance and IOP elevation [31‒33]. In addition, glaucoma filtration surgery such as trabeculectomy was found to cause tissue trauma inducing a localized inflammatory response such as release of cytokines, including TGF-β, recruitment or activation of macrophages, polymorphonuclear cells, and platelets [34].

The transdifferentiation of fibroblasts into myofibroblasts is a major step in wound healing and scar formation, and myofibroblast transdifferentiation is regulated by TGF-β [35]. Therefore, downregulation of TGF-β signalling is a principal strategy for preventing scar formation during wound healing after glaucoma surgery [36].

The amniotic membrane has anti-scarring and anti-inflammatory properties in fibrotic eye disease, including glaucoma surgery [10‒12, 37, 38]. The anti-scarring effect of the amniotic membrane is mediated by downregulating TGF-β signalling pathway and myofibroblast differentiation. The amniotic membrane also precludes polymorphonuclear cell infiltration and facilitates macrophage apoptosis [37, 39]. Therefore, we hypothesize that the fluid-filled space in the AMT group persisted after trabeculectomy for a long time due to the anti-scarring and anti-inflammatory effect of the amniotic membrane, although the AMT group revealed a similar interval between surgery and AS-OCT examination to that observed in the control group [14, 40]. Nakamura et al. [14] also suggested that the amniotic membrane may prevent adhesion between the conjunctiva and sclera, which can help maintain the fluid-filled space and subsequent subconjunctival aqueous drainage pathway in the posterior direction.

In a rabbit experiment, the preserved human amniotic membrane was decomposed after approximately 3–4 weeks after trabeculectomy [41]. However, in our study, AS-OCT was performed in patients with a minimum follow-up period of 1 year, and the transplanted amniotic membrane was noted in 52/72 (72.2%) in the successful IOP control group and 6/13 (46.2%) in the unsuccessful IOP control group (p = 0.065). Even though the presence of the amniotic membrane was not associated with surgical success (p = 0.069), transplanted amniotic membrane may maintain its anti-fibrosis, anti-inflammatory activity and improve the stability of the bleb wall as a supportive tissue for a considerable period of time.

The second explanation for our finding is that the prominent aqueous drainage route may be different between the control group and AMT group [14]. The function of the filtering bleb after trabeculectomy alone may be associated with transconjunctival outflow of aqueous humour [14]. Amar et al. [42] found that microcysts observed at the surface of functioning bleb corresponded to goblet cell containing aqueous humour. The finding of this study is consistent with those of previous AS-OCT imaging studies which evaluated intrableb structural parameters associated with IOP control after trabeculectomy alone [15, 20, 24, 43]. Tominaga et al. [24] demonstrated that thicker bleb wall with lower bleb wall reflectivity was associated with lower IOP. Kawana et al. [20] also reported that volume of the hyporeflective area and microcyst number was negatively related with IOP. Narita et al. [15] found that greater bleb height was significantly associated with surgical success in multivariate analysis. Singh et al. [43] reported that thickening of the bleb wall was found in the majority of successful blebs, whereas bleb wall thickening was typically absent in failed blebs.

The main aqueous drainage route may be the subconjunctival pathway in the AMT group [14]. Nakamura et al. [14] suggested that amniotic membrane located between the conjunctiva and sclera may act as a barrier against transconjunctival movement of aqueous humour. Therefore, the authors concluded that the blebs after trabeculectomy with AMT could maintain successful IOP control only when the fluid-filled space was formed posteriorly beyond the field of image view regardless of bleb wall reflectivity [14].

However, we found that a greater fluid-filled score, as well as lower bleb wall reflectivity and more frequent microcyst formation were associated with successful IOP control in the AMT group. Amniotic membrane has a high hydraulic conductivity and is semipermeable to water [44]. Our previous experimental study found that when the amniotic membrane was applied to the upper surface of the Ahmed glaucoma valve body, the fibrous capsule was looser and had a more disorganized collagen architecture in rabbit eyes with AMT compared to eyes without AMT [36]. Although successful IOP control is mainly associated with intact subconjunctival aqueous drainage pathway, transconjunctival drainage of aqueous humour may play a partial role in the AMT group.

In contrast to the previous study by Nakamura et al. [14] in the AMT group in our study, both the transconjunctival pathway (lower bleb wall reflectivity and microcyst formation) and subconjunctival pathway (greater fluid-filled space score) were significantly associated with good IOP control in multivariate logistic regression. Even if the eyes had a fluid-filled space score of one and the posterior margin of the fluid-filled space was closed, blebs with lower bleb wall reflectivity and microcyst formation showed successful IOP control. Therefore, in the AMT group, not only subconjunctival absorption but also transconjunctival pathway may play an important role in aqueous outflow.

The conflicting findings between this study and the study by Nakamura et al. [14] may be due to differences in demographic and ocular characteristics, surgical technique, and imaging modality. It is recognized that the conjunctival flap technique, such as fornix-based conjunctival flap and limbal-based conjunctival flap, affects bleb morphology [3, 21, 22]. A more diffuse posteriorly draining bleb was formed in the fornix-based trabeculectomy, whereas more cystic blebs and/or ring of steel with more anterior drainage were formed with limbal-based trabeculectomy [3]. The fact that a limbal-based conjunctival flap was made in the study by Nakamura et al. [14] and a fornix-based conjunctival flap was made in this study for trabeculectomy with AMT, suggest that different types of conjunctival incision may result in distinct features of the filtering bleb. In the study by Nakamura et al. [14] among the eyes with AMT, the median number of previous intraocular procedures was two and only 6/28 (21.4%) eyes had POAG. Conversely, in our study, there were no eyes with a history of intraocular surgery, and a diagnosis of POAG was made in 100% of the eyes with AMT. A history of previous ocular surgery and secondary glaucoma may have an impact on bleb morphology [45]. In the study by Nakamura et al. [14], bleb imaging was performed using UBM with a resolution of 50 μm. However, we analysed intrableb structures using AS-OCT with higher resolution than that of UBM, and AS-OCT enabled us to measure intrableb parameters and analyse bleb wall reflectivity quantitatively using the IMAGEJ software.

There are some limitations to the present study. As this was retrospective study, only a small sample of patients with unsuccessful IOP control was included (8 eyes in the control group and 13 eyes in the AMT group). Therefore, the findings of this study should be interpreted with caution. However, there was no significant difference in sex, age at trabeculectomy, IOP at AS-OCT test, preoperative IOP, number of preoperative glaucoma medications, preoperative BCVA, preoperative lens status, CCT, axial length, spherical equivalent, and visual field parameters between the control and AMT group.

Although the Spectralis AS-OCT used in this study has higher resolution and speed than UBM or time domain OCT such as the Visante OCT (Carl Zeiss AG, Germany) [46, 47], the Spectralis AS-OCT uses shorter wavelength light sources of 870 nm, resulting in a penetration depth of 1.9 mm compared with 6 mm of Visante OCT and 4–7 mm of UBM [46, 47]. Therefore, the possibility that the structure beyond 1.9 mm depth may not be included in the OCT image cannot be excluded. However, the AS-OCT image was taken with EDI and ART modes on to improve image quality, and only OCT images >25 dB were included in the analysis. Manual adjustment of the contrast setting allowed us to identify the scleral edge when the scleral margin was not apparent in the OCT image.

An earlier diagnosis of non-functioning bleb may help direct proper management in the postoperative period after trabeculectomy because bleb management procedures are more effective when administered as soon as possible if needed [48]. In the previous study, the tendency towards encapsulation was seen much earlier with AS-OCT, showing thinner bleb wall thickness and higher bleb cavity height (referred as fluid-filled space height in the present study) at 1–2 weeks postoperatively, compared with functioning bleb [17]. However, the interval between trabeculectomy and the time of AS-OCT examinations in this study was 2.48 ± 0.73 years in the control group and 2.23 ± 0.74 years in the AMT group. An early bleb evaluation with AS-OCT at 1–2 weeks was not performed in this study. For future research, we plan to evaluate early intrableb parameters associated with long-term IOP control and bleb management procedures.

We used a stringent success criterion for success in this study. Previous studies suggested that IOP <21 mm Hg may not be optimal success criteria for trabeculectomy [49, 50]. Recent clinical trials have adopted cut off points of IOP <18 mm Hg based on long-term outcomes of glaucoma surgery [51, 52]. Nevertheless, several post hoc analyses with more stringent IOP criteria may be needed to apply the results generalizable to patients with glaucoma.

Finally, the long-term follow-up periods might affect the intrableb parameters as described in the previous study by Narita et al. [16]. Bleb height, bleb wall thickness, striping layer thickness, and other parameters changed over time. Therefore, the findings of the intrableb parameters in this study should be interpreted with caution.

Bleb wall reflectivity was significantly associated with bleb function in both the control and AMT group. However, the fluid-filled space score was associated with IOP control only in the AMT group. The transplanted amniotic membrane in trabeculectomy may be involved in IOP control by enhancing aqueous flow through the subconjunctival absorption, as well as by maintaining flow through the transconjunctival pathway.

We would like to thank Editage (www.editage.co.kr) for English language editing.

The study was conducted according to the guidelines of the World Medical Association Declaration of Helsinki, and approved by the Institutional Review Boards of Pusan National University Hospital (approval No. 2111-004-108). All patients gave written informed consent for the surgical procedures and for their information to be stored in the hospital database and used for research.

The authors have no conflicts of interest to declare.

This research was supported by a grant from Medical big data and AI-based early detection of visual dysfunction funded by Busan and managed by Busan Techno Park and by the Patient-Centered Clinical Research Coordinating Center, funded by the Ministry of Health and Welfare, Republic of Korea (Grant no. HI19C0481, HC19C0276). The funding agencies have played no role in this research.

Jiwoong Lee contributed to conceptualization and involved in writing, review, and editing; Jiwoong Lee and Sangwoo Moon contributed to methodology; Sangwoo Moon contributed to investigation and involved in writing original draft preparation; Jiwoong Lee, Sangwoo Moon, and Jinmi Kim contributed to data curation. All authors have read and approved the final manuscript and agreed to publish the manuscript.

The data generated or analysed during this study are available from the corresponding author [Jiwoong Lee] upon reasonable request.