Changes in Anterior Segment Morphology After Laser Peripheral Iridotomy in Acute Primary Angle Closure

SASAN MOGHIMI, REBECCA CHEN, MOHAMMADKARIM JOHARI, FAEZEH BIJANI, MASSOOD MOHAMMADI, ALIREZA KHODABANDEH, MINGGUANG HE, AND SHAN C. LIN

PURPOSE: To evaluate the anterior segment biomet- different between groups with and without exaggerated ric changes measured by anterior segment optical LV after LPI. coherence tomography (AS-OCT) in acute primary CONCLUSION: This study confirms that LPI results in a angle closure (APAC) after laser peripheral iridotomy significant increase in the angle width, ACD, and ACA as (LPI). well as flattening of the in APAC eyes. The extent of DESIGN: Prospective interventional study. angle deepening is inversely related to baseline angle width. METHODS: In this clinic-based study, 52 eyes of 52 The shifts posteriorly after resolution of attack, espe- patients with resolved APAC attack who underwent cially in those with greater lens vault. (Am J LPI were enrolled. Subjects underwent complete Ophthalmol 2016;166:133–140. Ó 2016 Elsevier Inc. ophthalmic examination and AS-OCT imaging before All rights reserved.) and 6 weeks after LPI. Anterior chamber depth (ACD), anterior chamber area (ACA), iris thickness (IT), iris area, iris curvature, lens vault (LV), anterior vault, angle opening distance (AOD500, AOD750), and trabecular RIMARY ANGLE CLOSURE DISEASE IS CHARACTER- iris space area (TISA500, TISA750) were measured in ized by a crowded anterior segment and synechial qualified images and compared before and after LPI. A P or appositional closure of the angle. Angle closure linear mixed-model analysis was performed for potential can have several different presentations. Acute primary predictors of change in AOD750. Main outcome measure angle closure (APAC) is characterized by sudden intraoc- was change in AOD750 after LPI. ular pressure (IOP) rise and its consequences, such as RESULTS: The mean age of participants was 60.7 ± 9.2 corneal edema, decreased vision, seeing halos around years. Mean angle width (Shaffer grade) changed from lights, sluggish mid-dilated pupil, headache, eye pain, 1–3 0.25 ± 0.34 at baseline to 1.22 ± 0.86 after LPI (P < and eye redness. This condition is an ophthalmologic .001). However, 25 nasal angles (48.0%) and 28 tempo- emergency; untreated, APAC may lead to potentially ral angles (53%) had iridotrabecular contact after LPI. sight-threatening complications. Its treatment has 2 All angle parameters (AOD500, AOD750, TISA500, arms: IOP reduction and relief of angle closure. The TISA750; P £ .03), ACD (P [ .001), and ACA (P most common underlying mechanism of primary angle < .001) increased significantly after LPI. Iris curvature closure in this presentation is pupil block, which can be and LV were reduced (P [ .01 for both) after LPI, but relieved with laser peripheral iridotomy (LPI) in most 3,4 there was no significant change in IT and iris area. After cases. multivariate analysis, pre-LPI AOD750 was the only fac- A lens-induced mechanism in the development of APAC has tor associated with change in AOD750 (b [ L0.992, also been suggested. Lens vault (LV) is an important anatomic P [ .02). Exaggerated LV, defined as LV greater than risk factor and was previously regarded as one of the strongest pre- 3,5–8 one-third of the anterior vault (sum of LV and ACD), dictors of primary angle closure (PACG). This was present in 61.5% of the cases (32 eyes). The extent mechanism of angle closure, called the ‘‘exaggerated lens vault 8 of change in angle parameters was not significantly mechanism,’’ was found to be predominant in half of APAC eyes in a study by Moghimi and associates.9 Although various investigators have studied factors Supplemental Material available at AJO.com. predicting LPI success in drainage angle opening among 1,4,10–17 Accepted for publication Mar 23, 2016. patients with primary angle closure/suspect, few From Farabi Eye Hospital, Tehran University of Medical Sciences, studies are performed on APAC eyes with a high Tehran, Iran (S.M., M.J., F.B., M.M., A.K.); Koret Vision Center, 18 University of California, San Francisco Medical School, San Francisco, proportion of exaggerated LV. The present study eval- California (S.M., R.C., S.C.L.); and Department of Ophthalmology, uates the effect of LPI on anterior segment morphology University of Melbourne, Melbourne, Australia (M.H.). and investigates potential predictors of LPI success in Inquiries to Shan C. Lin, Koret Vision Center, University of California, San Francisco Medical School, San Francisco, CA 94143; e-mail: [email protected]. angle opening in terms of and biometric edu findings.

0002-9394/$36.00 Ó 2016 ELSEVIER INC.ALL RIGHTS RESERVED. 133 http://dx.doi.org/10.1016/j.ajo.2016.03.032 METHODS TABLE 1. Anterior Segment Parameters Measured THE STUDY PROTOCOL WAS APPROVED BY THE INSTITU- by Anterior Segment Optical Coherence Tomography and tional review board of Farabi Eye Hospital, Tehran, Iran. Their Definitions This was a prospective study in which, after complete Parameter Definition explanation, all patients gave written informed consent to participate in this research protocol. Angle opening The distance between the posterior The participants were recruited from the glaucoma clinic distance at 500 and corneal surface and the anterior iris 750 mm (AOD500, surface on a line perpendicular to the of Farabi Eye Hospital, Tehran, Iran, which is a tertiary care AOD750) , 500 and center, as part of the Farabi Angle Closure Study. All 750 mm from the scleral spur, patients had an APAC attack, defined by the presence of respectively the following: (1) at least 2 of the following symptoms of Trabecular iris space The surface area of a trapezoid with the an acute episode of IOP rise: ocular pain or headache, area at 500 and following boundaries: anteriorly, the nausea or vomiting, decreased vision, and rainbow- 750 mm (TISA500, angle opening distance at 500 or colored halos around lights; (2) IOP at presentation of at TISA750) 750 mm from the scleral spur; least 30 mm Hg by Goldmann applanation tonometry; posteriorly, a line drawn from the (3) examination findings such as conjunctival injection, scleral spur perpendicular to the corneal epithelial edema, fixed mid-dilated pupil, and plane of the inner scleral wall to the shallow anterior chamber; and (4) shallow anterior cham- iris; superiorly, the inner corneoscleral wall; and inferiorly, the iris surface ber and narrow angle in the other eye. APAC attacks were Anterior chamber The axial distance from the corneal broken within 24 hours of the onset of symptoms with depth (ACD) endothelium to the anterior lens intravenous mannitol or oral glycerin, oral acetazolamide, surface and topical timolol. The APAC attack was defined as Anterior chamber The cross-sectional area of the anterior broken when IOP was less than 21 mm Hg (with or without area (ACA) chamber bordered by the posterior medication) and when signs and symptoms of acute IOP surface of the cornea, the anterior rise had subsided. Eyes whose attack could not be broken surface of the iris, and the anterior with these medications were excluded from the study and surface of the lens within the pupil received further interventions. Iris area (I-Area) Region defined as the cross-sectional Individuals with history of ocular trauma, uveitis, sur- area of the iris from the scleral spur to gery, or laser therapy were excluded from the study. Eyes the pupil Iris curvature (I-Curv) The perpendicular distance from a line with iris or angle neovascularization, pseudoexfoliation, between the most central to the most secondary angle closure, or any iris or corneal abnormalities peripheral points of the iris pigment were also excluded. Miotic or mydriatic medications were epithelium to the posterior iris surface not used in any of the patients prior to imaging. at the point of greatest convexity Slit-lamp examination of the anterior segment, Iris thickness (IT) Iris thickness at 750 mm from the scleral Goldmann applanation tonometry, and gonioscopy (with spur (IT750) and without indentation) in dark conditions, using a Lens vault (LV) The perpendicular distance from the Zeiss-style 4-mirror goniolens (Model G-4; Volk Optical, anterior pole of the lens to the Mentor, Ohio, USA) with a narrow 1-mm beam of light horizontal line between the scleral were conducted for all the patients before LPI and 6 weeks spurs later. The Shaffer grading system was used to evaluate the Anterior vault (AV) The perpendicular distance from the corneal endothelium to the horizontal angle on gonioscopy. The angle was considered ‘‘closed’’ line between the scleral spurs when posterior trabecular meshwork was not visible on gonioscopy of that quadrant.

ANTERIOR SEGMENT OPTICAL COHERENCE TOMOGRA- Two experienced ophthalmologists (S.M., M.M) deter- phy: Anterior segment optical coherence tomography mined scleral spur location in each image. The principal (AS-OCT) (Visante OCT; Carl Zeiss Meditec, Dublin, investigator (S.M.) validated all the images for quality and California, USA) was performed before LPI and again scleral spur location. After identifying the scleral spur, 6 weeks later in dark ambient lighting. Scans were centered the software automatically measures anterior segment on the pupil and were obtained along the horizontal axis parameters, including anterior chamber depth (ACD), ante- using the Enhanced Anterior Segment Single protocol. rior chamber area (ACA), iris thickness (IT), iris area, iris Three images were captured, and the highest-quality image curvature, LV, lens thickness, and anterior vault (AV); and was used for analysis using the Zhongshan Angle Assess- angle parameters, including angle opening distance at 500 ment Program (ZAAP; Zhongshan Ophthalmic Center, and 750 mm from the scleral spur (AOD500, AOD750) Guangzhou, China). and trabecular iris space area at 500 and 750 mmfromthe

134 AMERICAN JOURNAL OF OPHTHALMOLOGY JUNE 2016 scleral spur (TISA500, TISA750). Table 1 demonstrates the definitions of the AS-OCT anterior segment and angle pa- TABLE 2. Changes in Mean Anterior Segment Parameters rameters analyzed in this study.6,7,19,20 Before and After Laser Peripheral Iridotomy in Acute Primary The AS-OCT images were also assessed grossly for the Angle Closure Eyes presence and extent of iridotrabecular contact. The angles Variable Pre-LPI Post-LPI P Value were classified as ‘‘closed’’ if there was iridotrabecular con- tact anterior to the scleral spur and ‘‘open’’ if there was Anterior segment parameters 6 6 visible space between the iris root and trabecular meshwork. ACD (mm) 1.81 0.24 1.86 0.19 .001* m 6 6 The eyes were classified as having an ‘‘exaggerated LV’’ if Lens vault ( m) 1081.1 265.9 1051.7 256.9 .01* ACA (mm2) 12.53 6 1.84 13.25 6 1.88 <.001* the LV was greater than one-third of the anterior vault, Anterior vault (mm) 2.89 6 0.23 2.91 6 0.20 .37 which is the distance between the posterior corneal surface Iris parameters 19,21 and the line connecting the scleral spurs. IT750 (mm2) 0.49 6 0.08 0.47 6 0.11 .97 All patients underwent LPI as standard medical care I-Curve (mm) 0.30 6 0.11 0.22 6 0.13 .01* within 48 hours of the onset of the APAC attack. I-Area (mm) 1.49 6 0.31 1.46 6 0.31 .98 Patients with prior LPI were excluded. The pupil was Nasal angle constricted with 2%, and LPI was performed AOD500 (mm) 0.030 6 0.052 0.066 6 0.069 <.001* using the ophthalmic neodymium–yttrium-aluminum-garnet AOD750 (mm) 0.053 6 0.074 0.112 6 0.094 .001* 2 (Nd:YAG) laser (Laserex Tango Nd:YAG; Ellex Medical, TISA500 (mm ) 0.017 6 0.028 0.030 6 0.030 .03* m 2 6 6 Adelaide, Australia) and an Abraham iridotomy contact TISA750 ( m ) 0.028 0.041 0.054 0.050 .009* lens. Finally, all eyes underwent dilated fundus examination Temporal angle AOD500 (mm) 0.031 6 0.049 0.093 6 0.082 <.001* including stereoscopic examination of the optic nerve head. AOD750 (mm) 0.057 6 0.068 0.142 6 0.108 .01* TISA500 (mm2) 0.013 6 0.021 0.039 6 0.035 <.001* STATISTICAL ANALYSIS: Statistical analysis was TISA750 (mm2) 0.025 6 0.034 0.070 6 0.055 <.001* performed using SPSS software version 17 (SPSS Inc, Chicago, Illinois, USA). ACA ¼ anterior chamber area; ACD ¼ anterior chamber depth; Pre-LPI and post-LPI gonioscopy grades were compared AOD500/AOD750 ¼ angle opening distance at 500/750 mm from using the x2 test. Nonparametric tests were used for all sta- the scleral spur; I-Area ¼ iris area; I-Curve ¼ iris curvature; ¼ m ¼ tistical analysis. Comparison of mean values before and IT750 iris thickness at 750 m from the scleral spur; LPI laser ¼ after LPI was performed using Wilcoxon signed rank for peripheral iridotomy; TISA500/TISA750 trabecular iris space area at 500/750 mm from the scleral spur. nonparametric data. Baseline AS-OCT parameters and P values with an asterisk are statistically significant. their changes at 6 weeks after LPI were compared between the Exaggerated and Nonexaggerated LV groups using the Mann-Whitney U test. A linear mixed-effects model was used to compare used for the analysis. The mean age was 60.7 6 9.2 years changes in AOD750 between the nasal and temporal (range, 45–88). Mean IOP was 45.0 6 12.1 mm Hg at pre- angles and to estimate the relationship between AOD750 sentation and reduced to 11.87 6 5.15 mm Hg after changes and potential influential factors. The random the attack was broken. Mean Lens Opacities Classification effect was the individual study participants; the fixed effects System score for lens nucleus opacification was 2.78 6 were age, sex, amount of peripheral anterior synechiae 0.84. The mean axial length and lens thickness were (PAS), AL, lens thickness, AOD750, AOD500, and other 21.91 6 1.11 and 5.03 6 0.45 mm, respectively. AS-OCT variables. Univariate regression was performed Prior to LPI, 49 of 52 (94%) nasal, 45 of 52 (87%) tem- between potential confounders and change in AOD750. poral, 45 of 52 (88%) superior, and 49 of 52 (94%) inferior Those variables with P < .20 and variance inflation factor angles were gonioscopically closed before LPI. After LPI, less than 3 were included in the final multivariate model, 28 of 52 (54%) nasal, 26 of 52 (50%) temporal, 42 of 52 along with age and sex. A P value of less than .05 was (72%) superior, and 29 of 52 (56%) inferior angles were considered statistically significant. still closed. The mean gonioscopic grade was 0.25 6 0.34 and increased to 1.22 6 0.86 after LPI (P < .001). Based on the extent of iridotrabecular contact in AS- OCT images, nasal and temporal angles were closed in 44 RESULTS of 48 (91%) and 43 of 51 (84%) angles, respectively. After LPI, 25 of 52 (48%) nasal quadrants and 28 of 52 (54%) FIFTY-NINE EYES OF 59 PATIENTS WITH BROKEN APAC temporal quadrants remained closed (P ¼ .007 and P ¼ attacks were enrolled in the study. Seven cases were .003, respectively). excluded owing to loss to follow-up (4 cases) or unclear Both nasal and temporal angle parameters increased visualization of the scleral spur in AS-OCT images (3 significantly after LPI (Table 2). Although ACD cases). Data from 52 patients (37 female, 15 male) were and ACA increased significantly after LPI (P ¼ .002 and

VOL. 166 PERIPHERAL IRIDOTOMY EFFECT ON ACUTE PRIMARY ANGLE CLOSURE 135 groups (Table 5). The extent of change in angle parameters after LPI was not significantly different between the 2 groups. Change in LV after LPI was greater in the Exagger- ated LV group than in the Nonexaggerated LV group (50.2 6 159.2 and 38.6 6 146.3 mm, respectively; P ¼ .04). All other parameters were not significantly different between groups.

DISCUSSION

IN THIS PROSPECTIVE STUDY, GONIOSCOPIC AND BIOMETRIC properties of 52 APAC eyes were evaluated before and after LPI. There was a significant decrease in iris curvature, and an increase in anterior chamber depth, anterior chamber FIGURE. Scatterplot demonstrating negative association of amount of change in angle opening distance at 750 mm and pre- area, and all angle parameters after LPI, as measured by operative angle opening distance at 750 mm in univariate AS-OCT. Approximately 50% of angles have still some (ß [ L0.622, P < .001) and multivariate model degree of iridotrabecular meshwork contact after LPI. After (ß [ L0.992, P [ .02). Solid line shows the univariate fitted multivariate analysis, pre-LPI AOD750 was the only base- line and dashed line shows the association after adjusting for age, line parameter that predicted change in AOD750. The sex, and confounders with P < .20 in univariate analysis and smaller the AOD750 before the procedure, the greater open- variance inflation factor less than 3. ing would be expected after LPI. Although the change in gonioscopic angle grading was statistically significant (0.25 6 0.34 pre-LPI to 1.22 6 P < .001, respectively), there was no significant change in 0.86 post-LPI), only 46% of nasal, 50% of temporal, 19% anterior vault. Lens vault decreased at post-LPI visit and of superior, and 44% of inferior angles were open 6 weeks the amount of change was associated with preoperative after LPI. Similar gonioscopic results have been previously LV (ß ¼0.127, P ¼ .04). The mean iris curvature reported by other investigators. Lim and associates followed was 0.30 6 0.11 mm before LPI and decreased to 44 eyes with APAC for 1 year after LPI and found a signif- 0.22 6 0.13 mm after LPI (P ¼ .01) (Table 2). However, icant increase in angle width 2 weeks after LPI and no there was no significant change in IT750 or iris area. further change thereafter.22 He and associates4 also demon- In the linear mixed model, only angle parameters strated that LPI results in a significant increase in the angle (AOD750 and AOD500) were significantly correlated width in eyes that were primary angle closure suspect with change in AOD750. In the final multivariate model (PACS). However, residual angle closure after LPI was adjusting for age, sex, and confounders with P < .20 in uni- seen in one-fifth of their PACS cases. Recent evidence variate analysis and variance inflation factor less than 3, suggests that the amount of residual angle closure is even the only significant predictor of angle opening was higher in APAC eyes. After relief of pupillary block by AOD750 (b ¼0.704, P ¼ .02) (Figure) Change in LPI, residual angle closure may occur in up to 40% of AOD was not associated with age, sex, axial length, iris these eyes.23 parameters (iris thickness, area, or curvature), anterior In our study, approximately 50% of the temporal or nasal segment parameters (ACD, ACA, LV, lens thickness, angles still had some appositional closure on AS-OCT after and anterior vault), or amount of PAS (Table 3). LPI, although this proportion is lower compared to the 80% Exaggerated LV was present in 61.5% of the cases (32 of seen preoperatively. An ultrasound biomicroscopy (UBM) 52 eyes) before LPI. The relative LV ratio, calculated as LV study among Indian eyes with PACG showed that 60% of divided by the sum of LV and ACD, was 0.41 6 0.05 in the angles in eyes with patent LPI persisted in being narrow, Exaggerated LV group and 0.26 6 0.03 in the Nonexagger- largely owing to anteriorly rotated ciliary processes.24 In a ated LV group (P < .001) (Table 4). large retrospective survey of Asian APAC cases, Aung Eyes with exaggerated LV had smaller ACD and ACA and associates reported that a high proportion (58.1%) of and greater lens thickness and LV than those without. these eyes developed an increase in IOP on long-term However, there was no significant difference in axial follow-up after resolution of the acute attack despite the length, anterior vault, iris parameters (thickness, area, presence of a patent LPI.2 and curvature), or the amount of PAS before LPI between Several studies on PACS have shown a significant in- the 2 groups. crease in all average angle parameters measured by UBM The change in anterior segment parameters after LPI was or AS-OCT.10,12–14,25–27 A recent study by Kansara and compared between Exaggerated and Nonexaggerated LV associates26 reported that changes in angle parameters

136 AMERICAN JOURNAL OF OPHTHALMOLOGY JUNE 2016 TABLE 3. Predictive Factors Related to Change in Angle Opening Distance at 750 mm After Laser Peripheral Iridotomy

Univariate Analysis Multivariate Analysisa

Variable B (95% Confidence Interval) P Value B (95% Confidence Interval) P Value

Age (y) 0.004 (0.002 to 0.001) .71 0.0008 (0.003 to 0.001) .44 Sex (F/M) 0.003 (0.057 to 0.050) .88 0.024 (0.072 to 0.024) .41 PAS 0.0002 (0.0001–0.0006) .19 0.0006 (0.0002 to 0.0003) .64 Axial length (mm) 0.005 (0.028–0.017) .62 - - ACD (mm2) 0.015 (0.094 to 0.126) .77 - - Lens thickness (mm) 0.030 (0.082 to 0.240) .26 - - ACA (mm2) 0.001 (0.012 to 0.009) .73 - - Lens vault (mm) 0.0002 (0.0007 to 0.0001) .27 - - Relative lens vault 0.089 (0.234 to 0.413) .58 - - Anterior vault 0.521 (0.544 to 0.158) .33 - - Iris parameters IT750 (mm) 0.005 (0.042 to 0.054) .80 - - I-Area (mm2) 0.002 (0.033 to 0.1029) .86 - - I-Curve (mm) 0.005 (0.110 to 0.098) .91 - - Angle parameters AOD750 (mm) 0.599 (0.895 to 0.303) <.001* 0.992 (1.937 to 0.048) .02* AOD500 (mm) 0.622 (1.045 to 0.199) .005* 0.907 (0.252 to 2.067) .12

ACA ¼ anterior chamber area; ACD ¼ anterior chamber depth; ACW ¼ anterior chamber width; AOD500/AOD750 ¼ angle opening distance at 500/750 mm from the scleral spur; I-Area ¼ iris area; I-Curve ¼ iris curvature; IT750 ¼ iris thickness at 750 mm from the scleral spur. P values with an asterisk are statistically significant. aIncluding age, sex, and those variable with P < .20 in univariate analysis and variance inflation factor less than 3.

were greater in the nasal quadrant and less in the superior findings, Unterlauft and associates18 demonstrated that quadrant. In this study, we observed a significant angle LPI led to increased ACD and volume in APAC cases by widening by biometric parameters (AOD500, AOD750, Scheimpflug photography. In the present study, lens vault TISA500, and TISA750) in APAC eyes. LPI is the stan- decreased at the post-LPI visit. The amount of decrease dard first-line intervention for APAC and their fellow was greater in eyes with greater baseline lens vault. We eyes.4,28 It prevents recurrence of acute episodes and postulate that the anterior chamber changes are related eliminates the risk of acute attacks in fellow eyes. LPI to changes in both iris and lens profiles, and posterior eliminates the pressure gradient between the anterior and movement of the lens may contribute to this increase. posterior chambers and flattens the iris, allowing the Choroidal expansion has been reported as one of the trigger peripheral iris to fall backward and widen the angle. Iris factors that could contribute to development of APAC. In flattening after LPI has been shown in a study by Huang some recent studies using AS-OCT, investigators demon- and associates25 on PACS patients. Our study demonstrates strated a shallower ACD in APAC eyes before therapeutic that iris curvature similarly decreases and that the iris pro- interventions when compared to the fellow PACS eyes. file becomes flatter in APAC eyes. Although many of these Choroidal thickening has also been observed in APAC eyes had an exaggerated LV with a ‘‘volcano-like’’ iris eyes by enhanced depth imaging OCT when the IOP is configuration and small iris curvature, LPI was effective reduced, which would favor the choroidal expansion in flattening the iris in these eyes. hypothesis.31,32 The shift in lens position may in fact be Although most studies on PACS do not support a signif- due not to the LPI but to a resolution of inflammation of icant change in axial ACD after LPI,4,13 anterior chamber and choroid. volume may increase.10,13 Jain and associates29 performed Kumar and associates also demonstrated uveal effusion Scheimpflug photography on Indian patients with PACS in a fourth of their cases that might explain ciliary body who were treated with LPI and found that LPI resulted in rotation and forward movement of the iris lens dia- significant increase of anterior chamber volume and phragm.33 However, Yang and associates used ultrasonog- peripheral ACD but not central ACD. A recent study, raphy to measure anterior segment dimensions and did however, has shown that ACD increased in Japanese pop- not observe significant changes in biometric ocular mea- ulation with variety of angle closure after LPI.30 Interest- surements before and 2 weeks after LPI in eyes affected ingly, we found a significant increase in both central by APAC.34 This discrepancy might be due to differences ACD and ACA in our APAC eyes. Consistent with our in anterior segment morphology in different ethnicities as

VOL. 166 PERIPHERAL IRIDOTOMY EFFECT ON ACUTE PRIMARY ANGLE CLOSURE 137 TABLE 4. Demographics and Anterior Segment Variables in TABLE 5. Changes in Mean Anterior Segment Parameters Exaggerated Lens Vault and Nonexaggerated Lens Vault Before and After Laser Peripheral Iridotomy in Exaggerated Groups Measured by A-Scan Ultrasonography or Anterior Lens Vault and Nonexaggerated Lens Vault Groups Segment Optical Coherence Tomography Measured by Anterior Segment Optical Coherence Tomography Exaggerated Lens Nonexaggerated Variable Vault (N ¼ 32) Lens Vault (N ¼ 20) P Value Exaggerated Nonexaggerated Variable (Changes) Lens Vault (N ¼ 32) Lens Vault (N ¼ 20) P Value Age (y) 63.2 6 9.7 56.3 6 6.7 .009* Sex (F/M) 25/7 12/8 .13 ACD (mm) change 0.06 6 0.12 0.02 6 0.06 .13 Nuclear opacity 3.00 6 0.79 2.46 6 0.83 .05* ACA (mm2) 0.94 6 1.13 0.72 6 0.80 .56 (LOCS III) Lens vault (mm) 74.4 6 126.4 10.6 6 74.9 .04* PAS 66.4 6 107.3 77.5 6 126.8 .97 Anterior vault 0.01 6 0.16 0.06 6 0.02 .58 Axial length (mm) 21.75 6 1.33 22.14 6 0.62 .32 Iris parameters ACD (mm) 1.68 6 0.16 2.03 6 0.19 <.001* I-curve change 0.06 6 0.16 0.07 6 0.14 .58 ACA (mm2) 11.70 6 1.32 13.84 6 1.82 <.001* Nasal angle Lens thickness (mm) 5.25 6 0.42 4.71 6 0.31 <.001* AOD500 (mm) 0.045 6 0.069 0.011 6 0.103 .40 Lens vault (mm) 1206.7 6 214.0 814.4 6 121.1 <.001* AOD750 (mm) 0.064 6 0.086 0.047 6 0.014 .92 Anterior vault (mm) 2.89 6 0.24 2.84 6 0.21 .48 TISA500 (mm2) 0.015 6 0.032 0.002 6 0.034 .55 Iris parameters TISA750 (mm2) 0.030 6 0.050 0.010 6 0.069 .73 IT750 (mm) 0.49 6 0.09 0.65 6 0.08 .92 Temporal angle change I-Area (mm2) 1.49 6 0.33 1.49 6 0.30 .19 AOD500 (mm) 0.057 6 0.056 0.073 6 0.105 .73 I-Curve (mm) 0.31 6 0.11 0.27 6 0.11 .90 AOD750 (mm) 0.085 6 0.080 0.093 6 0.017 .92 Nasal angle TISA500 (mm2) 0.024 6 0.025 0.029 6 0.051 .65 AOD500 (mm) 0.026 6 0.035 0.036 6 0.072 .40 TISA750 (mm2) 0.042 6 0.036 0.049 6 0.077 .98 AOD750 (mm) 0.041 6 0.040 0.073 6 0.108 .92 ¼ ¼ TISA500 (mm2) 0.019 6 0.026 0.015 6 0.032 .55 ACA anterior chamber area; ACD anterior chamber depth; ¼ m TISA750 (mm2) 0.028 6 0.032 0.030 6 0.053 .73 AOD500/AOD750 angle opening distance at 500/750 m from ¼ ¼ Temporal angle the scleral spur; I-Curve iris curvature; TISA500/TISA750 m AOD500 (mm) 0.033 6 0.041 0.028 6 0.061 .73 trabecular iris space area at 500/750 m from the scleral spur. P AOD750 (mm) 0.049 6 0.044 0.071 6 0.094 .92 values with an asterisk are statistically significant. TISA500 (mm2) 0.016 6 0.020 0.010 6 0.010 .65 TISA750 (mm2) 0.026 6 0.028 0.024 6 0.024 .98

ACA ¼ anterior chamber area; ACD ¼ anterior chamber depth; study by Huang and associates,25 angle deepening was posi- AOD500/AOD750 ¼ angle opening distance at 500/750 mm from tively correlated with age and iris curvature and negatively the scleral spur; I-Area ¼ iris area; I-Curve ¼ iris curvature; correlated with preoperative iris area. How and associates IT750 ¼ iris thickness at 750 mm from the scleral spur; LOCS did not find iris thickness at the root predictive of angle ¼ ¼ III Lens Opacities Classification System III; LPI laser periph- opening. They postulated that iris thickness is a dynamic eral iridotomy; PAS ¼ peripheral anterior synechia; TISA500/ ¼ m factor, so static measurement may lead to different re- TISA750 trabecular iris space area at 500/750 m from the 13 scleral spur. sults. The same authors also reported greater angle open- P values with an asterisk are statistically significant. ing with greater baseline LV. Similarly, Lee and associates showed that the extent of angle opening is inversely depen- dent on pre-LPI angle measurements.15 In the present study, the amount of angle opening in response to LPI well as in the severity of attack and timing of the second was only correlated inversely with pre-LPI AOD750. measurement.34 This discrepancy can be explained by differences in study AS-OCT provides a noncontact assessment of the populations and angle closure subtypes. Although iris anterior chamber parameters and quantification of angle, thickness, iris curvature, and LV are greater in APAC iris, and anterior segment structures.35 Quantitative mea- eyes than in other subtypes of angle closure,5,38 they do surement of anterior segment anatomy is important for not appear to influence angle change in these eyes. evaluating the effectiveness of a particular treatment in Han and associates performed cluster analysis of 88 different subtypes of angle closure, even if the exact clinical Korean subjects with primary angle closure spectrum dis- implications are still uncertain.35,36 To our knowledge, orders and reported 2 distinct clusters with different there are no studies in the literature evaluating biometric anatomic features. The first cluster had lower ACD, predictors of angle opening after LPI in APAC eyes. AOD750, and anterior chamber width and greater LV, However, several studies on PACS showed that iris while the second cluster had greater iris thickness. They thickness, iris curvature, baseline angle opening, and LV found that the proportional change in AOD750 after may affect the anatomic LPI efficacy.10,13,15,25,26,37 In a LPI differed between clusters, with a greater change

138 AMERICAN JOURNAL OF OPHTHALMOLOGY JUNE 2016 seen in the first cluster.12 We also classified our APAC of this study may not be applicable to other ethnic groups. subjects based on LV. Studies in Chinese, Iranian, and Another limitation is the low AS-OCT image quality of Singaporean ethnicities have confirmed the prominent the superior and inferior angles because of interference role of LV in APAC eyes.5,6,39,40 Exaggerated LV has with eyelids, which led us to use only the nasal and been reported to be more common in APAC eyes.19 In temporal angle parameters for analysis. Although images some of these cases, the iris configuration is ‘‘volcano- taken during the attack may better provide additional like’’ because of a normal or large lens pushing the iris for- data about risk factors, significant corneal edema prevented ward, leading to a small anterior chamber volume.8,9,19 In obtaining good-quality images for evaluation of the angle, our series of APAC eyes, the Exaggerated LV group had iris, and anterior segment. Finally, it was impossible to older age, shallower ACD, smaller ACA, and greater mask the observer to pre-LPI and post-LPI status when LV and thickness. All other iris and angle parameters collecting gonioscopic data, and there is the possibility of were not significantly different between the Exaggerated inaccurate evaluation of PAS in cases with very shallow and Nonexaggerated groups. We also did not observe anterior chambers or steep irises. any significant difference in amount of angle deepening In summary, laser peripheral iridotomy leads to a signif- between the 2 groups. The lens was found to move icant increase in the angle width in APAC eyes. Anterior more posteriorly after LPI in the Exaggerated LV group. chamber depth and area increased and iris curvature We speculate that vitreous/choroid expansion might be decreased after LPI. Although lens vault decreases 6 weeks a parallel mechanism to greater lens thickness that after resolution of attack, it may not influence the efficacy results in exaggerated LV. of LPI in these eyes. The only factor that predicts the This study has some limitations that should be kept in amount of angle widening after LPI in APAC eyes in our mind. Because patients were of Iranian descent, the results study is angle width.

FUNDING/SUPPORT: NO FUNDING OR GRANT SUPPORT. FINANCIAL DISCLOSURES: SHAN C. LIN IS A CONSULTANT FOR ALLER- gan, Alimera, and Iridex. The following authors have no financial disclosures: Sasan Moghimi, Rebecca Chen, Mohammadkarim Johari, Faezeh Bijani, Massood Mohammadi, Alireza Khodabandeh, and Mingguang He. All authors attest that they meet the current ICMJE criteria for authorship. The authors thank Mrs Nassim Khatibi for her assistance in gathering data.

REFERENCES optical coherence tomography in primary angle closure mech- anisms. Clin Experiment Ophthalmol 2012;40(8):792–801. 1. Ang GS, Wells AP. Changes in Caucasian eyes after laser 9. Moghimi S, Zandvakil N, Vahedian Z, et al. Acute angle peripheral iridotomy: an anterior segment optical coherence closure: qualitative and quantitative evaluation of the tomography study. Clin Experiment Ophthalmol 2010;38(8): anterior segment using anterior segment optical coherence to- 778–785. mography. Clin Experiment Ophthalmol 2014;42(7):615–622. 2. Aung T, Ang LP, Chan S-P, Chew PT. Acute primary angle- 10. Ang GS, Wells AP. Factors influencing laser peripheral closure: long-term intraocular pressure outcome in Asian iridotomy outcomes in white eyes: an anterior segment eyes. Am J Ophthalmol 2001;131(1):7–12. optical coherence tomography study. J Glaucoma 2011; 3. Lam DS, Leung DY, Tham CC, et al. Randomized trial of 20(9):577–583. early phacoemulsification versus peripheral iridotomy to pre- 11. Buckley S, Reeves B, Burdon M, et al. Acute angle closure vent intraocular pressure rise after acute primary angle glaucoma: relative failure of YAG iridotomy in affected closure. Ophthalmology 2008;115(7):1134–1140. eyes and factors influencing outcome. Br J Ophthalmol 1994; 4. He M, Friedman DS, Ge J, et al. Laser peripheral iridotomy in 78(7):529–533. primary angle-closure suspects: biometric and gonioscopic 12. Han S, Sung KR, Lee KS, Hong JW. Outcomes of laser outcomes: the Liwan Eye Study. Ophthalmology 2007; peripheral iridotomy in angle closure subgroups according 114(3):494–500. to anterior segment optical coherence tomography parame- 5. Moghimi S, Vahedian Z, Fakhraie G, et al. Ocular biometry ters. Invest Ophthalmol Vis Sci 2014;55(10):6795–6801. in the subtypes of angle closure: an anterior segment optical 13. How AC, Baskaran M, Kumar RS, et al. Changes in anterior coherence tomography study. Am J Ophthalmol 2013;155(4): segment morphology after laser peripheral iridotomy: an 664–673. anterior segment optical coherence tomography study. 6. Nongpiur ME, He M, Amerasinghe N, et al. Lens vault, thick- Ophthalmology 2012;119(7):1383–1387. ness, and position in Chinese subjects with angle closure. 14. Lee JW, Lee JH, Lee KW. Prognostic factors for the success of Ophthalmology 2011;118(3):474–479. laser iridotomy for acute primary angle closure glaucoma. 7. Nongpiur ME, Sakata LM, Friedman DS, et al. Novel associ- Korean J Ophthalmol 2009;23(4):286–290. ation of smaller anterior chamber width with angle closure in 15. Lee RY, Kasuga T, Cui QN, et al. Association between base- Singaporeans. Ophthalmology 2010;117(10):1967–1973. line iris thickness and prophylactic laser peripheral iridotomy 8. Shabana N, Aquino MC, See J, et al. Quantitative evaluation outcomes in primary angle-closure suspects. Ophthalmology of anterior chamber parameters using anterior segment 2014;121(6):1194–1202.

VOL. 166 PERIPHERAL IRIDOTOMY EFFECT ON ACUTE PRIMARY ANGLE CLOSURE 139 16. Saunders D. Acute closed-angle glaucoma and Nd-YAG laser 29. Jain R, Grewal D, Grewal S. Quantitative analysis of anterior iridotomy. Br J Ophthalmol 1990;74(9):523–525. chamber following peripheral laser iridotomy using Pentacam 17. Sawada A, Yamamoto T. Correlation between extent of in eyes with primary angle closure. Eur J Ophthalmol 2012; preexisting organic angle closure and long-term outcome http://dx.doi.org/10.5301/ejo.5000158. after laser peripheral iridotomy in eyes with primary angle 30. Ang BC, Nongpiur ME, Aung T, Mizoguchi T, Ozaki M. closure. J Glaucoma 2012;21(3):174–179. Changes in Japanese eyes after laser peripheral iridotomy: 18. Unterlauft J, Yafai Y, Wiedemann P. Changes of anterior an anterior segment optical coherence tomography study. chamber architecture induced by laser peripheral iridotomy Clin Experiment Ophthalmol 2015; http://dx.doi.org/10.1111/ in acute angle closure crisis. Int Ophthalmol 2015;35(4): ceo.12673. 549–556. 31. Wang W, Zhou M, Huang W, Chen S, Ding X, Zhang X. Does 19. Moghimi S, Vahedian Z, Zandvakil N, et al. Role of lens vault acute primary angle-closure cause an increased choroidal in subtypes of angle closure in Iranian subjects. Eye 2014; thickness? Invest Ophthalmol Vis Sci 2013;54(5):3538–3545. 28(3):337–343. 32. Zhang X, Wang W, Aung T, Jonas JB, Wang N. Choroidal 20. Sng CC, Allen JC, Nongpiur ME, et al. Associations of iris physiology and primary angle closure disease. Surv Ophthalmol structural measurements in a Chinese population: The 2015;60(6):547–556. Singapore Chinese Eye Study. Invest Ophthalmol Vis Sci 33. Kumar RS, Quek D, Lee KY, et al. Confirmation of the pres- 2013;54(4):2829–2835. ence of uveal effusion in Asian eyes with primary angle 21. Sihota R, Vashisht P, Sharma A, Chakraborty S, Gupta V, closure glaucoma: an ultrasound biomicroscopy study. Arch Pandey RM. Anterior segment optical coherence tomography Ophthalmol 2008;126(12):1647–1651. characteristics in an Asian population. J Glaucoma 2012; 34. Yang M, Aung T, Husain R, et al. Choroidal expansion as a 21(3):180–185. mechanism for acute primary angle closure: an investigation 22. Lim LS, Aung T, Husain R, Wu Y-J, Gazzard G, Seah SK. into the change of biometric parameters in the first 2 weeks. Acute primary angle closure: configuration of the drainage Br J Ophthalmol 2005;89(3):288–290. angle in the first year after laser peripheral iridotomy. 35. Radhakrishnan S, Huang D, Smith SD. Optical coherence to- Ophthalmology 2004;111(8):1470–1474. mography imaging of the anterior chamber angle. Ophthalmol 23. Nonaka A, Iwawaki T, Kikuchi M, Fujihara M, Nishida A, Clin North Am 2005;18(3):375–381. Kurimoto Y. Quantitative evaluation of iris convexity in pri- 36. Eslami Y, Latifi G, Moghimi S, et al. Effect of adjunctive mary angle closure. Am J Ophthalmol 2007;143(4):695–697. viscogonioplasty on drainage angle status in cataract surgery: 24. Garudadri CS, Chelerkar V, Nutheti R. An ultrasound bio- a randomized clinical trial. Clin Experiment Ophthalmol 2013; microscopic study of the anterior segment in Indian eyes 41(4):368–378. with primary angle-closure glaucoma. J Glaucoma 2002; 37. Lin Z, Liang Y, Wang N, et al. Peripheral anterior synechia 11(6):502–507. reduce extent of angle widening after laser peripheral iridot- 25. Huang G, Gonzalez E, Lee R, et al. Anatomic predictors for omy in eyes with primary angle closure. J Glaucoma 2013; anterior chamber angle opening after laser peripheral iridot- 22(5):374–379. omy in narrow angle eyes. Curr Eye Res 2012;37(7):575–582. 38. Guzman CP, Gong T, Nongpiur ME, et al. Anterior segment 26. Kansara S, Blieden LS, Chuang AZ, et al. Effect of laser optical coherence tomography parameters in subtypes of peripheral iridotomy on anterior chamber angle anatomy in primary angle closure. Invest Ophthalmol Vis Sci 2013;54(8): primary angle closure spectrum eyes. J Glaucoma 2015; 5281–5286. 25(5):e469–e474. 39. Gazzard G, Friedman DS, Devereux JG, Chew P, Seah SK. A 27. Yun SC, Hong JW, Sung KR, Lee JY. Effects of laser periph- prospective ultrasound biomicroscopy evaluation of changes eral iridotomy in subgroups of primary angle closure based on in anterior segment morphology after laser iridotomy in iris insertion. J Ophthalmol 2015;2015:581719. Asian eyes. Ophthalmology 2003;110(3):630–638. 28. Husain R, Gazzard G, Aung T, et al. Initial management of 40. Nongpiur ME, Gong T, Lee HK, et al. Subgrouping of primary acute primary angle closure: a randomized trial comparing angle-closure suspects based on anterior segment optical phacoemulsification with laser peripheral iridotomy. Ophthal- coherence tomography parameters. Ophthalmology 2013; mology 2012;119(11):2274–2281. 120(12):2525–2531.

140 AMERICAN JOURNAL OF OPHTHALMOLOGY JUNE 2016