Choroidal vascularity index is independent of ocular and image-based factors in healthy eyes: a systematic review and meta-analysis

Abstract

Choroidal vascularity index (CVI) has been extensively used to assess choroidal health in posterior eye disease; however, interpretation is hindered by conflicting evidence on effect of normal physiological factors on CVI. Thus, this review aimed to derive a normative value for CVI and understand its relationship with various normal ocular/image-based factors. Studies were screened for eligibility, defined as eyes with normal ocular parameters (refractive error (RxE) < ± 6 DS, intraocular pressure (IOP < 21 mmHg) and no posterior segment disease and 98 studies were included. Pooled weighted average of subfoveal CVI extracted from healthy eyes (n = 5332 eyes) was 66.50% [CI 65.67-67.32]. Secondary stratifications (time of scan, device wavelength, imaging mode, region of interest, systemic factors) and meta regression (age, RxE and IOP) were insignificant on CVI (p = 0.579-0.872), however Best corrected visual acuity showed positive correlation (p = 0.037). Unaltered, narrow confidence interval in sensitivity analysis, complemented by non-significant publication bias indicated robustness of the synthesised data. These results provide the highest-level evidence in hierarchy of the first normative subfoveal CVI value synthesised from literature and that it is not influenced by most person, eye and imaging factors in healthy eyes.

Keywords: CVI; Choroid; Normative value; Vessel lumen.

Fig. 1

Prisma flowchart of screening process.

Fig. 2

Results of primary outcome and stratification. The error bars indicate confidence intervals. SW short wavelength, LW long wavelength, STD standard imaging, EDI enhanced depth imaging.

Fig. 3

Galbraith plot showing studies within 95% confidence intervals.