October 2016
Volume 57, Issue 13
Open Access
Retina  |   October 2016
Macular Inner Retinal Layer Thickening and Outer Retinal Layer Damage Correlate With Visual Acuity During Remission in Behcet's Disease
Author Affiliations & Notes
  • Dan Cheng
    School of Ophthalmology and Optometry, Wenzhou Medical University, Wenzhou, Zhejiang, China
  • Yuqin Wang
    School of Ophthalmology and Optometry, Wenzhou Medical University, Wenzhou, Zhejiang, China
  • Shenghai Huang
    School of Ophthalmology and Optometry, Wenzhou Medical University, Wenzhou, Zhejiang, China
  • Qiuyan Wu
    School of Ophthalmology and Optometry, Wenzhou Medical University, Wenzhou, Zhejiang, China
  • Qi Chen
    School of Ophthalmology and Optometry, Wenzhou Medical University, Wenzhou, Zhejiang, China
  • Meixiao Shen
    School of Ophthalmology and Optometry, Wenzhou Medical University, Wenzhou, Zhejiang, China
  • Fan Lu
    School of Ophthalmology and Optometry, Wenzhou Medical University, Wenzhou, Zhejiang, China
  • Correspondence: Meixiao Shen, School of Ophthalmology and Optometry, Wenzhou Medical University, 270 Xueyuan Road, Wenzhou, Zhejiang, China, 325027; shenmxiao7@hotmail.com
  • Fan Lu, School of Ophthalmology and Optometry, Wenzhou Medical University, 270 Xueyuan Road, Wenzhou, Zhejiang, China, 325027; lufan62@mail.eye.ac.cn
  • Footnotes
     DC and YW contributed equally to the work presented here and should therefore be regarded as equivalent authors.
Investigative Ophthalmology & Visual Science October 2016, Vol.57, 5470-5478. doi:10.1167/iovs.16-19568
  • Views
  • PDF
  • Share
  • Tools
    • Alerts
      ×
      This feature is available to Subscribers Only
      Sign In or Create an Account ×
    • Get Citation

      Dan Cheng, Yuqin Wang, Shenghai Huang, Qiuyan Wu, Qi Chen, Meixiao Shen, Fan Lu; Macular Inner Retinal Layer Thickening and Outer Retinal Layer Damage Correlate With Visual Acuity During Remission in Behcet's Disease. Invest. Ophthalmol. Vis. Sci. 2016;57(13):5470-5478. doi: 10.1167/iovs.16-19568.

      Download citation file:


      © ARVO (1962-2015); The Authors (2016-present)

      ×
  • Supplements
Abstract

Purpose: To identify macular intraretinal layer changes of patients in remission from Behcet's disease (BD) of short and long duration and evaluate the associations with visual acuity (VA).

Methods: Thirty-two eyes from 26 BD patients were enrolled, including 16 eyes with a duration less than 3 years (0.5–2.5 years; BD1) and 16 eyes of longer duration (3–12 years; BD2). Their intraretinal layer thicknesses and integrity of ellipsoid zone (EZ) and interdigitation zone (IZ) were evaluated by spectral-domain optical coherence tomography (SD-OCT). Associations between VA and retinal structural changes were analyzed.

Results: Compared to controls, the inner retina was significantly thicker in BD groups, especially the nerve fiber layer (NFL). The outer retinal layer (ORL) was thicker in BD1 in the central and temporal regions and thinner in BD2 compared to controls in all regions. In BD2, there were more eyes with disruption of the EZ and IZ. Worsening VA was correlated with thickening of the NFL and inner nuclear layer (INL), thinning of the ORL, and greater disruption of the EZ and IZ. Multiple linear regression analysis revealed EZ disruption, nasal ORL, inferior NFL, and temporal and nasal INLs were independent predictors of best-corrected (BCVA).

Conclusions: Behcet's disease patients in remission had significant changes in the inner and outer retinal structures, associated with worse VA. Thickness and integrity of the intraretinal layers by SD-OCT and segmentation might be useful predictors for the degree of VA damage in BD remission.

Behcet's disease (BD) is a chronic, inflammatory systemic immune-mediated vasculitis affecting both arteries and veins in all organs.1 Ocular involvement presents in 60% to 80% of BD patients, mostly in the form of bilateral panuveitis, and may lead to blindness.2 Repeated ocular attacks and progressive retinal vasculitis result in permanent damage of the neurosensory retina, such as maculopathy and retinal atrophy.2,3 Due to the various clinical presentations of retinal complications, spectral-domain optical coherence tomography (SD-OCT) has become an important tool for routine examination, diagnosis, and treatment management of BD. 
With the improved imaging ability of retinal structure by OCT, several studies have used this tool to evaluate the effectiveness of treatment methods and disease-related retinal complications.46 Based on the previous studies and clinical outcomes, it has been noted that most affected eyes in BD patients lose useful vision within 5 to 10 years of disease onset despite successful treatment and complete resolution of retinal complications, such as macular edema.79 Moreover, the correlation between macular thickness and visual acuity (VA) in BD patients is an unresolved issue. A number of studies reported moderate to strong correlations between macular thickness and VA.6,10 Other investigators concluded that there is only a weak or no correlation between macular thickness and VA.11,12 These findings hint that VA maybe related to the changes or damage of the individual retinal layers. 
Recently, the use of high resolution SD-OCT and automatic segmentation techniques has allowed measurement of the thickness of individual retinal layers around the macula.1316 Several studies have revealed changes of intraretinal layer thicknesses in retinal diseases including diabetic retinopathy, retinitis pigmentosa, high myopia, and age-related macular degeneration.1719 Furthermore, the damage of intraretinal layers in these retinal pathologies has been correlated with poor VA or clinical outcome.17,19 
Limited information is available regarding the characterization of changes in intraretinal layers around the macula in BD during remission and the association of those changes with VA. Only two cross-sectional studies using OCT described the disruption of photoreceptor ellipsoid zone (EZ) in BD during remission phase.20,21 These studies reported that the disruption of photoreceptors is strongly related to the loss of VA. However, until now, no studies have quantified the changes of intraretinal thicknesses around the macula in BD that may also dramatically impact the VA. 
The purpose of this retrospective study of Behcet patients who were in remission was to characterize the specific structural changes with disease duration in the macular retina. Further, we evaluated the associations of the changes in intraretinal thicknesses with VA to explore which intraretinal layers would be good predictors of VA loss. 
Methods
Subjects
In this retrospective study, medical records of Behcet's uveitic patients who visited the uveitis clinic of the Eye Hospital, Wenzhou Medical University from December 2013 to July 2015 were reviewed. Patients were identified who fulfilled the diagnostic criteria of the International Study Group for Behcet's disease,22 experienced acute ocular inflammation, and were currently in a remission phase. We defined the remission phase as eyes at the current visit in an inactive status without any documented intraocular inflammation, such as retinitis, vasculitis, papillitis, macular edema, or retinal hemorrhage.1,20,23 Furthermore, patients with anterior uveitis only were excluded, but those with evidence of posterior segment involvement were included according to the Standardization of Uveitis Nomenclature (SUN) working group classification.23 Posterior segment involvement in Behcet's uveitis included cell infiltration of the vitreous body, retinal-, macular-, and optic disc edema, and retinal vasculitis. Exclusion criteria were eyes with spherical equivalent refractive error of more than ± 6.0 diopters, significant cataract or vitreous opacity, glaucoma, diabetic retinopathy, age-related macular degeneration, other marked complications, or a history of intraocular surgery including cataract, glaucoma, or vitrectomy within 1 year. 
We divided the BD patients into two groups according to disease duration from initial onset of uveitis.24,25 Patients in group BD1 had disease durations of less than 3 years and those in BD2 had disease durations of more than 3 years. 
We recruited normal subjects who were age-, sex-, and spherical equivalent-matched as the control group. These subjects had no ocular disease or systemic disease, such as diabetes or hypertension. The best-corrected VA (BCVA) of the control subjects was 20/20 or better, and the spherical equivalent refractive error was less than ± 6.0 diopters. 
This project was approved by the Ethics Committee of the Wenzhou Medical University and was performed in accordance with the tenets of the Declaration of Helsinki. Informed consent was obtained from the patients and subjects after explanation of the nature and possible consequences of the study. 
Data Collection
Each patient underwent a complete ocular examination, including slit-lamp examination, intraocular pressure by Goldman applanation tonometry, measurement of BCVA and spherical equivalent refractive error, funduscopy, fluorescein angiography (FA), and SD-OCT examination. Best-corrected VA was measured with a Snellen chart and was converted to the logarithm of the minimum angle of resolution (logMAR) units for statistical analysis. LogMAR values of 2.0 and 3.0 were assigned for counting fingers and hand motion vision, respectively.26 The spherical equivalent refractive error was measured by autorefractor (KR8900; Topcon, Tokyo, Japan). Demographic and clinical information included age, sex, duration of Behcet's uveitis, and the duration interval from the last ocular attack to the current remission phase. All patients were examined by the same uveitis specialist. 
Image Acquisition and Data Analysis
Optical coherence tomography examination was performed with the Spectralis OCT (Heidelberg Engineering, Heidelberg, Germany). One cross-sectional OCT image with a quality score ≥ 20 (defined by the machine) was selected along the horizontal and vertical meridians passing through the fovea, respectively. A single drop of Mydrin-P (tropicamide 5 mg/ml and phenylephrine 5 mg/mL; Santen Pharmaceuticals, Osaka, Japan) was administered to both eyes of each subject at least 30 minutes before OCT examination to ensure pupil dilation. 
Following data collection, the Spectralis OCT images were processed using custom-developed software.15,16 Based on these images, only five layers could be clearly visualized, especially in the BD group (Fig. 1). Therefore, to get precise thickness measurements for the intraretinal layers, the following five layers were segmented from each image (Figs. 1A, 1B): (1) the nerve fiber layer (NFL), (2) the ganglion cell layer and inner plexiform layer (GCL+IPL), (3) the inner nuclear layer (INL), (4) outer retinal layer (ORL), and (5) the retinal pigment epithelium (RPE)/Bruch complex.27 A good set (image quality score ≥20) of horizontal and vertical scans with a well-centered fovea for each subject was selected for analysis. Detection of the layer boundaries was achieved by automatic segmentation14,15 and checked by visual inspection performed by two of the authors (DC and QW). 
Figure 1
 
Segmentation of five intraretinal layers and macular regions analyzed by SD-OCT and automatic segmentation technology. (A) Cross-sectional image along the horizontal meridian of a Behcet uveitic eye. (B) Cross-sectional image along the vertical meridian of a Behcet uveitic eye. (C) The thickness profiles of five intraretinal layers were obtained within a 6-mm circular area around the macula that was divided into five regions for analysis. The central region included a 1-mm-diameter circle centered on the fovea. The temporal, superior, nasal, and inferior regions consisted of the horizontal or vertical scan lines extended from the intersection points to the outer diameter of 6 mm. Scale bar: 200 μm.
Figure 1
 
Segmentation of five intraretinal layers and macular regions analyzed by SD-OCT and automatic segmentation technology. (A) Cross-sectional image along the horizontal meridian of a Behcet uveitic eye. (B) Cross-sectional image along the vertical meridian of a Behcet uveitic eye. (C) The thickness profiles of five intraretinal layers were obtained within a 6-mm circular area around the macula that was divided into five regions for analysis. The central region included a 1-mm-diameter circle centered on the fovea. The temporal, superior, nasal, and inferior regions consisted of the horizontal or vertical scan lines extended from the intersection points to the outer diameter of 6 mm. Scale bar: 200 μm.
For each eye, a circle centered on the fovea with a radius of 0.5 mm was made, and the intersecting points of the circle with the horizontal and vertical scan lines were obtained (Fig. 1). The central thickness for each retinal layer was determined by the average along both the horizontal and vertical scans within the innermost circle. The thickness profiles along the horizontal and vertical scans from the intersecting points to the outer diameter of 6 mm were averaged to obtain the thicknesses in the temporal, nasal, inferior, and superior regions, respectively. 
Using these Spectralis images, the integrity of the EZ and interdigitation zone (IZ) was evaluated in each image for 500 μm in either direction of the fovea along the horizontal and vertical scans. The line disruption was graded from 0 to 2. Grade 0 was given when the line was intact (Fig. 2A). Grade 1 was assigned when there was a focal disruption of the line of 200 μm or less (Fig. 2B), and Grade 2 was assigned when the disruption was greater than 200 μm (Fig. 2C). Grades from each patient's horizontal and vertical scans were added to yield a global disruption scale.28 A longitudinal reflectivity profile (LRP) was also generated to show the structure of the outer retinal bands.29,30 Briefly, OCT reflectivity as a function of retinal depth was measured for approximately 40 A-scans over 500 μm in both directions of the central foveal horizontal and vertical meridians. The mean LRP was constructed by averaging all of the LRPs within the selected region. 
Figure 2
 
Representative examples of grading EZ and IZ band disruptions on Heidelberg Spectralis SD-OCT images through the fovea. The ellipsoid zone and IZ bands were evaluated over a span of 500 μm in either direction of the foveal horizontal and vertical meridians. The line disruption was graded from 0 to 2. (A) Grade 0 bands had no disruptions. (B) Grade 1 bands had focal disruptions ≤200 μm. (C) Grade 2 bands had focal disruptions >200 μm.
Figure 2
 
Representative examples of grading EZ and IZ band disruptions on Heidelberg Spectralis SD-OCT images through the fovea. The ellipsoid zone and IZ bands were evaluated over a span of 500 μm in either direction of the foveal horizontal and vertical meridians. The line disruption was graded from 0 to 2. (A) Grade 0 bands had no disruptions. (B) Grade 1 bands had focal disruptions ≤200 μm. (C) Grade 2 bands had focal disruptions >200 μm.
Statistical Analysis
Refraction data were converted to spherical equivalents, which were calculated as the spherical dioptric power plus one-half of the cylindrical dioptric power. The Kolmogorov-Smirnov test was used to verify distribution normality for continuous variables. To determine statistically significant differences between groups, we used independent t-tests for normally distributed continuous variables. The Mann-Whitney U test was used for nonnormally distributed continuous variables, and χ2 or Fisher's exact test was used for categorical variables. Univariate analysis of variance (ANOVA) was performed to test differences among the three groups in the global or regional thickness in specific intraretinal layers. Association of intraretinal thickness and BCVA was determined with the Pearson correlation test. The Spearman correlation test was used to evaluate associations between the disruption of EZ or IZ bands and BCVA. Multivariate linear regression analysis was further used to identify the associations between BCVA and independent variables including intraretinal layer thicknesses and the disruptions of EZ and IZ bands. Data were expressed as the means ± the standard deviations and were analyzed with SPSS software (version 22.0; SPSS, Inc., Chicago, IL, USA). P values < .05 indicated statistical significance. 
Results
Demographics
A total of 32 eyes from 26 Behcet's posterior uveitic patients were included in this study, including 16 eyes in BD1 and 16 eyes in BD2 (disease duration: 1.29 ± 0.63 years for BD1 and 6.60 ± 2.63 years for BD2). Sixteen eyes from 16 normal subjects were also included as the control group. There were no significant differences in age, sex, or spherical equivalents between controls and study groups (Table 1). There was no significant difference in interval between the last active and current remission phase for the two study groups. Compared with the BD1 group, the BD2 group had a longer average disease duration (1.29 ± 0.63 years versus 6.60 ± 2.63 years, P < 0.001) and worse VA (0.83 ± 0.86 vs. 0.22 ± 0.23, P = 0.008). 
Table 1
 
Subject Characteristics of Normal and BD Groups
Table 1
 
Subject Characteristics of Normal and BD Groups
Intergroup Differences: Retinal Structure Analysis
The intraretinal layers along the horizontal and vertical meridians for the three groups showed that each group exhibited different profile patterns for both the inner and ORLs (Fig. 3). Behcet's disease caused substantial alterations in the thicknesses of inner and ORLs (Figs. 4A, 4B, 5). The macular NFL thickness was significantly larger in both BD groups compared with controls in all regions (Fig. 5A). Global NFL and mean NFL at each location were thicker in BD2 than BD1 (P < 0.05). In BD2, the greatest percentage (92.7%) of increase in NFL thickness occurred in the nasal regions, compared to that in any of the other regions. The GCL+IPL was thicker in BD1 compared with controls in the inferior region (Fig. 5B, P = 0.01) and temporal region (Fig. 5B, P = 0.026) regions. Patients in group BD2 also tended to have thicker GCL+IPLs than controls, but this was significant only in the temporal region (Fig. 5B, P = 0.012). 
Figure 3
 
Thickness profiles of five intraretinal layers in control, BD1, and BD2 groups. Thickness profiles of five intraretinal layers along horizontal (top row) and vertical (bottom row) scans were measured from SD-OCT images and averaged over each group.
Figure 3
 
Thickness profiles of five intraretinal layers in control, BD1, and BD2 groups. Thickness profiles of five intraretinal layers along horizontal (top row) and vertical (bottom row) scans were measured from SD-OCT images and averaged over each group.
Figure 4
 
Representative Heidelberg Spectralis SD-OCT images of eyes for a normal subject and a BD1 and a BD2 patient. Horizontal scan (left). Vertical scan (right). (A, B) Spectral-domain optical coherence tomography images demonstrating the intraretinal layer thicknesses in a normal patient with a VA of 20/20 (top), a BD1 patient with a VA of 20/25 (middle), and a BD2 patient with a VA of 20/1000 (bottom). (C, D) The structure of the outer retinal bands around the central fovea. The longitudinal reflectivity profile (LRP) as a function of retinal depth was calculated over a span of 500 μm of the central fovea in either direction of the horizontal and vertical scans and overlaid on the image. The arrowheads correspond to the EZ and IZ bands. In the left panel, the EZ/IZ bands were intact along both the horizontal and vertical scans with an EZ disruption scale = 0 and IZ disruption scale = 0. The longitudinal reflectivity profile also showed that three peaks corresponding to the EZ, IZ, and RPE were visible. In the middle panel, the EZ and IZ bands were either partially or totally disrupted along the horizontal and vertical scans with an EZ disruption scale = 1 and IZ disruption scale = 3. The peak of the IZ was not visible in the LRP (D, middle panel) due to the total loss of the IZ band along the vertical scan (IZ disruption grade = 2). For this patient, there was a decrease in reflectivity of the EZ peak due to the partial disruption of this band. There was a total loss of EZ and IZ bands (right panel, arrows) with an EZ scale = 4 and IZ scale = 4. The peaks of the EZ and IZ bands were not visible in the LRP along either the horizontal or vertical scans. Scale bar: = 100 μm.
Figure 4
 
Representative Heidelberg Spectralis SD-OCT images of eyes for a normal subject and a BD1 and a BD2 patient. Horizontal scan (left). Vertical scan (right). (A, B) Spectral-domain optical coherence tomography images demonstrating the intraretinal layer thicknesses in a normal patient with a VA of 20/20 (top), a BD1 patient with a VA of 20/25 (middle), and a BD2 patient with a VA of 20/1000 (bottom). (C, D) The structure of the outer retinal bands around the central fovea. The longitudinal reflectivity profile (LRP) as a function of retinal depth was calculated over a span of 500 μm of the central fovea in either direction of the horizontal and vertical scans and overlaid on the image. The arrowheads correspond to the EZ and IZ bands. In the left panel, the EZ/IZ bands were intact along both the horizontal and vertical scans with an EZ disruption scale = 0 and IZ disruption scale = 0. The longitudinal reflectivity profile also showed that three peaks corresponding to the EZ, IZ, and RPE were visible. In the middle panel, the EZ and IZ bands were either partially or totally disrupted along the horizontal and vertical scans with an EZ disruption scale = 1 and IZ disruption scale = 3. The peak of the IZ was not visible in the LRP (D, middle panel) due to the total loss of the IZ band along the vertical scan (IZ disruption grade = 2). For this patient, there was a decrease in reflectivity of the EZ peak due to the partial disruption of this band. There was a total loss of EZ and IZ bands (right panel, arrows) with an EZ scale = 4 and IZ scale = 4. The peaks of the EZ and IZ bands were not visible in the LRP along either the horizontal or vertical scans. Scale bar: = 100 μm.
Figure 5
 
Five intraretinal layer thicknesses in the macular region determined by SD-OCT and automated segmentation technology. (A) NFL; (B) GCL+IPL; (C) INL; (D) ORL; (E) RPE. *P < 0.05, **P ≤ 0.001.
Figure 5
 
Five intraretinal layer thicknesses in the macular region determined by SD-OCT and automated segmentation technology. (A) NFL; (B) GCL+IPL; (C) INL; (D) ORL; (E) RPE. *P < 0.05, **P ≤ 0.001.
The inner nuclear layer was thicker in both BD groups compared to controls in all meridional regions except the nasal region (Fig. 5C, P < 0.05). The global INL was also increased in both BD groups compared to controls (P = 0.013, P = 0.006, respectively). Group BD2 tended to have a thicker INL compared to BD1, but this was significant only in the temporal region (Fig. 5C, P = 0.023). 
The outer retinal layer tended to be thicker in BD1 patients with shorter disease duration, although significant thickening compared to controls was present only in the central region and temporal regions (Fig. 5D, P = 0.008 and 0.029, respectively). In contrast, for BD2 patients with longer disease duration, the ORL was significantly thinner compared to controls in all regions (P ≤ 0.001). The greatest percentage (26.6%) of decrease in thickness occurred in the nasal regions. 
Further, we analyzed the status of the EZ and IZ bands in the OCT images to evaluate the integrity of the photoreceptor layer (Figs. 4C, 4D). In BD1, 8 out of the 16 eyes had no disruption of the EZ band and 2 out of the 16 had no disruption of the IZ band (50.0% and 12.5% for EZ and IZ bands, respectively; Table 3, Fig. 6). Among the 16 eyes, 8 had EZ disruption scales from 1 to 2 (Table 3) and 14 had IZ disruption scales from 1 to 3 (Table 3). There were significantly more eyes of patients in the BD2 group that had EZ and IZ disruptions compared to the BD1 group (χ2, P = 0.004 for EZ and P = 0.005 for IZ). In this group, only two eyes had no disruption of the EZ, and only one had no disruption of the IZ bands (12.5% and 6.3% for EZ and IZ, respectively; Table 3, Fig. 6). Among the 16 eyes, 8 and 10 eyes had EZ and IZ disruption grading scales of 4, respectively (Table 3). 
Table 2
 
Correlations of Global and Regional Intraretinal Layers (μm) with BCVA (Unit: logMAR, n = 32)
Table 2
 
Correlations of Global and Regional Intraretinal Layers (μm) with BCVA (Unit: logMAR, n = 32)
Table 3
 
Comparisons of EZ and IZ Disruption Scales Between the Two BD Groups
Table 3
 
Comparisons of EZ and IZ Disruption Scales Between the Two BD Groups
Figure 6
 
Percentage of intact EZ and IZ bands in normal subjects and BD patients. There were no disruptions of the EZ or IZ band in the control group. More eyes with EZ and IZ band disruptions were found in BD2 patients than in BD1 patients.
Figure 6
 
Percentage of intact EZ and IZ bands in normal subjects and BD patients. There were no disruptions of the EZ or IZ band in the control group. More eyes with EZ and IZ band disruptions were found in BD2 patients than in BD1 patients.
The thickness of the RPE was increased in BD1 patients compared with controls in the central region (Fig. 5E, P < 0.001), and in the superior (P = 0.027), inferior (P = 0.046), and nasal regions (P = 0.012). Global RPE was also increased in BD1 patients compared with controls (P = 0.004). In BD2 patients, RPE thickening was also present in the central region and all four meridional regions (Fig. 5E, P < 0.05) compared with controls. Global RPE was increased in BD2 patients compared with both controls (Fig. 5E, P < 0.001) and BD1 patients (P < 0.05). The RPE thickness was greater in BD2 patients compared with BD1 patients in the superior (Fig. 5E, P = 0.016), inferior (P = 0.018), and temporal regions (P = 0.02). 
Associations Between Changes in Intraretinal Structure and VA
For both BD groups, increased global and each regional NFL thicknesses were associated with reduced VA (Pearson r range: 0.355∼ 0.573, P < 0.05). Increased global and each regional INL thicknesses correlated with worse VA (Pearson r range: 0.350∼ 0.696, P ≤ 0.05). Reduced VA was also correlated with decreased global ORL thickness and thickness at each region, as well as greater grading scales of EZ and IZ band disruptions (Pearson r range: −0.773 ∼ −0.650 for ORL thickness, P < 0.001; Spearman r = 0.790 and 0.745 for EZ and IZ disruptions, respectively, P < 0.001). For the RPE and the GCL+IPL, there were no correlations of thickness changes in any of the regions with VA. 
Multiple stepwise regression analysis for BCVA (Table 4) revealed that only the EZ disruption scale (P < 0.001), nasal ORL (P < 0.001), temporal INL (P = 0.003), inferior NFL (P = 0.014), and nasal INL (P = 0.029) were independent predictors of BCVA. 
Table 4
 
Results of Multiple Linear Regression Analysis for Independence of Factors Contributing to BCVA
Table 4
 
Results of Multiple Linear Regression Analysis for Independence of Factors Contributing to BCVA
Discussion
The purpose of this study was to characterize the changes occurring within specific retinal layers in clinically quiescent BD eyes and identify their associations with VA. To the best of our knowledge, the thicknesses of the macular intraretinal layers have not been previously evaluated in BD eyes. Using SD-OCT with automatic segmentation techniques, we found significant thickening in the inner retinal layers including the NFL, GCL+IPL, and INL. Outer retinal layer thinning and RPE thickening were present in BD2, the group with longer disease duration. These results revealed that the repeated inflammatory episodes in BD affected both inner and ORL structures. Further, the duration of the Behcet's uveitis played an important role in the thickness profiles of specific intraretinal layers. 
A principal feature of BD is the recurrent inflammatory attacks followed by self-limiting remissions.31 In the active phase, there is concurrent retinal involvement, such as retinal vasculitis and macular edema. In addition, the occurrence of retinal thickening has been shown using OCT.46 Although our patients were in remission and showed no clinically obvious intraocular inflammation, thickening in the inner retinal layers during this period suggests that there may still exist some subclinical inflammatory activity, even after the inflammation is under control. Retinal thickening due to subclinical inflammation was also reported in previous studies of early diabetic retinopathy.32,33 We found the phenomenon was most pronounced in the NFL, especially on the nasal side. This NFL thickening on the nasal side was also observed by Yamamoto et al.,34 signifying a potential increased susceptibility of this anatomical region to thickening in response to inflammatory stimuli. This point should be validated in a follow-up study. 
We found that the increased thickness of the inner retinal layers tended to be greater with increased disease duration. Although we cannot verify if this tendency of increased subclinical inflammatory activity increases the susceptibility of an acute recurrent attack of uveitis, our results nonetheless suggest that SD-OCT measurement of the inner retinal layer thicknesses, especially the NFL, may be useful and sensitive in evaluating the subclinical macular involvement in response to inflammatory stimuli during the quiescent phases of Behcet's posterior uveitis. Identifying patients with subclinical inflammatory activity before development of overt uveitis may help clinicians to apply preventive procedures at this stage and may potentially be useful in predicting response to therapy. 
In our study, there was dramatically more disruption of the outer retinal bands and thinning of ORL in the BD2 patients, all of whom had longer disease duration. This may be attributed to permanent and progressive loss of the photoreceptors in eyes with long-standing BD.18,35 The progressive damage of the photoreceptors is probably the result of recurrent macular edema attacks due to inflammation of the retinal vessels, causing irreversible damage of the neurosensory retina.6,10,36 This finding agrees with previous pathology studies that have shown retinal damage in BD.37,38 
Additionally, we found RPE thickening in our quiescent BD patients. This finding agrees with previous clinical studies that reported RPE proliferation and dysfunction.4,39 To our knowledge, the finding of a thickened RPE on OCT examination has not been reported in BD patients. We found the global RPE thicknesses changed from 23.76 μm for normal subjects to 28.23 μm for BD2 patients, which is within the range of the 7-μm axial resolution of the Spectralis SD-OCT by the company. Further investigation with higher resolution, such as 3 μm available in ultra-high resolution OCT instruments, is needed to validate such subtle changes in the RPE and understand the thickness changes of the ORLs. 
Correlational analysis revealed that decreases in ORL thickness, as well as the disruption scale of the ORLs were strongly correlated with worse VA. Multiple stepwise regression analysis further revealed that the disruption scale of the EZ band and nasal ORL thickness were independent predictors of BCVA. These results support the hypothesis that the ORL thickness and continuity of the EZ band reflect the integrity of the photoreceptor layer. The relationships between VA and parameters reflecting the photoreceptor damage as seen by OCT have been shown previously in several other retinal diseases,18,19,40,41 and we now suggest that ORL thickness as well as EZ band disruption may be a useful predictor for VA damage in remission of BD. In addition, correlational analysis showed that increases in NFL and INL thicknesses in most regions were moderately correlated with decreases in VA, suggesting that with longer disease duration, increased thickening of the retinal NLF and INL in response to the inflammatory stimuli may also lead to photoreceptor damage. 
FA is useful in demonstrating the retinal vasculature, but it cannot be performed frequently due to its invasive nature. Studies using conventional OCT have previously shown retinal involvement for BD patients. However, most of the studies focused only on the total retinal thickness due to the limitation of OCT resolution, and the results have been controversial. Corrêa et al.12 reported thickening of the macula,10 while Garcher et al.42 reported significant retinal thinning.43 Based upon our data, the inconsistent change patterns of total retinal layer thickness reported in those previous studies may be explained by the different change patterns that we found in each intraretinal layer. Therefore, measuring the thickness of each intraretinal layer rather than the total retinal layer thickness may provide more useful information, as these are the layers thought to be inflamed or irreversibly damaged in BD patients after repeated inflammatory attacks. In this respect, SD-OCT with retinal layer segmentation algorithm has several advantages over conventional imaging methods in evaluating the changes of subtle macular intraretinal layers. 
This study has several limitations. It was a cross-sectional, retrospective study with a small sample size and not randomized. Both eyes in Behcet's uveitic patients were included in the analysis, and the correlation between both eyes may affect the interpretation on the findings.44,45 We did not correct the OCT lateral scale due to the retrospective nature of the study, and axial length information was not available for all patients. The high myopia with long axial length may affect the results regarding the retinal thickness measurement.46,47 Additionally, previous studies have demonstrated that the axial length may be correlated with retinal thickness in high myopia.48,49 Thus we only enrolled eyes with emmetropia or low myopia and matched the three groups with refractive error as well as possible to try to reduce the influence of axial length on the interpretation on the findings. Moreover, scanning of only the horizontal and vertical meridians did not provide any information to understated the changes in the regions between the meridians. Future studies will be composed of larger sample sizes and will use a longitudinal approach to fully understand the thickness changes in the intraretinal layers within the same patients over a long period time. 
In conclusion, we demonstrated that patients in remission of BD had a significant thickening in the inner retinal structure, especially in the NFL, compared to normal controls. Behcet's disease patients also exhibited changes of outer retinal structure, including reduced ORL thickness and increased disruption of the ORL bands. Inner retinal layer thickening and outer layer damage were associated with decreased VA. The results suggest that SD-OCT measurement of macular intraretinal layer structure with automatic segmentation may be useful in evaluating the macular involvement and predicting the degree of VA damage with the progression of Behcet's uveitic disease. 
Acknowledgments
The authors thank Britt Bromberg of Xenofile Editing (www.xenofileediting.com) for providing editing services for this manuscript. 
Supported by the National Major Equipment Program of China Grant 2012YQ12008004 (FL) and the National Nature Science Foundation of China Grants 81170869 (FL), 81570880 (FL), and 81400441 (MS). The authors alone are responsible for the content and writing of the paper. 
Disclosure: D. Cheng, None; Y. Wang, None; S. Huang, None; Q. Wu, None; Q. Chen, None; M. Shen, None; F. Lu None 
References
Deuter CM, Kotter I, Wallace GR, Murray PI, Stubiger N, Zierhut M. Behcet's disease: ocular effects and treatment. Prog Retin Eye Res. 2008; 27: 111–136.
Sakane T, Takeno M, Suzuki N, Inaba G. Behcet's disease. N Engl J Med. 1999; 341: 1284–1291.
Zierhut M, Stubiger N, Deuter CME, Kotter I. Behcet's disease. In: Pleyer U, Mondino B. eds. Uveitis and Immunological Disorders. Berlin, Heidelberg: Springer; 2005: 173–200.
Yang P, Fang W, Meng Q, Ren Y, Xing L, Kijlstra A. Clinical features of Chinese patients with Behcet's disease. Ophthalmology. 2008; 115: 312–318.
Kim M, Kim H, Kwon HJ, Kim SS, Koh HJ, Lee SC. Choroidal thickness in Behcet's uveitis: an enhanced depth imaging-optical coherence tomography and its association with angiographic changes. Invest Ophthalmol Vis Sci. 2013; 54: 6033–6039.
Markomichelakis NN, Halkiadakis I, Pantelia E, et al. Patterns of macular edema in patients with uveitis: qualitative and quantitative assessment using optical coherence tomography. Ophthalmology. 2004; 111: 946–953.
Benezra D, Cohen E. Treatment and visual prognosis in Behcet's disease. Br J Ophthalmol. 1986; 70: 589–592.
Zierhut M, Saal J, Pleyer U, Kotter I, Durk H, Fierlbeck G. Behcet's disease: epidemiology and eye manifestations in German and Mediterranean patients. German J Ophthalmol. 1995; 4: 246–251.
Paovic J, Paovic P, Sredovic V. Correlation between ocular manifestations and their complications as opposed to visual acuity and treatment in Behcet's disease. Autoimmune Dis. 2013; 2013: 842673.
Tran TH, de Smet MD, Bodaghi B, Fardeau C, Cassoux N, Lehoang P. Uveitic macular oedema: correlation between optical coherence tomography patterns with visual acuity and fluorescein angiography. Br J Ophthalmol. 2008; 92: 922–927.
Takeuchi M, Iwasaki T, Kezuka T, et al. Functional and morphological changes in the eyes of Behcet's patients with uveitis. Acta Ophthalmologica. 2010; 88: 257–262.
Correa TP, Arantes TE, Lima VC, Muccioli C. Structural evaluation of posterior pole in patients with Behcet's disease. Arquivos Brasileiros de Oftalmologia. 2011; 74: 395–399.
Bagci AM, Shahidi M, Ansari R, Blair M, Blair NP, Zelkha R. Thickness profiles of retinal layers by optical coherence tomography image segmentation. Am J Ophthalmol. 2008; 146: 679–687.
Chiu SJ, Li XT, Nicholas P, Toth CA, Izatt JA, Farsiu S. Automatic segmentation of seven retinal layers in SDOCT images congruent with expert manual segmentation. Optics Exp. 2010; 18: 19413–19428.
Liu X, Shen M, Huang S, Leng L, Zhu D, Lu F. Repeatability and reproducibility of eight macular intra-retinal layer thicknesses determined by an automated segmentation algorithm using two SD-OCT instruments. PLoS One. 2014; 9: e87996.
Liu X, Shen M, Yuan Y, et al. Macular thickness profiles of intraretinal layers in myopia evaluated by ultrahigh-resolution optical coherence tomography. Am J Ophthalmol. 2015; 160: 53–61.
Pilotto E, Benetti E, Convento E, et al. Microperimetry, fundus autofluorescence and retinal layer changes in progressing geographic atrophy. Canadian J Ophthalmol. 2013; 48: 386–393.
Lim JI, Tan O, Fawzi AA, Hopkins JJ, Gil-Flamer JH, Huang D. A pilot study of Fourier-domain optical coherence tomography of retinal dystrophy patients. Am J Ophthalmol. 2008; 146: 417–426.
Witkin AJ, Ko TH, Fujimoto JG, et al. Ultra-high resolution optical coherence tomography assessment of photoreceptors in retinitis pigmentosa and related diseases. Am J Ophthalmol. 2006; 142: 945–952.
Unoki N, Nishijima K, Kita M, Hayashi R, Yoshimura N. Structural changes of fovea during remission of Behcet's disease as imaged by spectral domain optical coherence tomography. Eye. 2010; 24: 969–975.
Yuksel H, Turkcu FM, Sahin M, et al. Inner and outer segment junction (IS/OS line) integrity in ocular Behcet's disease. Arquivos Brasileiros de Oftalmologia. 2014; 77: 219–221.
International Study Group for Behcet's Disease. Criteria for diagnosis of Behcet's disease. Lancet. 1990; 335: 1078–1080.
Standardization of Uveitis Nomenclature for Reporting Clinical Data. Results of the First International Workshop. Am J Ophthalmol. 2005; 140: 509–516.
Mamo JG. The rate of visual loss in Behcet's disease. Arch Ophthalmol. 1970; 84: 451–452.
Yazici H, Ozyazgan Y. Medical management of Behcet's syndrome. Develop Ophthalmol. 1999; 31: 118–131.
Holladay JT. Proper method for calculating average visual acuity. J Refract Surgery. 1997; 13: 388–391.
Staurenghi G, Sadda S, Chakravarthy U, Spaide RF;, International Nomenclature for Optical Coherence Tomography. . Proposed lexicon for anatomic landmarks in normal posterior segment spectral-domain optical coherence tomography: the IN*OCT consensus. Ophthalmology. 2014; 121: 1572–1578.
Maheshwary AS, Oster SF, Yuson RM, Cheng L, Mojana F, Freeman WR. The association between percent disruption of the photoreceptor inner segment-outer segment junction and visual acuity in diabetic macular edema. Am J Ophthalmol. 2010; 150: 63–67.
Park JC, Collison FT, Fishman GA, et al. Objective analysis of hyperreflective outer retinal bands imaged by optical coherence tomography in patients with Stargardt disease. Invest Ophthalmol Vis Sci. 2015; 56: 4662–4667.
Huang Y, Cideciyan AV, Papastergiou GI, et al. Relation of optical coherence tomography to microanatomy in normal and rd chickens. Invest Ophthalmol Vis Sci. 1998; 39: 2405–2416.
Tugal-Tutkun I, Onal S, Altan-Yaycioglu R, Huseyin Altunbas H, Urgancioglu M. Uveitis in Behcet disease: an analysis of 880 patients. Am J Ophthalmol. 2004; 138: 373–380.
Yanoff M. Ocular pathology of diabetes mellitus. Am J Ophthalmol. 1969; 67: 21–38.
vom Hagen F, Feng Y, Hillenbrand A, et al. Early loss of arteriolar smooth muscle cells: more than just a pericyte loss in diabetic retinopathy. Exp Clin Endocrinol Diabetes. 2005; 113: 573–576.
Yamamoto M, Nishijima K, Nakamura M, Yoshimura N. Inner retinal changes in acute-phase Vogt-Koyanagi-Harada disease measured by enhanced spectral domain optical coherence tomography. Japanese J Ophthalmol. 2011; 55: 1–6.
Forooghian F, Yeh S, Faia LJ, Nussenblatt RB. Uveitic foveal atrophy: clinical features and associations. Arch Ophthalmol. 2009; 127: 179–186.
Paovic J, Paovic P, Sredovic V. Behcet's disease: systemic and ocular manifestations. BioMed Res Int. 2013; 2013: 247345.
Yamamoto JH, Fujino Y, Lin C, Nieda M, Juji T, Masuda K. S-antigen specific T cell clones from a patient with Behcet's disease. Br J Ophthalmol. 1994; 78: 927–932.
Yamamoto JH, Minami M, Inaba G, Masuda K, Mochizuki M. Cellular autoimmunity to retinal specific antigens in patients with Behcet's disease. Bri J Ophthalmol. 1993; 77: 584–589.
Malamos P, Masaoutis P, Georgalas I, et al. The role of fundus autofluorescence imaging in the study of the course of posterior uveitis disorders. BioMed Res Int. 2015; 2015: 247469.
Hayashi H, Yamashiro K, Tsujikawa A, Ota M, Otani A, Yoshimura N. Association between foveal photoreceptor integrity and visual outcome in neovascular age-related macular degeneration. Am J Ophthalmol. 2009; 148: 83–89.
Wakabayashi T, Oshima Y, Fujimoto H, et al. Foveal microstructure and visual acuity after retinal detachment repair: imaging analysis by Fourier-domain optical coherence tomography. Ophthalmology. 2009; 116: 519–528.
Garcher C, Bielefeld P, Desvaux C, Besancenot JF, Bron A. Bilateral loss of vision and macular ischemia related to Behçet disease. Am J Ophthalmol. 1997; 124: 116–117.
Yilmaz G, Akova Y, Aydin P. Macular ischaemia in Behcet's disease. Eye. 2000; 14 pt 5: 717–720.
Armstrong RA. Statistical guidelines for the analysis of data obtained from one or both eyes. Ophthal Physiol Optics. 2013; 33: 7–14.
Karakosta A, Vassilaki M, Plainis S, Elfadl NH, Tsilimbaris M, Moschandreas J. Choice of analytic approach for eye-specific outcomes: one eye or two? Am J Ophthalmol. 2012; 153: 571–579.
Nowroozizadeh S, Cirineo N, Amini N, et al. Influence of correction of ocular magnification on spectral-domain OCT retinal nerve fiber layer measurement variability and performance. Invest Ophthalmol Vis Sci. 2014; 55: 3439–3446.
Kang SH, Hong SW, Im SK, Lee SH, Ahn MD. Effect of myopia on the thickness of the retinal nerve fiber layer measured by Cirrus HD optical coherence tomography. Invest Ophthalmol Vis Sci. 2010; 51: 4075–4083.
Lim MC, Hoh ST, Foster PJ, et al. Use of optical coherence tomography to assess variations in macular retinal thickness in myopia. Invest Ophthalmol Vis Sci. 2005; 46: 974–978.
Ooto S, Hangai M, Sakamoto A, et al. Three-dimensional profile of macular retinal thickness in normal Japanese eyes. Invest Ophthalmol Vis Sci. 2010; 51: 465–473.
Figure 1
 
Segmentation of five intraretinal layers and macular regions analyzed by SD-OCT and automatic segmentation technology. (A) Cross-sectional image along the horizontal meridian of a Behcet uveitic eye. (B) Cross-sectional image along the vertical meridian of a Behcet uveitic eye. (C) The thickness profiles of five intraretinal layers were obtained within a 6-mm circular area around the macula that was divided into five regions for analysis. The central region included a 1-mm-diameter circle centered on the fovea. The temporal, superior, nasal, and inferior regions consisted of the horizontal or vertical scan lines extended from the intersection points to the outer diameter of 6 mm. Scale bar: 200 μm.
Figure 1
 
Segmentation of five intraretinal layers and macular regions analyzed by SD-OCT and automatic segmentation technology. (A) Cross-sectional image along the horizontal meridian of a Behcet uveitic eye. (B) Cross-sectional image along the vertical meridian of a Behcet uveitic eye. (C) The thickness profiles of five intraretinal layers were obtained within a 6-mm circular area around the macula that was divided into five regions for analysis. The central region included a 1-mm-diameter circle centered on the fovea. The temporal, superior, nasal, and inferior regions consisted of the horizontal or vertical scan lines extended from the intersection points to the outer diameter of 6 mm. Scale bar: 200 μm.
Figure 2
 
Representative examples of grading EZ and IZ band disruptions on Heidelberg Spectralis SD-OCT images through the fovea. The ellipsoid zone and IZ bands were evaluated over a span of 500 μm in either direction of the foveal horizontal and vertical meridians. The line disruption was graded from 0 to 2. (A) Grade 0 bands had no disruptions. (B) Grade 1 bands had focal disruptions ≤200 μm. (C) Grade 2 bands had focal disruptions >200 μm.
Figure 2
 
Representative examples of grading EZ and IZ band disruptions on Heidelberg Spectralis SD-OCT images through the fovea. The ellipsoid zone and IZ bands were evaluated over a span of 500 μm in either direction of the foveal horizontal and vertical meridians. The line disruption was graded from 0 to 2. (A) Grade 0 bands had no disruptions. (B) Grade 1 bands had focal disruptions ≤200 μm. (C) Grade 2 bands had focal disruptions >200 μm.
Figure 3
 
Thickness profiles of five intraretinal layers in control, BD1, and BD2 groups. Thickness profiles of five intraretinal layers along horizontal (top row) and vertical (bottom row) scans were measured from SD-OCT images and averaged over each group.
Figure 3
 
Thickness profiles of five intraretinal layers in control, BD1, and BD2 groups. Thickness profiles of five intraretinal layers along horizontal (top row) and vertical (bottom row) scans were measured from SD-OCT images and averaged over each group.
Figure 4
 
Representative Heidelberg Spectralis SD-OCT images of eyes for a normal subject and a BD1 and a BD2 patient. Horizontal scan (left). Vertical scan (right). (A, B) Spectral-domain optical coherence tomography images demonstrating the intraretinal layer thicknesses in a normal patient with a VA of 20/20 (top), a BD1 patient with a VA of 20/25 (middle), and a BD2 patient with a VA of 20/1000 (bottom). (C, D) The structure of the outer retinal bands around the central fovea. The longitudinal reflectivity profile (LRP) as a function of retinal depth was calculated over a span of 500 μm of the central fovea in either direction of the horizontal and vertical scans and overlaid on the image. The arrowheads correspond to the EZ and IZ bands. In the left panel, the EZ/IZ bands were intact along both the horizontal and vertical scans with an EZ disruption scale = 0 and IZ disruption scale = 0. The longitudinal reflectivity profile also showed that three peaks corresponding to the EZ, IZ, and RPE were visible. In the middle panel, the EZ and IZ bands were either partially or totally disrupted along the horizontal and vertical scans with an EZ disruption scale = 1 and IZ disruption scale = 3. The peak of the IZ was not visible in the LRP (D, middle panel) due to the total loss of the IZ band along the vertical scan (IZ disruption grade = 2). For this patient, there was a decrease in reflectivity of the EZ peak due to the partial disruption of this band. There was a total loss of EZ and IZ bands (right panel, arrows) with an EZ scale = 4 and IZ scale = 4. The peaks of the EZ and IZ bands were not visible in the LRP along either the horizontal or vertical scans. Scale bar: = 100 μm.
Figure 4
 
Representative Heidelberg Spectralis SD-OCT images of eyes for a normal subject and a BD1 and a BD2 patient. Horizontal scan (left). Vertical scan (right). (A, B) Spectral-domain optical coherence tomography images demonstrating the intraretinal layer thicknesses in a normal patient with a VA of 20/20 (top), a BD1 patient with a VA of 20/25 (middle), and a BD2 patient with a VA of 20/1000 (bottom). (C, D) The structure of the outer retinal bands around the central fovea. The longitudinal reflectivity profile (LRP) as a function of retinal depth was calculated over a span of 500 μm of the central fovea in either direction of the horizontal and vertical scans and overlaid on the image. The arrowheads correspond to the EZ and IZ bands. In the left panel, the EZ/IZ bands were intact along both the horizontal and vertical scans with an EZ disruption scale = 0 and IZ disruption scale = 0. The longitudinal reflectivity profile also showed that three peaks corresponding to the EZ, IZ, and RPE were visible. In the middle panel, the EZ and IZ bands were either partially or totally disrupted along the horizontal and vertical scans with an EZ disruption scale = 1 and IZ disruption scale = 3. The peak of the IZ was not visible in the LRP (D, middle panel) due to the total loss of the IZ band along the vertical scan (IZ disruption grade = 2). For this patient, there was a decrease in reflectivity of the EZ peak due to the partial disruption of this band. There was a total loss of EZ and IZ bands (right panel, arrows) with an EZ scale = 4 and IZ scale = 4. The peaks of the EZ and IZ bands were not visible in the LRP along either the horizontal or vertical scans. Scale bar: = 100 μm.
Figure 5
 
Five intraretinal layer thicknesses in the macular region determined by SD-OCT and automated segmentation technology. (A) NFL; (B) GCL+IPL; (C) INL; (D) ORL; (E) RPE. *P < 0.05, **P ≤ 0.001.
Figure 5
 
Five intraretinal layer thicknesses in the macular region determined by SD-OCT and automated segmentation technology. (A) NFL; (B) GCL+IPL; (C) INL; (D) ORL; (E) RPE. *P < 0.05, **P ≤ 0.001.
Figure 6
 
Percentage of intact EZ and IZ bands in normal subjects and BD patients. There were no disruptions of the EZ or IZ band in the control group. More eyes with EZ and IZ band disruptions were found in BD2 patients than in BD1 patients.
Figure 6
 
Percentage of intact EZ and IZ bands in normal subjects and BD patients. There were no disruptions of the EZ or IZ band in the control group. More eyes with EZ and IZ band disruptions were found in BD2 patients than in BD1 patients.
Table 1
 
Subject Characteristics of Normal and BD Groups
Table 1
 
Subject Characteristics of Normal and BD Groups
Table 2
 
Correlations of Global and Regional Intraretinal Layers (μm) with BCVA (Unit: logMAR, n = 32)
Table 2
 
Correlations of Global and Regional Intraretinal Layers (μm) with BCVA (Unit: logMAR, n = 32)
Table 3
 
Comparisons of EZ and IZ Disruption Scales Between the Two BD Groups
Table 3
 
Comparisons of EZ and IZ Disruption Scales Between the Two BD Groups
Table 4
 
Results of Multiple Linear Regression Analysis for Independence of Factors Contributing to BCVA
Table 4
 
Results of Multiple Linear Regression Analysis for Independence of Factors Contributing to BCVA
×
×

This PDF is available to Subscribers Only

Sign in or purchase a subscription to access this content. ×

You must be signed into an individual account to use this feature.

×