https://orcid.org/0000-0002-2028-9935
Figures
Abstract
Purpose
Behcet disease (BD) is an autoimmune disease characterized by diffuse all-sized obliterative vasculitis, thrombotic vasculopathy, and common ocular involvement. This study aims to evaluate the retinal microvascular alterations in BD using optical coherence tomography angiography (OCTA).
Methods
PubMed, Embase, and Web of Sciences were systematically searched for relevant studies assessing OCTA measurements in BD and healthy controls (HCs). Meta-analysis was conducted on OCTA parameters with at least two studies using the same OCTA device, with similar case and control groups. A qualitative synthesis approach was used to report the data that could not be pooled.
Results
Twenty-eight related studies (769 BD subjects, 123 active BU eyes, 462 inactive BU eyes, 112 non-specified ocular BD, and 486 non-ocular BD eyes) were included. Patients with inactive Behcet uveitis (BU) and non-ocular BD showed a statistically significant greater FAZ size and lower superficial retinal capillary plexus and deep retinal capillary plexus vessel density (VD) in comparison to HCs, particularly in the parafoveal sector. Also, radial peripapillary capillary (RPC) VD was lower in patients with inactive BU than HCs. However, no significant difference in OCTA parameters was found in patients with active BU compared to HCs.
Conclusions
Meta-analysis demonstrated reduced VD in various regions and greater FAZ size in superficial and deep retinal plexuses, particularly in inactive BU and non-ocular BD, which are mostly studied so far. However, more studies with larger sample sizes should draw a more definite conclusion. OCTA can provide valuable insights into retinal microvasculature changes in BD.
Citation: Mozafar M, Amanollahi M, Samiee R, Jameie M, Mousavi A, Ghanbari Z, et al. (2025) OCTA measurements in Behcet’s disease across different stages of the disease activity: A systematic review and meta-analysis. PLoS One 20(7): e0323192. https://doi.org/10.1371/journal.pone.0323192
Editor: Ayman Elnahry,, National Eye Institute, UNITED STATES OF AMERICA
Received: October 17, 2024; Accepted: April 3, 2025; Published: July 2, 2025
Copyright: © 2025 Mozafar et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: The data supporting the findings is available in a compressed Supporting information as Supporting information 2.
Funding: The author(s) received no specific funding for this work.
Competing interests: The authors have declared that no competing interests exist.
1. Introduction
Behcet’s disease (BD) is characterized as a chronic, inflammatory condition that affects several organs such as skin, mucous membranes, eyes, joints, and lungs, as well as gastrointestinal, genital, and central nervous systems [1–4]. It is distinguished by immune-mediated vasculitis, which can affect blood vessels of various organs [5]. Ocular inflammation most commonly manifests as relapsing-remitting uveitis, affecting more than 70% of patients [6–9].
According to the European Alliance of Associations for Rheumatology (EULAR), multimodal imaging is recommended for managing patients with BD-associated severe eye disease. This is defined by the presence of retinal vasculitis affecting the macula or optic nerve leading to significant vision loss. Multimodal imaging typically includes fluorescein angiography (FA), indocyanine green angiography (ICGA), and spectral domain optical coherence tomography (SD-OCT) [10,11]. FA remains the gold standard for documenting and monitoring the involvement of the posterior segment of the eye in BD. However, it’s an invasive procedure requiring an injected dye and has limitations in visualizing deeper vascular structures of the retina [6,12].
Optical coherence tomography angiography (OCTA) offers a new approach to visualize the intricate network of microvasculature in the retina and choroid. Unlike traditional angiography, OCTA does not require intravenous contrast injection [13,14]. OCTA can create detailed, layered images of the retinal vasculature, such as the superficial retinal capillary plexus (SRCP), the deep retinal capillary plexus (DRCP), and the choriocapillaris. The detection of retinal or choroidal vascular changes, measurement of the foveal avascular zone (FAZ), and quantification of vascular density (VD) in the inner retina, outer retinal circulation, or Choriocapillaris are all advantages of OCTA [14]. OCTA has been widely used to assess the microvasculature structure in various diseases such as diabetic retinopathy [15, 16], age-related macular degeneration [17], optic neuropathies [18], and autoimmune diseases such as systemic lupus erythematous [19], Neuromyelitis optica spectrum disorders (NMOSD) and Myelin oligodendrocyte glycoprotein antibody disease (MOGAD) [20].
Previously, the results of ocular and non-ocular BD retinal microvasculature were assessed by Ji et al. by pooling relevant studies up to 2021 [21]. However, several gaps still exist that need to be addressed. A comprehensive systematic review incorporating all the OCTA parameters including choriocapillaris and radial peripapillary capillary VD was missing. Furthermore, despite the typical categorization of ocular BD into those with active or inactive uveitis by various studies, the previous study pooled both of the results together, which could justify the higher amount of heterogeneity throughout their analyses. In fact, uveitis can lead to retinal vasculitis and hence result in microvasculature alterations in OCTA imaging [22]. Thus, it is imperative to separately evaluate ocular BD based on uveitis activity status, which is the primary goal of the present study. Herein, we aim to comprehensively review the studies that evaluated the microvascular changes in the eyes of patients with BD (i.e., active or inactive ocular BD and non-ocular BD) using OCTA.
2. Methods
This systematic review and meta-analysis adhere to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guideline [23] (Tables S5 and S6 in S1 File). The study protocol was registered on the International Prospective Register of Systematic Reviews (PROSPERO) website (Registration code: CRD42023468476).
2.1. Search strategy
We searched PubMed, EMBASE, and Web of Sciences for relevant papers from the earliest to October 09, 2024. The keywords contain combinations of OCTA and Behcet headings as follows: (“optical coherence tomography angiography” OR “OCT angiography” OR “OCTA”) AND (“Behcet”). The detailed search strategy for the three databases is provided in Tables S1-3 in S1 File). Two independent reviewers did the initial screening (RS, MA), and disagreements were solved by consulting the senior author (EK). Included studies were not restricted to location, age, and gender discrepancies. We further explored the references of included studies for additional relevant articles.
2.2. Inclusion and exclusion criteria
Studies in which OCTA evaluated retinal microvasculature of patients with BD were incorporated into this review if they met the following criteria: (1) original peer-reviewed papers, (2) English language, (3) Including patients with BD, (4) Including a control group. Exclusion criteria were as follows: (1) reviews, conference abstracts, and case reports, (2) non-English, (3) non-human, (4) no comparative control group, and (5) not stating the presence of ocular involvement in BD patients. The detailed selection process is illustrated in Fig 1.
2.3. Data extraction
Following demographics and clinical information were collected from each article: author’s name, publication year, study design, status of ocular inflammation (active Behcet’s uveitis (BU), inactive BU, non-ocular Behcet), sample size, female percentage, age, disease duration, studies’ inclusion and exclusion criteria, axial length, intraocular pressure (IOP) of enrolled eyes, and best corrected visual acuity (BCVA). We also extracted OCTA-related data such as device type, software, field of view, VD in the area of SRCP, DRCP and their relative subregions (whole, foveal, parafoveal, perifoveal), Choriocapillaris flow area, radial peripapillary capillary (RPC). Two independent authors extracted the data (MM, MA), and any conflicts were resolved by the senior advisor.
Active BU was defined by included studies as clinical signs of active posterior or panuveitis on ophthalmologic examination [24–29], whereas inactive BUs were selected from patients who had a history of previous uveitis attacks without apparent anterior or posterior uveitis at the time of study [30–33]. In the context of non-ocular BD, patients were identified as those without signs of ocular involvement. This includes an absence of past inflammatory signs such as posterior synechiae, vitreous cells, vascular sheathing, and chorioretinal scars [28, 34–40]. Additionally, some studies did not specify the activity status of ocular involvement [37, 40]. We referred to this population as a non-specified ocular BD group.
2.4. OCTA parameters
The primary metric analyzed in these studies was VD. It represents the proportion of a retinal area occupied by blood vessels. Researchers measured VD in various retinal layers and regions throughout the studies. However, the wording for superficial and deep layers varied slightly between studies. To ensure consistency, we categorized these layers simply as either the SRCP or the DRCP. Importantly, we confirmed that the definitions of all reported layers and regions aligned with established standards. Furthermore, to minimize potential discrepancies caused by differences in imaging devices, we only compared VD measurements from studies using the same type of OCTA device. The Early Treatment Diabetic Retinopathy Study (ETDRS) grid, established in 1991, provides a standardized approach for segmenting the macula into specific regions like whole, foveal, parafoveal, and perifoveal. This grid is widely used when analyzing vessel density in the SRCP and DRCP layers of OCTA scans [41]. Thus, we referred to the foveal as the central macula circle (1 mm in diameter), parafoveal as the inner ring circle (3 mm in diameter), and perifoveal as the outer ring (6 mm in diameter) [41]. Some studies subdivided the regions into superior, inferior, nasal, and temporal. We also analyzed the whole image retinal layer’s VDs in the 3*3 and 6*6 field of view separately, for a more precise comparison.
SRCP is defined as a retinal layer extended from 3 μm below the inner limiting membrane (ILM) to 15 μm below the inner plexiform layer (IPL) [25,27–30, 32, 34, 37, 42–46]. SRCP was measured in 3*3 fields of view by six studies [25,27,33,37,42,44], and in 6*6 fields of view by thirteen studies [24,26,29,34–36,38,43,45,47–50].
DRCP is generally defined as the retinal layer spanning from IPL to the outer plexiform layer (OPL), with minor variations in the exact boundaries across the studies included in this review [25,27–30,32,34,37,42–46]. Six studies measured DRCP in 3*3 fields of view [25, 27, 33, 37, 42, 44], while in twelve studies, DRCP was viewed in 6*6 fields of view [24, 26, 29, 34–36, 43, 45, 47–50].
Choriocapillaris is a thin layer beneath the Bruch membrane [51]. Unlike other layers, the flow area is its general measurement metric, which can be measured in a 1 mm or 3 mm radius by twelve studies [24, 29–31, 34, 35, 43, 45, 47–50]. To reduce the risk of heterogeneity, we analyzed the choriocapillaris in 1 mm and 3 mm radius separately.
FAZ is a capillary-free area centered on the fovea. Its size is measured in square millimeters (mm²) and reflects the circulation status in various diseases [52]. It makes the most common variable reported by 26 included studies [25, 27–40, 42–50, 53]. Reportedly, previous studies demonstrated greater FAZ parameters in the superficial layer than in the deep layer; highlighting separate measurements of FAZ based on the retinal layer [54]. The included studies in the current review mostly reported the FAZ in the superficial layer [25, 27, 30–35, 38, 39, 45, 53], while some reported the metric in the deep layer [25, 27, 30, 32–34, 39, 43, 53]. Noteworthy, some did not specify the layer in which FAZ was measured [28, 29, 35, 37, 40, 42, 44, 45, 47–50]; hence we referred to those as FAZ in general and were analyzed separately.
RPC is the primary OCTA-related measurement of the optic disc, the vascular plexus in the retinal nerve fiber layer (RNFL). In the present review, seven studies measured RPC VD and defined it as the area that prolongs from the ILM to the RNFL [29, 31, 45, 47–50].
2.5. Quality assessment
Newcastle-Ottawa Scale (NOS) was used as a risk-of-bias assessment tool [55]. Studies with less than five scores are considered at high risk of bias, and those with more than five are considered “good” ones. Two authors (MM, MA) reviewed the included studies for this purpose, and disagreements were referred to the senior author (EK).
2.6. Statistical analysis
This meta-analysis employed a quantitative approach (synthesis) to analyze all measurable OCTA imaging outcomes. The analysis generated mean and standard deviation (SD) values for each outcome. It’s important to note that a qualitative approach (synthesis) was used to report findings for parameters that could not be combined in specific comparisons (Supplemental Results in S1 File)). Whenever possible, values reported using other metrics were converted to mean and SD for consistency.
Quantitative data synthesis was conducted on all measurements assessed in at least two studies. These studies needed to share similar patient characteristics (active or inactive BU and non-ocular BD) and utilize the same type of OCTA device. Exceptions were made when a qualitative approach was deemed necessary to report study findings. The False Discovery Rate (FDR) method was applied to correct for multiple comparisons. Stata version 16 software was used for the statistical analyses. To compare cases and controls for each OCTA outcome, the analysis employed Hedges’ g, a standardized mean difference statistic, along with a 95% confidence interval (CI) to represent the effect size. Heterogeneity, or variation, across studies, was assessed using Higgin’s I² statistic. A fixed-effects model was used when heterogeneity was below 40%, while a random-effects model was used for studies with higher heterogeneity. The data supporting the findings is provided in S2 File).
3. Results
3.1. Studies’ characteristics
Twenty-eight studies were included. Table 1 illustrates an overview of the studies’ characteristics. 769 BD subjects (112 non-specified eyes, 123 active BU eyes, 462 inactive BU eyes, and 486 non-ocular BD eyes) and 1100 healthy control (HC) subjects (1446 HC eyes) were analyzed for different parameters. The majority of the patients and HCs were male (58.7% across all studies) and in their 40s. One of the studies addressed patients’ right and left eyes; hence, we considered this study as two distinct ones when conducting a meta-analysis (two spreadsheet rows were dedicated to this study) [31]. Among the studies, six assessed active BU eyes [24, 26–29, 50], 11 recruited inactive BU eyes (during the remission period) [24, 25, 30–33, 36, 42–44, 46, 56], and 14 included BD patients with no present or history of uveitis (non-ocular BD) [28,34–40,42,45,47–49,53]. Two did not determine the active or past BU and were allocated as non-specified ocular BD [37, 40, 49]. Additionally, the studies’ inclusion and exclusion criteria for selecting cases and controls were demonstrated in Table 2.
Except for FAZ, all the analyses regarding different parameters (VD or flow area) were grouped based on the OCTA device because of the basic differences in the values. Twenty-one studies applied RTVue XR Avanti (Optovue Inc., Fremont, CA) along with AngioVue or AngioAnalytics software [25, 26, 29–31, 33–40, 44–50, 53], two used Topcon DRI OCT Triton Plus along with ImageJ or IMAGEnet software (Topcon Corp., Tokyo, Japan) [27, 32], and two used Heidelberg Spectralis OCTA (Heidelberg Engineering, Heidelberg, Germany) with internal Heidelberg software [28, 42]. Moreover, a study used PLEX Elite [24], one used BM-400K Bmizar [43], and one conducted OCTA with Svision [56] (Table 3).
3.2. Metrics
Table 3 represents the retinal layers and related regions evaluated by studies. Twenty-six studies measured FAZ size [25, 27–40, 42–50, 53]. The included studies considered VD as their measurement (rather than perfusion density (PD) or fractal dimension (FD)) across OCTA variables. All of the studies measured macular VD (superficial and deep layers) [24–40, 42–50, 53,56], which was the most common site measured among the studies. Only seven assessed RPC VD [29,31,45,47–50]. The included studies assessed macular parafoveal VD and RPC VD as the most common variables (Table 3). A summary of the study results is illustrated in Fig 2. In the following, we demonstrate the results of meta-analyses conducted on the included studies (Quantitative data synthesis) (Table 4). Noteworthy, each study’s detailed results are presented in Table 5. Moreover, the qualitative synthesis of the included studies is reported as Supplemental Results in S1 File).
Abbreviations: BU: Bechet’s uveitis, DRCP: deep retinal capillary plexus, HC: healthy control, RPC: retinal peripapillary capillary, SRCP: superficial retinal capillary plexus, VD: vessel density.
3.3. Active BU vs. HC
3.3.1. Whole (6*6) SRCP VD (active BU vs. HC).
Two studies compared SRCP whole (6*6) VD between active BU and HC subgroups using the RTVue XR Avanti device [26, 29]. The pooled results (50 active BU and 54 HC eyes) showed no significant difference between the two groups (Hedges g = −0.59, CI= [−2.07 to 0.88], I2 = 74.28%, P value = 0.43, Corrected P value = 0.43) (Fig 3A).
A: whole image (3*3) SRCP VD in active BU and HC eyes, B: whole image (3*3) DRCP VD in active BU and HC eyes, C: FAZ in active BU and HC eyes. Abbreviations: BU: Bechet’s uveitis, CI: confidence interval, DRCP: deep retinal capillary plexus, SRCP: superficial retinal capillary plexus, SD: standard deviation, VD: vessel density.
3.3.2. Whole (6*6) DRCP VD (active BU vs. HC).
Two studies compared DRCP whole (6*6) VD between active BU and HC subgroups using the RTVue XR Avanti device [26, 29]. The pooled results (50 active BU and 54 HC eyes) showed no significant difference between the two groups (Hedges g = −0.36, CI= [−1.12 to 0.40], I2 = 74.28%, P value = 0.36, Corrected P value = 0.36) (Fig 3B).
3.3.3. FAZ (active BU vs. HC).
Two studies compared FAZ between active BU and HC subgroups using the RTVue XR Avanti [29] and Heidelberg [28] devices. Meta-analysis (45 active BU and 46 HC eyes) revealed no significant difference between the two groups (Hedges g = 0.853, CI= [−0.05 to 1.76], I2 = 77.39%, P value = 0.06, Corrected P value = 0.06) (Fig 3C).
3.4. Inactive BU vs. HC
3.4.1. Whole (3*3) SRCP VD (inactive BU vs. HC).
Four studies evaluated the whole picture (3*3) SRCP VD in inactive BU and HC eyes using the RTVue XR Avanti device [25, 33, 46, 57]. (59 active BU and 76 HC eyes) revealed no significant difference between the two groups (Hedges g = −0.75, CI= [−6.53 to 1.03], I2 = 98.64%, P value = 0.15) (Fig 4A). Values for the whole SRCP VD in the study by Pei et al. [44] were excluded from this analysis, since they were not provided in mean and SD and also were not convertible to this form due to being skewed away from normality.
A: whole image (3*3) SRCP VD in inactive BU and HC eyes, B: foveal SRCP in inactive BU and HC eyes, C: parafoveal SRCP in inactive BU and HC eyes, D: whole image (3*3) DRCP VD in inactive BU and HC eyes, E: foveal DRCP in inactive BU and HC eyes, F: parafoveal DRCP in inactive BU and HC eyes. Abbreviations: BU: Bechet’s uveitis, CI: confidence interval, DRCP: deep retinal capillary plexus, SRCP: superficial retinal capillary plexus, SD: standard deviation, VD: vessel density.
3.4.2. Foveal SRCP VD (inactive BU vs. HC).
Two studies compared SRCP foveal VD between inactive BU and HC subgroups using the RTVue XR Avanti device [30, 33]. Meta-analysis (65 inactive BU and 61 HC eyes) showed no significant difference between the two groups (Hedges g = −0.24, CI= [−0.70 to 0.23], I2 = 43.80%, P value = 0.32, Corrected P value = 0.46) (Fig 4B).
3.4.3. Parafoveal SRCP VD (inactive BU vs. HC).
Two studies compared SRCP parafoveal VD between inactive BU and HC subgroups using the RTVue XR Avanti device [30, 36]. Meta-analysis (72 inactive BU and 74 HC eyes) showed a significant reduction in the active BU group compared to the HC (Hedges g = −1.10, CI= [−1.80 to −0.41], I2 = 74.30%, P value = 0.001, Corrected P value = 0.003) (Fig 4C).
3.4.4. DRCP whole (3*3) VD (inactive BU vs. HC).
Four studies compared DRCP whole (3*3) VD between inactive BU and HC subgroups using the RTVue XR Avanti device [25,33,44,46,57]. Meta-analysis (161 inactive BU and 200 HC eyes) revealed a significant reduction in the inactive BU group (Hedges g = −1.32, CI= [−1.68 to −0.95], I2 = 47.31%, P value = 0.00001) (Fig 4D).
3.4.5. Foveal DRCP VD (inactive BU vs. HC).
Two studies compared DRCP foveal VD between inactive BU and HC subgroups using the RTVue XR Avanti device [30, 33]. Meta-analysis (65 inactive BU and 61 HC eyes) revealed no significant difference between the groups (Hedges g = −0.37, CI= [−0.85 to 0.10], I2 = 45.84%, P value = 0.13, Corrected P value = 0.13) (Fig 4E).
3.4.6. Parafoveal DRCP VD (inactive BU vs. HC).
Two studies compared DRCP parafoveal VD between inactive BU and HC subgroups using the RTVue XR Avanti device [30, 36]. Meta-analysis (72 inactive BU and 74 HC eyes) revealed a significant reduction in the active BU group (Hedges g = −1.38, CI= [−1.97 to −0.78], I2 = 62.41%, P value = 0.00001, Corrected P value = 0.000015) (Fig 4F).
3.4.7. RPC VD (inactive BU vs. HC).
One study evaluating both right and left eyes compared RPC VD between inactive BU and HC subgroups using the RTVue XR Avanti device [31]. Meta-analysis of the left and right eyes’ data (40 inactive BU and 52 HC eyes) revealed a significant reduction in the inactive BU group (Hedges g = −1.22, CI= [−1.66 to −0.78], I2 = 0.00%, P value = 0.00001, Corrected P value = 0.00002) (Fig 5A).
A: RPC VD in inactive BU and HC eyes, B: Choriocapillaris flow area (1 mm) in inactive BU and HC eyes, C: FAZ in inactive BU and HC eyes, D: Superficial FAZ in inactive BU and HC eyes, E: Deep FAZ in inactive BU and HC eyes. Abbreviations: BU: Bechet’s uveitis, CI: confidence interval, FAZ: foveal avascular zone, RPC: radial peripapillary capillary, VD: vessel density.
3.4.8. Choriocapillaris flow area (inactive BU vs. HC).
Two studies, one assessing both eyes, compared the Choriocapillaris flow area (1 mm) between inactive BU and HC subgroups using the RTVue XR Avanti device [30, 31]. Meta-analysis (75 inactive BU and 82 HC eyes) revealed no significant difference between the two groups (Hedges g = −0.20, CI= [−0.51 to 0.11], I2 = 0.00%, P value = 0.22, Corrected P value = 0.66) (Fig 5B).
3.4.9. FAZ (inactive BU vs. HC).
Four studies compared FAZ between inactive BU and HC subgroups utilizing the RTVue XR Avanti [36, 44, 46] and Heidelberg devices [42]. Meta-analysis (172 inactive BU and 214 HC eyes) revealed no significant difference between the two groups (Hedges g = 1.22, CI= [−0.07 to 2.50], I2 = 96.37%, P value = 0.06, Corrected P value = 0.06) (Fig 5C). However, conducting a leave-one-out analysis (removing Mostafa et al. study [46]) revealed a significant difference between the two groups (P value <0.001).
3.4.10. FAZ superficial (inactive BU vs. HC).
Six studies compared superficial FAZ between inactive BU and HC subgroups applying RTVue XR Avanti [30, 31, 33, 46], Heidelberg [25], and Topcon [32] devices. Meta-analysis (166 inactive BU and 176 HC eyes) revealed a significant difference between the two groups (Hedges g = 0.27, CI= [0.05 to 0.48], I2 = 0.00%, P value = 0.01, Corrected P value = 0.02) (Fig 5D).
3.4.11. FAZ deep (inactive BU vs. HC).
Four studies compared deep FAZ between inactive BU and HC subgroups utilizing the RTVue XR Avanti [30 33], Heidelberg [25], and Topcon [32] devices. Meta-analysis of the four studies (126 inactive BU and 124 HC eyes) revealed a significant difference between the two groups, as inactive BU had greater metric compared to the HC (Hedges g = 0.48, CI= [0.08 to 0.87], I2 = 59.18%, P value = 0.02, Corrected P value = 0.04) (Fig 5E).
3.5. Non-ocular BD vs. HC
3.5.1. Whole (6*6) SRCP VD (non-ocular BD vs. HC).
Nine studies compared SRCP whole (6*6) VD between non-ocular BD and HC subgroups utilizing the RTVue XR Avanti device [24, 34–36, 38, 45, 47–49]. Meta-analysis (284 non-ocular BD and 310 HC eyes) revealed a significant reduction in the non-ocular BD group (Hedges g = −0.61, CI= [−1.11 to −0.12], I2 = 88.08%, P value = 0.01, Corrected P value = 0.02) (Fig 6A).
A: whole image (6*6) SRCP VD in non-ocular BD and HC eyes, B: Foveal SRCP VD in non-ocular BD and HC eyes, C: Parafoveal SRCP VD in non-ocular BD and HC eyes, D: Perifoveal SRCP VD in non-ocular BD and HC eyes. Abbreviations: BD: Bechet disease, CI: confidence interval, N: non-ocular BD, SRCP: superficial retinal capillary plexus, SD: standard deviation, VD: vessel density.
3.5.2. Foveal SRCP VD (non-ocular BD vs. HC).
Seven studies compared SRCP foveal VD between non-ocular BD and HC subgroups applying the RTVue XR Avanti device [35, 39, 40, 45, 47–49]. Meta-analysis (232 non-ocular BD and 271 HC eyes) revealed no statistically significant difference between the groups (Hedges g = −0.14, CI= [−0.31 to 0.04], I2 = 0.59%, P value = 0.13, Corrected P value = 0.39) (Fig 6B).
3.5.3. Parafoveal SRCP VD (non-ocular BD vs. HC).
Eight studies compared SRCP parafoveal VD between non-ocular BD and HC subgroups utilizing the RTVue XR Avanti device [35–37, 39, 40, 45, 47, 48]. Meta-analysis (331 non-ocular BD and 368 HC eyes) demonstrated a significant increase in favor of the HC group (Hedges g = −0.21, CI= [−0.36 to −0.06], I2 = 14.62%, P value = 0.005, Corrected P value = 0.005) (Fig 6C).
3.5.4. Perifoveal SRCP VD (non-ocular BD vs. HC).
Five studies compared SRCP perifoveal VD between non-ocular BD and HC subgroups utilizing the RTVue XR Avanti device [35, 40, 45, 48, 49]. Meta-analysis (138 non-ocular BD and 186 HC eyes) demonstrated significantly greater value in favor of the HC group (Hedges g = −0.28, CI= [−0.5 to −0.06], I2 = 0.00%, P value = 0.01, Corrected P value = 0.01) (Fig 6D).
3.5.5. Whole (6*6) DRCP VD (non-ocular BD vs. HC).
Seven studies compared DRCP whole (6*6) VD between non-ocular BD and HC subgroups utilizing the RTVue XR Avanti device [34–36, 45, 47‒49]. Meta-analysis of the studies (259 non-ocular BD and 300 HC eyes) revealed a significantly decreased value among non-ocular BD than HC (Hedges g = −0.6, CI= [−1.03 to −0.17], I2 = 83.33%, P value = 0.006, Corrected P value = 0.012) (Fig 7A).
A: whole image (6*6) DRCP VD in non-ocular BD and HC eyes, B: Foveal DRCP VD in non-ocular BU and HC eyes, C: Parafoveal DRCP VD in non-ocular BD and HC eyes, D: Perifoveal DRCP VD in non-ocular BD and HC eyes. Abbreviations: BD: Bechet disease, CI: confidence interval, DRCP: deep retinal capillary plexus, N: non-ocular BD, SD: standard deviation, VD: vessel density.
3.5.6. Foveal DRCP VD (non-ocular BD vs. HC).
Seven studies compared DRCP foveal VD between non-ocular BD and HC subgroups utilizing the RTVue XR Avanti device [35, 39, 40, 45, 47–49]. Meta-analysis (232 non-ocular BD and 271 HC eyes) revealed significantly greater metric for HC than patients (Hedges g = −0.7, CI= [−1.21 to −0.18], I2 = 86.58%, P value = 0.007, Corrected P value = 0.012) (Fig 7B).
3.5.7. Parafoveal DRCP VD (non-ocular BD vs. HC).
Ten studies compared DRCP parafoveal VD between non-ocular BD and HC subgroups utilizing the RTVue XR Avanti device [35–37, 39, 40, 45, 47–49, 53]. Meta-analysis (375 non-ocular BD and 417 HC eyes) revealed no significant difference between the two groups (Hedges g = −0.27, CI= [−0.56 to 0.02], I2 = 77.36%, P value = 0.06, Corrected P value = 0.06) (Fig 7C).
3.5.8. Perifoveal DRCP VD (non-ocular BD vs. HC).
Five studies compared DRCP perifoveal VD between non-ocular BD and HC subgroups utilizing the RTVue XR Avanti device [35, 40, 45, 48, 49]. Meta-analysis (138 non-ocular BD and 186 HC eyes) revealed a significant reduction in the non-ocular BD eyes (Hedges g = −0.45, CI = [−0.77 to −0.12], I2 = 49.75%, P value = 0.007, Corrected P value = 0.014) (Fig 7D).
3.5.9. RPC VD (non-ocular BD vs. HC).
Four studies compared RPC VD between non-ocular BD and HC subgroups utilizing the RTVue XR Avanti device [45, 47–49]. Meta-analysis (152 non-ocular BD and 178 HC eyes) showed no significant difference between the two groups (Hedges g = −0.48, CI = [−0.99 to 0.03], I2 = 79.41%, P value = 0.06, Corrected P value = 0.06) (Fig 8A).
A: RPC VD in non-ocular BD and HC eyes, B: Choriocapillaris flow area (1 mm) in non-ocular BD and HC eyes, C: Choriocapillaris flow area (3 mm) in non-ocular BD and HC eyes, D: FAZ in non-ocular BD and HC eyes, E: Superficial FAZ in non-ocular BD and HC eyes, F: Deep FAZ in non-ocular BD and HC eyes. Abbreviations: BD: Bechet disease, CI: confidence interval, FAZ: foveal avascular zone, N: non-ocular BD, RPC: radial peripapillary capillary, SD: standard deviation, VD: vessel density.
3.5.10. Choriocapillaris flow area (non-ocular BD vs. HC).
Three studies compared choriocapillaris flow area (1 mm) between non-ocular BD and HC subgroups utilizing the RTVue XR Avanti device [35, 45, 48]. Meta-analysis (100 non-ocular BD and 113 HC eyes) demonstrated no significant difference between the two groups (Hedges g = 0.13, CI = [−0.58 to 0.32], I2 = 60.84%, P value = 0.58, Corrected P value = 0.75) (Fig 8B). Four studies compared choriocapillaris flow area (3 mm) between non-ocular BD and HC subgroups utilizing the RTVue XR Avanti device [34, 35, 45, 48]. Meta-analysis (142 non-ocular BD and 153 HC eyes) demonstrated no significant difference between the two groups (Hedges g = 0.10, CI = [−0.51 to 0.70], I2 = 84.75%, P value = 0.75, Corrected P value = 0.75) (Fig 8C).
3.5.11. FAZ (non-ocular BD vs. HC).
Ten studies compared FAZ between Non-ocular BD and HC groups using RTVue XR Avanti [35–37, 40, 45, 47–49] and Heidelberg [28, 42] devices. Pooled results (398 non-ocular BD and 381 HC eyes) demonstrated a trend toward significant difference in the non-ocular BD group compared to HCs (Hedges g = 0.82, CI= [0.01 to 1.63], I2 = 96.12%, P value = 0.05, Corrected P value = 0.06) (Fig 8D).
3.5.12. FAZ superficial (non-ocular BD vs. HC).
Six studies made a comparison for FAZ superficial among Non-ocular BD and HC groups using RTVue XR Avanti [34, 35, 38, 45, 53] and Heidelberg [28] devices. Meta-analysis of the six studies (188 non-ocular BD and 196 HC eyes) showed a significant increase in the non-ocular BD group compared to HCs (Hedges g = 0.46, CI= [0.03 to 0.90], I2 = 77.26%, P value = 0.04, Corrected P value = 0.04) (Fig 8E).
3.5.13. FAZ deep (non-ocular BD vs. HC).
Three studies using RTVue XR Avanti addressed the comparison of FAZ deep between Non-ocular BD and HC groups [34, 39, 53]. Meta-analysis (124 non-ocular BD and 124 HC eyes) revealed demonstrated no significant difference between the groups (Hedges g = 0.45, CI= [−0.11 to 1.02], I2 = 80.02%, P value = 0.12, Corrected P value = 0.12) (Fig 8F).
3.6. Non-specified ocular BD vs. HC
3.6.1. Foveal SRCP (non-specified ocular BD vs. HC).
Two studies compared SRCP foveal VD between non-specified ocular BD and HC subgroups applying the RTVue XR Avanti device [40, 49]. Meta-analysis (69 non-specified ocular BD and 73 HC eyes) showed no significant difference in the metric between the two groups (Hedges g = −0.12, CI= [−0.45 to 0.21], I2 = 0%, P value = 0.46, Corrected P value = 0.46) (Fig 9A).
A: Foveal SRCP VD in non-specified BD and HC eyes, B: Parafoveal SRCP VD in non-specified BD and HC eyes, C: Perifoveal SRCP VD in non-specified BD and HC eyes, D: Foveal DRCP VD in non-specified BD and HC eyes, E: Parafoveal DRCP VD in non-specified BD and HC eyes, F: Perifoveal DRCP VD in non-specified BD and HC eyes, G: FAZ in non-specified BD and HC eyes. Abbreviations: BD: Bechet disease, HC: Healthy Controls, CI: confidence interval, FAZ: foveal avascular zone, SD: standard deviation, SRCP: superficial retinal capillary plexus, DRCP: deep retinal capillary plexus, VD: vessel density.
3.6.2. Parafoveal SRCP (non-specified ocular BD vs. HC).
Three studies compared SRCP parafoveal VD between non-specified ocular BD and HC subgroups applying the RTVue XR Avanti device [37, 40, 49]. Meta-analysis (112 non-specified ocular BD and 116 HC eyes) showed a significant reduction in the metric in the non-specified ocular BD group (Hedges g = −0.75, CI= [-.1.153 to −0.34], I2 = 56.01%, P value = 0.003, Corrected P value = 0.004) (Fig 9B).
3.6.3. Perifoveal SRCP (non-specified ocular BD vs. HC).
Two studies compared SRCP perifoveal VD between non-specified ocular BD and HC subgroups applying the RTVue XR Avanti device [40, 49]. Meta-analysis (69 non-specified ocular BD and 73 HC eyes) showed a significant reduction in the metric in the non-specified ocular BD group (Hedges g = −0.94, CI= [−1.29 to −0.59], I2 = 25.44%, P value = 0.00001, Corrected P value = 0.00002) (Fig 9C).
3.6.4. Foveal DRCP (non-specified ocular BD vs. HC).
Two studies compared DRCP foveal VD between non-specified ocular BD and HC subgroups applying the RTVue XR Avanti device [40, 49]. Meta-analysis (69 non-specified ocular BD and 73 HC eyes) revealed a significant difference between the two groups in favor of the HC group (Hedges g = −0.44, CI= [−0.78 to −0.11], I2 = 0.00%, P value = 0.008, Corrected P value = 0.012) (Fig 9D).
3.6.5. Parafoveal DRCP (non-specified ocular BD vs. HC).
Three studies compared DRCP parafoveal VD between non-specified ocular BD and HC subgroups applying the RTVue XR Avanti device [37, 40, 49]. Meta-analysis (112 non-specified ocular BD and 116 HC eyes) revealed a significant difference between the two groups in favor of the HC group (Hedges g = −0.81, CI= [−1.07 to −0.54], I2 = 0.00%, P value = 0.00001, Corrected P value = 0.000015) (Fig 9E).
3.6.6. Perifoveal DRCP (non-specified ocular BD vs. HC).
Two studies compared FAZ between non-specified ocular BD and HC subgroups applying the RTVue XR Avanti device [40, 49]. Meta-analysis (69 non-specified ocular BD and 73 HC eyes) revealed no significant difference between the two groups in favor of the HC group (Hedges g = 0.38, CI= [−0.08 to 0.84], I2 = 44.64%, P value = 0.1, Corrected P value = 0.1) (Fig 9F).
3.6.7. FAZ (non-specified ocular BD vs. HC).
Two studies compared DRCP perifoveal VD between non-specified ocular BD and HC subgroups applying the RTVue XR Avanti device [40, 49]. Meta-analysis (69 non-specified ocular BD and 73 HC eyes) revealed a significant difference between the two groups in favor of the HC group (Hedges g = −0.92, CI= [−1.62 to −0.21], I2 = 72.64%, P value = 0.01, Corrected P value = 0.04) (Fig 9G).
3.7. Meta-regression
We investigated the effect of age, disease duration, BCVA, and IOP on all eligible analyses using meta-regression. However, only analyses regarding non-ocular BD and HC yielded significant associations; hence, in the following, we showcased the significant associations in this regard. SRCP parafoveal meta-analysis showed a significant positive association with disease duration (Coefficients: 0.08, P value = 0.01). DRCP whole (6*6) had a significant positive relation with the IOP of patients (Coefficients: 0.51, P value = 0.008). Significantly, foveal DRCP was negatively influenced by BCVA of controls (Coefficients: −0.82, P value = 0.03), and IOP of patients (Coefficients: 0.73, P value = 0.01). Parafoveal DRCP meta-analysis significantly negatively affected the BCVA of the control (Coefficients: −1.32, P value = 0.008). The meta-analysis of RPC VD was affected positively by non-ocular BD IOP (Coefficients: 0.48, P value <0.001). Choriocapillaris at 1 mm diameter revealed significant associations with both patient and control age (Coefficients: 0.02, 0.02, P values = 0.03). Moreover, disease duration had significant positive impacts on the analysis of both 1 mm and 3 mm choriocapillaris (Coefficients: 0.1, 0.17, respectively. P values = 0.03, 0.003, respectively). Despite insignificant results for the IOP of controls, the case’s IOP significantly influenced the analysis of 3 mm choriocapillaris (Coefficients: 1.31, P values = 0.03). FAZ size showed associations with several factors. Significant negative relationships were found for FAZ analyses with IOP of subjects and controls (Coefficients: −1.9, −1.5, respectively. P values: < 0.001, 0.02, respectively). However, the factor exhibited positive associations when conducting a meta-analysis for the deep portion of FAZ (Coefficients: 0.26, 0.19, respectively. P values: 0.002, 0.001, respectively). Besides, deep FAZ was influenced by other factors such as the age of subjects and controls (Coefficients: 0.08, 0.05, respectively. P values: 0.01, 0.004, respectively) and strong negative associations between deep FAZ analyses and BCVA metrics of patients and HC (Coefficients: −8.5, −7.13, respectively. P values: 0.002, 0.002, respectively).
3.8. Risk of bias assessment
The studies were evaluated for the risk of bias according to the NOS checklist. Articles were evaluated in three main domains, including “selection,” “comparability,” and “exposure”. Seven studies scored 9/9 [28,35,39,42,43,48,56], six scored 8/9 [25,30,32,47,49,58], and six scored 7/9 [29,37,40,45,46,50], four scored 6/9 [27,34,38, 53], and five scored 5/9 [26, 31, 33, 36, 44]. Table S4 in S1 File)details the quality assessment of the studies.
4. Discussion
Investigating the novel OCTA modality, the present systematic review and meta-analysis incorporated 28 studies, including a total of (769 BD subjects, 123 active BU eyes, 462 inactive BU eyes, 112 non-specified ocular BD, and 486 non-ocular BD eyes). The meta-analysis revealed that patients with inactive BU and non-ocular BD exhibited significantly larger FAZ sizes and lower VD in both the superficial and deep retinal capillary plexuses (SRCP and DRCP) compared to HCs, particularly in the parafoveal sector. Additionally, RPC VD was lower in inactive BU patients than in HCs. However, no significant differences in OCTA parameters were found in patients with active BU compared to HCs, nor were there significant differences in choriocapillaris flow among the groups.
One of the explanations for the latter could rely on the nature of the underlying inflamed vessel in the active phase. It seems that inflammatory processes compromise the integrity of the vessel wall, leading to plasma leakage. This, in turn, decreases blood flow to levels undetectable by OCTA [59]. Apart from plasma leakage, given the smaller capillary diameter, low-velocity blood flow in the lumen falls below the detectable range of OCTA when vasculitis occurs [28]. Another reason could be simply due to the limitations of the imaging itself. OCTA image quality would be affected by media opacity, for instance, in the presence of cells and flare in the anterior chamber or vitreous [60], which raises another hypothesis for the mentioned insignificancy.
The present study confirmed that the inactive state of BU exhibits reduced VDs, which is reasonable considering BD is a necrotizing occlusive vasculitis affecting arteries and veins. In this accordance, whether inactive BU eyes were the fellow eyes during unilateral uveitis or the resolved previously active uveitis, reduced VDs were inevitable [44]. Consistent with meta-analysis results, inactive BU eyes showed a lower RPC VD compared to the control group [29, 31], despite insignificant differences observed in the non-ocular form of the disease [45, 47, 48]. RPC VD is prone to ischemia and was shown to be impaired in BU, highlighting the disease’s impact on optic nerve head circulation, which is supported by a study that showed a decrease in RPC VD contributed to poorer visual acuity [29]. In fact, the RPC network contains high metabolic demand for unmyelinated nerves, making it more susceptible to ischemia in vascular compromise conditions such as Behcet vasculitis [14].
This study postulated that BD, in the form of clinically unaffected eyes, can still induce microvascular impairment of the retina. Given that Behcet is a multisystem inflammatory disease with features of vascular obliteration [38], and due to the presence of inflammatory mediators such as C-reactive protein and erythrocyte sedimentation rate (ESR), immune complexes, and endothelial cell damage, some level of decrease in VD in retinal vascular layers has been observed in non-ocular BU [25,31,44,61,62]. Further evidence for microvascular involvement in non-ocular BD comes from the correlation between BD duration and subclinical worsening of retinal microvasculature in non-ocular BD patients using OCTA [28]. This supports the notion that BD duration itself induces microvascular alterations independent of uveitis attacks. With that in mind, an interesting extrapolation made by Smid et al. indicated that future longitudinal investigations should insist on whether non-ocular BD with detected impaired microvasculature will become more vulnerable to uveitis attacks in the future [28].
OCTA enables investigators to evaluate FAZ areas easily, in contrast to the conventional FA method in which, due to dye leakage, both eyes could not be studied simultaneously [26]. In line with the meta-analysis result, FAZ enlargement in BD, regardless of ocular involvement, is well documented in the literature [24, 25, 27, 28, 35]. Conversely, some investigations found no difference in FAZ size between BD and healthy subjects [34, 37, 38]. Several reasons could explain this inconsistency in the literature. Firstly, between-subject variability of FAZ size is considerably high among healthy eyes, and it also varies with age, sex, and ethnicity [34, 37, 38, 63–65]. Another bias may be considering the fellow clinically healthy eyes as the non-ocular BD while the contralateral eye is involved in BU, as in the Koca et al. study [37]. Since potential circulatory inflammatory factors affect the microvasculature of the fellow eyes, it is reasonable not to allocate the fellow eyes of unilateral BU as entirely healthy ones. Altogether, this observation underscores the significance of FAZ enlargement as a marker of retinal ischemia in BD, advocating for future studies.
The present study has several limitations. 1) The observed results could be classified by various Behcet’s treatment options; although treatment status was not well stated in the included studies. 2) Device-related limitations, such as possible segmentation errors and the inability to detect low flow, could also affect the outcomes. 3) Another limitation relies on the definitions of active and inactive BU. The differentiation might not be accurate as residual retinal vascular leakage may be present if the remission phase is only judged clinically. 4) The exact number of uveitis attacks and disease duration were ambiguous in studies, which can potentially affect the comparison of the studies.
5. Conclusions
Taken together, this meta-analysis highlights microvascular alterations in various retinal regions and different stages of BD. While our findings did not demonstrate a significant reduction in vessel density in active BD cases, this may be influenced by factors such as prior medical therapy, the varying severity of disease activity, and limitations in OCTA imaging due to poor visualization in severe cases. Future research should aim to clarify these confounding variables and explore whether other microvascular parameters may serve as more sensitive markers of disease activity.
Interestingly, the non-ocular form of BD exhibited retinal microvascular impairment in OCTA studies, suggesting that retinal screening could be beneficial for all BD patients, even in the absence of clinical ocular involvement. However, the clinical relevance of such screening remains uncertain, particularly in terms of management implications. Further longitudinal studies are required to assess the progression of vascular impairment over time and determine whether early detection through OCTA has practical consequences for patient care.
Supporting information
S1 File. Tables S1–S6 and Supplemental Results are compiled in this single. PDF.
https://doi.org/10.1371/journal.pone.0323192.s001
(PDF)
S2 File. The data supporting the findings is available in the compressed supplemental file.
https://doi.org/10.1371/journal.pone.0323192.s002
(RAR)
S3 File. PRISMA Guideline for Reporting Systematic Reviews.
https://doi.org/10.1371/journal.pone.0323192.s003
(DOCX)
S4 File. PRISMA Guideline for Reporting Abstract of Systematic Reviews.
https://doi.org/10.1371/journal.pone.0323192.s004
(DOCX)
References
- 1. Muruganandam M, Rolle NA, Sibbitt WL Jr, Cook GB, Emil NS, Fangtham M, et al. Characteristics of Behcet’s Disease in the American Southwest. Semin Arthritis Rheum. 2019;49(2):296–302. pmid:30952423
- 2. He Y, Wang C, Su G, Deng B, Ye Z, Huang Y, et al. Decreased expression of A20 is associated with ocular Behcet’s disease (BD) but not with Vogt-Koyanagi-Harada (VKH) disease. Br J Ophthalmol. 2018;102(8):1167–72. pmid:29699987
- 3. Tong B, Liu X, Xiao J, Su G. Immunopathogenesis of Behcet’s disease. Front Immunol. 2019;10:665.
- 4. Hatemi G, Seyahi E, Fresko I, Talarico R, Hamuryudan V. One year in review 2019: Behçet’s syndrome. Clin Exp Rheumatol. 2019;37 Suppl 121(6):3–17. pmid:31856939
- 5. Mattioli I, Bettiol A, Saruhan-Direskeneli G, Direskeneli H, Emmi G, et al. Pathogenesis of Behçet’s syndrome: genetic, environmental and immunological factors. Front Med (Lausanne). 2021;8:713052.
- 6. Tugal-Tutkun I, Onal S, Altan-Yaycioglu R, Altunbas HH, Urgancioglu M, et al. Uveitis in Behçet disease: an analysis of 880 patients. Am J Ophthalmol, 2004;138(3):373–80.
- 7. Yazici H, Tüzün Y, Pazarli H, Yurdakul S, Ozyazgan Y, Ozdoğan H, et al. Influence of age of onset and patient’s sex on the prevalence and severity of manifestations of Behçet’s syndrome. Ann Rheum Dis. 1984;43(6):783–9. pmid:6524980
- 8. Mishima S, Masuda K, Izawa Y, Mochizuki M, Namba K. The eighth Frederick H. Verhoeff Lecture. presented by saiichi mishima, MD Behçet’s disease in Japan: ophthalmologic aspects. Trans Am Ophthalmol Soc. 1979;77:225–79. pmid:397657
- 9. Deuter CME, Kötter I, Wallace GR, Murray PI, Stübiger N, Zierhut M. Behçet’s disease: ocular effects and treatment. Prog Retin Eye Res. 2008;27(1):111–36. pmid:18035584
- 10. Özdal PÇ. Posterior Segment Involvement in Ocular Behçet’s Disease. European Journal of Ophthalmology. 2002;12(5):424–31.
- 11. Atmaca LS, Sonmez PA. Fluorescein and indocyanine green angiography findings in Behçet’s disease. Br J Ophthalmol. 2003;87(12):1466–8. pmid:14660454
- 12. Kornblau IS, El-Annan JF. Adverse reactions to fluorescein angiography: A comprehensive review of the literature. Surv Ophthalmol. 2019;64(5):679–93. pmid:30772364
- 13. Spaide RF, Curcio CA. Evaluation of Segmentation of the Superficial and Deep Vascular Layers of the Retina by Optical Coherence Tomography Angiography Instruments in Normal Eyes. JAMA Ophthalmol. 2017;135(3):259–62. pmid:28097291
- 14. Spaide RF, Fujimoto JG, Waheed NK, Sadda SR, Staurenghi G. Optical coherence tomography angiography. Prog Retin Eye Res. 2018;64:1–55. pmid:29229445
- 15. Durbin MK, An L, Shemonski ND, Soares M, Santos T, Lopes M, et al. Quantification of Retinal Microvascular Density in Optical Coherence Tomographic Angiography Images in Diabetic Retinopathy. JAMA Ophthalmol. 2017;135(4):370–6. pmid:28301651
- 16. Al-Sheikh M, Akil H, Pfau M, Sadda SR. Swept-Source OCT Angiography Imaging of the Foveal Avascular Zone and Macular Capillary Network Density in Diabetic Retinopathy. Invest Ophthalmol Vis Sci. 2016;57(8):3907–13. pmid:27472076
- 17. de Carlo TE, Bonini Filho MA, Chin AT, Adhi M, Ferrara D, Baumal CR, et al. Spectral-domain optical coherence tomography angiography of choroidal neovascularization. Ophthalmology. 2015;122(6):1228–38. pmid:25795476
- 18. Fard MA, Ghahvechian H, Sahrayan A, Subramanian PS. Early Macular Vessel Density Loss in Acute Ischemic Optic Neuropathy Compared to Papilledema: Implications for Pathogenesis. Transl Vis Sci Technol. 2018;7(5):10. pmid:30271677
- 19. Fekrazad S, Hassanzadeh G, Salehi MA, Mozafar M, Farahani MS, Arevalo JF. et al. Optical coherence tomography angiography measurements in systemic lupus erythematosus: A systematic review and meta-analysis. Surv Ophthalmol. 2024.
- 20. Amanollahi M, Mozafar M, Rezaei S, Rafati A, Ashourizadeh H, Moheb N, et al. Optical coherence tomography angiography measurements in neuromyelitis optica spectrum disorder and myelin oligodendrocyte glycoprotein antibody disease: A systematic review and meta-analysis. Mult Scler Relat Disord. 2024;91:105864. pmid:39265270
- 21. Ji K-B, Hu Z, Zhang Q-L, Mei H-F, Xing Y-Q. Retinal microvasculature features in patients with Behcet’s disease: a systematic review and meta-analysis. Sci Rep. 2022;12(1):752. pmid:35031636
- 22. Ebrahimiadib N, Maleki A, Fadakar K, Manhapra A, Ghassemi F, Foster CS. Vascular abnormalities in uveitis. Surv Ophthalmol. 2021;66(4):653–67. pmid:33412171
- 23. Page MJ, McKenzie JE, Bossuyt PM, Boutron I, Hoffmann TC, Mulrow CD, et al. The PRISMA 2020 statement: an updated guideline for reporting systematic reviews. BMJ. 2021;372:n71. pmid:33782057
- 24. Accorinti M, Gilardi M, De Geronimo D, Iannetti L, Giannini D, Parravano M. Optical coherence tomography angiography findings in active and inactive ocular Behçet disease. Ocul Immunol Inflamm. 2020;28(4):589–600.
- 25. Cheng D, Shen M, Zhuang X, Lin D, Dai M, Chen S, et al. Inner Retinal Microvasculature Damage Correlates With Outer Retinal Disruption During Remission in Behçet’s Posterior Uveitis by Optical Coherence Tomography Angiography. Invest Ophthalmol Vis Sci. 2018;59(3):1295–304. pmid:29625452
- 26. Emre S, Güven-Yılmaz S, Ulusoy MO, Ateş H. Optical coherence tomography angiography findings in Behcet patients. Int Ophthalmol. 2019;39(10):2391–9. pmid:30710254
- 27. Khairallah M, Abroug N, Khochtali S, Mahmoud A, Jelliti B, Coscas G, et al. Optical coherence tomography angiography in patients with behçet uveitis. Retina. 2017;37(9):1678–91. pmid:28002270
- 28. Smid LM, Vermeer KA, Missotten TOAR, van Laar JAM, van Velthoven MEJ. Parafoveal Microvascular Alterations in Ocular and Non-Ocular Behҫet’s Disease Evaluated With Optical Coherence Tomography Angiography. Invest Ophthalmol Vis Sci. 2021;62(3):8. pmid:33666648
- 29. Yan C, Li F, Hou M, Ye X, Su L, Hu Y, et al. Vascular Abnormalities in Peripapillary and Macular Regions of Behcet’s Uveitis Patients Evaluated by Optical Coherence Tomography Angiography. Front Med (Lausanne). 2021;8:727151. pmid:34604263
- 30. Aksoy FE, Basarir B, Altan C, Pasaoglu I, İnal A, Tunç U, et al. Retinal microvasculature in the remission period of Behcet’s uveitis. Photodiagnosis Photodyn Ther. 2020;29:101646. pmid:31904550
- 31. Balikoglu Yilmaz M, Aydin E, Nalcacioglu P, Doganay Kumcu N, Saritepe Imrel S, et al. Assessment of optic nerve head parameters using optical coherence tomography angiography in Behçet uveitis patients with healthy controls. Retina-Vitreus. 2020.
- 32. Eser-Ozturk H, Ismayilova L, Yucel OE, Sullu Y. Quantitative measurements with optical coherence tomography angiography in Behçet uveitis. Eur J Ophthalmol. 2021;31(3):1047–55.
- 33. Türkcü FM, Şahin A, Karaalp Ü, Çınar Y, Şahin M, Özkurt ZG, et al. Automated quantification of foveal avascular zone and vascular density in Behçet’s disease. Ir J Med Sci. 2020;189(1):349–54. pmid:31267410
- 34. Çömez A, Beyoğlu A, Karaküçük Y. Quantitative analysis of retinal microcirculation in optical coherence tomography angiography in cases with Behçet’s disease without ocular involvement. Int Ophthalmol. 2019;39(10):2213–21. pmid:30875015
- 35. Goker YS, Yılmaz S, Kızıltoprak H, Tekin K, Demir G. Quantitative Analysis of Optical Coherence Tomography Angiography Features in Patients with Nonocular Behcet’s Disease. Curr Eye Res. 2019;44(2):212–8. pmid:30261150
- 36. Karaca D, Dıraçoğlu A, Önder F. Can optical coherence tomography angiography be a first line ophthalmological evaluation in patients with Behçet’s disease?. Arch Rheumatol. 2022;38(2):200–8. pmid:37680514
- 37. Koca S, Onan D, Kalaycı D, Allı N. Comparison of Optical Coherence Tomography Angiography Findings in Patients with Behçet’s Disease and Healthy Controls. Ocul Immunol Inflamm. 2020;28(5):806–13. pmid:31414919
- 38. Raafat KA, Allam RSHM, Medhat BM. Optical coherence tomography angiography findings in patients with nonocular behçet disease. Retina. 2019;39(8):1607–12. pmid:29689026
- 39. Simsek M, Aksoy M, Ulucakoy RK. Evaluation of retinal and choroidal microcirculation in Behçet’s disease. Eye (Lond). 2022;36(7):1494–9. pmid:35115719
- 40. Yilmaz PT, Ozdemir EY, Alp MN. Optical coherence tomography angiography findings in patients with ocular and non-ocular Behcet disease. Arq Bras Oftalmol. 2021;84(3):235–40. pmid:33567020
- 41. Grading diabetic retinopathy from stereoscopic color fundus photographs--an extension of the modified Airlie House classification. ETDRS report number 10. Early Treatment Diabetic Retinopathy Study Research Group. Ophthalmology. 1991;98(5 Suppl):786–806. pmid:2062513
- 42. Ferreira BFA, Higashi AH, Prado LL, Gonçalves CR, Haddad MAO, Zacharias LC, et al. Optical coherence tomography angiography biomarkers and microperimetry features in behçet’s uveitis. Retina. 2023;43(10):1680–90. pmid:37531971
- 43. Guo S, Liu H, Gao Y, Dai L, Xu J, Yang P. Analysis of vascular changes of fundus in behcet uveitis by widefield swept source optical coherence tomography angiography and fundus fluorescein angiography. Retina. 2023;43(5):841–50. pmid:36729551
- 44. Pei M, Zhao C, Gao F, Qu Y, Liang A, Xiao J, et al. Analysis of Parafoveal Microvascular Abnormalities in Behcet’s Uveitis Using Projection-Resolved Optical Coherence Tomographic Angiography. Ocul Immunol Inflamm. 2021;29(3):524–9. pmid:31743043
- 45. Yılmaz Tuğan B, Sönmez HE. Optical coherence tomography angiography of subclinical ocular features in pediatric Behçet disease. J AAPOS. 2022;26(1):24.e1-24.e6. pmid:35032652
- 46. Mostafa A, El-shayed AN, El-shehawy AH, Mayada M, Ahmed RA, Mohamed AE. OCTA Biomarkers in Sight-Threatening Behcet’s Disease-Related Uveitis Treated with Adalimumab. The Medical Journal of Cairo University. 2022;90(9):1889–96.
- 47. Karalezli A, Kaderli ST, Sul S, Pektas SD. Preclinical ocular features in patients with Behçet’s disease detected by optical coherence tomography angiography. Eye (Lond). 2021;35(10):2719–26. pmid:33235340
- 48. Küçük MF, Yaprak L, Erol MK, Ayan A, Kök M. Quantitative changes in peripapillary, macular, and choriocapillaris microvasculature of patients with non-ocular Behçet’s disease and relationship with systemic vascular involvement, an optical coherence tomography angiography study. Photodiagnosis Photodyn Ther. 2022;38:102749. pmid:35123013
- 49. Kianersi F, Bazvand M, Fatemi A, Naderi Beni A, Kianersi H. Comparative analysis of optical coherence tomography angiography (OCTA) results between Behçet’s disease patients and a healthy control group. Clin Rheumatol. 2024;43(3):1155–70. pmid:38285372
- 50. Shen J, Qin G, Feng L. Lipid profile alterations in non-infectious uveitis: correlation with quantitative optical coherence tomography angiography parameters. BMJ Open Ophthalmol. 2024;9(1):e001641. pmid:38754892
- 51. Chu Z, Gregori G, Rosenfeld PJ, Wang RK. Quantification of Choriocapillaris with Optical Coherence Tomography Angiography: A Comparison Study. Am J Ophthalmol. 2019;208:111–23. pmid:31323202
- 52. de Oliveira BMR, Nakayama LF, de Godoy BR, de Azevedo AGB, Hirai FE, Mitne S. Reliability of foveal avascular zone measurements in eyes with retinal vein occlusion using optical coherence tomography angiography. Int J Retina Vitreous. 2020;6:35. pmid:32774887
- 53. Degirmenci MFK, Yalcindag N, Temel E. Quantitative evaluation of vascular parameters with optical coherence tomography angiography (OCTA) in patients with Behcet disease. Clinical and Experimental Rheumatology. 2018;36(6):S168–S168.
- 54. Anvari P, Najafi A, Mirshahi R, Sardarinia M, Ashrafkhorasani M, Kazemi P, et al. Superficial and Deep Foveal Avascular Zone Area Measurement in Healthy Subjects Using Two Different Spectral Domain Optical Coherence Tomography Angiography Devices. J Ophthalmic Vis Res. 2020;15(4):517–23. pmid:33133443
- 55. Stang A. Critical evaluation of the Newcastle-Ottawa scale for the assessment of the quality of nonrandomized studies in meta-analyses. Eur J Epidemiol. 2010;25(9):603–5.
- 56. Dai L, Huang F, Jiang Q, Guo S, Tan S, Su G, et al. Sensitive optical coherence tomography angiography parameters detecting retinal vascular changes in Behcet’s uveitis. Photodiagnosis and Photodynamic Therapy. 2024;:104353.
- 57. International Study Group for Behçet’s Disease. Criteria for diagnosis of Behçet’s disease. Lancet. 1990;335(8697):1078–80.
- 58. Abucham-Neto JZ, Torricelli AAM, Lui ACF, Guimarães SN, Nascimento H, Regatieri CV. Comparison between optical coherence tomography angiography and fluorescein angiography findings in retinal vasculitis. Int J Retina Vitreous. 2018;4:15. pmid:29692926
- 59. Pichi F, Sarraf D, Morara M, Mazumdar S, Neri P, Gupta V. Pearls and pitfalls of optical coherence tomography angiography in the multimodal evaluation of uveitis. J Ophthalmic Inflamm Infect. 2017;7(1):20. pmid:28983872
- 60. Zhang J, Tang FY, Cheung CY, Chen H. Different Effect of Media Opacity on Vessel Density Measured by Different Optical Coherence Tomography Angiography Algorithms. Transl Vis Sci Technol. 2020;9(8):19. pmid:32855866
- 61. Mullaney J, Collum LM. Ocular vasculitis in Behçet’s disease. A pathological and immunohistochemical study. Int Ophthalmol. 1985;7(3–4):183–91.
- 62. Mesquida M, Molins B, Llorenç V, Sainz de la Maza M, Hernandez MV, Espinosa G, et al. Proinflammatory cytokines and C-reactive protein in uveitis associated with Behçet’s disease. Mediators Inflamm. 2014;2014:396204. pmid:24994946
- 63. Carpineto P, Mastropasqua R, Marchini G, Toto L, Di Nicola M, Di Antonio L. Reproducibility and repeatability of foveal avascular zone measurements in healthy subjects by optical coherence tomography angiography. Br J Ophthalmol. 2016;100(5):671–6. pmid:26377414
- 64. Lupidi M, Coscas F, Cagini C, Fiore T, Spaccini E, Fruttini D, et al. Automated quantitative analysis of retinal microvasculature in normal eyes on optical coherence tomography angiography. Am J Ophthalmol. 2016;169:9–23.
- 65. Samara WA, Say EAT, Khoo CTL, Higgins TP, Magrath G, Ferenczy S, et al. Correlation of foveal avascular zone size with foveal morphology in normal eyes using optical coherence tomography angiography. Retina. 2015;35(11):2188–95. pmid:26469536