- Email: [email protected]
Evaluation of Macular Vascular Abnormalities Identified by Optical Coherence Tomography Angiography in Sickle Cell Disease
Accepted Manuscript Evaluation of macular vascular abnormalities identified by optical coherence tomography angiography in sickle cell disease Ian C. Han, Mongkol Tadarati, Katia D. Pacheco, Adrienne W. Scott PII:
S0002-9394(17)30064-8
DOI:
10.1016/j.ajo.2017.02.007
Reference:
AJOPHT 10040
To appear in:
American Journal of Ophthalmology
Received Date: 30 September 2016 Revised Date:
4 February 2017
Accepted Date: 7 February 2017
Please cite this article as: Han IC, Tadarati M, Pacheco KD, Scott AW, Evaluation of macular vascular abnormalities identified by optical coherence tomography angiography in sickle cell disease, American Journal of Ophthalmology (2017), doi: 10.1016/j.ajo.2017.02.007. 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
Abstract Purpose: To evaluate macular vascular flow abnormalities identified by optical coherence tomography angiography (OCT-A) in patients with various sickle cell genotypes.
RI PT
Design: Prospective, observational case series.
SC
Methods: This is a single institution case series of adult patients with various sickle cell genotypes. All patients underwent macular OCT-A (Avanti RTVue XR, Optovue Inc, Fremont, CA). Images were analyzed qualitatively for areas of flow loss and quantitatively for measures of foveal avascular area, parafoveal flow, and vascular density. The findings were compared by sickle cell genotype and retinopathy stage and correlated to retinal thickness and visual acuity.
M AN U
Results: OCT-A scans of 82 eyes from 46 patients (60.9% female, mean age 33.5 years) were included. Sickle cell genotypes included 27 patients with SS (58.7%), 14 SC (30.4%), 4 betathalassemia (8.7%), and 1 sickle trait (2.2%). Discrete areas of flow loss were noted in 37.8% (31/82) of eyes overall and were common in both SS (40.0%, 20/50 eyes) and SC (41.7%, 10/24 eyes). Flow loss was more extensive in the temporal and nasal parafoveal subfields of the deep plexus with sickle SC or proliferative retinopathy. Retinal thickness measurements correlated with vascular density of the fovea, parafovea, temporal and superior subfields. Visual acuity correlated with foveal avascular zone area and parafoveal vascular density in the superficial and deep plexi.
AC C
EP
TE D
Conclusions: Areas of abnormal macular vascular flow are common in patients with various sickle cell genotypes. These areas may be seen at any retinopathy stage but may be more extensive with sickle SC or proliferative retinopathy.
ACCEPTED MANUSCRIPT
Title: Evaluation of macular vascular abnormalities identified by optical coherence tomography angiography in sickle cell disease Short Title: Macular OCT angiography of sickle cell retinopathy
RI PT
Authors: Ian C. Han1; Mongkol Tadarati1, Katia D. Pacheco1, Adrienne W. Scott1
M AN U
Corresponding Author: Adrienne W. Scott, MD Wilmer Eye Institute, Retina Division 600 North Wolfe Street Maumenee 719 Baltimore, MD 21287 [email protected]
SC
Author Affiliations: 1Retina Division, Wilmer Eye Institute, Johns Hopkins School of Medicine, Baltimore, Maryland
AC C
EP
TE D
Footnote for current author affiliations: ICH: Department of Ophthalmology and Visual Sciences, Carver College of Medicine, University of Iowa, Iowa City, Iowa MT: Retina Division, Department of Ophthalmology, Rajavithi Hospital, College of Medicine, Rangsit University, Bangkok, Thailand KDP: Retina Division, Department of Ophthalmology, Brazilian Center of Vision Eye Hospital, Brasilia, Brazil; Retina Division, Department of Ophthalmology, Armed Forces Hospital, Brasilia, Brazil
1 Han et al.
ACCEPTED MANUSCRIPT
Introduction
AC C
EP
TE D
M AN U
SC
RI PT
Sickle cell retinopathy is classically considered a peripheral retinopathy, with characteristic changes as described by Goldberg, including arteriolar occlusions, arteriovenous anastomoses, neovascularization, vitreous hemorrhage, or retinal detachment1,2. However, numerous macular vascular abnormalities can also be seen, including irregularities of the foveal avascular zone, microaneurysmal dots, and hairpin venular loops3. These macular vascular abnormalities are commonly noted temporal to the fovea because the temporal macular vessels are within the watershed zone along the horizontal raphe and are thus akin to terminal vessels and more susceptible to vascular occlusion3. Numerous reports have described morphologic changes in the macula using spectral domain optical coherence tomography (SD-OCT), including the presence of macular splaying and areas of inner retinal thinning, predominantly temporal to the macula4-12. While many of these patients are asymptomatic, decreased retinal sensitivity has been documented in areas of thinning5, and the presence of macular thinning has been correlated to more advanced peripheral changes7,13. Specifically, patients with temporal macular thinning on SD-OCT have higher rates of peripheral neovascularization in the same eye, suggesting that macular thinning may be a surrogate marker for peripheral retinopathy7. Although macular thinning is commonly observed on SD-OCT, these patients may not have corresponding abnormalities on fluorescein angiography (FA)10. Optical coherence tomography angiography (OCT-A) is a relatively nascent technology that utilizes subtle differences in specular patterns across temporally distinct signals to detect movement of blood cells, allowing for the digital reconstruction of retinal vasculature. Unlike FA, OCT-A does not allow dynamic evaluation of the speed of vascular filling or vascular leakage. However, OCT-A has several advantages over FA, including the lack of contrast dye, improved visualization of several vascular layers including the radial papillary network and deep capillary plexus, and the ability to optically segment the retina to analyze the various retinal vascular networks separately14. We recently reported that OCT-A can detect areas of abnormal vascular flow within the macula in patients with sickle cell disease and that these areas correspond to areas of retinal thinning seen on SD-OCT15. Areas of flow loss by OCT-A appear to affect the deep greater than superficial plexus and importantly, can be seen in patients who were clinically visually asymptomatic and have no apparent vascular abnormalities on FA. Other small case series and case reports have since supported our initial findings16-18. However, the frequency and severity of these macular vascular abnormalities and their relationship to sickle cell genotype or the severity of peripheral retinopathy have not yet been described. There are also minimal existing data correlating areas of abnormal vascular flow with retinal thickness or visual acuity in patients with sickle cell disease. This study aims to evaluate the frequency and severity of macular flow abnormalities as detected by OCT-A in patients with various sickle cell genotypes, and to compare these flow abnormalities by sickle cell genotype and stage of retinopathy. Additionally, we evaluate the correlation between retinal thickness and vascular density, as well as between visual acuity and measures of foveal avascular area and parafoveal vascular flow.
2 Han et al.
ACCEPTED MANUSCRIPT
Methods
AC C
EP
TE D
M AN U
SC
RI PT
This is a single-institution, prospective, observational case series evaluating clinical and multimodal imaging findings for consecutive patients with various sickle cell genotypes who presented to the investigators’ (ICH, AWS) retina sub-specialty clinics for eye examination. The study was approved by the Johns Hopkins Hospital Institutional Review Board and complied with regulations set forth by the Health Insurance Portability and Accountability Act. Patients with known sickle cell of any genotype and adequate view to the fundus were included via written informed consent for participation. Exclusion criteria included cataract, vitreous hemorrhage, or other media opacity prohibiting adequate image acquisition, known maculopathy from other causes including vitreomacular traction, or history of known retinal vascular disease, including diabetes, hypertension, or retinal artery or vein occlusion. All patients received Early Treatment Diabetic Retinopathy Study (ETDRS) visual acuity testing with habitual distance spectacle correction and complete ophthalmic examination including dilated fundus examination by the lead investigators (ICH, AWS), both fellowshiptrained retina specialists. For each eye, stage of retinopathy was determined based on clinical examination per Goldberg’s classification2 as 1) no retinopathy, 2) non-proliferative retinopathy (sun-burst lesions, salmon patches, arteriolar occlusions, peripheral anastomoses without neovascularization, i.e. Goldberg stage 2 and below), or 3) proliferative retinopathy (with neovascularization, i.e. Goldberg stage 3 and above). Macular SD-OCT and OCT-A (Optovue RTVue XR Avanti, Optovue Inc, Fremont, California, USA) was performed for all patients on the same day as the clinical examination. Scan protocols included 6 mm x 6 mm and 3 mm x 3 mm scans centered on the fovea. Given the propensity for retinal thinning in the temporal macula4,7,8,10, we also performed dedicated 3 mm x 3 mm scans through the temporal macula, with the nasal edge of the scan at or near the fovea (Figure 1). For each field, multiple scans were acquired as needed to ensure adequate quality for analysis. OCT-A images were then generated using the instrument’s split-spectrum amplitude decorrelation angiography (SSADA) software, including automated segmentation of the various vascular networks (superficial capillary plexus, deep capillary plexus, choriocapillaris). For multiple scans in the same area, the images were analyzed by the lead investigators (ICH, AWS), and the best quality scans selected for analysis. Images were excluded if the quality was insufficient for analysis owing to eye movement, blinking, or poor fixation during image acquisition. Images were then analyzed by the lead investigators (ICH, AWS), and eyes with areas of flow loss within either the superficial or deep plexus or both were recorded. The frequency and relative proportion of scans with identifiable flow abnormalities on OCT-A were then calculated. Automated retinal thickness measurements and vascular density data were recorded for the whole image, fovea, parafovea, and parafoveal subfields for 3 mm x 3 mm scans centered on the fovea (Figure 2). The non-flow area (mm2) within the foveal avascular zone (FAZ) and flow area within a 1 mm radius circle centered on the fovea were measured in an automated fashion using the instrument’s intrinsic software (RTVue XR version 2016.1.0.23). These measures were evaluated by the lead investigators (ICH, AWS) and excluded in eyes when pathologic disruption of the FAZ precluded the software’s ability to properly identify the boundaries of the non-flow or flow areas. Figure 3 shows an example of a pathologic FAZ, 3 Han et al.
ACCEPTED MANUSCRIPT
M AN U
SC
RI PT
where the software correctly identifies the FAZ in the superficial plexus but cannot determine the boundaries of the FAZ within the deep plexus, where a contiguous, adjacent area of flow loss extends beyond the edges of the scan. Visual acuities were converted to logarithm of minimal angle of resolution (logMAR) equivalents prior to statistical analysis. Pearson’s correlation coefficient was used to evaluate the association between visual acuity and measures of foveal flow including foveal non-flow area and flow within a 1 mm radius circle centered on the fovea. Retinal thickness measurements were compared for each parafoveal subfield using paired T-tests. Vessel density measurements for each subfield were compared in similar fashion. Pearson’s correlation coefficient was used to evaluate the association between retinal thickness measurements and vessel density within the various parafoveal subfields. Patients were then stratified by sickle cell genotype and scans for each eye stratified by stage of retinopathy. Based on pre-existing SDOCT data correlating areas of temporal thinning with proliferative retinopathy7, we selected eyes with proliferative retinopathy (Goldberg Stage 3 or above) or without as pre-set groups for sub-analysis. Chi-square tests of proportions were used to compare frequency of abnormal findings across sub-populations within the study, and two-tailed, unpaired T-tests used to compare vessel density data across the various subgroups. P-values of <0.05 were considered statistically significant for all statistical analyses, which were performed using Microsoft Office Excel 2007 (Microsoft Corp., Seattle, WA, USA) and GraphPad Prism 7 (GraphPad Software, Inc., La Jolla, CA, USA). Results
AC C
EP
TE D
Fifty-one consecutive patients with sickle cell disease were enrolled in the study. Of the 102 available eyes, 5 eyes had media opacity precluding imaging, and OCT-A scans of 97 eyes (50 right, 47 left) were obtained for the study. Of these, 82 eyes (42 right, 40 left) from 46 patients (28/46 female, 60.9%), had scans with image quality acceptable for qualitative analysis. Sickle cell genotypes included 27 patients with hemoglobin SS (58.7%), 14 with hemoglobin SC (30.4%), 4 with beta-thalassemia (8.7%), 1 with sickle trait (2.2%). The mean age was 33.5 years old (Table 1). Visual acuity ranged from 20/12 to 20/80, with the majority of eyes possessing visual acuity of 20/20 or better (63 out of 82 eyes, 76.8%). Discrete areas of foveal and parafoveal flow loss, in the superficial or deep plexus or both, were noted in 37.8% (31/82) of eyes overall. When stratified by genotype, these areas were common in eyes of both SS (40.0%, 20/50) and SC (41.7%, 10/24) patients (P=0.89). When stratified by presence of proliferative retinopathy as defined by neovascularization (Goldberg stage 3 and above), discrete areas of flow loss were seen in 53.8% of eyes (14/26) with neovascularization versus 32.1% of eyes (18/56) without (P=0.06) (Table 2). When evaluating quantitative measures of macular vascular flow, algorithm errors precluded accurate measurement of foveal non-flow area within the superficial plexus for 2 eyes (80 eyes included for analysis) and within the deep plexus for 4 eyes (78 eyes included). Similarly, for quantitative analysis of flow within a 1 mm circle centered on the fovea, algorithm errors precluded accurate measurement of the superficial plexus for 4 eyes (78 eyes included for analysis) and of the deep plexus for 7 eyes (75 eyes included). LogMAR visual acuity was positively correlated with foveal non-flow measures (P=0.001), i.e. greater non-flow area 4 Han et al.
ACCEPTED MANUSCRIPT
EP
TE D
M AN U
SC
RI PT
correlated to worse visual acuity, and negatively correlated with measures of flow within 1 mm disc centered on the fovea (P<0.001), i.e. greater parafovea flow correlated to better visual acuity (Table 3). When stratified by sickle cell genotype and retinopathy stage, the mean foveal non-flow area in both the superficial and deep plexi was greater for sickle SC eyes compared to sickle SS, and for eyes with neovascularization compared to those without. However, none of these differences reached statistical significance (Table 4). The mean flow area within a 1 mm radius circle centered on the fovea was greater for sickle SS eyes and those without neovascularization, but these differences also did not reach statistical significance (Table 4). Average retinal thickness measurements for the various subfields are shown in Table 5. When comparing retinal thickness values across the parafoveal subfields, the temporal subfield was noted to be thinner (P<0.001) relative to the superior, nasal, and inferior subfields. There were no statistically-significant differences in thickness measurements across the remaining subfields (P-values all >0.3). Vessel density within the temporal and nasal subfields were noted to be lower than the superior and inferior subfields in both the superficial and deep plexi (P<0.001 for all comparisons). When comparing the horizontal subfields, vessel density was noted to be lower in the temporal versus the nasal for both the superficial (P=0.03) and deep plexi (P=0.04). Retinal thickness was found to be correlated to vessel density in the fovea and parafovea overall (Table 5) for both the superficial and deep plexi. These measures were also strongly correlated in the temporal and superior parafoveal subfields (P<0.001) of the superior and deep plexi, whereas no statistically-significant correlation was found for the nasal and inferior subfields. When comparing vessel density measures by sickle cell genotype, the vessel density of the deep capillary plexus overall and multiple subfields (parafovea overall, and temporal and nasal parafoveal subfields) were found to be different to a statistically significant degree, with lower densities seen in sickle SC eyes compared to sickle SS eyes (Table 6). When stratified by severity of retinopathy, vessel density within these same fields was lower in eyes with neovascularization compared to those without. Within the superficial plexus, a statistically significant difference was encountered only when comparing densities of the temporal subfield in eyes with neovascularization versus those without.
Discussion
AC C
We previously reported that OCT-A can identify areas of macular vascular flow loss in eyes with sickle cell retinopathy, even in the absence of obvious abnormalities on fluorescein angiography15. These areas of flow loss appeared to correspond to areas of thinning as seen on SD-OCT and were most prominent within the deep plexus, which is not well visualized on FA. This larger case series confirms that discrete areas of flow loss as detected by OCT-A are common (37.8% of eyes overall), and provides further support of the relationship between vascular flow loss and retinal thinning. When evaluating by genotype, the frequencies of flow abnormalities in our study were similar for both sickle SS and SC (40.0% and 41.7% respectively). These percentages compare well to those reported for macular thinning as detected by SD-OCT in patients with sickle cell disease (43% overall, and 48% in SS versus 35% of SC eyes in the largest series to date)8. OCT-A allows for several different quantitative measures of vascular flow, including flow areas, or vessel density. When comparing retinal 5 Han et al.
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
thickness measurements with measured vessel density in the various parafoveal subfields, a strong statistical correlation was found between retinal thickness measurements and vessel density in the fovea, parafovea, temporal, and superior subfields of both the superficial and deep plexi (Table 5). Thus, while initial observations of macular thinning on SD-OCT were presumed to be due to vascular causes not well-visualized by FA, OCT-A provides definitive evidence of a correlation. Although macular abnormalities are observed frequently in patients with sickle cell disease, the clinical implications of these changes are unclear, particularly with regard to impact on visual acuity. Anatomically, sickle cell patients have been found to have wider foveal diameters as detected by FA than controls19. Studies utilizing SD-OCT have described a characteristic broad foveal contour (“foveal splaying”) in patients with sickle cell disease and showed that central macular and parafoveal thicknesses are less than that of control eyes6. Chow et al. used SD-OCT and microperimetry to demonstrate a structure-function relationship between areas of retinal thinning and decreased macular sensitivity in sickle cell patients5. However, these areas were predominantly temporal to the fovea, and visual acuity was good (20/25 or better) across the entire cohort, so the effect on distance visual acuity was unclear. A prior study utilizing FA in sickle cell disease patients to measure FAZ diameter was unable to show correlation to visual acuity19. OCT-A allows for quantitative analysis of vascular flow and vascular density, and these measures have been shown to have good reliability and repeatability in normal eyes20-22. Quantitative measures of foveal avascular area and macular vascular density by OCT-A have been correlated to visual acuity in conditions such as central retinal vein occlusion and diabetic retinopathy23-25. In this study, we found a strong correlation between visual acuity and measures of foveal non-flow area and parafoveal flow (P-values <0.001 and 0.001 respectively). Despite this strong correlation, the study was comprised of patients with very good distance visual acuity (median 20/20, range 20/12-20/80), many of whom were asymptomatic, so the effect of these vascular flow abnormalities may indeed be subclinical. We did not record near visual acuity, and it is possible that capillary loss in the macula may have more influence on near vision tasks. Future studies may attempt to correlate near visual acuity and assessment of functional visual impairment with OCT-A measures of macular vascular flow. Sickle cell retinopathy is classically considered a peripheral retinopathy, whereby peripheral vascular changes predominate. However, the vessels in the temporal macula along the horizontal raphe are akin to terminal branches and are known to be more susceptible to occlusion in sickle cell patients3. Data from SD-OCT studies demonstrated that retinal thinning occurs most commonly temporal to the fovea4,7,8. The results of this study confirm these prior reports, as retinal thickness measurements in our study were lowest in the temporal parafoveal subfield relative to all other subfields. Initial observations using OCT-A suggested that flow loss may account for these areas of temporal macular retinal thinning seen on SD-OCT, but these were qualitative assessments based on case reports and small case series15,17. In this study, we used quantitative measures of vessel density in the parafoveal subfields to assess the susceptibility of these various areas to vascular flow loss. Not surprisingly, average vessel density measurements in our cohort were lower across all parafoveal subfields compared to published normative data22. When comparing vessel density between the parafoveal subfields, vessel density in the temporal and nasal subfields were found to be less than those in the 6 Han et al.
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
superior and inferior subfields, also consistent with published normative data22. However, contrary to normative data, vessel density was lower in the temporal versus the nasal subfield, for both the superficial and deep plexi, providing quantitative evidence that the vessels within the temporal subfield may be more susceptible to flow loss in sickle cell disease. To further demonstrate the relationship between areas of vascular flow loss and retinal thinning, we analyzed retinal thickness measurements in various subfields relative to vessel density. A strong correlation was found in the fovea, parafovea overall, and the temporal and superior parafoveal subfields. No statistically-significant correlation was found for the nasal and inferior subfields, but this may be because these subfields were not commonly affected by flow loss. Natural history studies have shown that peripheral retinopathy findings are typically more common and more severe in those with SC than SS26, and the presence of temporal macular thinning on SD-OCT has been correlated with more advanced peripheral retinopathy7,13. In this study, we used OCT-A measures of vessel density in the various parafoveal subfields and compared these values by sickle cell genotype and retinopathy stage. Vessel density was found to be lower within the temporal and nasal parafoveal subfields of the deep capillary plexus in eyes with sickle SC and those with proliferative retinopathy (Table 6). Within the superficial plexus, a statistically significant difference was encountered only in the temporal parafoveal subfield for eyes with proliferative retinopathy compared to those without. These results suggest that decreased vessel density, particularly in the temporal subfield of both the superficial and deep retinal plexi, may be a convenient and useful surrogate marker for more advanced peripheral disease, specifically the presence of proliferative retinopathy. We did not obtain fluorescein angiograms as part of this study protocol, and future studies may correlate OCT-A macular flow loss with quantitative measures of peripheral ischemia on as seen with ultra-wide field FA. While retinal thinning and vascular flow loss in the temporal macula seems to correlate with more advanced peripheral retinopathy, existing evidence is limited regarding whether foveal changes also correlate, and the quantitative flow analyses in our study do not suggest a definitive correlation (Table 4). However, eyes were excluded from analysis if significant disruption of the FAZ precluded accurate quantitative measures (e.g. Figure 3, where the boundaries of a pathologic FAZ are indistinct in the deep plexus and contiguous with an adjacent area of flow loss that extends beyond the borders of the scan). As such, eyes with more severe macular vascular pathology may be underrepresented in this analysis, with unclear effect on the correlations to more severe peripheral disease. Similarly, though evidence suggests that peripheral retinopathy may be more severe in sickle SC patients than those with SS, any differences in the FAZ are less apparent. In this study, there were no statisticallysignificant differences in mean foveal non-flow area and mean flow area within a 1 mm radius circle centered on the fovea between eyes from sickle SC patients versus those from sickle SS. These results support prior studies using FA, which showed no difference in FAZ diameter when comparing across different sickle cell genotypes19. Thus, while our study demonstrated a correlation between measures of foveal flow and visual acuity, these same measures do not appear to correlate well with more advanced peripheral retinopathy or to be significantly different between sickle SC and SS patients. Our study had several limitations. Despite the exclusion of patients with significant media opacity, image artifacts were still common and precluded qualitative flow analysis in 7 Han et al.
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
15.5% of eyes (15/97 eyes), and quantitative analysis in up to 22.7% of eyes (22/97 eyes when analyzing parafoveal deep plexus flow). Few patients with severe peripheral retinopathy (e.g. stage 4, vitreous hemorrhage or stage 5, retinal detachment) were included in this study, in part because these patients may have media opacity or poor visual acuity prohibiting image acquisition. Similarly, eyes with severe macular vascular abnormalities precluding accurate measurements of non-flow/flow areas or vessel density were excluded from quantitative flow analyses. As such, the study may underestimate the frequency of flow abnormalities in eyes with more severe disease. Minvielle et al. have shown that foveal non-flow areas are larger in sickle cell disease patients than age-matched controls17, and the macular vessel density measurements in our cohort are lower than published age-matched controls22. We did not include a control group and instead performed comparisons within our cohort by stratifying into subgroups by genotypes and stage of retinopathy. Our study was performed at a single, urban, tertiary care center in the United States, where sickle SS and SC are most common27. Thus, even though our study included a relatively large number of eyes, we had sufficient numbers to compare only between the SS and SC genotypes and the pre-set analysis of eyes with neovascularization versus without. The study was not powered to further sub-stratify by individual stages of retinopathy, or stages of retinopathy within genotype comparisons. Although our results suggest that the SC genotype and presence of proliferative retinopathy are likely associated with more extensive macular vascular flow loss, the severity of sickle cell retinopathy can vary significantly between two eyes of the same individual, or between individuals of the same genotype. Further study is needed to elucidate the systemic and ocular risk factors that may predispose the retinal circulation of individuals with sickle disease to incur damage from these vascular insults. OCT-A is a relatively quick, non-invasive, and useful way to evaluate macular vascular changes in sickle cell disease patients. This study demonstrates that areas of flow loss as identified by OCT-A may be seen at any stage of retinopathy and are common in patients with sickle cell retinopathy of various genotypes. The study also shows that flow loss affecting the fovea is correlated with decreased visual acuity and provides further qualitative and quantitative evidence that decreased vascular flow correlates with areas of retinal thinning in sickle cell disease. The temporal macula appears to be most susceptible to macular vascular changes, and flow loss in the temporal subfield is seen more frequently in patients with sickle SC disease, or eyes with proliferative retinopathy. Clinicians should pay attention to macular vascular abnormalities, particularly within the temporal and nasal parafoveal subfields of the deep capillary plexus, and carefully evaluate the retinal periphery for neovascularization if flow loss in these areas is seen.
8 Han et al.
ACCEPTED MANUSCRIPT
ACKNOWLEDGMENT / DISCLOSURES
RI PT
A. Funding/Support: Supported through unrestricted contributions to the Johns Hopkins University Retina Division research fund, and in part by private philanthropy from Gail C. and Howard Woolley. B. Financial Disclosures: ICH: No financial disclosures; MT: No financial disclosures; KDP: No financial disclosures; AWS: No financial disclosures C. Other Acknowledegments: None.
AC C
EP
TE D
M AN U
SC
Presented in part at the American Society of Retina Specialists Annual Meeting 8/11/16, San Francisco, California.
9 Han et al.
ACCEPTED MANUSCRIPT
References
6.
7.
8. 9.
10.
11.
12.
13. 14.
15.
16.
17. 18.
RI PT
SC
5.
M AN U
4.
TE D
3.
EP
2.
Goldberg MF. Natural history of untreated proliferative sickle retinopathy. Arch Ophthalmol. 1971;85(4):428-437. Goldberg MF. Classification and pathogenesis of proliferative sickle retinopathy. Am J Ophthalmol. 1971;71(3):649-665. Stevens TS, Busse B, Lee CB, Woolf MB, Galinos SO, Goldberg MF. Sickling hemoglobinopathies; macular and perimacular vascular abnormalities. Arch Ophthalmol. 1974;92(6):455-463. Brasileiro F, Martins TT, Campos SB, et al. Macular and peripapillary spectral domain optical coherence tomography changes in sickle cell retinopathy. Retina. 2015;35(2):257-263. Chow CC, Genead MA, Anastasakis A, Chau FY, Fishman GA, Lim JI. Structural and functional correlation in sickle cell retinopathy using spectral-domain optical coherence tomography and scanning laser ophthalmoscope microperimetry. Am J Ophthalmol. 2011;152(4):704-711.e702. Hoang QV, Chau FY, Shahidi M, Lim JI. Central macular splaying and outer retinal thinning in asymptomatic sickle cell patients by spectral-domain optical coherence tomography. Am J Ophthalmol. 2011;151(6):990-994.e991. Hood MP, Diaz RI, Sigler EJ, Calzada JI. Temporal Macular Atrophy as a Predictor of Neovascularization in Sickle Cell Retinopathy. Ophthalmic Surg Lasers Imaging Retina. 2016;47(1):27-34. Mathew R, Bafiq R, Ramu J, et al. Spectral domain optical coherence tomography in patients with sickle cell disease. Br J Ophthalmol. 2015;99(7):967-972. Minnal VR, Tsui I, Rosenberg JB. Central macular splaying and outer retinal thinning in asymptomatic sickle cell patients by spectral-domain optical coherence tomography. Am J Ophthalmol. 2011;152(6):1074; author reply 1074-1075. Murthy RK, Grover S, Chalam KV. Temporal macular thinning on spectral-domain optical coherence tomography in proliferative sickle cell retinopathy. Arch Ophthalmol. 2011;129(2):247-249. Oltra EZ, Chow CC, Wubben T, Lim JI, Chau FY, Moss HE. Cross-Sectional Analysis of Neurocognitive Function, Retinopathy, and Retinal Thinning by Spectral-Domain Optical Coherence Tomography in Sickle Cell Patients. Middle East Afr J Ophthalmol. 2016;23(1):79-83. Witkin AJ, Rogers AH, Ko TH, Fujimoto JG, Schuman JS, Duker JS. Optical coherence tomography demonstration of macular infarction in sickle cell retinopathy. Arch Ophthalmol. 2006;124(5):746-747. Ghasemi Falavarjani K, Scott AW, Wang K, et al. Correlation of Multimodal Imaging in Sickle Cell Retinopathy. Retina. 2016;36 Suppl 1:S111-S117. Spaide RF, Klancnik JM, Jr., Cooney MJ. Retinal vascular layers imaged by fluorescein angiography and optical coherence tomography angiography. JAMA Ophthalmol. 2015;133(1):45-50. Han IC, Tadarati M, Scott AW. Macular Vascular Abnormalities Identified by Optical Coherence Tomographic Angiography in Patients With Sickle Cell Disease. JAMA Ophthalmol. 2015;133(11):1337-1340. Grover S, Sambhav K, Chalam KV. Capillary nonperfusion by novel technology of OCT angiography in a patient with sickle cell disease with normal fluorescein angiogram. Eur J Ophthalmol. 2016:0. Minvielle W, Caillaux V, Cohen SY, et al. Macular Microangiopathy in Sickle Cell Disease Using Optical Coherence Tomography Angiography. Am J Ophthalmol. 2016;164:137-144.e131. Sanfilippo CJ, Klufas MA, Sarraf D, Tsui I. OPTICAL COHERENCE TOMOGRAPHY ANGIOGRAPHY OF SICKLE CELL MACULOPATHY. Retin Cases Brief Rep. 2015;9(4):360-362.
AC C
1.
10 Han et al.
ACCEPTED MANUSCRIPT
22.
23.
24.
25. 26.
AC C
EP
TE D
27.
RI PT
21.
SC
20.
Sanders RJ, Brown GC, Rosenstein RB, Magargal L. Foveal avascular zone diameter and sickle cell disease. Arch Ophthalmol. 1991;109(6):812-815. Al-Sheikh M, Tepelus TC, Nazikyan T, Sadda SR. Repeatability of automated vessel density measurements using optical coherence tomography angiography. Br J Ophthalmol. 2016. Hwang TS, Gao SS, Liu L, et al. Automated Quantification of Capillary Nonperfusion Using Optical Coherence Tomography Angiography in Diabetic Retinopathy. JAMA Ophthalmol. 2016;134(4):367-373. Coscas F, Sellam A, Glacet-Bernard A, et al. Normative Data for Vascular Density in Superficial and Deep Capillary Plexuses of Healthy Adults Assessed by Optical Coherence Tomography Angiography. Invest Ophthalmol Vis Sci. 2016;57(9):OCT211-223. Samara WA, Shahlaee A, Adam MK, et al. Quantification of Diabetic Macular Ischemia Using Optical Coherence Tomography Angiography and Its Relationship with Visual Acuity. Ophthalmology. 2016. Casselholmde Salles M, Kvanta A, Amren U, Epstein D. Optical Coherence Tomography Angiography in Central Retinal Vein Occlusion: Correlation Between the Foveal Avascular Zone and Visual Acuity. Invest Ophthalmol Vis Sci. 2016;57(9):OCT242-246. Kang JW, Yoo R, Jo YH, Kim HC. Correlation of Microvascular Structures on Optical Coherence Tomography Angiography with Visual Acuity in Retinal Vein Occlusion. Retina. 2016. Downes SM, Hambleton IR, Chuang EL, Lois N, Serjeant GR, Bird AC. Incidence and natural history of proliferative sickle cell retinopathy: observations from a cohort study. Ophthalmology. 2005;112(11):1869-1875. Elagouz M, Jyothi S, Gupta B, Sivaprasad S. Sickle cell disease and the eye: old and new concepts. Surv Ophthalmol. 2010;55(4):359-377.
M AN U
19.
11 Han et al.
ACCEPTED MANUSCRIPT
Figure Captions
RI PT
Figure 1. Example images, 29-year-old man with sickle SC. First column, 3x3 mm optical coherence tomography angiography (OCT-A) scans temporal to fovea (dotted box, top left). Second column, OCT-A scans centered on fovea (solid box, top middle). Third column, montage of first two columns. Yellow arrows point to areas of flow loss.
SC
Figure 2. Example vessel density data, 3x3 mm scans, right eye, 32-year-old man with sickle SC. First column, en face OCT-A images. Second column, vessel density maps, with areas of lowest density in blue. Third column, automated vessel density values. Fourth column, OCT reference images of segmentation.
AC C
EP
TE D
M AN U
Figure 3. Example flow measurements, left eye, 51-year-old woman with sickle SC. Top row, without flow measurements. Note ill-defined foveal avascular zone that extends beyond edge of scan. Second row, foveal non-flow areas. Note contiguous area of flow loss in deep plexus, not incorporated by automated non-flow measurements (asterisks). Third row, flow area measurements within 1 mm radius circle centered on fovea.
12 Han et al.
ACCEPTED MANUSCRIPT
M AN U
27 (58.7%) 14 (30.4%) 4 (8.7%) 1 (2.2%) 102 5 15 82 42 (51.2%) 40 (48.8%)
EP AC C
SC
46 33.5 28 (60.9%)
TE D
Participant characteristics Number of patients Age, mean (years) Female, no (%) Sickle genotype, no (%) SS SC B-thal Trait Number of eyes total Unable to obtain Excluded with artifacts Eyes included for analysis, no (%) OD OS
RI PT
Table 1. Participant characteristics
ACCEPTED MANUSCRIPT
Table 2. Frequency of loss of flow, stratified by genotype, stages of retinopathy, and presence of proliferative retinopathy. NV=neovascularization
OD
OS
All genotypes
16 (38.1)
15 (37.5)
31 (37.8)
SS
10 (41.7)
10 (38.5)
20 (40.0)
SC
6 (46.2)
4 (36.4)
10 (41.7)
0 (0)
1 (50.0)
1 (16.7)
Trait
0 (0)
0 (0)
0 (0)
3 (27.3)
1 (12.5)
4 (21.1)
1 (11.1)
5 (55.6)
6 (33.3)
5 (45.5)
3 (37.5)
8 (42.1)
6 (66.7)
6 (42.9)
12 (52.2)
1 (100.0)
0 (0)
1 (50.0)
1 (100.0)
0 (0)
1 (100.0)
9 (29.0)
9 (36.0)
18 (32.1)
8 (72.7)
6 (40.0)
14 (53.8)
M AN U
B-thal
SC
Genotype
Stage of retinopathy No retinopathy Stage 1 Stage 2
Stage 4 Stage 5
TE D
Stage 3
Total (OU)
RI PT
Areas of flow loss, no (%)
Stratified by Proliferative Disease
EP
No NV (Stage 2 or below)
AC C
NV (Stage 3 or above)
ACCEPTED MANUSCRIPT
Table 3. Correlation between visual acuity and measures of foveal flow. r=Pearson’s correlation coefficient. r
P-value
r
P-value
0.285
0.01
0.360
0.001
-0.303
0.007
-0.380
<0.001
2
Foveal non-flow area (mm )
AC C
EP
TE D
M AN U
SC
Foveal flow area within 1 mm radius circle (mm )
RI PT
2
ACCEPTED MANUSCRIPT
Superficial Capillary Plexus
Deep Capillary Plexus
SC
Foveal Flow area Foveal Flow area non-flow (1 mm non-flow (1 mm area area radius) radius) 0.367
1.345
0.462
1.359
Sickle SS
0.353
1.366
0.426
1.397
Sickle SC
0.427
1.317
0.568
1.290
P-values
0.33
0.27
0.20
0.05
No NV (Stage 2 and below)
0.340
1.363
0.410
1.389
NV (Stage 3 and above)
0.427
1.31
0.575
1.300
P-values
0.18
0.21
0.09
0.10
EP
TE D
M AN U
All eyes
AC C
RI PT
Table 4. Non-flow area within foveal avascular zone, and flow area within 1 mm radius centered on the fovea, with comparisons according to genotype and presence of neovascularization. P-values <0.05 considered statistically significant. All values are mm2. NV=neovascularization.
ACCEPTED MANUSCRIPT
Table 5. Correlation between retinal thickness measurements and vessel density in various subfields of the superficial and deep plexi. r=Pearson’s correlation coefficient.
RI PT
P-value <0.001 0.002 <0.001 <0.001 0.06
SC
r 0.591 0.334 0.546 0.571 0.204
Deep Plexus Average Vessel Density (%) 26.45 57.22 56.02 57.73 56.90
AC C
EP
TE D
M AN U
Subfield (3x3 mm scan) Fovea Parafovea Temporal Superior Nasal
Average Retinal Thickness (microns) 240.6 311.0 295.8 315.3 317.5
Superficial Plexus Average Vessel Density (%) 27.81 52.64 51.27 53.43 52.16
r 0.562 0.258 0.522 0.435 0.095
P-value <0.001 0.02 <0.001 <0.001 0.38
ACCEPTED MANUSCRIPT
Plexus
Deep Capillary
whole
fovea
All eyes
50.52
27.81
52.64
51.27
53.43 52.16 53.69 54.36
Sickle SS
51.15
27.51
53.39
52.28
53.99 53.01 54.26 55.05
Sickle SC
49.50
28.43
51.44
49.34
52.86 50.57 52.97 52.41
0.13
0.57
0.11
0.05
0.36
51.02
27.95
53.22
49.23
27.87
51.09
0.11
0.96
0.09
26.45
parafovea temporal superior nasal inferior
57.22
56.02
57.73
56.90 58.22
56.95
58.47
58.00 58.78
27.52
54.91
53.38
55.94
54.14 56.16
0.04
0.34
0.03
0.04
0.09
0.04
52.22
53.73 52.85 54.07 55.21
26.29
58.24
57.30
58.75
57.97 58.93
48.87
52.47 49.91 53.07 52.19
27.14
54.52
52.69
55.08
53.63 56.66
0.59
0.01
0.01
0.02
0.04
M AN U
58.05
0.15
TE D
AC C
fovea
26.05
0.02
EP
P-values (SS vs SC) No NV (Stage 2 and below) NV (Stage 3 and above) P-values (no NV versus NV)
parafovea temporal superior nasal inferior whole
Plexus
SC
Superficial Capillary
RI PT
Table 6. Vessel density (%) for superficial and deep capillary plexus, with comparisons of density in various subfields according to genotype and presence of neovascularization (stage 3 or above). P values <0.05 considered statistically significant. NV=neovascularization.
0.34
0.12
0.20
0.28
0.02
0.04
0.06
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
S0002-9394(17)30064-8
DOI:
10.1016/j.ajo.2017.02.007
Reference:
AJOPHT 10040
To appear in:
American Journal of Ophthalmology
Received Date: 30 September 2016 Revised Date:
4 February 2017
Accepted Date: 7 February 2017
Please cite this article as: Han IC, Tadarati M, Pacheco KD, Scott AW, Evaluation of macular vascular abnormalities identified by optical coherence tomography angiography in sickle cell disease, American Journal of Ophthalmology (2017), doi: 10.1016/j.ajo.2017.02.007. 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
Abstract Purpose: To evaluate macular vascular flow abnormalities identified by optical coherence tomography angiography (OCT-A) in patients with various sickle cell genotypes.
RI PT
Design: Prospective, observational case series.
SC
Methods: This is a single institution case series of adult patients with various sickle cell genotypes. All patients underwent macular OCT-A (Avanti RTVue XR, Optovue Inc, Fremont, CA). Images were analyzed qualitatively for areas of flow loss and quantitatively for measures of foveal avascular area, parafoveal flow, and vascular density. The findings were compared by sickle cell genotype and retinopathy stage and correlated to retinal thickness and visual acuity.
M AN U
Results: OCT-A scans of 82 eyes from 46 patients (60.9% female, mean age 33.5 years) were included. Sickle cell genotypes included 27 patients with SS (58.7%), 14 SC (30.4%), 4 betathalassemia (8.7%), and 1 sickle trait (2.2%). Discrete areas of flow loss were noted in 37.8% (31/82) of eyes overall and were common in both SS (40.0%, 20/50 eyes) and SC (41.7%, 10/24 eyes). Flow loss was more extensive in the temporal and nasal parafoveal subfields of the deep plexus with sickle SC or proliferative retinopathy. Retinal thickness measurements correlated with vascular density of the fovea, parafovea, temporal and superior subfields. Visual acuity correlated with foveal avascular zone area and parafoveal vascular density in the superficial and deep plexi.
AC C
EP
TE D
Conclusions: Areas of abnormal macular vascular flow are common in patients with various sickle cell genotypes. These areas may be seen at any retinopathy stage but may be more extensive with sickle SC or proliferative retinopathy.
ACCEPTED MANUSCRIPT
Title: Evaluation of macular vascular abnormalities identified by optical coherence tomography angiography in sickle cell disease Short Title: Macular OCT angiography of sickle cell retinopathy
RI PT
Authors: Ian C. Han1; Mongkol Tadarati1, Katia D. Pacheco1, Adrienne W. Scott1
M AN U
Corresponding Author: Adrienne W. Scott, MD Wilmer Eye Institute, Retina Division 600 North Wolfe Street Maumenee 719 Baltimore, MD 21287 [email protected]
SC
Author Affiliations: 1Retina Division, Wilmer Eye Institute, Johns Hopkins School of Medicine, Baltimore, Maryland
AC C
EP
TE D
Footnote for current author affiliations: ICH: Department of Ophthalmology and Visual Sciences, Carver College of Medicine, University of Iowa, Iowa City, Iowa MT: Retina Division, Department of Ophthalmology, Rajavithi Hospital, College of Medicine, Rangsit University, Bangkok, Thailand KDP: Retina Division, Department of Ophthalmology, Brazilian Center of Vision Eye Hospital, Brasilia, Brazil; Retina Division, Department of Ophthalmology, Armed Forces Hospital, Brasilia, Brazil
1 Han et al.
ACCEPTED MANUSCRIPT
Introduction
AC C
EP
TE D
M AN U
SC
RI PT
Sickle cell retinopathy is classically considered a peripheral retinopathy, with characteristic changes as described by Goldberg, including arteriolar occlusions, arteriovenous anastomoses, neovascularization, vitreous hemorrhage, or retinal detachment1,2. However, numerous macular vascular abnormalities can also be seen, including irregularities of the foveal avascular zone, microaneurysmal dots, and hairpin venular loops3. These macular vascular abnormalities are commonly noted temporal to the fovea because the temporal macular vessels are within the watershed zone along the horizontal raphe and are thus akin to terminal vessels and more susceptible to vascular occlusion3. Numerous reports have described morphologic changes in the macula using spectral domain optical coherence tomography (SD-OCT), including the presence of macular splaying and areas of inner retinal thinning, predominantly temporal to the macula4-12. While many of these patients are asymptomatic, decreased retinal sensitivity has been documented in areas of thinning5, and the presence of macular thinning has been correlated to more advanced peripheral changes7,13. Specifically, patients with temporal macular thinning on SD-OCT have higher rates of peripheral neovascularization in the same eye, suggesting that macular thinning may be a surrogate marker for peripheral retinopathy7. Although macular thinning is commonly observed on SD-OCT, these patients may not have corresponding abnormalities on fluorescein angiography (FA)10. Optical coherence tomography angiography (OCT-A) is a relatively nascent technology that utilizes subtle differences in specular patterns across temporally distinct signals to detect movement of blood cells, allowing for the digital reconstruction of retinal vasculature. Unlike FA, OCT-A does not allow dynamic evaluation of the speed of vascular filling or vascular leakage. However, OCT-A has several advantages over FA, including the lack of contrast dye, improved visualization of several vascular layers including the radial papillary network and deep capillary plexus, and the ability to optically segment the retina to analyze the various retinal vascular networks separately14. We recently reported that OCT-A can detect areas of abnormal vascular flow within the macula in patients with sickle cell disease and that these areas correspond to areas of retinal thinning seen on SD-OCT15. Areas of flow loss by OCT-A appear to affect the deep greater than superficial plexus and importantly, can be seen in patients who were clinically visually asymptomatic and have no apparent vascular abnormalities on FA. Other small case series and case reports have since supported our initial findings16-18. However, the frequency and severity of these macular vascular abnormalities and their relationship to sickle cell genotype or the severity of peripheral retinopathy have not yet been described. There are also minimal existing data correlating areas of abnormal vascular flow with retinal thickness or visual acuity in patients with sickle cell disease. This study aims to evaluate the frequency and severity of macular flow abnormalities as detected by OCT-A in patients with various sickle cell genotypes, and to compare these flow abnormalities by sickle cell genotype and stage of retinopathy. Additionally, we evaluate the correlation between retinal thickness and vascular density, as well as between visual acuity and measures of foveal avascular area and parafoveal vascular flow.
2 Han et al.
ACCEPTED MANUSCRIPT
Methods
AC C
EP
TE D
M AN U
SC
RI PT
This is a single-institution, prospective, observational case series evaluating clinical and multimodal imaging findings for consecutive patients with various sickle cell genotypes who presented to the investigators’ (ICH, AWS) retina sub-specialty clinics for eye examination. The study was approved by the Johns Hopkins Hospital Institutional Review Board and complied with regulations set forth by the Health Insurance Portability and Accountability Act. Patients with known sickle cell of any genotype and adequate view to the fundus were included via written informed consent for participation. Exclusion criteria included cataract, vitreous hemorrhage, or other media opacity prohibiting adequate image acquisition, known maculopathy from other causes including vitreomacular traction, or history of known retinal vascular disease, including diabetes, hypertension, or retinal artery or vein occlusion. All patients received Early Treatment Diabetic Retinopathy Study (ETDRS) visual acuity testing with habitual distance spectacle correction and complete ophthalmic examination including dilated fundus examination by the lead investigators (ICH, AWS), both fellowshiptrained retina specialists. For each eye, stage of retinopathy was determined based on clinical examination per Goldberg’s classification2 as 1) no retinopathy, 2) non-proliferative retinopathy (sun-burst lesions, salmon patches, arteriolar occlusions, peripheral anastomoses without neovascularization, i.e. Goldberg stage 2 and below), or 3) proliferative retinopathy (with neovascularization, i.e. Goldberg stage 3 and above). Macular SD-OCT and OCT-A (Optovue RTVue XR Avanti, Optovue Inc, Fremont, California, USA) was performed for all patients on the same day as the clinical examination. Scan protocols included 6 mm x 6 mm and 3 mm x 3 mm scans centered on the fovea. Given the propensity for retinal thinning in the temporal macula4,7,8,10, we also performed dedicated 3 mm x 3 mm scans through the temporal macula, with the nasal edge of the scan at or near the fovea (Figure 1). For each field, multiple scans were acquired as needed to ensure adequate quality for analysis. OCT-A images were then generated using the instrument’s split-spectrum amplitude decorrelation angiography (SSADA) software, including automated segmentation of the various vascular networks (superficial capillary plexus, deep capillary plexus, choriocapillaris). For multiple scans in the same area, the images were analyzed by the lead investigators (ICH, AWS), and the best quality scans selected for analysis. Images were excluded if the quality was insufficient for analysis owing to eye movement, blinking, or poor fixation during image acquisition. Images were then analyzed by the lead investigators (ICH, AWS), and eyes with areas of flow loss within either the superficial or deep plexus or both were recorded. The frequency and relative proportion of scans with identifiable flow abnormalities on OCT-A were then calculated. Automated retinal thickness measurements and vascular density data were recorded for the whole image, fovea, parafovea, and parafoveal subfields for 3 mm x 3 mm scans centered on the fovea (Figure 2). The non-flow area (mm2) within the foveal avascular zone (FAZ) and flow area within a 1 mm radius circle centered on the fovea were measured in an automated fashion using the instrument’s intrinsic software (RTVue XR version 2016.1.0.23). These measures were evaluated by the lead investigators (ICH, AWS) and excluded in eyes when pathologic disruption of the FAZ precluded the software’s ability to properly identify the boundaries of the non-flow or flow areas. Figure 3 shows an example of a pathologic FAZ, 3 Han et al.
ACCEPTED MANUSCRIPT
M AN U
SC
RI PT
where the software correctly identifies the FAZ in the superficial plexus but cannot determine the boundaries of the FAZ within the deep plexus, where a contiguous, adjacent area of flow loss extends beyond the edges of the scan. Visual acuities were converted to logarithm of minimal angle of resolution (logMAR) equivalents prior to statistical analysis. Pearson’s correlation coefficient was used to evaluate the association between visual acuity and measures of foveal flow including foveal non-flow area and flow within a 1 mm radius circle centered on the fovea. Retinal thickness measurements were compared for each parafoveal subfield using paired T-tests. Vessel density measurements for each subfield were compared in similar fashion. Pearson’s correlation coefficient was used to evaluate the association between retinal thickness measurements and vessel density within the various parafoveal subfields. Patients were then stratified by sickle cell genotype and scans for each eye stratified by stage of retinopathy. Based on pre-existing SDOCT data correlating areas of temporal thinning with proliferative retinopathy7, we selected eyes with proliferative retinopathy (Goldberg Stage 3 or above) or without as pre-set groups for sub-analysis. Chi-square tests of proportions were used to compare frequency of abnormal findings across sub-populations within the study, and two-tailed, unpaired T-tests used to compare vessel density data across the various subgroups. P-values of <0.05 were considered statistically significant for all statistical analyses, which were performed using Microsoft Office Excel 2007 (Microsoft Corp., Seattle, WA, USA) and GraphPad Prism 7 (GraphPad Software, Inc., La Jolla, CA, USA). Results
AC C
EP
TE D
Fifty-one consecutive patients with sickle cell disease were enrolled in the study. Of the 102 available eyes, 5 eyes had media opacity precluding imaging, and OCT-A scans of 97 eyes (50 right, 47 left) were obtained for the study. Of these, 82 eyes (42 right, 40 left) from 46 patients (28/46 female, 60.9%), had scans with image quality acceptable for qualitative analysis. Sickle cell genotypes included 27 patients with hemoglobin SS (58.7%), 14 with hemoglobin SC (30.4%), 4 with beta-thalassemia (8.7%), 1 with sickle trait (2.2%). The mean age was 33.5 years old (Table 1). Visual acuity ranged from 20/12 to 20/80, with the majority of eyes possessing visual acuity of 20/20 or better (63 out of 82 eyes, 76.8%). Discrete areas of foveal and parafoveal flow loss, in the superficial or deep plexus or both, were noted in 37.8% (31/82) of eyes overall. When stratified by genotype, these areas were common in eyes of both SS (40.0%, 20/50) and SC (41.7%, 10/24) patients (P=0.89). When stratified by presence of proliferative retinopathy as defined by neovascularization (Goldberg stage 3 and above), discrete areas of flow loss were seen in 53.8% of eyes (14/26) with neovascularization versus 32.1% of eyes (18/56) without (P=0.06) (Table 2). When evaluating quantitative measures of macular vascular flow, algorithm errors precluded accurate measurement of foveal non-flow area within the superficial plexus for 2 eyes (80 eyes included for analysis) and within the deep plexus for 4 eyes (78 eyes included). Similarly, for quantitative analysis of flow within a 1 mm circle centered on the fovea, algorithm errors precluded accurate measurement of the superficial plexus for 4 eyes (78 eyes included for analysis) and of the deep plexus for 7 eyes (75 eyes included). LogMAR visual acuity was positively correlated with foveal non-flow measures (P=0.001), i.e. greater non-flow area 4 Han et al.
ACCEPTED MANUSCRIPT
EP
TE D
M AN U
SC
RI PT
correlated to worse visual acuity, and negatively correlated with measures of flow within 1 mm disc centered on the fovea (P<0.001), i.e. greater parafovea flow correlated to better visual acuity (Table 3). When stratified by sickle cell genotype and retinopathy stage, the mean foveal non-flow area in both the superficial and deep plexi was greater for sickle SC eyes compared to sickle SS, and for eyes with neovascularization compared to those without. However, none of these differences reached statistical significance (Table 4). The mean flow area within a 1 mm radius circle centered on the fovea was greater for sickle SS eyes and those without neovascularization, but these differences also did not reach statistical significance (Table 4). Average retinal thickness measurements for the various subfields are shown in Table 5. When comparing retinal thickness values across the parafoveal subfields, the temporal subfield was noted to be thinner (P<0.001) relative to the superior, nasal, and inferior subfields. There were no statistically-significant differences in thickness measurements across the remaining subfields (P-values all >0.3). Vessel density within the temporal and nasal subfields were noted to be lower than the superior and inferior subfields in both the superficial and deep plexi (P<0.001 for all comparisons). When comparing the horizontal subfields, vessel density was noted to be lower in the temporal versus the nasal for both the superficial (P=0.03) and deep plexi (P=0.04). Retinal thickness was found to be correlated to vessel density in the fovea and parafovea overall (Table 5) for both the superficial and deep plexi. These measures were also strongly correlated in the temporal and superior parafoveal subfields (P<0.001) of the superior and deep plexi, whereas no statistically-significant correlation was found for the nasal and inferior subfields. When comparing vessel density measures by sickle cell genotype, the vessel density of the deep capillary plexus overall and multiple subfields (parafovea overall, and temporal and nasal parafoveal subfields) were found to be different to a statistically significant degree, with lower densities seen in sickle SC eyes compared to sickle SS eyes (Table 6). When stratified by severity of retinopathy, vessel density within these same fields was lower in eyes with neovascularization compared to those without. Within the superficial plexus, a statistically significant difference was encountered only when comparing densities of the temporal subfield in eyes with neovascularization versus those without.
Discussion
AC C
We previously reported that OCT-A can identify areas of macular vascular flow loss in eyes with sickle cell retinopathy, even in the absence of obvious abnormalities on fluorescein angiography15. These areas of flow loss appeared to correspond to areas of thinning as seen on SD-OCT and were most prominent within the deep plexus, which is not well visualized on FA. This larger case series confirms that discrete areas of flow loss as detected by OCT-A are common (37.8% of eyes overall), and provides further support of the relationship between vascular flow loss and retinal thinning. When evaluating by genotype, the frequencies of flow abnormalities in our study were similar for both sickle SS and SC (40.0% and 41.7% respectively). These percentages compare well to those reported for macular thinning as detected by SD-OCT in patients with sickle cell disease (43% overall, and 48% in SS versus 35% of SC eyes in the largest series to date)8. OCT-A allows for several different quantitative measures of vascular flow, including flow areas, or vessel density. When comparing retinal 5 Han et al.
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
thickness measurements with measured vessel density in the various parafoveal subfields, a strong statistical correlation was found between retinal thickness measurements and vessel density in the fovea, parafovea, temporal, and superior subfields of both the superficial and deep plexi (Table 5). Thus, while initial observations of macular thinning on SD-OCT were presumed to be due to vascular causes not well-visualized by FA, OCT-A provides definitive evidence of a correlation. Although macular abnormalities are observed frequently in patients with sickle cell disease, the clinical implications of these changes are unclear, particularly with regard to impact on visual acuity. Anatomically, sickle cell patients have been found to have wider foveal diameters as detected by FA than controls19. Studies utilizing SD-OCT have described a characteristic broad foveal contour (“foveal splaying”) in patients with sickle cell disease and showed that central macular and parafoveal thicknesses are less than that of control eyes6. Chow et al. used SD-OCT and microperimetry to demonstrate a structure-function relationship between areas of retinal thinning and decreased macular sensitivity in sickle cell patients5. However, these areas were predominantly temporal to the fovea, and visual acuity was good (20/25 or better) across the entire cohort, so the effect on distance visual acuity was unclear. A prior study utilizing FA in sickle cell disease patients to measure FAZ diameter was unable to show correlation to visual acuity19. OCT-A allows for quantitative analysis of vascular flow and vascular density, and these measures have been shown to have good reliability and repeatability in normal eyes20-22. Quantitative measures of foveal avascular area and macular vascular density by OCT-A have been correlated to visual acuity in conditions such as central retinal vein occlusion and diabetic retinopathy23-25. In this study, we found a strong correlation between visual acuity and measures of foveal non-flow area and parafoveal flow (P-values <0.001 and 0.001 respectively). Despite this strong correlation, the study was comprised of patients with very good distance visual acuity (median 20/20, range 20/12-20/80), many of whom were asymptomatic, so the effect of these vascular flow abnormalities may indeed be subclinical. We did not record near visual acuity, and it is possible that capillary loss in the macula may have more influence on near vision tasks. Future studies may attempt to correlate near visual acuity and assessment of functional visual impairment with OCT-A measures of macular vascular flow. Sickle cell retinopathy is classically considered a peripheral retinopathy, whereby peripheral vascular changes predominate. However, the vessels in the temporal macula along the horizontal raphe are akin to terminal branches and are known to be more susceptible to occlusion in sickle cell patients3. Data from SD-OCT studies demonstrated that retinal thinning occurs most commonly temporal to the fovea4,7,8. The results of this study confirm these prior reports, as retinal thickness measurements in our study were lowest in the temporal parafoveal subfield relative to all other subfields. Initial observations using OCT-A suggested that flow loss may account for these areas of temporal macular retinal thinning seen on SD-OCT, but these were qualitative assessments based on case reports and small case series15,17. In this study, we used quantitative measures of vessel density in the parafoveal subfields to assess the susceptibility of these various areas to vascular flow loss. Not surprisingly, average vessel density measurements in our cohort were lower across all parafoveal subfields compared to published normative data22. When comparing vessel density between the parafoveal subfields, vessel density in the temporal and nasal subfields were found to be less than those in the 6 Han et al.
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
superior and inferior subfields, also consistent with published normative data22. However, contrary to normative data, vessel density was lower in the temporal versus the nasal subfield, for both the superficial and deep plexi, providing quantitative evidence that the vessels within the temporal subfield may be more susceptible to flow loss in sickle cell disease. To further demonstrate the relationship between areas of vascular flow loss and retinal thinning, we analyzed retinal thickness measurements in various subfields relative to vessel density. A strong correlation was found in the fovea, parafovea overall, and the temporal and superior parafoveal subfields. No statistically-significant correlation was found for the nasal and inferior subfields, but this may be because these subfields were not commonly affected by flow loss. Natural history studies have shown that peripheral retinopathy findings are typically more common and more severe in those with SC than SS26, and the presence of temporal macular thinning on SD-OCT has been correlated with more advanced peripheral retinopathy7,13. In this study, we used OCT-A measures of vessel density in the various parafoveal subfields and compared these values by sickle cell genotype and retinopathy stage. Vessel density was found to be lower within the temporal and nasal parafoveal subfields of the deep capillary plexus in eyes with sickle SC and those with proliferative retinopathy (Table 6). Within the superficial plexus, a statistically significant difference was encountered only in the temporal parafoveal subfield for eyes with proliferative retinopathy compared to those without. These results suggest that decreased vessel density, particularly in the temporal subfield of both the superficial and deep retinal plexi, may be a convenient and useful surrogate marker for more advanced peripheral disease, specifically the presence of proliferative retinopathy. We did not obtain fluorescein angiograms as part of this study protocol, and future studies may correlate OCT-A macular flow loss with quantitative measures of peripheral ischemia on as seen with ultra-wide field FA. While retinal thinning and vascular flow loss in the temporal macula seems to correlate with more advanced peripheral retinopathy, existing evidence is limited regarding whether foveal changes also correlate, and the quantitative flow analyses in our study do not suggest a definitive correlation (Table 4). However, eyes were excluded from analysis if significant disruption of the FAZ precluded accurate quantitative measures (e.g. Figure 3, where the boundaries of a pathologic FAZ are indistinct in the deep plexus and contiguous with an adjacent area of flow loss that extends beyond the borders of the scan). As such, eyes with more severe macular vascular pathology may be underrepresented in this analysis, with unclear effect on the correlations to more severe peripheral disease. Similarly, though evidence suggests that peripheral retinopathy may be more severe in sickle SC patients than those with SS, any differences in the FAZ are less apparent. In this study, there were no statisticallysignificant differences in mean foveal non-flow area and mean flow area within a 1 mm radius circle centered on the fovea between eyes from sickle SC patients versus those from sickle SS. These results support prior studies using FA, which showed no difference in FAZ diameter when comparing across different sickle cell genotypes19. Thus, while our study demonstrated a correlation between measures of foveal flow and visual acuity, these same measures do not appear to correlate well with more advanced peripheral retinopathy or to be significantly different between sickle SC and SS patients. Our study had several limitations. Despite the exclusion of patients with significant media opacity, image artifacts were still common and precluded qualitative flow analysis in 7 Han et al.
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
15.5% of eyes (15/97 eyes), and quantitative analysis in up to 22.7% of eyes (22/97 eyes when analyzing parafoveal deep plexus flow). Few patients with severe peripheral retinopathy (e.g. stage 4, vitreous hemorrhage or stage 5, retinal detachment) were included in this study, in part because these patients may have media opacity or poor visual acuity prohibiting image acquisition. Similarly, eyes with severe macular vascular abnormalities precluding accurate measurements of non-flow/flow areas or vessel density were excluded from quantitative flow analyses. As such, the study may underestimate the frequency of flow abnormalities in eyes with more severe disease. Minvielle et al. have shown that foveal non-flow areas are larger in sickle cell disease patients than age-matched controls17, and the macular vessel density measurements in our cohort are lower than published age-matched controls22. We did not include a control group and instead performed comparisons within our cohort by stratifying into subgroups by genotypes and stage of retinopathy. Our study was performed at a single, urban, tertiary care center in the United States, where sickle SS and SC are most common27. Thus, even though our study included a relatively large number of eyes, we had sufficient numbers to compare only between the SS and SC genotypes and the pre-set analysis of eyes with neovascularization versus without. The study was not powered to further sub-stratify by individual stages of retinopathy, or stages of retinopathy within genotype comparisons. Although our results suggest that the SC genotype and presence of proliferative retinopathy are likely associated with more extensive macular vascular flow loss, the severity of sickle cell retinopathy can vary significantly between two eyes of the same individual, or between individuals of the same genotype. Further study is needed to elucidate the systemic and ocular risk factors that may predispose the retinal circulation of individuals with sickle disease to incur damage from these vascular insults. OCT-A is a relatively quick, non-invasive, and useful way to evaluate macular vascular changes in sickle cell disease patients. This study demonstrates that areas of flow loss as identified by OCT-A may be seen at any stage of retinopathy and are common in patients with sickle cell retinopathy of various genotypes. The study also shows that flow loss affecting the fovea is correlated with decreased visual acuity and provides further qualitative and quantitative evidence that decreased vascular flow correlates with areas of retinal thinning in sickle cell disease. The temporal macula appears to be most susceptible to macular vascular changes, and flow loss in the temporal subfield is seen more frequently in patients with sickle SC disease, or eyes with proliferative retinopathy. Clinicians should pay attention to macular vascular abnormalities, particularly within the temporal and nasal parafoveal subfields of the deep capillary plexus, and carefully evaluate the retinal periphery for neovascularization if flow loss in these areas is seen.
8 Han et al.
ACCEPTED MANUSCRIPT
ACKNOWLEDGMENT / DISCLOSURES
RI PT
A. Funding/Support: Supported through unrestricted contributions to the Johns Hopkins University Retina Division research fund, and in part by private philanthropy from Gail C. and Howard Woolley. B. Financial Disclosures: ICH: No financial disclosures; MT: No financial disclosures; KDP: No financial disclosures; AWS: No financial disclosures C. Other Acknowledegments: None.
AC C
EP
TE D
M AN U
SC
Presented in part at the American Society of Retina Specialists Annual Meeting 8/11/16, San Francisco, California.
9 Han et al.
ACCEPTED MANUSCRIPT
References
6.
7.
8. 9.
10.
11.
12.
13. 14.
15.
16.
17. 18.
RI PT
SC
5.
M AN U
4.
TE D
3.
EP
2.
Goldberg MF. Natural history of untreated proliferative sickle retinopathy. Arch Ophthalmol. 1971;85(4):428-437. Goldberg MF. Classification and pathogenesis of proliferative sickle retinopathy. Am J Ophthalmol. 1971;71(3):649-665. Stevens TS, Busse B, Lee CB, Woolf MB, Galinos SO, Goldberg MF. Sickling hemoglobinopathies; macular and perimacular vascular abnormalities. Arch Ophthalmol. 1974;92(6):455-463. Brasileiro F, Martins TT, Campos SB, et al. Macular and peripapillary spectral domain optical coherence tomography changes in sickle cell retinopathy. Retina. 2015;35(2):257-263. Chow CC, Genead MA, Anastasakis A, Chau FY, Fishman GA, Lim JI. Structural and functional correlation in sickle cell retinopathy using spectral-domain optical coherence tomography and scanning laser ophthalmoscope microperimetry. Am J Ophthalmol. 2011;152(4):704-711.e702. Hoang QV, Chau FY, Shahidi M, Lim JI. Central macular splaying and outer retinal thinning in asymptomatic sickle cell patients by spectral-domain optical coherence tomography. Am J Ophthalmol. 2011;151(6):990-994.e991. Hood MP, Diaz RI, Sigler EJ, Calzada JI. Temporal Macular Atrophy as a Predictor of Neovascularization in Sickle Cell Retinopathy. Ophthalmic Surg Lasers Imaging Retina. 2016;47(1):27-34. Mathew R, Bafiq R, Ramu J, et al. Spectral domain optical coherence tomography in patients with sickle cell disease. Br J Ophthalmol. 2015;99(7):967-972. Minnal VR, Tsui I, Rosenberg JB. Central macular splaying and outer retinal thinning in asymptomatic sickle cell patients by spectral-domain optical coherence tomography. Am J Ophthalmol. 2011;152(6):1074; author reply 1074-1075. Murthy RK, Grover S, Chalam KV. Temporal macular thinning on spectral-domain optical coherence tomography in proliferative sickle cell retinopathy. Arch Ophthalmol. 2011;129(2):247-249. Oltra EZ, Chow CC, Wubben T, Lim JI, Chau FY, Moss HE. Cross-Sectional Analysis of Neurocognitive Function, Retinopathy, and Retinal Thinning by Spectral-Domain Optical Coherence Tomography in Sickle Cell Patients. Middle East Afr J Ophthalmol. 2016;23(1):79-83. Witkin AJ, Rogers AH, Ko TH, Fujimoto JG, Schuman JS, Duker JS. Optical coherence tomography demonstration of macular infarction in sickle cell retinopathy. Arch Ophthalmol. 2006;124(5):746-747. Ghasemi Falavarjani K, Scott AW, Wang K, et al. Correlation of Multimodal Imaging in Sickle Cell Retinopathy. Retina. 2016;36 Suppl 1:S111-S117. Spaide RF, Klancnik JM, Jr., Cooney MJ. Retinal vascular layers imaged by fluorescein angiography and optical coherence tomography angiography. JAMA Ophthalmol. 2015;133(1):45-50. Han IC, Tadarati M, Scott AW. Macular Vascular Abnormalities Identified by Optical Coherence Tomographic Angiography in Patients With Sickle Cell Disease. JAMA Ophthalmol. 2015;133(11):1337-1340. Grover S, Sambhav K, Chalam KV. Capillary nonperfusion by novel technology of OCT angiography in a patient with sickle cell disease with normal fluorescein angiogram. Eur J Ophthalmol. 2016:0. Minvielle W, Caillaux V, Cohen SY, et al. Macular Microangiopathy in Sickle Cell Disease Using Optical Coherence Tomography Angiography. Am J Ophthalmol. 2016;164:137-144.e131. Sanfilippo CJ, Klufas MA, Sarraf D, Tsui I. OPTICAL COHERENCE TOMOGRAPHY ANGIOGRAPHY OF SICKLE CELL MACULOPATHY. Retin Cases Brief Rep. 2015;9(4):360-362.
AC C
1.
10 Han et al.
ACCEPTED MANUSCRIPT
22.
23.
24.
25. 26.
AC C
EP
TE D
27.
RI PT
21.
SC
20.
Sanders RJ, Brown GC, Rosenstein RB, Magargal L. Foveal avascular zone diameter and sickle cell disease. Arch Ophthalmol. 1991;109(6):812-815. Al-Sheikh M, Tepelus TC, Nazikyan T, Sadda SR. Repeatability of automated vessel density measurements using optical coherence tomography angiography. Br J Ophthalmol. 2016. Hwang TS, Gao SS, Liu L, et al. Automated Quantification of Capillary Nonperfusion Using Optical Coherence Tomography Angiography in Diabetic Retinopathy. JAMA Ophthalmol. 2016;134(4):367-373. Coscas F, Sellam A, Glacet-Bernard A, et al. Normative Data for Vascular Density in Superficial and Deep Capillary Plexuses of Healthy Adults Assessed by Optical Coherence Tomography Angiography. Invest Ophthalmol Vis Sci. 2016;57(9):OCT211-223. Samara WA, Shahlaee A, Adam MK, et al. Quantification of Diabetic Macular Ischemia Using Optical Coherence Tomography Angiography and Its Relationship with Visual Acuity. Ophthalmology. 2016. Casselholmde Salles M, Kvanta A, Amren U, Epstein D. Optical Coherence Tomography Angiography in Central Retinal Vein Occlusion: Correlation Between the Foveal Avascular Zone and Visual Acuity. Invest Ophthalmol Vis Sci. 2016;57(9):OCT242-246. Kang JW, Yoo R, Jo YH, Kim HC. Correlation of Microvascular Structures on Optical Coherence Tomography Angiography with Visual Acuity in Retinal Vein Occlusion. Retina. 2016. Downes SM, Hambleton IR, Chuang EL, Lois N, Serjeant GR, Bird AC. Incidence and natural history of proliferative sickle cell retinopathy: observations from a cohort study. Ophthalmology. 2005;112(11):1869-1875. Elagouz M, Jyothi S, Gupta B, Sivaprasad S. Sickle cell disease and the eye: old and new concepts. Surv Ophthalmol. 2010;55(4):359-377.
M AN U
19.
11 Han et al.
ACCEPTED MANUSCRIPT
Figure Captions
RI PT
Figure 1. Example images, 29-year-old man with sickle SC. First column, 3x3 mm optical coherence tomography angiography (OCT-A) scans temporal to fovea (dotted box, top left). Second column, OCT-A scans centered on fovea (solid box, top middle). Third column, montage of first two columns. Yellow arrows point to areas of flow loss.
SC
Figure 2. Example vessel density data, 3x3 mm scans, right eye, 32-year-old man with sickle SC. First column, en face OCT-A images. Second column, vessel density maps, with areas of lowest density in blue. Third column, automated vessel density values. Fourth column, OCT reference images of segmentation.
AC C
EP
TE D
M AN U
Figure 3. Example flow measurements, left eye, 51-year-old woman with sickle SC. Top row, without flow measurements. Note ill-defined foveal avascular zone that extends beyond edge of scan. Second row, foveal non-flow areas. Note contiguous area of flow loss in deep plexus, not incorporated by automated non-flow measurements (asterisks). Third row, flow area measurements within 1 mm radius circle centered on fovea.
12 Han et al.
ACCEPTED MANUSCRIPT
M AN U
27 (58.7%) 14 (30.4%) 4 (8.7%) 1 (2.2%) 102 5 15 82 42 (51.2%) 40 (48.8%)
EP AC C
SC
46 33.5 28 (60.9%)
TE D
Participant characteristics Number of patients Age, mean (years) Female, no (%) Sickle genotype, no (%) SS SC B-thal Trait Number of eyes total Unable to obtain Excluded with artifacts Eyes included for analysis, no (%) OD OS
RI PT
Table 1. Participant characteristics
ACCEPTED MANUSCRIPT
Table 2. Frequency of loss of flow, stratified by genotype, stages of retinopathy, and presence of proliferative retinopathy. NV=neovascularization
OD
OS
All genotypes
16 (38.1)
15 (37.5)
31 (37.8)
SS
10 (41.7)
10 (38.5)
20 (40.0)
SC
6 (46.2)
4 (36.4)
10 (41.7)
0 (0)
1 (50.0)
1 (16.7)
Trait
0 (0)
0 (0)
0 (0)
3 (27.3)
1 (12.5)
4 (21.1)
1 (11.1)
5 (55.6)
6 (33.3)
5 (45.5)
3 (37.5)
8 (42.1)
6 (66.7)
6 (42.9)
12 (52.2)
1 (100.0)
0 (0)
1 (50.0)
1 (100.0)
0 (0)
1 (100.0)
9 (29.0)
9 (36.0)
18 (32.1)
8 (72.7)
6 (40.0)
14 (53.8)
M AN U
B-thal
SC
Genotype
Stage of retinopathy No retinopathy Stage 1 Stage 2
Stage 4 Stage 5
TE D
Stage 3
Total (OU)
RI PT
Areas of flow loss, no (%)
Stratified by Proliferative Disease
EP
No NV (Stage 2 or below)
AC C
NV (Stage 3 or above)
ACCEPTED MANUSCRIPT
Table 3. Correlation between visual acuity and measures of foveal flow. r=Pearson’s correlation coefficient. r
P-value
r
P-value
0.285
0.01
0.360
0.001
-0.303
0.007
-0.380
<0.001
2
Foveal non-flow area (mm )
AC C
EP
TE D
M AN U
SC
Foveal flow area within 1 mm radius circle (mm )
RI PT
2
ACCEPTED MANUSCRIPT
Superficial Capillary Plexus
Deep Capillary Plexus
SC
Foveal Flow area Foveal Flow area non-flow (1 mm non-flow (1 mm area area radius) radius) 0.367
1.345
0.462
1.359
Sickle SS
0.353
1.366
0.426
1.397
Sickle SC
0.427
1.317
0.568
1.290
P-values
0.33
0.27
0.20
0.05
No NV (Stage 2 and below)
0.340
1.363
0.410
1.389
NV (Stage 3 and above)
0.427
1.31
0.575
1.300
P-values
0.18
0.21
0.09
0.10
EP
TE D
M AN U
All eyes
AC C
RI PT
Table 4. Non-flow area within foveal avascular zone, and flow area within 1 mm radius centered on the fovea, with comparisons according to genotype and presence of neovascularization. P-values <0.05 considered statistically significant. All values are mm2. NV=neovascularization.
ACCEPTED MANUSCRIPT
Table 5. Correlation between retinal thickness measurements and vessel density in various subfields of the superficial and deep plexi. r=Pearson’s correlation coefficient.
RI PT
P-value <0.001 0.002 <0.001 <0.001 0.06
SC
r 0.591 0.334 0.546 0.571 0.204
Deep Plexus Average Vessel Density (%) 26.45 57.22 56.02 57.73 56.90
AC C
EP
TE D
M AN U
Subfield (3x3 mm scan) Fovea Parafovea Temporal Superior Nasal
Average Retinal Thickness (microns) 240.6 311.0 295.8 315.3 317.5
Superficial Plexus Average Vessel Density (%) 27.81 52.64 51.27 53.43 52.16
r 0.562 0.258 0.522 0.435 0.095
P-value <0.001 0.02 <0.001 <0.001 0.38
ACCEPTED MANUSCRIPT
Plexus
Deep Capillary
whole
fovea
All eyes
50.52
27.81
52.64
51.27
53.43 52.16 53.69 54.36
Sickle SS
51.15
27.51
53.39
52.28
53.99 53.01 54.26 55.05
Sickle SC
49.50
28.43
51.44
49.34
52.86 50.57 52.97 52.41
0.13
0.57
0.11
0.05
0.36
51.02
27.95
53.22
49.23
27.87
51.09
0.11
0.96
0.09
26.45
parafovea temporal superior nasal inferior
57.22
56.02
57.73
56.90 58.22
56.95
58.47
58.00 58.78
27.52
54.91
53.38
55.94
54.14 56.16
0.04
0.34
0.03
0.04
0.09
0.04
52.22
53.73 52.85 54.07 55.21
26.29
58.24
57.30
58.75
57.97 58.93
48.87
52.47 49.91 53.07 52.19
27.14
54.52
52.69
55.08
53.63 56.66
0.59
0.01
0.01
0.02
0.04
M AN U
58.05
0.15
TE D
AC C
fovea
26.05
0.02
EP
P-values (SS vs SC) No NV (Stage 2 and below) NV (Stage 3 and above) P-values (no NV versus NV)
parafovea temporal superior nasal inferior whole
Plexus
SC
Superficial Capillary
RI PT
Table 6. Vessel density (%) for superficial and deep capillary plexus, with comparisons of density in various subfields according to genotype and presence of neovascularization (stage 3 or above). P values <0.05 considered statistically significant. NV=neovascularization.
0.34
0.12
0.20
0.28
0.02
0.04
0.06
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT