Purpose
To measure foveal avascular zone (FAZ) dimensions in healthy eyes using optical coherence tomography angiography (OCTA) and calculate interobserver variability.
Design
Reliability analysis.
Methods
Thirty-four eyes of 17 healthy subjects underwent OCTA at the Retina Service of Wills Eye Hospital. Two masked graders performed measurements of FAZ dimensions including area, perimeter, and maximum horizontal and vertical diameters using ImageJ. Intraclass correlation coefficient (ICC) between graders was calculated.
Results
Mean area (mm 2 ), perimeter (mm), and maximum horizontal and vertical diameters (mm) were 0.27 ± 0.101, 2.21 ± 0.451, 0.59 ± 0.126, and 0.56 ± 0.118, respectively, at the superficial and 0.34 ± 0.116, 2.50 ± 0.462, 0.69 ± 0.123, and 0.63 ± 0.110 at the deep network. Interobserver agreement was high for all superficial FAZ measurements (ICC ≥0.90) but did not meet the lowest acceptable grader agreement for the deep vascular network (ICC <0.85). Fellow eyes had statistically similar values ( P > .05).
Conclusion
Manual measurement of FAZ dimensions using OCTA is a noninvasive and reliable method for quantifying FAZ at the superficial vascular network. Assessing FAZ alterations in the deep vascular network may be subject to greater interobserver variability.
The foveal avascular zone (FAZ) is the capillary-free area in the central macula. Various methods including histology, immunohistochemistry, fluorescein angiography (FA), and indocyanine green angiography (ICGA) have been used to evaluate the retinal vasculature. Such in vivo and in vitro assessments show variations in the dimensions of the FAZ in normal eyes. An accurate and reliable method of measuring FAZ dimensions could serve an important role in the diagnosis and management of various retinal vascular diseases that affect the FAZ, such as diabetic retinopathy or retinal vein occlusions.
Optical coherence tomography angiography (OCTA) is a novel noninvasive technology that uses motion contrast imaging to produce angiographic data across the retina and choroid. The obtained images have high-resolution capillary detail.
We used OCTA images to measure the dimensions of the FAZ. Given the higher-resolution images obtained from retinal capillary networks compared to conventional fluorescein angiography, this method may provide better visualization of capillaries around the fovea and enable more precise and reliable measurements of the FAZ. The purpose of this report is to establish normative data and to calculate interobserver variability in measurement of the characteristics of the FAZ using OCTA.
Methods
Healthy subjects with no previous ophthalmologic or medical history and a normal fundus examination who underwent OCTA at the Retina Service of Wills Eye Hospital between November 20, 2014 and March 20, 2015 were evaluated. The study was approved by the institutional review board of Wills Eye Hospital and complied with the Health Insurance Portability and Accountability Act. Informed consent was obtained prior to study enrollment.
The AngioVue OCTA system on the commercially available Avanti spectral-domain optical coherence tomography (SD OCT) device (Optovue Inc, Fremont, California, USA) was used for imaging. This system used a split-spectrum amplitude decorrelation angiography (SSADA) software algorithm (version 2014.2.0.93) and operated at 70 000 A-scans per second to acquire OCTA volumes consisting of 304 × 304 A-scans. Orthogonal registration and merging of 2 consecutive scans were used to obtain OCTA volume scans of a central 3 × 3-mm area of both eyes for each subject. OCTA images of the superficial and deep capillary networks were generated separately using the automated software algorithm. Based on these default settings, the superficial network extends from 3 μm below the internal limiting membrane to 15 μm below the inner plexiform layer (IPL). The deep capillary network extends from 15 to 70 μm below the IPL.
Using the acquired images, 2 graders (A.S. and M.P.) performed measurements of the FAZ. OCTA images at the levels of superficial and deep retinal capillary networks were loaded into an image analysis program (ImageJ 1.48v; National Institutes of Health, Bethesda, Maryland, USA). The graders used the set scale parameter of the software to define the 304-pixel width of the images as 3 mm. The area of the FAZ and maximum horizontal and vertical diameters were then marked manually as shown in Figure 1 . The graders performed the assessments independently and were masked to the results of each other. Reliability analysis was performed by calculating the intraclass correlation coefficient (ICC) between graders. The ICC measures the agreement between graders, ranging from 0, indicating no agreement, to 1, indicating perfect agreement. Lowest acceptable grader agreement was defined as 0.85. SAS software (version 9.4; SAS Institute Inc, Cary, North Carolina, USA) PROC NESTED was used to estimate the variance components for subject, eye (nested within subject), and grader (nested within subject and eye). The ICC was computed as follows:
VAR subj + VAR eye − VAR grader / VAR subj + VAR eye + VAR grader
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
