Diagnostic capacity and sectoral structure-function relationship of superficial circumpapillary and macular vessel density in early glaucomatous eyes
The protocols for this retrospective observational study have been approved by the Research Ethics Board of the Graduate School of Medicine and the University of Tokyo Medical School (ID: 2217). Patients gave written informed consent to have their information stored in the hospital’s database and used for research during their first visit. The requirement for additional written informed consent for this retrospective observational study has been waived by the Research Ethics Board of the Graduate School of Medicine and the University of Tokyo Medical School (ID: 2217 ). Instead, study participants were informed of the protocol posted at the outpatient clinic and given the option to withdraw from the study. The study protocol adhered to the principles of the Declaration of Helsinki.
Participants in this study included consecutive normal subjects who presented for routine eye examination or refractive error and patients with early onset glaucoma or PPG (DM > – 6 dB) from the hospital of the University of Tokyo (Tokyo, Japan), Yotsuya Shirato Eye Clinic (Tokyo, Japan) and Tajimi Iwase Eye Clinic (Gifu, Japan) between June 2020 and July 2021.
All participants included in this study underwent the following eye examinations: corneal refraction and radius of curvature measurements, best corrected visual activity (BCVA), axial length measurement, slit lamp examination, pressure measurement intraocular (IOP) with Goldmann applanation tonometry, gonioscopy, fundus examination including examination of the optic nerve head, stereophotography of the optic disc, OCT and OCTA imaging. Subjects also underwent Humphrey Field Analyzer (HFA) (Carl Zeiss Meditec, Dublin, CA, USA) measurements with the standard 24-2 Swedish Interactive Threshold Algorithm strategy within 3 months of OCT and OCTA imaging.
The inclusion criteria for this study were a BCVA of 20/25 or better, obtaining a good quality OCT/OCTA scan and reliable VF results (
The diagnosis of each eye was made by at least two glaucoma specialists (KA, HS). Normal eyes had normal optic disc appearances on fundus examination and fundus stereophotographs, IOP 34.
Glaucomatous eyes included in this study had open angles on gonioscopy, glaucomatous change (i.e. narrowing of the neuroretinal rim, notching, and the presence of retinal nerve fiber layer defects) of the head optic nerve on fundus examination and fundus stereophotographs, VF abnormality as previously defined34 with DM > − 6 dB consistent with a glaucomatous change in the optic nerve head. PPG eyes with glaucomatous optic nerve head changes but no apparent VF abnormalities on HFA 24-2 results were also included in this study.
OCT and OCTA imaging
OCT and OCTA imaging was performed using Cirrus HD-6000 with AngioPlex OCTA (Carl Zeiss Meditec, Dublin, CA) with a scan rate of 100,000 A-scans per second and eye tracking technology which reduces the risk of motion artifacts such as those caused by blinks and jerks. Cirrus HD-6000 is the newest OCTA model with a sweep speed approximately 1.5 times faster than previous models used in previous OCTA papers (Cirrus HD-5000 (Carl Zeiss Meditec, Dublin, CA): 68,000 A-scans per second and Angio Vue (Optovue, Inc, Fremont, CA, USA): 70,000 A-scans per second). All subjects underwent 6 × 6 mm optic nerve head (ONH) cubic scans, 6 × 6 mm macular cubic scans, 4.5 × 4.5 mm ONH angiographic scans, and 6 × macular angiographic scans. 6mm. Only images of optimal quality (signal strength indices > 7) without artifacts were selected. En face microvascular flow images were obtained by comparing differences in phase and intensity information contained in sequential B-scans performed in the same area. An automatic retinal layer segmentation program was applied to each scan to obtain layer-specific blood flow.
CpRNFLT, GCIPL, cpVD and mVD were measured in all subjects. CpVD was calculated as the density of the superficial capillary plexus between the inner limiting membrane (ILM) and the RNFL within a 4.5 mm diameter circle centered on the disc excluding the area of the optical disc. Vessel density is defined as the total area of perfused vasculature per unit area in the measurement region and large vessels were excluded from the analysis. Upper and lower averages provided by trading software and 6 Garway-Heath industry stocks35 Calculated from raw OCT data were used to assess sector structure function relationships for the cpRNFLT and cpVD parameters.
The MVD was calculated as the density of the superficial capillary plexus between the ILM and the inner plexiform layer within a circle 6 mm in diameter excluding the foveal avascular zone 1 mm in diameter. The measurement area was divided into parafoveal (3 mm outer diameter ring and 1 mm inner diameter) and perifoveal (6 mm outer ring diameter and 3 mm inner diameter) regions which were further subdivided into four quadrants at the continuation of the study on the early treatment of diabetic retinopathy. (ETDRS). The upper and lower parafoveal and perifoveal quadrants were used for segmental analysis.
Three consecutive ONH angiograms and macular angiograms were performed during the same visit on 20 normal eyes and 20 glaucomatous eyes to assess intra-visit reproducibility.
All data is reported as the mean ± standard deviation unless otherwise specified. ANCOVA (analysis of covariance) was used to compare parameters between normal and glaucomatous eyes by adjusting for age and axial length. Categorical variables were compared using chi-square tests. Intra-visit reproducibility was assessed by intra-class correlation coefficients. AUROCs were used to assess the diagnostic ability of each parameter. The comparison of AUROCs was carried out using Delong’s tests. One eye from each subject was randomly selected for AUROC analysis and unpaired t-tests of OCT/OCTA parameters. Structure-function relationships of sector parameters and FV sensitivity of corresponding regions in glaucomatous eyes were determined using linear mixed models accounting for inter-eye correlations. Statistical analyzes were performed with commercially available software (SPSS version 27.0; SPSS, Inc., Chicago, IL, USA). Delong tests were performed with EZR (Saitama Medical Center, Jichi Medical University, Saitama, Japan), which is a graphical user interface for R (The R Foundation for Statistical Computing, Vienna, Austria). P values