Purpose To investigate the 5-year changes in static and dynamic anterior segment optical coherence tomography (AS-OCT) parameters and their predictors.
Methods This was a prospective, population-based cohort study of people aged 50 years and older residing in the Liwan District, Guangzhou, China. Standardised AS-OCT scans were performed in November 2008 and November 2013 under dark and light conditions. Customised software was used to analyse horizontal AS-OCT images. Parameters in dark and measurements of light-to-dark changes were used for analyses.
Results A total of 186 (71.8%) subjects underwent AS-OCT twice, 5 years apart and were included for analyses. The mean age in 2008 was 64.7±7.0 years, and 60.2% were women. The anterior chamber width (ACW) decreased from 11.74±0.44 mm in 2008 to 11.60±0.37 mm in 2013 (p=0.001). There was a trend towards a decrease in dynamic capacity (light-to-dark changes) in the anterior segment, with decreased iris thickness at 750 µm (ΔIT750), ΔACW, Δ anterior chamber area (ACA) and Δ pupil diameter at 5 years (all p<0.05). After adjusting for age and sex, the following baseline parameters were associated with a greater decrease rate in trabecular iris space area at 500 µm (TISA500) at 5 years: TISA500, IT750 and ACA in dark (p<0.001 for all).
Conclusions Anterior chamber angle width decreased and the amount of light-to-dark changes declined during 5-year follow-up. Subjects with greater height, wider angle width and thicker iris at baseline have greater angle narrowing at follow-up.
- population-based study
- anterior segment optic coherence tomography
- angle closure
This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/.
Statistics from Altmetric.com
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.
Primary angle-closure glaucoma (PACG) is a leading cause of irreversible blindness, affecting 20.17 million people worldwide, 76.7% of whom are of Asian ethnicity.1 The number of people with PACG is expected to increase to 23.36 million in 2020 and 32.04 million in 2040 globally.1 Ocular biometric features such as a shallower anterior chamber, smaller central corneal diameter, a thicker and more anterior placed lens and shorter axial length have been identified as important risk factors for PACG.2 Population-based studies have also shown that women and those of Chinese descent have a higher risk of developing PACG.3 4
The introduction of anterior segment optical coherence tomography (AS-OCT) has expanded our understanding of biometric risk factors for PACG. This technique permits the capture of the entire anterior segment in a single image and provides repeatable and quantitative measurements of the anterior segment parameters with excellent accuracy.5 Cross-sectional studies using AS-OCT identified novel risk factors for angle closure, such as shorter anterior chamber width (ACW), smaller anterior chamber area (ACA), thicker iris with greater curvature and area, and greater lens vault (LV).6–9 Dynamic AS-OCT parameters such as increase in iris volume changes with physiological or pharmacological dilation were also shown to be strongly associated with PACG in cross-sectional analyses.10–13
Information on longitudinal changes in static and dynamic AS-OCT parameters over time is important for predicting future risk of PACG. However, very few cohort studies have produced longitudinal data, most of which was derived from clinics with relative short-term follow-up, small sample sizes and assessment was based on static measures only.14 15 To the best of our knowledge, there are no population-based studies evaluating the long-term changes in static and dynamic AS-OCT parameters. Therefore, the aim of this study was to assess the 5-year changes of AS-OCT parameters in dark and light-to-dark conditions in a cohort of older Chinese people, using data obtained from follow-up visits in the population-based Liwan Eye Study.
Statement of ethics
The study was approved by the Ethical Review Board of the Zhongshan Ophthalmic Centre, and the Research Governance Committee of Moorfield’s Eye Hospital. The study adhered to the tenets of the Helsinki Declaration. Written informed consent was obtained from all subjects.
The Liwan Eye study was a population-based cohort study, its methodology and details have been described previously.16 17 In brief, the Liwan district of Guangzhou was selected for the survey because of its relatively stable population and representative demographic and socioeconomic characteristics. Chinese persons aged 50 years and older were identified by cluster random sampling and were invited for a comprehensive eye examination in a research clinic set up nearby communities. The baseline eye examination was performed between September 2003 and February 2004, the 5-year and 10-year follow-up examinations were performed in November 2008 and November in 2013, respectively. At the 5-year and 10-year follow-up, one out of two subjects were systematically selected to undergo AS-OCT imaging under dark and light conditions as well as detailed ocular examinations. All examination equipment was the same at baseline and at follow-up. The same instrument (AS-OCT device) and image capture protocol were used at baseline and follow-up. In the current study, AS-OCT imaging data at 5-year and 10-year follow-up were analysed. Subjects at baseline and follow-up were excluded if they were aphakic/pseudo-phakic or had a history of intraocular surgery, laser therapy, ocular trauma and corneal disease that may affect AS-OCT imaging.
All subjects underwent a standardised interview and comprehensive examinations. Height was measured in centimetres using a wall-mounted measuring tape. Weight was measured in kilograms with a digital scale. Lens thickness (LT), axial length (AL), and central anterior chamber depth (ACD) were measured by A-scan ultrasound sonography (Model US-800; Nidek Co., Tokyo, Japan). Intraocular pressure was performed using a hand-held tonometer (Tonopen; Mentor, Norwell, MA, USA) device after instilling topical anaesthesia (0.4% Benoxil; Oxybuprocaine, Santen, Japan). Gonioscopy was performed with a Goldmann-type one-mirror lens (Haag Streit, Bern, Switzerland) at ×25 magnification in a dark room. The angle width for four quadrants was graded using the Shaffer method.
AS-OCT imaging and analyses
The standardised protocol for obtaining AS-OCT images has been reported previously.18–20 AS-OCT (Visante; Carl Zeiss Meditec, Dublin, CA, USA) imaging was performed under dark lighting conditions (<5 lux, to induce physiological mydriasis) and then under light conditions (350–400 lux) using a standardised light source. The single-scan protocol which produces 256 scans in 0.125 s was used to obtain AS-OCT images. Refractive correction was used to adjust the internal fixation target to prevent accommodation. To ensure scan alignment, the reflex saturation beam was used. The polarisation for each scan was optimised, and the image saturation and noise were adjusted to obtain the best quality image. Each AS-OCT image captured both the temporal and nasal quadrants (nasal-temporal 0°−180°) in a single image. The image with best quality and fewest artefacts was selected for analysis.
A customised version of the Zhongshan Angle Assessment Program (ZAAP, Guangzhou, China) was used to analyse the AS-OCT image measurements. After manually locating the sclera spur, the ZAAP automatically defines the border and curvatures of the anterior segment structures and calculates the AS-OCT parameters. Two experienced and independent graders were masked to the refraction data.
The AS-OCT parameters obtained by ZAAP has been shown to be highly reproducible, with an intraclass correlation coefficient exceeding 0.88.21 Trabecular iris space area at 500 µm (TISA500), angle recess area (ARA), iris thickness at 750 µm (IT750), iris area (Iarea), ACW, ACA, LV and pupil diameter (PD) are illustrated in figure 1. The same lighting conditions, scan protocols and ZAAP software version were used for AS-OCT imaging in 2008 and 2013.
Data from right eyes were used for analysis. Light-to-dark changes (Δ) were defined as the measurements of AS-OCT parameters in dark minus light conditions. Continuous variables were presented as mean±SD. Subject demographics and AS-OCT parameters at baseline between the included subjects and excluded subjects were compared with t-test for continuous variables and Fisher’s exact test for categorical variables. Included subjects were the patients who finish 5-year follow-up examinations without image problems and previous surgeries. And excluded subjects were the patients lost to follow-up or with image problems and surgeries from 2008 to 2013. The 5-year change of AS-OCT measurements (10-year follow-up minus 5-year follow-up) were assessed using a paired t-test. Associations between longitudinal change rate in TISA500 in dark conditions and potential baseline risk factors were assessed using two mixed-effect models with the assumption that data missing was random, and predictors of missing data were included in the models. Each visit in 2008 and 2013 was assigned a number from 0 to 1 accordingly and was used as a proxy for time. All of the model covariates were adjusted for baseline age and gender. Age, gender, weight, height, systolic blood pressure (SBP), diastolic blood pressure, spherical equivalent (SE), AL, ACD, LT, TISA500, ARA, IT750, larea, ACW, LV, PD, ΔARA, ΔIT750, ΔIarea, ΔACW, ΔACA, ΔLV and ΔPD were included as fixed effects. Individual subject was considered as random effect. Mean changes and 95% CIs were calculated from the mixed models. Model 1 was a univariate regression and model 2 was a multivariate regression. For each variable, the unstandardised regression coefficient (β) and its significance level were obtained. All statistical analysis was performed using Stata SE V.12.0 (Stata Corp, College Station, TX, USA) with a statistical significance set at 0.05.
A total of 504 subjects completed AS-OCT examination in 2008 in both dark and light conditions. Among them, 184 (36.5%) had a history of previous ocular surgery or imaging problems (such as indiscernible scleral spur, inaccurate identification anterior chamber angle, inaccurate identification posterior iris boundaries and lens) that precluded accurate measurement of AS-OCT parameters and were excluded from analysis. The remaining 320 (63.5%) eligible subjects had available data for analyses from 2008. Of those subjects, 259 (80.9%) presented in 2013, 73 (28.2%) of whom were ineligible because of the ocular surgery during the follow-up period or imaging problems. Finally, a total of 186 (71.8%) subjects were included in the final statistical analyses. The flowchart of participants selection is shown in figure 2.
Baseline characteristics of eligible participants included in and excluded from analysis are shown in table 1. Included subjects tended to be significantly younger (p<0.001), with lower SBP (p=0.023), greater ACA (p=0.004), smaller LV (p=0.008) and larger PD (p=0.009) in dark, but there were no significant differences in other demographic and ocular biometric parameters in comparison with the subjects who were not included in the analysis. The mean age of analysed subjects was 64.7±7.0 years, and 60.2% were women.
The 5-year changes of AS-OCT parameters in dark and light-to-dark during follow-up are shown in table 2. After 5 years, there was a trend towards narrowing of the angle in dark conditions with a significant decrease in ACW (11.74±0.44 mm vs 11.60±0.37 mm, p=0.001) compared with baseline measurements. As respect to light-to-dark changes, the dynamic capacity in angle parameters, iris thickness parameters and LV decreased over time. However, only ΔIT750 (p=0.001), ΔACW (p=0.003) and ΔACA (p<0.001) achieved statistically significance when compared with baseline. A significant decrease in ΔPD (p<0.001) was also observed at follow-up.
To identify potential associated parameters with 5-year change rate in TISA500, univariate and multiple regression analysis models were created. Table 3 shows the association of baseline factors with 5-year change rate of TISA500 in dark conditions. Having a greater height, wider angle and thicker iris at baseline was significantly associated with a greater chance of angle narrowing at follow-up, as TISA500, IT750 and ACA in dark were associated with larger mean TISA500 change rate at follow-up (p<0.001).
This is the population-based cohort study investigating the long-term changes in AS-OCT parameters in both dark and light-to-dark conditions. The results demonstrated a significant decrease in angle width and decrease in dynamic capacity (light-to-dark changes) over time. The level of anterior chamber narrowing over time was predicted by greater height, angle depth and thicker iris parameters in dark conditions at baseline.
The current study confirms the findings from cross-sectional studies which have previously reported that angle parameters slope negatively with age, with −0.0119 mm/year for ACD, −0.0845 mm2/year for Iarea; while iris parameters and LV had positive slopes with age, with 0.0084 mm/year for LV, 0.0019 mm/year for Icurv, 0.0006 mm/year for IT750, 0.0008 mm/year for IT500, 0.0131 mm2/year for Iarea.22 In another clinical-based studies, negative age-related changes of AS-OCT parameters were observed, and the normalised slopes of ACD, angle opening distance (AOD)500, ARA, TISA500 were −0.02396 mm/year, −0.00634 mm/year, −0.0019 mm2 /year, −0.00177 mm2 /year, respectively.23
The findings of the current study suggest that angle narrowing occurs over time and is consistent with previous longitudinal studies. These findings could partly explain the phenomenon that older people have narrower angles and prevalence of PACG increases with ageing suggested in previous literature.17 A study of Chinese participants aged ≥35 years found that AOD500 decreased from 0.25±0.13 mm at baseline to 0.21±13 mm at 2-year follow-up and ARA decreased from 21.5±3.73 to 21.0±3.64 mm2; however, no significant changes in LV or ACW were noted during the 2- year follow-up.24 In a community-based cohort of Singaporean adults with open angles, He et al 14 found a significant decrease in angle and anterior chamber parameters. However, the IT2000 and Iarea in nasal and temporal quadrants decreased significantly during the 4-year follow-up. Interestingly, our observed 5-year decrease in TISA500 (6.5% of the baseline values) was smaller than that reported in previous studies. These differences may be caused by different population characteristics, a narrower baseline angle and older age in the present cohort. The relatively small rate of narrowing of angle width was also observed in the eyes of primary angle-closure suspects (PACS).5
The lens probably moves anteriorly (estimated by LV) through the effect of cataract formation occupying the space in anterior chamber with age increasing. The effect of iris and lens interaction make the iris forward changing the iris volume and increasing pupillary block. Several clinic-based studies have demonstrated that the decrease in iris volume after pupil dilation was less in patients with PACG than normal patients or those with primary open-angle glaucoma (POAG).25–27 A population-based cross-sectional study of rural Chinese persons has shown that the decrease in Iarea and iris volume after physiological and pharmacological dilation was significantly less in PACS than normal eyes.12, 13 Additionally, Seager et al 28 found that Iarea was related to age, race, disease diagnosis and Iarea loss per millimetre of pupil dilation. The association between loss of Iarea on dilation, race and age has also been found to exist in Korean patients with PAC or POAG and was found to decline with age.29 Similarly, our study confirmed results above and suggested that dynamic AS-OCT parameters declined over time, which was corresponded with the higher prevalence of PACG in older persons. However, the longitudinal changes of dynamic AS-OCT parameters require further study.
The TISA500 has been described as an indicator of angle area because it is a more objective measure than the point distance (AOD500).30 The finding that the height was correlated with the narrowing of TISA500 was in accord with our previous analysis in this population that shorter height was significantly associated with narrower angle width (p<0.001).29 Eyes with a wider angle at baseline had a greater chance rate of TISA500 narrowing at follow-up in this cohort, which may be due to a ceiling effect. Similar findings were observed in our previous 2-year cohort study with younger subjects. Those who had a larger baseline AOD500 showed a greater decrease in AOD500 and ARA at follow-up.24 The decrease in ACD was shown to be associated with deeper baseline ACD and greater ACA. In a community-based study of 204 Singaporeans, the 4-year decrease in TISA750 was reported to be lower in eyes with shorter AL and greater Icurv at baseline.14 However, our study did not detect this correlation and this may be related to different population characteristics and study design.
We find that light-to-dark change of angle and iris parameters are not associated with angle narrowing. And in multivariate analysis, greater height, wider angle and thicker iris at baseline were the only independent factors associated with changes rate in TISA500 at follow-up. These findings may shed light on our understanding of the risk factors for PACG in the Chinese population. Meanwhile, it has implications in understanding the evolution of anterior segment changes in eyes with angle closure.
There are several advantages of this study; these include the population-based study design which minimises selection bias and the analysis of both static and dynamic AS-OCT parameters which were not collected in previous studies. And mixed-effect model was used to assess the change-to-change relationships after controlling for confounding factors.
Nevertheless, there are several limitations that warrant further consideration. First, only 186 (71.8%) subjects with complete data were included in the longitudinal analysis. However, this is similar to the 61% of participants with eligible images at both baseline and follow-up in the previous cohort study of Singaporean participants by He et al 14 The limitations of time-domain OCT in terms of poor axial resolution and motion artefacts. Reasons for the high percentage of image problems may be related to three primary reasons (cannot identify scleral spur, cannot identify anterior chamber angle, cannot identify posterior iris boundaries and lens). The fact that a comparison of longitudinal changes requires image quality to be acceptable at both baseline and follow-up visits. Second, stratifying analyses by age, sex or quartile of baseline angle parameters is not possible due to the relatively small sample size. Further cohort study with larger sample size are warrant to confirm our findings. Third, a single scan was used to measure AS-OCT parameters, regional variations of anterior chamber and iris structure may exist, and further studies are warranted. Finally, the findings may not be generalisable to other populations. Comparisons of static and dynamic changes in anterior segment structure between Chinese and Caucasian subjects have shown that Chinese have greater iris parameters and lower dynamic changes in AS-OCT parameters than Caucasians. Therefore, further research is required in other ethnic backgrounds to confirm these findings.
In conclusion, there was a significant decrease in angle parameters and a decline in light-to-dark changes during the 5-year follow-up in this population-based sample of Chinese subjects. The level of anterior chamber narrowing was associated with height, angle and iris parameters in dark conditions at baseline. Further studies are warranted to clarify the underlying mechanisms of these changes in anterior segment parameters.
YJ, DW, WW and FC are joint first authors.
Correction notice This paper has been corrected since it was published online. The project name,
Fundamental Research Funds of the State Key Laboratory in Ophthalmology was missing from the funding statement and this has now been added.
Contributors Conception and design MH; data collection YJ, WW, FC, WH; SH; analysis and interpretation YJ, DW, FC, WW, WH, SH, MH; writing of the article YJ, DW, WW, FC, JS; critical revision of the article: MH.
Funding This research was supported in part by the grant from the Fundamental Research Funds of the State Key of Ophthalmology, National Natural Science Foundation of China (81420108008 81570843H1204), Science and Technology Planning Project of Guangdong Province, China (2013B20400003). MH receives support from the University of Melbourne at Research Accelerator Program and the CERA Foundation. The Centre for Eye Research Australia receives Operational Infrastructure Support from the Victorian State Government. No additional external funding was received.
Competing interests None declared.
Patient consent for publication Not required.
Ethics approval Ethical approval was obtained from the zhongshan Ophthalmic Center Ethics Review Board and the Research Governance Committee of Moorfields Eye Hospital in London, England. ID:  No.12.
Provenance and peer review Not commissioned; externally peer reviewed.
Data availability statement Data are available upon request.