Anterior Segment Anatomic Parameters Based on the Scleral Spur and Cornea for Risk Profiling of Primary Angle Closure Glaucoma

Primary angle closure glaucoma (PACG) is a sight-threatening condition associated with an unfavorable prognosis. Compared to primary open-angle glaucoma, PACG presents a 2.5-fold higher risk of blindness [1, 2]. Timely implementation of interventions can effectively delay or prevent visual impairment among individuals diagnosed with PACG [3, 4]. Timely implementation of interventions can effectively delay or prevent visual impairment among individuals diagnosed with PACG [5]. However, only approximately 1 in five patients exhibiting narrow anterior chamber angle (ACA) subsequently experiences some form of angle closure [6]. Nonselective interventions for those presenting with narrow angles (NAs) demonstrate limited effectiveness [7‒9]. Therefore, it becomes crucial to promptly identify high-risk patients susceptible to developing PACG and administer suitable interventions.

The early detection of PACG has been extensively researched. Consequently, various ocular parameters, such as a reduced angle-open distance (AOD), increased iris thickness, elevated lens vault (LV), and shorter axial length (AL), have been established as biometric risk factors for PACG [10‒14]. However, these parameters are insufficient in accurately identifying individuals at high risk of developing PACG. Therefore, there is an urgent need to identify novel biological and anatomical parameters that can enhance the early detection of high-risk patients with PACG and facilitate timely intervention.

In this study, CASIA 2 was utilized to identify novel anatomical parameters that enable noninvasive and convenient testing. The close association between PACG and the crowding of the anterior segment structure is evident. Therefore, we considered anterior segment measurement parameters based on the scleral spur (SS) and cornea, encompassing parameters such as anterior chamber width, anterior vault (AV), as well as our customized metric parameters: corneal scleral-spur angle (CSA) and trabecular scleral-spur angle (TSA). The aforementioned parameters offer valuable insights into the anterior chamber’s horizontal size, depth, flattening, and angular dimensions, enabling a comprehensive assessment of crowding within the anterior chamber from various perspectives. Given that the anatomical position and morphology of the SS and cornea are less influenced by age and ocular physiological conditions [15‒17], these parameters also remained relatively stable over time. This may facilitate early prediction of PACG. Additionally, we investigated other parameters including LV and AL. Our findings indicate that patients with PACG exhibit smaller TSA compared to those with NA, along with a larger relative LV (rLV), which represents the proportion of LV within the anterior chamber. This suggests that besides subsequent cataract progression, preexisting narrower space in the ACA area may also be associated with the onset of PACG.

This was a retrospective controlled study. We reviewed the medical records of patients diagnosed with PACG and NA at the Department of Ophthalmology, Zhongnan Hospital, Wuhan University, from February 2022 to November 2022. All patients underwent comprehensive eye examinations, anterior segment optical coherence tomography (AS-OCT, CASIA 2; Tomey Corp., Nagoya, Japan), and IOLMaster 700 (Carl Zeiss, Germany) scans.

PACG includes acute PACG (APACG) and chronic PACG (CPACG). APACG was defined based on two criteria: (1) clinical symptoms including ocular or periocular pain, headache, nausea and vomiting, blurred vision, and halos around lights; and (2) ophthalmologic findings including intraocular pressure (IOP) over 21 mm Hg, conjunctival hyperemia, corneal epithelial edema, shallow anterior chamber with angle closure, iris bombe, and mid-dilated pupil. CPACG was defined when patients presented with more than three clock hours of peripheral anterior synechiae, elevated IOP (>21 mm Hg), and glaucomatous optic neuropathy or visual field defect [18].

NA comes from patients presenting with blurred vision because of cataract during the same period. And the diagnosed cataract stage may be between NO 1–6, C 1–4, and P 1–5 according to lens opacity classification system III [19]. The cataract stage was consistent between PACG group and NA group. NA was defined as a trabecular-iris angle <15° on AS-OCT, with the exclusion of glaucomatous optic neuropathy or visual field loss [20, 21]. Trabecular-iris angle is measured with its apex at the SS and the arms of the angle passing through a point on the trabecular meshwork at 500 μm from the SS and the point on the anterior surface of the iris perpendicularly opposite [22]. The exclusion criteria were as follows: (1) severe systemic disease; (2) secondary angle closure glaucoma; (3) any ocular diseases that affect the ocular anatomical configuration, such as ectopia lentis, lens dislocation, ciliary body cyst, etc.; (4) history of ocular trauma and eye surgery; (5) PACG patients who use pilocarpine to lower IOP; (6) unclear examination images.

AS-OCT (CASIA 2; Tomey, Nagoya, Japan) scans were centered on the pupil under the same light intensity. Each patient underwent a cross-sectional horizontal scan (nasal-temporal angle: 0°–180°). Images were analyzed using the built-in software. Specifically, the following parameters were obtained from the AS-OCT images (shown in Fig. 1): (1) AV – the maximum perpendicular distance from the posterior corneal surface to the horizontal line between the SS [23, 24]; (2) anterior chamber width – the horizontal SS-to-spur distance [25]; (3) LV – the perpendicular distance from the anterior pole of the lens to the horizontal line between the SS [26]; (4) CSA – measured with its apex at the SS and the arms of the angle passing through a point on the corneal endothelium apex and the point on the bilateral SS; (5) TSA – measured with its apex at the SS and the arms of the angle passing through a point on the trabecular meshwork at 750 μm from the SS and the point on the bilateral SS; (6) relative AV – the AV divided by the AL; (7) rLV – the LV divided by the AV; (8) rCSA – the CSA divided by the AL. The measurement of CSA and TSA was at temporal side.

IOLMaster 700 scans were performed at the same light intensity. The light color was used as a quality indicator of the data. We only included values obtained from successful measurements (indicated by a green light). The AL and lens thickness (LT) were measured. Relative LT (rLT) was defined as the LT divided by the AL.

Statistical analyses were performed using the SPSS software system (version 26, SPSS, Inc., Chicago, IL) and GraphPad Prism 8.0 software (GraphPad Software Inc.). Normality was assessed using the Kolmogorov-Smirnov test. The independent sample T test was used to compare normal distribution parameters between PACG and NA groups, while one-way ANOVA was employed to compare parameters among APACG, CPACG, and NA groups. Nonparametric tests were utilized for comparing non-normal distribution parameters. Univariate and multivariate logistic regression models were applied to evaluate the diagnostic utility of baseline horizontal parameter measurements. Parameters with a p value <0.05 in univariate regression analysis and a variance inflation factor <2 were included in the multivariate regression model. Receiver operating characteristic (ROC) curves were generated, and the area under ROC (AUROC) curve was calculated to assess the diagnostic performance of parameters. The statistical significance level was set at a two-tailed p value <0.05. The optimal cutoff point of each parameter was determined using Jorden index (J = max [sensitivity + specificity-1]).

We enrolled a total of 84 patients with PACG, with a mean age of 68.2 ± 8.4 years, and 84 patients with NA, with a mean age of 71.0 + 9.9 years. Among the PACG patients, there were 48 cases of APACG and 36 cases of CPACG. The number of female patients in the PACG and NA groups was 62 and 54, respectively, showing no statistically significant difference (p > 0.05).

The mean values, Kolmogorov-Smirnov test results, and corresponding p values for the biometric parameters of patients with PACG and NA are presented in Table 1. In comparison with NA patients, PACG patients exhibited significantly lower values for TSA (p < 0.001), CSA (p < 0.001), and AL (p = 0.026), while demonstrating significantly higher values for LV (p < 0.001) and rLV (p < 0.001).

The results of both univariate and multivariate logistic regression models are presented in Table 2. In the univariate logistic regression model, CSA (p = 0.006), TSA (p < 0.001), AL (p = 0.029), LV (p < 0.001), and rLV (p < 0.001) demonstrated statistical significance; thus, they were included in the subsequent multivariate logistic regression model. Within the multivariate logistic regression model, TSA exhibited statistical significance after adjusting for age and gender with an odds ratio (OR) of 0.813 per 1° increase (p < 0.001). Similarly, rLV also showed statistical significance with an OR of 1.112 per 1 percent increase (p < 0.001).

Figure 2 displays the ROC curves for TSA, rLV, and their combination. The AUROC values (Table 3) were 0.738 (p < 0.001) for TSA, 0.720 (p < 0.001) for rLV, and 0.807 (p < 0.001) for their combination. The cutoff points of TSA and rLV are listed as 48.8° and 32.67%, respectively, in Table 3. Additionally, the sensitivity and specificity values are provided.

The TSA was stratified into two groups based on the cutoff point: TSA ≥48.8° and TSA <48.8°, for further analysis. After adjusting for age and gender, multivariable binary logistic regression analysis revealed that the risk of PACG was 5.473 (95% CI: 2.747–10.903, p < 0.001) times higher in patients with TSA <48.8° compared to those with TSA ≥48.8° (Table 4).

The mean values and statistical analysis results for the biometric parameters of APACG, CPACG, and NA patients are presented in Table 5. In comparison with NA patients, APACG patients exhibited significantly smaller AV (p = 0.025), CSA (p = 0.002), and TSA (p < 0.001), while LV (p < 0.001), rLV (p < 0.001), and rLT (p = 0.026) were found to be larger. Furthermore, CPACG patients demonstrated a smaller TSA (p < 0.001) when compared to NA patients. Additionally, APACG patients displayed larger LV (p < 0.001), LT (p = 0.045), and rLV (p < 0.001) in comparison with CPACG patients.

Our findings suggest that both the TSA and rLV are significant risk factors for PACG, with TSA being a novel parameter. Notably, while there is a considerable number of individuals with NA, particularly in Asia, only a small subset of them progress to develop PACG. Hence, it is crucial to promptly identify NA patients who are at high risk of developing PACG and provide timely interventions.

A smaller AOD, more significant LV, and thicker iris thickness are significantly associated with PACG progression [11‒14]. However, these factors are closely linked to the morphological changes in the iris, lens, and ciliary body that occur with age; thus, they may not be optimal for early identification of high-risk PACG patients. Alternatively, we investigated factors based on SS and corneal, which are less influenced by age and ocular physiological status. Consequently, they possess an inherent advantage over previous parameters in terms of early risk assessment.

Our findings suggested that a smaller TSA is a risk factor for PACG. This parameter reflects the angular size of the trabecular meshwork area at the ACA. A smaller TSA suggests an initially narrow aqueous outflow pathway, which makes it easy for the iris to adhere to the trabecular meshwork and eventual PACG development. In the multivariate logistic regression model, the risk of PACG decreased by 19.0% for every degree increase in TSA (OR = 0.810). The AUROC value of TSA was 0.738 (95% CI: 0.663–0.812), which indicates that it is moderately predictive. In the previous studies, the AUROC value was 0.69 (95% CI: 0.63–0.75) of IOP and central and limbal anterior chamber depths [26], 0.698 (95% CI: 0.657–0.737) of mean angle width [14], and 0.72 of AOD500 and iris curvature [12]. The AUROC values of previous proved parameters exhibit a comparable performance to the TSA.

Additionally, our findings demonstrate that patients diagnosed with PACG exhibit significantly enlarged LV and rLV, corroborating previous studies highlighting the pivotal role of pupillary block in the development of PACG. [13, 27, 28]. Among them, rLV represents the proportion of the lens in the anterior segment, providing a more accurate reflection of lens expansion. The risk of PACG exhibits a 10.6% increase for every one percent rise in rLV (OR = 1.106). Furthermore, the AUROC was determined to be 0.720.

The AUROC for the combination of TSA with rLV increased to 0.807, indicating enhanced accuracy in assessing the risk of PACG. Consequently, integrating TSA with previously established parameters may enable more precise identification of patients at a heightened risk of PACG.

PACG eyes exhibit a smaller TSA, in addition to the greater expansion of the lens, which leans to anterior movement of the iris. Both factors contribute to angle closure, suggesting that PACG may arise from a combination of inherent small size and subsequent structural crowding.

According to the results presented in Table 5, patients with CPACG and NA exhibit significant similarity in their parameter values, except for a smaller TSA (p < 0.001) observed in CPACG patients. The extent of anterior lens expansion in CPACG is comparable to that seen in NA patients but not as pronounced as that observed in APACG patients. In comparison with NA, CPACG patients have a closer proximity between their iris and trabecular meshwork, rendering them more susceptible to chronic attachment. Except for a smaller TSA (p < 0.001), it can be noted that APACG patients demonstrate a relatively thicker and more anteriorly bulging lens characterized by larger LV (p < 0.001), rLV (p < 0.001), and rLT (p = 0.026) in comparison with NA patients. As cataract progression occurs, there is an increase in the degree of lens expansion which heightens the likelihood of acute angle closure.

Therefore, individuals displaying small TSA along with increased LT and inconspicuous anterior expansion should exercise vigilance toward potential occult development of CPACG. Similarly, for those exhibiting small TSA alongside evident lens expansion measures must be taken to prevent acute attacks of angle closure.

In summary, for patients with a shallow anterior chamber and narrow ACA, who exhibit good visual acuity and mild cataract progression, AS-OCT scan can be employed as an initial assessment tool to evaluate their risk of developing PACG. The optimal cutoff values identified in this study were 48.8° for TSA and 32.67% for rLV. Our findings demonstrate that patients with a TSA <48.8° have a significantly higher risk of PACG, with an OR of 5.473 (95% CI: 2.747–10.903, p < 0.001), compared to those with TSA ≥48.8° (Table 4). Therefore, it is recommended to closely monitor individuals with TSA <48.8° or rLV >32.67%, in order to facilitate timely intervention and prevent the onset of PACG.

Besides, our study demonstrated that eyes with PACG exhibited a significantly smaller CSA and shorter AL through univariate logistic regression analysis. The presence of a smaller CSA, which is an innovative parameter, indicates a flatter cornea morphology and contributes partially to the anterior segment crowding. Additionally, individuals with shorter AL may have an increased susceptibility to PACG, aligning with our research findings. However, it should be noted that the coexistence of PACG and short AL is not always guaranteed. Even among patients diagnosed with PACG who possess relatively shorter AL measurements, there can still be instances where the size of the anterior segment remains relatively small, although such cases are infrequent [18, 24]. Taken together, except for patients with PACG who exhibit a shorter AL [29], it holds greater significance that their anterior segment dimensions are comparatively smaller.

Several studies have identified other risk factors for PACG. For example, the ciliary body, whose morphology can affect the conformation of the ACA, is closely associated with PACG [30, 31].

Besides, the vitreous zonules, which are the bridging bundles of zonular fibers that extend from the zonular plexus region in the valleys of the posterior pars plicata to the vitreous in the ora serrata region, were less likely to be observed in PACG eyes [32, 33]. Furthermore, the anterior choroid is associated with PACG [34, 35]. Notably, these structures cannot be observed on AS-OCT. Therefore, future studies on the risk factors for PACG should combine AS-OCT with other examination techniques, such as ultrasound biomicroscopy.

This study is subject to several limitations. First, due to its retrospective nature, a causal relationship between the anatomical parameters and PACG could not be confirmed. Therefore, prospective studies are warranted to validate our findings. Second, the small-scale design of this study restricts the generalizability of our results. Further research involving larger sample sizes is imperative for validating our findings. Lastly, as this study primarily focused on Chinese individuals, caution should be exercised when extrapolating these findings to other ethnicities.

AS-OCT is a valuable tool for early assessment of progression risk in NA patients at risk for developing PACG. Our findings suggest that both TSA and rLV are significant risk factors for PACG. Moreover, the novel parameter TSA demonstrates promising potential as a predictive tool for early PACG screening.

The authors acknowledge the clinical program of Zhongnan Hospital of Wuhan University (No. ZNYYIIT20200507).

This study protocol was reviewed and approved by the Institutional Review Board of Zhongnan Hospital, Approval No. 2022227K, and has been granted an exemption from requiring written informed consent. This study adhered to the tenets of the Declaration of Helsinki.

The authors have no conflicts of interest to declare.

The authors received no funding for this study.

All authors contributed to the study’s conception and design. Y.F., B.J., L.L., Y.L., and Z.C. performed material preparation, data collection, and analysis. Manuscript drafting and literature searching by Y.F. Supervision and manuscript revision by Y.L., M.K., and W.Z. All authors commented on previous versions of the manuscript and, finally, read and approved the final manuscript.

Trial registration ID: NCT06143943.

The data that support the findings of this study are not publicly available due to their containing information that could compromise the privacy of research participants but are available from M.K. ([email protected]) upon reasonable request.