The Adjunctive Effect of Goniotomy Combined with Phacoemulsification and Goniosynechialysis for Advanced Primary Angle-Closure Glaucoma with 360° Angle Closure: 12-Month Outcomes

Abstract

Purpose

To compare phacoemulsification with intraocular lens implantation and goniosynechialysis, with or without adjunctive goniotomy, in advanced primary angle-closure glaucoma with complete circumferential angle closure.

Design

Retrospective comparative study.

Methods

This study included 61 eyes undergoing phacoemulsification and goniosynechialysis without (N=31) or with adjunctive goniotomy (N=30), with at least 12 months of follow-up. A peripheral anterior synechiae of 360° was confirmed by gonioscopy and ultrasound biomicroscopy. The primary outcome was month-12 intraocular pressure (IOP). Longitudinal generalized estimating equations adjusted for baseline IOP, age, sex, and vertical cup-to-disc ratio. Secondary outcomes included medication use, surgical success, postoperative angle status, and complications.

Results

At month 12, mean IOP was 16.26 ± 3.03 mmHg in the group without goniotomy and 16.63 ± 3.29 mmHg in the group with goniotomy. The adjusted between-group difference was −0.46 mmHg (95% CI −1.97 to 1.04, P=0.54), below the post hoc exploratory margins of 4, 3, and 2 mmHg. Medication use had no significant difference between groups (P=0.89). Complete and qualified success rates were similar at IOP thresholds of ≤21, ≤18, and ≤15 mmHg combined with at least 20% IOP reduction, with comparable cumulative success rate. Hyphema occurred in 1/31 eyes (3.2%) and 9/30 eyes (30.0%) in groups without and with adjunctive goniotomy (absolute risk difference, 26.8 percentage points, 95% CI, 7.7 to 45.8, P<0.01).

Conclusions

In advanced PACG eyes with 360° angle closure, the differences between the 12-month IOP control, surgical success, and medication reduction of the two groups showed no statistically significance. Routine addition of goniotomy may not provide additional benefit.

Introduction

Primary angle-closure glaucoma (PACG) is a major cause of irreversible blindness worldwide, particularly in Asian populations.¹ It is characterized by the formation of peripheral anterior synechiae (PAS), where the peripheral iris adheres to the trabecular meshwork, obstructing aqueous humor outflow and leading to elevated intraocular pressure (IOP). In advanced disease, extensive PAS can lead to 360° circumferential angle closure and persistent impairment of aqueous humor outflow, resulting in markedly elevated IOP and progressive optic nerve damage.^2,3 When PACG coexists with cataracts, the thickened lens further exacerbates angle crowding. Although phacoemulsification with intraocular lens implantation (PEI) alone can effectively deepen the anterior chamber and widen the angle and release iridotrabecular contact, it cannot reverse permanent synechial closure or restore trabecular meshwork dysfunction. By mechanically separating PAS, GSL could restore the physiological drainage by re-exposing the trabecular meshwork.⁴ Recent studies have demonstrated that combined PEI and goniosynechialysis (GSL) can deepen the anterior chamber and achieve high and stable success rate, independent of the preoperative extent of PAS.^4–6

Despite the established success of combined PEI and GSL, the concern remains that chronic synechial adhesion may cause irreversible damage to the trabecular meshwork, particularly in advanced cases with 360° closure. Goniotomy (GT) has been increasingly incorporated into angle-based procedures to directly incise dysfunctional trabecular meshwork (TM) and thereby enhance aqueous outflow. Emerging evidence in PACG, particularly when GT is combined with PEI and GSL (phacogoniotomy), suggests that adding GT may provide additional IOP lowering and reduce postoperative medication dependence compared with approaches relying on angle reopening alone.^7–9 Multicenter studies and randomized trials in PACG with cataract have further supported the efficacy of combined PEI, GSL, and GT, reporting substantial and sustained postoperative IOP reduction with acceptable safety profiles.^10,11 Zhang et al has shown that phacogoniotomy is effective and safe in treating advanced PACG with cataract over a 2-year follow-up period.¹² Collectively, these findings imply that, beyond restoring angle configuration, directly addressing TM resistance via GT may be important for achieving lower postoperative IOP targets.

However, whether the addition of GT confers incremental IOP-lowering benefit over PEI+GSL alone in the specific setting of complete (360°) angle closure remains unclear. Therefore, this study aimed to directly compare PEI+GSL and PEI+GSL+GT in advanced PACG eyes with complete angle closure to determine whether the addition of GT provides greater and more durable IOP reduction and medication sparing without compromising safety.

Methods

Study Design and Participants

This retrospective comparative study was conducted at Nanfang Hospital in accordance with the tenets of the Declaration of Helsinki. Participants were recruited between December 2023 to December 2025. The study protocol was approved by the Ethics Committee of Nanfang Hospital (NO:NFEC-2024-290), and the written informed consent was obtained from the participants.

Medical records of consecutive patients diagnosed with PACG who underwent PEI+GSL or PEI+GSL+GT were reviewed. Eligible eyes were required to have PAS of 360° confirmed preoperatively by both gonioscopy and ultrasound biomicroscopy (UBM) and to have completed at least 12 months of postoperative follow-up. PACG was diagnosed based on 1) PAS ≥180° range, including the nasal and inferior quadrants; 2) IOP higher than 21mmHg with or without antiglaucoma medications, taken with Goldmann applanation tonometer; 3) glaucomatous optic neuropathy (vertical cup-to-disc (C:D) ratio ≥ 0.7 or C:D asymmetry ≥ 0.2, or rim width at the superior and inferior temporal areas <0.1 of the vertical diameters of the optic disc); and 4) advanced glaucomatous visual field defects (nasal step, arcuate scotoma and paracentral scotoma on a reliable Humphrey analyser and a mean deviation of ≤−12 dB).¹³ Eyes with secondary angle-closure glaucoma, prior incisional glaucoma surgery, coexisting ocular conditions affecting angle assessment or IOP measurement, or incomplete clinical or imaging data were excluded. When both eyes met the eligibility criteria, both eyes were included, and inter-eye correlation was accounted for in statistical analyses.

Surgical Procedures

All surgeries were performed by one experienced glaucoma surgeon following a standardized protocol. Before intraocular manipulation, the endpoints of the planned 120° goniotomy arc were marked on the corneal limbus with a sterile marking pen using a Mendez degree gauge, which is commonly used for toric intraocular lens axis marking. For example, the 2- and 6-o’clock positions were marked in a right eye. After phacoemulsification and goniosynechialysis, the limbal reference marks were used under direct gonioscopic visualization to delimit the corresponding start and end points of the angle incision, and the trabecular meshwork between these points was incised with a microhook. In both groups, PEI was first performed. Additional viscoelastic was introduced to ensure adequate chamber depth and visualization. Complete 360° GSL was performed by gently sweeping peripheral anterior synechiae circumferentially using a chopper. In the PEI+GSL+GT group, 120° GT was additionally performed following GSL. Goniotomy was subsequently performed in either the nasal or inferior quadrant based on surgeon preference. For nasal goniotomy, the surgeon was positioned temporally, and the patient’s head was rotated 35°– 40° away from the surgeon. The operating microscope was angled approximately 30° toward the surgeon. Viscoelastic was applied to the corneal surface, and a surgical goniolens was placed to visualize the anterior chamber angle and identify the trabecular meshwork. A trabecular microhook (TMH) was then introduced into the anterior chamber through the main corneal incision. The hook tip was engaged within Schlemm canal and advanced to incise the inner wall and trabecular meshwork over an arc of approximately 120°. The 120° target was selected to align with protocols used in previous multicentre studies and a randomized trial of phacogoniotomy for advanced PACG, and because direct comparative evidence in POAG suggests that increasing the goniotomy extent beyond 120° does not improve IOP or medication outcomes but may increase postoperative hyphema.^12,14 Upon completion, viscoelastic material and intraoperative blood were thoroughly irrigated and aspirated, and the corneal incisions were sealed by stromal hydration.

Postoperative management was similar between groups and included topical antibiotics and corticosteroids, with antiglaucoma medications added as needed based on postoperative IOP.

Clinical Examinations and Follow-Up

Patients underwent comprehensive ophthalmic examinations preoperatively and at postoperative day 1, week 1, and months 1, 3, 6, and 12. IOP was measured using Goldmann applanation tonometry. An IOP spike was defined as an IOP >30 mmHg and/or an increase of >10 mmHg above baseline within the first postoperative week. The number of glaucoma medications, best-corrected visual acuity, slit-lamp findings, and postoperative complications were recorded.

Angle Assessment

Preoperative angle status was assessed using gonioscopy and confirmed by UBM to ensure complete angle closure. At 12 months postoperatively, angle configuration was evaluated using both gonioscopy and anterior segment optical coherence tomography (CASIA II; Tomey Corporation). Gonioscopic assessment documented angle opening in all quadrants and overall extent, with CASIA used to corroborate findings.

Anterior chamber imaging with the CASIA2 system was performed following established protocols.¹⁵ During acquisition, participants fixated on an internal target, and images were captured over a 5-second period. A total of 128 anterior segment OCT images covering the full 360° circumference were subjected to three-dimensional analysis. Images compromised by eyelid interference or motion artifacts were excluded. Circumferential angle parameters were automatically generated by the system. For quadrant-based evaluation of right eyes (nasal, superior, temporal, and inferior), images were rotated counterclockwise from 0° to 360°, while images from left eyes were horizontally mirrored to ensure consistency. Quantitative assessments included anterior chamber angle opening distance (AOD), trabecular–iris space area (TISA), trabecular–iris angle (TIA), angle recess area (ARA), iridotrabecular contact (ITC), and the iridotrabecular contact index (ITC index), all defined according to previously published criteria.^15–17 Identification of the scleral spur was independently performed by two experienced ophthalmologists. Interobserver agreement for CASIA-derived angle measurements obtained following independent scleral-spur identification by two ophthalmologists was assessed.

Outcome Measures

The primary outcome was IOP at postoperative 12 months. The adjusted between-group difference, defined as PEI+GSL minus PEI+GSL+GT (positive values indicating higher IOP with PEI+GSL), was estimated using a longitudinal generalized estimating equation model including treatment group, visit, and their interaction, with adjustment for baseline IOP, age, sex, and baseline vertical cup-to-disc ratio. Robust standard errors were clustered by patient to account for repeated measurements and the inclusion of both eyes from some patients. Because the study was retrospective and neither an equivalence or noninferiority margin nor a target sample size was prospectively prespecified, no equivalence or noninferiority testing was included in the revised analysis. Secondary outcomes included IOP reduction from baseline, number of glaucoma medications, surgical success rates, postoperative complications, and postoperative angle status. In accordance with WGA recommendations,¹⁸ failure is defined as IOP outside a prespecified range on two consecutive visits. In our study, percentage of IOP reduction was calculated as 100 × (baseline − postoperative IOP)/baseline IOP. Success required IOP of 6–15, 6-18 and 6-21 mmHg and ≥20% reduction, respectively. Complete success requires no medication, whereas qualified success permits medication. Exceeding the upper limit, <20% reduction, or IOP ≤5 mmHg on two consecutive visits from month 3 constituted failure. Reoperation meant an additional operating-room glaucoma procedure.¹⁹ Vision-threatening complications, including endophthalmitis, suprachoroidal hemorrhage, clinically significant choroidal detachment, hypotony maculopathy, malignant glaucoma, retinal detachment, and corneal decompensation, were reported separately and counted as failure if they caused loss of light perception or required reoperation.

Statistical Analysis

No priori sample-size calculation was performed because this was a retrospective review. All consecutive eligible eyes treated during the study period and completing the required follow-up were included. The present sample size yielded an approximate minimum detectable between-group difference of 2.15 mmHg at 80% power and a two-sided α of 0.05, assuming the observed patient-clustered robust standard error. These calculations describe the achieved precision of this dataset and do not convert the retrospective analysis into a confirmatory equivalence or noninferiority trial. Normality was assessed using the Shapiro–Wilk test. Normally distributed data are presented as mean ± SD and non–normally distributed data as median (IQR). Between-group comparisons were performed using generalized estimating equations (GEE) to account for inter-eye correlation. A longitudinal GEE with Gaussian identity link, exchangeable working correlation, and robust patient-clustered standard errors included treatment group, postoperative visit, a treatment-by-visit interaction, baseline IOP, age, sex, and baseline cup-to-disc ratio. The marginal risk difference for postoperative hyphema was estimated using an identity-link binomial GEE with an independence working correlation and robust standard errors clustered by patient to account for the inclusion of both eyes from some patients. Kaplan–Meier survival analysis with the log-rank test was used to compare success rates. Multivariate analysis determines the predictors of angle status and IOP at postoperative 12 months was performed using GEE. A 2-sided P < 0.05 was considered statistically significant. All the statistical analysis was performed by R (version 4.5.1) and Python (version 3.11.9). The figures were made by GraphPad Prism 10.

Results

A total of 61 eyes from 59 patients were included in the analysis. Adjunctive goniotomy was performed at the discretion of the operating surgeon, based on the gonioscopic results of the angle preoperatively and on the patients’ preference at preoperative counselling. No protocol-specified allocation rule was applied. Consequently, preferential selection of more complex eyes or selection based on unmeasured clinical or socioeconomic factors cannot be excluded.

Thirty-one eyes were assigned as PEI+GSL group and the other 30 eyes were assigned as PEI+GSL+GT group. Baseline demographic and ocular characteristics of the two groups were shown in Table 1. There were no statistically significant differences in sex, age, mean deviation and pattern standard deviation of visual field, axial length, anterior chamber depth, baseline IOP, or preoperative glaucoma medication use (all P > 0.05). Except for disease duration (absolute SMD=0.316), all measured preoperative characteristics had absolute SMDs below 0.25, indicating no substantial measured imbalance, although residual confounding cannot be excluded.

Table 1.

Baseline Characteristics

Variable	PEI+GSL (n=31)	PEI+GSL+GT (n=30)	P	Absolute SMD
Gender (Male), n (%)	14 (45.2%)	14 (46.7%)	0.71	0.030
Age, years	62.23 ± 10.55	62.20 ± 8.16	0.82	0.003
BCVA LogMAR	1.03 ± 1.11	0.92 ± 0.98	0.17	0.105
Vertical C/D ratio	0.80 (0.70–0.90)	0.80 (0.70–1.00)	0.87	0.000
MD (dB)	−15.34 ± 3.22 (−21.79 to −12.01)	−15.80 ± 2.75 (−21.68 to −12.01)	0.89	0.153
PSD (dB)	6.62 ± 2.13 (3.19 to 9.75)	6.72 ± 1.91 (3.61 to 9.53)	0.84	0.050
Duration (month)	5.00 (2.00–12.00)	2.50 (0.81–6.00)	0.18	0.316
AL (mm)	22.17 ± 0.94	22.35 ± 0.69	0.41	0.210
CCT (μm)	511 (504–538)	550 (532–562)	0.71	0.241
ACD-before (mm)	1.98 ± 0.46	2.03 ± 0.28	0.78	0.142
ACD-after (mm)	3.52 ± 0.26	3.46 ± 0.25	0.61	0.215
PEI open range (°)	150 (0–180)	30 (0–180)	0.08	0.390
Basic IOP (mmHg)	50 (39–58)	50 (40–60)	0.79	0.026
PRE-IOP (mmHg)	24 (17–31)	26 (17–35)	0.55	0.158
PRE-MED (n)	3.00 (2.00–4.00)	3.00 (3.00–4.00)	0.87	0.064

Notes: Table 1 presents baseline and intraoperative characteristics. Data are presented as mean ± SD for normally distributed continuous variables, median (IQR) for nonnormally distributed continuous variables, or n (%) for categorical variables. Between-group P values were obtained using generalized estimating equations with robust standard errors clustered by patient. The coefficient test compares PEI+GSL+GT with PEI+GSL. Absolute SMDs quantify between-group imbalance independently of sample size and should be interpreted alongside P values. Angle-opening extent after PEI was measured intraoperatively after lens extraction and intraocular lens implantation and before GSL. Two-sided P<0.05 was considered statistically significant; statistically significant P values are shown in bold. Baseline IOP was the highest preoperative IOP recorded for each eye.

Abbreviations: ACD, anterior chamber depth; AL, axial length; BCVA, best-corrected visual acuity; CCT, central corneal thickness; C:D ratio, cup-to-disc ratio; GEE, generalized estimating equation; GSL, goniosynechialysis; GT, goniotomy; IOP, intraocular pressure; IQR, interquartile range; logMAR, logarithm of the minimum angle of resolution; MD, mean deviation; PEI, phacoemulsification with intraocular lens implantation; PSD, pattern standard deviation; SD, standard deviation; SMD, standardized mean difference.

Postoperative IOP was significantly reduced from baseline in both groups at all follow-up visits (Table 2). In a longitudinal GEE, the treatment-by-visit interaction was not statistically significant (Wald chi-square=8.94, 5 df; P=0.11). At month 12, adjusted IOP was 16.14 mmHg with PEI+GSL and 16.60 mmHg with PEI+GSL+GT (difference −0.46 mmHg; 95% CI −1.97 to 1.04; P=0.54). From postoperative day 1 through 12 months, no significant differences in mean IOP were observed between the PEI+GSL group and the PEI+GSL+GT group (all P > 0.05), indicating comparable IOP control over the entire follow-up period (Figure 1). Both groups demonstrated a marked and sustained reduction in IOP, with mean decreases of approximately 30–32 mmHg, corresponding to a 61%–72% reduction from baseline through 12 months (Table 2). The magnitude of absolute and percentage IOP reduction was similar between groups at all postoperative time points (all P > 0.05).

Table 2.

Intraocular Pressure and Reduction from Baseline After Surgery in the Two Groups

Time Point	PEI+GSL (n=31)				PEI+GSL+GT (n=30)				Unified Longitudinal GEE Between-group comparison
	IOP	Abs. Red.	% Red.	P	IOP	Abs. Red.	% Red.	P	Difference (95% CI)	P
Baseline	50.00 (39.00–58.00)	-	-	-	50.00 (40.00–60.00)	-	-	-	-	-
Post-op Day 1	16.00 (12.50–20.00)	29.58 ± 12.32	64.43 (55.76–74.23)	<0.01	12.50 (10.00–19.25)	32.20 ± 16.16	71.67 (50.78–80.39)	<0.01	2.43 (−1.62 to 6.48)	0.24
Post-op Day 7	15.52 ± 3.61	31.94 ± 10.29	66.05 ± 10.17	<0.01	14.00 (13.00–15.75)	32.57 ± 14.35	70.21 (57.36–73.91)	<0.01	0.19 (−1.54 to 1.92)	0.83
Post-op 1 Month	15.23 ± 2.87	32.23 ± 10.77	69.27 (61.72–73.03)	<0.01	15.00 (13.25–16.75)	32.07 ± 14.94	69.80 (57.42–75.00)	<0.01	−0.50 (−2.52 to 1.52)	0.63
Post-op 3 Months	15.00 (14.00–18.00)	30.48 ± 11.15	65.96 (56.64–72.82)	<0.01	15.57 ± 3.57	31.83 ± 13.68	67.63 (55.04–74.83)	<0.01	1.05 (−1.38 to 3.49)	0.40
Post-op 6 Months	16.26 ± 3.17	31.13 ± 10.97	65.22 (60.38–73.08)	<0.01	16.00 (14.25–19.00)	30.63 ± 13.38	63.37 (54.90–73.00)	<0.01	−0.81 (−2.25 to 0.63)	0.27
Post-op 12 Months	16.26 ± 3.03	31.13 ± 11.16	67.30 (59.50–72.63)	<0.01	16.63 ± 3.29	30.77 ± 13.73	61.43 (56.31–71.65)	<0.01	−0.46 (−1.97 to 1.04)	0.54

Notes: Data are presented as mean ± SD for normally distributed variables or median (IQR) for nonnormally distributed variables, based on Shapiro–Wilk tests. Baseline IOP was the highest preoperative IOP recorded for each eye. Absolute IOP reduction was calculated as baseline IOP minus postoperative IOP, and percentage IOP reduction as 100 × (baseline IOP − postoperative IOP)/baseline IOP. P values were derived from a longitudinal GEE model including group, visit, and the group-by-visit interaction, adjusted for baseline IOP, age, sex, and baseline vertical C:D ratio, with robust standard errors clustered by patient. Between-group differences are PEI+GSL minus PEI+GSL+GT; positive values indicate higher IOP with PEI+GSL. Statistically significant P values are shown in bold (P<0.05).

Abbreviations: Abs. Red., absolute IOP reduction; CI, confidence interval; GEE, generalized estimating equation; GSL, goniosynechialysis; GT, goniotomy; IOP, intraocular pressure; IQR, interquartile range; PEI, phacoemulsification with intraocular lens implantation; % Red., percentage IOP reduction; SD, standard deviation.

Figure 1.

Comparison of intraocular pressure between the two groups after surgery. Figure 1 shows the changes in the intraocular pressure over postoperative follow-up. Values are shown at different postoperative time points for the two groups. Data are presented as mean ± standard deviation. Blue squares represent PEI+GSL group and red triangles represent PEI+GSL+GT group. No significant differences were found between the two group throughout the 12-months follow-up.

Abbreviations: PEI, phacoemulsification; GSL, goniosynechialysis; GT, goniotomy; Post-op, postoperative.

The number of glaucoma medications decreased significantly after surgery in both groups and remained low throughout follow-up (Table 3). No significant differences in postoperative medication use were observed between the two surgical groups at any visit (all P > 0.05), suggesting comparable medication-sparing effects up to 12 months (Supplementary Figure 1).

Table 3.

Number of Medications After Surgery in the Two Groups

Time Point	PEI+GSL (n=31)		PEI+GSL+GT (n=30)		Unified Longitudinal GEE Between-group Comparison
	Medications	P	Medications	P	Difference (95% CI)	P
Baseline	3.0 ± 1.0	-	2.9 ± 1.0	-	-	-
Post-op Day 1	0.13 ± 0.50	<0.01	0.20 ± 0.81	<0.01	−0.07 (−0.40 to 0.26)	0.68
Post-op Day 7	0.39 ± 0.80	<0.01	0.37 ± 0.93	<0.01	0.10 (−0.22 to 0.43)	0.53
Post-op 1 Month	0.32 ± 0.75	<0.01	0.33 ± 0.88	<0.01	0.07 (−0.26 to 0.40)	0.68
Post-op 3 Months	0.55 ± 1.09	<0.01	0.40 ± 0.93	<0.01	0.24 (−0.17 to 0.66)	0.25
Post-op 6 Months	0.71 ± 1.30	<0.01	0.53 ± 1.14	<0.01	0.28 (−0.23 to 0.78)	0.28
Post-op 12 Months	0.65 ± 1.17	<0.01	0.70 ± 1.24	<0.01	0.04 (−0.46 to 0.53)	0.89

Notes: Data are presented as mean ± SD. Within-group P values were derived from a longitudinal GEE model including baseline and all six postoperative visits, with group, visit, and the group-by-visit interaction and robust standard errors clustered by patient. Each P value compares a postoperative visit with baseline within that group and was Holm-adjusted across the six contrasts. Between-group estimates (PEI+GSL minus PEI+GSL+GT) were obtained from a separate longitudinal GEE model including all six postoperative visits, group, visit, and the group-by-visit interaction, adjusted for baseline glaucoma medication count, age, sex, and baseline vertical C:D ratio; positive values indicate a higher medication count with PEI+GSL. Visit-specific between-group P values are two-sided and unadjusted. Statistically significant P values are shown in bold (P<0.05).

Abbreviations: CI, confidence interval; GEE, generalized estimating equation; GSL, goniosynechialysis; GT, goniotomy; PEI, phacoemulsification with intraocular lens implantation; SD, standard deviation.

At the postoperative 12 months, no significant differences in surgical success rates were observed between groups (P>0.05). With IOP ≤ 21mmHg as the criterion combined with at least 20% IOP reduction, complete success rate was 64.5% in PEI+GSL group and 60.0% in PEI+GSL+GT group. The qualified success rate reached 80.6% and 86.7%, respectively. With IOP ≤ 18mmHg and at least 20% IOP reduction, complete success was decreased to 58.1% and 60.0%, qualified success decreased to 67.7% and 70.0% in PEI+GSL and PEI+GSL+GT group, respectively. While setting the IOP threshold to 15mmHg, complete success were 25.8% and 23.3%, and qualified success were 25.8% and 26.7%, respectively. Statistical analysis demonstrated comparable efficacy between PEI+GSL and PEI+GSL+GT groups in terms of success rate at the postoperative 12 months. Based on predefined IOP criteria, both complete and partial surgical success rates were similar between the PEI+GSL and PEI+GSL+GT groups (Figure 2). Kaplan–Meier survival analyses demonstrated a gradual decline in cumulative success probabilities over time in both groups. For both complete and qualified success, no statistically significant differences were observed between the two procedures at IOP thresholds of ≤21 mmHg, ≤18 mmHg, or ≤15 mmHg throughout the 12-month follow-up period (log-rank test, all P > 0.05). The survival curves showed largely overlapping trajectories across all success definitions, indicating comparable durability of surgical success between the two surgical approaches.

Figure 2.

Kaplan-Meier survival analysis of the two groups. Figure 2 shows the kaplan-Meier survival curves comparing treatment outcomes between PEI+GSL (blue line) and PEI+GSL+GT (red line) groups at different IOP thresholds. The figure displays six panels showing probability of survival over 12 months for both complete success and qualified success criteria at three different IOP cutoff values (≤15mmHg, ≤18mmHg, and ≤21mmHg) combined with at least 20% IOP reduction. Complete success indicates IOP control without additional medication, while qualified success allows for supplemental medical therapy. P-values represent log-rank test results comparing the two treatment groups. No statistically significant differences were observed between groups across any threshold (all P>0.05).

Abbreviations: PEI, phacoemulsification; GSL, goniosynechialysis; GT, goniotomy; Post-op, postoperative.

Gonioscopic evaluation at 12 months demonstrated comparable postoperative angle opening between the two groups across all quadrants as well as in overall extent (Supplementary Table 1). No statistically significant differences were observed in superior, nasal, inferior, temporal, or total angle opening between the PEI+GSL and PEI+GSL+GT groups (all P > 0.05), although a trend toward less superior angle opening was noted in the combined procedure group. The CASIA examination also demonstrated comparable anterior chamber angle parameters between the two groups (Supplementary Table 2). No statistically significant differences were observed between the PEI+GSL and PEI+GSL+GT groups in overall or quadrant-specific measurements of angle opening distance (AOD), trabecular–iris space area (TISA), angle recess area (ARA), iridotrabecular contact (ITC), or trabecular–iris angle (TIA) across the superior, nasal, inferior, and temporal quadrants (all P > 0.05). These findings indicate similar postoperative anatomic restoration of the anterior chamber angle between the two surgical approaches at 12 months. Interobserver ICC estimates ranged from 0.881 to 1.000 across the 56 quadrant-specific CASIA indicators, indicating good to excellent interobserver agreement (Supplementary Table 3).

After adjusted for age, gender, cup-to-disc ratio, and medication use at 12 months, the multivariate generalized estimating equation analyses (Table 4) showed that postoperative angle opening at 12 months was not significantly associated with baseline IOP, the range of angle opened after performing phacoemulsification without goniosynechialysis, or types of preoperative medication (all P > 0.05). However, a greater extent of goniotomy was independently associated with lower IOP at 12 months (β = −0.014, P = 0.01), indicating that each additional 10° of goniotomy was associated with an approximate 0.14 mmHg lower in IOP.

Table 4.

Multivariate Analysis of Factors Associated with the Angle Opening Range and Intraocular Pressure at Postoperative 12 months

	Post-op 12 Months Angle Open Range (n=61)		Post-op 12 Months IOP (n=61)
	β (95% CI)	P	β (95% CI)	P
Basic IOP	−0.07 (−1.794, 1.654)	0.94	−0.042 (−0.114, 0.030)	0.25
PEI Angle Open Range	0.134 (−0.15, 0.418)	0.36	0.006 (−0.005, 0.016)	0.28
Pre-op Medications	−6.395 (−29.474, 16.683)	0.59	0.322 (−0.377, 1.022)	0.37
GT Range	−0.115 (−0.476, 0.246)	0.53	−0.014 (−0.025, −0.003)	0.01

Notes: Exploratory multivariable GEE models were adjusted for baseline IOP, angle-opening extent after PEI, angle-opening extent after GSL, glaucoma medication count at month 12, preoperative glaucoma medication count, recorded goniotomy extent, age, sex, baseline vertical C:D ratio, and disease duration. Baseline IOP was the highest preoperative IOP recorded for each eye. Angle-opening extent after PEI was measured after PEI and before GSL. Statistically significant P values are shown in bold (P<0.05).

Abbreviations: β, regression coefficient; CI, confidence interval; C:D ratio, cup-to-disc ratio; GEE, generalized estimating equation; GSL, goniosynechialysis; GT, goniotomy; IOP, intraocular pressure; PEI, phacoemulsification with intraocular lens implantation.

As for postoperative complications, hyphema was more frequently observed in the PEI+GSL+GT group (30.0%) than in the PEI+GSL group (3.2%). The incidence of postoperative IOP spikes did not differ significantly between the two groups (6.0% [2/31]^10,20–22 vs 10.0% [3/30], P = 0.98). The absolute risk was 26.8% points higher with PEI+GSL+GT (95% CI 7.7 to 45.8, P<0.01). Corneal edema was common in both groups, with comparable rates between the PEI+GSL and PEI+GSL+GT groups (61.3% [19/31] vs 53.3% [16/30], P = 0.94). Similarly, postoperative exudation or adhesion showed no significant difference between groups (16.1% [5/31] vs 16.7% [5/30], P = 0.94).

Discussion

In this retrospective comparative study, we specifically enrolled 61 eyes from 59 patients with advanced PACG characterized by complete 360° circumferential angle closure, a population that has been largely underrepresented in the existing literature. Most prior studies investigating combined phacoemulsification and angle surgery have stratified patients by PAS extent ranging from 180° to 270°, or have included heterogeneous cohorts spanning a broad spectrum of angle closure severity.^4,10 While these studies have provided valuable insights, their inclusion of partially closed angles introduces a significant confound: residual open segments of the trabecular meshwork may independently contribute to IOP reduction following lens extraction alone, thereby obscuring the true incremental benefit of adjunctive angle procedures. By strictly selecting patients with complete circumferential angle closure, our study addresses this critical gap and provides a more rigorous assessment of whether the addition of goniotomy to PEI+GSL offers meaningful benefit in the most anatomically severe phenotype. Both procedures were associated with substantial reductions in IOP and medication use. At month 12, the adjusted PEI+GSL minus PEI+GSL+GT difference was −0.46 mmHg (95% CI −1.97 to 1.04). The upper confidence bound was below the post hoc exploratory margins of 4, 3, and 2 mmHg. Notably, these findings do not establish equivalence or confirmatory noninferiority and require validation in an adequately powered randomized trial. Secondary outcomes, including cumulative complete and qualified success rates across multiple IOP thresholds, and overall complication profiles, were also comparable between groups.

Extensive evidence supports the capacity of phacoemulsification to facilitate spontaneous angle opening across various degrees of PAS. Xiong et al demonstrated in a multicenter prospective study that PEI-GSL is safe and effective regardless of preoperative PAS extent.⁴ A systematic review and meta-analysis further confirmed the superiority of PEI-GSL over PEI alone in IOP reduction (MD: −1.73 mmHg, 95% CI: −3.53 to −0.13), though both approaches yielded comparable success rates and similar reductions in medication burden (MD: −0.14), with GSL primarily contributing through anatomic improvement of the angle.²³ In terms of preoperative PAS range, previous studies have consistently reported that the extent of PAS did not influence the final IOP or reduction in glaucoma medications.^4,10 One representative study categorized patients into three groups by PAS extent and found no statistically significant difference in final IOP among groups (13.8 ± 3.2 vs 14.3 ± 3.0 vs 13.6 ± 2.8 mmHg, P = 0.594), with comparable success rates regardless of initial PAS severity.⁴ These results collectively suggest that GSL combined with PEI is the dominant contributor to IOP control, and that routine addition of GT does not confer a clear advantage in eyes with 360° angle closure in advanced PACG.

The management of advanced PACG with complete angle closure remains a clinical challenge, and the role of adjunctive trabecular procedures in this setting continues to evolve. GT has been evaluated as a strategy to further enhance aqueous outflow by directly targeting trabecular resistance, and prior studies in primary open-angle glaucoma and juvenile open-angle glaucoma have demonstrated meaningful IOP reduction following this procedure.²⁴ Several multicenter and comparative studies have corroborated these benefits, reporting high qualified success rates, substantial medication sparing, and sustained IOP control at short- to mid-term follow-up.^12,25–28 In PACG eyes without coexisting cataract, combined GSL and goniotomy have similarly shown favorable outcomes.^22,29 Our findings should also be considered in the context of other trabecular procedures. In primary angle closure, GATT combined with goniosynechialysis lowered IOP and reduced medication use from baseline, but did not outperform previously reported outcomes after goniosynechialysis with phacoemulsification or phacoemulsification alone, which is consistent with our result.³⁰ Larger series have reported favourable outcomes for GATT alone in PACG (103 eyes; 78% success at 24 months),³¹ and KDB-assisted goniosynechialysis with excisional goniotomy has been shown to achieve sustained reduction through 24 months.³² These studies, however, were single-arm series without a non-goniotomy comparator, and therefore cannot separate the specific incremental effect of trabecular incision. Evidence for TrabEx+ and OMNI comes almost entirely from open-angle glaucoma: TrabEx+ achieved a ≥20% IOP reduction in 72% of eyes,³³ while GEMINI reported a 34% reduction in unmedicated mean diurnal IOP in mild-to-moderate OAG.³⁴ However, GT is a technically demanding procedure that requires specialized surgical skills and a steep learning curve. The procedure necessitates direct visualization and manipulation of the trabecular meshwork under gonioscopic guidance, demands precise instrument control, and carries an inherent risk of intraoperative hemorrhage from Schlemm’s canal.^27,28 Indeed, our study is consistent with prior reports of trabecular incision–based procedures. Other postoperative complications, including IOP spikes, corneal edema, and inflammatory reactions, occurred at similar rates and were generally self-limited. From a mechanistic perspective, PEI addresses lens-related contributors to angle closure by deepening the anterior chamber and relieving relative pupillary block, while GSL reopens the angle by mechanically separating peripheral anterior synechiae, restoring access to the trabecular meshwork.^4,35 The similar pressure outcomes may be consistent with residual conventional outflow after lens extraction and mechanical angle reopening.^4,10 However, trabecular function was not measured, and the present clinical data cannot establish the underlying mechanism. Only 25.8% and 23.3% of eyes remained continuously successful without medication at the ≤15-mmHg threshold through 12 months. For context, the PVP randomized trial reported an 84.6% complete-success rate after phacotrabeculectomy at 24 months, although it used the less stringent criterion of an IOP of 5–18 mmHg with at least a 20% reduction from baseline and no medication.¹⁴ This result indicates that many advanced PACG eyes require continued surveillance, medication, or additional pressure-lowering intervention despite substantial average IOP reduction.

An additional noteworthy finding of our study is that the intraoperative extent of angle opening achieved during PEI did not significantly influence postoperative IOP or angle outcomes. The association between goniotomy extent and postoperative-12-month IOP was small, corresponding to approximately 0.14 mmHg per additional 10° of goniotomy, based on the multivariate generalized estimating equation analyses. Given the retrospective design, this exploratory finding should not be interpreted as evidence of a clinically important or causal dose-response relationship. This finding is in line with previous reports indicating that the extent of PAS does not predict final IOP after combined phacoemulsification and angle surgery.^4,10 Collectively, these data support the interpretation that lens extraction and mechanical angle restoration represent the dominant determinants of long-term IOP control in eyes with complete angle closure, whereas the added value of GT after adequate GSL may be marginal at the cohort level.

Several limitations of this study should be acknowledged. First, the sample size was relatively small, and no a priori sample-size calculation was performed because of the retrospective design. Although the available sample yielded an approximate minimum detectable between-group difference of 2.15 mmHg at 80% power, based on the observed patient-clustered robust standard error, this post hoc estimate reflects only the achieved precision of the dataset and should not be interpreted as evidence of equivalence or noninferiority. Second, the study population consisted exclusively of Chinese patients, which may limit the generalizability of the findings to other ethnic groups with different anatomic and epidemiologic characteristics of PACG. Third, allocation was nonrandom and reflected the surgeon’s preoperative gonioscopic assessment, confounding by indication therefore cannot be excluded, and the absence of a detected difference should not be interpreted as evidence that adjunctive goniotomy confers no benefit. The potential selection bias underscore the need for future prospective studies or randomized clinical trials to confirm these findings and to further define the role of adjunctive goniotomy in eyes with complete angle closure.

Conclusion

Both procedures achieved substantial and durable IOP reduction in eyes with advanced PACG and coexisting cataract. Adding 120° goniotomy routinely did not provide further IOP lowering and was associated with a higher rate of transient hyphema. Its use is better reserved for eyes showing evidence of greater trabecular dysfunction, where goniosynechialysis alone cannot relieve trabecular outflow resistance. Given the retrospective, nonrandomized design, these findings do not establish either the absence of benefit or a causal increase in hyphema and require confirmation in randomized studies. Further prospective studies are needed to develop objective measures of trabecular dysfunction and determine whether these measures can guide a more individualized choice of surgical procedure.

Funding Statement

This work was supported by the Clinical Research Funding of Nanfang Hospital, Southern Medical University, China (2025CR009, HYZ), the Science and Technology Project in Baiyun District, Guangzhou (2025-YL-005, CXL) and President Foundation of the People’s Hospital of Baiyun District Guangzhou (BYYZ25009, CXL).

Ethics Approval and Consent to Participate

This study was performed in accordance with the tenets of the Declaration of Helsinki. It had been approved by Nanfang Hospital (NO:NFEC-2024-290). All the patients’ consent were obtained.

Author Contributions

Yuqiao Zhang and Wenhui Dong are co-first authors and contributed equally to this work. Di Wang and Hongyang Zhang are co-corresponding authors and contributed equally to this work as co-corresponding authors. All authors made a significant contribution to the work reported, whether that is in the conception, study design, execution, acquisition of data, analysis and interpretation, or in all these areas; took part in drafting, revising or critically reviewing the article; gave final approval of the version to be published; have agreed on the journal to which the article has been submitted; and agree to be accountable for all aspects of the work. HYZ and DW designed the study and had full access to all the data in this study. YQZ and WHD enrolled all the participants, while HYZ assigned participants to interventions. After that, HLL and CXL collected all the preoperative and postoperative data. SWZ and XYL processed the data and wrote the manuscript. LZ and YXL reviewed and edited the manuscript. YQZ participated in analysing and interpreting the data. YXL participated in revision of the manuscript.

Disclosure

This paper has been uploaded to Research Square as a preprint. https://www.researchsquare.com/article/rs-9359614/v1. The authors have no conflicting relationships to disclose.