Evaluation of Lamina Cribrosa and Choroid in Nonglaucomatous Patients With Pseudoexfoliation Syndrome Using Spectral-Domain Optical Coherence Tomography

Abstract

Purpose: To evaluate the lamina cribrosa (LC) and peripapillary choroid in patients with pseudoexfoliation syndrome (PXS).

Methods: In this cross-sectional study, one eye each of 32 nonglaucomatous PXS cases and 29 healthy volunteers were enrolled. The optic discs were scanned using enhanced depth imaging spectral-domain optical coherence tomography, and measurements were obtained using HEYEX software 6.0. LC and other related variables at three areas (mid-superior, center, and mid-inferior) and peripapillary choroidal thickness were determined. Linear mixed modeling was used to adjust the variables.

Results: After adjustment for age, sex, and axial length, there was no significant difference between the two groups in peripapillary choroidal thickness or in retinal nerve fiber layer thickness. The LC was significantly thinner in all three areas in the PXS group when compared with the control group, even after adjustment. Although no significant difference in central laminar depth was observed between the two groups (P = 0.74), the superior and inferior laminar depth were significantly deeper in the PXS group when compared with the control group (P = 0.04 and P = 0.006, respectively). Although there was a significant negative association between age and central choroidal thickness in the control group (β = −2.820, P = 0.02), this correlation was not significant in the PXS group.

Conclusions: We found that LC is significantly thinner in all three areas of the optic nerve head in nonglaucomatous PXS patients than in controls. Although no significant difference in peripapillary choroidal thickness was observed between the two groups, peripheral posterior displacement of LC in nonglaucomatous PXS eyes was noted.

Pseudoexfoliation syndrome (PXS) is an age-dependent generalized disorder characterized by abnormal elastosis and excessive accumulation of microfibrils throughout the body, including ocular tissues.¹ It has been reported that glaucomatous damage can occur in a significant proportion of normotensive patients with PXS.² Also, at a given level of intraocular pressure (IOP), the probability of having glaucomatous damage was shown to be higher in eyes with PXS than in those without.³ In the Early Manifest Glaucoma Trial, the presence of PXS was recognized as the most important independent risk factor for glaucoma progression.⁴ In another study, the glaucoma conversion rate was twice as high in patients with ocular hypertension and PXS as in control patients matched for IOP, age, and sex.⁵ In addition to these associations, pseudoexfoliation glaucoma (PXG) in general has a higher level of IOP and more severe fluctuation in IOP when compared with primary open-angle glaucoma (POAG), highlighting this type of glaucoma as having a faster progression and poorer prognosis than POAG.^6,7 Overall, it seems that factors other than IOP level may contribute to the development and progression of glaucoma damage in PXS patients. 

The lamina cribrosa (LC) of the optic nerve head (ONH) has been considered the principal site of retinal ganglion cell axonal injury in glaucomatous damage.⁸ Hence, one possible explanation for this kind of susceptibility may be related to structural alterations in this component of the ONH complex. It has been shown that in PXS, a primary disturbance of the elastic fiber homeostasis in this structure may lead to decreased stiffness and increased deformability of the ONH, rendering PEX eyes more vulnerable to pressure-induced optic nerve damage and glaucoma development and progression.^9–11 

The nutrition of the choroid could be another possible source of vulnerability of PXS eyes to develop glaucoma. As an analogy, some histologic studies have demonstrated decreased density of capillaries and large choroidal vessels as well as choroidal thinning in advanced primary angle closure glaucoma.^12,13 However, postmortem studies suffer significantly from a lack of repeatability and reproducibility because any in vitro processing method can potentially change the shape and thickness of tissues. Therefore, in vivo imaging modalities have gained popularity in addressing the exact pathophysiologic events underlying numerous ocular diseases. 

Spectral-domain optical coherence tomography (SD-OCT) can reliably capture high-resolution cross-sectional images of the posterior segment of the eye with in vivo status, providing qualitative and quantitative assessment and monitoring of glaucoma and other ocular diseases.¹⁴ However, the sensitivity of conventional OCT is reduced as the signal attempts to penetrate deeper structures including the LC and choroid because of light scattering.¹⁵ Enhanced depth imaging (EDI) is a newly emerging technique to overcome this drawback by placing the reference plane deeper and thereby enhancing the images from the deeper layers of the ONH and choroid.^16,17 

In this study, our goal was to evaluate the LC as well as the peripapillary choroid in nonglaucomatous pseudoexfoliative patients using EDI function on the SD-OCT. 

Methods

Patients

In this cross-sectional study at Farabi Eye Hospital in Tehran, Iran, 38 consecutive patients with PXS and 37 normal volunteers who were examined for healthy eye and/or refractive check-ups were enrolled. In cases where both eyes of the patient met the eligibility criteria for the study, only one eye was randomly chosen for inclusion. All participants underwent a complete ophthalmic examination, including measurement of best-corrected visual acuity (Early Treatment Diabetic Retinopathy Study chart), slit-lamp biomicroscopy, Goldmann applanation tonometry, gonioscopy, dilated stereoscopic fundus examination using a 90 or 78 diopter (D) lens, measurement of the central corneal thickness by pachymeter (Tomey Corporation, Nagoya, Japan), a visual field test (Humphrey Field Analyzer II 750; 24-2 Swedish interactive threshold algorithm; Carl Zeiss Meditec, Dublin, CA, USA), axial length measurement (IOLMaster; Carl Zeiss Meditec), and SD-OCT imaging of the ONH and macula (Spectralis OCT, Heidelberg Engineering, Inc., Dossenheim, Germany). The inclusion criteria were (1) best-corrected visual acuity of 20/40 or better with a spherical equivalent within 5 diopters (D) and a cylinder correction within 3 D; (2) IOP < 22 mm Hg without the use of glaucoma medications; and (3) reliable Humphrey Field Analyzer results with a false-positive error rate of less than 15%, a false-negative error rate of less than 20%, and a fixation loss rate of less than 20%. The exclusion criteria included (1) any other intraocular diseases or neurologic diseases that could cause visual field loss, (2) any history of previous ocular surgery, (3) a diagnosis of glaucoma in the fellow eye, (4) significant media opacity, or (5) a history of diabetes mellitus. 

The patients were enrolled into the PXS group if they possessed (1) visible pseudoexfoliation material on the anterior lens capsule or pupillary margin after mydriasis on slit-lamp, (2) having an IOP < 22 mm Hg, (3) no history of increased IOP, (4) an absence of glaucomatous disc appearance (an intact neuroretinal rim without cupping, notches, or localized pallor), and (5) a normal visual field defined by mean deviation (MD) and pattern standard deviation (PSD) within 95% confidence interval limits and a Glaucoma Hemifield Test within normal limits. The normal control group was defined as those having an IOP < 22 mm Hg, no history of increased IOP, an absence of glaucomatous disc appearance, normal standard automated perimetry results, and no evidence of pseudoexfoliation material on the anterior lens capsule or pupillary margin after mydriasis on slit-lamp in both eyes. 

The study was implemented in accordance with the tenets of the Declaration of Helsinki. The study protocol was approved by the local ethics review committee of Tehran University of Medical Sciences, and all participants provided written informed consent prior to inclusion. 

Spectral-Domain Optical Coherence Tomography

All OCT measurements were performed using SD-OCT (Heidelberg Spectralis SD-OCT; Spectralis software version 5.3.2; Heidelberg Engineering, Inc., Dossenheim, Germany) after pupillary dilation. A circular scan pattern, 3.4 mm in diameter, was used for peripapillary choroidal thickness and peripapillary retinal nerve fiber layer (RNFL) measurements. Scans with a quality score of less than 20 were excluded from the analysis. We also excluded scans with inadequate quality as determined by unclear fundus images, interruption of the RNFL, or unclear border of the LC or posterior choroid (Fig. 1). 

Figure 1

 

Upper Left: An enhanced depth imaging with SD-OCT (EDI-OCT) of a normal optic nerve shows Bruch's membrane opening (BMO), the anterior and posterior borders of the LC (arrows), and the anterior laminar depth (ALD) in a raster line across the center of the optic nerve. Upper Right: An EDI-OCT image of a pseudoexfoliative eye demonstrating posterior displacement of the LC in the upper part of the optive nerve with high ALD. Lower Left: A peripapillary OCT image shows the manual segmentation of Bruch's membrane (upper red line), the sclerochoroidal border (lower blue line), and the global thickness of the peripapillary choroid as well as six sectors of the Heidelberg layout.

Enhanced Depth Imaging SD-OCT of Peripapillary Choroid

For peripapillary choroidal measurements, two specialists who were blind to the clinical data of the examined eyes (SM, MKJ) manually delineated the lower and upper segmentation lines of the circular scan. The lines were set so that they corresponded to the sclerochoroidal interface and posterior edge of the retinal pigment epithelium to signify the outer and inner boundaries of the choroid, respectively. The choroidal thickness in the corresponding sectors was calculated automatically using the RNFL thickness sectors algorithm (Fig. 1). 

Enhanced Depth Imaging SD-OCT of the Optic Nerve Head

The method for enhanced depth imaging with SD-OCT (EDI-OCT) of the ONH was previously described.¹⁵ Briefly, the SD-OCT device was set to image a 15 × 10° rectangle centered on the optic disc. This rectangle was divided into approximately 65 sections, each of which had on average 42 OCT frames. From these horizontal B scans, three frames (center, mid-superior, mid-inferior) that passed through the ONH were selected, and parameters were measured in each of these frames and labeled Cen, Sup, or Inf, respectively. Figure 1 depicts the LC borders and Bruch's membrane opening (BMO) in an ONH OCT image. The line that connects both ends of Bruch's membrane was defined as BMO. All of the distances were measured on the line perpendicular to the reference line. Parameters were measured as close as possible to the vertical center of the ONH. When a vessel trunk made the measurement impossible, the measurements were taken at the temporal side of it. The anterior and posterior borders of the highly reflective region at the vertical center of the ONH in the horizontal SD-OCT cross-section were defined as the borders of the LC, and the distance between these two borders was defined as LC thickness. Anterior laminar depth (ALD) and posterior laminar depth (PLD) were defined as the distance between BMO and the anterior border and posterior border of the LC, respectively. The prelaminar thickness was defined as the distance between the anterior surface of the optic cup and the anterior border of the LC, and the distance between the BMO and the internal limiting membrane (surface of the optic cup) was labeled the prelaminar depth. All of the measurements were obtained using HEYEX software 6.0 (Heidelberg Engineering, Inc.). All of the images were analyzed by two specialists (SM, PA) in two different sessions. 

To evaluate the intraobserver and interobserver reproducibility of ONH and choroidal thickness measurements, 15 randomly selected EDI-OCT B-scans from 15 eyes were evaluated. The analysis was based on two independent series of reevaluations made by two independent observers. The absolute agreement of a single observer's measurements and the mean of the measurements conducted by the two observers were calculated with the intraclass correlation coefficient from a two-way mixed-effect model (Table 1). 

Table 1

 

Interclass Coefficient Correlation (ICC) for Peripapillary Choroid and Optic Nerve Head Measurements Shows Good Reproducibility for Peripapillary Choroid and Optic Nerve Head Measurements

 

Statistical Analysis

Data were analyzed using SPSS software (version 18 for Windows; SPSS, Inc., Chicago, IL, USA). Continuous variables are expressed as mean (standard deviation). Comparisons between the two groups were performed using the Student's independent samples t-test or Mann-Whitney U test for parametric and nonparametric continuous variables, respectively. Categorical variables were compared using the chi-square test. Linear mixed modeling was used to adjust the effects of age, sex, axial length, IOP, and confounders on the choroidal thickness. The model was built with choroidal thickness/RNFL as the outcome variable and the presence of PXS as the main predictor variable in the unadjusted analysis, adjusting for the effects of age, sex, IOP, and axial length. Univariate analysis was performed to assess any association between peripapillary choroidal thickness and the ONH parameters in each group. We also determined the factors associated with central LC thickness and central prelaminar thickness using the linear model. Multivariate regression was performed to determine the factors associated with choroidal thickness, LC thickness, and central prelaminar thickness, including age, sex, and those variables with P < 0.10 in univariate analysis. P ≤ 0.05 was considered statistically significant. 

Results

Of 38 and 37 participants who were PXS and controls who met the initial inclusion criteria, 6 (16%) and 8 (21%) eyes, respectively, were excluded because of poor image quality. Thus, 32 PXS and 29 control eyes were included in the final analysis. None of the patients used glaucoma medications. A total of 26 eyes had bilateral PXS, but none of the cases or their fellow eyes had PXG. There was no significant difference between the two groups regarding age, sex, mean central corneal thickness (CCT), or axial length (Table 2). 

Table 2

 

Baseline Characteristics of the Pseudoexfoliation and Control Groups

 

Intraobserver and interobserver reproducibility of choroidal and ONH measurements ranged from 0.827 to 0.998 for the various parameters (Table 1). There was no significant difference between the two groups regarding RNFL thickness (Table 3). There was also no significant difference in peripapillary choroidal thickness between the two groups before and after adjustment for confounders (Table 3). 

Table 3

 

Comparison of RNFL Thickness Between the Pseudoexfoliation and Control Groups With and Without Adjustments for Age, Sex, Intraocular Pressure, and Axial Length

 

Comparing the LC of the PXS group with the control group, there was no significant difference in the BMO, central prelaminar depth, or central prelaminar thickness. The LC was significantly thinner in all three areas in the PXS group when compared with the control group. After adjusting for age, sex, and axial length, this significance still remained. Although we did not find any significant difference in central anterior laminar depth (_CenALD) between the two groups (P = 0.74), the superior and inferior anterior laminar depth were significantly deeper in the PXS group (380.0 [81.51] and 350.82 [99.73] μm, respectively) when compared with the control group (324.05 [87.68] and 280.33 [92.38] μm, respectively; P = 0.04 and P = 0.006, respectively) (Table 4). 

Table 4

 

Comparison of Optic Nerve Head and Lamina Cribrosa Parameters Between Pseudoexfoliation and Control Groups With and Without Adjustments for Age, Sex, Intraocular Pressure, and Axial Length

 

Tables 5 and 6 show the factors associated with choroidal thickness, LC thickness, and central prelaminar thickness in the two groups. Although the correlation between age and peripapillary choroidal thickness in the control group was significant (β = −2.820, P = 0.02), there was no significant association between age and choroidal thickness in the PXS group (β = −0.089, P = 0.98) (Fig. 2). There was a significant positive association between peripapillary choroidal thickness and _CenALD in both the control and PXS groups before (β = 0.226, P = 0.02; β = 0.561, P = 0.03, respectively) and after (β = 0.193, P = 0.02; β = 0.205, P = 0.01, respectively) adjustments. There were no significant determinants of LC thickness and prelaminar thickness in either group. 

Table 5

 

Association Between Different Variables and Global Peripapillary Choroidal Thickness in Pseudoexfoliation and Control Groups

 

Table 6

 

Association Between Different Variables and Central Laminar Thickness and Central Prelaminar Thickness

 

Figure 2

 

Left: Scatterplot showing the association of age and peripapillary choroidal thickness (in the control group). There was a negative correlation between age and global peripapillary choroidal thickness (β = −2.820, P = 0.02) in the control group. Right: Scatterplot showing the correlation of age and peripapillary choroidal thickness (right) in the pseudoexfoliation group. No correlation was found between age and global peripapillary choroidal thickness (β = −0.089, P = 0.98) in the pseudoexfoliation group.

Discussion

The evaluation of potential risk factors of glaucomatous damage other than IOP would be of clinical utility in cases of PXG because high IOP fluctuation itself may not completely explain the poorer prognosis of this type of glaucoma when compared with POAG.⁴ In the present study, we found that LC is significantly thinner in all three areas of the ONH in PXS patients than in controls. However, the peripapillary choroid was not thinner in PXS eyes when compared with controls after adjustments. 

The thinning of the LC in various stages of glaucoma has been documented in previous reports.^18,19 In their study comparing the LC thickness of glaucomatous patients with controls using EDI, Park et al.¹⁷ observed that LC thickness was thinner in the normal tension glaucoma (NTG) group than in the POAG and control groups. They concluded that the thinner lamina in NTG patients may result in a greater risk of deformation of the lamina, thus leading to a block of the axonal transport. 

In a recent study, Kim et al.²⁰ compared the LC thickness of PXG and POAG patients. They observed that despite similar mean IOP in the two groups, the LC was significantly thinner in the PXG group when compared with the POAG group. When comparing the LC thickness in nine cases of unilateral PXG with their fellow eyes, no significant difference was found in the LC thickness. Consistent with this finding, we observed that the LC is thin in PXS eyes even before the development of glaucoma. Thinning of the LC, besides being a consequence of glaucomatous damage, could also have prognostic and mechanistic significance. It is possible that thinner laminae may be more vulnerable to glaucomatous progression. 

Several studies described generalized ultrastructural alterations in ocular tissues and other organs in patients with PXS.^21,22 Schlötzer-Schrehardt et al.⁹ found a significant downregulation of lysyl oxidase-like 1 and elastic fiber in LC tissues obtained from early and late stages of pseudoexfoliation syndrome without and with glaucoma when compared with normal and POAG specimens.⁹ These alterations may lead to reduced elasticity of the LC¹¹ and may place PXS patients at different stages of the disease at an increased risk of glaucomatous damage because of weakening and thinning of the LC. 

In our study, in line with the study by Kim et al.20, we observed no significant difference in central ALD between the two groups. However, we found a significant increase in ALD of the superior and inferior regions of LC in PXS patients when compared with controls. Some recent studies have reported posterior displacement of the peripheral LC in POAG eyes.^19,23–25 Lee et al.^26,27 found significant associations between the peripheral outward deformation of the LC and the presence of a disc hemorrhage with a faster rate of RNFL thinning in POAG patients. In our study, we found peripheral posterior displacement of the LC in nonglaucomatous PXS eyes. It has been postulated that a marked decrease in stiffness of the LC in PXS eyes would result in increased deformability of the ONH in these patients.¹¹ Peripheral deformation in the LC structure, either in PXS or in NTG patients, may lead to increased susceptibility of retinal ganglion cell (RGC) axons to pressure-induced damage, thus making it a potential risk factor of glaucoma development and progression. 

Some recent studies have focused on the role of the choroid in the pathogenesis of glaucoma.^28–31 The choroid may exert its effects via its nutritional and vascular support of the ONH. In a study by Hossieni et al.,³¹ no significant difference was found in peripapillary and macular choroidal thickness in POAG patients when compared with normal controls after adjusting for age and axial length. Although Park et al.²⁹ have not observed significant differences in peripapillary and macular choroidal thickness in patients with POAG when compared with controls, they have detected a significant decrease in peripapillary choroidal thickness in patients with NTG when compared with controls. In our study, we did not find any difference in macular choroidal thickness in the PXS group when compared with the control group after adjusting for age and axial length. In a recent study, Turan-Vural et al.³⁰ compared the subfoveal choroidal thickness of 35 PXS eyes with 26 healthy controls and found a significant decrease in the choroidal thickness in PXS eyes.³⁰ However, they did not adjust their analysis for confounding variables, including age and axial length, which might be a reason for this discrepancy. 

In agreement with previous studies, peripapillary choroidal thickness was negatively associated with age and a female sex in the control group.^28,32 However, we did not find a similar correlation in the PXS group. In both groups, choroidal thickness was related to anterior laminar depth. This finding might be explained by the higher position of the BMO in eyes with a thicker choroid.³³ 

In this study, we observed no association between LC thickness and CCT in either of the two groups. Some prior studies have confirmed our findings.^34–36 The different embryologic origins of these two tissues suggest that there would not be a connection between them. Some previous studies have detected an increased LC thickness with increasing age,^35,37 whereas others do not.^36,38 In the present study, we did not observe any change in LC thickness with aging. Laminar alteration is actually a dynamic-modulated remodeling process, that is, a biomechanical feedback mechanism through which laminar cells modify their environment in an attempt to return to a homeostatic mechanical environment. 

Our study has some limitations. A relatively small size is one of these limitations. Moreover, considerable variations exist in LC among patients, and thus the standard deviation of the LC thickness is large. Although we have a better visualization of the choroid and LC with the EDI technique than with conventional imaging, it does not provide satisfactory visualization of the LC in all eyes. Finally, measuring the anterior and posterior lamina depth from the BMO may provide a biased insight because choroidal thickness is included in the measurement.³³ We did find a positive association between choroidal thickness and anterior laminar depth. However, prelaminar and laminar thickness would not be affected by this bias. In addition, the peripapillary choroidal thickness was not different between the PXS and control groups. 

In conclusion, we demonstrated that there is a thinner LC in the central, mid-superior, and mid-inferior areas of PXS eyes when compared with normal control eyes. Our data showed that the anterior laminar depth is greater in mid-superior and mid-inferior portions of the optic nerve in PXS eyes. However, peripapillary choroidal thickness was not significantly different between PXS eyes and normal controls. 

Acknowledgments

The authors thank Nassim Khatibi for help with gathering data and Somaye Heidarzadeh and Sepideh Heydarzdeh for help with biometric and corneal thickness measurements. Shan C. Lin, MD, is a consultant for Allergan, Alimera, and Iridex. All other authors indicate no financial support. The protocol of the study was approved by the institutional review board of Farabi Eye Hospital, Tehran, Iran. Written informed consent was obtained for all the participants after complete explanation. Supported by a grant from Tehran University of Medical Sciences, Tehran, Iran (Project 19076). The authors alone are responsible for the content and writing of the paper. 

Disclosure: S. Moghimi, None; M. Mazloumi, None; M.K. Johari, None; P. Abdi, None; G. Fakhraie, None; M. Mohammadi, None; R. Zarei, None; Y. Eslami, None; M.A. Fard, None; S.C. Lin, Allergan (C), Alimera (C), Iridex (C)