Variations in Iris Volume with Physiologic Mydriasis in Subtypes of Primary Angle Closure Glaucoma

Abstract

Purpose.: To compare the changes in iris volume with pupil dilation using anterior segment optical coherence tomography (AS-OCT) in eyes of subjects with different subtypes of primary angle closure.

Methods.: This prospective study examined 44 fellow eyes (FA group) of subjects with previous acute primary angle closure (APAC), and 56 subjects (AC group) with chronic primary angle closure and/or primary angle closure glaucoma. All participants underwent gonioscopy and AS-OCT imaging. The iris volume, iris cross-sectional area, and pupil diameter were measured with custom semiautomated software. The main outcome variable analyzed was mean change in iris volume between light and dark conditions in a multivariate linear regression analysis.

Results.: Thirty-five eyes from the FA group (79.5%) and 50 eyes from the AC group (89.3%) were included in the final analysis. When going from light to dark, iris volume did not change significantly in eyes in the FA group (+1.50 ± 6.73 mm³; P = 0.19), but decreased in the AC group by 1.52 ± 3.07 mm³ (P < 0.001). This difference was significant (P = 0.01). On multivariate analysis after controlling for age, sex, baseline pupil diameter, and change in pupil diameter, age (β = −0.397; P < 0.001) and diagnostic category (AC versus FA group; β = 0.347; P < 0.001) were significant determinants of iris volume change.

Conclusions.: With physiologic mydriasis, the iris volume decreased in eyes with chronic angle closure but remained unchanged in fellow eyes of APAC. Such variations in iris volume responses may influence the subtype of angle closure that develops.

Introduction

About 30% of the 60 million people afflicted with glaucoma worldwide have primary angle closure glaucoma (PACG). ¹ The proportion of blindness among subjects with PACG was reported to be greater than 25%, and a large proportion of those with PACG live in Asia. ^2–5 Although previous studies have shown biometric factors such as shallower anterior chamber depth (ACD), thicker lens and shorter axial length (AL) are associated with angle closure, ^6–14 no specific factors are known that predict who will develop acute primary angle closure (APAC), which is often devastating. ¹⁵  

Quigley et al. ¹⁶ recently postulated that the pathogenesis of angle closure may be related to dynamic processes related to the iris and the choroid rather than just static anatomic differences. In a study using anterior segment optical coherence tomography (AS-OCT) to measure iris cross-sectional area, the authors found that compared with eyes with open angles, eyes with angle closure lost less iris cross-sectional area with pupil dilation (both under physiologic conditions and after pharmacologic dilation). When Aptel et al. ¹⁷ used AS-OCT to assess dynamic changes in iris volume with pharmacologic pupil dilation, the iris volume was found to increase with pupil dilation in the fellow eyes of subjects with APAC, whereas iris volume decreased in age-matched, normal eyes. The results of a further study by the same group suggest that this unique feature of iris behavior was specific to subjects with APAC. ¹⁸ This pattern of increased residual iris volume with pupil dilation in predisposed eyes may thus be an important determinant for development of APAC during mydriasis. However, there have been no studies that have compared the iris behavior in subjects with different subtypes of angle closure. Such variations in iris dynamics may determine the clinical spectrum of angle closure. 

The aim of this study was to compare fellow eyes of subjects with APAC to eyes of subjects with chronic PAC/PACG, to determine if they have different responses to physiologic pupil dilation. 

Methods

This was a prospective study of individuals 40 years of age and older who were recruited from the outpatient clinics of Singapore National Eye Centre. The study adhered to the tenets of Declaration of Helsinki and had the approval of the hospital's Institutional Review Board, and written informed consent was obtained from all subjects. 

Two subgroups of angle closure were recruited; the definitions of each subgroup in the study were as follows. ^19,20  

1.  
   FA Group: This group consisted of fellow (nonattack) eyes of subjects with previous APAC. We defined APAC as a documented episode having at least two of the following symptoms: ocular or periocular pain, nausea, vomiting, or an antecedent history of intermittent blurring of vision. In addition, APAC eyes also had a presenting intraocular pressure (IOP) of >28 mm Hg on Goldmann applanation tonometry, and the presence of at least three of the following signs: conjunctival injection, corneal epithelial edema, mid-dilated nonreactive pupil, and shallow anterior chamber (AC). ²⁰ The fellow eyes were included in this category irrespective of the gonioscopic status of the angle or stage of angle closure.
2.  
   AC Group: Subjects with chronic PAC and PACG. These patients were asymptomatic and had narrow angles, defined as having the posterior trabecular meshwork not visible for at least 180° on nonindentation gonioscopy. In addition, PAC eyes had either peripheral anterior synechiae (PAS) or IOP > 21 mm Hg, but without glaucomatous optic neuropathy (GON), defined as a vertical cup: disc (CD) ratio > 0.7, CD asymmetry > 0.2, and/or focal notching with compatible visual field loss on static automated perimetry (SITA Standard algorithm with a 24‐2 test pattern, Humphrey Visual Field Analyzer II; Carl Zeiss Meditec, Dublin, CA). The latter was defined as a Glaucoma Hemifield Test outside normal limits with an abnormal pattern SD with P < 5% occurring in the normal population and fulfilling the test reliability criteria (fixation losses < −20%, false positives < −33%, and/or false negatives < 33%). PASs were defined as abnormal adhesions of the iris to the angle that were at least a half clock hour in width and were present to the level of the anterior trabecular meshwork or higher, despite indentation gonioscopy. PACG eyes had narrow angles (as defined above) with GON.

All subjects had previously undergone laser peripheral iridotomy at least 4 weeks before the study visit and were not on any topical steroid medications. Subjects with a prior history of any intraocular surgery or evidence of any secondary glaucoma were excluded. Subjects on any medication that has an effect on the iris (specifically miotics, mydriatics, and systemic tamsulosin therapy) were also excluded. If both eyes were eligible (AC group), one randomly selected eye was included for imaging and analysis. 

All subjects were asked about their medical and ophthalmic history and underwent a standardized eye examination that included assessment of Snellen visual acuity, IOP measurement by Goldmann applanation tonometry, gonioscopy using a four-mirror Sussman lens and central corneal thickness (CCT) measurement by ultrasound pachymetry (Echoscan, US-1800; Nidek Co., Ltd., Aichi, Japan). Biometric measurements included corneal curvature, ACD, and AL (measured by the IOLMaster 500, Carl Zeiss Meditec). We also performed dilated ophthalmoscopy of the optic disc and macula and visual field examination (SITA Standard algorithm with a 24‐2 test pattern, Humphrey Field Analyzer II-750i; Carl Zeiss Meditec). 

AS-OCT Imaging

A single trained technician performed AS-OCT imaging using the quad-scan mode, which ensures that images of the anterior segment are captured simultaneously along the four meridians (45, 90, 135, and 180°). Subjects adapted to the dark (0 lux) for at least 5 minutes prior to imaging (Fig. 1, top) and were imaged again after a brief pause while exposing the fellow eye to bright light (pen light, 1700 Lux; Fig. 1, bottom). The fellow eye was continuously exposed to the pen light during this phase of image capture process. 

Figure 1

 

Top: Examples of image capture using the anterior segment OCT in a “quad mode.” All meridians (180, 90, 45, and 135°) are captured simultaneously in the dark. The quad-scan mode ensures that error due to pupil size fluctuation is minimal. Bottom: Quad scan mode images of the same eye after exposure to light in fellow eye.

We analyzed OCT images using custom semiautomated software (Zhongshan Angle Assessment Program [ZAAP], Guangzhou, China). ²¹ An experienced observer (CZ) marked the scleral spur in each image and from this the software generated the following parameters (Fig. 2A): iris cross-sectional area, iris volume, anterior chamber width (ACW), iris curvature, iris thickness (IT750 and IT2000), and lens vault. The iris cross-sectional area was calculated as the cumulative cross-sectional area of the full length (from spur to pupil) of the iris. Iris volume calculations were based on the principles of centroid theorem. ²² In brief, the centroid and its coordinate were identified as an average of all pixel coordinates within the cross-sectional area by the software. The distance from centroid to the AC axis was defined as the radius, and the iris volume was then calculated by rotating the iris cross-sectional area along this radius. The ACW was defined as the horizontal scleral spur-to-spur distance and the lens vault was measured by estimating the perpendicular distance between the anterior pole of the crystalline lens and the horizontal line joining the two scleral spurs (Fig. 2B). The iris curvature (I-Curv) was estimated by the software by drawing a line from the most peripheral to the most central points of iris pigment epithelium and then a perpendicular line was extended from this line to the iris pigment epithelium at the point of greatest convexity. The IT750 and IT2000 were defined as the iris thickness (IT) measured at 750 and 2000 μm from the scleral spur, respectively. Averages of iris volume and cross-sectional area were obtained from three meridians (45, 135, and 180°) in each condition of illumination. The sum of average iris volume and iris cross-sectional areas from the nasal temporal sectors of the iris were designated as total iris volume and cross-sectional area, respectively, for that eye and that specific illumination condition. The 90° images were not used for analysis due to frequent lid artifacts resulting in poor cross-sectional images along that meridian. Average pupil diameter was measured and calculated using the caliper function of the software from the three meridians used (45, 135, and 180°) for both light and dark conditions. Images obtained in the dark were designated as baseline and averages of ACW, I-Curv, IT750, and lens vault were determined from these images. Changes in total iris volume and cross-sectional area between conditions of light and dark were evaluated in a multivariate model that included diagnostic groups and associated ocular and demographic factors. 

Figure 2

 

(A) Iris parameter as measured by Zhongshan Angle Assessment Program (Guangzhou, China). Once the scleral spur is identified and marked in each image, the software generates the following parameters: iris cross-sectional area, iris volume, anterior chamber width, iris curvature, and iris thickness (IT750 and IT2000). (B) ACW is derived from the horizontal spur-to-spur distance and lens vault is measured by estimating the perpendicular distance between the anterior pole of the crystalline lens and the horizontal line joining the two scleral spurs.

Statistical Analysis

Throughout, χ² or independent sample t-tests were used to estimate statistical differences in categorical data. Pairwise t-tests were conducted for data associated with changes in light and dark. Univariate regression was conducted with iris volume change between conditions of light and dark as a dependent variable and the effects of age, sex, ACD, AL, baseline pupil diameter, pupil diameter change, lens vault, ACW, and diagnosis were evaluated. Age, sex, and variables that were significant at a level of P < 0.2 were included in a multivariate linear regression model. Sex and group were included in the linear regression by converting the categories into a dichotomous variable. Statistical analysis was performed using commercial analytical software (SPSS, version 17; SPSS Inc., Chicago, IL). 

Results

A total of 100 subjects were enrolled in the study, 44 subjects in the FA Group and 56 subjects in the AC Group (25 with PAC and 31 PACG). Nine eyes in the FA group (20%) and 6 eyes from the AC group (10.7%) were excluded due to poor image quality, delineation errors by the ZAAP software, and inability to locate the scleral spurs accurately. 

The majority (95.8%) of the subjects were Chinese (Table 1). FA group subjects were younger, had shallower ACD, shorter AL, and greater lens vault compared with parameters of the AC group. Baseline pupil diameters (light) were similar between the groups (P = 0.1) and the mean change in pupil diameter from light to dark was similar (P = 0.23). 

Table 1. 

 

Demographics and Mean Values of Biometric Parameters

Parameter	FA Group (n = 35)	AC Group (n = 50)	P Value
Age, y	62.0 ± 8.2	65.4 ± 6.4	0.04
Sex, % F	62.9	68.0	0.62
ACD, mm	2.4 ± 0.2	2.6 ± 0.3	0.001
Axial length, mm	22.6 ± 0.8	23.3 ± 0.8	0.002
CCT, μm	546.9 ± 27.8	549.0 ± 34.0	0.12
Average K, mm	7.67 ± 0.25	7.68 ± 0.26	0.98
Pupil diameter, mm (light)	2.5 ± 0.6	2.3 ± 0.6	0.12
Pupil diameter, mm (dark)	3.7 ± 0.8	3.6 ± 0.8	1.0
Diameter change, mm (light to dark)	1.2 ± 0.6	1.3 ± 0.6	0.23
Anterior chamber width, mm	11.3 ± 0.9	11.6 ± 0.3	0.06
Iris curvature in dark, μm	0.195 ± 0.057	0.170 ± 0.076	0.9
Iris thickness (dark) at 750 μm	0.501 ± 0.091	0.476 ± 0.072	0.17
Lens vault, μm	1031.6 ± 207.7	779.9 ± 245.8	<0.001

 K, keratometry.

Mean iris volume and iris cross-sectional area in light were similar in both groups (Table 2). Iris volume did not change in FA eyes when going from light to dark (mean change [SD] in iris volume = +1.50 mm³ [6.73]; P = 0.19), whereas iris volume decreased in AC eyes (mean change [SD] = −1.52 mm³ [3.07]; P < 0.001). The change in volume response to light differed between the two groups (P = 0.01). Iris cross-sectional area decreased significantly in both groups between conditions of light and dark (Table 2, P < 0.05 for both) and was similar in magnitude. After adjusting for age, sex, baseline pupil diameter, and change in pupil diameter, change in iris volume was significantly associated with age (β = −0.397; P < 0.001) and diagnostic category (AC versus FA group: β = 0.347; P < 0.001, Table 3). 

Table 2. 

 

Mean Iris Volume and Cross-Sectional Area

Mean	FA Group	AC Group	P Value
Iris volume in light, mm3	40.82 ± 5.83	40.90 ± 5.65	1.0
Iris volume in dark, mm3	42.33 ± 7.97	39.37 ± 6.23	0.15
Change in volume, mm3	↑1.50 ± 6.73	↓1.52 ± 3.07	0.01
Iris cross-sectional area in light, mm2	3.85 ± 0.46	3.99 ± 0.51	0.7
Iris cross-sectional area in dark, mm2	3.41 ± 0.80	3.43 ± 0.52	1.0
Change in cross-sectional area, mm2	0.44 ± 0.82	0.55 ± 0.26	1.0

 Arrow up = increase, arrow down = decrease.

Table 3. 

 

Factors Associated with Change in Iris Volume between Light and Dark

Variable	Univariate		Multivariate
	β		β
Age	−0.217	0.01	–0.351	0.001
Sex	0.078	0.39	0.181	0.07
ACD	0.01	0.89	—	—
Axial length	−0.03	0.70	—	—
CCT	0.093	0.39	—	—
Average K	−0.030	0.78	—	—
Anterior chamber width	−0.05	0.51	—	—
Lens vault	0.04	0.63	—	—
Pupil diameter (light)	−0.21	0.01	–0.154	0.14
Change in pupil diameter	0.138	0.13	0.151	0.15
Iris curvature in dark	−0.133	0.23	—	—
Iris thickness (dark) at 750 μm	−0.198	0.07	–0.127	0.23
AC group (vs. FA group)	0.294	0.006	0.319	0.003

We also evaluated the proportion of eyes that demonstrated a variable response to change in iris volume within the subgroups and noted that only 28% of eyes in the AC group demonstrated an increase in iris volume compared with 51.4% in the FA group and this difference was statistically significant (P = 0.028). Further, on linear regression analysis, factors associated with iris volume increase were age (P = 0.021), sex (P = 0.007), pupil diameter in light (P = 0.031), and lens vault (P = 0.027) in the FA group, whereas the average iris thickness (P = 0.01) was the only significant factor in the AC group. Iris volume decrease was associated with ACD (P = 0.017), pupil diameter in the dark (P = 0.023), and average iris curvature (P = 0.013) in the FA group and in the AC group, ACD (P = 0.001) and pupil diameter in light were significant factors (P = 0.002). 

Discussion

We found that iris volume changes in response to illumination differed between fellow eyes of subjects with APAC and those with chronic PAC/PACG. The irides of eyes in the AC group lost more volume in response to a physiologic increase in pupil size compared with eyes in the FA group, in which there was no significant change in residual volume of iris in the FA group. Age also had an important negative effect on iris volume change and the combined effects of these factors on iris volume change could be important triggers of acute angle closure in eyes that are predisposed. 

In contrast to our findings where the increase in iris volume in fellow eyes of subjects with APAC was not statistically significant, Aptel et al. ^17,18 found a significant increase in iris volume in their group of subjects with the same profile. They demonstrated this specific response of iris volume in this group both under conditions of pharmacologic and physiologic dilatation. The difference in response between the studies could be related to measurement variations using different software, true differences in response to illumination in Chinese subjects compared with White subjects, or limited power in the current study to detect a true difference. It would also be important to acknowledge that the variations in iris volume in eyes of the FA group are notably larger than eyes in the AC group (6.73 mm³ vs. 3.07 mm³). These variations could represent an inherent difference between these subgroups and may be one of the trigger factors for acute angle closure. The final common denominator established from these studies is that the residual iris volume in the mydriatic state and the fluctuation of iris volume is higher in fellow eyes of subjects with APAC and these factors could be important triggers for initiating an acute event. 

The response of iris cross-sectional area decreasing as pupil diameter increases has been reported in other racial/ethnic groups. ^16,17,23 A consistent finding from these studies is that eyes with angle closure lose less iris area in response to mydriasis compared with eyes that have open angles. Our study findings suggest that fellow eyes of subjects with acute angle closure had a lower loss in iris cross-sectional area compared with the eyes of subjects with chronic angle closure. It is possible that the loss in iris cross-sectional area with mydriasis follows a trend of least area loss noted among subjects with APAC, followed by subjects with PAC/PACG and then eyes with open angles. These responses combined with age and ocular biometric characteristics may determine the incidence of acute versus that of chronic disease in the subset of eyes at risk for angle closure. 

The change in iris volume with pupil dilation may result from an exchange of aqueous humor between the iris extracellular space and the anterior chamber. ²⁴ Some eyes may retain more volume during physiologic or pharmacologic pupil dilation. Such eyes may be more likely to develop angle closure if other anatomic characteristics are present. The presence of an altered fluid conductivity of the iris in angle closure eyes may also have a biological basis. Aquaporins are water-selective molecular channels that play an integral role in fluid conductivity across plasma membranes and are present abundantly in ciliary and iris epithelium. ²⁵ The level of aquaporin expression may determine fluid conductivity in a particular tissue and it would be interesting to explore if the degree of aquaporin expression or other components of iris cellular or molecular architecture might differ between acute and chronic variants of angle closure. Eyes with PACG have higher expression of secreted protein, acidic character, and are rich in cysteine (SPARC). ²⁶ SPARC promotes the deposition of fibrotic collagen–I, which may affect iris stromal behavior in angle closure eyes. 

Chinese people have an incidence of APAC twice that of other Asian races (Thai, Malay, Indian, or Indonesian). ^27–29 This could be due to the variation in anatomic or biometric differences between different races. However, some studies have found that biometric parameters among Chinese subjects were not vastly different from those of persons of African or European descent ³⁰ and simple anatomic parameters do not fully explain the dramatically higher prevalence of APAC among the Chinese. ¹⁵ We speculate that the varied light–dark response of the iris as seen in our study (our sample comprised predominantly Chinese subjects [95.8%]), may also have inherent interracial differences and may be a factor contributing to a varying incidence of angle closure disease. 

The strengths of this study were that our results were obtained after adjusting for biometric variables that could influence the light–dark changes in iris volume and we demonstrated the changes under physiologic conditions, without pharmacologic mydriasis. Furthermore, none of the eyes had incisional surgery nor were on any systemic therapy that could potentially affect iris dynamics and thereby confound results. The limitations of our study were that 15% of the eyes had to be excluded from the analysis due to poor image quality and inability to locate the scleral spur. The findings for the FA group may not be generalizable. In our study population, 68.5% of the eyes studied in the FA group were primary angle closure suspects, 25.7% had PAC, and the remaining were PACG (5.7%). These proportions might vary in other populations. The imaging cross-sections included only six sectors of iris and may not be representative of the entire iris and we have to be cautious about extrapolating this as the response of the iris in toto. In addition, 38% of the study subjects were on IOP-reducing medications, including prostaglandins that are known to be associated with tissue remodeling ³¹ and this could have an influence on iris dynamics. However, evidence from a prior study suggests this possibility to be remote. ¹⁶ All eyes in our study had a prior iridotomy and some of the scans may have included the iridotomy site and this could cause some bias in volume estimation. Furthermore, the potential damage to iris stroma induced by laser iridotomy can theoretically alter the blood–aqueous barrier characteristics and also influence its fluid conductivity. Finally, the presence of optic nerve damage in eyes with PACG could alter iris dynamics and the effect of this needs to be further evaluated. 

In summary, iris volume change with pupil dilation differs between fellow eyes of subjects with APAC compared with those with chronic asymptomatic angle closure. Such variations in iris volume responses may influence the subtype of angle closure that develops in subjects at risk. 

Footnotes

 Supported in part by grants from the National Medical Research Council, Singapore and the National Research Foundation, Singapore.

Footnotes

 Disclosure: A. Narayanaswamy, None; C. Zheng, None; S.A. Perera, None; H.M. Htoon, None; D.S. Friedman, None; T.A. Tun, None; M. He, None; M. Baskaran, None; T. Aung, None