September 2016
Volume 57, Issue 12
Open Access
ARVO Annual Meeting Abstract  |   September 2016
Automated quantitative analysis of SD-OCT scans to predict visual outcome after epiretinal membrane (ERM) removal surgery
Author Affiliations & Notes
  • Tiffany J Au
    Stanford University, Stanford, California, United States
  • Luis De Sisternes
    Stanford University, Stanford, California, United States
  • Theodore Leng
    Stanford University, Stanford, California, United States
  • Daniel Rubin
    Stanford University, Stanford, California, United States
  • Footnotes
    Commercial Relationships   Tiffany Au, None; Luis De Sisternes, None; Theodore Leng, None; Daniel Rubin, None
  • Footnotes
    Support  Stanford Medical Scholars grant
Investigative Ophthalmology & Visual Science September 2016, Vol.57, 5929. doi:
  • Views
  • Share
  • Tools
    • Alerts
      ×
      This feature is available to Subscribers Only
      Sign In or Create an Account ×
    • Get Citation

      Tiffany J Au, Luis De Sisternes, Theodore Leng, Daniel Rubin; Automated quantitative analysis of SD-OCT scans to predict visual outcome after epiretinal membrane (ERM) removal surgery. Invest. Ophthalmol. Vis. Sci. 2016;57(12):5929.

      Download citation file:


      © ARVO (1962-2015); The Authors (2016-present)

      ×
  • Supplements
Abstract

Purpose : To identify and assess the association of quantitative SD-OCT imaging biomarkers measured prior to and shortly after ERM removal surgery with long-term post-surgical visual outcome

Methods : SD-OCT macular volumes were scanned pre and post-operatively from 68 patients with 72 operated eyes, on which ERM removal surgery was performed by a single vitreoretinal surgeon. We developed automated methods to extract quantitative features from the images, including area, extent, and circularity of ellipsoid zone (EZ) band defects, photoreceptor loss, area and volume of internal limiting membrane (ILM) puckering, extent of foveal elevation, and mean axial thickness of full, inner, and outer retina. We correlated the features extracted pre-operatively (up to 2 months prior to surgery) and shortly after surgery (up to 2 months) with long-term (6+ months) post-operative visual acuity (VA) in LogMAR scale. Odds ratios were also computed, with VA as a binary outcome (improved/not improved), to quantify the risk of having no post-surgical improvement. An improved outcome was defined as maintaining a gain of 2+ Snellen lines, or better or equal to 20/40 VA.

Results : Of the 72 eyes analyzed, 43 had improved VA outcome. Lower values of full and inner retina thickness and higher values of outer retina thickness were seen in eyes with VA improvement, though differences were not statistically significant. Higher short-term post-operative values for ILM puckering volume and EZ band defect area, and pre-operative inner retina thickness in perifoveal-superior sector yielded the most significant (albeit moderate) correlation with worse long-term post-operative outcomes. A pre-surgical fovea pit elevation increase of 65μm was associated with a 1.68 times increased risk of no long-term VA improvement. A 60.5μm increase in post-surgical full retina thickness in parafoveal-inferior sector was associated with a 1.38 times increased risk of no VA improvement.

Conclusions : Among the pre-surgical features evaluated, higher foveal pit depression and thinner retina had the strongest association with long-term post-surgical improvement, whereas larger ILM puckering and EZ band defects shortly after surgery were most associated with poor long-term visual outcome. However, a larger study is needed to fully evaluate the potential of these features in quantifying benefits versus risks of ERM removal surgery.

This is an abstract that was submitted for the 2016 ARVO Annual Meeting, held in Seattle, Wash., May 1-5, 2016.

×
×

This PDF is available to Subscribers Only

Sign in or purchase a subscription to access this content. ×

You must be signed into an individual account to use this feature.

×