June 2022
Volume 63, Issue 7
Open Access
ARVO Annual Meeting Abstract  |   June 2022
Investigating the Utility of Near Infrared Reflectance (NIR) Imaging for Diabetic Retinopathy Screening
Author Affiliations & Notes
  • Sachi A Patil
    Ophthalmology, New York University, New York, New York, United States
  • Archana Nair
    Ophthalmology, Vanderbilt University Medical Center, Nashville, Tennessee, United States
  • Saagar Pandit
    Ophthalmology, New York University, New York, New York, United States
  • Nitish Mehta
    Ophthalmology, New York University, New York, New York, United States
  • Yasha Modi
    Ophthalmology, New York University, New York, New York, United States
  • Footnotes
    Commercial Relationships   Sachi Patil None; Archana Nair None; Saagar Pandit None; Nitish Mehta None; Yasha Modi Allergen,Alimera, Genentech, Thea, Zeiss, Code C (Consultant/Contractor)
  • Footnotes
    Support  None
Investigative Ophthalmology & Visual Science June 2022, Vol.63, 555 – A0120. doi:
  • Views
  • Share
  • Tools
    • Alerts
      ×
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      Sachi A Patil, Archana Nair, Saagar Pandit, Nitish Mehta, Yasha Modi; Investigating the Utility of Near Infrared Reflectance (NIR) Imaging for Diabetic Retinopathy Screening. Invest. Ophthalmol. Vis. Sci. 2022;63(7):555 – A0120.

      Download citation file:


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

      ×
  • Supplements
Abstract

Purpose : An investigation of the reliability of near infrared reflectance (NIR) imaging as a method of assessing severity and stage of diabetic retinopathy (DR) by quantitative assessment of hyporeflective foci.

Methods : Retrospective cohort study of spectral domain optical coherence tomography (SD-OCT) images from patients with type I and type II diabetes seen at a tertiary academic center. 195 near infrared images were reviewed by 2 independent graders to document the number of hypo reflective foci and presumptive DR staging based on the image. The benchmarked DR staging was made by the retinal specialist in clinic and recorded in the electronic medical record. Interrater reliability was confirmed via one-way random effects model of intraclass correlation coefficients. A third rater was consulted if there were discrepancies in number of hyporeflective spots between the two original raters. Demographic data was collected. Analysis of variance was conducted to estimate differences between sub-groups based on severity. Receiver operating curves were created to validate the reliability of the model.

Results : A statistically significant difference in mean number of hypo-reflective foci was found between none and moderate NPDR (p<0.0001*), none and severe NPDR (p<0.0001*), none and PDR (p<0.0001*), mild and moderate NPDR (p=0.008*), mild and severe NPDR (p<0.001*), mild NPDR and PDR (p<0.001*). A statistically significant difference did not exist between moderate and severe, moderate and proliferative disease, or between severe and proliferative disease. Upon creation of the receiver operating characteristic curve, the area under this curve was 0.849 (CI: 0.792, 0.905). Based on this curve, the ideal threshold for detection of moderate nonproliferative diabetic retinopathy or worse was 4.75 hypo reflective foci, with a sensitivity of 79.0% and a false positive rate of 20.0%.

Conclusions : Near infrared imaging may be a useful adjunct tool in screening for diabetic retinopathy as discrete hyporeflective foci can provide understanding of DR staging. These findings should be established on newly diagnosed DR patients with reading center evaluations of DR staging and hyporeflective foci counts to further establish the validity of NIR imaging as a screening tool.

This abstract was presented at the 2022 ARVO Annual Meeting, held in Denver, CO, May 1-4, 2022, and virtually.

×
×

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.

×