- Email: [email protected]
Age and gender influence on foveal avascular zone in healthy eyes
Experimental Eye Research 189 (2019) 107856
Contents lists available at ScienceDirect
Experimental Eye Research journal homepage: www.elsevier.com/locate/yexer
Age and gender influence on foveal avascular zone in healthy eyes a,b,c
a
a
T
a,b
Francisco Gómez-Ulla , Paula Cutrin , Paz Santos , Maribel Fernandez , Maximino Abraldesa,b, Jose Manuel Abalo-Lojoa,∗, Francisco Gonzaleza,b,c,∗∗ a b c
Service of Ophthalmology, Complejo Hospitalario Universitario de Santiago de Compostela, Santiago de Compostela, Spain Department of Surgery and CIMUS, University of Santiago de Compostela, Santiago de Compostela, Spain IDIS, Santiago de Compostela, Spain
A R T I C LE I N FO
A B S T R A C T
Keywords: Retina Foveal avascular zone OCT-A Macula
The foveal avascular zone (FAZ) is the capillary-free area in the central macula with high photoreceptor density and metabolic activity. In the present study we measured the superficial and deep macular foveal avascular zone (sFAZ, dFAZ) in the eyes of healthy adults of both sexes of various ages ranging from 10 to 69 years using optical coherence tomography angiography (OCT-A) in order to evaluate the influence of gender and age on FAZ size. A cross-sectional study was carried out in 240 eyes of 120 healthy subjects, OCT-A was performed by means of a Topcon swept source OCT. sFAZ and dFAZ areas were measured using the IMAGEnet6 software package. Subjects were grouped by age (six groups) and gender. The mean ± sd age of the subjects was 39.2 ± 17.4 years (50% women, 50% men), ranging from 10 to 69 years. The overall mean sFAZ size in women (0.297 ± 0.110 mm2) was significantly larger (p = 0.002) than in men (0.254 ± 0.098 mm2). Similarly, the overall mean dFAZ in women (0.322 ± 0.111 mm2) was significantly larger (p < 0.001) than in men (0.273 ± 0.099). However, when analyzed by age group, these gender differences appeared only in groups younger than 20 years old and older than 50 years old. Men did not show differences among the six age groups. In women, for both sFAZ and dFAZ, the 20–29 year old group had a smaller FAZ size than the 50–59 year old group. In conclusion for both sFAZ and dFAZ, women have larger areas than men, but this occurs only in the young and old age groups. In men, age does not seem to influence the size of the FAZ, but in women, both sFAZ and dFAZ were significantly smaller in younger than in older ages. These results suggest that retinal changes in retinal structure caused by aging may be different in woman than in men, probably reflecting the more hormonal variations known to exist with age in women.
1. Introduction Retinal imaging to assess retinal alterations has been widely used for disease diagnosis, grading and follow-up (Eladawi et al., 2018; Gómez-Ulla et al., 2002; Gonzalez et al., 2001). The optical coherence tomography angiography imaging modality (OCT-A) has recently emerged as an effective technique to obtain valuable information to assess macular disease. The foveal avascular zone (FAZ) is the capillaryfree area in the central macula with high photoreceptor density and metabolic activity (Jonas et al., 1992; Yu et al., 2005). The macular area has two separate capillary plexuses, known as the superficial and deep capillary plexuses (Spaide et al., 2015a,b). Although the size of the FAZ by itself does not seem to influence visual function in healthy eyes (Mammo et al., 2015; Samara et al., 2015; Laatikainen and Larinkari,
∗
1977), FAZ changes could be potential markers of foveal function and health status (Conrath et al., 2005; Arend et al., 1995; Bresnick et al., 1984). OCT-A lacks the undesirable problems of fluorescein angiography and allows for the visualization of the superficial and deep macular retinal capillary plexuses (Spaide et al., 2018). This makes it possible to neatly delimit the superficial (sFAZ) and deep (dFAZ) foveal avascular zones and to calculate their areas (de Carlo et al., 2015a,b). FAZ size reflects the status of the microcapillary bed in the macular area and has a strong correlation with the severity of capillary nonperfusion (Bresnick et al., 1984; Spaide et al., 2018; de Carlo et al., 2015a,b; Zeffren et al., 1990); changes in FAZ size are observed in retino-vascular diseases such as diabetic retinopathy, retinal vein occlusion or radiation retinopathy (Arend et al., 1995; de Carlo et al.,
Corresponding author. Correspondent author.Department of Surgery, University of Santiago de Compostela, Santiago de Compostela, Spain. E-mail addresses: [email protected] (J.M. Abalo-Lojo), [email protected] (F. Gonzalez).
∗∗
https://doi.org/10.1016/j.exer.2019.107856 Received 6 August 2019; Received in revised form 20 October 2019; Accepted 21 October 2019 Available online 22 October 2019 0014-4835/ © 2019 Elsevier Ltd. All rights reserved.
Experimental Eye Research 189 (2019) 107856
F. Gómez-Ulla, et al.
centered on the fovea in both eyes. This device has a scanning light wavelength of 1050 nm with a speed of 100,000 scans per second and involuntary eye movement compensation. An internal fixation light spot was used to center the scan in the fovea. The quality of the OCT-A was optimized with the “Auto” function included in the device in order to find the best position to acquire macular images. Images with a quality index below than 60 were discarded and subsequently repeated until adequate quality was obtained. Layer segmentation of the OCT-A images was performed automatically by the IMAGEnet6 software package (Topcon Medical Systems, Inc., Oakland, NJ). Two layer segmentations were used in this study, i.e. the superficial capillary plexus (SCP) and the deep capillary plexus (DCP). The SCP included the retinal volume from 2.6 to 15.6 μm below the inner limiting membrane. The DCP included the retinal volume from 15.6 to 70.2 μm below the inner limiting membrane. It has been shown in the retinas of monkeys that these two planes of capillaries bracket the inner nuclear layer (Snodderly et al., 1992). All images were obtained under mydriasis (1% tropicamide and 2.5% phenylephrine eyedrops).
2015a,b; Agemy et al., 2015; Veverka et al., 2015; Mansour et al., 1993) and are correlated with visual acuity in patients suffering from these diseases (Balaratnasingam et al., 2016). Age and sex seem to influence macular anatomy and vascularity. Early studies using OCT in healthy eyes reported changes in the macular profile in relation to age and sex (Alamouti and Funk, 2003; Chan et al., 2006; Eriksson and Alm, 2009; Grover et al., 2009). Subsequent studies reported significant differences in macular thickness amongst subjects of different ethnic backgrounds, gender and age (Adhi et al., 2012; Song et al., 2010; Ooto et al., 2011). Other reports however, failed to find any association between macular thickness and either gender or age (Chan et al., 2006; Grover et al., 2009; Sull et al., 2010). Therefore, demographic variations may be important parameters when assessing FAZ area for the diagnosis and follow-up of macular diseases. It has been reported that women have a larger FAZ area than men (Linderman et al., 2017), and that the macular vascular density is higher in women than in men aged older than 60 years. However, the sFAZ area is significantly smaller in older than in younger subjects (Coscas et al., 2016), which suggests the existence of an age-related reduction in the FAZ area. Tan et al. (2016) found significant correlations between sFAZ and dFAZ areas, central retinal thickness and sex, but not with age. Other authors reported that sFAZ is larger in women (Yu et al., 2015) while others have reported no differences in FAZ size between sexes (Samara et al., 2015). Takase et al. (2015) did not find any correlation in healthy eyes between age and the FAZ area for both sFAZ and dFAZ. However, Wu et al., (1985) and Laatikainen and Larinkari (1977) using fluorescein angiograms, found a positive correlation between FAZ size and age. In the present study we measured the sFAZ and dFAZ areas in the eyes of 120 healthy adults of both sexes of various ages ranging from 10 to 69 years using OCT-A in order to evaluate the influence of gender and age on FAZ size. To analyze in detail the effect of age on FAZ area, the subjects were divided in six age groups by decade.
2.3. Measurements of FAZ area Delimitation of the sFAZ and dFAZ areas was done manually in all cases by means of the IMAGEnet6 Angiographic Viewer (Topcon Medical Systems, Inc., Oakland, NJ). For this, the examiner first had to manually delimitate the perimeter of the FAZ on the image and then the software calculated the area of the delimited surface in mm2. The study subjects underwent two different OCT-A imaging sessions within a single visit with the images obtained by the same operator (PC, Observer 1). In the first session, a macular scan (OCT-A1) was obtained for each eye of each subject (120 subjects, 240 eyes) and data from both eyes of each subject were included in the analysis. In the second session, in 92 of these subjects (184 eyes), a second scan (OCT-A2) was obtained by the same operator. The measurements of the FAZ were performed by two independent examiners the same day the images were obtained (PC, Observer 1; PS, Observer 2). Observer 1 measured all scans (240 eyes, 120 subjects) obtained in the first session (OCT-A1, measurement T0), whereas Observer 2 measured the scans of 80 eyes of 40 of these subjects obtained in the same session (OCT-A1, measurement T1). Twenty days after the first session, Observer 1 performed a new measurement in the scans of 80 eyes of 40 subjects obtained in the first session (OCT-A1) (measurement T2). Additionally, Observer 1 performed a measurement in all scans obtained in the second session (OCT-A2, 184 eyes, 92 subjects, measurement T3). Table 1 summarizes these procedures.
2. Patients and methods 2.1. Subjects A total of 120 healthy Caucasian volunteers aged between 10 and 69 years (mean 39.2 ± 17.4 years, 50% women and 50% men) were enrolled in the study. In order to ensure age and sex homogeneity, subjects were divided into six age groups by decade, with 10 women (20 eyes) and 10 men (20 eyes) in each group (10–19, 20–29, 30–39, 40–49, 50–59, and 60–69 years old). None of the subjects were under the effect of any vasoactive substance for at least 12 h before OCT-A was performed. The inclusion criteria were, best corrected visual acuity equal or better than 20/20, refractive defect from −1.5 to +1.0 D of spherical equivalent, intraocular pressure < 21 mmHg, no media opacity, normal eye fundus, and normal macular OCT. The exclusion criteria were: previous or current history of ophthalmic disease related to the posterior segment, previous ocular surgery or traumatism, any systemic disease with ocular involvement, poor collaboration for complete ophthalmic examination, or resulting angiographic image with insufficient quality (blurring, artifacts) to perform reliable measurements. This study followed the tenets of the Declaration of Helsinki, and was authorized by the Comite de Etica de la Investigacion de SantiagoLugo. Informed consent was obtained from all subjects before ophthalmic exploration.
2.4. Statistical analysis To assess the reliability of our measurement procedures, we evaluated the intra-observer repeatability, the inter-observer Table 1 - Measurements made to evaluate the repeatability and reproducibility of the FAZ area assessment. Intra-observer repeatability was evaluated by calculating the ICC between T0 and T2. Inter-observer reproducibility was evaluated by calculating the CCC between T0 and T1. Inter-session reproducibility was evaluated by calculating the CCC between T0 and T3. The rightmost column indicates when the measurements were made after the images were obtained.
2.2. OCTA-A macular scans Retinal OCT-A macular scan images were obtained with a DRI OCT Triton Plus swept source OCT (Topcon Medical Systems, Inc., Oakland, NJ) that included a macular radial OCT and a 3 × 3 mm OCT-A
Measurement
OCT-A Session
N (eyes)
N (subjects)
Observer
T0 T1 T2 T3
OCT-A1 OCT-A1 OCT-A1 OCT-A2
240 80 80 184
120 40 40 92
Observer Observer Observer Observer
Measurement
1 2 1 1
Same day Same day 20 days later Same day
ICC: intraclass correlation coefficient; CCC: concordance correlation coefficient. 2
Experimental Eye Research 189 (2019) 107856
F. Gómez-Ulla, et al.
reproducibility and test-retest reliability. In order to evaluate the intra-observer repeatability of our measurements, we calculated the intraclass correlation coefficient (ICC) (Shrout and Fleiss, 1979) for measurements made in 80 eyes by Observer1 in T0 and T2. Values greater than 0.75 are indicative of good to excellent reliability (Koo and Li, 2016). To evaluate inter-observer reproducibility we calculated the concordance correlation coefficient (CCC) for the measurements performed in 80 eyes by Observer 1 (T0) and Observer 2 (T1) in images obtained in the first session (OCT-A1). The CCC measures the agreement between two variables (Lin, 1989) and ranges between 0 and 1; higher values indicate grater agreement and hence better reproducibility. To assess the test-retest reproducibility we calculated the CCC of the FAZ area measurements performed in 184 eyes by Observer 1(T0 and T3) in images obtained from two different imaging sessions (OCT-A1 and OCT-A2). The fitting of the data to a normal distribution was tested by the Kolmogorov-Smirnov test. To assess comparisons of data fitting a normal distribution, a two tailed Student's t-test was used. Otherwise a Mann-Whitney U test was used. For comparisons of more than two groups, the Kruskal-Wallis test was used. Differences were considered statistically significant when p < 0.05. Statistical analyses were performed with Microsoft Office Excel (Microsoft Corporation, Redmond, WA, USA), IBM SPSS Statistics v. 24 (IBM Corporation, Somers, NY, USA), and online calculators available on the internet. Statistical power calculations were made using STATISTICA software (StatSoft, Inc, Tulsa, OK, USA).
Fig. 1. - Correlation between superficial (sFAZ) and deep (dFAZ) FAZ in the 240 eyes investigated in our sample (r = 0.95). Deep FAZ was in most cases larger than superficial FAZ.
3.3. Gender and FAZ area The sFAZ area of women (mean 0.297 mm2, sd: 0.110) was significantly larger (two tailed t-test, p = 0.002, statistical power 0.91) than that of men (mean 0.254 mm2, sd: 0.098). Similarly, the dFAZ area in women (mean 0.322 mm2, sd: 0.111) was significantly larger (two tailed t-test, p < 0.001, statistical power 0.95) than that of men (mean 0.273 mm2, sd: 0.099). When the differences (Mann-Whitney test) between women and men were analyzed by age group, we found that the differences appeared only in those groups aged younger than 20 years and older than 50 years. Table 3 shows these results. For the 10–19 years of age group, the median sFAZ for women was 0.300 mm2, whereas for men it was 0.247 mm2. In the same age group, the median dFAZ for women was 0.314 mm2, whereas for men it was 0.255 mm2. Regarding the 50–59 years of age group the median sFAZ for women was 0.325 mm2 and that for men was 0.218 mm2, whereas for dFAZ, these values were 0.349 mm2 and 0.238 mm2, respectively. Finally, for the 60–69 years of age group, the median sFAZ for women was 0.312 mm2 and that for men was 0.222 mm2. These differences were statistically significant (p < 0.05) for both sFAZ and dFAZ areas and in all these instances, the FAZ area of women was larger than that of men. Fig. 2 shows the results related to age, gender, and sFAZ and dFAZ areas.
3. Results 3.1. Repeatability and reproducibility The repeatability and reproducibility assessment results are shown in Table 2. Intra-observer repeatability reached ICC values of 0.993 for sFAZ area and 0.985 for dFAZ area measurements. Inter-observer reproducibility reached CCC values of 0.924 for sFAZ area measurements, and 0.620 for dFAZ area measurements. Test-retest reproducibility showed CCC values of 0.974 for sFAZ area and 0.957 for dFAZ area.
3.2. Descriptive statistics The mean sFAZ area (n = 240 eyes) was 0.275 mm2 (sd: 0.106), and ranged from 0.060 to 0.545 mm2. The mean dFAZ area (n = 240 eyes) was 0.298 mm2 (sd: 0.108) and ranged from 0.068 to 0.576 mm2. Comparing all eyes, the dFAZ was significantly larger than the sFAZ (mean difference 0.023 mm2, two tailed paired t-test, p < 0.001, statistical power 0.73). The sFAZ and dFAZ area sizes were strongly correlated (r = 0.95) (Fig. 1). When comparing left and right eyes, we found no differences (paired two tailed t-test, p = 0.173) between the sFAZ of the right (mean 0.278 mm2, sd: 0.106) and left eyes (0.273 mm2, sd: 0.107). Similarly, there were no differences (paired two tailed t-test, p = 0.275) between the dFAZ of the right (mean 0.300 mm2, sd: 0.106) and left eyes (mean 0.295 mm2, sd: 0.109).
3.4. Age and FAZ area To assess the effect of age on FAZ area we grouped the subjects into six age groups as indicated previously, and used the Kruskal-Wallis test to see whether the distribution of FAZ areas was the same across these groups. In men, we found no differences between age groups. In women, however, the FAZ area distribution was not the same across groups for both sFAZ and dFAZ (p < 0.001). In both cases, there were statistically significant differences between the 20–29 year old group (sFAZ median: 0207 mm2, dFAZ median: 0.218 mm2) and the 50–59 year old group (sFAZ median: 0.325 mm2, dFAZ median: 0.349 mm2) (p = 0.031 for sFAZ and p = 0.022 for dFAZ). Fig. 2 shows the results related to age, gender, and sFAZ and dFAZ areas.
Table 2 - Intra-observer repeatability, and inter-observer and inter-session reproducibility of superficial (sFAZ) and deep (dFAZ) FAZ area measurements.
sFAZ dFAZ
Intra-observer repeatability ICC (95% CI)
Inter-observer reproducibility CCC (95% CI)
Test-retest reproducibility CCC (95% CI)
0.993 (0.988–0.995) 0.985 (0.958–0.993)
0.924 (0.887–0.949) 0.620 (0.512–0.709)
0.974 (0.966–0.981) 0.957 (0.943–0.967)
ICC: intraclass correlation coefficient; CCC: concordance correlation coefficient; CI: confidence interval. 3
Experimental Eye Research 189 (2019) 107856
F. Gómez-Ulla, et al.
Table 3 - Comparisons of FAZ between women and men by age group.
Table 4 -Reported superficial and deep FAZ areas calculated from OCT-A images. Values are mean mm2 ( ± sd).
Age group (years) 10–19
20–29
30–39
40–49
50–59
Authors
sFAZ
Carpineto et al.39 2016a
0.251 ( ± 0.096) 0.252 ( ± 0.096) 0.350 ( ± 0.110) 0.280 ( ± 0.100) 0.288 ( ± 0.136) 0.360 ( ± 0.110) 0.257 0.350 ( ± 0.013) 0.285 ( ± 0.150) 0.373 ( ± 0.109) 0.377 ( ± 0.112) 0.350 ( ± 0.013) 0.304 ( ± 0.132) 0.257 ( ± 0.104) 0.274 0.269 ( ± 0.092) 0.270 ( ± 0.090) 0.266 ( ± 0.097) 0.270 ( ± 0.101 0.250 ( ± 0.060) 0.240 0.474 ( ± 0.172) 0.301( ± 0.058) 0.275 ( ± 0.106)
dFAZ
60–69
Superficial FAZ: Women median Men median Mann-Whitney, p
0.300 0.247 0.011a
0.207 0.293 0.157
0.347 0.258 0.091
0.314 0.255 0.477
0.325 0.218 0.004a
0.312 0.222 0.016a
Deep FAZ: Women median Men median Mann-Whitney, p
0.314 0.255 0.011a
0.218 0.321 0.157
0.336 0.287 0.091
0.342 0.262 0.477
0.349 0.238 0.004a
0.338 0.228 0.016a
Choi et al.48 2017 Coscas et al.29 2016 de Carlo et al.40 2015 Di et al.57 2016 Freiberg et al.46 2016b Gadde et al.49 2016 Goudot et al.52 2017 Guo et al.41 2017a
a Indicates statistically significant differences (p < 0.05) between women and men.
Hussain and Hussain58 2016 Kuehlewein et al.42 2015 Linderman et al.28 2017 Magrath et al.47 2016 Mastropasqua et al.44 2017a
4. Discussion 4.1. Reliability of FAZ measurements
Samara et al.8 2015 Shahlaee et al.45 2016 Takase et al.32 2015 Tan et al.30 2016 Yu et al.31 2015 Mean of above results This report
Since OCT-A artifacts are common and can lead to misleading interpretations (Spaide et al., 2015a,b), the reliability when measuring FAZ areas using this technique has been addressed in many recent studies, which widely agree in reporting excellent repeatability and reproducibility as assessed by the ICC and CCC (Linderman et al., 2017; Coscas et al., 2016; Tan et al., 2016; Carpineto et al., 2016; de Carlo et al., 2015a,b; Guo et al., 2017; Kuehlewein et al., 2015; La Spina et al., 2017; Mastropasqua et al., 2017; Shahlaee et al., 2016; Freiberg et al., 2016). In a recent study performed on healthy eyes, the FAZ area measurements showed high repeatability and reproducibility (Carpineto et al., 2016) as well as high sensitivity and specificity (Freiberg et al., 2016). Linderman et al. (2017) found that manual and automatic segmentation of FAZ areas in OCT-A images also showed excellent repeatability and reliability. For sFAZ measurements, the reported values of ICC range from 0.850 (Coscas et al., 2016), to 0.998 (Carpineto et al., 2016), whereas for dFAZ these values range from 0.840 (Shahlee et al. 2016) to 0.960 (Coscas et al., 2016). Reported CCC values for sFAZ measurements range from 0.984 (Guo et al., 2017) to 0.999 (Carpineto et al., 2016), whereas for dFAZ the reported CCC is < 0.850 (La Spina et al., 2017). We found ICC values of 0.993 for sFAZ and 0.985 for dFAZ measurements, and CCC values of 0.924 for sFAZ and 0.620 for dFAZ measurements. Therefore our values are in agreement with those reported by other authors, indicating high reliability in our measurements. We found however a low CCC value when dFAZ areas were measured. La Spina et al. (2017) also reported a low CCC value for
0.370 (0.120)
0.341 0.490 ( ± 0.012) 0.398 ( ± 0.138)
0.490 ( ± 0.012) 0.486 ( ± 0.162) 0.364
0.495 ( ± 0.227) 0.340 ( ± 0.116) 0.380 ( ± 0.110) 0.380 0.424 ( ± 0.067) 0.298 ( ± 0.108)
a
The measurements were made by two observers. The area shown here was calculated from the reported diameters, assuming it was circular. b
dFAZ area measurements. This may occur because dFAZ is less well delimited than sFAZ, which makes the delineation of dFAZ much less sharp; therefore a higher variability should be expected in the measurements of dFAZ area. A recent study found consistent measurements of sFAZ and dFAZ size in OCT-A images taken with the same machine and technique, although significant differences were found when two different imaging devices were used (Magrath et al., 2017). We used a single machine to carry out the present study, but we obtained two different OCT-A scans of 184 eyes in which the FAZ areas were measured by the same observer. These measurements demonstrated excellent agreement (CCC values of 0.974 and 0.957 for sFAZ and dFAZ areas, respectively). Since we used a single machine we cannot provide results on the effect of using different machines on the variability of these measurements.
Figure 2. -Age and FAZ area in men and women. There were no differences among age groups in either sFAZ or dFAZ area in the men group. In women there were differences between the age groups of 20–29 years and 50–69 years for both sFAZ area (p = 0.030) and dFAZ area (p = 0.019). 4
Experimental Eye Research 189 (2019) 107856
F. Gómez-Ulla, et al.
this increase was faster in men than in women. Using fluorescein angiography and grouping the subjects by age, Laatikaninen and Larinkari found that the mean diameter of the FAZ was 0.530 mm in patients younger than 39 years and 0.610 mm after the age of 40 (Laatikainen and Larinkari, 1977). Our results show no significant changes with age in men, but women showed an increase in the size of both sFAZ and dFAZ with age. There is evidence that retinal changes occur with age. Although the origin of these changes is not clear, hormonal variations through life may play a role. An abrupt decline in hormonal activity causes the symptoms observed in menopausal women. A decrease in circulating testosterone induces a number of physical changes in aging men. Estrogen receptor expression has been found in bovine and rodent retinas (Kobayashi et al., 1998). There is evidence that the human eye has estrogen receptors and that reduced levels of circulating estrogens in women after menopause may lead to changes in the expression of estrogen receptors in ocular tissues (Ogueta et al., 1999). A large number of estrogen-induced proteins have been identified, and most of these proteins can be found in human eye tissues, including the retina (Yu et al., 1994). Since women have more drastic hormonal variations through life than men, this may explain why we observed changes in macular capillaries related to age in women but not in men.
4.2. FAZ area size Table 4 shows the reported FAZ areas by other authors along with our results. We found a mean sFAZ area of 0.275 mm2, which is in agreement with other reported values in healthy eyes for sFAZ area using OCT-A, i.e. from 0.250 mm2 (Takase et al., 2015) to 0.350 mm2 (Choi et al., 2017; Gadde et al., 2016). Studies using fluorescein angiography to analyze the FAZ size in healthy eyes demonstrated more variability those obtained using OCT-A, with areas ranging from 0.205 mm2 to 0.405 mm2 (Mansour et al., 1993; Zheng et al., 2010). The mean dFAZ area for our dataset was 0.298 mm2, which is lower than previously reported values, ranging from 340 mm2 (Shahlaee et al., 2016) to 0.495 mm2 (Samara et al., 2015). This difference may be explained because of the difficulties in delineating the dFAZ perimeter. Indeed, as other authors have reported (La Spina et al., 2017) this difficulty is reflected in the lower CCC found when assessing the reliability of dFAZ area measurements. All reported values of sFAZ and dFAZ areas showed higher values for the latter (Table 4). For instance, Coscas et al. reported a mean sFAZ area of 0.280 mm2 and a mean dFAZ area of 0.370 mm2; the latter was significantly larger (Coscas et al., 2016). In agreement with previous studies, our results also show that the dFAZ area is larger than the sFAZ area. The larger size of the dFAZ can be explained in part because in the deep capillary plexus the vessels are displaced into the more internal retinal layers as these layers narrow at the fovea (Coscas et al., 2016).
4.5. Limitations of the study Among the limitations of this study is the limited number of subjects; a larger number of subjects would likely corroborate the findings reported here. Because our sample size is small, future studies will need a larger population based study design, exploring the effects of age, sex, and important factors such as axial eye length. There are also some technical limitations to the currently available devices. For instance, when the capillary flow rate falls below the threshold detected by OCTA algorithms, some vessels may not be completely visualized and therefore FAZ area measurements may be inaccurate. These technical limitations are inherent to the OCT machines and currently cannot be resolved. The dFAZ has poorly defined borders when compared with sFAZ. Therefore, there is difficulty in manually measuring this area, showing a larger variability than the sFAZ area measurements. We believe however that this may not have influenced our results as there was consistency in the differences found between age groups in sFAZ and dFAZ. The refractive error also affects the dimension of the FAZ as measured in fundus images. It was found however that the correlation between refractive error and FAZ dimensions was low and not statistically significant (Bresnick et al., 1984). This study concluded that it was unlikely that the effect on the image magnification contributed meaningfully to the FAZ measurements. Tan et al. (2016) performed univariate linear regression analysis of factors affecting the FAZ size and found that emmetropes and low myopes do not show a significant relationship between spherical equivalent (p = 0.223) and axial length (p = 0.469) with the FAZ size. We did not correct our data for refractive error, but since our subjects were emmetropes or low myopes, we believe that these optical factors did not significantly influence our measurements. The axial length may also be a significant source of error for measuring FAZ area (Linderman et al., 2018), but unfortunately a correction method for axial length when calculating the FAZ area is not currently available on OCT-A devices. While axial length may not be critical when comparing multiple scans from the same subject, it may be a limitation when comparing FAZ area measurements across individuals. Lindermann et al. (Linderman et al., 2017) observed that the error in FAZ area estimates as a function of axial length has an average error of 8.29%, with an absolute maximum error of 0.07 mm2. Cruickshank and Logan (2018) found a relationship between axial length and refractive error (mean spherical equivalent). By using the
4.3. Gender effect on FAZ Macular structural characteristics may be influenced by gender. For instance, Adhi et al. (2012) using a spectral domain OCT system found that men had greater foveal and macular thickness compared to women. Song et al. reported that the inner macular thickness and the overall macular volume were significantly smaller in women (Song et al., 2010). Other authors, however, have found no significant gender differences in macular or foveal thickness (Chan et al., 2006). Some authors have failed to find changes related to gender in macular vasculature, in terms of flow density or FAZ area (Samara et al., 2015; Coscas et al., 2016; Alnawaiseh et al., 2018). On the contrary, other studies report that women have larger FAZ areas than men (Linderman et al., 2017; Tan et al., 2016). Yu et al. (2015) reported similar findings in the healthy eyes of Chinese individuals, with a larger FAZ area in women than in men. Our overall results support the observation that the FAZ area in women is significantly larger (mean 0.297 mm2) than in men (mean 0.254 mm2). 4.4. Age and FAZ Several authors analyzing OCT images have found that retinal thickness declines with age (Alamouti and Funk, 2003; Eriksson and Alm, 2009; Song et al., 2010). However other studies have failed to find such a relationship between age and macular or foveal thickness (Chan et al., 2006; Adhi et al., 2012; Sull et al., 2010). Age-related anatomical changes in the macular region should be concurrent with changes in the vasculature and hence in FAZ area; however, studies reporting on the influence of age on FAZ size are somehow confusing. Some have found no relationship (Samara et al., 2015; Linderman et al., 2017; Tan et al., 2016; Takase et al., 2015; Gadde et al., 2016; Goudot et al., 2017) while others have found a relationship between age and FAZ size. Using fluorescein angiography, Wu et al., (1985) found a positive correlation between FAZ area and age, but Coscas et al., (2016) found that the superficial FAZ area was significantly smaller in older subjects. Alnawaiseh et al., (2018) found that older subjects had lower macular flow density. Bresnick et al. (1984) using fluorescein angiography, found that the FAZ diameter decreases with increasing age in healthy control eyes. Yu et al. (2015) reported in healthy Chinese individuals that the FAZ area increased by an average of 1.48% annually, and that 5
Experimental Eye Research 189 (2019) 107856
F. Gómez-Ulla, et al.
equation reported by these authors, the estimated axial length of the eyes included in our study ranged from 23 to 24 mm. If our measurements were corrected for axial length according to the equations reported by Linderman et al. (2017) for eyes with 23 mm the FAZ area would have an underestimation of 0.018 mm2, whereas for eyes of 24 mm, this would be 0.000 mm2. Therefore we believe that the axial length should not have a significant impact on the results we have obtained from our data set.
s40942-015-0005-8. Eladawi, N., Elmogy, M.M., Ghazal, M., et al., 2018. Classification of retinal diseases based on OCT images. Front. Biosci. 23, 247–264. Eriksson, U., Alm, A., 2009. Macular thickness decreases with age in normal eyes: a study on the macular thickness map protocol in the Stratus OCT. Br. J. Ophthalmol. 93, 1448–1452. Freiberg, F.J., Pfau, M., Wons, J., 2016. Optical coherence tomography angiography of the foveal avascular zone in diabetic retinopathy. Graefes Arch. Clin. Exp. Ophthalmol. 254, 1051–1058. Gadde, S.G., Anegondi, N., Bhanushali, D., et al., 2016. Quantification of vessel density in retinal optical coherence tomography angiography images using local fractal dimension vessel density in OCTA images. Invest. Ophthalmol. Vis. Sci. 57, 246–252. Gómez-Ulla, F., Fernandez, M.I., Gonzalez, F., et al., 2002. Digital retinal images and teleophthalmology for detecting and grading diabetic retinopathy. Diabetes Care 25, 1384–1389. Gonzalez, F., Iglesias, R., Suarez, A., et al., 2001. Teleophthalmology link between a primary health care centre and a reference hospital. Med. Inform. Internet Med. 26, 251–263. Goudot, M.M., Sikorav, A., Semoun, O., et al., 2017. Parafoveal OCT angiography features in diabetic patients without clinical diabetic retinopathy: a qualitative and quantitative analysis. J. Ophthalmol. 2017https://doi.org/10.1155/2017/8676091. 8676091. Grover, S., Murhty, R.K., Brar, V.S., et al., 2009. Normative data for macular thickness by high-definition Spectral-domain optical coherence tomography (Spectralis). Am. J. Ophthalmol. 148, 266–271. Guo, J., She, X., Liu, X., et al., 2017. Repeatability and reproducibility of foveal avascular zone area measurements using angioplex spectral domain optical coherence tomography angiography in healthy subjects. Ophthalmologica 237, 21–28. Jonas, J.B., Schneider, U., Naumann, G.O., 1992. Count and density of human retinal photoreceptors. Graefes Arch. Clin. Exp. Ophthalmol. 230, 505–510. Kobayashi, K., Kobayashi, H., Ueda, M., et al., 1998. Estrogen receptor expression in bovine and rat retinas. Invest. Ophthalmol. Vis. Sci. 39, 2105–2110. Koo, T.K., Li, M.Y., 2016. A guideline of selecting and reporting intraclass correlation coefficients for reliability research. J. Chiropr. Med. 15, 155–163. Kuehlewein, L., Tepelus, T.C., An, L., et al., 2015. Noninvasive visualization and analysis of the human parafoveal capillary network using swept source OCT optical microangiography. Invest. Ophthalmol. Vis. Sci. 56, 3984–3988. La Spina, C., Carnevali, A., Marchese, A., et al., 2017. Reproducibility and reliability of optical coherence tomography angiography for foveal avascular zone evaluation and measurement in different settings. Retina 37, 1636–1641. Laatikainen, L., Larinkari, J., 1977. Capillary-free area of the fovea with advancing age. Invest. Ophthalmol. Vis. Sci. 16, 1154–1157. Lin, L.I., 1989. A concordance correlation coefficient to evaluate reproducibility. Biometrics 45, 255–268. Linderman, R., Salmon, A.E., Strampe, M.1, et al., 2017. Assessing the accuracy of foveal avascular zone measurements using optical coherence tomography angiography: segmentation and scaling. Trans. Vis. Sci. Tech. 6, 16. https://doi.org/10.1167/tvst. 6.3.16. Linderman, R.E., Muthiah, M.N., Omoba, S.B., et al., 2018. Variability of foveal avascular zone metrics derived from optical coherence tomography angiography images. Transl. Vis. Sci. Technol. 7, 20. Magrath, G.N., Say, E.A.T., Sioufi, K., et al., 2017. Variability in foveal avascular zone and capillary density using optical coherence tomography angiography machines in healthy eyes. Retina 37, 2102–2111. Mammo, Z., Balaratnasingam, C., Yu, P., et al., 2015. Quantitative noninvasive angiography of the fovea centralis using speckle variance optical coherence tomography. Invest. Ophthalmol. Vis. Sci. 56, 5074–5086. Mansour, A.M., Schachat, A., Bodiford, G., et al., 1993. Foveal avascular zone in diabetes mellitus. Retina 13, 125–128. Mastropasqua, R., Toto, L., Mattei, P.A., et al., 2017. Reproducibility and repeatability of foveal avascular zone area measurements using swept-source optical coherence tomography angiography in healthy subjects. Eur. J. Ophthalmol. 27, 336–341. Ogueta, S.B., Schwartz, S.D., Yamashita, C.K., et al., 1999. Estrogen receptor in the human eye: influence of gender and age on gene expression. Invest. Ophthalmol. Vis. Sci. 40, 1906–1911. Ooto, S.1., Hangai, M., Tomidokoro, A., et al., 2011. Effects of age, sex, and axial length on the three-dimensional profile of normal macular layer structures. Invest. Ophthalmol. Vis. Sci. 52, 8769–8779. Samara, W.A., Say, E.A., Khoo, C.T., et al., 2015. Correlation of foveal avascular zone size with foveal morphology in normal eyes using optical coherence tomography angiography. Retina 35, 2188–2195. Shahlaee, A., Pefkianaki, M., Hsu, J., et al., 2016. Measurement of foveal avascular zone dimensions and its reliability in healthy eyes using optical coherence tomography angiography. Am. J. Ophthalmol. 161, 50–55. Shrout, P.E., Fleiss, J.L., 1979. Intraclass correlations: uses in assessing rater reliability. Psychol. Bull. 86, 420–428. Snodderly, D.M., Weinhaus, R.S., Choi, J.C., 1992. Neural-vascular relationships in central retina of macaque monkeys (Macaca fascicularis). J. Neurosci. 12, 1169–1193. Song, W.K., Lee, S.C., Lee, E.S., et al., 2010. Macular thickness variations with sex, age, and axial length in healthy subjects: a spectral domain-optical coherence tomography study. Invest. Ophthalmol. Vis. Sci. 51, 3913–3918. Spaide, R.F., Fujimoto, J.G., Waheed, N.K., 2015a. Image artifacts in optical coherence angiography. Retina 35, 2163–2180. Spaide, R.F., Fujimoto, J.G., Waheed, N.K., et al., 2018. Optical coherence tomography angiography. Prog. Retin. Eye Res. 64, 1–55. Spaide, R.F., Klancnik Jr., J.M., Cooney, M.J., 2015b. Retinal vascular layers imaged by
5. Conclusions In our study the patients were grouped by age and gender and this allowed us to obtain data to demonstrate that both sFAZ and dFAZ are larger in women than in men only in young and older age groups. In men age does not influence FAZ size, but in women there is age effect in sFAZ and dFAZ in such way that their area is smaller in younger than in older ages. These changes could be induced by age related hormonal variations, which are more pronounced in women than in men. Since ocular tissues express estrogen receptors, future studies may clarify their possible role in the foveal microvascular age related changes we found in women. OCT-A is reproducible and widely available in ophthalmology clinics, therefore the results we report should be helpful for diagnosis and follow-up of macular diseases. Declaration of competing interest No conflicting relationship exists for any author. Acknowledgments Funded by the ISCIII-Subdireccion General de Evaluacion y Fomento de la Investigacion (RD16/0008/003), cofunded by the European Regional Development Fund (FEDER). References Adhi, M., Aziz, S., Muhammad, K., et al., 2012. Macular thickness by age and gender in healthy eyes using spectral domain optical coherence tomography. PLoS One 7, e37638. https://doi.org/10.1371/journal.pone.0037638. Agemy, S.A., Scripsema, N.K., Shah, C.M., et al., 2015. Retinal vascular perfusion density mapping using optical coherence tomography angiography in normals and diabetic retinopathy patients. Retina 35, 2353–2363. Alamouti, B., Funk, J., 2003. Retinal thickness decreases with age: an OCT study. Br. J. Ophthalmol. 87, 899–901. Alnawaiseh, M., Brand, C., Lauermann, J.L., et al., 2018. Messung der Flussdichte mittels OCT-angiographie. Der Ophthalmologe 115, 659–662. Arend, O., Wolf, S., Harris, A., et al., 1995. The relationship of macular microcirculation to visual acuity in diabetic patients. Arch. Ophthalmol. 113, 610–614. Balaratnasingam, C., Inoue, M., Ahn, S., et al., 2016. Visual acuity is correlated with the area of the foveal avascular zone in diabetic retinopathy and retinal vein occlusion. Ophthalmology 123, 2352–2367. Bresnick, G.H., Condit, R., Syrjala, S., et al., 1984. Abnormalities of the foveal avascular zone in diabetic retinopathy. Arch. Ophthalmol. 102, 1286–1293. Carpineto, P., Mastropasqua, R., Marchini, G., et al., 2016. Reproducibility and repeatability of foveal avascular zone measurements in healthy subjects by optical coherence tomography angiography. Br. J. Ophthalmol. 100, 671–676. Chan, A., Duker, J.S., Ko, T.H., et al., 2006. Normal macular thickness measurements in healthy eyes using Stratus optical coherence tomography. Arch. Ophthalmol. 124, 193–198. Choi, J., Kwon, J., Shin, J.W., et al., 2017. Quantitative optical coherence tomography angiography of macular vascular structure and foveal avascular zone in glaucoma. PLoS One 12, e0184948. https://doi.org/10.1371/journal.pone.0184948. Conrath, J., Giorgi, R., Raccah, D., et al., 2005. Foveal avascular zone in diabetic retinopathy: quantitative vs qualitative assessment. Eye (Lond) 19, 322–326. Coscas, F., Sellam, A., Glacet-Bernard, A., et al., 2016. Normative data for vascular density in superficial and deep capillary plexuses of healthy adults assessed by optical coherence tomography angiography. Invest. Ophthalmol. Vis. Sci. 57, 211–223. Cruickshank, F.E., Logan, N.S., 2018. Optical 'dampening' of the refractive error to axial length ratio: implications for outcome measures in myopia control studies. Ophthalmic Physiol. Opt. 38, 290–297. de Carlo, T.E., Chin, A.T., Bonini Filho, M.A., et al., 2015a. Detection of microvascular changes in eyes of patients with diabetes but not clinical diabetic retinopathy using optical coherence tomography angiography. Retina 35, 2229–2235. de Carlo, T.E., Romano, A., Waheed, N.K., et al., 2015b. A review of optical coherence tomography angiography (OCTA). Int. J. Retin. Vitr. 1, 5. https://doi.org/10.1186/
6
Experimental Eye Research 189 (2019) 107856
F. Gómez-Ulla, et al. fluorescein angiography and optical coherence tomography angiography. JAMA Ophthalmol. 133, 45–50. Sull, A.C., Vuong, L.N., Price, L.L., 2010. Comparison of spectral/Fourier domain optical coherence tomography instruments for assessment of normal macular thickness. Retina 30, 235–245. Takase, N., Nozaki, M., Kato, A., et al., 2015. Enlargement of foveal avascular zone in diabetic eyes evaluated by en face optical coherence tomography angiography. Retina 35, 2377–2383. Tan, C.S., Lim, L.W., Chow, V.S., et al., 2016. Optical coherence tomography angiography evaluation of the parafoveal vasculature and its relationship with ocular factors. Invest. Ophthalmol. Vis. Sci. 57, OCT224–O230. Veverka, K.K., AbouChehade, J.E., Iezzi Jr., R., et al., 2015. Noninvasive grading of radiation retinopathy: the use of optical coherence tomography angiography. Retina 35, 2400–2410. Wu, L.Z., Huang, Z.S., Wu, D.Z., et al., 1985. Characteristics of the capillary-free zone in
the normal human macula. Jpn. J. Ophthalmol. 29, 406–411. Yu, D.Y., Cringle, S.J., Su, E.N., 2005. Intraretinal oxygen distribution in the monkey retina and the response to systemic hyperoxia. Invest. Ophthalmol. Vis. Sci. 46, 4728–4733. Yu, J., Jiang, C., Wang, X., et al., 2015. Macular perfusion in healthy Chinese: an optical coherence tomography angiogram study. Invest. Ophthalmol. Vis. Sci. 56, 3212–3217. Yu, M.C., Li, W.W., Liu, K., et al., 1994. An immunohistochemical study of the c-fos protooncogene in the developing human retina. Neuroscience 60, 983–987. Zeffren, B.S.1., Applegate, R.A., Bradley, A., et al., 1990. Retinal fixation point location in the foveal avascular zone. Invest. Ophthalmol. Vis. Sci. 31 2099–105. Zheng, Y., Gandhi, J.S., Stangos, A.N., et al., 2010. Automated segmentation of foveal avascular zone in fundus fluorescein angiography. Invest. Ophthalmol. Vis. Sci. 51, 3653–3659.
7
Contents lists available at ScienceDirect
Experimental Eye Research journal homepage: www.elsevier.com/locate/yexer
Age and gender influence on foveal avascular zone in healthy eyes a,b,c
a
a
T
a,b
Francisco Gómez-Ulla , Paula Cutrin , Paz Santos , Maribel Fernandez , Maximino Abraldesa,b, Jose Manuel Abalo-Lojoa,∗, Francisco Gonzaleza,b,c,∗∗ a b c
Service of Ophthalmology, Complejo Hospitalario Universitario de Santiago de Compostela, Santiago de Compostela, Spain Department of Surgery and CIMUS, University of Santiago de Compostela, Santiago de Compostela, Spain IDIS, Santiago de Compostela, Spain
A R T I C LE I N FO
A B S T R A C T
Keywords: Retina Foveal avascular zone OCT-A Macula
The foveal avascular zone (FAZ) is the capillary-free area in the central macula with high photoreceptor density and metabolic activity. In the present study we measured the superficial and deep macular foveal avascular zone (sFAZ, dFAZ) in the eyes of healthy adults of both sexes of various ages ranging from 10 to 69 years using optical coherence tomography angiography (OCT-A) in order to evaluate the influence of gender and age on FAZ size. A cross-sectional study was carried out in 240 eyes of 120 healthy subjects, OCT-A was performed by means of a Topcon swept source OCT. sFAZ and dFAZ areas were measured using the IMAGEnet6 software package. Subjects were grouped by age (six groups) and gender. The mean ± sd age of the subjects was 39.2 ± 17.4 years (50% women, 50% men), ranging from 10 to 69 years. The overall mean sFAZ size in women (0.297 ± 0.110 mm2) was significantly larger (p = 0.002) than in men (0.254 ± 0.098 mm2). Similarly, the overall mean dFAZ in women (0.322 ± 0.111 mm2) was significantly larger (p < 0.001) than in men (0.273 ± 0.099). However, when analyzed by age group, these gender differences appeared only in groups younger than 20 years old and older than 50 years old. Men did not show differences among the six age groups. In women, for both sFAZ and dFAZ, the 20–29 year old group had a smaller FAZ size than the 50–59 year old group. In conclusion for both sFAZ and dFAZ, women have larger areas than men, but this occurs only in the young and old age groups. In men, age does not seem to influence the size of the FAZ, but in women, both sFAZ and dFAZ were significantly smaller in younger than in older ages. These results suggest that retinal changes in retinal structure caused by aging may be different in woman than in men, probably reflecting the more hormonal variations known to exist with age in women.
1. Introduction Retinal imaging to assess retinal alterations has been widely used for disease diagnosis, grading and follow-up (Eladawi et al., 2018; Gómez-Ulla et al., 2002; Gonzalez et al., 2001). The optical coherence tomography angiography imaging modality (OCT-A) has recently emerged as an effective technique to obtain valuable information to assess macular disease. The foveal avascular zone (FAZ) is the capillaryfree area in the central macula with high photoreceptor density and metabolic activity (Jonas et al., 1992; Yu et al., 2005). The macular area has two separate capillary plexuses, known as the superficial and deep capillary plexuses (Spaide et al., 2015a,b). Although the size of the FAZ by itself does not seem to influence visual function in healthy eyes (Mammo et al., 2015; Samara et al., 2015; Laatikainen and Larinkari,
∗
1977), FAZ changes could be potential markers of foveal function and health status (Conrath et al., 2005; Arend et al., 1995; Bresnick et al., 1984). OCT-A lacks the undesirable problems of fluorescein angiography and allows for the visualization of the superficial and deep macular retinal capillary plexuses (Spaide et al., 2018). This makes it possible to neatly delimit the superficial (sFAZ) and deep (dFAZ) foveal avascular zones and to calculate their areas (de Carlo et al., 2015a,b). FAZ size reflects the status of the microcapillary bed in the macular area and has a strong correlation with the severity of capillary nonperfusion (Bresnick et al., 1984; Spaide et al., 2018; de Carlo et al., 2015a,b; Zeffren et al., 1990); changes in FAZ size are observed in retino-vascular diseases such as diabetic retinopathy, retinal vein occlusion or radiation retinopathy (Arend et al., 1995; de Carlo et al.,
Corresponding author. Correspondent author.Department of Surgery, University of Santiago de Compostela, Santiago de Compostela, Spain. E-mail addresses: [email protected] (J.M. Abalo-Lojo), [email protected] (F. Gonzalez).
∗∗
https://doi.org/10.1016/j.exer.2019.107856 Received 6 August 2019; Received in revised form 20 October 2019; Accepted 21 October 2019 Available online 22 October 2019 0014-4835/ © 2019 Elsevier Ltd. All rights reserved.
Experimental Eye Research 189 (2019) 107856
F. Gómez-Ulla, et al.
centered on the fovea in both eyes. This device has a scanning light wavelength of 1050 nm with a speed of 100,000 scans per second and involuntary eye movement compensation. An internal fixation light spot was used to center the scan in the fovea. The quality of the OCT-A was optimized with the “Auto” function included in the device in order to find the best position to acquire macular images. Images with a quality index below than 60 were discarded and subsequently repeated until adequate quality was obtained. Layer segmentation of the OCT-A images was performed automatically by the IMAGEnet6 software package (Topcon Medical Systems, Inc., Oakland, NJ). Two layer segmentations were used in this study, i.e. the superficial capillary plexus (SCP) and the deep capillary plexus (DCP). The SCP included the retinal volume from 2.6 to 15.6 μm below the inner limiting membrane. The DCP included the retinal volume from 15.6 to 70.2 μm below the inner limiting membrane. It has been shown in the retinas of monkeys that these two planes of capillaries bracket the inner nuclear layer (Snodderly et al., 1992). All images were obtained under mydriasis (1% tropicamide and 2.5% phenylephrine eyedrops).
2015a,b; Agemy et al., 2015; Veverka et al., 2015; Mansour et al., 1993) and are correlated with visual acuity in patients suffering from these diseases (Balaratnasingam et al., 2016). Age and sex seem to influence macular anatomy and vascularity. Early studies using OCT in healthy eyes reported changes in the macular profile in relation to age and sex (Alamouti and Funk, 2003; Chan et al., 2006; Eriksson and Alm, 2009; Grover et al., 2009). Subsequent studies reported significant differences in macular thickness amongst subjects of different ethnic backgrounds, gender and age (Adhi et al., 2012; Song et al., 2010; Ooto et al., 2011). Other reports however, failed to find any association between macular thickness and either gender or age (Chan et al., 2006; Grover et al., 2009; Sull et al., 2010). Therefore, demographic variations may be important parameters when assessing FAZ area for the diagnosis and follow-up of macular diseases. It has been reported that women have a larger FAZ area than men (Linderman et al., 2017), and that the macular vascular density is higher in women than in men aged older than 60 years. However, the sFAZ area is significantly smaller in older than in younger subjects (Coscas et al., 2016), which suggests the existence of an age-related reduction in the FAZ area. Tan et al. (2016) found significant correlations between sFAZ and dFAZ areas, central retinal thickness and sex, but not with age. Other authors reported that sFAZ is larger in women (Yu et al., 2015) while others have reported no differences in FAZ size between sexes (Samara et al., 2015). Takase et al. (2015) did not find any correlation in healthy eyes between age and the FAZ area for both sFAZ and dFAZ. However, Wu et al., (1985) and Laatikainen and Larinkari (1977) using fluorescein angiograms, found a positive correlation between FAZ size and age. In the present study we measured the sFAZ and dFAZ areas in the eyes of 120 healthy adults of both sexes of various ages ranging from 10 to 69 years using OCT-A in order to evaluate the influence of gender and age on FAZ size. To analyze in detail the effect of age on FAZ area, the subjects were divided in six age groups by decade.
2.3. Measurements of FAZ area Delimitation of the sFAZ and dFAZ areas was done manually in all cases by means of the IMAGEnet6 Angiographic Viewer (Topcon Medical Systems, Inc., Oakland, NJ). For this, the examiner first had to manually delimitate the perimeter of the FAZ on the image and then the software calculated the area of the delimited surface in mm2. The study subjects underwent two different OCT-A imaging sessions within a single visit with the images obtained by the same operator (PC, Observer 1). In the first session, a macular scan (OCT-A1) was obtained for each eye of each subject (120 subjects, 240 eyes) and data from both eyes of each subject were included in the analysis. In the second session, in 92 of these subjects (184 eyes), a second scan (OCT-A2) was obtained by the same operator. The measurements of the FAZ were performed by two independent examiners the same day the images were obtained (PC, Observer 1; PS, Observer 2). Observer 1 measured all scans (240 eyes, 120 subjects) obtained in the first session (OCT-A1, measurement T0), whereas Observer 2 measured the scans of 80 eyes of 40 of these subjects obtained in the same session (OCT-A1, measurement T1). Twenty days after the first session, Observer 1 performed a new measurement in the scans of 80 eyes of 40 subjects obtained in the first session (OCT-A1) (measurement T2). Additionally, Observer 1 performed a measurement in all scans obtained in the second session (OCT-A2, 184 eyes, 92 subjects, measurement T3). Table 1 summarizes these procedures.
2. Patients and methods 2.1. Subjects A total of 120 healthy Caucasian volunteers aged between 10 and 69 years (mean 39.2 ± 17.4 years, 50% women and 50% men) were enrolled in the study. In order to ensure age and sex homogeneity, subjects were divided into six age groups by decade, with 10 women (20 eyes) and 10 men (20 eyes) in each group (10–19, 20–29, 30–39, 40–49, 50–59, and 60–69 years old). None of the subjects were under the effect of any vasoactive substance for at least 12 h before OCT-A was performed. The inclusion criteria were, best corrected visual acuity equal or better than 20/20, refractive defect from −1.5 to +1.0 D of spherical equivalent, intraocular pressure < 21 mmHg, no media opacity, normal eye fundus, and normal macular OCT. The exclusion criteria were: previous or current history of ophthalmic disease related to the posterior segment, previous ocular surgery or traumatism, any systemic disease with ocular involvement, poor collaboration for complete ophthalmic examination, or resulting angiographic image with insufficient quality (blurring, artifacts) to perform reliable measurements. This study followed the tenets of the Declaration of Helsinki, and was authorized by the Comite de Etica de la Investigacion de SantiagoLugo. Informed consent was obtained from all subjects before ophthalmic exploration.
2.4. Statistical analysis To assess the reliability of our measurement procedures, we evaluated the intra-observer repeatability, the inter-observer Table 1 - Measurements made to evaluate the repeatability and reproducibility of the FAZ area assessment. Intra-observer repeatability was evaluated by calculating the ICC between T0 and T2. Inter-observer reproducibility was evaluated by calculating the CCC between T0 and T1. Inter-session reproducibility was evaluated by calculating the CCC between T0 and T3. The rightmost column indicates when the measurements were made after the images were obtained.
2.2. OCTA-A macular scans Retinal OCT-A macular scan images were obtained with a DRI OCT Triton Plus swept source OCT (Topcon Medical Systems, Inc., Oakland, NJ) that included a macular radial OCT and a 3 × 3 mm OCT-A
Measurement
OCT-A Session
N (eyes)
N (subjects)
Observer
T0 T1 T2 T3
OCT-A1 OCT-A1 OCT-A1 OCT-A2
240 80 80 184
120 40 40 92
Observer Observer Observer Observer
Measurement
1 2 1 1
Same day Same day 20 days later Same day
ICC: intraclass correlation coefficient; CCC: concordance correlation coefficient. 2
Experimental Eye Research 189 (2019) 107856
F. Gómez-Ulla, et al.
reproducibility and test-retest reliability. In order to evaluate the intra-observer repeatability of our measurements, we calculated the intraclass correlation coefficient (ICC) (Shrout and Fleiss, 1979) for measurements made in 80 eyes by Observer1 in T0 and T2. Values greater than 0.75 are indicative of good to excellent reliability (Koo and Li, 2016). To evaluate inter-observer reproducibility we calculated the concordance correlation coefficient (CCC) for the measurements performed in 80 eyes by Observer 1 (T0) and Observer 2 (T1) in images obtained in the first session (OCT-A1). The CCC measures the agreement between two variables (Lin, 1989) and ranges between 0 and 1; higher values indicate grater agreement and hence better reproducibility. To assess the test-retest reproducibility we calculated the CCC of the FAZ area measurements performed in 184 eyes by Observer 1(T0 and T3) in images obtained from two different imaging sessions (OCT-A1 and OCT-A2). The fitting of the data to a normal distribution was tested by the Kolmogorov-Smirnov test. To assess comparisons of data fitting a normal distribution, a two tailed Student's t-test was used. Otherwise a Mann-Whitney U test was used. For comparisons of more than two groups, the Kruskal-Wallis test was used. Differences were considered statistically significant when p < 0.05. Statistical analyses were performed with Microsoft Office Excel (Microsoft Corporation, Redmond, WA, USA), IBM SPSS Statistics v. 24 (IBM Corporation, Somers, NY, USA), and online calculators available on the internet. Statistical power calculations were made using STATISTICA software (StatSoft, Inc, Tulsa, OK, USA).
Fig. 1. - Correlation between superficial (sFAZ) and deep (dFAZ) FAZ in the 240 eyes investigated in our sample (r = 0.95). Deep FAZ was in most cases larger than superficial FAZ.
3.3. Gender and FAZ area The sFAZ area of women (mean 0.297 mm2, sd: 0.110) was significantly larger (two tailed t-test, p = 0.002, statistical power 0.91) than that of men (mean 0.254 mm2, sd: 0.098). Similarly, the dFAZ area in women (mean 0.322 mm2, sd: 0.111) was significantly larger (two tailed t-test, p < 0.001, statistical power 0.95) than that of men (mean 0.273 mm2, sd: 0.099). When the differences (Mann-Whitney test) between women and men were analyzed by age group, we found that the differences appeared only in those groups aged younger than 20 years and older than 50 years. Table 3 shows these results. For the 10–19 years of age group, the median sFAZ for women was 0.300 mm2, whereas for men it was 0.247 mm2. In the same age group, the median dFAZ for women was 0.314 mm2, whereas for men it was 0.255 mm2. Regarding the 50–59 years of age group the median sFAZ for women was 0.325 mm2 and that for men was 0.218 mm2, whereas for dFAZ, these values were 0.349 mm2 and 0.238 mm2, respectively. Finally, for the 60–69 years of age group, the median sFAZ for women was 0.312 mm2 and that for men was 0.222 mm2. These differences were statistically significant (p < 0.05) for both sFAZ and dFAZ areas and in all these instances, the FAZ area of women was larger than that of men. Fig. 2 shows the results related to age, gender, and sFAZ and dFAZ areas.
3. Results 3.1. Repeatability and reproducibility The repeatability and reproducibility assessment results are shown in Table 2. Intra-observer repeatability reached ICC values of 0.993 for sFAZ area and 0.985 for dFAZ area measurements. Inter-observer reproducibility reached CCC values of 0.924 for sFAZ area measurements, and 0.620 for dFAZ area measurements. Test-retest reproducibility showed CCC values of 0.974 for sFAZ area and 0.957 for dFAZ area.
3.2. Descriptive statistics The mean sFAZ area (n = 240 eyes) was 0.275 mm2 (sd: 0.106), and ranged from 0.060 to 0.545 mm2. The mean dFAZ area (n = 240 eyes) was 0.298 mm2 (sd: 0.108) and ranged from 0.068 to 0.576 mm2. Comparing all eyes, the dFAZ was significantly larger than the sFAZ (mean difference 0.023 mm2, two tailed paired t-test, p < 0.001, statistical power 0.73). The sFAZ and dFAZ area sizes were strongly correlated (r = 0.95) (Fig. 1). When comparing left and right eyes, we found no differences (paired two tailed t-test, p = 0.173) between the sFAZ of the right (mean 0.278 mm2, sd: 0.106) and left eyes (0.273 mm2, sd: 0.107). Similarly, there were no differences (paired two tailed t-test, p = 0.275) between the dFAZ of the right (mean 0.300 mm2, sd: 0.106) and left eyes (mean 0.295 mm2, sd: 0.109).
3.4. Age and FAZ area To assess the effect of age on FAZ area we grouped the subjects into six age groups as indicated previously, and used the Kruskal-Wallis test to see whether the distribution of FAZ areas was the same across these groups. In men, we found no differences between age groups. In women, however, the FAZ area distribution was not the same across groups for both sFAZ and dFAZ (p < 0.001). In both cases, there were statistically significant differences between the 20–29 year old group (sFAZ median: 0207 mm2, dFAZ median: 0.218 mm2) and the 50–59 year old group (sFAZ median: 0.325 mm2, dFAZ median: 0.349 mm2) (p = 0.031 for sFAZ and p = 0.022 for dFAZ). Fig. 2 shows the results related to age, gender, and sFAZ and dFAZ areas.
Table 2 - Intra-observer repeatability, and inter-observer and inter-session reproducibility of superficial (sFAZ) and deep (dFAZ) FAZ area measurements.
sFAZ dFAZ
Intra-observer repeatability ICC (95% CI)
Inter-observer reproducibility CCC (95% CI)
Test-retest reproducibility CCC (95% CI)
0.993 (0.988–0.995) 0.985 (0.958–0.993)
0.924 (0.887–0.949) 0.620 (0.512–0.709)
0.974 (0.966–0.981) 0.957 (0.943–0.967)
ICC: intraclass correlation coefficient; CCC: concordance correlation coefficient; CI: confidence interval. 3
Experimental Eye Research 189 (2019) 107856
F. Gómez-Ulla, et al.
Table 3 - Comparisons of FAZ between women and men by age group.
Table 4 -Reported superficial and deep FAZ areas calculated from OCT-A images. Values are mean mm2 ( ± sd).
Age group (years) 10–19
20–29
30–39
40–49
50–59
Authors
sFAZ
Carpineto et al.39 2016a
0.251 ( ± 0.096) 0.252 ( ± 0.096) 0.350 ( ± 0.110) 0.280 ( ± 0.100) 0.288 ( ± 0.136) 0.360 ( ± 0.110) 0.257 0.350 ( ± 0.013) 0.285 ( ± 0.150) 0.373 ( ± 0.109) 0.377 ( ± 0.112) 0.350 ( ± 0.013) 0.304 ( ± 0.132) 0.257 ( ± 0.104) 0.274 0.269 ( ± 0.092) 0.270 ( ± 0.090) 0.266 ( ± 0.097) 0.270 ( ± 0.101 0.250 ( ± 0.060) 0.240 0.474 ( ± 0.172) 0.301( ± 0.058) 0.275 ( ± 0.106)
dFAZ
60–69
Superficial FAZ: Women median Men median Mann-Whitney, p
0.300 0.247 0.011a
0.207 0.293 0.157
0.347 0.258 0.091
0.314 0.255 0.477
0.325 0.218 0.004a
0.312 0.222 0.016a
Deep FAZ: Women median Men median Mann-Whitney, p
0.314 0.255 0.011a
0.218 0.321 0.157
0.336 0.287 0.091
0.342 0.262 0.477
0.349 0.238 0.004a
0.338 0.228 0.016a
Choi et al.48 2017 Coscas et al.29 2016 de Carlo et al.40 2015 Di et al.57 2016 Freiberg et al.46 2016b Gadde et al.49 2016 Goudot et al.52 2017 Guo et al.41 2017a
a Indicates statistically significant differences (p < 0.05) between women and men.
Hussain and Hussain58 2016 Kuehlewein et al.42 2015 Linderman et al.28 2017 Magrath et al.47 2016 Mastropasqua et al.44 2017a
4. Discussion 4.1. Reliability of FAZ measurements
Samara et al.8 2015 Shahlaee et al.45 2016 Takase et al.32 2015 Tan et al.30 2016 Yu et al.31 2015 Mean of above results This report
Since OCT-A artifacts are common and can lead to misleading interpretations (Spaide et al., 2015a,b), the reliability when measuring FAZ areas using this technique has been addressed in many recent studies, which widely agree in reporting excellent repeatability and reproducibility as assessed by the ICC and CCC (Linderman et al., 2017; Coscas et al., 2016; Tan et al., 2016; Carpineto et al., 2016; de Carlo et al., 2015a,b; Guo et al., 2017; Kuehlewein et al., 2015; La Spina et al., 2017; Mastropasqua et al., 2017; Shahlaee et al., 2016; Freiberg et al., 2016). In a recent study performed on healthy eyes, the FAZ area measurements showed high repeatability and reproducibility (Carpineto et al., 2016) as well as high sensitivity and specificity (Freiberg et al., 2016). Linderman et al. (2017) found that manual and automatic segmentation of FAZ areas in OCT-A images also showed excellent repeatability and reliability. For sFAZ measurements, the reported values of ICC range from 0.850 (Coscas et al., 2016), to 0.998 (Carpineto et al., 2016), whereas for dFAZ these values range from 0.840 (Shahlee et al. 2016) to 0.960 (Coscas et al., 2016). Reported CCC values for sFAZ measurements range from 0.984 (Guo et al., 2017) to 0.999 (Carpineto et al., 2016), whereas for dFAZ the reported CCC is < 0.850 (La Spina et al., 2017). We found ICC values of 0.993 for sFAZ and 0.985 for dFAZ measurements, and CCC values of 0.924 for sFAZ and 0.620 for dFAZ measurements. Therefore our values are in agreement with those reported by other authors, indicating high reliability in our measurements. We found however a low CCC value when dFAZ areas were measured. La Spina et al. (2017) also reported a low CCC value for
0.370 (0.120)
0.341 0.490 ( ± 0.012) 0.398 ( ± 0.138)
0.490 ( ± 0.012) 0.486 ( ± 0.162) 0.364
0.495 ( ± 0.227) 0.340 ( ± 0.116) 0.380 ( ± 0.110) 0.380 0.424 ( ± 0.067) 0.298 ( ± 0.108)
a
The measurements were made by two observers. The area shown here was calculated from the reported diameters, assuming it was circular. b
dFAZ area measurements. This may occur because dFAZ is less well delimited than sFAZ, which makes the delineation of dFAZ much less sharp; therefore a higher variability should be expected in the measurements of dFAZ area. A recent study found consistent measurements of sFAZ and dFAZ size in OCT-A images taken with the same machine and technique, although significant differences were found when two different imaging devices were used (Magrath et al., 2017). We used a single machine to carry out the present study, but we obtained two different OCT-A scans of 184 eyes in which the FAZ areas were measured by the same observer. These measurements demonstrated excellent agreement (CCC values of 0.974 and 0.957 for sFAZ and dFAZ areas, respectively). Since we used a single machine we cannot provide results on the effect of using different machines on the variability of these measurements.
Figure 2. -Age and FAZ area in men and women. There were no differences among age groups in either sFAZ or dFAZ area in the men group. In women there were differences between the age groups of 20–29 years and 50–69 years for both sFAZ area (p = 0.030) and dFAZ area (p = 0.019). 4
Experimental Eye Research 189 (2019) 107856
F. Gómez-Ulla, et al.
this increase was faster in men than in women. Using fluorescein angiography and grouping the subjects by age, Laatikaninen and Larinkari found that the mean diameter of the FAZ was 0.530 mm in patients younger than 39 years and 0.610 mm after the age of 40 (Laatikainen and Larinkari, 1977). Our results show no significant changes with age in men, but women showed an increase in the size of both sFAZ and dFAZ with age. There is evidence that retinal changes occur with age. Although the origin of these changes is not clear, hormonal variations through life may play a role. An abrupt decline in hormonal activity causes the symptoms observed in menopausal women. A decrease in circulating testosterone induces a number of physical changes in aging men. Estrogen receptor expression has been found in bovine and rodent retinas (Kobayashi et al., 1998). There is evidence that the human eye has estrogen receptors and that reduced levels of circulating estrogens in women after menopause may lead to changes in the expression of estrogen receptors in ocular tissues (Ogueta et al., 1999). A large number of estrogen-induced proteins have been identified, and most of these proteins can be found in human eye tissues, including the retina (Yu et al., 1994). Since women have more drastic hormonal variations through life than men, this may explain why we observed changes in macular capillaries related to age in women but not in men.
4.2. FAZ area size Table 4 shows the reported FAZ areas by other authors along with our results. We found a mean sFAZ area of 0.275 mm2, which is in agreement with other reported values in healthy eyes for sFAZ area using OCT-A, i.e. from 0.250 mm2 (Takase et al., 2015) to 0.350 mm2 (Choi et al., 2017; Gadde et al., 2016). Studies using fluorescein angiography to analyze the FAZ size in healthy eyes demonstrated more variability those obtained using OCT-A, with areas ranging from 0.205 mm2 to 0.405 mm2 (Mansour et al., 1993; Zheng et al., 2010). The mean dFAZ area for our dataset was 0.298 mm2, which is lower than previously reported values, ranging from 340 mm2 (Shahlaee et al., 2016) to 0.495 mm2 (Samara et al., 2015). This difference may be explained because of the difficulties in delineating the dFAZ perimeter. Indeed, as other authors have reported (La Spina et al., 2017) this difficulty is reflected in the lower CCC found when assessing the reliability of dFAZ area measurements. All reported values of sFAZ and dFAZ areas showed higher values for the latter (Table 4). For instance, Coscas et al. reported a mean sFAZ area of 0.280 mm2 and a mean dFAZ area of 0.370 mm2; the latter was significantly larger (Coscas et al., 2016). In agreement with previous studies, our results also show that the dFAZ area is larger than the sFAZ area. The larger size of the dFAZ can be explained in part because in the deep capillary plexus the vessels are displaced into the more internal retinal layers as these layers narrow at the fovea (Coscas et al., 2016).
4.5. Limitations of the study Among the limitations of this study is the limited number of subjects; a larger number of subjects would likely corroborate the findings reported here. Because our sample size is small, future studies will need a larger population based study design, exploring the effects of age, sex, and important factors such as axial eye length. There are also some technical limitations to the currently available devices. For instance, when the capillary flow rate falls below the threshold detected by OCTA algorithms, some vessels may not be completely visualized and therefore FAZ area measurements may be inaccurate. These technical limitations are inherent to the OCT machines and currently cannot be resolved. The dFAZ has poorly defined borders when compared with sFAZ. Therefore, there is difficulty in manually measuring this area, showing a larger variability than the sFAZ area measurements. We believe however that this may not have influenced our results as there was consistency in the differences found between age groups in sFAZ and dFAZ. The refractive error also affects the dimension of the FAZ as measured in fundus images. It was found however that the correlation between refractive error and FAZ dimensions was low and not statistically significant (Bresnick et al., 1984). This study concluded that it was unlikely that the effect on the image magnification contributed meaningfully to the FAZ measurements. Tan et al. (2016) performed univariate linear regression analysis of factors affecting the FAZ size and found that emmetropes and low myopes do not show a significant relationship between spherical equivalent (p = 0.223) and axial length (p = 0.469) with the FAZ size. We did not correct our data for refractive error, but since our subjects were emmetropes or low myopes, we believe that these optical factors did not significantly influence our measurements. The axial length may also be a significant source of error for measuring FAZ area (Linderman et al., 2018), but unfortunately a correction method for axial length when calculating the FAZ area is not currently available on OCT-A devices. While axial length may not be critical when comparing multiple scans from the same subject, it may be a limitation when comparing FAZ area measurements across individuals. Lindermann et al. (Linderman et al., 2017) observed that the error in FAZ area estimates as a function of axial length has an average error of 8.29%, with an absolute maximum error of 0.07 mm2. Cruickshank and Logan (2018) found a relationship between axial length and refractive error (mean spherical equivalent). By using the
4.3. Gender effect on FAZ Macular structural characteristics may be influenced by gender. For instance, Adhi et al. (2012) using a spectral domain OCT system found that men had greater foveal and macular thickness compared to women. Song et al. reported that the inner macular thickness and the overall macular volume were significantly smaller in women (Song et al., 2010). Other authors, however, have found no significant gender differences in macular or foveal thickness (Chan et al., 2006). Some authors have failed to find changes related to gender in macular vasculature, in terms of flow density or FAZ area (Samara et al., 2015; Coscas et al., 2016; Alnawaiseh et al., 2018). On the contrary, other studies report that women have larger FAZ areas than men (Linderman et al., 2017; Tan et al., 2016). Yu et al. (2015) reported similar findings in the healthy eyes of Chinese individuals, with a larger FAZ area in women than in men. Our overall results support the observation that the FAZ area in women is significantly larger (mean 0.297 mm2) than in men (mean 0.254 mm2). 4.4. Age and FAZ Several authors analyzing OCT images have found that retinal thickness declines with age (Alamouti and Funk, 2003; Eriksson and Alm, 2009; Song et al., 2010). However other studies have failed to find such a relationship between age and macular or foveal thickness (Chan et al., 2006; Adhi et al., 2012; Sull et al., 2010). Age-related anatomical changes in the macular region should be concurrent with changes in the vasculature and hence in FAZ area; however, studies reporting on the influence of age on FAZ size are somehow confusing. Some have found no relationship (Samara et al., 2015; Linderman et al., 2017; Tan et al., 2016; Takase et al., 2015; Gadde et al., 2016; Goudot et al., 2017) while others have found a relationship between age and FAZ size. Using fluorescein angiography, Wu et al., (1985) found a positive correlation between FAZ area and age, but Coscas et al., (2016) found that the superficial FAZ area was significantly smaller in older subjects. Alnawaiseh et al., (2018) found that older subjects had lower macular flow density. Bresnick et al. (1984) using fluorescein angiography, found that the FAZ diameter decreases with increasing age in healthy control eyes. Yu et al. (2015) reported in healthy Chinese individuals that the FAZ area increased by an average of 1.48% annually, and that 5
Experimental Eye Research 189 (2019) 107856
F. Gómez-Ulla, et al.
equation reported by these authors, the estimated axial length of the eyes included in our study ranged from 23 to 24 mm. If our measurements were corrected for axial length according to the equations reported by Linderman et al. (2017) for eyes with 23 mm the FAZ area would have an underestimation of 0.018 mm2, whereas for eyes of 24 mm, this would be 0.000 mm2. Therefore we believe that the axial length should not have a significant impact on the results we have obtained from our data set.
s40942-015-0005-8. Eladawi, N., Elmogy, M.M., Ghazal, M., et al., 2018. Classification of retinal diseases based on OCT images. Front. Biosci. 23, 247–264. Eriksson, U., Alm, A., 2009. Macular thickness decreases with age in normal eyes: a study on the macular thickness map protocol in the Stratus OCT. Br. J. Ophthalmol. 93, 1448–1452. Freiberg, F.J., Pfau, M., Wons, J., 2016. Optical coherence tomography angiography of the foveal avascular zone in diabetic retinopathy. Graefes Arch. Clin. Exp. Ophthalmol. 254, 1051–1058. Gadde, S.G., Anegondi, N., Bhanushali, D., et al., 2016. Quantification of vessel density in retinal optical coherence tomography angiography images using local fractal dimension vessel density in OCTA images. Invest. Ophthalmol. Vis. Sci. 57, 246–252. Gómez-Ulla, F., Fernandez, M.I., Gonzalez, F., et al., 2002. Digital retinal images and teleophthalmology for detecting and grading diabetic retinopathy. Diabetes Care 25, 1384–1389. Gonzalez, F., Iglesias, R., Suarez, A., et al., 2001. Teleophthalmology link between a primary health care centre and a reference hospital. Med. Inform. Internet Med. 26, 251–263. Goudot, M.M., Sikorav, A., Semoun, O., et al., 2017. Parafoveal OCT angiography features in diabetic patients without clinical diabetic retinopathy: a qualitative and quantitative analysis. J. Ophthalmol. 2017https://doi.org/10.1155/2017/8676091. 8676091. Grover, S., Murhty, R.K., Brar, V.S., et al., 2009. Normative data for macular thickness by high-definition Spectral-domain optical coherence tomography (Spectralis). Am. J. Ophthalmol. 148, 266–271. Guo, J., She, X., Liu, X., et al., 2017. Repeatability and reproducibility of foveal avascular zone area measurements using angioplex spectral domain optical coherence tomography angiography in healthy subjects. Ophthalmologica 237, 21–28. Jonas, J.B., Schneider, U., Naumann, G.O., 1992. Count and density of human retinal photoreceptors. Graefes Arch. Clin. Exp. Ophthalmol. 230, 505–510. Kobayashi, K., Kobayashi, H., Ueda, M., et al., 1998. Estrogen receptor expression in bovine and rat retinas. Invest. Ophthalmol. Vis. Sci. 39, 2105–2110. Koo, T.K., Li, M.Y., 2016. A guideline of selecting and reporting intraclass correlation coefficients for reliability research. J. Chiropr. Med. 15, 155–163. Kuehlewein, L., Tepelus, T.C., An, L., et al., 2015. Noninvasive visualization and analysis of the human parafoveal capillary network using swept source OCT optical microangiography. Invest. Ophthalmol. Vis. Sci. 56, 3984–3988. La Spina, C., Carnevali, A., Marchese, A., et al., 2017. Reproducibility and reliability of optical coherence tomography angiography for foveal avascular zone evaluation and measurement in different settings. Retina 37, 1636–1641. Laatikainen, L., Larinkari, J., 1977. Capillary-free area of the fovea with advancing age. Invest. Ophthalmol. Vis. Sci. 16, 1154–1157. Lin, L.I., 1989. A concordance correlation coefficient to evaluate reproducibility. Biometrics 45, 255–268. Linderman, R., Salmon, A.E., Strampe, M.1, et al., 2017. Assessing the accuracy of foveal avascular zone measurements using optical coherence tomography angiography: segmentation and scaling. Trans. Vis. Sci. Tech. 6, 16. https://doi.org/10.1167/tvst. 6.3.16. Linderman, R.E., Muthiah, M.N., Omoba, S.B., et al., 2018. Variability of foveal avascular zone metrics derived from optical coherence tomography angiography images. Transl. Vis. Sci. Technol. 7, 20. Magrath, G.N., Say, E.A.T., Sioufi, K., et al., 2017. Variability in foveal avascular zone and capillary density using optical coherence tomography angiography machines in healthy eyes. Retina 37, 2102–2111. Mammo, Z., Balaratnasingam, C., Yu, P., et al., 2015. Quantitative noninvasive angiography of the fovea centralis using speckle variance optical coherence tomography. Invest. Ophthalmol. Vis. Sci. 56, 5074–5086. Mansour, A.M., Schachat, A., Bodiford, G., et al., 1993. Foveal avascular zone in diabetes mellitus. Retina 13, 125–128. Mastropasqua, R., Toto, L., Mattei, P.A., et al., 2017. Reproducibility and repeatability of foveal avascular zone area measurements using swept-source optical coherence tomography angiography in healthy subjects. Eur. J. Ophthalmol. 27, 336–341. Ogueta, S.B., Schwartz, S.D., Yamashita, C.K., et al., 1999. Estrogen receptor in the human eye: influence of gender and age on gene expression. Invest. Ophthalmol. Vis. Sci. 40, 1906–1911. Ooto, S.1., Hangai, M., Tomidokoro, A., et al., 2011. Effects of age, sex, and axial length on the three-dimensional profile of normal macular layer structures. Invest. Ophthalmol. Vis. Sci. 52, 8769–8779. Samara, W.A., Say, E.A., Khoo, C.T., et al., 2015. Correlation of foveal avascular zone size with foveal morphology in normal eyes using optical coherence tomography angiography. Retina 35, 2188–2195. Shahlaee, A., Pefkianaki, M., Hsu, J., et al., 2016. Measurement of foveal avascular zone dimensions and its reliability in healthy eyes using optical coherence tomography angiography. Am. J. Ophthalmol. 161, 50–55. Shrout, P.E., Fleiss, J.L., 1979. Intraclass correlations: uses in assessing rater reliability. Psychol. Bull. 86, 420–428. Snodderly, D.M., Weinhaus, R.S., Choi, J.C., 1992. Neural-vascular relationships in central retina of macaque monkeys (Macaca fascicularis). J. Neurosci. 12, 1169–1193. Song, W.K., Lee, S.C., Lee, E.S., et al., 2010. Macular thickness variations with sex, age, and axial length in healthy subjects: a spectral domain-optical coherence tomography study. Invest. Ophthalmol. Vis. Sci. 51, 3913–3918. Spaide, R.F., Fujimoto, J.G., Waheed, N.K., 2015a. Image artifacts in optical coherence angiography. Retina 35, 2163–2180. Spaide, R.F., Fujimoto, J.G., Waheed, N.K., et al., 2018. Optical coherence tomography angiography. Prog. Retin. Eye Res. 64, 1–55. Spaide, R.F., Klancnik Jr., J.M., Cooney, M.J., 2015b. Retinal vascular layers imaged by
5. Conclusions In our study the patients were grouped by age and gender and this allowed us to obtain data to demonstrate that both sFAZ and dFAZ are larger in women than in men only in young and older age groups. In men age does not influence FAZ size, but in women there is age effect in sFAZ and dFAZ in such way that their area is smaller in younger than in older ages. These changes could be induced by age related hormonal variations, which are more pronounced in women than in men. Since ocular tissues express estrogen receptors, future studies may clarify their possible role in the foveal microvascular age related changes we found in women. OCT-A is reproducible and widely available in ophthalmology clinics, therefore the results we report should be helpful for diagnosis and follow-up of macular diseases. Declaration of competing interest No conflicting relationship exists for any author. Acknowledgments Funded by the ISCIII-Subdireccion General de Evaluacion y Fomento de la Investigacion (RD16/0008/003), cofunded by the European Regional Development Fund (FEDER). References Adhi, M., Aziz, S., Muhammad, K., et al., 2012. Macular thickness by age and gender in healthy eyes using spectral domain optical coherence tomography. PLoS One 7, e37638. https://doi.org/10.1371/journal.pone.0037638. Agemy, S.A., Scripsema, N.K., Shah, C.M., et al., 2015. Retinal vascular perfusion density mapping using optical coherence tomography angiography in normals and diabetic retinopathy patients. Retina 35, 2353–2363. Alamouti, B., Funk, J., 2003. Retinal thickness decreases with age: an OCT study. Br. J. Ophthalmol. 87, 899–901. Alnawaiseh, M., Brand, C., Lauermann, J.L., et al., 2018. Messung der Flussdichte mittels OCT-angiographie. Der Ophthalmologe 115, 659–662. Arend, O., Wolf, S., Harris, A., et al., 1995. The relationship of macular microcirculation to visual acuity in diabetic patients. Arch. Ophthalmol. 113, 610–614. Balaratnasingam, C., Inoue, M., Ahn, S., et al., 2016. Visual acuity is correlated with the area of the foveal avascular zone in diabetic retinopathy and retinal vein occlusion. Ophthalmology 123, 2352–2367. Bresnick, G.H., Condit, R., Syrjala, S., et al., 1984. Abnormalities of the foveal avascular zone in diabetic retinopathy. Arch. Ophthalmol. 102, 1286–1293. Carpineto, P., Mastropasqua, R., Marchini, G., et al., 2016. Reproducibility and repeatability of foveal avascular zone measurements in healthy subjects by optical coherence tomography angiography. Br. J. Ophthalmol. 100, 671–676. Chan, A., Duker, J.S., Ko, T.H., et al., 2006. Normal macular thickness measurements in healthy eyes using Stratus optical coherence tomography. Arch. Ophthalmol. 124, 193–198. Choi, J., Kwon, J., Shin, J.W., et al., 2017. Quantitative optical coherence tomography angiography of macular vascular structure and foveal avascular zone in glaucoma. PLoS One 12, e0184948. https://doi.org/10.1371/journal.pone.0184948. Conrath, J., Giorgi, R., Raccah, D., et al., 2005. Foveal avascular zone in diabetic retinopathy: quantitative vs qualitative assessment. Eye (Lond) 19, 322–326. Coscas, F., Sellam, A., Glacet-Bernard, A., et al., 2016. Normative data for vascular density in superficial and deep capillary plexuses of healthy adults assessed by optical coherence tomography angiography. Invest. Ophthalmol. Vis. Sci. 57, 211–223. Cruickshank, F.E., Logan, N.S., 2018. Optical 'dampening' of the refractive error to axial length ratio: implications for outcome measures in myopia control studies. Ophthalmic Physiol. Opt. 38, 290–297. de Carlo, T.E., Chin, A.T., Bonini Filho, M.A., et al., 2015a. Detection of microvascular changes in eyes of patients with diabetes but not clinical diabetic retinopathy using optical coherence tomography angiography. Retina 35, 2229–2235. de Carlo, T.E., Romano, A., Waheed, N.K., et al., 2015b. A review of optical coherence tomography angiography (OCTA). Int. J. Retin. Vitr. 1, 5. https://doi.org/10.1186/
6
Experimental Eye Research 189 (2019) 107856
F. Gómez-Ulla, et al. fluorescein angiography and optical coherence tomography angiography. JAMA Ophthalmol. 133, 45–50. Sull, A.C., Vuong, L.N., Price, L.L., 2010. Comparison of spectral/Fourier domain optical coherence tomography instruments for assessment of normal macular thickness. Retina 30, 235–245. Takase, N., Nozaki, M., Kato, A., et al., 2015. Enlargement of foveal avascular zone in diabetic eyes evaluated by en face optical coherence tomography angiography. Retina 35, 2377–2383. Tan, C.S., Lim, L.W., Chow, V.S., et al., 2016. Optical coherence tomography angiography evaluation of the parafoveal vasculature and its relationship with ocular factors. Invest. Ophthalmol. Vis. Sci. 57, OCT224–O230. Veverka, K.K., AbouChehade, J.E., Iezzi Jr., R., et al., 2015. Noninvasive grading of radiation retinopathy: the use of optical coherence tomography angiography. Retina 35, 2400–2410. Wu, L.Z., Huang, Z.S., Wu, D.Z., et al., 1985. Characteristics of the capillary-free zone in
the normal human macula. Jpn. J. Ophthalmol. 29, 406–411. Yu, D.Y., Cringle, S.J., Su, E.N., 2005. Intraretinal oxygen distribution in the monkey retina and the response to systemic hyperoxia. Invest. Ophthalmol. Vis. Sci. 46, 4728–4733. Yu, J., Jiang, C., Wang, X., et al., 2015. Macular perfusion in healthy Chinese: an optical coherence tomography angiogram study. Invest. Ophthalmol. Vis. Sci. 56, 3212–3217. Yu, M.C., Li, W.W., Liu, K., et al., 1994. An immunohistochemical study of the c-fos protooncogene in the developing human retina. Neuroscience 60, 983–987. Zeffren, B.S.1., Applegate, R.A., Bradley, A., et al., 1990. Retinal fixation point location in the foveal avascular zone. Invest. Ophthalmol. Vis. Sci. 31 2099–105. Zheng, Y., Gandhi, J.S., Stangos, A.N., et al., 2010. Automated segmentation of foveal avascular zone in fundus fluorescein angiography. Invest. Ophthalmol. Vis. Sci. 51, 3653–3659.
7