The Blooming Effect: A Potential Source of Variability in OCT-Derived Choroidal Vascularity Measurements [Letter]

Dear editor

We read with great interest the systematic review by Iovino et al¹ on choroidal structural and vascular changes during myopia progression and in response to myopia-control interventions. The authors provide a comprehensive synthesis of changes in choroidal thickness (CT) and choroidal vascularity index (CVI), while appropriately highlighting the substantial methodological heterogeneity among the included studies, particularly with regard to OCT acquisition protocols, segmentation procedures, and image-binarization algorithms.

In this context, we would like to draw attention to a further technical issue that may affect the quantitative assessment of choroidal vascularity but is not explicitly addressed in the review: the potential impact of the blooming effect on OCT-derived measurements of the choroidal vascular compartment.

CVI is calculated as the ratio of luminal choroidal area (LCA) to total choroidal area (TCA), with LCA and stromal choroidal area (SCA) being determined through OCT image binarization. The distinction between luminal and stromal components therefore relies critically on the distribution of OCT signal intensity and on the thresholding method used to classify pixels within the selected region of interest.

The blooming effect, well known in the echographic field,^2–7 may represent an additional source of measurement bias in this context. Hyperreflective signals originating from vessel walls and surrounding tissue interfaces may extend beyond their actual anatomical boundaries into adjacent hyporeflective luminal spaces. Consequently, portions of the vascular lumen may be incorrectly classified as stromal tissue during image binarization, resulting in an apparent reduction in LCA and a corresponding increase in SCA. Because CVI is directly dependent on the relative proportion of LCA within the TCA, such an effect could lead to an underestimation of the measured CVI.

This issue may be particularly relevant when evaluating myopia-control interventions associated with substantial or relatively rapid changes in choroidal structure. As highlighted by Iovino et al, several interventions, particularly repeated low-level red-light (RLRL) therapy and orthokeratology, have been associated with choroidal thickening that appears to be predominantly driven by expansion of the luminal component. Therefore, when longitudinal changes in choroidal vascular parameters are assessed, modifications in vessel caliber, signal intensity, or the spatial distribution of OCT reflectivity could potentially influence the extent of blooming and, consequently, the segmentation of the vascular compartment.

Importantly, we do not suggest that the vascular changes reported in the studies included in the review are necessarily artifactual. Rather, we propose that the blooming effect should be considered as a potential contributor to measurement variability when interpreting changes in LCA, SCA, and CVI. This distinction may be particularly important when relatively small changes in CVI are interpreted as evidence of biological vascular remodeling or as potential mechanisms underlying the effects of myopia-control interventions.

The methodological heterogeneity emphasized by Iovino et al—including differences in OCT platforms, image-processing procedures, segmentation approaches, and binarization algorithms—further supports the need to consider this potential source of variability. Although the review appropriately discusses approaches aimed at compensating for shadowing artifacts and notes the development of automated methods for choroidal vascular quantification, other optical and image-processing phenomena, such as blooming, may also influence the identification of the luminal boundaries.

Future studies may therefore benefit from standardized image-processing protocols that explicitly account for the blooming effect, together with appropriate compensation for shadowing artifacts and robust segmentation methods. Sensitivity analyses using different thresholding strategies could also help determine the stability of LCA, SCA, and CVI measurements. In addition, reporting OCT acquisition parameters and image-processing procedures in sufficient detail would improve reproducibility and facilitate comparisons across studies.

We believe that considering the potential impact of the blooming effect would complement the methodological limitations discussed by Iovino et al and may help distinguish genuine biological changes in choroidal vascularity from variability introduced by OCT image acquisition and processing. This consideration may become increasingly relevant as OCT-derived choroidal vascular parameters are investigated as biomarkers of treatment response and as potential indicators of the mechanisms underlying myopia progression and its control.

Disclosure

The authors report no conflicts of interest in this communication.