- Email: [email protected]
Anterior Segment Imaging Predicts Incident Gonioscopic Angle Closure
Anterior Segment Imaging Predicts Incident Gonioscopic Angle Closure Mani Baskaran, DNB,1,2,3 Jayant V. Iyer, MBBS, MMed,1 Arun K. Narayanaswamy, DNB, MMed,1 Yingke He, BSc(Hons),2 Lisandro M. Sakata, MD, PhD,4 Renyi Wu, MD, PhD,1,5 Dianna Liu, MD,6 Monisha E. Nongpiur, MD,1,2,3 David S. Friedman, MD, PhD,6,* Tin Aung, FRCS(Ed), PhD1,2,3,* Purpose: To investigate the incidence of gonioscopic angle closure after 4 years in subjects with gonioscopically open angles but varying degrees of angle closure detected on anterior segment optical coherence tomography (AS OCT; Visante; Carl Zeiss Meditec, Dublin, CA) at baseline. Design: Prospective, observational study. Participants: Three hundred forty-two subjects, mostly Chinese, 50 years of age or older, were recruited, of whom 65 were controls with open angles on gonioscopy and AS OCT at baseline, and 277 were cases with baseline open angles on gonioscopy but closed angles (1e4 quadrants) on AS OCT scans. Methods: All subjects underwent gonioscopy and AS OCT at baseline (horizontal and vertical single scans) and after 4 years. The examiner performing gonioscopy was masked to the baseline and AS OCT data. Angle closure in a quadrant was defined as nonvisibility of the posterior trabecular meshwork by gonioscopy and visible iridotrabecular contact beyond the scleral spur in AS OCT scans. Main Outcome Measures: Gonioscopic angle closure in 2 or 3 quadrants after 4 years. Results: There were no statistically significant differences in age, ethnicity, or gender between cases and controls. None of the control subjects demonstrated gonioscopic angle closure after 4 years. Forty-eight of the 277 subjects (17.3%; 95% confidence interval [CI], 12.8e23; P < 0.0001) with at least 1 quadrant of angle closure on AS OCT at baseline demonstrated gonioscopic angle closure in 2 or more quadrants, whereas 28 subjects (10.1%; 95% CI, 6.7e14.6; P < 0.004) demonstrated gonioscopic angle closure in 3 or more quadrants after 4 years. Individuals with more quadrants of angle closure on baseline AS OCT scans had a greater likelihood of gonioscopic angle closure developing after 4 years (P < 0.0001, chi-square test for trend for both definitions of angle closure). Conclusions: Anterior segment OCT imaging at baseline predicts incident gonioscopic angle closure after 4 years among subjects who have gonioscopically open angles and iridotrabecular contact on AS OCT at baseline. Ophthalmology 2015;-:1e5 ª 2015 by the American Academy of Ophthalmology. See Editorial on page xxx.
Primary angle-closure glaucoma is a major cause of irreversible blindness in Asia, especially in people of Chinese ethnicity.1,2 The reference standard for angle assessment remains gonioscopy, which requires topical anesthesia and a contact lens and is cumbersome and time consuming. In fact, nearly half of those with diagnosed glaucoma have no documentation of gonioscopy in the medical charts, indicating that there are barriers to carrying out the examination.3 Anterior segment optical coherence tomography (AS OCT) is a noncontact technique for imaging the AS and iridocorneal angle and can be performed by a trained technician. We previously reported that AS OCT identified more eyes as having angle closure than gonioscopy.4 Although this finding could indicate a high false-positive rate for AS OCT, it also could be the case that AS OCT is better able to identify milder forms of angle closure than gonioscopy because no external illumination is required and no contact is made with the eye, which can cause angle widening. Furthermore, there may be low iridotrabecular contact just above the scleral spur that can be seen as angle 2015 by the American Academy of Ophthalmology Published by Elsevier Inc.
closure on AS OCT but would appear to be open on gonioscopy.5 To determine whether those identified as closed on AS OCT have an increased risk of developing angle closure, we examined subjects with gonioscopically open angles but varying degrees of iridotrabecular contact as measured by AS OCT 4 years after the initial examination.
Methods Study Population The study was approved by the institutional review boards of the Singapore Eye Research Institute and Johns Hopkins University and adhered to the tenets of the Declaration of Helsinki. All subjects provided written informed consent. The primary study population consisted of phakic subjects 50 years of age or older who were examined in a community-based study of Singaporeans visiting a government-based polyclinic for nonophthalmic medical problems in 2007.4 The study methodology and details of the study population have been described previously.4 http://dx.doi.org/10.1016/j.ophtha.2015.07.030 ISSN 0161-6420/15
1
Ophthalmology Volume -, Number -, Month 2015 A total of 2052 subjects completed all tests in 2007, and 1630 had open angles on gonioscopy. Of those with open angles on gonioscopy, 591 subjects (36.3%) had angle closure in 1 or more quadrants on AS OCT (defined as irisetrabecular contact beyond the scleral spur in any quadrant). For the current study, we examined 485 of 591 randomly selected individuals (82.1%; stratified by number of quadrants of AS OCT closure) with open angles on gonioscopy and angle closure in any quadrant on AS OCT (considered to be cases). We also examined a sample of subjects with open angles on both gonioscopy and AS OCT (considered the control group). The control group was chosen by computer-generated random selection from 738 of 1039 subjects (71%; with gonioscopic open angles with visible posterior trabecular meshwork in all quadrants, as well as open angles in all quadrants with reliable AS OCT scans) eligible from the baseline cohort. After an initial interview about medical and ophthalmic history, all respondents underwent the following examinations on the same day: visual acuity, AS imaging by AS OCT in the dark (Visante AS OCT; Carl Zeiss Meditec, Dublin, CA), and anterior chamber depth (ACD) and axial length (AL) measurements (IOLMaster, software version 3.02; Carl Zeiss Meditec) by a single operator. The AS OCT and IOLMaster hardware and software versions were the same at baseline and at the 4-year follow-up study. Goldman applanation tonometry, gonioscopy (details below), and optic disc assessment using a 78-diopter (D) lens by slit-lamp biomicroscopic examination were performed. Individuals were excluded at baseline and follow-up if they had a history of intraocular surgery, previous AS laser treatment, or penetrating trauma to the eye. In addition, those with aphakia or pseudophakia on examination as well as those with corneal disorders that could influence imaging by AS OCT, such as corneal endothelial dystrophy, corneal opacity, or severe pterygium, were excluded.
Anterior Segment Optical Coherence Tomography Imaging Imaging with AS OCT was performed in dark room conditions (0 lux) by a single operator who was masked to the results of all the other tests performed. The standard AS single-scan protocol, which produces 256 scans in 0.125 seconds, was used, and scans were centered at the pupil. To obtain the best-quality image, the examiner adjusted the saturation and noise and optimized the polarization for each scan during the examination. The examiner chose the best image with the least motion or image artifacts resulting from the eyelids, masked to the patient’s clinical data, including gonioscopic and baseline imaging results. Similar lighting conditions, scan protocols, and analyses were used for baseline and follow-up visits. A glaucoma fellowshipetrained ophthalmologist (L.M.S.) assessed all AS OCT images (both at baseline and after 4 years), masked to gonioscopic and AS OCT findings, to determine the number of quadrants that were open or closed based on presence of iridotrabecular contact. Baseline and follow-up AS OCT images were assessed over 2 different sessions to avoid bias.
Gonioscopy Gonioscopy was performed in the dark in all cases by a single examiner (M.B.) masked to AS OCT findings. Nonindentation gonioscopy was performed using a Goldmann 2-mirror lens (Ocular Instruments, Inc., Bellevue, WA) at high magnification (16) with the eye aligned to the goniolens in the primary gaze position. Care was taken to avoid light falling on the pupil. Indentation gonioscopy also was performed using a Sussman 4mirror lens (Ocular Instruments, Inc.). The angle in each
2
quadrant was graded based on the anatomic structures observed during gonioscopy (grade 0 ¼ no angle structures, grade 1 ¼ Schwalbe’s line, grade 2 ¼ anterior trabecular meshwork, grade 3 ¼ posterior trabecular meshwork or scleral spur, grade 4 ¼ visible ciliary body). This grading system was derived from the modified Scheie’s angle grading system.6 A quadrant was considered to be closed if the posterior trabecular meshwork could not be seen in the primary position without indentation (grade 0, 1, or 2). Gonioscopic angle closure in an eye was defined as closure in 2 or more quadrants.
Measurements of the Other Ocular Variables A trained ophthalmic technician measured the ACD and AL with the IOLMaster. The IOLMaster measures ACD from the corneal epithelium to the anterior lens surface with lateral slit illumination. The averages of 5 readings obtained for ACD and AL, and 3 consecutive readings obtained for corneal curvature, were calculated and used for subsequent analyses. All readings for each parameter were required to be within 0.05 mm of the reading within the highest signal-to-noise ratio. Goldmann applanation tonometry was used to measure intraocular pressure (IOP), and the vertical cup-to-disc ratio was determined clinically using a 78-D lens at the slit lamp with a graticule (Haag-Streit Model BQ-900; Haag Streit, Koeniz, Switzerland). The median of 3 IOP readings was used for analysis. All clinical measurements were performed by a glaucoma fellowshipetrained ophthalmologist (M.B.) masked to the baseline and follow-up AS OCT findings.
Statistical Analysis Statistical analyses were performed using SPSS software version 18.0 for Windows (SPSS, Inc., Chicago, IL). Demographic and clinical parameters at the baseline evaluation between the participants and nonparticipants were compared with the t test for continuous variables and the chi-square test for categorical variables. The incidence of angle closure on gonioscopy and the association between incident angle closure and baseline AS OCT findings were compared using the Fisher exact test and the chisquare test for trend. Statistical significance was found at P < 0.05.
Results Demographics Telephone calls were made to a total of 585 subjects, of whom 53 were ineligible (23 underwent bilateral cataract surgery during the follow-up period and 30 had medical conditions such as stroke and heart attack that prevented them from coming for the examination) and 12 had died. Of the remaining 520 subjects, 342 (65.8%) responded and were able to make a follow-up visit, 118 (22.8%) declined to come in for a follow-up visit, and 60 (11.6%) subjects could not be contacted (Fig 1). Two hundred seventyseven of the 342 responding subjects were considered cases (having 2 or more angles closed on baseline AS OCT), whereas 65 were controls. Of the eligible subjects, men (59.7% vs. 40.3% women; P ¼ 0.003) were less likely to participate, as were older individuals. Participants and nonparticipants were similar in ethnicity and baseline ocular clinical parameters such as AL, ACD, IOP, and mean gonioscopic grading (Table 1). There were also no statistically significant differences in baseline age, ethnicity, gender, spherical equivalent, visual acuity, and IOP between participating cases and controls, whereas cases had shorter AL and ACD (Table 2).
Baskaran et al
Predicting Gonioscopic Angle Closure with AS OCT proportion with incident gonioscopic angle closure (for both 2- and 3-quadrant definitions) with an increasing number of closed quadrants (P < 0.0001, chi-square test for trend) on baseline AS OCT. None of the baseline demographics (age, ethnicity, gender) or clinical (visual acuity, spherical equivalent, IOP) and A-scan (AL, ACD) parameters showed statistical significance between gonioscopically open and closed angles, except mean gonioscopic grade (2.9 [0.3 SD] vs. 2.6 [0.4 SD]; P ¼ 0.0001) and AS OCT closure at baseline (0% vs. 17.3%; P ¼ 0.0001). Plateau iris configuration was noted more often in closed angles (12/48) in comparison with open angles (24/298) on gonioscopy at 4 years of follow-up (25% vs. 8.2%; P ¼ 0.001).
Discussion
Figure 1. Flowchart showing study recruitment.
Anterior Segment Optical Coherence Tomography and Gonioscopy after 4 Years None of the 65 control subjects had incident gonioscopic angle closure (defined as 2 or more closed quadrants) after 4 years, whereas 48 of the 277 cases (17.3%; 95% confidence interval, 12.8e23) demonstrated gonioscopic angle closure. There was a significant trend of a higher likelihood of incidence of gonioscopic angle closure for those with more quadrants of angle closure as assessed by AS OCT at baseline (Table 3). On follow-up AS OCT, a greater number of eyes had 2-quadrant angle closure compared with baseline (183/333 [55%] vs. 165/342 [48.2%]; P ¼ 0.0001). When angle closure was defined as 3 or more closed quadrants on gonioscopy, 28 of the 277 cases (10.1%; 95% confidence interval, 6.7e14.6; P < 0.004) were found to have gonioscopic angle closure. With this definition, 41.2% of those with 4-quadrant AS OCT angle closure at baseline demonstrated gonioscopic angle closure at 4 years, compared with only 4.5% of those with 1quadrant AS OCT angle closure at baseline (Table 4). On follow-up AS OCT, a larger number of eyes had 3-quadrant closure compared with baseline (53/342 eyes [15%] vs. 91/333 eyes [27.3%];P ¼ 0.0001). There was a monotonic increase in the Table 1. Demographics Comparison between Respondents and Nonrespondents for Follow-up Study of Angle Closure
Mean age (yrs)* Chinese ethnicity (%)z Female gender (%)y Axial length (mm)z IOP (mmHg)z Mean Scheie’s anglez
Respondents (n [ 365)
Nonrespondents (n [ 220)
61.9 (7) 86.6 53.8 24. (1.3) 14.9 (2.4) 2.8 (0.4)
64.1 (7) 85.1 41.7 23.8 (1.1) 14.7 (2.3) 2.8 (0.3)
IOP ¼ intraocular pressure. Data are mean (standard deviation) unless otherwise indicated. *P ¼ 0.001. y P ¼ 0.002 z P > 0.05.
We report novel findings on the incidence of gonioscopic angle closure in a cohort of subjects with gonioscopic open angles but varying degrees of angle closure detected on AS OCT at baseline. The few reports of AS changes in relation to aging and angle closure have been cross-sectional in nature.7e9 Other studies have focused on morphologic changes in AS parameters after laser iridotomy or phacoemulsification.10e12 Some of these studies found AS OCT to be more sensitive in detecting angle closure than gonioscopy.4,13 In this study, we aimed to determine whether angle closure detected on AS OCT in individuals with open angles on gonioscopy represented a false-positive finding, or if it was an indication of increased likelihood of future angle closure. Although gonioscopy is considered the gold standard for judging the angle, it is certainly possible that newer imaging technologies may be more sensitive and less prone to artifact than what is arguably a subjective examination technique in detecting angle closure. Gonioscopy can result in angle opening because of the lens pressing on the cornea, as well as angle opening when incident light lands on the pupil. There is a need for better standardization of gonioscopic technique or documentation to avoid such artifacts. We found that angle closure on AS OCT at baseline was Table 2. Comparison between Cases and Controls for Baseline Parameters
Mean age (yrs) Chinese ethnicity (%) Female gender (%) Visual acuity (logMAR) Spherical equivalent Axial length (mm) ACD (mm) IOP (mm Hg)
Controls (n [ 65)*
Cases (n [ 277)y
60.9 (7.1) 85.6 53.5 0.39 (0.2) 0.31 (1.7) 24.38 (1.4) 3.34 (0.37) 15.2 (2.5)
61.4 (7) 89 55 0.36 (0.2) 0.65 (1.32) 23.89 (1.2) 3.14 (0.32) 14.9 (2.3)
ACD ¼ anterior chamber depth; IOP ¼ intraocular pressure; logMAR ¼ logarithm of the minimum angle of resolution. Data are mean (standard deviation) unless otherwise indicated. All P > 0.05 except axial length (P ¼ 0.009) and ACD (P ¼ 0.001). *Open angles on gonioscopy and anterior segment optical coherence tomography at baseline. y Baseline open angles on gonioscopy, but closed angles (1e4 quadrants) on anterior segment optical coherence tomography scans.
3
Ophthalmology Volume -, Number -, Month 2015 Table 3. Gonioscopic and Anterior Segment Optical Coherence Tomography Angle Closure after 4 Years (Defined as 2 or More Quadrants Closed)* No. of Quadrants Closed on Anterior Segment Optical Coherence Tomography at Baseline 0 1 2 3 4
No. of Subjects
No. (%) with Gonioscopic Angle Closure after 4 Yearsy
65 112 112 36 17
0 12 16 12 8
(0) (10.7) (14.3) (33.3) (47.5)
No. (%) with Anterior Segment Optical Coherence Tomography Angle Closure after 4 Yearsy 8/63 50/110 78/109 31/35 16/16
(12.7) (45.5) (78) (88.5) (100)
*All subjects had no quadrants closed at baseline on gonioscopy. y P < 0.0001, chi-square test for trend.
predictive of later angle closure in this cohort of patients. Furthermore, those with more quadrants of angle closure on AS OCT at baseline had the highest likelihood of demonstrating angle closure on gonioscopy. We noted that the follow-up AS OCT images showed persistence of angle closure in most eyes and that angle closure progression was seen in more quadrants. These findings should be considered in the context of AS OCT angle width varying with time of the day, Valsalva maneuver, and accommodation, conditions that have not been studied in detail so far. We found that most baseline demographic and A-scan parameters were not significantly different between the incident gonioscopic closed angles and the rest of the subjects in this study, except mean gonioscopic grade and angle closure found on baseline AS OCT. However, we found that AL and ACD were shorter in those eyes with baseline AS OCT angle closure. The above findings highlight the relevance of these factors in the overall mechanical environment of the AS. Careful studies with adequate power need to be conducted to elicit the interaction of biometric parameters to the AS structures in determining the risk of angle closure over time. We had to exclude 26 patients because of cataract surgery since baseline. In this context, the narrowing angle width in correlation to age and appositional growth of the lens over time, which can put older individuals at risk of angle closure, may be relevant. This study has several limitations. Baseline and follow-up examinations were carried out at different locations, and therefore lighting conditions may have varied, and this could have led to some of the incident changes in angle
configuration seen. Efforts were made to maintain similar low lighting conditions, and we were careful to center AS OCT images (pupil diameter was not significantly different between baseline and follow-up; data not shown). That said, some subtle differences may have remained, and this could have led to higher rates of incident angle closure. Gonioscopy at baseline and after 4 years was performed by different gonioscopists, which may have led to some variability in results. However, the finding that AS OCT imaging predicts angle closure on gonioscopy is supported by the fact that none of the subjects with open angles on AS OCT demonstrated gonioscopic angle closure. Bias toward diagnosing angle closure on follow-up gonioscopy therefore is unlikely. Finally, the response rate of 62.4% was less than ideal, and although there were no statistically significant differences noted among respondents and nonrespondents in parameters such as ethnicity, AL, baseline IOP, and gonioscopic grade, there were significantly fewer men among the respondents than among the nonrespondents, which could bias toward a higher rate of angle closure because women are more predisposed to angle closure development. However, this bias toward a higher incidence rate may have been offset by a tendency for younger patients to participate. In conclusion, we found that angle closure on AS OCT is predictive of incident gonioscopic angle closure in patients with gonioscopically open angles at baseline. The more quadrants that were closed on AS OCT at baseline, the higher the risk of gonioscopic angle closure developing. The implication of this study’s findings for clinical practice is that ophthalmologists should consider closer monitoring of
Table 4. Gonioscopic and Anterior Segment Optical Coherence Tomography Angle Closure after 4 Years (Defined as 3 or More Quadrants Closed)* No. of Quadrants Closed on Baseline Anterior Segment Optical Coherence Tomography 0 1 2 3 4
No. of Subjects 65 112 112 36 17
*All subjects had no quadrants closed at baseline on gonioscopy. y P < 0.0001, chi-square test for trend.
4
No. (%) with Gonioscopic Angle Closure after 4 Yearsy 0 5 9 7 7
(0) (4.5) (8.0) (19.4) (41.2)
No. (%) with Anterior Segment Optical Coherence Tomography Angle Closure after 4 Yearsy 1/63 22/110 31/109 22/35 15/16
(1.6) (20) (28.4) (62.9) (93.8)
Baskaran et al
Predicting Gonioscopic Angle Closure with AS OCT
patients with signs of angle closure on AS OCT, even if the angles appear open on gonioscopy. Further work is needed to determine whether clinical outcomes other than angle closure are better predicted by imaging than by clinical evaluation.
7.
References
8.
1. Foster PJ, Oen FT, Machin D, et al. The prevalence of glaucoma in Chinese residents of Singapore: a cross-sectional population survey of the Tanjong Pagar district. Arch Ophthalmol 2000;118:1105–11. 2. Quigley HA, Broman AT. The number of people with glaucoma worldwide in 2010 and 2020. Br J Ophthalmol 2006;90:262–7. 3. Quigley HA, Friedman DS, Hahn SR. Evaluation of practice pattern for the care of open angle glaucoma compared with claims data: the Glaucoma Adherence and Persistence Study. Ophthalmology 2007;11:1599–606. 4. Lavanya R, Foster PJ, Sakata LM, et al. Screening for narrow angles in the Singapore population: evaluation of new noncontact screening methods. Ophthalmology 2008;115:1720–7. 7e1e2. 5. Nolan WP, Baasanhu J, Undraa A, et al. Screening for primary angle closure in Mongolia: a randomised controlled trial to determine whether screening and prophylactic treatment will reduce the incidence of primary angle closure glaucoma in an east Asian population. Br J Ophthalmol 2003;87:271–4. 6. Stamper RL, Lieberman MF, Drake MV. Clinical interpretation of gonioscopic findings. In: Becker-Shaffer’s Diagnosis
9. 10.
11.
12.
13.
and Therapy of the Glaucomas. 7th ed. St Louis: Mosby; 1999: 101–13. Lee Y, Sung KR, Na JH, Sun JH. Dynamic changes in anterior segment (AS) parameters in eyes with primary angle closure (PAC) and PAC glaucoma and open-angle eyes assessed using AS optical coherence tomography. Invest Ophthalmol Vis Sci 2012;53:693–7. Yan PS, Lin HT, Wang QL, Zhang ZP. Anterior segment variations with age and accommodation demonstrated by slitlamp-adapted optical coherence tomography. Ophthalmology 2010;117:2301–7. Sihota R, Ghate D, Mohan S, et al. Study of biometric parameters in family members of primary angle closure glaucoma patients. Eye 2008;22:521–7. Ramani KK, Mani B, George RJ, Lingam V. Follow-up of primary angle closure suspects after laser peripheral iridotomy using ultrasound biomicroscopy and A-scan biometry for a period of 2 years. J Glaucoma 2009;18:521–7. Dooley I, Charalampidou S, Malik A, et al. Changes in intraocular pressure and anterior segment morphometry after uneventful phacoemulsification cataract surgery. Eye 2010;24: 519–26. Fleck BW, Wright E, Fairley EA. A randomised prospective comparison of operative peripheral iridectomy and Nd:YAG laser iridotomy treatment of acute angle closure glaucoma: 3 year visual acuity and intraocular pressure control outcome. Br J Ophthalmol 1997;81:884–8. Nolan WP, See JL, Chew PT, et al. Detection of primary angle closure using anterior segment optical coherence tomography in Asian eyes. Ophthalmology 2007;114:33–9.
Footnotes and Financial Disclosures Originally received: November 4, 2014. Final revision: July 20, 2015. Accepted: July 27, 2015. Available online: ---.
Supported by the SingHealth Foundation, Singapore, Republic of Singapore; and the National Medical Research Council, Singapore, Republic of Singapore. Manuscript no. 2014-1762.
1
Author Contributions:
Singapore Eye Research Institute and Singapore National Eye Centre, Singapore, Republic of Singapore.
Conception and design: Baskaran, Friedman, Aung
2
Data collection: Baskaran, Iyer, Narayanaswamy, He, Wu, Liu, Nongpiur,
Duke-NUS Graduate Medical School, Singapore, Republic of Singapore.
Analysis and interpretation: Baskaran, Sakata, Friedman, Aung
3
Obtained funding: Aung
4
Federal University of Parana, Curitiba, Brazil.
Overall responsibility: Baskaran, Iyer, Narayanaswamy, Sakata, Wu, Nongpiur, Friedman, Aung
5
Xiamen Eye Center, Xiamen University, Xiamen, China.
Yong Loo Lin School of Medicine, National University of Singapore, Singapore, Republic of Singapore.
6
Wilmer Eye Institute, Dana Center for Preventive Ophthalmology, Johns Hopkins University, Baltimore, Maryland. *Both Dr. Friedman and Dr. Aung contributed equally as last authors. Financial Disclosure(s): The author(s) have no proprietary or commercial interest in any materials discussed in this article.
Abbreviations and Acronyms: ACD ¼ anterior chamber depth; AL ¼ axial length; AS OCT ¼ anterior segment optical coherence tomography; D ¼ diopter; IOP ¼ intraocular pressure. Correspondence: Tin Aung, FRCS(Ed), PhD, Singapore National Eye Centre, 11 Third Hospital Avenue, Singapore 168751, Republic of Singapore. E-mail: aung. [email protected].
5
Primary angle-closure glaucoma is a major cause of irreversible blindness in Asia, especially in people of Chinese ethnicity.1,2 The reference standard for angle assessment remains gonioscopy, which requires topical anesthesia and a contact lens and is cumbersome and time consuming. In fact, nearly half of those with diagnosed glaucoma have no documentation of gonioscopy in the medical charts, indicating that there are barriers to carrying out the examination.3 Anterior segment optical coherence tomography (AS OCT) is a noncontact technique for imaging the AS and iridocorneal angle and can be performed by a trained technician. We previously reported that AS OCT identified more eyes as having angle closure than gonioscopy.4 Although this finding could indicate a high false-positive rate for AS OCT, it also could be the case that AS OCT is better able to identify milder forms of angle closure than gonioscopy because no external illumination is required and no contact is made with the eye, which can cause angle widening. Furthermore, there may be low iridotrabecular contact just above the scleral spur that can be seen as angle 2015 by the American Academy of Ophthalmology Published by Elsevier Inc.
closure on AS OCT but would appear to be open on gonioscopy.5 To determine whether those identified as closed on AS OCT have an increased risk of developing angle closure, we examined subjects with gonioscopically open angles but varying degrees of iridotrabecular contact as measured by AS OCT 4 years after the initial examination.
Methods Study Population The study was approved by the institutional review boards of the Singapore Eye Research Institute and Johns Hopkins University and adhered to the tenets of the Declaration of Helsinki. All subjects provided written informed consent. The primary study population consisted of phakic subjects 50 years of age or older who were examined in a community-based study of Singaporeans visiting a government-based polyclinic for nonophthalmic medical problems in 2007.4 The study methodology and details of the study population have been described previously.4 http://dx.doi.org/10.1016/j.ophtha.2015.07.030 ISSN 0161-6420/15
1
Ophthalmology Volume -, Number -, Month 2015 A total of 2052 subjects completed all tests in 2007, and 1630 had open angles on gonioscopy. Of those with open angles on gonioscopy, 591 subjects (36.3%) had angle closure in 1 or more quadrants on AS OCT (defined as irisetrabecular contact beyond the scleral spur in any quadrant). For the current study, we examined 485 of 591 randomly selected individuals (82.1%; stratified by number of quadrants of AS OCT closure) with open angles on gonioscopy and angle closure in any quadrant on AS OCT (considered to be cases). We also examined a sample of subjects with open angles on both gonioscopy and AS OCT (considered the control group). The control group was chosen by computer-generated random selection from 738 of 1039 subjects (71%; with gonioscopic open angles with visible posterior trabecular meshwork in all quadrants, as well as open angles in all quadrants with reliable AS OCT scans) eligible from the baseline cohort. After an initial interview about medical and ophthalmic history, all respondents underwent the following examinations on the same day: visual acuity, AS imaging by AS OCT in the dark (Visante AS OCT; Carl Zeiss Meditec, Dublin, CA), and anterior chamber depth (ACD) and axial length (AL) measurements (IOLMaster, software version 3.02; Carl Zeiss Meditec) by a single operator. The AS OCT and IOLMaster hardware and software versions were the same at baseline and at the 4-year follow-up study. Goldman applanation tonometry, gonioscopy (details below), and optic disc assessment using a 78-diopter (D) lens by slit-lamp biomicroscopic examination were performed. Individuals were excluded at baseline and follow-up if they had a history of intraocular surgery, previous AS laser treatment, or penetrating trauma to the eye. In addition, those with aphakia or pseudophakia on examination as well as those with corneal disorders that could influence imaging by AS OCT, such as corneal endothelial dystrophy, corneal opacity, or severe pterygium, were excluded.
Anterior Segment Optical Coherence Tomography Imaging Imaging with AS OCT was performed in dark room conditions (0 lux) by a single operator who was masked to the results of all the other tests performed. The standard AS single-scan protocol, which produces 256 scans in 0.125 seconds, was used, and scans were centered at the pupil. To obtain the best-quality image, the examiner adjusted the saturation and noise and optimized the polarization for each scan during the examination. The examiner chose the best image with the least motion or image artifacts resulting from the eyelids, masked to the patient’s clinical data, including gonioscopic and baseline imaging results. Similar lighting conditions, scan protocols, and analyses were used for baseline and follow-up visits. A glaucoma fellowshipetrained ophthalmologist (L.M.S.) assessed all AS OCT images (both at baseline and after 4 years), masked to gonioscopic and AS OCT findings, to determine the number of quadrants that were open or closed based on presence of iridotrabecular contact. Baseline and follow-up AS OCT images were assessed over 2 different sessions to avoid bias.
Gonioscopy Gonioscopy was performed in the dark in all cases by a single examiner (M.B.) masked to AS OCT findings. Nonindentation gonioscopy was performed using a Goldmann 2-mirror lens (Ocular Instruments, Inc., Bellevue, WA) at high magnification (16) with the eye aligned to the goniolens in the primary gaze position. Care was taken to avoid light falling on the pupil. Indentation gonioscopy also was performed using a Sussman 4mirror lens (Ocular Instruments, Inc.). The angle in each
2
quadrant was graded based on the anatomic structures observed during gonioscopy (grade 0 ¼ no angle structures, grade 1 ¼ Schwalbe’s line, grade 2 ¼ anterior trabecular meshwork, grade 3 ¼ posterior trabecular meshwork or scleral spur, grade 4 ¼ visible ciliary body). This grading system was derived from the modified Scheie’s angle grading system.6 A quadrant was considered to be closed if the posterior trabecular meshwork could not be seen in the primary position without indentation (grade 0, 1, or 2). Gonioscopic angle closure in an eye was defined as closure in 2 or more quadrants.
Measurements of the Other Ocular Variables A trained ophthalmic technician measured the ACD and AL with the IOLMaster. The IOLMaster measures ACD from the corneal epithelium to the anterior lens surface with lateral slit illumination. The averages of 5 readings obtained for ACD and AL, and 3 consecutive readings obtained for corneal curvature, were calculated and used for subsequent analyses. All readings for each parameter were required to be within 0.05 mm of the reading within the highest signal-to-noise ratio. Goldmann applanation tonometry was used to measure intraocular pressure (IOP), and the vertical cup-to-disc ratio was determined clinically using a 78-D lens at the slit lamp with a graticule (Haag-Streit Model BQ-900; Haag Streit, Koeniz, Switzerland). The median of 3 IOP readings was used for analysis. All clinical measurements were performed by a glaucoma fellowshipetrained ophthalmologist (M.B.) masked to the baseline and follow-up AS OCT findings.
Statistical Analysis Statistical analyses were performed using SPSS software version 18.0 for Windows (SPSS, Inc., Chicago, IL). Demographic and clinical parameters at the baseline evaluation between the participants and nonparticipants were compared with the t test for continuous variables and the chi-square test for categorical variables. The incidence of angle closure on gonioscopy and the association between incident angle closure and baseline AS OCT findings were compared using the Fisher exact test and the chisquare test for trend. Statistical significance was found at P < 0.05.
Results Demographics Telephone calls were made to a total of 585 subjects, of whom 53 were ineligible (23 underwent bilateral cataract surgery during the follow-up period and 30 had medical conditions such as stroke and heart attack that prevented them from coming for the examination) and 12 had died. Of the remaining 520 subjects, 342 (65.8%) responded and were able to make a follow-up visit, 118 (22.8%) declined to come in for a follow-up visit, and 60 (11.6%) subjects could not be contacted (Fig 1). Two hundred seventyseven of the 342 responding subjects were considered cases (having 2 or more angles closed on baseline AS OCT), whereas 65 were controls. Of the eligible subjects, men (59.7% vs. 40.3% women; P ¼ 0.003) were less likely to participate, as were older individuals. Participants and nonparticipants were similar in ethnicity and baseline ocular clinical parameters such as AL, ACD, IOP, and mean gonioscopic grading (Table 1). There were also no statistically significant differences in baseline age, ethnicity, gender, spherical equivalent, visual acuity, and IOP between participating cases and controls, whereas cases had shorter AL and ACD (Table 2).
Baskaran et al
Predicting Gonioscopic Angle Closure with AS OCT proportion with incident gonioscopic angle closure (for both 2- and 3-quadrant definitions) with an increasing number of closed quadrants (P < 0.0001, chi-square test for trend) on baseline AS OCT. None of the baseline demographics (age, ethnicity, gender) or clinical (visual acuity, spherical equivalent, IOP) and A-scan (AL, ACD) parameters showed statistical significance between gonioscopically open and closed angles, except mean gonioscopic grade (2.9 [0.3 SD] vs. 2.6 [0.4 SD]; P ¼ 0.0001) and AS OCT closure at baseline (0% vs. 17.3%; P ¼ 0.0001). Plateau iris configuration was noted more often in closed angles (12/48) in comparison with open angles (24/298) on gonioscopy at 4 years of follow-up (25% vs. 8.2%; P ¼ 0.001).
Discussion
Figure 1. Flowchart showing study recruitment.
Anterior Segment Optical Coherence Tomography and Gonioscopy after 4 Years None of the 65 control subjects had incident gonioscopic angle closure (defined as 2 or more closed quadrants) after 4 years, whereas 48 of the 277 cases (17.3%; 95% confidence interval, 12.8e23) demonstrated gonioscopic angle closure. There was a significant trend of a higher likelihood of incidence of gonioscopic angle closure for those with more quadrants of angle closure as assessed by AS OCT at baseline (Table 3). On follow-up AS OCT, a greater number of eyes had 2-quadrant angle closure compared with baseline (183/333 [55%] vs. 165/342 [48.2%]; P ¼ 0.0001). When angle closure was defined as 3 or more closed quadrants on gonioscopy, 28 of the 277 cases (10.1%; 95% confidence interval, 6.7e14.6; P < 0.004) were found to have gonioscopic angle closure. With this definition, 41.2% of those with 4-quadrant AS OCT angle closure at baseline demonstrated gonioscopic angle closure at 4 years, compared with only 4.5% of those with 1quadrant AS OCT angle closure at baseline (Table 4). On follow-up AS OCT, a larger number of eyes had 3-quadrant closure compared with baseline (53/342 eyes [15%] vs. 91/333 eyes [27.3%];P ¼ 0.0001). There was a monotonic increase in the Table 1. Demographics Comparison between Respondents and Nonrespondents for Follow-up Study of Angle Closure
Mean age (yrs)* Chinese ethnicity (%)z Female gender (%)y Axial length (mm)z IOP (mmHg)z Mean Scheie’s anglez
Respondents (n [ 365)
Nonrespondents (n [ 220)
61.9 (7) 86.6 53.8 24. (1.3) 14.9 (2.4) 2.8 (0.4)
64.1 (7) 85.1 41.7 23.8 (1.1) 14.7 (2.3) 2.8 (0.3)
IOP ¼ intraocular pressure. Data are mean (standard deviation) unless otherwise indicated. *P ¼ 0.001. y P ¼ 0.002 z P > 0.05.
We report novel findings on the incidence of gonioscopic angle closure in a cohort of subjects with gonioscopic open angles but varying degrees of angle closure detected on AS OCT at baseline. The few reports of AS changes in relation to aging and angle closure have been cross-sectional in nature.7e9 Other studies have focused on morphologic changes in AS parameters after laser iridotomy or phacoemulsification.10e12 Some of these studies found AS OCT to be more sensitive in detecting angle closure than gonioscopy.4,13 In this study, we aimed to determine whether angle closure detected on AS OCT in individuals with open angles on gonioscopy represented a false-positive finding, or if it was an indication of increased likelihood of future angle closure. Although gonioscopy is considered the gold standard for judging the angle, it is certainly possible that newer imaging technologies may be more sensitive and less prone to artifact than what is arguably a subjective examination technique in detecting angle closure. Gonioscopy can result in angle opening because of the lens pressing on the cornea, as well as angle opening when incident light lands on the pupil. There is a need for better standardization of gonioscopic technique or documentation to avoid such artifacts. We found that angle closure on AS OCT at baseline was Table 2. Comparison between Cases and Controls for Baseline Parameters
Mean age (yrs) Chinese ethnicity (%) Female gender (%) Visual acuity (logMAR) Spherical equivalent Axial length (mm) ACD (mm) IOP (mm Hg)
Controls (n [ 65)*
Cases (n [ 277)y
60.9 (7.1) 85.6 53.5 0.39 (0.2) 0.31 (1.7) 24.38 (1.4) 3.34 (0.37) 15.2 (2.5)
61.4 (7) 89 55 0.36 (0.2) 0.65 (1.32) 23.89 (1.2) 3.14 (0.32) 14.9 (2.3)
ACD ¼ anterior chamber depth; IOP ¼ intraocular pressure; logMAR ¼ logarithm of the minimum angle of resolution. Data are mean (standard deviation) unless otherwise indicated. All P > 0.05 except axial length (P ¼ 0.009) and ACD (P ¼ 0.001). *Open angles on gonioscopy and anterior segment optical coherence tomography at baseline. y Baseline open angles on gonioscopy, but closed angles (1e4 quadrants) on anterior segment optical coherence tomography scans.
3
Ophthalmology Volume -, Number -, Month 2015 Table 3. Gonioscopic and Anterior Segment Optical Coherence Tomography Angle Closure after 4 Years (Defined as 2 or More Quadrants Closed)* No. of Quadrants Closed on Anterior Segment Optical Coherence Tomography at Baseline 0 1 2 3 4
No. of Subjects
No. (%) with Gonioscopic Angle Closure after 4 Yearsy
65 112 112 36 17
0 12 16 12 8
(0) (10.7) (14.3) (33.3) (47.5)
No. (%) with Anterior Segment Optical Coherence Tomography Angle Closure after 4 Yearsy 8/63 50/110 78/109 31/35 16/16
(12.7) (45.5) (78) (88.5) (100)
*All subjects had no quadrants closed at baseline on gonioscopy. y P < 0.0001, chi-square test for trend.
predictive of later angle closure in this cohort of patients. Furthermore, those with more quadrants of angle closure on AS OCT at baseline had the highest likelihood of demonstrating angle closure on gonioscopy. We noted that the follow-up AS OCT images showed persistence of angle closure in most eyes and that angle closure progression was seen in more quadrants. These findings should be considered in the context of AS OCT angle width varying with time of the day, Valsalva maneuver, and accommodation, conditions that have not been studied in detail so far. We found that most baseline demographic and A-scan parameters were not significantly different between the incident gonioscopic closed angles and the rest of the subjects in this study, except mean gonioscopic grade and angle closure found on baseline AS OCT. However, we found that AL and ACD were shorter in those eyes with baseline AS OCT angle closure. The above findings highlight the relevance of these factors in the overall mechanical environment of the AS. Careful studies with adequate power need to be conducted to elicit the interaction of biometric parameters to the AS structures in determining the risk of angle closure over time. We had to exclude 26 patients because of cataract surgery since baseline. In this context, the narrowing angle width in correlation to age and appositional growth of the lens over time, which can put older individuals at risk of angle closure, may be relevant. This study has several limitations. Baseline and follow-up examinations were carried out at different locations, and therefore lighting conditions may have varied, and this could have led to some of the incident changes in angle
configuration seen. Efforts were made to maintain similar low lighting conditions, and we were careful to center AS OCT images (pupil diameter was not significantly different between baseline and follow-up; data not shown). That said, some subtle differences may have remained, and this could have led to higher rates of incident angle closure. Gonioscopy at baseline and after 4 years was performed by different gonioscopists, which may have led to some variability in results. However, the finding that AS OCT imaging predicts angle closure on gonioscopy is supported by the fact that none of the subjects with open angles on AS OCT demonstrated gonioscopic angle closure. Bias toward diagnosing angle closure on follow-up gonioscopy therefore is unlikely. Finally, the response rate of 62.4% was less than ideal, and although there were no statistically significant differences noted among respondents and nonrespondents in parameters such as ethnicity, AL, baseline IOP, and gonioscopic grade, there were significantly fewer men among the respondents than among the nonrespondents, which could bias toward a higher rate of angle closure because women are more predisposed to angle closure development. However, this bias toward a higher incidence rate may have been offset by a tendency for younger patients to participate. In conclusion, we found that angle closure on AS OCT is predictive of incident gonioscopic angle closure in patients with gonioscopically open angles at baseline. The more quadrants that were closed on AS OCT at baseline, the higher the risk of gonioscopic angle closure developing. The implication of this study’s findings for clinical practice is that ophthalmologists should consider closer monitoring of
Table 4. Gonioscopic and Anterior Segment Optical Coherence Tomography Angle Closure after 4 Years (Defined as 3 or More Quadrants Closed)* No. of Quadrants Closed on Baseline Anterior Segment Optical Coherence Tomography 0 1 2 3 4
No. of Subjects 65 112 112 36 17
*All subjects had no quadrants closed at baseline on gonioscopy. y P < 0.0001, chi-square test for trend.
4
No. (%) with Gonioscopic Angle Closure after 4 Yearsy 0 5 9 7 7
(0) (4.5) (8.0) (19.4) (41.2)
No. (%) with Anterior Segment Optical Coherence Tomography Angle Closure after 4 Yearsy 1/63 22/110 31/109 22/35 15/16
(1.6) (20) (28.4) (62.9) (93.8)
Baskaran et al
Predicting Gonioscopic Angle Closure with AS OCT
patients with signs of angle closure on AS OCT, even if the angles appear open on gonioscopy. Further work is needed to determine whether clinical outcomes other than angle closure are better predicted by imaging than by clinical evaluation.
7.
References
8.
1. Foster PJ, Oen FT, Machin D, et al. The prevalence of glaucoma in Chinese residents of Singapore: a cross-sectional population survey of the Tanjong Pagar district. Arch Ophthalmol 2000;118:1105–11. 2. Quigley HA, Broman AT. The number of people with glaucoma worldwide in 2010 and 2020. Br J Ophthalmol 2006;90:262–7. 3. Quigley HA, Friedman DS, Hahn SR. Evaluation of practice pattern for the care of open angle glaucoma compared with claims data: the Glaucoma Adherence and Persistence Study. Ophthalmology 2007;11:1599–606. 4. Lavanya R, Foster PJ, Sakata LM, et al. Screening for narrow angles in the Singapore population: evaluation of new noncontact screening methods. Ophthalmology 2008;115:1720–7. 7e1e2. 5. Nolan WP, Baasanhu J, Undraa A, et al. Screening for primary angle closure in Mongolia: a randomised controlled trial to determine whether screening and prophylactic treatment will reduce the incidence of primary angle closure glaucoma in an east Asian population. Br J Ophthalmol 2003;87:271–4. 6. Stamper RL, Lieberman MF, Drake MV. Clinical interpretation of gonioscopic findings. In: Becker-Shaffer’s Diagnosis
9. 10.
11.
12.
13.
and Therapy of the Glaucomas. 7th ed. St Louis: Mosby; 1999: 101–13. Lee Y, Sung KR, Na JH, Sun JH. Dynamic changes in anterior segment (AS) parameters in eyes with primary angle closure (PAC) and PAC glaucoma and open-angle eyes assessed using AS optical coherence tomography. Invest Ophthalmol Vis Sci 2012;53:693–7. Yan PS, Lin HT, Wang QL, Zhang ZP. Anterior segment variations with age and accommodation demonstrated by slitlamp-adapted optical coherence tomography. Ophthalmology 2010;117:2301–7. Sihota R, Ghate D, Mohan S, et al. Study of biometric parameters in family members of primary angle closure glaucoma patients. Eye 2008;22:521–7. Ramani KK, Mani B, George RJ, Lingam V. Follow-up of primary angle closure suspects after laser peripheral iridotomy using ultrasound biomicroscopy and A-scan biometry for a period of 2 years. J Glaucoma 2009;18:521–7. Dooley I, Charalampidou S, Malik A, et al. Changes in intraocular pressure and anterior segment morphometry after uneventful phacoemulsification cataract surgery. Eye 2010;24: 519–26. Fleck BW, Wright E, Fairley EA. A randomised prospective comparison of operative peripheral iridectomy and Nd:YAG laser iridotomy treatment of acute angle closure glaucoma: 3 year visual acuity and intraocular pressure control outcome. Br J Ophthalmol 1997;81:884–8. Nolan WP, See JL, Chew PT, et al. Detection of primary angle closure using anterior segment optical coherence tomography in Asian eyes. Ophthalmology 2007;114:33–9.
Footnotes and Financial Disclosures Originally received: November 4, 2014. Final revision: July 20, 2015. Accepted: July 27, 2015. Available online: ---.
Supported by the SingHealth Foundation, Singapore, Republic of Singapore; and the National Medical Research Council, Singapore, Republic of Singapore. Manuscript no. 2014-1762.
1
Author Contributions:
Singapore Eye Research Institute and Singapore National Eye Centre, Singapore, Republic of Singapore.
Conception and design: Baskaran, Friedman, Aung
2
Data collection: Baskaran, Iyer, Narayanaswamy, He, Wu, Liu, Nongpiur,
Duke-NUS Graduate Medical School, Singapore, Republic of Singapore.
Analysis and interpretation: Baskaran, Sakata, Friedman, Aung
3
Obtained funding: Aung
4
Federal University of Parana, Curitiba, Brazil.
Overall responsibility: Baskaran, Iyer, Narayanaswamy, Sakata, Wu, Nongpiur, Friedman, Aung
5
Xiamen Eye Center, Xiamen University, Xiamen, China.
Yong Loo Lin School of Medicine, National University of Singapore, Singapore, Republic of Singapore.
6
Wilmer Eye Institute, Dana Center for Preventive Ophthalmology, Johns Hopkins University, Baltimore, Maryland. *Both Dr. Friedman and Dr. Aung contributed equally as last authors. Financial Disclosure(s): The author(s) have no proprietary or commercial interest in any materials discussed in this article.
Abbreviations and Acronyms: ACD ¼ anterior chamber depth; AL ¼ axial length; AS OCT ¼ anterior segment optical coherence tomography; D ¼ diopter; IOP ¼ intraocular pressure. Correspondence: Tin Aung, FRCS(Ed), PhD, Singapore National Eye Centre, 11 Third Hospital Avenue, Singapore 168751, Republic of Singapore. E-mail: aung. [email protected].
5