Choroidoscleral Interface Irregularity Index: A novel optical coherence tomographybased parameter in patients with epiretinal membrane

Abstract

Purpose : This study aimed to assess the regularity of the choroidoscleral interface (CSI) using a novel parameter, CSI-irregularity index, before and after epiretinal membrane (ERM) surgery.

Methods : This retrospective cohort study included 36 patients with idiopathic ERM who underwent uncomplicated pars plana vitrectomy and ERM removal. All subjects underwent ocular examinations at baseline and at 1, 2, 4, and 6 months after surgery.

Results : The bowl-shaped (regular) contour of the CSI was found in 14 patients (38.9%); mean CSI-irregularity index was 14.8 ± 11.0 in this group. The inflective (irregular) contour of the CSI was found in 22 patients (61.1%) and mean CSI-irregularity index was 34.0 ± 20.6. The CSI irregularity index decreased gradually after ERM surgery, and was correlated with postoperative best-corrected visual acuity.

Conclusions : The CSI-irregularity index could serve as a surrogate marker to quantitatively represent the CSI morphology. We observed the gradual decrease of the CSI-irregularity index after ERM surgery in quantitative manner. This study revealed correlations between the CSI-irregularity index and visual outcomes after ERM surgery. Our results suggest that the CSI-irregularity index might be an intuitive anatomic indicator of the CSI and might be useful as a possible prognostic marker for patients undergoing ERM surgery.

This abstract was presented at the 2019 ARVO Annual Meeting, held in Vancouver, Canada, April 28 - May 2, 2019.

 

The definition and a representative image of measurement of the CSI irregularity index. (A) The hatched area between the choroidoscleral interface (CSI) on optical coherence tomography and its best-fit spherocylinder demonstrate the concept of the CSI irregularity index, which is defined as the weighted sum of differences at each measured point between the CSI on optical coherence tomographic image and its best-fit spherocylinder. (B) Before importing images into Matlab software, the CSI was manually delineated. (C) After removing background noise with the thresholding filter in Matlab software, the contour of the CSI was subsequently extracted. (D) The polynomial curve was used for curve fitting with the Curve Fitting Tool in Matlab software. The order of polynomial function was determined to identify the best approximation to the shape of the CSI. The Polyfit function was used to fit the coordinates via the least squares principle. CSI irregularity index = 12.6665.