Accepted Manuscript Morphological and Functional Retinal Vessel Changes in Branch Retinal Vein Occlusion: An Optical Coherence Tomography Angiography Study Yuto Iida, Yuki Muraoka, Sotaro Ooto, Kiyoshi Suzuma, Tomoaki Murakami, Yuko Miwa, Rima Ghashut, Akitaka Tsujikawa PII:
S0002-9394(17)30338-0
DOI:
10.1016/j.ajo.2017.08.004
Reference:
AJOPHT 10229
To appear in:
American Journal of Ophthalmology
Received Date: 17 May 2017 Revised Date:
27 July 2017
Accepted Date: 9 August 2017
Please cite this article as: Iida Y, Muraoka Y, Ooto S, Suzuma K, Murakami T, Miwa Y, Ghashut R, Tsujikawa A, Morphological and Functional Retinal Vessel Changes in Branch Retinal Vein Occlusion: An Optical Coherence Tomography Angiography Study, American Journal of Ophthalmology (2017), doi: 10.1016/j.ajo.2017.08.004. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT
Purpose: The present study aimed to investigate the morphology of the retinal vasculature in eyes with branch retinal vein occlusion (BRVO) using optical
Design: Observational case series.
RI PT
coherence tomography (OCT) angiography.
Methods: Fluorescein angiography (FA) and OCT angiography were used to
examine macular area and the retinal vasculature at the affected arteriovenous
SC
(AV) crossing in 46 patients with BRVO.
Results: FA revealed that the affected AV crossing pattern involved arterial
M AN U
overcrossing in 23 eyes (50.0%) and venous overcrossing in 11 eyes (23.9%). However, FA failed to detect the crossing pattern in 10 eyes (21.7%). OCT angiography was significantly more effective for identification of the AV crossing pattern than FA (44 eyes; 95.7%; P= .013). The number of eyes with venous
TE D
overcrossing detected via OCT angiography (20 eyes, 43.5%) was also higher than that detected via FA (P= .047). OCT angiography revealed that venous narrowing (25.5±21.1 µm) was significantly greater in instances of venous
EP
overcrossing than in those of arterial overcrossing (46.4±23.7 µm, P= .005). Macular nonperfusion areas (NPAs) were larger in eyes with venous
AC C
overcrossing than in those with arterial overcrossing (P= .011 for superficial plexus, P=.049 for deep plexus). The peripheral NPA was significantly larger in eyes with venous overcrossing (65.1 ± 35.3 disc area (DA)) than in those with arterial overcrossing (17.2 ± 24.1 DA; P< .001). Conclusions: Our findings suggest that BRVO characterized by venous overcrossing may be more prevalent than previously reported, and that there is a significant association between NPA size and AV crossing pattern.
ACCEPTED MANUSCRIPT
Morphological and Functional Retinal Vessel Changes in Branch Retinal Vein Occlusion: An Optical Coherence Tomography Angiography Study
RI PT
Yuto Iida, Yuki Muraoka, Sotaro Ooto, Kiyoshi Suzuma, Tomoaki Murakami, Yuko Miwa, Rima Ghashut, Akitaka Tsujikawa
Department of Ophthalmology and Visual Sciences, Kyoto University Graduate
SC
School of Medicine, 54 Shogoin Kawahara-cho, Sakyo-ku, Kyoto, Japan
M AN U
Short title: Retinal vessel changes in branch retinal vein occlusion Address correspondence to: Yuki Muraoka, MD, PhD,
Department of Ophthalmology, Kyoto University Graduate School of Medicine, 54 Shogoin Kawahara-cho, Sakyo-ku, Kyoto 606-8507, Japan. E-mail: [email protected] Tel: +81-75-751-3248
AC C
EP
TE D
Fax: +81-75-752-0933
1
ACCEPTED MANUSCRIPT
Branch retinal vein occlusion (BRVO) is the second most frequent major disease of the retinal vasculature.1-3 Since the advent of anti-vascular endothelial growth factor (VEGF) agents for the treatment of macular edema (ME), the visual prognosis of
RI PT
eyes with BRVO has significantly improved.4-7 However, several randomized clinical trials have reported that most patients require several injections of anti-VEGF
agents due to recurrence of ME.5,8-10 In contrast, a systematic review of more than 1,500 eyes with BRVO revealed that ME resolved in approximately 20-40% of patients over the natural course of one year.11
SC
BRVO-associated retinal nonperfusion can lead to severe visual impairment due to a marked reduction in retinal sensitivity,12-14 as well as vitreous hemorrhage
M AN U
due to retinal neovascularization.15,16 In such severe cases, large retinal nonperfusion areas (NPAs) have been observed using fluorescein angiography (FA), although limited NPAs have also been observed in mild cases. However, it remains unclear why the severity of these retinal vascular pathologies varies among patients.
BRVO, which occurs most frequently at an arteriovenous (AV) crossing, is defined as focal occlusion of a first- or second-order retinal vein.17-22 Clinical studies
TE D
have suggested that mechanical compression of rigid arterial walls contributes to disruptions in venous flow.23-26 However, subsequent observational studies utilizing FA have indicated that most venous lesions occurred downstream, rather than upstream, of the affected AV crossing, suggesting that hemodynamic changes play a critical role in the development of BRVO.27,28
EP
Using optical coherence tomography (OCT), Muraoka and associates visualized the three-dimensional retinal vasculature of BRVO-associated AV
AC C
crossings on sequential thin sections, observing that the venous lumen was preserved in arterial overcrossings. However, in venous overcrossings, the retinal vein was compressed between the arterial wall and the internal limiting membrane (ILM).29 The authors speculated that the mechanism underlying BRVO may differ depending on the relative anatomical positions of the crossing vessels.29 However, research regarding the pathophysiological characteristics of AV crossings in patients with BRVO is limited. Recent advances in OCT angiography have enabled researchers to observe capillary changes in each retinal layer, including BRVO-associated capillary dropout 2
ACCEPTED MANUSCRIPT
and collateral formation.30-34 Additionally, OCT angiography allows for the monitoring of longitudinal changes in the retinal vasculature, as the procedure does not require injections of dye and is relatively easy to perform.33 In the present study,
RI PT
we utilized OCT angiography to investigate morphological and functional
characteristics in the affected AV crossing site in patients with BRVO in order to
examine the association between such characteristics and BRVO pathophysiology. PATIENTS AND METHODS
SC
・PATIENTS： ： The present observational study was approved by the Institutional Review Board of
M AN U
Kyoto University Graduate School of Medicine (Kyoto, Japan) and adhered to the tenets outlined in the Declaration of Helsinki. Written informed consent was obtained from each participant during the initial visit prior to the initiation of the study.
The present study included patients with major BRVO involving the temporal sector and the macular area. However, patients with macular vein occlusion or BRVO in which the occluded site was located within the optic disc or on the disc
TE D
margin were excluded. Additionally, eyes with hemi-central retinal vein occlusion, multiple occlusions of the retinal veins, any interventions for ME prior to the study period, and co-existing ocular diseases (diabetic retinopathy, retinal arterial occlusion, retinal macroaneurysm, glaucoma, or epiretinal membrane) were excluded, as were eyes with keratoconus, high myopia (more severe than -6
EP
diopters), or high astigmatism (more severe than ± 3 diopters). Eyes for which poor-quality OCT angiograms (signal strength index < 50) had been obtained due to
AC C
eye movement or media opacities were also excluded. A final total of 46 patients with BRVO who had visited the Department of Ophthalmology at Kyoto University Hospital between February 2015 and December 2016 met the aforementioned criteria for eligibility.
At the initial visit, patients underwent a comprehensive ophthalmic examination,
which included measurement of best-corrected visual acuity (BCVA) using a Landolt chart and 45º digital fundus photography (TRC-50LX, Topcon, Tokyo, Japan; 3,216 × 2,136 pixels). Following pupil dilation, FA was performed in each patient using an Optos 200Tx imaging system (Optos PLC, Dunfermline, UK). 3
ACCEPTED MANUSCRIPT
Patients received treatment for ME via monthly injections of intravitreal ranibizumab (IVR) in the affected eye and underwent follow-up examination at our clinic each month. Initially, patients received three such injections, although
RI PT
additional injections were provided when ME and/or serous retinal detachment was evident at the fovea on OCT sections. No eyes received any other form of treatment, such as scatter or grid laser photocoagulation, steroid therapy, surgical intervention, or intravitreal injections of anti-VEGF agents other than ranibizumab. Morphological and functional alterations in the retinal microvasculature of the macular area and
SC
the affected AV crossing were examined using OCT angiography (RTVue XR
Avanti, Optovue Inc., Fremont, CA) at every visit. FA was reperformed 1 year after
M AN U
the start of ranibizumab treatment, or when fresh retinal hemorrhage, peripheral white vessels, or neovascular changes had been detected during routine ophthalmic examinations.
・OCT EVALUATION OF FOVEAL THICKNESS： ： Foveal thickness was measured using macular volume scans obtained via OCT (Spectralis HRA+OCT). A whole-retinal thickness map centered on the foveal
TE D
center was created using the Early Treatment Diabetic Retinopathy Study grid. Foveal thickness was defined as the average value calculated from the retinal thickness of the central grid. Retinal thickness calculations for each grid were performed using the manufacturer’s built-in software (Spectralis Acquisition and
EP
Viewing Modules, version 6.0, Heidelberg Engineering, Heidelberg, Germany).
AC C
・CLASSIFICATION OF AV CROSSING PATTERN AT THE AFFECTED AV CROSSING SITE: We classified the relative anatomical vessel position at the affected AV crossing into four types: arterial overcrossing, venous overcrossing, helical overcrossing, and undetermined. Arterial overcrossings were defined as those in which the affected artery coursed over the adjacent vein, while venous overcrossings were defined as those in which the affected vein coursed over the adjacent artery. Helical overcrossings were defined as those in which the affected vein twisted around the adjacent major artery. AV crossing patterns were classified using both OCT angiography and FA. FA classification was determined based on sequential images 4
ACCEPTED MANUSCRIPT
obtained between 20 seconds and 5 minutes after dye injection. OCT angiography classifications were determined based on OCT angiograms and B-scan images of the original OCT scans. For each imaging modality, classification was performed by
RI PT
two independent retinal specialists (YI and YMi). When classifications of crossing patterns differed between the raters, a senior retinal specialist (YMu) determined the final classification.
SC
・OCT ANGIOGRAPHY OF MACULAR AREA AND AFFECTED AV CROSSING: The scanning area was captured in 3 mm × 3 mm sections, centered on the foveal center and the affected AV crossing. The built-in software in RTVue XR Avanti
M AN U
automatically recognizes the layers of the ILM and inner plexiform layer (IPL). Images of the superficial capillary layer were obtained as follows: The slab containing the superficial capillary plexus extended from the ILM to 15 µm below the IPL, while that containing the deep capillary plexus extended from 15 µm below the IPL to 70 µm below the IPL, as described in previous reports.13,35,36
TE D
・MEASUREMENT OF MACULAR NPA AND VENOUS LUMINAL DIAMETER VIA OCT ANGIOGRAPHY: To avoid segmentation errors in each retinal layer, the macular NPA was measured using OCT angiography following complete resolution of ME, macular detachment, and macular hemorrhage on OCT sections. The macular NPA was defined as the capillary dropout area within a 3 mm × 3 mm section, including the foveal avascular
EP
zone. The foveal avascular zone and the parafoveal capillary dropouts (parafoveal NPA) were independently reviewed by three fully trained retina specialists (YI, YMi,
AC C
and RG) blinded to all clinical information regarding the study eyes. The OCT angiography measurements of macular NPA were performed using the manufacturer’s built-in software (AngioVue Clinical Release, v2015.100.0.3, Optovue Inc.). Measurements were then corrected for axial length-related magnification using the modified Littmann formula, in which measurements of axial length, flatter and steeper meridians, and the spherical equivalent were applied (Bennett procedure).37 OCT angiograms of the affected AV crossing were used to evaluate venous luminal diameter at the crossing site. Measurements of luminal diameter were 5
ACCEPTED MANUSCRIPT
obtained for the following three locations using ImageJ software (Wayne Rasband, National Institutes of Health, Bethesda, MD; available at http://rsb.info.nih.gov/ij/index.html): the center of the crossing site, 5 pixels proximal
RI PT
to the center of the crossing site, and 5 pixels distal from the center of the crossing site. Venous luminal diameter was calculated as the average of these three values, which was also corrected using the Littmann formula.
SC
・MEASUREMENT OF PERIPHERAL NPA: Peripheral NPA and disc area (DA) were measured on FA images obtained 60 s
after dye injection using a plugin software program in ImageJ. The peripheral NPA
M AN U
was expressed in units of DA.
・STATISTICAL ANALYSIS: Statistical analysis was performed using PASW Statistics version 18.0 (SPSS, Chicago, IL). Values are presented as the mean ± standard deviation. For statistical analysis, BCVA measured using a Landolt chart was converted to a logarithm of the minimum angle of resolution (logMAR). Comparisons between the two groups were
TE D
performed using unpaired t-tests, while longitudinal changes within each group were analyzed using paired t-tests. Significant differences in the sampling distributions were determined using chi-square tests. Interclass correlation coefficients (ICCs) were calculated to examine the reproducibility and validity of NPA measurements obtained via FA and OCT angiography. The level of statistical
EP
significance was set at P
S0002-9394(17)30338-0
DOI:
10.1016/j.ajo.2017.08.004
Reference:
AJOPHT 10229
To appear in:
American Journal of Ophthalmology
Received Date: 17 May 2017 Revised Date:
27 July 2017
Accepted Date: 9 August 2017
Please cite this article as: Iida Y, Muraoka Y, Ooto S, Suzuma K, Murakami T, Miwa Y, Ghashut R, Tsujikawa A, Morphological and Functional Retinal Vessel Changes in Branch Retinal Vein Occlusion: An Optical Coherence Tomography Angiography Study, American Journal of Ophthalmology (2017), doi: 10.1016/j.ajo.2017.08.004. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT
Purpose: The present study aimed to investigate the morphology of the retinal vasculature in eyes with branch retinal vein occlusion (BRVO) using optical
Design: Observational case series.
RI PT
coherence tomography (OCT) angiography.
Methods: Fluorescein angiography (FA) and OCT angiography were used to
examine macular area and the retinal vasculature at the affected arteriovenous
SC
(AV) crossing in 46 patients with BRVO.
Results: FA revealed that the affected AV crossing pattern involved arterial
M AN U
overcrossing in 23 eyes (50.0%) and venous overcrossing in 11 eyes (23.9%). However, FA failed to detect the crossing pattern in 10 eyes (21.7%). OCT angiography was significantly more effective for identification of the AV crossing pattern than FA (44 eyes; 95.7%; P= .013). The number of eyes with venous
TE D
overcrossing detected via OCT angiography (20 eyes, 43.5%) was also higher than that detected via FA (P= .047). OCT angiography revealed that venous narrowing (25.5±21.1 µm) was significantly greater in instances of venous
EP
overcrossing than in those of arterial overcrossing (46.4±23.7 µm, P= .005). Macular nonperfusion areas (NPAs) were larger in eyes with venous
AC C
overcrossing than in those with arterial overcrossing (P= .011 for superficial plexus, P=.049 for deep plexus). The peripheral NPA was significantly larger in eyes with venous overcrossing (65.1 ± 35.3 disc area (DA)) than in those with arterial overcrossing (17.2 ± 24.1 DA; P< .001). Conclusions: Our findings suggest that BRVO characterized by venous overcrossing may be more prevalent than previously reported, and that there is a significant association between NPA size and AV crossing pattern.
ACCEPTED MANUSCRIPT
Morphological and Functional Retinal Vessel Changes in Branch Retinal Vein Occlusion: An Optical Coherence Tomography Angiography Study
RI PT
Yuto Iida, Yuki Muraoka, Sotaro Ooto, Kiyoshi Suzuma, Tomoaki Murakami, Yuko Miwa, Rima Ghashut, Akitaka Tsujikawa
Department of Ophthalmology and Visual Sciences, Kyoto University Graduate
SC
School of Medicine, 54 Shogoin Kawahara-cho, Sakyo-ku, Kyoto, Japan
M AN U
Short title: Retinal vessel changes in branch retinal vein occlusion Address correspondence to: Yuki Muraoka, MD, PhD,
Department of Ophthalmology, Kyoto University Graduate School of Medicine, 54 Shogoin Kawahara-cho, Sakyo-ku, Kyoto 606-8507, Japan. E-mail: [email protected] Tel: +81-75-751-3248
AC C
EP
TE D
Fax: +81-75-752-0933
1
ACCEPTED MANUSCRIPT
Branch retinal vein occlusion (BRVO) is the second most frequent major disease of the retinal vasculature.1-3 Since the advent of anti-vascular endothelial growth factor (VEGF) agents for the treatment of macular edema (ME), the visual prognosis of
RI PT
eyes with BRVO has significantly improved.4-7 However, several randomized clinical trials have reported that most patients require several injections of anti-VEGF
agents due to recurrence of ME.5,8-10 In contrast, a systematic review of more than 1,500 eyes with BRVO revealed that ME resolved in approximately 20-40% of patients over the natural course of one year.11
SC
BRVO-associated retinal nonperfusion can lead to severe visual impairment due to a marked reduction in retinal sensitivity,12-14 as well as vitreous hemorrhage
M AN U
due to retinal neovascularization.15,16 In such severe cases, large retinal nonperfusion areas (NPAs) have been observed using fluorescein angiography (FA), although limited NPAs have also been observed in mild cases. However, it remains unclear why the severity of these retinal vascular pathologies varies among patients.
BRVO, which occurs most frequently at an arteriovenous (AV) crossing, is defined as focal occlusion of a first- or second-order retinal vein.17-22 Clinical studies
TE D
have suggested that mechanical compression of rigid arterial walls contributes to disruptions in venous flow.23-26 However, subsequent observational studies utilizing FA have indicated that most venous lesions occurred downstream, rather than upstream, of the affected AV crossing, suggesting that hemodynamic changes play a critical role in the development of BRVO.27,28
EP
Using optical coherence tomography (OCT), Muraoka and associates visualized the three-dimensional retinal vasculature of BRVO-associated AV
AC C
crossings on sequential thin sections, observing that the venous lumen was preserved in arterial overcrossings. However, in venous overcrossings, the retinal vein was compressed between the arterial wall and the internal limiting membrane (ILM).29 The authors speculated that the mechanism underlying BRVO may differ depending on the relative anatomical positions of the crossing vessels.29 However, research regarding the pathophysiological characteristics of AV crossings in patients with BRVO is limited. Recent advances in OCT angiography have enabled researchers to observe capillary changes in each retinal layer, including BRVO-associated capillary dropout 2
ACCEPTED MANUSCRIPT
and collateral formation.30-34 Additionally, OCT angiography allows for the monitoring of longitudinal changes in the retinal vasculature, as the procedure does not require injections of dye and is relatively easy to perform.33 In the present study,
RI PT
we utilized OCT angiography to investigate morphological and functional
characteristics in the affected AV crossing site in patients with BRVO in order to
examine the association between such characteristics and BRVO pathophysiology. PATIENTS AND METHODS
SC
・PATIENTS： ： The present observational study was approved by the Institutional Review Board of
M AN U
Kyoto University Graduate School of Medicine (Kyoto, Japan) and adhered to the tenets outlined in the Declaration of Helsinki. Written informed consent was obtained from each participant during the initial visit prior to the initiation of the study.
The present study included patients with major BRVO involving the temporal sector and the macular area. However, patients with macular vein occlusion or BRVO in which the occluded site was located within the optic disc or on the disc
TE D
margin were excluded. Additionally, eyes with hemi-central retinal vein occlusion, multiple occlusions of the retinal veins, any interventions for ME prior to the study period, and co-existing ocular diseases (diabetic retinopathy, retinal arterial occlusion, retinal macroaneurysm, glaucoma, or epiretinal membrane) were excluded, as were eyes with keratoconus, high myopia (more severe than -6
EP
diopters), or high astigmatism (more severe than ± 3 diopters). Eyes for which poor-quality OCT angiograms (signal strength index < 50) had been obtained due to
AC C
eye movement or media opacities were also excluded. A final total of 46 patients with BRVO who had visited the Department of Ophthalmology at Kyoto University Hospital between February 2015 and December 2016 met the aforementioned criteria for eligibility.
At the initial visit, patients underwent a comprehensive ophthalmic examination,
which included measurement of best-corrected visual acuity (BCVA) using a Landolt chart and 45º digital fundus photography (TRC-50LX, Topcon, Tokyo, Japan; 3,216 × 2,136 pixels). Following pupil dilation, FA was performed in each patient using an Optos 200Tx imaging system (Optos PLC, Dunfermline, UK). 3
ACCEPTED MANUSCRIPT
Patients received treatment for ME via monthly injections of intravitreal ranibizumab (IVR) in the affected eye and underwent follow-up examination at our clinic each month. Initially, patients received three such injections, although
RI PT
additional injections were provided when ME and/or serous retinal detachment was evident at the fovea on OCT sections. No eyes received any other form of treatment, such as scatter or grid laser photocoagulation, steroid therapy, surgical intervention, or intravitreal injections of anti-VEGF agents other than ranibizumab. Morphological and functional alterations in the retinal microvasculature of the macular area and
SC
the affected AV crossing were examined using OCT angiography (RTVue XR
Avanti, Optovue Inc., Fremont, CA) at every visit. FA was reperformed 1 year after
M AN U
the start of ranibizumab treatment, or when fresh retinal hemorrhage, peripheral white vessels, or neovascular changes had been detected during routine ophthalmic examinations.
・OCT EVALUATION OF FOVEAL THICKNESS： ： Foveal thickness was measured using macular volume scans obtained via OCT (Spectralis HRA+OCT). A whole-retinal thickness map centered on the foveal
TE D
center was created using the Early Treatment Diabetic Retinopathy Study grid. Foveal thickness was defined as the average value calculated from the retinal thickness of the central grid. Retinal thickness calculations for each grid were performed using the manufacturer’s built-in software (Spectralis Acquisition and
EP
Viewing Modules, version 6.0, Heidelberg Engineering, Heidelberg, Germany).
AC C
・CLASSIFICATION OF AV CROSSING PATTERN AT THE AFFECTED AV CROSSING SITE: We classified the relative anatomical vessel position at the affected AV crossing into four types: arterial overcrossing, venous overcrossing, helical overcrossing, and undetermined. Arterial overcrossings were defined as those in which the affected artery coursed over the adjacent vein, while venous overcrossings were defined as those in which the affected vein coursed over the adjacent artery. Helical overcrossings were defined as those in which the affected vein twisted around the adjacent major artery. AV crossing patterns were classified using both OCT angiography and FA. FA classification was determined based on sequential images 4
ACCEPTED MANUSCRIPT
obtained between 20 seconds and 5 minutes after dye injection. OCT angiography classifications were determined based on OCT angiograms and B-scan images of the original OCT scans. For each imaging modality, classification was performed by
RI PT
two independent retinal specialists (YI and YMi). When classifications of crossing patterns differed between the raters, a senior retinal specialist (YMu) determined the final classification.
SC
・OCT ANGIOGRAPHY OF MACULAR AREA AND AFFECTED AV CROSSING: The scanning area was captured in 3 mm × 3 mm sections, centered on the foveal center and the affected AV crossing. The built-in software in RTVue XR Avanti
M AN U
automatically recognizes the layers of the ILM and inner plexiform layer (IPL). Images of the superficial capillary layer were obtained as follows: The slab containing the superficial capillary plexus extended from the ILM to 15 µm below the IPL, while that containing the deep capillary plexus extended from 15 µm below the IPL to 70 µm below the IPL, as described in previous reports.13,35,36
TE D
・MEASUREMENT OF MACULAR NPA AND VENOUS LUMINAL DIAMETER VIA OCT ANGIOGRAPHY: To avoid segmentation errors in each retinal layer, the macular NPA was measured using OCT angiography following complete resolution of ME, macular detachment, and macular hemorrhage on OCT sections. The macular NPA was defined as the capillary dropout area within a 3 mm × 3 mm section, including the foveal avascular
EP
zone. The foveal avascular zone and the parafoveal capillary dropouts (parafoveal NPA) were independently reviewed by three fully trained retina specialists (YI, YMi,
AC C
and RG) blinded to all clinical information regarding the study eyes. The OCT angiography measurements of macular NPA were performed using the manufacturer’s built-in software (AngioVue Clinical Release, v2015.100.0.3, Optovue Inc.). Measurements were then corrected for axial length-related magnification using the modified Littmann formula, in which measurements of axial length, flatter and steeper meridians, and the spherical equivalent were applied (Bennett procedure).37 OCT angiograms of the affected AV crossing were used to evaluate venous luminal diameter at the crossing site. Measurements of luminal diameter were 5
ACCEPTED MANUSCRIPT
obtained for the following three locations using ImageJ software (Wayne Rasband, National Institutes of Health, Bethesda, MD; available at http://rsb.info.nih.gov/ij/index.html): the center of the crossing site, 5 pixels proximal
RI PT
to the center of the crossing site, and 5 pixels distal from the center of the crossing site. Venous luminal diameter was calculated as the average of these three values, which was also corrected using the Littmann formula.
SC
・MEASUREMENT OF PERIPHERAL NPA: Peripheral NPA and disc area (DA) were measured on FA images obtained 60 s
after dye injection using a plugin software program in ImageJ. The peripheral NPA
M AN U
was expressed in units of DA.
・STATISTICAL ANALYSIS: Statistical analysis was performed using PASW Statistics version 18.0 (SPSS, Chicago, IL). Values are presented as the mean ± standard deviation. For statistical analysis, BCVA measured using a Landolt chart was converted to a logarithm of the minimum angle of resolution (logMAR). Comparisons between the two groups were
TE D
performed using unpaired t-tests, while longitudinal changes within each group were analyzed using paired t-tests. Significant differences in the sampling distributions were determined using chi-square tests. Interclass correlation coefficients (ICCs) were calculated to examine the reproducibility and validity of NPA measurements obtained via FA and OCT angiography. The level of statistical
EP
significance was set at P
